i
Managing Nutrient Cycles to Sustain Soil
Fertility in Sub-Saharan Africa
Managing Nutrient Cycles to
Sustain Soil Fertility in SubSaharan Africa
Edited by
André Bationo
AfNet Coordinator, TSBF-CIAT
Academy Science Publishers (ASP)
in association with the
Tropical Soil Biology and Fertility Institute of CIAT
iv
Academy Science Publishers (ASP)
A Division of the African Academy of Sciences (AAS)
P.O. Box 24916 Nairobi
Tel: 884401-5
Fax: 884406
E-mail: asp@africanonline.co.ke
Website: www.aasciences.org
©AfNet-CIAT All rights reserved, 2004
No Part of this publication may be reproduced in any form or by any
means, electronically, mechanically by photocopying, recording or
otherwise, without the prior permission of the copyright owner.
A catalogue of this book is available from: African Academy of Sciences
ISBN: 9966-24-075-6
Typeset in Nairobi by: Hi-Tech Typesetters, P.O. Box 51709 Nairobi
00100; Tel: 254-020-3742046
Printed and bound in Kenya by: Triscope Consulting Publishers,
P.O. Box 45306 Nairobi 00100; Tel: 254-020-571796/711704
v
Dedication
At the African Network for Soil Biology and Fertility (AfNet) scientific
committee meeting on the 15th December 2002 at Naivasha, Kenya,
the committee members decided to have this book dedicated to Professor
Mike Swift and make him an AfNet life member.
It is with much gratitude that AfNet scientific committee wishes to
dedicate this book to Professor Mike Swift for his many years of
dedicated service to the TSBF Institute. Mike, as we have popularly
referred to him, served TSBF and established AfNet as the single most
important implementing agent of this institute.
Professor Mike Swift has been instrumental in career development
and professional growth of many scientists and academicians on the
African continent, and beyond. He is committed to the concept that the
fertility of tropical soils is controlled by biological processes and can be
managed by the manipulation of these processes.
We all wish Professor Mike Swift and his family a happy retirement
and success in all their future endeavors.
André Bationo
Susan Ikerra
Stephen Kimani
Daniel Mugendi
Martins Odendo
Mary Silver
vi
Preface
Enormous research has been done within the African continent in
various areas including soil organic matter, soil biota, synchrony and
resource integration. All this research is geared towards gaining more
understanding on soil processes which have direct or indirect influence
on soil fertility and land productivity as a whole.
The Tropical Soil Biology and Fertility (TSBF) Institute of CIAT is a
research programme whose main aim is to contribute to human welfare
and environmental conservation in the tropics by developing adoptable
and suitable soil management practices that integrate the biological,
chemical and socio-economic processes that regulate soil fertility and
optimize the use of organic and inorganic resources available to the landusers. The African Network for Soil Biology and Fertility (AfNet) being a
network of scientists in Africa is the single most important implementing
agency of TSBF in Africa. AfNet’s main goal is to strengthen and sustain
stakeholder capacity to generate, share and apply soil fertility management
knowledge and skills to contribute to the welfare of farming communities.
It is a mechanism to facilitate and promote collaboration in research and
development among scientists in Africa for the purpose of developing
innovative and practicable resource management practices for sustainable
food production in the African continent.
AfNet’s overall target outputs are:
1) To exchange information and combine collective experience of
professionals in the same field;
2) To achieve economies of scale and efficiency by concentrating scarce
human, financial and other resources on key national and regional
problems;
3) To carry out collaborative research through network experiments;
4) To minimize duplication;
5) To provide increased bargaining power with external partners; and
6) To undertake joint capacity building.
In order to enhance these objectives of collaborative research, the
network members were offered the opportunity to participate in a
conference that brought together all partners and stakeholders to share,
exchange and publish results emanating from their research activities
in soil biology and fertility in Africa. This book on Managing Nutrient
Cycle to Sustain Soil Fertility in sub-Saharan Africa is a synthesis of
AfNet member research results of the past few years.
Preface
vii
Soil fertility degradation still remains the single most important
constraint to food production in sub-Saharan Africa and an efficient
cycling of nutrients among crops, animals and soil is crucial to the
sustained productivity of the farming systems. Emerging evidence
indicate that there is considerable consensus on guiding principles for
integrated soil fertility management (ISFM) as the more pragmatic and
feasible approach to overcome the limitations of past research
approaches. As a holistic approach to research on soil fertility, ISFM
embraces responses to the full range of driving factors and consequences
namely biological, physical, chemical, social, economic and political
aspects of soil fertility decline. The approach encompasses nutrient
deficiencies, inappropriate germplasm and cropping system design, pestdisease interaction with soil fertility, linkage between land degradation
and poverty and global policies, incentives as well as institutional failures.
Such long-term soil fertility management strategy requires an
evolutionary, knowledge intensive process, participatory research and
development focus rather than a purely technical focus.
After the introduction in chapter 1 on new challenges and
opportunities of AfNet, this book is divided in three broad parts. Part
one ranges from chapter 2 to chapter 28 and deals with the issues on
integrated soil fertility management. The second part is from chapter
29 to chapter 34 and is on belowground biodiversity. Part three, from
chapter 35 to chapter 42 is on participatory research and scaling up of
soil fertility restoration technologies.
AfNet recently published a book on “Soil Fertility Management in
Africa: A Regional Perspective”. AfNet also intends to publish another
book on “Fighting Poverty in Sub-Saharan Africa: The Multiple Roles
of Legumes in Integrated Soil Fertility Management”.
We are grateful to the Rockefeller Foundation for their continual
support to AfNet and particularly for financial support towards the
successful organization of the conference leading to the publication of
this book. AfNet also wishes to acknowledge the financial support from
Regional Land Management Unit (RELMA) in the publishing this book.
André Bationo
The African Network for Soil Biology and Fertility (AfNet) of
the Tropical Soil Biology and Fertility (TSBF) Institute of CIAT
c/o ICRAF, United Nations Avenue, Gigiri
P.O. Box 30677 Nairobi, Kenya
Tel: (254)-2-524755/6; Fax: (254)-2-524763/4
The Tropical Soil Biology and Fertility (TSBF) Institute of CIAT is a
research programme whose main aim is to contribute to human welfare
and environmental conservation in the tropics by developing adoptable
and suitable soil management practices that integrate the biological,
chemical and socio-economic processes that regulate soil fertility and
optimize the use of organic and inorganic resources available to the
land users. TSBF research basically targets the empowerment of farmers
so as to effectively (i) manage nutrient cycles; (ii) manage below ground
biodiversity and (iii) manage ecosystem services, so as to achieve the
necessarry sustainable Agro-ecosystem management.
The African Network for Soil Biology and Fertility (AfNet) is the single
most important implementing agency of TSBF in Africa. Its main goal is
to strengthen and sustain stakeholder capacity to generate, share and
apply soil fertility and biology management knowledge and skills to
contribute to the welfare of farming communities. It is a mechanism to
facilitate and promote collaboration in research and development among
scientists in Africa for the purpose of developing innovative and practical
resources management interventions for sustainable food production.
AfNet has membership from National Agricultural Research and
Extension Services (NARES) and universities from various disciplines
mainly soil science, social science, agronomy and technology exchange.
Contents
Preface
.................................................................................. vi
Contributors ............................................................................... xiv
Chapter 1:
The African Network for Soil Biology and Fertility:
New Challenges and Opportunities
Bationo, A., Kimetu, J., Ikerra, S., Kimani,
S., Mugendi, D., Odendo, M., Silver, M.,
Swift, M.J. and Sanginga, N. .......................................... 1
PART I
INTEGRATED SOIL FERTILITY MANAGEMENT
Chapter 2:
Integrated Soil Fertility Management Research
at TSBF: The Framework, the Principles, and their
Application
Vanlauwe B. ................................................................. 25
Chapter 3:
Guidelines for Integration of Legumes into the
Farming Systems of East African Highlands
Amede T. and Kirkby R. .............................................. 43
Chapter 4:
Effect of Organic and Inorganic Nutrient Sources
on Soil Mineral Nitrogen and Maize Yields in Western
Kenya
Ayuke F.O, Rao M.R., Swift M.J. and
Opondo-Mbai M.L. ........................................................ 65
Chapter 5:
Long Term Effects of Mineral Fertilisers, Phosphate
Rock, Dolomite and Manure on the Characteristics
of an Ultisol and Maize Yield in Burkina Faso
Bado, B.V., Sedogo, M.P. and Lompo, F. ....................... 77
Chapter 6:
Changes in Soil Properties and their Effects on
Maize Productivity Following Sesbania Sesban
and Cajanus Cajan Improved Fallow Systems in
Eastern Zambia
Chirwa T. S., Mafongoya P. L., Mbewe D.N.M.
and Chishala B. H........................................................ 89
Contents
Chapter 7:
Tillage Effects on Soil Organic Carbon and
Nitrogen Distribution in Particle Size Fractions
of a Red Clayey Soil Profile in Zimbabwe
Chivenge P.P., Murwira H.K. and
Giller K.E. ................................................................... 113
Chapter 8:
Combating Nutrient Depletion in East Africa –
the work of the SWNM program
Delve R.J. ................................................................... 127
Chapter 9:
Effects of Farmyard Manure, Potassium and their
Combinations on Maize Yields in the High and
Medium rainfall Areas of Kenya
Gikonyo E.W. and Smithson P.C................................. 137
Chapter 10: Effects of Nitrogen and Phosphorus Fertilizer
Addition on Wheat Straw Carbon Decomposition
in a Burundi Acidic Soil
Kaboneka S., Nivyiza J.C. and Sibomana L............... 151
Chapter 11: Evaluation of Crop Availability of K and Mg in
Organic Materials under Greenhouse Conditions
Kaboneka S. and Sabbe W.E. .................................... 163
Chapter 12: The Influence of Goat Manure Application on
Crop Yield and Soil Nitrate Variations in SemiArid Eastern Kenya
Kihanda F.M., Warren G.P. and Atwal S.S. ................ 173
Chapter 13: Managing Manures Throughout their Production
Cycle Enhances their Usefulness as Fertilisers: A Review
Kimani S.K. and Lekasi J.K. ...................................... 187
Chapter 14: Simulated Partitioning Coefficients for Manure
Quality Compared With Measured C:N Ratio Effects
Kimani S.K., Gachengo C. and Delve R...................... 199
Chapter 15: Nitrogen Fertilizer Equivalency Values for Different
Organic Materials Based on Maize Performance at
Kabete, Kenya
Kimetu J.M., Mugendi D.N., Palm C.A., Mutuo P.K.,
Gachengo C.N., Nandwa S. and Kungu J.B............... 207
Chapter 16: Base Nutrient Dynamics and Productivity of Sandy
Soils Under Maize-Pigeonpea Rotational Systems in
Zimbabwe
Mapfumo P. and Mtambanengwe F. ........................... 225
Chapter 17: Soil Organic Matter (SOM): The Basis for Improved
Contents
Crop Production in Arid and Semi-Arid Climates of
Eastern Kenya
Micheni A., Kihanda F. and
Irungu J. ..................................................................... 239
Chapter 18: Response of Tephrosia vogelii to Minjingu
Phosphate Rock Application on a Ferralsol of
Varying Soil pH
Mkangwa C.Z., Semoka J.M.R. and
Maliondo S.M.S. ......................................................... 249
Chapter 19: Decomposition of Organic Matter in Soil as
Influenced by Texture and Pore Size Distribution
Mtambanengwe F. , Mapfumo P. and Kirchmann H. .. 261
Chapter 20: Soil Conservation and Fertility Improvement Using
Leguminous Shrubs in Central Highlands of Kenya:
NARFP Case Study
Mugwe J., Mugendi D., Okoba B.,
Tuwei P. and O’Neill M. .............................................. 277
Chatper 21: The Relationship Between Nitrogen Mineralization
Patterns and Quality Indices of Cattle Manures
from Different Smallholder Farms in Zimbabwe
Nhamo N., Murwira H.K., Giller K.E. .......................... 299
Chapter 22: Effect of Cattle Manure and N Fertiliser on Nitrate
Leaching Losses in Smallholder Maize Production
Systems of Zimbabwe Measured in Field Lysimeters
Nyamangara J. and Bergström L.F. ........................... 317
Chapter 23: Combined use of Tithonia diversifolia and Inorganic
Fertilizers for Improving Maize Production in a
Phosphorus Deficient soil in Western Kenya
Nziguheba G., Merckx R., Palm C.A.
and Mutuo P.K. ........................................................... 329
Chapter 24: Effect of Combining Organic and Inorganic
Phosphorus Sources on Maize Grain Yield in
a humic-Nitosol in Western Kenya
Ojiem J. O., Palm C. A., Okwuosa E. A. and Mudeheri
M.A. ............................................................................ 247
Chapter 25: Use of Organic and Inorganic Resources to
Increase Maize Yields in some Kenyan Infertile
Soils: A Five-Year Experience
Okalebo J.R., Palm C.A., Lekasi J.K., Nandwa S.M.,
Othieno C.O., Waigwa M. and Ndungu K.W. .............. 359
Contents
Chapter 26: The Potential of Green Manures to Increase Soil
Fertility and Maize Yields in Malawi
Sakala W.D., Kumwenda J.D.T. and Saka A.R.......... 373
Chapter 27: Effects of Ramial Chipped Wood and Litter Compost
of Casuarina Equisetifolia Tomato Growth and Soil
Properties in Niayes, Senegal
Soumare M. D., Mnkeni P.N.S. and Khouma M. ......... 385
Chapter 28: The Use of Pigeon Pea (Cajanus cajan) for
Amelioration of Ultisols in Ghana
Yeboah E., Fening J.O. and Ampontuah E.O.............. 401
PART II
BELOWGROUND BIODIVERSITY
Chapter 29: Assessment of Biomass Transfer from Green
Manure to Soil Macrofauna in Agroecosystem-Soil
Macrofauna Biomass
Ayuke F.O., Rao M.R., Swift M.J. and
Opondo-Mbai M.L. ...................................................... 411
Chapter 30: Dual Inoculation of Woody Legumes and
Phosphorus Uptake from Insoluble Phosphate Rock
Kimiti J.M. and Smithson P.C. .................................... 423
Chapter 31: Effect of Vesicular-arbuscular Mycorrhiza (vam)
Inoculation on Growth Performance of Senna
Spectabilis
Kung'u J.B.................................................................. 433
Chapter 32: Soil Invertebrate Macrofauna Composition within
Agroforestry and Forested Ecosystems and their
Role in Litter Decomposition in Embu, Kenya
Mwangi M., Mugendi D. N., Kung’u J.B., Swift M.J.,
and Albrecht A. .......................................................... 447
Chapter 33: Selection of Arbuscular Mycorrhizal Fungi for
Inoculating Maize and Sorghum Grown in
Oxisol/Ultisol and Vertisol in Cameroon
Nwaga D., The C., Ambassa-Kiki R., Ngonkeu
Mangaptché E.L. and Tchiegang-Megueni C..............467
Chapter 34: Macrofaunal Abundance and Diversity in
Selected Farmer Perceived Soil Fertility Niches
in Western Kenya
Tabu I.M., Obura R.K. and Swift M.J. ........................ 487
Contents
PART III
PARTICIPATORY RESEARCH AND SCALING UP OF
SOIL FERTILITY RESTORATION TECHNOLOGIES
Chapter 35: Understanding Soil in its Social Context:
Integrating Social and Natural Science Research
within AFNET
Ramisch J.J................................................................ 501
Chapter 36: Linking Research Results with Rural Development
Projects: Experiences from Southern Africa
Murwira H.K............................................................... 523
Chapter 37: Economic Analysis of Non-Conventional Fertilizers
in Vihiga District, Western Kenya
Kipsat M.J., Maritim H.K., and Okalebo J.R. ............ 535
Chapter 38: Early farmer evaluation of Integrated Nutrient
Management Technologies in Eastern Uganda
Miiro R., Kabuye F., Jama B.A., Musenero E.,
Zake J.Y.K., Nkwiine C., Kakinda M.J., Onyango O.,
and Delve R.J. ............................................................ 545
Chapter 39: Potential for Adoption of Legume Green Manure
on Smallholder Farms in Western Kenya
Odendo M., Ojiem J. and Okwosa E .......................... 557
Chapter 40: The Profitability of Manure Use on Maize in the
Small-holder Sector of Zimbabwe
Mutiro K. and Murwira H. K. ...................................... 571
Chapter 41: Improved Food Production by Use of Soil Fertility
Amendment Strategies in The Central Highlands of
Kenya
Mucheru M., Mugendi D.N., Micheni A.,
Mugwe J., Kung'u J.B., Otor S.
and Gitari J. ............................................................... 583
Chapter 42: Impact of Adopting Soil Conservation Practices
on Wheat Yield in Lesotho
Kaliba A.R.M. and Rabele T. ...................................... 593
xiv
Contributors
Albrecht, A., International Centre for Research in Agroforestry (ICRAF),
P.O. Box 30677, Nairobi, Kenya
Ambassa-Kiki, R., Institute of Research for Agriculture & Development,
P.O. Box 2123, Yaoundé-Messa, Cameroon
Amede, T., Research Fellow, Tropical Soils Biology & Fertility Institute
of CIAT/Africa Highlands Initiative, P.O. Box 1412, code 1110, Addis
Ababa, Ethiopia, t.amede@cgiar.org
Ampontuah, E. O., Soil Research Institute Academy Post Office,
Kwadaso, Kumasi, Ghana
Atwal, S. S., Department of Soil Science, University of Reading, P.O.
Box 233, READING, RG6 6DW, U.K.
Ayuke, F. O., Department of Forestry, Moi University, P.O. Box 1125,
Eldoret, Kenya; Tropical Soil Biology and Fertility Programme,
Institute of CIAT, P.O. Box 30677, Nairobi, Kenya, Email:
Fayuke2002@yahoo.co.uk; Fayuke@cgiar.org
Bado, B. V., INERA, BP 910 Bobo-Dioulasso, Burkina Faso
Bationo, A., TSBF -CIAT c/o ICRAF, P.O. Box 30677, Nairobi, Kenya
Bergström, L .F., Division of Water Quality Management, Department
of Soil Sciences, Swedish University of Agricultural Sciences, P.O.
Box
7072,
S-750
07
Uppsala,
Sweden,
E-mail:
Lars.Bergstrom@mv.slu.se Phone: 46-18-67 10 00; Fax: 46-18-67
34 30
Chirwa, T. S., Msekera Research Station, P.O. Box 510089, Chipata,
Zambia
Chishala, B. H., University of Zambia, School of Agricultural Sciences,
P.O. Box 32379, Lusaka, Zambia
Chivenge, P. P., Tropical Soil Biology and Fertility (TSBF), P.O. Box MP
228, Mt Pleasant, Harare, Zimbabwe
xiv
Contributors
xv
Delve, R. J., Tropical Soil Biology and Fertility Programme (TSBF), UN
Complex Gigiri, P.O. Box 30592, Nairobi, Kenya; TSBF Institute of
CIAT, c/o CIAT-Uganda, P.O. Box 6247, Kawanda, Kampala, Uganda
Fening, J. O., Soil Research Institute Academy Post Office, Kwadaso,
Kumasi, Ghana
Gachengo, C., Tropical Soil Biology and Fertility Programme (TSBF),
UN Complex Gigiri, P.O. Box 30592, Nairobi, Kenya
Gikonyo, E. W., Kenya Agricultural Research Institute, P.O. Box 14733,
Nairobi, Kenya, E-mail: est.gikonyo@cgiar.org, Fax: 444144
Giller, K. E., Soil Science and Agricultural Engineering Department,
Faculty of Agriculture, University of Zimbabwe, Box MP167, Mt
Pleasant, Harare, Zimbabwe
Gitari, J., Kenya Agricultural Research Institute, P.O. Box 27, Embu,
Kenya
Ikerra, S., Department of Soil Science, Sokoine University of Agriculture,
P.O. Box 3008, Morogoro, Tanzania
Irungu, J., Kenya Agricultural Research Institute, P.O. Box 27 Embu,
Kenya; Tel: 068 20116/20873; Fax: 068 30064; E-mail:
kariembu@salpha.co.ke
Jama, B. A., Bashir Jama, International Centre for Research in
Agroforestry (ICRAF), Nairobi, Kenya, b.jama@cgiar.org
Kaboneka, S., Institut des Sciences Agronomiques du Burundi (ISABU).
B.P. 795, Bujumbura, Burundi. Université du Burundi, Faculté des
Sciences Agronomiques. B.P. 2900, Bujumbura, Burundi
Kabuye, F., Africa 2000 Network, P.O. Box 7184, Kampala, Uganda,
anetwork@imul.com
Kakinda, M. J., Africa 2000 Network, P.O. Box 7184, Kampala, Uganda,
anetwork@imul.com
Kaliba, A. R. M., Formerly, Lecturer and Undergraduate Student,
Department of Economics, National University of Lesotho, P.O. Roma,
180, Lesotho, Tel: 266 12 3593; Fax: 266 12 340000, E-mail:
akaliba@uaex.edu
Khouma, M., National Agricultural Research Center, ISRA/CNRA, B.
P. 53 Bambey, Senegal. E-mail: mkhouma@isra.sn
Kihanda, F., Kenya Agricultural Research Institute, P.O. Box 27 Embu,
Kenya; Tel: 068 20116/20873; Fax: 068 30064; E-mail:
kariembu@salpha.co.ke
xvi
Contributors
Kimani, S. K., Kenya Agriculture Research Institute, NARC Muguga
P.O. Box 30148 Nairobi; Email: skimani@net2000ke.com
Kimetu, J. M., TSBF -CIAT c/o ICRAF, P.O. Box 30677, Nairobi, Kenya
Kimiti, J. M., Kenya Forestry Research Institute, P.O. Box 20412,
Nairobi, Kenya
Kipsat, M. J., Department of Agricultural Resource Economics, Moi
University, P.O. Box 1125, Eldoret, Kenya
Kirchmann, H., Swedish University of Agricultural Sciences,
Department of Soil Science, Box 7014, Uppsala, 750 07, Sweden
Kirkby, R., CIAT, Pan-African Coordinator, P.O. Box 6247, Kampala,
Uganda. ciat-africa @cgiar.org
Kumwenda, J.D.T., Chitedze Agricultural Research Station, P.O. Box
158, Lilongwe, Malawi
Kung'u, J. B., School of Pure and Applied Sciences, Kenyatta University,
P.O. Box 43844 Nairobi Kenya; E-mail: kungu@avu.org
Lekasi, J. K., National Agricultural Research Centre, Muguga Kenya
Agricultural Research Institute (KARI), P.O. Box 30148, Nairobi,
Kenya
Lompo, F., INERA, 01BP 476 Ouagadougou 01, Burkina Faso
Mafongoya, P. L., ICRAF-Zambia Agroforestry Project, P.O. Box 510046,
Chipata, Zambia
Maliondo, S. M. S., Forest Biology Department, P.O. Box 3000, Sokoine
University of Agriculture, Tanzania
Mapfumo, P., Department of Soil Science and Agricultural Engineering,
University of Zimbabwe, P.O. Box MP 167, Mount Pleasant, Harare,
Zimbabwe; E-mail: pmapfumo@agric.uz.ac.zw
Maritim, H. K., Department of Agricultural Resource Economics, Moi
University, P.O. Box 1125, Eldoret, Kenya
Mbewe, D. N. M., University of Zambia, School of Agricultural Sciences,
P.O. Box 32379, Lusaka, Zambia
Merckx, R., Laboratory of Soil Fertility and Soil Biology, K.U. Leuven,
Kasteelpark Arenberg 20, B 3001 Heverlee, Belgium
Micheni, A., Kenya Agricultural Research Institute, P.O. Box 27 Embu,
Kenya; Tel: 068 20116/20873; Fax: 068 30064; E-mail:
kariembu@salpha.co.ke
Miiro, R., Department of Agricultural Extension/Education, Faculty of
Agriculture, Makerere University, P.O. Box 7062, Kampala, Uganda,
rfmiiro@yahoo.com; rfmiiro@agric.mak.ac.ug
Contributors
xvii
Mkangwa, C. Z., Soil Science Department, P.O. Box 3008, Sokoine
University of Agriculture, Tanzania
Mnkeni, P. N. S., University of Fort Hare, Faculty of Agricultural and
Environmental Sciences, Department of Agronomy, Private Bag
X1314, Alice 5700, South Africa, Tel: +27 40 602 2139 Fax: +27 40
653 1730; E mail: pmnkeni@ufh.ac.za
Mtambanengwe, F., Department of Soil Science and Agricultural
Engineering, University of Zimbabwe, Box MP 167, Mt Pleasant,
Harare, Zimbabwe
Mucheru, M., Kenyatta University, Faculty of Environmental Studies,
P.O. Box 43844, Nairobi, Kenya Email:moniquechiku@yahoo.com
Mudeheri, M. A., Kenya Agricultural Research Institute, Regional
Research Centre, P.O. Box 169, Kakamega, Kenya; E-mail:
hi@swiftkisumu.com
Mugendi, D. N., Senior Lecturer, Faculty of Environmental Studies
Kenyatta University. P.O. Box Box 43844 Ext 216/214 Nairobi, Kenya
Tel: 2-811622/812722. E-Mail: dmugendi@yahoo.com
Mugwe, J., Kenya Forestry Research Institute, P.O. Box 20412, Nairobi,
Kenya
Murwira, H. K., Tropical Soil Biology and Fertility (TSBF), P.O. Box
MP288, Mount Pleasant, Harare, Zimbabwe, Email:
hmurwira@zambezi.net
Musenero, E., Department of Production, Tororo District, Uganda
Mutiro, K., Tropical Soil Biology and Fertility (TSBF), P.O. Box MP288,
Mount Pleasant, Harare, Zimbabwe, Email: kvmutiro@zambezi.net
Mutuo, P. K., Tropical Soil Biology and Fertility Programme (TSBF),
P.O. Box 30592, Nairobi, Kenya
Mwangi, M., International Centre for Research in Agroforestry (ICRAF),
P.O. Box 30677, Nairobi, Kenya; Kenyatta University, Department
of Environmental Foundations, P.O. Box 43844, Nairobi, Kenya
Nandwa, S. M., National Agricultural Research Laboratories, KARI, P.O.
Box 14733, Nairobi, Kenya
Ndungu, K. W., Department of Soil Science, Moi University, Chepkoilel
Campus, P.O. Box 1125, Eldoret, Kenya
Ngonkeu Mangaptché, E. L., Biotechnology Centre & Plant Biology
Department, University of Yaoundé I, Cameroon, P.O. Box 812,
Yaoundé, Cameroon; E-mail: dnwaga@uycdc.uninet.cm
Nhamo, N., Soil Productivity Research Laboratory (SPRL), Private Bag
3757, Marondera, Zimbabwe, Telephone: 263-79-23621 or 26311614367; Fax: 263-79-24279; E-mail: nnsprl@mweb.co.zw
xviii
Contributors
Nivyiza, J.C., Université du Burundi. Faculté des Sciences
Agronomiques. B.P. 2940. Bujumbura, Burundi. Fax: 257-22-2500.
E-mail: facagro@cni.cbinf.com
Nkwiine, C., Makerere University, Department of Soil Science, Faculty
of Agriculture, P.O. Box 7062, Kampala, Uganda,
acss@starcom.co.ug or plectumu@imul.com
Nwaga, D., Biotechnology Centre & Plant Biology Department, University
of Yaoundé I, Cameroon, P.O. Box 812, Yaoundé, Cameroon; E-mail:
dnwaga@uycdc.uninet.cm
Nyamangara, J., Department of Soil Science and Agricultural
Engineering, University of Zimbabwe, P.O. Box MP167, Mount
Pleasant, Harare, Zimbabwe. Email: jnyamangara@agric.uz.ac.zw
Phone: 263-4-30 32 11; Fax: 263-4-33 28 53
Nziguheba, G., Laboratory of Soil Fertility and Soil Biology, K.U. Leuven,
Kasteelpark Arenberg 20, B 3001 Heverlee, Belgium
O’Neill, M, International Centre for Research in Agroforestry (ICRAF);
KARI-Regional Research Centre, Embu, Kenya. P.O. Box 27, Embu;
E-mail: icraf-embu@cgiar.org; presently: Asst. Prof. and
Superintendent, New Mexico State University, Agricultural Science
Center, P.O. Box 1018, Farmington, NM, USA, 87499; Email:
moneill@nmsu.edu
Obura, R. K., Department of Agronomy, Egerton Univesity, Box 536
Njoro, Kenya
Odendo, M., Kenya Agricultural Research Institute, Regional Research
Centre, P.O. Box 169, Kakamega, Kenya
Ojiem, J. O., Kenya Agricultural Research Institute, Regional Research
Centre, P.O. Box 169, Kakamega, Kenya; E-mail:
hi@swiftkisumu.com
Okalebo, J. R., Department of Soil Science, Moi University, Chepkoilel
Campus, P.O. Box 1125, Eldoret, Kenya
Okoba, B., Soil Engineer, Kenya Agricultural Research Institute (KARI).
National Agroforestry Research Project; KARI-Regional Research
Centre, Embu, Kenya. P.O. Box 27, Embu. Tel: +254-161-20116/
20873 Fax: +254-161-30064 E-mail: kariembu@salpha.co.ke
Okwuosa, E. A., Kenya Agricultural Research Institute, Regional
Research Centre, P.O. Box 169, Kakamega, Kenya; E-mail:
hi@swiftkisumu.com
Onyango, O., Tororo District Local Government, Tororo, Uganda
Contributors
xix
Opondo-Mbai, M. L., Department of Forestry, Moi University, P.O. Box
1125, Eldoret, Kenya
Othieno, C. O., Department of Soil Science, Moi University, Chepkoilel
Campus, P.O. Box 1125, Eldoret, Kenya
Otor, S., Kenyatta University, Faculty of Environmental Studies, P.O.
Box 43844, Nairobi, Kenya
Palm, C. A., Tropical Soil Biology and Fertility Programme, TSBF/
UNESCO, UN Complex, P.O. Box 30592, Nairobi, Kenya
Rabele, T., Formerly, Lecturer and Undergraduate Student, Department
of Economics, National University of Lesotho, P.O. Roma, 180,
Lesotho, Tel: 266 12 3593; Fax: 266 12 340000, E-mail:
trabele@yahoo.com
Ramisch, J. J., Social Science Officer, TSBF, P.O. Box 30677 Nairobi,
Kenya j.ramisch@cgiar.org
Rao, M. R., International Centre for Research in Agroforestry, P.O. Box
30677, Nairobi, Kenya; ICRISAT Colony (plot II, phase 1),
Secunderabad-500 009, Andhra Pradesh, India
Sabbe, W. E., Professor of Agronomy (Deceased). Department of
Agronomy, University of Arkansas, PTSC 115, Fayetteville, AR 72701
Saka, A. R., Chitedze Agricultural Research Station, P.O. Box 158,
Lilongwe, Malawi
Sakala, W. D., Chitedze Agricultural Research Station, P.O. Box 158,
Lilongwe, Malawi
Sanginga, N., TSBF -CIAT c/o ICRAF, P.O. Box 30677, Nairobi, Kenya
Sedogo, M. P., INERA, 01BP 476 Ouagadougou 01, Burkina Faso
Semoka, J. M. R., Soil Science Department, P.O. Box 3008, Sokoine
University of Agriculture, Tanzania
Sibomana, L., Université du Burundi. Faculté des Sciences
Agronomiques. B.P. 2940. Bujumbura, Burundi. Fax: 257-22-2500.
E-mail: facagro@cni.cbinf.com
Silver, M., Department of Soil Science, Makerere University, P.O. Box
7062, Kampala, Uganda
Smithson, P. C., International Centre for Research in Agroforestry, P.O.
Box 30677, Nairobi, Kenya, E-mail: psmithson@cgiar.org
Soumare, M. D., B P: 22087, Dakar – Ponty, Dakar, Senegal
Swift, M. J., Tropical Soil Biology and Fertility Programme, Institute of
CIAT, P.O. Box 30677, Nairobi, Kenya
xx
Contributors
Tabu, I. M., Department of Agronomy, Egerton Univesity, Box 536 Njoro,
Kenya
Tchiegang-Megueni, C., Department of Biological Sciences, University
of Ngaoundéré, P.O. Box 455, Ngaoundéré, Cameroon
The, C., Institute of Research for Agriculture & Development, P.O. Box
2123, Yaoundé-Messa, Cameroon
Tuwei, P., Senior Lecturer, Faculty of Environmental Studies Kenyatta
University. P.O. Box Box 43844 Ext 216/214 Nairobi, Kenya Tel: 2811622/812722. E-Mail: dmugendi@yahoo.com
Vanlauwe, B., Scientific Officer, Tropical Soil Biology and Fertility
Institute of CIAT, P.O. Box 30677, Nairobi, Kenya
Waigwa, M., Department of Soil Science, Moi University, Chepkoilel
Campus, P.O. Box 1125, Eldoret, Kenya
Warren, G. P., Department of Soil Science, University of Reading, P.O.
Box 233, READING, RG6 6DW, U.K.
Yeboah, E., Soil Research Institute Academy Post Office, Kwadaso,
Kumasi, Ghana
Zake, J. Y. K., Makerere University, Department of Soil Science, Faculty
of Agriculture, P.O. Box 7062, Kampala, Uganda,
acss@starcom.co.ug or plectumu@imul.com
xxi
PART I
INTEGRATED SOIL
FERTILITY
MANAGEMENT
PART II
BELOWGROUND
BIODIVERSITY
PART III
PARTICIPATORY
RESEARCH AND
SCALING UP OF SOIL
FERTILITY
RESTORATION
TECHNOLOGIES
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
The African Network for Soil
Biology and Fertility: New
Challenges and
Opportunities
1
1
Bationo, A.1, Kimetu, J.1, Ikerra, S.2,
Kimani, S.3, Mugendi, D.4, Odendo, M.5,
Silver, M.6, Swift, M.J.1 and Sanginga, N.1
TSBF -CIAT c/o ICRAF, P.O. Box 30677, Nairobi, Kenya
Department of Soil Science, Sokoine University of
Agriculture, P.O. Box 3008, Morogoro, Tanzania
3
Kenya Agricultural Research Institute, P.O. Box 30148,
Nairobi, Kenya
4
School of Pure and Applied Sciences, Kenyatta University,
P.O. Box 43844, Nairobi, Kenya
5
Kenya Agricultural Research Institute, Regional Research
Centre, P.O. Box 169, Kakamega, Kenya
6
Department of Soil Science, Makerere University, P.O. Box
7062, Kampala, Uganda
1
2
Abstract
Soil fertility degradation has been described as the single most
important constraint to food security in sub-Saharan Africa
(SSA). Soil fertility decline is not just a problem of nutrient
deficiency but also of 1) Inappropriate germplasm and cropping
system design, 2) Interactions with pests and diseases, 3) The
linkage between poverty and land degradation, 4) Often perverse
Bationo, A. et al
2
national and global policies with respect to incentives, and 5)
Institutional failures. Tackling soil fertility issues thus requires
a long-term perspective and a holistic approach. The African
Network for Soil Biology and Fertility (AfNet) of Tropical Soil
Biology and Fertility institute of CIAT whose ultimate goal is to
strengthen and sustain stakeholder capacity to generate, share
and apply soil fertility management knowledge and skills to
contribute to the welfare of farming communities is devoted to
overcoming this challenge. This African-wide network has over
200 members from National Agricultural Research and
Extension Services (NARES) and universities from various
disciplines mainly soil science, social science and technology
exchange. This paper is an highlight of AfNet’s main activities
which include: Network field research activities, information
and documentation, training and capacity building.
Introduction
Africa has 340 million people, over a half of its population living on less
than USD 1 per day, a mortality rate of children under 5 years of age of
140 per 1000 and life expectancy of only 54 years. The latest figures
show that some 200 million people, or 28% of Africa population are
chronically hungry. The average African consumes only about 87% of
the calories needed for a healthy and productive life. At present, over
USD 18 billion is spent annually on food imports and in the year 2000,
Africa received 2.8 million tons of food aid, a quarter of the world’s
total. Over half of the African population is rural, and directly dependent
on locally grown crops or foods harvested from the immediate
environment. Macro-policy changes imposed externally in the last
decade, such as structural adjustment and the removal of fertilizer
subsidies, were executed without any clear understanding of the likely
consequences at a micro-level and hidden effect on continued erosion
of the natural resource base. Structural adjustment policies resulted in
the reduction of the use of external inputs, extensification of agriculture
through the opening of new lands and the reduction of the farmers’
potential for investment in soil fertility restoration. Technological,
environmental, socio-cultural, economic, institutional and policy
constraints have been identified to hamper agricultural development in
Africa. These constraints are: (i) low soil fertility (ii) fragile ecosystems
(iii) over dependence on rainfall (iv) aging rural population and thus
limited physical energies for production (v) underdeveloped and degraded
rural infrastructure (vi) insufficient research due to lack of motivation
and inadequate facilities (vii) inadequate training and extension services
(viii) high post harvest losses (ix) insufficient market (x) lack of credit
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
3
and insufficient agri-input delivery systems (xi) limited farmers’
education and know-how (xii) continental brain-drain of African
intellectuals (xiii) policy instability (xiv) inconsistent agricultural policies
and efficient land tenure. This led to the New Partnership for Africa’s
Development (NEPAD) to recognize that agriculture-led development is
fundamental to cutting hunger, reducing poverty, generating economic
growth, reducing burden of food imports and opening the way to an
expansion of exports. Per capita food production in Africa has been
declining over the past two decades, contrary to the global trend. The
result is widespread malnutrition, a recurrent need for emergency food
and an increasing dependence on food grown outside the region. The
average annual increase of cereal yield in Africa is about 10 kg/ha, the
rate known as the one for extensive agriculture neglecting external inputs
like improved seeds and plant nutrients. The growth rate for cereal
grain yield is about 1% while population growth will be about 3%. During
the last 35 years, cereals production per capita has decreased from 150
to 130 kg/person, whereas in Asia and Latin America an increase from
about 200 to 250 kg/person have been observed. Both labor and land
productivity are among the lowest of the world. The Forum for
Agricultural Research in Africa (FARA) with its member sub-regional
organizations (SRO) has developed a vision for African Agricultural
Research, which calls for 6% annual growth in agricultural productivity.
Land degradation is one of the most serious threats to food production in
the continent. The population is thus trapped in a vicious poverty cycle between
land degradation, and the lack of resources or knowledge to generate adequate
income and opportunities to overcome the degradation and it is urgent to
invest to combat land degradation to revert this vicious circle (Figure 1.1).
Figure 1.1: Combating land degradation to improve rural livelihoods
Bationo, A. et al
4
Scientists have reported that soil loss through erosion is about 10
times greater than the rate of natural formation, while the rate of
deforestation is 30 times higher than that of planned reforestation.
Although large areas of forests, wetlands, river valley bottoms and
grassland savanna have been put under food crops, the food gap
(requirements minus production) keeps widening. Soil nutrient
depletion is a major bottleneck to increased productivity in Africa and
has largely contributed to poverty and food insecurity. Soil nutrient
depletion occurs when nutrient inflows are less than outflows. Nutrient
balances for many cropping systems are negative indicating that
farmers are mining their soil. The data in Figure 1.2 clearly illustrate
the level of nutrient mining in African agro-ecosystems. For nitrogen
as an example, whereas 4.4 million tons is lost per year, only 0.8 million
tons is applied.
Figure 1.2: Macronutrient application versus loss in Africa
123
123
123 Loss
5.0
4.5
Million tons per year
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
4.4
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678 0.8
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
N
Applied
0.5
123456789
123456789
123456789 0.3
123456789
123456789
123456789
p
3.0
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789
123456789 0.2
123456789
123456789
123456789
k
Nutrients
Source: Sanchez et al. (1997)
The different biophysical, chemical and socio-economic factors
contributing to low soil fertility and poor productivity are reported in
Figure 1.3.
At present, fertilizer use in Africa is about 9 kg ha-1 as compared to
87 kg ha-1 for the developed countries (Table 1.1). With 9% of the world’s
population, SSA account for less than 1.8% of global fertilizer use and
less than 0.1% of global fertilizer production.
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
5
Table 1.1: Population, cropped land and fertilizer use (1961-97) in some African countries
as compared to some developed ones
1961
1997
Pop.
Crop land
Fertilizer use
Pop.
Crop land
Fertilizer use
(Million)
(Million ha)
(Kg ha-1)
(Million)
(Million ha)
(Kg ha-1)
World
3136
1352.0
23.0
5823
1501.0
90
Dev.
987
654
42
1294
640.0
87
S.S Africa
219
120
0.15
578
154
9
D.R. Congo
16
7.0
0.04
48
7.9
0.8
Kenya
9
28.8
2.8
28
4.5
29
Nigeria
38
0.6
0.5
104
30.7
4.5
Egypt
29
2.6
93
65
3.2
313
France
46
21.4
113
58
19.5
260
India
452
160.9
21
966
169.8
95
USA
189
182.5
41
272
177.0
114
countries
Source: FAO 1999
Figure 1.3: Biophysical, chemical and socio-economic factors contributing to low soil
fertility and poor productivity in Sub-Saharan Africa
• Low CEC
• Low organic
matter
• Low WHC
• Unfavourable pH
• Nutrients
toxities
• Leaching
• Nutrients mining
• Nutrients fixation
Weakened
ability to
maintain
fertility
Low inherent
fertility
Low
soil
fertility
Nutrient
depletion
Source: Murwira, 2003
Low returns
on
investment in
raised fertility
• Traditional strategies
undermined or
inappropriate
• Increased
pressure on land
• Lack of labour
• Inadequate finances
•
•
•
•
•
Low prices
Poor infrastructures
Lack of information
Weak market
Unfavourable policy
environment
6
Bationo, A. et al
The gradual degradation of the land is a menace to rural
communities, in terms of food security and a continued exploitation of
the fragile resource base depleted from many plant nutrients. There is,
therefore, a critical need to develop and implement management options
that both mitigate soil degradation, deforestation and biological resources
losses and enhance local economies while protecting the natural resource
base.
Transforming African Agriculture and expanding its production
capacity are prerequisites for alleviating rural poverty, household food
deficits and environmental exploitation in the continent. Because
opportunities for expanding the cultivated area are rapidly being
exhausted, as much as four-fifths of future production increases must
come from higher yields. The use of effective strategies to combat land
degradation is one of the key components of the higher productivity.
The African Network for soil biology and fertility (AfNet) of the Tropical
Soil Biology and Fertility Institute of CIAT is established to overcome
the challenge of soil fertility degradation in the African continent. In
this paper, after a brief presentation of AfNet objective and management,
we will present the new challenges and opportunities of this network in
field research activities, information and documentation, training and
capacity building.
AfNet Objectives and Management
Networking may be defined as a strategy by stakeholders in a given
area of interest to work together to achieve a common objective. The
building blocks of a network are the participating individuals or
institutions. These stakeholders collaborate on the hypothesis that
working together is more beneficial and effective than working
independently, and that there is a need to go outside the organization
in order to accomplish their goals. Through networking, participants (a)
build-up their knowledge base, (b) understand the processes through
which they can promote values and (c) translate their understanding
into action. Several achievements are possible in research through
networking. The collaborating institutions or individuals are in a position
to exchange information and combine collective experience of
professionalism in the same field as partners. Figure 1.4 gives the
different elements of partnerships and these elements are considered
by AfNet in order to increase the network effectiveness and efficiency.
The advantages of networking include:
i) To achieve economies of scale and efficiency in research by
concentrating scarce human, financial and other resources on key
national and regional problems;
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
7
Figure 1.4: Elements of Partnerships
Credit &
Recognition
Trust &
Commitment
Power &
Equity
Communication
Compelling
Vision
Explicit Decision
Making Process
Mutual
Accountability
Leadership
Attention to
Process
Shared
Understanding
of Problems
Interdependency
&
Complementarity
ii) To provide increased bargaining power with external partners;
iii) To minimize duplication;
iv) To exchange information and combine collective experience of
professionals in the same field;
v) To carry out collaborative research through network experiments;
vi) To undertake joint capacity building;
vii) To capture research spill-over/ spill-in effects;
viii)To rationalize human resource development;
ix) To mobilize research efforts on trans-national problems that require
collaboration between countries;
x) To exploit a larger market for agricultural research technologies
through regional collaboration;
xi) To demonstrate impact despite the declining investment in
agricultural research through regional cooperation;
xii) To achieve lower transaction costs;
xiii)To facilitate better and more access by all stakeholders of available
technologies at regional and international levels.
Bationo, A. et al
8
The African Network for Soil Biology and Fertility (AfNet) was
established in 1988 and is the single most important implementing
agency of TSBF in Africa. Its main goal is to strengthen and sustain
stakeholder capacity to generate, share and apply soil fertility and biology
management knowledge and skills to contribute to the welfare of farming
communities. It is a mechanism to facilitate and promote collaboration
in research and development among scientists in Africa for the purpose
of developing innovative and practical resource management
interventions for sustainable food production. AfNet has membership
from National Agricultural Research and Extension services (NARES)
and Universities from various disciplines mainly soil science, social
science, agronomy and technology exchange.
Figure 1.5: AfNet member registrations since inception
210
200
195
101 from East and Central Africa
68 from Southern Africa
31 from West Africa
180
Number of scientists
165
150
135
120
105
90
80
75
60
45
30
15
48
51
91-95
96-98
35
23
10
0
<89
89-90
91-95
99-00
>02
Year
With a total number of 10 researchers in 1989, AfNet has now a
total number of over 200 persons in 2003. It is an African-wide network
with 101 members from East and Central Africa, 68 from Southern
Africa and 31 from West Africa (Figure 1.5). The data in Figure 1.6 gives
the AfNet participating countries.
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
9
Figure 1.6: AfNet partitipating countries, 2003: Number in parenthesis represent the
number of AfNet participants in the particular country
East and Central
Africa
1. Uganda (28)
2. Kenya (55)
3. Tanzania (11)
4. Rwanda (1)
5. Burundi (9)
6. Ethiopia (2)
7. DRC (8)
8. Madagascar (1)
Southern Africa
9. Zambia (12)
10. Malawi (1)
11. Zimbabwe (14)
12. South Africa (5)
West Africa
13. Mali (2)
14. Niger (1)
15. Nigeria (6)
16. Burkina Faso (4)
17. Cote D'Ivoire (4)
18. Ghana (4)
19. Togo (1)
20. Cameroon (5)
21. Sierra Leone (1)
22. Senegal (2)
23. Benin (2)
AfNet is under the auspices of the Tropical Soil Biology and Fertility
Institute of CIAT, who implement most of its activities in Africa through
AfNet. The AfNet members share TSBF goals and approaches. TSBF
conduct research in a variety of tropical countries, but always in
collaboration with national scientists. This implementation of TSBF
agenda through partnership utilizes a range of approaches with
particular emphasis on the following:
i) Catalysis: Ensuring that AfNet members are kept at the forefront of
conceptual and methodological advances by conducting and
promoting review, synthesis and dissemination of knowledge and
information. This is done through workshops, training and sabbatical
and short exchange visits.
ii) Facilitation: Co-ordinating actions by members to achieve progress
10
Bationo, A. et al
and success in research. This is done by providing backstopping
support in the preparation, submission, implementation and
publication of research results.
iii) Collaboration: Developing appropriate alliances with institutions
across the research, educational and development spectrum,
including linkages between institutions in the North and those in
the South.
AfNet has a coordination unit comprising of a secretariat, research
assistants and the coordinator. AfNet is managed by a scientific
committee comprising of the director of TSBF, the AfNet coordinator
and five members from the national programmes elected during general
assemblies by AfNet members.
AfNet is dedicated to work more closely with other networks, systemwide ecoregional initiative such as AHI, SoilFertiNet, ANAFE, DMP,
SWMNet, ECABREN, MIS and is planning to have an active role in the
various challenge programmes of the CGIAR.
Network field activities
Predictive interdisciplinary research across environments, using
standard methods and experimental designs, reinforces results, enables
the drawing and extrapolation of generalized conclusions and enhances
modeling capacity, all leading to accelerate progress in essential research
areas. AfNet works with partners to identify key research themes or
problems of regional or international importance and then develops
appropriate experimental methods and protocols for addressing those
topics. There will be a special focus on the use of decision support
systems, GIS and modeling for the extrapolation of research results to
other recommendation domains.
AfNet field research activities addresses the same research outputs
of the TSBF institute of CIAT (Figure 1.7) with the overall goal of
empowering farmers for sustainable agro ecosystem management.
Output 1 on Integrated Soil Fertility Management (ISFM), output 2 on
belowground biodiversity and agro-ecosystem health and output 3 on
soil-based ecosystem services are the technical outputs for the
development of alternative options. In Africa all research institutions
are confronted with the challenge of extending their research findings
for successful impact on farm. The fourth output on strategies for scaling
up/ out will focus on evaluation of management options, on pathways
of knowledge interchange and on policies for sustainable soil
management by using the technical options developed by the other
outputs.
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
11
Figure 1.7: The Tropical Soil Biology and Fertility Institute of CIAT research outputs
Goal: Empowering farmers for sustainable
agroecosystem management
Output 4:
Strategies for scaling up/out
Output 1:
ISFM
constraints
and processes
Output 2:
BGBD and
agroecosystem
health
Output 5:
Capacity
Building
Output 3:
Soil-based
ecosystem
services
During the past three decades, the paradigms underlying soil fertility
management research and development efforts in SSA have undergone
substantial change. From the nutrient replacement paradigm to Low Input
Sustainable Agriculture (LISA) AfNet adopted the Integrated Soil Fertility
Management (ISFM) paradigm that forms an integral part of the Integrated
Natural Resource Management (INRM) research approach with a focus
on appropriate management of the soil resource (Figure 1.8).
Figure 1.8: Integrated Soil Fertility Management strategies with wider natural
management concerns
Integrated Soil Fertility Management Strategy
Integrated Pest
Management
lSFM
Policy
En
tr
y
po
i
nt
Soil Conservation
Water Management
Resilient GermPlasm
/Fertilizer (org+inorg)
Market
Ecosystem
Services
12
Bationo, A. et al
In essence, ISFM is the adoption of a holistic approach to research
on soil fertility that embraces the full range of driving factors and
consequences- biological, physical, chemical, social, economic and
political.
The need has been recognized for integration of socio-economic and
policy research besides technical research. Soil fertility can no longer
be regarded as a simple squared by the issue of organic and inorganic
nutrient sources. The holistic approach encompasses nutrient
deficiencies, inappropriate germplasm and cropping systems, pest and
disease interaction with soil fertility, linkage between land degradation
and poverty and global policies, incentives as well as institutional failures.
Such long-term soil fertility management strategy requires an
evolutionary and knowledge intensive process and participatory research
and development focus rather than a purely technical focus.
AfNet will focus in the years to come on the following research topics
and projects for the implementation of its field research activities.
Nutrient budgets of agroecosystems
Past research focuses on N, P, and K, there is need to target other
macronutrient besides nitrogen, phosphorus and potassium and
micronutrients and soil carbon. There will be need to focus more on
methodologies for extrapolation of results in the time and space scales.
The validation of transfer functions leading to better estimates of leaching
losses, gaseous and erosion losses and the need to link nutrient balances
data to other soil productivity indicators, total or available nutrient
stocks, fertilizer needs and response functions and nutrient budgets to
farmers’ perception and knowledge systems.
Economic, policy and dissemination issues
In most of the research projects, economic policy and dissemination
issues are incorporated with focus of economic analysis of soil fertility
technologies with special emphasis on the trade-offs of alternative
strategies of soil fertility management (eg food, feed, soil fertility
management, social functions), the need to incorporate economic and
bio-physical modeling to capture long-term sustainability and risk
perspectives.
On adoption and impact assessment, special attention is put on the
assessment of socio-economic and agronomic factors affecting farmers’
adoption of best bet technologies, the measurement, the understanding
of the potential and constraints and the economics of different
dissemination channels. Research on ways to increase farmers’ access
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
13
to external input through the establishment of appropriate credit and
saving schemes. Policy research and advocacy to create an enabling
environment to accelerate adoption of best bet technologies and
establishment of policy briefs and studies on economic of different
dissemination channels will be emphasized.
Long-term soil fertility management trials
AfNet is contributing with NARS to maintain long-term soil fertility
management trials in the sub-humid highlands of Kenya at Kabete
(established since 1976), the sub-humid zone of Burkina Faso at
Farakoba (established since 1990), the dry savannah of Burkina Faso
at Saria (established since 1960) and at Fada (established since 1990)
and the Sahelian zone of Niger at Sadore (established since 1982). The
overall goal is to access sustainability indicators from the different inputs
(organic and inorganic) and cropping systems.
Combining organic and inorganic nutrient sources for
increased soil quality
The overall goal of the work on organic-inorganic interactions is to (i)
empower farmers (Including extension workers and stakeholders) to
use organic and inorganic resources with optimal efficiency; (ii)
understand the long-term effects on nutrient recovery efficiencies and
(iii) better understand the non-N effects of organic amendments (weed
suppression eg striga, other nutrients (Ca, Mg, K, S, P etc…), moisture
retention and use).
The role of legumes in soil fertility restoration
The general objective for this network study is to foster strategic research
on issues that increase efficiency of legume cover crops (LCCs) for
enhancement and sustaining soil fertility and hence crop yields in
smallholder farms in the Sub-Saharan Africa. The derived specific
objectives are: 1) Review and document current information on the use
of LCCs for soil fertility improvement in the Africa region 2) Determine
the contribution of above and below ground biomass from LCC on the
subsequent food crops 3) Determine the relationship between source
and quantity of N from cover crops and its recovery in the subsequent
crop 4) Evaluate tradeoffs in gains and losses in food production, land
availability, labour constraints and capital that may affect the adoption
of LCCs 5) Develop a decision support guide for dissemination of LCC
technologies.
14
Bationo, A. et al
Livestock and soil fertility issues
For this research theme we will emphasize on the assessment of manure
production, livestock rangeland ratios for sustainable production.
Strategies that minimize competition between crops and livestock. The
overall objective for the improvement of manure management aims at
reviewing past manure work in the individual countries and identifying
technologies that could be disseminated without doing basic research;
testing and validating, various composting/storage techniques on crop
yield, soil fertility maintenance and economics with farmers’
participation; and the contribution of manure use to soil organic matter.
Use of rock phosphate as capital investment to replenish
soil fertility
The use of fertilizers is a possible option to reverse the soil fertility decline
trend but their high costs constitute an handicap mainly to resource
poor farmers. There is therefore a need for alternative, affordable P
sources. Rock phosphates which are found in most parts of Africa have
low reactivity. They, however offer a cheaper source of P for resource
poor farmers. For this issue, there is need to extend the agronomic
evaluation of suitability of PR to perennial crops and other crops than
the traditional cereals and to investigate on the interaction between
soil, climate and water conservation on PR effectiveness. The screening
of plant species and association with Vascular Arbuscular Mychorrizae
(VAM) for efficient use of PR need more attention. The economics of
compacted products with PR and development of decision support
systems (DSS) need to be emphasized. Solubility of these phosphate
rocks can be improved by using combination of the rock and organic/
green manures. Besides their solubilization effects, the organic materials
influence soil P availability by altering some processes governing soil P
pools such as microbial activities and P sorption. Different organic inputs
are likely to impart different effects on rock phosphate dissolution and
soil P availability depending on their composition, rate of application,
and type of soil and agro- ecological zone. This network research theme
therefore is to do on-farm testing of Phosphate rocks P dissolution as
influenced by different types and rates of organic materials and the
subsequent crop yield. Specifically, the research is intended to identify,
characterize and evaluate locally available organic materials for their
potential to enhance phosphate rock solubilization under farmer’s
conditions. Also establish the effect of local organic materials on PR
dissolution and its relative agronomic effectiveness. In addition the
research will assess the effects of organic and inorganic P sources on
soil P dynamics and fractions.
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
15
Nutrient and water interaction
African dry lands soils have low inherent fertility, and this combined
with high inter-annual variability and erratic rainfall distribution in
space result in water limiting conditions and poor crop yields. The use
of effective strategies to control nutrient mining and improve water and
nutrient use efficiency in dry land Africa is one of the key components
for higher productivity. In light of large initial investment in material
machinery and labor for water harvesting, there is need to focus research
to increase the profitability in the farming systems. Although water and
nutrient interaction research is essential for increasing and stabilizing
crop production, and for maximizing the returns from investments on
fertilizer and water harvesting techniques, far less studies on these
interactions have been carried out in the dry lands of Africa, compared
to studies on nutrients or those of water separately. A win win situation
will occur when water and nutrients are combined as this will increase
the efficiency of these inputs and therefore improve their profitability to
the small-scale farmer.
Land tenure
Land tenure has a critical impact on market values and thus on economic
decision making as to the uses to which land should be put and how to
utilize the natural resource. Nature of land rights affects use; duration
of right affects nature of long-term investment. World Bank estimates
suggest that the capital value of land and natural resources constitutes
half to three quarters of a nation’s wealth: the less domestic capital and
the less developed the economy, the higher this proportion. What is
true for the nation is also typically true for the family and individual.
Land and natural resources are therefore likely to be by far the largest
class of asset in most economies. Its efficient use and management
must be one of the keys to successful economic development. Secure
land rights will move the key economic resource of land towards the
highest and economically most efficient use.
Lack of secure land tenure is associated with overexploitation of
resources. In turn, overexploitation of land and natural resources
critically affects the economic welfare and food security. With insecure
land tenure, farmers have no incentive to commit long-term investments
for sustainable farming and livelihood.
The main goal of this research theme is to contribute to poverty
reduction through increased land productivity to improve food security,
while conserving the natural resource base for sustainable production.
The purpose is to provide farming communities, policy-makers and other
stakeholders with land tenure policy options that will improve adoption
16
Bationo, A. et al
of land management and conservation technologies. The research agenda
seek to achieve the following specific objectives: (1) Review and compile
existing land and natural resource use and investment policies that
has direct implication to smallholder farmers’ decision-making process,
(2) Identify categories of tenure that play major role in adopting available
technologies for integrated land management and natural resource use,
(3) Estimate socioeconomic gains associated with secure land tenure
through bio-economic modeling/simulation, (4) Suggest policy
instruments that would encourage secure land tenure and maximize
national goals of improving smallholder farmers’ welfare.
Conservation tillage
Combination of soil fertility restoration technologies and conservation
tillage practices offer opportunities to sustainable land use. However,
little has been done to integrate these approaches within existing crop
production systems. It is hypothesized that combining soil fertility
technologies with conservation tillage practices is one of the best
strategies to increase food security, sustain rural livelihoods in subSaharan Africa and maintain soil organic matter.
In this research theme, the following specific objectives are sought
to be achieved: (i) Evaluate the productivity of different cropping systems
following conservation tillage practices, (ii) Determine the effect of
conservation tillage on sustainability and soil health indicators,
(iii) Promote conservation tillage practices as a means to restore the
productivity of degraded soils.
Belowground biodiversity
The soil biota constitutes a major fraction of global terrestrial biodiversity
and is responsible for key ecosystem functions such as decomposition;
nutrient acquisition, storage and cycling; soil organic matter synthesis
and mineralization; soil structural modification; regulation of
atmospheric composition; and the biological control of soil-borne pests
and diseases. These functions remain largely under-exploited by humans
for services and products in agriculture because little has been
understood on the biological processes of soil unlike physical and
chemical management of soil.
The strategic research in AfNet to realise this potential by:
• Developing quantitative techniques for monitoring and manipulating
key functional groups of soil biota and their relationship to ecosystem
service functions and plant health.
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
17
•
Developing and validate management practices for key groups of
beneficial soil organisms for small-scale farms.
•
Linking local knowledge about biological indicators of soil quality
with scientific knowledge to develop robust soil quality monitoring
systems that combine precision and relevance.
This research agenda will also seek to establish the relationship
between organic residue quality (resource quality), farmers’ management
strategies and diversity, populations and activities of biotic community
(macrofauna: ecosystem engineers-earthworms, termites, ants and
others) associated with biomass transfer technologies. The main aim is
to come out with the best-bet approach that promote soil biotic activities,
increase and sustain soil productivity and minimize pest incidence in
tropical agroecosystems.
Also to sustainably increase crop yield for small-scale farmers in
sub-Sahara Africa by using mycorrhiza as bio-fertilizer and build farmers’
understanding on the importance of termites and other macro and meso
faunal communities in African farming systems.
Low quality organic resource management
This research theme examines the functional role of low quality organic
resources on soil organic matter (SOM) and the ultimate influence in
sustaining crop productivity and environmental service functions in
tropical agro-ecosystems as affected by management of quality and
quantities of low quality organic resources available to smallholder
farmers in the sub-Saharan region of Africa. The research will i)
Characterize the quantity and quality of organic materials available to
smallholder farmers in benchmark areas, and determine how these have
influenced SOM status and dynamics under different management
practices and biophysical environments; ii) determine the quantitative
effects of continuous application of low quality organic inputs on SOM
build up, soil nutrient supply patterns and soil physico-chemical
properties; iii) quantify the differential contribution of distinct SOM
functional pools (fractions) to soil properties essential for maintenance
of crop productivity and environmental quality under different
management systems, soils types and climatic environments in selected
benchmark areas; iv) define the biophysical and socio-economic
boundaries within which SOM management can be enhanced for
increased soil productivity and environmental services in tropical farming
systems in sub-Saharan Africa.
Bationo, A. et al
18
Site selection
The field activities are carried out on benchmark and satellite sites. The
benchmark sites are selected according to several factors such as soil
types, rainfall regime, farming systems, type of market, land tenure,
etc…
At present, AfNet is implementing field research in about 50
representative benchmark sites in Africa. Figure 1.9 gives some selected
benchmark sites where experiments were carried out in 2002. AfNet
encourages multi-disciplinary approach for the implementation of its
field activities but individual research projects are also supported by
the network. In addition to the thematic research, focus is put now on
the development of country proposals using an holistic approach to
Integrate Soil Fertility Management (ISFM).
Figure 1.9: Selected AfNet research sites in 2002
4
Koulikoro**,
Fana** Niono
(ICRISAT, IER)
1
2
Farakou-Ba**,
Kouare** (INERA)
Sadore**, Banizoumbou**
Karabedji**, Gaya**, Gobery** (ICRISAT)
IRAD, University of Yaounde
8
9
Ndere dance Troupe
10 Makerere university
NARO
University of
Abidjan-Cocody
Lamto**
5
Kumasi**
7
6
Lome**
AfNet participating countries
and research sites
1.
2.
3.
4.
5.
6.
7.
8.
9.
Mali
Niger
Nigeria
Burkina Faso
Cote DÕIvoire
Ghana
Togo
Cameroon
Congo (DR)
University of Kisangani
10.
11.
12.
13.
14.
15.
16.
11
3
Zaria**
Uganda
Kenya
Tanzania
Zambia
Zimbabwe
South Africa
Madagascar
** Established 2002
Egerton university,
Nairobi university,
KAR-KEFRI-ICRAF
12
Mlingano,
17 ChitedzeSokoine university
FOFIFA
16
13
Misamfu Research
Centre, Chilanga
14 University of Zimbabwe
15
University of Witwatersrand
University of Fort Hare
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
19
Funding mechanism
The funding of network trials is on a competitive basis and the criteria
used for the attribution of funds are based on: (i) the level of contribution
to food security and self sufficiency (ii) equity (number of beneficiaries,
poverty alleviation, gender/ age consideration) (iii) efficiency (iv)
sustainability (v) effectiveness (probability of success, cost of adoption)
(vi) regional collaboration.
Information and documentation
One of the main constraints to soil biological research experienced by
many national scientists is limited access to current research findings.
It is important not only that current research developments are accessible
to members of the network but that the results of their own work are
effectively disseminated. In addition, farmers in SSA are attempting to
improve soils, but their efforts are constrained by limited access to useful
information, low resource endowments, and lack of incentives. Wealthier
households having access to information and with more options
available, are more likely to manage their soils better. Poor households
lack knowledge of soil management options, the capacity to invest in
soils (especially in fertilizer), and have less ability to bear risk and wait
for future payoffs from investment. For example, in Western Kenya,
resource-poor households, with no access to information, were found
to make only 5% of the farm investments, had over twice the erosion
rates as compared to the wealthy farmers, and obtained only 28% of
maize yields. Tragically, these resource-poor households constitute about
90% of the population. Compounding the problem are poor price
incentives, land and labour constraints, and the weakness or complete
lack of rural institutions for supporting information and other services.
The network will collaborate with other institutions to develop
information easy-to-read by farmers on transferable technologies for
soil fertility restoration.
A major function of AfNet is to publish, synthesis and disseminate
research results relevant to its programme goals. AfNet is publishing
twice yearly the comminutor (TSBF newsletter) as a link between network
members. Literature search is done as needed on specific subjects for
distribution to network members. In addition to publication to refereed
journals, AfNet has committed to produce three books.
(i) Soil Fertility Management in Africa: A Regional Perspective.
(ii) Managing Nutrient Cycles to Sustain Soil Fertility in Africa:
Proceedings of the 8th Meeting of The African Network for Soil Biology
and Fertility, Arusha, Tanzania.
(iii) Fighting Poverty in sub-Saharan Africa: The Multiple Roles of
Legumes in Integrated Soil Fertility Management.
20
Bationo, A. et al
Training and capacity building
The capacity for ISFM research in sub-Saharan Africa is insufficient
both in terms of the numbers of professional personnel and the essential
laboratory facilities. ISFM is a knowledge intensive approach to soil
management. Professional staff and students alike suffer from isolation
and lack of access to up-to-date educational opportunities. Networks
run by sub-Regional Organisations and CGIAR Centres, such as the
TSBF African Network for Soil Biology and Fertility (AfNet) provide a
vehicle of opportunity to correct this situation. Priority actions include:
• Strengthen networking to engage a wide range of stakeholders and
enhance the efficiency of ISFM research.
• In particular, strengthen links between research and extension
(including NGOs) using a “learning by doing” approach, which
includes local knowledge and builds on existing networks.
• Develop strategic partnerships in capacity building that identify and
utilise the range of comparative expertise.
• Improve the dissemination of knowledge on ISFM through a wide
range of methods including electronic sharing and training of trainers.
• Promote programmatic linkages with Universities and other
educational institutions to strengthen curricula with appropriate
and up-to-date information and teaching materials.
• Raise awareness of ISFM issues with policy and decision-makers at
all levels.
Table 1.2 below showing the percentage literacy rates reveals an
average (57%) literacy rate in the African continent as compared to 97%
in the European continent. In some countries like Niger, the literacy
rate is as low as 16%. This is a clear indication that more than half of
the African population cannot read neither write hence imposing a great
impairment to the implementation and dissemination of the research
results.
Table 1.2: Selected education statistics for some countries in Africa and other world
regions
Country
Burkina Faso
Kenya
Uganda
Niger
World
SSA
Europe
Source: ADB 1999
Literacy rate (%)
Tertiary school
enrolment (%)
24
83
67
16
74
57
97
0.9
1.7
1
14
5
27
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
21
In sub-Saharan Africa tertiary school enrolment has gone as low as
5% as compared to the European continent, which has 27%. However,
in addition to low literacy rate in the African continent, there has been
a great concern that institutions of higher learning are not making a
significant contribution to the national agricultural research agenda.
This is due in part to the limited funding of agricultural higher education
(Table 1.3). From 1987-97, World Bank global support to agricultural
extension was 46.3% as compared to 2.2% for agricultural higher
education. The common trend in the African continent has been decline
in support for research in these institutions. This trend has to change
especially with the realization that many universities in Africa have a
large stock of agricultural scientists with M.Sc. and PhD degrees. For
example, in 1995, there were 547 African scientists with a PhD in
agriculture employed by universities and 357 in the National Agricultural
Research Systems (NARSs) in Eastern and Southern Africa.
Table 1.3: World Bank Global Support for Agricultural Research, Extension and
Agricultural Higher Education, 1987-97
Million $
Percent
Agricultural research
Agricultural extension
Agricultural higher education
2,482
2,229
108
51.5
46.3
2.2
Total
4,819
100
Source: Willett 1998
Lack of administrative, managerial and scientific capacity has been
noted as the weak link in African development. Therefore, it is of great
importance to launch capacity building initiatives in the African
continent. The availability of personnel suitably trained in the appropriate
techniques is essential for sustainable agricultural development and
research. Since investment in knowledge and human resources is central
to sustained development, capacity building should help to rehabilitate
and strengthen research and higher education in the African region.
TSBF promotes interest in soil biology and fertility among scientists by
providing experience and orientation in TSBF methods through short
courses, internships and attendance at professional meetings.
Universities and other institutions of higher learning represent the
only sustainable option that can, in the long-term, reduce the over
dependency on overseas training in the African continent. Therefore,
the managers of agricultural research and extension systems in Africa
should have a deep concern on improving the quality of local graduate
programs because, after phasing out scholarships for overseas training,
22
Bationo, A. et al
African universities remain the primary source of human capital for
agricultural research and extension agencies in the continent. The
African Network for Soil Biology and Fertility (AfNet) has taken this
challenge and is in the process of developing a soil biology curriculum
support in African Universities. Some of the needs highlighted by 13
African Universities include: lack of critical mass, limited access to
information, limited access to teaching material, poor laboratory facilities,
and limited examples from African environments.
AfNet will organize short term training courses which will address
the following issues: (i) TSBF Standard methods for process and applied
research in Soil Biology and Fertility; (ii) data collection, statistical
analysis and interpretation; (iii) methodology for on-farm research; (iv)
scientific paper writing; and (v) development of research proposals. It
will also liase with universities in Europe and the United States of
America to have students do their thesis with TSBF officers in Africa for
co-supervision of students for MSc and PhD from local universities on
topics relevant to TSBF research
Conclusion
Land degradation is one of the most serious threats to food production
in the African continent. The population is thus trapped in a vicious
poverty cycle between land degradation, and the lack of resources or
knowledge to generate adequate income and opportunities to overcome
the degradation and it is urgent to invest to combat land degradation to
revert this vicious circle. Soil fertility can no longer be regarded as a
simple issue squared by the issue of organic and inorganic sources of
nutrients. Integrated soil fertility management embraces responses to
the full range of driving factors and consequences, namely biological,
physical, chemical, social, economic and political aspects. The holistic
approach encompasses nutrient deficiencies, inappropriate germplasm
and cropping system design, pest, disease interaction with soil fertility,
linkage between land degradation and poverty and global policies,
incentives as well as institutional failures. Such long-term soil fertility
management strategy requires an evolutionary and knowledge intensive
process and participatory research and development focus rather than
a purely technical focus.
AfNet developed several research projects on Integrated Soil Fertility
Management (ISFM), Belowground Biodiversity (BGBD) and agro
ecosystem health, soil based ecosystem services and strategies for scaling
up/ out to empower farmers for sustainable agro-ecosystems’
management. Information and documentation, training and capacity
building are among the main strategies of AfNet for sustainable
agricultural development in Africa.
The African Network for Soil Biology and Fertility: New Challenges and Opportunities
23
References
African Development Bank (ADB). African Development Fund (1999) Education
Sector Policy Paper (ESPP), OCOD, December 1999.
FAO (1999) AGROVOC multilingual agricultural thesaurus, Rome, Italy.
Murwira, H.K. (2003) Managing Africa’s soils: Approaches and challenges. In:
Gichuru, M.P., Bationo, A., Bekunda, M.A., Goma, H.C., Mafangoya, P.L.,
Mugendi, D.N., Murwira, H.M., Nandwa, S.M., Nyathi, P. and Swift, M. Soil
Fertility Management in Africa: A regional Perspective. Academy Science
Publishers, Nairobi, Kenya, 306 pp.
Palm, C., Bationo, A. and Waddington, S. (2001). Integration of soil research
activities in Eastern and Southern Africa, Nairobi, Kenya, 48 pp.
Sanchez, P.A., Shepherd, K.D., Soule, M.J., Place, F.M., Buresh, R.J., Izaac,
A.M.N.,Mokwunye, A.Z., Kwesiga, F.R., Ndiritu, C.G., Woomer, P.L. (1997)
Soil fertility replenishment in Africa: An investment in natural resource
capital. In: Buresh, et al. Replenishing Soil Fertility in Africa. SSSA Special
Publication Number 51:1-46.
Willet, A (1998) Agricultural education review: Support for agricultural education
in the bank and by other donor executive summary. Washington D.C., World
Bank.
24
Bationo, A. et al
Integrated Soil Fertility Management Research at TSBF: The Framework, the Principles,
and their Application
Integrated Soil Fertility
Management Research at
TSBF: The Framework, the
Principles, and their
Application
25
2
Vanlauwe, B.
Scientific Officer, Tropical Soil Biology and Fertility Institute of
CIAT, PO Box 30677, Nairobi, Kenya
Abstract
Integrated Soil Fertility Management (ISFM) has been
adopted by the Tropical Soil Biology and Fertility (TSBF)
Institute, its African Network (AfNet), and various other
organisations as the paradigm for tropical soil fertility
management research and development. The development
of ISFM is the result of a series of paradigm shifts generated
through experience in the field and changes in the overall
socio-economic and political environment the various
stakeholders, including farmers and researchers, are facing.
A first part of the paper illustrates these shifts and sketches
how the science of organic matter management has
developed in the framework of the various paradigms. The
second part focuses on the technical backbone of ISFM
strategies by illustrating the roles of organic resources,
mineral fertilizer, and soil organic matter (SOM) in providing
soil-related goods and services. Special attention is given
to the potential occurrence of positive interactions between
26
Vanlauwe, B.
these three factors, leading to added benefits in terms of
more crop yield, improved soil fertility status, and/or
reduced losses of C and nutrients to the environment. A
third part aims at confronting the principles and
mechanisms for soil fertility management, highlighted in
the second section, with reality and focuses on the impact
of other realms of capital on soil management opportunities
and the potential of decision aids to translate all knowledge
and information in a format accessible to the various
stakeholders.
Paradigm shifts related to tropical soil fertility
management: From a Nutrient Replenishment to an
Integrated Soil Fertility Management agenda
During the past 3 decades, the paradigms underlying soil fertility
management research and development efforts have undergone
substantial change because of experiences gained with specific
approaches and changes in the overall social, economic, and political
environment the various stakeholders are facing. TSBF has traditionally
put a lot of emphasis on the appropriate management of organic
resources and the conceptualisation of the role of organic resources in
tropical soil fertility management has obviously been adapted to the
various underlying paradigms.
During the 1960s and 1970s, an external input paradigm was driving
the research and development agenda. The appropriate use of external
inputs, be it fertilizers, lime, or irrigation water, was believed to be able
to alleviate any constraint to crop production. Following this paradigm
together with the use of improved cereal germplasm, the ‘Green
Revolution’ boosted agricultural production in Asia and Latin America
in ways not seen before. Organic resources were considered less
essential. Sanchez (1976) stated that when mechanization is feasible
and fertilizers are available at reasonable cost, there is no reason to
consider the maintenance of SOM as a major management goal. However,
application of the ‘Green Revolution’ strategy in sub-Saharan Africa
(SSA) resulted only in minor achievements because of a variety of reasons
(IITA, 1992). This, together with environmental degradation resulting
from massive applications of fertilizers and pesticides in Asia and LatinAmerica between the mid-1980’s and early-1990’s (Theng, 1991) and
the abolition of the fertilizer subsidies in SSA (Smaling, 1993), imposed
by structural adjustment programs led to a renewed interest in organic
resources in the early 1980s. The balance shifted from mineral inputs
only to low mineral input sustainable agriculture (LISA) where organic
resources were believed to enable sustainable agricultural production.
Integrated Soil Fertility Management Research at TSBF: The Framework, the Principles,
and their Application
27
After a number of years of investment in research activities evaluating
the potential of LISA technologies, such as alley cropping or live-mulch
systems, several constraints were identified both at the technical (e.g.,
lack of sufficient organic resources) and the socio-economic level (e.g.,
labour intensive technologies).
In this context, Sanchez (1994) revised his earlier statement by
formulating the Second Paradigm for tropical soil fertility research: ‘Rely
more on biological processes by adapting germplasm to adverse soil
conditions, enhancing soil biological activity and optimizing nutrient
cycling to minimize external inputs and maximize the efficiency of their
use’. This paradigm did recognize the need for both mineral and organic
inputs to sustain crop production, and emphasized the need for all
inputs to be used efficiently. The need for both organic and mineral
inputs was advocated because (i) both resources fulfil different functions
to maintain plant growth, (ii) under most small-scale farming conditions,
neither of them is available or affordable in sufficient quantities to be
applied alone, and (iii) several hypotheses could be formulated leading
to added benefits when applying both inputs in combination. The second
paradigm also highlighted the need for improved germplasm, as in earlier
days, more emphasis was put on the nutrient supply side without
worrying too much about the demand for these nutrients. Obviously,
optimal synchrony or use efficiency requires both supply and demand
to function optimally.
From the mid-1980s to the mid-1990s the shift in paradigm towards
the combined use of organic and mineral inputs was accompanied by a
shift in approaches towards involvement of the various stakeholders in
the research and development process, mainly driving by the
‘participatory’ movement. One of the important lessons learnt was that
the farmers’ decision making process was not merely driven by the soil
and climate but by a whole set of factors cutting across the biophysical,
socio-economic, and political domain. The Sustainable Livelihoods
Approach (DFID, 2000) recognizes the existence of five realms of capital
(natural, manufactured, financial, human, and social) that constitute
the livelihoods of farmers. It was also recognized that natural capital,
such as soil, water, atmosphere, or biota does not only create services
which generate goods with a market value (e.g., crops and livestock)
but also services which generate amenities essential for the maintenance
of life (e.g., clean air and water). Due to the wide array of services provided
by natural capital, different stakeholders may have conflicting interests
in natural capital. The Integrated Natural Resource Management (INRM)
research approach (Figure 2.1) aims at developing interventions that
take all the above into account (Izac, 2000). The Integrated Soil Fertility
Management (ISFM) paradigm, that forms and integral part of the INRM
research approach with a focus on appropriate management of the soil
resource, is currently adopted in the soil fertility research and
28
Vanlauwe, B.
development community. Although technically ISFM adopts the Second
Paradigm, it recognizes the important role of social, cultural, and
economic processes regulating soil fertility management strategies. ISFM
is also broader than Integrated Nutrient Management (INM) as it
recognizes the need of an appropriate physical and chemical environment
for plant to grow optimally, besides a sufficient and timely supply of
available nutrients.
Figure 2.1: The Integrated Natural Resource Management research approach
Problem analysis
Food insecurity
Increasing poverty
Degrading environments
Policy constraints
Enhanced productivity
Quantity
Quality products
G x X matching efficiency
Enhanced well being
Risk management
Resource users
Participation in decisions
Enhanced resilience
Nutrient cycling
C sequestration
Biodiversity
Water balance
Trade-off analysis
Identification of rantes of flexible adaptive options
Extrapolation
Dissemination
Policy implementation
Wide-scale adoption
Impact assessment
Source: Izac, 2000
The science of organic matter management as affected by
shifts in soil fertility management paradigms
Although organic inputs had not been new to tropical agriculture, the
first seminal synthesis on organic matter management and
decomposition was written only in 1979 by Swift et al. (1979) (Table
2.1). Between 1984 and 1986, a set of hypotheses was formulated based
on 2 broad themes, ‘synchrony’ and ‘SOM’ (Swift, 1984, 1985, and 1986),
building on the concepts and principles formulated in 1979. Under the
first theme, especially the O(rganisms)-P(hysical environment)-Q(uality)
framework for OM decomposition and nutrient release (Swift et al., 1979),
formulated earlier, was worked out and translated into hypotheses
driving management options to improve nutrient acquisition and crop
growth. Under the second theme, the role of OM in the formation of
functional SOM fractions was stressed. During the 1990s, the
Integrated Soil Fertility Management Research at TSBF: The Framework, the Principles,
and their Application
29
formulation of the research hypotheses related to residue quality and N
release led to a vast amount of projects aiming at validation of these
hypotheses, both within AfNet and other research groups dealing with
tropical soil fertility. This information has been very instrumental
for proper evaluation of the sustainability of LISA systems. As such
systems did not emphasize the need for mineral inputs, organic resources
were merely considered as short-terms sources of nutrients and
especially N.
Table 2.1: A brief summary of the science of tropical organic resource management
Period
Observation
Reference
< 1970s
Organic matter as a ‘blob’
Palm, personal
communication
1979
Organisms - Physical environment –
Quality framework for organic matter
decomposition
Swift et al., 1979
1984-1986
Development of the ‘synchrony’ research
theme within the Tropical Soil Biology
and Fertility programme
Swift, 1984; Swift,
1985; Swift, 1986
1990s
Various experiments addressing the
‘synchrony’ hypothesis
Various
1995
International Symposium on ‘Plant Litter
Quality and Decomposition’
Cadisch and Giller,
1997
2000
Development of the ‘Organic Resource
Database’ and the Decision Support System
for organic N management
Palm et al., 2001
> 2001
Quantification of the Decision Support
System for organic N management
The current and
future publications
Two major events further accentuated the relevance of the topic in
tropical soil fertility management. Firstly, a workshop was held in 1995
with the theme ‘Plant litter quality and decomposition’ resulting in a
book summarizing the state of the art of the topic (Cadisch and Giller,
1997). Secondly, TSBF in collaboration with its national partners and
Wye College developed the Organic Resource Database (ORD) and related
Decision Support System (DSS) for OM management (Figure 2.2) (Palm
et al., 2001). The Organic Resource Database contains information on
organic resource quality parameters including macronutrient, lignin
and polyphenol contents of fresh leaves, litter, stems and/or roots from
almost 300 species found in tropical agroecosystems. Careful analysis
of the information contained in the ORD led to the development of the
30
Vanlauwe, B.
DSS which makes practical recommendations for appropriate use of
organic materials, based on their N, polyphenol, and lignin contents
resulting in four categories of materials (Figure 2.2). Recently, a farmerfriendly version of the DSS has been proposed by Giller (2000).
Figure 2.2: The Decision Support System for organic matter management
Yes
Yes
Lignin < 15%
Polyphenols
< 4%
No
Incorporate directly
Mix with N
fertilizer
or
high quality organic matter
%N
> 2.5
No
Yes
Mix with N
fertilizer
or add to compost
No
Apply at the soil surface
Lignin < 15%
Source: Palm et al., 2001
The DSS recognizes the need for certain organic resource to be
applied together with mineral inputs, consistent with the Second
Paradigm. Organic resources are seen as complimentary inputs to
mineral fertilizers and their potential role has consequently been
broadened from a short term source of N to a wide array of benefits
both in the short and long term (Vanlauwe et al., 2002a). The ISFM
paradigm has also led to increased emphasis on the social, economic,
and policy dimensions of organic and mineral input management (TSBF,
2002). In this context, it is important to note the full-time involvement
of a social scientist in TSBF and the recognition for more social input
need in AfNet.
The technical backbone of ISFM: optimal management of
organic resources, mineral inputs, and the soil organic
matter pool
Optimum management of the soil resource for provision of goods and
services requires the optimum management of organic resources, mineral
inputs, and the SOM pool (Figure 2.3). Each of these resources
contributes to the provision of goods and services individually, but more
interestingly, these various resources can be hypothesized to interact
with each other and generate added benefits in terms of extra crop
yield, an improved soil fertility status, and/or reduced losses of nutrients
to the environment.
Integrated Soil Fertility Management Research at TSBF: The Framework, the Principles,
and their Application
31
Figure 2.3: The goods and environmental services generated by the soil are the result
of the management of organic resources, mineral inputs, and the SOM pool and the
interactions between these various factors
Organic
Mineral
inputs
inputs
Goods
Services
Soil
organic
matter
Impact of individual factors on the provision of goods and
services
Numerous studies have looked at crop responses to applied fertilizer in
sub-Saharan Africa and reported substantial increases in crop yield.
Results from the FAO Fertilizer Program have shown an average response
of 750 kg maize grain ha -1 to medium NPK applications (FAO, 1989).
Value–to–Cost ratios (VCR) varied between 1.1 and 8.9, and were usually
above the required minimum ratio of 2. National fertilizer
recommendations exist for most countries, but actual application rates
are nearly always much lower to nil due to constraints of a socioeconomic rather than a technical nature. For a variety of reasons,
fertilizers are relatively expensive in SSA, certainly if compared to - often
subsidized - prices in, for example, Western Europe ($7.5 per 50 kg bag
of urea in Germany, 1999, vs $13-17 per 50 kg bag of urea in Nigeria
in, 1999, — S Schulz, personal communication, 2000). This is further
aggravated by the lack of credit schemes to purchase these inputs as
there is often a large time-gap between revenue collection from selling
harvested products and fertilizer purchase. In terms of environmental
services, mineral inputs have relatively little potential to enhance the
SOM status (Vanlauwe et al., 2001a) and may, in the case of N fertilizer,
contaminate (ground)water resources when not used efficiently. The
production of N fertilizer itself requires a substantial amount of energy,
usually derived from fossil fuels, and contributes to the CO2 load of the
atmosphere.
32
Vanlauwe, B.
In cropping systems with sole inputs of organic resources, shortterm data reveal a wide range of increases in maize grain yield compared
to the control systems without inputs (Figure 2.4). With higher soil fertility
status, the maximum increases were observed to decrease to virtually nil
at control grain yields of about 3000 kg ha-1. Although yields on fields
with a low soil fertility status, e.g., with control yields below 1000 kg ha1
, can easily be increased up to 140% after incorporation of a source of
OM in the cropping system, this would lead to absolute yields hardly
exceeding 1500 kg ha-1 (Figure 2.4). In most cropping systems, absolute
yield increases in the OM-based treatments are far below 1000 kg ha-1,
while significant investments in labour and land are needed to produce
and manage the OM. This is partly related to the low N use efficiency of
OM to be low (Vanlauwe and Sanginga, 1995; Cadisch and Giller, 1997).
Other problems related to the sole use of organic inputs are low and/or
imbalanced nutrient content, unfavorable quality, or high labor demand
for transporting bulky materials (Palm et al., 1997).
Figure 2.4: Increase in maize grain yield relative to the control in cropping systems
based on organic matter management (legume-maize rotation, alley cropping, systems
with application of external organic matter) without inputs of fertilizer N as influenced by
the initial soil fertility status, expressed as yield in the control plots. The linear regression
line shows the estimated maximal increases in grain yield. The curved lines show the
absolute yields in the treatments receiving organic matter (in kg ha-1)
160
InIncrease
crease in maize
maizegrain
grainyield
yield
1000 1500 2000 2500 3000
Legume-maize rotation
Alley cropping
140
Application of external residues
y = -0.053x + 171
120
100
80
60
40
20
0
-20
0
500
1000
1500
2000
2500
Maize grain yield of the control
-1)
control (kg
(kg ha
ha –1
)
Source: Vanlauwe et al., 2001a
3000
3500
Integrated Soil Fertility Management Research at TSBF: The Framework, the Principles,
and their Application
33
Although most of the organic resources show limited increases in
crop growth, they do increase the soil organic C status (Vanlauwe et al.,
2001a) and have a positive impact on the environmental service functions
of the soil resource. This is evidenced by the existence of steep gradients
in soil organic C status between fields at the farm scale caused by longterm site-specific soil management by the farmer (Table 2.2). Soil organic
matter is not only a major regulator of various processes underlying the
supply of nutrients and the creation of a favourable environment for
plant growth but also regulates various processes governing the creation
of soil-based environmental services (Figure 2.5). Consequently, the high
SOM status in the homestead fields is often observed to be related
positively with crop yield (Figure 2.6).
Table 2.2: Soil fertility status of various fields within a farm in Burkina Faso. Home
gardens are near the homstead, bush fields furthest away from the homestead and
village fields at intermediate distances
Field
Home garden
Village field
Bush field
Organic C
(g kg-1)
Total N
(g kg-1)
Available P
(mg kg -1)
Exchangeable K
(mmol kg -1)
11 – 22
5 – 10
2– 5
0.9 – 1.8
0.5 – 0.9
0.2 – 0.5
20 – 220
13 – 16
5 – 16
4.0 – 24
4.1 – 11
0.6 – 1
Source: Prudencio et al., 1993
Figure 2.5: Regulating nutrient supply and soil-based environmental services
Nutrient
Supply
Clean water resources;
reduced NH3 losses?
Water availability
Crops and
environmental
services
Water use efficiency
Soil structure maintenance
C sequestration
Nutrient buffering
Value of SOM
Miscellaneous (sorption,...)
No input
Systems
Nutrient
Inputs
Irrigation
Crop
production
Artif.
Support
Land use intensification
Hydroponics
34
Vanlauwe, B.
Figure 2.6: Relationship between the soil organic C content and maize grain yield for a
set of fields varying in distance to the homestead in Northern Nigeria
5
4.5
4
3.5
Compound
3
fields
2.5
Long distance
2
1.5
1
Carsky et al., 1998
0.5
0
0
1
Soil Organic C (%)
2
3
Source: Carsky et al., 1998
From the crop production point of view, the relevance of SOM in
regulating soil fertility decreases (plain horizontal arrows on figure 2.5)
as natural capital is being replaced by manufactured or financial capital
with increasing land use intensification. From an ISFM point of view,
that also considers environmental service functions besides crop
production functions, one could argue that the relevance of SOM does
not decrease (dashed horizontal arrows on Figure 2.5).
Potential interactions between the various factors on the
provision of goods and services
The Second Paradigm initiated a substantial effort on evaluating the
impact of combined applications of organic resources and mineral inputs
as positive interactions between both inputs could potentially result in
added benefits. A Direct and Indirect Hypothesis which could form the
Integrated Soil Fertility Management Research at TSBF: The Framework, the Principles,
and their Application
35
basis for the occurrence of such benefits has been formulated by
Vanlauwe et al. (2001a). The Direct Hypothesis was formulated as:
Temporary immobilization of applied fertilizer N may improve the
synchrony between the supply of and demand for N and reduce losses to
the environment. The Indirect Hypothesis was formulated for N supplied
as fertilizer as: Any organic matter-related improvement in soil conditions
affecting plant growth (except N) may lead to better plant growth and
consequently enhanced efficiency of the applied N. The Indirect Hypothesis
recognizes that organic resources can have multiple benefits besides
the short-term supply of available N. Such benefits could be an improved
soil P status by reducing the soil P sorption capacity, improved soil
moisture conditions, less pest and disease pressure in legume-cereal
rotations, or other mechanisms. Both hypotheses, when proven, lead to
an enhancement in N use efficiency, processes following the Direct
Hypothesis through improvement of the N supply and processes following
the Indirect Hypothesis through an increase in the demand for N.
Obviously, mechanisms supporting both hypotheses may occur
simultaneously.
Testing the Direct Hypothesis with 15N labelled fertilizer, Vanlauwe
et al. (2002b) concluded that direct interactions between OM and
fertilizer-N not only exist in the laboratory but also under field
conditions. The importance of residue quality and way of incorporation
in the overall size of these interactions was also demonstrated. In a
multilocational trial with external inputs of organic matter, Vanlauwe
et al. (2001b) observed added benefits from the combined treatments
in 2 of the 4 sites, which experienced serious moisture stress during
the early phases of grain filling. The positive interaction in these 2
sites was attributed to the reduced moisture stress in the ‘mixed’
treatments compared to the sole urea treatments because of the
presence of organic materials (surface and sub-surface placed) and
constitutes evidence for the occurrence of mechanisms supporting the
Indirect Hypothesis. Although more examples can be found in literature
supporting the Indirect Hypothesis, it is clear that a wide range of
mechanisms could lead to an improved use efficiency of applied external
inputs. These mechanisms may also be site-specific, e.g., an
improvement in soil moisture conditions is of little relevance in the
humid forest zone. Unravelling these, where feasible, as a function of
easily quantifiable soil characteristics is a major challenge and needs
to be done in order to optimize the efficiency of external inputs. On
the other hand, when applying organic resources and mineral fertilizer
simultaneously, one hardly ever observes negative interactions,
indicating that even without clearly understanding the mechanisms
underlying positive interactions, applying organic resources in
combination with mineral inputs stands as an appropriate fertility
management principle.
36
Vanlauwe, B.
Because SOM affects a series of factors supporting plant growth and
because of the observed within-farm variability in soil fertility and SOM
status, interest has been recently developed in relating the use efficiency
of mineral N inputs to the SOM status. A set of hypotheses follows the
general principles behind the Indirect Hypothesis outlined above and result
in positive relationships between SOM content and fertilizer use efficiency.
On the other hand, SOM also release available N that may be better
synchronized with the demand for N by the plant than fertilizer N and
consequently a larger SOM pool may result in lower use efficiencies of
the applied fertilizer N. A preliminary investigation, carried out in a longterm alley cropping trial showed a negative correlation between the
proportion of maize N derived from the applied fertilizer and the topsoil
organic C content and supports the latter hypothesis (Vanlauwe et al.,
Unpublished data). Other reports show higher use efficiency of N fertilizer
(Breman, personal communication) and P fertilizer (Bationo, personal
communication) for homestead fields with a higher SOM content.
Finally, application of organic resources is the easiest way to enhance
the SOM pool. Although it is only possible in the medium to long term to
induce substantial changes in soil organic C content in experimental
trials using realistic organic matter application rates, the above-mentioned
often drastic differences in SOM between fields within one farm prove
that farmers are already managing the SOM status. While residue quality
has been shown to significantly affect the short-term decomposition/
mineralization dynamics (Palm et al., 2001), it is unclear whether quality
is still an important modifier of the long-term decomposition dynamics.
Several hypotheses have been formulated, most of them postulating that
slowly decomposing, low quality organic inputs with relatively high lignin
and polyphenol content will have a more pronounced effect on the SOM
pool than rapidly decomposing, high quality organic inputs (Figure 2.2).
The C stabilization potential could be an equivalent index to the N fertilizer
equivalency index used to describe the short term N release dynamics.
The few trials that have shown significant increases in SOM have used
farmyard manure as organic input, which may be related to the presence
of resistant C in the manure as the available C is digested while passing
through the digestive track of the animal.
Production of organic matter in existing cropping systems:
the bottleneck in implementing ISFM practices
Although there is a wide range of potential niches to produce organic
resources within existing cropping systems (Table 2.3), introducing an
organic matter production phase in a cropping system creates problems
with adaptability and adoptability of such technologies, especially if this
fallow production phase does not yield any commercial product, such as
grain or fodder. Although a significant amount of organic matter can
Integrated Soil Fertility Management Research at TSBF: The Framework, the Principles,
and their Application
37
potentially be produced in cropping systems with in-situ organic matter
production, adoption of such cropping systems by the farmer community
is low and often driven by other than soil-fertility regeneration arguments.
Dual-purpose grain legumes, on the other hand, have a large proportion
of their N derived from biological N fixation, a low N harvest index, and
produce a substantial amount of both grain and biomass, have a great
potential to become part of such cropping systems (Sanginga et al., 2001).
Further advantages besides a substantial amount of N fixation from the
atmosphere associated with growing high biomass producing legumes in
rotation with cereal are, among others, potential improvement of the soil
available P status through rhizosphere processes operating near the rootzone of the legume crop (Lyasse et al., 2002), reduction in pest and disease
pressure by e.g., Striga spp, (iii) improved soil physical properties. These
processes yield benefits to a cereal crop beyond available N but are often
translated into N fertilizer equivalency values. Obviously, values greater
than 100% should be sometimes expected.
Table 2.3: Place and time of production of organic matter (fallow species) relative to crop
growth and the respective advantages/disadvantages of the mentioned organic matter
production systems with respect to soil fertility management and crop growth. ‘Same place’
and ‘same time’ mean ‘in the same place as the crop’ and ‘during crop growth’
Place and time of organic
matter production
- example of farming system
Advantages
Disadvantages
Same place, same time
- alley cropping
- ‘Safety-net’ hypothesis
(complementary rooting depths)
- Possible direct transfer from N2
fixed by legume species
- Potential competition
between crop and
fallow species
- Reduction of available
crop land
Same place, different time
- crop residues
- legume-cereal rotation
- improved tree fallows
- manure, derived from
livestock fed from
residues collected from
same field
- ‘Rotation’ effects (N transfer,
improvement of soil P status,...);
- Potential inclusion of ‘dual
purpose’ legumes
- In-situ recycling of less mobile
nutrients
- No competition between fallow
species and crops
- Land out of crop
production for a
certain period
- Decomposition of
organic matter may
start before crop
growth (potential
losses of mobile
nutrients, e.g., N, K,...)
- Extra labour needed
to move organic
matter (manure)
Different place
- cut-and-carry systems
- household waste
- animal manure, not
originating from same
field
- Utilization of land/nutrients
otherwise not used
- No competition between fallow
- Extra labour needed to
move organic matter
- No recycling of nutrients
on crop land
- Need for access
to extra land
- Manure and household
waste often
have low quality
Source: Adapted from Vanlauwe et al., 2001a
38
Vanlauwe, B.
In cut-and carry systems, which involve the transfer of nutrients
from one area to another, it is necessary to determine how long soils
can sustain vegetation removal before collapsing, especially soils which
are relatively poor and where vegetative production can be rapid. Cutand-carry systems without use of external inputs may be a ‘stay of
execution’ rather than a sustainable form of soil fertility management.
Of further importance is the vegetation succession that will occur after
vegetative removal. It is possible that undesirable species could take
over the cut-and-carry field once it is no longer able to sustain removal
of the vegetation of the selected species. Where an intentionally planted
species is used, the natural fallow species needs to be compared to
determine what advantage, if any, is being derived from the extra effort
to establish and maintain the planted species.
From theory to practice: Implementation of ISFM practices
at the farm level
Having focussed on the principles and technical issues underlying the
ISFM research agenda, these need to be put into the wider context this
paper started off with. This section aims at looking at ISFM options
from the farmer perspective and considers ways to disseminate these
options to the various stakeholders.
Beyond the soil: Links with other realms of capital
So far, the paper mainly focussed on the management of natural capital
with some inclusion of manufactured capital in the form of mineral
inputs. However, as stated above, farmers’ livelihoods consist of various
realms of capital which all contribute to their decision-making process
regarding soil fertility management. One obvious factor affecting the
way farmers manage their soils is related to their wealth in terms of
access to other realms of capital, such as cash, labour, or knowledge.
Rommelse (2001) reported that in a set of villages in Western Kenya,
wealthy farmers spend 102 USD on farm inputs per year compared to 5
USD for poor farmers. Besides having an overall impact on the means
to invest in soil fertility replenishment, farmers’ wealth also affects the
strategies preferred to address soil fertility decline. In two districts in
Western Kenya, Place et al. (2002) observed that wealthy farmers do not
only use more frequently mineral fertilizers compared to poor farmers,
but also a wider range of soil management practices. Farmer production
objectives, which depend on a whole set of biophysical, but also social,
cultural, and economic factors, also take into account the fertility
gradients existing within their farm boundaries. Most soil fertility
research has been targeted at the plot level, but decisions are made at
the farm level, taking into account the production potential of all plots.
Integrated Soil Fertility Management Research at TSBF: The Framework, the Principles,
and their Application
39
In Western Kenya, e.g., farmers will preferably grow sweet potato on the
most degraded fields, while banana’s and cocoyam occupy the most
fertile fields (Tittonell, personal communication).
Finally, farmers are not the only stakeholders benefiting from proper
land management. As stated earlier soils provide and regulate a series
of important ecosystem services that affect every living organism and
society as a whole and maintaining those ecosystem service functions
may be equally or more vital than maintaining the crop production
functions. Unfortunately, little information is available on the potential
trade-offs between the use of land for either of both functions, on the
most appropriate way to create a dialogue between the various
stakeholders benefiting from a healthy soil fertility status, and on the
role policy needs to assume to resolve above questions. The INRM
research approach is aiming at creating a basis for such trade-off analysis
and stakeholder dialogue.
Putting it all together: User-friendly decision aids for ISFM
After having obtained relevant information as described above, two extra
steps may be required to complete the development of a user-friendly
decision aid: (i) all above information needs to be synthesized in a
quantitative framework and (ii) that framework needs to be translated
in a format accessible to the end-users. The level of accuracy of such
quantitative framework is an important point to consider. The generation
of a set of rules of thumb is likely to be more feasible than softwarebased aids that generate predictive information for a large set of
environments. The level of complexity is another essential point to take
into consideration. For instance, if variation between fields within one
farm is large and affects ISFM practices, then this may justify having
this factor included in decision aids. Other aspects that will influence
the way information and knowledge is condensed into a workable
package are: (i) the targeted end-user community, (ii) the level of
specificity required by the decisions to be supported, and (iii) level of
understanding generated related to the technologies targeted. Van
Noordwijk et al. (2001) prefer the term ‘negotiation support systems’
because the term ‘decision support systems’ suggests that a single
authority makes decisions that will then be imposed on the various
stakeholders. In an INRM context, it is recognized that different
stakeholders may have conflicting interests related to certain specific
soil management strategies and that a certain level of negotiation may
be required.
The final format of the decision aid should take into account the
realities on the field. Some of these realities, among others, are: (i) large
scale soil analyses are not feasible, so local soil quality indicators need
to be included in decision aids as farmers use those to appreciate existing
40
Vanlauwe, B.
soil fertility gradients within a farm; (ii) conditions within farms vary as
does the availability of organic resources and fertilizer, therefore rules
of thumb rather than detailed quantitative recommendations would be
more useful to convey the message to farmers; (iii) farmers decision
making processes involve more than just soil and crop management;
and (iv) access to computers, software and even electricity is limited at
the farm level, necessitating hard copy-based products.
References
Cadisch, G. and Giller, K.E. (1997) Driven by nature: plant litter quality and
decomposition. CAB International, Wallingford, UK.
Carsky, R.J., Jagtap, S., Tian, G., Sanginga, N. and Vanlauwe, B. (1998)
Maintenance of soil organic matter and N supply in the moist savanna zone
of West Africa. In: Soil Quality and Agricultural Sustainability (Ed. R Lal),
Ann Arbor Press, Chelsea, Michigan, pp. 223-236.
DFID (2000) Sustainable livelihoods guidance sheets. The Department for
International Development, London, UK.
FAO (Food and Agriculture Organization) (1989) Fertilizers and Food Production.
The FAO Fertilizer Programme 1961-1986. FAO, Rome, Italy.
Giller, K.E. (2000) Translating sience into action for agricultural development
in the tropics: an example from decomposition studies. Applied Soil Ecology
14: 1-3.
IITA (International Institute of Tropical Agricultue) (1992) Sustainable food
production in sub-Saharan Africa: 1. IITA’s contributions. IITA, Ibadan,
Nigeria.
Izac, A-M.N (2000) What Paradigm for Linking Poverty Alleviation to Natural
Resources Management? Proceedings of an International Workshop on
Integrated Natural Resource Management in the CGIAR: Approaches and
Lessons, 21-25 August 2000, Penang – Malaysia.
Lyasse, O., Tossah, B.K., Vanlauwe, B., Diels, J., Sanginga, N. and Merckx, R.
(2002) Options for increasing P availabilitiy from low reactive Rock
Phosphate. In: Integrated Plant Nutrient Management in sub-Saharan Africa:
From Concept to Practice (Eds B Vanlauwe, J Diels, N Sanginga and R
Merckx). CABI, Wallingford, UK, pp 225-237.
Palm, C.A., Myers, R.J.K. and Nandwa, S.N. (1997) Combined use of organic
and inorganic nutrient sources for soil fertility maintenance and
replenishment. P. 193-217. In R. J. Buresh, P. A. Sanchez and F. Calhoun
(ed.) Replenishing soil fertility in Africa. SSSA, ASA, Wadington, USA.
Palm, C.A., Gachengo, C.N., Delve, R.J., Cadisch, G. and Giller, K.E. (2001)
Organic inputs for soil fertility management in tropical agroecosystems:
application of an organic resource database. Agriculture, Ecosystems and
Environment 83, 27-42.
Place, F., Franzel, S., Dewolf, J., Rommelse, R., Kwesiga, F., Niang, A. and
Jama, B. (2002) Agroforestry for Soil Fertility Replenishment: Evidence on
Integrated Soil Fertility Management Research at TSBF: The Framework, the Principles,
and their Application
41
Adoption Processes in Kenya and Zambia, in Barrett,C.B., Place, F., Abdillahi,
A.A., (eds.), Natural Resources Management in African Agriculture:
Understanding and Improving Current Practices, CABI, Wallingford, UK,
155-168.
Prudencio, C.Y. (1993) Ring management of soils and crops in the West African
semi-arid tropics: the case of the mossi farming system in Burkina Faso.
Agriculture, Ecosystems and Environment 47: 237-264.
Rommelse, R. (2001) Economic analyses of on-farm biomass transfer and
improved fallow trials in western Kenya. Natural Resources Problems,
Priorities and Policies Programme Working Paper 2001-2, International
Centre for Research in Agroforestry, Nairobi.
Sanchez, P.A. (1976) Properties and Management of Soils in the Tropics. John
Wiley and Sons, New York, USA.
Sanchez, P.A. (1994) Tropical soil fertility research: Towards the second
paradigm. P. 65-88. In Inaugural and state of the art conferences.
Transactions 15th World Congress of Soil Science. Acapulco, Mexico.
Sanginga, N., Okogun, J.A., Vanlauwe, B., Diels, J., Carsky, R.J. and Dashiell,
K. (2001) Nitrogen contribution of promiscuous soybeans in maize-based
cropping systems (Eds G Tian, F Ishida and J D H Keatinge), SSSA Special
Publication Number 58, Madison, USA, pp 157-178.
Smaling, E.M.A. (1993) An Agro-ecological Framework of Integrated Nutrient
Management with Special Reference to Kenya. Wageningen Agricultural
University, Wageningen, The Netherlands.
Swift, M.J. (1984) Soil biological processes and tropical soil fertility: a proposal
for a collaborative programme of research, Biology International Special
Issue 5, International Union of Biological Sciences, Paris, France.
Swift, M.J. (1985) Tropical soil biology and rertility: planning for research.
International Union of Biological Sciences, Paris, France.
Swift, M.J. (1986) Tropical soil biology and fertility : inter-regional research
planning workshop, Biology International Special Issue 13, International
Union of Biological Sciences, Paris, France.
Swift, M.J., Heal, O.W. and Anderson, J.M. (1979) Decomposition in Terrestrial
Ecosystems, Studies in Ecology Volume 5, Blackwell Scientific Publications,
Oxfort, UK.
Theng, B.K.G. (1991) Soil science in the tropics - the next 75 years. Soil Science
151, 76-90.
TSBF (2002) Soil Fertility Degradation in sub-Saharan Africa: Leveraging Lasting
Solutions to a Long-Term Problem. Conclusions from a Workshop held at
the Rockefeller Foundation Bellagio Study and Conference Centre, March
4th to 8th 2002.
Vanlauwe, B. and Sanginga, N. (1995) Efficiency of the use of N from pruning
and soil organic matter dynamics in Leucaena leucocephala alley cropping
in Southwestern Nigeria. P. 279-292. In R. Dudal and R. N. Roy (ed.)
Integrated plant nutrition systems. FAO, Rome, Italy.
Vanlauwe, B., Wendt, J. and Diels, J. (2001a) Combined application of organic
matter and fertilizer. In: Sustaining Soil Fertility in West-Africa (Eds G Tian,
42
Vanlauwe, B.
F Ishida and J D H Keatinge), SSSA Special Publication Number 58, Madison,
USA, pp 247-280.
Vanlauwe, B., Aihou, K., Aman, S., Iwuafor, E.N.O., Tossah, B.K., Diels, J.,
Sanginga, N., Merckx, R. and Deckers, S. (2001b) Maize yield as affected by
organic inputs and urea in the West-African moist savanna. Agronomy
Journal 93, 1191-1199.
Vanlauwe, B., Diels, J., Sanginga, N. and Merckx, R. (2002a) Integrated Plant
Nutrient Management in sub-Saharan Africa: From Concept to Practice.
CABI, Wallingford, UK, 352 pp.
Vanlauwe, B., Diels, J., Aihou, K., Iwuafor, E.N.O., Lyasse, O., Sanginga, N.
and Merckx, R. (2002b) Direct interactions between N fertilizer and organic
matter: evidence from trials with 15N labelled fertilizer. In: Integrated Plant
Nutrient Management in sub-Saharan Africa: From Concept to Practice
(Eds B Vanlauwe, J Diels, N Sanginga and R Merckx). CABI, Wallingford,
UK, 173-184.
Van Noordwijk, M., Toomich, M.T.P. and Verbist, B. (2001) Negotiation support
models for integrated natural resource management in tropical forest
margins. Conservation Ecology 5, 21.
Guidelines for Integration of Legumes into the Farming Systems of East African Highlands
Guidelines for Integration of
Legumes into the Farming
Systems of East African
Highlands
3
Amede, T1 and Kirkby, R2
1
Research Fellow, Tropical Soils Biology & Fertility Institute
of CIAT/Africa Highlands Initiative, P.O. Box 1412, code
1110, Addis Ababa, Ethiopia, t.amede@cgiar.org
2
CIAT, Pan-African Coordinator, P.O. Box 6247, Kampala,
Uganda. ciat-africa @cgiar.org
Abstract
Grain legumes are major protein sources for animals and
humans. Given that farmers export both grain and stover
from the fields, the amount of residue left to the soil is too
small to have a profound effect on soil fertility. Participatory
research was conducted to evaluate the performance of six
legume cover crops (Vetch, Stylosanthus, Crotalaria, Mucuna,
Canavalia, and Tephrosia) and two food crops (Pea and
Common bean) in southern Ethiopian Highlands, one of the
African Highlands Initiative (AHI) sites called Areka, to be
used for soil fertility improvement. Besides evaluating the
biomass productivity of legumes, the objective of this research
was to learn about the perception of farmers to LCC, feed
and food legumes, to identify socio-economic factors affecting
adoption and also to identify potential niches for their
integration. For short term fallow (three months or less),
Crotalaria gave significantly higher biomass yield (4.2 t ha-1)
43
Amede, T. and Kirkby, R.
44
followed by Vetch and Mucuna (2 t ha-1), while for mediumterm fallow (six months or more) Tephrosia was the best
performing species (13.5 t ha-1) followed by Crotalaria (8.5 t
ha-1). The selection criterion of farmers was far beyond
biomass production, and differed from the selection criteria
of researchers. Farmers identified firm root system, early
soil cover, biomass yield, decomposition rate, soil moisture
conservation, drought resistance and feed value as important
biophysical criteria. Soil moisture conservation was
mentioned as one important criterion and decreased in order
of Mucuna (22.8%), Vetch (20.8 %), Stylosanthus (20.2 %),
bare soil (17.1 %), Crotalaria (14 %), Canavalia (14 %) and
Tephrosia (11.9 %), respectively. The overall sum of farmers’
ranking showed that Mucuna followed by Croletaria are
potentially fitting species. However, Vetch was the most
preferred legume by farmers regardless of low biomass, due
to its’ early growth, high feed value and fast decomposition
when incorporated into the soil. The most important socioeconomic criteria of farmers for decision-making on which
legumes to integrate into their temporal & spatial niches of
the system were land productivity, farm size, land ownership,
access to market and need for livestock feed. These indicators
were used for the development of draft decision guides for
integration of legumes into multiple cropping systems of East
African Highlands.
Key words: Participatory research; soil degradation; legume cover crops;
integration; decision-guide
Introduction
Grain legumes are important components of the farming systems of the
East African highlands as they are the sole protein sources for animals
and humans. Besides restoring soil fertility, legumes are grown in
rotation with cereals mainly because, besides restoring soil fertility, they
also accompany the staple cereals in the local dishes. However, as
farmers export both grain yield and stover from the field, the amount of
legume residue left to the soil is too small to have a profound effect on
restoration of soil fertility.
Degradation of arable lands became the major constraint of
production in East African Highlands, due mainly to nutrient loss
resulting from soil erosion, lack of soil fertility restoring resources,
and unbalanced nutrient mining (Amede et al., 2001). However, most
farmers in the region have very low financial resources to combat
Guidelines for Integration of Legumes into the Farming Systems of East African Highlands
45
nutrient depletion, and hence research should be directed to seek
affordable and least risky, but profitable amendments necessary to
keep nutrient balance neutral (Versteeg et al., 1998). In 1999 and 2000,
researchers of the African Highlands Initiative (AHI) conducted farmers
participatory research on maize varieties on a degraded arable land in
Southern Ethiopia, Areka, by applying different inorganic sources of
fertilisers. Although the soil is an Eutric Nitisol deficient in nitrogen
and phosphorus (Waigel, 1986), high level application of inorganic N
and P did not improve maize yield. The land was highly degraded and
the organic matter was totally depleted. Lack of response to inorganic
fertilisers because of low soil organic matter content was also reported
elsewhere (Swift and Woomer, 1993). Organic inputs in the form of
green manuring or otherwise could increase the total amount of
nutrients added, and also influence availability of nutrients (Palm et
al., 1997). However, more than 50% of the organic resource available
in the region is maize stalk, of which 80% is used as fuel wood (Amede
et al., 2001). The strong competition for crop residues between livestock
feed, soil fertility and fuel wood in the area limits the use of organic
ferilizers unless a suitable strategy that builds the organic resource
capital is designed. Fallowing for restoration of soil fertility is no more
practised in the region due to extreme land shortage.
One strategy could be systematic integration of legume cover crops
into the farming system. Organic inputs from legumes could increase
crop yield through improved nutrient supply/availability and/or
improved soil-water holding capacity. Moreover, legumes offer other
benefits such as providing cover to reduce soil erosion, maintenance &
improvement of soil physical properties, increasing soil organic matter,
cation exchange capacity, microbial activity and reduction of soil
temperature (Abayomi et al., 2001) and weed suppression (Versteeg
et al., 1998). There are several studies in Africa that showed positive
effects of Legume Cover Crops (LCCs) on subsequent crops (Abayomi
et al., 2001; Fishler & Wortmann, 1999; Gachene et al., 1999; Wortmann
et al., 1994). Studies in Uganda with Crotalaria (Wortmann, et al., 1994;
Fishler and Wortmann, 1999), and in Benin with Mucuna (Versteeg et
al., 1998) showed that maize grown following LCCs produced significantly
higher yield than those without green manure. The positive effect was
due to high N and P benefits and nutrient pumping ability of legumes
from deeper horizons. However, the success rate in achieving effective
adoption of LCCs and forage legumes in sub-Saharan Africa has been
low (Thomas and Sumberg, 1995) since farmers prefer food legumes
over forage or/legume cover crops in that the opportunity cost is so
high to allocate part of the resources of food legumes to LCC. Therefore,
there is a need to develop an effective guideline that targets different
legumes types into different niches of different agro-ecologies and socioeconomic strata.
46
Amede, T. and Kirkby, R.
The objective of this paper was, therefore a) to analyse the distribution
of legumes in the perennial- based (Enset-based) systems, b) test the
performance of legumes under short term and medium term periods, c)
identify the potential causes of non-adoption of LCC, and d) develop
preliminary decision guides that could be used to integrate LCC in small
scale farms with various socio-economic settings.
Materials and Methods
Location, Climate and Soil
The research was conducted at the Gununo site (Areka), Southern
Ethiopian Highlands. It is situated on 37° 39’ E and 6° 51’ N, at an
altitude range between 1880 and 1960 m.a.s.l. The topography of the
area is characterised by undulating slopes divided by V-shaped valleys
of seasonal and intermittent streams, surrounded by steep slopes.
The mean annual rainfall and temperature is about 1350 mm and
19.5°C, respectively. The rainfall is unimodal with extended growing
periods from March to the end of October, with short dry spell in June
(Figure 3.1). The highest rainfall is experienced during the months of
July and August and caused soil loss of 27 to 48 t ha-1 (SCRP, 1996).
The dominant soils in the study area are Eutric Nitiosols, very deep
(>130 m), acidic in nature, and are characterised by higher concentration
of nutrients and organic matter within the top few centimetres of the
soil horizon (Table 3.1). These soils originated from kaolinitic minerals
which are inherently low in nitrogen and phosphorus (Waigel, 1986).
Soil fertility gradient decreases from homestead to the outfield due to
management effects. The chemical properties of the Gununo soils are
presented in Table 3.1.
Participatory evaluation of LCCs
The research site has relatively very high human population density
with an average land holding of 0.5 ha household-1. Using LCCs for
soil fertility purposes is not a common practise in the area. LCCs were
introduced into the system in 2000 following a farmers field school
(FFS) approach so as to allow farmers to learn and appreciate various
legumes uncommon to the area. The farmers research group (FRG)
was mainly composed of men, despite the repeated temptation of
researchers to include women. The legumes were planted in two
planting dates. The on-farm experiments, also used for FFS, were
planted on April 25, 2000 and July 1, 2000 and harvested on October
6, 2000 and January 6, 2001, respectively, using recommended seed
Guidelines for Integration of Legumes into the Farming Systems of East African Highlands
47
rates. The interest of the farmers was to evaluate the effect of planting
dates and length of fallow period on biomass productivity of respected
species, and to identify the best fitting legumes for a short-term fallow
(three months) or medium term (six months) fallow. Long-term fallow
became impractical due to land scarcity. Thirty interested farmers,
who were organised under one farmers research group (FRG), have
studied six different species namely, Stylosanthus (Stylosanthus
guianensis), Crotalaria (Crotalaria ochroleuca), Mucuna (Mucuna
pruriens), Tephrosia (Tephrosia vogelii), Vetch (Vicia dasycarpa) and
Canavalia (Canavalia ensiformis). All LCC were exotic species to the
system except Stylosanthus. We also included two food legumes, namely
common bean (Phaseolus vulgaris) and Pea (Pisum sativum), in the
study that were existing in the farming system.
Rainfall (mm)
Figure 3.1: Crop calendar, rainfall amount and distribution, and crops grown in the farming
system of Areka
sweet
Amede, T. and Kirkby, R.
48
The FRG studied and monitored growth and biomass productivity
in short and long seasons of 2000. The researchers were involved mainly
in facilitation of continual visits and stimulation of discussions among
farmers. Farmers and researchers were recording their own data
independently. After intensive discussion, the FRG identified six major
criteria to propose one or the other legume to be integrated into the
system. Since farmers considered soil water conservation as one
important criterion for selecting LCCs, soil water content was determined
under the canopy of each species at top 25-cm depth gravimetrically.
Sampling was done in relatively dry weeks of November 2000, five months
after planting. We considered four samples per plot, weighed immediately
after sampling, oven dried the samples at 120°C for a week before taking
dry weight. Legume ground cover was determined using the beaded
string method, knotted at 10-cm interval and laid across the diagonals
of each plot, 12 weeks after planting.
Table 3.1: Chemical Properties of Nitisols at Gununo site at the depth of top 20 cm
Soil fertility parameters
Total N (%)
Available P (ppm), Olsen
Organic matter (%)
pH (H2O)
CEC (me/100g soil)
Exchangeable cations (me 100g-1 soil)
Na+
K+
Ca2+
Mg 2+
Analytical value
0.05
7
1.2
5.9
15
0.22
0.96
14.04
2.93
Source: Waigel (1986)
In August 2002, after farmers monitored the introduced legumes,
26 farmers from four villages selected species of their choice LCC and
tested them in their farms together with a food legume, Pea. During the
growing seasons of 2000 and 2001, we monitored which farmer selected
what, how did they manage the LCCs in comparison to the food legume
and for what purpose the legumes were used. Biomass production of
the various legumes under farmers’ management was also recorded.
Besides structured questionnaire and formal survey (Pretty et al., 1995),
an informal repeated on-field discussion using transect walks were used
to identify the socio-economic factors that dictated farmers to choose
one or the other option and to prioritise the most important criteria of
decision making using pair wise analysis matrix. Moreover, farmers
Guidelines for Integration of Legumes into the Farming Systems of East African Highlands
49
invited non-participating neighbouring farmers for discussion; hence
the decision made is expected to represent the community.
We have conducted an additional replicated experiment to evaluate
biomass production of LCCs under partially controlled replicated
experiment to verify earlier obtained results. It was also meant to identify
the most promising species for short term fallow, as farmers were
reluctant to allocate land for LCCs beyond three months. The species
were planted on October 12, 2001 and harvested on January 10, 2001.
The tested species were those most favoured by farmers for further
integration namely Crotalaria (Crotalaria ochroleuca), Mucuna (Mucuna
pruriens), Tephrosia (Tephrosia vogelii), Vetch (Vicia dasycarpa) and
Canavalia (Canavalia ensiformis) replicated three times arranged in a
randomised block design. The plot size was 12 m2, with one-meter
gangway between treatments. The field was weed free throughout the
season by hand weeding. In all cases, phosphorus was applied at a rate
of 13-Kg ha-1 to facilitate growth and productivity. Data on biomass
production of the species was analysed by ANOVA using statistical
packages (Jandel Scientific, 1998).
Using the qualitative and quantitative data obtained from the site,
and by considering the hierarchy of indicators identified by farmers, we
developed draft decision guides on the integration of legumes into the
farming systems of the Ethiopian Highlands.
Crop management
The technology, green manuring, in Gununo was first tested in a
researcher/farmer- managed participatory research on farmers fields,
who were interested to try Legume Cover crops and select the appropriate
green manure species that could be adapted to their agro-ecology and
also fit into their farming systems. Seven species of legume cover crops
namely Trifolium (Trifolium quartinianum), Stylosanthus (Stylosanthus
guianensis), Crotalaria (Crotalaria ochroleuca), Mucuna (Mucuna
pruriens), Tephrosia (Tephrosia vogelii), Vetch (Vicia dasycarpa) and
Canavalia (Canavalia ensiformis) were planted on April 25, 2000 (Belg
season), July 1, 2000 (Meher season) and August 28, 2000 (Birra), to
evaluate the performance of those legumes under different planting
dates. The legumes were harvested on October 6, 2000, January 6,
2000 and March 5, 2001, for Belg, Meher and Birra planting, respectively.
The plot size was 2 x 10m and with 1m gang way between each treatment.
The recommended seed rate was used. We have also considered two
food legumes, namely common bean (Phaseolus vulgaris) and Pea (Pisum
sativum), in the study but they are already in the farming system. The
trial also served as farmers field school to introduce farmer communities
to alternative soil improving legume cover crops. The farms used for
50
Amede, T. and Kirkby, R.
Belg planting are known by the community as the most degraded, and
hence the fate of the LCCs to be accepted or rejected by the community
relied on whether the LCC could improve the productivity of those farms.
Short before maturity the LCCs of Belg planting were chopped and
incorporated after three weeks time. A sweet potato crop, cultivar Gadisa
was planted following the LCCs on October, 15, 2000 and harvested on
March 10, 2001.
As soil water conservation was considered by farmers as one of the
important criteria of LCCs, we determined soil water content under the
canopy of each species at 25 cm depth gravimetrically in the relatively
dry weeks of November, five months after planting.
In July 2000, after farmers had repeatedly visited the Belg-planted
green manure, we distributed seeds of their choice LCC together with
improved Pea variety to 19 interested farmers to see what farmers were
doing with those LCCs, where did they grow the food legume (pea) or
LCC, and the type of management they were doing for the Pea or LCCs
field. Besides structured questionnaires, we used participatory
procedures of Pretty et al (1995) for data collection and follow-up.
Results and Discussion
Land use and Soil fertility management
Farming communities in Gununo prefer to build their homes on the top
of the hills, in scattered hamlets surrounded by plantations of Enset,
also called ‘false banana’ (Enset ventricosum) and coffee. The hamlets
face towards the open communal fields, which people use for social
occasions. The Wollaytas (which also includes Gununo communities)
are reputed to be fond of trees for their own sake, growing trees and
shrubs around their farm for spices, medicine, aromatic use, shade,
farm implements and fuel wood. The farming system is a perennial
based (Enset-based system) highly intensive system with a possibility
of up to three cropping per year. Enset is a carbohydrate rich perennial
crop, with strong spurious stem and edible bulbs and corm. Unlike the
land holdings in the northern parts of Ethiopia which is characterised
by land fragmentation, land holding in Areka is consolidated. Multiple
cropping, in the form of intercropping, relay cropping and crop diversity,
are practised by farmers thanks to the long growing season with unimodal pattern of rainfall distribution (Figure 3.1). The farmers of Wollayta
have divided their land into several plots for various purposes (Figure
3.1 and Figure 3.2). Trees are planted on valley bottoms, sloppy area,
farm boundaries, in front of house and gully areas. Grazing land
(tithering) are found in front of house (Deje). Some plots are left for cut
Guidelines for Integration of Legumes into the Farming Systems of East African Highlands
51
and carry for livestock feeding. These plots have also different inherent
soil fertility status (Figure 3.2), that is soil fertility declines with distance
from houses (Eyasu, 1998).
Figure 3.2: Biomass production of various legume cover crops grown in Nitisols for
three and six months of growing period under highland conditions (n=3)
a)
b)
Vetch
Canavalia
Tephrosia
Crotalaria
6 months
old
14 12 10 8 6
Mucuna
3 months
old
4 2 0 12 10
8
6
4
2
0
-1
Dry matter production (t ha )
The major land use systems in the community include homestead
farms, (plot A in Figure 3.2) which are characterised by soils with high
organic matter content due to continuos application of organic residue.
These soils are dark brown to black in colour mainly due to high organic
matter content. This part of the farm was used to grow most important
crops such as enset, coffee, vegetables, planting materials for sweet
potato and raise tree seedlings. In the system our two years survey
showed that only about 3% of the homestead is occupied by legumes
intercropped under the enset/ coffee plants (data not presented).
Farmers are not applying inorganic fertiliser in this part of the farm.
Homestead soils (Kareta) are characterised by high organic matter
content due to continuous application of organic residue. Soils of the
neighbouring field except the Kareta types are red in colour. Even the
Kareta soils changed their colour due to organic matter application (PRA
report, 1997). Red soils are less fertile and since the organic fertiliser
sources are limited they require application of inorganic fertilisers. The
homestead field is followed by the main field (plot B), which is
characterised by red soils. Red soils are considered by the farmers as
less fertile due to limited application of organic inputs, hence require
52
Amede, T. and Kirkby, R.
application of inorganic fertiliser to get a reasonable amount of yield. In
this part of the farm, farmers grow maize in association with taro, beans
and sweet potato. This is also the part of land where legumes are growing
most (Figure 3.2). Sweet potato is also planted as sole crop in this part
of the land following long maturing maize during the small rainy season.
The outfield (plot B in Figure 3.2) is the most depleted and commonly
allocated for growing maize or potato using inorganic fertlizers. This
plot does not receive any organic manure, legumes are rarely planted
and the crop residue is even exported for different purposes. Farmers
do not practice intercropping in this part of the land.
Although legumes are the major components of the system, the
primary objective of the farmers is production of food grains as sources
of protein followed by feed production as a secondary product, and not
soil fertility. That is also partly the reason why the amount of land
allocated for legumes decreases with distance from the homestead
(decreasing soil fertility), excluding the enset field (Figure 3.2).
Participatory Evaluation of Legume Cover Crops
Seven green manuring cover crops were evaluated on-farm under three
planting dates at the beginning (Belg), in the middle (Meher) and at the
end of the growing season (Birra) of Areka, in 2000/2001. The rainfall
amount and distribution is presented in Figure 1. The rainfall distribution
was favorable and there was no extended dry spell within the growing
season of 2000 and 2001. For the medium-term fallow, Tephrosia
produced the highest dry matter biomass yield, 13.5 t ha-1 followed by
Crotalaria, 9 t ha-1 (Figure 3.2). Most of the biomass accumulation in
Tephrosia was observed four months after planting. The lowest yield
was observed from Vetch, but it showed early vigour and matured much
earlier than the other species. For the short-term fallow, Crotalaria was
the best performing species followed by Mucuna and Vetch. On individual
farmer’s field, Crotalaria was the best performing species regardless of
soil fertility. Similar results were reported from Uganda (Wortmann
et al., 1994). On the other hand, vetch and mucuna were performing
best in fertile corners of the farms. For the Belg planting, the highest
biomass yield (about 5t of dry matter ha-1) was obtained from Crotalaria
followed by Stylosanthsu and Trifolium (Table 3.2). The smallest biomass
yield was obtained from Vetch. Crotalaria was the best performing species
under this degraded soil. In the Meher planting, the highest dry matter
yield (13.5 t ha-1) was obtained from Tephrosia followed by Crotalaria.
Like that of the Belg planting, the smallest yield was obtained from
Vetch. This did not agree with the findings of Birra planting, when the
amount of rainfall sharply declined two months after planting of the
green manure, the highest dry matter yield was obtained by Crotalaria
Guidelines for Integration of Legumes into the Farming Systems of East African Highlands
53
and Mucuna (about 2.9 t ha-1 dry matter). In this experiment plants
were exposed to drought for extended period, and hence the yield
obtained from Birra planting was relatively smaller than in the other
two experiments. Although the Belg planting and Meher planting received
about equal amount of rainfall (Table 3.2) dry matter production was
the highest in Meher than in Belg planting. It could be explained by
differences in soil fertility status of the two farms, whereby the land of
Mr. Demeke (Belg planting) was highly degraded with a slope of about
18%. Unlike the results of Versteeg et al., (1998), which indicated that
mucuna performed better than other green manures (including
crotalaria) to recover completely degraded soils, our data showed that
Crotalaria performed much better than Mucuna in the degraded field of
Mr. Demeke. When those species were planted in the driest part of the
season, crotalaria and mucuna performed best and produced up to 2.9
t ha-1 dry matter within three months of time (data not presented).
Although farmers who own livestock considered Stylos and Vetch for
integration, resource-poor farmers went for Crotalaria and Canavalia.
Table 3.2: Tuber yield of Sweet potato following LCCs, Soil water content and ground
cover of Legume Cover Cropsgreen manure in an on-farm trial, 2000. Data on ground
cover (1 the least and 10 the highest) and soil water content (%) was taken when the
plants were five months old (n= 4), and soil water was determined at harvesting
Species
Canavalia
Vetch
Tephrosia
Mucuna
Crotalaria
Stylosanthus
Undisturbed soil
Mean
SED
Soil water
(%)
Ground cover
(1-10) rating
13.98
20.78
11.91
22.72
14.05
20.22
17.12
17.25
4.10
7
5
6
10
7
9
1
6.43
2.94
Besides dry matter yield, we measured soil water content under the
canopies of LCCs. The data from Meher planting showed that, the highest
soil water content was obtained from mucuna and stylosanthus, which
could be due to the self-mulching (Table 3.2). The ground cover (%) was
the highest for Mucuna (100 %), and the lowest for vetch (60%). A similar
result was obtained for mucuna in western Nigeria (Abayomi et al., 2001).
Higher soil water content under mucuna and stylosanthus (Table 3.3)
implies that these species could improve soil water availability through
reduction of evaporative loss. They also do not compete for water strongly
if grown in combination with food crops.
Amede, T. and Kirkby, R.
54
Table 3.3: Farmers’ criteria of selection of legume cover crops. According to farmers’
ranking 6 was the highest and 1 the lowest (n=25)
Species
Firm Eary soil Bio
roots cover mass
Rate of Moisture Drought
decomp- conser- resistaosition
vation
nce
Feed
value
Sum
Total
Crotalaria
2
6
6
6
2
2
2
26
Vetch
1
5
5
4
1
1
6
23
Mucuna
6
4
3
3
6
6
4
32
Canavalia
5
3
4
1
4
5
2
24
Tephrosia
3
2
2
2
5
3
2
19
Stylosanthus
4
1
1
5
3
4
5
23
In Belg planting, all tested LCC were chopped and incorporated
into the plots where they were grown. Mr. Demeke planted Sweet potato
following the green manures with the residual moisture of 2000/2001,
and obtained relatively higher tuber yield as an after effect of legumes
(Table 3.3). The best performance was observed from those planted
after Tephrosia and Mucuna followed by Canavalia & Crotalaria.
Interestingly crop yield under the best performed legume (Crotalaria)
was not the highest. Organic inputs from green manuring increased
tuber yield possibly because it could increase the total amount of
nutrients added, improve the soil-water holding capacity and also
influence nutrient availability. Palm et al., (1997) indicated that organic
fertilisers could serve (i) as sources of carbon and energy to enhance
microbial activity (ii) by controlling the net mineralisation immobilization patterns (iii) as precursors to soil organic matter
fractions and (iv) in complexing toxic cations and reducing the P
sorption capacity of the soil.
Farmers evaluated the performance of LCCs in the fields
individually or in groups through repeated visits. The selection criteria
of farmers were beyond biomass production (Table 3.3). After intensive
discussion among themselves, the FRG farmers agreed on seven types
of biophysical criteria to be considered for selection of LCCs (Table
3.3). However, the criteria of choice had different weights for farmers
of different socio-economic categories. None of the farmers mentioned
labour demand as an important criterion. They considered firm root
system (based on the strength of the plant during uprooting), rate of
decomposition (the strength of the stalk and or the leaf to be broken),
moisture conservation (moistness of the soil under the canopy of each
species), drought resistance (wilting or non-wilting character of the
leaf during warm days), feed value (livestock preference), biomass
Guidelines for Integration of Legumes into the Farming Systems of East African Highlands
55
production ( the combination of early aggressive growth and dry matter
production) and early soil cover. For resource poor farmers (who
commonly did not own animal or own few) food legumes green manure
crops with fast biomass production (Crotalaria and Mucuna) were the
best choices. For farmers who own sloppy lands with erosion problems
mucuna and canavalia were considered to be the best: Mucuna for its
mulching behaviour and canavalia for its firm root system that reduced
the risk of rill erosion. Farmers with exhausted land selected crotalaria,
as all the other legumes were not growing well in the degraded corners
of their farms. On the other hand, farmers with livestock selected
legumes with feed value and fast growth (Vetch and Stylosanths). In
general, Vetch was the most favoured legume despite low dry matter
production, as it produced a considerable amount of dry matter within
a short period of time to be used for livestock feed. It was also easy to
incorporate into the soil and found it to be easily decomposable due to
its early aggressive growth. None of the farmers mentioned labour
shortage as a potential constraint. The overall sum of farmers’ ranking,
however, showed that mucuna followed by crotalaria are the best
candidates for the current farming system of Areka. Since Mucuna is
aggressive in competition when grown in combination with other crops
(Versteeg et al., 1998) it could be used to increase soil fertility in well
established Enset/Coffee fields, while Crotalaria and Canavaia could
be used to ameliorate exhausted outfields. Canavalia is found to be
best fitting as an intercrop under maize as it has deep root system
and did not hang on the stocks of the companion crop (personal
observation). The herbaceous LCCs feed and green manure legumes
are reported as high quality organic resources (Gachene, et al., 1999)
to be used directly as organic fertilizers to improve the grain yield of
subsequent crops (Abayomi et al., 2001).
Farmers’ management of experimentation with LCCs green
manure
After thorough monitoring about the productivity and growth behaviour
of LCCs in the experimental plots, about 19 farmers have tested various
LCCs in their own farm. They tried mainly Canavalia, Crotalaria,
Mucuna, Stylosanthus and Vetch. We observed that farmers have
selected the most degraded corners of the farm for growing the LCCs
and the fertile parts of their land for growing Pea (Table 3.4). About 50%
of the trial farmers allocated depleted lands (degraded and abandoned)
for the LCC. Further discussion with farmers revealed that they took
this type of decision partly due to fear of risk, and partly not to occupy
land that could be used for growing food crops.
Amede, T. and Kirkby, R.
56
Table 3.4: Spatial niches identified by farmers for growing Legume Cover Crops or
Food legumes (Pea) in the growing seasons of 2000. Data shows number of involved
farmers (%) grew legumes at different spatial niches (n=26)
Crop type
Legume cover
crops
Pea
Sole in
fertile
soil
Sole in
degraded
soil
Relay
under
maize
Steepy
Land
0
64.29
28.6
0
7.1
35.7
14.3
0
Border abandoned
strips
land
21.43
0
21.42
0
From the total respondents, 86.6% of the farmers knew about the
role of green manures as soil fertility restorers (Figure 3.3). However
only 63% of them tested LCCs and of those who tested the green manures
only 21 % responded LCCs were effective in improving the fertility status
of the soil. About 79% believed that LCCs may not fit into their system
mainly because they did not emerge well, or showed poor performance
under depleted soils or are competing with food legumes for resources
(labour, water and, land) (Figure 3.3). This could be explained by the
fact that almost all of the farmers planted the LCCs on the degraded
corners of their farm (Table 3.4), which in turn caused low biomass
production and generally poor performance of LCCs (data not presented),
especially at the initial stage of growth.
Figure 3.3: Guidelines for identification of factors of adoption or non-adoption of legume
cover crops in multiple cropping systems of Areka
Guidelines for Integration of Legumes into the Farming Systems of East African Highlands
57
Socio-economic factors dictating guidelines for integration
of legumes
Results from PRA studies augmented by structured questionnaire
showed that there are 21 different factors that affect the integration of
legumes. When farmers were asked to prioritise the most important
factors affecting integration and adoption of legumes they mentioned a)
farm size b) suitability of the species for intecropping with other crops
for space and resources c) soil productivity of their land d) suitability
for livestock feed e) marketability of the product f) toxicity of the pod to
children and animals g) who manages the farm (self or share cropping)
h)ownership of the farm i) length of time needed to grow the species and
j) risk associated with growing LCCs in terms of introduction of pests
and diseases. Earlier works suggested that farm size and land ownership
affect integration of LCCs into small holder farms (Wortmann & Kirungu,
1999). After comparing those factors in a pair-wise analysis, four major
indicators of different hierarchy were identified (Figure 3.4).
Figure 3.4: Tools for determining degree of integration of legumes into multiple cropping
systems of Areka
58
Amede, T. and Kirkby, R.
1) Degree of land productivity: Farmers in Gununo associated land
productivity mainly with the fertility status of the soil and distance
of the plot from the homestead. The homestead field is commonly
fertile due to continual supply of organic resources. Farmers did
not apply inorganic fertiliser in this part of the farm. They remained
reluctant to allocate a portion of this land to grow LCCs for biomass
transfer or otherwise, but they grow food legumes, mainly beans, as
intercrops in the coffee and enset fields. The potential niche that
farmers were willing to allocate for LCCs is the most out field.They
are well aware of the role of legumes in crop rotation, though they
give priority to food legumes with immediate benefits. When it comes
to integration of LCCs solely for the sake of soil fertility maintenance,
farmers are unwilling to allocate the land which otherwise could be
used to grow food crops.
2) Farm size: Despite very high interest of farmers to get alternative
sources of inorganic fertilizers, the probability of farmers to allocate
land for growing LCCs depended on the size of their land holdings.
For Areka conditions, a farm size of 0.75 ha is considered large.
Farmers with very small land holdings did not grow legumes as sole
crops, but may integrate them into their system as intercrops or
relay crops. Therefore, the potential niches for LCCs are partly
occupied unless the farm is highly depleted.
3) Ownership of the farm: Whether a legume (mainly LCCs) could be
grown by farmers or not depended also on the authority of the person
to decide on the existing land resources, which is linked to land
ownership and management. Those farmers who did not have enough
farm inputs (seed, fertilizer, labour and/or oxen) are obliged to give
their land for share cropping. In this type of arrangement, the
probability of growing LCCs on that farm is minimal. Instead farmers
who contracted the land preferred to grow high yielding cereals (maize
& wheat) or root crops (sweet potato). As share cropping is an
exhaustive profit-making arrangement, the chance of growing LCCs
in such type of contract was almost nil. Without ownership or security
of tenure, farmers are unlikely to invest in new soil fertility amendment
technology (Thomas and Sumberg, 1995)
4) Livestock feed: In mixed farming systems of Ethiopia livestock is a
very important enterprise. Farmers select crop species/varieties not
only based on grain yield but also straw yield (as a crop residue or
forage) when evaluating new variety or crop. Similarly legumes with
multiple use were more favoured than those legumes that were
appropriate solely for green manure purposes.
These socio-economic criteria of farmers together with the
productivity experimental data from the field were used to develop
Guidelines for Integration of Legumes into the Farming Systems of East African Highlands
59
decision guides to help farmers in selecting legumes to be incorporated
into their land use systems as presented in Figure 3.5 and Table 3.5. As
presented in Figure 3.4, farmers considered the degree of land
productivity as the most important factor (placed at the highest
heirarchy) for possible integration of legumes. Farmers who own
degraded arable lands were willing to integrate more LCCs green manures
while those who own productive lands of large size wanted to grow food
legumes with additional feed values. However, all farmers decided to
have food legumes in their system regardless of farm size or land
productivity. Beans and Pea are already in the system and farmers
found niches to grow them as they are also part of the local dishes.
From the feed legumes, farmers favoured stylosanthus and vetch as
mentioned above. Those farmers who wanted soil improving LCCs,
selected crotalaria, as they found it better performing even under
extremely degraded conditions. However, about 45% of the farmers with
degraded arable lands are not willing to integrate LCCs, or grow green
manures either because they did not manage their own farm, and hence
share cropping /contract or have limited options of household income.
Figure 3.5: Guideline for integrating food, feed legumes and legume cover crops in
small-scale farms
Don’t own
Own
Fertile land
Large/small farm
Good market
Food & feed
legumes
Fertile land
Small farm size
Good/poor market
Food legumes
Decreasing soil fertility
with distance from
homestead
Non-fertile land
Large farm size
Good market
Food & feed
legumes, cover
crops
Non-fertile land
Small land size
Cover crops
Amede, T. and Kirkby, R.
60
Table 3.5: Tools for identification of potential legume niches for possible integration into
the multiple cropping systems of Areka developed in consultation with farmers
Position within
the farm
Land
size
Soil fertility Demand for
status
fodder
High
Large
Low
Homestead
Fertile
Small
High
Low
Intercrop under Stylosanthus
enset/coffee
Desmodium
vetch
same
Beans/pea
Intercrop under Beans/pea
ensey/coffee
same
Same
a) Intercrop
a) Beans & Peas
with maize
b) Relay under b)Vetch
maize
Low
a) Sole
b) Intercrop
under
maize
a) Beans/Pea
b)Crotalaria/
Mucuna
/Tephrosia
High
Relay crop/
short fallow
a) Relay crop
b) Intercrop
Vetch
Stylosanthus
a) Crotalaria
b)Canavalia/
Tephrosia
a) Beans/Pea
b)Vetch
Stlosanthu
Large
Less
Fertile
Low
High
Relay/Inter
Under maize
Low
High
same
Relay crop
Short fallow
Relay crop
Short fallow
Fertile
Small
Less
Fertile
Best-best
High
Fertile
Outfield
Available
niche
Low
Pea/Beans
Stylosanthus/
Mucuna
Crotalaria/
Canavalia
In general, given very high population pressure and associated land
shortage, farmers in Areka may not allocate full season for LCC, but
preferred fast growing LCCs for short term fallow. The probability of
integrating LCCs into the system became even less when the land is
relatively fertile. As the homestead Enset and darkua fields are relatively
fertile (Figure 3.2) and used for intercropping/relay cropping purposes,
growing LCC on that part of the land may not be the choice of farmers.
On the other hand, farmers with large farm size and high degree of land
Guidelines for Integration of Legumes into the Farming Systems of East African Highlands
61
degradation may go for selected LCCs. The potential niche available in
the system would be the least fertile most-out field (Figures 3.1 and 3.2)
where intercropping is not practised. The most out field is commonly
occupied by potato in rotation with maize (Figure 3.1) with relatively
less vegetative cover over the years.
The length of the growing period together with the amount and
distribution of the rainfall dictates whether the system may allow growing
legumes intercroped with maize, intercroped with perennials, or relay
cropped with maize or sweet potato. In regions, where the growing season
is extended up to eight months, and where the outfield became depleted
to sustain crop production, LCCs green manures that could grow under
poor soil fertility conditions in drought-prone months would be
appreciated. Indeed, crotalaria performed very well under such
conditions.
The Decision Guides
We are presenting three guidelines for integration of legumes into the
farming systems of multiple cropping, perennial-based systems.
Maize is the major staple crop in the region, and about 45% of the
arable land in Areka is allocated for maize. Table 3.3 shows a decision
tree developed to improve nutrient availability of Maize with decreasing
costs using organic resources in combination with inorganic fertilizers
or sole. The decision trees were developed based on the following background information from the site.
1) Farmers preferred food legumes over non-food legumes regardless
of soil fertility status of their farm.
2) The above ground biomass of grain legumes (grain & stover) is
exported to the homestead for feed and food while the below ground
biomass of grain legumes (beans and pea) is small to effect soil
fertility. The probability of the manure to be returned to the same
plot is less as farmers prefer to apply manure to the perennial crops
(Enset & Coffee) growing in the home-stead.
3) The tested legumes may fix nitrogen to fulfil their partial demand
(we have observed nodules in all although we did not quantify Nfixation), but in conditions where the biomass is exported, most of
the crop residue of legumes or green manure are used as feed
sources. Therefore, we did not expect significant effect on soil fertility.
It is not clear whether the organic resource circulates back to the
field in the form of manure.
62
Amede, T. and Kirkby, R.
4) LCCsGreen manures produced much higher biomass when planted
as relay crops in the middle of the growing season than when planted
at the end of the growing season as short-term fallows due to possible
effects of end-of season drought.
5) The homestead field is much more fertile than the outfield,; hence
those legumes sensitive to water and nutrient deficiency will do better
in the homestead than in the outfield.
The first guide (Figure 3.3) is intended to assist researchers to get
feed back information about technologies that were accepted or rejected
by the farmers or farmer research groups. This guide will assist researchers
not only to identify the major reasons for the technology to be accepted or
rejected, but also to prioritise the reasons of resistance by farmers not to
adopt the technology. This type of feed back will help to modify/improve
the technology through consultative research to make technologies
compatible to the socio-economic conditions of the community.
The second guide (Table 3.5) is intended to assist farmers and
researchers in identification of potential legumes that could be
compatible to the existing spatial and temporal niches. This guide was
developed based on the fact that the outfield is larger in size than the
homestead field, and land size, soil fertility status, feed demand and
available niches in the system (see also Figure 3.4) determined the bestbets that could fit into the current land use system.
The third guide (Figure 3.5) is developed based on the data presented
in Figure 3.4, and by taking into account the market effects. The most
important criteria at the lowest level is the presence or absence of
livestock in the household followed by who manages the farm, market
access, the size of the land holding and the land quality. The factor that
dictates the decision at the highest level was land productivity, which
was governed mainly by soil fertility status. Growing food legumes was
the priority of every farmer regardless of wealth (land size, land quality
& number of livestock). Farmers with livestock integrated feed crops
regardless of land size, land productivity and market access to products.
However, the size and quality of land allocated for growing feed legumes
depended on market access to livestock products (milk, butter and meat).
Those farmers with good market access are expected to invest part of
their income on external inputs, i.e. inorganic fertilisers. Hence farmers
of this category did not allocate much land for growing LCCs, but applied
inorganic fertilisers. In the homestead field, there was no land allocated
for LCCs in the system, not only because farmers gave priority to food
legumes, but it also became very expensive for farmers to allocate the
fertile plot of the farm for growing LCCs. The most clear spatial niche
for growing LCCs is the most out field, especially in poor farmers’ field
with exhausted land and limited market-driven farm products.
Guidelines for Integration of Legumes into the Farming Systems of East African Highlands
63
Acknowledgement
The first author would like to thank Dr Ann Stroud for her continual
support throughout the research work, Dr. Matette Bekunda for his
constructive comments, Dr. Rob Delve for improving the presentation
of the guide, Mr. Wondimu Wallelu researcher of Areka Research Centre
for their valuable inputs in the field work, and Gununo farmers for
their direct involvement in the research process.
References
Abayomi, Y.A., Fadayomi, O., Babatola, J.O., Tian, G. (2001) Evaluation of
selected legume cover crops for biomass production, dry season survival
and soil fertility improvement in a moist savannah location in Nigeria. African
Crop Science Journal 9(4): 615-627.
Amede, T. , Geta, E. and Belachew, T. (2001) Reversing degradation of arable
lands in Ethiopia highlands. Managing African Soils series no. 23., IIED,
London.
Areka PRA report (1997) Unpublished.
Eyasu, E., 1998. Is soil fertility declining? Perspectives on environmental change
in southern Ethiopia.
Managing Africa’s Soils, series no. 2, IIED, London.
Fishler, M. and Wortmann, C. (1999) Crotaelaria (C. ochroleuca) as a green
manure crop in maize-bean cropping systems in Uganda. Field Crops
Research 61: 97-107.
Gachene, C.K., Palm, C, Mureithi, J. (1999) Legume Cover Crops for soil fertility
improvement in the East African Region. Report of an AHI Workshop, TSBF,
Nairobi, 18-19 February, 1999. 26p.
Palm, C., Myers, R.J. and Nandwa, S.M. (1997) Combined use of organic and
inorganic sources for soil fertility maintenance and replenishment. SSSA
Special publication No. 51: 193-218.
Pretty, J., Guijt, I., Thompson, J., and Scoons, I. (1995) A trainer’s guide for
participatory approaches. IIED, London.
Soil Conservation Research Program (SCRP) (1996) Data base report (19821993), Series II: Gununo Research Unit. University of Berne, Switzerland.
Swift, M.J. and Woomer, P. (1993) Organic matter and the sustainability of
agricultural systems: definition and measurement. In: Mulongoy, K. and
Merck, R. (eds). Soil organic matter dynamics and sustainablity of tropical
agriculture. Wiley-Sayce, Chichester, UK. Pp. 3-18.
Thomas, D. and Sumberg, J. (1995) A review of the evaluation and use of tropical
forage legumes in Sub-saharan Africa. Agriculture, Ecosystem & Environment
54: 151-163.
Waigel, G. (1986) The soils of Gununo area. Soil Conservation Research Project
(SCRP). Research report 8, University of Berne, Switzerland.
64
Amede, T. and Kirkby, R.
Wortmann, C. Isabirye, M., Musa, S. (1994) Crotalaria ochreleuca as a green
manure crop in Uganda. Afri. Crop Sci. J. 2: 55-61.
Wortmann, C. , Kirungu, B. (1999) Adoption of soil improving and forage legumes
by small holder farmers in Africa. Conference on: Working with farmers:
The key to adoption of forage technologies. Cagayan de oro, Mindano, The
Philipines. 12-15 Oct., 1999.
Versteeg, M.N., Amadji, F., Eteka, A., Gogan, A., and Koudokpon, V., 1998.
Farmers adaptability of Mucuna fallowing and Agroforestry technologies in
the coastal savannah of Benin. Agricultural Systems 56 (3)) 269-287.
65
Effect of Organic and
Inorganic Nutrient Sources
on Soil Mineral Nitrogen and
Maize Yields in Western
Kenya
4
Ayuke, F.O.1,4*, Rao, M.R.2,3*, Swift, M.J.4
and Opondo-Mbai, M.L.1
Department of Forestry, Moi University, P.O. Box 1125,
Eldoret, Kenya
2
International Centre for Research in Agroforestry, P.O. Box
30677, Nairobi, Kenya
3
ICRISAT Colony (plot II, phase 1), Secunderabad-500 009,
Andhra Pradesh, India
4
Tropical Soil Biology and Fertility Programme, Institute of
CIAT, P.O. Box 30677, Nairobi, Kenya
1
*Corresponding address: P.O. Box 30677, Nairobi, Tel. 524000, Ext 4771,
Email: Fayuke2002@yahoo.co.uk; Fayuke@cgiar.org
Abstract
The effects of organic and inorganic fertilizers on soil mineral
N and maize yields were evaluated in a Kandiudalfic
Eutrodox soil of western Kenya.
Leaf biomass of tithonia (Tithonia diversifolia [Hemseley]
A. Grey) and senna (Senna spectabilis D.C. & H.S. Irwin) at 5
t ha-1 dry weight were incorporated into the soil and compared
with the response obtained from control without any input
Ayuke, F.O. et al
66
and fertilizer at 120 kg N, 150 kg P and 100 kg K ha-1 from
urea and triple super phosphate (TSP). Soil mineral (inorganic),
N, was measured at the beginning of the trial and subsequently
at 1, 2, 4, 8 and 12 weeks after applying the treatments. Maize
grain and stover yields were estimated at harvest.
Total inorganic nitrogen in the soil at the beginning of the
season was at a similar level in all treatments. It increased
rapidly after applying the materials and at the onset of rains
for all treatments probably because of rapid nitrogen
mineralisation in all treatments. After four weeks, inorganic
nitrogen decreased progressively until end of the experiment
in all the treatments. The highest contribution of mineral N to
the soil by the organic residues was noted at four weeks stage
and this was significantly higher with tithonia than senna.
This could be due to rapid N mineralization by these residues.
Senna treatment that had the lowest mineral N during the
first weeks of the trial, showed that N mineralization was slow
with the mineral N reaching highest level at four-week stage.
However, it is interesting to note that while soil N under tithonia
was statistically higher than in senna at four weeks, it was
higher under senna at later stage observations. Thus tithonia
decomposed completely in about four weeks, while senna was
still mineralizing at 8 weeks.
Fertilizer use increased maize grain yield by 63% over
the control. Although tithonia biomass increased maize grain
yield by 38% over the control and did not differ significantly
from fertilizer treatment, senna increased maize yield by only
6% over the no input control. Higher yield with tithonia than
senna was partly because of higher nutrient concentration
and hence greater amounts of nutrients added for the same
quantity of material applied. The study indicates that high
quality residues such as tithonia can be used as sources of
nutrients to improve crop yields.
Keywords: Biomass transfer, Tithonia diversifolia, Senna spectabilis,
mineral nitrogen, maize yield.
Introduction
Crop yields in large parts of Kenya are low due to declining soil fertility
as a result of continuous cropping and non-application of fertilizers
by farmers. For example, soils in western Kenya, (Acrisols, Ferralsols
and Nitisols)(FAO, 1965) are poor in organic matter content and have
low reserves of nitrogen (N), phosphorus (P) and some trace elements
Effect of Organic and Inorganic Nutrient Sources on Soil Mineral Nitrogen and Maize Yields
67
(ICRAF, 1994; ICRAF, 1997; Mwiinga et al., 1994; Mugendi, 1996;
Sanchez et al., 1997; Rao et al., 1998). In addition they are easily
compacted and are prone to erosion. As soon as the vegetative cover is
removed and land intensely cropped with grain crops, the soil’s
physical, chemical and biological properties are readily degraded
(ICRAF, 1993; Sanchez et al., 1997).
With the liberalization of trade and introduction of structural
adjustment programmes (SAP), fertilizer costs have increased to a level
unaffordable to small-scale farmers. How to increase and maintain crop
yields to meet the needs of the growing population has become a major
national problem. Agroforestry technologies such as short duration
planted tree fallows and green manuring (biomass transfer) with tree
residues have been demonstrated to increase crop yields (Niang et al.,
1996; ICRAF, 1997). These technologies have also been found to be
economically attractive to farmers (Sanchez et al., 1997). In the absence
of fertilizers, crop production relies largely on nutrient management
through organic residues (Vanlauwe et al., 1996; Rao et al., 1998).
In western Kenya, farmers have live fences around their farms and
grow shrub and tree hedges on contours, but rarely use the biomass
from these trees and shrubs for soil fertility improvement. Several studies
have shown that tree residues can be used as a source of nutrients to
crops (Niang et al., 1996; Palm, 1996; ICRAF, 1997). The residues serve
mainly as source of organic matter and nitrogen, but may also contribute
significant amounts of other essential nutrients. These residues upon
incorporation into the soil can help increase crop yields. For example,
experiments conducted in western Kenya, have demonstrated that higher
yields can be obtained with leaf biomass of Tithonia diversifolia (Hemsley)
A. Gray than even with commercial urea fertilizer (ICRAF, 1996; ICRAF,
1997; Rao et al., 1998). Tithonia diversifolia is a soft and succulent
shrub belonging to the family Asteraceae (Compositae), and is commonly
referred to as wild sunflower. Tithonia at 5 t ha-1 rate (on fresh weight
basis) increased maize grain yield about one and half times higher than
without inputs (Gachengo, 1996).
The capacity of any agroforestry system to enhance nutrient cycling
depends both on soil fauna, environmental conditions (e.g. temperature,
moisture, and aeration) and on management factors. Management
aspects include the selection of tree species with appropriate phenology,
rooting patterns and litter quality. Scientists need to understand the
complex interactions among the above in order to realize the potential
benefits of introducing agroforestry in a given environment (ICRAF, 1993).
In this study we seek to:
1) compare the effect of adding organic residues from agroforestry trees
and shrubs on soil mineral N to that of inorganic source of N (fertilizer).
2) assess how these inputs of organic and inorganic sources of nutrients
influence crop yields.
68
Ayuke, F.O. et al
Materials and Methods
Study site description
The study was conducted on farm near Maseno (0°6' N, 34°35' E, and
1560 m above sea level), in Vihiga District of western Kenya. The area
receives an average annual rainfall of 1800 mm in two rainy seasons;
‘long rains’ (March to July) and ‘short rains’ (September to January).
However, during 1997, a total rainfall of 2037 mm was recorded with
1200 mm in the short rains, received because of the El nino phenomenon.
Mean monthly temperature ranges between 14.6°C and 30.7°C. The
soil at the experimental site was classified as Kandiudalfic Eutrodox
(USDA, 1992). At the start of the study, the field had the following soil
physical and chemical characteristics at 0-15 cm and 15-30 cm depths
respectively: pH (1:2.5 soil water) 5.5, 5.5; organic carbon (g kg-1soil)
15.5, 14.5; extractable soil inorganic P (mg kg-1) 1.3, 0.9; exchangeable
calcium (cmolc kg-1) 4.03, 3.85; exchangeable potassium (cmolc kg-1)
0.15, 0.13; clay (%) 41, 42; sand 33%, 33%; silt % 26%, 25%; porosity
ranged between 50% and 60%. The soil is considered to be moderately
P fixing with a soil P concentration corresponding to 310 mg P kg-1
adsorbed by the soil (Nziguheba et al., 1998).
Experimental set-up and management
The present study was superimposed on an on-going larger experiment
that was initiated in 1995, during the short rain season to evaluate six
organic tree and shrub residues (Tithonia diversifolia, Lantana camara,
Calliandra calothyrsus, Senna spectabilis, Sesbania sesban and Croton
megalocarpus), as sources of nutrients in comparison with inorganic
nutrients at six different N and P levels. The treatments were replicated
four times in a randomized complete block design in plots of 7.5 m wide
and 7 m long. The present study was conducted during the 1997 short
rains with the following treatments using maize (Zea mays L.) hybrid as
the test crop:
1) control: with no external inputs (Farmers’ practice),
2) fertilizer input at: 120 kg N, 150 kg P and 100 kg K ha-1,
3) fresh biomass of Tithonia diversifolia at 5 tonnes (dry weight) ha-1
and
4) fresh biomass of Senna spectabilis at 5 tonnes (dry weight) ha-1.
The trial initially did not include a “no input” control (no N and P),
so a farmers’ no fertilizer control was randomly assigned to one of the
Effect of Organic and Inorganic Nutrient Sources on Soil Mineral Nitrogen and Maize Yields
69
unutilized blank plots in each replicate. The site was relatively flat and
there was no particular problem of runoff from plot to plot.
The amount of N and P added by the organic residues, depends on
the chemical composition. Chemical composition was determined every
season at the time of application. All the selected material contained
fairly high N and P, but differed with respect to tannin, lignin, polyphenol
levels (Table 4.1). In the fertilized plots, 120 kg N ha-1 rate was chosen
as it is close to the total N applied for the different materials ranging
between 136 kg N ha-1 to 183 kg N ha-1. The rate is also sufficient to
overcome N limitation to maize growth in these soils. The choice of the
two residues (tithonia and senna) was based on the nutrient (N and P)
concentration, plant residue quality index (PRQI)(Tian et al., 1995) and
availability in the region for potential use by farmers.
Table 4.1: Chemical composition and plant residue quality index (PRQI) of tithonia and
senna foliage
Plant residue
%N
%P
%Lignin
% Polyphenols
Senna spectabilis
Tithonia diversifolia
3.3
3.5
0.21
0.28
9.0
9.0
1.03
3.20
C/N ratio PRQI(%)
10.89
10.10
10.26
10.59
The difference between the two test materials as measured by PRQI,
has turned out to be much smaller than initially thought (Table 4.1).
However, the experience of many researchers indicate that tithonia
decomposes faster than senna and represents high quality residues
(Jama and Palm, Personal communications). In western Kenya,
particularly around Maseno area, farmers grow tithonia as a part of live
fences around their farms to mark boundaries or as hedges on contour.
Senna spectabilis trees are also common. The two residues were therefore
readily available.
Soil sampling
Soil samples were taken using 2-inch wide auger, from five different
locations at two depth intervals (0-15 cm and 15-30 cm), within each
plot. One composite sample was prepared for each 0-15 cm and 15-30
cm depths and they were analyzed for nitrate-N and ammonium-N
contents of the soil using standard methods/procedures (Anderson and
Ingram, 1993; Weaver et al, 1994). Inorganic N content was measured
at the start of season and subsequently at 1, 2, 4, 8 and 12 weeks after
treatments were applied.
70
Ayuke, F.O. et al
Maize yield measurements
Maize grain and stover yields were estimated by harvesting the four
central rows (3.0 m wide and 5.5 m long) leaving three guard rows on
either sides and one metre each on either end. Within each row, two
maize plants were left on either end as guard. The maize cobs were
harvested, weighed and sub-samples obtained. The sub-samples (about
0.5 kg from each plot) were oven-dried and the cobs threshed. The
threshing percentage was used to estimate the maize grain yield in tonnes
per hectare. The maize stover from the net plot was harvested, weighed
and sub-samples obtained. The sub-samples of stover were chopped
into smaller pieces and were then oven-dried at 70°C. The ratio of dry
weight to fresh weight and plot fresh weights were used to estimate the
maize stover yield in tonnes per hectare.
Data analyses
The data collected were subjected to analyses of variance (ANOVA), to
compare treatment effects on soil mineral N and maize yields. ANOVA
was conducted using the GENSTAT 5 Committee (1993) statistical
package. While sampling was conducted at different periods, the data
were analyzed in a split-plot design with the applied treatments as the
main plot factor and sampling period as the sub-plot factor. Treatment
differences were evaluated using the least significance difference (LSD)
at P<0.05. Standard error of difference of means (SED) was given.
Results
Effect of fertilizer and organic residues on mineral nitrogen
(N) in the soil
Total inorganic (mineral) nitrogen in the soil at the beginning of the
season was at a similar level in all treatments. It increased rapidly after
applying the materials and the onset of rains probably because of rapid
nitrogen mineralisation in all treatments. At one week stage after addition
of inputs, the highest amount of soil mineral N was observed in the
urea fertilized plot (48 kg N ha-1), followed by tithonia treated plots (40
kg N ha-1), but lowest under senna (4.5 kg N ha-1) (Figure 4.1).
After four weeks inorganic nitrogen decreased progressively until
end of the experiment in all the treatments. However, for urea-fertilized
plots, progressive decrease in mineral N was noted after one week. The
highest contribution of mineral N to the soil by the organic residues
was noted after four weeks and this was significantly higher with tithonia
Effect of Organic and Inorganic Nutrient Sources on Soil Mineral Nitrogen and Maize Yields
71
than senna. This could be due to rapid N mineralization by these
residues. However, it is interesting to note that while soil N under tithonia
was statistically higher than in senna at four weeks, it was higher under
senna at 8 weeks after application (Figure 4.1).
Increase in total inorganic nitrogen (kgNha-1)
Figure 4.1: Increase in mineral nitrogen (total mineral nitrogen in the top 0-30 cm soil
depth) above the control (no input) over 12 weeks under different inputs of organic and
inorganic nutrient sources
70
60
Fertilizer
Senna
Tithonia
SED
50
40
30
20
10
ncrease in total inorganic nitrogen (kg N ha
0
0
2
4
6
8
10
12
Sampling time (weeks after application)
SED- Standard error of difference of means.
Effect of organic and inorganic sources of nutrients on
maize yields
The treatments affected maize grain and stover yields in a similar way
(Figure 4.2).
Maize without inputs (i.e. control) produced the lowest yields of 0.8
t grain and 1.9 t stover ha-1. Application of senna residue did not increase
the yields significantly. However, addition of fertilizer and tithonia
biomass increased maize yields significantly over the control (Figure
4.2). Whereas the fertilized maize produced 1.3 t ha-1 grain and 3.1 t ha1
stover, which represented about 63% increase over the respective yields
in the control, maize yield following application of tithonia biomass
yielded 1.1 t ha-1 grain and 3.0 t ha-1 stover per ha, which represented
38% and 58% respectively over the control. The fertilizer and tithonia
treatments did not differ significantly between them. Senna treatment
increased grain yield by only 6% over the no input control.
Ayuke, F.O. et al
72
Figure 4.2: Maize yields affected by organic residues and fertilizer compared with no
inputs during 1997 short rains in western Kenya
3.5
3.0
SED
Grain
Maize yield (t ha -1)
Stover
2.5
2.0
1.5
aize yield (t ha
1.0
0.5
0
Control
Fertilizer
Senna
Tithonia
Treatment
SED- Standard error of difference of means.
Discussion
Effect of organic and inorganic fertilizer inputs on soil
mineral nitrogen
Tithonia decomposed and mineralized nutrients faster than senna
probably because of its higher N and P concentration and lower C:N
ratio (Table 4.1). The overall level of secondary compounds (lignin and
polyphenols) in tithonia and senna were low compared with foliage of
many trees and shrubs (Chesson, 1997; Palm and Rowland, 1997). It
has been shown that tithonia residue has high microbial biomass hence
higher microbial activities resulting in higher decomposition rate
(Nziguheba and Palm personal communication). The high soluble carbon
in the tissue of tithonia provides the necessary substrate for higher
microbial activity. Tithonia contains 80% water that further contributes
to rapid decomposition. Senna has comparatively high C:N ratio and
therefore soil fauna has greater role to play in its decomposition (TSBF,
1996). Thus, decomposition of senna proceeded at a slow rate because
Effect of Organic and Inorganic Nutrient Sources on Soil Mineral Nitrogen and Maize Yields
73
of its overall low quality relative to that of tithonia. Senna released more
N than tithonia towards the end of the season, i.e. it asynchronized to
plant uptake. It has been shown that the presence of a low quality
material with low N and P contents at the onset of rains extends the
time period of nutrient availability to the plants (Myers et al., 1994).
Asynchrony has undesirable effects on the crops because nutrients are
released when their demand by crops is low. The benefits of such residues
to the crop may be through the long-term build-up of N rather than the
direct use of N from the decomposing residues (Palm, 1995). Application
of senna and tithonia did not significantly affect the soil mineral N among
the treatments (Figure 4.1). Lack of treatment related differences in the
soil could be due to:
1) plant uptake of the nutrients during the growing season, and
2) loss of nutrients from the soil by leaching and also by surface runoff after the release of the nutrients following mineralization.
Nitrogen mineralization for tithonia was high at the beginning of the
trial and this decreased toward the end of the season. It is possible that
this being a high quality residue, the fauna promoted early release of N
and leaching took place at the onset of rains.
Effect of organic and inorganic fertilizer inputs on maize
yield
The yield differences among treatments could be related to N and P
availability to crops and release patterns by the organic residues. Higher
yields obtained in the fertilizer treatments could be attributed to the
nutrients being readily available from the fertilizers. Nutrients from
organic residues must first undergo decomposition before they are
available for crop uptake. In the organic residue treatments, nutrient
availability depended on nutrient concentration and release in synchrony
with crop needs. Tithonia had a higher N and P concentration and
underwent rapid mineralization, while senna, which has low
concentration of N and P, exhibited slow mineralization and/or
immobilization during early stages of maize. Maize yields with tithonia
were therefore significantly higher than with senna. Higher yields with
use of organic residue have been reported. For instance, experiments
conducted in western Kenya have demonstrated that higher yields can
be obtained when organic residues have been incorporated (Gachengo
et al., 1999; Palm, 1996). Gachengo (1996) showed that tithonia can
increase maize yields by one and half times higher than without tithonia
input. Furthermore, tithonia was found to reduce P sorption capacity of
the soil and increase crop yields particularly in P limited soils by making
P available to crops (Nziguheba et al., 1998; Palm, 1996). As the
74
Ayuke, F.O. et al
experimental site was deficient in P, the increased yield in tithonia greenmanure treatment was probably related to the combined effect of rapid
N and P mineralization and their increased availability to crops.
Phosphorus availability might have also increased through reduction of
P sorption by tithonia (Nziguheba et al., 1998).
Conclusions
This is a one-season study conducted during the 1997 short rains
(October 1997 to February 1998). The experimental period was
characterized by above normal rainfall due to El nino effect. A rainfall of
1200 mm was received during this season and therefore maize did not
grow well. Crop yields were low and variable. The results of the study
and recommendations made should be considered in the above
background of El nino effect, poor crop growth and high variability of
observations.
Based on the results of this study, foliage of tithonia is a better
organic residue for soil nutrient management than that of senna.
Resource-poor farmers who cannot afford fertilizers may be encouraged
to plant tithonia in hedges or in contours and use this organic residue
to improve the soil nutrient status.
Organic residues such as senna, which release nutrients slowly,
can be considered for long-term build up of soil fertility. It should
preferably be incorporated into the soil much before crop sowing to
synchronize nutrient release with the crop needs.
Because only two organic residues, tithonia and senna, were studied,
there is still need to investigate and test more organic residues to identify
potential alternatives to tithonia for different agroclimatic condition.
Acknowledgements
The Principal investigator was supported by ICRAF-ANAFE Postgraduate
fellowship. The authors wish to thank Ms Eva Gacheru for managing
the main field experiment and Mr. Dominic Tumbo and Mr. Romanus
Ouma for assistance in the coordination of field activities.
References
Anderson, J.M. and Ingram J.S.I. (1993) Tropical Soil Biology and Fertility: A
Handbook on Methods, CAB International, Wallingford UK.
Chesson, A. (1997). Plant degradation by ruminants: Parallels with litter
decomposition in soils. In: Cadisch, G. and Giller, K.E. (eds.), Driven by
Effect of Organic and Inorganic Nutrient Sources on Soil Mineral Nitrogen and Maize Yields
75
nature: Plant litter quality and Decomposition. CAB International, Wallingford,
U.K, pp. 47-66.
Food and Agricultural Organization (FAO). (1965) Guidelines for soil description.
Soil Survey and Fertility Branch, Land and Water Development Division.
Gachengo, C.N. (1996). Phosphorus (P) release and availability on addition of
organic materials to phosphorus fixing soils. M.Phil. thesis, Moi University,
Eldoret, Kenya.
Gachengo, C.N., Palm, C.A., Jama, B. and Othieno, C. (1999) Tithonia and
senna green manures and inorganic fertilizers as phosphorus sources for
maize in western Kenya. Agroforestry Systems 44: 21-36.
Genstat 5 Committee (1993) Genstat release reference manual, Oxford University
Press, Oxford UK.
ICRAF., (1993). International Council for Research in Agroforestry. Annual Report,
ICRAF., Nairobi, Kenya, 185pp.
ICRAF. (1994) International Council for Research in Agroforestry. Annual Report,
ICRAF, Nairobi, Kenya, 185pp.
ICRAF. (1996) International Council for Research in Agroforestry. Annual Report,
ICRAF, Nairobi, Kenya, 286pp.
ICRAF. (1997) International Council for Research in Agroforestry. Annual Report,
ICRAF, Nairobi, Kenya, 165pp.
Mugendi, D.N. (1996) Resource integration: productivity aspects of alley cropping
Calliandra calothyrsus and Leucaena leucocephala with maize in the
highlands of Kenya. Annual Report of the Tropical Soil Biology and Fertility
Programme (TSBF), Nairobi, Kenya.
Mwiinga, R.D., Kwesiga, F.R. and Kamara, C.S. (1994) Decomposition of leaves
of six mulitipupose tree species in Chipata, Zambia. Forest Ecology and
Management 64: 209-216.
Myers, R.J.K., Palm, C.A., Cuevas, E., Gunatilleke, I.U.N. and Brossard, M.
(1994) The synchronization of nutrient mineralization and plant nutrient
demand. In: Woomer, P.L. and Swift, M.J. (eds.). The Biological Management
of Tropical Soil Fertility, pp. 81-116.
Niang, A., Amadalo, B., Gathumbi, S. and Obonyo, C. (1996) Maize yield response
to green manure application from selected shrubs and tree species in Western
Kenya: A preliminary assessment. In: Mugah, J.O. (ed.), People and
Institutional Participation in Agroforestry for Susainable Development, KEFRI,
Nairobi, Kenya, pp. 35-358.
Nziguheba, G., Palm, C.A., Buresh, R.J. and Smithson, P.C. (1998) Soil
phosphorus fractions and adsorption as affected by organic and inorganic
sources. Plant and Soil 198: 159-168.
Palm, C.A. (1995) Contribution of agroforestry trees to nutrient requirements of
intercropped plants. Agroforestry Systems 30: 105-124.
Palm, C.A. (1996) Nutrient management: Combined use of organic and inorganic
fertilizers for increasing soil phosphorus availability. Annual Report of
Tropical Soil Biology and Fertility Programme (TSBF), Nairobi, Kenya.
76
Ayuke, F.O. et al
Palm, C.A. and Rowland, A.P. (1997) A minimum dataset for characterization of
plant quality for decomposition. In Cadisch, G. and Giller, K.E. (eds.). Driven
by nature: Plant litter quality and Decomposition. CAB International,
Wallingford, U.K, pp. 379-392.
Rao, M.R., Niang, A., Kwesiga, F.R., Duguma, B., Franzel, S., Jama, B. and
Buresh, R. (1998) Soil fertility replenishment in sub-Saharan Africa: New
techniques and the spread of their use on farms. Agroforestry Today, 10(2):
38.
Sanchez, P.A., Shepherd, K.D., Soule, M.J., Place, F.M., Buresh, R.J., AnneMarie, N.I., Mokwunje, A.U., Kwesiga, F.R., Ndiritu, C.G. and Woomer, P.L.
(1997) Soil fertility replenishment: An investment in Natural Resource
Capital. In: Buresh, R.J., Sanchez, P.A. and Calhoun, F. (eds.). Replenishing
Soil Fertility in Africa. Soil Science Society of America (SSSA). Special
publication, No. 51.
Tian, G., Kang, B.T. and Brussard, L. (1995) An index for assesssing the quality
of plant residues and evaluating their effects on soil and crop in the subhumid tropics. Applied Soil Ecology 2: 25-32.
TSBF. (1996) Tropical Soil Biology and Fertility programme. Annual Report, pp
48, Nairobi, Kenya.
USDA. (1992) Soil taxonomy. Agricultural Handbook no. 294, Washington D.C.
Van den Bosch, S. (1994) The influence of residue quality on nitrogen dynamics
in the litter fraction. An MSc. Thesis abstract, Catholic University Leuven,
Belgium.
Vanlauwe, B., Dendooven, L., Merckx, R., Vanlangenhove, G., Vlassak, K. and
Sanginga, N. (1996) Residue quality and decomposition under controlled
and field conditions. International Institute of Tropical Agriculture (IITA)
Research Highlights and Annual Report, 12, 1-6.
Weaver, R.W., Angle, J.S. and Bottomley, P.S. (1994) Methods of Soil Analyses:
Microbiological and Biochemical Properties, 5, 2.
Long Term Effects of Mineral Fertilisers, Phosphate Rock, Dolomite and Manure on the
Characteristics of an Ultisol and Maize Yield in Burkina Faso
Long Term Effects of Mineral
Fertilisers, Phosphate Rock,
Dolomite and Manure on the
Characteristics of an Ultisol
and Maize Yield in Burkina
Faso
Bado, B.V.1, Sedogo, M.P.2 and
Lompo, F.2
INERA, BP 910 Bobo-Dioulasso, Burkina Faso
INERA, 01BP 476 Ouagadougou 01, Burkina Faso
1
2
Abstract
The effects of soil liming, mineral and organic fertilisation
on soil characteristics and maize (Zea mays) productivity
was studied in an Ultisol in the South Savannah zone of
Burkina Faso. The experiments were carried out in Farakôba research station located in the South Sudanian zone at
11°6’ N latitude, 4°20’ W longitude and 405 m altitude.
Two experimental designs were used. In a long-term
experiment with different treatments of fertilizers, a single
rate of 571 kg ha-1 of the local Kodjari phosphate rock (PR)
containing 200 kg ha-1 CaO and 143 kg ha-1 P was applied
the first year (1983). The liming effect of organic fertilisation
was also evaluated with an application of 5 t ha-1 of manure
every two years. In the second experiment, the liming effect
of a local dolomitic limestone was tested with 3 rates of
potassium (0, 30 and 60 kg ha-1 K2O) in a split-plot design.
The results of the long term experiment showed that mineral
77
5
Bado, B.V. et al
78
fertilisers might increase maize yield from 250 to 350%
during the first 5 years indicating that nutrient deficiency
is one of the main constraints that limit crop productivity.
However, mineral fertilisers induced soil acidification and
became less efficient after 5 years of cultivation. Manure
increased and maintained mineral fertiliser effectiveness
during 6 years of cultivation. Manure had a significant
liming effect on soil acidity by increasing pH and reducing
exchangeable acidity and Al saturation. PR increased
exchangeable Ca, and base saturation. Soil liming with PR
increased P availability, maize yield and mineral fertiliser
effectiveness during 6 years of cultivation indicating that P
deficiency is an important limiting factor in this soil. The
critical soil fertility limit of available P (Bray I) for maize
using Cate and Nelson graphical method was found to be
15.9 mg kg-1 soil. Dolomite also increased base saturation
and soil pH and reduced Al saturation and exchange acidity.
However, a significant interaction between dolomite and
potassium fertiliser was observed. Dolomite effectiveness
was affected by K rate. The higher yield of maize was
obtained with 370 kg ha-1 of dolomite combined with 42 kg
K ha-1 indicating that Ca, Mg and K ratios have to be
considered when dolomite is used for soil liming.
Keywords: acidity, soil, phosphorus, manure, phosphate rock, dolomite,
maize
Introduction
Agriculture in sub-Saharan Africa (SSA) is characterized by its poor
productivity. Several factors related to soil fertility limit agricultural
production. Many factors such as soil type, farmer’s practices, crop
residues and mineral fertilizers management influence crop yields.
Alfisols, Oxisols and Ultisols dominate sub-Saharan Africa zones. The
Sahel zone of West Africa is particularly covered with sandy acidic soils
with low buffering capacities. Acidity in these soils is probably a
consequence of parent sands derived from acid continental terminal
deposits, strong paleoclimate and contemporaneous leaching and basecycling processes (Wilding and Hossner, 1989). Majority of West African
soils belong to the Alfisols soil order according to the United State Soil
Taxonomy and Ferruginous tropical group in the French classification
(Pieri, 1985). In the savannah zone with low rainfall (500 - 1000 mm
annually), base leaching is limited; hence soils have relatively high soil
pH and base saturation (Ssali et al., 1985). Particularly Alfisols of
Long Term Effects of Mineral Fertilisers, Phosphate Rock, Dolomite and Manure on the
Characteristics of an Ultisol and Maize Yield in Burkina Faso
79
Savannah zone have a low inherent acidity (pH 6.0 to 6.5). Considering
the pH values, West African soils are not excessively acid. However, soil
acidity may rapidly increase with farmers cultural practices (Pieri, 1985).
Traditional fallow system reduction due to population growth, intensive
cropping, nutrient losses by erosion and runoff, cations Ca2+ and Mg2+
losses with nitrate leaching, all tend to induce acidification in lowbuffered soils. So, the acidification of cultivated soils and P deficiency
due to high P fixation could significantly affect crop yields.
Considering P deficiency and soil acidification induced by farmers
practices, organic amendments, rock phosphates and dolomite are
interesting alternatives that could be exploited to improve traditional
farming system productivity in SSA. The objective of this study was to
test PR and dolomite ability to alleviate both P deficiency and soil acidity
constraints.
Materials and Methods
Two trails were established in the research station of Farakô-ba in
Burkina Faso. The site is located in the South Sudanian zone at 11°6' N
latitude, 4°20’ W longitude and 405 m altitude. The average annual
rainfall varies from 900 to 1000 mm. Ultisols and Alfisols are the main
soil types of Farakô-ba. The two experiments were established on Ultisols.
These soils have a low inherent acidity (pH 6.0 to 6.5), which may rapidly
increase with cultural practices. The major properties of the soil are
presented in Table 5.6. Before the establishment of the experiments,
the land was under several years of fallow.
The dolomite effectiveness was studied in a randomised block design
in a split-plot arrangement with six replications. The main plot treatments
were four levels of dolomite (0, 100, 200 and 400 kg CaO ha-1). The
potassium was applied in the sub plots. Three levels of K (0; 21 and 42
kg K ha-1) were applied. All plots received 16 kg P ha-1 in the form of
triplesuperphosphate except the control. A uniform rate of 90 kg N ha1
was applied on all subplots except the control. The N was applied in
the form of urea and was split. Three splits were applied, with one third
at the sowing, one-third 30 days after planting (DAP) and the last third
60 DAP. An improved maize variety SR 22 (120 days) was used at the
recommended planting density of 62500 plants ha-1.
PR liming effect was studied in a long-term experiment started in
1983. The experimental design was a randomised complete block design
in a split-plot treatment arrangement with six replications. The main
plot treatment was six levels of mineral, organic and organo-mineral
fertilisers. The PR was applied in the sub plots. Two levels of mineral
fertilisers (weak annual mineral fertilizer: 60N-10P-10K-6S-1B (fm) and
high annual mineral fertiliser: 90N-15P-36K-9S-1.5B (FM) was applied
Bado, B.V. et al
80
alone or in combination with 5 tonnes ha-1 of manure every two years
(fmo and FMO). The nutrients P, K, S and B were applied as NPKSB
fertiliser and KCl. Nitrogen in the form of urea and NPKSB fertiliser was
split. Three splits were applied, with one third at the sowing, one-third
30 days after planting (DAP) and the last third 60 DAP. Each main plot
was split in two subplots and one of them received a basal application
(1983) of 571 kg ha-1 of PR corresponding to 200 kg ha-1 CaO and 62 kg
ha-1 P. The main characteristics of PR are presented in Table 5.1. On
the two experiments all fertilisers were broadcast and incorporated.
An improved maize variety, IRAT 171 (120 days) was used at the
recommended planting density of 62,500 plants ha-1. The dates of
planting varied according to the start of the rains. In general, plating
was in June and harvesting occurred in November. The monthly rainfall
distribution during the experimentation is showed in Table 5.2. Maize
gain and stover yield were measured. Consistent with traditional
practices, the crop residues were removed each year.
Table 5.1: Element content (%) of phosphate rock and dolomite
Phosphate rock *
Dolomite
25.5
3.1
3.4
34.5
2.5
26.24
0.27
35.5
19.0
P2O5
Al solubility (HCl)
Fe2O3solubility HCl
CaO
F
SiO2
MgO
* Source: Mc Clellan et al ( 1986 )
Table 5.2: Monthly rainfall (mm) distribution during the crop cycle of the experiments
Month
Year
May
June
July
August
September
October
Total
1983
1984
1985
1986
1987
1988
1989
122
102
112
118
43
83
59
104
104
290
83
151
99
126
166
123
272
215
199
194
155
194
274
429
233
372
196
366
131
157
169
168
74
305
144
3
13
57
53
22
62
41
720
773
1329
870
861
939
891
In 1989, soil samples were taken after harvest from the top 20 cm
depth of all subplots for chemical characterisation. Organic carbon was
measured by the procedure of Walkley & Black (1934). Soil pH was
measured in 1 N KCl using 2:1 solution to soil ratio and exchangeable
Long Term Effects of Mineral Fertilisers, Phosphate Rock, Dolomite and Manure on the
Characteristics of an Ultisol and Maize Yield in Burkina Faso
81
acidity was measured using McLean method (1982). Exchangeable bases
(Ca, Mg and Na) were displaced with NH4O. Ca and Mg were determined
by atomic absorption spectrometry, while K and Na were determined
using flame photometry. The data were analysed as split-plot with
SYSTAT using analysis of variance.
Results and Discussions
The effects of fertilisers and PR on maize yield are presented in Tables
5.3 and 5.4. Compared to the control, all treatments increased the maize
yields. All Fertilisers improved maize grain and stover yields during the
eight years of experimentation. The mineral fertilisers highly increased
maize grain and stover yields particularly during the first four years. The
highest yields of maize were obtained with the application of mineral
fertilisers associated with organic manure. As showed by other works
(Berger et al., 1987; Pichot et al, 1981; Sedogo, 1981; Bationo and
Mokwunge 1991; Bado et al., 1997) these high responses to fertilisers
may be explained by the poverty in nutrients and the low content in
organic mater of west African weakly acidic Ultisols. So, all additions of
fertilisers or high quality organic mater can significantly increase crop
yields. As showed on Table 5.6, mineral fertilisers reduced soil base
saturation and pH. They increase exchangeable acidity and Al saturation.
Mineral fertilisers not only increased nutrients availability in soil but
also increased soil acidity at the same time. This acidification effects of
the mineral fertilisers are reduced or suppressed when manure was
simultaneously applied with mineral fertilisers (Table 5.6), explaining
the beneficial effect of organic and mineral fertilisers on soil fertility and
maize yields. Similar results relative to the beneficial effect of the
simultaneous application of organic amendments and mineral fertilisers
on crop yields were also obtained by Sedogo (1981) and Bado et al. (1997).
The basal application of PR in the first year (1983) had a significant
effect (<0.01) on maize grain yield, particularly during the first two years
(Table 5.3). The PR also increased maize stover yield during the first
three years (Table 5.4). The beneficial effect of the PR on maize yields
may probably be due to it’s effect on P availability and soil acidity. PR
application involved an increasing of soil available P (Table 5.6). As
indicated by the relationship between soil available P and maize yield
calculated in 1989 using the data of all treatments, maize yield was
significantly affected by soil P availability (Figure 5.1). Maize yields and
P-Bray are related by positive correlations (<0,01) indicating that 94 %
of maize yield variations were due to soil P availability. By using the
graphical methodology of Cate and Nelson (1965) we saw that the critical
limit of P-Bray I for maize production in this soil was 15.9 mg P kg-1 and
16.5 mg P kg-1 respectively for grain and stover yield indicating that P is
Bado, B.V. et al
82
an important limiting factor for maize yield in this soil as shown by
Saharawat et al (1997) in Côte d’Ivoire.
Table 5.3: Effect of organic and mineral fertilizers and basal application of rock phosphate
(only in 1983) on maize grain yield over six years (1983-1989)
Organic and mineral fertilisers
Year
1983
1984
1985
1986
1987
1988
1989
Control
Weak
mineral
fertiliser
(fm)
High
mineral
feriliser
(FM)
fm +
5 t ha-1
manure
FM +
5 t ha-1
manure
0 PR
571 kg PR
PR (a)
Fertiliser(b)
a*b
505
1353
1766
2413
2159
1889
*
**
*
1566
1775
2175
2230
0
PR
PR*(a)
Fertiliser(b)
a*b
294
1185
2036
2730
1901
3089
**
**
ns
1812
2397
2486
3240
0
PR
PR*(a)
Fertiliser(b)
a*b
847
1386
2276
2247
2994
2903
NS
**
ns
2861
3224
3242
3641
0
PR
PR*(a)
Fertiliser (b)
a*b
679
1263
2413
2233
2543
2985
NS
**
ns
2758
2873
3538
3428
0
PR
PR*(a)
Fertiliser (b)
a*b
453
1022
2285
2310
2796
3223
NS
**
ns
1673
1867
2857
2708
0
PR
PR*(a)
Fertiliser (b)
a*b
64
284
1108
1160
1734
2438
**
**
ns
1308
1745
2923
3024
0
PR
PR*(a)
Fertiliser (b)
a*b
35
227
472
657
869
878
NS
**
ns
479
733
1069
1203
*, **, ns: indicate significant at 0.05, 0.01 probability or not significant (> 0.05)
Long Term Effects of Mineral Fertilisers, Phosphate Rock, Dolomite and Manure on the
Characteristics of an Ultisol and Maize Yield in Burkina Faso
83
Table 5.4: Effect of organic and mineral fertilizers and basal application of rock phosphate
(only in 1983) on maize stover yield over six years (1983-1989)
Organic and mineral fertilisers
Year
1983
1984
1985
1986
1987
1988
1989
Control
Weak
mineral
fertiliser
(fm)
High
mineral
feriliser
(FM)
fm +
5 t ha-1
manure
FM +
5 t ha-1
manure
0 PR
571 kg PR
PR (a)
Fertiliser(b)
a*b
1595
3424
3328
4581
3762
4292
**
**
**
3086
3617
4099
3906
0
PR
PR*(a)
Fertiliser(b)
a*b
1109
1591
3376
4003
2893
3810
**
**
ns
2894
3231
3906
3762
0
PR
PR*(a)
Fertiliser(b)
a*b
1254
2083
4128
4109
4417
4900
*
**
ns
4552
4687
4630
5112
0
PR
PR*(a)
Fertiliser (b)
a*b
-
-
-
-
-
0
PR
PR*(a)
Fertiliser (b)
a*b
1234
2180
4687
4649
4784
5035
NS
**
ns
3974
3569
4321
5343
0
PR
PR*(a)
Fertiliser (b)
a*b
249
532
1742
1835
2656
3351
NS
**
NS
2098
2264
3583
3313
0
PR
PR*(a)
Fertiliser (b)
a*b
260
583
1138
1205
1428
1755
**
**
ns
776
1635
1726
2030
*, **, ns: indicate significant at 0.05, 0.01 probability or not significant (P> 0.05)
Bado, B.V. et al
84
On the soil acidity parameters, the basal application of PR
significantly reduced aluminium saturation (Table 5.6). The basal
application of PR in 1983 had a residual effect on the reduction of Al
saturation until 1989. By using the data of all treatments we found that
soil pH and soil exchange acidity are related by a significant (P<0.05)
exponential relationship (Figure 5.2).
Figure 5.1: Relationship between P-Bray I and maize grain yield
3500
-1
Yield (kg
(kgha
ha-1)
523,74
- 568,6
) = =523.74
PP
- 568.6
Yield
2
= 0,94
RR2
= 0.94
3000
Grain (kg ha-1)
2500
2000
1500
1000
500
0
0
1
2
3
4
5
6
7
P (mg kg-1 soil)
Figure 5.2: Relationship between soil pH and exchangeable acidity
0.9
Exchangeable acidity (meq/100g)
0.8
0.7
-2,4724pH
0.6
EA EA
= 5307,8e
= 5307.8e-2.4724pH
0.5
R = 0,89
R2 = 0.89
2
0.4
0.3
0.2
0.1
0
3.5
4
4.5
pH-KCl
5
Long Term Effects of Mineral Fertilisers, Phosphate Rock, Dolomite and Manure on the
Characteristics of an Ultisol and Maize Yield in Burkina Faso
85
The effects of dolomite and potassium on maize yields are presented
in Table 5.5. Dolomite increased maize yields. The efficiency of dolomite
was affected by potassium rates. When potassium wasn’t applied, the
highest yield of maize was obtained with 200 kg CaO ha-1. Potassium
had a significant effect on maize yield when it was applied at 42 kg K
ha-1 indicating that Ca, Mg and K ratios are to be considered for a better
use of dolomitic limestone as observed by Bado et al. (1993). The best
combination of dolomite and potassium providing highest yield is 100
kg ha-1 CaO as dolomite combined with 42 kg K ha-1.
Table 5.5: Effect of dolomite and potassium on maize grain yield (kg ha-1)
Potassium (kg K ha-1)
Dolomite
0 kg ha-1
21 kg ha-1
42 kg ha-1
3557 a
3831 a
4286 b
4140 b
3284 a
3509 a
4293 b
4234 b
3143 a
4736 b
3309 a
4493 b
(kg CaO ha -1)
0
100
200
400
Yields affected by the same letter are significantly different (P< 0.05).
Table 5.6: Effects of organic and mineral fertiliser and phosphate rock on soil
characteristics after 9 years of maize cultivation.
P-Bray I pHKCl
Control
4.9
Control + PR 10.5
4.2
4.2
Al+H
Al sat.
(%)
Ca
Mg
K
0.22
0.16
11
8
0.38
0.40
0.32
0.28
0.12
0.12
0.98
0.96
1.20
1.12
0.32
0.33
0.21
0.17
0.11
0.10
0.82
0.77
1.14
0.97
0.18
0.33
0.19
0.21
0.14
0.16
0.68
0.86
1.06
1.20
0.32
0.41
0.27
0.27
0.14
0.16
0.91
1.00
1.13
1.18
0.32
0.20
Bases CEC
fm
fm + PR
13.1
15.7
FM
FM+PR
19.4
23.2
fmo
fmo+PR
14.5
25.1
FMO
FMO+PR
16.1
24.4
4.0
4.1
0.24
0.20
13
8
0.30
0.39
0.26
0.29
0.23
0.18
0.96
1.00
1.20
1.23
Original Soil
4.4
4.3
-
9
0.63
-
-
1.48
1.71
3.8
3.8
0.38
0.35
26
20
0.22
0.17
86
Bado, B.V. et al
Conclusion
In this weakly acid soil, low organic matter content and low nutrient
content soil, particularly for P, mineral and organic fertilisers are the
main constraints limiting maize yield. This explains the good response
of maize to mineral fertilisers and organic manure applications. Mineral
fertilizers alone may induce soil acidification, a decrease in exchangeable
cations and an increase in aluminium dissolution. Thus, the decrease
in productivity is associated to soil acidification as a consequence of
soil organic matter declining due to long-term cultivation and crop
residues exportation, bases absorption by plants and bases leaching
over years. To solve these problems, an economic solution may be to
use local agro mineral resources (rock phosphate and dolomite) to supply
P and to correct soil acidification over time.
References
Bado, B. V., Dakyo, D., N’Dayegamiye, A. and Cescas, M. (1993) Effet de la
dolomie sur la production et les propriétés chimiques d’un sol ferralitique.
Agrosol. VI (2): 21-24.
Bado, B. V., Sedogo, M. P., Cescas, M. P., Lompo, F. and Bationo., A. (1997)
Effet à long terme des fumures sur le sol et les rendements du maïs au
Burkina Faso. Agricultures. Vol. 6 No 6 : 571 - 575.
Bationo, A. and Mokwunye, A.U. (1991) Role of manures and crop residue in
alleviating soil fertility constraints to crop production with special reference
to the Sahelian and Sudanian zones of West Africa. In: A. U. Mokwunye
(Ed.) Alleviating soil fertility constraints to increase crop production in West
Africa.. Kluwer Academic Publishers, 217 - 225.
Berger, M., Belem, P. C., Dakouo, D., and Hien, V. (1987) Le maintien de la
fertilité des sols dans l’Ouest du Burkina Faso et la nécessité de l’association
agriculture élevage. Cot. et Fib. Trop.; vol. XLII Fasc 3 pp 10.
Cate, R.B., and Nelson, L.A. (1965) A rapid method for correlation of soil test
analyses with plant response data. North Carolina State Univ. Int. Soil
Testing series Tech. Bull. 1.
Pichot J., Sédogo, M. P., and Poulain, J.F. (1981) Évolution de la fertilité d’un
sol ferrugineux tropical sous l’influence des fumures minérales et organiques.
Agronomie Tropicale N° 36: 122-133.
Pieri, C. (1985) Management of Acid Tropical Soils in Africa. In: Management of
Acid Tropical Soils for Sustainable Agriculture. 41-61, Proc. Of an IBSRAM
Inaugural Workshop. Yurimaguass, Peru.
Sahrawat, K. L., Jones, M. P., and Diatta, S., (1997) Extractable phosphorous
and rice yield in an Ultisol of the humid forest zone in West Africa. Comm.
Soil Sci. Plant Anal. 28: 711-716.
Long Term Effects of Mineral Fertilisers, Phosphate Rock, Dolomite and Manure on the
Characteristics of an Ultisol and Maize Yield in Burkina Faso
87
Sédogo, M.P. (1981) Contribution à l’étude de la valorisation des résidus
culturaux en sol ferrugineux et sous climat tropical semi-aride. Matière
organique du sol, nutrition azotée des cultures. Thèse de Docteur Ingénieur,
INPL Nancy. 135 P.
Ssali, H., Ahn, P.M. and Mokwunye, U. (1985) Fertility of soils of tropical Africa:
a historical perspective. In: Mokwunye, A.U., Vlek, L. G., (Eds), Management
of nitrogen and phosphorus fertilizers in Sub-Saharan Africa. 59-82.
Walkley, A., and Black, J.A. (1934) An examination of the Detjareff method for
determining soil organic matter and a proposed modification of the chromic
acid titration method. Soil Science 37: 29-38.
Wilding, L. P., Hossner, L.R. (1989) Causes and effects of acidity in Sahelian
Soils. In: Soil, Crop and Water Management in the Sudano Sahelian Zone.
215-227. Proc of International Workshop. ICRISAT, Sahelian Center, Niamey,
Niger.
88
Bado, B.V. et al
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
Changes in Soil Properties
and their Effects on Maize
Productivity Following
Sesbania Sesban and
Cajanus Cajan Improved
Fallow Systems in Eastern
Zambia
89
6
Chirwam T. S.1*, Mafongoya, P. L.2,
Mbewe, D. N. M.3 and Chishala, B. H.3
Msekera Research Station, P.O. Box 510089, Chipata,
Zambia
2
ICRAF-Zambia Agroforestry Project, P.O. Box 510046,
Chipata, Zambia,
3
University of Zambia, School of Agricultural Sciences, P.O.
Box 32379, Lusaka, Zambia.
1
*Corresponding author’s Tel/fax 26 062 21725/21404 and E-mail:
zamicraf@zamnet.zm
Abstract
Changes in soil properties and their effects on maize
productivity following Sesbania sesban and Cajanus cajan
improved fallow system were measured on a Typic kandiustalf
in eastern Zambia. The treatments used in the study were
two-year planted improved fallows of Sesbania sesban (L.)
90
Chirwa, T. S. et al
Merr. (sesbania) and Cajanus cajan (L.) Millsp (pigeonpea);
natural fallow and continuous fertilized (m+f) and unfertilized
maize (m-f) (Zea mays L.) mono culture. At the end of 10week incubation period, the cumulative N mineralization of
sesbania (fresh leaves + litter), reached 59.4 mg N kg-1 soil
as compared to 5.1 mg N kg-1 soil for pigeonpea litter. Grass
fallow litter had a cumulative net immobilization of 0.8 mg N
kg-1 soil. Maize with fertilizer had the highest pre-season soil
nitrate-N at all soil depths. A polynomial regression model
between maize grain yield and pre-season inorganic nitrateN for 0-20 cm, 0-40 cm and 0-60 cm soil layers showed that
the amount of pre-season inorganic nitrate-N in the soil layer
accounted for 71%, 68% and 71%, respectively of the maize
yield. Total inorganic N in the top 0-20 cm soil depths was in
the order of: m+f > cajanus > sesbania > m-f > natural fallow.
As was the case with pre-season soil nitrate-N, total inorganic
N in 0-20 cm, 0-40 cm and 0-60 cm soil depths was
significantly correlated to grain yield (R2 = 0.70, 0.67 and
0.71). DM accumulation ranged from 0.2 t ha-1 to 9.5 t ha-1
for m-f (at 4 WAP) and m+f (at 24 WAP), respectively. The
maximum N accumulation in maize tops at 24 weeks after
planting (WAP) averaged 156.9 kg N ha-1 and 77.0 kg N ha-1
for m+f and sesbania land use system (LUS), respectively,
with grain yields of 5.51 and 3.02 t ha-1, correspondingly.
The lowest penetrometer resistance measured at 4 WAP for
0 -40 cm soil depth was recorded in sesbania LUS (2.2 Mpa).
On the other hand fertilized maize had the highest resistance
of 3.9 Mpa. The highest percentage of water stable aggregates
> 2.00 mm at fallow termination was recorded in sesbania
LUS (83.3 %), followed by pigeonpea LUS (80.8 %). At crop
harvest the highest percentage of water stable aggregates >
2.00 mm was recorded in pigeonpea LUS (76.9 %), followed
by natural fallow LUS (65.8 %). At fallow termination, the
average cumulative water intake after 3 hours was 233, 315,
465, 485 and 572 mm for continuous maize without fertilizer,
continuous maize with fertilizer, sesbania, pigeonpea and
natural fallow, respectively. Soil water sorptivity at fallow
termination was in the order of pigeonpea > natural fallow >
sesbania > m+f > m-f. On the other hand soil water sorptivity
at crop harvest was in the order of sesbania > natural fallow
> pigeonpea > m+f > m-f. At crop harvest the average
cumulative water intake at 3 hours was 173, 184, 221, 246
and 399 mm for unfertilized maize, fertilized maize,
pigeonpea, natural fallow and sesbania, respectively. The
improved soil condition and nitrogen contribution of sesbania
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
91
and pigeon pea fallows to subsequent crop was evidenced by
increased maize yields after these fallows as compared with
no tree treatments. Mixing of litter (low quality) with fresh
leaves (high quality) from the same tree species at fallow
termination had an effect on maize N uptake. Therefore there
is need to carefully manipulate the quantities of materials
(fresh leaves and litter) at fallow termination so as to get the
maximum N utilization by maize plants in improved planted
fallow systems.
Key words: Mineralization, immobilization, stable aggregates, penetration
resistance, cumulative water intake
Introduction
Under traditional farming methods, farmers have relied on short natural
or shrub fallows to grow maize and other crops. In eastern part of Zambia
this fallow system is known locally as 'cisala' (Kwesiga et al. 1997). Nye
and Greenland (1960) also reported that natural fallows have long been a
way to overcome soil fertility depletion that results from continuous
cropping with no nutrient inputs. The fallow period may vary from five to
twenty years. However, long fallow periods have become impractical
because of increasing human and livestock populations. Losses of mineral
nutrients during the cultivation phase, through runoff, erosion, leaching
and crop removal, can no longer be restored by short periods of bush
fallow (Brady, 1996). The processes of natural soil fertility restoration are
not completed with bush short duration fallows of between 1-5 years and
this has necessitated the need for improved fallows.
Intensive cultivation and cropping may have negative effects on the
chemical, physical, and biological properties of the soil due to the induction
of changes in temperature, water, and aeration fluxes, decreasing organic
matter content and increasing aggregate disruptions and soil erosion
(Migliena et al. 1988). Nitrogen limits crop production over large areas of
Zambia and the main sources of plant-available N are mineralization of
soil organic matter (SOM), biological N2 fixation, fertilizers and organic
inputs (e.g., plant residues, composts and manures (Giller et al. 1997).
The improvement in soil physical properties could be another reason for
yield improvement but little quantitative data exist on these changes.
Recent reviews (Rhoades, 1997 and Young, 1997) on the soil improvement
effects of trees have largely concentrated on studies of soils under forest
stands or along transects under individual trees. Studies by Mafongoya
and Nair (1997) under field conditions showed that lignin, polyphenols
and nitrogen content had a significant effect on N release and maize
yield. Research on mixing of legume tree prunings from different species
of high quality with low quality has been done by many researchers
92
Chirwa, T. S. et al
(Handayanto et al. 1995 and Mafongoya et al. 1997). The residual effect
on nutrient release and long-term changes in soil fertility resulting from
mixing of prunings of different quality from the same tree legume species
is a subject, which has received little attention to date.
Soil physical properties, such as aggregate stability and infiltration,
are difficult, time consuming, and expensive to measure, hence their
importance often receives insufficient research attention. Whilst the
response of maize growth and yield in improved fallow systems has
received much attention, the processes in tree and post fallow phase
have not been understood. Therefore, the objectives of the study were:
1) To quantify some changes in soil properties that are responsible for
improvement in crop productivity under fallow cultivation systems
compared to the continuously cropped maize system. 2) To quantify the
nitrogen mineralization patterns of mixing litter and fresh leaves from
the same tree species.
Materials and Methods
The study was conducted in Eastern province of Zambia at Msekera
research station during 1996/97 to 1998/99 season. Msekera research
station is situated between latitudes 13038' S and longitudes 32034' E.
The soils at experimental site in 0-20 cm soil layer are composed of
1.2% carbon content, pH (CaCl2) of 4.5, 25% clay, 67% sand and receives
an average rainfall of 1092 mm per annum. In general, the surface
texture for the experimental site is sandy clay loam with reddish brown
top and subsoils, classified as Typic kandiustalf (USDA, 1975) or Haplic
luvisols (FAO, 1988).
A randomised complete block design (RCBD) comprising of five landuse systems (LUS) replicated three times was used, with gross plots of 10
m x 10 m. The LUS were Sesbania sesban (L.) Merr. (sesbania) and Cajanus
cajan (L.) Millsp (pigeonpea); natural fallow and maize (Zea mays L.)
monoculture with and without fertilizer. Sesbania sesban (prov. Chipata
dam) fallow trees were planted from nursery raised bare rooted seedlings
at the age of 5 weeks at a spacing of 1.0 m x 1.0 m (10 000 plants ha-1).
While Cajanus cajan (cv. ICP 9145) was direct seeded in the plots at the
same time the Sesbania sesban seedlings were transplanted into the field
in November 1996 at a spacing of 1.0 m x 0.50 m (20 000 plants ha-1).
Trees were clear felled at collar (ground) level in November 1998 after 2
years of fallow, while stumps and root system were left below ground.
Data collection and observations
Total above ground biomass of trees (leaves, twigs and wood) was
measured at fallow clearing by separating the biomass components into
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
93
foliage (leaves and twigs), branches and stems. These components were
then weighed as green after which samples of each component were
collected on plot basis and oven dried at 70°C to equilibrium moisture
content. This data was used to estimate dry weight on plot basis and
extrapolated to a hectare basis. The tree biomass (leaf + twig) and natural
grass fallow of hyparrhenia sp. was incorporated in the soil by hand
hoeing. After land preparation, hybrid maize (Zea mays L.) (variety MM
604) was sown by hand at 30 cm within-row and 75 cm between-row
spacing (44 444 plant ha-1). Fertilizer was applied to the fertilized control
plots at the recommended rates of 20 kg N ha-1, 40 kg P2O5 ha-1, and 20
kg K2O ha-1 of Compound-D at sowing and 92 kg N ha-1 as urea at 4
weeks after sowing (WAS). All the plots were managed following the
recommended agronomic practices for weeding and harvesting.
Soil ammonium and nitrate nitrogen
Soil sampling for soil ammonium and nitrate was done at fallow clearing
(pre-season, November 1998) using a metal sampler (4.2 cm diameter
G. I. Pipe) from 0-20, 20-40 and 40-60 cm soil depths. For determination
of ammonium and nitrate nitrogen, about 20 g of field moist soil was
extracted with 100 ml of 2 M KCl. The samples were shaken on a
horizontal shaker for 1 hour at 150 oscillations min-1 followed by gravity
filtering with pre-washed Whatman No. 5 filter paper. A second subsample of soil was dried at 105 oC for 24 hours to determine the dry
weight of the extracted soil. Ammonium was determined by colorimetric
method (Anderson and Ingram, 1993). Nitrate and nitrite concentrations
were determined by cadmium reduction (Dorich and Nelson, 1984). The
sum of inorganic ammonium-N and inorganic nitrate-N constituted the
total inorganic nitrogen.
Laboratory incubation
Laboratory incubation was done to characterize the nutrient release
patterns of sesbania (fresh leaves + litter), sesbania litter alone, pigeonpea
(fresh leaves + litter), pigeonpea litter alone and dry grass littter. Chemical
compositions of the organic materials used are shown in Table 6.2.
Fresh leaves collected from the two species were sun dried for 2-3 days
and oven-dried at 65 0C for 48 hours to determine the dry matter (DM)
content. Soil was air-dried and sieved through a 2-mm mesh screen.
The soil was first leached with deionised water at a water-to-soil ratio of
1 to 3 and left to drain until 50 % water holding capacity was achieved
by constant weighing. 1.35g of ground organic material was mixed with
270g of soil in 350 ml aluminium moisture cans. This rate is equal to 5
Chirwa, T. S. et al
94
t ha-1, which is applied in the field. The treatments were 1) Soil + sesbania
(fresh leaves + litter). 2) Soil + sesbania litter alone. 3) Pigeonpea (fresh
leaves + litter). 4) Pigeonpea litter alone. 5) Soil + dry grass. 6) Soil alone
(control). Fresh leaves and litter were mixed in a 1:1 w/w basis. Moisture
cans were covered with aluminium foil that was perforated to allow air
movement and put in the incubator at 28 0C throughout the experiment
period. Soil moisture content in the cans was maintained at 50% water
holding capacity throughout the experimental period by periodic
additions of deionised water using a syringe and constant weight
adjustment. Each treatment was replicated three times in a completely
randomised design. Sub samples (20g) were analyzed for exchangeable
NH4-N and NO3-N immediately after addition of plant material (at week
0) and once every week for 10 weeks. Ammonium was analysed by the
modified calorimetric method of Dorich and Nelson (1984) and nitrate
by the method of Cataldo et al. (1975). Results were reported as
cumulative net mineralizable total inorganic nitrogen (ammonium-N +
nitrate-N).
Soil penetration resistance
Penetration resistance was measured with a hand penetrometer, Bush
soil penetrometer SP1000, version 1.0, supplied by ELE International,
England. The penetrometer probe of 12.83 mm diameter with a cone
semi-angle of 600 was pushed to a depth of 50 cm, and the resistance
offered by the soil was recorded at 2 cm interval by a digital balance.
Five insertions in the net plot were measured at 4 WAP and 24 WAP.
Cumulative water intake
Cumulative water intakes were monitored at fallow clearing and at
crop harvest during the dry season. Two standard infiltrometer rings
(double ring) per plot were used according to the procedure described
by Bouwer (1986). Water measurements were recorded for three hours
at 0, 5, 10, 15, 20, 30, 45, 60, 90, 120, 150 and 180-minute intervals.
The average readings were used to calculate cumulative water intake
per plot using Kostiakov (1932) and Philip (1957) models.
The Kostiakov (1932) model is described by equations 1 and 2:
1) Equation for cumulative depth is described by:
z = kta
2) Equation for infiltration rate is described by:
i = akta-1
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
95
Where: z = cumulative depth infiltrated
t = time
i = infiltration rate
a and k are constants determined empirically.
The Philip’s model is described by equations 3 and 4:
3) Equation for cumulative depth is described by:
z = St1/2+At
4) Equation for infiltration rate is described by:
i = 1/2St-1/2+ A
Where: z = cumulative depth infiltrated
t = time
i = infiltration rate
S =sorptivity which indicates the capacity of a soil to absorb water.
A = transmissivity
Aggregate size distribution
Aggregate size distribution and stability was determined by the methods
of De Leenheer and De Boodt (1958). Soil clods were dug at random
from 0-20 cm depth at fallow clearing and at crop harvest using a hand
hoe. Soil clods were hand broken to a maximum aggregate size of 50
mm then each soil sample was air dried at room temperature to stimulate
some forces involved in aggregation. These forces are those related to
cultivation, erosion (wind and water), and wetting of soils, respectively.
The sample was allowed to pass through 9.50 mm and retained on the
0.30 mm opening sieve. A Yoder (1936) type-sieving machine, which
raises and lowers the nests of sieves, through water with a stroke length
of 1.5 inches approximately 30 cycles per minute was used with a set of
4.75, 2.0, 1.0, 0.50 and 0.30 mm openings sieves with a receiver at the
bottom. An air-dry sample of 500g was placed gently in the sieve with
4.75 mm opening. The set of sieves were lowered in water of sieving
machine and the machine was made to run for five minutes. Fractions
obtained on each sieve and retainer was oven dried at 105oC for 48
hours and weighed. The oven dry aggregates were expressed as mean
weight diameter (MWD) of aggregates and percent of aggregates retained
on each sieve.
Trees, natural grass and crop biomass
Total above ground biomass of natural grass and tree components
including leaves, twigs and litter were measured at fallow clearing.
Total inorganic N in grass, leaves and litter was analysed by microkjeldahl digestion followed by distillation and titration (Anderson and
96
Chirwa, T. S. et al
Ingram, 1993). Nitrogen uptake was assessed in the plant dry matter
(DM) measured at 4, 6, 8 and 24 WAP (at harvest in grain and stover).
Five plants were cut at ground level and oven-dried at 70°C for 72
hours for DM determination. N in the plant tissue was analysed by
micro-kjeldahl digestion followed by distillation and titration (Anderson
and Ingram, 1993). Maize grain and stover yields were measured at
harvest (24 WAP).
Data analysis
The data were subjected to analysis of variance using GENSTAT version
5 (Genstat 5 committee, 1988). For all mean comparisons significance
was tested at P ≤ 0.05, using Duncan’s Multiple Range Test (Gomez and
Gomez, 1984). Simple linear and curvilinear regressions were used to
determine the relationship between maize grain yield and pre-season
soil inorganic nitrate-N and total inorganic N.
Results
Tree growth
Significant differences (P<0.05) were observed in the survival rates. The
highest survival rates were recorded in sesbania (91.7%), while pigeonpea
had only 31.0%. As was the case with survival, high total biomass was
recorded in sesbania fallows (Table 6.2).
Table 6.1: Growth performance of Sesbania sesban and Cajanus cajan fallow species
at 24 months after fallow establishment at Msekera, Chipata-Zambia (November 1998)
Land-use system
Survival (%)
Leaf + Twig
(t ha-1)
Total biomass
(t ha-1)
Pigeonpea
Sesbania sesban
Mean
SED
31.0b
91.7a
61.30
10.68
0.19a
0.23a
0.21
0.18
8.50
16.8
12.6
2.94
Nitrogen mineralization and immobilization
The organic materials used had C-to-N ratio ranging from 14.7 to 69 for
sesbania fresh leaves and natural grass, respectively (Table 6.1). The N
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
97
concentration ranged from 0.62 to 3.09% for natural grass and sesbania,
respectively (Table 6.1). The quality of leaves and litter significantly
affected the N release pattern throughout the 10-week incubation period.
After 10 weeks of incubation, the cumulative net N mineralization ranged
from 5.1 to 59.4 mg N kg-1 soil for pigeonpea litter only and sesbania
(fresh leaves + litter) mixture, respectively (Figure 6.1). Cumulative net
immobilization of 0.8 mg N kg-1 soil was observed at end of the incubation
period in natural grass litter (Figure 6.1). Between week 1 and 4 there
was net immobilization in pigeonpea (fresh leaves + litter) mixture and
sesbania litter alone. Pigeonpea litter alone had a cumulative net
immobilization from week 1 to 5.
Table 6.2: Chemical compositions of organic materials used for incubation study at
Msekera, Chipata-Zambia
Land-use system
Pigeonpea fresh alone
Pigeonpea litter alone
Pigeonpea (fresh leaves + litter) mixture
Sesbania fresh leaves alone
Sesbania litter alone
Sesbania (fresh leaves + litter) mixture
Natural grass litter alone
Carbon (%)
Nitrogen (%)
Carbon to
Nitrogen ratio
45
45
45
45
45
45
43
2.98
1.36
2.10
3.09
1.28
2.4
0.62
15.1
33.1
21.4
14.7
35
18
69
Pre-season soil mineralizable nitrogen
Pre-season nitrate-N and total inorganic N was significantly (P≤ 0.05)
affected by soil depth and LUS (Table 6.3). Ammonium-N was not
significantly affected by either soil depth or LUS (Table 6.3). Maize with
fertilizer had the highest soil nitrate-N in all soil depths, which was not
significantly different from pigeonpea and sesbania LUS. At 40-60 cm
soil depth soil nitrate-N had the following trend: m+f > sesbania > m-f >
pigeonpea > natural fallow (Table 6.3). A polynomial regression model
between maize grain yield and pre-season inorganic nitrate-N for 0-20
cm, 0-40 cm and 0-60 cm soil layers showed that the amount of preseason inorganic nitrate-N in the soil layer accounted for 71%, 68%
and 71%, respectively of the maize yield. Total mineral N in the top 0-20
cm soil depths was in the order of: m+f > pigeonpea > sesbania > m-f >
natural fallow (Table 6.3). As was the case with pre-season soil nitrateN, total mineral N in 0-20 cm, 0-40 cm and 0-60 cm soil depths was
significantly correlated to grain yield (R2 = 0.70, 0.67 and 0.71).
Chirwa, T. S. et al
98
Table 6.3: Pre-season soil mineralizable nitrogen as affected by Land-use system and
soil depth at Msekera, Chipata-Zambia (November 1998)
Treatment
Ammonium (mg N kg-1)
0-20
20-40
40-60
0-20
20-40
40-60
0-20
20-40
40-60
Pigeonpea
3.65a
2.52a
2.10a
3.49ab
2.37ab
1.25bc
7.14ab
4.88ab
3.36b
Natural fallow 3.09a
2.14a
1.64a
1.49b
0.74c
0.59c
4.58b
2.88b
2.23b
a
a
a
a
a
a
a
a
7.73a
ab
3.41b
ab
4.04b
Maize +fert
Maize – fert
Sesbania
3.11
a
4.07
a
4.21
2.53
a
2.00
a
1.85
Nitrate (mg N kg-1)
1.59
4.24
a
2.99
a
1.97
3.52
b
2.49
4.74
bc
2.18
8.32
bc
1.70
ab
Total N (mg N kg-1)
1.81
bc
b
1.52
ab
5.29
ab
2.07
6.70
5.52
4.23
3.37
Mean
3.36
2.21
2.06
2.78
1.97
2.09
6.40
4.18
4.15
SED
0.83
0.71
0.62
0.84
0.58
0.56
1.29
0.94
0.98
Means in a column followed by the same letter or letters are not significantly different at
P<0.05 based on the Duncan’s Multiple Range Test
Figure 6.1: Cumulative amount of net N mineralised as affected by quality of multipurpose
tree leaves and litter during 10-week incubation period at Msekera, Chipata-Zambia
Cumulative net mineralization (mg N kg-1 soil)
80
70
bars = sed
60
50
40
30
20
10
0
-10
-20
0
1
2
3
4
5
6
7
8
9
Time in weeks after aplication
Cajanus litter
Cajanus litter+fresh
Grass litter
Soil control
Sesbania litter
Sesbania litter+fresh
10
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
99
Dry matter and seasonal nitrogen accumulation in maize
topmass
DM accumulation during the growing season ranged from 0.2 t ha-1 to
9.5 t ha-1 for maize without fertilizer (at 4 WAP) and with fertilizer (at 24
WAP), respectively (Figure 6.2). High N accumulation in maize above
ground biomass was observed from 4 to 6 WAP in sesbania LUS (13.9
kg N ha-1), as compared to fertilized plot that only accumulated 2.4 kg N
ha-1 (Figure 6.3). Between 6 WAP to 8 WAP, there was a sharp rise of N
accumulation in fertilized maize. Fertilized maize accumulated the
highest amount of N (39 kg N ha-1). On the other hand, sesbania and
pigeonpea had only 7.0 kg N ha-1 and 15.8 kg N ha-1, respectively (Figure
6.4). The maximum N accumulation in maize aboveground biomass at
24 WAP averaged 156.9 kg N ha-1 and 77.0 kg N ha-1 for maize with
fertilizer and sesbania LUS, respectively. A polynomial regression model
between maize dry matter accumulation and nitrogen uptake at 8 WAP
and 24 WAP showed that the amount of nitrogen uptake accounted for
93% and 98%, respectively of the dry matter accumulation in maize
plant.
Figure 6.2: Maize dry matter (DM) accumulation during the growing season under different
land-use systems at Msekera, Chipata-Zambia (1998/99 season)
Dry matter accumulation (t ha-1)
16
14
12
bars = sed
10
8
6
4
2
0
4
6
8
24
Time in weeks after planting
Cajanus
Natural fallow
M+f
M-f
Sesbania
Chirwa, T. S. et al
100
Figure 6.3: Seasonal Nitrogen accumulation in maize above ground biomass during the
growing season under different land-use systems
N accumulation (Kg ha-1)
250
bars = sed
200
150
100
50
0
4
6
8
24
Time in weeks after planting
Cajanus
Natural fallow
Sesbania
M+f
M-f
Figure 6.4: Effects of Land-use system on cumulative water intake (mm) of the soil at
Msekera, Chipata-Zambia (at crop harvest May 1999)
Cummulative water intake (mm)
700
600
bars = sed
500
400
300
200
100
0
0
20
40
60
80
100
120
140
160
180
200
Time (minutes)
Natural fallow
M+f
M-f
Sesbania
Cajanus
Maize yields
Analysis of variance of maize stover, grain yield and total biomass showed
significant differences (P≤0.05) due to land-use systems. The highest
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
101
stover yield of 4.01 t ha-1 was in continuous maize with fertilizer followed
by sesbania (3.01 t ha-1) and the least was in natural fallow (0.93 t ha-1).
As was the case with stover, the highest grain yield of 5.51 t ha-1 was
recorded in maize with fertilizer followed by sesbania (3.02 t ha-1) and
the least was in natural fallow (0.85 t ha-1)(Table 6.4).
Table 6.4: Maize yields as affected by different land-use system at Msekera, ChipataZambia (May 1999)
Land-use
system
Stover yield
(t ha-1)
Grain yield
(t ha-1)
3.01a
2.79a
0.93a
4.01a
1.25b
2.40
0.67
3.02a
2.69bc
0.85c
5.51a
1.01c
2.61
0.82
Sesbania sesban
Cajanus cajan
Natural fallow
Cont. M+F
Cont. M-F
Mean
SED
Total biomass
(t ha-1)
6.02b
5.48bc
1.78d
9.52a
2.26cd
5.01
1.46
Means in a column followed by the same letter or letters are not significantly different at
P≤0.05 based on the Duncan’s Multiple Range Test
Penetration resistance
Average cone penetrometer resistance measured at both 4 WAP
(November 1998) and 24 WAP (May 1999) was significantly affected by
the LUS. Average cone penetrometer resistance measured at 4 WAP
ranged from 2.2 to 3.9 Mpa for sesbania and maize with fertilizer LUS,
respectively (Table 6.5). At 24 WAP no significant difference was observed
in cone penetrometer resistance among the LUS (Table 6.5).
Table 6.5: Effects of Land use system on some soil physical properties after 2 years of
improved fallow system at Msekera, Chipata-Zambia (November 1998 and May 1999)
Land-use system
Sesbania sesban
Cajanus cajan
Natural fallow
Continuous M+F
Continuous M-F
Mean
SED
Average penetrometer Average water
resistance at
stable aggregates
at 40 cm soil
>2.00mm (%)
depth (Mpa)
Average cumulative
water intake at 3
hours(mm)
Nov.
1998
May
1999
Nov.
1998
May
1999
Nov.
1998
May
1999
2.2c
2.9b
2.9b
3.9a
3.2b
3.1
0.19
3.3a
3.0a
4.3a
4.0a
3.3a
3.6
0.57
83.3a
80.8a
65.7b
65.6b
61.2a
71.5
3.13
65.1b
76.9a
65.8b
58.4c
44.0d
62.0
2.74
465ab
485ab
572a
315bc
233c
414
93.3
399a
221b
246ab
184b
173b
245
72.5
Means in a column followed by the same letter or letters are not significantly different at
P<0.05 based on the Duncan’s Multiple Range Test
Chirwa, T. S. et al
102
Aggregate stability
The percentages of aggregates bigger than 2.00 mm (aggregates > 2.00
mm) were significantly affected by the LUS at fallow termination (p<0.05).
The highest percentage of water stable aggregates > 2.00 mm at fallow
termination was recorded in sesbania LUS (83.3 %) followed by pigeonpea
LUS (80.8 %). The least was recorded in maize without fertilizer (61.2
%) (Table 6.5). The highest percentage of aggregates at crop harvest
(May 1999) greater than 2.00 mm was recorded in pigeonpea LUS (76.9
%) followed by natural fallow LUS (65.8 %). The least was recorded in
maize without fertilizer (44.0 %) (Table 6.5).
Infiltration rate and cumulative water intake
Significant differences (P≤0.05) were observed at both fallow termination
and crop harvest stages in cumulative water intake. At fallow
termination, the highest average cumulative water intake at 3 hours
was 572 mm in natural fallow followed by pigeonpea (485 mm). The
lowest cumulative water intake was recorded in maize without fertilizer
(233 mm) (Table 6.5). At crop harvest, the maximum (399 mm) and
least (173 mm) average cumulative water intake were recorded in
Sesbania and continuous maize without fertilizer, respectively (Table
6.5). Soil water sorptivity at fallow termination was in the order of:
pigeonpea > natural fallow > sesbania > m+f > m-f (Table 6.6). On the
other hand soil water sorptivity at crop harvest was in the order of:
sesbania > natural fallow > pigeonpea > m+f > m-f (Table 6.6).
Table 6.6: Prediction Equations and Correlation Coefficients (r2) relating to equilibrium
infiltration rate (i) in mm min-1 with the time (t) in minutes at fallow termination and crop
harvest at Msekera, Chipata-Zambia (November 1998 and May 1999)
Land-use system
Sesbania (Nov 1998)
Sesbania (May 1999)
Cajanus (Nov 1998)
Cajanus (May 1999)
Natural fallow(Nov 1998)
Natural fallow(May 1999)
M+F(Nov 1998)
M+F(May 1999)
M-F (Novy 1998)
M-F (May 1999)
S= Sorptivity
A= Transmissivity
Kostiakov’s model of
infiltration rate
(mm min-1)
i = akta-1
a
0.67
0.65
0.63
0.59
0.66
0.61
0.63
0.66
0.69
0.67
k
13.52
13.02
18.50
9.76
16.89
10.26
10.97
7.98
6.25
5.42
r2
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
Philip’s model of
infiltration rate
(mm min-1)
i = 1/2St-1/2 + A
S
15.92
14.97
22.30
10.92
19.05
11.68
12.41
8.82
7.74
6.64
A
1.44
1.09
1.12
0.40
1.66
0.49
0.79
0.34
0.72
0.50
r2
0.98
0.99
0.88
0.95
0.98
0.95
0.96
0.98
0.94
0.96
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
103
Discussion
Tree growth
Results on the growth performance of pigeonpea fallows showed that
survival was very poor and this was probably due to the method of
establishment and drought. Pigeonpea was directly seeded as compared
to sesbania that was raised from nursery seedlings. Soon after sowing
of pigeonpea there was a period of drought that might have contributed
to high mortality. Kwesiga et al. (1993) reported similar results of poor
establishment and high mortality in pigeonpea fallow under similar
environmental conditions.
Nitrogen mineralization and immobilization
The large C-to-N ratio (69) and low N (0.62%) in grass litter resulted in
immediate immobilization of all N available in the soil. All the treatments,
which had litter, showed some form of immobilization except for Sesbania
(fresh leaves + litter) mixture. This was because of the low C-to N ratio
of 18. Palm et al. (1997) showed that sesbania fresh leaves which have
3-4 % N decompose faster than those species with high C-to-N ratio.
The increased growth and grain yield in the sesbania LUS can be
attributed to the high concentrations of N, fast nutrient release and
decomposition of the fresh leaves and litter. The 4 weeks of N
immobilization in pigeonpea (fresh leaves + litter) delayed the release of
N to the maize crop. Sakala et al. (2000) also reported that senesced
cajanus leaves have a short period of N immobilization despite having a
narrow C-to-N ratio. Similarly, Mafongoya et al. (2000) reported low
quality materials initially immobilize nutrients, but later they mineralise
and make the nutrients available to the crop for uptake. Therefore a
mixture of pigeonpea fresh leaves and litter will only start adding N to
the maize crop after a period of 4 weeks as compared to the mixture of
sesbania fresh leaves and litter. The other reason of high N mineralization
for sesbania fresh leaves + litter could be ascribed to low lignin content
as compared to other materials used in this study (Mafongoya et al.
(1998).
Plant materials with high lignin concentration decompose more
slowly than those with low lignin (Melillo et al. 1982). Similarly, the low
release of nitrogen in natural grass fallow litter or pigeonpea litter could
be as a result of high lignin and low N contents in these materials.
While Mafongoya et al. (2000) reported that nutrient release from these
organic inputs depends on their chemical composition and soil
104
Chirwa, T. S. et al
properties. However, work done by Mafongoya et al. (1998), Handayanto
et al. (1995) and Constantinides and Fownes (1994) on nitrogen release
patterns of MPT leaves say that the ratios of NDF-N:N, Soluble
polyphenols:N, and (Lignin + polyphenol):N are a good predictor of net
N release patterns on MPT leaves. Whilst Palm and Sanchez, (1991)
showed that nitrogen concentration, lignin and polyphenolic contents
are considered to control N release rates of decomposing plant residues.
Our results indicate that mixing of litter of low quality with fresh leaves
at fallow termination will cause the nitrogen to immobilize for a few
weeks except for those species, which have low C-to- N ratio. Under
field conditions there is either more of the litter or fresh leaves depending
at what time you terminate the fallow. In most cases fallows are
terminated in November or December at that time there is less fresh
leaves on the trees. Therefore, the N mineralization patterns will depend
on the ratios of these organic materials (fresh leaves to litter). Athough
no data is available on the polyphenols and lignin composition of these
organic inputs used, our results suggest that not only the C-to-N ratio
played a major role in the N mineralization pattern, but also other
chemical characteristics of these materials as reported by many workers
(Mafongoya et al. 1998, Handayanto et al. 1995 and Constantinides
and Fownes 1994).
Pre-season soil mineralizable nitrogen
Our results show that pre-season soil nitrate-N and total inorganic N at
lower depths was higher in fertilized plots than the tree or natural fallow
plots. This is because most of the nitrate-N in the top layers is bound to
be leached to lower depths quickly after heavy rains as compared to the
tree based system which releases nitrogen slowly. Buresh (1995) also
reported that most of the nitrate-N in the top layers is bound to be
leached to lower depths that are beyond the rooting depth of most annual
crops. Tree fallows are best since trees are able to capture lost nutrients
and transfer them back to surface soil in form of litterfall and leafy
biomass which subsequently is made available to the maize crop as
compared to the natural fallow (Mekonnen et al. 1997). Higher nitrateN in the topsoil was observed under pigeonpea and sesbania fallow
than the natural fallow. Mekonnen et al. (1997) and Onim et al. (1990)
also reported similar results. They attributed this higher topsoil nitrate
in pigeonpea and sesbania as being due to faster mineralization under
N-fixing trees than under natural fallow. The high level of nitrate-N in
the lower depth of maize with fertilizer was probably due to leaching.
Similarly, Hartemink et al. (1996) and Mekonnen et al. (1997) found
greater accumulation of subsoil nitrate under maize monoculture on
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
105
the Oxisol and they attributed this to higher rainfall and leaching of
nutrients to lower depths. Low levels of subsoil nitrate were also observed
in natural fallow and sesbania LUS.
Dry matter and seasonal nitrogen accumulation in maize
topmass
The high N accumulation in maize with fertilizer above ground biomass
was probably due to the addition of mineral fertilizer and rapid
assimilation of nutrients by the maize plants. Low N accumulation in
sesbania, pigeonpea, natural fallow and maize without fertilizer was
probably due to low rate of inorganic-N mineralization and lack of
synchrony of N release to N demand by the crop. On the other hand,
Mafongoya et al. (2000) attributed the mechanism contributing to
synchrony as being the action of chemical constituents in organic inputs
which slow or delay the release of nutrients, thus reducing leaching
and asynchrony between nutrient release and crop uptake. The other
reasons for low N accumulation in maize without fertiliser and natural
fallow LUS could be: 1) low levels of soil nutrients to influence plant
uptake and growth, and 2) the limited utilization of soil nitrate from the
subsoil by maize due to poorly developed root system resulting from the
rapid deterioration of soil physical properties (high penetration
resistance, low infiltration rate, and low aggregate stability). A polynomial
regression model between maize dry matter accumulation and nitrogen
uptake at 8 WAP and 24 WAP showed that the amount of nitrogen uptake
accounted for 93% and 98%, respectively of the dry matter accumulation
in maize plant.
Maize yields
The increase of grain yields in the sesbania and pigeonpea fallow system
was a result of plant-available N from decomposing aboveground biomass
(fresh leaves and litter). Other sources of nitrogen was probably from
the decomposition of root biomass of sesbania and pigeonpea fallow
species. Similar results of increased maize yields after 2 years of sesbania
fallows and pigeonpea fallows have been reported (Kwesiga and Coe,
1994; MacColl, 1989). Maroko et al. (1997) attributed increase in crop
yield after sesbania fallow to rapid release of plant-available N from
sesbania litter and leaves resulting in an increased supply of inorganic
N at crop planting after fallow period, and increased soil N mineralization
rates. On the other hand, the decline in yield in unfertilised and natural
fallow plots could be soil fertility depletion and deterioration of soil
106
Chirwa, T. S. et al
physical properties such as resistance to root penetration, aggregate
stability and infiltration. The other reason for yield component decline
is water stress during pollination (Claassen and Shaw, 1970). Sanchez
(1976) also reported that the main reason for the decline in yield is soil
fertility depletion, increased weed infestation, deterioration of soil
physical properties, and increased insect and disease attacks. Similarly,
the data from this experiment confirms the decline in yield of
continuously cropped maize without fertilizer as being due to soil fertility
depletion and deterioration of soil physical properties.
Penetration resistance
The major reason for low penetration resistance in natural fallow,
sesbania and pigeonpea LUS at fallow termination, can be attributed to
addition of aboveground biomass during fallow phase and improved
soil aggregation. On the other hand Harris et al. (1996) attributed low
penetration resistance to the addition organic matter to soil which
increases soil microbial activity and together with the decomposed soil
organic matter, this microbial activity promotes aggregation, hence the
soil is more porous and as a result, soil penetration resistance is
decreased. Decrease in penetration resistance under agroforestry
systems have been reported by Torquebiau and Kwesiga (1996), Lal
(1989) and Dalland et al. (1993). There was an increase in penetration
resistance after cropping. This could be as a result of reduced pore
space and loss of soil aggregation.
Aggregate stability
The high percent of water stable aggregates >2.00 mm in pigeonpea
sesbania and natural fallow at both fallow termination and at crop
harvest was probably due to high organic matter content as compared
to maize with and without fertilizer LUS. Mapa and Gunasena (1995)
and Yamoah et al. (1986) reported similar results in hedgerow inter
cropping. The importance of soil organic matter in stabilizing soil has
been well documented (Tisdall and Oades 1983, and Chaney and Swift
1984). Continuous cultivation breaks large aggregates into smaller
aggregates as was evidenced at crop harvest of this experiment. There
was a decrease in percent of water stable aggregates >2.00 mm after
cropping at crop harvest. The improved size aggregation in sesbania,
pigeonpea and natural fallow LUS has an effect on increased water
infiltration and water holding capacity, which reduces surface water
runoff and hence decreased erosion as compared to the maize mono
cropping system.
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
107
Infiltration rate and Cumulative water intake
The high cumulative water intake in the natural fallow, sesbania and
pigeonpea could have been due to the improvement in the soil physical
properties (improved soil aggregation and decreased resistance to
penetration). Mapa and Gunasena, (1995) reported that higher wet stable
aggregates facilitate higher macro-porosities, higher infiltration rate and
reduce soil erosion which is a major contributing factor in degrading
soil physical properties under shifting cultivation. Similar results in
hedgerow intercropping were reported by Lal, (1989), and Hulugalle
and Ndi, (1993). Soil water sorptivity in November 1998 and May 1999
was highest in tree-based system than maize with and without fertilizer.
These results show that pigeonpea and sesbania tree based LUS will
have a higher affinity for water by soil matrix. Which means that during
periods of water stress maize under pigeonpea and sesbania LUS will
perform better than the maize with and without fertilizer.
Conclusion
This study shows the importance of improved fallow technology in
maintaining soil fertility. Sesbania and pigeonpea have a potential to
supply inorganic soil nitrogen through leafy biomass and litter. The
nitrogen contribution of sesbania and pigeonpea fallows to subsequent
crop was evidenced by increased maize yields after these fallows as
compared to no tree treatments. Improved fallows have the potential of
improving soil physical conditions as compared to maize mono-cropping
systems as shown from high soil aggregation, greater water infiltration,
higher soil water sorptivity and reduced resistance to penetration.
Continuous cultivation causes the breakdown of numerous soil
processes associated with crop productivity.
The results from the incubation study under laboratory condition
indicate that mixing of litter (low quality) with fresh leaves (high quality)
from the same tree species at fallow termination had an effect on maize
N uptake. Maize planted after sesbania fallows will have an immediate
benefit from the prunings than maize planted after pigeonpea. This is
because pigeonpea mixture (fresh leaves +litter) starts to release nitrogen
to the crop after a period of 4 weeks. However there is need to carryout
this study under field condition to support the results found under
laboratory condition.
Acknowledgements
The authors would like to thank the staff at Zambia/ICRAF Agroforestry
project in Chipata district for their co-operation during the laboratory
and fieldwork. Finally, thanks go to our sponsors; African Network for
108
Chirwa, T. S. et al
Agroforestry Education/International Centre for Research in Agroforestry
(ANAFE/ICRAF), for the financial support.
References
Anderson, J. M. and Ingram, J. S. I. (1993) Tropical Soil Biology and Fertility: A
Handbook of Methods. 2nd edition. CAB International, Wallingford, UK. pp
221.
Bouwer, H. (1986) Intake rate: Cylinder infiltrometer. p825-844. In: Klute, A.(ed).
Methods of soil analysis, part 1, Physical and mineralogical methodsAgronomy Monograph Number 9. Am. Soc. Agr. Madison, Wisconsin, USA.
Brady, N. C. (1996) Alternatives to slash-and burn: a global imperative.
Agriculture Ecosystems and Environment 58: 3-11.
Buresh, R. J. and Tian, G. (1997) Soil improvement by trees in sub-Saharan
Africa. Agroforestry systems 38: 51-76.
Buresh, R.J. (1995) Nutrient cycling and nutrient supply in agroforestry systems
p155-164. In: Dudal, R. and Roy, R.N.(Eds). Integrated plant nutrition
systems. FAO fertilizer and plant nutrition Bulletin number 12. Rome, Italy.
Cataldo, D. A., Harron, M., Schrader, L. E. and Youngs, V. L. (1975) Rapid
colorimetric determination of nitrate in plant tissue by nitration of salicylic
acid. Commun Soil Sci. Plant Anal. 6: 71-80.
Chaney, K. and Swift, R. S. (1984) The influence of organic matter on aggregate
stability in some British soils. Journal of Soil science 35: 223-236.
Claassen, M. M. and Shaw, R. H. (1970) Water deficit effects on corn: II. Grain
components. Agron. J. 62: 652-655.
Commissaris, A.L.T.M. (1975) Detailed Soil Survey of Msekera Research Station.
Soil Survey report no. 19.
Constantinides, M. and Fownes, S. H. (1994) Nitrogen mineralization from leaves
and litter of tropical plants. Relationship to nitrogen, lignin and soluble
polyphenol concentrations. Soil Biol. Biochem 26: 49-55.
Dalland, A., Våje, P. I., Mathews, R. B. and Singh, B. R. (1993) The potential of
alley cropping in improvement of cultivation systems in the high rainfall
areas of Zambia III. Effects on soil chemical and physical properties.
Agroforestry Systems. 21: 117-132.
De Leenheer, L. and De Boodt, M. (1958) Determination of aggregate stability
by the change in mean weight diameter. pp.260-300. In: Proceedings of
International Symposium of Soil Structure. Ghent, Belgie.
Dorich, R. A. and Nelson, D. W. (1984) Evaluation of manual cadmium reduction
methods for determination of nitrate in potassium chloride extracts of soil.
Soil Sci. Soc. Am. J. 48: 72-75.
FAO (1988) Soil Map of the World-Revised Legend. World Soil Resources Report
60 FAO/UNESCO, Rome.
Genstat 5 Committee. (1988) Genstat 5 reference manual. Oxford Univ. Press,
Oxford, England. 749pp.
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
109
Giller, K. E., Cadisch, G., Ehaliotis, C., Adams, E., Sakala, W. D. and Mafongoya,
P. L. (1997) Building soil nitrogen capital in sub-Saharan Africa. In: Buresh,
R. J., Sanchez, P. A. and Calhoum, F. (Eds) Replenishing Soil Fertility in
Africa p 151-192. SSSA Special Publication 51. SSSA and ASA, Madison,
WI, USA.
Gomez, K. A. and Gomez, A. A. (1984) Statistical procedures for agricultural
research 2nd edition. Jony Wiley and Sons Inc.
Handayanto, E. Cadisch, G. Giller, K. E. (1995) Manipulation of quality and
mineralization of tropical legume tree prunings by varying nitrogen supply.
Plant Soil 176: 149-160.
Harris, R. F., Chesters, G. and Allen, O. N. (1996) Dynamics of soil aggregation.
Adv. Agron 18: 107-169.
Hartemink, A. E. Buresh, R. J. Jama, B. Janssen, B. H. (1996) Soil nitrate and
water dynamics in sesbania fallows, weed fallows and maize. Soil Sci Soc
Am J. 60: 568-574.
Hulugalle, N. R. and Ndi, J. N. (1993) Effects of no-tillage and alley cropping on
soil properties and crop yields in a Typic Kandiudult of Southern Cameroon.
Agroforestry Systems. 22: 207-220.
Kostiakov, A. N. (1932) On the dynamics of the coefficient of water-percolation
in soils and on the necessity for studying it from a dynamic point of view for
purposes of amelioration. TRANS. Sixth Intnl. Society of Soil Science, Russian
Part A: 17-21.
Kwesiga F. and Coe, R. (1994) The effect of short rotation Sesbania sesban
planted fallows on maize yield. Forest Ecology and Management 64: 199208.
Kwesiga, F., Phiri, D. M. and Rauno, A. (1997) Improved fallows with Sesbania
in Eastern Zambia. Summary proceedings of a consultative workshop 2226 April, 1996, Chipata, Zambia. ICRAF Nairobi, Kenya pp. 101.
Kwesiga, F., Phiri, D. M., Simwanza, C. P. and Mwanza, S. (1993) Zambia/
ICRAF Agroforestry Research Project, 1993 Annual report. AFRENA Report
no. 71.
Lal, R. Wilson, G. F. and Okigbo, B. N. (1978) No-tillage farming after various
grasses and leguminous cover crops in tropical Alfisols I. Crop performance.
Field Crops Res. 1: 71-84.
Lal, R. Wilson, G. F. and Okigbo, B. N. (1979) Changes in properties of an Alfisol
produced by various crop covers. Soil Sci. 127: 377-382.
Lal, R. (1989) Agroforestry systems and soil surface management of a tropical
Alfisol: IV. Effects on soil physical and mechanical properties. Agroforestry
systems 8: 197-215.
MacColl, D. (1989) Studies on maize (Zea mays L.) at Bunda, Malawi: II. Yield in
short rotation with legumes. Exp. Agric. 25: 367-374.
Mafongoya, P. L. and Nair, P .K. R. (1997) Multipurpose tree prunings as a
source of nitrogen to maize (Zea mays L.) under semiarid conditions in
Zimbabwe. I. Nitrogen recovery rates in relation to pruning quality and
method of application. Agroforestry system 35: 31-46.
110
Chirwa, T. S. et al
Mafongoya, P. L., Barak, P. and Reed, J. D. (2000) Carbon, nitrogen and
phosphorus mineralization of tree leaves and manure. Biology Fertility Soils
30: 298-305.
Mafongoya, P. L., Nair, P. K. R. and Dzowela, B. H. (1997) Effect of multipurpose
trees, age of cutting and drying method on pruning quality. In: Cadisch, G.
and Giller, K.E. (Eds) Driven by nature: plant litter quality and decomposition.
CAB International, Wallingford, UK, pp 167-174.
Mafongoya, P. L., Nair, P. K. R. and Dzowela, B. H. (1998) Mineralization of
nitrogen from decomposing leaves of multipurpose trees as affected by their
chemical composition. Biology Fertility Soils 27: 143-148.
Mapa, R. B. and Gunasena, H. P. M. (1995) Effect of alley cropping on soil
aggregate stability of a tropical Alfisol. Agroforestry systems 32: 237-245.
Maroko, J. B., Buresh, R. J. and Smithson, P. C. (1997) Soil nitrogen availability
as affected by fallow-maize systems on two soils in Kenya. Biol. Fert. Soils.
26: 229-234.
Mekonnen, K., Buresh, R. J. and Jama. (1997) Root and inorganic nitrogen
distributions in sesbania fallow, natural fallow and maize fields. Plant and
soil 188: 319-327.
Melillo, J. M., Aber, J. D. and Muratone, J. F. (1982) Nitrogen and Lignin control
of hard wood leaf litter decomposition dynamics. Ecology 63: 621-626.
Migliena, A. M., Rosell, R. A. and Glave, A. E. (1988) Changes of chemical
properties of a Haplustoll soil under cultivation in Semi-arid Argentina. p
421-425. In: Unger, P. W. Jordan, W.R. Sneed, T. V. and Jensen R. W. (Eds).
Challenge in dryland agriculture- Grobal perspective- proceedings of the
International Conference on Dryland Farming, August 15-19, 1988. Texas,
USA.
Nye, P. H. and Greenland, D. J. (1960) The soil under shifting cultivation.
Technical Communication No. 51. Commonwealth Bureau of Soils,
Harpenden, UK.
Onim, J. F. M., Mathuva, M., Otieno, K. and Fitzhugh, H. A. (1990) Soil fertility
changes and response of maize and deans to green manures of leucaena,
sesbania and pigeonpea. Agroforestry Systems. 12: 197-215.
Palm, C. A., Myers, R.J. K. and Nandwa, S. M. (1997) Combined use of organic
and inorganic nutrient sources for soil fertility maintenance and
replenishment. In: Replenishing Soil Fertility in Africa. Buresh, R. J.,
Sanchez, P. A. and Calhoun, F. G. (Eds). pp196-217. SSSA Special
Publication No.5!: Soil Science society of America, Madison, WI, USA.
Proceedings of an International Symposium.
Palm, C. A. and Sanchez, P. A. (1991) Nitrogen release from the leaves of some
tropical legumes as affected by their lignin and polyphenolic contents. Soil
Biol. Biochem 23: 83-88.
Pereira, H. C., Wood, R. A., Brzostowski, H. W. and Hosegood, P.A. (1958) Water
conservation by fallowing in semi-arid tropical East Africa. Emp. J. Exp.
Agr. 26: 203-228.
Philip, J. R. (1957) The theory of infiltration: 4. Sorptivity and algebraic infiltration
equations. Soil Science 84: 257-264.
Changes in Soil Properties and Their Effects on Maize Productivity Following Sesbania
Sesban and Cajanus Cajan Improved Fallow Systems in Eastern Zambia
111
Rhoades, C. C. (1997) Single-tree influences on soil properties in tropical
agroforestry systems. Agroforesty Systems 38: 3-50.
Sakala, W. D., Gadish, G. and Giller, K. E. (2000) Interactions between residues
of maize and pigeonpea and mineral N fertilizers during decomposition and
N mineralization. Soil Biology and Biochemistry. 32: 679-688.
Sanchez, P. A. (1976) Properties and management of soils in the tropics. New
York: John Wiley, pp 618.
Tisdall, J. M. and Oades, J. M. (1983) Organic matter and water stable aggregates
in soils. Journal of Soil Science 33: 141-163.
Torquebiau, E. F. and Kwesiga, F. (1996) Root development in Sesbania sesban
fallow-maize system in eastern Zambia. Agroforestry systems 34: 193-211.
USDA (United States Department of Agriculture) (1975) Soil Taxonomy,
agricultural handbook number 436, Soil Conservation Service, USA.
Yamoah, C. F., Agboola, A. A., Wilson, G. F. and Mulongoy, K. (1986) Soil
properties as affected by the use of leguminous shrubs for alley cropping
with maize. Agriculture Ecosystems and Environment. 18: 167-177.
Yoder, R. E. (1936) A direct method of aggregate analysis of soils and a study of
the physical nature of erosion losses. J. Am. Soc. Agron. 28: 337-351.
Young, A. (1997) Agroforestry for soil management. CAB International,
Wallingford, UK and ICRAF, Nairobi, Kenya. pp 276.
112
Chirwa, T. S. et al
Tillage Effects on Soil Organic Carbon and Nitrogen Distribution in Particle Size
Fractions of a Red Clayey Soil Profile in Zimbabwe
Tillage Effects on Soil
Organic Carbon and Nitrogen
Distribution in Particle Size
Fractions of a Red Clay
Soil Profile in Zimbabwe
113
7
Chivenge, P.P1*, Murwira, H.K.1 and
Giller, K.E.2
1
Tropical Soil Biology and Fertility (TSBF), P O Box MP 228,
Mt Pleasant, Harare, Zimbabwe.
2
Soil Science and Agricultural Engineering Department,
Faculty of Agriculture, University of Zimbabwe, Box MP167,
Mt Pleasant, Harare, Zimbabwe (E-mail:ken.giller@wur.nl)
* Corresponding author: Tel: 263-4-333243/4, Fax: 263-4-333244, Email: tsbfzim@zambezi.net
Abstract
The long-term tillage effects on soil organic C and N
distribution in particle size fractions of a chromic luvisol
(FAO soil classification) soil profile was evaluated. Three
treatments, conventional tillage, mulch ripping and weedy
fallow from a long-term tillage experiment established in
1988/89 season were compared.
Relative to the weedy fallow, conventional tillage showed
a more marked decline in organic C and total N than mulch
ripping. There was a general decline in organic C, total N,
and C and N in the soil organic matter fractions with depth
114
Chivenge, P. P. et al
for mulch ripping and the weedy fallow while there was a
more or less uniform distribution with depth under
conventional tillage. The decline in organic C and total N
was more pronounced in the surface horizons with about
20% and 50% decline for mulch ripping and weedy fallow,
respectively, from the 0-2 cm to the 2-5 cm depths. There
were no treatment differences in organic C and total N
distribution below the plough layer (30 cm).
For conventional tillage and mulch ripping, the largest
decline in organic C was in the coarse sand fractions (2122000 µm), whereas the least decline in organic C was observed
in the clay fractions of 22% and 13% for conventional tillage
and mulch ripping respectively, when compared with the
weedy fallow. There was a decline in total N in the organic
matter fractions with depth. The sand fractions had the least
organic C and total N at all depths than silt and clay fractions
for conventional tillage and mulch ripping, and the silt
fractions had intermediate concentrations. All the size
fractions from the weedy fallow had high N content in the 02 cm soil layer where silt, medium sand and coarse sand
fractions, had even higher N contents than the clay fraction.
We concluded that mulch ripping promotes soil organic
matter accumulation and reduces soil organic matter loss.
This build up in soil organic matter from the decomposition
of residue mulch is significant when compared with
conventional tillage. It was also concluded that higher organic
C losses under conventional tillage were due to intensive
cultivation and associated soil losses through erosion.
Keywords: soil organic matter, conventional tillage, mulch ripping and
weedy fallow
Introduction
Type and length of tillage practice influence the amount of soil organic
matter (SOM), present in the soil and the rate of SOM turnover and its
distribution among size fractions down the profile (Cambardella and
Elliot, 1994). Conventional tillage mixes upper and lower horizon soils
and disrupts aggregate protected organic matter (Hassink, 1995;
Jastrow, 1996). This results in faster decomposition and loss of organic
matter and more or less uniform distribution of organic matter in the
plough layer (Etana et al., 1999; Stockfisch et al., 1999).
The change in total organic C pool size as affected by soil management
practice can be expressed using the carbon pool index (CPI), which is
Tillage Effects on Soil Organic Carbon and Nitrogen Distribution in Particle Size
Fractions of a Red Clayey Soil Profile in Zimbabwe
115
calculated from sample total C expressed as a fraction of the reference
total C (Blair et al., 1997). The higher the organic C loss, the lower the
carbon pool index and the more difficult it is to rehabilitate, especially
for soil which has a small initial total carbon pool.
Conventional tillage systems are known to increase organic matter
losses from the soil whereas no till practices can improve soil aggregation
thus increasing physical protection and maintenance of SOM (Doran et
al., 1987; Feller and Beare, 1997). Accumulation of SOM under minimum
tillage is limited to the surface while conventional tillage, which
incorporates organic inputs, may affect SOM and other properties to a
greater depth (Fernandes et al., 1997). Work done by Beare et al. (1994)
showed that under similar management, soil organic carbon content of
no till surface soils (0-5 cm) was 18% higher than that of conventional
tillage after eleven years of continuous treatment.
The objective of this experiment was to assess the effects of
conventional and conservation tillage in relation to a reference point
(weedy fallow) on:
a) total N and reduction of organic C as reflected by the C pool index
(CPI);
b) SOM content and distribution of C among SOM fractions; and
c) distribution of organic C, N and SOM fractions down the profile.
It was hypothesised that conservation tillage results in the
accumulation of SOM especially in the surface horizons whereas
conventional tillage results in lower SOM amounts that is uniformly
distributed down the profile.
Materials and Methods
Measurements for this experiment were taken from a tillage experiment,
which was established in 1991/92 season at the Institute of Agricultural
Engineering (IAE), Harare, Zimbabwe (17°45’S; 31°10’E). The IAE site is
in an agro-ecological zone that receives an annual rainfall of 800-1000
mm and is on a red clay soil derived from gabbro parent material and is
classified as rhodic paleustalf (USDA) or chromic luvisol (FAO).
Runoff plots of 10 m x 30 m planed at 4.5% slope were laid out for
five treatments; conventional tillage, mulch ripping, clean ripping, no
till tied ridging and no till strip cropping. Work reported in this paper is
from two treatments, mulch ripping and conventional tillage, and from
a weedy fallow found close to the treatment plots for use as a reference
point. Conventional tillage consisted of annual ox-mouldboard ploughing
to 25 cm depth. Mulch ripping involved ripping between rows into
residues to a depth of about 27 cm. Perennial grass was growing naturally
with no tillage operations and no fertiliser additions for the weedy fallow.
However, the weedy fallow had been cropped since 1980 up to the onset
116
Chivenge, P. P. et al
of the experiment when it was converted to be a grass fallow. There
were annual fertiliser additions of 115 kg N ha –1, 22 kg P ha -1 and
25 kg K ha-1 to mulch ripping and conventional tillage. Maize was grown
annually as the test crop.
Soil samples were collected from the treatment plots and from a
weedy fallow in July 1999 at the following depths; 0-2, 2-5, 5-10, 1020, 20-30 and 30-60 cm. Another set of samples was collected from the
0-30 cm depth (plough layer) from the treatment plots and the weedy
fallow, using an auger. Soil was sieved through a 2 mm sieve and
analysed for organic C using the Walkley Black method and total N
using the Kjeldahl analysis method (Anderson and Ingram, 1993).
Soil was fractionated according to the method by Feller et al. (1996).
Fifty grams of soil was shaken overnight in 200 ml of 0.5% sodium
hexametaphosphate after soaking the soil in water overnight. Soil was wet
sieved through a series of sieves to separate 212-2000 µm (coarse sand),
53-212 µm (medium sand) and 20-53 µm (fine sand) fractions. The 0-2 µm
(clay) and 2-20 µm (silt) fractions, were separated by the sedimentation
method based on Stoke’s Law (Hillel, 1982). The fractions were dried in an
oven at 50°C and analysed for organic C using the modified Walkley-Black
method and total N using the Kjeldahl analysis method. Statistical analysis
was done using Genstat for analysis of variance (ANOVA).
Results
Tillage effects on total organic C, C pool index and total N
Conventional tillage and mulch ripping resulted in a decrease in soil
organic C content, compared with the weedy fallow (Table 7.1).
Table 7.1: Tillage effects on total organic carbon (%), total nitrogen (%) and CPI in the 0-30 cm
depth of a red clay soil from the Institute of Agricultural Engineering, Harare, Zimbabwe
Tillage practice
Conventional tillage
Mulch ripping
Weedy fallow
Total organic C (%) Total N (%)
1.49
1.72
2.79
0.11
0.15
0.19
C/N ratio
CPI
13.6
11.5
14.7
0.53
0.62
1.00
Soil under conventional tillage had smaller amounts of % organic C
and % total N compared with mulch ripping (Table 7.1). Weedy fallow
had almost twice as much total organic C and total N contents as
conventional tillage and mulch ripping. The C/N ratio was highest under
the weedy fallow while it was lowest for the mulch ripping treatment.
Conventional tillage had a lower soil organic carbon pool index than
mulch ripping, reflecting a higher reduction in the total C pool under
Tillage Effects on Soil Organic Carbon and Nitrogen Distribution in Particle Size
Fractions of a Red Clayey Soil Profile in Zimbabwe
117
conventional tillage compared with mulch ripping (Table 7.1). The carbon
pool index (CPI), of conventional tillage was almost half that of the weedy
fallow as was shown by a 47% decline in organic C. There were however
no significant differences in maize yield under mulch ripping and
conventional tillage over the seasons (Figure 7.1).
Conventional tillage caused higher surface runoff losses over the
seasons compared with mulch ripping which had low surface runoff
losses even in years with high rainfall (Figure 7.2). Surface runoff and
soil erosion are other pathways in which soil organic matter was lost.
Figure 7.1: Tillage effects on maize yield over nine seasons at the Institute of Agricultural
Engineering, Harare, Zimbabwe
Figure 7.2: Tillage effects on runoff over nine seasons at the Institute of Agricultural
Engineering, Harare, Zimbabwe
118
Chivenge, P. P. et al
Tillage effects on organic C in the SOM fractions
Soil under mulch ripping had higher organic C content in the organic
matter size fractions compared with conventional tillage except for the
coarse sand organic matter fractions which was not significantly different
for conventional tillage and mulch ripping (Table 7.2).
Table 7.2: Tillage effects on organic carbon distribution in soil organic matter size fractions
of a red clayey soil in Harare, Zimbabwe
Tillage
treatment
Organic C in SOM size fractions (mg C g -1 soil)
212532000 µm 212 µm
coarse Medium
sand
sand
20- 5-20 µm 0-5 µm
53 µm
Silt
Clay
Fine
sand
Sum
Total
%
measured Recovery
Conventional
tillage
0.97
0.95
0.84
1.69
8.1
12.6
14.9
84.6
Mulch ripping
1.04
1.34
1.00
2.09
9.0
14.5
17.2
84.3
Weedy fallow 4.47
SED
0.167
3.57
0.187
1.90
0.091
3.15
0.118
10.4
0.242
23.5
27.9
84.2
NB n=3 except for the weedy fallow where n=1
There was however more than a four fold decrease in organic C in
the coarse sand organic matter fraction for mulch ripping (1.04 mg C g1
soil) and conventional tillage (0.97 mg C g-1 soil), when compared with
the reference point, the weedy fallow (4.47 mg C g-1 soil) (Table 7.2). The
smallest decline in organic C was in the clay fractions. Relative to the
weedy fallow, conventional tillage caused a larger decline in organic C
in the clay (22%) and silt (46%) fractions compared with mulch ripping
which caused a 13% and 34% decrease in organic C in clay and silt
fractions, respectively (Table 7.2). There were no complete recoveries of
organic C for all the treatments indicating that the fractionation method
does not account for the entire C in the soil (Table 7.2). Some of the
soluble organic C was lost during fractionation.
Tillage effects on total organic C and total N distribution
down the soil profile
There was a general decline in total organic C and N with depth for all
the treatments but this was less pronounced for conventional tillage
(Figure 7.3).
Tillage Effects on Soil Organic Carbon and Nitrogen Distribution in Particle Size
Fractions of a Red Clayey Soil Profile in Zimbabwe
119
Figure 7.3: Tillage effects on a) organic C and b) total N distribution down the profile of
a red clayey soil at the Institute of Agricultural Engineering, Harare, Zimbabwe
TotalN %
O rganic C %
0
1
2
3
4
0.0
5
0-2
2-5
0-2
2-5
5-10
5-10
10-20
10-20
20-30
20-30
0.1
0.2
0.3
0.4
0.5
D epth (cm )
Conventional
Conventional
M ulch
30-60
W eedy fallow
M ulch
30-60
W eedy fallow
The decrease in total N and organic C was more pronounced in the
surface horizons (0-10 cm) of mulch ripping and weedy fallow, where
there was about 20% decline in organic C and total N from 0-2 cm to
2-5 cm depth for the mulch ripping treatment. Total N and organic C
contents decreased by about 50% from the 0-2 cm to the 2-5 cm depth
for the weedy fallow. In the upper 20 cm weedy fallow had the highest
total N and organic C contents while mulch ripping had higher total N
and organic C contents than conventional tillage (Figure 7.3). At depths
below 20 cm, there were no treatment differences on total N and organic
C content. The 30-60 cm depth had the lowest organic C and total N
contents for all the treatments. The C/N ratio did not change with
depth.
Tillage effects on organic carbon and nitrogen distribution
in size fractions down the profile
There were no significant differences in the distribution of organic C
and N in the coarse, medium and fine sand fractions for the
conventional tillage and mulch ripping treatments at all depths (Figures
7.4 and 7.5).
120
Chivenge, P. P. et al
Figure 7.4: Effects of a) conventional tillage, b) mulch ripping and c) weedy fallow on
the distribution of organic C in the organic matter size fractions down the profile of a red
clay soil from Harare, Zimbabwe
a) CT
0
2
C in fraction (mg C g-1 soil)
4
6 8 10 12 14 16
0-2
2-5
5-10
10-20
20-30
30-60
c) WF
0
2
4
6
8
10
12
14 16
0-2
2-5
5-10
10-20
30-60
Figure 7.5: Effects of a) conventional tillage, b) mulch ripping and c) weedy fallow on
the distribution of N in the organic matter size fractions down the profile of a red clay soil
from Harare, Zimbabwe
Fraction N (mg g-1 soil)
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
b) MR
0.0 0.2
Fraction N (mg g-1 soil)
0.4
0.6 0.8 1.0
0-2
2-5
5-10
10-20
20-30
20-30
30-60
c) WF
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.41.6
0-2
2-5
5-10
10-20
20-30
30-60
Coarse sand
Medium sand
Fine sand
Silt
Clay
1.2
1.4 1.6
Tillage Effects on Soil Organic Carbon and Nitrogen Distribution in Particle Size
Fractions of a Red Clayey Soil Profile in Zimbabwe
121
There were no significant changes in organic C and N in the sand
fractions with depth for mulch ripping and conventional tillage. The sand
fractions had the lowest amounts of organic C and N at all depths followed
by silt, while clay had the highest for the conventional and mulch ripping
treatments. Total N and organic C in the silt fractions of conventional
tillage and mulch ripping did not change significantly with depth in the
upper 30 cm (Figure 7.4 and Figure 7.5). There was a decline of organic
C and N in the silt fraction from the 20-30 cm to 30-60 cm depth by
about 40% and 50% for conventional tillage and mulch ripping respectively
(Figure 7.4 and Figure 7.5). There were increases in the organic C and N
content of the clay fraction from the 0-2 cm depth to the 5-10 cm depth
for conventional tillage and from 0-2 cm depth to 20-30 cm depth for
mulch ripping followed by a decline at depths below (Figure 4 and Figure
5). The biggest decline was from the 20-30 cm depth to the 30-60 cm
depth where there was almost a 50% decline for mulch ripping and
conventional tillage (Figure 7.4 and Figure 7.5).
For the weedy fallow, all the fractions had high organic C and N
contents in the surface layer (0-2 cm) (Figure 7.4 and Figure 7.5). Unlike
the other treatments the coarse sand and the silt fractions for the weedy
fallow had high organic C and N contents in the 0-2 cm depth. Organic
C and N contents in the silt fraction were almost twice that in the clay
fraction in the 0-2 cm depth for the weedy fallow. There was almost a
four-fold decline in organic C and N contents in the coarse sand fraction,
from the 0-2 cm to the 2-5 cm depth. At depths below 2 cm, there were
no differences in organic C and N contents in the sand fractions. The
organic C and N contents in the sand fractions did not significantly
change with depth. There was a decrease in organic C and N contents
in the silt fraction with depth with the biggest decline from the 20-30
cm depth to the 30-60 cm depth. There was an increase in organic C
and N contents in the clay fraction from the 0-2 cm depth to the 2-5 cm
depth followed by a decline at depths below. Organic C and N contents
were higher than in the silt fraction in the clay fraction in the depths
above 5 cm and was lower at depths below for the weedy fallow.
Discussion
Tillage effects on total organic C, total N and CPI
The lower organic C and total N for conventional tillage was probably a
result of high organic matter decomposition enhanced by disruption of
aggregates (Table 7.1) (Hassink, 1995). This could have been enhanced
by the removal of residues under conventional tillage compared with
mulch ripping where residues are left on the surface increasing organic
matter inputs. The lower CPI for conventional tillage means that
122
Chivenge, P. P. et al
conventional tillage stimulates organic matter decomposition while
mulch tillage enhances organic matter retention (Table 7.1) (Blair et al.,
1997). Conventional tillage also had lower total N than mulch ripping
most probably due to low organic matter content under conventional
tillage (Table 7.1).
The additions of organic matter due to litter fall and lack of tillage
resulted in higher organic C under the weedy fallow than cultivated soil
(Table 7.1). The weedy fallow had the highest total N value but its C/N
ratio was much higher than that of mulch tillage. This was probably
because tillage enhances soil organic matter decomposition lowering
the C/N ratio of cultivated soil. Under the weedy fallow, there is
predominantly grass litter, which could have resulted in a high C/N
ratio. The lack of differences in maize yield under conventional tillage
and mulch ripping could have been due to annual fertilizer additions
such that there were no nutrient limitations (Figure 7.1). The higher
surface runoff losses under conventional tillage could have been a result
of disruption of soil aggregates, increasing the susceptibility of soil to
raindrop detachment and associated runoff and soil erosion (Figure
7.2).
Tillage effects on SOM fractions
The higher organic C in the soil organic matter size fractions under
mulch ripping compared with conventional tillage, was caused by higher
additions of soil organic matter through returning of residues and
reduced tillage intensity (Table 7.2). Organic C in the coarse sand fraction
in mulch ripping was expected to be higher than conventional tillage
because of additions of organic residues to the soil. The lack of difference
could have been due to the fact that sampling was done at the end of
the season such that much of the added litter had broken down into
smaller sizes. The decrease of organic C in the coarse sand fraction
from the weedy fallow to conventional tillage and mulch ripping was
probably a result of little organic residue additions under cultivated
soil when compared with all year round additions of organic residues
under the weedy fallow (Table 7.2). Lower organic C contents in the fine
sand, silt and clay fractions in conventional tillage compared with mulch
ripping resulted from disruption of aggregates (Table 7.2) (Hassink, 1995)
and the resulting decomposition of organic matter and dilution of organic
matter due to mixing of plough layer.
Tillage effects on organic C and total N distribution down
the soil profile
The higher organic C and total N in the soil surface was a result of
additions of organic residues on the soil surface for mulch ripping and
Tillage Effects on Soil Organic Carbon and Nitrogen Distribution in Particle Size
Fractions of a Red Clayey Soil Profile in Zimbabwe
123
weedy fallow (Figure 7.3). Total N and organic C was low in the surface
horizon for conventional tillage because of little or no organic residues
added in the soil and tillage disrupts aggregate protected organic matter
(Hassink, 1995). Beare et al (1994), found that there was 18% more
organic C for a no tillage treatment than conventional tillage after 11
years of continuous treatment. There was a uniform distribution of
organic C and total N in the upper 20 cm of soil under conventional
tillage caused by mixing of upper and lower horizon soil (Etana et al.,
1999).
In this study, the sharp decline in organic C and total N under the
weedy fallow from the 0-2 cm to the 2-5 cm depth, was probably because
of the large additions of litter to the surface horizon (Figs. 3 and 4). Soil
under the weedy fallow was not tilled, such that there was no mixing of
the litter layer with the subsurface horizon. For depths below 20 cm,
there were no treatment differences even for the conventional tillage
treatment because there was little or no mixing of soil beyond that depth.
The 30-60 cm depth had the lowest organic C and total N contents for
all the treatments. This is because, at this depth there are limited
additions of organic residues. The C/N ratio was not affected by depth
for all the treatments because as organic C content decreased down the
profile so did the total N content.
Tillage effects on organic carbon and nitrogen distribution
in size fractions down the profile
The high C and N contents in the clay fractions for the conventional
tillage and mulch ripping treatments was probably due to the high clay
content in the soil, which could have protected organic matter from
microbial decomposition and hence high C and N. The decline in organic
C and total N contents in the silt fractions from the 20-30 cm depth to
the 30-60 cm depth reflected changes in total N with depth.
High litter fall under the weedy fallow resulted in high organic C
and total N contents associated with all the fractions in the surface
layer. Litter on the soil surface was probably in different stages of
decomposition, such that all the fractions had high organic C and N
contents. Organic C and total N in the clay fraction was lower in the silt
than coarse and fine sand fractions on the soil surface probably because
of high litter content such that there was a lower proportion of organic
matter associated with the clay fraction compared to the other fractions.
Conclusion
Conventional tillage results in faster decomposition and loss of soil
organic matter than mulch ripping, as was shown by the 47% decline
124
Chivenge, P. P. et al
in organic carbon under conventional tillage, compared to 38% organic
carbon decline under mulch ripping. Mulch ripping promotes physical
protection and soil organic matter accumulation in the plough layer,
compared with conventional tillage. Conventional tillage results in the
mixing of the surface soil horizon, diluting the organic layer and resulting
in a more uniform distribution of organic matter in the plough layer.
Mulch ripping on the other hand involves minimum tillage and promotes
organic matter in the surface horizons.
Acknowledgements
This work was sponsored by a grant from IFAD to TSBF. Special
acknowledgements also go to staff from the Soil and Water Conservation
Branch, Institute of Agricultural Engineering (IAE), namely, Mr. I.
Nyagumbo, Mr. M. Munyati and Dr. G. Nehanda, for providing us with
data on soil loss, surface runoff and maize yields. We also thank them
for allowing us to work on their site.
References
Anderson, J. M. and Ingram, J. S. I. (1993) Tropical Soil Biology and Fertility: A
Handbook of Methods. Second edition. CAB International, Wallingford, UK.
Beare, M., Cabrera, M. L., Hendrix H., P. F. and Coleman, D. C. (1994) Aggregateprotected and unprotected pools of organic matter in conventional and notillage soils. Soil Sci. Soc. Am. J. 58:787-795.
Blair, G. J., Lefroy, R. D. B., Singh, B. P. and Till, A. R. (1997) Development and
use of a carbon management index to monitor changes in soil C pool size
and turnover rate. In: Cadisch, G. and Giller, K. E. (eds) Driven by Nature:
Plant Litter Quality and Decomposition. CAB International.
Cambardella, C. A. and Elliot, E. T. (1994) Carbon and nitrogen dynamics of
soil organic matter fractions from cultivated Grassland soils. Soil Sci. Am.
J. 58: 123-130.
Doran, J. W., Mielke, L. N. and Power, J. F. (1987) Tillage/Residue management
interactions with the soil environment, organic matter and nutrient cycling.
In Cooley J. H. (ed), Soil Organic Matter Dynamics and Soil Productivity.
INTECOL Bulletin. 1987:15.
Etana, A., Hakansson, I., Zagal, E. and Bucas, S. (1999) Effects of depth on
organic carbon content and physical properties in five Swedish soils. Soil &
Tillage Research 52:129-139.
Feller C., Albrecht, A.. and Tessier, D. (1996) Aggregation and organic matter
storage in kaolinitic and smectitic tropical soils. In Carter, M. R. and Stewart,
B. A. (eds), Structure and Organic Matter Storage in Agricultural Soils. Lewis
Publishers, London.
Tillage Effects on Soil Organic Carbon and Nitrogen Distribution in Particle Size
Fractions of a Red Clayey Soil Profile in Zimbabwe
125
Feller, C. and Beare, M. H. (1997) Physical control of soil organic matter dynamics
in the tropics. Geoderma 79: 69-116.
Fernandes, E. C. M., Motavalli, P. P., Castilla, C. and Mukurumbira, L. (1997)
Management of soil organic matter dynamics is tropical land-use systems.
Geoderma 79:49-67.
Hassink, J. (1995) Density fractions of soil macroorganic matter and microbial
biomass as predictors of C and N mineralization. Soil Biol. Biochem. 27:10991108.
Hillel, D. (1982) Introduction to Soil Physics. Second edition. Academic Press,
London.
Jastrow, J. D. (1996) Soil aggregate formation and the accrual of particulate
and mineral-associated organic matter. Soil Biol. Biochem. 28: 665-676.
Stockfisch, N., Forstreuter, T. and Ehlers, W. (1999) Ploughing effects on soil
organic matter after twenty years of conservation tillage in Lower Saxony,
Germany. Soil & Tillage Research 52: 91-101.
126
Chivenge, P. P. et al
Combating Nutrient Depletion in East Africa – the Work of the SWNM Program
Combating Nutrient Depletion
in East Africa – the work of
the SWNM program
127
8
Delve, R.J.
TSBF Institute of CIAT, c/o CIAT-Uganda, P.O. Box 6247,
Kawanda, Kampala, Uganda
Introduction
A workshop was held 11-16 February 1996 at CMRT, Egerton University
to launch the Combating Nutrient Depletion (CND) theme of the Soil,
Water, and Nutrient Management Program (SWNMP) of the CGIAR. The
meeting at Egerton, convened by Tropical Soil Biology and Fertility
Programme (TSBF), brought together partners from national institutions
of the countries participating in the African Highlands Program, Ethiopia,
Kenya, Madagascar, and Uganda as well as international centers
collaborating in the region, and expertise from outside the region. The
project is a joint action of the African Highlands Initiative (AHI) and the
Soil Water and Nutrient Management Programme (SWNMP) of the CGIAR.
The project is implemented by a consortium of national, regional and
international agencies with a wide range of expertise and experience in
nutrient management research, farmer participatory research,
agricultural problem-solving and development actions. The project is
co-ordinated by TSBF on behalf of the SWNMP and represented within
AHI through the technical support group.
The objective of the CND – East Africa Consortium are:
1. Integrating nutrient management practices that redress nutrient
imbalances and environmental degradation.
128
Delve, R.J.
2. Enabling policies for combating nutrient depletion.
3. Assisting farmers to adopt improved nutrient management practices.
Erosion and nutrient depletion are major causes of land degradation
and loss of productivity in the infertile, mainly acid soils of the humid
and sub-humid tropics. These two constraints threaten the livelihood
and food security of up to 630 million people who occupy most of the
1.8 billion ha of marginal or fragile areas where nutrient depletion and
soil erosion are most prevalent.
Unlike soil erosion nutrient depletion is generally a reversible
constraint because diverse nutrient resources are usually available to
resource poor farmers but are often underutilised because of a lack of
knowledge or other constraint such as labour. Available resources may
include commercial fertilizers, agro-mineral deposits, agro-industrial
by-products and locally produced or harvested organic materials.
Income generation by smallholders, and subsequent investment in
soil amendments and erosion control measures, are essential
components to combating soil depletion and erosion. Nutrient access
operates at the national level and is policy-driven while nutrient
availability operates at local levels and is strongly influenced by farmer
opportunities and decision-making. Improvements in both nutrient
access and availability require innovative solutions achieving nutrient
recapitalization, better fertilizer distribution and marketing, better
integrated nutrient management technologies and refinements in farmer
participatory techniques. Erosion control measures generally need to
provide a product of immediate financial or other benefit in order to be
accepted, adopted and adapted by land users.
Project purpose
The stated purpose of the project was to develop for use by farmers,
researchers and policy makers, effective decision support systems for
the integrated management of nutrient resources and soil erosion,
including a simple cost-benefit analysis for their on and off-site effects.
Development of Decision Support Tools
Fertilizer equivalency of organic materials
Research over the past century has related N release patterns to the
resource quality, or chemical characteristics of organic materials (Heal
Combating Nutrient Depletion in East Africa – the Work of the SWNM Program
129
et al., 1997). The N concentration and the C-to-N ratio of the material
still probably serve as the most robust indices when all plant materials
are concerned (Constantinides and Fownes, 1994). Lignin and
polyphenols are, however, important modifiers of N release for the fresh,
non-senescent leaves of high-quality materials (Constantinides &
Fownes, 1994). The delayed N release resulting from polyphenolics,
particularly condensed tannins, may be much longer than the temporally
N immobilization resulting from high C-to-N ratios in cereal crop
residuals (Giller et al., 1997). Beneficial effects of the individual and
combined use of organic and inorganic nutrients on soil fertility, crop
yields, and maintenance of SOM have been repeatedly shown in
laboratory and field trials, yet there are no predictive guidelines for
their management, such as those that exist for inorganic
fertilizers.
A set of hypothesis has been placed in a decision tree for selecting
organic materials for soil N management based on their quality (Palm et
al., 1997). Organic materials with N content above 2.5%, lignin and
polyphenol contents less than 15% and 4%, respectively can be
incorporated directly with annual crops. The recommendation for organic
materials with N content above 2.5% and lignin and polyphenol contents
more than 15% and 4% is to mix with N fertilizer or high quality organic
material. For organic materials with N content <2.5% and lignin content
<15%, the recommendation is to mix with N fertilizer or add to compost.
Organic materials recommended for surface application for weed,
erosion, and water control are those with N content <2.5% and lignin
content >15% (Figure 8.1).
Figure 8.1: Decision Tree for Selecting Organic Imputs for Nitrogen Management
Organic Resource Database
% N > 2.5
yes
no
Lignin < 15%
Phenol < 4%
yes
Incorporate
directly
with
no
Mix with
fertilizer or
high quality
Lignin < 15%
yes
Mix with
fertilizer or
add to compost
no
Surface apply
for erosion and
water control
130
Delve, R.J.
Network trials using a wide range of mainly high quality organic
materials in combinations with inorganic N fertilizers were established
in eastern and southern Africa. The objectives of the trials were (i)
establishing fertilizer equivalency values of organic materials based
on quality and (ii) to investigate benefits and trade-offs of combining
organic and inorganic N sources. Field trials were carried out in five
countries in sub-Saharan Africa to establish fertilizer equivalency
values of organic materials based on their resource quality and to
investigate the effects of combining organic and inorganic N sources
on maize yield. Organic materials used were leaves of Tithonia, Senna,
Calliandra, Sesbania, Tephrosia, and pigeon pea litter. Tithonia, Senna
and Tephrosia (all with %N > 3.5%) had percent fertilizer equivalencies
of about 100% and would be classified as high quality organic materials
that can be recommended for direct application as N sources. The
good performance of Tephrosia in increasing maize yield indicates that
the lignin content of 18 to 19% may be a good critical value. The fairly
poor performance of Calliandra, Sesbania, pigeon pea litter and Neem
(all with %N > 2.5) was due to high levels of polyphenols and/or lignin.
The management recommendation that is suggested for these materials
is to mix with N fertilizer or high quality organic material. Following
application of maize stover (%N < 1.5), yields were lower or comparable
to that from the control (no inputs) and hence, can be best used when
mixed with N fertilizer or added to compost. The fertilizer equivalencies
for organic materials with %N > 2.5 were positively correlated (r =
0.86, P = 0.01) to their N content, showing the dependency of the
fertilizer equivalency of an organic material on its N content. The linear
function indicated that with an increase of 0.1% N in the tissue of the
plant material, there is 8% increase in the fertilizer value. The critical
level of N content of organic materials for the transition for increasing
crop yield relative to 0 N was 2.4% (Figure 8.2). There were positive
interactions when Sesbania and Tephrosia were applied in combination
with inorganic fertilizers, while it was more beneficial to apply Tithonia
and Senna than both the N fertilizer and the organic-N fertilizer
combinations. Better returns were obtained with the addition of
inorganic N fertilizer rather than pigeon pea litter. More organic
materials need to be tested against a set of hypothesis that has been
placed in a decision tree for selecting organic materials for soil N
management and to improve the predictability of the
relationship between fertilizer equivalencies and N content of organic
inputs.
These quality parameters were further tested against data held in
the Organic Resource Database. It was found for a wide range of
incubation experiments and organic resource quality that there was
very close agreement to this initial decision tree (Figure 8.3).
Combating Nutrient Depletion in East Africa – the Work of the SWNM Program
131
Figure 8.2: The relationship between percent fertilizer equivalencies and N content of
organic materials (Regression lines excludes Calliandra and maize stover)
Fertilizer Equivalency (%)
140
Sen
Slope = 56
R2 = 0.45
100
TV
TD
Ri
CG
60
TD
CG
Cy
Ag
GS
Ses
CjC
20
0
-20
Sc
Ses
CC
AI
AI
ZM
ZM
ZM
1
2
3
4
N content
% of initial N released
Figure 8.3: Nitrogen mineralized or immobilized after 8 weeks from organic materials
from 11 incubation studies as determined by the N concentration of the materials and
modified by high lignin or polyphenol concentrations. The regression equation is for all
materials. Filled squares or circles represent materials with %N > 2.5m open square or
circles %N < 2.5; squares represent materials with % < lignin and %polyphenol 4; circles
represent materials with %lignin > 15 or %polyphenol > 4
Initial N concentration (%) of organic material
(Palm et al., 2001)
Delve, R.J.
132
Management of legume cover crop and biomass transfer
species
The above example shows how we can select a particular type of
management practice depending on the organic resource quality.
Farmers also need to know that once they have selected a technology
based on these criteria how do they manage that technology. For example,
if they have a high quality resource how much of it should they
incorporate it directly into the soil. Or, which species should they choose
given their production objectives and available resources? Much onfarm evaluation of legumes in central Uganda has led to the production
of a legume use decision guide (Figure 8.4). This selects a particular
legume species based on the intended benefit the farmer wants from
using a legume.
Figure 8.4: Decision aid for LCC use to allow farmers to select the best species for their
production system and to address their production objectives
Decision Tree for Green Manure Use
If you want to produce a sole crop
Plant Mucuna
If you want to intercrop with maize
If you want to suppress weeds
Plant Lablab
If you want to produce fodder
Plant Canavalia
If you want to reduce nematode numbers
If you want a durable mulch
Plant
Clotalaria
Adaptive research to test and refine decision support tools
The above decision guide on selecting a legume has been tested in eastern
Uganda since 1999 through adaptive research with a range of
governmental, non-governmental and research partners. This work has
just started and the results will be used to update the decision guide as
Combating Nutrient Depletion in East Africa – the Work of the SWNM Program
133
more data is collected. In addition, other organic resources were
investigated, e.g. Tithonia for biomass transfer. Some research highlights
are:
•
Incorporation of 50% and 100% of one season Mucuna and
canavalia fallow biomass was assessed to look at the fallow
management options. Both incorporation rates of Mucuna and
100% incorporation of canavalia increased grain yields by more
than 200% compared to the control in two areas of eastern Uganda
(Figure 8.5 and Figure 8.6). Incorporation of 100% rather than
50% of the biomass produced in the plot did not significantly
increase maize grain yields compared to the control. This would
allow the farmer to produce the biomass in one place and to apply
the biomass over twice the area, i.e. to use one part for incorporation
and the other for biomass transfer or livestock feed. This is crucial
where land sizes and fallow areas are limited, and little area is
available for non-food crop production.
•
Tithonia biomass, which is abundant in the district on roadsides
and farm boundaries, can increase yields considerably over no input
situation (Figure 8.7). In this trial, the treatments were balanced to
add 60 kg N ha-1.
•
A mechanism for on-farm seed production has been implemented.
As a back-up, seed stands at the DFI /DATIC have also been
established for the seeds in greatest demand. Over 200 kg of starter
seeds of legumes has been supplied to the farmers between 1999
and 2000, with a further 1200 kg distributed so far in 2001. Seeing
the potential of this work, NARO’s Outreach programme is now
funding seed production at the DFI and in all the ARDCs in Uganda
in order to get enough seeds for Tororo and other districts where it
wants to promote the technologies. These figures do not include
the amount that farmers themselves have used from their own
stands.
•
Strong linkages with National Agricultural Research Organisation
(NARO) for transfer of knowledge and materials to other parts of the
country have been developed.
•
A strong partnership for scaling up impact to the whole district and
other neighbouring districts in eastern agroecological zone of Uganda.
This has been through distribution of seed to farmers through
government extension staff and field exchange visits.
Delve, R.J.
134
Figures 8.5: Maize grain yield following a one season fallow in Kisoko sub-county
Isd = 1.4
Maize yield (t/ha)
5.0
4.0
4.5
3.7
3.8
3.2
3.0
2.0
1.9
1.9
Control
F.P
Control
PK
1.0
100%
Canavalia
50%
Canavalia
100%
Mucuna
50%
Mucuna
Figures 8.6: Maize grain yield following a one season fallow in Osukuru sub-county
Isd = 1.0
4.0
Maize yield (t/ha)
3.3
3.0
3.0
3.4
3.4
1.9
2.0
1.6
1.0
0.0
100%
Canavalia
50%
Canavalia
100%
Mucuna
50%
Mucuna
Control
F.P
Control
PK
Combating Nutrient Depletion in East Africa – the Work of the SWNM Program
135
Figures 8.7: Maize grain yield following biomass transfer with Tithonia
LR2000 - application Isd = 1.18
SR2000 - residual
Maize yield (t ha-1)
4.0
Isd = 0.98
3.0
2.0
1.0
0.0
Control
1/2 Tithonia +
NPK
NPK
PK
Tithonia
Tithonia +
PK
The Way Forward
The challenge now is to revise the decision aids in response to on-farm
adaptive research and feedback from the farmer evaluations of the
technologies and this will form the major research work for the next
couple of seasons. One of the key success factors for this will be to translate
these decision aids and information into formats that can be used by
extension agents and understood by farmers. An example, is where Figure
8.1 has been translated into simple assessments of resource quality to
be used in Farmer Field Schools and farmer training (Figure 8.8).
Figures 8.8: Farmer decision guide for selecting organic resource management options
136
Delve, R.J.
Other areas of on-going research to ensure effective research to
extension linkages are:
(a) Mobilisation and training of more farmers and extension staff
(b) Mobilising resources to synergise partnership, farmer to farmer
exchange visits from neighbouring districts and countries and to
keep the secretariat at A2N running
(c) Development and packaging of information to aid dissemination for
farmers, NGOs, extension agents and policy makers
(d) Enhancing the ability of NARS scientists to conduct relevant and
participatory on-farm research
(e) Strengthening research to extension linkages
References
Constantinides, M. and Fownes, J.H. (1994) Nitrogen mineralization from leaves
and litter of tropical plants: Relationship to nitrogen, lignin and soluble
polyphenol concentrations. Soil Biol. Biochem. 26, 49-55.
Giller, K.E., Cadisch, G., Ehaliotis, C., Adams, E., Sakala, W.D. and Mafongoya,
P.L. (1997) Building soil nitrogen capital in Africa. In: Buresh, R.J., et al.
(Eds.) Replenishing soil fertility in Africa. SSSA Spec. Publ. 51. SSA, Madison,
WI, USA, pp. 151-192.
Heal, O.W., Anderson, J.M. and Swift, M.J. (1997) Plant litter quality and
decomposition: An historical overview. In: Cadisch, G., Giller, K.E., (Eds.),
Driven by Nature: Plant liter quality and decomposition. CAB International,
Wallingford, UK. pp 3-30
Palm, C.A., Myers, R.J.K. and Nandwa, S.M. (1997) Combined use of organic
and inorganic nutrient sources for soil fertility maintenance and
replenishment. In: In: Buresh, R.J., Sanchez, P.A., Calhoun, F. (Eds.),
Replenishing soil fertility in Africa, SSSA, American Society of Agronomy,
Madison, Wisconsin, USA, pp 193-217.
Effects of Farmyard Manure, Potassium and their Combinations on Maize Yields in the
High and Medium rainfall Areas of Kenya
Effects of Farmyard Manure,
Potassium and their
Combinations on Maize
Yields in the High and
Medium rainfall Areas of
Kenya
137
9
Gikonyo, E.W.1 and Smithson, P.C.2
Kenya Agricultural Research Institute, P.O. Box 14733,
Nairobi, Kenya, E-mail: est.gikonyo@cgiar.org, Fax: 444144
2
International Centre for Research in Agroforestry, P.O. Box
30677, Nairobi, Kenya, E-mail: psmithson@cgiar.org,
1
Abstract
A national Fertilizer Use Recommendation Project (FURP),
conducted trials in 71 sites in 32 districts of Kenya to
monitor the effects of potassium (K) and farmyard manure
(fym) on maize yields. The trials were in 36 sites in different
agro-ecological zones with widely varying soil characteristics.
Out of 36 sites, 8 responded to K, with 4 sites responding
positively and an equal number negatively. Maize yield
increases due to K ranged between 0.6 to 1.0 t ha -1.
Farmyard manure significantly increased yields in 13 sites
by
about
0.46
to
Gikonyo, E.W. and Smithson, P.C.
138
1.3 t ha-1 (P= 0.01 – 4.58). The yields were significantly
improved by a combination of fym and K in only 4 sites
(P=2.5 - 4.95). Response to inorganic K was in most cases
reduced by manure application probably due to excessive
K resulting from the additional K supplied from the manure.
Results from one site that received linear K levels from 0 to
75 kg K2O ha-1 indicated that K response to 50 kg K2O ha-1
was almost equaled by 5t ha -1 FYM. This suggests a
supplemental effect of K by manure, hence no yield benefits
of applying K fertilizers and manure in combination.
Introduction
In Kenya, like many other sub-Saharan African countries, soil fertility
and hence productivity is declining at an alarming rate because areas
of high agricultural potential are densely populated and in most cases
farm holdings are less than 1 ha. Farmers practice intensive continuous
cropping with limited or no replenishment of nutrients through fertilizer
application due to the high cost of the inorganic fertilizers. This has
resulted in high nutrient depletions. For example, in Kisii, western Kenya,
an annual depletion of –112 kg N ha-1, -3 kg P ha-1 and -70 kg K ha-1,
was reported (Smaling, 1993).
Although K is depleted at such high rates, it has long been assumed
by most Kenyans that potassium is non-limited in most Kenyan soils.
However, results from a long-term cropping experiment at Kabete
indicated a decline in soil K with time, but crop residue return or manure
application helped to decelerate the depletion (Kanyanjua et. al., 1999;
Kapkiyai et al., 1999). Besides, limited amounts of K containing fertilizers
have been used in the country. Potassium fertilizers constitute only 7%
of NPK fertilizers imported into the country (Ministry of Agriculture
Annual Reports). This may result in K mining in soils, leading to K
deficiencies, particularly in food crops that rarely receive K fertilizers.
Low K levels and in some cases K deficiency symptoms were observed
in Vihiga, Kakamega and Bungoma districts of Western Kenya (Gikonyo
et al., 1998, unpublished data; Gikonyo et al., 2000; Kanyanjua
unpublished data; Sambili, personal communication). Potassium
deficiency symptoms were also extensively observed in maize (Zea mays
L.) in Bungoma. Application of K fertilizers in maize resulted in good
response to K (Sambili, 1998, personal communication). Potassium
fertilization is therefore emerging as an important crop production
constraint in Kenyan agriculture.
Effects of Farmyard Manure, Potassium and their Combinations on Maize Yields in the
High and Medium rainfall Areas of Kenya
139
Due to economic hardships, the small-scale farmers cannot afford
the potash fertilizers making it important to conduct research on organic
resource utilization as an alternative or complementary source of plant
potassium. It has been adequately demonstrated that application of
organic manures can improve crop yields and soil properties (Probert et
al., 1995; Kihanda, 1996; Nandwa et al., 2000). However, most manure
experiments have been conducted in just a few sites. Manures contain
about 2.6 cmol. kg-1 of potassium and could therefore supplement K in
K deficient soils.
Research on manure is by no means exhaustive. Manure has been
shown to decelerate K depletion but the interaction between manure
and potassium fertilizers has not been addressed. This is becoming
important due to the fact that soils are becoming more K deficient
and at the same time, manure use is increasing. This paper discusses
the effects of farmyard manure and potash fertilizers and their
interactions on maize performance in different soils and agroecological zones.
Materials and Methods
Data used in this study were generated from the Kenya-wide Fertilizer
Use Recommendation Project (FURP), conducted from 1987 to 1993.
The project conducted trials in 71 sites in 31 districts of Kenya. Most of
these trials were conducted on farmers’ fields and represented different
agro-ecological zones and soil types. Detailed descriptions of the sites
were presented in 30 district volumes (FURP, 1987).
At each site three crop sequences were tested. The sequences were
designated as modules 1, 2 and 3. Module 1 involved growing maize in
monoculture continuously, season after season on the same plot. An
intercrop of maize with beans (Phaseolus vulgaris L.), cowpeas (Vigna
unguiculata (L.) Walp) or pigeonpea (Cajanus cajan (L.) Millsp.), formed
module 2. Area specific crops such as potatoes (Solanum tuberosum L.),
cabbages (Brassica oleracea L.), sorghum (Sorghum bicolor L.) and millet
(Pennisetum americanum (L.) Leeke), were tested in seasonal rotations
in module 3.
Two experiments, designated as trials 1 and 2, were carried out
concurrently at each site. Trial 1 was a 4-level N by P factorial experiment
laid out in a randomized complete block design with two replications
and will not further be discussed in this paper. Trial 2 investigated the
effects of presence/absence of N, P, K, S, lime and farmyard manure
(FYM) in various combinations in different sites. A total of 36 sites
included K and manure in the experimental design.
Gikonyo, E.W. and Smithson, P.C.
140
Table 9.1: Selected sites and soil data from Kenya FURP trials
Site
District
Soil Classification
(According to FAO)
Otamba
Kiamokama
Kisii NARS
Ukwala
Siaya Obambo
Yala Swamp
Bukiri Buburi
Alupe ARS
Kamakoiwa
Tongaren
Mumias
Kakamega ARS
Vihiga Maragoli
Mwihila
Baraton
Chepkumia
Sosiot
Eldoret Moi TTC
Turbo
Kapenguria
Charagita
Tulaga
Kandara Kareti
Makuyu
Chehe
Kerugoya
Kavutiri
Gachoka
Embu NARS
Kaguru FTC
Tunyai
Mitunguu
Mpeketoni
Mtondia Tezo
Weruga
Kichakasimba
Kisii
Kisii
Kisii
Siaya
Siaya
Siaya
Busia
Busia
Bungoma
Bungoma
Kakamega
Kakamega
Kakamega
Kakamega
Nandi
Nandi
Kericho
Uasin Gishu
Uasin Gishu
West Pokot
Nyandarua
Nyandarua
Muranga
Muranga
Nyeri
Kirinyaga
Embu
Embu
Embu
Meru
Meru
Meru
Lamu
Kilifi
Taita-Taveta
Kwale
Mollic Nitisol
Humic Nitisol
Mollic Nitisol
Orthic Acrisol
Chromic Luvisol
Humic Gleysol
Ferralo-chromic Acrisol
Ferralo-orthic Acrisol
Rhodic Ferralsol
Ferralo-chromic Acrisol
Ferralo-orthic Acrisol
Dystro-mollic Nitisol
Nito-humic Ferralsol
Dystric Nitisol
Humic Nitisol
Humic Acrisol
Dystro-mollic Nitisol
Ferralic Cambisol
Ferralo-chromic Acrisol
Humic Cambisol
Nito-chromic Luvisol
Eutric Planosol
Humic Nitisol
Dystric Nitisol
Ando-humic Nitisol
Humic Nitisol
Ando-humic Nitisol
Rhodic Ferralsol
Humic Nitisol
Humic Nitisol
Nito-rhodic Ferralsol
Nito-rhodic Ferralsol
Chromic Luvisol
Chromic Luvisol
Chromic Acrisol
Humic Nitisol
pH Modified
Organic
(H2O) Olsen K
Carbon
(cmolc kg-1) (WalkleyBlack)
(%)
5.8
5.3
5.4
5.0
6.3
4.8
5.6
5.4
4.9
4.9
4.8
5.2
5.3
5.3
5.3
5.6
5.6
5.3
5.2
5.9
5.0
5.4
5.8
5.6
4.6
5.6
4.0
5.8
5.6
5.6
5.9
5.5
7.0
7.7
5.4
6.4
1.21
0.46
0.50
0.16
0.86
0.66
0.62
0.47
0.19
0.41
0.12
0.42
0.19
0.20
0.21
0.82
0.12
1.15
0.48
1.85
0.64
0.25
0.24
0.72
0.73
0.07
0.20
0.92
0.97
1.07
0.52
0.65
0.34
0.49
0.23
0.20
2.7
1.8
1.8
0.6
1.3
1.7
1.3
1.4
1.6
1.0
2.1
2.5
1.9
1.7
3.1
3.6
3.3
1.4
1.3
2.7
3.0
2.2
2.2
2.0
3.1
1.4
3.0
1.7
2.2
1.2
1.1
1.5
0.7
0.7
1.6
0.5
Effects of Farmyard Manure, Potassium and their Combinations on Maize Yields in the
High and Medium rainfall Areas of Kenya
141
Nitrogen and potassium were applied at 0 and 50 kg N and K2O
ha-1 in all sites except Kerugoya, where K levels of 0, 25, 50 and 75K2O
kg ha-1 were applied. Farmyard manure was applied at two levels of 0
and 5 t ha-1 in all the sites, except Embu Regional Research Center
where manure was applied at linear rates of 0, 2.5, 5.0 and 7.5 t ha-1.
The experiment was laid out in a randomized complete block design.
The plot sizes were 6 m x 6 m and maize was planted at 0.75 m x 0.60
m. The ultimate aim was to have two plants after thinning at
knee-height to have a population of 44,000 plants ha-1. Harvesting
was done from an area of 21.6 m2 and was used to compute the yields
in kg ha-1.
Maize varieties grown differed from one area to the other depending
on agro-climatic conditions. The “late maturing” Kitale Hybrids were
planted in the highlands and humid midlands at altitudes about 1500
– 2300 m. These in descending order were H614, H 622, and H625.
The “medium maturing” Embu Hybrids were grown in the sub-humid
midlands with lengths of growing periods between 130 - 160 days at
1200 m–1800 m (i.e. H512 and H511). The “early maturing”
composites (Katumani composite B (KCB) and Makueni Composite
(MC), were grown at the semi-arid Midlands to Lowlands (0-1600 m).
Makueni replaced Katumani composite if the length of growing period
was less than 85 days. The “early to medium maturing lowland
maize”(Coast composite (CC) and Pwani hybrid), were grown at the
coastal lowlands. Experimental details are available in the Final FURP
Methodology (FURP, 1988). The experiment was conducted from 1987
to 1992, though most sites were not studied for all years of the FURP
study.
Statistical analysis was done by analysis of variance using the
Statistical Analysis System (SAS Institute, 1990). Reference to statistical
significance refers to a probability level of 0.05 unless otherwise noted.
Results
In majority of the sites, 29 out of the 36 trial sites maize responded
positively to NP application. Maize responded positively to manure
application in 14 sites while response to K was recorded in only 8 out of
36 sites (Table 9.2). Maize responses to manure, K fertilizers and K
fertilizer and manure combinations, varied from site to site but a few
generalized categories are presented (Figures 9.1a-1f). In some, there
were no significant responses to K (1d and 1e) while significant responses
to K were observed in others. Some responded to K but not N or NP (1a
and 1b) while others responded to K, manure and NP combinations.
142
Gikonyo, E.W. and Smithson, P.C.
Table 9.2: Sites and respective maize response to Farmyard manure (FYM), Potash
fertilizer (K) and N/NP applications
Maize Responses (t ha-1) to
Site
Kisii NARS
Ukwala
Bukiri Buburi
Kamakoiwa
Tongaren
Mumias
Kakamega ARS
Vihiga Maragoli
Chepkumia
Sosiot
Turbo
Kapenguria
Charagita
Tulaga
Makuyu
Chehe
Kerugoya
Kavutiri
Gachoka
Kaguru FTC
Tunyai
Mpeketoni
Mtondia Teso
Weruga
FYM
Potassium (K)
Nitrogen X
Phosphorus
Ns
Ns
Ns
0.94
1.07
0.51
1.3
0.7
Ns
1.0
0.58
Ns
Ns
Ns
1.3
0.83
Ns
0.46
0.54
Ns
0.47
0.56
0.7
Ns
-1.02
ns
ns
ns
ns
0.45
ns
ns
ns
ns
ns
-0.66
ns
0.4
-0.33
ns
+vea
0.61
ns
-0.44
ns
ns
ns
ns
1.2
1.2
0.9
2.9
1.9
1.5
0.78
1.93
1.67
ns
2.2
ns
1.21
ns
1.15
0.28
1.4
1.1
0.88
1.64
1.06
0.56
1.7
0.7
Manure increased maize grain yields in all the manure responsive
sites. Manure increased yields by 0.46 - 1.3 t ha-1 compared to the control
plots. The highest yield increases were from Kakamega ARS and Makuyu
site, while the lowest were from Kavutiri site. It was noted that all the
manure responsive sites with an exception of Sosiot, were NP responsive.
However, out of 10 manure non-responsive sites, only Kapenguria and
Tulaga were not responsive to NP. Maize yield increases due to NP
application was much higher (0.56 – 2.9 t ha-1) than that resulting from
manure application. In Embu, a site with linear manure applications
(Figure 9.2), it was observed that yields increased linearly with manure
applications from 3.2 t ha-1 at zero manure rate to 4.5 t ha-1 at manure
rate of 7.5 t ha-1. When NP was applied, the yields were shifted at each
level of manure application by about 2t ha-1. Potash fertilizer increased
yields particularly at zero manure rate, but yields decreased
Effects of Farmyard Manure, Potassium and their Combinations on Maize Yields in the
High and Medium rainfall Areas of Kenya
143
(c)
Maize yields (t ha-1)
(b)
Maize yields (t ha-1)
(a)
Maize yields (t ha-1)
Figure 9.1 (a - c): Typical patterns of responses to various fertilizer additions at selected
sites in multi-locational fertilizer trials in Kenya
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
Gikonyo, E.W. and Smithson, P.C.
144
(e)
Maize yields (t ha-1)
(f)
Maize yields (t ha-1)
(d)
Maize yields (t ha-1)
Figure 9.1 (d - f): Typical patterns of responses to various fertilizer additions at selected
sites in multi-locational fertilizer trials in Kenya
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
123456
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
12345678
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
1234567
Effects of Farmyard Manure, Potassium and their Combinations on Maize Yields in the
High and Medium rainfall Areas of Kenya
145
with increasing manure rates in combination. With NP application there
was a large yield increase that was further improved by Potash fertilizer
application alone, but addition of manure to that combination lowered
the yields. This implies that in both cases, i.e. Potash fertilizer alone or
in combination with NP fertilizer, inclusion of manure to K had negative
effects on maize yields. Manure and Potash fertilizer combinations
resulted to no increases or depressed yields in all the sites tested (Figure
9.1a-f).
Maize responded significantly to potash fertilizer application in only
8 sites. However, both negative and positive responses to K were observed,
with an equal number of sites showing positive and negative responses
(Table 9.2). Four out of thirty five sites (11%) responded positively while
four responded negatively. Maize yields increased by up to 0.66t ha-1,
but apparently real decreases of up to 1 t ha-1 were also recorded. Positive
K responses were observed in Mumias, Tulaga, Kerugoya and Kavutiri
sites all with extractable modified Olsen K levels of 0.20 cmol kg-1. The
four sites showing negative responses were Kisii NARS, Kapenguria,
Makuyu and Kaguru FTC and they all had Olsen K levels of 0.5 cmolc
kg-1 soil. Negative K responses were also reported in maize on granitic
and phonolithic soils that were slightly acidic to acidic (Weiss, 1973).
In Kerugoya, an extremely K-deficient site at which linear levels of K
were applied in the presence and absence of NP and FYM, response to K
alone was quadratic with the maximum yields occurring at K rates
around 50 kg K2O ha-1 (Figure 9.3).
Figure 9.2: Effect on maize yields of linear rates of manure or manure + NP fertilizer,
plus or minus K at 50 kg K2O ha-1, at Embu site
6
manure (fym)
fym + K
fym + NP
fym + NP + K
Yield (t ha -1)
5
4
3
0
2
4
Manure (t ha -1)
6
8
Gikonyo, E.W. and Smithson, P.C.
146
Figure 9.3: Effect on maize yields of linear rates of K fertilizer (0 to 75 kg K2O ha-1), plus
or minus manure at 5 t ha-1 at Kerugoya site
Maize gain yield (t ha-1)
4
3
2
1
NP0, FYM0
NP0, FYM5
NP50, FYM0
NP50, FYM5
0
0
50
25
75
-1
K rate (kg K 2 0 ha )
In the same site, K response was greatly reduced to almost 0 by
application of 5t ha-1 FYM suggesting that manure supplied K to the
maize. Manure alone gave slightly higher yields than the potash fertilizer.
Addition of NP fertilizer together with potash fertilizer increased yields
by more than double (Figure 9.3). Such a synergistic effect was also
observed in legumes in Western Kenya where beans, groundnuts and
soybeans, gave greater responses when K was applied in the presence
of P (Qureshi, 1979).
Discussion
Most Kenya soils are N and P deficient, which, probably explains why
most sites responded to NP application. The benefits of manure as a
source of soil nutrient is well documented.
Besides addition of nutrients, manure increases the water holding
capacity, PH and infiltration of water and decreases bulk density of the
soil (Azevedo and Stout, 1997). Maize response to manure was probably
an indirect response to NP and sometimes K in K deficient soils and this
is probably why most of the NP responsive sites also responded to manure
application. However in a number of sites maize did not respond to
manure but responded to inorganic NP application. This is probably
due to the variable quality of manure used in terms of nutrients content
(Wanjekeche et al., 1999; Micheni et al., 2000; Lekasi et al., 1997).
Effects of Farmyard Manure, Potassium and their Combinations on Maize Yields in the
High and Medium rainfall Areas of Kenya
147
Higher maize yield increases were obtained with NP as compared to
manure probably due to low nutrients in low quality manure as shown
in SMP (1996, 1997) that supplementing some fertilizers for some manure
produced a response in cases where manure alone had no response. In
Embu site, addition of NP to manure increased yields by twofold, maybe
due to the improved nutrient status with addition of NP fertilizers. Some
soils were non-responsive to manure but they were responsive to N or
NP (Table 9.2). This is probably due to the low nutrient content of some
of the manures used (Micheni et al., 2000), hence, manure quality should
therefore be considered in manure recommendations.
Response to K was observed in only a few sites since as indicated by
soils analysis, most sites had high K levels. They were above the K
critical level for most crops (> 0.2 cmol kg-1) and therefore, response to
K was unlikely (Gikonyo et al., 2000). The negative responses at high
soil K may be associated with Mg to K ratio imbalance. Considerable
evidence indicates that heavy applications of potash fertilizer or high
level of K in the soil can lead to low-magnesium content in the plant.
Doll and Hossner (1964), reported that fertilizer K decreased potato
(Solanum tuberosum L.) yields at every level of magnesium fertilization.
Differential absorption of K was the controlling factor in the uptake of
magnesium in corn seedlings (Stout and Baker, 1981). Mg deficiency
was observed in plants growing in soils well supplied with Mg as a result
of K-induced Mg deficiency (Messing, 1974).
Lack of response or negative response resulting from manure and
potash fertilizer application is probably due to excessive K from the two
sources. Manure at 5t ha -1 seems to supply the maize crop’s K
requirement and the application of the two sources not beneficial in all
cases.
Acknowlegement
The authors wish to express their gratitude to; the German Agency for
Technical Cooperation (GTZ), for providing funds for the FURP studies
(1987-1993); staff members of the Kenya Agricultural Research Institute
(KARI), for collecting and compiling the original field and laboratory
data; P. Angala of ICRAF for providing computer training and logistical
support during the attachment of the first author; the European Unionfunded Soil Fertility and Plant Nutrition Research Programme, KARI for
providing funds to allow the first author’s attachment during which
data for this work was analysed; TSBF/AfNet for sponsoring my
attendance to this conference; Last and not least God almighty for his
continued grace and everlasting love.
148
Gikonyo, E.W. and Smithson, P.C.
References
Azevedo, J. and Stout, P.R. (1997) Farm animal manures. An overview of their
role in the agricultural environment. Califonia Agricultural Experiment station,
Extension service.
Doll, E.C., and Hossner. (1964) Magnesium deficiency as as related to liming
and potassium levels in acid sandy podzols. Int.Congr. Soil Sci., Trans
8th,1964. 4: 907 – 912.
FURP. (1987) The Fertilizer Use Recommendation Project, Final Report Annex1:
Fertilizer Trial Documentation (FERDOC). Min. of Agriculture, Nairobi,
Kenya.
FURP. (1988) The fertilizer Use Recommendation Project (Phase 1). Main Repoort,
Methodology and inventory of existing Information. Fertilizer Trial Documentation
(FERDOC). Min. of Agriculture, Nairobi, Kenya.
Gikonyo, E.W., Smithson, P.C., Nandwa, S.M., Lijzenga, M. and B. Janssen.
(2000) Effects of potassium fertilizer on some major food crops in Kenya. A
paper presented at the 18th AGM of Soil Science Society of East Africa
meeting held in December 2000 at Mombasa, Kenya.
Kanyanjua, S.M., Buresh, R.J., Recke, H. and Nandwa, S.M. (1999) Potassium
status in Kenyan soils and its influence on crop yields: A study to challenge
the convention that soils in Kenya have adequate potassium to sustain
productivity of major crops. A poster paper presented at the conference on
Managing Plant Nutrition: Towards Maximum Resource Efficiency,
Barcelona, Spain, 29th July 1999.
Kapkiyai, J. J., Karanja, N.K., Qureshi, J.N., Smithson, P.C. and Woomer, P.L.
(1999) Soil organic matter and nutrient dynamics in a Kenyan Nitisol under
long-term fertilizer and organic input management. Soil Biol. Biochem. 31,
1773-1782.
Kihanda, F.M. (1996) The role of farmyard manure in improving maize production
in sub-humid highlands of central Kenya. Ph. D. Thesis. University of
Reading. UK.
Lekasi, J. C., Tanner, J.C., Kimani, S.K. and harris, P.J.C. (1998) Manure
management in the Kenya highlands: Practices and potential. ADRA
Publications. Pp. 35.
Messing, J.H.L. (1974) Long term changes in potassium and calcium content of
banana plants and soils in the Windward Islands. Trop. Agric. (Trinidad)
51: 154-160.
Micheni, A. Mugendi, D., Mugwe, J., kung’u, J. and Gitari, J. (2000) Low cost
soil fertility amendment strategies for improved food production in maize
cropping systems in Central highlands of Kenya. Paper presented in the
SSSEA, Dec. 2000, Mombasa, Kenya.
Ministry of Agriculture. Annual reports of fertilizer imports to Kenya. Kenya
Ministry of Agriculture, Nairobi.
Nandwa, S.M., Thuranira, E.G., Wamaitha, J., Gikonyo, E., Kibunja, C.N. and
Onyango, J. W. (2000) Changes on soil fertility and crop yields in a
continuous maize-based cropping system with low-and-high external
Effects of Farmyard Manure, Potassium and their Combinations on Maize Yields in the
High and Medium rainfall Areas of Kenya
149
nutrient inputs at semi humid site, Kabete (Nairobi) in Kenya. Paper
presented at the International Symposium on balanced Nutrient
Management Systems for moist savana and Humid Fores Zones of Africa,
Cotonou, Republic of Benin, Oct. 2000.
Probert, M. Okalebo, E., Simpson, J.R. (1995) The use of manure on smallholders
in semi-arid eastern Kenya. Experimental Agriculture 31: 371-381.
Qureshi, J.N. (1979) Potassium workshop organized by the International Potash
Institute in Nairobi, Kenya.
SAS Institute (1990) SAS/STAT User’s Guide, Vol. 2. Version 6.1.
SAS Institute, Cary, NC, USA
Smaling, E.M.A. (1993) Soil nutrient depletion in sub-Saharan Africa. In: The
role of plant nutrients for sustainable food production in sub-Saharan Africa
pp. 53-67. In Van Reuler, H. and Prins W.H. (eds). Leidschendam: VKP, The
Netherlands.
Stout, W.L and Baker, D.I. (1981) Effect of differencial absorption of potassium
and magnesium in soils on magnesium uptake by corn. Soil Sci. Soc. Am. J.
45: 996 – 997.
SMP 1996, (1997) Soil Management Project Annual Report. Wanjekeche, E.,
Mwangi, T., Powon, P. and Khaemba, J. (1999) Survey of management
practices of Boma manure and its nutrient in West Pokot district in Kenya.
SSSEA, 17th conference, pp. 335-359.
150
Gikonyo, E.W. and Smithson, P.C.
Effects of Nitrogen and Phosphorus Fertilizer Addition on Wheat Straw Carbon
Decomposition in a Burundi Acidic Soil
Effects of Nitrogen and
Phosphorus Fertilizer
Addition on Wheat Straw
Carbon Decomposition in a
Burundi Acidic Soil
151
10
Kaboneka, S.1,2, Nivyiza, J.C.2 and
Sibomana, L.2
1
FAO Burundi. B.P. 1250 Bujumbura, Burundi. Fax: 257-227547. E-mail : fao-urg-bdi@usan-bu.net
2
Université du Burundi. Faculté des Sciences Agronomiques.
B.P. 2940. Bujumbura, Burundi. Fax: 257-22-2500. E-mail:
facagro@cni.cbinf.com
Abstract
Two laboratory incubation studies were conducted for 56
days in an acidic high altitude Burundi soil, to evaluate the
effect of increasing application rates of N and P fertilizers
on carbon (C) decomposition from wheat straw (Triticum
aestivum L.). Nitrogen was applied as NH4NO3 at 0, 40, 80
and 120 kg N ha-1, while P was applied at 0, 17.6, 35.2 and
52.8 kg P ha-1 as K 2HPO4. Carbon dioxide (CO2) evolution
was regularly monitored using the alkali absorption method.
At the completion of the studies, % C decomposition from
wheat straw was equal to 18.8, 34.1, 45.4 and 48.1 % for
straw + 0, straw + 40, straw + 80 and straw + 120 kg N ha-1,
respectively. Comparatively, similar values were 21.4,
152
Kaboneka, S. et al
40.4, 51 and 58.6 % for straw + 0, straw + 17.6, straw +
35.2 and straw + 52.8 kg P ha -1, respectively. Straw C
decomposition kinetics was described by a simple
exponential model and decomposition rate constants (k),
half-lives (t 1/2), together with the time periods required for
90 and 99 % straw C mineralization were evaluated. The
data indicated that % straw C mineralization increased
with increasing application rates of N and P fertilizers.
The highest decomposition rates were obtained with P
fertilizers. The investigations illustrated the benefit of the
combined use of low-quality organic materials and
inorganic nutrient sources in enhancing decomposition
and implicitly increasing total available nutrients for plant
uptake.
Introduction
Carbon (C) decomposition from organic residues is controlled by as
many factors as soil environmental conditions (temperature,
moisture, aeration, soil pH, nutrient availability, etc), substrate
quality (chemical composition) and quantity, soil residue pretreatment, application methods and their potential interactions
(Marion and Black, 1987).
Organic materials are used in many conflicting ways in developing
countries in general and in Burundi in particular. They can serve as
cooking fuel, livestock feed, building materials, animal litter, substrate
for edible mushroom production or mulch, when they are not used in
soil fertility replenishment either by direct application or through
composting. Subsistence agriculture in Burundi is very much dependent
on organic materials (which include crop residues, animal manure and
agroforestry species biomass) to replenish soil organic matter and supply
nutrients.
Fertilization could offset the negative effects of low-quality organic
materials (Palm et al., 1997) and accelerate their decomposition, thereby
releasing over a relatively short time nutrients, which normally would
be cycled over a more extended period (Kelly and Henderson, 1978).
These released nutrients added to those contributed by fertilizer
applications, would increase total available nutrients for plant uptake
(Palm et al., 1997).
The objective of the present laboratory studies was to evaluate the
short-term effects of N and P mineral fertilizer addition on the
decomposition of wheat straw incorporated in a Burundi acidic soil.
Effects of Nitrogen and Phosphorus Fertilizer Addition on Wheat Straw Carbon
Decomposition in a Burundi Acidic Soil
153
Materials and Methods
Soil
The soil used in the studies was collected in April 1999 from the Ruzibazi
Seed Center in Mukike District, Rural Bujumbura Province. Selected
physical and chemical properties of the soil are given in Table 10.1.
Niyongabo (1986) described a similar soil.
Table 10.1: Physical and chemical properties of the soil used in the study
Parameter
Value
% clay
% sand
pHwater
pHKCl
Electrical conductivity (dS/m)
%C
%N
C/N
CEC (cmolc kg-1 soil)
Al3+ (cmolc kg-1 soil)
Exchangeable acidity (cmol c kg-1 soil)
% Al saturation
P-Olsen (mg kg -1 soil)
64.9
24.5
4.7
4.0
0.0845
5.65
0.61
9.18
33.38
4.91
5.50
4.71
Wheat straw
Wheat straw was collected after crop harvest. The material was dried at
70° C to a constant weight and subsamples were ground in a Wiley mill
before chemical analyses were performed. Total N was determined by a
Leco N analyser model FP 428 (Leco Corporation, St Joseph, MI). Total
C was determined by dry combustion (Nelson and Sommers, 1982).
Total P, S, K, Ca and Mg were analyzed by ICP spectrometry after
digestion of a 0.2-g sample with HNO 3 and H2O2 at 120° C for 3 hours
(Zarcinas et al., 1987). Selected properties of the wheat straw used in
the study are shown in Table 10.2.
Incubation Procedure
Each incubation vessel (250-mL) was fitted with 2 test tubes each
containing 5 mL of 2 N NaOH to capture evolved CO2.
The laboratory incubations were conducted at room temperature
(25 ± 1° C) in the soil laboratory facilities of the Faculty of Agricultural
154
Kaboneka, S. et al
Sciences, University of Burundi. Soil samples (50 g d.w. basis) were mixed
with wheat straw to approximate a field application of 3000 kg ha -1,
corresponding to wheat straw production in the wheat producing region
of Mugamba, where wheat straw is used as animal litter or composting
materials (Niyongabo, 1986). Plant materials were cut into approximately
0.5 cm-long sections and were incorporated in soil in combination with
N and P fertilizers. In the first study, N was applied as NH4NO3 at 0, 40,
80 and 120 kg N ha -1. In the second study, P was applied at 0, 17.6,
35.2 and 52.8 kg P ha -1 as K 2HPO 4, which simultaneously brought 0,
44.3, 88.6 and 132.9 kg K ha-1, respectively. In the latter investigation,
fertilizer K was adjusted in all treatments using KCl in order to nullify
the effect of K on straw decomposition.
Table 10.2: Chemical Composition of the wheat straw used in the study
Parameter
g kg -1
C
N
P
S
K
Ca
Mg
C/N
C/P
C/S
N/P
N/S
420
5.5
0.4
0.9
10.4
2.9
0.6
76.4
1050
466.7
13.8
6.1
CO2 sampling was performed at 3, 7, 14, 21, 28, 42 and 56 days of
incubation. All incubation vessels were opened and aerated for about 5
minutes at each sampling period to maintain aerobic conditions, while
test tubes containing the alkali solution were simultaneously changed
and titrated (Stotzky, 1965; Zibilske, 1994).
Control soils without straw and fertilizer were run, and empty
incubation vessels were used as controls for CO 2 absorbed from the
atmosphere during the incubation procedure. Soil moisture was adjusted
to 60 % water holding capacity (WHC). The total quantity of CO2 collected
in the dilute NaOH solution was determined by titration to a
phenolphthalein indicator endpoint with standardized HCl following
addition of BaCl2, according to the following reactions (Stevenson, 1986):
2NaOH + CO2
Na2CO3 + BaCl2
NaOH (exc.) + HCl
-
Na2CO3 + H2O
BaCO3 (_) + 2 NaCl
NaCl + H2O
(1)
(2)
(3)
Effects of Nitrogen and Phosphorus Fertilizer Addition on Wheat Straw Carbon
Decomposition in a Burundi Acidic Soil
155
Calculations
Carbon evolved as CO2 was estimated by the following formula (Stotzky,
1965):
(mg C as CO2 = ( B – V ) x N x E
(4 )
where;
B = mL of standard acid for the blank;
V = mL of standard acid for amended treatments;
N = normality of standard acid;
E = equivalent weight of C (= 6).
The evolution of wheat organic C evolved as CO 2 from soils
amended with straw was determined by subtracting the quantity of
CO 2–C evolved from control samples from the quantity of CO 2 –C
evolved, from wheat-amended soils. This is the usual method of
determining decomposition of unlabelled substrate in soils. As in
many other studies of this nature, the decomposition rate of native
soil organic matter C in the presence of wheat straw was assumed to
be the same for each treatment, meaning that there was no priming
effect (Ajwa and Tabatabai, 1994).
Percentage decomposition was estimated by calculating the
percentage of C added evolved as CO2 after correction for the CO2 evolved
from unamended soils according to the following equation:
(% C decomposition = [( X – Y) / Z)] x 100
(5)
where
X = mg of C evolved as CO2 from wheat-fertilizer treatments;
Y = mg of C evolved as CO2 from unamended soil (control);
Z = mg of C added in the wheat straw.
Decomposition model
Numerous mathematical models have been tested to describe C
mineralization from soil organic matter or plant materials. Most models
follow the first-order kinetics, for which the magnitude of decomposition
is assumed proportional to the quantity of mineralizable C. The simple
or one-component exponential model is the oldest (Stanford and Smith,
1972). The model assumes that only one form of potentially
mineralizable C exists and mineralizes at a rate proportional to its
concentration.
156
Kaboneka, S. et al
Statistical analyses
The incubation studies were conducted in a completely randomized
design (CRD). Experimental treatments consisted of a blank, a control
(soil only) and wheat-amended treatments. In each one of the two studies,
each treatment was replicated three times to make a total of 18
experimental units. Straw C decomposition can be described by the
following equation:
dt = -k C
Ct = Co exp (-kt)
(6)
(7)
where
Ct = carbon content at time t (day),
Co = initial C content,
k = first-order rate constant,
t = time, days.
Decomposition rate constants (k) were estimated by using the Linear
Least Squares (LLS) procedure by plotting the natural logarithm of % C
remaining versus time of incubation in days. The software used to
evaluate the fitness of different models of C decomposition kinetics from
wheat straw was version 3.2 of SAS JMP IN (SAS, 1996). Mean separation
of % C decomposition was performed with the Newman and Keuls test.
The 0.05 level of probability was used as the criterion for accepting or
rejecting null hypotheses in all statistical analyses.
Using decay coefficient (k) values and assuming constant decay rates
for specific treatments, half-lives (t 1/2) together with the time periods
required for 90 and 99 % straw C mineralization were estimated
according to equations 8, 9 and 10, respectively.
t
t
t
= ln 2 / k = 0.693 / k
= ln 10 / k = 2.303 / k
0.9
= ln 100 / k = 4.605 / k
0.99
0.5
(8)
(9)
(10)
Results and Discussion
Data were fitted to the simple (one-component) exponential model and
decomposition rate constants, t 0.5, t 0.9 and t 0.99 were estimated. For
convenience, the results obtained from the two laboratory studies are
discussed separately.
Effects of Nitrogen and Phosphorus Fertilizer Addition on Wheat Straw Carbon
Decomposition in a Burundi Acidic Soil
157
Effect of N fertilizer on straw C Mineralization
The addition of N fertilizer significantly increased wheat straw C
decomposition. At the completion of the study, % straw C decomposition
as affected by N fertilizer ranged from 18.8 with no N addition to 48.1 %
with 120 kg N ha -1 as shown in Table 10.3. No significant differences
were observed between treatments fertilized with 80 and 120 kg N ha-1.
Overall, % straw C decomposition followed the order: Straw + 120 kg N
ha-1 = Straw + 80 kg N ha-1 > Straw + 40 kg N ha-1 > Straw alone.
Table 10.3: Percent wheat straw decomposition as affected by N fertilizer rates
Treatment
Straw + 120 kg N ha -1
Straw + 80 kg N ha -1
Straw + 40 kg N ha -1
Straw + 0 kg N ha-1
% Decomposition
% Increase
48.10 ± 1.53a*
45.40 ± 2.28a
34.11 ± 1.47b
18.80 ± 2.41c
+ 155.9
+ 141.5
+ 81.4
–
*Values followed by the same letter are not significantly different at 5 % probability level.
Table 10.4: Decomposition model and rate constants of wheat straw fertilized with N
Treatment
Decomposition Model
Straw + 0 kg N ha -1
Y = 4.576626 (± 0.009675) –
0.003497 (± 0.000320)
Y = 4.536913 (± 0.001655) –
0.006944 (± 0.005480)
Y = 4.492028 (± 0.002539) –
0.009955 (± 0.008410)
Y = 4.474588 (± 0.002411) –
0.010637 (± 0.007990)
Straw + 40 kg N ha -1
Straw + 80 kg N ha -1
Straw + 120 kg N ha -1
Prob.
R2
Decomposition
Time (days)
t 0.5 t 0.9 t 0.99
< 0.0001 0.86 198 658 1317
< 0.0001 0.89 100 332 663
< 0.0001 0.88
70
231 463
< 0.0001 0.90
65
216 433
Table 10.4 shows that N application hastened straw decomposition.
When compared to straw alone, addition of 40 kg N ha -1 doubled the
decomposition rate, while application of higher N rates (80 and 120 kg
N ha-1) tripled it (Table 10.4).
Effect of P fertilizer on straw C Mineralization
At the completion of the study, % wheat straw C decomposition as
affected by P fertilizer ranged from 21.4 with no P addition to 58.6 %
with 52.8 kg P ha-1) as shown in Table 10.5. Overall, percent straw C
158
Kaboneka, S. et al
decomposition followed the order: Straw + 52.8 kg P ha-1 > Straw + 35.2
kg P ha-1 > Straw + 17.6 kg P ha-1 > Straw alone.
It can be observed that P addition brought about higher percent
straw C decomposition when compared to N fertilizer addition. When
compared to straw alone, addition of 17.6 kg P ha -1 doubled the
decomposition rate, while application of 35.2 and 52.8 kg P ha-1 tripled
and quadrupled it, respectively (Table 10.6).
Table 10.5: Percent wheat straw decomposition as affected by P fertilizer rates
Treatment
Straw + 52.8 kg P ha -1
Straw + 35.2 kg P ha -1
Straw + 17.6 kg P ha-1
Straw + 0 kg P ha-1
% Decomposition
% Increase
58.6 ± 0.2a*
51.0 ± 3.3b
40.4 ± 0.6c
21.4 ± 3.4d
+ 173.8
+ 138.3
+ 88.8
–
Table 10.6: Decomposition model and rate constants of wheat straw fertilized with P
Treatment
Decomposition Model
Y = 4.567626 (± 0.151300) –
0.004290 (± 0.000500)
Straw + 17.6 kg P Y = 4.546650 (± 0.016548) –
0.009040 (± 0.005480)
Straw + 35.2 kg P Y = 4.531140 (± 0.030420) –
0.012942 (± 0.001008)
Straw + 52.8 kg P Y = 4.522903 (± 0.037912) –
0.016430 (± 0.001260)
Prob.
R2
< 0.0001
0.78
162
537 1073
< 0.0001
0.91
77
255
509
< 0.0001
0.89
54
178
356
< 0.0001
0.89
42
140
280
Decomposition
Time (days)
t 0.5 t 0.9 t 0.99
Straw + 0 kg P
Results obtained in this study were in agreement with those found
by Smith and Douglas (1971), who reported greater decomposition of
wheat straw with N than without it. Also, in a study conducted in a
Kenyan ferralsol, Munyampundu e t al . (1997) reported that the
combination of wheat straw and increasing rates of inorganic N and P
enhanced growth and yields of Leucaena leucocephala, Sesbania sesban,
maize grain and stover.
However, conflicting results have also been reported elsewhere. For
example, Kelly and Henderson (1978) found that urea application had
little effect and superphosphate addition depressed the decomposition
of white oak leaves (Quercus alba L.) in litter bags. Also, recent field
experiments conducted in India (Goyal et al., 1992) and in Malawi (Jones
Effects of Nitrogen and Phosphorus Fertilizer Addition on Wheat Straw Carbon
Decomposition in a Burundi Acidic Soil
159
et al., 1997), have shown mixed results with regard to the effect of
combining organic and inorganic inputs on yield and nutrient uptake
(Palm et al., 1997). In a 4-year experiment using pearl millet [Pennisetum
glaucum (L.) R. Br.], Goyal et al. (1992), indicated that inorganic N
addition to wheat straw did not improve yields, N uptake and N recovery
by the test plant. On the contrary, Jones et al. (1997) reported a
significant increase in yields and N-use efficiency when Leucaena
[Leucaena leucocephala (Lam.) de Wit.] leaf residues were combined with
urea in the 1: 3 ratio (Palm et al., 1997).
The study reported in this paper concerned CO2 evolution. One can
argue that this parameter does not have a practical agronomic meaning.
However, it has been proved that C decomposition controls nutrient
release from organic materials. In particular, Clark and Gilmour (1983)
and Castellanos and Pratt (1981) proposed CO2 evolution as a means of
providing a general estimate of net N mineralization. In fact, Gilmour et
al. (1985) and Moorhead et al. (1987) showed significant linear
relationships between N mineralization (g kg-1 N) and carbon dioxide
evolution (g kg-1 C) for a number of organic substrates (Table 10.7).
Table 10.7: Selected linear relationships between carbon evolution as carbon dioxide
(C) and N mineralization (N) for some organic substrates.
Substrate
Sewage sludge
Alfalfa
Clover
Bermudagrass
Ryegrass
Low-N plant biomass
Fresh
Digested
High-N plant biomass
Fresh
Digested
Equation
Correlation coefficient
N = 1.01 C + 48
N = 0.98 C + 79
N = 0.97 C + 84
N = 0.43 C + 396
N = 0.53 C + 291
r = 0.90
r = 0.96
r = 0.995
r = 0.94
r = 0.97
N = 0.15 C - 36
N = 0.32 C + 15
r = 0.73
r = 0.94
N = 0.52 C + 89
N = 0.30 C + 9
r = 0.77
r = 0.95
Source: Gilmour et al., 1985 Moorhead et al., 1987
Conclusion
The two laboratory studies have indicated that N and P fertilizers
significantly increase percent straw C decomposition and decomposition
rate. The response of straw C decomposition to N fertilizer addition
followed the order: Straw + 120 kg N ha-1 = Straw + 80 kg N ha-1 > Straw
+ 40 kg N ha-1 > Straw alone. On the other hand, straw C decomposition
160
Kaboneka, S. et al
as affected by P fertilizer rates followed the order: Straw + 52.8
kg P ha-1 > Straw + 35.2 kg P ha-1 > Straw + 17.6 kg P ha-1 > Straw alone.
Straw C decomposition was higher in P- than N-fertilized treatments,
presumably because P is generally the most limiting nutrient in acid,
weathered soils of the subhumid and humid tropics in general (Buresh
et al., 1997) and in Burundi high altitude soils in particular (Ntiburumusi
et al., 1998).
Although our results could not simultaneously be validated under
field conditions, they are in agreement with field experiments conducted
with similar organic materials and soils in Kenya (Munyampundu
et al., 1997). Both studies illustrate the benefit of the combined use of
low-quality organic materials and inorganic nutrient sources in
enhancing organic material decomposition and increasing total available
nutrients for plant uptake.
References
Ajwa, H.A. and Tabatabai, M. A (1994) Decomposition of different organic
materials in soils. Biol. Fert. Soils 18: 175-182.
Buresh, R.J., Smithson, P.C. and Hellums, D.T. (1997) Building soil phosphorus
capital in Africa. p. 111-149. In. R.J. Buresh et al. (eds.). Replenishing soil
fertility in Africa SSSA Spec. Publ. 51. SSSA, Madison, WI.
Castellanos, J.Z., and Pratt, P.F. (1981) Mineralization of manure nitrogencorrelation with laboratory indexes. Soil Sci. Soc. Am. J. 45: 354-357.
Clark, M.D. and Gilmour, J.T. (1983) The effect of temperature on decomposition
at optimum and saturated water content. Soil Sci. Soc. Am. J. 47: 927-929.
Gilmour, J.T., Clark, M.D. and Sigua, G.C. (1985) Estimating net nitrogen
mineralization from carbon dioxide evolution. Soil Sci. Soc. Am. J. 49: 13981402.
Goyal, S., Mishra, M.M., Hooda, E.S. and Singh, R. (1992) Organic matermicrobial relationships in field experiments under tropical conditions: effects
of inorganic fertilization and organic amendments. Soil Biol. Biochem. 24:
1081-1084.
Jones, R.B., Snapp, S.S. and Phombeya, H.S.K. (1997) Management of
leguminous leaf residues to improve nutrient use efficiency in sub-humid
tropics. p. 239-250. In Cadisch, G. and Giller, K.E. (ed.). Driven by nature:
Plant litter quality and decomposition CAB Int., Wallingford, England.
Kelly, J.M., and Henderson, G.S. (1978) Effects of nitrogen and phosphorus
additions on deciduous liter decomposition. Soil Sci. Soc. Am. J. 42: 972976.
Marion, G.M., and Black, C.H. (1987) The effect of time and temperature on
nitrogen mineralization in Arctic Tundra soils. Soil Sci. Soc. Am. J. 15011508.
Effects of Nitrogen and Phosphorus Fertilizer Addition on Wheat Straw Carbon
Decomposition in a Burundi Acidic Soil
161
Moorhead, K.K., Graetz, D.A. and Reddy, K.R. (1987) Decomposition of fresh
and anaerobically digested plant biomass in soil. J. Environ. Qual. 16: 25-28.
Munyampundu, A., Okalebo, J.R., Othieno, C.O., Wesonga, J. and Njoroge, R.
(1997) Maize, Leucaena leucocephala, Sesbania sesban growth and yield
improvement through combination of crop residues with nitrogen fertilizer in
a Kenyan ferralsol. 13. Paper presented at the 7th TSBF – AfNet Workshop.
10-15 September. Kampala, Uganda.
Nelson, D.W., and Sommers, L.E. (1982) Total carbon, organic carbon, and
organic matter. p.539-579. In A.L. Page et al. (ed.). Methods of soil analysis.
Part 2. 2nd ed. Agronomy 9. ASA, Madison, W.I.
Niyongabo, D. (1986) Etude d’un système cultural: cycle de la biomasse et des
éléments biogènes. Cas de la colline de recensement de Nyarushanga.
Mémoire. Faculté des Sciences Agronomiques. Université du Burundi.
Nivyiza, J.C. (2001) Effet de l’application des engrais phosphatés sur la
décomposition de la paille de blé appliquée dans un sol de haute altitude
du Burundi. Mémoire. Faculté des Sciences Agronomiques. Université du
Burundi.
Ntiburumusi, F., Ntibashirwa, S. et Nzohabonayo, Z. (1998) Guide sur la
fertilisation des cultures au Burundi. Publication ISABU. p. 95.
Olsen, C.R., C.V. Cole, F.S. Watanabe, and L.A. Dean. (1954) Estimation of
available phosphorus in soils by extraction with sodium bicarbonate. USDA
Circ. 939. Washington, DC. USDA.
Palm, C.A., Myers, R.J.K. and Nandwa, S.M. (1997) Combined use of organic
and inorganic nutrient sources for soil fertility maintenance and
replenishment. p. 193-217. In. R.J. Buresh et al. (ed.). Replenishing soil
fertility in Africa. SSSA Spec. Publ. 51. SSSA, Madison, W.I.
SAS. (1996) A guide to statistical and data analysis using JMP and JMP IN
software. Statistical Analysis System Institute, Inc., Cary, NC.
Sibomana, L. (2001) Effet de l’application des engrais azotés sur la décomposition
de la paille de blé dans un sol d’altitude du Burundi. Mémoire. Faculté des
Sciences Agronomiques. Université du Burundi.
Stanford, G., and Smith, S.J. (1972) Nitrogen mineralization potentials of soils.
Soil Sci. Soc. Am. Proc. 36: 465-472.
Stevenson, F.J. (1986) Cycles of Soil. Carbon, Nitrogen, Phosphorus, Sulfur,
Micronutrients. John Wiley & Sons, New York, NY.
Stotzky, G. (1965) Microbial respiration. p. 1550-1569. In Black, C.A. e t al.
(ed.), Methods of soil analysis. Part 2. Agronomy 9. ASA, Madison, WI.
Zarcinas, B.A., Cartwright, B. and Spouncer, L.R. (1987) Nitric acid digestion
and multi-element analysis of plant material by inductively coupled plasma
spectrometry. Commun. Soil Sci. Plant Anal. 180: 131-146.
Zibilske, L.M. (1994) Carbon decomposition. In: Weaver et al. (ed.). Methods of
soil analysis. Part 2. Microbial and biochemical properties. SSSA. Madison,
WI.
162
Kaboneka, S. et al
Evaluation of Crop Availability of K and Mg in Organic Materials under Greenhouse Conditions
Evaluation of Crop
Availability of K and Mg in
Organic Materials under
Greenhouse Conditions
163
11
Kaboneka, S.1 and Sabbe, W.E.2
1
Institut des Sciences Agronomiques du Burundi (ISABU). B.P.
795, Bujumbura, Burundi. Université du Burundi, Faculté des
Sciences Agronomiques. B.P. 2900, Bujumbura, Burundi
2
Professor of Agronomy (Deceased). Department of Agronomy,
University of Arkansas, PTSC 115, Fayetteville, AR 72701
Abstract
A greenhouse investigation using sorghum sudan [(Sorghum
bicolor (L.) Moench] was conducted to evaluate the
availability of K and Mg added in organic materials applied
to a Leadvale (fine-silty, siliceous, thermic typic Fragiudult)
and Taloka (fine, mixed, thermic mollic Albaqualf) soil series.
Organic materials were compared to an N-P 2O5-K2O (1313-13) fertilizer during a 70 days greenhouse study. They
were added based on two N rates: 25 and 50 mg N kg-1 soil.
Percent Mg recovery from added organic materials was
substantially lower than that of K. Soybean, corn and wheat
residues were found to be important sources of K, as about
40-50 % of K was available from these residues in a 70
days period. Therefore, we recommend that the contribution
of these organic materials should be taken into account
when formulating K fertilizer programs.
164
Kaboneka, S. and Sabbe, W.E.
Introduction
Most studies on organic residue decomposition have been done on N,
and to some extent P and S mineralization (Kaboneka et al., 1997;
Blackmer and Green, 1995; Aoyama and Nozawa, 1993; Janzen and
Kucey, 1988; Enwezor, 1976). The few studies on K and Mg release
have been performed on foliar litter decomposition in forestry (Bockheim
and Leide, 1986; Blair, 1988). Little is known on the availability of K
and Mg from organic materials applied to soils of different fertility
levels.
The objectives of this greenhouse study were:
i)
to evaluate plant uptake of K and Mg from soil-incorporated cattle
manure, soybean [Glycine max (L.) Merr.], wheat (Triticum aestivum
L.) and corn (Zea mays L.) residues;
ii) to evaluate the effect of soil fertility status on K and Mg release from
added organic materials.
Materials and Methods
Soil physical and chemical properties have been described elsewhere
(Kaboneka et al ., 1997; Kaboneka and Sabbe, 1995) and are
summarized in Table 11.1. The Leadvale soil was limed with 9.0 g of
limestone per pot (2.5 metric tons ha-1) and pots were regularly watered
to approximately field capacity for three weeks prior to sorghum-sudan
planting to activate the added limestone. Characterization of residues
was done by analyzing them for total N by digestion with H 2SO4 and
H2O2, followed by steam distillation of appropriate aliquots (Bremner
and Mulvaney, 1982). Total plant P, S, Ca, Mg and K were analyzed by
inductively coupled plasma (ICP) spectrometry after digestion with
HNO3 and H2O2 of a 0.2 g sample at 120° C (Zarcinas et al., 1987). Ash
content of residues was determined by dry combustion in a muffle
furnace at 550° C for 4 hours.
Results of the analyses expressed on a dry weight basis are presented
in Table 11.2. Of these organic materials, cattle manure provided the
highest total amount of Ca and Mg, whereas corn and wheat residues
provided the highest amount of K.
Polyethylene pots were filled with 3.53 kg of 2-mm sieved air-dried
Taloka and Leadvale soil series. Residues were added based on two N
rates: 25 mg N kg-1 soil and 50 mg N kg-1 soil (56 and 112 kg N ha-1,
respectively). They were compared to an N-P2O5-K2O (13-13-13) fertilizer.
A control (soil only) treatment was also included.
Evaluation of Crop Availability of K and Mg in Organic Materials under Greenhouse Conditions
165
Table 11.1: Selected physico-chemical properties of Taloka and Leadvale soil series
Soil
Parameter
PH (H2O)
K (mg kg-1)
Ca (mg kg -1)
Mg (mg kg -1)
Clay (%)
Silt ( %)
Sand ( %)
O.M. (g kg -1)
Taloka
Leadvale
5.9
121
976
63.5
5.9
68.8
25.3
20
4.6
20
188
46.5
6.7
75.4
17.9
20
Source: Kaboneka and Sabbe, 1995.
Table 11.2: Chemical composition of the plant residues, manure and inorganic fertilizer
Material
Parameter
manure
wheat
soybean
corn
Fertilizer
15.98
2.29
1.24
15.45
3.70
1.75
130.00
56.80
130.00
107.90
9.00
3.00
g kg-1
N
P
S
K
Ca
Mg
3.43
0.97
0.49
1.75
6.40
1.70
9.03
1.33
1.38
9.75
4.30
0.80
28.47
3.19
2.33
17.40
15.30
2.65
Two successive sorghum-sudan crops were harvested from each pot
after 35 days of growth, corresponding to a total of 70 days of nutrient
uptake. Twenty seeds of sorghum-sudan were planted each time. Plants
were thinned to 10 plants after germination. During plant growth, pots
were regularly watered to approximately field capacity. After 35 days,
above-ground plant tissues were harvested, dried for 48 hours at 70° C
and the dry weight recorded. Dry plant samples were ground for chemical
analyses. Experimental treatments consisted of a control (soil only) and
five sources of nutrients (chemical fertilizer, manure, corn, wheat, and
soybean residues). Treatments were applied to Leadvale and Taloka
soil series in a completely randomized design with three replications.
Statistical analyses were performed on each of the two sorghum-sudan
harvests and on their combination by using the SAS-GLM procedures
of the Statistical Analysis System (SAS, 1985). Treatment mean
166
Kaboneka, S. and Sabbe, W.E.
separation was performed by using LSD at the 0.05 level of probability.
The LSD values indicated in different tables were used to compare
treatments within and across soil series.
Total plant uptake of K and Mg was estimated by multiplying the
respective concentration of nutrients by above ground dry matter yield.
Net uptake of the two cation nutrients were estimated by difference
between the nutrient uptake from respective materials and the control
treatment. Percentage recovery of each nutrient was estimated as
follows :
Recovery (%) = [(X - Y) / Z] x 100
where
X = Nutrient uptake from fertilized or amended treatments;
Y = Nutrient uptake from the control treatment;
Z = Total amounts of nutrients added in organic or mineral fertilizer
materials.
Results and Discussion
Nutrient uptake and recovery did not include nutrients in roots. It was
assumed that addition of root biomass and their nutrient content would
not affect treatment differences, as was suggested by Broadbent and
Nakashima (1965). Effects of soil series, residue type, rate of residue
application and their interactions on K and Mg uptake were evaluated
by using orthogonal contrast comparisons (Table 11.3).
Table 11.3. Selected orthogonal contrast comparisons and effect of soil, residue,
application rate and their interactions on total K and Mg uptake from added residues.
Pr > F
Contrast
NPK vs residue
Control vs others
Soil
Residue
Rate
Soil * residue
Soil * rate
Residue * rate
Soil * residue * rate
K
Mg
0.0001
0.0026
0.0001
0.5420
0.0001
0.0001
0.2310
0.8320
0.0340
0.0001
0.6440
0.0001
0.0001
0.3970
0.2490
0.4600
0.5920
0.3050
Evaluation of Crop Availability of K and Mg in Organic Materials under Greenhouse Conditions
167
Potassium Uptake
Potassium uptake for individual and combined harvests are presented in
Table 11.4, which indicates higher K uptake in the first harvest. A two-to
three-fold decrease between the first and the second harvest was observed
for both soils, with the greater decrease occurring in Taloka soil. A similar
trend was recorded with sorghum-sudan dry matter yields (Kaboneka and
Sabbe, 1995). Analysis of variance and orthogonal contrast comparisons
performed on combined K uptake showed significant effects of soil series,
residue application rate, soil series × residue type and soil × residue type
× application rate interactions on total K uptake. Total K uptake from
chemical fertilizer was significantly higher than that from residue
treatments. Both fertilizer and residue effects on total K uptake were
significantly higher in Taloka than in Leadvale soil. These differences were
probably due to the initial exchangeable high K levels in Taloka soil, which
was originally six times higher than that of Leadvale soil. K net uptake
from all added materials was uniformly positive in Leadvale soil, suggesting
an important K release from residues in this soil series. On the contrary,
the only positive net K uptake values in Taloka soil occurred with chemical
fertilizer and with the higher manure and soybean residue rates.
Table 11.4: Effect of plant residues, manure and chemical fertilizer on K uptake by
sorghum-sudan
Harvest I
Material
Control
NPK
Corn
Manure
Soybean
Wheat
LSD (5 %)
Harvest II
N rate K applied Leadvale Taloka Leadvale Taloka
0
25
50
25
50
25
50
25
50
25
50
0
75.4
150.8
87.8
175.6
46.3
92.6
55.0
110.0
98.0
196.0
50.5
199.9
126.1 253.4
148.3 290.5
87.8
168.4
108.1 185.9
74.3
184.6
80.7
232.8
79.7
170.9
96.1
185.6
78.0
154.6
122.6 145.5
41.5
29.5
40.3
59.5
29.7
53.6
26.7
27.8
29.9
39.7
38.0
54.0
62.9
61.8
68.5
64.1
71.9
60.3
69.8
91.1
88.4
68.7
61.8
23.5
Total Harvest
Leadvale Taloka
80.0
262.8
166.4
335.2
208.8
359.0
117.5
232.5
161.7
257.8
101.0
244.9
108.5
302.6
109.6
262.0
135.8
274.0
116.0
223.3
176.6
207.3
40.4
The high K uptake in the limed Leadvale soil could be due to the
complementary effect of the two cation nutrients on soil cation exchange
sites. It has indeed been demonstrated that lime and the Ca ion in
general increases soil solution K by exchanging K from soil exchange
sites (Tisdale et al., 1985).
168
Kaboneka, S. and Sabbe, W.E.
Total K recovery from the lower fertilizer rate was 96 % or more in
both soils, with the higher percent in Leadvale soil (115 %). K recoveries
were 85 % and 64 % with the higher fertilizer rate, respectively. No K
was recovered from the lower manure rate nor from other residues in
Taloka soil, with the exception of the higher manure rate (43 %). K
percent recovery occurred in all residue treatments in Leadvale soil.
However, it decreased from 45 to 31 % from the lower to the higher
manure application rates. A similar trend was observed with the
soybean residues where K recovery decreased from 54 to 51 % from
the lower to the higher residue application rate. On the contrary, the
opposite trend was observed with corn and wheat residues in the same
soil, where high K recoveries occurred with higher residue rates. As a
matter of fact, K recovery from corn residue was 43 and 47 % in
Leadvale soil, whereas 37 and 49 % of wheat K were taken up by
sorghum-sudan from the same soil. However, the data were not
significantly different.
Magnesium Uptake
Similarly to K, total Mg uptake decreased from the first to the second
harvest, particularly in Taloka soil. Analysis of variance and orthogonal
contrast comparisons performed on combined Mg uptake showed
significant effects of soil series and residue application on total Mg uptake
(Table 11.3). Total Mg uptake from chemical fertilizer was significantly
higher than that recorded from residue treatments. Fertilizer effect on
total Mg uptake was significantly higher than residue treatments in
Leadvale soil. Net Mg uptake was positive with manure in both soils
and with the higher soybean rate in Leadvale soil. All other treatments
were characterized by negative net Mg uptake, suggesting that Mg
immobilization possibly occurred during microbial degradation of added
residues. Residue Mg was recovered from manure at both application
rates in Leadvale soil and with the lower soybean application rate in the
same soil. In particular, as much as 24 and 6 % Mg were recovered from
the lower and the higher manure rates in Leadvale soil, respectively.
Substantial Mg (31%) was also recovered from the lower soybean residue
rate in Leadvale soil. No Mg was recovered from Taloka soil from either
applied organic material. In general, percent Mg recovery from added
organic materials was substantially lower than that of K.
Although the chemical fertilizer contained little Mg (Table 11.5), the
highest Mg uptake was observed with the fertilized treatments as
compared to treatments receiving organic materials. One can hypothesize
that fertilizer application released Mg from available forms into soil
solution. The competition theory as discussed for K can also be applied
to Mg. K applied in the chemical fertilizer could have displaced Mg from
Evaluation of Crop Availability of K and Mg in Organic Materials under Greenhouse Conditions
169
soil exchange sites into soil solution, where it would have been actively
absorbed by the sorghum-sudan crops. The same competition theory
could also explain the higher Mg uptake from the fertilizer in Leadvale
soil, since this soil received Ca from applied lime. Therefore, we suspect
that a double competition between Mg and both Ca and K for exchange
sites would have resulted in higher available Mg in soil solution in
fertilized and limed Leadvale soil series (Tisdale et al., 1985).
Table 11.5: Effect of plant residues, manure and chemical fertilizer on Mg uptake by
sorghum-sudan
Harvest I
Material
Control
NPK
Corn
Manure
Soybean
Wheat
LSD (5 %)
Harvest II
N rateMg applied Leadvale Taloka Leadvale Taloka
0
25
50
25
50
25
50
25
50
25
50
0
2.1
4.2
9.9
19.8
45.0
90.0
8.5
17.0
8.0
16.0
13.8
38.6
29.9
17.3
16.1
6.5
25.1
17.4
16.9
10.5
3.9
19.7
21.2
31.2
12.0
10.3
17.3
19.9
12.4
15.1
11.7
10.0
9.6
22.1
21.8
27.0
12.7
19.3
20.1
16.4
21.1
12.7
11.2
13.6
7.8
8.0
5.5
4.9
7.6
6.6
8.8
8.7
7.9
5.2
5.9
8.6
Total Harvest
Leadvale Takola
35.9
59.4
6.9
30.0
35.4
46.6
41.5
38.5
29.6
21.7
27.5
27.5
29.2
36.7
16.9
17.9
23.9
28.7
21.1
23.0
16.9
15.9
10.6
From our data, we observed that K and Mg release from applied
residues followed the order K > Mg. Our results agree with those reported
by other investigators. For example, in a decomposition study of dogwood
(Cornus florida L.), red maple (Acer rubrum L.) and chestnut (Quercus
prinus L.) forest litter, Blair (1988) found that 91 % of K and 58 % of Mg
were released after 2 years of decomposition. For all three species,
nutrients were released in the order K > Mg. In a similar study on foliar
litter and forest floor dynamics in a Pinus resinosa stand, Bockheim
and Leide (1986) found that cation nutrients were also released in the
order K > Mg.
The difference in the two cation nutrient release pattern from plant
materials is governed by their structural nature in the plant matrix.
Whether a nutrient is a structural or non-structural component of plant
tissues will affect its release dynamics during residue decomposition
(Blair, 1988; Budelman, 1988). Among the six major plant nutrients,
namely N, P, K, Ca, Mg and S, K is the only non-structural component
of plant tissue. It is present as a cation freely moving in the cell fluid.
Therefore, K release from crop residues is less correlated with biotic
170
Kaboneka, S. and Sabbe, W.E.
factors as compared to other nutrients. Thus, it is easily leached out
and rendered available to crops when the cell membranes disintegrate
during residue decay.
Conclusion
This study indicated that direct soil application of organic materials
could provide substantial amounts of K. In particular, soybean, corn
and wheat residues were found to be important sources of K, as about
40-50 % of K was available from these residues in a 70 days period.
From the findings of this study, we recommend that K added as crop
residues is readily available for plant uptake and should definitely be
taken into account when formulating K fertilizer programs.
References
Aoyama, M. and Nozawa T (1993) Microbial biomass nitrogen and mineralization
– immobilization processes of nitrogen in soils incubated with various organic
materials. Soil Sci. Plant Nutr. 39: 23-32.
Blackmer, A.M. and GreenC.J. (1995) Nitrogen turnover by sequential
immobilization and mineralization during residue decomposition in soils.
Soil Sci. Soc. Am. J. 59: 1052-1058.
Blair, J.M. (1988) Nutrient release from decomposing foliar litter of three species
with special reference to calcium, magnesium and potassium dynamics.
Plant Soil 110: 49-55.
Bockheim, J.G. and Leide, J.E. (1986) Litter and forest floor dynamics in a
Pinus resinosa plantation in Wisconsin. Plant Soil 96: 393-406.
Bremner, J.M. and Mulvaney, C.S. (1982) Nitrogen - Total. In Page A.L. et al.
(ed.). Methods of soil analysis. Part 2. 2nd ed. Agronomy. 9: 595-624.
Broadbent, F.E. and Nakashima T. (1965) Plant recovery of immobilized nitrogen
in greenhouse experiments. Soil Sci. Soc. Am. Proc. 29: 55-60.
Budelman, A. (1988) The decomposition of the leaf mulches of L eucaena
leucocephala, Gliricidia sepium and Flemingia macrophylla under humid
tropical conditions. Agroforestry Systems 7: 33-45.
Enwezor, W.O. (1976) The mineralization of nitrogen and phosphorus in organic
materials of varying C: N and C:P ratios. Pl ant Soil 44: 237-240.
Janzen, H.H. and Kucey R.M.N. (1988) C, N, S mineralization of crop residues
as influenced by crop species and nutrient regime. Plant Soil 106: 35-41.
Kaboneka, S., Sabbe W.E. and Mauromoustakos A. (1997) Carbon decomposition
kinetics and N mineralization from corn, soybean and wheat residues.
Commun. Soil Sci. Plant Anal. 28: 1359-1373.
Evaluation of Crop Availability of K and Mg in Organic Materials under Greenhouse Conditions
171
Kaboneka, S., and Sabbe W.E. (1995) Evaluation of the fertilizer value and
nutrient releasefrom corn and soybean residues under laboratory and
greenhouse conditions. Commun. Soil Sci. Plant Anal. 28: 1359-1373.
SAS Institute. (1985) SAS user’s guide: Statistics. Version 5 ed. SAS Institute,
Cary, NC.
Tisdale, S.L., Nelson W.L, and Beaton J.D (1985) Soil fertility and fertilizers.
MacMillan Publishing Company. New York.
Zarcinas, B.A. and Cartwright B, and Spouncer L.R. (1987). Nitric acid digestion
and multi-element analysis of plant material by inductively coupled plasma
spectrometry. Commun. Soil Sci. Plant Anal. 18: 131-146.
172
Kaboneka, S. and Sabbe, W.E.
The Influence of Goat Manure Application on Crop Yield and Soil Nitrate Variations in
Semi-Arid Eastern Kenya
The Influence of Goat Manure
Application on Crop Yield
and Soil Nitrate Variations in
Semi-Arid Eastern Kenya
173
12
Kihanda, F. M.1, Warren, G. P.2 and
Atwal, S. S.2
KARI Embu Regional Research Center, P.O. Box 27, Embu
Department of Soil Science, University of Reading, P.O.
Box 233, READING, RG6 6DW, U.K.
1
2
Abstract
The effect of manure on crop yield and soil nitrate -N was
measured during the 1994-5 cropping season at three sites
in Mbeere and Tharaka-Nithi districts of eastern Kenya.
The sites were Machang’a in Mbeere district and Mutuobare
and Kajiampau in Tharaka Nithi district. The treatments
were goat manure applied at rates of 0,5 and 10 t ha-1 (dry
weight basis), annually since 1989, and 5 and 10 t ha-1
annually from 1989 to 1992. The plots were intercropped
with sorghum (Sorghum bicolor) and cowpea (Vigna
unguiculata). The grain yields of sorghum varied widely from
0.3 to 3.9 t ha -1 , depending on site and manuring.
Differences were most marked at Machang’a site. For
example, where manure was continuously applied at10 t
ha-1, sorghum yield of more than 3 t ha-1 was realized as
Kihanda, F. M. et al
174
compared to 1.5 t ha-1 from the residual plots. Topsoil nitrate
concentrations were highest at the start of the season and
within about 15 days, most nitrate had been lost. For
example, at the Mutuobare site, nitrate N concentration at
the start of the season was about 78 mg kg-1 in the recently
manured plots but decreased to less than 10 mg kg-1 20
days after onset of the rains. A high manure rate of 10 t ha-1
resulted in surplus nitrate at the end of the season, hence
a maximum manure rate of 5 t ha-1, is suggested to reduce
losses of N. Nitrate N correlated well with N taken up by
crops at Kajiampau, where N was the main limiting nutrient,
but not at Mutuobare and Machang’a, where P was the
limiting nutrient. The wide variations of measured nitrate
between sites and sampling dates make it unsuitable as a
practical indicator of N requirements for crops. Total soil N
was correlated with N taken up by the sorghum at all sites
and is an appropriate parameter to assess crop requirement
to N.
Key words: Goat manure, sorghum, cowpea , yields, soil nitrate
Introduction
It has long been recognized that manure application is one of the most
effective ways of improving soil fertility and crop production in tropical
African conditions (Dennison, 1961; Watts-Padwick, 1983). For example,
in the Nigerian savanna, application of farmyard manure (FYM) at 2.0 t
ha-1 increased cotton yield by more than 100 % (Hartley, 1937). Similarly,
it has been observed that the production, distribution and application
of manure has been a vital role in sustaining small-holder arable farming
around Kano in northern Nigeria, which is a semi-arid area with a high
population and a long history of arable farming.
In semi-arid Kenya, inadequate soil nitrogen (N) and phosphorus
(P), restrict crop production (Siderius and Muchena, 1977), so manure
and fertilizers can give yield increases in this climate (Ikombo, 1983).
Improved cropping systems are needed, which will be both more efficient
with plant nutrients and more attractive to farmers. In many areas,
such as the semi-arid zones of eastern Kenya, little use has been made
of this resource until recently (Gibberd, 1995a). Development projects
to improve agricultural production have been undertaken in this region
(Gibberd, 1995b) and a part of this program was a multi-location field
experiment with manure initiated in nine sites in the semi-arid eastern
Kenya.
The Influence of Goat Manure Application on Crop Yield and Soil Nitrate Variations in
Semi-Arid Eastern Kenya
175
Nitrate is the main form in which plants take N from soil. It is released
from soil organic matter by mineralization, but is then subject to several
processes by which it can be lost before plant roots can take it up. In
semi-arid climates, nitrate-N concentration is normally highest at the
start of the season (Semb and Robinson, 1969; Wong and Nortcliff, 1995).
The purpose of carrying out this study was firstly to determine the effect
of manure (residual or continuous application) on sorghum and cowpea
grain yield. The second purpose was to monitor changes in nitrate
concentration in the different manuring regimes and thirdly to establish
whether a relationship exists between soil N (total or NO3- N) and crop
yield. The results presented in this paper are part of a long-term trial on
soil fertility improvement conducted in several locations in Mbeere and
Tharaka- Nithi districts in semi-arid Kenya.
Materials and Methods
Field experiment
The sites were located in Mbeere and Tharaka-Nithi Districts of
eastern Kenya. Information on their location, climate and soil is given
in Table 12.1.
Table 12.1: Selected characteristics of the sites and unmanured soils in september 1994
Latitude
Longtude
Altitude (m)
Rainfall (mm)
Mean Annual
October to January season
October 1994 to Jan.1995
pH (water)
pH (CaCl2)
Total N (Kjeldahl)(%)
Olsen –P (mg/kg)
Texture
Machang’a
Mutuobare
Kajiampau
0°45’S
37°40’E
1050
0°45’S
37° 45’E
900
0° 15’S
37° 45’E
750
740
423
439
809
497
616
1040
683
862
6.55
6.84
7.10
5.75
0.066
0.98
Sandy clay loam
6.18
0.092
26.3
Sandy clay loam
6.39
0.059
10.2
Sandy loam
176
Kihanda, F. M. et al
Rainfall was bimodally distributed with peaks in November and April,
exceeding evaporation only in November and December. The soils are
classified as Chromic Cambisols. Machang’a and Mutuobare sites both
in Mbeere district were cleared from long-term bush, while Kajiampau
in Tharaka-Nithi district had been under arable cropping for five years.
The field experiment was set up in 1988, but the measurements
described in this study were made in 1994/95 season. There were nine
sites altogether but three sites were selected for this investigation because
of their contrasting soil properties. Machang’a had low extractable soil
P (0.98 mg kg-1, Table 12.1), Kajiampau had adequate extractable P but
the lowest total soil N, while Mutuobare had adequate N and P. In the
season under study, the crops grown were
(i) sorghum (sorghum bicolor, var.954066) planted in rows 0.7 m apart
of 5 m long and 0.25 m within rows
(ii) cowpea (Vigna unguiculata, var. M66) planted at the same spacings
in rows placed midway between sorghum rows.
There were seven rows of sorghum of 5.0 m long. The treatments
were:
Code Treatments
A1 Goat manure applied annually since 1989 to 1994 at 5 t ha-1
A2 Goat manure applied annually since 1989 to 1994 at 10 t ha-1
B1 Goat manure applied annually from 1989 to 1992 at 5 t ha-1
B2 Goat manure applied annually from 1989 to 1992 at 10 t ha-1
C Control (no manure)
The five treatment were arranged in a randomised complete block
design replicated three times. Treatments A1 and A2 therefore, had
continuous manuring whereas treatments B1 and B2 showed residual
effects of manure. The goat manure was obtained from the Ministry of
Agriculture (Marimanti station) in September each year and broadcast
and incorporated immediately. The cumulative amount of manure applied
from 1989 to October 1994 were 30, 60, 20 and 40 t ha-1 in treatments
A1,A2, B1 and B2, respectively. The manure applied in September 1994
was analyzed and it was found that the amounts of N, P and K applied
with 10 t ha-1 manure were 189, 47 and 372 kg ha-1, respectively.
Soil sampling and analysis
The characterisation data of soils presented in Table 12.1 were for bulk
samples collected to a depth of 20 cm in September 1994, before
cultivation, manure application and sowing. The first sampling was done
before the onset of rains while subsequent sampling were done at
intervals of 25 days and continued for a period of three months. For
measurements of nitrate during the season, five soil cores from 0 to 15
The Influence of Goat Manure Application on Crop Yield and Soil Nitrate Variations in
Semi-Arid Eastern Kenya
177
cm were taken using an auger in each plot. This soil was mixed by hand
in a bucket, a sample of about 1kg taken and placed in a polyethylene
bag. Fresh soil (45 g), was weighed into a polypropylene bottle (250 ml),
1M KCI (200 ml) added and shaken end-over-end (10 revolutions/
minute) for 1 hour on a rotary shaker. Nitrate was measured with a
“Heloflow” portable flow injection analyser (WPA Ltd., Ware, UK), using
the Griess-Llosvay procedure (Keeney, 1982). In this method, nitrate-N
is reduced to nitrate by copperised Cd metal, reacted with N- (1-naphthyl)
ethylene diamine dihydrochloride (NEDD) and measured colorimetrically
(at 540 nm). The rest of the soil sample was air-dried (25-30°C), for 2 to
5 days in a freely ventilated shed at Machang’a. Dried soils for Machang’a,
Mutuobare and Kajiampau under these conditions contain
approximately 2% water on an oven-dry basis. Results here are presented
on the basis of air dry soil.
Crop harvests
Cowpeas were harvested in January 1995 and sorghum in February
1995. For each crop, the above-ground residues (leaves, stalk and
threshing residues) were collected separately. The grains and residues
were air-dried (25-30°C) for each plot separately, weighed and subsamples taken for anlaysis.
Statistical analyses
Initially, data for each site was treated separately. Where required
comparisons between sites were made by pooling Standard Errors. Grain
yields, N uptakes and soil nitrate at each sampling time was analyzed
by ANOVA (Manure treatment × Replicate blocks in the field). Correlations
were made between N uptake by sorghum and cowpea in each plot and
measured soil total N and soil nitrate at each sampling time. Statistical
calculations were done using INSTAT version 5.31.
Results
Crop yields and N uptake
The yields of sorghum depended on site and manuring. In general, yields
and N uptake increased in the order of treatments C<B1< B2<A1<A2 (Tables
12.2 and 12.3). Differences were most marked at Machang’a, where, for
continuously applied manure, 10 t ha-1 manure, whereas for residual
manure the earlier application of 10 t ha-1 was no more effective than
Kihanda, F. M. et al
178
5 t ha-1 for increasing yield. Differences were least marked at Kajiampau,
where the manured treatments A1, A2 and B2 slightly differed in yield.
However, Kajiampau was the only place cowpeas responded significantly
to manure.
Table 12.2: Grain yield (kg ha-1) of sorghum and cowpeas at the three sites
Treatment
Machang’a
Mutuobare
Kajiampau
Sorghum
C
B1
B2
A1
A2
(SE; df=8)
360
1580
1560
3153
3893
(248)
920
2267
1853
2067
2653
(124)
627
853
1360
1480
1920
(259)
73
400
233
113
527
(190)
1653
1733
1653
1720
1827
(87)
1147
1133
1307
1347
1520
(85)
Cowpea
C
B1
B2
A1
A2
(SE;df=8)
Standard errors are given in paretheses
Table 12.3: Total nitrogen uptake in grain plus above-ground residues (kg N ha-1) by
sorghum and cowpeas at the three sites
Treatment
Machang’a
Mutuobare
Kajiampau
Sorghum N uptake
C
B1
B2
A1
A2
(SE;df=8)
24.4
68.6
64.8
123.4
140.0
(15.3)
53.7
114.8
101.6
111.0
141.1
(12.3)
34.9
53.4
81.3
74.2
85.3
(10.4)
Cowpea N uptake
C
B1
B2
A1
A2
(SE;df=8)
5.27
24.47
12.96
8.65
29.42
(11.85)
79.5
81.5
77.2
89.4
92.0
(8.3)
54.9
52.5
56.9
68.3
76.2
(4.5)
Standard errors are given in parentheses
The Influence of Goat Manure Application on Crop Yield and Soil Nitrate Variations in
Semi-Arid Eastern Kenya
179
Temporal variations in soil nitrate
Machang’a site
Nitrate concentrations were high at the start of the season and declined
in the next 15 days (Figure 12.1), due to leaching or more probably,
denitrification. The concentration was higher at day 56, but remained
generally low for the rest of the season. It rose again towards harvest
time, suggesting that soil N was being mineralized but no longer required
by the crop. Although nitrate values were higher in manured soil at the
beginning and end of the season, there were no significant differences
between treatments on any one occasion (Table 12.4).
Nitrate - N (mg kg-1)
Figure 12.1: Nitrate N concentration at Machang’a for different sampling dates
Time (days) (Day 1 = 1 October)
Table 12.4: Effect of manure treatment on soil nitrate N (mg kg-1) in the three sites at the
final soil sampling of the season
Treatment
Machang’a
Mutuobare
Kajiampau
C
B1
B2
A1
A2
(SE;df=8)
2.89
1.29
2.10
1.66
4.56
(0.79)
4.39
4.51
6.11
7.09
9.90
(1.08)
1.03
0.62
0.59
1.15
1.93
(0.16)
Standard errors are given in parentheses
180
Kihanda, F. M. et al
Mutuobare site
Nitrate was highest at the start of the season and followed a pattern
similar to that at Machang’a. At the first two samplings, soil nitrate
concentration increased in the order of treatments: C<B1<B2<A1<A2
(Figure 12.2). As would be expected, the higher the manure rate, the
more nitrate in recently manured soil (A1 and A2) than in the residual
manure plots (B1 and B2). At the second sampling, the difference
between treatment groups were significant as follows: C<B<A. In addition,
nitrate was significantly less in the control (C) than the other treatments
on days 75 and 120. By the end of the season, nitrate was still
significantly higher in treatment A2 compared to C (Table 12.4).
Nitrate - N (mg kg-1)
Figure 12.2: Nitrate N concentration at Mutuobre for different sampling dates
Time (days) (Day 1 = 1 October)
Kajiampau site
Nitrate was highest at the start of the season and changed in a pattern
similar to that at Mutuobare. At the first two samplings, soil nitrate
concentration increased as follows: C<B<A (Figure 12.3). As at
Mutuobare, the differences were not significant at the first sampling,
but were, at the second sampling. Freshly manured soil (treatment A)
also contained significantly more nitrate than other soils at days 75
and 116. By the end of the season, the differences between treatments
appeared small (Table 12.4), but the difference between A2 and the
other treatments was highly significant (P=0.01).
The Influence of Goat Manure Application on Crop Yield and Soil Nitrate Variations in
Semi-Arid Eastern Kenya
181
Nitrate - N (mg kg-1)
Figure 12.3: Nitrate N concentration at Kajiampau for different sampling date
Time (days) (Day 1 = 1 October)
Table 12.5: Correlation coefficients between (i) total N uptake (grain plus residues) by
sorghum or cowpeas and (ii) soil nitrate or soil total N
Sorghum N uptake (kg N ha-1)
Planned
Sampling Day
Machang’a
Mutuobare
Kajiampau
0
10
25
50
75
100
120
0.628*
0.166
0.326
-0.337
-0.503
0.102
0.077
0.657**
0.423
0.656**
0.406
0.640*
0.450
0.442
0.760***
0.444
0.645**
0.397
0.678*
0.310
0.328
Cowpea N uptake (kg N ha-1)
Planned
Sampling Day
Machang’a
Mutuobare
Kajiampau
0
10
25
50
75
100
120
0.171
0.608*
0.609*
-0.102
-0.035
0.355
0.163
-0.049
0.345
0.253
0.147
0.359
0.374
0.137
0.524*
0.578*
0.603*
0.641**
0.762**
0.544*
0.764**
182
Kihanda, F. M. et al
Correlations between N uptake and soil total N and nitrate N
At Machang’a, there were no good correlations of N uptake with soil
nitrate although significant correlations at the 5% level were found for
cowpeas on day 20 (Table 12.5).
At Mutuobare, N uptake by sorghum and total N uptake were
correlated highly significantly with total soil N and nitrate measured at
days 26 and 76. Cowpea N uptake did not correlate with soil nitrate.
For Kajiampau, N uptakes by both sorghum and cowpeas, total N uptake,
correlated significantly with soil total N and also with soil nitrate on
many sampling days (Table 12.5). For sorghum N uptake, the closeness
of correlation with soil total N increased in the order Machang’a
<Mutuobare< Kajiampau, and this correlation was always better than
with any nitrate measurements.
Discussion
Temporal variation in soil nitrate
At all sites, large losses of nitrate were observed between the first and
second sampling. In the companion experiment at Machang’a,
denitrification seemed the most likely loss mechanism, because of the
speed of loss (Warren et al. 1997). The possibility of substantial
denitrification in Kenya soils is supported by laboratory measurements
made on a Vertisol and a Phaoezem from western Kenya (Sigunga 1997).
Denitrification occurred when the soils were maintained at 60% water
holding capacity for four days. Because of periods of prolonged rainfall,
these conditions were almost certainly achieved during the season of
the field experiment here. The general pattern of nitrate concentration
was high at the start of the season, followed by rapid losses at the onset
of rain, low levels for a couple of months and a rise towards the end of
the season. This is of a particular relevance to the timing of fertility
improvement. Mineral fertilizers applied soon before a time of high losses
will also be subject to rapid loss and low efficiency. For this reason,
basal applications of mineral N are inappropriate, but if manure is used
in conjunction with late applied mineral N fertilizer, some synchrony of
supply and demand can be achieved as suggested by Murwira and
Kirchmann (1993).
Effect of manuring history on soil nitrate
Analyses of the manure applied in October 1994 showed that treatments
A1 and A2 received 94 and 189 kg N ha-1. Similar amounts would have
The Influence of Goat Manure Application on Crop Yield and Soil Nitrate Variations in
Semi-Arid Eastern Kenya
183
been applied in previous years because the source of the manure was
the same every year from October 1989. Recently manured soil was
able to supply nitrate throughout the growing season as indicated by
the often higher amounts of nitrate in manured soils. The clearest
evidence came at the end of the season, when at Mutuobare and
Kajiampau, nitrate remained significantly higher in treatment A2
compared to C and A1. For the other treatments, there were no significant
treatment differences, but the yields and N uptakes, especially for
sorghum, were improved in treatments A1, B1 and B2. This suggested
that by the end of the season, the readily mineralised N had been taken
up by the crops and only more resistant organic matter remained.
For treatment A2, surplus available N was left over at the end of the
season, indicated by the raised nitrate concentration. Probably it would
be lost at the next rains as mentioned in the preceeding sub-section
and two further facts supported this hypothesis. First, measurements
of soil total N for all plots in 1993 after 4 years’ manuring, showed that
the higher rate of manure gave little extra soil N over the low rate (Warren
et al., 1997). Second, in the season from October 1994, there was no
significant difference between treatments B1 and B2 in yield or N uptake
suggesting the same reserve of available N in these two treatments. It is
therefore proposed that manure applications may not provide more than
94 kg N ha-1 per year (the amount in this experiment) because the extra
N is lost.
Transport of manure to the fields presents a problem to farmers.
For this reason, there has been several examinations of the relative
benefits of applying a small amount of manure frequently or a larger
amount at intervals of several years. Grimes and Clark (1962), concluded
that in Coast Province, Kenya, an annual application of 3 t ha-1 every 3
years were equally effective. In contrast, Gatheca (1970), concluded that
5 or 6 t ha-1 every year was better than 20 to 30 t ha-1 every 4 or 5 years
at Embu, although the latter was a particularly high rate. In the present
work, the residual manure did provide extra yield in the fifth season
after the last application. We suggest that as long as the maximum rate
applied in one season is not so large as to result in losses in the current
season, then the rest will be stabilized in soil organic matter. Therefore,
it does not matter whether manure is applied every season or in some
seasons only.
Differences between sites
Concentrations of nitrate generally increased in the order Kajiampau
<Machang’a<Mutuobare (Table 12.3) and since non-leguminous plants
obtain most N via nitrate, this confirms that Kajiampau had the lowest
available soil N. Differences between soils in the importance of N as a
184
Kihanda, F. M. et al
fertility constraint, were also indicated by the closeness of correlations
between N uptake and nitrate or total soil N. Particularly for sorghum,
crop N uptake was best related to soil N at Kajiampau but least at
Machang’a, suggesting that N was the main deficiency at Kajiampau
unlike Machang’a. At Machang’a the unexpected situation arose where
soil nitrate was consistently highest in the unmanured treatment from
days 49 to 80. Extractable P was very low here (0.98 mg kg-1, Table
12.1) and it is postulated that because P was so deficient, there was
surplus N in treatment C. But when extra P was added in manure, the
crops were able to utilise N from both manure and soil. Although
sorghum N uptake was increased significantly by manure (Table 12.4),
the N uptake was not strongly related to soil N.
An intercropped system had been chosen because this is the normal
farming practice in most localities of the region. However, this makes
interpretation of results more difficult, since the cowpeas would be able
to supply at least a part of the N by fixation from the atmosphere.
Nevertheless, cowpeas at Kajiampau gave the most significant
correlations between nitrate and yield (Table 12.5) and the potential for
N fixation here would not be restricted by lack of P. These results indicates
that legumes are not net providers of N to soil but also utilise mineralised
soil N.
Prediction of N requirements
Worldwide, there is no generally accepted method to test soil for avilable
N. In dry climates, the soil nitrate at the start of the season can be a
major contributor to the N supply. If measured, the rest of the N required
as fertilizer can be calculated , ignoring mineralization. The method
was successfully used in the dry climate of southwest USA (Dahnke
and Johnson 1990). However, this measurement is of little help in
assessing requirements for manure or fertilizer in the circumstances of
agriculture in semi-arid Kenya. Nitrate is still the major fraction of soil
N that plants can take up, but because of large losses at the start of the
rains, nitrate-N had no relationship to the available N later on.
In contrast, the total soil N at the start of the season was always
significantly correlated with yield and N uptake for sorghum at every
site and is therefore the most reliable indicator of available soil N in
these soils. There is a paradox here in that much of the plants N should
have been supplied by the manure applied in October, but this N was
not included in the total N, which was measured in samples taken in
September. The situation may be explained because the manured soils
had a supply of N built up by previous manure applications. Each season,
a relatively constant proportion of soil organic N is mineralised, between
1 and 3% (Bremner 1967). This is substituted by freshly immobilised N
The Influence of Goat Manure Application on Crop Yield and Soil Nitrate Variations in
Semi-Arid Eastern Kenya
185
from organic residues. Provided the manuring regime is maintained,
total soil N remains proportional to the “steady state” mineralised N
which feeds the plants.
Conclusions
The effect of manure on sorghum and cowpea yield depended on the
initial soil fertility status of the sites. The highest yield response was
recorded in Machanga where soil P was very low. To conserve the fertility
value of manure, it is suggested that no more than 5 t ha-1 per year of
manure should be applied, because at higher rates there was surplus
nitrate at the end of the season and this nitrate would probably be lost
at the start of the next season. At lower application rates, manure had
significant effects in the fifth season after application, so the frequency
of application should be of little consequence for overall efficiency. This
gives opportunities for labour saving management in scheduling of
applications.
At Kajiampau, the crops responded mainly to N in the manure,
indicated by close correlations of soil N with crop N uptake, whereas at
Machang’a, the crops responded mainly to P in the manure. Although N
is supplied from soil to plant roots in the form of nitrate, the amounts of
nitrate at any time were not well related to yield and N uptake. Total N
is the better indicator of the fertility of soil with regard to N.
Acknowledgements
We thank the Directors of Embu Regional Research Center and the
National Agricultural Research Laboratories (NARL), Nairobi for provision
of field and laboratory facilities. We are grateful to P. Mutwiri, P. Njoroge
and J. Odhiambo, for maintaining the field experiments and technical
assistance. This publication is a result of research funded by the
Department for International Development (DFID) in the UK, but the
DFID accepts no responsibility for any information provided or views
expressed.
References
Bremner, J.M. (1967) Nitrogenous compounds. In: Mc Laren, A.D., Peterson,
G.H. (eds), Soil Biochemistry Vol. 1. Marcel Dekker, New York.
Dahnke, W.C., Johnson, G.C. (1990) Testing soils for available nitrogen. In: Soil
Testing and Plant Analysis, 3rd edition RL Westerman (Ed). Soil Science
Society of America, Madison, USA. pp 127-139.
Dennison, E.B. (1961) The value of farmyard manure in maintaining fertility in
northern Nigeria. Empire Journal of Experimental Agriculture 29: 330-336.
186
Kihanda,
F. M. Kihanda
F. M. et
et al
al
Gatheca, T.W. (1970) The maintenance and improvement of soil fertility under
arable crops and grass leys in the first and second rotation cycles of a
fertiliser trial at Embu. East African Agriculture and Forestry Journal 35:264252.
Gibberd, V. (1995a) A farmer-friendly research project in semi-arid Kenya.
Tropical Science 35:308-320.
Gibberd, V. (1995b) Yield responses of food crops to animal manure in semiarid Kenya. Tropical Science 35:418-426.
Grimes, R.C., Clarke, R.T. (1962) Continuous arable cropping with the use of
manure and fertilisers. East African Agriculture and Forestry Journal 28:7480.
Hartley, K.T. (1937). An explanation of the effect of farmyard manure in northern
Nigeria. Empire Journal of Experimental Agriculture 19: 244-263.
Ikombo, B.M. (1983) Effects of FYM and fertilisers on maize in semi-arid areas
of Eastern Kenya. Dryland Farming Research in Kenya, East African
Agriculture and Forestry Journal (Special Issue) 44:266-274.
Keeney, D.R. (1982) Nitrogen availability indices. In: Methods of Soil Analyses,
Part 2. Page, A., Miller, R.H. and Keeney, D.R. Agronomy Monograph No.9
(2nd Edition). American Society of Agronomy. Madison, W1, USA.
Murwira, H.K., Kirchmann, H. (1993) Nitrogen dynamics and maize growth in a
Zimbabwean sandy soil under manure fertilization. Communications in Soil
Science and Plant Analysis 24:2343-2359.
Semb, G., Robinson, J.B.D. (1969) The natural nitrogen flush in different arable
soils and climates in East Africa. East African Agricultural Forestry Journal
34:350-370.
Siderius, W., Muchena, F.N. (1977) Soils and environmental conditions at
agricultural research stations in Kenya. Miscellaneous paper no.M5, Ministry
of Agricultural University.
Sigunga, D.O. (1997) Fertilizer nitrogen use efficiency and nutrient up take by
maize (Zea mays L.) in Vertisols. Chapter 5. PhD Thesis, Wageningen
Agricultural University.
Warren, G.P., Atwal, S.S., Irungu, J.W. (1997) Soil nitrate variations under grass,
sorghum and bare fallow in semi-arid Kenya. Experimental Agriculture
33:321-333
Watts-Padwick, G. (1983) Fifty years Experimental Agriculture.II. The
maintenance of soil fertility in tropical Africa: A review. Experimental
Agriculture 19: 293-310.
Wong, M.T.F., Nortcliff, S. (1995) Seasonal fluctuations of available N and soil
management implications. Fertilizer Research 42:13-26.
Managing Manures Throughout Their Production Cycle Enhances Their Usefulness as
Fertilisers: A Review
Managing Manures
Throughout their Production
Cycle Enhances their
Usefulness as Fertilisers: A
Review
187
13
Kimani, S.K.* and Lekasi, J.K.
Kenya Agriculture Research Institute, NARC Muguga P.O.
Box 30148 Nairobi Tel: 0154-32590; 0154-33190 Fax: 015432348; Email: skimani@net2000ke.com
* Corresponding author
Abstract
Per capital food production lags behind population growth
in most parts of sub-Saharan Africa. One of the reasons for
this situation is the decline in soil fertility, arising from
continuous cultivation where the levels of soil
replenishment, by whatever means, are too low to redress
the process of nutrient mining. One of the ways to address
the problem of the low and declining soil fertility is by using
inorganic fertilisers, but use of these inputs in most of subSaharan Africa is currently low. High costs, unavailability,
marketing problems, poor infrastructure, and the absence
of enabling policy environment are major reasons for the
low use of inorganic fertilisers. The use of other resources
188
Kimani, S.K. and Lekasi, J.K.
available on farm is therefore increasingly gaining
importance. These include green manures, farmyard
manure, crop residues and composts. Of these resources,
farmyard manure is by far the most important. However, a
major limitation in the effectiveness of organic resources is
the quality and quantity of these materials. This review
paper attempts to define cattle manure quality and
discusses some management factors that influence its
quality. The paper also suggests some way forward in better
use of cattle manures for enhanced crop production.
Introduction
Per capital food production lags behind population growth in most parts
of sub-Saharan Africa. One of the reasons for this situation is the decline
in soil fertility. The low and declining soil fertility arises from continuous
cultivation where the levels of soil replenishment, by whatever means,
are too low to mitigate the process of soil mining, whereby the soil fertility
is not replaced by new inputs. One of the effective ways to address this
problem is by using inorganic fertilisers. This is, however, beset by several
problems. Africa’s average annual fertiliser use is only 20 kg ha-1 against
a world average of 96 kg ha -1 (Heisey and Mwangi, 1996). In central
highlands of Kenya, farmers who use inorganic nitrogen (N) fertilisers
do so at rates between 15-25 kg N ha-1, which is far below recommended
rates at 40 kg N ha-1 and above (Kimani et al, 2001). High costs, marketing
problems, and poor infrastructure are major reasons for the low use of
inorganic fertilisers. The use of other resources available on farm is
therefore increasingly gaining importance. These resources include green
manures, farmyard manure, crop residues and composts.
Of these resources, farmyard manure is by far the most important.
In most farms in central Kenya the manure used is mainly cattle (65%)
with the rest comprising sources such as shoats (6%) and poultry (4%)
(Kimani et al., 2000). Most of the manures are from own sources (83%)
with a very small proportion of farmers (2%) purchasing manure (Kimani
et al., 2000). Manures, or other organic inputs applied to the soil control
the rate, pattern and extent of growth and activity of soil organisms and
provide the source of carbon, energy and nutrients for the synthesis of
soil organic matter. Manure can increase the humus content of soils by
15-50%, depending on soil type, in addition to increasing soil aggregate
stability and root permeability (Klapp, 1967). In the longer term, as
shown in an ongoing experiment (20 years by 1996) in Kabete, Kenya
Managing Manures Throughout Their Production Cycle Enhances Their Usefulness as
Fertilisers: A Review
189
(Swift et al., 1994; Kapkiyai et al., 1996), manuring restocks the
particulate organic matter fraction better than fresh crop residues.
Manure also acts as a buffer, thus improving nutrient uptake for crops
grown in acid soils. Using manures alleviates aluminium toxicity and
improves the availability of nutrients such as P, particularly in soils
with a high P fixation, and sulphur (S) (Simpson, 1986). Manure also
supplies essential elements such as Mg, and trace elements which may
not be available in commonly used inorganic fertilisers (Simpson, 1986).
However, the use of manure has several drawbacks. Firstly, the
farmers cite quantity as a problem, that the manure is usually not
enough. Secondly, the quality of manure with regard to nutrient release
and crop uptake is poor. In some instances manure has alternative
uses such as fuel and house construction material. Despite these
drawbacks, manure continues to be an important source of nutrients.
This paper attempts, firstly, to define manure quality. This is followed
by a discussion on the improvement of manure quality through better
management of animal feed, coupled with improved collection and
storage methods. The paper also provides a brief discussion on the effects
of combining manures with inorganic fertilisers.
What is manure quality?
Manure quality may simply be defined as the value of manure in
improving soil properties and enhancing crop yields. Scientists have
used laboratory analysis for nutrient contents as a measure of quality.
The perception has been that the higher the nutrient levels, the better
the manure quality. More recently the use of nutrient release patterns,
using laboratory incubations of manures, and how the nutrient release
can be synchronised with crop uptake has been considered a better
measure of manure quality. On the other hand farmers have traditionally
used their own yardsticks to determine what quality manure is. The
challenge is therefore to match the scientist and farmer perceptions to
come up with simple decision making tools for defining quality manure
without expensive laboratory analysis. An example of laboratory analysis
for manure quality determination is given in Table 13.1. While the values
given are means, the range is quite variable and wide. For instance, N
contents for cattle manures from Kenya range from 0.20-2.2%N, whilst
P contents range from 0.08-0.95%.
The farmers of central Kenya use texture, longevity of composting,
homogeneity, presence of fungi spores/hyphae, as some of the quality
characteristics (Lekasi et al., 1998; Wanjekeche et al., 1999). In Ethiopia,
Tigray region farmers distinguish between two types of manure, the
‘husse’ and ‘aleba’, based on the degree of decomposition. The ‘husse’ is
190
Kimani, S.K. and Lekasi, J.K.
well-decomposed and rich in plant nutrients while ‘aleba’ is less
decomposed and has less nutrients (Kihanda and Gichuru, 1999). Table
13.2 shows indicators of manure quality as determined by farmers in
West Pokot district, northern Kenya.
Table 13.1: Nutrient contents of farmyard manure samples collected from different countries.
Nutrient content (%)
Country
UK
Kenya
Zimbabwe
Madagascar
N
P
K
Ca
Mg
1.76
1.62
0.80
1.10
0.24
0.50
0.20
0.80
1.29
1.34
0.85
0.86
0.74
0.26
0.25
0.85
0.34
0.26
0.15
0.40
Source: Manure management for soil fertility improvement (Kihanda and Gichuru, 1999)
Table 13.2: Indicators of good quality manures used by farmers in Cheptuya village,
West Pokot district, northern Kenya.
Indicator
Fine soil-like texture
Black-grey colour
Longer time of composting
Appearance of white caterpillars
Lack of heat in the manure
Frequency of farmers
10
12
3
5
2
Source: Wanjekeche et al., 1999
Management factors influence manure quality: Effects of
animal feed
Animals fed on high quality supplements produce high quality manures.
The high quality supplement would range from feeds concentrates
(Lekasi, 2000; Odongo, 1999), to high N content legumes (Delve et al.,
1999). Tables 13.3 and 13.4 show the effects of diet supplementation
on manure quality, with regard to P and N. The practicality of these
findings at the smallholder farm level is doubtful. This is because in
most situations farmers feed their livestock opportunistically. This is
the feeding situation where a farmer feeds the livestock with whatever
feed may be available at a particular time. In general, feeds fluctuate
with the rain patterns, where large quantities of high quality are available
during wet periods, and low quantities of poor quality dominating during
Managing Manures Throughout Their Production Cycle Enhances Their Usefulness as
Fertilisers: A Review
191
the dry periods. The improved quality associated with improved diets is
therefore more practical in the more intensive systems for instance under
Zero-grazing units, which are generally associated with farmers with
medium to high income.
Table 13.3: Effect of feeding different P supplements on manure P contents.
Diet
% P of the collected manure
Basal diet (Napier grass)
Busumbu Rock Phosphate
Mijingu Rock Phosphate
Bone meal
Unga commercial feed
0.24
0.70
0.45
0.50
0.95
Source: Odongo, 1999, (unpublished)
Table 13.4: Total N content and C:N ratio of faeces obtained from feeding supplemented
diets. Percentages indicate supplementation in relation to the total dry matter offered as
feed
Faecal sample
Barley straw basal diet
15% C. calothyrsus diet
30% C. calothyrsus diet
15% M. axillare diet
30% M. axillare diet
15% poultry manure diet
30% poultry manure diet
Nitrogen (%)
C:N ratio
0.9
1.4
1.7
1.1
1.2
1.2
1.3
27
23
22
20
23
27
23
Source: Delve et al., 1999.
Collection and Storage methods
In the extensive systems, where animals graze freely manuring is done
in-situ as the animals graze. Where they are confined overnight at the
‘Kraal’, the manure collected usually comprises faeces only. The dung
is heaped besides the kraal as a continuous process throughout the
year. This system is common in pastoral areas of Kenya, such as Maasai
land and West Pokot (Wanjekeche et al., 1999). Similar methods of
manure collection have been reported for the communal grazing areas
of Zimbabwe (Nzuma et al., 1998). A limitation in this method of manure
collection is that most of the urinary N is leached down the soil profile.
192
Kimani, S.K. and Lekasi, J.K.
It is also suffices to say that a considerable amount of N is lost via
volatilization. During the wet season, the soggy anaerobic conditions
may result in denitrification. Where possible, there is therefore need to
improve on composting methods, for instance in areas characterised by
intensive farming.
Methods of Composting
The purpose of composting is to allow further microbial decomposition.
Methods of composting include surface heaps, pitting and deep litter
systems. In Zimbabwe, Nzuma et al., (1998) compared pit and heap
composting, with or without straw additions. They showed that manures
composted for three months using pit method was of higher quality (N
content) than the surface heap. These differences of N contents in the
manure could be related to the pH of the manures during the composting
process. In the heaps, where conditions are aerobic, the manure pH is
normally high (8-9). This tends to stimulate N losses via volatilization.
On the other hand, manure stored under anaerobic conditions tends to
produce organic acids that lead to a lower pH (<7), and therefore fewer
losses of N via volatilization (Kihanda and Gichuru, 1999). A threshold
moisture content of 40-60% is recommended for composting with a
view to enhancing fertiliser value. Table 13.5 shows different composting
methods and the effects on quality. Farmer cattle manure from a
traditional system of central Kenya has manures with a lower N
concentration compared with improved composting systems, other than
the Maasai manure. The stable system is where the animal is confined
throughout. Feed is provided on the stable floor, and the animal feeds
on what is necessary and tramples on the rest, mixing it with the urine
and faeces. In faeces, urine and feed refusals (F+U+FR), the animal is
fed from a trough and the refusals are collected on a daily basis and put
in the zero-grazing unit, where they are mixed by the animal through
trampling. F+FR refers to the system where animals are fed from a
trough. The feed refusals and faeces are collected daily and heaped in a
covered storage area, outside the stable, where they are mixed manually.
F+U refer to the system where the faeces and urine only are mixed by
the animal in the stable. Faeces alone (F) is the system where only the
faeces are composted, in a heap outside the stable, with no urine. Other
than the farmer practice and the Maasai, the other composting systems
were done at a research station, where composting period was 90 days.
Details of the procedures of all these systems are provided by Lekasi
(2000).
Managing Manures Throughout Their Production Cycle Enhances Their Usefulness as
Fertilisers: A Review
193
Table 13.5: N and C:N ratios of manures under different composting systems
Type of manure/composting
N%
C:N ratio
Farmer cattle manure
Stable
F+U+FR
F+FR
F+U
F
Maasai cattle manure
1.1
1.6
1.7
1.9
1.5
1.6
0.8
31
23
21
19
24
25
32
Source: Lekasi, 2000. Faeces (F), Urine (U) and Feed Refusals (FR)
The results show that composting process affects C:N ratios.
Subsequent work in the field showed that manures with high N contents
resulted in higher N mineralisation and a better crop performance, with
the exception of Maasai kraal manure (Lekasi, 2000).
Composting in Zero-grazing Systems
In zero grazing units, it is important to manage manures to enhance
fertiliser quality. Where animals are confined this way, the land sizes
are generally small, for some of them as small as 0.01 ha, for instance
in Kiambu and Muranga districts of central Kenya. In the high potential
areas, manures combined with feed refusals are of superior quality
compared with faeces alone (Table 13.5). The added feed refusals help
to conserve urinary N, by minimising leaching losses, apart from
providing a conducive environment for aerobic decomposition. This
composting method thus reduces environmental problems associated
with leaching of nutrients.
According to studies by Lekasi (2000), a small scale farm can produce
cattle manures that are able to supply 100 kg N within a period of six
months, and this supply may be in excess of the farm requirement for a
0.01 ha small farm in densely populated Kiambu district of central Kenya.
This raises a question as to why farmers in the central Kenya highlands
continue to indicate that they do not have enough manures (Makokha
et al., 2001). It is probable that manures produced may be of low quality,
rather than quantity, and therefore not effective in raising crop yields,
to the satisfaction of farmers. The finding thus highlight the possibility
that what may be required in the high potential areas of central Kenya
may be a better focus on manure quality and management. Such results
also indicate the possibility of selling the excess manures elsewhere,
thus providing a source of income to those farmers with surplus manure.
194
Kimani, S.K. and Lekasi, J.K.
The studies by Lekasi (2000) also bring another dimension to manure
quality. For instance, Maasai cattle manure with a low N concentration
0.8% was more superior to increasing maize yields when compared to
feedlot Zero-grazing manure produced under experimental conditions
(N = 1.6%) . The Maasai manure is usually collected from a kraal, where
animals stay overnight after grazing. It is usually dry and dusty and of
a finer tilth compared with the manure collected from zero-grazing units
which is wet (usually 70% moisture), and aggregated into clods. It is
possible that the fine tilth of the Maasai cattle manure provides a high
surface area of application (when the manure is broadcast). The higher
microbial activity releases nutrients readily for crop uptake in these
Maasai manures compared to the feedlot ones , which comprise of moist
clods and are more difficult to distribute in the soil. This raises questions
as to whether farmers need to dry and grind manure for enhanced
effectiveness.
Improving manure effectiveness by placement methods
The placement method also influences the effectiveness of manures. In
a trial conducted at Thika, central Kenya, placing manure in a planting
hole, as farmers commonly do, resulted in higher maize yields compared
with broadcast manures. The yields were 3.5 t ha -1 and 1.3 t ha -1 for
hole and broadcast treatments respectively (Kimani, 1999); Lekasi, 2000),
where manures had been applied at an iso-N level of 75 kg ha-1.
Improving manure effectiveness by combining with mineral
fertilisers
The interaction between manures and inorganic fertilisers is increasingly
becoming an important subject of research. A combination of organicinorganic nutrient sources is thought to improve the synchronization of
nutrient release and subsequent uptake by the crop. For example, the
synchrony between N release and uptake is thought to be best achieved
under a combined application of manures and inorganic fertilisers. This
is particularly so when the manures are available on-farm, where only
modest application of inorganic fertiliser are applied. The concept of
organic-inorganic combinations (Table 13.6) has been demonstrated in
central Kenya by Kimani et al., (2001), where the combinations resulted
in higher maize grain yields. The increased maize yields above an
unfertilised control were 60%, 50% and 40% for mineral fertiliser alone,
fertiliser-manure combination, and manure alone, respectively, in a
single season (Table 13.6). Manure 1 and 2 had N contents of 1.8% and
Managing Manures Throughout Their Production Cycle Enhances Their Usefulness as
Fertilisers: A Review
195
1.6% respectively, and differed in composting methods whereby manure
1 consisted of faeces (F) alone plus feed refusals (FR) and was manually
turned in a heap. On the other hand manure 2 consisted of faeces and
feed refusals and was mixed by the animal via trampling in the animal
housing. Both manures were composted over a period of four months
and were applied at the rate of 75 kg N ha-1. The small differences in N
contents may therefore not result in significant differences in grain yields
associated with the two manures, though both differed significantly
compared with the control.
Table 13.6: Effects of manures when applied singly or in combination with mineral
fertilisers on maize yields in Kariti, central Kenya long rains 1998. Manures were applied
to supply 80 kg N ha -1
Treatment
Grain yields t ha -1
Control (no soil amendments)
100 kg N ha-1
40 kg N ha-1
20 kg N ha-1
Manure 1 (F+FR manual mix)
Manure 2 (F+FR animal mix)
Manure 1 + 20 kg N ha -1
Manure 1 + 40 kg N ha -1
Manure 2 + 20 kg N ha -1
Manure 2 + 40 kg N ha -1
2.53
7.51
5.36
5.09
4.82
5.08
5.64
6.01
5.75
6.33
Lsd 0.05
1.03
Source: Kimani et al., 2001. F, faeces alone; FR, feed refusals, either manually mixed or
mixed by the animal in the stable, and composted for 4 months
Conclusions
This review shows that managing manures, through animal feed sources,
or by composting can enhance fertiliser value. Quality parameters, based
on laboratory measurements are available, although there is dire need
to relate them to on-farm management systems. This would lead to
production of simple extension manuals that relate management factors
with quality, on-farm. Manure placement method will influence its
effectiveness as a fertiliser. Hill placement produces a higher yield, at
least during the first season of application. Recent studies indicate that
in small farms (less than 1 ha), there is possibly excess manure, which
196
Kimani, S.K. and Lekasi, J.K.
is an opportunity to increase the financial situation of farmers through
off-farm sales for the better-managed manures. There is need to monitor
effects of excess nutrients with regard to environmental conservation.
Acknowledgements
We wish to thank the Director KARI, Muguga for facilitating this work.
We wish to thank the Rockefeller Foundation for financial support for
the referred Kenyan studies. We also thank the technical staff at KARI
Muguga for the excellent field and laboratory work.
References
Delve, R.J., Tanner, J.C., Kimani, S.K., Giller, K.E., Cadisch G. and Thorne,
P.J. (1999) Effects of dietary nitrogen source fed to steers on faecal nitrogen,
its mineralisation, and the vegetative responses in maize to the incorporation
of the faeces in soil. 1999. In: CIMMYT and EARO 1999. Maize production
Technologies for the future: Challenges and Opportunities. Proceedings of
the 6th Regional maize conference for Eastern and Southern Africa. Addis
Ababa, Ethiopia September 21-25, 1998. pp. 282-285.
Heisey, P.W. and Mwangi, W.M. (1996) Fertiliser use and maize production in
sub-Saharan Africa. Economics Working Paper 96-10. CIMMYT, Mexica.
Kapkiyai J., Woomer, P. Qureshi, J., Smithson, P. and Karanja N. (1996) Effects
of Fertiliser and Organic Inputs on Soil Organic Matter and Nutrient
Dynamics in Kenyan Nitisol. Paper presented to the International Symposium
“Carbon and Nutrient Dynamics in Natural and Agricultural Tropical
Ecosystems”, Harare, Zimbabwe, 29 April-4 May 1996.
Kihanda, F.M. and Gichuru, M. (1999) Manure management for soil fertility
improvement. TSBF/AHI Report.
Kimani, S.K. (1999) Integrated use and effects of manures with modest
application of inorganic fertiliser in maize production on central Kenya.
Progress Report No. 3 to the Rockefeller Foundation, Nairobi.
Kimani S.K., Odera M.M and Musembi, F. (2000) Factors influencing adoption
of integrated use of manures and fertilisers in central Kenya. Proceedings
of KARI Scientific Conference. November, 2000.
Kimani SK, Mangale, N. Gichuru, M., Palm, C.and Wamuongo, J. (2001)
Integrated Use and Effects of Manures with modest Application of Inorganic
Fertiliser on soil properties and maize production in the central Kenya
Highlands. Final Technical Report to the Rockefeller Foundation. May 2001.
Klapp, E. (1967) Lehbuch des Acker- und Pflanzenbaus. 6. Aufl. Parey, Berlin.
603 pp. As cited in Karl et al., (1994). pp. 381-437.
Managing Manures Throughout Their Production Cycle Enhances Their Usefulness as
Fertilisers: A Review
197
Lekasi, J.K., Tanner, J.C., Kimani, S.K. and Harris P.J.C. (1998) Manure
management in the Kenya highlands: Practices and Potential. Emmerson
Press, Kenilworth, UK.
Lekasi, J.K. (2000) Manure management in the Kenya highlands: collection,
storage and composting strategies to enhance fertiliser quality. PhD Thesis.
Coventry University, UK
Makokha, S, Kimani, S.K., Mwangi, W., Verkuijl H. and Musembi, F. (2001)
Determinants of fertiliser and manure use in maize production in Kiambu
district, Kenya. Mexico, D.F.: International Maize and Wheat Improvement
Centre (CIMMYT) and Kenya Agricultural Research Institute (KARI).
Nzuma, J.K., Murwira, H.K., and Mpepereki,J. (1998) Cattle manure
management options for reducing nutrient losses. Farmer perception and
solutions in Magwanda, Zimbabwe. In: Waddington, S.R., Murwira, H.K.,
Kumwenda, J.P.T., Hikwa, D. and Tagwira, F. (eds). The soil fertility research
for maize-based systems in Malawi and Zimbabwe. CIMMYT. Pp. 183-190.
Odongo, N. (1999) Effects of Rock P fed to cattle as supplement and the resultant
manure quality. Unpublished PhD study. University of Guelph, Canada.
Simpson, K. (1986) Manures. In: Fertilisers and Manures. Longman, London
pp. 83-108.
Swift, M.J., Seward, P.G.H., Frost, J.N., Qureshi, J. and Muchena, F.N. (1994)
Long-term experiments in Africa: developing a database for sustainable land
use under global change. p. 229-251. In: R.A. Leigh and A.E. Johnson (eds.).
Long-term Experiments in Agricultural and Ecological Sciences. CAB
International, Wallingford, UK.
Wanjekeche, E., Mwangi, T., Powon, P. and Khaemba, J. (1999) Management
practices and their effects on nutrient status of farmyard manure in West
Pokot district, Kenya. Paper presented at the 17th Conference of the Soil
Science Society of East Africa, Kampala, Uganda. 6th -10th August 1999.
198
Kimani, S.K. and Lekasi, J.K.
Simulated Partitioning Coefficients for Manure Quality Compared with Measured C:N
Ratio Effects
Simulated Partitioning
Coefficients for Manure
Quality Compared With
Measured C:N Ratio Effects
199
14
Kimani, S.K.1, Gachengo, C.2 and
Delve, R.2
1
Kenya Agricultural Research Institute, Muguga,
P.O. Box 30148, Nairobi, Kenya
2
Tropical Soil Biology and Fertility Institute of CIAT,
P.O. Box 30677, Nairobi, Kenya
Abstract
Livestock manures comprise an important source of
nutrients in many farming systems. However, the quality
of manure is generally variable depending on composting,
storage and handling. This in turn results in different
manure responses with regard to the release of nutrients
for crop uptake. Recent studies have shown that there is a
wide scope to better manage manures to improve quality,
and the field performance. Simulation modelling, for
example using APSIM, is a useful tool in exploring improved
strategies for management of manures, with regard to
200
Kimani, S.K. et al
enhancing fertiliser quality. This paper discusses some
concepts behind APSIM Soil-N module and attempts to
simulate responses for different quality manures. The
manures were collected from farmer’s fields in central Kenya
and compared with a feedlot-managed manure. The C:N
ratios of these 9 manures ranged from 13 to 32. Fresh
organic matter (FOM) in manures was partitioned into
different pools of carbohydrate, cellulose and lignin
fractions. Simulations on net N mineralisation for C:N ratios
of 10, 15 and 25, using APSIM were done. The measured
responses were compared for manures with different C:N
ratios of these manures, against net N mineralisation as
determined in a laboratory incubation study. Whilst the
measured responses showed an initial nitrogen
immobilisation, the simulated responses showed net N
mineralisation from time 0 up to 100 days, for manures
with a C:N ratio of 13-22. Manures with C:N ratio above 30
showed net immobilisation throughout the experimental
period for the simulated responses.
Introduction
The use of inorganic fertilisers can overcome most of the soil fertility
decline in Kenya but the use of inorganic fertilisers in smallholder
farms is associated with several constraints. A survey in Kiambu
district, Central Kenya listed high cost of fertiliser and unavailability
as major constraints to fertiliser use in maize production (Makokha et
al., 2001). An alternative to inorganic fertilisers is the utilisation of
farm-derived sources of crop nutrients such as crop residues, composts
and farmyard manure. In most smallholder farms of central Kenya
cattle manure is the most widely used organic fertiliser, at
approximately 80% of the households (Makokha et al., 2001). This
arises from the higher livestock populations in these areas. About 90%
of farmers use manure from their farms, while 10% either purchase or
are given free. However, in majority of farms, the manure is not enough
to fertilise the farms (Makokha et al., 2001; Kihanda and Gichuru,
2000; Kagwanja, 1996). These manures are usually of poor quality
and are particularly low in total N, with most having less than 1% N .
In comparison, legume cover crops can have over 3% N (Palm et al.,
2000). Animal manures are of major importance in nutrient cycling
but generally of poor quality to supply plant nutrients (Giller et al.,
1997). In addition other factors, for example, improvement in soil
physical conditions such as improved moisture retention and addition
Simulated Partitioning Coefficients for Manure Quality Compared with Measured C:N
Ratio Effects
201
of nutrients, in addition to N, play an important role in farming systems
(Kihanda and Gichuru, 2000). Simulation models can be useful in
predicting, designing and evaluating effects of applied treatments.
However, quality data is required in order for simulation models to be
validated. This work was an attempt to use Agricultural Systems
Simulation Model (APSIM) for defining manure quality, with regard to
N release.
This study compared simulated partitioning coefficients for
manure quality compared with measured C:N ratio effects, using
APSIM model
Simulation modelling
The residue module of APSIM has provision for surface residues to be
incorporated into the soil, or to decompose in situ. This enabled the
simulation of the laboratory incubation experiment. Manures vary in
composition, being a complex mixture of animal excreta and plant
residues that has undergone varying degrees of composting/
decomposition and may be mixed with varying amount of soil. Some of
the nutrients contained in manure will be in forms that are immediately
available for crop uptake, and some will undergo further decomposition
before they become available. This concept of nutrient availability has a
time dimension, therefore nutrients in manure show a wide range of
availabilities.
In this study, the manure fresh organic matter pools were
fractionated into three components, carbohydrate-like, cellulose-like
and lignin-like fractions. These components depicted nutrient
compositions ranging from those that are water soluble to very
recalcitrant. The APSIM residue module fractionates the manure pools
into percentages of 20:70:10 respectively for carbohydrate, cellulose
and lignin.
For our study, low quality manure was fractionated at 0:1:99
respectively for carbohydrate, cellulose and lignin. Improved quality
manure was fractionated at 0:50:50, and best quality manure had
33:33:34 respectively for carbohydrate, cellulose and lignin. These three
partitioning coefficients were used as a scenario to simulate net N
mineralisation at C:N ratios of 10, 15 and 25. The model showed a
higher sensitivity to C:N ratio of organic inputs, compared with the
fractionated components (Figures 14.1 and 14.2). The model also
simulated immobilisation of N for C:N ratio of 25 up to 100 days of
incubation. However, the model did not simulate the initial mineral N
immobilisation, commonly observed in incubation experiments
(Figure 14.3).
202
Kimani, S.K. et al
Figure 14.1: Partitioning coefficient effect on simulated net mineral N (Treatment –
Unfertilised control). Part 1 (0:1:99), Part 2 (0:50:50), Part 3 (33:33:34), at CN ratios of
10 (CN10), 15 (CN15) and 25 (CN25).
Simulated Partitioning Coefficients for Manure Quality Compared with Measured C:N
Ratio Effects
203
Figure 14.2: Simulated net mineral N (Treatment-unfertilised control) for partition 2
(0:50:50), at CN ratios of 10, 15 and 25
Figure 14.3: Measured responses for N release in the Muguga incubation experiment
204
Kimani, S.K. et al
Figure 14.4: Simulated net nitrate N (Treatment – Unfertilised control) for the Manure
incubation experiment at Muguga
APSIM simulation of an incubation experiment conducted
at Muguga
Nine manures were selected for this study. Their characteristics are
briefly described in Table 14.1. Carbon and Nitrogen were measured at
various extraction times during the incubation using methods described
by Anderson and Ingram (1993). The simulated responses showed net
N mineralisation from time 0 up to 100 days, for manures with a C:N
ratio of 13-22. Manures with C:N ratio above 30 showed net
immobilisation throughout the experimental period for the simulated
responses.
Table 14.1: Manures used in the simulation of the Muguga incubation study
Manure
M1
M2
M3
M4
M5
M6
M7
M8
M9
Description
%N
%C
C:N ratio
on-station
on-farm
on-farm
on-farm
on-farm
on-farm
on-farm
on-farm
on-farm
1.7
1.1
1.3
1.5
1.2
1.9
1.6
0.9
1.8
34.6
32.7
28.4
29.1
17.7
24.9
30.3
28.5
30.6
20
30
22
19
15
13
19
32
17
% N mineralisation
Simulated Partitioning Coefficients for Manure Quality Compared with Measured C:N
Ratio Effects
205
Manures with the high C:N ratio immobilised N throughout the 100
days of incubation (Figure 14.4). There was again no initial N
immobilisation for these runs, suggesting the need to make some
improvements on the model parameters.
Conclusions
APSIM is a model that has been widely used, for instance to analyse
management options and help improve farmers’ and scientists’
understanding of the soil-crop system (McCown et al., 1998). It is likely
that the wider application of APSIM has been for field data, for example
in New South Wales, Australia (Turpin et al., 1998), and other areas in
Australia (Probert et al., 1995; Probert et al., 1998). For these laboratory
experiments, whilst the measured responses showed an initial nitrogen
immobilisation, the simulated responses showed net N mineralisation
from time 0 up to 100 days, for manures with a C:N ratio of 13-22.
Manures with C:N ratio above 30 showed net immobilisation throughout
the experimental period for the simulated responses. This probably
suggests the need to improve the model to show some initial N
immobilisation.
Acknowledgements
We wish to thank the Director KARI, Muguga for facilitating this work.
We wish to thank the Rockefeller Foundation for financial support for
this study through the grant RF 97 002 131. We also thank the technical
staff at TSBF laboratory at ICRAF for the excellent lab incubation work.
References
Anderson, J.M. and Ingram, J.S.I. (1993) Tropical Soil Biology and Fertility. A
Handbook of Methods. C.A.B. International. Wallingford, U.K. Bull. No.9.
The Sulphur Institute, Washington, U.S.A.
Giller, K.E., Cadisch, G., Ehaliotis, C., Adams, E., Sakala, W.D., and Mafongoya,
P.L. (1997) Building soil nitrogen capital in Africa, In: Buresh, R.J., et al.,
(Eds.). Replenishing Soil Fertility in Africa. SSSA Special Publication No.
51. SSSA, Madison, WI, pp. 151-192.
Kagwanja, J.C. (1996) Determinants of Farm Level Soil Conservation Technology
Adoption: Lessons from the High rainfall, Highly Populated, Steep Sloped
Mt. Kenya Highlands, Embu, Kenya. PhD Thesis, University of Missouri,
Columbia.
206
Kimani, S.K. et al
Kihanda, F.M. and M. Gichuru. (2000) Manure management in East and
Southern Africa: A synthesis report. AHI/TSBF.
Makokha, S., S. Kimani, W. Mwangi. H. Verkuijl, and F. Musembi. (2001)
Determinants of Fertilizer and Manure Use in Maize Production in Kiambu
District, Kenya. Mexico, D.F.: International Maize and Wheat Improvement
Centre (CIMMYT) and Kenya Agricultural Research Institute (KARI).
McCown, R.L., Hammer, G.L., Hargreaves, J.N.G., Holzworth, D.P., and
Freebairn, D.M. (1996) APSIM: a novel software system for model
development, model testing and simulation in agricultural systems research.
Agric. Systems 50:255-271.
McCown, R.L., Carberry, P.S., Foale, M.A., Hochman, Z., Coutts, J.A. and
Dalgeish, N.P. (1998) The FARMSCAPE approach to farming systems
research. In: Michalk, D.L. and Prately, J.E. (eds.) Agronomy – Growing a
greener future. Proceedings of the 9th Australian Agronomy Conference.
Charles Sturt University, Wagga Wagga, pp. 633-666.
Palm, C.A., Giller, K.E., Mafongoya, P.L., and Swift, M.J. (2000). Management
of organic matter in the tropics: translating theory into practice Nutr. Cyc.
Agroecosystems. In Press:
Probert, M.E., Keating, B.A., Thomson, J.P. and Parton, W.J. (1995) Modelling
water, nitrogen and crop yield for a long-term fallow management experiment.
Australian Journal of Experimental Agriculture 348:941-950.
Probert, M.E., Carberry, P.S., McCown, R.L. and Turpin, J.E. (1998) Simulation
of legume-cereal systems using APSIM. Australian Journal of Agricultural
Research 49:317-327.
Turpin, J.E., Hayman, P.T., Marcellos, H. and Freebairn, D.F. (1998) Nitrogen
fertiliser decisions for wheat on the Liverpool Plains NSW 1. Should farmers
consider paddock history and soil tests? In: Michalk, D.L. and Prately, J.E.
(eds.) Agronomy – Growing a greener future. Proceedings of the 9th Australian
Agronomy Conference. Charles Sturt University, Wagga Wagga, pp. 761764.
Nitrogen Fertilizer Equivalency Values for Different Organic Materials Based on Maize
Performance at Kabete, Kenya
Nitrogen Fertilizer
Equivalency Values for
Different Organic Materials
Based on Maize Performance
at Kabete, Kenya
207
15
Kimetu, J.M.1, Mugendi, D.N.2, Palm,
C.A.1, Mutuo, P.K.1, Gachengo, C.N.1,
Nandwa, S.3 and Kungu, J.B.2
1
Tropical Soil Biology and Fertility Institute of CIAT,
P.O. Box 30677, Nairobi, Kenya
2
Department of Environmental Foundations, Kenyatta
University, P.O. Box 43844, Nairobi, Kenya
3
National Agricultural Research Laboratories (NARL),
P.O. Box 14733, Nairobi, Kenya
Abstract
Decline in crop yields has been a major problem facing small
holder farming in Kenya and the entire sub-Saharan region.
This is attributed mainly to the mining of macronutrients
due to cropping without external addition of adequate
nutrients. Inorganic fertilizers are expensive hence
unaffordable by most small holder farmers. Although
organic nutrient sources are available, information about
the right proportions of application is scanty.
An experiment was set up in 1999 at the National
Kimetu, J.M. et al
208
Agricultural Research Laboratories (NARL) at Kabete, with
the overall objective of determining nitrogen fertilizer
equivalencies based on high quality organic inputs. The
specific objectives of the study included determination of
the nitrogen fertilizer equivalency values of Tithonia
diversifolia, Senna spectabilis and Calliandra calothyrsus
and the investigation of nitrogen use efficiency from
combined organic and inorganic inputs.
The experiment consisted of maize plots to which freshly
collected leaves of Tithonia diversifolia (tithonia), Senna
spectabilis (senna) and Calliandra calothyrsus (calliandra)
(all with % N >3) obtained from hedgerows grown ex situ
(biomass transfer from outside) and urea (inorganic nitrogen
source) were applied. Results obtained indicated that a
combination of both organic and inorganic nutrient sources
gave higher maize grain yield than when each is applied
separately, except for tithonia whose sole application gave
better grain yield than a combination of the same with
mineral fertilizer. Maize grain yield production after organic
and inorganic application was in the order of tithonia >
tithonia+urea = calliandra+urea > urea > senna+urea >
calliandra > senna > control. The percentage N recovery
was highest in sole application of urea followed by a
combination of both urea and tithonia while sole application
of tithonia biomass had relatively lower percentage N
recoveries. In both seasons, the mineral N content was high
in sole application of tithonia than in senna and calliandra
treatments. The three organic materials (senna, calliandra
and tithonia) gave fertilizer equivalency values of 68%, 72%
and 119% respectively.
Key words: N fertilizer equivalency, mineral-N, N-recovery
Introduction
Decline in soil fertility is an acute problem facing small holder farming
in Kenya. Due to the high cost and uncertain availability of inorganic
fertilizers, it is important to provide alternative sources of nutrients
such as organic materials. In the recent past there has been increased
interest in the use of leaf biomass from woody perennials as a source of
nutrients to annual crops (Kang et al., 1990; Palm et al., 1997; Mugendi
et al., 1999). The big challenge to this approach is ensuring that crops
efficiently utilize nutrients from the applied organic materials.
Synchronizing release of nutrients from decomposing biomass with crop
Nitrogen Fertilizer Equivalency Values for Different Organic Materials Based on Maize
Performance at Kabete, Kenya
209
demand could lead to increased nutrient-use efficiency (Becker et al.,
1994; Mwale et al., 2000b), and this in turn could minimize nutrient
loss (Swift, 1987; Myers et al., 1994; Mugendi et al., 1999). The use of
organic materials of differing quality in combination with inorganic
fertilizers to optimize nutrient availability to annual crops is still a
challenge to scientists currently.
Much research has been done to determine the use of organic plant
materials as a source of nutrients in place of inorganic fertilizers and
most of this research has revealed both advantages and disadvantages
of combining nutrient sources (Palm et al., 1997). However, little
predictive understanding for the management of organic inputs especially
in tropical agroecosystems is available (Palm et al., 2001). It has been
therefore difficult to give valid advice to farmers on the best organic N
source for direct application and the right combinations with inorganic
N source.
Although organic N sources have the potential to supply large
quantities of N required by growing crops, to obtain maximum production
and for more sustainability, they should be supplemented with inorganic
fertilizers (Mugendi, 1997; Jama et al., 2000; Vanlauwe et al., 2001).
The combination of inorganic N fertilizer with organic N sources is said
to increase the rate of decomposition and mineralization (Mugendi et
al., 1999) of low quality materials. This coupled with the right time of
application can improve synchrony of the N released from the
decomposing biomass and nutrient requirement by annual crop, thereby,
reducing N losses. This postulation is however, yet to be ascertained.
This research was therefore aimed at shedding light on the combined
use of organic (tithonia, senna and calliandra) and inorganic N sources
for farmers in the central region of Kenya. In addition, the study will
provide information to link the fertilizer equivalency of organic materials
(specific amount of an organic material that can have same effect on
crop yield as a certain amount of inorganic fertilizer) with the resource
quality as well as investigating the influence of N source on N uptake by
maize.
Materials and Methods
Site description
The experiment was carried out at the National Agricultural Research
Laboratories (NARL), Kabete, Kenya. The station is located at 36°46'E and
01°15'S and an altitude of 1650 m above sea level. The soils are mainly
Humic Nitisols (FAO, 1990) that are deep and well weathered, and with the
following chemical characteristics: pH =5.4; total N = 1.35g kg-1; extractable
P = 27mg kg-1; carbon = 1.6%; exchangeable Ca, Mg, and K (cmol kg-1) of
Kimetu, J.M. et al
210
5.8, 1.7, and 0.7 respectively; clay = 40%; sand = 23%; and silt = 37%. The
mean annual rainfall is about 950 mm received in two distinct rainy seasons;
the long rains (LR) received mid March to June and the short rains (SR)
received mid October to December. The rainfall amount during the study
period is shown in figure 15.5. The average monthly maximum and
minimum temperature is 23.8°C and 12.6°C respectively.
Experimental design and treatments
The experiment was designed and established by Tropical Soil Biology
and Fertility (TSBF) Programme in 1999 with the aim of determining
fertilizer equivalency values based on high quality organic materials.
The experiment was a completely randomised block design (CRBD) with
10 treatments replicated 4 times. The plot size was 5.25 m by 5 m with
an interplot spacing of 0.75 m. Urea and freshly collected leaves of
tithonia, senna and calliandra (Table 15.1) were applied directly into
the plots after which maize was planted. Collection of the organic
materials was done by hand at the same location for both seasons. The
leaves included the petioles, and in the case of senna and calliandra,
they also included the rachis since these two have compound leaves.
The calculation of the application amount of organic materials (that
would give 60 kg N ha-1)(Table 15.2) was done on dry matter basis giving
1.3, 1.8 and 1.9 t ha -1 for tithonia, senna and calliandra respectively.
The percentages (%) indicated in Table 15.2 refer to the specific amount
of N that was applied as different treatments in kg ha -1.
Table 15.1: Chemical properties of plant materials used at NARL, Kabete, Kenya
Season 1 (1999 short rains)
Sample
%N
%P
%K
%Ca
%Mg
%PP
%Lignin
Tithonia
Senna
Calliandra
4.7
3.7
3.2
0.5
0.2
0.1
5.1
2.0
1.0
3.0
0.9
1.1
0.2
0.2
0.3
2.5
3.4
9.9
5.2
10.7
14.4
sed
0.2
0.03
0.3
0.3
0.1
1.0
1.8
Season 2 (2000 long rains)
Tithonia
Senna
Calliandra
4.0
3.1
2.4
0.4
0.1
0.1
5.5
1.8
0.7
2.2
0.9
1.1
0.4
0.2
0.3
1.9
1.8
12.4
9.3
10.9
17.5
sed
0.2
0.03
0.3
0.3
0.1
1.0
1.8
Abbreviations: PP= Polyphenols
sed = Standard error of differences
Nitrogen Fertilizer Equivalency Values for Different Organic Materials Based on Maize
Performance at Kabete, Kenya
211
Table 15.2: Experimental Treatments at NARL Kabete, Kenya.
Trt.
Inorganic N
(kg ha-1)
Organic N
(kg ha-1)
1. * (Control)
0
0
2.
30
30 (50% tithonia)
3.
4. *
0
60
60 (100% tithonia)
0
5.
6.
30
0
30 (50% senna)
60 (100% senna)
7.
30
30 (50% calliandra)
8.
9. *
0
35
60 (100% calliandra)
0
10. *
100
0
* These are the treatments that were used in plotting of the response curve, which was
used in the calculation of the nitrogen fertilizer equivalency values of the three organic
inputs.
Selection of tithonia, senna and calliandra as the organic N sources
was based on their contrasting qualities with respect to polyphenols
and rates of decomposition (Gachengo et al., 1999; Mutuo et al., 1999).
The chemical characteristics of these three organic inputs used for the
two cropping seasons are shown in Table 15.1.
Sampling and analyses
Plant samples were oven-dried at 35°C for 48 hours then ground to
pass through a 1.0 mm sieve and analyzed for total N, P, K, Ca, and Mg
by Kjeldahl digestion with concentrated sulfuric acid (Anderson and
Ingram, 1993; ICRAF, 1995). Nitrogen and phosphorus were determined
colorimetrically (Parkinson and Allen, 1975) while potassium was by
flame photometry (Anderson and Ingram, 1993). Magnesium and calcium
was by atomic absorption spectrophotometer at wavelength of 2852
and 4227 respectively. Determination of lignin was done using the acid
detergent fiber (ADF) method as described by Van Soest (1963). Total
soluble polyphenols were analyzed by extraction using 50% aqueous
methanol (Anderson and Ingram, 1993). The plant material to extractant
ratio was 0.1 g / 50 ml and phenols were analyzed colorimetrically
using the Folin-Ciocalteu reagent as described by Constantinides and
Fownes (1994).
212
Kimetu, J.M. et al
Data Analysis
Fertilizer equivalency value
Fertilizer equivalencies (FE) of organic materials were obtained by
comparing the yield from the organic material treatments to that of the
nitrogen (N) response curve from inorganic N fertilizer (Mutuo et al.,
1999). Calculation for the corresponding N fertilizer equivalent for an
organic material was obtained from the quadratic equation (Y = aFE2 +
bFE + c) exhibited by the N response curves. The following formula for
solving quadratic equations was used:
FE = -b ± b2 - 4ac)
2a
Where a, b, and c are constants, with values -0.0001, 0.0252, and 1.8297
respectively for 1999 short rains (Figure 15.1) and -0.0001, 0.0284,
and 2.0827 respectively for 2000 long rains (Figure 15.2).
Figure 15.1: Season 1 biomass yield response to levels of N at NARL – Kabete,
Kenya, 1999
y = -0.0001x2 + 0.0252x + 1.8297
R2 = 0.9704
N level (kg N ha -1)
Nitrogen Fertilizer Equivalency Values for Different Organic Materials Based on Maize
Performance at Kabete, Kenya
213
Figure 15.2: Season 2 grain yield response to N levels at NARL – Kabete, Kenya, 2000
N level (kg N ha -1)
In order to compare the fertilizer equivalencies of organic materials,
the fertilizer equivalency (FE) % values were calculated as follows:
%FE =
FE
x 100
N applied
Where: N applied = actual amount of N applied (100% organic/inorganic).
Source: Mutuo et al. (1999).
Maize yields
To compare treatment effects on maize grain yield, yields were converted
to relative increase compared to the control:
Yield increase (%) = (Yield treatment – Yield control )
x 100
Yield control
(Source: Gachengo et al., 1999)
Nitrogen uptake
Nitrogen uptake by the maize crop was determined by multiplying the
grain, stover and core yields with the nitrogen concentration in the
specific components. Nitrogen recovery was determined as shown below:
214
Kimetu, J.M. et al
Nitrogen recovery (%) = (N uptake
treatment
– N uptake
control )
Amount of nitrogen applied
x 100
Statistical comparisons
Treatment effects on soil N availability and maize yield were analyzed
using Genstat 5 for windows (Release 4.1) computer package. Treatment
means found to be significantly different from each other were separated
by Least Significant Differences (LSD) at P < 0.05.
Results and Discussions
Nitrogen fertilizer equivalencies of tithonia, calliandra and
senna
The study sought to attain this by investigating the performance of maize
crop supplied with green leaves from the organics as compared to maize
grown with urea as N source. Due to poor rainfall distribution during the
1999 short rains, maize crop was harvested six weeks before maturity
hence no grain yields were obtained. Biomass yield data was therefore
used in calculating the fertilizer equivalencies for the organics. Results
obtained showed that maize biomass yields were 3.3, 3.6 and 3.9 t ha-1
for 60 kg N ha-1 of calliandra, tithonia and senna treatments respectively.
As shown in Figure 15.1, these yields were higher than the biomass yields
from any of the inorganic N source treatments whose highest yield was
only 3.0 t ha-1. Thus, the values for the yields obtained from the three
organic materials fell high above the response curve. Hence, the fertilizer
equivalencies for the organic materials could not be estimated from the N
response curve. These differences in the yields obtained from the organic
and inorganic N sources could be attributed to the poor rainfall distribution
during that growing season and the timing of the N application. Much of
the rainfall was received late November, 1999 and early December, 1999
and scarcely any rainfall in January, 2000 which was the tussling stage
for the maize. Also, the fact that all the 60 kg N from the organics was
applied when there was moisture (at planting) unlike for urea which was
applied in split (20 kg N at planting and 40 kg N applied after five weeks)
could also be a partial explanation for the better performance of the maize
crop supplied with the organics. This is mainly because the application
of the second split of the urea was followed by a dry spell. Hence, the
growing maize crop might not have utilized this portion of the urea, thus
leading to the low maize biomass yields from urea treatment. The relatively
high biomass yield from organic treatments could also be due to other
Nitrogen Fertilizer Equivalency Values for Different Organic Materials Based on Maize
Performance at Kabete, Kenya
215
positive effects of the organic materials on soil physical properties (like
moisture retention) and chemical properties (other micronutrients like
calcium and magnesium) (Chen and Avnimelech, 1986; Wallace, 1996;
Mutuo et al., 1999).
A better maize performance was observed during the 2000 long rains
season. Maize grain yields from the organic treatments were 3.6, 3.1 and
3.0 t ha-1 for 60 kg N ha -1 tithonia, calliandra and senna respectively
compared to the highest yields from urea treatments of about 3.6 t ha-1
(Figure 15.2). This gave fertilizer equivalency values of 119%, 72% and
68% for tithonia, calliandra and senna. The implication was that tithonia
biomass performed better than an equivalent amount of inorganic fertilizer
in improving maize grain yield while calliandra and senna performed
relatively lower to an equivalent amount of inorganic N source. The high
fertilizer equivalency value for tithonia compared to the other two organic
materials (senna and calliandra) could be attributed to its low polyphenol
content compared to senna and calliandra. Hence, decomposition rate
and subsequent N release is higher in tithonia green biomass (Gachengo
et al., 1999) as compared to senna and calliandra (Lehmann et al., 1995).
The N content in the material also influence decomposition and N release
as Mutuo et al. (1999) noted in different sites in East and Southern Africa.
The conclusion was that fertilizer equivalency value of organic materials
is proportional to the N content. However, from the results we obtained,
the fertilizer equivalency values for senna and calliandra (68% and 72%
respectively) did not differ significantly despite the 3.1 and 2.4% N content
in the two organic materials. This could be an indication of more
conspicuous residual effect (from season one) in the calliandra treatment
than in senna treatment.
Fertilizer equivalency values for tithonia and calliandra were almost
twice the values reported by Mutuo et al. (1999) for the same organic
materials in their trial in Western Kenya. This could be attributed to the
difference in the climatic conditions. Western Kenya received adequate
rains, while the Central region (Kabete trial site) was characterized by
poor rainfall distribution during the two seasons when this research
was carried out.
As per the study findings, tithonia green biomass can be
recommended for direct application while senna green biomass can be
applied in combination with inorganic fertilizer. Calliandra leaf biomass
on the other hand may not be recommended for direct application due
to the high polyphenol content (11.1%) as compared to the suggested
critical level of 4.0% (Palm et al., 1997; Palm et al., 2001) and also
because of it’s low nitrogen content (2.4% N). Therefore, as suggested in
the organic matter management decision tree (Delve et al., 2000; Palm
et al., 2001), calliandra leaf biomass may give better results when mixed
with inorganic N fertilizer.
216
Kimetu, J.M. et al
Maize performance as influenced by the N source
Maize yields were dependent on the N source (Figure 15.3 and Figure
15.4). During the 1999 short rains, biomass yield obtained from a
combination of either of the three organic inputs with inorganic N
source differed significantly from biomass yield obtained from sole
application of inorganic N source (Figure 15.3). It was also found that
maize biomass yield obtained from tithonia + urea treatment was
significantly higher compared to maize biomass yield obtained from
sole tithonia and sole urea treatments. Approximately twice as much
maize biomass yield was obtained with combination of tithonia and
urea as compared to urea applied alone. This could be an indication of
better results in combining organic and inorganic N source, which
could be attributed to better synchrony of nutrient availability to maize
crop demand. Separate application of either tithonia or urea did not
show significant differences.
Sole application of senna green biomass and a combination of the
same with urea had significantly higher maize biomass yield than urea
applied separately. Calliandra, calliandra + urea and sole urea treatments
did not show any significant differences from each other. It was also
found out that the control gave significantly lower maize biomass yield
compared to all the other treatments.
Figure 15.3: The effect of combining organic-inorganic N sources on maize biomass
yield during 1999 short rains at NARL – Kabete, Kenya
Nitrogen Fertilizer Equivalency Values for Different Organic Materials Based on Maize
Performance at Kabete, Kenya
217
Figure 15.4: The effect of combining organic-inorganic N sources on maize grain yield
during 2000 long rains at NARL – Kabete, Kenya
Figure 15.5: Rainfall data (1999-2000) at Kabete, Kenya
218
Kimetu, J.M. et al
Similar results were reported by Jama et al. (2000) who observed
higher maize yields obtained from a combination of tithonia and
phosphorus fertilizer in their work in western Kenya. Other researchers
have observed greater maize production through application of highquality organic inputs like tithonia in combination with inorganic
fertilizer as compared to sole application of mineral fertilizers (Gachengo,
1996; Palm et al., 1997).
Results obtained during the 2000 long rains season revealed that
all other treatments had significant increase on maize grain yield above
the control (Figure 15.4). Tithonia green manure increased maize grain
yield by about 71.4% while calliandra and senna increased grain yield
by 48% and 43% respectively. A percentage grain yield increase of about
52% above control was realized from sole urea treatment. Maize grain
yields from combined use of organic-inorganic N sources were dependent
on the organic material used. Although there was significantly higher
maize grain yield from tithonia green biomass as compared to senna
and calliandra, grain yield obtained from sole application of any of the
organic materials and a combination with mineral fertilizer did not
significantly differ from each other.
The relatively better results realized from tithonia sole application
than a combination with mineral N source and sole application of urea
(though not significantly different) could still be attributed to other
indirect effects to the soil such as moisture retention (Wallace, 1996;
Lehmann et al., 1999) and addition of other macro- and micronutrients.
Nziguheba (2001), also reported increase in maize growth with
application of tithonia green biomass which (in addition to increased N
availability), was attributed to increased labile P as compared to inorganic
inputs.
As noted also during 1999 short rains, time of application might
have as well played a major role in maize performance. All the tithonia
green biomass was applied at once at planting when there was rain
while only one third of the urea was applied at planting. Two thirds of
the urea was applied five weeks later, which was followed by a dry spell,
hence, insufficient amounts of the urea N were available to the growing
crop.
Lower maize grain yields obtained from sole application of either
calliandra or senna, could be attributed to N immobilization or reduced
N release as Mwale et al . (2000b) also noted in their study at
Chalimbana, Zambia. Other researchers also observed that, large
portion of N from a slowly decomposing biomass may be incorporated
into soil organic matter fractions (Lehmann et al., 1999) or immobilized
into forms not readily available to annual crops (Mugendi et al., 1999).
Therefore tithonia green biomass can be recommended for direct
incorporation for soil fertility improvement (Delve et al., 2000; Palm et
al., 2001).
Nitrogen Fertilizer Equivalency Values for Different Organic Materials Based on Maize
Performance at Kabete, Kenya
219
Nitrogen uptake and total %N recovery by maize
The results revealed that, nitrogen concentration in the grain, stover
and core yields differed significantly among N sources (Table 15.3).
Nitrogen uptake ranged from 93.3 to 131.9 kg ha -1. From the study
findings, it was noted that the inorganic fertilizer (urea) applied treatment
gave the highest N uptake while control had the lowest. Tithonia + urea
and urea treatments were significantly higher than the control. Above
ground yield from urea sole application had about 131.9 kg ha-1 total N
uptake while tithonia + urea gave 114.3 kg ha-1. This relatively high N
uptake from the two treatments could be attributed to the readily
available N from the urea. The N uptake by maize that received tithonia
green biomass alone as nitrogen source was about 97.6 kg ha-1, which
was not significantly different from the control.
Table 15.3: Nitrogen recovery by maize crop in long rains 2000 at NARL, Kabete, Kenya
Treatment
Control
Tithonia + Urea
Tithonia
Urea
sed
N applied
(kg N ha-1)
Nitrogen uptake
(kg N ha-1)
Grain
%N
Stover
Total % N
recovery
0
60
60
60
-
93.3
114.3
97.6
131.9
16.4
1.7
1.9
1.8
2.0
-
0.63
0.9
0.8
1.1
-
N/A
35
7.2
64.3
N/A
The apparent percentage N recovery by maize crop that received sole
tithonia green biomass was found to be 7.2% while 35% was recovered in
tithonia + urea treatment. Sole urea treatment had a 64.3% nitrogen
recovery. However, these values might not have reflected the actual N
recoveries by the maize. This is because the material used was not labelled
hence it was not possible to follow up the applied nitrogen (either in
organic or inorganic form). Therefore, calculated total % N recovery values
obtained in the study were meant to be estimates to the actual recoveries.
Nitrogen recovery by the maize crop that received sole application of
urea and the one that received a combination (inorganic-organic N
source), was significantly higher compared to nitrogen recovered by maize
that received sole tithonia green biomass. The high N recoveries by maize
crop planted in sole urea and tithonia + urea applications were an
indication that there was less N loss from soil-plant system. Therefore,
the growing maize crop took up a large percentage of the N supplied by
either the inorganic or inorganic-organic inputs. This justifies split
application of urea.
Grain yield accounted for a greater portion of the recovered N than
either stover yield or the core. This was also noted by Mugendi et al.
(2000) in their work in the subhumid highlands of Kenya.
220
Kimetu, J.M. et al
Other researchers working on different N sources (organic and
inorganic inputs) also reported a percentage N recovery ranging from
25% to 111% (Westerman et al., 1972; Kruijs et al., 1988; Christianson
et al., 1990; Gachengo et al., 1999). In this study, nitrogen recovery
values from tithonia green biomass was found to be relatively lower
than the values Gachengo et al. (1999) observed using the same
organic material in a study in Western Kenya. This could be due to
differences in environmental conditions especially rainfall distribution
between the two sites. However, the N recovery value from inorganic
fertilizer (urea) agrees with the findings of Chabrol et al. (1988) in a
study in Bedfordshire, England as well as what Mugendi et al. (1999)
found out in their studies in the subhumid highlands of Kenya.
Conclusions
Tithonia had a fertilizer equivalency value of 119% while calliandra and
senna had 72% and 68% respectively. The extent to which an organic
material will perform comparable to mineral fertilizer, is dependent on
several factors especially the quality of the organic materials, climatic
factors and site characteristics. Although higher biomass and grain yields
were obtained from tithonia sole application compared to sole urea
application, maize crop supplied with sole urea was found to recover
nitrogen at a higher rate than maize crop supplied with tithonia biomass.
It is evident that the effect of external inputs on crop N use efficiency is
dependent on the organic material used and climatic conditions
(especially rainfall amount) prevailing throughout the growing period of
the annual crop.
Tithonia diversifolia can be used as a source of nitrogen in place of
mineral fertilizer and smallholder farmers should be encouraged to use
tithonia green biomass for annual crops especially in areas of inadequate
rainfall. Senna spectabilis and Calliandra calothyrsus green biomass
should be recommended for use in combination with inorganic N source
for better results. A similar research should be recommended for other
organic materials and at different agroecological zones as well as to
establish other specific beneficial effects of organic inputs on annual
crop yields.
Acknowledgements
The authors would like to thank the Rockefeller Foundation, which
funded this research through the TSBF Programme. We are also grateful
to all TSBF Nairobi staff led by Prof. Mike Swift for their support
throughout this study.
Nitrogen Fertilizer Equivalency Values for Different Organic Materials Based on Maize
Performance at Kabete, Kenya
221
References
Anderson, J.M. and Ingram, J.S.L. (1993) Tropical Soil Biology and Fertility: A
Handbook of Methods. CAB International, Wallingford, U.K.
Becker, M., Ladha, J.K. and Ottow, J.C.G. (1994) Nitrogen losses and lowland
rice yield as affected by residue nitrogen release. Soil Science Society of
America Journal 58: 1660-1665.
Chabrol, M., Powlson, D.S. and Hornby, D. (1988) Uptake by maize (Zea mays
L.) of nitrogen from 15N-labelled dazomet, 15N-labelled fertilizer and from
the soil microbial biomass. Soil Biology and Biochemistry 20: 517-523.
Chen, Y., Avnimelech, Y., (Eds) (1986) The role of organic matter in modern
Agriculture. Martinus Nijhoff Publishers, Dordrecht. The Netherlands.
Christianson, C.B., Bationo, A., Henao, J. and Vlek, P.L.G. (1990) Fate and
efficiency of N fertilizers applied to millet in Niger. Plant and Soil 125: 221231.
Constantinides, M. and Fownes, J.H. (1994) Nitrogen mineralization from leaves
and litter of tropical plants: relationship to nitrogen, lignin and polyphenol
concentrations. Soil Biology and Biochemistry 26: 49-55.
Delve, R., Gachengo, C., Adams, E., Palm, C., Cadisch, G. and Giller, K.E. (2000)
The Organic Resource Database. In: The Biology and fertility of Tropical
soils: A TSBF Report 1997- 1998. PP 20-22.
FAO (Food and Agriculture Organization) (1990) New Environmental Threat:
Declining Soil Fertility. Land and Water Technical Newsletter.
Gachengo, C.N. (1996) Phosphorus release and availability on addition of organic
materials to phosphorus fixing soils. M.Phil. thesis. Moi University, Eldoret,
Kenya.
Gachengo, C.N., Palm, C.A., Jama, B. and Otieno, C. (1999) Tithonia and senna
green manures and inorganic fertilizers as phosphorus sources for maize
in Western Kenya. Agroforestry Systems 44, 21-36.
ICRAF (1995) Laboratory Methods for Soil and Plant Analysis: Version 1.1.
Nairobi.
Jama, B., Palm, C.A., Buresh, R.J., Niang, A., Gachengo, C., Nziguheba, G. and
Amadalo, B. (2000) Tithonia diversifolia as a green manure for soil fertility
improvement in western Kenya: A review. Agroforestry systems 49: 201221.
Kang, B.T., Reynolds, L. and Atta-Krah, A.N. (1990) Alley farming. Advances in
Agronomy 43: 315-359.
Kruijs, A. C. B. M. van., Wong, M. T. F., Juo, A.S.R. and Wild, A., (1988) Recovery
of 15N-labelled fertilizer in crops, drainage water and soil using monolith
lysimeters in south-east Nigeria. Journal of Soil Science 39: 483-492.
Lehmann, J., Schroth, G., Zech, W. (1995) Decomposition and nutrient release
from leaves, twigs and roots of three alley-cropped tree legumes in Central
Togo. Agrofor Syst. 29: 21-36.
Lehmann, J., Feilner, T., Gebauer, G., Zech, W. (1999) Nitrogen uptake of
sorghum (Sorghum bicolor L.) from tree mulch and mineral fertilizer under
222
Kimetu, J.M. et al
high leaching conditions estimated by nitrogen-15 enrichment. Biology and
Fertility of Soils 30: 90-95.
Mugendi, D.N. (1997) Tree Biomass decomposition, Nitrogen Dynamics and
Maize growth Under Agroforestry conditions in the sub-humid highlands of
Kenya. PhD Thesis, University of Florida, Gainesville, Florida, U.S.A.
Mugendi, D.N., Nair, P.K.R., Mugwe, J.N., O’Neill, M.K., Swift, M.J. and Woomer,
P. (1999) Alley cropping of maize with Calliandra and Leucaena in the Subhumid highlands of Kenya. Part 2: Biomass decomposition, N mineralization,
and N uptake by Maize. Agroforestry Syst. 46: 51-64.
Mugendi, D.N., Nair, P.K.R., Graetz, D.A., Mugwe, J.N. and O’Neill, M.K. (2000)
Nitrogen recovery by alley-cropped maize and trees from 15N-labeled tree
biomass in the subhumid highlands of Kenya. Biol. Fertil. Soils 31: 97-101.
Mutuo, P. K., Marandu, A.E., Rabeson R. Mwale, M. Snapp, S. and Palm, C.A.
(1999) Nitrogen fertilizer equivalencies based on organic input quality and
optimum combinations of organic and inorganic N sources: Network trial
results from East and Southern Africa. In SWNM Report on the combating
Nutrient Depletion – East Africa Highlands consortium.
Mwale, M., Mapiki, A. and Phiri, L.K. (2000a) Optimal combinations of organic
and inorganic N sources. In: The Biology and fertility of Tropical soils: A
TSBF Report 1997- 1998. PP 38-40.
Mwale, M., Mapiki, A. and Phiri, L.K. (2000b. To synchronize nutrient availability
with plant uptake. In: The Biology and fertility of Tropical soils: A TSBF
Report 1997- 1998. PP 40-41.
Myers, R.J.K., Palm, C.A., Cuevas, E., Gunatilleke, I.U.N. and Brossard, M.
(1994) The synchronization of nutrient mineralization and Plant nutrient
demand, In the Biological management of soil fertility, Woomer, P.L. and
Swift, M.J., (eds) John Wiley & Sons, Chichester.
Nziguheba, G. (2001) Improving phosphorus availability and maize production
through organic and inorganic amendments in phosphorus deficient soils
in Western Kenya. PhD thesis, Katholieke University, Leuven.
Palm, C. A. Myers, R. J. K. and Nandwa, S. M. (1997) Combined use of organic
and inorganic nutrient Sources for soil fertility Maintenance and
replenishment. In Buresh, R. J., and Sanchez, P.A., (eds). Replenishing soil
fertility in Africa SSSA spec. Publ. 51. SSSA, Madison, WI.
Palm, C.A., Gachengo, C.N., Dalve, R.J., Cadisch, G. and Giller, K.E. (2001)
Organic inputs for soil fertility management in tropical agroecosystems:
application of an organic resource database. Agriculture, Ecosystems and
Environment. 83: 27-42.
Parkinson, J.A. and Allen, S.E. (1975) A wet oxidation procedure suitable for
the determination of nitrogen and mineral nutrients in biological materials.
Communications in Soil Science and Plant Analysis 6: 1-11.
Shepherd, G., Buresh, R.J. and Gregory, P.J. (2000) Land use affects the
distribution of soil inorganic nitrogen in smallholder production systems in
Kenya. Biol. Fertil. Soils 31: 348-355.
Swift, M.J. (1987) Tropical Soil Biology and Fertility: Interregional research
planning workshop, special issue 13, in biology international IUBS, Paris.
Nitrogen Fertilizer Equivalency Values for Different Organic Materials Based on Maize
Performance at Kabete, Kenya
223
Thonnissen, C., Midmore, D.J., Ladha, J.K., Olk, D.C. and Schmidhalter, U.
(2000) Legume decomposition and nitrogen release when applied as green
manures to tropical vegetable production systems. Agron. J. 92: 253-260.
Vanlauwe, B., Aihou, K., Houngnandan, P., Diels, J., Sanginga, N. and Merckx,
R. (2001) Nitrogen management in ‘adequate’ input maize-based agriculture
in the derived savanna benchmark zone of Benin Republic. Plant and Soil
228: 61-71.
Van Soest, P.J. (1963) Use of detergents in the analysis of fibrous feeds. II: A
rapid method for the determination of fibre and lignin. Journal of the
Association of Agricultural Chemists 46: 829-835.
Wallace, J.S. (1996) The water balance of mixed tree-crop systems. In: Ong,
C.K., and Huxley, P. (eds) Tree-crop interactions, a physiological approach.
CAB International, Wallingford, UK, pp 73-158.
Westerman, R.L., Kurtz, L.T. and Hauck, R.D. (1972) Division s-4 – soil fertility
and plant nutrition: Recovery of 15N-labeled fertilizers in field experiments.
Soil Science Society of America Proc., Vol. 36: 82-86.
224
Kimetu, J.M. et al
Base Nutrient Dynamics and Productivity of Sandy Soils Under Maize-Pigeonpea
Rotational Systems in Zimbabwe
Base Nutrient Dynamics and
Productivity of Sandy Soils
Under Maize-Pigeonpea
Rotational Systems in
Zimbabwe
225
16
Mapfumo, P.* and Mtambanengwe, F.
Department of Soil Science and Agricultural Engineering,
University of Zimbabwe, P.O. Box MP 167, Mount Pleasant,
Harare
*Corresponding Author: pmapfumo@agric.uz.ac.zw
Abstract
A two-year study was conducted in a smallholder farming
area in northeast Zimbabwe to determine the rotational
effects of pigeonpea (Cajanus cajan L. Millsp.) of different
maturity genotypes on maize (Zea mays L.) yields.
Researcher and farmer-managed on-farm rotation
experiments were established on ten sites in farmers’ fields,
with a long history of maize monocropping. A maize crop
receiving different rates of mineral N fertilizer followed long,
medium and short duration pigeonpea genotypes plus
control maize. Relationships between N management factors
and nutrient uptake by maize under the three pigeonpea
genotypes were examined in light of the current soil fertility
management practices by the farmers. Significant (p<0.05)
226
Mapfumo, P. and Mtambanengwe, F.
maize yield responses were obtained after pigeonpea despite
low productivity of these legumes at all sites. Relative to
the control, medium and long duration pigeonpea
treatments resulted in maize yield increases of 46% to 37%
for biomass and 20% to 28% for grain, respectively, while
there was a 6-19% decrease in yield after short-duration
pigeonpea. The low N contributions from pigeonpea, which
ranged from 6-18 kg N ha-1 could not account for the observed
yield responses. Yields across all treatments increased with
increasing mineral N application. Maize tissue analyses at 6
and 15 weeks after emergence (WAE), showed significant
increases in P, K, Ca and Mg uptake. Regression analysis
showed highly significant (P<0.001) linear relationships
between N uptake and Mg (R2 = 0.74; DF = 78), Ca (R2 = 0.62;
DF = 78) and K (R2 = 0.53; DF = 78) uptake. Based on multiple
regression models, N and Mg uptake accounted for most of
the maize grain yield increases observed. Because of the low
N contributions from pigeonpea, increases in maize yields
were largely attributed to improved availability of base
nutrients, particularly Mg. Pronounced maize yield responses
to mineral N observed under pigeonpea systems, were likely
due to increased N use efficiency.
We concluded that the residual benefits of pigeonpea
in these cropping systems are largely due to their capacity
to remobilize and recycle base nutrients, and that the
productivity potential of granitic sandy soils is undermined
by continued depletion of these cations under current
management practices.
Introduction
Positive residual effects of N 2-fixing legumes on subsequent cereals in
rotations have been widely reported in both olden and modern agriculture
(Giller and Wilson, 1991; Kumwenda et al., 1995; Peoples et al., 1995).
The yield increases have been primarily attributed to an improvement
in N economy of the soils. However, beneficial effects may also arise
from breaks in disease and pest cycles, changes in soil microbial and
faunal activity, and chemical and physical attributes (Peoples and
Craswell, 1992), although these factors have seldom been quantified.
In order to optimize the ecological contribution of legumes in low-input
agricultural systems, it is imperative that the differential effects of these
factors are clearly understood and the benefits due to their interaction
quantified. This is particularly important in tropical savanna agro
ecosystems in which soil nutrient stocks are inherently low.
Base Nutrient Dynamics and Productivity of Sandy Soils Under Maize-Pigeonpea
Rotational Systems in Zimbabwe
227
Studies on the predominantly sandy soils of Southern Africa have
shown the complexity of soil fertility problems on smallholder farms
and the challenges in developing sustainable management options
(Scoones et al., 1996; Scoones, 1998; Snapp et al., 1998; Giller, 2001).
There are slim chances of building soil organic matter and hence nutrient
stocks (Giller et al., 1997), rendering farmers to rely heavily on external
nutrient inputs on a seasonal basis. However, most of the smallholder
farmers use sub-optimal amounts of fertilizers due to cash limitations
and poor access to fertilizer markets (Kumwenda et al., 1995; Ahmed et
al., 1996). This, therefore, calls for increased efficiency in use and
recycling of both exogenous and endogenous nutrient pools in the
cropping systems. Although problems of multiple nutrient deficiency
are often apparent under most continuously cropped soils (Grant, 1981;
Mukurumbira and Nemasasi, 1998), little has been done to determine
the influence of the various soil fertility technologies on availability of
nutrients such as K, Mg and Ca. In this study, we examined the residual
effect of pigeonpea (Cajanus cajan (L.) Millsp.) cropping has on maize
(Zea mays L.) yields on a sandy soil in Zimbabwe, with particular
attention on N, P, K, Mg and Ca uptake.
Pigeon pea, a relatively new crop in Zimbabwe, was chosen for its
ability to grow on relatively infertile soils, and tolerance to drought and
other environmental stress (Whiteman et al., 1985; van der Maesen, 1990).
Study site
The study was conducted in Murewa Communal Area, 130 km north
east of Harare (17°45’S and 31°31’E). The area has a unimodal rainfall
pattern, receiving an average of 750-1000 mm annually, between
November and March and is about 1300 m above sea level. The mean
annual temperature is 22°C. Soils are predominantly Lixisols (FAO
Classification), derived from granitic parent material. Because of
intensive agricultural activity and population pressure, most of the
natural Julbernardia globiflora and Brachystegia spiciformis (miombo)
tree vegetation has disappeared.
Materials and Methods
The experiment was conducted for two seasons on 10 farm sites, with
each farm site considered as a replicate. All sites had been under maize
monocropping with no manure application for at least five years. Soil
samples were taken from 0-20 cm depths for physical and chemical
analysis, at the beginning of each growing season. Three pigeonpea
228
Mapfumo, P. and Mtambanengwe, F.
maturity types, namely short (cv. ICPL 87109 – 90 days to maturity),
medium (cv. ICP 9145 – 150 5ays), long duration (cv. Ex-Marondera –
180 days), were grown during season one. Maize (cv. SC 501) was
included as a fourth treatment to serve as a control in season two. Each
plot measured 18 m x 4.5 m in gross area. Pigeonpea was spaced at 0.9
m between rows and 0.2 m within rows while maize was spaced at 0.9
m x 0.3 m. Land preparation was done by farmers using an ox-drawn
plough. The pigeonpea received 12.5 kg P ha-1 and 18 kg S ha-1 in form
of single superphosphate (SSP) incorporated just before planting. The
maize received a basal dressing of Compound D at 200 kg ha-1 (16.0 kg
N ha -1; 12.4 kg P ha-1; 11.6 kg K ha-1 and 13 kg S ha-1), and was topdressed with ammonium nitrate at 63.8 kg N ha-1. The short duration
pigeonpea was harvested for grain at 101 days after planting (DAP) and
residues retained in the field for incorporation. The medium and long
duration pigeonpea were incorporated as green manure at flowering at
the request of farmers who feared interference from livestock. An oxdrawn plough was used for incorporation. For each pigeonpea cultivar,
a net plot of 17 m x 2.7 m was harvested to determine fresh shoot
biomass before incorporation. Sub-samples of three whole plants each
were taken in replicates for moisture correction and quality analysis.
Maize harvesting was done at physiological maturity. The samples were
oven-dried at 60°C to constant mass, for dry matter measurements.
Dried samples were ground and passed through a 1 mm sieve in a
Wiley Mill. Total C, N, lignin and polyphenol contents were determined
using methods described by Anderson and Ingram (1993).
In the second season, all plots were planted with maize. The crop
was given a basal P, K and S fertilizer dressing in form of Compound D
as described above for year one. However, each of the season one plot
was divided into two, with one half receiving no mineral N application
and the other getting 60 kg N ha-1 in form of ammonium nitrate. This
gave rise to a split plot design, with the different rotation treatments
(pigeonpea maturity types + maize control), providing for main plots
and the two N fertilizer rates as subplots. The N was split applied, with
30 percent being applied at 2 WAE, 50% at 6 WAE and other 20% at 9
WAE. Three randomly selected plants were taken for biomass estimates
at 2, 6 and 15 WAE. At each sampling time, the biomass was analyzed
for N, P, K, Mg and Ca uptake. Individual farmers did the weeding
whenever it was necessary. Grain yield was determined from a net plot
of 4 m x 2.7 m and measured at 12.5% moisture content.
Pigeonpea productivity was also determined in the second season
by planting the three maturity types on areas adjacent to the season
one plots. All the areas had been put under maize by farmers in the
previous season and the plots measured 4.5 m x 5 m. Shoot biomass
were determined at flowering and physiological maturity by destructive
sampling from a 1.8 m 2 area. At maturity, cumulative litter was
Base Nutrient Dynamics and Productivity of Sandy Soils Under Maize-Pigeonpea
Rotational Systems in Zimbabwe
229
handpicked from the sampled area and quantified for dry matter. The
plant shoot samples were analyzed for total C, N, lignin and polyphenol
content as described above. An area measuring 4 m2 was harvested for
grain yield determination.
All soils were analysed using methods given by Anderson and Ingram
(1993). Organic C was determined using a modified Wakley-Black method
while the resin method was used to measure available P. Ammonium
and nitrate N were determined using the indophenol and cadmiumreduction methods respectively, with the N measured colourimetrically.
Soil exchangeable K was determined by flame photometry and Mg and
Ca by atomic absorption spectrophotometry following leaching of soil
with ammonium acetate.
Treatment differences were tested using ANOVA and treatment
means compared by LSD at P<0.05. Linear and multiple regression
analyses were used to determine the relationships between maize yields
and nutrient uptake.
Soil properties
All the farm sites used in the study were low in major plant nutrients,
especially N and P (Table 16.1).
Table 16.1: Soil characteristics for the ten farm sites used for a pigeonpea study in
Murewa Communal Area, Zimbabwe
Site
Mukarakate
Shangwa
Gutu
Chikurunhe
Chawanda
Marume
Chiroodza T
Chiroodza M
Chamboko
Chituwu
Clay
(%)
4
4
4
3
4
5
9
6
9
6
Sand
pH Avail. N Resin Organic Total N
(%) (CaCl 2) (ppm) P(ppm) C (%
(%)
89
90
92
92
86
88
82
90
86
92
4.2
4.3
4.1
4.6
4.4
4.2
4.5
4.6
4.4
4.2
17
30
17
34
26
27
31
18
27
16
8
2
6
10
3
2
2
11
4
8
0.26
0.35
0.28
0.23
0.40
0.24
0.44
0.32
0.47
0.32
0.02
0.02
0.01
0.02
0.02
0.02
0.01
0.02
0.02
0.02
K
(cmolckg-1)
0.07
0.09
0.06
0.06
0.07
0.06
0.14
0.08
0.13
0.10
Ca
Mg
(cmolc- (cmolckg-1)
kg-1)
0.39
0.68
0.52
0.96
1.19
0.60
1.39
2.06
0.94
0.85
0.15
0.35
0.21
0.30
0.40
0.20
0.59
0.75
0.26
0.30
Clay content ranged from 3 to 9%. Available P ranged from 2 to 11
mg kg-1, while N ranged from 17 to 34 mg kg-1 soil. The soils were strongly
acidic and had a mean organic C of 0.33% (Table 16.1).
Pigeonpea productivity and quality attributes
Pigeonpea biomass yields recorded during the season one were highly
variable across the different farm sites. The yields ranged from 264 kg
230
Mapfumo, P. and Mtambanengwe, F.
ha-1 for the short duration type to 1104 kg ha -1 for the long duration
type. High variability was partly due to excessive rainfall, as severe
waterlogging was experienced in some of the farms. The yields for Shortduration pigeonpea yields are likely to have been undermined by the
relatively low plant population used. Maize biomass and grain yields in
control maize plots averaged 2856 and 450 kg ha -1 respectively. The
long duration pigeonpea yielded 1.5 times more biomass than the
medium duration variety (Table 16.2).
Table 16.2: Biomass yields and resource quality characteristics for pigeonpea of different
maturity types (at flowering) grown on poor sandy soils in Murewa Communal Area,
Zimbabwe
Parameter
Cultivar /duration
ICPL 87109
(short)
Season 1
*biomass (kg ha -1)
Shoot N (kg ha -1)
N (%)
C (%)
C:N
Polyphenols
Lignin
264 (95)
6
(2)
2.35 (0.05)
43 (0.5)
18 (0.4)
2.8 (0.04)
10.7 (0.3)
Season 2
*biomass (kg ha -1)
Shoot N (kg ha -1)
N (%)
C (%)
C:N
Polyphenols
Lignin
1884 (259)
46
(7)
2.48 (0.05)
43 (0.4)
17 (0.3)
2.2 (0.04)
11.2 (0.2)
ICP 9145 (medium)
(medium)
733
13
1.78
44
25
2.8
13.2
Ex-Marondera
(long)
(251)
(5)
(0.08)
(0.2)
(1.2)
(0.05)
(0.6)
1104 (302)
18
(5)
1.65 (0.07)
44 (0.5)
27 (1.3)
3.3 (0.05)
12.7 (0.4)
6071 (721)
130
(16)
2.20 (0.10)
43 (0.6)
20 (0.9)
2.4 (0.04)
10.8 (0.4)
7619 (900)
157 (19)
2.16 (0.08)
44 (0.5)
20 (0.8)
3.3 (0.03)
13.0 (0.6)
*biomass = shoot biomass determined at flowering; Numbers in parentheses indicate
standard errors
Because of poor biomass yields and low shoot N concentration, the
total N produced by all the pigeonpea types were low. In season two,
pigeonpea yields were about 6-8 times more than those obtained in
season one, partly due to a favourable rainfall pattern. The long duration
genotype, Ex-Marondera, gave four times more biomass than the short
duration variety at flowering stage (Table 16.2). Potential N contribution
to soil, as measured at flowering stage, ranged from 46 kg ha-1 for the
short duration to 150 kg ha-1 for the long duration type. There was only
a small increase in biomass between flowering and maturity stages due
to terminal drought (Table 16.3). There was also a rapid increase in
litterfall between the two growth stages.
Base Nutrient Dynamics and Productivity of Sandy Soils Under Maize-Pigeonpea
Rotational Systems in Zimbabwe
231
Table 16.3: Biomass (at physiological maturity) and grain yields for pigeonpea of different
maturity types grown on a sandy soil in Murewa Communal Area, Zimbabwe
Pigeonpea Maturity Type Biomass (kg ha -1)
Short (ICPL 87109)
Medium (ICP 9145)
Long (Ex-Marondera)
Litterfall (kg ha -1)
1939 (263)
6436 (701)
8106 (879)
112 ( ±13 )
745 ( ±108)
994 ( ±131)
Grain Yield (kg ha-1)
725 ( ±68)
284 ( ±28)
391 ( ±67)
*Numbers in parentheses indicate standard errors.
Although most of the litterfall was observed to occur soon after
maturity, this could not be quantified due to disturbances of plots by
livestock. The terminal drought also contributed to poor grain yields by
the medium and late pigeonpea (Table 16.3). Grain filling was apparently
more affected by moisture stress than pod set.
Residual effects of pigeonpea on yields of subsequent
maize
An analysis of variance on grain yield and biomass yields at 2, 6 and 15
WAE, showed significant (P<0.05) rotation and mineral N fertilizer effects.
There was no significant interaction between rotation and mineral N
fertilizer. The medium and long duration pigeonpea gave rise to
significantly higher maize yields compared with the maize-maize control
and short duration pigeonpea (Table 16.4).
Table 16.4: Maize biomass at 2, 6 and 15 weeks after emergence, and grain yields
obtained following pigeonpea of different maturity types grown under two mineral N
fertilizer rates on sandy soils in Murewa Communal Area, Zimbabwe
Biomass (kg ha -1)
Pigeonpea rotation/
mineral N fertilizer
Maturity Type
Short (ICPL 87109)
Medium (ICP 9145)
Long (Ex-Marondera)
Maize control (SC 501)
SED
Fertilizer Level
0 (kg N ha -1)
60 (kg N ha -1)
SED
2 WAE
49ab
70c
59bc
45a
6
nd
nd
nd
Grain
(kg ha-1)
6 WAE
717a
1064bc
1069b
803ac
129
15 WAE
3015a
4690b
4442b
3203a
526
733a
1156b
1229b
962ab
173
709a
1118b
57
2499a
5176b
290
615a
1425b
125
Numbers in the same column followed by the same letter are not significantly different;
WAE = weeks after maize emergence; nd = not determined
232
Mapfumo, P. and Mtambanengwe, F.
At all harvesting stages, there were no significant differences in yield
between short duration pigeonpea and maize-maize control plots,
although the latter gave numerically higher yields at 6 and 15 WAE.
The maize grown after medium and long duration pigeonpea
treatments yielded 46% and 37% more biomass, respectively, relative
to the control. Similarly, there was 20% and 28% more grain under the
two treatments, respectively. Mineral N fertilizer had a highly significant
effect on maize yields, with yield increase ranging from 58% for biomass
at 6 WAE to 132% for grain relative to the unfertilized treatments (Table
16.4). Without mineral N fertilizer application, the maize grain yields
ranged from 427 kg ha -1 after short duration pigeonpea to 779 kg ha-1
after the long duration genotype. Grain yield for the control maize was
530 kg ha-1.
Effect of pigeonpea cropping on soil N availability and
maize nutrient uptake
Rotation treatments had no significant (P<0.05) effect on N and P uptake
at 2 WAE, while only P and K uptake were influenced by the rotation
systems at 6 WAE. While P, Ca and Mg uptake were significantly
increased by mineral N fertilizer application at this stage, N and K uptake
were not affected (Table 16.5).
Table 16.5: Rotational effects of pigeonpea of different maturity types on maize nutrient
uptake under a sandy soil in Murewa Communal Area, Zimbabwe
Nutrient (kg ha -1) /sampling stage (WAE)
Rotation system /
N level
N
P
K
Ca
Mg
6
15
6
15
6
15
6
15
6
15
Maturity Type
Short
Medium
Long
Maize control
SED
18
17
16
13
ns
22a
38b
37b
24a
4
0.2a
0.3b
0.3b
0.2a
0.02
0.7
0.9
0.9
0.6
ns
16ab
15ab
23a
9a
6
14a
30b
24ab
15a
5
3
4
4
4
ns
5a
8b
8b
5a
0.5
1
2
2
1
ns
2a
4b
4b
2a
1.4
Fertilizer Level
0 (kg N ha -1)
60 (kg N ha -1)
SED
19
18
ns
18a
43b
3
0.2a
0.3b
0.01
0.5a
1.1b
0.07
13
19
ns
12a
29b
4
2
4
0.5
4a
9b
1
1
2
0.2
2a
4b
0.3
Short duration = variety ICPL 87109; medium duration = ICP 9145; Long duration = ExMarondera; Maize = cv. SC 501. WAE = weeks after maize emergence. Numbers in the
same column followed by the same letter are not significantly different at P<0.05.
Base Nutrient Dynamics and Productivity of Sandy Soils Under Maize-Pigeonpea
Rotational Systems in Zimbabwe
233
There was a 50% increase in P uptake after medium and long
duration pigeonpea, relative to the control maize. Although maize yield
after short duration pigeonpea was generally lower than that of the
control, this treatment resulted in a 90% increase in K uptake (Table
16.5). Unlike early in the maize growth stages, both rotation systems
and mineral N had a highly significant effect on uptake of all the
measured nutrients at 15 WAE.
In general, there were no significant (P<0.05) differences in nutrient
uptake between the medium and long duration pigeonpea rotation
systems and between the control and the short duration. Regression
analyses showed highly significant (P<0.001) linear relationships between
N uptake and Mg (R 2 = 0.74; DF = 78), Ca (R 2 = 0.62; DF = 78), and
K (R2 = 0.53; DF = 78) uptake. Uptake of these nutrients increased with
increased N application rates and uptake. There were significant (P<0.05)
relationships between maize grain yield and nutrient uptake, particularly
N and Mg (Table 16.6).
Table 16.6: Linear relationships between maize grain yield and nutrient uptake rotated
with pigeonpea of different maturity types on sandy soils in Zimbabwe
Regression equation
DF
R2
P-level
Y (Grain yield) = 34 X (total N uptake)
Grain yield = 1429 (total P uptake)
Grain yield = 131 + 34 (total K uptake)
Grain yield = 256 + 117 (total Ca uptake)
Grain yield = 313 (total Mg uptake)
78
78
78
78
78
0.76
0.57
0.55
0.44
0.75
0.001
0.001
0.001
0.001
0.001
Discussion
Pigeonpea productivity under poor soil fertility
Low pigeonpea biomass yields during the first season was mainly
attributed to waterlogging and poor soil fertility of the soils at the study
sites. Waterlogging occurred during the early vegetative phase. This could
have interfered with N2-fixation resulting in poor N nutrition and hence
growth of the plants (Mapfumo et al., 1999). Because of their relatively
long growth duration, the medium and long maturity pigeonpea were
able to recover from the adverse effects of excessive soil moisture
resulting in relatively high yields. Wide plant spacing could also have
contributed to low biomass for short duration pigeonpea. While the high
biomass yields obtained in the second season strongly indicated
pigeonpea tolerance to poor soil fertility, the results also suggest that
there could be significant interactions between soil moisture and crop
234
Mapfumo, P. and Mtambanengwe, F.
nutrition under sandy soils. According to Grant (1981), soil acidity may
contribute to soil infertility in high rainfall areas. About 50% of the
farm sites used in this study had a pH of 4.3 (CaCl2), a value considered
critically low for most crops, particularly legumes (Grant, 1981). Liming
is a possible solution in a number of communal areas including Murewa
(Dhliwayo et al., 1998). However, the dynamics of soil acidity under
different soil moisture regimes in sandy soils in Zimbabwe is not well
understood.
Low grain yields found for the medium and long duration pigeonpea
were indicative of problems likely to be encountered by farmers in
regenerating seed. Although pigeonpea was able to sustain growth on
residual moisture, the soil moisture reserves were apparently not
sufficient to promote grain filling. Because of the poor water-holding
capacity of sandy soils, there is usually a rapid soil moisture decline
soon after the end of the rainy season (Mapfumo, 1995). Terminal
drought has been reported as a major constraint to grain production in
long duration pigeonpea (Whiteman et al., 1985; ICRISAT, 1993). This
may be a disincentive for farmers whose primary interest is grain
production. The short duration type may be attractive to farmers in
that respect. There is need to identify germplasm of appropriate maturity
types to minimize terminal drought effects while ensuring optimum
biomass accumulation.
Rotational effects on subsequent maize
Because of the low levels of pigeonpea biomass produced during season
one, the potential N contribution to the cropping system was low despite
the high quality of the incorporated biomass. Only 6-18 kg N ha-1 could
be added to the system from pigeonpea shoots. Given that only 20% of
this N was likely to be taken up in maize (Palm, 1995; Palm et al.,
1997), it was therefore highly likely that the observed treatment
differences were due to rotational effects other than N. While there were
significant treatment effects on maize biomass yield at 6 WAE, both
rotation systems and mineral N did not significantly affect N uptake
during the same period. This further suggests that N per se had no
significant effect on early maize development.
During the same period (6 WAE), P, K and Ca uptake were
significantly higher under medium and long duration pigeonpea rotations
than under the other two treatments. The yield increases may, therefore,
be partly attributable to the increased availability of these nutrients
after pigeonpea.
Rotation and mineral N effects on maize N uptake only became
significant at crop maturity. This could have been due to increased
nutrient use efficiency (NUE) under the pigeonpea treatments between
Base Nutrient Dynamics and Productivity of Sandy Soils Under Maize-Pigeonpea
Rotational Systems in Zimbabwe
235
6 and 9 WAE. In sandy soils of the Sahel region, there was a 20% NUE,
under pearl millet monocropping compared with 28% after a cowpea
rotation (Bationo et al., 1991; Bationo and Vlek, 1997). The results in
this study also showed a highly significant linear relationship between
maize grain yield and Mg uptake, suggesting that Mg may be a major
limiting nutrient for maize production in these sandy soils. Magnesium
uptake was also linearly related to N uptake, suggesting that increased
use of mineral N fertilizers in maize monocropping may accelerate Mg
depletion in sandy soils. Magnesium deficiency on continuously
cropped sandy soils has been historically reported from smallholder
farms in Southern Africa (Grant, 1981). Application of pigeonpea
residues did not only improve Mg nutrition, but also increased P, K
and Ca uptake. Although it may be difficult to account for the observed
yield depression under short duration pigeonpea based on our limited
results, it is likely that low potential of the short duration pigeonpea
to remobilize and recycle these nutrients was a factor. The genotype
may therefore not be suitable for soil fertility enhancement on sandy
soils.
Practical significance of rotational benefits and pigeonpea
productivity on-farm
Maize grain yields of about 1000 kg ha -1 obtained after pigeonpea,
may fall short of farmers' expectations despite their statistical
significance. Currently, most successful farmers achieve high yields
(about 3000 kg ha -1), by applying more than the recommended rates
of mineral N (Mapfumo and Mtambanengwe, 1999). Athough there is
little documentation on the economics of such production systems,
low N use efficiency may be eroding profits. Pigeonpea rotations, may
help to improve NUE and increase the yield potential of the cropping
system through remobilization and cycling of other nutrients such as
Mg, Ca and K. Our limited results indicate that there is merit in
directing our focus on non-N benefits of organic resources, particularly
the influence on availability of base nutrients, which may be
undermining the productivity of sandy soils even in cases where N
becomes available. The productivity potential of the granitic sandy
soils in Southern Africa might be declining due to base nutrient
depletion.
Acknowledgements
Funding for this study was provided by the European Union through
the Institute of Enviromental Studies of the University of Zimbambwe.
236
Mapfumo, P. and Mtambanengwe, F.
References
Ahmed, M. M., Rohrbach D. D., Gono L. T., Mazhangara E. P., Mugwira L.,
Masendeke D. D. and Alibaba S. (1996) Soil Fertility Management in the
Communal Areas of Zimbabwe: Current Practices, Constraints and
Opportunities for Changes. Results of a diagnostic survey. Southern and
Eastern Africa Regional Paper number 6. ICRISAT-Southern and Eastern
Africa Region. 27 pp.
Anderson, J.M. and Ingram, J.S.I. (1993) Tropical Soil Biology and Fertility: A
handbook of Methods. Second Edition. C. A. B. International. Wallington,
UK. 221pp.
Bationo, A., Ndunguru, B.J., Ntare, B.R., Christianson, C.B. and Mokwunye,
A.U. (1991) Fertilizer management strategies for legume-based cropping
systems in the West African Semi-Arid Tropics. In: Johansen, C., Lee, K.K.,
Sahwat K.L (eds.), Phosphorus Nutrition of Grain Legumes in the Semi-arid
Tropics. ICRISAT, 1991. Patancheru, A.P. 502 324, India:ICRISAT.
Bationo, A. and Vlek, P.L.G. (1991) The role of mitrogen fertilizers applied to
food crops in the Sudano-Sahelian zone of West Africa. In: Renard, G.,
Neef, A., Becker, K., and von Oppen, M. (eds.). Soil fertility management in
West African land use systems. Wekersheim, Germany: Margraf Verlag. pp.
437–444.
Dhliwayo, D.K.C., Sithole, T. and Nemasase, H. (1998) Soil acidity - Is it a problem
in Zimbabwe? In: Waddington, S.R., Murwira, H.K. Kumwenda J.D.T., Hikwa
D., Tagwira F. (eds.), Soil Fertility Research for Maize-Based Farming Systems
in Malawi and Zimbabwe. Soil Fert Net and CIMMYT-Zimbabwe, Harare,
Zimbabwe. pp. 217-221
Giller, K.E. (2001) Targeting management of organic resources and mineral
fertilizers: Can we match scientists’ fantasies with farmers’ realities? In:
Vanlauwe, B., Sanginga, N., Diels, J., Merckx, R. (eds.), Balanced Nutrient
Management Systems for the Moist Savanna and Humid Forest Zones of
Africa. CAB International, Wallingford. pp. 155 -168.
Giller, K.E. and Wilson, K.J. (1991) Nitrogen Fixation in Tropical Cropping Systems.
CAB International, Wallingford, UK. 313 pp.
Giller, K.E., Cadisch, G., Ehaliotis, C., Adams, E., Sakala, W.D. and Mafongoya,
P.L. (1997) Building soil nitrogen capital in Africa. In: Buresh, R.J., Sanchez,
P.A. (eds.), Replenishing Soil Retility in Africa. SSSA Special Publication 51.
SSSA, Madison, WI, USA. pp. 151-192.
Grant, P. M., (1981) The fertilization of sandy soils in peasant agriculture. Zim.
Agric. J. 78(5):169-175.
ICRISAT, (1993) Improvement of Pigeonpea in Eastern and Southern Africa.
Biannual Progress Report submitted to African Development Bank. 20 pp.
Kumwenda, J.D.T., Waddington, S.R., Snapp, S.S., Jones, R.B. and Blackie,
M.J. (1995) Soil Fertility Management for Smallholder Maize-Based Cropping
Base Nutrient Dynamics and Productivity of Sandy Soils Under Maize-Pigeonpea
Rotational Systems in Zimbabwe
237
Systems of Southern Africa: A Review. Network Working Paper No. 1. Soil
Fertility Network for Maize-Based Cropping Systems in Countries of Southern
Africa. CIMMYT, Harare, Zimbabwe. 34 pp.
Mapfumo, P. (1995) Agronomic and ecological adaptation of Lupinus spp for
Communal Area farming systems in Zimbabwe. M. Phil Thesis. Department
of Crop Science, University of Zimbabwe.Harare, Zimbabwe. 218 pp.
Mapfumo, P. and Mtambanengwe, F., (1999) Nutrient mining in maize-based
systems of rural Zimbabwe. In: CIMMYT and EARO, 1999. Maize Production
Technology for the Future: Challenges and Opportunities. Proceedings of the
Sixth Eastern and Southern Africa Regional Maize Conference, 21-25
September, 1998, Addis Ababa, Ethiopia. pp. 274-277.
Mapfumo, P. Giller, K.E. Mpepereki S. and Mafongoya, P.L. (1999) Dinitrogen
fixation by pigeonpea of different maturity types on granitic sandy soils in
Zimbabwe. Symbiosis 27:305-318.
Mtambanengwe, F. and Kirchmann, H. (1995) Litter from a tropical savanna
woodland (Miombo): chemical composition and C and N mineralization. Soil
Biol. Biochem. 27:1639-1651.
Mukurumbira, L. and Nemasasi, H. (1998) Micronutrients in the maize-based
system of the communal areas of Zimbabwe. In: Waddington, S.R. Murwira,
H.K. Kumwenda, J.D.T. Hikwa, D. Tagwira F. (eds.), Soil Fertility Research
for Maize-Based Farming Systems in Malawi and Zimbabwe. Soil Fert Net
and CIMMYT-Zimbabwe, Harare, Zimbabwe. pp. 229-232.
Palm, C.A. (1995) Contribution of agroforestry trees to nutrient requirements of
intercropped plants. Agrofor. Syst. 30:105-24.
Palm, C.A., Myers, R.J.K. and Nandwa, S.M. (1997) Combined use of organic
and inorganic nutrient sources for soil fertility maintenance and
replenishment. In: Buresh, R.J. Sanchez, P.A. (eds.), Replenishing Soil
Fertility in Africa. SSSA Special Publication 51. SSSA, Madison, WI, USA.
pp. 193-217.
Peoples, M. B. and Craswell, E.T. (1992) Biological nitrogen fixation: Investments,
expectations and actual contributions to agriculture. Plant and Soil 141:13-39.
Peoples, M.B., Herridge, D.F. and Ladha, J.K. (1995) Biological nitrogen fixation:
An efficient source of nitrogen for sustainable agricultural production. Plant
and Soil 174:3-28.
Scoones, I. (1998) Investigating soil fertility in Africa: Some reflections from
research in Ethiopia and Zimbabwe. In: Bergstrom, L. Kirchmann, H. (eds.),
Carbon and Nutrient Dynamics in Natural and Agricultural Tropical
Ecosystems. CAB International, Wallingford, UK. pp. 245-259.
Scoones, I., Chibudu, C., Chikura, S., Jeranyama, P., Machaka, D., Machanja,
W., Mavedzenge, B., Mombeshora, B., Mudhara, M., Mudziwo, C.,
Murimbarimba, F., and Zirereza, B. (1996) ha zards and opportunities.
Farming livelihoods in dry land Africa: Lessons from Zimbabwe. London, UK
and New Jersey, USA: Zed Books Ltd, in association with International
Institute for Environment and Development (IIED). 267 pp.
238
Mapfumo, P. and Mtambanengwe, F.
Snapp, S.S., Mafongoya, P.L. and Waddington, S. (1998) Organic matter
technologies for integrated nutrient management in smallholder cropping
systems of southern Africa. Agric. Ecosyst. Environ. 71:185-200.
van der Maesen, L.J.G. (1990) Pigeonpea: Origin, History, Evolution and
Taxonomy. In: Nene, Y.L. hall S.D., Sheila U.K. (eds.), The Pigeonpea. C.A.B.
International, UK. ICRISAT. pp. 15-46.
Whiteman, P.C., Byth, D.E. and Wallis, F.S. (1985) Pigeonpea (Cajanus cajan
L.). In: Summerfield, R.J. and Roberts, E.H. (Eds.), Grain legume crops.
Collins, London. pp. 658-698.
Soil Organic Matter (SOM): The Basis for Improved Crop Production in Arid and SemiArid Climates of Eastern Kenya
Soil Organic Matter (SOM):
The Basis for Improved Crop
Production in Arid and SemiArid Climates of Eastern
Kenya
239
17
Micheni, A., Kihanda, F. and Irungu, J.
Kenya Agricultural Research Institute, P.O. Box 27 Embu, Kenya.
E-mail: kariembu@salpha.co.ke
Abstract
Soil organic matter (SOM) plays an important role in
maintaining physical, chemical and biological properties
of the soil, and therefore the crop productivity. A study
was conducted in arid and semi-arid lands (ASALs) of
eastern Kenya to assess the influence of SOM on crop
productivity after 10 years of application of high quality
goat manure. The manure was acquired from a single
source where same breeds and flock management were
maintained throughout the experimentation period. The
manure contained 0.48 % P, 2.04% N and 25.62% C, and
was annually applied at 0, 5 and 10 tons ha-1 in soils where
continuous cultivation was a common practice. The
residual effects of manure were monitored after
discontinuation of 4 years manure application. Also,
240
Micheni, A. et al
inorganic fertilizers to supply phosphorus (P) and nitrogen
(N) were applied to compare the potency of long-term SOM
maintenance and inorganic fertilizers on crop performance.
The observed maize yields were compared with simulated
(predicted) values from modelling using the Agricultural
Production Systems Simulator (APSIM) model. The results
showed that both the application of manure and mineral
fertilizers improved crop total dry matter, and
discontinuation of annual manure application led to rundown trends in crop yields. A general conclusion made
from the study was that, it is worthwhile in terms of crop
productivity to maintain SOM through annual application
of high quality manure at 5 tons ha -1 in ASALs where
continuous cultivation is practiced.
Key words: Soil organic matter (SOM), crop yields, arid and semi-arid
lands (ASALs), manure, inorganic fertilizers, modelling
Introduction
Enhancement and maintenance of soil productivity is one of the essential
aspects for sustained agricultural production in sub-Saharan Africa
(Bunting, 1992). This is an important aspects, especially when the aim
is to achieve one of the most important objectives of our time, overcoming
hunger and poverty amongst the smallholder farmers who are the
majority among the stakeholders in agricultural production systems
(Micheni, 1996). Soil organic matter (SOM) serves as an indispensable
source of plant nutrients and enhances soil biological, chemical and
physical properties (Mokwunye et al., 1996). Almost all soil nitrogen
and other important soil properties such as moisture retention, cation
exchange capacity (CEC) and stabilization of soil aggregates are related
to SOM. The amount of SOM in the soil is dependent on the annual
inputs of organic materials and the rate of decomposition, the later
being the highest in hot, humid climatic regions (De Ridder and Van
Keulen, 1990; Rowell, 1994). Plant residues are the main source of soil
organic matter while animal remains and their waste are secondary
sources (Rowell, 1994).
According to Jaetzold and Schmidt (1983), the ASALs of eastern
Kenya are characterized by frequent droughts due to erratic and
unreliable rainfall, which is bimodal with first and second rains coming
in April and November, respectively. The average annual rainfall is about
750 mm with poor distribution within and between seasons. The soils
are generally sandy-loam, shallow and are low in organic matter (Jaetzold
Soil Organic Matter (SOM): The Basis for Improved Crop Production in Arid and SemiArid Climates of Eastern Kenya
241
and Schmidt, 1983; Warren et al., 1998). They are also deficient in major
plant nutrients such as nitrogen and phosphorus, a situation
significantly influencing crop yields and land biodiversity (Smaling et
al., 1997; Warren, 1998). Similarly, Ikombo (1984) noted that the soils
of semi-arid eastern Kenya have low soil organic carbon compared to
those of high rainfall areas. The situation is worsened by the methods
of cultivation that may be described as more of nutrients mining, rather
than nutrient build-up (Ikombo, 1984; Micheni, 1996). The farming
practices amongst some farmers involve burning crop residues, weeds
and other plant materials to make way for grazing and crop production
(Gibberd, 1995; Irungu et al., 1997). The problem is aggravated by tree
harvesting for timber, charcoal burning, and failure by farmers to apply
sufficient external soil fertility improvement inputs (Okoba and Altshul,
1995; Lal and Stewart, 1995). Wind and water erosions also causes
significant decline in soil organic matter and nutrients (DAREP, 1995;
Okoba and Altshul, 1995).
Use of mineral fertilizers has been recommended and popularized
to farmers, but the adoption of fertilizer based technologies is constrained
by the high costs, low farm returns and unavailability of the right
fertilizers to the resource poor farmers in arid and semi-arid areas
(Micheni, 1996). Most farmers apply insufficient or no soil fertility
improvement inputs to refurnish the removed soil nutrients (DAREP,
1995). Nitrogen is also lost through volatilization during prolonged dry
spells that are common phenomenon in arid and semi-arid climates
(Coen et al., 1992).
Indigenous shifting cultivation system characterized by long fallow
periods, thereby restoring soil fertility through build-up of SOM are
currently not applicable due to high pressure on land caused by the
ever-increasing human population. Organic nutrient resources (crop
residues, biomass transfer and livestock manure) may be an
alternative to mineral fertilizers. However, the low quality and labour
required for transporting, spreading and incorporating manure in
the field are major limitations (Ikombo, 1984; Kihanda, 1998). Another
major constraint regarding the use of organic inputs is their bulkiness.
For example, large quantities (5 – 10 tons ha -1 ) of farm-yard manure
(FYM) are required to provide a fraction of what would be needed to
maintain agricultural production at a desirable level (Kihanda, 1998).
Farmers in Machang’a are smallholders and keep livestock and grow
dryland crops for food and cash generation. Because of small family
land sizes, continuous cultivation, even on sloppy and fragile fields
is common.
To effectively improve the level of SOM in soils where continuous
cropping is practiced, large quantities of organic inputs should be
continuously applied in erosion free cultivated fields (De Ridder and
Van Keulen, 1990). A long-term manure application trial was initiated
242
Micheni, A. et al
in 1989 with the aim of establishing crop performance resulting from
SOM improvement through manure application. With a realization that
the long term trials are costly and take a lot of time, resources and
manpower, a decision was made to use the long-term Machang’a data
in modelling using Agricultural Production Systems Simulator (APSIM)
model to predict the future agronomic systems. The model considers
the soil as the central focus and allows for simulations of agricultural
production scenarios using pre-prepared weather, crop type and
management templates.
Materials and Methods
This study was carried out at Machang’a (0 0400 S-0 0450 S, 370350 E370450, 1050 m above sea level and 230 mean annual temperatures).
The site is in ASALs of eastern Kenya and has an average annual rainfall
of 750 mm, coming in two rainy (crops growing periods) seasons. The
rainy seasons are identified by the month that effective rainfall occurs.
They are the “April season” that runs from March to June/July and the
“November season” falling from October to January. The soils are
generally sandy clay loam (Chromic cambisol) with 6.45 pH (water),
0.67% organic carbon, 0.94mg kg-1 extractable P (Olsen method; 0.5M
Na HCO 3) and 0.06% total nitrogen. The soils are shallow (about 1m
deep) and lose their organic matter, including nutrient rich aggregates
within 3-4 years of cultivation with inadequate internal/external organic
material inputs and soil protection from water erosion. They have poor
structures and are easily compacted and eroded especially during heavy
storms that characterize the area.
A long-term (approximately 10 years) manure experiment was
initiated in 1989 with the aim of assessing the crop yields performance
following continuous cropping and improvement of SOM through manure
application. The manure was obtained from a single source where the
same breed and flock management were maintained throughout the
experimentation period and was applied at the rate of 5 and 10 tons ha1
. It (manure) was considered to be of high quality with 2% N, 26% C
and 0.5% P and was applied in October, prior to November rains by
evenly spreading and incorporating it within the cultivation depth (00.15m) of the soil.
Initially, from April 1989 to April 1993 the trial was based on a
complete factorial design with three replicates and 3 manure treatments
and three cropping systems. The net plots measured 5.0 m x 5.0 m
and were well protected from run-off or external water erosion by having
a cut-off drains on the upper side of the experimental field. Some
terraces were also constructed along the contours between blocks to
control the soil movement from the upper to lower blocks/plots. The
Soil Organic Matter (SOM): The Basis for Improved Crop Production in Arid and SemiArid Climates of Eastern Kenya
243
treatments initially adapted for the study were, no inputs (C3M0),
manure application at 10 tons ha -1 (C3M1) and 5 tons ha -1 (C3M2)
continuous cropping systems. The cropping systems (C3M0, C3M1
and C3M2) were rotations of [sorghums (Sorghum bicolor, var. 954066)
+ cowpea (Vigna unguiculata, var. M66)] and [peal millets (Pennisentum
typhides, var. KPM-1) + grams (Vigna aures, var. N26)] intercrops in
November and April seasons, respectively. Maize (Zea maize, var.
Katumani) as a test crop was introduced in 1999 November season
and was grown both seasons replacing sorghum and millet as test
crops. In 1993, four years after the start of the trial, annual manure
application was discontinued in C3M1 (10 tons ha-1 manure) and C3M2
(5 tons ha-1 manure) to form treatments C3R1 and C3R2 to respectively
assess the 10 and 5 tons ha -1 residual manure effects on crop dry
matter yields (Table 17.1). Another treatment (C3F) of annual inorganic
fertilizer at the rate of 51 and 12kg ha -1 of N and P yearly, but splitted
in equal amounts between April and November rains was also
introduced in the former C3M0 (control) to assess the benefits of crop
production using organic over inorganic fertilizers. The rate of applied
N and P was equivalent to nutrient contribution by 5 tons ha-1 manure
treatment.
Table 17.1: Soil fertility management treatments for both field observations and APSIM
model simulation on crop yields performance
Field
Code
Treatment description
C3M0
C3M2
C3M1
C3F (ex- C1M0)
C3R2 (ex-C1M2)
C3R1 (ex-C1M1)
No external input
5 tons ha -1 annual manure application
10 tons ha -1 annual manure application
Inorganic fertilizer (N and P) from 1993
Residual manure at 5 tons ha -1 (from 1993)
Residual manure at 10 tons ha -1 (from 1993)
Planting of all crops was done at the on-set of rains to make sure
that the crops benefited from the low and erratic rains experienced within
the trial site. Other agronomic practices (weeding, pest control and
harvesting) were carried out as per local recommendations and except
for the grains, other crop residues were returned into their respective
plots at the end of every season. The aboveground biomass (stovers)
were cut at the ground level, chopped before being incorporated into
the soil during land preparation. Data on weather, crop biophysical and
soil physical and chemical parameters were collected as part of APSIM
model inputs.
244
Micheni, A. et al
Results
Cumulative crop dry matter (DM) responses were improved by manure
application at 5 and 10 tons ha -1 (Table 17.2). Over time cultivation
without application of manure showed a decline in crop DM (Figure
17.1a). After a period of 20 growing seasons, the cumulative mean DM
from 5 tons ha-1 (C3M2) was 3,435 kg ha -1 and 4,141kg ha-1 from 10
tons ha-1 manure year-1 (Figure 17.1b). Continuous cultivation without
application of manure (C3M0) had the lowest average crop DM of 989
kg ha -1. This was about four times less than the highest recorded DM
(4,141 kg ha-1) from 10 tons ha-1 continuous annual manure application.
The problem was associated to overtime run-down of nutrients during
crop removal and aggravated soil erosion. Like the cumulative (10 years)
average crop yield response. There was no significant (p=0.05) difference
in crop yields between 5 and 10 tons ha -1 annual manure rates. The
average crop DM was significantly (p=0.05) different between all
treatments (C3F, C3M2 and C3M1) that had received external fertility
inputs, including manure residuals (C3R1 and C3R2) and absolute
control treatment (C3M0). The responses to residual manure at 10 and
5 tons treatments (C3R1 and C3R2) were also not significantly different
(p=0.05). There was a general decline trends in DM production from
1993 when manure application was stopped to November 2000 cropping
season when the last observations were done (Figure 17.1d).
Table 17.2: Cumulative average crop dry matter (DM) yields under different soil fertility
managements
Treatment description
Years manure
applied
Cumulative DM yields
No. seasons kg ha -1
observed
No external input (C3M0)
0
20
989
5 tons ha-1 annual manure
application (C3M2)
10
20
3435
10 tons ha -1 annual manure
application (C3M1)
10
20
4141
Inorganic fertilizer (C3F)
0
11
3723
Residual manure at 5 tons ha -1 (C3R2)
Residual manure at 10 tons ha -1 (C3R1)
4~
4~
11
11
3499
2677
~ Before manure residual effect study was initiated
Soil Organic Matter (SOM): The Basis for Improved Crop Production in Arid and SemiArid Climates of Eastern Kenya
245
Figure 17.1a: Comparison of field observations and APSIM simulated crop dry matter
(DM) for 1989-2000 Machang’a/cultivation without application of organic or mineral
fertilizers
3.5
C3M0 (Observed)
C3M0 (Simulated)
3
2.5
2
1.5
1
0.5
0
1989n 1990a 1990n 1991a 1991n 1992a 1992n 1993a 1993n 1994a 1994n 1995a 1995n 1996a 1996n 1997a 1997n 1998a 1998n 1999a 1999n 2000a
Season (a = April rains; n = November rains)
Figure 17.1b: Comparison of field observations and APSIM simulated crop dry matter
(DM) for 1989-2000 10 tons/ ha long-term manure application experiment
18
16
14
12
C3M1 (Observed)
10
C3M1 (Simulated)
8
6
4
2
0
1989n1990a1990n1991a1991n1992a1992n1993a1993n1994a1994n1995a1995n1996a1996n1997a1997n1998a1998n1999a1999n2000a
Season (a=April rains; n=november rains)
Figure 17.1c: Comparison of field observations and APSIM simulated crop dry matter
(DM) for 1989-2000 cultivation without manure application followed by mineral fertilizer
application (1994-2000)
14
12
10
8
C3M0 (Observed)
C3F (Observed)
C3F (Simulated)
6
4
2
0
1989n1990a1990n1991a1991n1992a1992n1993a1993n1994a1994n1995a1995n1996a1996n1997a1997n1998a1998n1999a1999n2000a
Season (a=April rains; n=november rains)
246
Micheni, A. et al
Figure 17.1d: Comparison of field observations and APSIM simulated crop dry matter
(DM) for 1989-2000 cultivation with manure application at 10 ton/ha followed by
assessment of manure residue effect (1994-2000) on crop performance
18
16
14
C3M1 (Observed)
12
C3R1 (Observed)
10
C3R1 (Simulated)
8
6
Crop DM
4 (tons/ha)
2
0
Yields in the residue treatments remained higher compared to the
control (C3M0) treatment whose yields were the least (Figure 17.1d).
Application of inorganic fertilizer (C3F) tremendously increased crop
DM to almost three folds relative to the no inputs treatment (Figure
17.1c). Observed crop DM yields from all the treatments were compared
with simulations from APSIM model. Crop DM yields of 15.6 tons ha-1
was recorded from the fields while the model predicted 7.5 tons ha-1 for
the 10 tons ha-1 annual manure application in 1994. This shows that
the model cannot fully be depended on in prediction of crop responses
to various soil fertility management options for the smallholder farmers.
However, the model may be used to provide the trends on future scientific
expected output(s) and predictions for on going or to be initiated studies.
Conclusions
Enhancement of soil productivity through the improvement of SOM is
essential for sustained agricultural production systems. This is
particularly important in ASAL where rainfall is erratic and soils are
low in most of the major nutrients needed by plants, and continuous
cultivation with little or no external soil fertility inputs is a widespread
practice. The study indicated that the annual manure application had
positive response to crop dry matter (DM) production. Cumulative mean
crop DM production after 20 seasons from 5 tons ha-1 and 10 tons ha-1
manure application did not differ significantly and therefore a
recommendation was put forwards to ASALs farmers to apply 5 tons
ha-1 manure in erosion free continuously cultivated lands. Manure
Soil Organic Matter (SOM): The Basis for Improved Crop Production in Arid and SemiArid Climates of Eastern Kenya
247
residual effects were monitored for 11 seasons after four years of annual
manure application and 5 tons ha-1 residual recorded a cumulative crop
DM of 3499 kg ha-1 compared to 2677 kg ha-1 from 10 tons ha-1 manure
residual. Discontinuation of manure application led to a decrease in
crop yields. This is probably due to the effect of nutrient run-down
through continuous cropping without application of manure or mineral
fertilizers. Non-application of mineral fertilizer or manure had negative
response to crop yields. Field crop yield observations and APSIM model
simulations had some positive correlation in terms of trends but not on
the actual values. This suggested that the model cannot fully be relied
upon in provision of true field situations, but on general trends of
scientific scenarios.
Acknowledgements
We wish to thank the DFID/NRI (UK), the Rockefeller foundation and
the University of Reading (UK) for their financial and technical support.
Much thanks to Tropical Soil Biology and Fertility (TSBF)/AfNet for their
contributions and collaborative efforts. We also appreciate the technical
support from ACIAR/CSIRO, the APSIM modellers and others that in
one way or the other played a positive role in trial implementation and
preparation of this paper.
References
Bunting, A. H. 1992 Feeding the world in the future. In: Spending, C. R. W.
(eds). Farmer’s Principles of Food and Agricultural (17th edition), pp. 256 –
290. Oxford: Blackkwell Scientific Publications.
Cambardella, C.A. and Elliot, E.T. 1993) Carbon and nitrogen distribution in
aggregates from cultivated and native grassland soils. Soil Science Society
of America Journal 57: 1071 – 1076.
Coen, R., Haverkort, B. and Ann, W.B. (1992) Farming for the future: An
introduction to low-external input and sustainable agriculture (ILEIA),
Macmillan, pp. 24-208.
DAREP, 1995. Farming in Tharaka; an informal diagnostic survey report of the
semi-arid areas of Tharaka/Nithi District. KARI/KEFRI/NRI, DAREP-Embu,
Kenya.
De Ridder, N. and Van Keulen, H. (1990) Some aspects of the role of organic
matter in sustainable intensified arable farming systems in the West African
semi-arid tropics. Fert. Res. 26: 299-310.
Gibberd, V. (1995) Crop Production Workbook for Arid and Semi- arid Land.
Chatham, UK: Natural Resources Institute.
Irungu, J.W., Micheni, A.N., Sutherland A.J. and Warren G.P. (1997) An appraisal
of soil fertility in smallholder farms in the semi-arid areas of Tharaka-Nithi
248
Micheni, A. et al
District: Farmers assessment compared to laboratory analysis. In:
Proceedings of the Conference on participatory Dryland Agricultural Research
East of Mount Kenya, January 21-24, 1997. KARI: Kitale, Kenya. Pages
243-248.
Ikombo, B.M. (1984) Effects of farmyard manure and fertilizers on maize in
semi-arid areas of eastern Kenya. East. Afr. Agric. for. J. 44: 266-274.
Jaetzold, R. and Schemidt H. (1983) Farm management hardbook of Kenya.
Natural conditions and farm information. vol.11/ C. East Kenya. Ministry of
agriculture, Kenya.
Kihanda, F.M. (1998) Improvement of FYM quality through composing with
high quality organic residues. Research proposal funded by the African
Science Career Awards of the Rockefeller Foundation (1998-1999). KARI,
Nairobi, Kenya.
Lal, R. and Stewart, B.A. (1995) Soil management, Experimental Basis for
Sustainability and Environmental Quality. Advances in Soil Science, Lewis,
London.
Micheni, A.N. (1996) The role of legumes in soil fertility maintenance in arid
and semi-arid areas (Reference to Tharaka, Kenya); MSc. thesis, University
of Reading, UK.
Mokwunye, A.U., de Jager, A and Smaling, E.M.A., (Eds.) (1996) Restoring and
Maintaining the Productivity of West African Soil: Key to sustainable
Development. In: Miscellaneous Fertilizer Studies No. 14.
Okoba, B.O. and Altshul, H.J. (1995) A report of a PRA in Gachoka Division,
Kenya. KARI - Embu, Kenya
Rowell, D.L. (1994) Soil science: Methods and applications. Longman. U.K., pp.
266-276.
Smaling, E.M.A., Nandwa, S.M. and Janssen, B.H. (1997) Soil fertility in Africa
is at stake. In: Buresh,R.J, Sanchez, P.A. and Calhoum, F.(eds.) ,
Replenishing Soil Fertility in Africa. SSSA Special Publication No.51.
Wisconsin, USA. Pp. 47-61.
Warren, G.P. (1998) Effects of continuous manure application on grain yields at
seven sites of trial 1, from 1993 to 1996. In: Maintenance of soil fertility and
organic matter in dryland areas. (Eds: Kihanda, F.M. and Warren G.P.). The
department of soil science, the University of Reading Occasional publication
No. 3. The University of Reading, UK.
Warren, G.P. (1998) Effects of native phosphorus on biodiversity, productivity
and the dynamics of soil carbon and nitrogen at Machang’a and Mutuobare
sites. In: Maintenance of soil fertility and organic matter in dryland areas.
(Eds: Kihanda, F.M. and Warren G.P.). The department of soil science, the
University of Reading, Occasional publication No. 3. The University of
Reading, UK.
Warren, G.P. Irungu J.W. and Muthamia J. (1998) Residual value of manure in
Trial 1. In: Maintenance of soil fertility and organic matter in dryland areas.
(Eds: Kihanda, F.M. and Warren G.P.). The department of soil science, the
University of Reading,. Occasional publication No. 3. The University of
Reading, UK.
Response of Tephrosia vogelii to Minjingu Phosphate Rock Application on a Ferralsol
of Varying Soil pH
Response of Tephrosia
vogelii to Minjingu Phosphate
Rock Application on a
Ferralsol of Varying Soil pH
249
18
Mkangwa, C.Z.1, Semoka, J.M.R.1 and
Maliondo, S.M.S.2
1
Soil Science Department, P.O. Box 3008, Sokoine
University of Agriculture, Tanzania,
2
Forest Biology Department, P.O. Box 3000, Sokoine
University of Agriculture, Tanzania
Abstract
Low available P and low soil pH are among soil conditions,
which could lead to poor growth of some N2 fixing plants.
The effect of these two soil conditions to Tephrosia vogelii
(agroforestry leguminous shrub) growth parameters needs
to be determined. A Glasshouse study using T. vogelii was
conducted on a Ferralsol with soil pH 5.0 and 5.9 and low
available P (2.5-2.9 mgP kg -1). The objectives of this study
were to assess the effects of soil pH and Minjingu Phosphate
Rock (MPR) on T. vogelii plant height, dry matter production
and shoot N and P contents. The two soils were treated
with P as MPR and planted with T. vogelii for twelve weeks.
The pots were replicated six times and arranged in a
completely randomized design. The parameters assessed
include plant height and dry matter yields at three and
twelve weeks and shoot N and P contents at twelve weeks.
250
Mkangwa, C.Z. et al
Soil pH 5.9 had significant (P≤ 0.05) effect on plant height,
dry matter yields, shoot N and P concentration and shoot N
and P uptakes. Application of P as MPR significantly (P≤ 0.05)
increased these parameters on both soil pH values. The results
suggest that in strongly acid soils with low available P, T.
vogelii biomass will be improved by addition of MPR.
Key words: T. vogelii, soil pH, Ferralsol, Minjingu phosphate rock, plant
height, dry matter yields, shoot N and P concentrations and shoot N
and P uptakes
Introduction
Low soil pH and inadequate levels of N and P are common features in
highly weathered tropical soils (Fox et al., 1985; National Soil Service,
1989), and are among major factors limiting crop production in subSaharan Africa (Eswaran et al., 1997). Soil pH per se is not the growth
limiting factor but rather one or more secondary factors, which are pH
dependent (Mengel and Kirkby, 1982). Some of the pH dependent
secondary factors, which limit plant growth at low soil pH include Al
and Mn toxicity, fixation and hence unavailability of P and deficiency of
Ca and Mo. Others are depression of the activity of microorganisms
involved in nitrification and N fixation and inhibition of the activity of
symbiotic microorganisms, which lead to poor nodulation of some
legumes (Brady, 1984). The situation is even worse when such soils are
continuously cultivated without addition of any fertilizers. The overall
consequences are low crop yields, persistent food insecurity, malnutrition
and wide spread rural poverty.
In order to remain productive, such soils need to be carefully
managed. Use of fallows (particularly improved fallows) - a common
feature of small scale farming throughout the tropics can achieve this
goal. Improved fallows consists of deliberately planted species-usually
legumes with the primary purpose of fixing N2 as part of a crop-fallow
rotation (Sanchez, 1999). Improved fallows have the advantage of insitu accumulation of biomass, optimisation of nutrient cycling, enhancing
soil biological activities and maximising the use efficiency of minimal
external inputs (Sanchez, 1994). T. vogelii Hook. F. is one of the
indigenous leguminous shrubs with high potential of improving soil
fertility when used in improved fallow situations (Balasubramanian and
Sekayange, 1992). This species is a valuable leguminous cover crop in
grass areas, relatively resistant to periodic fires, unpalatable to animals
and hence a suitable fallow species in places where livestock are
traditionally allowed to graze crop residues in the fields after harvest
Response of Tephrosia vogelii to Minjingu Phosphate Rock Application on a Ferralsol
of Varying Soil pH
251
(ICRAF, 1993). As a fallow species, T. vogelii has been reported to increase
significantly yields of maize in Tanzania (Mgangamundo, 2000), maize
in Malawi and Zambia (Kwesiga et al., 1999) and maize, sorghum and
beans in Rwanda (Balasubramanian and Sekayange, 1992; Hagedorn
et al., 1997).
The soil fertility improvement potential of T. vogelii fallow has been
noted in Nyabisindu, Rwanda (Prinz, 1986) and in Gairo, Tanzania
(Mgangamundo, 2000). The chemical constituents of T. vogelii foliar
biomass as analysed in different laboratories are 2.85-4.0 % N, 0.38 %
P, 1.03 % K, 1.89 % Ca, 0.16 % Mg, 8.0-8.3 % lignin, 2.37 % polyphenols,
21.1 % cellulose and 0.97 % retonones (Hagedorn et al., 1997; Mutuo et
al., 1998; TSBF, 1999). However, poor growth of T. vogelii at Morogoro
in Tanzania has been reported (Mugasha, 1999 Personal
Communication). Also, Ngazi and Kapinga (1998) reported low cotton
and cassava yields following a one year of T. vogelii fallow at Ukiriguru,
Tanzania. Low cotton and cassava yields were attributed to low biomass
produced by yield T. vogelii. Inadequate levels of available P at low soil
pH could be the causal factors. Donald and Williams (1955) noted that
generally legume growth and Rhizobium symbiosis are sensitive to
available P. Phosphorus deficiency reduces nodulation, N2 fixation and
plant growth. Research elsewhere reported that low available P limits
N2 fixation and growth of Leucaena lucocephala and Sesbania goetzei
(Luyindula and Haque, 1992). Hence in order to optimise the benefits
from T. vogelii fallows, it is important to identify soil factors that may
limit its performance. Based on this background, a glasshouse
experiment was conducted using two soils with Bray-I low available P
(2.5-2.9-mgP kg -1) and low soil pH (pH 5.0 and 5.9), applied with and
without P as MPR and T. vogelii as a test agroforestry species. The
objectives of the experiment were as follows:
• To assess the effect of low soil pH on T. vogelii growth and shoot N
and P contents.
• To evaluate the effect of MPR applications on such soils on T. vogelii
growth and shoot N and P contents.
Materials and Methods
General description of the study area
The soils for the glasshouse trial were obtained from experimental site
located in the Sokoine University of Agriculture (SUA) Farm, in Tanzania
at longitude 37°39'12.4''E and latitude 06°50'24.5''S, and an elevation of
540 m above sea level. The experimental site has a slope of about 1.5 2%. Prior to field experimentation, the soils of the site were classified by
using both World Reference Base for Soil Resources (1998) as Hyperdystri-
252
Mkangwa, C.Z. et al
Umbric Ferralsol and by Soil Taxonomy system (Soil Survey Staff, 1998)
as Typic Haplustox (Table 18.1). The soil pH of the site is variable and
thus two soil samples, one with pH 5.0 and another with pH 5.9 were
collected from different areas of the field and used for the glasshouse
study. The two soils were analysed for selected chemical and physical
properties using standard analytical methods as described by Okalebo et
al. (1993) and the results are presented on Table 18.2. The glasshouse
study was conducted between December and March 2001.
Table 18.1: Initial soil physical and chemical properties of trial site
Parameter
Sand (%)
Silt (%)
Clay (%)
Textural class
Soil pH (water)
Soil pH (CaCl 2)
Organic Carbon (%)
Total N (%)
C:N ratio
P (Bray-1 method) (mg kg -1)
Exchangeable Bases (me 100g -1)
Ca
Mg
K
Na
Cu (mg kg-1)
Zn (mg kg-1)
B (mg kg-1)
Exch. Al3+ (me100g -1)
Exch. H+ (me100g -1 )
Total Acidity (me100g-1 )
Magnitude
36
10
54
Clay
5.06
4.60
1.3
0.07
18.6
2.1
4.4
2.3
0.68
0.18
1.13
0.63
Nd
0.005
0.11
0.115
Glasshouse study
Five-litre plastic pots were filled with 5kg soil (air-dry weight) that was
sieved through 8mm sieve. Two levels of P viz. 0 and 400mg P kg-1 soil
were tested as MPR in a completely randomized design replicated six
times. Basal applications of potassium (K), magnesium (Mg), and zinc
(Zn) were made at the rate of 50 mg kg-1 soil K as K2SO4, 25 mg Mg kg-1
soil as MgSO4 and 5 mg Zn kg-1 soil as ZnSO4 kg-1 soil. Also, a starter N
dose as NH4SO4 at 20 mg N kg-1 soil was applied because the N and O.C
levels were < 0.12 and < 1.5 %, respectively. The fertilizers were
thoroughly mixed with soil by hand and five T. vogelii seeds planted in
Response of Tephrosia vogelii to Minjingu Phosphate Rock Application on a Ferralsol
of Varying Soil pH
253
each pot and then watered with deionized water to 80 % of the field
capacity by weight as described by Klute (1986). The seedlings were
thinned to two plants per pot one week after germination (WAG). T.
vogelii plant height and dry matter yield data were collected at 3 and 12
WAG and N and P contents analysed in the plant samples collected at
12 WAG. Dry matter yield was obtained by cutting the plants at 2.0 cm
above the soil, then washed to remove the adhering soil particles, weighed
and then dried in an oven at 70°C to constant weight. The dried plant
materials were weighed, cut into small pieces of about 1.0 x 0.5 cm and
ground by Sample Mill to pass through 0.5mm sieve.
Table 18.2: Initial properties of two soils used in the Glasshouse study
Soil property
Soil 1
Soil 2
Sand (%)
Silt (%)
Clay (%)
Textural class
Soil pH (water)
Soil pH (CaCl 2)
Organic carbon (%)
Total N (%)
C:N ratio
P (mgkg-1)
Exchangeable bases (me100g -1)
Ca (me100g-1)
Mg (me100g -1)
Na (me100g-1)
K (me100g-1)
Cu (mgkg-1)
Zn (mgkg-1)
36
10
54
clay
3.0
1.63
1.3
0.07
18.6
2.5
34
10
56
clay
3.9
1.60
1.2
0.07
17.1
2.9
3.0
1.63
0.23
0.9
1.3
0.64
3.9
1.60
0.19
0.68
1.2
0.63
Laboratory Analyses
Laboratory methods as described by Okalebo et al. (1993) were used for
both soil and plant analysis. Soil pH was measured using 1:2.5 soilwater and 1M-potassium chloride mixture using a pH meter. Available
P was extracted using the Bray-1 reagent and was determined
colourimetrically after developing colour with ascorbic acid. Organic
carbon was determined by the wet digestion method of Walkley-Black
method. Total N was determined by the macro-Kjeldahl digestiondistillation method. Cation exchange capacity (CEC) was determined
by the ammonium saturated method. Available Cu and Zn were extracted
by the DTPA and their concentrations determined by AAS. Particle size
analysis was done by the hydrometer method.
254
Mkangwa, C.Z. et al
Phosphorus in plant samples was analysed using dry ashing method
and determined colourimetrically while calcium was determined by AAS.
Statistical analyses
The data were analysed by MSTAT-C using completely randomized
design. Significant means were separated using Duncan’s New Multiple
Range Test.
Results and Discussion
Plant height
Plant height of T. vogelii at three and twelve weeks as affected by soil pH
and P application is given on Table 18.3. At three weeks, plant height
on the soils without P applications was lower (11.62 cm plant-1), on soil
with pH 5.0 compared to soil with pH 5.9 (17.97 cm plant -1), and at
twelve weeks plant heights were 81.33 and 99.76 cm plant-1for soil pH
5.0 and 5.9, respectively. The plant height for the soil with pH 5.9 without
P application at 3WAG was higher (17.97 cm plant-1) than that obtained
on soil with pH 5.0 (15.6 cm plant-1) with P application. The plant heights
observed on soil that had pH 5.0 without P application were significantly
(P≤0.05) different both at 3WAG and at 12WAG. Also, the plant heights
for the soil pH 5.9 without P application and pH 5.0 with P application
at 3WAG were statistically (P≤0.05) different. Low plant heights at soil
pH 5.0 with P application at 3WAG is probably due to inadequate levels
of some nutrients especially Ca (3.0 me100g-1) as compared to soil
with pH 5.9 (3.9 me100g-1) (Table 18.2). Application of P on soil with
slightly higher levels of Ca and a lower level of Al 3+ at 3WAG led to
higher plant height (23.2 cm plant-1) which was significantly (P≤ 0.05)
different from the other treatments.
Table 18.3: Effect of soil pH and MPR applications on plant height (cm plant-1) of T. vogelii
Treatment
3WAG
12WAG
Soil pH 5.0 -MPR
Soil pH 5.0 +MPR
Soil pH 5.9 -MPR
Soil pH 5.9 +MPR
LSD (0.05)
Std error
C.V. (%)
11.62 c
15.60 c
17.97 b
23.20 a
1.53
0.44
5.81
81.33 b
93.00 a
99.77 a
104.37a
11.56
3.341
6.12
Response of Tephrosia vogelii to Minjingu Phosphate Rock Application on a Ferralsol
of Varying Soil pH
255
The T. vogelii plant heights at three weeks on soil with pH 5.0 were
15.6 and 11.62 cm plant –1 for with and without P application,
respectively. The plant heights for twelve weeks on soil with pH 5.0
were 93.03 cm plant–1 with P and 81.3 cm plant–1 without P application.
P applications did not significantly increase plant heights at three weeks
but significantly (P ≤ 0.05) increased it at twelve weeks possibly due to
increased Ca uptake and P dissolution from MPR (Rajan et al., 1996) at
12 weeks.
At twelve weeks, application of P on soil with pH 5.9 increased plant
height from 99.7 to 104.3 cm plant–1. The increase in plant height for
soil with pH 5.0 due to P application was statistically comparable to
plant height obtained on soil with pH 5.9 without P application. The
small increase in plant height on soil pH 5.0 that received P was caused
by strong soil acidity, which depressed the activity of microorganisms
involved in various activities (including nitrification) in the rhizosphere
(Brady, 1984; Mengel and Kirkby, 1982).
Dry matter yield
The dry matter yield (DYM) of T. vogelii accumulated at three and twelve
weeks as affected by soil pH and P application is shown on Table 18.4.
At three weeks, P application on soils with pH 5.0 increased DYM from
0.75 to 1.2 gpot-1 and for the soil with pH 5.9 DYM increased from 1.2
and 1.8 gpot-1. Soil at pH 5.0 without P application had significantly
(P≤0.05) lower dry matter production than the other treatments. The
dry matter production from the soil with pH 5.9 applied with P was
significantly (P≤0.05) higher than the other treatments.
Table 18.4: Effect of soil pH and MPR applications on dry matter production (gpot-1) of T.
vogelii
Treatment
3WAG
12WAG
Soil pH 5.0 -MPR
Soil pH 5.0 +MPR
Soil pH 5.9 -MPR
Soil pH 5.9 +MPR
LSD (0.05)
Std error
C.V. (%)
0.75c
1.20 b
1.20 b
1.80 a
0.21
0.061
8.40
17.90c
34.80b
37.63b
51.33a
7.98
2.306
11.28
At twelve weeks, the DMY was doubled by the application of P on
the soil with pH 5.0. The dry matter production were 17.9 and 34.8g pot-1
with and without P application, respectively. The DMY at soil pH 5.9
was also increased from 37.63 without P treatment to 51.3 gpot -1 in
pots that were applied with P. The difference was significant (P≤0.05).
256
Mkangwa, C.Z. et al
The observations made by Fox et al. (1985), Aggarwal, (1994) and
Giller et al. (1998) using different N2 fixing species are similar to the
results obtained in this study with T. vogelii. Fox et al. (1985) in Hawaii
assessed the growth response of L. lucocephala to varying soil pH that
ranged from < 5 to > 7 and found that relative yield increased with
increasing soil pH until above pH 7. The increased relative yield was
attributed to improved Ca nutrition that was associated with increasing
soil pH in the range of 6 to 7. Aggarwal (1994), assessed 15 bean varieties
on limed soils with soil pH 4.6-5.0 and low available P 0.35-1.40 in
Malawi and reported a linear increase in the nodule number and grain
yield with increased lime application up-to 75 % level of Al neutralisation.
Giller et al. (1998) observed that addition of P fertilizer (26 kgPha-1) on
soils with pH ranging from 5.8-7.0 and available P 0.2-6.6 mg Pkg -1
dramatically increased the number of root nodules, N and seed yields of
Phaseolus vulgaris in farmers fields in the West Usambara Mountains
in northern Tanzania.
Balasubramanian and Sekayange (1992) in Rwanda obtained
contradicting results with other N 2 fixing species. These workers
compared the responses of Crotalaria ochroleuca, Mucuna pruriens,
Cajanus cajan and Sesbania sesban to P applications (9-40 kgPha-1) on
a soil with pH 5.1 and 7 mg kg –1 of Bray II P and reported that P
applications had no effect on the biomass production.
Shoot N and P concentration and uptake at twelve weeks
Table 18.5 gives the shoot N and P contents (%) and their uptakes
(g pot-1) as influenced by soil pH and P application of twelve weeks old T.
vogelii. MPR application increased shoots N content both at pH 5.0 and
pH 5.9. At pH 5.0, N concentration increased from 2.3% in the pots that
were not applied with P to 3.5 % in the pots that were applied with P
while at pH 5.9 the corresponding increase was from 2.5 to 3.2 %. The
N concentration for the soil of pH 5.9 treated with MPR was lower than
the comparable treatment of pH 5.0. However, the N uptakes (g pot -1)
for the soil of pH 5.9 treated with MPR was higher than the comparable
treatment of pH 5.0 confirming that there was more available P after
MPR application. The N uptakes (g pot-1) for the soil of pH 5.9 treated
with MPR was higher because at this soil pH level the activities of microorganisms involved in various processes like nitrification are increased
(Brady, 1984; Mengel and Kirkby, 1982).
On both soil pH levels, the shoot N concentrations were
significantly (P ≤0.05) increased due to P application. Similar
observation was made on P. vulgaris. Addition of P fertilizer (26kgPha 1
) on soils with pH ranging from 5.8-7.0 and available P (0.2-6.6 mg
Pkg -1) dramatically increased N content of P. vulgaris (Giller et al.,
Response of Tephrosia vogelii to Minjingu Phosphate Rock Application on a Ferralsol
of Varying Soil pH
257
1998). The shoot N content for both soils when P was not applied
was statistically the same, which suggests that N 2 fixation was
depressed due to possibly one or more secondary factors which are
pH dependent (Mengel and Kirkby, 1982). Contrary to the observation
made in this study Balasubramanian and Sekayange (1992), reported
that P applications to C. ochroleuca, M. pruriens, C. cajan and S. sesban
(9-40 kgPha -1) on a soil with pH 5.1 and 7 mg kg –1 of Bray II P had no
effect on N content of these fallow species.
Table 18.5: Effect of soil pH and MPR applications on shoot N and P concentration (%)
and uptake (g pot -1) of T. vogelii after twelve weeks
Treatment
Soil pH 5.0 -MPR
Soil pH 5.0 +MPR
Soil pH 5.9 -MPR
Soil pH 5.9 +MPR
LSD (0.05)
Std error
C.V. (%)
Shoot concentrations (%)
N and P uptake (g pot -1)
N
P
N
P
2.32c
3.53a
2.53c
3.21b
0.32
0.091
5.47
0.22d
0.32b
0.25c
0.35a
0.02
0.006
7.30
0.43c
1.26a
0.95b
1.28a
0.039
0.018
2.01
0.04c
0.12ab
0.09bc
0.18a
0.006
0.018
2.62
The shoot P concentrations and P uptakes as influenced by soil pH
and P application at twelfth week are presented on Table 18.5. MPR
application increased shoot P concentration at soil pH 5.0 from 0.22
% to 0.32 % for pots that were not treated with P and for pots that
were treated with P, respectively. At soil pH 5.9, MPR application
increased shoot P concentration from 0.25 % for pots that were not
treated with P to 0.35 % for pots that were treated with P. These shoot
P concentrations were significantly (P≤0.05) different, with the highest
shoot P concentration on soil with 5.9. T. vogelii P uptake at twelfth
week was also increased with MPR application on both soil pH levels
tested. However, the P uptakes at both soil pH levels when MPR was
applied were not significantly (P≤0.05) different suggesting that P is
necessary for T. vogelii in soils with low pH. The P uptakes for the soil
of pH 5.0 in pots that were treated with P were statistically similar to
pots that were not treated with P at soil pH 5.9. Similar observations
were made on Tritcum aestivum and Lupinus albus. Kamh et al. (1999)
reported significant (P≤0.05) N and P uptakes and shoot dry weight of
T. aestivum and L. albus after P application as Ca(H 2PHO 4)2.
258
Mkangwa, C.Z. et al
Conclusion
From this pot experiment, the following conclusions can be made:
• On soils low in pH (strongly acid soil) and low available P, T. vogelii
performance is appreciably reduced.
• On soils with low soil pH and low levels of available P, improved T.
vogelii performance is observed when P levels are increased.
• Applications of P as MPR significantly increase T. vogelii plant height,
dry matter yield, shoot N and P concentrations and shoot N and P
uptakes. However, the increase of these parameters on a more
strongly acid soil (pH 5.0) applied with MPR was lower than that
obtained on soil pH 5.9, suggesting that P was not the only nutrient
element limiting T. vogelii performance on soil with pH 5.0.
Acknowledgements
The authors sincerely acknowledge the Tanzania Agricultural Research
Project Phase II for financing this research work.
References
Aggarwal, V.D. (1994) An overview of SADC/CIAT bean activities in Malawi.
Bean Research 8, 15 - 31.
Balasubramanian, V. and Sekayange, L. (1992) Five years research on improved
fallow in the semi-arid highlands of Rwanda. In: (eds Mulongoy, K., Gueye,
M. and Spencer, D.S.C.) Biological Nitrogen Fixation and Sustainability of
Tropical Agriculture. International Institute of Tropical Agriculture. A WileySayce Co-Publication, pp 405-422.
Brady, N.C. (1984) The Nature and properties of soils. 9th Edition, MacMillan
Publishing Company, New York. pp 206-209.
Donald, C.M. and Williams, C.H. (1955) Fertility and productivity of a podsolic
soil, as influenced by subterranean clover (Trifolium subterraneum L.) and
superphosphate. Soil Biology and Biochemistry 5: 83-89.
Eswaran, H., Almaraz, R., van den Berg, E. and Reich, P. (1997) An assessment
of the soil resources of Africa in relation to productivity. Geoderma 77, 1 18.
Fox, R.L., Yost, R.S., Saidy, N.A. and Kang, B.T. (1985) Nutritional complexities
associated with pH variables in humid tropical soils. Soil Science Society of
America Journal 49:1475-1480.
Kamh, M., Horst, W.J., Amer, F., Mostafa, H. and Maier, P. (1999) Mobilization
of soil and fertilizer phosphate by cover crops. Plant and Soil 211: 19-27.
Giller, K.E., Amijee, F., Brodrick, S.J. and Edje, O.T. (1998) Environmental
constraints to nodulation and nitrogen fixation of Phaseolus vulgaris L. in
Response of Tephrosia vogelii to Minjingu Phosphate Rock Application on a Ferralsol
of Varying Soil pH
259
Tanzania. II Response to N and P fertilizers and inoculation with Rhizobium.
African Crop Science Journal 6: 171-178.
Hagedorn, F., Steiner, K.G., Sekayange, L. and Zech, W. (1997) Effect of rainfall
pattern on N mineralization and leaching in a greenhouse experiment in
Rwanda. Plant and Soil 195: 365-375.
Annual report 1992. ICRAF, (1993) International Centre for Research in
Agroforestry, Nairobi, Kenya. pp 208.
Klute, A. (1986) Water Retention: Laboratory Methods. In: (ed) Methods of soil
Analysis. Part 1. Physical and Mineralogical Methods. 2nd edition (Klute,
A.) Number 9 (1) in Series of Agronomy. pp 635 - 660.
Kwesiga, F.R., Franzel, S., Place, F., Phiri, D. and Simwanza, C.P. (1999) Sesbania
sesban improved fallows in Eastern Zambia: Their inception, development
and farmer enthusiasm. Agroforestry Systems 47: 49-66.
Luyindula, N. and Haque, I. (1992) Effect of Rhizobium innoculation and P on
growth and N fixation in tree legumes grown on highland Vertisols. In: eds
Mulongoy, K., Gueye, M. and Spencer, D.S.C.) Biological N Fixation and
Sustainability of Tropical Agriculture. IITA, Nigeria. pp 109-112.
Mengel, K. and Kirkby, E.A. (1982) Principles of plant nutrition. 3rd Edition,
International Potash Institute – Bern, Switzerland. pp 54-61.
Mgangamundo, M.A.(2000) Effect of fallow periods of Cajanus cajan, Sesbania
sesban and Tephrosia vogelii on soil fertility improvement and maize and
fuelwood production at Gairo in Morogoro, Tanzania. MSc Thesis, Sokoine
University of Agriculture, Morogoro, Tanzania. pp 51-67.
Mutuo, P., Marandu, A.E., Rabeson, R. and Mwale, M. (1998) N fertilizer
equivalency values of organic materials of differing quality: Network trial
results from East and southern Africa, season 1. In: Proceedings of the 24th
East Africa Soil Science Society Annual General Meeting. 13 - 19 December,
Tanga, Tanzania. pp 88 -101.
National Soil Service, (1989) Effect of continuous cultivation and fertilizer
applications on yields of maize and soil characteristics of a Ferralsol-Luvisol
catena on Gneiss in North eastern Tanzania. Mlingano Agricultural Research
Institute, Tanga Tanzania. Soil Fertility Report F 5. pp 1-10.
Ngazi, H. and Kapinga, R. (1998) Effects of fallow species and length of fallow
on soil fertility and cotton yield improvements on sandy loam soils. Annual
Progress Report 1998, Ukiriguru, Mwanza, Tanzania. pp 8-9.
Okalebo, J.R., Gathua, K.W. and Woomer, P.L. (1993) Laboratory Methods of
soil and Plant analysis. A working manual. KARI, SSEA, TSBF, UNESCOROSTA, Nairobi. pp 88.
Prinz, D. (1986) Increasing productivity of smallholder farming
system by introduction of planted fallows. Plant Research Division 24: 3156.
Rajan, S.S.S., Watkinson, J.H. and Sinclair, A.G. (1996) Phosphate rocks for
direct application to soils. Advances in Agronomy 57: 77-159.
Sanchez, P.A. (1994) Tropical Soil Fertility research: Toward the second paradigm.
In: Proceedings of the 15th Transactions of the World Congress of Soil Science.
10 - 16 July 1994, Acapulo, Mexico. pp 65 - 88.
260
Mkangwa, C.Z. et al
Sanchez, P.A. (1999) Improved fallows come of age in the tropics. Agroforestry
Systems 47: 3-12.
Soil Survey Staff (1998) Keys to Soil Taxonomy. United States Department of
Agriculture, Natural Resources Conservation Unit (8th Edition). pp 328.
Tropical Soil Biology and Fertility (1999) Organic Resource Database: A Guide
for Selecting Organic Inputs for Soil Fertility Management. Tropical Soil Biology
and Fertility, Nairobi. pp 5.
World Reference Base for Soil Resources (1998) Food and Agriculture
Organization of the United Nations. World Soil Resources Reports No. 84,
International Society of Soil Science, International Soil Reference and
Information Centre, Rome. pp 88.
Decomposition of Organic Matter in Soil as Influenced by Texture and Pore Size Distribution
Decomposition of Organic
Matter in Soil as Influenced
by Texture and Pore Size
Distribution
261
19
Mtambanengwe, F.1*, Mapfumo, P.1 and
Kirchmann, H.2
Department of Soil Science and Agricultural Engineering,
University of Zimbabwe, Box MP 167, Mt Pleasant, Harare,
Zimbabwe
2
Swedish University of Agricultural Sciences, Department of
Soil Science, Box 7014, Uppsala, 750 07, Sweden
1
*Corresponding author: fmtamba@agric.uz.ac.zw
Abstract
The carbon mineralization of fresh organic matter in soils
of different texture was determined using 14C-labelled
tobacco (Nicotiana tobacum) starch solution and 14C-labelled
barley (Hordeum vulgare) straw particles. A heavy clay soil
(56% illite clay) was mixed with acid washed sand to come
up with a range of different soil textures, with clay contents
varying between 5.6 and 56%. For each resultant texture,
three treatments were imposed, namely, barley straw
amendment, tobacco starch solution amendment and an
unamended control. Total porosity ranged from 60.3% in
soils with 56% clay content to 43.7% in soils with a clay
Mtambanengwe, F. et al
262
content of 5.6%. Total clay content, porosity, bulk density
and pore-volume (0.75 to 6 m in diameters) were all
positively correlated to 14CO2-C evolution (a measure of
decomposition) of the amended soils. In both amended and
unamended soils, the C mineralization decreased with
increasing clay content, with ranges of 42-121 mg C g-1 soil
with 5.6% clay and 34-107 mg C g-1 soil with 56% clay.
Regression analysis showed that pores <75 m in diameter
had the greatest influence on the amounts of CO2-C released
(R2 =0.91 for starch and 0.94 for straw). Our findings provide
empirical evidence to support the theory that decomposition
of fresh organic matter is governed by its physical
accessibility by microbes as determined by soil texture and
pore size distribution. We concluded that pores of <75 µm
are responsible for the protection of organic substrates
against microbial decomposition in soils. The study provides
insights on the role of clays in organic matter stabilization
and hence the vulnerability of the limited amounts of organic
inputs often available to smallholder farmers in tropical
environments where soils are predominantly low in clay.
Key words: organic substrate, carbon mineralization, soil texture, pore
size distribution, organic matter decomposition
Introduction
The influence of soil texture on organic matter decomposition has been
widely studied and results indicate that the rate of decomposition and
net mineralization depend on the accessibility of organic substrates to
soil organisms (Oades, 1984; Christensen, 1987; Amato and Ladd, 1992,
Hassink, 1996). Although organic matter decomposition studies are
numerous, few have addressed the relative importance of direct and
indirect mechanisms of soil texture control on organic matter
stabilization (Srensen, 1981; van Veen et al., 1985). In general, the
quantity and nature of the soil clay affect the amount of C stabilized in
soil. Fine textured soils (clays) often contain higher amounts of organic
matter than sandy soils. Two mechanisms have beet put forward to
explain the effect of soil texture on organic matter decomposition:
• the protective action by clays against organic matter degradation
through the formation of complexes between metal ions associated
with large clay surfaces and high CEC (Giller et al., 1997); and
• accessibility by soil microbes (van Veen and Kuikman, 1990).
Decomposition of Organic Matter in Soil as Influenced by Texture and Pore Size Distribution
263
Clay particles are believed to protect some of the more easily
decomposable organic compounds from rapid microbial breakdown
through encrustation and entrapment (Paul and van Veen, 1978;
Anderson, 1979; Tisdall and Oades, 1982).
The soil pore size distribution is one factor determining the
microbial habitat since it is assumed that microorganisms mainly live
in pores of a certain size. Considering the size of bacterial cells, pores
of a neck diameter of 2-6 µm are favourable microhabitats for soil
bacteria (Hattori and Hattori, 1976). Hassink (1992) also found a good
correlation between the habitable pore size fraction and N
mineralization. Killham et al. (1993) showed that substrate utilization
by microbes in soil was strongly affected by its location, both in terms
of pore size and the matric water potential under which turnover takes
place. It should, however, be noted that only a very small fraction of
organic material in soil is likely to be at close proximity to soil organisms
at any one time (Adu and Oades, 1978).
Organic matter may be physically protected in soil such that large
amounts of decomposable compounds can be found in the vicinity of
starving microbial populations (van Veen and Kuikman, 1990). Elliott
et al. (1980) used a combination of bacteria, amoebae and nematodes
to demonstrate the importance of microbial trophic structure in relation
to soil texture and habitable pore space. Ladd et al. (1985) found a
significant linear relationship between residual labelled C in topsoil and
clay contents ranging from 5 to 42%. The higher the clay content in the
different soils, the higher the residual C content after 8 years of
experimentation. On the assumption that bacterial cells were found in
pores with a diameter of between 0.8 and 6 µm (Kilbertus, 1980; Hattori
and Hattori, 1976), the aim of the experiment was to test how soil texture
and habitable pore space affect decomposition of fresh organic matter
applied to soil.
Materials and Methods
Soil-sand mixtures
The topsoil of a heavy clay from a site in Uppsala, Kungsängen soil,
classified as a fine, illitic, frigid Gleyic Cambisol in the FAO-system
(Kirchmann, 1991) was used. The same amount of air-dried soil
comprising of 56% clay (<0.002 mm), 39.9% silt (0.002-0.06 mm) and
4.1 % sand (0.06-2 mm) and with a pH of 6.9, 2.14% C and 0.26% N,
was mixed with different quantities of acid-washed quartz sand (0.30.5 mm) to create a range of textures resulting in different soil pore-size
distribution. As the same amount of soil was mixed with increasing
proportions of sand, it was assumed that the soil mixtures under
Mtambanengwe, F. et al
264
investigation were uniform with respect to:
a) the number of exchange sites which was the same in all the mixtures,
b) the initial amount of organic matter, and
c) the initial number of microbes present.
The only physical aspect that was changed was the pore size in the
soil-sand mixtures. Bulk densities and volumetric water contents of
the mixtures were determined at -500 KPa, -1 500 KPa wilting point)
and -4 000 KPa water pressures (Table 19.1). The pore size distribution
of each soil mixture was obtained from the moisture characteristic curves
relating volumetric water content to soil metric potential. The porosity
of the soil-sand mixtures was determined assuming a particle density
of the sand of 2.65 g cm-3 and for the Kungsängen soil, of 2.62 g cm-3
(Kirchmann, 1991).
Table 19.1: Particle size distribution, bulk density and soil pore space of the different soil
treatments
10g Kungssangen
soil+ xg sand
Soil texture(%)
Bulk density
Mgm-3
mixtures
Sand
Silt
Clay Control soil+
soil only starch
0.0
2.5
6.7
10.0
15.0
23.3
40.0
90.0
4.1
23.3
42.5
52.0
61.6
71.2
80.8
90.4
39.9
31.9
23.9
20.0
16.0
12.0
8.0
4.0
56.0
44.8
33.6
28.0
22.4
16.8
11.2
5.6
1.1
1.2
1.3
1.3
1.4
1.4
1.4
1.5
1.1
1.2
1.3
1.3
1.4
1.4
1.4
1.5
Total Porosity
soil+ Control soil+ soil+
straw soil only starch straw
1.0
1.1
1.2
1.2
1.3
1.4
1.4
1.5
60.4
56.0
51.3
50.2
48.8
46.9
45.1
43.3
60.4
56.0
51.3
50.2
48.8
46.9
45.1
43.3
62.6
58.5
54.7
53.6
49.8
47.7
45.5
43.3
Soil amendments
Triplicate samples of each clay content (10 g of Kungsängen soil plus
sand as described in Table 19.1), were moistened to 45% of their
respective water-holding capacities. The samples were then amended
with either 2 ml of a tobacco (Nicotiana tobacum) starch solution
containing 44.4 mg starch C, 45.5 µCig-1 starch 14C or 2 ml distilled
water plus 100 mg ground (<2 mm) barley straw (Hordeum vulgare)
containing 100 µCig-1C plant 14C and 2.1 mg plant N. The moistened
soil treatments were incubated at 25°C for 45 days. Incubation vials of
corresponding moistened soil-sand mixtures without organic
Decomposition of Organic Matter in Soil as Influenced by Texture and Pore Size Distribution
265
amendments were used as controls. Determination of CO2 evolution
was done using 10 ml traps with 0.1 M NaOH (Stotzky, 1965), that were
titrated with 0.1 M HCl after 1, 3, 7, 11, 17, 24, 31, 38 and 45 days.
Radioactivity of absorbed 14CO2 was determined by scintillation counting
(Beckman liquid scintillation systems LS 1801) and residual organic
14
C was determined by wet combustion of the oven dried soil subsamples.
Statistical Analyses
Data were subjected to a two-factor (time and treatment) analysis
of variance (ANOVA) to determine if the materials mineralized
differently with time. Possible mean differences of the cumulative
mineralization data were tested using independent Student t-tests
and residual 14C was correlated to the different soil properties using
linear regression analysis with the statistical package of MINITAB
(Ryan et al., 1985).
Results
Changes in soil pore size distribution upon amendments
Changing the texture through sand amendments increased the bulk
densities of the soil but lowered the total soil porosity (Table 19.1). The
addition of tobacco starch solution to the soils did not alter the physical
structure of the soils resulting in these soils having the properties similar
to the control soils. Bulk densities ranged from 1.05 Mg m-3 (soil 56%
clay) to 1.50 Mg m-3 (soil 5.6% clay) in control and starch-amended
soils. However, adding straw particles (Hordeum vulgare) to the same
soils lowered the bulk densities in high clay soils but not in more sandier
soils which had almost equal bulk densities to control soils. Total porosity
increased with increasing clay content from 43.3 to 60.4% in the soilsand mixtures and from 43.3 to 62.6% in the straw-amended soil-sand
mixtures.
The absolute pore volume in the soil-sand mixtures was affected
when sand was added with increasing quantities to the Kungsängen
soil (Table 19.2). The volume of pores with diameters of <6 m was higher
in straw amended soils than in starch amendments and the control
soil-sand mixtures. Differences were statistically significant only in soilsand mixtures of between 5.6 and 28% clay content (P<0.05) while in
all the high clay soils (>28% clay), the volume of pores <6 µm occupied
between 44 to 50% of total porosity (Table 19.2). Soil pore volumes of
diameters between 0.75 and 6 µm in both straw and starch amended
soils decreased exponentially with increasing clay content. In the same
Mtambanengwe, F. et al
266
soils, the pore volume of the smaller, supposedly inaccessible pores of
<0.75 m in diameter increased linearly and showed significant
relationships (R2 = 0.91 for starch amended and 0.94 for straw amended
soils.
Table 19.2: Soil pore space, pore volume and CO2-C evolution in amended and
unamended soils
Soil treatment and
pore size
56.0% 44.8% 33.6% 28.0% 22.4% 16.8%11.2%5.6%
clay
clay
clay
clay
clay
clay
clay clay
Pores >6 µm (%)*
50
52
55
60
66
71
78
85
Pores >6 µm (mL)
3
3
4
5
6
8
12
24
Control soil
Pores <0.75 µm (%)*
44
41
36
30
25
19
12
6
Pores <0.75 µm (mL)
3
2
2
2
2
2
2
2
Pores (0.75-6) µm (%)*
6
6
9
10
9
10
9
9
Pores (0.75-6) µm (mL)
0.4
0.4
0.6
0.8
0.8
1
1
3
CO2-C evolution (mg C g-1 soil)
29
31
35
35
36
37
38
40
50
52
55
60
66
71
78
85
Tobacco starch- amended soil
Pores >6 µm (%)*
Pores >6 µm (mL)
3
3
4
5
6
8
12
24
Pores <0.75 µm (%)*
44
41
36
30
25
19
12
6
Pores <0.75 µm (mL)
3
2
2
2
2
2
2
2
Pores (0.75-6) µm (%)*
6
6
9
10
9
10
9
9
Pores (0.75-6) µm (mL)
0.4
0.4
0.6
0.8
0.8
1
1
3
10.3
114
120
121
118
120
Pores >6 µm (%)*
51
52
56
61
62
67
74
82
Pores >6 µm (mL)
3
3
4
5
6
8
12
24
Pores <0.75 µm (%)*
41
39
32
27
24
19
14
7
Pores <0.75 µm (mL)
3
3
2
3
2
2
2
2
Pores (0.75-6) µm (%)*
8
10
12
12
14
14
13
11
0.6
103
0.9
108
1
100
1
101
2
103
2
3
110 118
-1
CO2-C evolution (mg C g soil)
127 141
Barley straw-amended soil
Pores (0.75-6) µm (mL)
0.5
14CO2-C evolution (mg C g-1 soil) 97
*expressed as a percentage of total porosity
Decomposition of Organic Matter in Soil as Influenced by Texture and Pore Size Distribution
267
Figure 19.1: The relationship between soil-pore volume and clay content of (a)
unamended control soil and (b) the same soils following addition of barley straw
4
a) Control soil pore space
3
r <0.75µm
h diamete
Pores wit
2
Soil pore volume (%)
1
Pores w
ith diam
eter betw
een <0.7
5 and 6µ
m
0
4
b) Straw-amended soil pore space
3
Pores with
0.75µm
diameter <
2
Pores w
ith diam
1
eter betw
een <0.7
5 and 6µ
m
0
5.6
11.2
16.5
22.4
28.0
Soil clay content (%)
33.6
44.8
56.0
268
Mtambanengwe, F. et al
Carbon Dioxide Evolution
In the control soil-sand mixtures, there was very little difference in the
amount of native soil organic matter evolved as CO2 after 45 days of
incubation. Significant differences were apparent at high clay contents
(P<0.05). Total cumulative values ranged from 29.3 mg C g-1 soil (56%
clay) to 40.1 mg C g-1 soil (5.6% clay). The same trend of C mineralization
was observed when soils were amended with fresh organic matter (Table
19.2). Both barley straw and starch were decomposed to a higher extent
in soils with low clay contents of 5.6 to 16.8%. The CO2 evolution from
starch amended treatments was greater than from straw amended ones.
At the end of the incubation period, more than quarter of the added C
had been mineralized in all the treatments. Cumulative CO2-C evolution
data from starch-amended soils showed differences between high and
low clay content within one week and these differences became
statistically significant from the third week onwards (P<0.05) (Figure
19.2a).
In the straw-amended soils, there was very little difference in the
evolution of 14CO2-C in six of the eight mixtures (P>0.05) (Figure 19.2b).
The pattern of C mineralization observed was very similar to that of the
control soil-sand mixtures. Cumulative 14CO2-C evolution data showed
that C mineralization was initially rapid and after about 3 weeks,
decreased rates of 14CO2-C production following a first-order function
were observed. At the end of the incubation period, the highest percentage
of labelled C evolved (39.1% of added straw 14 C) was noted in soil with
5.6% clay, the other 7 soil mixtures released between 29.2 and 36.5%
of the C added.
Relationships between C mineralization and soil physical
properties
There were good correlations between labelled substrate C mineralization
and clay content, bulk density and soil pore spaces. Bulk density was
positively related to C mineralization whereas total porosity was
negatively related (Table 19.3). Differentiating soil pore diameters into
three possible groups (<0.75 m, 0.75-6 m and >6 m), the highest
correlation with C evolution was obtained with the volume of pores of
diameters <0.75 m (R2 = 0.914 for starch and 0.935 for straw; p < 0.001)
(Figure 19.3). The results showed that the higher the concentration of
these small pores, the less the 14CO2-C evolved. Linear regression
analysis showed that translating percentage porosity into absolute pore
volume per given soil did not improve correlations between the different
pore groups and 14CO2-C evolution.
Decomposition of Organic Matter in Soil as Influenced by Texture and Pore Size Distribution
269
Figure 19.2: Net cumulative 14CO2-C evolution of soil-sand mixtures amended with a)
tobacco starch solution and b) <2mm granulated barley straw. Means are based on n = 3
60
a) Starch-amended soils
50
5.6% clay
11.2% clay
16.5% clay
22.5% clay
33.6% clay
28.0% clay
56.0% clay
44.8% clay
40
Cumulative 14CO2-C evolution (% of added 14C)
30
20
10
0
60
b) Straw-amended soil
50
40
5.6% clay
11.2% clay
16.5% clay
22.5% clay
33.6% clay
28.0% clay
56.0% clay
44.8% clay
30
20
10
0
0
5
10
15
20
25
30
Incubation period (days)
35
40
45
Mtambanengwe, F. et al
270
Table 19.3: Linear regressions for 14CO2-C mineralization between (a) tobacco (Nicotiana
tobacum) starch and (b) barley (Hordeum vulgare) straw and soil properties after 45
days of soil incubation
Quality variable (x)
Dependent
variable (y)
Clay (%)
Bulk density
Total porosity (%)
Total pore volume (mL)
Porosity (<0.75 m) (%)
Pore volume (<0.75 m) (mL)
Porosity (>6 m) (%)
Pore volume (>6 m) (mL)
Intercept
Slope
R2
46.3
18.1
65.7
38.2
47.6
50.5
27.1
39.1
-0.2
18.1
-0.5
0.3
-0.2
-0.1
0.2
0.4
0.81
0.76**
0.78**
0.75**
0.91***
0.91***
0.91***
0.75**
38.5
13.4
56.2
29.8
40.2
45.2
17.1
30.8
0.2
16.2
-0.4
0.4
-0.2
-1.1
0.6
0.4
0.91***
0.90***
0.90***
0.79**
0.92***
0.94***
0.93***
0.79***
a
Clay (%)
Bulk density
Total porosity (%)
Total pore volume (mL)
Porosity (<0.75 m) (%)
Pore volume (<0.75 m) (mL)
Porosity (>6 m) (%)
Pore volume (>6 m) (mL)
b
*** - significant at p < 0.001, ** - significant at p < 0.01, * - significant at p < 0.05
Figure 19.3: Relationship between cumulative 14CO2-C evolution and soil pore volume
of diameters of <0.75m for starch-amended soils (a) and straw-amended soils (b)
48
a) Straw-amended soil
43
Y=50.5-11.0x R2=0.914
Y=45.2-11.0x R2=0.935
38
33
14
CO2-C evolution (% of added 14C)
a) Starch-amended soil
28
1.7 1.8
1.9
2
2.1
2.2
2.3
2.4
2.5
2.6
2.7 1.7
1.8
1.9
2
Pore volume < 0.75 µm diameter (mL)
2.1
2.2
2.3
2.4
2.5
2.6
2.7
Decomposition of Organic Matter in Soil as Influenced by Texture and Pore Size Distribution
271
Discussion
The addition of a readily decomposable organic C sources stimulated
the mineralization of native soil organic C in the soil-sand mixtures.
The extent to which the two different substrates 14C was decomposed
was consistent in both amendments being higher in low clay than in
high clay soils. Given that the Kungsangen soils used in treatments
already had high initial C and N, sampling possibly opened up some
labile C that would have been otherwise protected in soil aggregates
resulting in increased C evolution. However, the nature of the
decomposer population and available pore space have been found to
influence the rate of mineralization of organic substrates added to
soil (Srensen, 1975; Elliott et al., 1980; Hassink et al., 1993). Pores
with diameters of less than 0.75 µm were presumed to be responsible
for the protection of microbial decomposition of the fresh organic
substrates added to soil-sand mixtures. These pores were found in
higher proportions in high clay soils and thus may explain why high
clay soils are better able to protect organic matter from decomposition.
The concept of microbial accessibility in soil seem to be most
meaningful if it is used in relation to the size of the microbial
inaccessible pores. The structure of the decomposer community
available in the Kungsängen soil was not investigated in this study.
In our study, we sampled soils that had been under intensive cereal
production for over 25 years (Kirchmann, 1991) thus the amount
and quality of native organic matter in the soil was assumed to be
similar and was not disturbed during sample preparation. The number
of exchange sites of the clay were kept constant and the effect of pore
size could be investigated without interaction of other media. Although
soil texture has been shown to correlate with long-term C dynamics
(Jenkinson, 1971; Ladd et al., 1985), our results show that differences
in soil texture in the control soils had no significant effect on CO2-C
mineralization of native soil organic matter. We therefore suggest the
key factor controlling organic matter decomposition in different
textured soils is the soil pore size and distribution. Since soil texture
controls the pore size distribution and it is in this way that conclusions
on texture affecting organic matter decomposition have been made
(Hassink, 1992; Hassink et al, 1993). Our results have shown that
as the soils became more sandy, the degree of soil porosity decreased.
Carbon dioxide evolution from the two added organic substrates
appeared to be strongly affected by the soil pore-size distribution
and pore continuity.
There is no general agreement on the critical factors influencing
movement of organisms in soil, but larger organisms are probably more
restricted than smaller ones. Pore-size distribution and continuity are
272
Mtambanengwe, F. et al
known to influence soil water availability, gas diffusion and the
movement of soil organisms (Coleman et al., 1984; Scott et al., 1996).
Pore-neck size determines whether an organism can enter a given pore,
thus whether a substrate located within the soil pore is available to
microorganisms. We found that the best correlation was the negative
relationship between soil pores of less than 0.75 µm and the amount
of 14CO2-C evolved for both starch- and straw-amended treatments.
These results indicate the importance of those pores that are
inaccessible for microbes in the stabilization of organic matter. The
theory of physical accessibility and its linkage between soil texture as
described by soil pore size distribution to decomposition of fresh organic
matter was affirmed. Hassink et al. (1993) found good relationships
between pore volumes of between 0.2 and 1.2 µm diameter and bacterial
biomass. The difference between 14CO2 released from the high clay soils
and the low clay soils in both starch and straw amendments may be
considered to be a measure of the proportion of starch or straw in
micropores potentially protected from microbial degradation. Adu and
Oades (1978) attributed physical protection from microbial
decomposition when a portion of soluble starch in micropores was left
unmineralized.
The concept of substrate quality has also been identified as an
important factor in soil organic matter stabilization (McClaugherty et
al., 1985; Duxbury et al., 1989; Melillo et al., 1989). Although the
contribution of native soil organic matter to total carbon mineralization
was significant as was shown by the control soils, several studies have
shown what Jenkinson (1971) described as the 'priming action'
following the addition of fresh organic matter to soil. Carbon dioxide
evolution from the control (unamended) soils was significantly lower
than that of amended soil signifying the stimulation of ‘microbial
priming action’ following soil amendments. Higher mineralization and
decomposition rates are known to be stimulated by increased N
availability (Palm and Sanchez, 1991; Tian et al., 1995). Our study
showed that incubation of starch-amended soils showed greater
utilization than high N-containing, thus higher quality barley straw
(Cadisch and Giller, 1997). This was probably due to an abundance of
readily available C for maximum utilization for microbial growth relative
to N. The delayed differences in C mineralization of straw amended
soils observed towards the end of the study imply a gradual narrowing
of the C:N ratio when at some point, N becomes no longer limiting to
microbial growth. In addition, accessible pore space also played a major
role during the mineralization of a presumably distributed (soluble
starch) organic substrate and more strongly of unevenly distributed
one (granulated straw).
Decomposition of Organic Matter in Soil as Influenced by Texture and Pore Size Distribution
273
Conclusion
Our findings provide a challenge to soil fertility research focussing on
soil organic matter build-up and maintenance using organic
amendments. Given that most of the smallholder farming areas in
tropical environments are dominated by sandy soils, the vulnerability
of the limited amounts of organic inputs available in most of these
farming systems is implied. The study provides empirical evidence to
support the theory that decomposition of fresh organic matter is governed
by its physical accessibility by microbes as determined by soil texture
and pore size distribution. We therefore concluded that pores of diameters
of <75 µm were responsible for the protection of organic substrates
against microbial decomposition in soils. To be able to understand fully
the importance of clay in organic matter stabilization, there is a need
for more research in the soil pore system and the mechanisms that
take place within and develop organic matter management options for
soils of different textures.
Acknowledgements
We are grateful to SAREC-SIDA for financial support. Thanks are also
due to the technical staff at the Department of Soil Science, Swedish
University of Agricultural Sciences, for assistance with the Scintillation
counter and Professors K. Giller and S. Feresu for critically reviewing
the manuscript.
References
Adu, J.K., and Oades, J.M. (1978) Physical factors influencing decomposition
of organic materials in soil aggregates. Soil Biol. Biochem. 10: 109-115.
Amato, M., and Ladd, J.N. (1992) Decomposition of 14C-labelled glucose and
legume material in soils. Properties influencing the accumulation of organic
residue C and microbial biomass. Soil Biol. Biochem. 24: 455-464.
Anderson, D.W., (1979) Processes of humus formation and transformation in
the soils of the Canadian Great Plains. J. Soil Sci. 30, 77-84.
Cadisch, G., Giller, K. (1997) Driven by Nature: Plant Litter Quality and
Decomposition. CAB International, Wallingford.
Christensen, B.T. (1987) Decomposability of organic matter in particle size
fractions from field soils with straw incorporation. Soil Biol. Biochem. 19,
429-435.
Coleman, D.C., Anderson, R.V., Cole, C.V., McClellan, J.F., Woods, L.E.,
Trofymow, J.A., and Elliott, E.T. (1984) Roles of protozoa and nematodes in
274
Mtambanengwe, F. et al
nutrient cycling. In: Giddens, J.E. and Todd, R.L. (Eds.), Microbial-Plant
Interactions. ASA Special Publication 47. SSSA, Madison, pp. 17-28.
Duxbury, J.M., Smith, M.S. and Doran, J.W. (1989) Soil organic matter as a
source and sink of plant nutrients. In: Coleman, D.C., Oades, J.M. and
Uehara, G. (Eds.), Dynamics of Soil Organic Matter in Tropical Ecosystems.
NifTAL Project, University of Hawaii, Honolulu, pp. 33-67.
Elliott, T.E., Anderson, R.V., Coleman, D.C., and Cole, C.V. (1980) Habitable
pore space and microbial trophic interactions. Oikos 35: 327-335.
Giller, K.E., Cadish, G., Ehaliotis, C., Adams, E., Sakala, W. and Mafongoya, P.
(1997) Building soil nitrogen capital in Africa. In: Buresh, R.J., Sanchez,
P.A., Calhoun, F. (Eds.), Replenishing Soil Fertility in Africa. SSSA Special
Publication No. 51, Madison, pp. 151-192.
Hassink, J. (1992) Effects of soil texture and structure on carbon and nitrogen
mineralization in grassland soil. Biol. Fert. Soils 14: 126-134.
Hassink, J. (1996) Presevation of plant residues in soils differing in unsaturated
protective capacity. Soil Sci. Soc. Am. J. 60: 487-91.
Hassink, J., Bouman L.A., K.B. Zwart, K.B. and Brussard, L., (1993).
Relationship between habitable pore space, soil biota and mineralization
rates in grassland soils. Soil Biol. Biochem. 25: 47-55.
Hattori, T. and Hattori, R. (1976) The physical environment in soil microbiology:
An attempt to extend principles of microbiology to soil microorganisms.
CRC Crit. Rev. Microbiol. pp. 423-461.
Jenkinson, D.S. (1971) Studies on the decomposition of 14C labelled organic
matter. Soil Sci. 111: 64-72.
Kilbertus, G., (1980) Etude des microhabitats contenus dans les agregats du
sol. Leur relation avec la biomasse becterienne et la taille des procaryotes
presents. Rev. d'Ecol. Biol. Sol 17: 543-557.
Killham, K., Amato, M. and Ladd, J.N. (1993) Effect of substrate location in soil
and soil pore-water regime on carbon turnover. Soil Biol. Biochem. 25: 5762.
Kirchmann, H. (1991) Properties and classification of soils of the Swedish longterm fertility experiments. 1. Sites at Fors and Kungsängen. Acta Agric.
Scand. 41: 227-242.
Ladd, J.N., Amato, M. and Oades, J.M. (1985) Decomposition of plant material
in Australian soils. III. Residual organic and microbial biomass C and N
from isotope labelled plant material and organic matter decomposition under
field conditions. Aust. J. Soil Res. 23: 603-611.
McClaugherty, C., Pastor, J. and. Muratore, J.F. (1985) Forest litter
decomposition in relation to soil nitrogen dynamics and quality. Ecol.
66: 266-275.
Melillo, J.M., Aber, J.D., Linkins, A.E., Ricca, A., Fry, B. and K.J. Nadelhoffer.,
K.J. (1989) Carbon and nitrogen dynamics along a decay continuum: Plant
litter to soil organic matter. In: Clarholm, M., Bergstrm, L. (eds.), Ecology of
Arable Land. Kluwer, Dordrecht, pp. 53-62.
Oades, J.M. (1984) Soil organic matter and structural stability: mechanisms
and implications for management. Plant and Soil 76: 319-337.
Decomposition of Organic Matter in Soil as Influenced by Texture and Pore Size Distribution
275
Oades, J.M. (1988). The retention of organic matter in soils. Biogeochem.
5: 35-70.
Palm, C. and Sanchez, P. (1991) Nitrogen release from the leaves of some tropical
legumes as affected by their lignin and polyphenolic contents. Soil Biol.
Biochem. 23: 83-88.
Paul, E.A., van Veen, J.A. (1978) The use of tracers to determine the dynamic
nature of organic matter. International Congress of Soil Science, Transactions
of the 11th Symposia Papers. 3: 61-102, Edmonton.
Ryan, F.R., Joiner, B.L. and Ryan, T.A. (1985) MINITAB handbook. Prindle, Weber
and Schimdt Publishers, Boston.
Scott, N.A., Cole,V.C., Elliott, E.T. and Huffman, S.A. (1996) Soil textural control
on decompostion and soil organic matter dynamics. Soil Sci. Soc. Am. J.
60: 1102-1109.
Srensen, L.H. (1975) The influence of clay on the rate of decay of amino acid
metabolites synthesized in soils during decomposition of cellulose. Soil Biol.
Biochem. 7: 171-177.
Srensen , L.H. (1981) Carbon-nitrogen relationships during the humification of
cellulose in soils containing different amounts of clay. Soil Biol. Biochem.
13: 313-321.
Stotzky, G. (1965) Microbial respiration. In: Black, C.A. (ed.), Methods of Soil
Analyses Part 2, 2nd Edition, Agronomy Monograph 9, ASA, SSSA, Madison,
pp. 1550-1572
Tian, G., Kang, B.T. and Brussard, L. (1995) Mulching effect of plant residues
with chemically contrasting composition on maize growth and nutrient
accumulation. Plant and Soil 153: 179-187.
Tisdall , J.M. and Oades, J.M. (1982) Organic matter and water-stable aggregates
in soils. J. Soil Sci. 33: 141-163.
van Veen, J.A. and Kuikman, P.J. (1990) Soil structural aspects of decomposition
of organic matter by micro-organisms. Biogeochem. 11: 213-233.
van Veen, J.A. and J.N. Ladd, J.N. and Amato, M. (1985) Turnover of carbon
and nitrogen through the microbial biomass in sandy loam and a clay soil
incubated with [14C(U)] glucose ([15N](NH4)2)SO4 under different moisture
regimes. Soil Biol. Biochem. 17: 747-756.
276
Mtambanengwe, F. et al
Soil Conservation and Fertility Improvement Using Leguminous Shrubs in Central
Highlands of Kenya: NARFP Case Study
Soil Conservation and
Fertility Improvement Using
Leguminous Shrubs in
Central Highlands of Kenya:
NARFP Case Study*
277
20
Mugwe, J.1, Mugendi, D.2, Okoba, B.3,
Tuwei, P.1 and O’Neill, M.4
1
Kenya Forestry Research Institute (KEFRI). Scientist attached
to National Agroforestry Research Project; KARI-Regional
Research Centre, Embu. P.O. Box 27, Embu, Kenya. Tel: +254161-20116/20873 Fax: +254-161-30064 E-mail:
kariembu@salpha.co.ke
2
Faculty of Environmental Studies Kenyatta University. P.O.
Box Box 43844 Ext 216/214 Nairobi, Kenya Tel: 2-811622/
812722. E-Mail: dmugendi@yahoo.com
3
Kenya Agricultural Research Institute (KARI). National
Agroforestry Research Project; KARI-Regional Research
Centre, Embu, Kenya. P.O. Box 27, Embu. Tel: +254-16120116/20873 Fax: +254-161-30064 E-mail:
kariembu@salpha.co.ke
4
International Centre for Research in Agroforestry (ICRAF)
attached to National Agroforestry Research Project; KARIRegional Research Centre, Embu, Kenya. P.O. Box 27, Embu.
Tel: +254-161-20116/20873 Fax: +254-161-30064 E-mail: icrafembu@cgiar.orgpresently: Asst. Prof. and Superintendent, New
Mexico State University, Agricultural Science Center, P.O. Box
1018, Farmington, NM, USA, 87499; Tel: +1-505-327-7757;
Email: moneill@nmsu.edu
*NAFRP = National Agroforestry Research Project, Embu, Kenya
Mugwe, J. et al
278
Abstract
Declining land productivity with reduced crop yields has
been a major problem facing smallholder farmers in the
central highlands of Kenya. The major factors contributing
to the reduced land productivity is soil impoverishment
caused by continuos cropping without addition of adequate
fertilizer and manure, and soil erosion on steep slopes. The
National Agroforestry Research Project (NAFRP) initiated
research work in 1992 to try and address these problems.
The research work investigated the potential of using two
leguminous shrubs (Leucaena leucocephala and Calliandra
calothyrsus) for improving soil fertility and soil conservation
on steep slopes. The studies were carried out both at onstation and on-farm.
Treatments where leafy prunings of calliandra and
leucaena were incorporated yielded higher than the control
treatments without prunings incorporation. Leucaena alley
cropping system was beneficial and maintained crop yields
at 4 t ha -1 in most seasons. Calliandra hedgerow
intercropping system on the other hand depressed crop
yields. However calliandra was effective in controlling soil
erosion when planted as a contour-hedgerow system. The
contour hedgerows in addition to conserving soil produced
additional benefits in terms of high quality animal fodder.
␣ This study concluded that in the central highlands of Kenya
where land is slopy, and similar areas, it is advisable for the
smallholder farmers to plant leguminous fodder trees on
terraces as contour hedgerows for both soil conservation and
biomass production. The resulting biomass could be
incorporated into the soil to improve soil fertility for farmers
without livestock, or fed to livestock for farmers who own
livestock. If the biomass is fed to livestock, possibilities of
recycling nutrient through animal manure should be explored
to ensure soil nutrient replenishment.
Key words: Alley-cropping, Calliandra calothyrsus, Leucaena
leucocephala, Contour hedgerows, Leafy prunings (prunings), Fodder
Introduction
The central highlands of Kenya are densely populated with more than
500 persons Km-2 (Government of Kenya, 1994) and declining land
productivity with reduced crop yields has been a major problem facing
smallholder farmers in the region (Kipkiyai et al., 1998). The major factors
Soil Conservation and Fertility Improvement Using Leguminous Shrubs in Central
Highlands of Kenya: NARFP Case Study
279
contributing to the reduced land productivity is soil impoverishment
caused by continuous cropping with inadequate addition of fertilizer
and/or manure, and soil erosion on steep slopes (Minae and Nyamai,
1988).
Land sizes are small, averaging 1.2 ha per household, and this promotes
continuous cropping with limited scope for crop rotation and inadequate
soil fertility replenishment. Nitrogen and phosphorus are the most limiting
nutrients to crop production and high costs of inorganic fertilizers limit
their sufficient use by majority of the smallholder farmers. Indeed a
substantial number of farmers do not use fertilizers and the ones who use
fertilizers apply below the recommended rates (Kihanda, 1996). Muriithi et
al (1994) reported use of fertilizers by farmer to be less than 20 kg N and
10 kg P ha-1 against recommended rate of 60 kg N and Pha-1.
The topography of the central highlands of Kenya is gently to steeply
rolling with a medium to high erosion hazard as determined by FAO
(Kassam et al., 1992). The sloping topography and high rainfall has
resulted to soils in the region being highly prone to water erosion (O’Neill,
1997; Ongwenyi et al 1993). Water erosion carries away soil nutrients
and, soil nutrient depletion has been reported to be taking place at an
alarming rate in all the agroecological zones (Pieri et al., 1995).
Leguminous trees species have shown some potential for soil fertility
improvement and soil conservation. Soil fertility improvement can be
achieved through biomass transfer, short term fallows, nitrogen fixation,
re-activation of the ‘N bulge’ and phosphorus scavenging ( Rosecrance
et al., 1992; Amadalo et al., 1995; Jama et al., 1998; Hartemink et al.,
2000). The leguminous trees have similarly shown potential of reducing
soil erosion through five processes; interception of rainfall impact by
tree canopy, surface runoff impediment by tree stems, soil surface cover
by litter mulch, promotion of water infiltration and formation of erosionresistant blocky soil structure (Nair, 1987; Young, 1989; Young and
Sinclair, 1997). Researchers at other sites within East Africa AFRENA
(ICRAF, 1991) have had success of varying degrees by incorporating
fodder grasses and trees along the contours to both reduce soil erosion
and provide products which would help convince the farmers that soil
conservation can be profitable.
Despite the potentials for the use of tree shrubs for soil conservation
and soil fertility improvement, their use in the region is limited (ICRAF,
1992). A survey carried out in Meru District (Murithi et al., 1994) found
that 83 percent of the farmers surveyed had soil fertility problems while
91 percent had soil erosion problems. In the survey majority of farmers
(93 percent) used manure for improving soil fertility with limited use of
tree prunings while none of the farmers used trees for soil erosion control.
To address the aforementioned problems of soil fertility decline and
soil erosion on steep slopes, the National Agroforestry Research Project
(NAFRP), based at the Kenya Agricultural Research Institute (KARI),
280
Mugwe, J. et al
Regional Research Centre-Embu, Kenya, conducted some research
activities to investigate feasibility of using leguminous trees for soil
fertility improvement and soil conservation. The research work was
initiated in 1992 and this manuscript review consolidates the research
work, highlighting the major findings and also discusses the
dissemination potential. The leguminous shrubs used are Leucaena
leucocephala (Lam.) de Wit and Calliandra calothyrsus Meissn. The two
species have been shown to be the most appropriate species for soil
improvement and crop sustainability through agroforestry research at
Maseno, Kenya (Heinemann et al., 1990).
Methodology
The study area
The highlands of Central Kenya are characterized by a bimodal rainfall
distribution, which ranges from 600 mm to 2000 mm annually in the
mandate region of the National Agroforestry Research Project based in
Embu. Agriculture is characterized by smallholder mixed farming
activities, which include cash crops (coffee, tea and horticultural crops),
food crops (Maize, beans, bananas and Irish potatoes) trees and livestock
(dairy and beef cattle, goats, sheep, poultry and pigs). Nearly all farmers
in the region practice dairy farming under zero and/or semi-zero grazing
and the need for fodder is a main constraint (Minae and Nyamai, 1988).
Napier grass (Pennisetum purpureum) is the major fodder and is generally
produced in fodder plots or along the bunds of terraces where it is also
used as stabilizer for the erosion control structures.
High population pressure (> 500 persons Km-2) has led to the subdivision of family farms into small units (app. 1.2 ha) which require
intensive agricultural practices to produce enough food for home
consumption and outside sales. This has led to the exploitation of lands
of decreasing productivity and increasing soil erosion potential.
The soils are locally known as “Kikuyu Red Clay Loams”. They are
deep (>2m), well drained, dusky red to dark reddish brown in colour
with moderate structure (Mwangi, 1997). They are derived from rich,
basic volcanic rocks and have been classified as Typic Palehumult (Humic
Nitisols according to FAO-UNESCO, 1975).
Research Activities
Within the period 1992 and 1998 two on-station experiments and one
on-farm experiment were carried out to address the problems of declining
Soil Conservation and Fertility Improvement Using Leguminous Shrubs in Central
Highlands of Kenya: NARFP Case Study
281
soil fertility and soil erosion on steep slopes. These experiments are
described in the following sections.
Alley-cropping experiment
An alley-cropping was installed during 1992 at the Embu Regional
Research Centre. The aim was to evaluate feasibility of using leafy
prunings of calliandra and leucaena in a maize production system for
soil productivity enhancement in both alley-cropping and monocropping
systems. The experimental design was a Randomized Complete Block
with four replicates. The plot dimensions were 9 x 10 m while the sample
plot was 6 x 4.5 m. Maize (Zea mays L. ) variety Hybrid 511 was the test
crop. The experiment consisted of ten treatments. Six of these had fresh
leaf prunings of tree species (leucaena and calliandra) applied. The
prunings were obtained from hedgerows grown in situ (alley-cropped) or
ex-situ (cut and carry) from other sources. The treatments were as follows:
Alley-cropping; no fertilizer
1.
2.
3.
4.
Calliandra; prunings incorporated
Leucaena; prunings incorporated
Calliandra; prunings removed to treatment 5
Leucaena; prunings removed to treatment 6
Maize only; no alley cropping; prunings from outside incorporated
5. Calliandra prunings from 3; no fertilizer
6. Leucaena prunings from 4; no fertilizer
7. Calliandra prunings + fertilizer (25 kg N ha-1)
8. Leucaena prunings + fertilizer (25 kg N ha-1)
Maize only; no alley cropping; no prunings
9. With fertilizer (50 kg N ha-1)
10. Without fertilizer
The prunings were always lopped and soil-incorporated using hand
hoes immediately before maize was planted. The weight of prunings
applied to treatments 5 and 7, and 6 and 8 was equal to the weight of
prunings obtained from Treatments 3 and 4.
Harvesting of maize was done by cutting maize plants at soil level.
Maize cobs were manually separated from the stover, sun-dried and
packed in paper-bags before threshing. After threshing, moisture content
of the grains was determined using a moisture meter and grain weights
adjusted to 12% moisture content.
282
Mugwe, J. et al
Contour hedgerow experiments (on-station and on-farm)
An on-station contour hedgerow experiment was set up in 1993 with
two objectives. First to determine the degree to which contour hedges of
grasses and trees, in combination and alone, can reduce soil erosion
and; secondly to determine the amount and quality of fodder the various
combinations could provide over time. Due to the limited size of available
land, the maximum number of sufficiently large plots were eight. Hence
there were two replications of four treatments: the control (with no hedge),
grass hedge (consisting of two rows of Napier grass) (Pennisetum
purpureum)), tree hedge (consisting of two rows of calliandra), and
combination hedge (consisting of one row of calliandra and one row of
Napier grass). The runoff plots measured 5 x 30 m on an 18% slope and
runoff was measured by a tipping bucket system (Khan and Ong, 1997).
Maize was grown on all the plots at the recommended density and the
crop’s agronomic practices of the area followed as recommended.
Following promising results from the on-station experiment (O’Neill
et al., 1995), an on-farm experiment was initiated in 1996, with an
objective of assessing the impact of contour hedges on soil and water
conservation under direct farmer management practices. Treatments
consisted (1) tree hedge consisting of two rows of calliandra (2) A grass
hedge consisting of two rows of napier grass and (3) control with no
hedges. They were replicated thrice on two different slopes (20% slope
and 40% slope) within Kianjuki Catchment of Embu District, Kenya.
Runoff was collected in drums/barrels at the lower part of runoff plots.
Results and Discussions
Alley-cropping experiment
In the alley cropping experiment highest mean yields were obtained
from treatment 2 (leucaena alley crop + prunings) and treatments 5
and 6, (maize monocrop + prunings) and 7 and 8 (maize monocrop with
prunings + 25 Kg N ha-1) in most seasons over the 11 seasons under
study (1993 – 1998) (Table 20.1). These treatments had significantly
higher mean maize yields at 5 % probability level than all the other
treatments. This is an indication that the soil-incorporated leafy prunings
of calliandra and leucaena improved maize growth resulting to increased
maize grain yields. The results agree with findings of Guevara (1976),
Kang (1981) Evensen (1984) and Attah-Krah (1990) who reported
significant maize yield increases following application of green manure.
In the present study the leafy prunings incorporated into the soil (as
green manure) at the beginning of the season decomposed and released
nutrients especially nitrogen which enhanced crop performance
(Mugendi et al., 1999a).
TRT
LR 93
SR 93
LR 94
SR 94
LR 95
SR 95
LR 96
LR 97
SR 97
LR 98
SR 98
Mean
1
2.4 a
0.1 b
0.2 cd
3.2 ab
3.4 b
2.4 e
2.0 c
2.3 b
4.3 d
2.7 b
0.7 ab
2.6 bc
2
2.2
a
ab
d
a
a
bc
b
a
c
a
b
3.5 a
3
1.7 a
0.01 b
0.3 cd
1.6 c
1.1 e
1.0 g
1.3 d
1.9 b
3.6 e
1.3 d
0.5 b
1.6 d
4
1.5 a
0.2 ab
0.5 cd
2.7 b
1.8 de
1.7 f
1.5 cd
1.7 c
4.6 cd
2.2 c
0.7 ab
2.1 c
5
1.9 a
0.6 a
0.3 cd
3.6 ab
4.0 ab
4.9 a
3.8 b
3.0 a
4.0 d
2.8 b
1.0 a
3.4 a
6
1.6 a
0.3 ab
0.9 b
3.3 ab
4.2 a
4.7 ab
3.9 ab
2.5 b
7.1 b
2.7 b
1.2 a
3.7 a
7
2.1
a
a
b
ab
cd
a
a
b
a
a
a
3.8 a
8
1.8 a
0.3 ab
2.5 ab
3.2 ab
3.1 bc
5.0 a
4.0 ab
1.8 bc
7.5 ab
2.6 bc
1.1 a
3.5 a
9
1.6 a
0.3 ab
3.0 a
3.1 ab
3.1 bc
3.5 c
3.6 b
2.0 bc
4.1 de
2.9 b
0.7 ab
2.9 b
10
1.4 a
0.2 ab
1.1 c
3.0 ab
2.8 c
1.8 cd
1.8 c
1.9 bc
3.2 e
2.3 c
0.6 b
1.9 c
0.2
0.6
0.2
2.1
3.8
3.6
4.4
2.2
3.9
5.6
3.6
4.2
3.1
2.2
Means followed by the same letter within a column are not significantly different at P< 0.05
Abbreviations: SR = short rains; LR = long rains
5.0
7.9
3.8
3.4
0.6
0.9
Soil Conservation and Fertility Improvement Using Leguminous Shrubs in Central
Highlands of Kenya: NARFP Case Study
Table 20.1: Mean maize grain yield (t ha-1) for 1993-1998 seasons from various treatments in an on-station study at Embu, Kenya
Source: (Mugwe and Mugendi, 1999)
283
284
Mugwe, J. et al
During the first phase of this experiment, from 1993 long rains (LR)
to 1994 LR, nutrient deficiencies especially nitrogen and phosphorus
(P) occurred and resulted in low crop yields (Mugwe et al., 1997). During
1994 LR, Mwangi (1997) reported phosphorus deficiencies in all plots
except where fertilization was done. This was attributed to low native
phosphorus as a result of high P-fixing capacity of these soils and that
the biomass harvested from the hedgerows was low and contained little
amount of P (about 0.2 %) which could not supply adequate amounts of
P into the soil to meet the crop demand (Mugwe et al., 1999). This agrees
with findings of Palm (1995) and Salazar et al.(1993) who found
insufficient amount of P in prunings of most tree species. As such,
supplementation of P through the use of inorganic fertilizers was
recommended.
Biomass production over the study period was generally low in the
range of 1.1 to 3.5 Mg ha-1 season-1 (Table 20.2). Highest biomass
production was obtained during 1997 SR and 1998 LR which was
attributed to higher rainfall during the seasons due to El-nino
phenomenon (Mugwe et al., 2000). The low biomass incorporated
resulted in low nutrient contribution (containing approximately 60 kg
N ha-1 season-1 or 120 kg N ha-1 yr-1). This could not contribute sufficient
amounts of nutrients to compensate fully for those lost in crop harvests
(Mugendi et al., 1999a; Mugwe and Mugendi, 1999) as nitrogen removal
by crop harvests in the plots that received prunings ranged from 150
kg to 269 kg ha-1 year-1 (Mugendi et al., 1999b). The results agree with
those reported by Kang (1993), Nair (1993) and Scroth et al. (1995)
where a small decline in soil fertility was observed in plots that had
prunings applied. The findings, however did not agree with reports
from the humid tropics where application of prunings to the soil
resulted in increased soil organic matter and higher N, P, K, Ca, and
Mg (Kang et al., 1985; Kang et al., 1990; Tian et al., 1993). The
difference between the current study and studies in the humid tropics
is that, whereas hedgerow species in the humid tropics produced 8 to
10 Mg ha-1 yr-1 of biomass (Young 1989) the trial at Embu produced
less than half this amount (Table 20.3). The low biomass production
of hedgerow species in alley cropping systems in most areas is one
major drawback that limits the potential of prunings to improve fertility
and productivity of soils (Mathews et al., 1992; Yadvinder et al., 1992;
Young 1989; Nair, 1993).
Consistently higher yields were obtained in leucaena alley crop with
prunings incorporated treatment (treatment 2) than the fertilizer alone
and control treatments (Treatments 9 and 10) (Table 20.1). This is an
indication that leucaena can be used effectively in alley cropping
arrangements to improve crop yields (Mugendi et al., 1999a; Mugwe et
al., 1999). This corroborates with findings of other studies (Kang, 1993;
Xu, 1993) where biomass incorporation in alley cropping system
Soil Conservation and Fertility Improvement Using Leguminous Shrubs in Central
Highlands of Kenya: NARFP Case Study
285
increased crop yields. Calliandra alley crop on the other hand gave
significantly lower yields than leucaena alley cropping in the present
study. Other researchers working with calliandra reported mixed
performance. For instance, some have reported improved crop yields
(Heinneman,1992; Rosecrance et al., 1992), while Gutteridge (1992)
reported depressed or marginal yields.
The poor performance of calliandra in alley cropping may be
explained by the root morphology of the two species. Mugendi et al.
(1999c) in this experiment showed that over 95 % of all maize roots
were located in the top 90 cm while for calliandra and leucaena it was
60 % and 25 % respectively (Table 20.3). This agrees with findings of
other authors, for example, NAS (1993) who reported that calliandra
trees develop strong superficial root system in addition to the taproot
and Jama et al.(1997a) who demonstrated that calliandra had the highest
root density in the top 0-5 cm compared to other tree species in Western
Kenya.
Results from this experiment indicated that incorporation of prunings
of calliandra and leucaena in a maize monocrop system improves crop
yields. Also alley cropping with leucaena for soil productivity
improvement is advantageous but not with calliandra. However, in
already phosphorus deficient soils, soil productivity improvement
through alley cropping using Leucaena leucocephala or Calliandra
calothyrsus is advantageous. This is mainly because biomass produced
from the hedgerows is low and contains low amounts of P that is
insufficient to meet crop demands.
Contour hedgerow experiments (On-station and on-farm)
From the on-station contour hedgerow experiment O’Neill et al. (1998)
reported that maize grain yields from 1993 long rains to 1997/98 short
rains fluctuated with rainfall with a mean ranging from 0.64 t ha-1 during
1996 short rains (drought period) to 7.2 t ha-1 during 1997/98 (El Nino
period). However there were no significant differences between treatments
for maize grain yield during individual seasons (O’Neill et al., 1998).
This is an indication that competition between the hedgerows of Napier/
calliandra was not significant. Competition was however observed in
the alley cropping experiment where calliandra was found to lower maize
grain yields. Other studies, for example, those by Evensen, (1989),
Fernandes (1990) and Rosecrance et al. (1992) have shown competition
between hedgerows of trees and foodcrops. In the present study, lack of
significant competition can be explained by the fact that the hedges
were widely spaced with an inter-row spacing of 15 m compared to alley
cropping system which had an inter-row spacing of 4.5 m.
286
Table 20.2: Amount of prunings incorporated into the soil over the study period (1993-1998) in an on-station study at Embu, Kenya
Trt
SR 93
LR 94
SR 94
LR 95
SR 95
LR 96
LR 97
SR 97
……………...Biomass t ha-1……………..
LR 98
SR 98
Average N
supplied per
season
average P
supplied per
season
…....…..Kg ha-1….……..
2.2
1.4
2.6
1.1
2.3
1.5
2.3
3.3
3.1
2.0
62
4
2
1.7
1.2
2.5
2.2
1.3
1.3
1.6
1.2
1.0
1.2
42
3
5
2.3
1.3
2.3
1.5
3.1
2.3
2.5
3.5
3.4
2.4
70
5
6
2.3
1.2
2.6
2.4
1.9
2.1
1.9
1.5
1.5
1.4
53
4
7
2.2
1.3
2.3
1.5
3.1
2.3
2.5
3.5
3.1
2.0
70
5
8
2.3
1.2
2.6
2.4
1.9
2.1
1.9
1.5
1.5
1.4
53
4
Abbreviations:Trt = treatment (refer to page 5 for treatments description)
Nutrient concentration of prunings incorporated: N = 2.8 %P = 0.2 % (Source: Mwangi, 1997)
Nutrient contribution = quantity of prunings * nutrient in the prunings
Mugwe, J. et al
1
Soil Conservation and Fertility Improvement Using Leguminous Shrubs in Central
Highlands of Kenya: NARFP Case Study
287
Table 20.3: Total root length for maize, leucaena and calliandra at various soil depths at
maize grain-filling stage in the first season 1998 at Embu, Kenya
Depth (cm)
Maize
Calliandra
Leucaena
SED
---------- Total root length (m cm-2) -------0-30
30-60
60-90
90-120
120-150
150-180
180-210
210-240
240-270
270-300
65.5
53.4
14.3
3.4
1.8
0.4
0.1
0.1
0.1
0.1
25.7
30.3
13.7
11.8
9.9
6.3
5.1
3.4
4.2
3.8
9.4
11.3
14.7
31.2
18.6
12.8
11.6
8.2
7.4
6.6
5.4
7.2
2.1
3.6
2.7
3.3
1.3
2.6
3.2
1.2
Source: Murithi et al. (1999)
The results of runoff, soil loss and fodder production presented in
Table 20.4 indicate that contour hedges were effective in reducing soil
loss. Soil loss in the control was substantially greater than in any of the
control hedge treatments. This agrees with findings of other authors, for
example, contour hedges of Inga edulis planted in a 16% slope at
Yurimaguas, Peru with 2200 mm annual rainfall reduced soil loss on an
Ultisol from 53 to 1 t ha –1 yr-1 and runoff form 12 % of the annual rainfall
to only 2% (ICRAF, 1994). In Philippines, it was shown that contourhedgerow intercropping technology was capable of achieving a 50 to 58%
reduction in soil erosion on a 17-18% (Comia et al., 1994; Watson et al.,
1995). Kieppe (1995) demonstrated that, hedgerows reduced soil erosion
by 94% and runoff by 78% and that a combination of hedgerows and
mulch conserved 98% of the soil and 88% of water at Machakos, Kenya.
In addition to conserving soil and promoting terrace formation, the contour
hedges produced biomass that could be used as fodder source (O’Neill et
al., 1998). Indeed, Table 20.4 shows that the combination hedge (napier
plus calliandra) produced the highest biomass in all seasons and
consequently the highest crude protein.
Cumulative runoff and soil loss for three years in the on-farm trial
(Table 20.5) was higher on the 20% slope than on the 40% slope. Angima
(2000) attributed these difference to individuals farmers practices that
dated back 30 years. The farmer on the 40% slope practiced better crop
husbandry and crop rotation and periodically adds manure to his farm
while the farmer on 20% slope did not. Manure has been found to
contribute greatly to stability of soil aggregates making them less
susceptible to erosion.
Mugwe, J. et al
288
Table 20.4: Runoff, soil loss and fodder production in an on-station study at Embu,
Kenya for the 1997 long rains and the 1997/98 short rains
Runoff (mm)
1997 long rains
1997/98 short rains
Soil loss (t/ha)
1996 long rains
1997/98 short rains
Fodder - leafy dry matter (t/ha)
1997 long rains
1997/98 short rains
Fodder - crude protein* (kg/ha)
1997 long rains
1997/98 short rains
Control
Napier
Calliandra
Combination
108.0
33.0
49.0
9.0
48.0
15.0
40.0
14.0
51.0
21.1
10.0
7.5
38.0
6.4
-
1.5
4.0
1.0
1.9
-
103.0
280.0
255.0
466.0
20.0
8.3
Grass + Shrub
3.6 + 0.7 = 4.3
8.2 + 1.4 = 9.6
Grass + Shrub
252 + 170 = 422
574 + 343 = 917
Assume 7 % CP for Napier and 25 % CP for calliandra.
Source: O’Neill, et al. (1999)
Table 20.5: Cumulative runoff and soil loss for 3 years and total N and P losses in
eroded sediments in an on-farm study at the Kianjuki catchment Embu, Kenya
Treatment
20% slope
40 % Slope
.……Runoff (mm)…….
Control
Hedge
356 a
298 b
Hedge
Control
578 a
410 b
122 a
186 b
…….Soil loss (Mg ha -1)……
539 a
393 b
Hedge
Control
1.6 a
0.9 b
……P loss (Mg ha -1 yr -1 )……
2.0 a
1.4 b
Hedge
Control
1.5 a
1.0 b
……N (Mg ha -1 yr -1 )……
1.8 a
1.4 b
Source: Adapted from Angima (2000) and Murithi et al. (1998)
The hedges were found to be effective in reducing both runoff, soil
loss and nutrient loss (Table 20.5). The hedges reduced both runoff and
soil loss by 30% over the control treatment which had no hedges (Angima,
2000). This agrees with studies carried out in Ibadan, Nigeria, where
contour hedgerows of Leucaena leucocephala and Gliricidia sepium on a
7 % slope showed 85 % reduction in both soil and nutrient loss compared
Soil Conservation and Fertility Improvement Using Leguminous Shrubs in Central
Highlands of Kenya: NARFP Case Study
289
to conventional plowing (Young, 1989). Also, in Columbia, hedgerows of
Gliricidia sepium reduced soil losses from 23-35 t-1 ha-1 yr-1 under maize
to 13 Mg-1 ha-1 yr-1 on both 45% and 75% slope, resulting in a 48% soil
loss reduction (Young, 1989).
In this study, there were more nutrients lost on the 40% slope than
the 20% slope with the control plots loosing more than the hedge plots.
Murithi et al. (1999) estimated an equivalent of over 200 kg ha-1 of TSP
and 300 Kg ha-1 TSP lost by the control plots over the hedge plots for the
20% slope and 40% slope respectively. This also supports the farmer
practices of applying farmyard manure on 40% slope than on gentle slopes
of 20%. This underlines the need to control runoff and soil loss in these
zones to retain nutrients for crop production and reduce pollution of
rivers and reservoirs from eutrofication. Young (1997) has recently
documented further experimental evidence on the effectiveness of contour
hedgerows in controlling soil erosion. The evidence that covers diverse
countries, varying slopes, different climates and soils indicate the contourhedgerow systems reduced soil erosion by factors ranging from 2 to 58.
Results of this contour hedgerow experiments showed that calliandra
planted along contour hedgerows is effective in controlling soil erosion.
In addition, it produces biomass that contains high crude protein and
can therefore be used effectively as animal protein supplement.
Dissemination Potential
These studies demonstrated beneficial effects on crop yields of
incorporating prunings of calliandra and leucaena especially when
biomass is brought from outside (ex-situ). It is also evident that alleycropping with leucaena have great potential as a method of improving
sustainable yields at about 4 t ha-1 in the region. However, it has been
pointed out that the advantages of alley cropping seems to rest in the
complementarity of resource capture (Ong and Black, 1995); while it
has disadvantages in establishment costs and labour requirements.
Therefore, despite this promising results shown by alley cropping with
leucaena, the question of labour availability needs to be addressed
properly before a wide adoption by farmers can be envisaged. This
technology is labour-intensive with much of the demand for labour
occurring during the start of the rainy season which is the busiest time
of the year.
Calliandra alley-cropping system adversely affected crop yields and
should not be recommended as an alley-crop species. Reasons advanced
for this was that calliandra developed a strong superficial system that
competed with associated foodcrops for growth resources. However,
calliandra was found to be effective in controlling soil erosion when
used as a contour hedge possibly because of the strong root system
290
Mugwe, J. et al
that holds the soil together. In addition to conserving the soil, calliandra
provides up to 24 % crude protein and the large tonnage of biomass
yield from using hedges as a soil conservation resource can supplement
and in some cases substitute input protein rations in animals. Results
from western Kenya (Jama et al., 1997) indicate that fodder trees like
calliandra are most profitable when utilized as a protein supplement
for livestock.
Further research by NAFRP have have shown potential for calliandra
adoption by smallholder farmers for use in dairy production (Tuwei et
al., 1999). On-farm feeding trials have confirmed the effectiveness of
calliandra both as supplement to the basal diet and as a substitute for
dairy meal. The trials found that one kilogram of dry calliandra had
about the same amount of digestible protein as one kilogram of dairy
meal; both increased milk production by roughly 0.75 kg under farm
conditions (O’Neill et al., 1995,( Paterson et al., 1996a). Following these
findings a dissemination programme for calliandra was initiated in 1997
where the main focus was the use of participatory methods and
involvement of partners (Tuwei and Mugwe, 1998).
The dissemination procedure involved working with farmer group
nurseries. The group nurseries were provided with calliandra seeds to
raise and transplant following the onset of the rains. An evaluation of
planting niches adopted by farmers in Meru and Embu District showed
that terraces were the most preferred niche by farmers for planting
calliandra (Table 20.6). Majority (46%) indicated that they preferred
planting calliandra along terraces for soil conservation on the steep
areas. The calliandra planted on the contours for soil conservation
could help retain and cycle N in the soil for sustainable agriculture.
Studies by Jama et al. (1998) showed that calliandra roots develop
and grow rapidly into the subsoil and capture NO3– that accumulate in
the subsoil even at low available phosphorus.
Table 20.6: Niches where farmers planted calliandra during 1999 in Embu and Meru
Districts, Kenya
Niche
Terrace
External border
Homestead
Cropland
Coffee
N=178
Source: Tuwei and Mugwe (1999)
Frequency
Percentage (%)
83
63
38
19
1
46
35
21
11
0.5
Soil Conservation and Fertility Improvement Using Leguminous Shrubs in Central
Highlands of Kenya: NARFP Case Study
291
They also found that calliandra and sesbania reduced soil NO3– in
the top 2m by about 150 to 200 Kg kg N ha-1 within 11 months after
establishment and effectively captured subsoil NO3– in western Kenya.
At Embu (Experiment in this study), monocropped treatments (Maize
only) accumulated higher amounts of mineral N (averaging 15-30 Mg
NO3– N kg-1) in the 200-300 cm depth layer than treatments with
calliandra and leucaena which had an average of 1-3 mg NO-N kg-1 in
the same depth (Mugwe, 1999).
In this region, where dairy production is predominant, farmers have
the option of using prunings for direct incorporation into the soil to
improve their soils or may be better off using the calliandra for fodder.
If they use the prunings as fodder, they are then likely to benefit from
increased milk production and possibly increased incomes (O’Neill et al
1997; Franzel et al., 1999). The calliandra could be pruned for fodder 9
to 12 months after planting, and pruning continues at the rate of four
or five times per year (Roothaert et al., 1998). However, harvesting of
the biomass and removal from the site mines soil nutrients and a means
of replenishing the soil nutrients is necessary. This could be achieved
through recycling manure. This practice is feasible and economical and
is likely to have high adoption rates. If adopted, soil conservation and
fertility improvement could be enhanced.
Conclusions
These studies showed great potential for using Calliandra calothyrsus
and Leucaena leucocephala for soil fertility improvement and soil
conservation. Soil conservation improvement can be achieved through
application of leafy prunings from the leguminous shrubs into the soil.
In this study application of Calliandra and leucaena prunings into the
soil significantly increased maize grain yields over the control treatment.
The use of leguminous shrubs in an alley-cropping system however
presented mixed results. Leucaena can be used successfully as an alley
species and maintained maize grain yields at about 4 t ha-1 in the region.
Calliandra on the other hand was more competitive and depressed maize
grain yields. However calliandra contour hedges were found to be effective
in reducing soil loss, runoff and loss of soil nitrogen and phosphorus in
eroded sediments. In addition, calliandra on contour hedges produced
a large tonnage of biomass with high crude protein that could be used
as animal fodder.
Considering that smallholdings in the central highlands of Kenya
occur on sloping lands, and the importance of dairy production in the
region, there is great potential for using calliandra contour hedges for
soil conservation. The harvested prunings could either be used for soil
fertility improvement through direct soil-incorporation or could be used
292
Mugwe, J. et al
for fodder. Farmers in the region and similar areas should therefore be
encouraged to adopt this practice. If used for fodder, feasibility of
replenishing soil nutrients through recycling manure from the animals
needs to be explored.
Acknowledgements
The authors would like to thank the Swedish International Development
Agency (SIDA) and Rockefeller Foundation for funding this research
work. The Sida funded this work through the National Agroforestry
Research Project based at KARI, Embu. We would also like to
acknowledge the contribution of collaborators from KARI, KEFRI and
ICRAF in Kenya; and the staff of National Agroforestry Research Project
at Embu, Kenya. We are indebted to the many farmers who provided
their time and resources for this research.
References
Amadalo, B., Gathumbi. S., Niang, A. and Otieno, H. (1995) The effect of six
shrub species on grain yield of maize at Maseno. Biomass transfer. AFRENA,
Maseno Progress report. ICRAF, Nairobi, Kenya.
Agus, F., Cassel, D.K. and Garrity, D.P. (1998) Bromide transport under contour
hedgerow systems in sloping Oxisols. Soil Sci. Soc. Am. J. 62 (4): 10421048.
Akyeampong, E. and Muringa, K. (1994) Cutting management of Calliandra
calothyrsus in the wet season to maximise dry season fodder production in
the central highlands of Burundi. Agroforestry Systems 27: 101-105.
Angima, S.D. (2000) Soil erosion prediction using RUSLE in central Kenya. PhD
thesis. Faculty of Purdue University graduate School. USA.
Attah Krah, A.N. (1990) Alley farming with leucaena. Effect of short grazed
fallows on soil fertility and crop yields. Experimental Agriculture 26: 1-10
Comia, R.A., Paningbatan, E.P. and Hakansson, I. (1994) Erosion and crop
yield response to soil conditions under alley cropping systems in the
Philippines. Soil and Tillage Research 31: 249-261.
Evensen, C.L. (1984) Seasonal yield variation, green leaf manuring and
eradication of Leucaena leucocephala (L) de Wit. MSc. Thesis, Department
of Agronomy. University of Hawaii, USA.
Evensen, C.L. (1989) Alley cropping and green manuring for upland crop
production in West Sumatra. Ph.D Dissertation, Department of Agronomy,
University of Hawaii, USA.
FAO-UNESCO (1988) Food and Agriculture Organization of the United Nations/
UNESCO, Soil Map of the World. Revised Legend. World Soil Resources.
Report 60. FAO. Rome.
Soil Conservation and Fertility Improvement Using Leguminous Shrubs in Central
Highlands of Kenya: NARFP Case Study
293
Fernandes, E.C.M. 1990. Alley cropping on acid soils. Ph.D. Dissertation. North
Carolina State University. Raleigh, North Carolina.
Franzel, S., Arimi, H., Murithi, F.M. and Karanja, J. (1999) Calliandra
calothyrsus: Assessing the early stages of a fodder tree in the highlands of
central Kenya. AFRENA Report No. 127. International Centre for Research
in Agroforestry (ICRAF) Nairobi, Kenya.
Government of Kenya (1994) The 1989 Kenya National census Census
resultsResults. Ministry of home Home Affairs. Nairobi, Kenya.
Guevara, A.B. (1976) Management of Leucaena leucocephala (Lam) de wit for
maximum yield and N-contribution to intercropped corn. Ph.D. Dissertation,
University of Hawaii, Department of agronomy. Honolulu, Hawaii.
Gutteridge, R.C. (1992) Evaluation of a range of tree legumes as a source of
nitrogen for crop growth. Experimental Agriculture 28: 195-202.
Hartemink, A.E., Buresh, R.J., Jama, B. and Janssen, B.H. (1996) Soil nitrate
and water dynamics in sesbania fallow, weed fallows, and maize. Soil Sci.
Soc,. Am. J. 60: 568-574.
Heinnemann, M. Mengich, E.K., Olang, A.D. and Otieno, H.J.O. (1990)
Agroforestry Research Network for Africa Report No. 21. Maseno progress
report for the period January 1988 to January 1990. ICRAF, Nairobi, Kenya.
Heinnemann, A.M. (1997) Growth and yield of eight agroforestry tree species in
line plantings in Western Kenya and their effects on maize yields and soil
properties. Forest Ecology and Management 91: 103-135.
ICRAF (1992) Annual Report. International Centre for Research in Agroforestry,
Nairobi, Kenya.
ICRAF (1994) International Centre for Research in Agroforestry. Annual Report
1993. ICRAF, Nairobi, Kenya.
Jaetzold, R. and Schmidt H. (1983) Farm management handbook of Kenya. Vol
II/C, East Kenya. Natural condition and farm management information,
Ministry of agricultureAgriculture, Nairobi.
Jama., B., Buresh, R.J., Ndufa, J.K. and Shepherd, K.D. (1998) Vertical
distribution of roots and soil nitrate; tree species and phosphorus effects.
Soil Sci. Soc. Am. J. 62(1): 280-286.
Jama, B., Swinkles, R.A. and Buresh, R.J. (1997) Agronomic and economic
evaluation of organic and inorganic sources of phosphorus in western Kenya.
Agronomy Journal 89: 597-604.
Kang, B.T. (1993) Alley cropping. Past achievements and future directions.
Agroforestry systems 23: 141 – 155.
Kang, B.T, Wilson. G.F. and Sipkens, L. (1981) Alley cropping maize (Zea mays)
and Leucaena leucocephala (Lam) in Southern Nigeria. Plant and soil 63:
165 –179.
Kang, B.T., Grime, H. and Lawson T.L. (1985) Alley-cropping sequentially cropped
maize with Leucaena leucocephala on a sandy soil in southern Nigeria. Plant
and Soil 85: 267-277.
Kang, B.T., Reynolds, L. and Atta-Krah, A.N. (1990) Alley farming. Advances in
Agronomy 43: 315 - 359.
294
Mugwe, J. et al
Khan, A.H.A. and Ong, C.K. (1997 Design and calibration of tipping bucket
system for field runoff and sediment quantification. Journal of soil Soil and
water Water conservation Conservation 52 (6) 437-443.
Kapkiyai, J.J., Karanja, N.K., Woomer, P. and Quereish, J.N. (1998) Soil organic
carbon fractions in a long-term experiment and the potential for their use
as a diagnostic assay in highland farming systems of central Kenya. African
crop Crop science Science journal Journal 6: 9-28.
Kieppe, P. (1995) Effects of Cassia siamea hedgerow barriers on soil physical
properties. Geoderma 66: 113-120.
Kihanda, F.M. (1996) The role of farmyard manure in improving maize production
in the sub-humid highlands of central Kenya. Ph. D Thesis. University of
Reading, UK.
Mathews, R.B., Holden, S.T., Volk, J. and Lengu, S. (1992) The potential of
alley-cropping in improvement of cultivation systems in the high rainfall
areas of Zambia I. Chitemene and Fundikila. Agroforestry Systems 17: 219240.
Minae, S. and Nyamai, D. (1988) Agroforestry research proposal for the Coffee
based land use system in the bimodal highlands, Central and Eastern
provinces, Kenya. AFRENA Report No. 16. International Centre for Research
in Agroforestry, Nairobi, Kenya. pp 58.
Mugendi, D.N., Mugwe, J.N. O’Neill, M.K. (1999) Nitrogen movement in an Ultisol
in the central highlands of Kenya. Paper presented at the International Soil
Conservation Organization conference at Purdue University, USA, May 1999.
Mugendi, D.N., Nair, P.K.R., Mugwe, J.N., O’Neil,l M.K., Swift, M.J. and Woomer,
P.L. (1999b) Calliandra and leucaena alley cropped with maize. Part 2;
Biomass decomposition, N Minerilization, and N uptake by maize.
Agroforestry Systems 46:51-64.
Mugendi, D.N., Nair P.K.R., Mugwe, J.N., O’Neill, M.K. and Woomer, P.L. (1999a)
Calliandra and leucaena alley cropped with maize. Part 1; Soil fertility
changes and maize production in the subhumid highlands of Kenya.
Agroforestry Systems 46:39-50.
Mugwe, J.N. and Mugendi, D.N. (1999) Use of leguminous shrubs for soil fertility
improvement in sub-humid highlands of Kenya. Soil Science Society of East
Africa Proceedings of the 17th conference, Pg 194-198 (Eds) Tenywa JS,
Zake JYK, Ebayat P, Semalulu O, Nkalubo ST. 6-10 September 1999,
Kampala, Uganda.
Mugwe, J.N., O’Neill, M.K. and Mugendi, D.N. (1997) Leucaena leucocephala
and Calliandra calothyrsus for Soil Productivity Improvement in Sub-Humid
Highlands of Kenya. Pgs 181-184 ACIAR Proceedings No. 86. Leucaena
adaptation, Quality and Farming systems. Australian Centre for
International Research, Canberra, Australia.
Mugwe, J.N., Mugendi, D.N. and Gachanja, S.P. (2000) Soil productivity
Improvement through use of leguminous shrubs in subhumid highlands of
Kenya. Paper presented at the International Soil Conservation Organisation
(ISCO) conference. Held in Buenos Aires, Argentina. 22-27 October, 2000.
Muriithi, F.M., Thijssen, H.J.C., Mugendi, D.N., Mwangi, JN., O’Neill, M.K. and
Nyaata, O.Z. (1993) Report of a survey on Agroforestry technologies used
Soil Conservation and Fertility Improvement Using Leguminous Shrubs in Central
Highlands of Kenya: NARFP Case Study
295
for fodder production and soil fertility improvement in Meru District, Kenya.
National Agroforestry Research Project, Regional Research Centre, Embu,
Kenya.
Murithi, F.M., O’Neill, M,.K., Gachanja, S., Karanja, G., Kariuki, I., Mugwe,
J.N., Tuwei, P. and Roothaert, R. (1999) National Agroforestry Research
Project. KARI-Regional Research Centre – Embu, Annual Report: March
1998 – March 1999.
Mwangi, J.N., O’Neill, M.K. and Mugendi D.N. (1996) Effect of Leaf leaf Prunings
prunings incorporation on Crop crop Yields yields in Alley alley cropping
and sole Cropping cropping ssystems. East African Agricultural and Forestry
Journal. 62: 209-218.
Mwangi, J,N. (1997) Effect of soil incorporated leaf prunings on soil productivity
in alley and sole cropping systems using Leucaena leucocephala (Lam) de
Wit and Calliandra calothyrsus Meissn. M. Phil. thesisThesis,. Moi
University.,, Eldoret, Kenya.
Nair, P.K.R. (1993) An introduction to Agroforestry. Kluwer publishers, Dordrecht,
the Netherlands.
Nair,P.K.R. (1987) Soil productivity under agroforestry. In: Gholz, HL (ed),
Agroforestry, Realities, Possibilities and Potentials., pp 21-30. Martinus
Nijholf. Dordrecht. The Netherlands.
NAS, National Academy of Sciences. (1993) Innovations in tropical reforestation.
Calliandra: A versatile tree for the humid tropics. National Academy Press.
Washington DC, USA.
O’Neill, M., Gachanja, S., Karanja, G., Kariuki, I., Kanampiu, F., Kiruiro, E.,
Murithi, F.,Okoba, B., Nyaata, Z., Mwangi, J., Tuwei, P., Roothaert, R. and
Paterson, R. (1996) National Agroforestry Research Project. KARI-Regional
Research Centre – Embu, Annual Report: March 1995 – March 1996.
AFRENA Report No. 108. ICRAF, Nairobi, Kenya.
O’Neill, M., Gachanja, S., Karanja, G., Kariuki, I., Murithi, F., Nyaata, Z., Mwangi,
J., Tuwei, P. and Roothaert, R. (1997) National Agroforestry Research Project.
KARI-Regional Research Centre – Embu, Annual Report: March 1997 –
March 1998. AFRENA Report No. 120. ICRAF, Nairobi, Kenya.
O’Neill, M., Gachanja, S., Karanja, G., Kariuki, I., Kanampiu, F., Kiruiro, E.,
Murithi, F.,Okoba, B., Nyaata, Z., Mwangi, J., Tuwei, P., Roothaert, R. and
Paterson, R. (1997) National Agroforestry Research Project. KARI-Regional
Research Centre – Embu, Annual Report: March 1995 – March 1996.
AFRENA Report No. 108. ICRAF, Nairobi, Kenya.
O’NiellO’Neill, M.K., Gachanja, S.P., Karanja, G.M., Kariuki, I.W., Murithi, F.M.,
Mugwe, J.N. Tuwei, P.K and Roothaert, R.L. (1998) National Agroforestry
Research Project. KARI – Regional Research Centre – Embu. Annual Report:
March 1997 – March 1998.Agrforestry Research Networks for Africa,
AFRENA Report 120. International Centre for Research in Agroforestry,
ICRAF, Nairobi, Kenya.
O’Neill, M., Murithi, F., Mwangi, J., Nyaata, Z., Paterson, R., Roothaert, R. and
Kihanda, F. (1995) National Agroforestry Research Project. KARI Regional
Research Centre- Embu, Annual Report: March 1994-March 1995, AFRENA
Report No. 92, ICRAF, Nairobi, Kenya.
296
Mugwe, J. et al
Palm, C.A. (1995) Contribution of agroforestry trees to nutrient requirements of
intercropped plants. Agroforestry systems.
Paterson, R.T., Kiruiro, E.M. and Arimi, H.K. (1996) Input and Output data
monitoring at the zero-grazing model farm unit at RRC-Embu; the case for
Dairy Cattle Performance on Agroforestry based Feeding System. Proceedings
of Livestock Feeding Systems Project, End of Phase 1 Workshop. Izaak Walton
Inn, 30th July, 1996.
Rosecrance, R.C., Brewbaker, J.L. and Fownes, J.H. (1992a) Alley cropping of
maize with nine leguminous trees. Agroforestry systems 17: 159 –168.
Salazar, A., Scott, L.T. and Palm, C.A. (1993) Crop-tree interactions in alley
cropping systems on alluvial soils of the upper Amazon Basin. Agroforestry
systems 22: 67-82.
Ogwenyi, G.S., Kithara, S.M., and Denga, F.O. (1993) An overview of the soil
erosion and sedimentation problems in Kenya; Strategies for monitoring,
predicting and control. Proceedings of the Yokohama Symposium, July 1993.
IAHS Publication No. 21: 217-223.
O’Neill, M., Murithi, F., Mwangi, J., Nyaata, Z., Paterson, R., Roothaert, R. and
Kihanda, F. (1995) National Agroforestry Research Project. KARI Regional
Research Centre-Embu, Annual Report: March 1994-March 1995, AFRENA
Report No. 92, ICRAF, Nairobi, Kenya.
Roothaert, R., Karanja, G.M., and Kariuki, I.W., Paterson, P., Tuwei, P., Kiruiro,
E., Mugwe, J. and Franzel, S. (1998) Calliandra for livestock, Technical
Bulletin No. 1, National Agroforestry Research Project, KARI-regional
Research Centre, Embu, Kenya.
Schroth, G. (1995) Tree root characteristics as criteria for species selection and
systems design in agroforestry. Agroforestry Systems 19: 67-78.
Tuwei, P.K. and Mugwe, J.N. (1999) Adaptive on-farm technology Development;
Experience of the National Agroforestry Research Project, Embu. A case
study of Calliandra calothyrsus as an animal fodder. Paper presented in the
social forestry extension seminar for the promotion of tree planting in arid
and semi-arid areas of Kenya at KEFRI, Muguga 27-30 September 1999.
Watson, H.R., Laquihon, W.A. and Palmer, J.J. (1995) SALT vs SALT alley farms
in the Southern Philippines. In Kang et al (1995), q.v pp 366-371.
Tuwei, P.K., Wambugu, C., Karanja, G.M., and Mugwe, J. (1999) Adaptive onfarm technology development experiences of regional research centre, Embu;
A case study of Calliandra calothyrsus as an animal fodder. Paper presented
during AFRENA conference. “Is Agroforestry making a difference?”. ICRAF,
Nairobi, Kenya.
Xu, Z.H., Saffigna, R.J.K., Myers, R.J., and Chapman, A,L. (1992) Nitrogen
fertilizer in leucaena alley cropping and maize response to nitrogen fertilizer
and fate of fertilizer-15N. Fertilizer Research. Dordrecht; Kluwer academic
Academic Publishers 33 (3); 219-227.
Yadvider, S., Bijay, S. and Khind, C.S. (1992) Nutrient transformations in soils
amended with green manures. Advances in Soil Science 20:38-298.
Young, A. (1997) Agroforestry for soil management (2nd Edition). CAB
International, Wallingford, UK.
Soil Conservation and Fertility Improvement Using Leguminous Shrubs in Central
Highlands of Kenya: NARFP Case Study
297
Young, A. (1989) Agroforestry for soil conservation. International Ccentre for
Research in Agroforestry, ICRAF. Nairobi, Kenya.
Young, A. and Sinclair, F.L. (1997) The effectiveness of contour hedgerows for
soil and water conservation. Agroforestry systems 8:2-4.
298
Mugwe, J. et al
The Relationship Between Nitrogen Mineralization Patterns and Quality Indices af Cattle
Manures From Different Smallholder Farms af Zimbabwe
The Relationship Between
Nitrogen Mineralization
Patterns and Quality Indices
of Cattle Manures from
Different Smallholder Farms
in Zimbabwe
299
21
Nhamo, N.1*, Murwira, H.K.2,
Giller, K.E.3
Soil Productivity Research Laboratory (SPRL), Private Bag
3757, Marondera, Zimbabwe, Telephone: 263-79-23621 or
263-16-14367; Fax: 263-79-24279;
E-mail: nnsprl@mweb.co.zw
2
Tropical Soil Biology and Fertility (TSBF), P.O. Box MP288,
Mount Pleasant, Harare, Zimbabwe,
Email: hmurwira@zambezi.net
3
Department of Soil Science and Agricultural Engineering,
University of Zimbabwe, P.O. Box MP167, Mount Pleasant,
Harare, Zimbabwe, E-mail: ken.giller@wur.nl
1
*Corresponding author
Abstract
The use of cattle manure as a source of nutrients is
widespread among the smallholder-farmers of Zimbabwe.
Benefits of cattle manure on crop yields depend on the
amount, quality, nutrient release patterns and uptake by
the crops. Characterization of a large range of manure
Nhamo, N. et al
300
samples from Chivi, Mangwende, Shurugwi and Tsholotsho,
showed that the quality was variable with N content ranging
from 0.1% to 2.76%. Most manures had C:N ratios ranging
between 10 and 20. A laboratory study was conducted using
a selection of cattle manures with N contents ranging from
0.45 to 1.5% to determine their N release patterns, indices
of quality that correlated to N mineralization and the effect
of rate of manure application. The leaching tube
methodology was used to study the mineralization patterns
of (1) ten different manures with N contents of 0.45% to
1.5% applied at 100 kg N ha-1 equivalent and (2) two
manures with low (0.89% N) and high N content (1.5% N)
applied at 0, 5, 10, 20 and 40 t ha-1 equivalent.
The N release patterns were similar for the ten manures,
with a net immobilization phase occurring in six of the
manures in the first 14 days. Net mineralization occurred in
all the manures after the initial 14-day period. The %N, %C,
C:N, ash and lignin contents were significantly (P<0.05)
correlated to N mineralization after 56 days of incubation. A
linear model was proposed where net mineralized N was given
as 27.7N + 2.7Lignin + 2.34C:N + 1.49Ash –1.39C –86.2.
The rate of application did not cause a change in the
mineralization pattern in the low N manure (0.89% N) except
for the 40 t ha -1 treatment where there was no net N
immobilization throughout. With the high N manure content
(1.5% N), the high rates of 20 and 40 t ha-1 did not exhibit
the N immobilization phase during the initial stages of
mineralization which were seen with the lower rates. Mineral
N was positively correlated to the rate of application during
immobilization and mineralization phases.
Key words: Cattle manure quality, N mineralization, N immobilization,
manure decomposition indices
Introduction
The amounts of nutrients released from organic materials is a function
of their physical/chemical composition, the amounts applied and
environmental factors. The efficacy with which N in manure is used
depends on the synchrony between rate of mineralization and the crop
demand. Manure types used by farmers are variable due to different
management strategies. Determining N release patterns of the cattle
manures can give an estimate of the potential amount of N that a given
manure can contribute to crops.
The Relationship Between Nitrogen Mineralization Patterns and Quality Indices af Cattle
Manures From Different Smallholder Farms af Zimbabwe
301
Currently, there are several recommendations on the amount of
manure to use on maize crops in Zimbabwe (Mugwira and Murwira,
1997). The recommended quantities were derived from simple manure
response trials in different regions of the country. Addition of cattle
manure to infertile sandy soils increases nutrients for crop uptake. The
minimum amount of manure, which is required to give economical crop
yields, has not been determined.
Relationships between the mineralization of N from manure and
indices to describe manure quality have not been examined in detail.
Crop response work done in a greenhouse demonstrated that crop uptake
gives a good account of the quality of manure (Mugwira, 1984). Studies
done on green manures and agroforestry species show that it is possible
to use the release patterns, laboratory chemical indices and textural
indices to describe quality of materials and predict rates of decomposition
and N release (Melilo et al., 1982; Frankenberg and Albdelmagid, 1985;
Palm and Sanchez, 1991; Handayanto et al., 1997; Mafongoya et al.,
1997). The rate of net N mineralization of manures and other organics
must be known to optimise use and predict supplementation rates of
mineral fertilizers (Constantinides and Fownes, 1994; Hadas and
Portnoy, 1994). The paucity of such information on cattle manures makes
guidelines on their effective use, in the short and long term, difficult to
derive.
The ‘lignin’ content measured in manure depends on the cattle diet
and other constituents like the microbial remains. A diet containing
material with large proportions of high molecular weight, recalcitrant
carbon compounds will lead to high lignin contents in manures. Work
done by Mafongoya et al. (1997) on the effect of drying on litter quality
suggests that measured lignin contents also increases through the
browning reactions between polyphenols, proteins and carbohydrates.
Similar effects may also occur in the cattle rumen. This fibre or ‘lignin’
material degrades slowly in the soil. The physical protection caused by
lignin on other organic constituents causes a general reduction in the
rate at which the organic material is degraded (Haider, 1986). Manures
collected from smallholder farms in Zimbabwe also have other
contaminants like sand, twigs and wood particles, and other recalcitrant
materials.
This study aimed to characterise manures from four smallholder
farming areas, to determine N mineralization patterns of selected
manures, and examining relationships between N release and the
initial % N, %C, C:N ratio, lignin, % ash and % polyphenol (indices of
quality). The effect of rate of application of different manures on N
release was also studied following a laboratory characterization of
manures from Chivi, Mangwende, Shurugwi and Tsholotsho districts
of Zimbabwe.
302
Nhamo, N. et al
Materials and Methods
A total of 329 manure samples were collected prior to the onset of rains
in 1997 from Chivi, Mangwende, Shurugwi and Tsholotsho and analysed
for %N, %C, lignin content, %ash and polyphenols (Anderson and Ingram,
1993). Some of the manure samples were selected for further use in the
mineralization studies described below.
Nitrogen mineralization of cattle manures
N mineralization was studied using the aerobic leaching tube method
(Stanford and Smith, 1972). Glass tubes of 300 mm length with a
diameter of 50 mm and a thickness of 1 mm were used. A rubber stopper
inserted with a glass tube for drainage at the centre was used to close
the lower end of the tube.
A mass of 150 g of soil was used in each tube. This was mixed with
manure quantities equivalent to 100 kg N ha-1 per tube. Ten different
manures CM1-CM10 were used in this experiment. A set of control
tubes was included in the design, where no manure was added to the
150 g of soil. The treatments were each replicated four times.
Effect of rate of application on N release
In a parallel experiment, the effect of rate of application on N release
was investigated by using manure rates of 0, 5, 10, 20 and 40 t ha-1
equivalent as treatments. Two manures were used in this experiment
with low N (0.89% N) and high N (1.5% N). The samples were mixed with
sandy soil and added to leaching tubes.
A completely randomised design was used in both experiments with
four replicates of each treatment. Leaching tubes were randomised on
racks in the constant temperature room.
Sampling procedure
Leaching and collection of samples was done on day 0, 3, 7, 14, 28, 42
and 56 days. A leaching solution made up of 0.9 mM KCl, 1 mM CaCl2,
1mM MgSO4 and 0.1mM KH2PO4 was used. After each leaching event
the tubes were subjected to mild suction to bring the water content of
each tube to 70-80% WHC. The leachate collected was analysed for
mineral N (NH4-N and NO3-N) colorimetrically. The net mineralization
or immobilization was calculated using the difference between total N
in the amended soils and the control.
Glasswool was placed at the base to prevent soil and manure particles
from washing out of the tube. To ensure an even distribution of suction
and no further loss of the materials a layer of 10 g of acid-washed sand
The Relationship Between Nitrogen Mineralization Patterns and Quality Indices af Cattle
Manures From Different Smallholder Farms af Zimbabwe
303
was added on top of the glass wool before adding the soil-manure
mixture. Another 10 g of sand was added at the top of the soil-manure
mixture to avoid disturbance of particles from the sample on pouring in
the leaching solution. Each treatment mixture was thoroughly mixed in
a glass jar before transferring to the tube. To prevent separation of
different soil fractions and lumping up of the manure added on
transferring to the tube, about 5 ml of distilled water was added to the
soil and manure during mixing. The amount of water added was enough
to enable the mixing process without soil particles sticking to the walls
of the glass jar. Distilled water was then added to adjust each leaching
tube to 70-80 % of water holding capacity. Aluminium foil was used to
loosely cover the tops of each tube to minimise moisture losses.
Carbon was analysed using the Walkley Black method while nitrogen
was analysed after digestion using colorimetric methods (the cadmium
reduction and the salycilate methods). Lignin and ash were done via
Acid Detergent Fibre method and polyphenols were analysed using the
Folin-Ciocalteau method (Anderson and Ingram, 1993). A bulk soil
sample was collected from the plough layer (0-20 cm) from Chinonda
site in Mangwende for use in the laboratory incubations. The soil was
air dried and passed through a 2 mm sieve.
Results
Manure characteristics
The characterisation of manure samples from four smallholder farming
areas of Zimbabwe showed that the average N contents of manure at
the time of application is 0.89% (Table 21.1). The frequency of distribution
was strongly skewed with the majority of samples with N content ranging
between 0.4% and 1.2% (Figure 21.1). A similar trend was observed
with samples from each of the communal areas.
Table 21.1: Means of characteristics of manures from Chivi, Mangwende, Shurugwi and
Tsholotsho
Variable
Mean
Std Dev
Minimum
%N
%C
C:N
% Lignin
%Ash
%Polyphenols
0.89
13.1
13.7
8.7
70
0.13
0.43
6.2
3.8
5.4
17
0.12
0.10
5.4
4.2
0.4
27
0.04
Maximum No. of samples
2.76
28.8
25.3
28.7
92
0.45
329
81
81
108
92
16
Nhamo, N. et al
304
Figure 21.1: Frequency distribution of the measured N content (%) of manures from
Chivi, Mangwende, Shurugwi and Tsholotsho
25
Frequency (%)
20
15
10
5
00.
2
00.
2
00.
2
00.
2
00.
2
00.
2
00.
2
00.
2
00.
2
00.
2
00.
2
00.
2
00.
2
00.
2
0
%N
The ash contents of the manures were high ranging from 27 % to
92% with an average of 70% (Table 21.1). Only 7% of the manure
analysed had ash contents of less than 40%. About 60% had values
between 60 and 80%. A plot of the frequencies of the ash contents was
heavily skewed towards higher values. The high values in ash contents
are an indication of the high contamination from sand, which is a
common problem with manures from smallholder farmers. The C:N
ratios of the manure were all less than 30. Manure with C:N ratios
between 10 and 20 constituted 77% of the samples analysed. Only 6%
of the manures had a ratio more than 20 (Figure 21.2). The polyphenol
contents of the manures were small. A few samples analyzed for
polyphenols showed that they are not present in significant amounts.
The average ‘lignin’ content of the manures was low but some manures
had significantly high amounts of up to 29% (Table 21.1). There was a
significant positive correlation between the %N and the %C of the
manures. However, ash content and lignin explained only a small
proportion, 42% and 30% respectively, of the variability in N content
of the manures (Figure 21.3).
The Relationship Between Nitrogen Mineralization Patterns and Quality Indices af Cattle
Manures From Different Smallholder Farms af Zimbabwe
305
Figure 21.2: Frequency distribution of the measured C/N ratios of manures from Chivi,
Mangwende, Shurugwi and Tsholotsho
60
Frequency (%)
50
40
30
20
10
0
0-5
5-10
10-15
15-20
20-25
25-30
C/N ratio
Figure 21.3: Some relationships between measured variables of manures from Chivi,
Mangwende, Shurugwi and Tsholotsho
30
3
Y = 10.6x + 2.65
R 2 = 0.6
25
R 2 = 0.42
2
%N
20
15
1
10
5
0
0
0
3
2
1
20
30
40
50
60
%Lignin
%ash
%N
3
Y = 0.61 + 0.04x
R 2 = 0.3
2
%N
%C
Y = 2.4 – 0.02x
1
0
0
5
10
15
20
%Lignin
25
30
35
70
80
90
100
Nhamo, N. et al
306
Nitrogen mineralization patterns
There was net mineral N from all the manures at the start of the
incubation, the trend was however downward. This was followed by
immobilization of N for all the manures except for manure CM8 (Figure
21.4 and 21.5). Of the ten manures, four had immobilization phases.
Those which did not give net N immobilization had less mineralization
at this point compared to previous measurements. There was a
significant positive correlation between the initial %N, C:N, ash content
and %C and net N mineralised after 7 days of incubation and a weak
correlation with lignin (Table 21.3).
Table 21.2: Characteristics of manures used in the laboratory incubation experiments
Sample %
%
%
ID
Total N Org. C Lignin
% Ca
% Mg
%K
C:N
%
% Ash
ratio Polyphenol
CM1
CM2
CM3
CM4
CM5
CM6
CM7
CM8
CM9
CM10
0.22
0.18
0.23
0.27
0.30
0.41
0.70
0.82
0.86
1.19
0.07
0.04
0.06
0.07
0.10
0.10
0.15
0.18
0.20
0.29
0.53
0.33
0.78
0.45
0.70
0.66
1.10
1.02
1.31
1.59
18.0
21.1
14.5
11.5
20.2
14.5
15.3
21.5
14.9
19.2
0.45
0.47
0.56
0.60
0.89
0.89
1.10
1.24
1.29
1.50
8.10
9.90
8.10
6.90
18.00
12.90
16.80
26.70
19.20
28.80
3.8
6.1
6.1
6.5
13.8
21.1
9.2
6.5
14.5
22.9
0.04
0.07
0.05
0.04
0.07
0.14
0.33
0.12
0.16
0.45
87
83
82
84
68
56
70
78
58
31
Table 21.3: Regression equations for the two phases, the initial immobilization and the
mineralization phase
Equation
R2 value
Y = 8.13x – 7.7
Y = 0.33x – 5.2
Y = 0.15x – 2.14
Y = 0.47x – 7.8
Y = -0.2x + 13
0.73*
0.60*
0.15
0.54*
0.67*
Y = 17.6x – 35
Y = 0.86x – 32
Y = -0.15x + 31
Y = -1.21x – 0.88
Y = -0.4x + 9.2
0.60*
0.50*
0.66*
0.50*
0.70*
Indices
Immobilization phase
%N
%C
%Lignin
C/N
%Ash
Net mineralization phase
%N
%C
%Lignin
C:N
%Ash
*Significantly different from zero at P<0.05
The Relationship Between Nitrogen Mineralization Patterns and Quality Indices af Cattle
Manures From Different Smallholder Farms af Zimbabwe
307
Figure 21.4: Total N mineralization patterns for ten manures with initial N contents of
0.45-1.5% applied at a rate of 100 kg N ha-1
100
Control
CM 1
CM 2
CM 3
CM 4
CM 5
CM 6
CM 7
CM 8
CM 9
CM 10
-1
Total N Mineralized (mg kg )
80
60
40
20
0
0
10
20
30
40
50
60
Time (days)
Figure 21.5: Net N mineralization patterns for ten cattle manures with initial total N
contents of 0.45-1.5% applied at a rate of 100 kg N ha-1
CM 1
CM 2
CM 3
CM 4
CM 5
35
30
Net N mineralized (mg kg -1)
25
CM 6
CM 7
CM 8
CM 9
CM 10
20
15
10
5
0
-5
0
10
20
30
40
50
60
10
20
30
-10
Time (days)
Time (days)
40
50
60
308
Nhamo, N. et al
The results showed that between day 3 and 14 there was a N
immobilization in six of the ten manures used in the initial phase of
incubation. All the manures showed net N mineralization from the 14th
day of incubation (Figure 21.5). With the exception of CM8, the manure
treatments exhibited an increase in the net amount of mineralised N in
the first two weeks. After four weeks of incubation the C:N ratio of the
manures had significant negative correlation with the net mineralised
N (Table 21.3).
A multifactor regression on the net mineral N values obtained
showed that the indices influenced the manure decomposition process.
However different sets of indices were important for different phases
of decomposition during the period of incubation (Table 21.3). The
lignin content could not explain the variance in mineralization during
the early stages of decomposition but became important in the later
phase. Availability of easily degradable C during the beginning of the
experiment implies that more resistant C fractions would be utilized
later on in the decomposition process. For the N mineralization within
56 days a model with indices and their relative weights as a measure
of the overall influence on the N mineralization of manures was
developed where;
Net mineralized N = 27.7N + 2.7Lignin + 2.34C:N + 1.49Ash –
1.39C – 86.2. Percent N far outweighed the other indices in terms of
its weight and influence on the decomposition processes of the
manures.
Effect of manure rates on N mineralization patterns
The low N (LN) manure, rates of up to 20 t ha-1 gave an initial net N
immobilization period during the first 7 days of incubation with positive
net values only measured after 14 days (Figure 21.6). The 40 t ha-1
treatment had positive net mineralised N for the entire incubation period.
The general pattern of N release did not change with increasing quantities
of manure applied. However, for the 5, 10 and 20 t ha-1 treatments
there was no initial positive flush of mineralization with the low N
manure.
Release patterns from incubating manure with high N (HN) (1.50 %)
showed a similar initial immobilization period between days 3 and 14
for only the 5 and 10 t ha-1 treatments. Treatments with high quantities
of 20 and 40 t ha-1 did not show any immobilization phase at the
beginning (Figure 21. 6).
Figure 21.6: Net N mineralization of low (LN) and high N (HN) manure applied at different
The Relationship Between Nitrogen Mineralization Patterns and Quality Indices af Cattle
Manures From Different Smallholder Farms af Zimbabwe
309
rates equivalent to 5, 10, 20 and 40 t ha-1
35
LN5t
LN10t
LN20t
LN40t
30
HN5t
HN10t
HN20t
HN40t
Net N mineralized (mg kg -1)
25
20
15
10
5
0
-5
-10
0
10
20
30
40
50
60
10
20
30
40
50
60
Time (days)
Time (days)
Figure 21.7: The relationship between rate of application of manure and mineralized N
Net mineral N (mg kg -1)
20
35
Immobilization phase
Y = 0.45x – 6.6 r 2 = 0.66
15
Net mineralization phase
Y = 0.42x + 9 r2 = 0.72
30
25
10
20
5
15
0
10
-5
5
-10
0
0
10
20
30
40
0
10
20
30
40
Rate of manure application (t ha -1)
Discussion
Characteristics of cattle manure from study areas
Manure is managed in different ways by farmers, hence the variation in
manure quality. The frequency distribution graph shows the N content
of most manures varied between 0.4 and 1.2% (Figure 21.1). The
variations in the N contents of manures are mainly due to differences in
losses during handling, contamination with sand in the kraal and
310
Nhamo, N. et al
addition of legume and other residues to dung. The fact that less than
7% of the manure had N content greater than 1.5% while 67% had N
content less than 1% imply that the current management practices by
most farmers do not result in manures with high N content.
The ash content values were high, averaging 70%. This indicated
that sand contamination was high in all the manures from the different
areas. This is mainly because new cattle kraals in which manure
accumulates are built on loose sandy soils. Accumulation of manure on
a loose earth base results in manures mixing with the sand. High sand
content in cattle manure in Zimbabwe has also been reported by Mugwira
(1984) as an indicator of the manure quality. Contamination also occurs
when digging out the manure. Some farmers however have developed
interventions of building kraals on anthills or putting concrete floors in
order to reduce the loss in quality of the manure. There was an inverse
relationship between %N and ash content, indicating that sand
contamination causes part of this variability (Figure 21.3).
Net N mineralization pattern and manure indices
Manures with N contents higher than 0.89% (CM6-CM9) mineralised
through out the entire incubation period with the exception of CM10
(Figures 21.4 and 21.5). N immobilization was observed with six of the
manures after incubating the manured soil for two weeks. The findings
are similar to green house studies by Mugwira and Mukurumbira (1984)
in which they reported a depression in yields in the first two weeks
followed by significant plant growth increases after two weeks of planting
in manured pots. Tanner and Mugwira (1984) also observed a crop yield
depression in the first 4 weeks in a greenhouse study using manures
from other smallholder areas. Much longer periods of immobilization
from cattle manure of up to 105 days have been observed (Fauci and
Dick, 1994). These results are contrary to observation by Pathak and
Sarkar (1994) who reported cattle manure with %N of 0.79, C:N of 26
and an ash content of 27.5% mineralised throughout the entire study
period. The observation that manure with %N less than 0.89 (Figure
21. 5) immobilized more during the first 14 days is not in line with the
general conclusion drawn by Mugwira and Mukurumbira (1986). When
they classified manure into three groups they did not find any difference
in the effects of manures with %N ranging from 0.62% to 1.22% that
were all classified as medium and low quality groups.
The C:N ratio of the manure was an important index for indicating
the occurrence of a net N immobilization period. Manures with C:N
ratio less than 20 immobilized N, consistent with results of Castellanos
and Pratt (1981) who reported immobilization from manures with C:N
ratio of 15.9 and N contents of 2%. However this contrasts with reports
from workers who used plant materials who report C:N ratio of 23 as
The Relationship Between Nitrogen Mineralization Patterns and Quality Indices af Cattle
Manures From Different Smallholder Farms af Zimbabwe
311
the threshold for net mineralization (Frankenberger and Abdelmagid,
1985; Janssen, 1996; Quemada and Cabrera, 1995; Swift et al., 1979).
The implication is that neither %N nor C:N can on its own be used to
explain the mineralization patterns in manures.
An initial flush of net mineral N was found at day 3 in all the manures
studied, (Figures 21.5 and 21.6). Under field conditions this flush which
occurs after the initial wetting of the soil by the rains will provide N to
young plants early in the season. It is the balance between this initial
mineralization and leaching losses together with crop uptake that
determines the initial benefits of manures on crops.
The positive correlation between mineral N and the %N and the %C
during the immobilization period shows their importance in determining
the extent of mineralization in manures (Table 21.3). The organic carbon
content of manures is an important determinant of the mineralization/
immobilisation processes. This is shown by positive correlation with
mineralization at these initial stages of decomposition. Nitrogen and
carbon are the primary needs for soil microbes for energy and biomass
accumulation. These results corroborate studies on other organic
materials in which N and C have been identified as important indices of
mineralization (Constantinides and Fownes, 1994; Jansen, 1996). Hadas
and Portnoy (1994) also report that %N, %C and efficacy with which C
is assimilated are important in determining mineralization patterns of
organic materials.
The low percentage of the variance that is explained suggests that
indices have different overall effects on the mineralization trends with
time as the substrate concentration and composition changes. After 14
and 28 days of incubation the % ash, lignin and C:N ratio become
important indices of mineralization of these manures together with the
%C and %N. This shows that indices change in importance as the
decomposition process progresses. In studying green manures Oglesby
and Fownes (1992) found correlation of indices significant at different
times and phases of decomposition. The N content of the manure explains
most of the variance in the N release patterns of manure throughout
the incubation period thus it outweighs other indices in importance.
Whilst lignin and polyphenols are important intrinsic factors in plant
materials, the relatively low values in the manures analysed (Tables
21.1 and 21.2), renders the parameter less influential in the
decomposition processes. Palm and Sanchez (1991) reported threshold
values of lignin as 15% and 3% for polyphenols in leaves of tropical
legumes for net N mineralization to occur immediately. The average
from the four areas studied of lignin is 8.74% and that of polyphenols is
0.13%. Mostly, cattle feed on the grass from the grass velds in the
different areas and this, together with the rumen digestion process,
explains the low concentrations for these two parameters. In the short
term the lignin content did not affect the N mineralization but became
312
Nhamo, N. et al
important in the later stages. Cattle browsing on shrubs may have
manure of higher lignin contents.
The polyphenol contents of manures were low. As manure passes
through the alimentary canal, polyphenols react and form other products
through condensation reactions. Measured values from samples collected
from communal grazing areas are all less than 1% (Palm et al., 2001).
The effect of polyphenols in plant materials applied to the soil has been
those of reduced N release through their protein binding capacity. They
also bind enzymes which catalyse mineralization, in the nitrification
process in particular. However, under aerobic conditions most
polyphenols are quickly degraded (Paul et al., 1994).
Rates of manure application
As expected, increasing rate of manure application led to increases in
the overall amount of N mineralised from manure (Figure 21.4). There
was a significant positive correlation between the rate and the amount
of net mineralization during both the immobilization period and the
positive net mineralization afterwards (Figure 21.7). This is in line with
findings by Chang and Janzen (1996), who found a similar relationship
from a long term study on rates of manure application. Schmitt et al.
(1992) also found a linear relationship between release of NH4-N and
manure application rate.
The net mineralization pattern itself of the manures used in this
study did not change. Except for the 20 t ha-1 and the 40 t ha-1 treatments
of high N manure where there was no N immobilizaion phase throughout.
Other treatments in the high and low N manures all showed
immobilization within seven days of incubation. Lack of differences in
the mineralization pattern for the low rates of application implies that
the amount of substrate added to the soil did not alter mineralization in
the low N manures. However, the apparent difference in pattern with
the high N manure indicates that the C and N loading in the 40 t ha-1
treatment overrides any effects of immobilization.
Conclusions
The study on the characteristics of cattle manures from four smallholder
farming areas showed that they are highly variable, with N content that
ranged between 0.10 and 2.76% and that ash content was high averaging
70%. Most manures had a C:N ratio between 10 and 20. The %N, %C,
%ash, lignin and the C:N ratio were the most reliable indices of manure
N release. The indices however influenced the release of N from manure
at different stages of the decomposition. Whilst %N, %C, C:N and ash
The Relationship Between Nitrogen Mineralization Patterns and Quality Indices af Cattle
Manures From Different Smallholder Farms af Zimbabwe
313
content are important in the initial stages of mineralization, the lignin
content became important at a later stage. Polyphenol contents were low
in cattle manure thus the Browning reaction (reactions between
polyphenols and carbohydrates or soluble carbon) did not explain the
immobilization phase observed with the manures. It should be noted
that the ash content as an index of quality in manure is important in that
it can be estimated by hand texturing of materials in the field unlike
others that require chemical analysis. This could be important for use by
farmers as a quick test of quality. The regression model developed in this
study containing five indices, %N, %C, C:N, lignin and ash content, needs
further validation.
For all the manures used, the period of reduced mineralization was
the same, in the initial stages of incubation. It is only the extent of
immobilization not the duration which depended on quality. High N
manures, with N content greater than 1% mineralised while low N
manures immobilized N during the early stages of incubation.
The rate at which manure is added to the soil affects the amount of
net mineralised N. In both high N and low N manures, increasing the
quantities applied reduced the amount of immobilization and increased
net mineralization. The high N manure released higher net N amounts
compared with low N at the same rate of application. The rate of N
release was also high with increase in quantities of manure. Cattle
manure quality was important in determining N release pattern. The
linear relationship between quantities and release show that quantity
did not affect the mineralization pattern. The 40 t ha-1 treatment was
an exception possibly because of the amount of carbon and nitrogen
loading associated with application of large amounts of organic materials.
Acknowledgements
The authors wish to thank the International Fund for Agricultural
Development (IFAD) for providing financial support through the Tropical
Soil Biology and Fertility Programme (TSBF). We are also thankful to
staff in the Department of Soil Science and Agricultural Engineering
and the Institute of Environmental Studies, University of Zimbabwe for
providing the space and laboratory facilities during the period of the
study.
References
Anderson, J..M., Ingram, J.S.I. (1993) Tropical Soil Biology and Fertility. A
Handbook of Methods. CAB International, Wallingford.
Bremmer, J.M. (1965) Nitrogen availability indexes. In C.A. Black (Ed.) Methods
of Soil Analyses. Agronomy 10, 1324-1345, American Society of Agronomy,
Madison, Wisconsin.
314
Nhamo, N. et al
Castellanos, J.Z., Pratt, P.F. (1981) Mineralization of manure nitrogenCorrelation with laboratory indexes. Soil Science Society of America Journal
45: 354-357.
Chang, C., Janzen, H.H. (1996) Long term fate of N from animal feedlot manure
application. Journal of Environmental Quality 25: 785-790.
Constantinides, M., Fownes, J.H. (1994) Nitrogen mineralization from leaves
and litter of tropical plants: relationship to nitrogen, lignin and soluble
polyphenol concentrations. Soil Biology and Biochemistry 26: 49-55.
Fauci, M.F., Dick, R.P. (1994) Plant response to organic amendments and
decreasing inorganic nitrgen rates in soils from a long term experiment.
Soil Science Society of America Journal 58: 134-138.
Frankenberger, W.T., Abdelmagid, H.M. (1985) Kinetic parameters of nitrogen
mineralization rates of leguminous crops incorporated into soil. Plant and
Soil 87: 257-271.
Hadas, A., Portnoy, R. (1994) Nitrogen and Carbon mineralization rates of
composted manures incubated in soil. Journal of Environmental Quality 23:
1184-1189.
Haider, K. (1986) Changes in substrate composition during the incubation of
plant residues in soil. In: Jensen, V., Joun, M., Sprensin, L.H. (Eds.),
Microbial Communities in Soil. FEMS Symposium Number 33. Elsevier.
Handayanto, E., Giller, K.E., Cadisch, G. (1997) Regulating N release from tree
legume prunings by mixing residues of different quality. Soil Bioliology and
Biochemistry 29: 1417-1426.
Jansen, B.H. (1996) Nitrogen mineralization in relation to C:N ratio and
decomposibility of organic materials. Plant and Soil 181: 29-45.
Mafongoya, P.L., Dzowela, B.H., Nair, P.K. (1997) Effect of multipurpose trees,
age of cutting and drying methods on pruning quality. In: Cadisch, G.,
Giller, K.E. (Eds.), Driven by Nature: Plant Litter Quality and Decomposition.
CAB International, Wallingford, UK, pp. 167-174.
Melillo, J.M., Aber, J.D., Muratore, J.F. (1982) Nitrogen and lignin control of
hard wood leaf litter decomposition dynamics. Ecology 63: 621-626.
Mugwira, L.M. (1984) Relative effectiveness of fertilizers and communal area
manures as plant nutrient sources. Zimbabwe Agricultural Journal 81: 8589.
Mugwira, L.M., Mukurumbira, L.M. (1984) Comparative effectiveness of manure
from communal areas and commercial feedlots as plant nutrient sources.
Zimbabwe Agricultural Journal 81: 241-250.
Mugwira, L.M., Mukurumbira, L.M. (1986) Nutrient supplying power of different
groups of manure from the communal areas and commercial feedlots.
Zimbabwe Agricultural Journal 83: 25-29.
Mugwira, L.M., Murwira, H.K. (1997) Use of cattle manure to improve soil fertility
in Zimbabwe: Past, Current Research and Future Needs. Network Research
Results Working Paper No.2. Department of Research and Specialist Services,
Chemistry and Soil Research Institute, Harare, Zimbabwe.
Oglesby, K.A., Fownes, J.H. (1992) Effects of chemical composition on nitrogen
The Relationship Between Nitrogen Mineralization Patterns and Quality Indices af Cattle
Manures From Different Smallholder Farms af Zimbabwe
315
mineralization from green manures of seven tropical leguminous trees. Plant
and Soil 143: 127-132.
Palm, C.A., Gachengo, C.N., Delve, R.J., Cadisch, G., Giller, K.E. (2001) Organic
inputs for soil fertility management in tropical agroecosystems: application
of an organic resource database. Agriculture, Ecosystems and Environment
83: 27-42.
Palm, C.A., Sanchez, P.A. (1991) Nitrogen release from some tropical legumes
as affected by lignin and polyphenol content. Soil Biology and Biochemistry
23: 83-88.
Pathak, H., Sarkar, M.C. (1994) Nitrogen supplying capacity of an Ustochrept
amended in manure, urea and their combinations. Journal of the Indian
Society of Soil Science 42: 261-267.
Paul, J.W., Covert, J.A., Beauchamp, E.G. (1994) Influence of soil temperature
and moisture on water soluble phenolic compounds in manured soil.
Canadian Journal of Soil Science 74: 111-114.
Quemada, M., Cabrera, M.L. (1995) Carbon and Nitrogen mineralization from
leaves and stems of four cover crops. Soil Science Society of America Journal
59: 471-477.
Schmitt, M.A., Sawyer, J.E., Hoeft, R.G. (1992) Incubation of injected liquid
beef manure: effect of time and manure rate. Agronomy Journal 84: 224228.
Stanford, G., Smith, S.J. (1972) Nitrogen mineralization potentials of soils. Soil
Science Society American Proceedings 36: 465-472.
Swift, M.J., Heal, O.W., Anderson, J.M. (1979) Decomposition in Terrestrial
Ecosystems. Blackwell, Oxford, UK. 372pp.
Tanner, P.D., Mugwira, L. (1984) Effectiveness of communal area manures as
sources of nutrients for young maize plants. Zimbabwe Agricultural Journal
81: 31-35.
316
Nhamo, N. et al
Effect of Cattle Manure and N Fertiliser on Nitrate Leaching Losses in Smallholder
Maize Production Systems of Zimbabwe Measured in Field Lysimeters
Effect of Cattle Manure and N
Fertiliser on Nitrate Leaching
Losses in Smallholder Maize
Production Systems of
Zimbabwe Measured in Field
Lysimeters
317
22
Nyamangara, J.1* and Bergström, L.F.2
Department of Soil Science and Agricultural Engineering,
University of Zimbabwe, P.O. Box MP167, Mount Pleasant,
Harare, Zimbabwe. Email: jnyamangara@agric.uz.ac.zw
1
Division of Water Quality Management, Department of Soil
Sciences, Swedish University of Agricultural Sciences,
P.O. Box 7072, S-750 07 Uppsala, Sweden,
E-mail: Lars.Bergstrom@mv.slu.se
2
*Corresponding author
Abstract
Maize (Zea mays L.) production in the smallholder farming
areas of Zimbabwe is based on both organic and inorganic
nutrient sources. A study was conducted to determine the
effect of cattle manure, N fertiliser, and their combinations
on nitrate concentration in leachate leaving the root zone
Nyamangara, J. and Bergström, L.F.
318
and to establish N fertilisation levels which minimise N
leaching. Maize was grown for two seasons (1996/97 and
1997/98) in field lysimeters repacked with a coarse grained
sandy soil. Average leachate volumes over the two seasons
were quite similar between different treatments, and ranged
from 388 to 418 mm yr-1. Nitrogen fertiliser, especially the
high rate (120 kg N ha-1), and manure plus N fertiliser
combinations resulted in high nitrate leachate
concentrations (up to 37 mg N L-1) and nitrate losses (up to
56 kg N ha-1 yr-1) which represent both environmental and
economic concerns. Nitrate leaching from manure only
treatments was relatively low (av. less than 23 kg N ha-1 yr-1),
and plant availability in these treatments tended to be higher
in the second season. The partitioning between N leaching
and uptake depended on the intensity and timing of the
rainfall after fertiliser application. It was concluded that N
leaching at high inorganic N fertiliser rates posed a serious
economic and environmental risk when all the fertiliser was
applied at planting. It was also concluded that the risk of N
leaching from aerobically composted cattle manure was low
in the short-term.
Key words: Nitrate leaching, N uptake, lysimeter, manure, N fertiliser
Introduction
The smallholder cropping systems in Zimbabwe are based on maize,
the staple food crop, which accounts for about 50% of the calories
consumed. Cattle manure remains the major source of nutrients for
plant growth in the smallholder farming sector although some inorganic
fertiliser is also used. The Alvord system, recommended for the
smallholder farming sector of Zimbabwe, has widely been adopted by
the farmers and is based on the application of 30-40 t ha-1 of manure to
a four-course rotation of two maize crops, followed by a legume and
finally a small grain crop (Grant, 1976). However the low efficiency of
smallholder manures as sources of N (Murwira and Kirchmann, 1993;
Nyamangara, et al., 1999) has prompted farmers to supplement the
manures with inorganic N fertiliser. There is need to improve synchrony
between N release and plant uptake in order to optimise yield and
minimise N leaching losses which may occur when organic and inorganic
fertilisers are used in combination.
Effect of Cattle Manure and N Fertiliser on Nitrate Leaching Losses in Smallholder
Maize Production Systems of Zimbabwe Measured in Field Lysimeters
319
Nitrogen leaching from agricultural soil represents both an economic
loss to farmers and environmental pollutant to natural water systems.
The concentration of nitrate in groundwater, rivers and lakes has been
increasing steadily for the past thirty years in large parts of the world,
and agriculture is considered to be the major contributor (Addiscott et
al., 1991; Beckwith et al., 1998). However, the situation is quite different
to most smallholder farmers in developing countries in Africa and
elsewhere, for whom inorganic fertilisers are often unaffordable
(Kamukondiwa & Bergström, 1994a), and hence their efficient use is of
both agronomic and socio-economic importance.
Most studies on N leaching from soils amended with manure and/
or inorganic fertilisers have focussed on humid temperate regions
(Beckwith et al., 1998; Thomsen et al., 1993; Unwin, 1986), and
overall, very few quantitative measurements of N leaching have been
made in tropical and subtropical regions of Africa (Arora & Juo, 1982;
Omoti et al., 1983; Wong et al., 1987). In Zimbabwe, N leaching losses
of up to 39 kg N ha- 1 yr -1 have been reported on a sandy soil
(Kamukondiwa & Bergström, 1994b). However, the above study was
carried out during a sequence of very dry years, which limits the
representativeness of the results. Other studies, also on sandy soils
in Zimbabwe (Hagmann, 1994; Vogel et al., 1994), indicated that most
of the fertiliser (up to 54 % of applied N) was leached out of the
plough layer (0-0.5 m) when heavy rains followed N fertiliser
application. However, some of the leached nitrogen can probably be
recovered by roots later in the season.
In addition to being a potential environmental threat, large leaching
losses of N may also cause nitrate-related health problems. The World
Health Organisation (WHO) of the United Nations, the European
Community (EC) and the US Environmental Protection Agency (USEPA)
limit concentrations for nitrate in potable water at 22 (Killham, 1994),
11.3 (Addiscott et al., 1991) and 10 (Spalding & Exner, 1993) mg NO3N L-1, respectively. However, Addiscott and Benjamin (2000) recently
reported that nitrate is important in the control of gastroenteritis in
humans. Although concentrations greater than 10 mg NO3-N L-1 have
been reported in some districts in Zimbabwe (Interconsult A/S - NORAD,
1985) their link to agricultural activities is unclear.
Although the rainfall is seasonal, highly variable and generally
insufficient in most smallholder farming areas of Zimbabwe (Piha,
1993), its intensity is often very high and this may trigger N leaching
in the predominantly coarse-textured soils used for agriculture
(Twomlow, 1994). This led us to design a study in which the objective
was to measure nitrate in water leaving the root zone in agricultural
fields typical of smallholder cropping systems of Zimbabwe, and to
establish fertilisation levels which minimise N leaching losses, but
maintaining crop yields.
Nyamangara, J. and Bergström, L.F.
320
Materials and Methods
Experimental location and soil properties (Table 22.1)
The 2-year study was conducted at Domboshawa Training Centre
(17035’S, 31010’E), about 35 km north of Harare, Zimbabwe, where
average rainfall is 900 mm per annum (Agroecological Region IIa), mostly
restricted to the summer season (November-April). The soil is a well
drained, loamy sand (Typic Kandiustalf in the USDA soil classification
system, or Haplic Lixisol in the FAO system) (Nyamapfene, 1991) with a
low water holding capacity (AWC = 9% vol.) (Vogel et al., 1994).
Table 22.1: Chemical and physical properties of the experimental soil, Domboshawa,
Zimbabwe
Soil
depth
cm
0-20
20-60
pH
org-C
N1
Clay
(CaCl2)
%
mg kg-1 %
4.7
4.6
0.4
0.2
23
21
6
10
Silt
%
3
3
Fine sand Med.
%
Sand
23
20
51
52
C. sand Bulk
%
density
Mg m-3
17
15
1625
1620
1 Soil brought to field capacity using monocalcium phosphate and incubated for 14 days
at 350C before KCl-extraction (Saunder et al., 1957).
Lysimeter installations
A lysimeter station consisting of 27 repacked lysimeters was established
in the autumn of 1995 at the field site. A trench at the centre of the
lysimeter station contained twenty-seven buckets to collect leachate. The
lysimeters, square-shaped (1 m2) and 1.1 m deep, were constructed from
1.6 mm thick galvanized steel sheets. The lysimeter walls were painted to
provide a rough surface that would prevent water from channelling
between the soil and the tank walls. The lysimeter boxes were surrounded
by field soil to prevent excessive heating of the lysimeter soil.
A 1 mm wire mesh was fixed at the lysimeter outlet and covered
with a 10-cm layer of gravel before the soil was placed, reducing the
effective depth of the lysimeters to 1 m. The gravel improved drainage
(Stevens et al., 1992) and also prevented the fine soil material from
washing into the 10 mm steel outflow pipes. The pipes were laid at a
slope of about 2% to ensure rapid water flow to the collecting vessels.
The topsoil was a coarse loamy sand (0-0.3 m) overlying a sandy
loam subsoil (0.3-1.0 m). The layers were repacked to original density
following the sequence of the soil profiles identified during site
Effect of Cattle Manure and N Fertiliser on Nitrate Leaching Losses in Smallholder
Maize Production Systems of Zimbabwe Measured in Field Lysimeters
321
characterisation. Repacking the soil was considered appropriate because
it only introduces small changes in water transport and nitrogen
behaviour in course textured soils (Bergström, 1990). The lysimeters
were water saturated from the bottom end and thereafter allowed to
drain freely, and left to settle for 14 months before the experiment was
started in the summer of 1996.
Experimental design and layout
The treatments were manure (0, 12.5, and 37.5 t ha-1, which contained
0, 116 and 348 kg N ha-1) and N fertiliser (0, 60 and 120 kg N ha-1 as
NH4N03) replicated three times in a 2-factor randomised complete block
design. The manure rates were based on current recommendations for
maize in smallholder farming areas of Zimbabwe where about 37 t ha1
is applied every fourth year, or annually at about 12 t ha-1 (Mugwira &
Murwira, 1998). The 12.5 t ha-1 manure and N fertiliser treatments
were applied in both years but the larger manure treatment was applied
only in the first year.
The manure was aerobically composted and contained 0.93% N,
8.37% C (C:N = 9) and 73.7% soil. The low N and high soil contents of
the manure is typical of manures from smallholder farming areas in
Zimbabwe (Mugwira & Murwira, 1998). The manure and fertiliser, all
applied before planting, were incorporated into the top 0.1 m of the soil
at planting. Two maize plants were grown in each lysimeter during both
summer seasons, which were seeded on 3 December 1996 and 24
November 1997, respectively. The above-ground maize parts were
harvested after 12 weeks each year (milk dough stage) and the fresh
weight, dry weight and N content determined.
Leachate sampling and measurement of nitrate
concentrations
Leachate volume was recorded following each rain event when breakthrough of leachate was expected. Representative samples were taken
1-3 times each week depending on volume of leachate for nitrate-N
determination by colorimetric analysis (Keeney & Nelson, 1982). It was
assumed that the concentration of ammonium-N in leachate was
negligible.
Statistical analysis
A two-way ANOVA model in MStat (MSTAT, 1988) was used to determine
the significance of leachate volume, N leaching losses and maize N uptake
between treatments, and to determine whether there were manure x N
fertiliser interaction effects.
322
Nyamangara, J. and Bergström, L.F.
Results and Discussion
Soil, weather and drainage conditions
The 1996/97 growing season was wetter (1395 mm) than the long-term
seasonal average for the area, whereas in 1997/98 seasonal rainfall
was close to average (840 mm). Compared to 1997/98, most of the
rainfall received in 1996/97 was in the form of high intensity storms
(up to 120 mm day-1), which contributed to the larger total leachate
volumes recorded in 1996/97 (average 496 mm) compared to 1997/98
(average 311 mm) (Table 22.2). In both seasons leachate volumes
accounted for about a third of the total seasonal rainfall. The possible
effect of repacking on leachate volume was not estimated.
Above-ground N uptake
Total above-ground N uptake by maize from lysimeters which received
manure or N fertiliser was significantly (P<0.001) greater than that from
the control in both seasons (Table 22.3). There was a positive manure
and N fertiliser interaction which was more significant in the first growing
season (1996/97) (P=0.0000) than in the second growing season (1997/
98) (P=0.0004).
Table 22.2: Cumulative leachate volumes for the study period, Domboshawa, Zimbabwe
Leachate (mm)
Treatment
1996/97
1997/98
Average
Control NO
60 kg N ha-1 N60
120 kg N ha-1 N120
12.5 t ha-1 Manure ML
37.5 t ha-1 Manure MH
12.5 t ha-1 Manure + 60 kg N ha-1 N60ML
12.5 t ha-1 Manure + 120 kg N ha-1 N120ML
37.5 t ha-1 Manure + 60 kg N ha-1 N60MH
37.5 t ha-1 Manure + 120 kg N ha-1 N120MH
509
480
495
493
494
496
489
508
501
325
296
294
300
335
319
317
310
303
418
388
395
398
415
408
412
409
402
Significance
CV (%)
LSD (P<0.05)
NS
9.8
86.1
NS
11.9
64.6
NS
7.8
53.6
1The high rate of manure in the zero fertiliser treatment was only applied in the first
season.
Effect of Cattle Manure and N Fertiliser on Nitrate Leaching Losses in Smallholder
Maize Production Systems of Zimbabwe Measured in Field Lysimeters
323
Table 22.3: Total N uptake in above-ground plant parts during 1996/97 and 1997/98
growing seasons, Domboshawa, Zimbabwe
N uptake (kg N ha-1)
Treatment
1996/97
1997/98
N0
N60
N120
ML
MH
N60ML
N120ML
N60MH
N120MH
25.5
45.4
95.2
48.7
70.8
77.9
153.6
90.2
189.1
22.5
56.3
63.4
78.1
96.6
105.0
133.6
116.0
144.7
Significance
Interaction (MxF)
CV
LSD (P<0.05)
***
***
17.8
23.4
***
***
17.4
25.9
M – Manure, F – Fertiliser
Nitrogen uptake from the manure only rates (12.5 and 37.5 t ha-1)
was greater (60 and 36 % respectively) in the second season compared
to the first season. This implied that manure N became more available
(through mineralisation) for plant uptake in the second year compared
to the first year.
Nitrate concentration in leachate
Nitrate concentrations in the leachate of treatments during the
experimental period are shown in Figure 22.1. Concentrations from
lysimeters amended with manure were comparable to the control (< 15
mg N L-1) during the two growing seasons. Concentration from lysimeters
amended with N fertiliser were higher than the control in both growing
seasons. Kamukondiwa et al. (1996) also reported lower N concentration
in leachates from lysimeters receiving similar manure as used in this
study, compared to corresponding lysimeters receiving equal amounts
of N in inorganic fertilizer. However, there are examples showing the
opposite situation. For example, in a study performed under cold climate
conditions, N concentrations in drainage water was higher in poultry
manured soils than in soils receiving equal amounts of N with inorganic
fertiliser (Bergström & Kirchmann, 1999). However, it is important to
keep in mind that the manure used in that study and our study were
quite different, which precludes a direct comparison of the results.
Nyamangara, J. and Bergström, L.F.
324
Figure 22.1: Average nitrate concentration in leachate collected from replicate lysimeters
during the study period, Domboshawa, Zimbabwe. Bars represent standard errors
Control
NO3-N (mg/l)
40
30
20
10
0
D
J
NO3-N (mg/l)
60 kg N ha-1F
40
30
30
20
20
10
10
J
J
F
D
J
J
F
J
F
37.5 t ha-1M
12.5 t ha-1M
NO3-N (mg/l)
F
J
0
D
40
40
30
30
20
20
10
10
0
0
D
J
J
D
F
J
120 kg N ha-1 M + 37.5 t ha-1M
60 kg N ha-1 F + 12.5 t ha-1M
NO3-N (mg/l)
D
120 kg N ha-1F
40
0
40
40
30
30
20
20
10
10
0
0
D
J
J
D
F
J
J
F
120 kg N ha-1 F + 37.5 t ha-1M
120 kg N ha-1 F + 12.5 t ha-1M
NO3-N (mg/l)
F
40
40
30
30
20
20
10
10
0
0
D
J
J
F
D
J
J
F
Effect of Cattle Manure and N Fertiliser on Nitrate Leaching Losses in Smallholder
Maize Production Systems of Zimbabwe Measured in Field Lysimeters
325
Manure and N fertiliser treatment combinations resulted in nitrate
concentrations in leachate reaching 37 mg N L-1 when the high N fertiliser
rate (N120) was used. The depressed nitrate concentration in the high
manure plus high N fertiliser (N120MH) treatment during the second
season was not expected.
High intensity rainfall recorded soon after application of the N
fertiliser may have leached the soluble N resulting in high nitrate
concentrations in N fertiliser treatments, although much less than one
pore volume of water had leached through the profile between fertilization
and this break through of water. Therefore, there is reason to believe
that some of the fertilizer-N was displaced by preferential flow induced
by an unstable wetting front (Steenhuis & Parlange, 1991). This flow
behaviour was likely enhanced by the high rainfall intensity.
Nitrate leaching losses
The annual nitrate leaching losses ranged from 18.9 to 56.3 kg N ha-1 in
the first year and from 11.8 kg to 24.2 kg N ha-1 in the second season
(Table 22.4). The addition of N fertiliser only, or in combination with
manure, significantly (P<0.01) increased nitrate leaching in the first
(1996/97) growing season, while the addition of manure only had no
significant effect on N leaching compared to the control. In the second
season, only combinations of N fertiliser with the high (MH) manure
rate significantly increased nitrate leaching compared to the control.
There was a positive manure and N fertiliser interaction which was only
significant in the first growing season (P<0.001).
Table 22.4: Nitrate leaching losses during 1996/97 and 1997/98 growing seasons,
Domboshawa, Zimbabwe
N leached (kg N ha-1)
Treatments
1996/97
1997/98
NO
N60
N120
ML
MH
N60ML
N120ML
N60MH
N120MH
18.9
41.9
56.3
24.2
27.5
48.0
53.6
47.5
56.1
11.8
14.3
15.8
14.0
16.7
16.7
16.9
22.1
24.2
Significance
Interaction (MxF)
LSD (P<0.05)
CV (%)
**
***
19.7
13.5
*
NS
5.8
22.9
326
Nyamangara, J. and Bergström, L.F.
The higher nitrate leaching in the fertiliser treatments was attributed
to the large amounts of readily available N early in the season, whereas
most of the N derived from manure had to undergo mineralisation before
it became available for uptake and leaching. However, smallholder farmers
typically apply one third of the N fertiliser at planting and the balance at
4-6 weeks after planting, and under these conditions nitrate leaching
may be lower. In this study all the N fertiliser was applied at planting.
Overall, N uptake in all treatments was weakly correlated to N losses
for both growing seasons (R2<0.20). For manure only treatments,
correlation was only strong in the first season (R2= 0.97), whereas for N
fertiliser treatments the correlation was relatively strong for both seasons
(1996/97, R2 = 0.80; 1997/98, R2 = 0.94).
Conclusions
Due to the high intensity nature of rainfall experienced at the study site
and the low water holding capacity of the soil, the effects of differences
in plant growth on evapotranspiration were not significant. The rainfall
pattern (distribution and intensity) determined the partitioning between
N plant uptake and leaching. The increased N uptake at higher N fertiliser
rates means that high fertiliser addition within reasonable limits does
not necessarily result in more N leaching. The application of aerobically
composted manure from the smallholder farming areas of Zimbabwe to
soil does not pose an economic and environmental concern due to nitrate
leaching in the short-term. The application of inorganic N fertiliser to
sandy soils can result in high nitrate leaching losses when all the fertiliser
is applied at planting. The low manure (12.5 t ha-1) plus 60 kg N ha-1
fertiliser treatment was the best treatment in terms of maintaining dry
matter yield and minimising N leaching losses. Further studies are
required to determine the effect of manure quality and split application
of N fertiliser on plant uptake and leaching losses, and also the effect of
soil type.
Acknowledgements
This study was conducted within the Zimbabwe Soil Biology and Fertility
project sponsored by Swedish Agency for Research and Co-operation
with Developing Countries (SAREC-SIDA) to which the authors are
grateful. We would also like to thank the Institute of Environmental
Studies of the University of Zimbabwe for co-ordinating the project in
Zimbabwe and colleagues in the project for logistical support, and to
technical staff in the Crop Nutrition Section of the Department of
Research and Specialist Services, Zimbabwe, for their assistance during
soil and plant analysis.
Effect of Cattle Manure and N Fertiliser on Nitrate Leaching Losses in Smallholder
Maize Production Systems of Zimbabwe Measured in Field Lysimeters
327
References
Addiscott, T.M. and Benjamin, N. (2000) Are you taking your nitrate? Food Sci.
Tech. Today 14: 59-61.
Addiscott, T.M., Whitmore, A.P. and Powlson, D.S. (1991) Farming, fertilisers
and the nitrate problem. C.A.B. International, pp. 1-156.
Arora, Y., Juo, A.S.R. (1982) Leaching of fertiliser ions in a kaolinitic Utisol in
the high rainfall tropics: leaching of nitrate in field plots under cropping
and bare fallow. Soil Sci. Soc. Am. J. 46: 1212-1218.
Beckwith, C.P., Cooper, J., Smith, K.A., Shepherd, M.A. (1998) Nitrate leaching
losses following application or organic manures to sandy soils in arable
cropping. 1. Effects of application time, manure type, overwinter crop cover
and nitrification inhibition. Soil Use and Mgmt. 14: 123-130.
Bergström, L.F., Kirchmann, H. (1999) Leaching of total from 15N-labelled poultry
manure and inorganic N fertiliser. J. Environ. Qual. 28: 1283-1290.
Bergström, L. (1990) Use of lysimeters to estimate leaching of pesticides in
agricultural soils. Environ. Pollution 67: 325-347.
Grant, P.M. (1976) Peasant farming on infertilie soils. Rhod. Sci. News 10: 252254.
Hagmann, J. (1994) Lysimeter measurements of nutrient losses from from a
sandy soil under conventional-till and ridge-till. In: B.E. Jensen, P.
Schjonning and Mikkelsen (Eds.) Soil tillage for crop production and
protection of the environment. Proceedings of the 13th International
Conference, International Soil Tillage Research Organisation (ISTRO),
Aalbourg, Denmark, 24 to 29 July, pp. 305-310.
Interconsult A/S-NORAD, (1985) Water quality report on nation masterplan for
rural water supply and sanitation for the Republic of Zimbabwe, Harare,
Zimbabwe.
Kamukondiwa, W., Bergström, L., Campbell, B.M., Frost, P.G.H., Swift, M.J.
(1996) N leaching in manured and ammonium nitrate fertilised lysimeters
in Zimbabwe. In: van Cleemput et al (Eds.) Progress in nitrogen cycling
studies, Kluwer Academic Publishers, The Netherlands, pp. 675-680.
Kamukondiwa, W., Bergström, L. (1994a) Nitrate leaching in field lysimeters at
an agricultural site in Zimbabwe. Soil use and Mgmt. 10: 118-124.
Kamukondiwa, W., Bergström, L. (1994b) Leaching and crop recovery of 15N
from ammonium sulphate and labelled maize (Zea mays) material in
lysimeters at a site in Zimbabwe. Plant and Soil 161: 193-201.
Keeney, D.R., Nelson, D.W. (1982) Nitrogen-inorganic forms. In: A.L. Page (Ed.)
Methods of soil analysis Agronomy 9, Part II (2nd edn.), Am. Soc. Agron.
and Soil Sci. Soc. Am., Madison pp. 643-698.
Killham, K. (1994) Soil ecology. Cambridge University Press.
MSTAT, (1988) A microcomputer programme for design, management, and
analysis of agronomic research. Michigan State University, USA.
Murwira, H.K. and Kirchmann, H. (1993) Comparison of carbon and nitrogen
mineralisation of cattle manure subjected to different treatments, in
Zimbabwean and Swedish soils. In: R. Merrkx and J. Mulongoy (Eds.) The
dynamics of soil organic matter in relation to the sustainability of tropical
agriculture, John Wiley & Sons. Leuven, Belgium.
328
Nyamangara, J. and Bergström, L.F.
Mugwira, L.M., Murwira, H.K. (1998) A review of research on cattle manure as
a soil fertility amendment in Zimbabwe: Some perspectives. In: S.R.
Waddington et al. (Eds.) Soil fertility research for maize-based farming
systems in Malawi and Zimbabwe, Soil Fert Net and CIMMYT-Zimbabwe,
Harare, pp. 195-201.
Nyamapfene, K.W. (1991) The Soils of Zimbabwe. Nehanda Publishers (Pvt) Ltd.,
Harare.
Nyamangara, J., Piha, M.I. and Kirchamann, H. (1999) Interactions of aerobically
decomposed cattle manure and nitrogen fertiliser applied to soil. Nutr. Cycling
in Agroecosystems 54: 183-188.
Omoti, U., Ataga, D.O., Isenmila, A.E. (1983) Leaching of nutrients in oil palm
plantations determined by tension lysimeters. Plant and Soil 73: 365-376.
Piha, M.I. (1993) Optimising fertiliser use and practical rainfall capture in a
semi-arid environment with variable rainfall. Exp. Agric. 29: 405-415.
Saunder, D.H., Ellis, B.S., Hall, A. 1957. Estimation of available nitrogen for
advisory purposes in Southern Africa. J. Soil Sci. 8: 301-312.
Spalding, R.F., Exner, M.E. (1993) Occurrence of nitrate in ground water-a
review. J. Environ. Qual. 22: 392-402.
Steenhuis, T.S., Parlange, J.-Y. (1991) Preferential flow in structured and sandy
soils. In: T.J. Gish and A. Shirmohammadi (Eds) Preferential flow,
Proceedings of the national symposium, 16-17 December 1991, Chicago,
pp. 12-21.
Stevens, W., Mittelstaedt, W., Stock, A., Fuhr, F. (1992) Experimental approaches
to lysimeter studies: description of systems. In: Fuhr, F and R.J. Hance
(Eds) Lysimeter studies of the fate of pesticides in the soil. BCPC monograph
no. 53, pp. 21-34.
Thomsen, I.K., Hansen, J.K.., Kjeller, U.P.V., Christensen, B.T. (1993) Effects of
cropping system and rates of nitrogen in animal slurry and mineral fertiliser
on nitrate leaching from a sandy loam. Soil Use and Mgmt. 9: 53-58.
Unwin, R.J. (1986) Leaching of nitrate after application of organic manures:
lysimeter studies. In: J.H. Williams and P.L. Hermite (Eds.) Efficient land
used sludge and manure, Elsevier, London, pp. 158-167.
Vogel, H., Nyagumbo, I., Olsen, K. (1994) Effect of tied ridging and mulch ripping
on water conservation in maize production on sandveld soils. Der
Tropenlandwirt 95: 33-44.
Wong, M.T.F., Wild, A., Juo, A.S.R. (1987) Retarded leaching of nitrate measured
in monolith lysimeters in south-east Nigeria. J. Soil Sci. 38: 511-518.
Combined use of Tithonia diversifolia and Inorganic Fertilizers for Improving Maize
Production in a Phosphorus Deficient soil in Western Kenya
Combined use of Tithonia
diversifolia and Inorganic
Fertilizers for Improving
Maize Production in a
Phosphorus Deficient soil in
Western Kenya
329
23
Nziguheba, G.1, Merckx, R.1,
Palm, C.A.2 and Mutuo, P.2
1
Laboratory of Soil Fertility and Soil Biology, K.U. Leuven,
Kasteelpark Arenberg 20, B 3001 Heverlee, Belgium
2
Tropical Soil Biology and Fertility Programme (TSBF), P. O.
Box 30592, Nairobi, Kenya
Abstract
The ability of Tithonia diversifolia, fertilizers and their
combination to improve maize production in a
phosphorous (P) deficient ferralsol, was compared in
western Kenya. Tithonia and fertilizers were applied
separately or combined in different proportions to give equal
rates of 165 kg N ha -1, 15.5 kg P ha -1 and 155 kg K ha -1 in
two consecutive maize growing seasons, followed by two
residual maize crops. Maize grain yields and P recovered in
the above-ground biomass were higher in sole Tithonia than
sole fertilizer treatments. Maize yields increased with
330
Nziguheba, G. et al
increasing rate of Tithonia in the mixed treatments. When
less than 36% of the total P applied in the mixture were
supplied by Tithonia, there was no added yield benefit in
the combined treatments compared to the sole fertilizer
treatments. However, an added value ranging from 18 to
24 % increase in yields, was observed at higher Tithonia
rates. Economic returns were larger from the application of
Tithonia alone than from the application of sole fertilizers,
with larger profit when Tithonia was collected from existing
niches than when produced on site. Collecting Tithonia from
current niches resulted also in larger net returns from all
combined treatments than from fertilizers. The results of
this study indicate that a high quality organic residue such
as Tithonia can increase maize production to a greater extent
than fertilizers. The combination of Tithonia and fertilizers
can be an alternative to scarce resources and an added
benefit can be obtained by maximizing the proportion of
Tithonia in the mixture.
Key words: economic returns, leaf biomass, maize, phosphorus recovery,
relative agronomic effectiveness.
Introduction
Phosphorus has been identified as one of the major limiting nutrients
for crop production in many soils of East Africa. The use of fertilizers to
improve soil fertility in smallholder farming systems such as those found
in the East African highlands, will continue to be constrained by the
high cost of fertilizers, the low purchasing power of smallholders and
the restricted access to credit.
Although organic resources such as leaf biomass of agroforestry
tree (shrub) species do not provide sufficient P and have no effect in
increasing the total P of the system (Palm et al., 1997; Buresh, 1999),
they may increase the P availability of the already present P by rendering
it more accessible to crops. The contribution of organics as P sources
for crop production is limited by their low P content, thus requiring
large amounts to meet moderate yield increases (Palm, 1995). In densely
populated areas such as western Kenya, large amounts of organic
residues cannot be produced on small farms averaging 0.6 ha (David
and Swinkels, 1994). The limited land therefore has to be allocated to
other uses than the production of organic materials for soil fertility
replenishment. Where the materials can be found, the labour required
for collection, transport and incorporation becomes another handicap
to the use of large amounts of organic inputs (Jama et al., 2000).
Combined use of Tithonia diversifolia and Inorganic Fertilizers for Improving Maize
Production in a Phosphorus Deficient soil in Western Kenya
331
A supplementation of organic inputs with P fertilizers may be
envisaged as it addresses both the problem of insufficient fertilizer supply
and the large amount of organic material required for P supply. The
success of this strategy however will depend on many factors such as
the quality of the organic material used and the proportions of nutrient
applied from either source (Palm et al., 1997). Most trials studying the
combination of organic materials and mineral fertilizers have failed to
provide conclusive guidelines of the interactive effects of nutrients
supplied by the various sources in combination because nutrients were
not balanced. Total nutrients in the combined treatments were often
the sum of the nutrients supplied by each nutrient source applied alone,
explaining the higher yields from the combination compared to either
source (Gachengo et al., 1999).
Substitution type of experiments in which total nutrients supplied
by organic and inorganic inputs added separately or combined in
different proportions are equal (Mittal et al.,1992) provide the appropriate
design for investigating the effects of combining organic and inorganic
nutrient sources. Considering the current knowledge on the role of
organic residues in reducing the soil P adsorption capacity (Easterwood
and Sartain, 1990), increasing the pH (Kretzschmar et al, 1991) and
increasing soil biological activity (Smith et al, 1993), we hypothesize the
combination of organic and inorganic nutrient sources to be more
beneficial than the sole application of fertilizers.
Nziguheba et al (1998), reported that the combination of Tithonia
diversifolia (Hemsley A. Gray referred to as Tithonia in the text) and
TSP at 15 kg P ha-1 had a similar or larger effect on available P pools
than the sources applied alone at equal P rates. Whether crop response
to the combination of Tithonia and fertilizers reflects the observed effects
on soil P needs to be confirmed.
Therefore, a field experiment was conducted to:
(i) assess the ability of leaf biomass of Tithonia to substitute for equal
amounts of NPK mineral fertilizers for maize production,
(ii) test possible added benefits of the combined use of fertilizer and
Tithonia as opposed to sole application of either P source,
(iii) to determine the residual effects of the various sources and their
combination on maize production, and
(iv) to compare the economic returns of maize produced using Tithonia
and inorganic fertilizers applied alone or in combination.
Materials and Methods
Study site
The field experiment was conducted in the highlands of western Kenya
(altitude 1450 m). The area has 2 growing seasons per year (a long
332
Nziguheba, G. et al
rainy season from March to August and a short rainy season from
September to January), with a mean annual rainfall of 1800 mm. The
soil was classified as a ferralsol (FAO, 1990), with the following
characteristics in the top 0.15 m:
pH (soil/water suspension 1.25) = 5.4,
organic C = 15 g kg-1, exchangeable Ca = 4.6 cmolc kg-1,
exchangeable Mg = 1.9 cmolc kg-1,
exchangeable K = 0.08 cmolc kg-1,
exchangeable acidity = 0.25 cmolc kg-1.
The bicarbonate extractable P = 0.9 mg kg-1 (Olsen and
Sommers, 1982).
The soil has a clay content of 55%, silt 25%, and sand 20%.
Experiment design and management
This nutrient substitution trial was established in the short rainy season
of 1997, in a randomized complete block design with four replications
and 8 treatments. The treatments consist of Tithonia fresh leaf biomass
and inorganic fertilizers (urea, TSP and KCl), applied separately or
combined in different proportions to supply equal N, P and K rates of
165 kg N ha-1, 15.5 kg P ha-1 and 155 kg K ha-1 (Table 23.1).
Table 23.1: Description of treatments used to assess the combination of Fertilizer and
Tithonia for maize production in a field trial on a Ferralsol in western Kenya
Treatments
Code
Amount of nutrient added
(kg ha-1)
From Tithonia
Control
NOK
NPK
NPK + 0.45 Mg Tithonia
NPK + 0.9 Mg Tithonia
NPK + 1.8 Mg Tithonia
NPK + 3.6 Mg Tithonia
Tithonia
F1 + T1
F2 + T2
F3 + T3
F4 + T4
N
0
0
0
15
30
60
120
165
P
0
0
0
1.4
2.8
5.6
11.2
15.5
K
0
0
0
14
28
56
112
155
% P from
From fertilizer
N
P
0
0
165 0
165 15.5
150 14.1
135 12.7
105 9.9
45 4.3
0
0
K
0
155
155
141
127
99
43
0
0
9
18
36
72
100
At these rates phosphorus would be the only limiting nutrient. Six
rates of Tithonia were applied, 0, 0.45, 0.9, 1.8, 3.6, 5 Mg ha-1 on a dry
matter basis. A control treatment (no inputs) and a treatment with the
full rates of N and K but without P addition (NOK), were included as
references. Treatments were broadcasted and incorporated with hoes
in the top 0.15 m for 2 consecutive cropping seasons (input phase).
Combined use of Tithonia diversifolia and Inorganic Fertilizers for Improving Maize
Production in a Phosphorus Deficient soil in Western Kenya
333
This was followed by 2 consecutive maize growing seasons without
treatment additions to study the residual effect of the different inputs
(residual phase). Plot sizes were 5.25 m x 5 m. The average nutrient
concentrations of Tithonia leaves during the input phase were 33 g N
kg-1, 3.1 g P kg-1 and 31g K kg-1.
Maize ( Zea mays L.) hybrids 511 and 512 were planted respectively
in the short and long rainy seasons at a spacing of 0.75 m x 0.25 m.
Two seeds were sown per hole and thinned to one after germination.
Weeding was done whenever appropriate. At maturity, maize was
harvested and the fresh weight taken. Subsamples of cobs and stover
were taken from each plot and air-dried. At the end, maize grain yields
were expressed on a 15% water content. The above-ground maize
biomass and weeds were removed from the plots at each harvest.
Plant analyses
Phosphorus concentrations in grain and stover at the harvest of crop
2 were analyzed and used to calculate the amounts of nutrient held
in the above-ground biomass for the different crops. It was shown
from earlier studies in the same area that P concentrations do not
change significantly during seasons within treatments (Gachengo et
al., 1999).
Samples from maize stover and grain collected at the harvest of the
second crop were air-dried and ground to pass a 0.5 mm sieve.
Phosphorus in the samples was extracted using the sulphuric acid
Kjeldahl digestion method (Anderson and Ingram, 1993) and determined
colorimetrically by the method of Parkinson and Allen (1975).
In order to compare the P source effect for different cropping seasons,
maize yields were converted to relative increase compared to the NOK
treatments. Yield increase was calculated using the following formula:
Yield increase (%)
Yield treatment Yield NOK
Yield NOK
x 100
(1)
Relative agronomic effectiveness (RAE) values of the P sources relative
to yield obtained in the sole fertilizer treatment were calculated using
the formula:
RAE (%)
Yield treatment Yield NOK
Yield NPK – Yield NOK
x 100
(2)
The efficiency of P applied in the different treatments was estimated
by calculating the P recovered in the above-ground biomass of maize
(stover, core, grain) from the P applied in the treatments using the
formula:
334
Phosphorus recovered (%)
Nziguheba, G. et al
(P uptake
treatment
P uptake
NOK)
(3)
x 100
P added
Economic analysis
The economic returns from the application of each treatment were
calculated based on the partial budgeting, which included only added
costs and added benefits from the control treatment (CIMMYT, 1988).
Added costs included all the expenses for buying, collecting, transporting
and applying the inputs, while the added benefits referred to the gain
obtained by selling the harvested maize grain at the local market (Table
23.2).
Table 23.2: Parameters used to calculate the economic returns of fertilizers and Tithonia
applied alone or in combination in a maize-based system of western Kenya
Parameter
Actual values
Price of TSP
Price of urea
Price of KCl
Labor cost
Labor cost for planting
Baseline labor for application of fertilizers
Price of Tithonia
Labor for application of Tithonia collected within
the homestead
Price of maize
0.41 USD kg -1
0.38 USD kg -1
0.44 USD kg -1
0.16 USD h -1
17.36 USD ha -1
1.8 USD ha -1 (a)
0.04 USD kg -1 DM(b)
2.9 USD 100 kg-1 DM (c)
0.20 USD kg -1
Key
DM= dry matter basis
Baseline labor cost for the application of fertilizers correspond to the application of 44 kg
N ha -1 of urea, 10 kg P ha -1, and 50 kg K ha -1. For additional fertilizers application, an
extra cost of 2% was added per kg of nutrient (Jama et al, 1997). The price is based on
the value of maize which would be produced on the land used for Tithonia production.
Collection of Tithonia out of the homestead required an additional transport cost which
depends on the distance of collection. In this case, the extra cost was assumed to take
20 % of the labor cost of Tithonia collected within the homestead.
Price of fertilizers and their transport cost were determined through
a local market survey. Three scenarios were used in the determination
of the cost of Tithonia. The first scenario was based on the current
situation where Tithonia is collected from existing niches. Only the labor
for collection, transport and application is counted as Tithonia cost.
Combined use of Tithonia diversifolia and Inorganic Fertilizers for Improving Maize
Production in a Phosphorus Deficient soil in Western Kenya
335
If the use of Tithonia is adopted at large scale, it is unlikely that
Tithonia collected from current niches will satisfy the demand of farmers.
In such scenario, farmers will need to grow Tithonia. Growing Tithonia
requires that farmers sacrifice part of their land normally used for crop
production (e.g maize), for Tithonia production. The production cost is
therefore estimated by the price of maize, which would be produced on
the same plot without application of fertilizers (scenario 2). On very
depleted land, a minimum of P fertilization may be required for Tithonia
establishment. In this case the cost of production of Tithonia will also
include the cost of fertilizers applied on Tithonia (10 kg P ha-1) (Scenario
3). It was assumed that 5 Mg ha -1 of Tithonia on dry matter basis (2
cuttings) can be produced per year (Jama et al., 2000), requiring a
sacrifice of two maize crops estimated to 1 Mg ha-1 of grain under farmers’
conditions (Shepherd and Soule, 1998).
The application of urea and TSP at 44 kg N ha-1 and 10 kg P ha-1 was
estimated to take an extra 7% of the total labor cost required for planting
(Jama et al., 1997). The labor for collection, transport and application of
Tithonia within the homestead was estimated to 2.9 USD 100 kg-1 on dry
matter basis (Rommelse, AFRENA, pers. communication, 2000). The
collection of Tithonia from existing niches (scenario 1) requires an
additional transport cost, which depends on the distance of collection.
For this reason, an extra 20 % of labor cost was added for Tithonia in
scenario 1 compared to scenarios 2 and 3 to adjust for the added transport
cost. The labor was valued at the local wage rate of 0.16 USD per hour.
Harvested yields in each treatment were reduced by 10% to adjust to
realistic values if the experiment was to be managed by the farmer.
Monetary values were converted to US dollars (USD) at the exchange
rate of 74 Ksh=1 US$ (May, 2000).
The net benefit from each treatment was then determined as the
difference between added benefits and added costs. No dominance test
for checking the marginal rate of return was done because treatments
with highest net benefit had the lowest added cost, thus dominating all
other treatments (CIMMYT, 1988)
Data analysis
Analysis of variance was conducted using the general linear model
procedure (GLM) of SAS (SAS institute, 1995), to compare treatment
effects on the parameters studied. Standard errors of difference in means
(SED), were used for treatment comparison. Statistical significance refers
to α = 0.05. In order to check the interaction effect from combining
Tithonia and fertilizers, maize response to the integrated sources was
compared to the expected response determined by the equation:
Yi = ai Yf + bi Yt,
336
Nziguheba, G. et al
where:
Yi = the expected maize yield from treatment i,
Yf = maize yield obtained from the application of fertilizers alone,
Yt= maize yield obtained from the application of Tithonia alone, ai = the
proportion of P applied from the fertilizers in treatment i, and
bi= the proportion of P applied from Tithonia in treatment i (bi=1-ai).
Single degree freedom contrasts were also run to check the significance
of the benefit or reduction of yield from the interaction.
Results
Maize grain yield
Maize yields of the control treatment were significantly increased by the
addition of phosphorus sources. Despite the heavy rains (El Niño)
experienced during the first season, addition of P sources more than
tripled the yields obtained from the control treatments (Table 23.3).
Table 23.3: Maize grain yields over 4 consecutive seasons (2 fertilized, 2 residual) as
affected by application of organic and inorganic sources of nutrients to a Ferralsol in
western Kenya
Mg ha-1
Treatments
Input phase
Control
NOK
NPK
F1 + T1
F2 + T2
F3 +T3
F4 +T4
Tithonia
SED
Residual phase
Crop 1
Crop 2
Crop 3
Crop 4
0.3
0.8
1.1
1.1
1.1
1.3
1.3
1.4
0.3
0.7
0.8
3.6
3.6
3.6
4.0
4.2
4.3
0.3
0.00
0.01
0.08
0.06
0.12
0.12
0.14
0.17
0.04
1.0
1.7
2.6
2.2
2.8
2.8
2.2
3.0
0.4
Total
2.0
3.3
7.4
7.0
7.6
8.2
7.8
8.9
0.7
However, there was no significant difference between treatments
receiving a P source, although the highest increase was observed from
the sole application of Tithonia. The same trend was observed at the
second season but with much more yield increases from treatments
receiving P compared to the control (414% to 514 %).
The total maize grain yields during the input phase were 1.0 Mg ha-1
in the control treatment, and 1.6 Mg ha-1 in the NOK treatment, while
Combined use of Tithonia diversifolia and Inorganic Fertilizers for Improving Maize
Production in a Phosphorus Deficient soil in Western Kenya
337
they ranged from 4.7 Mg ha-1 to 5.7 Mg ha-1 in treatments receiving P.
Maize grain yield tended to increase with increasing amount of
Tithonia in the combined treatments with the sequence:
sole fertilizer = fertilizer + (0.45 Mg) Tithonia = fertilizers + (0.9 Mg)
Tithonia < fertilizers + (1.8 Mg) Tithonia < fertilizers + (3.6 Mg)
Tithonia < sole Tithonia.
All treatments having more than 36 % of the total P supplied by
Tithonia more than tripled the yields of the NOK treatment. The total
yield from sole fertilizers (4.7 Mg ha-1), was significantly lower than that
from sole Tithonia (5.7 Mg ha -1).
The integration of resources with less than 36% of the total P supplied
by Tithonia resulted in similar grain yields as the fertilizers applied
alone, but lower than the sole application of Tithonia. However, when
the P supplied by Tithonia accounted for 36% or more of the total P
applied in the combination, a yield increase of at least 0.6 Mg ha-1 was
observed compared to the addition of fertilizers alone, although this
increase was not significant.
The first season of the residual phase was affected by a drought
resulting in a nil grain harvest in the control treatment (Table 23.3).
However, treatments followed the same trend as observed during the
phase of treatment additions. For the second residual season,
treatments receiving a P source still more than doubled the yield of
the control. However, yields tended to be lower in the treatment
receiving 9% P from Tithonia than in sole fertilizers. Two reps of the
combination of fertilizers with 3.6 Mg ha -1 of Tithonia gave very low
yields in the second residual crops. As a result, the average yield in
this treatment was very low. The reason for such low yields could not
be identified and was most probably not due to treatment effect. For
this reason, the treatment will not be considered when discussing the
residual effect of treatments.
Total maize yields obtained in the 2 seasons of residual crops were
1.0 Mg ha-1 in the control and 1.7 Mg ha-1 in NOK. Yield increases in
treatments which previously received a P source relative to the NOK
treatment ranged from 38 % to 90% and were less than half the increases
obtained after the 2 seasons of treatment additions. Although not
significant, an extra 0.5 Mg ha -1 grain yield was obtained from the
residual maize after sole Tithonia addition compared to residual maize
from sole fertilizers. The only significant difference observed between
the treatments which previously received P source was from the
combination of fertilizers and Tithonia with 9% of total P supplied by
Tithonia, which resulted in lower maize yield than the yield obtained
from the addition of sole Tithonia.
338
Nziguheba, G. et al
Relative agronomic effectiveness of the different P sources
Addition of sole Tithonia resulted in an added maize yield of 32 %
compared to the addition of fertilizers alone at the end of the input
phase (Table 23.4).
Table 23.4: Relative Agronomic Effectiveness (RAE) of Tithonia and its combination
with TSP compared to TSP in a maize-based system in the highlands of western Kenya
Treatments
F1 + T1
F2 + T2
F3 + T3
F4 + T4
Tithonia
* RAE (%)
P added per
season (kg ha -1)
15.5
15.5
15.5
15.5
15.5
Yield treatment Yield NOK
Yield NPK – Yield NOK
RAE* (%)
Input
Residual
Total
100
100
118
124
132
60
116
122
61
141
90
104
119
108
135
x 100
The combination of fertilizers and Tithonia with 9 % or 18% of P
supplied by Tithonia resulted in maize yields equivalent to those obtained
by applying fertilizers only (RAE = 100 %). Addition of 15.5 kg P ha-1 from
the combination of fertilizers and Tithonia with 36 % and 72 % of P supplied
by Tithonia had REA values of 118 % and 124 % respectively (Table 23.4).
The benefit from the addition of Tithonia compared to fertilizers was
still observed during the residual phase where an added yield of 41 % was
harvested from Tithonia applied alone. The combination of fertilizers and
Tithonia at 9 % P from Tithonia reduced the residual yield from fertilizers
alone by 40%, while a benefit ranging from 16% to 22% was obtained
from other combined treatments.
There was no significant interaction between the fertilizers and
Tithonia. However, combining fertilizers and Tithonia with less than 36 %
P supplied by Tithonia tended to reduce the yields predicted from the
yields obtained by Tithonia and fertilizers applied separately (Figure 23.1).
When the proportion of P from Tithonia increased to 36% and above,
maize yields tended to increase above the predicted values.
Phosphorus recovered in the above-ground biomass of
maize
Differences in phosphorus recovered from P added in the different
treatments were hardly significant. However, P recovered after the first
application of treatments from Tithonia applied alone was the highest
and twice the value obtained from the sole fertilizer additions (Figure 23.2).
Combined use of Tithonia diversifolia and Inorganic Fertilizers for Improving Maize
Production in a Phosphorus Deficient soil in Western Kenya
339
Figure 23.1: Cumulative maize grain yields from the P added in Tithonia, fertilizers and
their combination. Bars indicated the standard errors of differences in means. Number
of replicates = 4.
Figure 23.2: Cumulative P recoverd in the above-ground biomass of maize from the P
added in Tithonia, fertilizers and their combination. Bars indicated the standard errors of
differences in means. Number of replicated = 4.
The total P recovered at the end of the experiment in the 4 maize
crops, was 41% of the P applied in the sole Tithonia treatments compared
to 28% in the sole fertilizer treatments.
The combination of Tithonia with fertilizers resulted in larger values
of P recovered compared to the pure fertilizer treatments, but were
smaller compared to the sole Tithonia treatment, although the differences
were not statistically significant.
340
Nziguheba, G. et al
Economic analysis
The net economic return from the application of N and K fertilizers without
P was negative even after including the residual yields (Table 23.5).
Table 23.5: Economic returns from addition of Tithonia, fertilizers or their combination to
a P deficient Ferralsol in a maize-based system
USD ha -1
Treatments
Added Cost
Added Values
Net Benefits
Input
Labor Total cost 2 seasons 4 seasons 2 seasons4 seasons
1 season 1 season 2 seasons
NOK
NPK
Tithonia (1)
Tithonia (2)
Tithonia (3)
Combine
200
222
2.6
2.6
176
147
147
F1 + T1
F2 + T2
F3 + T3
F4 + T4
276
249
194
84
18
34
66
128
F1 + T1
F2 + T2
F3 + T3
F4 + T4
294
285
266
228
16
29
55
108
F1 + T1
F2 + T2
F3 + T3
F4 + T4
296
289
274
244
16
29
55
108
(1)
(2)
(3)
273
304
551
614
352
694
738
Tithonia (1)
590
567
519
425
Tithonia (2)
620
628
642
671
Tithonia (3)
624
635
657
703
102.6
664.2
846
846
846
219.6
972
1233
1233
1233
-448
50
494
152
108
-331
358
881
539
495
664
662
765
799
895
1001
1114
1031
75
96
246
375
305
434
596
607
664
662
765
799
895
1001
1114
1031
44
35
123
128
274
373
473
361
664
662
765
799
895
1001
1114
1031
40
27
108
97
270
365
457
329
no production,
production of Tithonia without fertilizers,
production of Tithonia with P fertilization
The application of a small dose of P fertilizers (15.5 kg P ha-1) reversed
the trend to a positive net benefit of 50 USD.
Net benefits from Tithonia and the combined treatments were
positive, but depended on the strategy adopted for Tithonia production.
For the current situation where only the labor cost for cutting, carrying
and applying Tithonia are involved, higher net benefits were obtained
in treatments receiving Tithonia than in the sole application of fertilizers.
The benefits increased with increasing proportion of P from Tithonia,
Combined use of Tithonia diversifolia and Inorganic Fertilizers for Improving Maize
Production in a Phosphorus Deficient soil in Western Kenya
341
the highest being obtained from Tithonia applied alone (494 USD after
2 growing seasons) (Table 23.5, Figure 23.3).
Figure 23.3: Nets benefits from addition of Tithonia, fertilizers and their combination as
affected by the proportion of P added in Tithonia and the strategy of Tithonia production.
(1): Tithonia collected from existing niches, (2): Production of Tithonia without fertilizers,
(3): Production of Tithonia with P fertilization
Growing Tithonia, however, reduced its benefits. Although net
benefits in treatments receiving Tithonia were still positive, they were
larger than those obtained from fertilizers alone only when Tithonia
was applied at 1.8 Mg ha-1 or more.
When the residual crops were included in the economic analysis,
net benefits were larger in treatments which received Tithonia than in
sole fertilizer treatments for all strategies, except for the combination of
fertilizers with 0.45 Mg of Tithonia (Table 23.5).
Discussion
One of the major constraints to a proper management of fertilizers in
smallscale farming systems is the lack of information on limiting
nutrients. The results of this study highlighted the importance of P
fertilization on this site. Addition of N and K without P did not increase
significantly the yield of the unfertilized plots and resulted to negative
net benefit. Phosphorus either supplied by TSP, Tithonia or their
combination dramatically increased the yields compared to the control
treatments. However, Tithonia proved to be a more efficient P source
than fertilizers by providing the largest increase in maize grain yields,
the largest P recovery and an added value at least 32% higher compared
342
Nziguheba, G. et al
to sole fertilizers. These results are consistent with the work of Nziguheba
et al. (2000), who observed an added benefit going up to 112% from the
addition of Tithonia compared to TSP on resin extractable P during one
maize growing season. The results suggested that either the addition of
Tithonia converted part of non-available P forms into available P forms,
or the P added from fertilizers is easily transformed into non-available
forms, reducing their efficiency as P sources. Nziguheba et al. (1998)
found a decrease in P sorption from the application of Tithonia but not
from TSP at equal P rates. Phan Thi Công (2000) reported a reduction of
extractable aluminium and an increase of soil pH after addition of
Tithonia. Aluminium plays an important role for soil phosphorus fixation
(Mokwunye et al, 1986). In addition, an increase in microbial biomass
was observed in Tithonia treatments and not in TSP treatment
(Nziguheba et al,1998). The microbial biomass constitutes a potential
source of nutrients to the crop through turnover.
The integration of fertilizers with organic inputs has been regarded
as a more profitable alternative in low input systems, countering the
large costs of fertilizers (Janssen, 1994). This study confirmed that
the integration of fertilizers with Tithonia can be an alternative to the
limited use of fertilizers. However, higher proportions of P from organic
materials were required to get a small benefit from the combination.
From the results here, it can be deduced that the final goal is to
maximize the proportion of P supplied by Tithonia in the combination.
However, this strategy has many limitations. Large amounts of Tithonia
are not only difficult to produce but also require much labor for cutting,
carrying and incorporating (Jama et al., 2000). It is also important to
consider that addition of Tithonia does not constitute an added input
of P in the system but rather enhances the reutilization of P already in
the system. For soil P replenishment, an addition of P fertilizers remains
unavoidable.
Although the rate of P added was small, its effects were still observed
in the second residual crop. However combining fertilizers and Tithonia
with only 9% P supplied by Tithonia resulted in a reduction of the maize
yield in the pure fertilizer treatment, during the residual phase. This
therefore means that not any proportion of combination is advantageous
in an integrated nutrient management. No reduction of yields was
observed, (at any phase), from the combination of fertilizers and Tithonia
when 36% of P were supplied by Tithonia. Maize yields from this
combination tended to move above the expected values during the input
phase (Figure 23.1) and net benefits were higher than in sole fertilizer
treatments. Therefore this combination appears to be more advantageous
than the others. This combination also is in line with the quantity of
Tithonia biomass available to farmers in western Kenya, estimated
between one and two Mg ha-1 on a dry matter basis (Buresh and Niang,
1997).
Combined use of Tithonia diversifolia and Inorganic Fertilizers for Improving Maize
Production in a Phosphorus Deficient soil in Western Kenya
343
The major constraint to the use of biomass transfer for P fertilization
is the labor required for collection, transport and application of organic
residues. This study showed that this labor accounted for almost half
of the total cost of Tithonia treatments, for each season. Despite this
large cost, net benefits in treatments receiving Tithonia were positive
but depended on the strategy adopted for Tithonia production.
Application of Tithonia from existing niches resulted in larger net benefits
than the application of fertilizers. Although this strategy appeared to be
economically beneficial, it can only be envisaged at a short-term, before
the full adoption of the technology. In the long-term, the production of
Tithonia by farmers themselves needs to be envisaged. Growing Tithonia
on land normally used for maize production reduced the net benefit
obtained from collecting Tithonia from existing niches, particularly if
Tithonia is assumed to be produced with a minimum of fertilization.
However, the net benefits at high Tithonia rates were still larger than
with the sole application of fertilizers.
Due to limited land availability, the probability of growing Tithonia
on land reserved for crop production is very low. In small-scale farming
systems, the most likely scenario will be to grow Tithonia on marginal
areas or on field boundaries. In this case, the cost of production will be
limited to establishment and maintenance costs.
Conclusion
In P deficient soils, applying P in the form of soluble fertilizers, Tithonia
or their combination is very crucial for maize production. Substantial
maize yields and economic returns were obtained from Tithonia applied
alone. This study showed that a high quality organic resource such as
Tithonia can play an important role in supplying P to a growing crop.
However, considering the constraints related to the availability of Tithonia
biomass and the need for soil P replenishment, a combination of Tithonia
and fertilizers will be a more sustainable strategy, the goal being to
maximize the proportion of P derived from Tithonia in the combination.
Acknowledgement
We would like to thank John Mukalama for his assistance in the
management of the experiment and Catherine Gachengo for her
contribution in plant analysis. Financial support for this work was from
International Development Research Center (IDRC) and the Soil, Water
and Nutrient Management Programme of the Consultative Group on
International Agricultural Research (CGIAR) through grants to the
Tropical Soil Biology and Fertility Programme (TSBF), and a scholarship
from the Catholic University of Leuven, Belgium.
344
Nziguheba, G. et al
References
Anderson, J.M. and Ingram, J.S.I. (1993) Tropical Soil Biology and Fertility: A
Handbook on Methods. CAB International, Wallingford, UK.
Buresh, J.R. (1999) Agroforestry strategies for increasing the efficiency of
phosphorus use in tropical uplands. Special issue on phosphorus
availability, uptake and cycling in tropical agroforestry. Agrofor. For. 9: 812.
Buresh, J.R. and Niang, I.A. (1997) Tithonia diversifolia as a green manure:
awareness, expectations and realities. Agrofor. For. 8:29-30.
CIMMYT, (1988). From agronomic data to farmer recommendation: An economics
training manual. Mexico CIMMYT.
David, S. and Swinkels, R.A. (1994) Socioeconomic characteristics of households
engaged in agroforestry technology testing in Western Kenya. AFRENA report
no 78. International Centre for Research in Agroforestry Nairobi.
Easterwood, G.W. and Sartain, J.B. (1990) Clover residue effectiveness in
reducing orthophosphate sorption on ferric hydroxide coated soil. Soil Sci.
Soc. Am. J. 54: 1345-1350.
FAO, (1990) FAO-UNESCO Soil Map of the World Revised Legend. Soils Bulletin
60, FAO, Rome.
Gachengo, C.N. Palm, C.A., Jama, B. and Othieno, C. (1999) Combined use of
trees, shrubs, and inorganic fertilizers for soil fertility improvement. Agrofor.
Syst. 44: 21-36.
Jama, B., Swinkels, A.R. and Buresh, J.R. (1997) Agronomic and economic
evaluation of organic and inorganic sources of phosphorus in Western Kenya.
Agron. J. 89: 597-604.
Jama, B., Palm, C.A., Buresh, R.J., Niang, A.I., Gachengo, C., Nziguheba, G.,
Amadalo, B., 2000. Tithonia as a green manure for soil fertility improvement
in western Kenya: a Review. Agrofor. syst. 49: 201-221.
Janssen, B.H. (1994) Integrated nutrient management: the use of organic and
mineral fertilizers. In: Van Reuler, H., Prins, W.H. (Eds), The role of plant
nutrients for sustainable food crop production in sub-sahara Africa.
Wageningen, The Netherlands, pp 89-105.
Kretzschmar, R.M., Hafner, H., Bationo, A.. and Marschner, H. (1991) Longand short-term effects of crop residues on aluminium toxicity, phosphorus
availability and growth of pearl millet in a acid sandy soil. Plant Soil. 136:
215-223
Mittal, S.P., Grewal, S.S., Agnihotri, Y. and Sud, A.D. (1992) Substitution of
nitrogen requirement of maize through leaf biomass of Leucaena
leucocephala: Agronomic and economic considerations. Agrofor. Syst.
19: 207-216.
Mokwunye, A.U., Chien, S.H. and Rhodes, E. (1986) Phosphate reactions with
tropical African soils. In: Mokwunye, A.U., Vlek, P.L.G. (Eds.), Management
of Nitrogen and Phosphorus Fertilizers in Sub-Saharan Africa. Martinus Nijhoff
Publishers, Dordrecht, The Netherlands, pp. 253-281.
Combined use of Tithonia diversifolia and Inorganic Fertilizers for Improving Maize
Production in a Phosphorus Deficient soil in Western Kenya
345
Nziguheba, G., Palm, C.A., Buresh, R.J. and Smithson, P.C. (1998) Soil
phosphorus fractions and adsorption as affected by organic and inorganic
sources. Plant Soil 198, 159-168.
Nziguheba, G., Merckx, R., Palm, C.A.. and Rao, M. (2000) Organic residues
affect phosphorus availability and maize yields in a Nitisol of western Kenya.
Biol. Fertil. Soils 32: 328-339.
Olsen, S.R. and Sommers, L.E. (1982) Phosphorus. In. Page: A.L., Miller, R.H.,
Keeney, D.R. (Eds), Methods of Soil Analysis Part 2. ASA and SSSA Madison
,WI, USA, pp 403-430.
Palm, C.A. (1995) Contribution of agroforestry trees to nutrient requirements of
intercropped plants. Agrofor. Syst. 30: 105-124.
Palm, C.A., Myers, R.J.K. and Nandwa, S.M. (1997) Combined use of organic
and inorganic nutrient sources for soil fertility maintenance and
replenishment. In: Buresh, R.J., Sanchez, P.A. and Calhoun, F. (Eds),
Replenishing Soil Fertility in Africa. SSSA Spec. Publ. 51. ASA and SSSA,
Madison, Wi, USA, pp 193-217.
Parkinson, J.A. and Allen, S.E. (1975) A wet oxidation procedure suitable for
the determination of nitrogen and mineral nutrients in biological materials.
Commun. Soil Sci. Plant Anal. 6: 1-11.
Phan Thi Công, (2000) Improving phosphorus availability in selected soils from
the uplands of south Vietnam by residue management. A case study: Tithonia
diversifolia. PhD thesis No 439. Katholieke Universtiteit Leuven, Belgium.
SAS Institute (1995) SAS/Stat User Guide 2 Version 6.1. SAS Inst Cary NC,
USA.
Shepherd, D.K. and Soule, J.M. (1998) Soil fertility management in western
Kenya: dynamic simulation of productivity and sustainability at different
resource endowment levels. Agric. Ecosyst. Environ. 71:131-145.
Smith, J.L., Papendick, R.I., Bezdicek, D.F. and Lynch, J.M. (1993) Soil organic
matter dynamics and crop residue management. In: Blaine, M.F. Jr. (ed.),
Soil microbial ecology: Applications in agricultural and environmental
management. Marcel Decker, New York, pp 65-94.
346
Nziguheba, G. et al
Effect of Combining Organic and Inorganic Phosphorus Sources on Maize Grain
Yield in a humic-Nitosol in Western Kenya
Effect of Combining Organic
and Inorganic Phosphorus
Sources on Maize Grain Yield
in a humic-Nitisol in Western
Kenya
347
24
Ojiem, J. O.*1, Palm, C. A.2,
Okwuosa, E. A.1 and Mudeheri, M.A.1
Kenya Agricultural Research Institute, Regional Research
Centre, P.O. Box 169, Kakamega, Kenya;
E-mail: ahi@swiftkisumu.com
2
Tropical Soil Biology and Fertility Programme,
P.O. Box 30592, Nairobi, Kenya
1
Abstract
The western Kenya soils are typically low in fertility due to
continuous cropping with inadequate fertilizer use. A threeyear experiment was conducted to investigate the effect of
combining organic and inorganic Phosphorus (P) sources
on maize grain yield in a P deficient (2ppm) experimental
site at the Regional Research Centre, Kakamega, western
Kenya. The design was a Randomized Complete Block,
replicated three times. Farmyard manure (FYM) and Triple
Super Phosphate (TSP) were combined at the ratios: 0:100,
25:75, 50:50, 75:25, and 100:0 to attain 30 kg P ha-1 and
Ojiem, J.O. et al
348
applied at planting to all plots except the control plot which
received no P. To prevent Nitrogen (N) and Potassium (K)
deficiencies confounding P responses, N was toped up to
100 kg N ha-1 and K to 120 kg K ha-1 in all plots. Urea N was
top-dressed in equal splits at 3 and 6 weeks after maize
emerged, while K (KCl) was applied at planting. Maize grain
yield was determined at 13% moisture content and plotted
against organic-inorganic P treatments to determine the
response pattern. Non-linear regression analysis was then
performed to estimate the effects of organic and inorganic
P. Grain yield was significantly higher (p=0.05) with P than
without and sole organic P was comparable to sole inorganic
P. Grain yield responses best fitted quadratic functions and
the regression coefficients estimating organic P, inorganic
P and the interaction were significant (p=0.05), indicating
real FYM and TSP effects and synergy between them. The
results demonstrated grain yield benefits of integrating
organic and inorganic P sources.
Introduction
The western Kenya region, supporting between 500 and 1200 inhabitants
per km2 (Hoekstra and Corbett, 1995) is one of the most populated
rural regions of the world. However, the region is endowed with good
agricultural climate and has the potential to produce sufficient food to
meet the demand by the population. Despite the high agricultural
potential, food production is low. Farming is mainly subsistence, with
smallholder farmers growing two crops per year, with little or no fertilizer
inputs. This practice has, over the years, resulted in the depletion of
native soil fertility and a decline in productivity. A survey in western
Kenya (Onim et al., 1986) reported P deficiency in over 90% of the
smallholder farms surveyed.
P inputs in smallholder fields consist primarily of inorganic fertilizers
and organic sources such as biomass, animal manure, compost and
crop residues. However, low quality organic materials such as maize
stover, may not supply sufficient amounts of plant-available P (Palm et
al., 1997). While inorganic fertilizers can restore the fertility of soils and
improve crop yields, their use in the East African highlands is limited
(Hoekstra and Corbett, 1995) and alternative strategies for supplying P
to the P-deficient smallholder systems are necessary. Studies in western
Kenya indicate that the incorporation of higher quality organic manures,
like Tithonia diversifolia and Lantana camara, along with TSP, increases
Effect of Combining Organic and Inorganic Phosphorus Sources on Maize Grain
Yield in a humic-Nitosol in Western Kenya
349
the effectiveness of fertilizer phosphorus (Gachengo, 1996; Nziguheba
et al., 1998). Such integration of organic and inorganic resources would
have agronomic advantage, if the organic material enhances the
availability of added P (Palm et al., 1997).
The processes responsible for better response from the integration
of organic and inorganic P sources are not yet clearly established, mainly
because of the complex nature of P dynamics in the soil. However, there
are suggestions that the interactions resulting from this integration
reduces P-sorption capacity of the soil (Palm et al., 1997), thereby
increasing P availability to plants. Other benefits include immobilization
of excess nutrients that would otherwise be lost through leaching and
positive physical effects associated with improved soil structure. Addition
of organic residues also enhances microbial pool sizes activity (Smith et
al., 1993). These chemical and biological processes influence both
availability and utilization of nutrients.
The objectives of this experiment were:
1) To investigate the effect of combining organic and inorganic P sources
on maize grain yield
2) To determine the optimum ratio for combining organic and inorganic
P, and
3) To assess the synergy resulting from integrated use of organic and
inorganic P sources.
Materials and Methods
The experiment was conducted at Kakamega, western Kenya, during
the long rain seasons (March to August) of 1997, 1998 and 1999. The
experimental site was within the Kenya Agricultural Research Institute’s
Regional Research Centre, located 00° 16' N (latitude) and 34° 45' E
(longitude). The altitude is 1585 m above sea level, mean annual
temperature is 18-20°C, the average annual rainfall is 2012 mm and
the soil is classified as dystro-mollic Nitisol (Jaetzold and Schmidt, 1983).
Characterization previously conducted at the experimental site (FURP,
1987) indicated the soil is well drained, extremely deep, with a thick
humic top layer. The top soil reaction is in the strong to moderately acid
range (pH 4.5) and exchangeable Al is low, while organic matter is high
in the top soil (2.4%).
The experimental design was a Randomized Complete Block with
three replicates. Plots were 4.5 by 5.0 m (22.5 m2), in which 6 rows of
maize were planted with a row spacing of 75 cm. Five P treatments were
evaluated alongside a control that received no P. In each treated plot, P
350
Ojiem, J.O. et al
was applied at the rate of 30 kg P ha-1 by combining FYM and TSP in
different ratios as follows:
1. 30 kg P ha-1 (100% inorganic P)
2. 30 kg P ha-1 (100% organic P)
3. 15 kg P ha-1 (50% inorganic P) + 15 kg P ha-1 (50% organic P)
4. 7.5 kg P ha-1 (25% inorganic P) + 22.5 kg P ha-1 (75% organic P)
5. 22.5 kg P ha-1 (75% inorganic P) + 7.5 kg P ha-1 (25% organic P)
6. Control (no P)
A sample of the FYM was analyzed each planting season to determine
the P content, which was then used to compute the amount of the
material to add to the respective plots. Since FYM also supplied up to
73 kg N ha-1 and 120 kg K ha-1 to the organic P treated plots, urea and
muriate of potash (KCl) were applied to all plots, including the control,
to balance N and K at 100 kg N ha-1 and 120 kg K ha-1, respectively.
Apart from correcting the N and K imbalances, the high rates were
intended to prevent N and K deficiencies, which could confound P
responses. Urea was applied as top-dress in two equal splits. The first
split was applied at 21 days after emergence and the second one 42
days after emergence. KCl was broadcast applied at planting. Soil was
sampled in all plots (0-15cm depth) for P determination prior to
application of treatments. The plots were maintained throughout the
trial duration.
Maize grain yield was determined at harvest. Yield was adjusted to
13% moisture content and grain yield data plotted against treatments
to determine the P response pattern. To fit the curves, treatments were
arranged in order of increasing proportion of inorganic P (decreasing
proportion of organic P). Both linear and quadratic functions were fitted
in turn, to determine the best-fit model, based on the correlation
coefficient (R2) value. The model with the highest R2 value was selected
as the best function describing grain yield response pattern. Grain yield
data was then subjected to the Analysis of Variance (ANOVA). A nonlinear regression model: (Yield=a+bX1+gX2+dX1*X2) was fitted to the data
to test the effects of inorganic P, organic P and the interaction between
them, on maize grain yield. The regression coefficients in the model are
described below:
a = intercept
b = coefficient estimating inorganic P (X1) effect
g = coefficient estimating organic P (X2) effect
d = coefficient estimating inorganic P and organic P interaction (X1*X2)
effect.
The effects of organic P, inorganic P, and the interaction between
them, were evaluated by testing the null hypothesis (HO) that the
coefficient estimates were equal to zero. This was done by t-tests.
Effect of Combining Organic and Inorganic Phosphorus Sources on Maize Grain
Yield in a humic-Nitosol in Western Kenya
351
Results and Discussion
Maize grain yield response to P
Significant (p=0.05) grain yield responses to both organic and inorganic
P were observed in 1997, 1998 and 1999 (Table 24.1). The highest
response was recorded in 1997 when application of P increased grain
yield by 3.2 t ha-1 compared with 2.8 and 2.3 t ha-1 in 1998 and 1999,
respectively. Generally, yield following sole addition of FYM was
equivalent to those from sole inorganic P. The FYM used was of relatively
high quality (Table 24.2). The difference in grain yield between sole
inorganic P and organic P was highest in 1997 (0.8 t ha-1) but the
difference was not significant. In 1999, grain yield from sole organic P
was slightly higher than that from sole inorganic P. Grain yield responses
were more variable in 1999 due to poor rainfall distribution that year
(Figure 24.1). Much of the rainfall in 1999 long rain season was received
during the February to April period, the beginning of the season. Rainfall
sharply declined between April and June before picking up in July,
which was towards the end of the growing season.
Table 24.1: Effect of organic and inorganic P on maize grain yield at Kakamega Research
Station, western Kenya, in 1997, 1998 and 1999
Maize grain yield (kg ha-1)
Treatment combination
% Inorganic P
% Organic P
1997
1998
1999
Mean
100
75
50
25
0
0
0
25
50
75
100
0
6.22
6.25
6.67
5.75
5.39
3.45
5.38
3.83
4.34
5.13
3.12
2.57
2.84
2.92
4.38
4.35
2.97
2.05
4.81
4.33
5.13
5.07
3.82
2.69
1.50
18.2
0.95
15.6
1.2
24.8
1.30
20.2
LSD (0.05)
CV
Table 24.2: Nutrient and lignin content of the FYM used in the trial (%) in 1997, 1998 and 1999
Year
N
P
K
Lignin
1997
1998
1999
Mean
1.31
1.45
1.25
1.33
0.35
0.37
0.39
0.37
2.14
2.16
2.11
2.13
10.95
10.05
11.05
10.68
Ojiem, J.O. et al
352
Combination of fertilizer P with FYM in different proportions did not
result in significant grain yield differences in 1997 (Table 24.1). However,
in 1998, the treatment combination receiving 25% inorganic P and 75%
organic P (25:75) performed similar to the 50:50 treatment but
significantly better than the 75:25. In 1999, the 50:50 combination
performed significantly better than the 75:25.
The grain yield results were particularly significant in 1999 given
that it was the driest of the three trial years. Grain yield sharply declined
compared to the two previous years. However, significant and synergistic
effects of combining FYM and P fertilizer were demonstrated by the highest
yields obtained with the 50:0 and 25:75 combinations (Figure 24.2). The
superior performance may have been due to added benefits of the organic
material (FYM). Besides the direct benefits of nutrient supply, organic
materials have effect on soil physical properties that in turn influence
nutrient acquisition and plant growth (Palm et al., 1997). Principal among
these, is the soil moisture holding capacity. By influencing moisture
storage and promoting root growth, FYM may have greatly improved the
efficiency with which the available P was used during the drier 1999
season. Based on analysis of combined data over the three years, no
significant differences were detected between the different organic and
inorganic P combinations but the 50:50 treatment combination was
significantly better than sole organic P treatment (0:100).
The results indicated that the optimal organic to inorganic P combination
is close to the 50:50 ratio. Combining fertilizer P with FYM at this ratio is
likely to be beneficial to smallhoder farmers, who have limited access to
inorganic fertilizers and are typically unable to generate sufficient quantities
of high quality organic materials for fertility improvement. Suggestions for
integration of fertilizer P with available organic resources have been made
in the past by Janssen (1993) and Palm et al. (1997).
Figure 24.1: Rainfall distribution at Kakamega Research Station in 1997 , 1998 and
1999 growing seasons
350
Rainfall (mm)
1997
300
1998
250
1999
200
150
100
50
0
Jan.
Feb.
March April
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Effect of Combining Organic and Inorganic Phosphorus Sources on Maize Grain
Yield in a humic-Nitosol in Western Kenya
353
Figure 24.2: Performance of organic and inorganic P combinations across three growing
seasons at Kakamega Research Center, western Kenya
8
Maize grain yield (t ha-1)
7
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
6
5
4
3
2
1
0
123 100:0
123
123 75:25
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
123
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
12 50:50
12 25:75
12
0:100
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1998
1997
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1999
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
1234
Mean
Assessment of organic and inorganic P effects
The grain yield responses were best described by quadratic functions
(Figures 24.3 - 24.6). The R2 values for the functions were greater than
0.6 and were much higher than the values obtained when linear
functions were fitted to the data. This indicated that the effects were
largely quadratic in nature. The apparent non-linearity (curvature) in
grain yield responses indicated some degree of interaction or synergistic
effects between organic and inorganic P.
Figure 24.3: Mean effect of organic and inorganic P on maize grain yield (1997-1999)
5
Maize grain yield (t ha-1)
4.5
4
3.5
2
y = -0.1479x + 0.6201x + 3.884
2
yR
= 2-0.1479x
= 0.802 +0.6201x + 3.8
3
R 2 = 0.802
2.5
2
1.5
1
0.5
0
Organic P
Inorganic P
0
25
50
75
100
100
75
50
25
0
Percentage of organic and inorganic P
Ojiem, J.O. et al
354
Figure 24.4: Effect of organic and inorganic P on maize grain yield in 1988
6
Maize grain yield (t/ha)
5
4
2
= -0.1171x2 ++0.2909x
+ 4.966
y = y-0.1171x
0.2909x
+ 4.9
2
R = 0.604
R 2 = 0.604
3
2
1
0
Organic P
0
25
50
75
100
Inorganic P
100
75
50
25
0
Percentage of organic and inorganic P
Figure 24.5: Effect of organic and inorganic P on maize grain yield in 1999
Maize grain yield (t/ha)
5
4.5
4
3.5
2
y = -0.315x + 2.059x + 0.78
3
2
R2 = 0.6569
y = -0.315x
+ 2.059x + 0.7
R 2 = 0.6569
2.5
2
1.5
1
0.5
0
Organic P
0
Inorganic P 100
25
50
75
100
75
50
25
0
Percentage of organic and inorganic P
Figure 24.6: Mean effect of organic and inorganic P on maize grain yield
Maize grain yield (t/ha)
6
5
2
y = -0.1929x + 1.0031x + 3.804
4
y = -0.1929xR2 2+= 0.681
1.0031x + 3.8
R 2 = 0.681
3
2
1
0
Organic P
Inorganic
0
25
50
75
100
100
75
50
25
0
Percentage of organic and inorganic P
Effect of Combining Organic and Inorganic Phosphorus Sources on Maize Grain
Yield in a humic-Nitosol in Western Kenya
355
The curvature and the organic and inorganic P effects on maize
grain yield were further tested by fitting a non-linear regression model:
(Yield=a+bX1+gX2+dX1*X2) to the data. The regression coefficients
estimating the effects of organic and inorganic P and the interaction
between them are shown in Table 24.3. In 1997, both organic and
inorganic P had significant effects on maize grain yield. These results
were in agreement with those of the Analysis of Variance (Table 24.1).
However, the inorganic P coefficient was larger than organic P, indicating
that inorganic P had greater effect on grain yield. The coefficient for the
interaction between organic and inorganic P was the lowest in value
and the t-test was not significant. This suggested that there was no real
Table 24.3: Regression coefficients estimating the effects of organic P, inorganic P, and
the interaction, on maize grain yield
Parameter
a
b
g
d
Estimate
3.444
0.02742
0.01874
0.000241
1997
S.E
0.586
0.00804
0.00804
0.000250
t value (for HO)
5.878***
3.410**
2.330*
0.96 NS
Parameter
a
b
g
d
Estimate
2.565
0.02353
0.01137
0.00061
1998
S.E
0.497
0.00682
0.00682
0.000212
t value (for HO)
5.160***
3.450**
1.667 NS
0.47 NS
Parameter
a
b
g
d
Estimate
2.048
0.00475
0.01171
0.000499
1999
S.E
0.454
0.00623
0.00623
0.000194
t value (for HO)
4.511***
0.762 NS
1.879 NS
2.572*
Parameter
a
b
g
d
Estimate
2.253
0.01412
0.00997
0.000210
Mean
S.E
0.236
0.00324
0.00324
0.000101
t value (for HO)
9.546***
4.358***
3.077**
2.079 *
*** = significant (p <0.001)
** = significant (p <0.01)
* = significant (p <0.05)
NS=Not significant (p = 0.05)
S.E = Standard Error
a= intercept
b= coefficient estimating inorganic P (X1) effect.
g= coefficient estimating organic P (X2) effect.
d= coefficient estimating inorganic P and organic P
interaction (X1* X2) effect.
356
Ojiem, J.O. et al
interaction between organic and inorganic P and the effects were probably
largely additive. Similar to 1997, inorganic P had a significant effect on
maize grain yield. However, the organic P and the interaction coefficients
were not significant, suggesting no real effect on grain yield. In contrast
to 1997 and 1998, the effects of organic and inorganic P on grain yield
were small and insignificant in 1999 but the interaction coefficient was
relatively large and significant. These results confirmed the conclusions
drawn from the analysis of variance that addition of fertilizer P to FYM
had synergistic effect on maize grain yield that year.
Based on the results of regression analysis performed on grain yield
data averaged over three years (1997 to1999), the coefficients estimating
the effects of organic P, inorganic P and the interaction were significant
(Table 24.3). However, the coefficient of inorganic P was much larger
than that of organic P, indicating greater effect of inorganic P compared
to organic P. The significant interaction coefficient demonstrated synergy
between FYM and fertilizer P.
Conclusions
The results of this study indicate that applying P at a modest rate of 30
kg P ha-1, either in the organic or inorganic form, can substantially
increase maize grain yield. Provided the FYM rate supplies equivalent
amount of P, FYM appears to be nearly as effective as TSP. However,
since the quality of FYM determines the quantities to be applied to attain
the required P rate, low quality material could mean more labour for
application, probably making the practice economically unattractive.
Combining organic and inorganic P results in synergistic effects,
particularly in drier, moisture-stressed growing seasons. This synergy,
and other extra benefits of FYM, should be exploited by smallholder
resource poor farmers. A 50/50 organic-inorganic combination ratio
appears to be the optimum
Acknowledgements
The Authors would like to thank the European Union (EU), the Tropical
Soil Biology and Fertility Programme (TSBF) and the Kenya Agricultural
Research Institute (KARI) for providing the funds and other resources
that made this study possible. All collaborators who contributed to the
design and implementation of this study are also greatly acknowledged.
We particularly thank Janet Riley of Rothamstead Experimental Station,
U.K. for ideas on statistical analysis of similar experimental designs.
Effect of Combining Organic and Inorganic Phosphorus Sources on Maize Grain
Yield in a humic-Nitosol in Western Kenya
357
References
FURP. (1987) Fertilizer Use Recommendation Project (FURP). Final Report Annex
III. Phase I. Ministry of Agriculture, Nairobi, Kenya/National Agricultural
Research Laboratories, Nairobi, Kenya.
Gachengo, C. (1996) Phosphorus release and availability from organic materials
amended to phosphorus deficient soils of Western Kenya. M.Sc. Thesis,
Moi University, Eldoret, Kenya.
Hoekstra, D. and Corbett, J.D. (1995) Sustainable agricultural growth for the
highlands of East and Central Africa: Prospects to 2020. Int. Food Policy
Res. Inst. Washington DC.
Jaetzold, R. and Schmidt, H. (1983) Farm management handbook of Kenya.
Vol.iii. west Kenya. Ministry of Agriculture, Nairobi, Kenya.
Janssen, B.H. (1993) Integrated nutrient management: The use of organic and
mineral fertilizers. P. 89-105. In: Van Reuler, H. and Pins, W.H.(eds) The
role of plant nutrients and sustainable food production in Sub-Saharan Africa.
Vereniging van Kuunstmest Producenten, Leidschendam, the Netherlands.
Nziguheba, G., Palm,C.A., Buresh, R.J. and Smithson, P.C. (1998) Soil
phosphorus fractions and adsorption as affected by organic and inorganic
sources. Plant Soil 198: 1365-1373.
Onim, J.F.M., Mathuva, M., Hart, R., Fitzhugh, H. A. and Otieno. K. (1986)
Recommendation domains for dual purpose goat research in Nyanza and
Western provinces of Kenya. Proceedings of the 5th Small ruminant CRSP
Kenya Workshop, Kabate, Kenya. Nov. 4-6, 1986. P 40-47.
Palm, C. A., Myers, R.J.K. and Nandwa, S.M. (1997) Combined use of organic
and inorganic nutrient sources for soil fertility maintenance and
replenishment. P. 193-217. In R.J. Buresh et al. (eds.) Replenishing Soil
Fertility in Africa. SSSA Spec. Publ. 51. SSSA, Madison, WI.
Smith, J.L., R.I. Pependick, D.F. Bezdicek, and J.M. Lynch. 1993. Soil organic
matter dynamics and crop residue management. p. 65 -94. In M.F. Blaine
Jr. (ed) Soil microbial ecology: applications in agricultural and environmental
management. Marcel Decker, New York.
358
Ojiem, J.O. et al
Use of Organic and Inorganic Resources to Increase Maize Yields in some Kenyan
Infertile Soils: A Five-Year Experience
Use of Organic and Inorganic
Resources to Increase Maize
Yields in some Kenyan
Infertile Soils: A Five-Year
Experience
359
25
Okalebo, J.R.1, Palm, C.A.2, Lekasi,
J.K.3, Nandwa, S.M.4, Othieno, C.O.1,
Waigwa, M.1 and Ndungu, K.W.1
1
Department of Soil Science, Moi University, Chepkoilel
Campus, P.O. Box 1125, Eldoret, Kenya,
2
Tropical Soil Biology and Fertility Programme (TSBF),
P.O. Box 30592, Nairobi, Kenya,
3
National Agricultural Research Centre, Muguga Kenya
Agricultural Research Institute (KARI), P.O. Box 30148,
Nairobi, Kenya,
4
National Agricultural Research Laboratories, KARI,
P.O. Box 14733, Nairobi, Kenya.
Abstract
Use of organic and inorganic resources to increase and
sustain agricultural productivity of soils has been practiced
worldwide over a long period. Positive effects from these
materials are known to be the enhanced nutrient inputs to
soils and improved soil physical and biological properties.
Effects of separate or individual and combined applications
360
Okalebo, J.R. et al
of organic and inorganic materials to soils have been studied
rather extensively and the results are complex. But, it
appears for certain, that the quality and quantity attributes
are the driving forces towards basic processes in soils such
as nutrient mineralisation and release and the overall
effectiveness of added materials on crop yields. However,
many studies have considered mainly the immediate or one
seasonal effects of organics and inorganics on crop yield.
Therefore the monitoring of soil process studies in relation
to crop growth and yield, as well as considerations on
economic benefits arising from the use of these external
resources seem to have been slighted. In this paper we
present results of field studies at four on-farm, researcher
managed sites that vary widely with climate and soils
(acrisols, ferralsols and luvisols).
In 1994 first rains, the effectiveness of crop residues
(maize stover, groundnut trash and acacia mearnzii
prunings) on-farm manures and Minjingu phosphate rock
(PR) was tested on maize yields at Ndeiya, Gatuanyaga and
Malava sites. The organics above were incorporated into
soils individually or in combinations giving a target or
economical rate of 60 kg Nha-1, while the PR was added at a
uniform rate of 40 kg P ha-1 in various combinations of the
organics. Maize yields in that season ranged from 1256 3761 Kgha-1 (at Ndeiya and Malava sites only with adequate
rainfall). Although maize yield increases did not attain
statistical significance, the high N organics (poultry manure,
Acacia mearnzii and groundnut trash), with PR
combinations, gave overall high yield increases. The study
period was too short to monitor residual effects of treatments
and the solubilisation of PR. But in the Chepkoilel Campus
ferralsol (Moi University), maize has been cropped over four
consecutive years (1997 - 2000) in plots receiving annual
maize stover, wheat straw and initial superphoshate
application of 100 kg Pha -1 plus combined urea
combinations from 20-100 kg Nha -1. Maize yields over the
entire study period have ranged from 751-6836 kg ha-1 with
significant variations occurring from rainfall variations.
Nevertheless, again the combined applications of organic
and inorganic resources favoured maize production. There
are favourable effects of the materials used to improve the
soil fertility status of soils. The results suggest an economic
potential, to the smallhold farmers, arising from combined
use of organic and inorganic resources.
Use of Organic and Inorganic Resources to Increase Maize Yields in some Kenyan
Infertile Soils: A Five-Year Experience
361
Introduction
Low input agriculture mainly explains the cause of low and declining
crop yields in many countries south of the Sahara (Makken, 1993;
Simpson et al., 1996; World Bank, 1996). But specifically the prices of
imported (inorganic) fertilizers, without subsidies, are unfavourable to
most smallhold farmers in the region. Thus in Kenya, where nitrogen
(N) and phosphorus (P) nutrients are widely deficient in cropped soils
(Okalebo et al., 1992; FURP, 1994; TSBF, 1994), attempts have been
and continue to be made to find affordable technologies to correct
nutrient deficiencies in soils. These include agroforestry practices
especially the use of improved fallows (Jama et al., 1997; Sanchez et al.,
1997), combinations of organic and inorganic resources (Palm et al.,
1997; Okalebo et al.,1999) and use of direct and combined applications
of the reactive Minjingu (Tanzania) phosphate rock (PR) with organic
materials (Okalebo et al., 1991/1995; Okalebo and Woomer, 1994;
Okalebo and Nandwa, 1997; Jama et al., 1997; Woomer et al., 1997).
These PR and organics combinations have focussed on the provision of
cheap N inputs from organics and the solubilisation of PR through
formation of favourable acid environments that result when organics
(in contact with PR) decompose in soils (Nahas et al, 1996; Mutuo et al.,
1999; Nyambati, 2000; Waigwa et al., 2000). In this paper we present
results of field studies whereby organic and inorganics were combined
and tested on maize yields across the Kenyan infertile soils; mainly the
acrisols (ultisols), ferralsols (oxisols) and luvisols (alfisols). The trial sites
also varied with climatological characteristics, mainly the rainfall and
its distribution.
Materials and Methods
Two types of field experiments were conducted:
Experiment 1
A randomised complete block design (RCBD) experiment with 4 replicates
per site, was set up on smallholder farms at Ndeiya (Lat 10 14'S, Long
360 28'E), Kiambu district; Gatuanyaga (Lat 10, 22'S, 340 45'E), Thika
district; and Malava-Kabras (Lat 00 18'N, Long 34° 45'E), Kakamega
district. Both Ndeiya and Gatuanyaga receive annual rainfall of about
800 mm distributed within 2 seasons, March to May and October to
December. Malava site is within the highlands of western Kenya, with a
rainfall of 1000-1800 mm falling from March to September (Jaetzald
and Schmidt, 1983). Some properties of surface (0-20cm) soils for the
sites including Chepkoilel (for Experiment 2), taken before treatment
applications are given in Table 25.1.
362
Okalebo, J.R. et al
Table 25.1: Some properties of surface (0-20cm) soils from on-farm field trials in Kiambu,
Thika, Kakamega and Uasin Gishu districts, Kenya
Properties
Kiambu
Ndeiya
Thika
Gatuanyaga
Kakamega
Malava
1994
Uasin Gishu
Chepkoilel
1997
pH (H2O, 1:2.5)
Total Carbon (%)
Total Nitrogen (%)
Olsen P (mg/kg)
5.31
2.59
0.223
11
4.84
1.48
0.138
10
4.10
2.15
0.214
7
4.85
1.30
0.110
4
Soil Order
Luvisol
Acrisol
Acrisol
Ferralsol
Soil properties for the sites indicate low pH levels (<5.5) favourable
for PR dissolution (Sanchez et al., 1997 and also low soil test P values
close to or below 10mg P kg-1, the level below which P responses are
expected (Okalebo et al., 1993).
In this on-farm, researcher-managed experiment, the main objective
was to identify the readily available organic materials (crop residues,
tree prunings, manures) at farm level that would increase the N and P
levels in soils when incorporated with PR. Target N and P rates from
combinations of these materials were 60 Kg N ha -1 and 40 Kg P ha -1
(applied as PR). These 2 rates appear to be economical for annual high
and sustained maize production for the test sites (Okalebo, 1987).
Table 25.2: Treatment combination used in Experiment 1 in 1994 at Ndeiya, Gatuanyaga
and Malava sites
Treatment
Control
AM + MS
MS + PM
PM
UREA
PR
UREA + PR
MS +PR
AM + PR
FYM + PM
Nitrogen
Combination
(kg N ha-1)
Total (desired) N
(kg N ha-1)
Desired P
(kg P ha-1 PR)
0
30N, AM + 30N, MS
30N, MS + 30N, PM
60N, PM
60N, UREA
60N, UREA
60N, MS
60N, AM
30N, FYM + 30N, PM
0
60
60
60
60
60
60
60
60
0
40
40
40
40
-
Notes:
AM = Acacia mearnzii (wattle prunings) used in Ndeiya and Gatuanyaga but groundnut
trash used at Malava only.
MS = maize stover
PM = Poultry manure (broiler, from NARC Muguga/KARI)
FYM = on-farm yard manure or 'boma' manure; but compost was used at Gatuanyaga
site only.
PR = Minjingu phosphate rock (0-30-0).
Use of Organic and Inorganic Resources to Increase Maize Yields in some Kenyan
Infertile Soils: A Five-Year Experience
363
It is noted that the quantities of dry organic materials were calculated
to give the above N combinations (Table 25.2) using their total N contents
from prior laboratory analysis at NARC Muguga (Table 25.3). It was
also assumed that the materials (with low P contents) had negligible P
inputs. Overall, the experiment sought information from a wide range
of treatment combinations.
At all three sites, all materials (above) were broadcast to fine seedbeds
and incorporated into soil using a hoe. Maize (H511) was planted as a
test crop with a plant density to 4.4 plants m-2. Recommended husbandry
practices (weeding, stalkborer control) were followed. A parallel N and P
(inorganics) response trial was conducted at Gatuanyaga site only) with
adequate on-farm land for experimentation.
Experiment 2
A randomised complete block design (RCBD) trial was set with 4 replicates
at Chepkoilel Campus, Moi University, Eldoret, Uasin Gishu district (Lat
00 35'N, Long 350 18'E). This site receives annual rainfall of 1124 mm in
one season from March to September (Jaetzold and Schmidt, 1983). At
this site, treatments included incorporating N fertilizer with wheat straw
and soybean. Treatments consisted of chopped (15 cm length) crop
residues (above) applications at a uniform rate of 2 t ha-1 for each organic
material. Urea (fertilizer N) was combined with these two organics at the
rates of 0, 20, 40, 80 and 100 kg N ha-1 each. These treatments were
applied at maize (H 614D hybrid) planting, first in March 1997, and were
repeated in the subsequent years upto March 2000. To eliminate P and K
limitations, 100 kgP ha-1 singlesuperphosphate and 100 kg Kha-1 muriate
of potash were applied only at the start of the experiment.
Table 25.3: Some characteristics of organics applied in 1994 field experiments
(Expt. 1): The nitrogen data were used to calculate N inputs
Material
Maize Stover-Muguga
Maize Stover-Malava
Acacia mearnzii-Muguga
Groundnut trash - Malava
Poultry Manure - Muguga
FYM - Malava
Compost - Ndeiya
ash
N
% dry matter
P
K
Abbreviation
9.5
5.3
5.0
6.1
17.8
53.5
62.9
0.79
0.76
2.21
1.56
3.12
1.30
1.24
0.085
0.084
0.092
0.072
1.733
0.227
0.193
1.94
0.56
0.85
0.43
1.91
1.14
1.09
MS
MS
AM
AM
PM
FYM
FYM
Source: NARC/KARI Muguga, Soil Chemistry Station, 1994
Notes:
a) Apart from Acacia mearnzii (AM) and poultry manure (PM) from Muguga applied to
all 3 sites in 1994 Expt. 1 all other organics originated at on-farm level per site.
b) All materials vary in characteristics.
364
Okalebo, J.R. et al
All inputs for both experiment 1 and 2 were incorporated into the
seedbed by hand tillage and starter N for all nitrogen treatments was
applied at 20 KgNha -1 at planting. At harvest, cobs and stover were
separated. Sub-samples of these components were dried (40-50oC) until
constant weight, to obtain yield measurements.
Soil Sampling and Analysis
In all sites, surface (0-20cm) soils were sampled at random across all
experimental fields before treatment applications. Samples (30 auger
borings for each site) were bulked and mixed thoroughly and composite
soils taken. After maize harvest at Chepkoilel, soils were sampled from
each plot (9 borings per plot at random) and analysed to take
measurements on changes in soil properties due to treatments. Soils
were processed and analyzed for pH (H2O), total carbon and extractable
phosphorus (Olsen), following the procedures outlined in Okalebo et al.
(1993).
Results and Discussion
Maize Yields
In 1994, maize grain was harvested only at the Ndeiya and Malava sites
(Experiment 1), while due to severe drought at Gatuanyaga, only the
biomass (above ground parts) data was obtained as given in Table 25.4.
In that year grain yields ranged from 1256 to 3761 Kg ha-1 at the 2 sites
where maize grew to maturity. On the average, the organics and
inorganics applied individually or in combinations, tended to increase
maize yields above the treatment with highest yield increases being
obtained at the lowest soil test P site of Malava in Expt. 1 (Tables 25.1
and 25.4). At this site, the groundnut trash plus phosphate rock (AM +
PR) treatment gave the highest grain yield, increase of 102% above
control, while the poultry manure (PM) treatment resulted in an increase
of 48%, followed by a PR yield increase of 43%.
Further, in this site, significant yield increases have been reported
(Okalebo, 1987). In an earlier field study at this site (Okalebo and Lekasi,
1993), in which maize stover, groundnut trash and on-farmyard manure
were applied at 0, 2, 4 and 6 t ha-1 each, in combination with 40 KgPha-1
PR for each organic resource rate, the groundnut trash significantly
outyielded the maize stover organic material in terms of maize grain
yield increases. At Ndeiya semi-arid site, the maize stover (MS) plus
poultry manure (MS+PM) treatment gave the highest grain yield increase
of 73% above control, followed by the PM treatment alone, with a yield
increase of 42%.
Use of Organic and Inorganic Resources to Increase Maize Yields in some Kenyan
Infertile Soils: A Five-Year Experience
365
Table 25.4: Effect of crop/tree residues, manure and compost in combinations with
phosphate rock on maize yield (kg ha -1) in 1994
Treatment
Control
AM + MS
MS + PM
PM
UREA
PR
UREA + PR
MS + PR
AM + PR
FYM + PM
Mean
LSD (p = 0.05)
Ndeiya (grain)
Gatuanyaga (biomass)
Malava (grain)
2172
2536
3761
3091
2550
2580
2425
2590
2549
2646
(2690)
NS
2924
3804
4034
4030
4567
3163
4145
4241
3037
2789
(3673)
NS
1384
1256
1750
2048
1543
1974
1567
1708
2802
1570
(1760)
NS
AM = Acacia mearnzii (wattle prunings) used in Ndeiya and Gatuanyaga, but groundnut
trash used at Malava only.
MS = Maize stover
PM = Poultry manure (broiler, from NARC, Muguga, KARI)
FYM = On-farm yard manure (boma manure), but compost used at Gatuanyaga.
In a study in semi-arid eastern Kenya (Probert et al., 1992), addition
of PM at 5 t ha-1 (NARC Muguga broiler quality) gave the highest maize
yield increases compared to the on-farm yard manure applied at 10
t ha-1 and combined inorganic fertilizers: 60 kgNha-1 calcium ammonium
nitrate plus 40 kg P ha -1 triplesuperphosphate. Thus on infertile soils,
PM appears to boost the nutrient and organic matter levels and also to
improve soil physical properties, such as infiltration and soil moisture
retention. The highest maize biomass yield at Gatuanyaga site was
obtained from the urea alone treatment (at 60 kg N ha-1 ) where a yield
increase of 56% was obtained, implying a nitrogen limitation in this low
C and N site (Table 25.1). This observation is supported from the data
in a parallel experiment (Expt. 1) at this site whereby a significant N
response was found from urea applications at 0, 30 and 60 kg N ha -1
rates (Figure 25.1). In summary effect of treatments varied with site
and Figure 25.2 shows an overall picture/overview of the performance
of treatments across the 2 sites (Ndeiya and Malava) harvested for grain
in 1994 (Expt. 1). The rather overall outstanding performance from MS
+ PM, AM + PR and PM treatments is noted. Despite these guidelines or
trends, Expt. 1 in this research has illustrated the complexity in studying
and adopting the use of organic resources for soil fertility restoration/
replenishment, particularly in the TSBF AfNet region, where factors such
as the availability of the resource, its quality and quantity and methods
of application, play significant roles.
366
Okalebo, J.R. et al
Figure 25.1: Maize biomass (kg ha-1) as affected by nitrogen and phosphorus fertilizers
at Gatuanyaga (on farm), Thika, long rains
N source was urea
P source was TSP
kg P ha-1
Figure 25.2: An overall effect of crop residues, tree prunnings, manure with Phosphate
Rock combinations on maize grain yields (kg ha -1) at 2 sites in 1994
3000
2500
2000
1500
1000
500
0
Treatments
NB:
Nitrogen as applied at 60 kg/ha for each material and urea and phosphurus was added
at 40kg P/ha.
AM – Acacia mearnzii
MS – Maize stover
PM – Poultry manure
FYM – Farmyard manure
PR
– Phosphate Rock (Minjingu)
Use of Organic and Inorganic Resources to Increase Maize Yields in some Kenyan
Infertile Soils: A Five-Year Experience
367
In the Chepkoilel Campus experiment (Expt. 2), where the low quality
wheat straw (0.67%N) and higher quality soybean trash (1.07% N) were
compared from 1997 to 2000, maize grain yields varied with the
treatments and years of cropping (Okalebo et al., 1999). In 1997 grain
yields ranged from 875 to 1876 kg ha-1; from 2832 to 6836 kg ha-1 in
1998; and from 1363 to 2272 kg ha-1 in 2000 (Table 25.5).
Table 25.5. Effect of continued addition of crop residues and nitrogen fertilizers on maize
grain yield (kg ha -1) in Chepkoilel soils (Ferralsols)
Treatment
Control
80 N
WS + 0N
WS + 20N
WS + 40N
WS + 80N
WS + 100N
SYT + 0N
SYT + 20N
SYT + 40N
SYT + 80N
SYT + 100N
Overall LSD (p=0.05)
Year
1997
1998
2000
875a
1016a
960a
1321a
1304a
1666b
1677b
751a
1465b
1444b
1500b
1876b
2832a
4883b
2051a
2930a
3223a
4785b
5469b
2832a
2500a
3711a
5567b
6836b
1363
1704
1477
1363
1363
1704
1591
1818
2159
2272
2272
1931
555
1030
926 NS
Means followed by the same letters or none in a column are not significantly (p = 0.05)
different (using Duncan's multiple range test).
WS = Wheat Straw applied annually at 2 t ha -1
SYT = Soybean trash applied annually at 2 t ha-1
N
= Nitrogen applied as urea at 0, 20, 40, 80 and 100 kg N ha-1
Grain yield variations are partly explained in terms of low and
poor rainfall distribution (considering the 10-day periods), particularly
at the maize maturity months of August and September. This is
illustrated from yields obtained in 1997, 1998 and 2000 when total
rainfall received in these two months was 232, 380 and 283 mm for
the 3 respective years (Chepkoilel Campus Meteorological Records).
This magnitude of rainfall in 1997 and 2000 did not probably favor
adequate soil moisture and nutrient availability, contributing to overall
low maize yield in those 2 years. Nevertheless, many treatments gave
significant increases in maize yield (P<0.05), particularly from 2 tha -1
wheat straw and soybean trash combined with fertilizer N (urea) above
80 kgN ha-1 rate of incorporation (Table 25.5). Favourable rainfall and
368
Okalebo, J.R. et al
its distribution in 1998 most likely contributed to larger maize yields
in that year. Again higher yields were found from the higher quality
soybean trash and N fertilizer applied above 80 kgNha -1 . Past work in
field nutrient limiting study has pinpointed a nitrogen limitation in
Chepkoilel soils (Mwaura, 1998).
Changes in soil properties
Data for soil pH, C and available P parameters obtained in surface
(0-20 cm) soils sampled before application of treatments in March 1997
and soils taken soon after harvesting the fourth maize crop in November,
2000 are summarised in Table 25.6. There were positive changes from
organic matter inputs (and possibly continued urea addition) to increase
the levels of these three parameters in soils.
Table 25.6: Changes in soil (0-20cm) properties as influenced by cumulative incorporation
of crop residues and urea at Chepkoilel, Kenya, during 4 years of maize cropping
Treatment
After maize harvest, November, 2000
Control (No inputs)
80 kgNha-1 as urea
Wheat straw + N*
Soybean trash + N*
Initial contributions, March 1997
Soil parameter
pH
Local C (%)
Available P
(mg/kg)
4.80
5.06
5.32
5.41
4.85
2.16
2.26
2.47
2.78
1.30
11.6
10.4
14.4
14.3
3.9
*Wheat straw and soybean trash data include all treatments receiving urea at 0, 20, 40,
80 and 100 kg N ha -1
A similar trend was found from soils sampled soon after harvesting
the second maize crop in October 1998 (Okalebo et al., 1999). Marked
increases were found in C and P levels. These are very likely due to
their accumulation from consecutive organic and mineral nutrient
additions made from 1997 to 2000 except P which was added only
once at 100kg P ha-1 at the beginning of the experiment in 1997. Other
additions of P probably originated from the decay and release of the
organically bound P held in organic resources applied as previously
described by Rusell (1973). As reported by Okalebo et al. (1999), there
were large amounts of total N found from surface soil in all treatments
Use of Organic and Inorganic Resources to Increase Maize Yields in some Kenyan
Infertile Soils: A Five-Year Experience
369
(including the control) after four years of cumulative organic matter
and urea incorporations into soil. Probably analytical errors contributed
to these large N levels. Nevertheless, for better estimates of the N status
of soils, analysis for mineral N (NH4 and NO3 mainly) could have been
done. Mineral N gives better measurements of N released from
decomposing organic resources. It also gives better estimates of N
leached down the soil profile. Unfortunately the Moi University
laboratory does not have facilities or equipments to analyse mineral N
components in soils. Other indicators of soil fertility build-up from
organic matter additions (such as an increase in particulate light
fraction in soils) would also be useful.
Conclusions and Recommendations
1. In the one year (1994) of field experimentation at three on-farm
researcher-managed trials (at Ndeiya, Gatuanyaga and Malava),
maize yield increases were obtained from combined application of
organics and inorganics. But the largest yield increases were found
from incorporation of high quality poultry manure (3.1% N),
groundnut trash (1.6% N) and phosphate rock (13.2% P). The
effectiveness of each resource varied with site.
2. The longer-term (4 years) study of Chepkoilel site (with a major N
limitation has demonstrated the positive effect of incorporating crop
residues with N fertilizer into the seedbed to improve their
decomposition and nutrient release characteristics. The two forms
of residues tested (wheat straw and soybean trash) were each applied
at a uniform rate of 2 tha-1. It is quite feasible that this rate of residue
be retained in croplands even while alternative requirements (for
example fuel, feed are also being met)
3. Comparisons of different rates of residues are suggested to obtain
responses to both residues and N fertilizer rates.
4. Organic matter fractionation needs to be done after cropping in the
long-term trial, to determine the dynamics of important labile
fractions (such as the proportions and contents of C, N and P in
those fractions).
5. Investigations on economics of combined organic and mineral
sources will provide useful information on crop/tree residue and
manure management practices.
6. The study in one year (1994) was not able to pinpoint the enhanced
dissolution of PR and the associated mechanisms through combined
organic matter and PR incorporations into soil. This task needs
pursuing.
370
Okalebo, J.R. et al
Acknowledgements
We acknowledge the financial support of the Tropical Soil Biology and
Fertility Programme (TSBF) towards research expenses in 1994 and
1997, including the funding for the Senior Author to attend the 8th
AFNet Workshop in Arusha, Tanzania. We also acknowledge funding
from the Kenya Agricultural Research Institute (KARI) that enabled us
to execute our activities partly in 1994 (under the European Economic
Community then) and from 1998 - 2000 under their Agricultural
Research Fund (ARF) financial support. We are grateful to the Technical
staff at NARC Muguga, KARI and Moi University for their assistance in
field and Laboratory tasks.
References
Fertilizer use Recommendation Project (FURP) (1994) Fertilizer Use
Recommendations: Volumes 1-22, Kenya Agricultural Research Institute
(KARI) Nairobi, Kenya.
Gachengo, C.N. (1996) Phosphorus release and availability on addition of organic
materials to phosphorus fixing soils. M.Phil Thesis, Moi University, Eldoret,
Kenya.
Jaetzold, R. and Schmidt, H. (1983) Farm Management Handbook of Kenya.
Natural Conditions and Farm Management Information. Ministry of
Agriculture, Kenya.
Jama, B, Swinkels, R.A. and Buresh, R.J. (1997) Agronomic and economic
evaluation of organic and inorganic sources of phosphorus for maize. Agron.
J. 89: 597-604.
Makken, E. (1993) Overview of the agricultural sector in sub-Saharan Africa. In
Van Reuler, H. and: The role of plant nutrients for sustainable food crop
production in sub-Saharan Africa Prins, W.H. (eds) pp. 25-36. Leidschendem.
The Netherlands Dutch Association of Fertilizer Production.
Muasya, R.M. (1996) Wheat response to fortified wheat straw compost. M.Phil
Thesis, Moi University, Eldoret Kenya.
Muasya, R.M., Mwakha, E., Othieno, C.O., and Okalebo, J.R. (1996) Initial wheat
response to low and high rates of phosphate fortified compost. Proc. Of the
15th Meeting of the Soil Science Society of East Africa (SSSEA). Nanyuki,
Kenya. 19-23 August, 1996, pp. 79-87.
Mutuo, P.K., Smithson, P.C., Buresh, R.J. and Okalebo, J.R. (1999) Comparison
of phosphate rock and triplesuperphosphate on a phosphorus deficient
Kenyan soil. Commun. Soil Sci. Plant Anal. 30: 1091-1103.
Mwaura, H.W. (1998) Determination of nutrients limiting crop/tree growth on
a ferralsol at Chepkoilel Campus, Uasin Gishu district, Kenya. Fourth Year
B.Sc. (For) Dissertation. Moi University, Eldoret, Kenya.
Use of Organic and Inorganic Resources to Increase Maize Yields in some Kenyan
Infertile Soils: A Five-Year Experience
371
Nahas, E., Banzato, D.A. and Asis, L.C. (1996) Fluorapatite solubilisation by
Aspergillus niger in vinasse medium. Soil Biology and Biochemistry 22: 10971101.
Nyambati, R.O. (2000) Soil phosphorus fractions as influenced by P and N
sources on two sites in Western Kenya. M.Phil Thesis. Moi University, Eldoret,
Kenya.
Okalebo, J.R. (1987) A study of the effect of phosphate fertilizers on maize and
sorghum in some East African Soils. Ph.D Thesis, University of Nairobi,
Kenya.
Okalebo, J.R. and Lekasi, J.K. (1993) Effect of combined groundnut trash, maize
stover and farmyard manure with phosphate rock on maize yield at Malava,
Kakamega. In: NARC Muguga (KARI) Record of Research, Annual Report 1993.
Okalebo, J.R., Lekasi, J. K. and Schnier, H.F. (1991/95) Phosphate response
studies at Malava smallholder site, Kakamega, Kenya. In: NARC Muguga
(KARI) Records of Research. Annual Reports. 1991-1995.
Okalebo, J.R., Simpson, J.R. and Probert, M.E. (1992) Phosphorus status of
cropland soils in the semi-arid areas of Machakos and Kitui districts, Kenya.
In (M.E. Probert ed.): A search for strategies for sustainable dryland cropping
in Eastern Kenya. ACIAR (Australia) Proceedings, 41. 50-54.
Okalebo, J.R. and Woomer, P.L. (1994) Use of rock and fertilizer phosphorus in
Eastern and Southern Africa: A data summary. Proc. Of the 14th Conference
of the Soil Science Society of East Africa, Mbarara, Uganda, November,
1994.
Okalebo, J.R. and Nandwa, S. (1997) Effect of organic materials with and without
inorganic P fertilizer on maize yield mainly in P deficient soil. Report on
Activity 3.1.1. Soil Fertility and Plant Nutrition Research Programme, KARI,
EU Funded pp. 10
Okalebo, J.R., Gathua, K.W. and Woomer, P.L. (1993) Laboratory Methods of
Soil and Plant Analysis: A working manual. TSBF-UNESCO, Nairobi, EPZ
Publishers, Nairobi, Kenya, 88p.
Okalebo, J.R., Palm, C.A., Gichuru, M., Owuor, J.O., Othieno, C.O.,
Munyampundu, A., Muasya, R.M. and Woomer, P.L. (1999) Use of wheat
straw, soybean trash and nitrogen fertilizer for maize production in the
Kenyan Highlands. African Crop Science Journal . Vol. 7 No. 4. 423-431.
Palm. C.A., Myers, R.J.K. and Nandwa, S.M. (1997) Combined use of organic
and inorganic nutrient sources for soil fertility maintenance and
replenishment. In: R.J. Buresh et al. (eds). Replenishing Soil Fertility in Africa.
SSA Special Publ. 51. SSA. Madison. USA, p. 193-217.
Probert, M.E., Okalebo, J.R., Simpson, J.R. and Jones, R.K. (1992) The role of
boma manure for improving soil fertility. In: M.E. Probert et al ACIAR Proc.
41, 63-70. A search for strategies for sustainable dryland cropping in Eastern
Kenya. Symposium Proceedings, 1990.
Russell, E.W. (1973) Soil Conditions and Plant Growth. 10th ed. The English
Language book Society and Longman, U.K.
Sanchez, P.A., Shepherd, K.D., Soule, M.I., Place F.M., Buresh R.J., A.M.N.
Izac, Mokunye A.U., Kweriga F.B., Ndiritu C.G. and Woomer, P.L. (1997)
Soil fertility replenishment in Africa . An investment in natural capital.
372
Okalebo, J.R. et al
Proceedings of a Workshop 26-30 September 1996, Addis Ababa, Ethiopia.
SSA Special Publication No. 51, Madison USA, p 1-46.
Simpson, J.R. Okalebo, J.R., and Lubulwa, G. (1996) The problem of Maintaining
soil Fertility in Eastern Kenya: A Review of Relevent Research. ACIAR
(Australia) Monograph No. 41 pp. 60.
Symth, T.D. and Sanchez, P.A. (1982) Phosphate rock dissolution and availability
in Cerrado soils as affected by phosphorus sorption capacity. Soil Sci. Soc.
Am. J. 46: 339-345.
Tropical Soil Biology and Fertility Programme (TSBF). (1994) Nutrient Limiting
Experiment in Western Kenya by TSBF staff. 1994 - A Report.
Waigwa, M., Othieno, C.O and Okalebo, J.R. (2000) The use of manure and crop
residues to improve the solubility of Minjingu phosphate rock for phosphorus
replenishment in acid soils of Western Kenya. Presented at the 4th FORUM
Meeting, Malawi, 2000.
Woomer, P.L., Okalebo, J.R. and Sanchez, P.A. (1997) Phosphorus replenishment
in Western Kenya: from field experiments to an operational strategy: African
Crop Science Conference Proceedings 3(1), 559-570.
World Bank (1996) Natural Resource Degradation in sub-Saharan Africa .
Restoration of soil fertility, Africa Region. World Bank: Washington D.C.
USA.
The Potential of Green Manures to Increase Soil Fertility and Maize Yields in Malawi
The Potential of Green
Manures to Increase Soil
Fertility and Maize Yields in
Malawi
373
26
Sakala, W.D.*, Kumwenda, J.D.T. and
Saka, A.R.
Chitedze Agricultural Research Station, P.O. Box 158,
Lilongwe, Malawi
*Corresponding author
Abstract
The effect of sole maize and green manures ( Mucuna
pruriens, Crotalaria juncea and Lablab purpureus) on maize
for two successive cropping seasons was determined in an
on-farm experiment at five locations in Malawi, from 1996
to 1999. Legume residues were incorporated in two different
manners; “early” at peak biomass and “late” when the plants
started to senesce.
After growing and incorporating the green manures after
the end of the 1996/97 growing season, maize was planted
in 1997/98 and 1998/99 to test the effect of the legumes
on maize yields compared with continuous maize. Biomass
production from “early” incorporated legume residues were
6.7 t ha-1 for Mucuna, 4.9 t ha-1 for Crotalaria, and 4.9 t ha1
for Lablab purpureus; and for “late” incorporated legume
residues were 5.9, 5.2 and 4.1 t ha-1 for the same legumes,
374
Sakala, W.D. et al
respectively. Of the three legumes Lablab purpureus
produced less biomass (averaged 4.2 t ha-1), compared with
the other two green manures and Mucuna produced the
highest seed yield. Over the two seasons and across the
five sites, the application of inorganic fertiliser (35 or 69 kg
N ha -1) to maize significantly increased maize yields in all
the sites. Maize yields following green manures without
inorganic fertiliser additions were much higher than yields
from continuous maize with no fertiliser addition. Addition
of inorganic fertiliser to legume crop residues resulted in
increased maize yields at all the sites, but the highest
fertilizer use efficiency was obtained from the addition of
35 kg N ha -1 . There were no significant maize yield
differences when maize followed early or late incorporated
green manures across season and sites for all the three
legumes. Results indicate that all the three green manures
have potential to increase maize yields when used as sole
green manures or in combination with inorganic fertilizers
compared to sole maize alone, following each other.
Introduction
Smallholder farmers in Malawi are confronting a difficult set of conditions
that threaten not only their livelihoods, but also their abilities to feed
themselves and their families. Over the past twenty-four years, Malawi
has experienced a series of periodic drought that requires food relief by
international agencies. Structural reforms imposed, as a condition for
development aid, required the removal of subsidies for farm inputs,
which in turn reduced farmers access to materials required to modernise
agriculture. Continuous cropping and soil erosion has led to soil
degradation. But despite these setbacks, several promising new
technologies have emerged that offer promise to better manage
agricultural resources. One such technology is the use of nitrogen-fixing
green manure legumes (Giller and Wilson, 1991). A green manure legume
is one that is grown specifically for use as organic manure and maximises
the amount of N from the legume that is available for a subsequent
crop.
Apart from periodic drought and effects of structural adjustments
in Southern Africa, and in particular Malawi, some of the major causes
of low maize yields are declining soil fertility and insufficient use of
fertilisers resulting in severe nutrient depletion of soils (Buresh et al.,
1997). Inorganic fertilizers are unaffordable by most smallholder farmers
in Malawi (Kumwenda et al., 1997). Use of green manures in combination
with inorganic fertilizers is a less expensive way of increasing soil fertility
The Potential of Green Manures to Increase Soil Fertility and Maize Yields in Malawi
375
and consequently maize yields compared with the use of inorganic
fertilizers alone. Although the use of green manures in Malawi was first
initiated more than 70 years ago (Sakala, 2000), intensive research on
green manures gained momentum in the 1990s.
Large increases in growth and yield of crops sown after incorporation
of green manures have been reported, for example, maize yields have
more than doubled by incorporation of a 3-month-old green manure of
Mucuna pruriens var. utilis or Crotalaria juncea grown in alluvial soil on
the island of Java, Indonesia (Hairiah and van Noordwijk, 1989). Simple
decision aids based on chemical and physical characteristics of the green
manures have been developed to guide scientists, as well as farmers, in
choosing the legumes which are suitable for improving soil fertility (Palm
et al., 1997). This paper summarises the results of an experiment which
was initiated to:
(i) determine biomass production of early and late incorporated legume
manures
(ii) determine the effects of legume manures on the subsequent maize
yields, and
(iii) test the effect of combining organic and inorganic fertilisers on maize
yield.
Materials and Methods
An experiment with three legumes that were to be incorporated late
(after seed was harvested) was started in the 1995/96 cropping season.
The experiment was sited both on station and on-farm. During this
season all the on-farm sites were not successful because animals ate
all the legume crops as incorporation was scheduled after maize harvest.
This led to a new experiment, which included an early incorporation
treatment (before seed harvest) in 1996/1997 cropping season. This
trial was initiated at Bembeke, Mathambi, Mbawa, Chitedze and
Kamwendo. The site location characteristics are presented in Table 26.1.
A split plot design was used and seven treatments were arranged in
randomised complete block, with 3 replications. Each plot was 18 rows,
10 m long. The treatments for the 1996/97 seasons were: Sole Mucuna
(Mucuna pruriens), with early incorporation at maximum flowering and
pod initiation incorporation, sole Mucuna, with late incorporation after
harvest of Mucuna seed or grain, sole Crotalaria (Crotalaria juncea), with
early incorporation as in I, Sole Crotalaria with late incorporation as in II,
sole Lablab purpureus with early incorporation as in I, sole Lablab
purpureus with late incorporation as in II and sole Maize crop (NSCM 51).
In the 1997/98 and 1998/99 cropping season, maize a hybrid maize
NCM51 was planted on the 7 main-plots following Mucuna, Crotalaria or
Lablab purpureus legume crop residue, each incorporated early or late,
376
Sakala, W.D. et al
Table 26.1: Experimental site location characteristics where the experiment was
conducted between 1996 to 1999
Trial site
location
Evaluation
(masl)
Latitude
Longitude
Soils
Rainfall
range (mm)
Rainfall
variabilty
Bembeke
1219-1585
34° 25’
14° 21’
800-1,270
Low
Mathambi 1200-1810
35° 21’
16° 01’
2,000-2,400
Low
Mbawa
1219-1250
33° 25’
12° 07’
700-1,200
Medium
Chitedze
1097-1372
33° 38’
13° 59’
700-1,200
Medium
Kamwendo 1067 – 1158 38’ 02’
13° 50’
Ferallitic
Latosols
Ferallitic
Latosols
Weakly
Ferallitic
Lotosols,
Ferruginous
Latosols
Weakly
Ferallitic
Latosols
700-1,100
Medium
and a pure continuous maize crop. Each main-plot had 3 sub-plots
consisting of inorganic fertilisers which were added as packages as follows:
1) no fertiliser, 2) 35:10:0:+2S kg ha -1, and 3) 69:21:0+4S kg ha -1
(N:P2O5:0+S). A split plot design was used and these treatments were
arranged in a randomised complete block design with three replications.
Each main plot had 18 rows while each sub-plot had 6 rows, each 10 m
long. In plots, which had fertilizer, the basal fertiliser was applied at
planting and the source was 23:21:0+4S and the top-dressing source
was urea (46%N) that was applied three weeks after planting. Maize was
planted at a rate of 44,000 plants per hectare. Maize yield was determined
by harvesting 4 middle rows of each sub-plot, 9.1 m long each and the
yield was adjusted to 12.5% moisture content. Mucuna and Lablab
purpureus were planted at 74,407 seeds per hectare (90 cm x 15 cm x 1
plant). Crotalaria was drilled on the ridge, double row per ridge, at 45 kg
of seed per hectare. Maize seed was planted at 44,000 seeds per ha-1 (0.9
m x 0.75 m x 3 plants). The pure crop of maize received 35:10:0+2S
(N:P2O5+S) per hectare of fertiliser from 23:21:4S as a basal fertiliser and
from urea as a top dressing fertiliser. The aboveground biomass was
measured at each incorporation time, early or late (after seed harvest).
Legume seed or grain yield of late incorporated legume manures was also
determined. Seed yield of legume manures was determined by harvesting
18 rows x 10 m long each. Legume seed yields are reported for Mucuna
and Crotalaria only because Lablab purpureus started flowering late and
therefore seed was not harvested.
Maize yield was determined by harvesting 16 middle rows, 9.1 m
long each, and adjusted to 12.5% moisture content. Maize and green
manure yields were analysed by use of general statistical program
The Potential of Green Manures to Increase Soil Fertility and Maize Yields in Malawi
377
(GENSTAT), developed at Rothamsted experimental station (Payne,
1978). Analysis of variance was the main procedure used for testing
significances of differences between means.
Results
The soils from the experimental sites were slightly acidic and pH ranged
from 5.8 to 6.2 (Table 26.2). Bembeke was more acidic compared to the
other sites. All sites had low organic carbon, which is a common
characteristic of most soils from farmer’s fields in Malawi. The
exchangeable cations were also on the low side.
Table 26.2: Soil chemical characteristics for some experimental sites
Sites
pH
Total N
ppm
OC
%
P
mg kg-1
Ca
mg kg-1
K
mg kg-1
Mg
mg kg-1
Bembeke
Kamwendo
Mathambi
Chitedze
5.9
5.8
5.3
6.2
45.1
18.5
43.7
19.5
1.8
1.0
1.8
2.1
5.7
0.4
8.8
1.5
1.8
0.8
1.4
3.3
0.7
0.3
0.4
0.3
0.3
0.2
0.3
0.2
Crotalaria took a shorter time than the other legumes to attain
maximum flowering stage, which ranged from 64 to 85 days after
planting, followed by Mucuna, which ranged from 122 to 142 days after
planting. Mucuna and Crotalaria were incorporated when the soil was
still wet because they reached maximum flowering while the rainy season
was still in progress. Lablab purpureus flowered very late, hence, early
incorporation was thus done before flowering. At the date of early
incorporation, Mucuna had accumulated the highest dry matter (5 to
11 t ha-1 and averaged 6.7 t ha-1) followed by Crotalaria, which averaged
4.9 t ha-1 (Figure 26.1).
The biomass for late incorporated Mucuna ranged from 3.8 to 7.9
and averaged 5.9 t ha-1, Crotalaria biomass ranged from 3.8 to 8.6 and
averaged 5.2 t ha-1 and Lablab ranged from 0 to 8.9 and averaged 4.0 t
ha-1. Total nitrogen contribution to the soil ranged from 164 kg N ha-1
for late incorporated Lablab to 367 kg N ha -1 for early-incorporated
Mucuna.
The highest maize grain yield 2.1 t ha-1 and seed yield of Crotalaria
1.7 t ha -1 were obtained at Chitedze in 1997/98 season (Table 26.3).
The highest seed yields of Mucuna were obtained at Chitedze (1.8 t ha
1
). Mucuna seed yield was also higher than for maize or Crotalaria at
Kamwendo, Bembeke, and Mbawa. This shows that Mucuna like other
legume crops can grow well even on poor soils.
378
Sakala, W.D. et al
Figure 26.1: Mean dry matter (t ha-1) yields of early and late incorporated legume green
manures at five sites in 1996/97 cropping season
8
SE = 0.74
Early incorporation
Late incorporation
6
4
2
0
Mucuna
Crotalaria
Lablab
Green manure legumes
Table 26.3: Seed and grain yield (t ha-1 ) of legume crops and maize at test sites, 1996/
97 cropping season
Site
Chitedze
Kamwendo
Bembeke
Mbawa
Mathambi
Mean
Maize
Mucuna
Crotalaria
2.1
0.1
0.6
0.9
0.9*
1.8
1.2
1.1
2.5
1.7**
1.7
0.5
0.5
0.1
0.5
0.7
* Mean from 4 sites only; the farmer harvested the maize.
** Mean from 4 sites only. Mucuna did not flower at Chitala and Bembeke.
All the three green manures used in this experiment had a high N
content at flowering time (4%) (Table 26.4). The N content for the three
legumes were within the ranges of these green manures, which were
extracted from the organic resource database compiled by the Tropical
Soil Biology and fertility programme indicated in (Table 26.4).
379
The Potential of Green Manures to Increase Soil Fertility and Maize Yields in Malawi
Table 26.4: Chemical quality characteristics of Mucuna, Crotalaria and Lablab at flowering
time
Legume
%N
C:N ratio
Lignin (%)
Lignin:
N ratio
Mucuna
Crotalaria
(5.5) 1.4-6.5
(5.3) 1.6-5.7
9.8-30.8
8.0-32.1
5.5-16.8
3.8-9.8
1.3-8.3
1.0-6.3
Lablab
(4.1) 1.7-6.3
7.4-29.1
2.6-11.5
0.4-9.8
Source: Organic Resource Database Manual
( ) %N measured from the experiment.
When maize was planted following the three legumes and sole maize
for two successive seasons at five sites, maize grain yield differed with
maize following green manures having similar yields which were
significantly higher (P 0.05) than maize yield from plots where maize
was being grown continuously (Table 26.5). There were no significant
maize yield differences for each legume due to time of green manure
incorporation. Bembeke site had the least average maize yields during
the two seasons and Chitedze had the largest average maize yield when
grown after the green manures.
Table 26.5: Average maize grain yield following three legumes incorporated early or late
at five sites in 1997/98 and 1998/1999 seasons
Legume
incorporation
Bembeke Mathambi
Mbawa
(t ha-1)
Chitedze
Kamwendo Legume
Mean
Mucuna early
1.3
2.4
3.1
4.4
1.7
2.7
Mucuna late
1.8
2.3
3.4
3.5
1.9
2.9
Crotaralia early
Crotaralia late
1.9
0.9
2.8
2.6
3.7
3.5
4.2
3.6
1.9
2.0
3.0
2.8
Lablab early
Lablab Late
2.0
1.3
1.4
1.8
3.4
3.1
4.2
4.0
2.0
1.9
2.7
2.7
Maize
1.1
1.9
2.6
2.5
15
2.1
Site Mean
1.5
2.2
3.3
3.8
1.8
Legume
0.001
0.09
6.1
Site
0.001
0.09
36
Interaction
0.001
0.02
36
Significance
SED
CV (%)
380
Sakala, W.D. et al
Cross season and cross sites analysis showed that at all the five sites,
the application of inorganic fertilisers increased maize yield (Table 26.6).
An increase in fertilizer application resulted in greater maize yield increases
in plots where maize had followed a legume compared with plots where
maize was grown continuously. Legume residue incorporation resulted
in large maize yield increases for the two seasons that maize followed the
green manure planted in the first year at all the five sites compared with
continuous maize with no fertiliser additions. Again larger responses were
obtained when fertilizer was combined with inorganic fertilizer. There
were no significant differences in maize yields due to different types of
legumes that were planted in the first season. The mean for two seasons
showed no significant yield differences between maize yield following their
incorporation. The effect of time of incorporating legume residue was not
evident across the five sites for the two seasons.
Table 26.6: The effect of three different rates of inorganic fertilizer following legumes or
sole maize on the grain yield of maize across five sites for two seasons 1997-1999
Legume time
incorporation
Mucuna early
Mucuna late
Crotaralia early
Crotaralia late
Lablab early
Lablab late
Maize (Control)
Fertiliser Mean
Significance
SED
CV (%)
0
35:10:0+2S
69:21:0+4S
(t ha -1)
2.1
2.2
2.2
1.8
2.0
1.7
1.2
1.9
2.6
2.6
3.1
2.9
2.8
2.9
2.2
2.7
3.5
3.5
3.7
3.6
3.4
3.4
2.7
3.4
Legume
0.001
0.09
6.1
Fertilizer
0.001
0.07
8.7
Interaction
Ns
0.02
8.7
Legume
Mean
2.7
2.8
3.0
2.8
2.7
2.7
2.0
2.7
A combination of organic and inorganic fertilizer increased fertilizer
use efficiency at both rates of inorganic fertilizer (35 and 69 kgNha -1)
that were applied together with the green manures that were incorporated
during the first season compared to where inorganic fertilizers were
applied alone. Although fertilizer use efficiency was increased by
combining organic and inorganic fertilizer, the fertilizer use efficiency
was much higher when lower rates of 35 kgNha-1 were applied for both
year one (when the green manures had just been incorporated) and
year two (when the fertilizer was added after one season from the time
when green manures were incorporated). There were no marked
differences in fertilizer use efficiency due to combination of organic and
The Potential of Green Manures to Increase Soil Fertility and Maize Yields in Malawi
381
inorganic between year one and year two at each level of fertilizer added
to incorporated green manure, except for early Mucuna incorporation
which had a significantly reduced fertilizer use efficiency in the second
year when 35 kgN ha -1 of inorganic fertilizer were added. On the other
hand, late incorporated Mucuna had a slightly increased fertilizer use
efficiency in the second season at both rates of inorganic fertilizer added
to the green manures. Highest fertilizer use efficiency was realised from
early-incorporated Crotalaria at both rates of fertilizer combination over
the two season combined compared with the other two green manures
that were used. For Lablab purpureus both early and late incorporation
had similar fertilizer use efficiency (in the first and second season for
each rate of fertilizer combined with the green manure).
Discussion
Dry matter accumulation of the green manures and their
effect on maize yield
Dry matter accumulation varied among legume species because sampling
for dry matter and incorporation was done when each legume had reached
its maximum flowering stage. Late incorporation tended to have lower
biomass because biomass was measured after seed had been harvested
and during this time most leaves had senesced and started falling down
This affected the final biomass yield. Again, aphids severely infestedLablab
purpureus at Kamwendo and Mbawa and consequently resulted in reduced
dry matter yield at these sites (averaged 4.2 t ha-1). Lablab purpureus had
the lowest total nitrogen contribution to the soil due to low biomass yield
as well as low nitrogen content compared to the other legumes. Mucuna
seed yield was also highest compared with Lablab purpureus and Crotalaria
at Kamwendo, Bembeke, and Mbawa. This suggests that Mucuna unlike
other green manure crops can grow well even on poor soils.
Significance of high N content on the three green manures
All the three green manures had N content of more than 4% N, which is
an indication of better quality (Palm et al., 1991). Nitrogen release from
crop residues can be slow or rapid, depending on the quality of the
residues. With the high N content, it is likely that these legumes will be
able to release nitrogen for use by the plant when incorporated into the
field. Both slow and rapid release of nutrients from organic fertilisers can
have positive effects on nitrogen management in the soil (Sakala et al.,
2000). Rapid release enhances early nitrogen uptake by the crop but
may lead to nitrogen loss through leaching if crop demand is less than
382
Sakala, W.D. et al
the amount of nitrogen being released. Slow release would guarantee a
continuous supply of nitrogen during most of the growing period of the
crop, although, if the amount of nitrogen released is small its contribution
to crop growth may not significantly boost crop performance
Nitrogen use efficiency
Maize yields increased with the application of green manures and
inorganic fertilizers separately or in combination. However, the
combination of 35 kg N ha-1 and any of the three green manures produced
the highest efficiency in the first year of green manure application as
well as in the second year, when residual effects of the green manures
were being tested.
Conclusion
Results from this study have shown that early-incorporated Mucuna
produced the largest biomass, which averaged 6.7 t ha -1, followed by
Crotalaria, which produced 4.9 t ha -1. Biomass from late incorporated
legume manures were lower than from early incorporated legume
manures, due to the loss of leaves by harvest time. Application of organic
and inorganic fertilisers separately or in combination, increased maize
yields at all sites, but higher fertilizer use efficiency was obtained when
the green manures were combined with 35 kg N ha -1. Growing maize
after legume residues resulted in increased maize yields compared with
growing maize after maize. There was no difference between early and
late incorporation of legume residues, when the data were pulled for
two seasons. These results show that these three legume residues can
be potential alternative sources of fertilisers for a following maize crop.
Acknowledgments
We thank the Rockefeller Foundation for providing financial support
and field technicians for excellent assistance in the field.
References
Giller, K. E., and Wilson, K. J. (1991) Nitrogen fixation in tropical cropping systems.
CAB ,International, Wallingford.
Hairiah, K. and van Noordwijk, M. (1989) Root distribution of leguminous cover
crops in the humids tropics and effects on a subsequent maize crop. In:
The Potential of Green Manures to Increase Soil Fertility and Maize Yields in Malawi
383
van der Heide, J. (ed.) Nutrient Management for Food Crop Production in
Tropical Farming Sytems. Institute for Soil Fertility, Haren, The Nertherlands,
pp. 157-170.
Kumwenda, J.D.T, Waddington, S.R., Snapp, S.S., Jones, R.B. and Blackie,
M.J. (1997) Soil fertility management in the smallholder maize-based
cropping systems of Africa. In. The Emerging Maize Revolution in Africa:
The Role of Technology, Institution and Policy, Michigan State University,
USA, 9-12 July, 1995.
Kumwenda, J.D.T., Saka A.R., Kabambe V.H. and Ganunga R.P. (1998) Legume
manure crops-maize rotation trial for the 1997/98 cropping season. In 1997/
98 Annual Report, Chitedze Research Station, Lilongwe, Malawi.
Palm, C.A., Nandwa, S. and Myers, R.J. (1997) Combined use of organic and
inorganic nutrient sources for soil fertility maintenance and nutrient
replenishment. In: R.J. Buresh and P. A. Sanchez, (eds.) Replenishing Soil
Fertility in Africa Special Publ Vol. 51, pp. 193-217 . ASSA, CSSA, SSSA,
Madison, Wisconsin.
Palm, C.A. and Sanchez, P.A. (1991) Nitrogen release from the leaves of some
tropical legumes as affected by their lignin and polyphenolic contents. Soil
Biology and Biochemistry 23: 83-88.
Payne, R. W. (1978) Genstat 5 Reference Manual. Clarendon Press, Oxford.
Sakala, W.D. (2000) A review of potential legumes for integrated nutrient
management in maize-based systems in Malawi with emphasis on
pigeonpeas. Ritchie, J.M. (ed) (2000) Integrated crop management research
in Malawi. Proceedings of final workshop on the Farming systems integrated
pest management project, Club Makokola, Mangochi, Malawi, 29 November3 December, 1999. Chatham, UK: Natural Resources Institute.
Sakala, W., Cadisch, G. and Giller, K.E. (2000) Interactions between residues of
maize and pigeonpea and mineral N fertilizers during decomposition and N
mineralization. Soil Biol. Biochem. 32 (2000): 679-68.
384
Sakala, W.D. et al
Effects of Ramial Chipped Wood and Litter Compost of Casuarina Equisetifolia
Forst & Forst on Tomato Growth and Soil Properties in Niayes, Senegal
Effects of Ramial Chipped
Wood and Litter Compost of
Casuarina equisetifolia
Tomato Growth and Soil
Properties in Niayes, Senegal
385
27
Soumare, M.D.1, Mnkeni, P.N.S.2 and
Khouma, M.3
1
B P: 22087, Dakar – Ponty, Dakar, Senegal.
University of Fort Hare, Faculty of Agricultural and
Environmental Sciences, Department of Agronomy, Private
Bag X1314, Alice 5700, South Africa,
E mail: pmnkeni@ufh.ac.za ,Corresponding Author
3
National Agricultural Research Center, ISRA/CNRA, B. P. 53
Bambey, Senegal. E-mail: mkhouma@isra.sn
2
Abstract
Plantations of Casuarina equisetifolia in Niayes, Senegal
produce litter and ramial wood as by-products. These
organic materials were suspected to be potential sources of
plant nutrients upon decomposition in soils. However, this
potential remained to be established in the infertile sandy
soils of the Niayes area, Senegal. A field experiment was
therefore, conducted to study the effects of these forestry
386
Soumare, M.D. et al
by-products on tomato growth and soil properties. The
ramial wood was fragmented to produce ramial chipped
wood (RCW), while the litter was composted before being
applied to a sandy soil at three different levels: 10, 20, and
40 t ha-1 and compared to a reference control and locally
recommended fertilizer mixture. Soil and plant samples were
taken at 45 days of tomato growth and at harvest time for
analysis. Residual effects of the materials were also
evaluated through the establishment of a second tomato
crop on the same plots.
Application of RCW depressed tomato growth and yield
during the first cropping and this was attributed to the
effect of RCW to induce intense N immobilization in the soil
due to its wide C:N ratio. Improvements in growth and yield
were, however, observed during the second cropping and
were ascribed to improved nutrient release and especially
nitrogen as a result of the mineralization of earlier
immobilized nutrients, following the extended incubation
of the RCW in the soil. These results indicated that in order
to derive short-term benefits from RCW application, it may
have to be applied in combination with experimentally
determined amounts of mineral fertilizers.
Incorporation of litter compost (LC) improved tomato
growth and yield during both the first and second
croppings. The LC improved soil levels and tomato uptake
of N, P and K and possibly other nutrients that were not
measured. This was attributed to the narrower C:N ratio
of LC that encouraged its faster decomposition in the soil.
The observed effects were greater in the first than the
second crop indicating that LC had limited residual
nutrient value. Application of LC also improved soil organic
matter content, dry bulk density and water holding
capacity, suggesting that its regular use could result in
the long-term improvement of the productivity of the
experimental soil.
Key words: Casuarina equisetifolia, Litter compost, Nutrients, Ramialchipped wood, tomato, Senegal.
Introduction
Plantations of Casuarina equisetifolia (Forst & Forst), were established
in the Niayes region of Senegal through a re-forestation program begun
in 1948 (Maheut and Dommergues, 1963). These plantations play an
Effects of Ramial Chipped Wood and Litter Compost of Casuarina Equisetifolia
Forst & Forst on Tomato Growth and Soil Properties in Niayes, Senegal
387
important role in stabilizing the coastal sand dunes and help to protect
adjacent agricultural areas by acting as windbreaker. The C. equisetifolia
trees produce large quantities of organic materials in the form of leaf
litter as well as male and female spikes that accumulate on the plantation
floor. Mailly and Margolis (1992) estimated that a 13-year-old plantation
produces up to 3.3 kg m -2 year-1 of leaf litter. The accumulated litter is
so thick that it is suspected to prevent Casuarina seeds from germinating
by isolating them from the mineral soil. In addition, the decomposition
of the litter releases acids that further inhibit the germination of the
seeds. The partial removal of the litter from the forest floor may help to
solve the problem of germination of the C. equisetifolia seeds and thereby,
facilitate the natural regeneration of the plantations. One way of
encouraging its removal from the forest floor is for it to be used as a soil
amendment in crop or vegetable farming. Information on the agronomic
value of this amendment is, however, lacking.
Since C. equisetifolia plantations do not regenerate naturally, the
only way by which these plantations are regenerated in Senegal is by
cutting down 45-50 years old trees just before the rainy season begins.
After it rains, shoots sprout from the stumps, which upon pruning
grow to form new plants. This method of plantation management
results in the production of large quantities of trash in the form of
pruned small branches and twigs. Such branches and brushwood
have been viewed for centuries as having no value. However, there is
growing evidence that such brushwood and twigs, when applied to
soil as ramial chipped wood (RCW), could result in the improvement
of soil fertility and general soil productivity (Lemieux, 1993; Seck
and Lemieux, 1993; Sylvestre and Despatie, 1995). The trash from
the regeneration of C. equisetifolia plantation is, therefore, a byproduct that should be given consideration as a soil amendment in
the form of ramial chipped wood.
The Niayes area accounts for more than 80% of the vegetables
produced in Senegal for local consumption and export. The vegetables
are mainly produced under irrigation on sandy soils that are inherently
low in fertility. Due to continuous cultivation with the same crop
species, nematode infestation is reported to be a problem in the area
(Netscher, 1970). Farmers maintain their production levels by using
chemical fertilizers and pesticides, often in excessive amounts. This
practice is thought to be an unsustainable method of land use because
it reduces organic matter levels and thus enhances soil erosion,
especially in drifting sand dune soils. Farmers continue to use chemical
fertilizers and pesticides because of their highly visible short-term
benefits, but are generally not conscious of environmental pollution,
accumulation of pesticides and nitrates in the ecosystems and high
levels of pesticides in their produce. These apparent limitations in the
use of chemical fertilizers and their secondary effects, point to the
388
Soumare, M.D. et al
importance of alternative soil fertility management methods such as
the use of organic materials wherever these are readily available. The
objective of this study was, therefore, to evaluate the effectiveness of
the locally available litter compost and RCW of C. equisetifolia as
possible alternative sources of nutrients for tomato farming in Niayes,
Senegal.
Materials and Methods
Materials
The experiments were carried out on a field of the Developing Center for
Horticulture (CDH) farm. The farm covers an area of 40 ha, divided into
many fields in the Niayes area, Dakar, Senegal (12°30'E, 17°30'W). This
region consists of a coastal band, 15 to 20 km wide along the Atlantic
Ocean to the North of Dakar. The soil is classified as a dystric regosol
according to the FAO system of soil classification (FAO-Unesco-ISRIC,
1990). The site was previously cropped with onions from March to June
1999, after which it was left uncultivated until the planting time for this
experiment on October 28th 1999.
Fresh ligneous twigs less than 3 cm in diameter, were harvested
from a C. equisetifolia plantation and transported to the experimental
field where they were fragmented manually using a bush knife, into
small pieces of around 15 to 20 mm long. The resulting product is what
was used as ramial-chipped wood in this study.
The C. equisetifolia litter was collected from the plantation floor and
composted for three months (July to October). The heap method, utilizing
natural aeration (passive aeration), was used. The dimensions of each
of the three heaps used were 3 m long, 1.5 m wide and 1 m high. The
humidity was kept between 30% and 50% and the heaps were watered
whenever humidity was less than 30%.
A germination bioassay trial was conducted to monitor litter compost
maturity as the composting process progressed. In this trial, the
decomposing compost was sampled at the end of every week and put in
containers. Twenty-five lettuce seeds were then sown and their
germination percentage determined after 3 days. The results obtained
showed that seed germination was almost completely inhibited during
the first five weeks. However, the germination percentage increased
steadily thereafter and reached 100% during the 10th week indicating
that compost maturity had been achieved. The process was allowed to
continue for a further two weeks at which time the compost was used
for the field trials.
Effects of Ramial Chipped Wood and Litter Compost of Casuarina Equisetifolia
Forst & Forst on Tomato Growth and Soil Properties in Niayes, Senegal
389
Methods
A randomized complete block experimental design (RCBD), with four
replications was used for the field study. Treatments included a
control; recommended fertilizer (RF) rate for N, P and K; RF rate for
N, P and K plus mocap powder (nematicide); LC and RCW at three
levels each (10 t ha-1, 20 t ha-1 and 40 t ha-1) for a total of 9 treatments.
The recommended fertilizer included 20t ha -1 of horse manure (on
air-dry basis). The nematicide used was Ethoprophos (O-ethyl-S,Sdipropyl-phosphorodithioate) at a rate of 2 g m -2 . The materials were
applied to plots measuring 5 m long by 3.5 m wide. The RCW was
applied fresh and mixed using a spade in the top 5 cm to 7 cm of soil
while litter compost was applied and mixed with the top 15 cm of
soil.
Tomato (Lycospersicum esculenta) (CDH tomato variety XINA) seeds
were sown and raised in a nursery until the three-leaf stage (around
10-15 cm tall) at which time seedlings were transplanted to the preirrigated plots. The seedlings were transplanted at a spacing of 50 cm
within rows and 50 cm between rows. The tomato plants were regularly
monitored and treated against insect pests as well as fungus and bacteria
infections. Maneb (C4H6N2S4.Mn) was the fungicide used at a rate of 2 g
m-2 while Copac (ammoniacal copper sulphate) at a concentration of 10
ml l -1 was used for bacterial control. Maneb was applied once a week
and after rain events while Copac was applied every two weeks during
the growing period.
Treatment effects on plant growth were evaluated by measuring plant
height at 45 days of growth and tomato fruit yield at harvest time.
Harvesting began 80 to 90 days after planting and was done three times
a week. The harvest area was 4.5 m long and 3 m wide for each plot,
0.5m from each side of the plot were left as guard rows. The number of
fruits was counted and weights determined.
Leaf, soil and root samples were taken at specific times to assess
treatment effects on soil and plant nutrient contents. Leaf sampling
was done at 45 days of growth and at harvest time. This was done by
taking the third tomato leaf from the growing tip of each of 20 tomato
plants selected randomly from every plot. Soil sampling for purposes of
assessing treatment effects on soil properties was done after the second
harvest.
In order to evaluate the residual effects of the organic amendments,
a second tomato crop was grown on the same treatment plots after
clearing, but without further organic amendment applications. The RF
and RF+N treatments were, however, reapplied. The planting and the
management procedures were done exactly as for the first crop.
The pre-cropping soil and soil samples taken from treatment plots
after harvest were characterized for pH (McLean, 1982), organic carbon
390
Soumare, M.D. et al
(Nelson and Sommers, 1982), Total-N (Bremner and Mulvaney, 1982),
extractable P (Olsen and Sommers, 1982) and exchangeable K (Knudsen,
Peterson and Pratt, 1982), dry bulk density (Blake and Hartge, 1986),
particle size analysis (Gee and Bauder, 1986), water holding capacity
(Cassel and Nielsen, 1986), and cation exchange capacity (Rhoades and
Miyamoto 1990).
Litter compost and RCW samples were analyzed for pH and for carbon
by the wet oxidation method utilizing acidified dichromate as described
by Nelson and Sommers (1982). In addition, RCW and LC along with
tomato leaf samples were analyzed for N by the Kjeldahl method
(Bremmer and Mulvaney, 1982) as well as for P by the vanado-molybdophosphorus method (Okalebo et al., 1993), K by flame photometer and
Ca and Mg by EDTA titration (Lanyon and Heald, 1982). The results for
the organic amendments together with those of the pre-cropping soil
are summarized in Table 27.1.
Table 27.1: Selected properties of the experimental soil, ramial chipped wood (RCW)
and litter compost (LC)
Chemical or physical property
Soil
LC
RCW
C (%)
N (%)
C:N ratio
P (%)
C:P ratio
K (%)
Ca (%)
Mg (%)
EC(dS/m)
OM (%)
CEC (cmol(+)kg -1)
pH (H20)
DBD (Mg/m 3)
Particle size analysis (%)
-Sand
-Silt
-Clay
WHC (mm/m)
1.2
0.08
15
0.02
60
0.08
1.52
2.10
7
5.4
1.8
40
1.3
31
2.3
17
2.8
2.6
0.4
64
6.8
-
58.8
1.15
51
0.11
535
0.53
1.39
0.12
78
5.2
-
95
4.5
0.5
80
-
-
Data Analysis
The raw data obtained were statistically analyzed following procedures
described by Gomez and Gomez (1984). Analysis of variance (ANOVA)
was performed to evaluate treatment effects on the different parameters
that were measured. The least significant difference (LSD) test was used
Effects of Ramial Chipped Wood and Litter Compost of Casuarina Equisetifolia
Forst & Forst on Tomato Growth and Soil Properties in Niayes, Senegal
391
to separate treatment means and means were declared as significantly
different at P< 0.05.
The relative effects of the amendments on plant height (RH) and
fruit yield (RY) were calculated using a formula described by Engelstad
et al. (1974), viz:
RH =
HA - HC
x 100
HF - HC
RY =
YA - YC
x 100
YF - YC
Where:
HF and YF = plant height and fruit yield observed in the reference
fertilizer treatment plots, respectively.
HA and YA = plant height and fruit yield observed in a given
amendment treatment plots, respectively
HC and YC = plant height and fruit yield observed in the control
treatment plots, respectively
Results and Discussion
Effect of amendments on tomato growth and yield
Effects of ramial chipped wood
The relative effects of treatments on height (RH) at low, medium and
high rates of RCW application were – 37%, 10% and –78%, respectively
(Table 27.2).
The RY values were 9.75 %, -3.43 % and –25% for low, medium and
high rates of RCW application, respectively. These results showed that
RCW had a negative effect on plant growth and yield during the first
cropping. This depression in tomato growth and yield was most likely
the result of N immobilization by soil microorganisms.
According to Bartholomew (1965), addition of organic materials with
a total N content of 1.5 % can trigger N immobilization in soil. The RCW
that was applied in this study had a total N content of only 1.15 %
(Table 27.1), which was below the critical level suggested by Bartholomew
(1965). The suspected nutrient immobilization could also be explained
by the C:N and C:P ratios of the applied materials. The optimum C:N
ratio for a rapid decomposition and N mineralization was found to be
equal or less than 30 (Brady and Weil, 1999) or about 30 to 35
(Mustin,1981). With respect to C:P ratios, Rustad and Cronan (1988),
cited by Tremblay and Beauchamp (1998), reported that the critical C:P
392
Soumare, M.D. et al
ratio of organic residues above which net immobilization occurs is
between 350 to 480. The C:N and C:P ratios of the RCW used in the
present study were 51 and 535, respectively. Both values were above
the critical values suggesting that RCW could have induced N and P
immobilization thus reducing the levels of these nutrients in the soil
and their subsequent uptake by plants. The N immobilization is
confirmed by leaf N concentrations data observed at 45 days of growth
for the first crop (Table 27.3) which show that leaf N values in RCW
treatments were less than those observed in control treatments.
Table 27.2: The effect of RCW and LC on the growth and yield of two successive tomato
crops
Treatments
First crop
Height
(cm)
Control
RF
RF+N
RCW10 t ha-1
RCW20 t ha-1
RCW40 t ha-1
LC10 t ha-1
LC20 t ha-1
LC40 t ha-1
LSD (0.05)
50.1e**
62.3b
61.1c
45.6g
48.8f
40.6h
55.8d
61.1c
65.3a
0.5
RH
(%)
-37
-10
-78
46
90
100
Second crop
Yield
(t ha-1)
19.8e
26.6b
28.5b
19.0f
19.5ef
17.7g
23.8d
27.7c
29.2a
1.0
RY
(%)
9.7
-3.4
-25
47
82
115
Height
(cm)
34.9h
65.2b
64.8a
51.7e
55.2d
50.3f
49.5g
60.7c
64.2a
0.2
RH
(%)
56
67
51
48
85
97
Yield RY
(t ha-1) (%)
16.5g
28.6b
28.7b
24.1e
25.9d
22.9f
26.9c
28.6b
29.1a
0.3
62
78
52
87
100
132
*The recommended fertilizer included 20 t ha-1 horse manure.
**Means in each column followed by the same letter are not significantly different at p
0.05 according to the LSD test.
The application of RCW, however, had no effect on the leaf
concentrations of P in the first crop (Table 27.3), possibly because the P
immobilization was not intense enough to bring significant changes in
spite of the wide C:P ratio of 535. Similarly, RCW incorporation had no
effect on the leaf concentrations of K for the first crop (Table 27.3),
indicating that the observed reductions on tomato growth were largely
a result of the effects of added RCW on soil N.
The incorporated RCW had positive effects on the growth and yield
of the second tomato crop. The relative effects on height (RH) at low,
medium and high rates of RCW application were 56%, 67% and 51%,
respectively (Table 27.2). The corresponding RY values for fruit yield
were 62%, 78%, and 53% for low, medium and high rates of RCW
application, respectively. Based on these results, it is evident that
Effects of Ramial Chipped Wood and Litter Compost of Casuarina Equisetifolia
Forst & Forst on Tomato Growth and Soil Properties in Niayes, Senegal
393
20 t ha-1 of RCW application resulted in the highest tomato yields. The
observed improvement in plant growth appears to be related to improved
nutrient supply and availability. In contrast to the first crop, the
incorporated RCW resulted in increased leaf N, P and K concentrations
(Table 27.3). The most remarkable increases were observed with leaf N
which changed from a depressed situation in the first crop to a situation
where significant increases in leaf N were observed at each level of RCW
application (Table 27.3). This suggests that by the time the second crop
was planted the C:N ratio of the incorporated RCW had narrowed
sufficiently to result in net N mineralization instead of immobilization.
Table 27.3: The effect of RCW and L C on tomato leaf contents of N, P and K after 45
days of growth for the first and second crops
Treatment
Control
RF
RF+N
RCW10 t ha-1
RCW20 t ha-1
RCW40 t ha-1
LC10 t ha-1
LC20 t ha-1
LC40 t ha-1
LSD (0.05)
CV (%)
First crop
Second crop
N(%)
P(%)
K(%)
N(%)
P(%)
K(%)
0.95d**
3.90c
4.10c
0.22e
0.22e
0.14e
3.46c
4.72b
7.27a
0.66
15
0.23d
0.49b
0.48b
0.31cd
0.34c
0.21c
0.51b
0.71a
0.75a
0.10
14
1.11c
3.31b
2.61b
1.12c
1.25c
1.19c
1.58c
2.66b
4.33a
0.71
19.3
0.21d
1.37c
1.41c
1.51c
2.80b
3.88a
1.42c
1.20c
2.71b
0.65
17.6
0.01d
0.49a
0.44a
0.42a
0.45a
0.47a
0.21c
0.25bc
0.31b
0.08
16
0.01d
3.34a
3.12a
1.25c
2.35b
2.29b
0.97c
1.03c
2.28b
0.35
11.4
*The recommended fertilizer contained also horse manure (20 t ha-1)
**Means in each column followed by the same letter are not significantly
different at p 0.05 according to the LSD test.
These results imply that, in order to derive maximum short-term
crop benefits from the RCW of C. equisetifolia, it ought to be allowed to
undergo some decomposition in the soil first before a crop is planted.
Studies are therefore required to determine how long, before planting,
should the RCW be incorporated. If the incubation period of RCW before
planting is too long, other studies can be conducted to find ways of
shortening the period, possibly through co-application with limited
quantities of inorganic fertilizers to overcome the deleterious effects of
low quality organics. However, for longer-term crops, the RCW can be
used without co-application with other materials.
The results of this study indicate that RCW is potentially a good
organic amendment that should be seriously considered for use in the
Niayes area. If this idea is adopted, it will be necessary to work out a
394
Soumare, M.D. et al
sustainable ramial wood harvesting regime. At the moment, the ramial
wood becomes available only after the trees have been coppiced. However,
it is also practically possible to provide a regular supply of the ramial
wood through regular pruning of small branches in between coppicing
periods.
Effects of litter compost
Application of LC to soil stimulated plant growth. The relative effects on
height for LC at low, medium and high levels were 46%, 90% and 100
%, respectively (Table 27.2). The corresponding relative effects on yield
during the first crop were 47% at low level, 81% at medium level and
115% at high LC levels of application. The effects were greater during
the second crop where RY values of 86% for the LC at low level, 100%
for the medium level and 132% for the high level, were observed.
Table 27.4: The effect of RCW and LC on leaf contents of N, P and K at harvest time of
the first and second tomato crops
Treatment
Control
RF
RF+N
RCW10 t ha -1
RCW20 t ha -1
RCW40 t ha -1
LC10 t ha -1
LC20 t ha -1
LC40 t ha -1
LSD (0.05)
First crop
Second crop
N(%)
P(%)
K(%)
N(%)
P(%)
K(%)
0.65 d**
1.32c
1.27c
0.55e
0.72d
0.31f
1.28c
2.15b
3.67a
0.09
0.09g
0.10f
0.10f
0.16d
0.16e
0.18c
0.10f
0.21b
0.31a
0.27
0.08f
1.53d
1.59d
1.94c
2.18ab
1.39de
2.04bc
1.29e
2.37a
0.18
0.10f
0.36d
0.35d
1.76b
2.12a
1.45c
0.13e
0.14e
0.24de
0.16
0.01i
0.12d
0.12d
0.41c
0.53b
0.54a
0.03g
0.05f
0.09e
0.02
0.01e
0.17d
0.16d
1.27c
2.29a
2.13b
0.17d
0.16d
0.13d
0.07
*The recommended fertilizer contained also some horse manure (20 t ha-1)
** Means in each column followed by the same letter are not significantly different at
p 0.05 according to the LSD test.
During the first crop, tomato growth and yield were increased with
each increment in LC application, especially when the rate of applied
LC was increased from 20 to 40 t ha -1. This indicated that maximum
tomato growth and yield was not achieved with the rates of LC application
used in the present study.
The positive effects of LC on the growth of the first tomato crop were
associated with its effect to increase soil levels of N, P and K relative to
the control and RCW treatments (Table 27.5).
Effects of Ramial Chipped Wood and Litter Compost of Casuarina Equisetifolia
Forst & Forst on Tomato Growth and Soil Properties in Niayes, Senegal
395
Table 27.5: The effect of RCW and LC on soil N, P and K contents after harvesting the
first and second tomato crops
Treatment
Control
RF
RF+N
RCW10 t ha-1
RCW20 t ha-1
RCW40 t ha-1
LC10 t ha-1
LC20 t ha-1
LC40 t ha-1
LSD (0.05)
CV (%)
First crop
Second crop
N(%)
P(mg kg-1)
K(%)
N(%)
P(mg kg-1)
K(%)
0.15d
0.29c
0.27c
0.12d
0.15d
0.13d
0.36b
0.40b
0.51a
0.05
13
8.8c
11.68ab
11.32b
6.21d
5.95d
6.25d
7.8c
10.40b
13a
1.34
9
0.08e
0.18e
0.17e
1.10e
1.18d
1.17d
1.943c
2.78b
3.51a
0.040
2
0.03c
0.27a
0.29a
0.33a
0.33a
0.39a
0.16b
0.16b
0.17b
0.06
19
9.37c
12.68b
13.32b
7.22de
6.45e
6.94de
8.75cd
11.82b
17.05a
2
12
0.01e
0.17c
0.17c
1.93a
1.96a
1.35b
0.15cd
0.17c
0.11d
0.05
5
* The recommended fertilizer contained also horse manure (20 t ha-1)
** Means in each column followed by the same letter are not significantly different at p
0.05 according to the LSD test.
This was in turn reflected in corresponding increases in the plant uptake
of the nutrients (Tables 27.3 and 27.4). According to Foth and Ellis
(1988), young mature leaves of tomato are considered to have adequate
levels of N, P and K when they contain at least 1.2% N, 0.3% P and 0.3%
K. The concentration of nutrients in tomato leaves after 45 days growth
in plots treated with LC (Table 27.3) was higher than the critical levels
reported by Foth and Ellis (1988).
The greater effect of LC compared to RCW to increase soil nutrients
and tomato growth could be explained by the differences in the C:N and
C:P ratios of the two materials. The LC amendment had a C:N ratio of
31 and a C:P ratio of 17, which is within the ranges that allow organic
materials to decompose easily and mineralize. By contrast, the C:N and
C:P ratios of RCW were 51 and 535, respectively. These values were
much higher than those considered as optimum for easy decomposition
of organic materials.
The effect of LC on nutrient supply in the soil was much less during
the second cropping compared to the first (Table 27.5). However, soil
levels of N, P and K associated with the LC treatments were still greater
than those observed for the control. The leaf concentration of these
nutrients was even lower (Tables 27.3 and 27.4). Interestingly, tomato
growth was not affected by this negative nutrient trend. This is possibly
because, despite the negative nutrient trend, the leaf N concentration did
not drop below the critical level of 1.2%. These results conclusively indicate
396
Soumare, M.D. et al
that the litter compost of C. equisetifolia, unlike its ramial chipped wood,
releases most of its nutrients soon after incorporation into soil, and that
at least for the first two crops, it can be used as the sole source of nutrients
for tomatoes. More work is needed to establish the residual effects of this
amendment as well as the additional amounts that may have to be applied
regularly in order to maintain yields at reasonable levels. Nevertheless, it
is clear from the results of the present study that the litter compost of C.
equisetifolia can be effectively used as a source of nutrients for tomato
and possibly other vegetables in the Niayes area. If this amendment is
adopted for regular use in the area, it will be necessary to determine
sustainable levels of the litter harvesting taking into account its local
contribution to nutrient cycling in the plantation ecosystem.
Effect on soil physical and chemical properties
At the final harvest for the second crop, soil was sampled and analyzed
for organic matter content, cation exchange capacity, water holding
capacity and bulk density. The results obtained are shown in Table 27.6.
Table 27.6: Effect of RCW and litter compost on selected soil properties after the second
harvest
Treatment
Control
RF
RF+N
RCW10 t ha-1
RCW20 t ha-1
RCW40 t ha-1
LC10 t ha-1
LC20 t ha-1
LC40 t ha-1
LSD (0.05)
OM(%)
CEC (cmol(+) kg -1)
WHC (mm m -1)
1.93e
4.38b
4.39b
2.80e
2.68e
2.75e
3.94d
4.16c
4.85a
1.3
5.2h
7.5d
7.2e
5.8g
6.1f
5.8d
10.1c
15.1b
20.4a
0.19
99.5c
103.7c
102.5c
125.0b
127.5b
125.5b
131.5b
148.7a
143.2a
10.1
DBD (Mg m -3)
1.60a
1.39bc
1.39bc
1.40bc
1.39bc
1.47b
1.34cd
1.30cd
1.31cd
0.08
*The recommended fertilizer contained also horse manure (20 t ha -1)
** Means in each column followed by the same letter are not significantly differen at p
0.05 according to the LSD test
The incorporation of RCW did not increase the amount of organic
matter in soil after the second harvest. However, LC application increased
the amount of organic matter in soil significantly relative to the control
and RF treatments. The organic matter increases observed in plots
treated with composted litter could be attributed to its advanced state
of decomposition due to composting, which increased the proportion of
Effects of Ramial Chipped Wood and Litter Compost of Casuarina Equisetifolia
Forst & Forst on Tomato Growth and Soil Properties in Niayes, Senegal
397
oxidizable organic matter in the soil. Ramial chipped wood was not
substantially decomposed in the soil and consequently did not increase
soil organic matter levels.
The cation exchange and water holding capacities of the soil were
increased by the application of RCW and LC (Table 27.6). However, the
effects of RCW on these parameters were much smaller compared to
LC. This was consistent with the observed greater effect of LC to increase
organic matter levels in the experimental soil.
All amendments reduced soil bulk density significantly relative to
the control but litter compost treatments resulted in the lowest bulk
density values (Table 27.6). The significantly lower soil bulk density
values in plots treated with LC could be linked to the higher proportion
of decomposed organic residues introduced by this amendment. In a
non-aggregated soil such as the one used in the present study, it is
likely that any change in bulk density was due to the effect of soil
amendments. In this respect it was primarily due to polysaccharides
present in the decomposing amendments and the resulting humus,
which are able to cement soil particles together (Mustin, 1981), resulting
in improved aggregation and subsequent reduction in soil bulk density.
Conclusions
The application of RCW depressed tomato growth and yield during the
first cropping. This was attributed to the effect of RCW to induce intense
N immobilization in the soil due its wide C:N ratio, which resulted in
reduced N uptake by tomato plants. Improvements in tomato growth
and yield were observed during the second cropping and this was
ascribed to improved nutrient release, especially nitrogen, from RCW
following its extended incubation in the soil resulting in the
mineralization of earlier immobilized nutrients. These results suggested
that in order to derive short-term benefits from RCW application it may
have to be applied in combination with experimentally determined
amounts of mineral fertilizers. A longer-term investigation is, however,
necessary to establish the long-term effects of this amendment on the
productivity of the experimental soil.
Incorporation of LC resulted in improved tomato growth and
production as reflected by increased tomato height and yield relative
to the absolute control and recommended fertilizer (RF) treatments
during the two croppings. The positive effect of LC to improve soil and
tomato uptake of N, P and K, and possibly other nutrients that were
not measured by tomato, was associated with the narrower C:N ratio
of the composted litter. The observed effects were greater in the first
than the second crop indicating that LC had limited residual nutrient
value.
398
Soumare, M.D. et al
Application of LC also improved soil organic matter content and
water holding capacity; and reduced the soil dry bulk density suggesting
that its regular use could result in the long-term improvement of the
productivity of the experimental site.
Acknowledgements
The lead author is grateful to Winrock International for sponsoring this
research as part of her M.Sc. studies at the University of Fort Hare.
References
Bartholomew, W.V. (1965) Mineralization and immobilization of nitrogen in the
decomposition of plant and animal residues. Soil Nitrogen. Agronomy
10: 285-290.
Blake, G.R. and Hartge, K.H. (1986) Particle Density. In A. Klute. Methods of
Soil Analysis. Physical and mineralogical Methods. Agronomy no. 9 Part 1
2nd Edition. 377-382. Soil Science Society of America, Madison, Wisconsin,
U.S.A.
Brady, N.C. and Weil, R.R. (1999) The nature and properties of soils. 12th Edition.
466-489.
Bremner, J.M. and Mulvey, C.S. (1982) Nitrogen-Total. In Pages A. L. Methods
of Soil Analysis. Chemical and Microbiological Properties. Agronomy no. 9
Part 2. 2nd Edition. 595-624. Soil Science Society of America, Madison,
Wisconsin, U.S.A..
Cassel, D.K. and Nielsen, D.R. (1986) Field capacity and available water capacity.
In methods of Soil Analysis. Physical and mineralogical Methods. Agronomy
no. 9. Part 1. 2nd Edition. 901-926. Soil Science Society of America, Madison,
Wisconsin, U.S.A.
Engelstad, O.P, Jugsujinda and De Datta, S. (1974) Response by floded rice at
phosphate rocks varying in citrate solubility. Soil Sci. Soc. Am. Proc .
38: 524-529.
FAO-UNESCO-ISRIC (1990) Soil map of the world. Revised legend. Reprinted
with corrections. World Soil Resources Report no. 60. Food and Agriculture
Organization of the United Nations, Rome Italy.
Foth, H.D. and Ellis, B. (1988) Soil fertility. Edt Wiley & Son. New York. pp 212.
Gee, G.W. and Bauder, J.W. (1986) Particle size analysis. In Methods of soil
Analysis. Chemical and Microbiological Properties. Agronomy no. 9. 2nd
Edition. 383-411. Soil Science Society of America, Madison, Wisconsin,
U.S.A.
Gomez, K.A. and Gomez, R.A. (1984) Statistical Procedures for Agricultural
Research. Second Edition. 657 pp.
Effects of Ramial Chipped Wood and Litter Compost of Casuarina Equisetifolia
Forst & Forst on Tomato Growth and Soil Properties in Niayes, Senegal
399
Knudsen, D., Peterson, G.A. and Pratt, P.F. (1982) Lithium, sodium and
Potassium. In Pages A..L. Methods of Soil Analysis. Chemical and
Microbiological Properties. Agronomy no. 9 Part 2 2nd Edition. 225-246.
Soil Science Society of America, Madison, Wisconsin, U.S.A.
Lanyon, L.E. and Heald, W.R. (1982) Magnesium, Calcium, Strontium and
Barium. In Methods of Soil Analysis. Chemical and Microbiological Properties.
Agronomy no. 9 Part 2, 2nd Edition. 247-262. Soil Science Society of America,
Madison, Wisconsin, U.S.A.
Lemieux, G. (1993) Le bois raméal fragmenté et la méthode expérimentale: une
voie vers un institut international de pédogénese» In les actes du quatrieme
collogue international sur les bois raméaux fragmentés. Edité par le groupe
de coordination sur les bois raméaux. Departement des Sciences Forestières.
Université Laval. Quebec. 124-138.
Maheut, J. and Dommergues, Y. (1963) Evaluation du taux de fixation d’azote
dans un sol dunaire reboise en filoa. Agrochimica 7: 335-340.
Mailly, D. and Margolis, H. A. (1992) Forest floor and mineral soil development
in Casuarina equisetifolia plantation on the coastal sand dune of Senegal.
For. Ecol. and Manage. 55: 259-278.
Mclean, E.O. (1982) Soil and pH and lime requirements. In Pages A. L. Methods
of Soil Analysis. Chemical and Microbiological Properties. Agronomy no. 9.
Part 2. 2nd Edition. 199-224. Soil Science Society of America, Madison,
Wisconsin, U.S.A.
Mustin, M. (1981) Le compost. Editor Francois Dubusc. Paris, France.
N’dayegamiye, A. and Dube, A. (1986) L’effet de l’incorpporation de matières
ligneuses sur l’évolution des propriétés chimique du sol et sur la croissance
des plantes. Can. J. Soil Sci. 66: 623-631.
Nelson, D.W. and Sommers, L.E. (1982) Total carbon, organic carbon and organic
matter. In Methods of Soil Analysis. Chemical and Microbiological Properties.
Agronomy no. 9. Part 2. 2nd Edition. 539-579. Soil Science Society of
America, Madison, Wisconsin, U.S.A.
Netscher, P. (1970) Les nématodes parasites des cultures maraichères au Sénégal.
Cah. ORSTOM Ser. Biol. 11: 209-283.
Okalebo, J. R., Guathua, K. W. and Woomer, P. L. (1993) Laboratory methods
and Plant Analysis. Working manual. Karhsobf, Nairobi Kenya. pp 55.
Olsen, S.R. and Sommers, L.E. (1982) Phosphorus. In Soil Testing and Plant
Analysis. Book series no.3. Third edition. SSSA, Madison, Winconsin, USA.
pp: 403-430. Soil Science Society of America, Madison, Wisconsin, U.S.A.
Rhoades, J.D. and Miyamoto, S. (1990) Testing Soil for salinity. In Soil Testing
and Plant Analysis, 3rd Edition. SSSA Book series no 3:299-336.
Rustad, L.E. and Cronan, C. (1988) Element loss and retention during litter
decay in a red spruce stand in Maine. Can. J. For. Res. 18: 947-953.
Seck, A.. et Lemieux, G. (1993) Essai de fertilisation avec le boi raméal fragmenté
dans les cuvette de Niayes: In les actes du quatrieme collogue international
sur les bois raméaux fragmentés. Edite par le groupe de coordination sur
les bois raméaux. Departement des Sciences Forestieres. Universite Laval.
36-41.
400
M.
D. Soumare
Soumare,
M.D. et al
Sylvestre, A. and Despatie, S. (1995) Effet des bois raméaux fragmenté sur la
croissance et les rendements du mais en zone de transition foret-savane de
la Cote d’Ivoire, 5pp.
Tremblay, J. and Beauchamp, C.J. (1998) Fractionnement de la fertilisation
azotée d’appoint a la suite de l’incorporation au sol de bois raméaux
fragmentés : modifications de certaines propriétes biologiques et chimiques
d’un sol cultivé en pomme de terre. Can. J. Soil Sci. 78: 275-282.
401
The Use of Pigeon Pea (Cajanus Cajan) for Amelioration of Ultisols in Ghana
The Use of Pigeon Pea
(Cajanus cajan) for
Amelioration of Ultisols in
Ghana
28
Yeboah, E.*, J.O. Fening and
E.O. Ampontuah
Soil Research Institute Academy Post Office, Kwadaso,
Kumasi, Ghana
* Corresponding author: soils@africaonline.com.gh
Abstract
Pigeon pea, a multi-purpose species, is extensively used as
food grain and green manure crop for soil fertility
amelioration in local cropping systems. Recently, pigeon
pea root exudates have been found to contain phenolic
compounds (e.g. piscidic acid), which chelate Fe to free P in
Fe bound P in soils for crop uptake. It is also reported that
pigeon pea root exudates dissolve phosphate-containing
rocks (e.g. phosphate rocks) to make P available for crop
use. There are however, a few instances in West Africa where
the use of pigeon pea has become unpopular among farmers
due to its low and variable yield as well as its inability to
redress soil fertility sufficiently in the long-term. In this
study, the nutrient cycling, moisture storage of pigeon pea
collections is reported.
Key words: pigeon pea, nutrient cycling, moisture storage
402
Yeboah, E. et al
Introduction
Crop production in peasant cropping systems in the semi-arid areas of
West Africa is generally constrained by low and uncertain rainfall, poor
soil fertility (low nutrients content and structural degradation), lack of
credit facilities to purchase inputs such as fertilizers and improved
varieties (Weber et al., 1996; Bationo et al., 1993). Of these, phosphorus
and nitrogen usually limit crop yields on farm fields. Under favourable
soil and weather conditions, yields of improved maize ranged from 4.0
to 7.0 t ha-1 (Elemo et al., 1990). However, yields of improved varieties in
nutrient and moisture-stressed conditions, typical of farmers’
circumstances ranged from 0.30 to 4.0 t ha-1 (Carsky et al., 1998). Crop
yields will continue to decline in so far as appropriate remedial measures
are not put in place to conserve moisture and restore soil fertility (Mermut
and Eswaran, 1987).
Scientists in West Africa and elsewhere in the tropics, have developed
biological management practices which have the potential to address
the problem of low soil productivity in the region (Peoples and Craswell,
1992; Sanchez and Salinas, 1981).
Corrective measures that have been developed by local and
international research groups to address soil fertility related problems
include:
1) the use of organic and inorganic mineral fertilizers (Smyth et al.,
1993; Manu et al., 1988),
2) intercropping of cereals and legumes (Sanginga et al., 1996),
3) legume based cropping systems e.g. herbaceous green manuring,
agroforestry, improved fallow (Mafongoya et al., 1997; Barnes, 1995)
4) crop residue mulch management (Tian et al., 1993; Adeoye, 1984;
De Vleeschauwer et al., 1980);
5) the use of local rock phosphate (Ankomah et al., 1995; Zapata and
Axmann, 1991; Hammond et al., 1986 and
6) integrated soil husbandry comprising a combination of the above
(Mugwira and Mukurumbira, 1984).
These options were designed to increase nutrient use efficiency, make
the environment less harmful as well as to reduce costs of production.
Pigeon pea, a multipurpose species, is extensively used as food grain
and green manure crop for soil fertility amelioration in cropping systems
(Adu-Gyamfi et al., 1996; Tobita et al., 1994). Pigeon pea root exudates
have been found to contain phenolic compounds (e.g. piscidic acid),
which chelate Fe to free P in Fe bound P in soils for crop uptake
(ICRISAT, 1999). It has also been reported that pigeon pea root exudates
dissolve phosphate-containing rocks (e.g. phosphate rocks) to make P
available for crop use (Ae et al., 1990). There are however, a few instances
The Use of Pigeon Pea (Cajanus Cajan) for Amelioration of Ultisols in Ghana
403
in West Africa where the use of pigeon pea has become unpopular among
farmers due to its low and variable yield as well as its inability to redress
soil fertility sufficiently in the long-term (Juo et al., 1996).
The objective of the study is to validate the hypothesis that cultivation
of pigeon pea results in nutrient contribution to the soil.
Methodology
Site characterisation
The field study was carried at the Soil Research Institute experimental
field at Kwadaso, Kumasi (6 o40’N, 1 o4W; 255 m above sea level) (Soil
Survey Staff, 1990) in the semi deciduous forest zone of Ghana. The
mean annual rainfall in the area is 1473 mm per annum; the rainfall
pattern is bimodal, the rainy season starts in March and ends in October,
with a short dry spell in August with peaks in June and September. The
soil is classified as Ferric Acrisol (FAO-UNESCO, 1990).
Soil sampling
Composite samples from the 0-20 cm depth were taken from the
experimental sites following the method of Anderson and Ingram,
(1993) The samples were transported to the laboratory at the Soil
Research Institute, Kumasi and air-dried. Un-decomposed plant
materials were sorted out and the samples crushed to pass a 2-mm
sieve. The sieved soil samples were stored in thick polythene bags for
laboratory analyses.
Soil analyses
Soil Particle size distribution was determined by the modified Bouyoucos
hydrometer method as described by Day (1965). Soil pH was determined
in distilled water using a Glass electrode-calomel electrode (McLean,
1982), MV Pracitronic pH meter at a soil solution ratio of 1:1. Organic
carbon was determined by the method of Bremner (1965) and soil
available phosphorus was by Bray 1. Exchangeable Ca and Mg in the
extract were determined by an atomic absorption spectrophotometer.
The K and Na were determined by flame photometry. The effective cation
exchange capacity (ECEC) was calculated as the sum of the exchangeable
potassium, calcium, magnesium and sodium. All analyses were carried
out in duplicate.
404
Yeboah, E. et al
Pot experiment
Three seeds of each collection of pigeon pea were sown into pots
containing 5.0 kg soil. The treatments were replicated four times and
pots were arranged in a completely randomized design. Seedlings were
thinned to two per pot one week after germination. The moisture contents
in the pots were kept at filled capacity throughout the experimental
period with demineralized water. Plants were harvested 36 days after
planting and the above-ground plant material as well as the belowground material washed in distilled water. Harvested plant materials
were oven-dried and weighed.
Establishment of pigeon pea for field study
The study site was a three- year fallow field with Chromolaena odorata
as the dominant weed. The site was hand cleared with cutlass and the
thick biomass was burnt thereafter.
Nine (9) pigeon pea cultivars of 90 % germination were planted on
13th June 2000 at two seeds per hill. The collections were:
1) 82/492;
2) 82/472;
3) GJ 93/207;
4) 82/137;
5) 82/021;
6) 82/433;
7) 82/491;
8) 82/486 and
9) 82/481.
These collections represent local pigeon pea germplasm from Ghana
collected by the Plant Genetic and Research centre (PGRC, Bunso of
the Council for Scientific and Industrial Research, (CSIR). The collections
belong to the family Papilionaceae and the scientific name is Cajanus
cajan and the common name is pigeon pea. The locations where
collections were made are shown in Table 28.1.
Table 28.1: Some basic information of the pigeon pea cultivars
Collection number
82/492
82/472
GJ 93/207
82/137
82/021
82/433
82/491
82/486
82/481
Locality
Vernacular name
Source of sample
Gomoa Akropong
Ejura
Tanoboase
Kwahu Tafo
Norre
Bonuntong
Aboabo
Ayere
Kontonso
Adua
Asedua
Akye
Adua
Adua
Adua
Ase
Ase
Ase
Field
Field
Farm store
Farm store
Farm store
Farm store
Farm store
Farm store
Farm store
The Use of Pigeon Pea (Cajanus Cajan) for Amelioration of Ultisols in Ghana
405
Experimental design for the field study.
The experimental design was a randomised complete block with four
replications. Plot size was 4 m x 4 m and the distance between plants
were 1 m apart.
Trial maintenance
Efforts were made to ensure 100 % seed establishment by refilling
withered seedlings. Weeding was done mechanically with a hoe as often
as was necessary.
N and P were applied at the rate of 50 kg N ha -1 and 20 kg P ha -1
respectively, to each treatment four weeks after planting.
Results and Discussion
Table 28.1 shows locations in Ghana where pigeon pea commonly grown
were collected. The locations indicate the adaptability of the crop across
the major agro-ecological zones of the country. The collections vary in
seed colour and have benefited very little from morphological
characterisation.
Table 28.2: Shoot and root dry matter yield of pigeon pea collections.
Collection
Shoot Dry matter
yield g pot -1
Root dry matter
yield g pot -1
82/492
82/472
GJ 93/207
82/137
82/021
82/433
82/491
82/486
82/481
LSD (0.05)
S.E
4.49
3.96
2.96
2.49
3.28
5.07
4.41
4.98
2.65
NS
1.16
1.14
0.73
0.55
0.46
0.54
0.77
0.58
0.73
0.43
NS
0.22
The production of above-ground as well as below-ground biomass
from the pot study did not differ significantly among the collections
(Table 28.2). Again all the collections showed few tiny ineffective nodules
at harvest.
406
Yeboah, E. et al
Table 28.3: The effect of pigeon pea cultivation on some soil properties after one year of
cultivation.
Soil properties
pH 1:1 H20
Organic carbon (%)
% Nitrogen
C/N ratio
Exchangeable calcium (ppm)
Exchangeable magnesium (ppm)
Exchangeable potassium (ppm)
Exchangeable sodium (ppm)
Total exchangeable bases
ECEC (C.mol/kg)
Available P (ppm)
Available K (ppm)
Bulk density (g/cm 3)
Moisture content(g/g)
% sand
% silt
% clay
Uncultivated
soil
Pigeon pea
cultivated soil
Standard
Error
4.96
1.98
0.23
8.5
5.92
1.0
0.16
0.02
7.10
7.2
2.99
57.88
1.28
16.1
38.37
38.88
20.00
4.73
1.93
0.23
8.5
5.2
2.08
0.17
0.03
7.47
7.58
2.21
77.88
1.40
20.43
41.13
46.25
15.38
0.11
0.06
0.16
0.39
0.42
0.01
0.43
0.43
0.28
5.68
0.03
1.44
1.44
1.57
1.15
Table 28.3 shows the properties of the 0-20 cm layer of soils under
pigeon pea and in the uncultivated sites after one year of cultivation.
The data indicated a decline in soil pH with cultivation. The mean organic
carbon content of the pigeon pea sites is about 2.5 % lower than the
mean of the uncultivated sites. There are no differences between the
uncultivated soil and the cultivated soil with respect to the levels of
total nitrogen and C/N ratio.
In general with the exception of exchangeable calcium which was
almost 12 % more in the uncultivated site compared to the pigeon pea
cultivated sites, exchangeable cations were higher in the pigeon pea
cultivated sites than in the uncultivated sites. Exchangeable magnesium
was more than 100 % higher in the pigeon pea cultivated sites compared
to the uncultivated sites. Similarly, total exchangeable bases were about
5 % higher in the cultivated sites compared to the uncultivated sites.
The cation exchange capacity of the pigeon pea cultivated sites increased
by 5 % with respect to the uncultivated site within one (1) year. Also,
available potassium increased by 25 % in the cultivated site. Available
phosphorus, however, declined by almost 26 % in the cultivated sites
with respect to the uncultivated site. The texture of the soil is loam in
both the uncultivated and the cultivated sites. With the exception of
percentage clay , pigeon pea cultivated site indicated a higher % particle
size than the uncultivated sites. More moisture was stored under pigeon
The Use of Pigeon Pea (Cajanus Cajan) for Amelioration of Ultisols in Ghana
407
pea cultivated site which also showed a higher bulk density (1.40 g cm), than the uncultivated site (1.28 g cm-3).
The decrease in soil pH with cultivation could be attributed to erosion
of the exposed sites prior to canopy closure of pigeon pea. It is possible
that the amount of N derived from biological nitrogen fixation was utilised
by the cultivated crop rather than soil N. Pigeon pea like all legumes,
has a high phosphorus need and its performance can be affected by low
soil P as is the case in most tropical soils. Under the present management
of low fertilizer inputs, P shows a decline with cultivation. Soils under
fallow however, had high levels of P. This is a reflection of the biological
biochemical mineralization processes during which organic matter is
mineralized. It is also a reflection of biocycling of P through deeper plant
roots causing a relative enrichment in the topsoil (Barber, 1979).
Considering the fact that moisture storage improved under pigeon
pea cultivated sites, the crop could be considered as a possible candidate
in a moisture-stressed agro-ecological environment.
3
Conclusion
The one year study clearly demonstrated that there is a high potential
for pigeon pea as a good vehicle for magnesium, potassium and sodium
cycling in soils. The most valuable information for management decisions
is the phosphorus mining under pigeon pea cultivation, where seed
grain is harvested. Given the high cost of chemical fertilizers and the
fact that most peasant farmers cannot afford their purchase, the need
to test the efficiency of other sources such as phosphate rocks with
pigeon pea based farming systems is great.
Acknowledgements
The work reported here was made possible through grants from Prof.
S.K.A. Danso, The Director, Ecological Laboratory, University of Ghana,
Legon.
References
Adu-Gyamfi, J.J., Katayama, K., Gayatri, D., Rao, T. P. and Oto, O. (1996)
Improvement of fertilizer and nitrogen use efficiency in intercropping. In
Ito, O.
Johansen, C., Adu-Gyamfi, J.J., Katayama, K., Rao, J.V.D.K and Rego, T.J
(eds.)
408
Yeboah, E. et al
Dynamics of roots and nitrogen in cropping systems of the semi-arid tropics. pp
493-506. Japan International Center for Agricultural Sciences (JIRCAS),
JIRCAS International Agriculture Series No. 3. Tsukuba
Adeoye, K.B. (1984) Influence of grass mulch on soil temperature, soil moisture
and yield of maize and gero millet in a savanna zone soil. Samaru Journal of
Agriculrural Research 2: 87-97.
Ae, N, Arihara, J., Okada, K., Yoshihara, T. and Johansen, V. (1990) Phosphorus
uptake by Pigeonpea and its role in cropping systems of the Indian
subcontinent. Science 284: 477-480.
Anderson, J.M and Ingram, J.S.I. (1993) Tropical Soil Biology and Fertility: A
handbook of Methods. CAB International, Wallington, U.K
Ankomah, A..B, .Zapata, F., Danso, S.K.A. and Axmann. H. (1995) Cowpea
varietal
differences in uptake of phosphorus from Gafsa rock in a low-P Ultisol. Fertilizer
Res. 41: 219-225.
Barnes, P. (1995) Dry matter of herbage productivity and aspects of chemical
composition in four forage shrub species at a semihumid site in Ghana.
Agroforestry Systems 31: 223-227.
Bationo, A., Christianson, C.B. and Klaij, M.C. (1993) The effect of crop residues
and fertilizer on use of pearl millet yields in Niger. Fertilizer Research
34: 251-258.
Bremner, J.M. (1965) Total nitrogen. In: Methods of soil Analysis Part 2. Chemical
and Microbiological properties. (eds) Black, C.A. Agronomy Monograph 9.
Pp 1149-1178.
Carsky, R.J., Oyelwole, B. and Tian, G. (1998) Integrated soil management for
the savanna zone of West Africa: legume rotation and fertilizer N. Nutrient
Cycling in agroecosytems 55: 95-105
Day, P.R. (1965) Particle fractionation and particle size analysis. In C.A. Black
ed., Methods of Soil analysis, 545-67. Madison, WI: American Society of
Agronomy.
De Vleeschauwer, D., Lal, R. and Malafa, R. (1980) Effects of amount of surface
mulch on physical and chemical prpperties of an Alfisol from Nigeria. Journal
of Science food and agriculture 31: 730-738.
Elemo, K.A, Kumar, V., olukosi, J.O. and Ogunbile, A.O. (1990) Review of research
work on mixed cropping in the Nigeria savanna. Samaru Miscellaneous
Paper No. 127. IAR, Samaru, Nigeria.
FAO-UNESCO. 1990. Soil map of the world. Revised Legend. F.A.O, Rome. 119
pp.
Hammond, L.L., Chien, S.H. and Mokwunye, A.U. (1986) Agronomic value of
unacidulated and partially acidulated phosphate rocks indigenous to the
tropics. Adv. Agro. 40: 89-140.
ICRISAT, (1999) 15 Years of Progress: Government of Japan Project at ICRISAT
1984-99 (Adu-gyamfi, J.J., ed.). 1-2 Ohwashi, Tsukuba, Ibaraki 305-9851,
Japan and Patancheru 502 324, Andhra Pradesh, JIRCAS and ICRISAT.
32pp.
The Use of Pigeon Pea (Cajanus Cajan) for Amelioration of Ultisols in Ghana
409
Mafongoya, P.L, Nair, P.K., Dzowela, B.H. (1997) Mutipurpose tree prunings as
a source of nitrogen to maize under semiarid conditions in Zimbabwe.1.
Nitrogen recovery rates in relation to pruning quality and the method of
application. Agrofor. Syst. 35: 31-46.
Manu, A, Geiger, S.C. and Bationo, A. (1988) The interactive effect of P fertilizers
and millet residue on P dynamics of a sandy Nigerian soil. Agronomy
Abstracts p 59 Amerrican Society of Agronomy.
McLean, E.O. (1982) Soil pH and lime requirement. Pp 199-224. In Page, A.L.,
R.H. Miller and D.R. Keeney (eds) 1982. Methods of soil analysis. Part 2
(2nd edition). Chemical and Microbiological Properties- Agronomy Monograph
no. 9
Mermut, A.R., and Eswaran, H. (1997) Opportunity for soil science in a milieu
of reduced funds. Can. J. Soil Sc. 77: 1-7.
Mugwira, L.M and Mukurumbira, I.M. (1984) Comparative effectiveness of
manures from the communal areas and commercial feedlots as plant nutrient
sources. Zimbabwe Agriculture Journal 81: 241-250.
Peoples, M.B and Craswell, E.T. (1992) Biological nitrogen fixation: investments,
expectation and actual contribution to agriculture. Plant and Soil
141:13-40.
Sanchez, P.A. and Salinas, J.G. (1981) Low input technology for managing Oxisols
and Ultisols in Tropical America. Advances in Agronomy 34: 279-406.
Sanginga, N., Okogun, J.A., Akonbundu, I.O.and Kang, B.T. (1996) Phosphorus
requirement and nodulation of herbaceous and shrub legumes in low P
soils of a Guinea Savanna in Nigeria. Applied Soil Ecol. 3: 247-255.
Smith, A.J., Dumanski, J., Spendjian, G., Swift, M.J. and Thorton, P.K (1993)
FESLM: An international framework for evaluating sustainable land
management. A discussion paper. FAO, Rome.
Soil survey Staff. (1990) Keys to Soil Taxonomy USDA. SMSS Tech. Mon. No. 19
(4edn) Blacksburns, Virginia, USA: SMSS.
Tian, G, Kang, B.T. and Brussaard, L. (1993) Mulchiong effect of plant residues
with chemically contrasting compositions on maize growth and nutrients
accumulation. Pl.Soil. 153: 179-187.
Tobita, S, Ito, O., Matsunaga, R., Rao, T. P , Rego, T. J., Johansen,C. and
Yoneyama, T. (1994) Field evaluation of nitrogen fixation and use of nitrogen
fertilizer by sorghum/pigeon pea intercropping on an Alfisol in Indian semiarid tropics. Biol.Fert. Soils 17: 241-248.
Weber,G., Smith, J. and Manyong, V.M. (1996) System dynamics and the
definition of research domains for the northern Guinea Savanna of West
Africa. Agriculture, Ecosystems and Environment. 57: 133-148.
410
Yeboah, E. et al
Assessment of Biomass Transfer from Green Manure to Soil Macrofauna in Agroecosystem
Assessment of Biomass
Transfer from Green Manure
to Soil Macrofauna in
Agroecosystem-Soil
Macrofauna Biomass
411
29
Ayuke, F.O.1, 4, Rao, M.R.2, 3, Swift, M.J.4
and Opondo-Mbai, M.L.1
1
Department of Forestry, Moi University, P.O. Box 1125,
Eldoret, Kenya
2
International Centre for Research in Agroforestry, P.O. Box
30677, Nairobi, Kenya
3
ICRISAT Colony (plot II, phase 1), Secunderabad-500 009,
Andhra Pradesh, India
4
Tropical Soil Biology and Fertility Programme, Institute of
CIAT, P.O. Box 30677, Nairobi, Kenya
Corresponding address: P.O. Box 30677, Nairobi,
Tel. 524000, Ext 4771,
Email: Fayuke2002@yahoo.co.uk; Fayuke@cgiar.org
Abstract
During 1997 short rains (Oct 1997-Feb 1998), a study was
undertaken to assess how biomass transfer within
agroecosystem influence soil biodiversity (soil macrofauna
biomass).
This was part of a larger experiment conducted to test
the hypothesis that diversity, abundance and function of
412
Ayuke, F.O. et al
soil invertebrate fauna are related to the quality of organic
residues used. Leaf biomass of tithonia (Tithonia diversifolia
[Hemseley) A. Grey) biomass and senna (Senna spectabilis
D.C. & H.S. Irwin) biomass at 5 t ha -1 dry weight were
incorporated into the soil and these were compared with
the control without any input and fertilizer at 120 kg N,
150 kg P and 100 kg K ha -1 from urea and triple super
phosphate (TSP). Macrofauna biomass (fresh weight), was
monitored in soil monoliths (25cm x 25cm x 30cm) at the
beginning of the season, six weeks after sowing maize and
at maize harvest.
Addition of organic residues increased faunal biomass
substantially over the fertilized and unfertilized controls.
Whereas senna increased total biomass by 45% and tithonia
by 49%, the two organic residues did not differ significantly
between them. Addition of either senna or tithonia
significantly increased earthworm biomass by 390% over
no input control. Even though termite biomass increased
by 160% in senna and 120% in tithonia over no input
control, F test was not significant because of high variability
between replications of the same treatment. Fertilizer use
did not change biomass of termites and earthworms.
This study shows that:
(1) addition of organic residues significantly increase
faunal biomass indicating a likelihood that soil
invertebrate functions can be manipulated by external
inputs of organic residues
(2) under arable land use system characterized by low
amount, range and diversity of food resources, quality
of organic residues do not play a significant role in
influencing foraging behaviour of soil invertebrates. It
therefore remains to be demonstrated whether mixing
litter of organic residues of different quality may change
this foraging behaviour and consequently the
invertebrate functions in agroecosystem.
Key words: Biomass transfer, macrofauna, biomass, earthworms,
termites
Introduction
Soil fauna comprises a large variety of organisms with contrasted sizes
and adaptive strategies. Their abundance, and composition, hence
impact on soil processes vary greatly depending on vegetation and land
Assessment of Biomass Transfer from Green Manure to Soil Macrofauna in Agroecosystem
413
use practices (Lavelle et al., 1994a). Management practices such as
continuous tillage can cause alterations in the population structure,
elimination or reduction of key species and in some cases extremely low
abundance or biomass (Dangerfield, 1993; Beare et al., 1997). These
negative effects created by management practices may last for years.
House and Parmelee (1985), found that soil arthropods and
earthworm densities were higher under no tillage than in conventional
tillage practices, an expanded and beneficial involvement for this fauna
in crop residue decomposition processes. Studies conducted by Brown
et al. (1996) showed that under agro-ecosystem, earthworms were the
most dominant organism in terms of biomass, while in terms of numbers,
ants and termites predominated. The faunal biomass was however low,
compared with other tropical sites. In terms of diversity of faunal groups,
they found that natural sites were richer than cultivated sites.
Dangerfield (1993) found similar results and asserted that change of
natural forest, for instance, to arable agriculture resulted in a dramatic
decrease in faunal biomass, and diversity and a shift in dominance
from millipedes to beetle larvae and earthworms. The change in habitat
structure (removal of vegetation), the reduced range and abundance of
food resources and the more extreme climatic conditions at the soil
surface, combine to create an environment beyond the tolerance limits
of most soil animals (Dangerfield, 1993). Only those species that are
buffered from climatic extremes by building nests (e.g. termites) or living
in deeper soil layers (e.g. beetle larvae), are not immediately affected,
but may eventually suffer from the reduction in food resources (House
and Parmelee, 1985; Dangerfield, 1993; Tian et al., 1997). This explains
why a severe depletion of soil fauna has been observed in highly degraded
soils (Lavelle et al., 1994b).
Mafongoya et al. (1996), found that changes in microbial community
could be manipulated by applying prunings of different quality such
that processes of litter decomposition and nutrient dynamics are
enhanced. The major aim of the study was therefore to find out whether
through inputs of locally available organic residues of different quality
one could manipulate diversity, populations and ‘biomass’ of soil
invertebrate fauna in order to enhance nutrient cycling, improve soil
physical properties and also regulate decomposition processes.
Materials and Methods
Study site description
The study was conducted at Maseno (0°6' N, 34°35' E, and 1560 m
above sea level), in Vihiga District of western Kenya (Jaetzold and Smith,
1982). The area receives an average annual rainfall of 1800 mm in two
414
Ayuke, F.O. et al
rainy seasons; ‘long rains’ (March to July) and ‘short rains’ (September
to January). However during 1997 a total rainfall of 2037 mm was
received, with 1200 mm in the short rains because of El nino
phenomenon. Mean monthly temperature ranges between 14.6°C and
30.7°C. The soil at the experimental site was classified as Kandiudalfic
Eutrodox (USDA, 1992). At the start of the study, the field had the
following soil physical and chemical characteristics at 0-15cm and 1530 cm depths respectively: pH (1:2.5 soil water) 5.5, 5.5; organic carbon
(g kg -1 soil) 15.5, 14.5; extractable soil inorganic P (mg kg -1) 1.3, 0.9;
exchangeable calcium (cmolc kg-1) 4.03, 3.85; exchangeable potassium
(cmolc kg-1) 0.15, 0.13; clay (%) 41, 42; sand (%) 33, 33; silt (%) 26, 25;
porosity ranged between 50% and 60%. The soil is considered to be
moderately P fixing with a soil P concentration corresponding to 310
mg P kg-1 adsorbed by the soil (Nziguheba et al., 1998).
Experimental set up and management
The present study was superimposed on an on-going larger experiment
that was initiated in 1995 during the short rains season to evaluate six
organic tree and shrub residues (Tithonia diversifolia, Lantana camara,
Calliandra calothyrsus, Senna spectabilis, Sesbania sesban and Croton
megalocarpus), as sources of nutrients in comparison with inorganic
nutrients at six different N and P levels. The treatments were replicated
four times in a randomized complete block design in plots of 7.5 m wide
and 7 m long.
The study was conducted during the 1997 short rains in the following
treatments using maize as the test crop:
(1) Control: maize with no external inputs (Farmers’ practice),
(2) Maize + fertilizer input at: 120 kg N, 150 kg P and 100 kg K ha-1,
(3) Maize + fresh biomass of Tithonia diversifolia at 5 tonnes (dry weight)
ha-1 and
(4) Maize + fresh biomass of Senna spectabilis at 5 tonnes (dry weight)
ha-1.
The trial initially did not include “absolute control” (no N and P), so
a farmers’ no input control was randomly assigned to one of the
unutilized blank plots in each replication. The site was relatively flat
and there was no particular problem of runoff from plot to plot.
The amount of N and P added by the organic residues depends on
the chemical composition. Chemical composition was determined every
season at the time of application. All the selected material contained
fairly high N and P, but differed with respect to tannin, lignin, polyphenol
levels (Table 29.1). In the fertilized plots, 120 kg N ha-1 rate was chosen
as it is close to the total N applied for the different materials ranging
Assessment of Biomass Transfer from Green Manure to Soil Macrofauna in Agroecosystem
415
between 136 Kg N ha-1 to 183 Kg N ha-1. The rate is also sufficient to
overcome N limitation to maize growth in these soils. The choice of the
two residues (tithonia and senna) was based on:
1) the nutrient (N and P) concentration,
2) plant residue quality index (PRQI) (Tian et al., 1995) and
3) availability in the region for potential use by farmers.
The difference between the two test materials as measured by PRQI
has turned out to be much smaller than initially thought. However, the
experience of many researchers indicates that tithonia decomposes faster
than senna and represents high quality residues (Gachengo, 1999; Palm
et al., 2001). In western Kenya, particularly around Maseno area, farmers
grow tithonia as part of live fence around their farms to mark boundaries
or as hedges on contour. Senna spectabilis trees are also common. The
two residues were therefore readily available.
Table 29.1: Chemical composition and plant residue quality index (PRQI) of tithonia and
senna foliage
Plant residue
%N
%P
%Lignin
% Polyphenols
C/N ratio
PRQI(%)
Senna spectabilis
Tithonia diversifolia
3.3
3.5
0.21
0.28
9.0
9.0
1.03
3.20
10.89
10.10
10.26
10.59
Crop management
The entire field was tilled manually at the beginning of the season.
Tithonia and senna biomass were incorporated into the topsoil during
land preparation. The materials were collected a day before land
preparation. The required quantity of the fresh materials at 5 t ha -1,
was weighed (based on predetermined fresh weight to dry weight ratio;
16:1), and distributed uniformly on the ground before working into the
soil. Maize (hybrid 511), was sown on October 9, 1997 (a day after
incorporating treatments), at a spacing of 0.75m between rows and
0.25m between plants. Two seeds were placed into each hole, but the
crop was thinned to one plant per hole 14 days after emergence, during
first weeding. In the fertilizer treatment, the entire quantity of P as triple
super phosphate and K as muriate of potash and half quantity of N as
urea required for the plot were weighed and incorporated into the soil
during land preparation. The balance of N was top dressed one and half
months (42 days), after crop emergence during second weeding. In the
field, no plant protection for both pests and diseases was applied as the
study involved faunal observations.
416
Ayuke, F.O. et al
Macrofauna biomass assessment
Using a monolith unit of size 25 cm x 25 cm x 30 cm, samples were
taken at three periods during the season (Anderson and Ingram, 1993):
1) at the start of the experiment before the treatments were applied
(October 6, 1997),
2). six weeks after treatments were applied (November 19, 1997) and
3) at the end of the season (February 18, 1998).
At each observation, two samples were taken randomly from each
plot. The monolith was placed over a randomly selected spot and using
a metallic mallet, it was driven into the soil until it was level with the
ground. The soil from the monolith was removed by hand depthwise (010, 10-20 and 20-30 cm) into plastic buckets. The soil from each depth
was placed in different plastic trays (20 cm by 30 cm) and gently sorted
out to locate the animals. The animals were separated into major
taxonomic groups, recorded and then collected in glass and plastic
bottles using a pooter. In the laboratory, counting and weighing (for
biomass), were done. The fresh weight (in grams) determination took
place within 12 hours from the time of sampling. Biomass of different
category of animals was expressed per metre square (Anderson and
Ingram, 1993).
Data analyses
The data collected were subjected to analyses of variance (ANOVA),
to compare treatment effects on diversity, populations and biomass of
soil invertebrate fauna ANOVA was conducted using the GENSTAT 5
Committee (1993) statistical package. Where sampling was conducted
at different periods, the data were analyzed in a split-plot design
with the applied treatments as the main plot factor and sampling
period as the sub-plot factor. Treatment differences were evaluated
using the least significance difference (LSD) at P<0.05. Standard error
of difference of means (SED) was given.
Results
Total faunal biomass
Addition of organic residues increased faunal biomass substantially over
the fertilized and no input controls. Whereas senna treatment increased
total biomass by 45% and tithonia by 49%, the two organic residues did
not differ significantly.
Assessment of Biomass Transfer from Green Manure to Soil Macrofauna in Agroecosystem
417
Faunal biomass varied significantly over time between green manure,
fertilizer and no input control. At the beginning of the season, senna
and tithonia green manure treatments recorded 17% and 32% higher
faunal biomass than the no input control and 28% and 43% than
fertilized control, respectively. At six weeks after applying the materials,
senna treatment recorded 100% higher biomass than the fertilizer and
no input control treatments. Tithonia treatment recorded 96% higher
biomass than both fertilizer and no input control treatments (Figure
29.1). While the biomass in the fertilized and no input treatments
decreased continuously as the season progressed, it increased in the
green manure treatments by 48% for senna and 29% for tithonia at six
weeks stage and then declined to about 50% of the initial values at crop
harvest in both treatments (Figure 29.1).
Figure 29.1: Total biomass (fresh weight) of soil fauna in maize green manured with
organic residues compared with fertilized and unfertilized control at different periods
during 1997 short rains at Maseno, Western Kenya
140
120
SED
Beginning
6 weeks after
At harvest
100
80
60
40
20
0
Control
Fertilizer
Senna
Tithonia
Treatment
Earthworm biomass
ANOVA of earthworm biomass indicated significant differences due to
treatments, sampling period and interaction between them.
The average earthworm biomass across treatments was low at
2.1 g m-2 for both fertilizer and no input control, but addition of both
senna and tithonia green manures, significantly increased the biomass
418
Ayuke, F.O. et al
by five times. The two organic materials did not differ in their effect on
earthworm biomass. For no input control and fertilizer treatments, the
earthworm biomass was highest at the beginning of the season and it
decreased considerably in course of the season. Green manuring with
senna and tithonia increased the earthworm biomass by 100% and
72% respectively, at six weeks after applying the material, but the
biomass at final crop harvest was low similar to that in other treatments
(Table 29.2).
Table 29.2: Fresh weight of earthworms in maize green-manured with organic residues
compared with fertilized and unfertilized control at different periods during 1997 short
rains at Maseno, western Kenya
Sampling time
Treatment
Before
sowing
6 weeks after
sowing
At harvest
Mean
2.1 (1.9)b
2.1 (2.0)b
10.3 (3.7)a
10.3 (3.7)a
(weight g m -2)
Control
Fertilizer
Senna spectabilis
Tithonia diversifolia
9.8 (3.6)
8.5 (3.4)
21.6 (5.2)
8.0 (3.3)
0.5 (1.2)
0.4 (1.1)
10.7 (3.8)
13.8 (4.2)
0.3 (1.0)
0.7 (1.3)
3.0 (2.2)
1.7 (1.8)
Mean
11.4 (3.9)
4.3 (2.6)
1.2 (1.6)
SED (treatment)
SED (sampling time)
SED (interaction) 1
SED (interaction)2
(0.4)
(0.3)
(0.6)
(0.6)
F test: Treatment = p<0.001; Sampling time = p<0.001; Treatment sampling time =
p<0.001.
Means followed by the same letter within a column are not significantly different at 5%
level of probability. Values in parentheses are square-root { ( x + 0.5) } transformed.
SED (interaction)1 = Standard error of difference of means for sampling time in any
treatment.
SED (interaction)2 = Standard error of difference of treatment means at a given sampling
time.
Termite biomass
Addition of either fertilizer or organic materials, increased the biomass
of termites significantly compared with the no input control.
Assessment of Biomass Transfer from Green Manure to Soil Macrofauna in Agroecosystem
419
Table 29.3: Fresh weight of termites in maize green-manured with organic residues
compared with fertilized and unfertilized control at different periods during 1997 short
rains at Maseno, western Kenya.
Sampling time
Treatment
Before
sowing
6 weeks after
sowing
At harvest
Mean
0.5 (1.2)
0.5 (1.2)
1.3 (1.6)
1.1 (1.6)
(weight g m -2)
Control
Fertilizer
Senna spectabilis
Tithonia diversifolia
2.1 (1.9)
2.0 (1.9)
1.4 (1.7)
0.7 (1.3)
0.1 (0.8)
0.1 (0.8)
2.5 (2.1)
2.8 (2.2)
0.1 (0.8)
0.2 (0.9)
0.3 (1.1)
0.5 (1.2)
Mean
1.5 (1.7)
0.9 (1.5)
0.2 (1.0)
SED (treatment)
SED (sampling time)
SED (interaction)1
SED (interaction)2
(0.3)
(0.3)
(0.6)
(0.6)
F test: Treatment = NS; Sampling time = NS; Treatment sampling time = NS.
Values in parentheses are square-root { ( x + 0.5) } transformed.
SED (interaction)1 = Standard error of difference of means for sampling time in any
treatment.
SED (interaction)2 = Standard error of difference of treatment means at a given sampling
time.
NS = Not significant at 5% level of probability.
Termite biomass in fertilizer and no input control was highest at
the beginning of the season, which decreased to very low levels in course
of the season. In contrast to this, there was a 2-4 increase in termite
biomass six weeks after applying senna and tithonia green manures,
respectively. However, the biomass declined thereafter to similar levels
to the other treatments. Treatment differences were not significant
because of high variability among replicates.
Discussion
Microclimate, food resources and land use practices (e.g. pesticide
application, burning and clearing of land), are major factors affecting
the diversity, abundance and biomass of soil fauna communities (Warren
et al., 1987). Management practices such as continous tillage can cause
alterations in the population structure, elimination/reduction of key
groups and species of soil fauna and in some cases, low abundance or
420
Ayuke, F.O. et al
biomass (Dangerfield, 1993; Beare et al., 1997). Studies have shown
that cultivated sites are usually poorer than natural sites in terms of
faunal diversity and biomass (Brown et al., 1996).
Biomass of the soil fauna was low within the agroecosystem. This is
similar to results observed elsewhere in arable fields (Dangerfield, 1993;
Brown et al., 1996). In arable land use systems, the change in habitat
structure where vegetation are removed, reduced range and abundance
of food resources and the extreme climatic conditions at the soil surface
combine to create an environment beyond tolerance limits of most soil
fauna groups. The low diversity, abundance and biomass of the soil
invertebrate fauna observed, particularly in the no input control, typically
represent the status of soil fauna in the fields of resource poor farmers.
Most small-scale farmers clear and burn the land and rarely add external
inorganic inputs to the soil for nutrient replenishment. The implication
is that a change to continuous cropping decreases plant richness, thereby
reducing the diversity of food resources and residue quality. Studies
have shown that such changes in the land use systems lead to reduced
abundance, biomass and diversity of soil fauna communities (Warren
et al., 1987; Dangerfield, 1993).
Application of senna and tithonia residues increased the biomass of
the soil fauna groups for example earthworms. Studies have also shown
that addition of organic residues such as senna and tithonia increase
the faunal population by 100% over no input control (Ayuke, 2000).
Organic inputs such as crop residues, tree prunings and manures,
provide food to soil organisms. Greater faunal biomass in residue applied
treatments may be the result of a greater accumulation of organic matter.
Accumulation of organic matter from these residues (senna and tithonia),
may provide resource base for the invertebrates. Coleman et al. (2000)
states that soil organisms are strongly limited by available energy sources
and are in a state of starvation much of the time. The increased supply
of organic matter may possibly eliminate this state, in turn allowing
their consumers, i.e. earthworms and termites, to subsequently increase
in numbers hence increase in biomass. Surface applied residues preserve
soil water from evaporation, reduce soil temperature and provide
conducive niches for certain faunal groups. However, the insignificant
differences observed in faunal biomass between senna and tithonia,
could be due to reduced structural complexity and low diversity in food
brought about by changes in the arable land use system.
Conclusions
In intensively cropped and nutrient depleted soils such as the
Kandiudalfic-Eutrodox soil of this experiment, addition of organic
residues increase the faunal biomass, for example earthworms within
Assessment of Biomass Transfer from Green Manure to Soil Macrofauna in Agroecosystem
421
the cropping seasons. However under arable land use system
characterized by low amount, range and diversity of food resources and
the type and quality of organic residues do not play a significant role in
influencing foraging behaviour, hence biomass of soil invertebrates. Even
though faunal biomass was high in senna and tithonia treatments than
under fertilizer and no input controls, they did not significantly differ in
their effect of biomass. It therefore remains to be demonstrated whether
mixing litter of organic residues of different quality may change this
foraging behaviour resulting in increased biomass and consequently
the invertebrate functions in agroecosystem.
Acknowledgements
The lead author was supported by ICRAF-ANAFE Postgraduate
fellowship. The authors wish to thank Ms Eva Gacheru for managing
the main field experiment and Mr. Stephen Muoki for assistance in
faunal sampling.
References
Anderson, J.M. and Ingram, J.S.I. (1993) Tropical Soil Biology and Fertility: A
Handbook on Methods, CAB International, Wallingford UK.
Ayuke, F.O. (2000). Diversity, abundance and function of soil invertebrate fauna
in relation to quality of organic residues. M. Phil. thesis, Moi University,
Eldoret, Kenya.
Beare, M.H., Reddy, M.V., Tian, G. and Scrivasta, S.C. (1997) Agricultural
intensification, soil biodiversity and agroecosystem function in the tropics:
the role of decomposer biota. Applied soil biology, 6: 87-108.
Brown, G.G., Moreno, A.G. and Lavelle, P. (1996) Soil macrofauna under different
Agricultural Systems and native vegetation in four countries of East Africa.
Biological Management of Soil Fertility in Small-Scale Farming Systems in
Tropical Africa, 2nd Annual report.
Coleman, D.C., Hendrix, P.F., Beare, M.H., Cheng, W.X. and Crossley, D.A.
(2000) Microbial and faunal interations a sthey affect soil organic matter
dynamics in subtropical agroecosystems. In: Paoletti, M.G., Foissner, W.,
Coleman, D. (eds), Soil Biota, Nutrient Cycling, and Farming Systems. CRC
Press, Boca Raton, FL, pp. 1-14.
Dangerfield, J.M. (1993) Characterization of soil fauna communities. In: Rao,
M.R. and Scholes, R.J. (eds), Report on Characterization of an Experimental
Field in KARI Farm, Muguga, Kenya. ICRAF, Nairobi, Kenya, pp. 51-67.
Gachengo, C.N., Palm, C.A., Jama, B. and Othieno, C. (1999) Tithonia and
senna green manures and inorganic fertilizers as phosphorus sources for
maize in western Kenya. Agroforestry Systems 44: 21-36.
422
Ayuke, F.O. et al
Genstat 5 Committee (1993) Genstat release reference manual, Oxford University
Press, Oxford UK.
House, G.J. and Parmele, R.W. (1985) Comparison of soil arthropods and
earthworms from conventional, and no-tillage agroecosystems. Soil and
Tillage Research 5: 351-360.
Jaetzold, R. and Schmidt, H. (1982) Farm handbook of Kenya: National conditions
and farm management information of west Kenya (Nyanza and western
provinces, Volume IIA: 179-222. Typo-druck, Russedorf, West Germany.
Lavelle, P., Dangerfield, M., Fragoso, C., Eschnebrenner, V., Hernadez, D.L.,
Pashanasi, B. and Brussard, L. (1994a) The relationship between soil
macrofauna and tropical soil fertility. In: Woomer, P.L. and Swift, M.J. (eds).
The Biological Management of Tropical Soil Fertility . A Wiley- Sayce
publication, pp. 136-169.
Lavelle, P., Gilot, C., Fragoso, C. and Pashanasi, B. (1994b) Soil fauna and
sustainable land use in the humid tropics. In: Greenland, G.A. and
Szalbolcs, I. (eds), Soil Resilience and Sustainable Land use . CAB
International, Wallingford 18: 291-307.
Mafongoya, P.L., Mpepereki, S., Dzowela, B.H., Mangwayana, E. and Makonese,
F. (1996) Soil biota: Effects of pruning quality on soil microbial composition.
In: Swift, M.J. (ed). Report of the Tropical Soil Biology and Fertility Programme
(TSBF), Nairobi, Kenya.
Nziguheba, G., Palm, C.A., Buresh, R.J. and Smithson, P.C. (1998) Soil
phosphorus fractions and adsorption as affected by organic and inorganic
sources. Plant and Soil 198: 159-168.
Palm, C.A., Gachengo, C.N., Delve, R.J., Cadisch, G and Giller, K. (2001) Organic
inputs for soil fertility management in tropical agroecosystems: application
of an organic resource database. Agriculture, Ecosystems and Environment
83: 27-42.
Tian, G., Kang, B.T., Brussard, L. and Swift, M.J. (1997) Soil fauna mediated
decomposition of plant residues under constrained environmental and
residue quality and conditions. In: Cadisch, G. and Giller, K.E. (eds), Driven
by nature: Plant litter quality and decomposition . CAB International
Wallingford, U.K, pp 125-134,.
Tian, G., Kang, B.T and Brussard, L. (1995) An index for assessing the quality
of plant residues and evaluating their effects on soil and crop in the semihumid tropics. Applied Soil Ecology 2:25-32.
USDA. (1992) Soil taxonomy. Agricultural Handbook no. 294, Washington D.C.
Warren, S.D., Scifres, C.J. and Teel, P.D. (1987) Response of grassland arthropods
to burning: A review. Agricultural Ecosystem Environment 19: 105-130.
Dual Inoculation of Woody Legumes and Phosphorus Uptake from Insoluble
Phosphate Rock
Dual Inoculation of Woody
Legumes and Phosphorus
Uptake from Insoluble
Phosphate Rock
423
30
Kimiti, J.M.1* and Smithson, P.C.2
Kenya Forestry Research Institute, P.O. Box 20412, Nairobi,
Kenya
2
International Centre for Research in Agroforestry (ICRAF),
P.O. Box 30677, Nairobi, Kenya
Present address: University of Berea, Department of
chemistry, Berea, Kentucky USA
1
*Corresponding author
Abstract
Phosphorus deficiency limits crop production in western
Kenya, and there is need for more affordable sources of P for
resource-limited smallholder farmers in the region. Indigenous
phosphate rocks (PRs) from Uganda are abundant but
unreactive, so some means of increasing their effectiveness is
needed. We tested two agroforestry fallow species for their
ability to grow in sand culture with P supplied as Ugandan
Busumbu phosphate rock (BPR) or triple superphosphate
(TSP), and with or without rhizobial and/or mycorrhizal
inoculation. The test species were Crotalaria grahamiana
Kimiti, J.M. and Smithson, P.C.
424
and Tephrosia vogelii, two promising improved fallow species
in western Kenya. The experiment was conducted in a
greenhouse for three months. After the three months the
plants were harvested and shoot, root and nodule dry
biomass were determined. Addition of BPR had no effect on
biomass production of the test species compared to no-P
control, regardless of inoculation treatment. With P supplied
as TSP, Crotalaria total biomass production and nodule dry
weight was on average higher than that of Tephrosia in TSP
treatments. No growth parameters were affected by
inoculation with rhizobia, though uninoculated plants also
nodulated strongly in the presence of TSP. Mycorrhizae had
a small positive effect on root and shoot dry weight
(p = 0.033 to 0.057), but had no major effect on BPR
performance.
Introduction
Many soils in the tropical and subtropical regions are low in both
total and available P (Chien and Menon, 1995; Rao et al., 1999). As a
result, P is normally the most limiting nutrient for growth of
leguminous crops in tropical and subtropical regions. Continuous
cropping in cultivated areas with little or no use of fertilizers can
lead to depletion of soil P fertility. In many tropical regions and
particularly sub-Saharan Africa, soil P is declining as a result of
greater export of P through removal of harvested plant products and
erosion, than inputs of P from fertilizers and manures (Smaling et
al., 1997). Soil P deficiency may be due to either inherent low levels
of P or depletion of soils (Buresh et al., 1997). This is particularly so
for highly weathered acidic Oxisols and Ultisols in the humid tropics
which contain high levels of iron (Fe) or aluminium (Al) oxides where
P is strongly bound and thus less available for uptake by crops.
Therefore, P is critically needed to improve soil fertility for sustainable
crop production in acid soils (Chien and Menon, 1995).
Soil P can be replenished by addition of inorganic fertilizers or organic
matter in the form of plant and animal residues. However, the application
of soluble P fertilizers alone in acid soils is uneconomical due to high
production costs and low utilization efficiencies, which arise from the
high P-fixing abilities of the acid soils (Juo and Fox, 1977). Phosphorus
fixation limits crop production in high rainfall acid soils. In addition,
inorganic P fertilizers are expensive especially to smallholder farmers
who usually have low-income resource base.
In recent years, phosphate rock (PR) for direct application has been
tested in tropical acid soils as a potential alternative source of P to
Dual Inoculation of Woody Legumes and Phosphorus Uptake from Insoluble
Phosphate Rock
425
conventional water-soluble P fertilizers. Direct application of PR may
be an economically attractive alternative to the use of more expensive
imported soluble P fertilizers in these soils for certain crops. Soil
properties of agronomic importance in the effectiveness of PR are soil
pH, soil texture, P-sorption capacity and organic matter content.
However, the effectiveness of PR greatly depends on soil pH (Chien and
Menon, 1995). It is also documented that P release from PR may be
increased by inoculation with certain bacterial types, fungi as well as
root exudates from certain plant species (Tian and Kolawole 1999; Jones
and Farrar, 1999; Toro et al., 1996).
Plant species differ in their ability to take up nutrients from the
soil. As a result, higher plants have developed various mechanisms to
enhance nutrient acquisition from soils low in available nutrients. For
example, in response to P deficiency, certain plant species have
developed several complex mechanisms to take up P from the
rhizosphere. These mechanisms are both physiological and
morphological and vary from species to species (Zoysa et al., 1998;
Dinkelaker et al., 1988). Some of these changes include: increased
root hair length/density, enhanced symbioses with vesicular arbuscular mycorrhizae (VAM), formation of proteoid roots, release of
phosphatases to solubilize organically bound soil P and release of
organic acids and H+ to solubilize inorganic P (Jones and Farrar 1999;
Rao et al., 1999; Dinkelaker et al., 1988).
Organisms that cause increases in plant-available P in the soil
belong to a diverse group including bacteria, actinomycetes and several
groups of fungi (Kucey et al., 1989). For example, roots of most crop
and pasture species can be colonized by naturally occurring symbiotic
fungi to form vesicular-arbuscular mycorrhizae (VAM). The fungus
obligatorily depends on living plant roots for essential organic
compounds and in return increase the inflow of organic P from the
soil to the plant roots. Different crop species depend on varying degrees
of adequate colonization of their roots with VAM fungi. Consequently,
the inoculum density of VAM fungi in the soil when a crop is sown can
be an important factor in determining the P nutritional status of that
crop (Thompson, 1991).
Mycorrhizal plants have been shown to produce more dry matter
and remove more P from the soil than non-mycorrhizal plants (Raj et
al., 1980). For example, in 1989 Kucey and Leggett found that the
inoculation of canola (Brassica napus L.) with VAM under greenhouse
conditions increased straw and pod P concentration over uninoculated
control, while VAM inoculation in the field increased canola yields. In
the same study, it was found that combination of rock phosphate at 20
mg P kg -1 soil with VAM increased P uptake by canola to a level
comparable to that obtained by the addition of monoammonium
phosphate (MAP) at 20 mg P kg-1 soil. Mycorrhizal plants have also been
426
Kimiti, J.M. and Smithson, P.C.
shown to increase depletion of aluminium phosphate (Al-P), iron
phosphate (Fe-P) and calcium phosphate (Ca-P). In another study, Asea
et al. (1987) found that VAM fungi, Penicillium bilaji and Penicillium cf
fuscum could solubilize different amounts of rock phosphate in liquid
culture.
In an earlier study, Kimiti and Smithson ( 2001) grew 5 leguminous
agroforestry fallow species and 2 legume grain crops in unsterilised,
uninoculated sand culture, with P added as TSP, BPR or pure waterinsoluble Al-P, Fe-P and Ca-P compounds. Growth and P uptake of
species treated with BPR was poorer than that with all other P sources,
and was not different from no-P control. In the current study, we
hypothesised that inoculation with rhizobia and/or mycorrhizae could
improve growth and P uptake of leguminous fallow species from
unreactive PRs, through rhizosphere acidification or increased absorptive
surface area. Our main objective was to test the effect of separate or
dual inoculation with rhizobia and mycorrhizae on growth and P
acquisition by the short-fallow legumes Crotalaria grahamiana and
Tephrosia vogelii.
Materials and Methods
We grew Crotalaria grahamiana and Tephrosia vogelii in sand in pots
at the International Centre for Research in Agroforestry (ICRAF) nursery
in Nairobi. The study was a completely randomised design, with 24
treatments in four replications (Table 30.1). Seeds of C. grahamiana
and T. vogelii were obtained from ICRAF, while mycorrhizae and
rhizobial inocula were obtained from Kenya Forestry Research Institute
(KEFRI). Rhizobial inoculum was supplied in form of broth culture
and mycorrhizae as Glomus monosporum spores contained in chopped
roots of sorghum, which had been inoculated with Glomus and raised
in sterile potting mix in order to multiply the spores. River sand, sieved
to 1-2 mm diameter, was thoroughly washed with tap water and after it
was confirmed to be P-free, was sterilized in an autoclave and packed
into one litre plastic pots. Phosphorus as BPR or TSP (plus no-P control)
was applied at a rate of 50 kg P ha-1. Seeds of C. grahamiana and T.
vogelii were surface-sterilized with 70% ethanol and soaked in deionized
water for 24 hours. The seeds were sown directly in the pots. Rhizobial
and mycorrhizal inocula were placed over the seeds and then covered
with sand. All pots were placed on raised greenhouse benches and kept
moist with deionized water, and a P-free nutrient solution (Ae at al.,
1996) was added once per week. After three months, the plants were
harvested and shoot, root and nodule dry biomass was determined. Data
were analysed by analysis of variance and single degree of freedom
contrasts, using the Genstat 6th edition statistical package.
Dual Inoculation of Woody Legumes and Phosphorus Uptake from Insoluble
Phosphate Rock
427
Table 30.1: Treatment descriptions for a sand culture experiment testing effects of P
source and microbial inoculation on woody legume growth and nodulation
Trt
Crotalaria
Trt
No. Phosphorus Rhizobia Mycorrhizae No.
Tephrosia
Phosphorus Rhizobia Mycorrhizae
1
No P
0
0
13
No P
0
0
2
No P
1
0
14
No P
1
0
3
No P
0
1
15
No P
0
1
4
No P
1
1
16
No P
1
1
5
TSP
0
0
17
TSP
0
0
6
TSP
1
0
18
TSP
1
0
7
TSP
0
1
19
TSP
0
1
8
TSP
1
1
20
TSP
1
1
9
BPR
0
0
21
BPR
0
0
10
BPR
1
0
22
BPR
1
0
11
BPR
0
1
23
BPR
0
1
12
BPR
1
1
24
BPR
1
1
Results and Discussion
This experiment was conducted after an earlier study where the two
test species and others had been tested for their ability to access P from
BPR, TSP, aluminium phosphate, calcium phosphate and iron phosphate
(Kimiti and Smithson, 2001). In this earlier study, we observed that
BPR, which has 13% total P, had a poor solubility in water (0.7% of total
P) and neutral ammonium citrate (2.5% of total P). All 7-legume species
grew poorly in BPR compared to other P sources.
In this study rhizobia and/or mycorrhizae were included to test
whether BPR with microbial inclusion could improve plant growth relative
to BPR alone. With TSP, growth of both C. grahamiana and T. vogelli
was improved. Root and shoot biomass, nodule number and nodule dry
weight were all increased with TSP relative to BPR or No P (p < 0.0001,
Table 30.2, Figures 30.1 and 30.2) Growth with BPR was not significantly
different from No P control (p = 0.5 to 0.9). In addition, Crotalaria
outperformed Tephrosia in terms of all the growth parameters (Table
30.2). Statistical analysis revealed a significant difference (p < 0.0001)
in root and shoot biomass, nodule number and nodule dry weight
between C. grahamiana and T. vogelii (Table 30.2, Figure 30.1).
Kimiti, J.M. and Smithson, P.C.
428
Table 30.2: Nodulation, root and shoot biomass per pot with different P sources and
inoculation treatments. SED = standard error of the difference in means
Species
Crotalaria
P source
BPR
No P
TSP
Tephrosia
BPR
No P
TSP
SED
Contrasts (Prob.)
TSP vs other
BPR vs No P
Rhizobia vs none
Mycorrhizae vs none
Rhizobia Mycorrhizae
0
0
1
1
0
0
1
1
0
0
1
1
0
0
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
No.of
nodules
per pot
Shoot Dry Root Dry wt
wt (g per
(g per pot)
pot)
28
35
35
43
37
17
40
40
276
267
292
312
6
6
9
9
5
10
11
6
21
21
30
51
31
0.9
0.8
0.8
0.9
0.7
0.7
0.7
0.9
6.0
6.5
6.2
5.7
0.8
1.1
1.0
0.9
0.6
0.9
0.8
1.2
2.8
3.5
3.4
3.8
0.33
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.5
2.1
3.1
2.3
2.1
0.5
0.6
0.6
0.6
0.4
0.5
0.5
0.6
0.9
1.2
1.0
1.1
0.21
< 0.0001
0.940
0.168
0.778
< 0.0001
0.512
0.512
0.057
< 0.0001
0.579
0.572
0.033
While P source strongly affected nodulation, rhizobial inoculation
had almost no effect and in addition uninoculated plants nodulated
almost as profusely as those inoculated with rhizobia (Table 30.2, Figure
30.2). This study was conducted in a relatively open area, and pots
were exposed to airborne dust and microbes, as we did not take special
precautions to avoid exposure to atmospheric dust.
Mycorrhizal inoculation had a slight positive effect on root and shoot
dry weight (p = 0.033 for roots, 0.057 for shoots). In practical terms,
however, the differences were extremely small, and the hypothesized
improvement in BPR effectiveness was not realized.
Dual Inoculation of Woody Legumes and Phosphorus Uptake from Insoluble
Phosphate Rock
429
Figure 30.1: Total dry weight production (root plus shoot, g per pot) in Crotalaria
grahamiana (Cro.) and Tephrosia vogelii (Teph.), with or without inoculation with rhizobia
(R1 and R0) and/or mycorrhizae (M1 and M0), and with P absent (No P), or supplied at
50 kg ha-1 as triple superphosphate (TSP) or Ugandan Busumbu phosphate rock (BPR).
(Error bars are standard errors).
12.0
Total dry wt (g per pot)
10.0
123 No P
123
123 BPR
TSP
8.0
6.0
4.0
2.0
0.0
123
123
123
123
123
123
123
123
123
123
123
123
Cro.
M0
R0
Cro.
M1
R0
123
123
123
123
123
123
Cro.
M0
R1
123
123
123
123
123
123
Cro.
M1
R1
123
123
123
123
123
123
123
123
123
123
123
123
123
123
Teph.
M0
R0
123
123
123
123
123
123
123
Teph. Teph.
M1
M0
R0
R1
123
123
123
123
123
123
123
123
Teph.
M1
R1
Figure 30.2: Nodule dry weight production (g per pot) in Crotalaria grahamiana (Cro.)
and Tephrosia vogelii (Teph.), with or without inoculation with rhizobia (R1 and R0) and/
or mycorrhizae (M1 and M0) , and with P absent (No P), or supplied at 50 kg ha-1 as
triple superphosphate (TSP) or Ugandan Busumbu phosphate rock (BPR). (Error bars
are standard errors).
Nodule dry wt (g per pot)
0.6
No P
123
123
123 BPR
0.5
TSP
0.4
0.3
0.2
0.1
123
123
123
123
Cro.
M0
R0
Cro.
M1
R0
123
123
123
Cro.
M0
R1
123
Cro.
M1
R1
123
Teph.
M0
R0
123
123
12
12
12
12
Teph.
M1
R0
Teph.
M0
R1
Teph.
M1
R1
430
Kimiti, J.M. and Smithson, P.C.
It should be emphasized that this experiment was carried out in
pots in sand culture. Given these experimental conditions, the results
obtained may not translate into what might happen under actual field
conditions where the plants are not restricted in root growth and there
is interaction between the plants, soil, other soil micro-organisms and
other environmental factors. Recent field results from western Kenya
(Smithson and Kimiti, unpublished data) with BPR and TSP treatments,
using C. grahamiana and T. vogelii as fallows, show better maize
performance with BPR when used in combination with fallow legumes
than BPR alone, with N and K at equal rates throughout. This confirms
that sand culture may not be a good standard to predict the results
obtained from the field and that data from such studies should be
interpreted with caution.
Conclusion
We found little evidence in this study of increased growth of two
agroforestry legumes under P-limited conditions as a result of inoculation
with rhizobia and/or mycorrhizae. Growth and nodulation were
increased mainly by soluble P fertilizer and under P sufficient conditions,
Crotalaria grahamiana grew better and nodulated more profusely than
Tephrosia vogelii. These results, based on a pot study in sand culture,
should be extrapolated cautiously, however. Ongoing field studies with
BPR in combination with these same species show improved performance
of BPR when combined with legumes.
Acknowledgements
The authors thank The Ford Foundation, Nairobi, for financial support
for the first author’s participation in a research fellowship; ICRAF for
providing research facilities and staff time; KEFRI for providing microbial
inocula and release from duties for the first author; and Mercy Nyambura
for assistance in greenhouse and laboratory work.
References
Ae, N., Otani, T., Makino, T. and Tazawa, J. (1996) Role of cell wall of groundnut
roots in solubilizing sparingly soluble phosphorus in soil. Plant and Soil
186:197-204.
Asea, P.E.A., Kucey, R.M.N. and Stewart, J.W.B. (1987) Inorganic phosphate
solubilization by two Penicillium species in solution culture and soil. Soil
Biology and Biochemistry 20:459-464.
Dual Inoculation of Woody Legumes and Phosphorus Uptake from Insoluble
Phosphate Rock
431
Buresh, R.J., Smithson, P.C. and Hellums, D.T. (1997) Building soil phosphorus
capital in Africa. In: Buresh, R.J., P.A. Sanchez and F. Calhoun (eds).
Replenishing Soil Fertility in Africa. SSSA Special Publication 51. Soil Science
Society of America, Madison, Wisconsin, USA. pp. 111-149.
Chien, S.H. and Menon, R.G. (1995) Factors affecting the agronomic
effectiveness of phosphate rock for direct application. Fertilizer Research
41:227-234.
Dinkelaker, D., Römheld, V. and Marschner, H. (1988) Citric acid excretion
and precipitation of calcium citrate in the rhizosphere of white lupin (Lupinus
albus L.). Plant Cell and Environment 12:285-292.
Jones, D. and Farrar, J. (1999) Phosphorus mobilization by root exudates in
rhizosphere: fact or fiction. Agroforestry Forum 9(4):20-24.
Juo, A.R.S. and Fox, R.S. (1977) Phosphate sorption characteristics of some
benchmark soils of West Africa. Soil Science 124:370-376.
Kimiti, J.M. and Smithson, P.C. (2001) Phosphorus uptake by fallow and grain
legumes from sparingly soluble P sources. 5th Intl. Symposium on PlantSoil Interactions at Low pH, 11-16 March 2001, Alpine Heath, KwaZulu
Natal, South Africa
Kucey, R.M.N (1983) Phosphate-solubilizing bacteria and fungi in various
cultivated and virgin Alberta soils. Canadian Journal of Soil Science 63:671678.
Kucey, R.M.N. and Leggett, M.E. (1989) Increased yields and phosphorus uptake
by Westar canola (Brasssica napus L.) inoculated with a phosphatesolubilizing isolate of Penicillium bilaji. Canadian Journal of Soil Science
69:425-432.
Raj, J., Bagyaraj, D.J. and Manjunath, A. (1980) Influence of soil inoculation
with vesicular-arbuscular mycorrhiza and a phosphate-dissolving bacterium
on plant growth and 32P uptake. Soil Biology and Biochemistry 13:105-108.
Rao, I.M., Friesen, D.K. and Horst, W.J. (1999) Opportunities for germplasm
selection to influence phosphorus acquisition from low-phosphorus soils.
Agroforestry Forum. 9(4):13-17.
Smaling, E.M.A, Nandwa, S.M. and Janssen, B.M. (1997) Soil fertility in Africa
is at stake. In: Replenishing soil fertility in Africa. Buresh, R.J. Sanchez P.A
and Calhoun, F. (eds.). Soil Science Society of America Special Publication
51., Soil Science Society of America, Madison, Wisconsin, pp. 47-61.
Tian, G. and Kolawole, G.O. (1999) Phosphorus availability of phosphate rock
incubated with plant residues with various chemical composition.
Agroforestry Forum 9(4):40-42.
Toro, M., Azcon, R. and Herrera, R. (1996) Effects on yield and nutrition of
mycorrhizal and nodulated Pueraria phaseoloides exerted by P-solubilizing
rhizobacteria. Biology and Fertility of Soils 21:23-29.
Thompson, J.P. (1991) Improving the mycorrhizal condition of the soil
through cultural practices and effects of growth and phosphorus uptake by
432
Kimiti, J.M. and Smithson, P.C.
plants. In: Johansen, C., Lee K.K. and Sahrawat, K.L. (eds). Phosphorus
Nutrition of Grain Legumes in the Semi-Arid Tropics. International
Crops Research Institute for the Semi-Arid Tropics. Patancheru, India.
pp. 117-137.
Zoysa, A.K.N., Janzen H.H. and Leggett, M.E. (1998) Microbially mediated
increases in plant available phosphorus. Advances in Agronomy
42:199-227.
Effect of Vesicular-arbuscular Mycorrhiza (vam) Inoculation on Growth Performance of
Senna spectabilis
Effect of Vesicular-arbuscular
Mycorrhiza (vam) Inoculation
on Growth Performance of
Senna spectabilis
433
31
Kung’u, J.B.
School of Pure and Applied Sciences, Kenyatta University,
P.O. Box 43844 Nairobi Kenya
Email: kungu@avu.org
Abstract
The influence of vesicular-arbuscular mycorrhiza (VAM)
fungi inoculation on growth performance of Senna
spectabilis was studied in a screen house experiment. The
results obtained indicated the dependence of Senna
spectabilis on mycorhizal symbiosis. Inoculation with
vesicular-arbuscular mycorrhiza significantly improved the
growth performance of Senna spectabilis. The height growth
increased significantly by 85% after only three months while
the root collar diameter increased by 71%. Shoot production
increased by 213% while root biomass increased by 241%.
Inoculation with vesicular-arbuscular mycorrhiza increased
plant tissue phosphorus, nitrogen and potassium content.
The better growth response of mycorrhizal plants were
attributed to improvement in nutrient uptake, especially
phosphorous, nitrogen and potassium. Vesiculararbuscular mycorrhiza inoculation has a high potential in
agroforestry as a bio-fertilizer.
434
Kung’u, J.B.
Introduction
The intense exploitation of tropical forests has led to degradation of
once stable ecosystems. There has been changes in abiotic and biotic
soil properties, which hampers the re-establishment of proper vegetation
cover (Miller, 1987). Soils in these areas are very infertile and are acidic
in nature. The soils are characterized by low effective cation exchange
capacity, low available water and nutrient reserve, low soil pH, low
organic matter and phosphorus content and are highly susceptible to
soil erosion. The deforested and degraded areas no longer regenerate
into woody perennials due to lack of mycorrhiza propagules for
recolonization but rather into the so-called "derived savannas" which
now occupy million of hectares in Africa in form of Imperata cylindrica
and Themeda triandra grasslands (Janos, 1980a). This is because grasses
are the most independent of mycotrophic plants and they can tolerate
low soil fertility inspite of their low ineffectiveness (Baylis, 1975).
Agroforestry, a land-use system and technology in which trees are
deliberately planted on the same units of land with agricultural crop
and /or animals, has been recognized as one of the most promising
strategy for rehabilitating the already degraded areas. The benefits of
agroforestry includes the amelioration of soil chemical and physical
properties, the reduction of soil erosion, improved weed control and
increased availability of fuel wood and /or fodder (Young, 1997; Chin
and Huxley, 1996). The degree to which an agroforestry system can
provide the above benefits partially depends on the quantity of biomass
an agroforestry tree species can produce.
Acid soils are known particularly to be unfavorable for legumes due
to iron, aluminum and /or manganese toxicities, as well as molybdenum,
calcium, and/or magnesium deficiencies. Molybdenum is an essential
nutrient in nitrogen fixation, while calcium requirements in legumes
are high and therefore deficiencies of either of these elements can cause
low biomass yields in an agroforestry leguminous tree species.
Mycorrhizal fungi are known to affect growth of most plant species
through various ways. They increase phosphorus uptake, enhance
uptake of other plant nutrients by root system and are beneficial in the
biological nitrogen fixation of Rhizobium, biological control of root
pathogens and drought resistance (Harley and Smith, 1983; Sieverding,
1991; Dela Cruz, 1987; Janos, 1980b). The potential benefit of
mycorrhizal fungi in rehabilitation of degraded areas by use of
agroforestry system is more apparent than ever before. The need to
increase food, fibre, and fuel wood production to keep pace with the
fast growing population in Africa is crucial. The low biomass production
of agroforestry tree species in degraded areas can, therefore, be
circumvented by the use of mycorrhizal fungi. Unfortunately, there seems
to be very little research in using mycorrhizal fungi in an agroforestry
Effect of Vesicular-arbuscular Mycorrhiza (vam) Inoculation on Growth Performance of
Senna spectabilis
435
setting. This paper reports a green house experiment that tested the
effect of vesicular-arbuscular mycorrhiza inoculation on growth
performance of Senna spectabilis. The plant is an important agroforestry
tree species, which has passed the tests of practicability and acceptability
in the eyes of researchers and farmers. The tree is widely recommended
as an agroforestry tree species for degraded areas in many parts of the
tropics but its main problem lies in slow growth rate in acidic soils.
Materials and Methods
The experiment was conducted in a screen house in the University of
the Philippines Los Baños. The experiment was laid out in a randomized
complete block (RCB) design, with four replicates and four treatments.
Each treatment consisted of five 20cm clay pots. A total of eighty clay
pots were used and a total of 240 plants were planted. Top soil (015cm) was collected from a degraded grassland area that was dominated
by Imperata cylindrica. The soil was air dried, pulverized and passed
through a 2mm sieve. The soil was then sterilized with hot air at 100 °C
for 48 hours. The soil had an initial pH of 5.14 (Potentiometric Method),
organic matter content of 1.67% (Walkley-Black Method), total nitrogen
0.18% (Modified Kjedahl Method), potassium 4.11me/100g (Flame
Photometer Method) and available phosphorus 70.18 ppm (Bray No.2
Method).
The soil was then put into the 20cm top diameter clay pots. The VAmycorrhizal fungi inoculants consisting of spores, mycorrhizal root
fragments and infected soil was collected from pot cultures of trap plants
(Pensacola bahia ) grass which had been grown for five months after
being inoculated with mycorrhiza fungus species of Glomus tunicatum
and Glomus macrocarpum. The inoculants were added to some pots, at
the rate of one table spoon per pot which consisted of 23 spores per
gram of soil added. The rate of spores per gram of soil was determined
by wet sieving and decanting, surface sterilized in 2% sodium hypochlorite and then washed. The non vesicular arbuscular control pots
were left uninoculated. Seeds of Senna spectabilis were pre-treated with
hot water for three minutes. The seeds were then germinated in sterilized
river sand. After the seedlings had developed two leaves each, three
seedlings were transplanted to each clay pot containing the sterilized
soil, plus or minus the vesicular arbuscular mycorrhiza inoculum.
Seedlings were then watered twice a day for the first week and then
once a day in the following weeks.
To determine the effect of vesicular arbuscular mycorrhiza
inoculation on growth performance of Senna spectabilis, inoculated and
non-inoculated plants were raised in a screen house for three months.
Height growth was measured after every 15 days, except during the
436
Kung’u, J.B.
first months. Root collar diameter was measured at the end of three
months. After four months, 50% of the plants per block were harvested
using destructive sampling and vesicular arbuscular mycorrhiza
colonization above and below ground biomass production, root number
and root length were determined. At the end of fifth month, some plants
were harvested randomly per treatment and vesicular arbuscular
mycorrhiza infection level was assessed by clearing the roots for 2 hours
at 90°C in 10% KOH, neutralizing them in lactoglycerol for 20 minutes.
Infection was determined by the grid-line intersect method (Giovanetti
and Mosse, 1980). Biomass increment due to mycorrhiza inoculation
was computed as dry weight of inoculated plants minus dry weight of
non-inoculated plants divided by dry weight of non-inoculated plants
multiplied by 100%.
For the plant tissue nutrient content, above ground biomass was
harvested and was oven dried at 70 oC. The plant tissue were then
analyzed for total nitrogen (Micro-kjedahl method), Total phosphorus
(Vanadomolybdate method) and Potassium (Flame photometer method).
The numbers and length of primary roots per plants were assessed and
determined. The measured plants parameters were analysed using
IRRISTAT version 92-1 computer software. Analysis of variance was
used to describe the data.
Results and discussion
Plant Height
The results obtained indicated the dependence of Senna spectabilis on
mycorrhiza symbiosis. The effect of vesicular-arbuscular mycorrhiza
inoculation on the height increment was obvious on visual comparison
at the end of 90 days. As Table 31.1 shows, a significant height
increment in inoculated Senna spectabilis was recorded after only 60
days. The enhanced height increment in Senna spectabilis could be
attributed to the vesicular arbuscular mycorrhiza colonization.
Mycorrhiza infection is known to enhance plant growth by increasing
nutrients uptake (Marschner et al., 1994). Nye et al. (1977) reported
that the uptake of nitrogen, phosphorus and potassium is limited by
the rate of diffusion of each nutrient through the soil. It seems likely
that vesicular arbuscular mycorrhiza in this study increased nutrient
uptake by shortening the distance nutrients diffused through the soil
to the roots. During the first 45 days, there was no significant difference
in height increment between inoculated and non inoculated plants,
although the height increment in inoculated plants was higher. This
could be due to the "lag phase" effect of mycorrhiza inoculation. Many
studies have shown that there is a lag phase between mycorrhiza
Effect of Vesicular-arbuscular Mycorrhiza (vam) Inoculation on Growth Performance of
Senna spectabilis
437
inoculation and the time period when its effect is manifested in the
plant (Brandon and Shelton, 1993).
Table 31.1: Effects of VA mycorrhizal fungi inoculation on shoot height (cm) of Senna
spectabilis after 90 days in the screen house.
Days after
planting
Senna spectabilis
Treatment with
vesicular arbuscular
mycorrhiza
Treatment without
vesicular arbuscular
mycorrhiza
Difference
7.06e
9.90d
14.21 c
17.16b
19.80a
6.51c
8.20bc
8.29bc
9.03ab
10.72a
0.55ns
1.70ns
6.08**
8.13**
9.08**
30
45
60
75
90
Means in columns followed by the same letter are not significantly different at 5% level
based on DMRT test.
** = significant at 1% level
ns = not significant
At the end of ninety days, height growth of inoculated Senna.
spectabilis was highly significant as compared to the non inoculated
plants. The higher height increment registered with inoculated plants
could be as a result of enhanced inorganic nutrient absorption (Cooper,
1984) and greater rates of photosynthesis (Allen et al., 1981). Vesiculararbuscular mycorrhiza are known to affect both the uptake and
accumulation of nutrients and therefore, act as an important biological
factor that contribute to efficiency of both nutrient uptake and use.
Researchers have demonstrated that vesicular-arbuscular mycorrhiza
fungi, not only increases phosphorus uptake, but also plays an important
role in the uptake of other plant nutrients and water (Huang et al.,
1985; Ellis et al., 1985). Sander et al. (1983), reported that the inflows
of phosphorus to mycorrhiza roots can be greater than inflows to
comparable non-mycorrhiza roots by up to 2-5 times.
Shoot Biomass
Inoculating Senna spectabilis with vesicular-arbuscular mycorrhizal
fungi, increased significantly the shoot biomass yield. As shown in
Table 31.2, the shoot biomass production increased by 213% and
was highly significant. The highly significant shoot biomass
production by the inoculated plants, could be attributed to enhanced
438
Kung’u, J.B.
inorganic nutrition absorption and greater rates of photosynthesis
in inoculated plants (Allen et al., 1981; Cooper, 1984). Vesiculararbuscular mycorrhiza have been said to affect both the uptake
and accumulation of nutrients. Chulan and Martin (1992), reported
a significant shoot dry weight increment when Theobroma cacao
was inoculated with VA-mycorrhiza. Aggangan and Dela Cruz (1991),
reported a dry matter yield increment of up to 631% when L.
leucocephala was inoculated with vesicular-arbuscular mycorrhiza.
Zajicek et al. (1987) reported a significant increment in dry matter
yield when two forbs were inoculated with vesicular-arbuscular
mycorrhizal fungi. Vesicular-arbuscular mycorrhizal fungi are
reported to enhance plant growth rate through an increase in
nutrient uptake, especially phosphorus which is relatively immobile
in soils (Kormanik et al. , 1981, 1982; Dela Cruz, 1987; Janos,
1980a). Vesicular-arbuscular mycorrhiza inoculation could have
enhanced Senna spectabilis to absorb more nutrients via an increase
in the absorbing surface area. Similar observation has been reported
by Marschner and Dell (1994).
The movement of nutrients to plant roots and the rate of absorption
of nutrients by roots, especially nitrogen, phosphorus and potassium,
is known to be limited by the rate of diffusion of each nutrient through
the soil and not by the ability of the root to absorb the nutrient from low
concentration in the soil solutions (Abbott and Robson, 1982). In the
present study, since the soil used was not very fertile, inoculation with
vesicular-arbuscular mycorrhiza could have resulted in an increase in
nutrient uptake by merely shortening the distance that the nutrients
had to diffuse from the soil to the roots. This in turn, could have
enhanced a higher shoot biomass production in the inoculated Senna
spectabilis .
Root Biomass
As Table 31.2 shows, inoculating Senna spectabilis with vesiculararbuscular mycorrhiza significantly increased the root biomass
production. Vesicular abuscular mycorrhiza infection has been reported
to increase both the uptake of nutrients by the roots and the
concentration of nutrients in the plant tissues (Smith et al., 1979). An
increase in nutrient uptake, especially phosphorus in the infertile soil
used, could have resulted in relief of nutrients stress and an increase in
photosynthetic rate, which obviously could have given rise to an increase
in plant growth. Research has shown that when root exploration is
restricted, up to 80% of the plant phosphorus can be delivered by the
external vesicular-arbuscular mycorrhizal hyphae to the host plant over
a distance of more than 10 cm from the root surface (Li et al., 1991) .
Effect of Vesicular-arbuscular Mycorrhiza (vam) Inoculation on Growth Performance of
Senna spectabilis
439
Hattingh et al. (1973) found that vesicular arbuscular mycorrhizal
hyphae, could intercept labelled phosphorus, placed 27mm from a
mycorrhizal root, whereas it remained unavailable to non-mycorrhizal
roots. This confirms that vesicular-arbuscular mycorrhizal hyphae could
have increased the volume of soil available to the Senna specabilis for
nutrient uptake.
Table 31.2: Effect of vesicular-arbuscular mycorrhiza inoculation on growth performance
of Senna spectabilis after 90 days in a screen house.
Growth
Treatment
Parameter
VAM Innoculation
Measurement
Increment
percentage
Shoot dry weight
Non inoculated
Inoculated
Non inoculated
Inoculated
Non inoculated
Inoculated
Non inoculated
Inoculated
Non inoculated
Inoculated
Non inoculated
Inoculated
Not inoculated
Inoculated
Non inoculated
Inoculated
Non inoculated
Inoculated
2.82
8.83**
1.66
5.66**
10.72
19.80**
0.21
0.36**
4.30
8.80**
0.60
0.65ns
24.33
30.41**
10.0
10.75 ns
0
67.75**
213
Root dry weight (g pot -1)
Total shoot length (cm)
Root collar diameter (cm)
Leaf number
Root/shoot ratio (R/S)
Root length (cm)
Root Number/plant
Roots colonized (%age)
241
85
71
105
8
25
7.5
67.8
** = significant at 1% level,
ns = not significant
Mycorrhizal roots have been known to absorb phosphorus faster
per gram of root than non-mycorrhizal plants (Jakobsen et al., 1992).
This may relate to the greater surface area per gram of mycorrhiza roots.
It therefore follows that mycorrhiza were able to enhance the absorption
of nutrients from the soil, which could have moved to the roots principally
by mass flow, in addition to those which could have diffused through
the soil slowly. This could have resulted in a higher root biomass in
inoculated plants.
440
Kung’u, J.B.
Root collar diameter
Vesicular-arbuscular mycorrhiza inoculation increased the root collar
diameter of Senna spectabilis by 74%. As shown in Table 31.2, the
increment of the root collar diameter of the vesicular-arbuscular
mycorrhiza inoculated, plants was highly significant. The higher diameter
increment of the inoculated plants could be attributed to enhanced
inorganic nutrition absorption and greater rates of photosynthesis of
inoculated plants (Allen et al., 1981; Cooper, 1984). Vesicular- arbuscular
mycorrhiza have been said to affect both the uptake and accumulation
of nutrients. Researchers have demonstrated that vesicular-arbuscular
mycorrhiza fungi not only increases phosphorus uptake, but also plays
an important role in the uptake of other plant nutrients (Huang et al.,
1985; Sieverding, 1991).
Many authors have reported a significant increment in root collar
diameter, after inoculating the plants with vesicular-arbuscular
mycorrhiza. Reid et al. (1988), reported an increment in root collar
diameter when sugar maple seedlings were inoculated with vesiculararbuscular mycorrhiza. Osonubi et al. (1989), while working with
inoculated Gmelina seedlings, reported a significant biomass increment.
Huang et al. (1985) while working with inoculated Leucaena leucocephala,
reported a significant increment in plant growth parameters. Aggangan
and Dela Cruz (1991), while working with Acacia auriculiformis and
Leucaena leucocephala, reported a diameter increment of between 18%
to 123% when the two plants were inoculated with different types of
vesicular-arbuscular mycorrhizal fungi. Castillo (1993), while working
with Pterocarpus indicus, reported a significant diameter increment when
the plants were inoculated with vesicular-arbuscular mycorrhizal fungi.
Kormanik et al. (1981) reported a significant increment in root collar
diameter when sweetgum seedlings were inoculated with vesiculararbuscular mycorrhizal fungi. He reported that inoculation with
vesicular-arbuscular mycorrhiza increased the root collar diameter by
268%.
Root to Shoot Ratio
As shown in Table 31.2, the difference between the root to shoot ratio of
inoculated and non-inoculated Senna spectabilis, was not statistically
significant at 5% level though the inoculated Senna spectabilis had a
higher root to shoot ratio as compared to non inoculated plants. The
higher root to shoot ratio of the inoculated plants could be attributed to
the effect of mycorrhiza infection, which could have increased nutrients
absorption, giving rise to a higher root and shoot biomass increment
with a uniform growth. Clapperton and Reid (1992) while researching
Effect of Vesicular-arbuscular Mycorrhiza (vam) Inoculation on Growth Performance of
Senna spectabilis
441
on the relationship between plant growth and increasing vesiculararbuscular mycorrhizal inoculum density, reported that as the
colonization by vesicular-arbuscular mycorrhizal fungi increased, so
did root to shoot ratios. They concluded that this was due to the
vesicular-arbuscular mycorrhizal plants being able to translocate more
carbon to the roots than non-mycorrhiza plants. The same has been
reported by Kucey and Paul (1982); Douds et al. (1988) and Wang et al.
(1989). Tree seedlings with higher root to shoot ratios are able to have a
higher survival percentage when planted in the field.
Root number and length
As Table 31.2 shows, inoculating Senna spectabilis with vesiculararbuscular mycorhiza fungi, significantly increased the root length. The
inoculation with VAM increased the root length by 25%. Huang et al.
(1985) reported a root length increment of up to 80% when Leucaena
leucocephala was inoculated with vesicular-arbuscular mycorrhiza. Levy
and Syvertsen (1983) while working on the effect of drought stress on
citrus, reported that, although plant to plant variations obscured
significant differences, vesicular-arbuscular mycorrhiza plants did tend
to have greater total feeder root length per plant than control plants. In
addition to the mycorrhiza inoculation enhancing the plants absorption
of more nutrients, especially phosphorus, via an increase in the
absorbing surface area (Marschner and Dell, 1994), mycorrhiza
colonization could have protected roots from soil pathogen (Perrin, 1990),
and therefore increased root growth and nutrients acquisition of Senna
spectabilis. Inoculated plants had higher number of roots than non
inoculated ones, though the increment was not significant at 5% level.
Mycorrhiza inoculation is known to enhance the plants absorption of
more nutrients especially phosphorus via an increase in the absorbing
surface area (Marschner and Dell, 1994). This in turn could have
enhanced a higher plant growth rate resulting to more roots per plant.
Mycorrhiza colonization also protect the roots from the soil pathogens
(Perrin, 1990) and, therefore could have lead to an increase in not only
the root growth and nutrient acquisition of the host roots, but also the
number of surviving roots.
Root Colonization Percentage
As shown in Table 31.2, inoculating Senna spectabilis with vesiculararbuscular mycorrhiza fungi resulted into a 67.8% colonization. There
was no vesicular-arbuscular mycorrhiza contamination as evident in
the non inoculated plants (control) which showed a 0% colonization.
442
Kung’u, J.B.
Mycorrhiza colonization is normally attributed to the tree species and
environmental factors. Smith et al. (1979) reported that the extent to
which typical vesicular-arbuscular mycorrhiza fungi colonize root
systems varies with species of plant. It has also been noted that there
are differences in the extent of infection between genotypes of the same
species. The extent of mycorrhiza infection in root systems is also known
to be influenced by environmental conditions; the most important being
the age of the plants, the level of phosphate (P) in the soil relative to the
requirements of the plant and the capacity of the population of
mycorrhiza propagules in the soil to form mycorrhiza. Senna spectabilis
is a non nodulating legume (Ladha et al., 1993) and rhizobium bacteria
could not have posed any threat in competing with mycorrhiza fungi for
carbohydrates. The time period of the seedlings (five months) could have
been too short to record a higher colonization percentage since the root
system infected normally increases with time sigmoidally. Seasonal
patterns in the formation of mycorrhiza have also been said to vary
considerably from year to year (Allen et al., 1989).
Plant Tissue Nutrients Concentration
Inoculating Senna spectabilis with vesicular-arbuscular mycorrhiza,
increased plant tissue nutrients concentration. As Table 31.3 shows,
plant tissue phosphorus, nitrogen and potassium concentration was
much higher in the inoculated plants than non inoculated ones. The
higher phosphorus concentration in the inoculated plants could be
attributed to a higher nutrients absorption rate by mycorrhiza plants.
Table 31.3: Effect of vesicular-arbuscular mycorrhiza inoculation on nutrient concentration
(NPK) in shoot of Senna spectabilis after 90 days in a screen house
Plant
Concentration
Tissue
Nutrient
VAM Innoculation
Phosphorus (%)
Nitrogen (%)
Potassium (%)
Inoculated
non-inoculated
0.46**
0.19
3.05 ns
2.99
1.64ns
1.53
** = significant at 1% level,
ns = not significant
Several authors have reported that mycorrhizal roots are able to
absorb several times more phosphate than non inoculated roots from
soils and from solutions (Pearson and Gianinazzi, 1983; Michelsen and
Effect of Vesicular-arbuscular Mycorrhiza (vam) Inoculation on Growth Performance of
Senna spectabilis
443
Rosendahl, 1990; Fitter, 1988; Dela Cruz et al., 1988; Nielsen, 1983).
Increased efficiency of phosphorus uptake by mycorrhizal plants could
have led to higher concentrations of P in the plant tissues. The greater
phosphate absorption by vesicular-arbuscular mycorrhizae has been
suggested to have arisen due to superior efficiency of uptake from labile
forms of soil phosphate, which is not attributable to a capacity to mobilize
phosphate sources unavailable to non mycorrhizal roots (Pearson and
Gianinaazzi, 1983). Under certain conditions, mycorrhiza is known to
absorb fixed phosphate and even to stimulate root phytase activities
(Pearson and Gianinazzi, 1983). Mycorrhizal roots are known to have
not only a considerably greater phosphate inflow rates, but also to
possess a pathway of phosphate uptake with a much higher affinity for
phosphate than non mycorrhizal roots.
The higher plant tissue nitrogen content in inoculated plants could
be attributed to hyphae uptake. It has been reported that the existence
of extra-radical hyphal bridges between individual plants permits transfer
of nutrients such as nitrogen (Marschner and Dell, 1994). The two have
reported that about 24% of the total nitrogen uptake in mycorrhizal
plants could be atributed to uptake and delivery by the external hyphae.
There is also evidence that nitrogen is taken up by vesicular-arbuscular
mycorrhiza hyphae from inorganic sources of ammonium (Ames et al.,
1983) and therefore, the higher nitrogen concentration in mycorrhizal
plants could be attributed to the hyphae uptake. The same could be
said of the higher potassium concentration in inoculated plants. In a
compartment pots experiment, Li et al. (1991), demonstrated that about
10% of the total potassium uptake in mycorrhizal coach grass was due
to hyphal uptake and transport.
Conclusion
The current study had shown that inoculating Senna spectabilis with
vesicular-arbuscular mycorrhyza enhances growth performance. The
inoculation resulted in an increment in height growth by 85% and root
collar diameter by 71% within three months. Shoot biomass increased
significantly by 213% while root biomass increased by 241%. Inoculated
plants subequently produced more leaves per plant, which could have
increased the rate of photosynthesis. Inoculated plants produced also
more roots per plant which were longer than in the non inoculated
plants. This improvement in plant growth could be attributed to the
enhancement of the plant to absorb more nutrients, via an increase in
the absorbing surface area. Vesicular-arbuscular mycorrhiza
colonization also protects roots from soil pathogens and thereby increase
root growth and nutrients acquisition of the host plants.
444
Kung’u, J.B.
References
Abbott, L. K. and Robson, A. D. (1982) The role of Vesicular- Arbuscular
Mycorrhizal fungi in agriculture and the selection of fungi for inoculation.
Aust. J. Res. 33: 389-408.
Aggangan, N. S. and Dela, R. E. Cruz. (1991) Growth improvement of two forest
tree legumes by VA mycorrhizal inoculations. The Phillipines journal of
Biotechnology 2(1): 72-80).
Allen, M. F., Smith, W.K., Moore jr, T.S. and Christensen, M. (1981) Comparative
water relations and photosynthesis of mycorrhizal Bouteloua gracilis H.B.K.
lag ex Steud. New Phytologist 88: 683-693.
Allen, M. F., Allen, E. B. and Friese, C.F. (1989) Responses of the non-mycotrophic
plant Salsola kali to invasion by vesicular-arbuscular mycorrhizal fungi.
New Phytol. 111: 45-49.
Ames, R.N., Reid, C.P.P., Porter, L.K. and Cambaedella, C. (1983) Hyphal uptake
and transport of nitrogen from two 15N-labelled sources by Glomus mosseae,
a vesicular-arbuscular mycorrhizal fungus. New Phytology 95: 381-396.
Baylis, G.T.S. (1975) The magnoliod mycorrhiza and mycotrophy in root systems
derived from it. In, Sanders, F.E., Mosse, B. and Tinker, P.B. (Eds).
Endomycorrhizas. Pp 373-389. Academic press, London.
Brandon, N.J. and Shelton, H.M. (1993) Role of vesicular-arbuscular mycorrhizae
in Leucaena establishment. Proceedings of the XVII International Grassland
congress 2064-2065.
Castillo, E.T. (1993). Physiological responses of Narra (Pterocarpus indicus wild.)
inoculated with VA Mycorrhiza and rhizobium in macolod soils. Ph.D. Thesis.
UP Los Banos.
Chin K. Ong and Huxley, P. (editors). (1996) Tree-crop interactions. A physiological
approach. CAB International.386 pp.
Chulan, H.A. and martin, K. (1992) The vesicular-arbuscular (VA) mycorrhiza
and its effects on growth of vegetatively propagated Theobroma cacao L. Plant
and Soil 144. 227-233.
Clapperton, M.J. and Reid, D.M. (1992) A relationship between plant growth
and increasing VA mycorrhizal inoculum density. New phytologist 120. 227234.
Cooper, K.M. (1984) Physiology of VA mycorrhizal associations. In: VA Mycorrhiza
(Ed. By C. L. Powell and D. J. Bagyaraj), pp. 155-186. CRC Press, Inc.,
Boca Raton, Florida.
Dela Cruz, R.E. (1987) Status of Mycorrhizal Research in the Philippines and
Its implication to national development, SEARCA Professional Chair Lecture
given at SEARCA Auditorium March 11, 1987.
Dela Cruz, R.E., Manalo, M.Q., Agangan, N.S. and Tambalo, J.D. (1988) Growth
of three legume trees inoculated with VA mycorrhizal fungi and Rhizobium.
Plant and Soil 108: 111-115.
Dela Cruz, R.E. (1991) Status of Bio-reforestation in the Philippines. Paper
presented during the pre-Workshop on Bio-Reforestation sponsored by
Effect of Vesicular-arbuscular Mycorrhiza (vam) Inoculation on Growth Performance of
Senna spectabilis
445
IUFRO Japan, Bogor, Indonesia March 25-29.
Douds, J.R., Johnson, C.R. and koch, K.E. (1988) Carbon cost of the fungal
symbiont relative to net leaf P accumulation in split-root Va mycorrhizal
symbiosis. Plant Physiology 86: 491-496.
Ellis, J.R., larsen, H. J. and Boosalis, M.G. (1985) Drought resistance of wheat
plants inoculated with vesicular mycorrhizae. Plant and Soil 86: 369-378.
Fitter, A.H. (1988) Water relations of red clover Trifolium pratence L. as affected
by VA mycorrhizal colonization of phosphorus supply before and during
drought. J. Exp. Bot. 39: 595-603.
Giovannetti, M. and Mosse, B. (1980) An evaluation of techniques for measuring
vesicular-arbuscular mycorrhizal infection in roots. New Phytol. 84: 489500.
Harley, J.L. and Smith, S.E. (1983) Mycorrhizal Symbiosis. Academic Press,
London 483 pp.
Hattingh, M.J., Gray, L. E. and Gerdenmann, J. W. (1973) Uptake and
translocation of 32P-labelled phosphate to onion roots by endomycorrhizal
fungi. Soil Sci. 116, 383-387.
Huang, R.S., Smith, W. K. and Yost, R.E. (1985) Influence of vesicular-arbuscular
mycorrhizae on growth, water relation and leaf orientation in Leucaena
leucocephala (Lam.) De wit. New Phytol. 99: 229-243.
Jakobsen, I., Abbott, L.K. and Robson, A.D. (1992) External hyphae of arbuscular
mycorrhizal fungi associated with Trifolium subterraneum L. I. Spread of
hyphae and phosphorus inflow into roots. New Phytol. 120: 372-380.
Janos, D.P. (1980a). Mycorrhizae influence tropical succession. Biotropica.
Special edition. (1980) 56-64.
Janos, D.P. 1980b. Vesicular-arbuscular mycorrhizae affect lowland tropical
rain forest plant growth. Ecology 61(1):151-162.
Kormanik, P.P., Bryan, W.C. and Schultz, R. C. (1981) Effects of three vesiculararbuscular mycorrhizal fungi on sweetgum seedlings from nine mother trees.
Forest Sci, 27 No. 2. 327-335.
Kormanik, P.P. and Mcgraw, A..C. (1982) Quantification of vesicular-arbuscular
mycorrhizae in plant roots. In Methods and principles of mycorrhizal research,
by N.C. Schenck (Ed), University of Florida. pp 37-45.
Kucy, R.M.N. and Paul, E.A. (1982) Carbon flow, photosynthesis and nitrogen
fixation in mycorrhizal and nodulated faba beans (Vicia faba L.) Soil biology
Biochemistry 14: 407-410.
Ladha, J.K., Peoples, M. B., Garrity, D. P., Capuno, V. T. and Dart, P. J. (1993)
Estimating dinitrogen fixation of hedgerow vegetation using the nitrogen15 natural abundance method. Soil Sci. Soc. Am. J. 57: 732-737.
Levy, Y. and Syvertsen, J. P. (1983) Effect of drought stress and vesiculararbuscular mycorrhiza on citrus transpiration and hydraulic conductivity
of roots, New Phytologist 93: 61-66.
Li X-L, George, E. and Marschner, H. (1991) Extension of the phosphorus
depletion zone in VA-mycorrhizal white clover in calcareous soil. Plant and
Soil. 136: 41-48.
446
Kung’u, J.B.
Marschner, H. and Dell, B. (1994) Nutrient uptake in mycorrhizal symbiosis.
Plant and Soil 159:89-102.
Michelsen, A. and Rosendahl, S. (1990) The effect of VA mycorrhizal fungi,
phosphorus and drought stress on the growth of Acacia nilotica and Leucaena
leucocephala seedlings. Plant and soil 124: 7-14.
Miller, R.M. (1987) The ecology of vesicular-arbuscular mycorrhizae in grass
and shrublands. In: G.R. Safir (editor), Ecophysiology of VA Mycorrhizal
Plants. CRC. Press, Boca T=Raton, pp. 135-170.
Nielsen, J.D. (1983) Influence of vesicular-arbuscular mycorrhiza fungi on growth
and uptake of varius nutrients as well as uptake ratio of fertilizer P for
lucerne (Medicago sativa). Plant and Soil 70: 165-172.
Nye, P.H. and Tinker, P.B. (1977) "Solute movements in the root-soil systems"
Blockwell. Oxford.
Osunubi, O., Okon, I.E. and Bamiduro, T.A. (1990) Effect of different fungal
inoculation periods on performance of Gmelina seedlings under dry
conditions. Forest Ecology and Management 37(4): 223-232.
Pearson, V.G. and Gianinazzi, S. (1983) The Physiology of vesicular-arbuscular
mycorrhizal roots. Plant and Soil 71: 197-209.
Perrin, R. (1990) Interactions between mycorrhizae and diseases caused by soilborne fungi. Soil Use Management 6: 189-195.
Reid, R.K., Parker, W.C., Mitchell, R.J., Garrett, H.E., henderson, G.S and Cox,
G.S. (1988) Vesicular-arbuscular mycorrhizal influences on growth and
aluminium tolerance of sugar maple seedlings. In proceedings, 10th North
American Forest Biology Workshop, Physiology and genetics of reforestation;
vancouver British Columbia, July 10-22.
Sanders, F.E. and Sheikh, N..A. (1983) The development of vesicular-arbuscular
mycorrhizal infection in plant root systems. Plant and Soil 71: 223-246.
Sieverding, E. (1991) Vesicular-arbuscular mycorrhiza management in tropical
Agrosystems. GTZ Press. Germany.
Smith, S.E., Nicholas, D. J. D. and Smith, F. A. (1979) Effect of early mycorrhizal
infection on nodulating and nitrogen fixation in Trifolium subterraneum L.
Aust, Plant J. Physiol. 6: 305-316.
Wang, G.M., Coleman, D. C., Freckman, D. W., Dyer, M. I., Mcnaughton, S.J.
Acra, M. A. and Goeschl, J. D. (1989) Carbon partitioning patterns of
mycorrhizal versus non- mycorrhizal plants: real time dynamic
measurements using 11CO2. New Phytologist 112, 489-493.
Young, A. (1997) Agroforestry for soil management.CAB International. 320pp.
Zajicek, J. M., Hetrick, B. A. D. and Albrecht, M.L. (1987) Influence of drought
stress and mycorrhizae on growth of two native forbs. J. Amer. Soc. Hor. Sci.
112(3): 454-459.
Soil Invertebrate Macrofauna Composition Within Agroforestry and Forested
Ecosystems and Their Role in Litter Decomposition in Embu, Kenya
Soil Invertebrate Macrofauna
Composition within
Agroforestry and Forested
Ecosystems and their Role in
Litter Decomposition in
Embu, Kenya
447
32
Mwangi, M.1,2*, Mugendi, D.N.2, Kung’u,
J.B2, Swift, M.J3, and Albrecht, A.1
1
International Centre for Research in Agroforestry (ICRAF),
P.O. Box 30677, Nairobi, Kenya
2
Kenyatta University, Department of Environmental
Foundations, P.O. Box 43844, Nairobi, Kenya
3
Tropical Soil Biology and Fertility Institute, P.O. Box 30677,
Nairobi, Kenya
*Corresponding author (email: mmwangi@lycos.com;t.mwangi@cgiar.org)
Abstract
Adequate food to meet the needs of an ever-increasing
population is a major challenge for most developing
countries, especially in the tropics. Despite this, few new
technical packages capable of increasing net returns without
deteriorating the environment have been developed. Crop
yields in Embu, Kenya are poor due to declining soil fertility
Mwangi, M. et al
448
prompted by continuous cropping and application of
fertilizers in non-sufficient quantities by farmers. Studies
have shown that soil biota provides the means and regulates
the transformation of organically bound nutrients into plantavailable forms through mineralization.
An experiment was conducted to investigate soil
macrofauna composition within agroforestry and forested
ecosystems and their role in litter decomposition. This was
anticipated to address poor crop yields in the study region.
The study was conducted during the long and the short rains
of the year 2000 on-station at Embu in an ongoing hedgerow
intercropping experiment. Two types of Standard PVC
litterbags with mesh size 7 mm and 1mm, were used. The 7
mm mesh size allowed macrofauna to enter while the 1 mm
excluded the macrofauna. Two types of litter: Calliandra
calothyrsus (low quality) and Leucaena leucocephala (high
quality), were placed in the litterbags in duplicate in selected
treatments of the Embu trials and were sampled at 1, 2, 4,
8, and 16 weeks. Decomposition rate constants (k) were
estimated using a non-linear module in the EXCEL
spreadsheet upon fitting first order exponential equations.
Results from the study depicted that different
management practice and/or land use affect soil
macrofauna in varied manner. Soil invertebrate macrofauna
enhanced the rate of decomposition of C. calothyrsus and
L. leucocephala litter.
Keywords: Agroforestry, Hedgerow intercropping, Litter decomposition,
Macrofauna.
Introduction
Production of adequate food to meet the needs of an ever-increasing
population is a major challenge for most developing countries and in
particular those of tropical Africa, Borlaug (2000). The focus on food
production should therefore, be widened to include the problem of how
best to conserve natural resources and biodiversity while achieving
optimum sustainable yields.
Soil fauna may affect soil function in a variety of ways, and could be
used as indicators of nutrient status of soil in a given site (Doube, 1997;
Rao et al., 1998 and Vanlauwe et al., 1996). Soil invertebrates are the
major determinants of soil processes in tropical ecosystems, whereas
pest management is an integral part of crop production, the potential
for manipulating the beneficial soil animals has rarely been considered
Soil Invertebrate Macrofauna Composition Within Agroforestry and Forested
Ecosystems and Their Role in Litter Decomposition in Embu, Kenya
449
in designing management practices (Lavelle et al., 1994b). Practices
that eliminate beneficial soil faunal communities are unlikely to
contribute to the sustainable production in the long term, especially in
low-input systems based on organic residues. Thus focus on food
production should be widened to include the problem of how best to
conserve natural resources and biodiversity while achieving optimum
sustainable yields.
According to ICIPE (1997), the diversity and role of soil fauna have
been largely ignored by traditional and conventional agriculturists due
to limited knowledge on their impact on crop yields. In recent years,
many well-documented articles and reports have established the
importance and urgency of improved knowledge and management
practice for tropical soils. There is relatively little data and information
available regarding tropical soil biology. Moreover, technological
developments in temperate zones may not be applicable or appropriate
in the tropics. According to Bruyn (1997), soil degradation in the tropics
is related to drastic decline in activity and diversity of soil fauna among
other aspects. The challenge in the future will therefore be to shift the
emphasis of soil fauna research towards understanding their function
in soil processes essential to ecosystem functioning. The soil biota,
including soil microbial biomass and soil fauna provide the means and
regulate the transformation of organically bound nutrients into plantavailable forms through mineralization (Vanlauwe et al., 1996; Lavelle
et al., 1994 and Tian et al., 1997).
The process of litter decomposition is critical for maintaining the
functioning of natural and managed ecosystems. This process occurs
with partial involvement of soil invertebrates in the terrestrial ecosystems.
Mugendi (1997) pointed out that studies on how litter quality affects
decomposition in agroforestry systems are scanty. Studies done
elsewhere depicts that the attributes of litter decomposition are
determined by litter traits and climatic conditions (Thomas et al., 1993;
and Kochy, et al., 1997). According to Upadhyay and Singh (1989),
decomposition could be regulated by variables such as decomposer
communities among others. Studies have also shown that rates and
patterns of litter decomposition can be described as a function of season,
climate and the conditions within the soil environment (Kwabiah et al.,
1999 and Mafongoya et al., 2000).
This study mainly investigated the role of soil invertebrate
macrofauna in litter decomposition within a hedgerow intercropping.
Little research has been done on this aspect thus a need to undertake a
study on the same. With this sort of experimental evidence, scientists
can indicate to the farmer the state of the soil resource. The study
specifically investigated the role of soil macrofauna on the rate of litter
decomposition and compared the rate of litter decomposition of C.
calothyrsus and L. leucocephala.
450
Mwangi, M. et al
Materials and Methods
Experimental site
The study was conducted at the National Agroforestry Research Project
(NAFRP) site at the Kenya Agricultural Research Institute (KARI) Regional
Research Centre, Embu district in the Eastern province of Kenya. The
centre is in the central highlands of Kenya on the southeastern slopes
of Mt. Kenya at 0° 30’S, 37o 30’E and an altitude of 1480 m. The average
maximum temperature is 25°C; the minimum is 14°C while the longterm monthly temperature is 19.5°C. The area receives a total annual
rainfall of between 1200 and 1500 mm in two distinct seasons: long
rains (March to June) average of 650mm and the short rains (mid October
to December) average of 450 mm. The soils are mainly Humic Nitisols
(FAO-UNESCO, 1989), derived from basic volcanic rocks (Jaetzold and
Schmidt, 1983). They are deep, well weathered with friable clay texture
with moderate to high inherent fertility.
Experimental treatments
Calliandra calothyrsus and L. leucocephala, were the two tree species
selected for this experiment. The two hedgerow species had been
identified as two of the most appropriate species for soil fertility
management (Heinemann et al., 1990). The hedgerows were planted in
April 1992 while the application of experimental treatments started in
the long rain season of March 1993. There were ten (10) treatments
replicated three (3) times in randomized complete block design as shown
in Figure 32.1.
Management of tree hedges and pruning incorporation
Calliandra calothyrsus and L. leucocephala tree hedges were lopped
two days before maize was planted. Hedges were lopped at a height of
50 cm using sharp knives. Leafy biomass and succulent stems were
separated from hardened stems that were removed for firewood. The
leafy biomass were weighed, chopped into smaller pieces (5 to 10 cm)
and spread evenly on the ground over the plot area. They were then
incorporated in the soil using hand hoes in the plots that were designed
to receive pruning (Figure 32.1) as the land was being prepared for
maize planting. The rate of leafy biomass of each tree species applied
to different treatments was approximately 2 Mg ha -1 season -1 on dry
weight basis.
Soil Invertebrate Macrofauna Composition Within Agroforestry and Forested
Ecosystems and Their Role in Litter Decomposition in Embu, Kenya
451
Figure 32.1: Field and plot layout, and experimental treatments within hedgerows at the
National Agroforestry Research Project (NAFRP) site in Embu, Kenya
REP 1
P1
T8
P2
T5
P3
T9
P4
T4
P5
T6
P6
T3
P7
T1
P8
T2
P9
T10
P10
T7
REP 2
P20
T1
P19
T4
P18
T6
P17
T3
P16
T10
P15
T2
P14
T5
P13
T8
P12
T7
P11
T9
REP 3
P21
T6
P22
T3
P23
T2
P24
T1
P25
T7
P26
T10
P27
T4
P28
T9
P29
T5
P30
T8
N
MET
STATION
Crop Cropping
System
Tree Species
T1
T2
T3
T4
T5
T6
T7
Maize
Maize
Maize
Maize
Maize
Maize
Maize
Intercrop
Intercrop
Intercrop
Intercrop
Monocrop
Monocrop
Monocrop
Incorporation
Yes
Yes
Removed to trt 5
Removed to trt 6
Imported from trt 3
Imported from trt 4
Imported from outside
at rate of trt 6
T8 Maize Monocrop L. leucocephala Imported from outside
at rate of trt 6
T9 Maize Monocrop None
None
T10 Maize Monocrop None
None
C. calothyrsus
L. leucocephala
C. calothyrsus
L. leucocephala
C. calothyrsus
L. leucocephala
C. calothyrsus
Fertilizer
9m
4.5m
None
None
None
None
None
None
10m
50kg.N ha-1
50kg.N ha-1
50kg.N ha-1
None
6 maize
rows
Hedgerows
NB: Diagram not drawn to scale
P = Plot; T = Treatment
Sampling of soil invertebrate macrofauna
Sampling for macrofauna was done six weeks after the incorporation of
the litter biomass. Using a monolith of size 25 cm x 25 cm x 30 cm,
samples were taken in two seasons; the long and the short rainy seasons.
At each observation, five samples were taken randomly from each plot
three times per season. The monolith was placed over a randomly selected
spot and using a metallic mallet, it was driven into the soil to the ground
level. The soil from the monolith was removed by hand depthwise at 010 cm, 10-20 cm and 20-30 cm depths into plastic buckets. The soil
from each depth was placed in different plastic trays (20 cm by 30 cm)
and gently sorted out to locate the organisms. The organisms were
separated into major taxonomic groups and then collected in glass and
plastic bottles using a pooter. After sorting, soil was returned to the
sampling sites to minimize site degradation. In the laboratory, counting
and recording was done. Numbers of different category of organisms
were expressed per metre square. After counting, the soil fauna were
preserved in 75% alcohol for subsequent identification at the Department
of Entomology, National Museums of Kenya, Nairobi.
452
Mwangi, M. et al
Decomposition of incorporated litter within hedgerow
agroecosystem
The experiment on litter decomposition was conducted within a hedgerow
intercrop to investigate the role of soil macrofauna in litter decomposition
and the relationship between resource quality and the rate of
decomposition.
Two types of plastic (polyvinyl) bags with mesh size 1 mm and 7 mm
respectively with an envelope configuration were used. The 7 mm mesh
size allowed macrofauna to enter while the 1 mm excluded the macrofauna.
The sides of the litterbags were bent to retain the shape of shallow box
like container to prevent compression of the enclosed litter and also to
allow or exclude free access to most macrofauna groups. 100 g (fresh
weight) of L. leucocephala or C. calothyrsus was placed into the bags and
the open edges of the bags then sealed with nylon thread and the litter
spread evenly within the bags. Ninety (90) bags of each mesh size were
buried in a completely randomized design (CRD), horizontally in the soil
at a depth of 15 cm with the subtreatments replicated twice to allow
retrieval of two litterbags (one of 1mm and another of 7mm), per each
plot at 1, 2, 4, 8, and 16 weeks after incorporating the litter for dry matter
analyses within the plots whose treatments involved litter incorporation
(treatments 1, 2, 5, 6, 7, and 8). Throughout the period of the experiment,
the experimental area was kept free of weed by hand weeding.
Dry matter loss analyses of Calliandra calothyrsus and
Leucaena leucocephala
At each sampling, the soil attached to the litterbag was carefully removed
and the litter was put in polythene bags and taken to the laboratory,
where soil and organic debris were sorted out by hand from the
decomposing plant materials. Samples were then cleaned and oven dried
at 65° C to a constant weight for dry weight determination (Anderson
and Ingram, 1993). The dry weights were expressed as percentage of
the initial sample weight at time zero. Decomposition rate constants (k)
were estimated using Wieder and Lang (1982), first order exponential
equation:
LR/LI = e - kt
Where: LR = litter remaining after a given time.
LI = initial litter weight at time zero.
t = time interval of sampling LR expressed in weeks.
k = rate constant (decomposition rate constant per week).
e = base of natural logarithm.
The k values were estimated using a non-linear module in the EXCEL
spreadsheet.
Soil Invertebrate Macrofauna Composition Within Agroforestry and Forested
Ecosystems and Their Role in Litter Decomposition in Embu, Kenya
453
This exponential model was considered to be close to the biological
reality where the decomposition rate of fresh litter is rapid when
hydrosoluble compounds are leached, but subsequently decrease over time.
Nutrients attributes of Calliandra calothyrsus and Leucaena
leucocephala
The oven-dry samples of fresh plant samples taken at the onset (time zero)
and during the experiment, were ground in a Wiley mill to pass through a
0.5 mm sieve. Sub-samples of the ground litter were analyzed for total
nitrogen, phosphorus, potassium, calcium, and magnesium using ICRAF
laboratory methods (ICRAF, 2000). Lignin contents were analyzed according
to the methods of Rowland and Roberts (1994) and polyphenols by
procedures detailed in the TSBF Handbook (Anderson and Ingram, 1993).
One gram (1g) of ground plant samples was ashed in a muffle furnace
at 500°C for four hours to correct for soil contamination, when the
samples were buried in the soil. The ashed samples were reweighed
and percentage ash content determined as shown below:
Percentage Ash = [(crucible + unashed sample) - (crucible + ashed
sample)] [(crucible + unashed sample) - (crucible weight)] x 100
The nutrient values were corrected on the basis of ash-free weight:
Ash free weight per g of material = 1 - 0.01x % ash.
Percentage Corrected value for nutrient (N, P, K, Ca, and Mg) = 0.01
x % nutrient (N, P, K, Ca, and Mg)
Decomposition over time was calculated following the formula by
Giashudin et al. (1993):
Percentage of dry weight remaining = (DWt)/(Dwi) ◊ 100
Where: DWt = oven dry weight at time t and
Dwi = initial oven dry weight.
Results and Discussions
Soil invertebrate macrofauna abundance within the
hedgerow agroecosystem
The macrofauna observed during the period of study were identified
into their respective groups/orders. Whenever possible the fauna were
identified up to the species level, but for some it was not possible as
they were still in their juvenile stage and therefore indicated as not
identified (NI). Different macrofauna groups were observed in varying
Mwangi, M. et al
454
numbers during the study period as shown in Table 32.1. The hedgerow
agroecosystem recorded several distinct groups of macrofauna. This
could have been as a result of the region offered a wide range of habitat
for diverse faunal groups and therefore, it could be a rich ecosystem.
Table 32.1: Macrofauna groups observed within hedgerow agroecosystem during the
long and the short rain seasons of the year 2000 and 2000/2001 respectively in Embu,
Kenya
Group/Order
Family/Subfamily
1. Myriapoda
(Millipedes)
2. Coleoptera
(Beetles)
1.
2.
3.
3. Hymenoptera
(Ants)
Scarabidae/aphodina
Staphylinidae.
Carabidae.
Formicidae/Myrmacinae
Genera/Species
NI*
NI*
1. Aphodius ividus L.
(chaffer grub).
2. Philanthus sp.
(dark tiny beetles).
3. Hyparpulus ornatus Per.
1.
2.
3.
4. Acarina (Mites)
5. Chilopoda
(centipedes)
6. Aranae (Spiders)
7. Isoptera (Termites)
8. Diptera (Flies)
9. Lepidoptera (Moths)
NI*
NI*
Agriopidae
Termitinae/
Macrotermitinae
NI*
NI*
Bothroponera sp.
(big dark ants)
Euponera sp.
(brownish and small).
Anoma sp. (red ants).
NI*
NI*
Araneus dradematus L.
Microtermes pusillas
Wasmann (tiny termites)
NI*
NI*
* Not identified
Relatively higher numbers of fauna were observed within the
hedgerows agroecosystem (Table 32.2) compared to the forested site
(Table 32.3). This could have been because the macrofauna were able
to utilize the benefits accrued from combining trees with crops than
with trees alone as observed within the forested site.
Isopterans were the most abundant of the macrofaunal observed
followed by Hymenopterans, Lepidopterans, Coleopterans, Chilopoda,
Aranae, Myriapoda, Acarinas and Dipterans in that order. It was evident
that in the hedgerow agroecosystem, termites formed the major
macrofaunal group contributing 76.5% of the total macrofauna observed
as depicted in Table 32.2.
Soil Invertebrate Macrofauna Composition Within Agroforestry and Forested
Ecosystems and Their Role in Litter Decomposition in Embu, Kenya
455
Table 32.2: Total macrofaunal counts and percentages observed within hedgerow
agroecosystem during the long and the short rain seasons of the year 2000 and 2000/
2001 respectively in Embu, Kenya
Faunal group
Total counts (m-2)
Total counts (%)
Isoptera
Hymenoptera
Lepidoptera
Coleoptera
Chilopoda
Aranae
Myriapoda
Acarina
Diptera
22406
3344
1466
1126
403
205
166
106
77
76.5
11.4
5.0
3.8
1.4
0.7
0.6
0.4
0.3
Table 32.3: Total macrofaunal counts and percentages observed within the forest
ecosystem during the long and the short rain seasons of the year 2000 and 2000/2001
respectively in Embu, Kenya.
Faunal group
Total counts (m-2)
% Total counts
Isoptera
Hymenoptera
Lepidoptera
Coleoptera
Chilopoda
Aranae
Myriapoda
Acarina
Diptera
8470
705
1568
934
368
213
596
106
96
64.87
5.4
12
7.2
2.8
1.6
4.6
0.8
0.7
The presence of high number of termites in the hedgerow
agroecosystem could imply that they were better able to withstand
disturbed conditions as well as diminishing food resources resulting
from such disturbances. It could also have been that the termites being
ecosystem engineers, influenced the access of litter to other faunal groups
hence their abundance over the rest. Termites may as such be able to
survive a wide range of conditions. This corroborates with the work
done by Christopher (1994) that showed that the influence of termites
is not confined to certain litter qualities and that they control the
accessibility of litter to other decomposers to an extent that exceeds
their influence by direct consumption.
Total macrofauna abundance varied significantly (p<0.05), across
treatments with higher numbers of fauna being recorded for treatments
with C. calothyrsus than with L. leucocephala over the sampling period
456
Mwangi, M. et al
(Figure 32.2) within the hedgerows. Treatments involving C. calothyrsus,
biomass incorporation plus fertilizer (Treatment 7), recorded highest
numbers of fauna.
Figure 32.2: Macrofauna counts within hedgerows during the long and the short rain
seasons of the year 2000 and 2000/2001 respectively in Embu, Kenya
Key:
1 & 2 = Alley cropping of Calliandra calothyrsus and Leucaena leucocephala respectively
with their respective prunings incorporated and no fertilizer applied.
3 & 4 = Alley cropping of Calliandra calothyrsus and Leucaena leucocephala respectively
with no prunings incorporated and no fertilizer applied.
5 & 6 = Maize only, no alley cropping, prunings of Calliandra calothyrsus and Leucaena
leucocephala respectively with their respective prunings incorporated from outside and
no fertilizer applied.
7 & 8 = Maize only, no alley cropping, prunings of Calliandra calothyrsus and Leucaena
leucocephala respectively with their respective prunings incorporated from outside and
fertilizer applied
9 &10 = Maize only, no alley cropping with and without fertilizer applied respectively
fertilizer applied
Litter decomposition as influenced by soil invertebrate
macrofauna within hedgerow intercropping in Embu, Kenya
Decomposition and nutrient release of the litter biomass are the key
processes by which nutrients locked up in plant parts eventually become
available to crops. The processes are regulated by variables such as the
quality of the litter, climate, soil properties and decomposer communities
(Upadhyay and Singh, 1989). Therefore, understanding the influence of
these variables on biomass decomposition and nutrient release is a vital
step to better management of organic inputs that are applied in different
agroecosystems (Mafongoya et al., 1997).
Soil Invertebrate Macrofauna Composition Within Agroforestry and Forested
Ecosystems and Their Role in Litter Decomposition in Embu, Kenya
457
Nutrient analyses of the litter used in the decomposition study
depicted that C. calothyrsus and L. leucocephala had varied nutrient
content as shown in Table 32.4 and therefore, they could be varied in
terms of nutritional value hence resource quality. The two species had
varied lignin and/or polyphenols to nitrogen ratios (Table 32.5). T-test
indicated that nutrient concentrations of the two species were
significantly different (p < 0.05).
Table 32.4: Average chemical composition of Calliandra calothyrsus and Leucaena
leucocephala within hedgerows in the year 2000/2001in Embu, Kenya.
Material
%N
%P
%K
%Ca
%Mg
%
Lignin
%Polyphenol
Calliandra calothyrsus
Leucaena leucocephala
SED
2.8
2.8
0.02
0.1
0.1
0.01
0.6
1.9
0.06
1.2
1.3
0.04
0.4
0.3
0.01
13.4
9.5
0.10
11.2
8.1
0.11
Table 32.5: Lignin and/or polyphenols to Nitrogen ratios of Calliandra calothyrsus and
Leucaena leucocephala within hedgerows in the year 2000/2001in Embu, Kenya
Plant Material
Lig/N
Pp/N
(Lig +Pp)/N
Calliandra calothyrsus
Leucaena leucocephala
SED
4.8
3.4
0.02
4.0
2.9
0.06
8.9
6.3
0.01
Key:
N = Nitrogen, Lig = Lignin, Pp = Polyphenol
In Figure 32.3, it is evident that L. leucocephala decomposed at
relatively faster rate than C. calothyrsus, save for the second season
when C. calothyrsus decomposed at faster rate than L. leucocephala
during the second and the fourth week . The pattern of litter
decomposition was gradual in season one and drastic in season two.
This could have been due to different abiotic conditions in terms of
moisture and temperatures experienced and therefore varied faunal
population within treatments involving the two species. These findings
agree with the observations made by Mugendi et al. (1994) in Machakos
district of Kenya, during the short rains and the long rains, which gave
an indication that faunal decomposition could be having some relation
to climatic conditions and resource quality.
There was some litter remaining for both the species even after the
16th week for the first season, whereas all the litter had decomposed by
the 8th week in the second season. The rate of litter decomposition was
458
Mwangi, M. et al
relatively slow within the first four weeks of the first season, after which
it became fast. During the second season, the rate was slower for the
first two weeks after which it proceeded at a faster pace.
Figure 32.3: Calliandra calothyrsus and Leucaena leucocephala decomposition within
hedgerows during the long and the short rain seasons of the year 2000 and 2000/2001
respectively in Embu, Kenya
The varied rate of decomposition of L. leucocephala and C.
calothyrsus could have been due to the varied litter substrate quality
in which case, the former may be seen to be of higher quality than the
latter as it had higher levels of nitrogen and phosphorus, and lower
ratios of lignin and/or polyphenols to nitrogen. Lignin is known to be
highly resistant to microbial decomposition, according to studies by
Melillo et al. (1982) and Chesson (1997). This is also in agreement for
instance with observation made by Thomas et al. (2000) that litter
with low levels of lignin decomposes faster. According to Vityakon et
al. (2000), Polyphenols exhibits a significant influence on nitrogen
release from litter biomass hence decomposition. These findings are
in agreement with studies conducted by Bubb et al. (1998), which
indicated that litter-mass loss is strongly correlated with litter quality
indicators such as nitrogen, phosphorus, carbon to nitrogen ratio,
lignin and polyphenolics.
Overall, the rate of decomposition of L. leucocephala and C.
calothyrsus was faster in the second season than in the first season.
This could have been due to the presence of fully established crop that
Soil Invertebrate Macrofauna Composition Within Agroforestry and Forested
Ecosystems and Their Role in Litter Decomposition in Embu, Kenya
459
might have increased the decomposition of the residues of the prunings.
This corroborates to findings by Vanlauwe et al. (1997) that crop cover
may increase decomposition and nitrogen release of the residues.
The litter enclosed in 7mm litterbags, decomposed at a faster rate
than that in the 1-mm litterbags and the rates were higher in season
two than in season one for both C. calothyrsus and L. leucocephala as
depicted in Figure 32.4 and 32.5 respectively. These variations in
decomposition could have been due to varied effects of the decomposers
giving an indication that the presence of soil invertebrate macrofauna
could have promoted the rate of litter decomposition. Season one was
a dry season, and although the faunal biomass and counts was high
during this specific season, the diversity was low (Table 32.6) as
opposed to the second season where higher diversity of organisms
was recorded. Therefore, higher faunal population and biomass could
have enhanced decomposition in season one. This corroborates the
studies by Rusek (1998), Gupta et al. (1998) and Beck, (2000) that
fauna play an important role in plant litter decomposition processes.
Figure 32.4: The influence of litterbag mesh size on Calliandra calothyrsus decomposition
within hedgerows during the long and the short rain seasons of the year 2000 and 2000/
2001 respectively in Embu, Kenya
Key: LL = Leucaena leucocephala;
S1 = season 1;
S2 = season 2
460
Mwangi, M. et al
Figure 32.5: The influence of litterbag mesh size on Leucaena leucocephala
decomposition within hedgerows during the long and the short rain seasons of the year
2000 and 2000/2001 respectively in Embu, Kenya
Key: LL = Leucaena leucocephala;
S1 = season 1;
S2 = season 2
The moist conditions and favorable temperatures (Table 32.7),
coupled with community composition in terms of diversity, may have
played a great role in enhancing the rate of decomposition in the second
season as opposed to the first season. Therefore, both the biotic and the
abiotic factors could have influenced litter decomposition, though,
decomposition may have been influenced more by the faunal diversity
as opposed to faunal abundance. Different feeding habits resulting from
the higher diversity recorded in the second season may have also
promoted the rate of litter decomposition unlike in season one where
the diversity was low.
During the second season, it could be that decomposition of C.
calothyrsus was enhanced by more suitable moisture, temperature and
soil condition that prevailed. Therefore, with favorable states of these
abiotic factors, C. calothyrsus would have decomposed at a rate close or
similar to that of Leucaena leucocephala. Overall results depicted that
L. leucocephala decomposed and released nutrients faster than C.
calothyrsus. This could be due to the varied chemical concentration as
previously depicted in Table 32.4, hence varied nature of resource
quality. Leucaena leucocephala had higher levels of nitrogen,
phosphorus, potassium and calcium but low levels of magnesium and
both polyphenols and lignin as opposed to C. calothyrsus. Higher lignin
levels in C. calothyrsus than in Leucaena leucocephala may have slowed
its rate of decomposition and may be even the release of nutrients.
Soil Invertebrate Macrofauna Composition Within Agroforestry and Forested
Ecosystems and Their Role in Litter Decomposition in Embu, Kenya
461
Table 32.6: Macrofauna diversity as indicated by Shannon-Wiener index within hedgerows
during the long and the short rain seasons of the year 2000 and 2000/2001 respectively
in Embu, Kenya.
Treatment
Season 1
Shannon-Wiener Index
Season 2
Shannon-Wiener index
-0.043(0.675)
-0.064(0.658)
-0.037(0.679)
-0.000(0.707)
-0.055(0.666)
-0.085(0.642)
-0.034(0.682)
-0.076(0.649)
-0.032(0.683)
-0.013(0.698)
-0.071(0.654)
-0.119(0.616)
-0.061(0.661)
-0.090(0.639)
-0.093(0.636)
-0.095(0.635)
-0.037(0.681)
-0.087(0.641)
-0.087(0.619)
-0.133(0.605)
(0.020)
(0.019)
1
2
3
4
5
6
7
8
9
10
SED (treatment)
Season 1: F test: P = 0.05;
Season 2: F test: P = 0.047
Values in parentheses are square root {(x + 0.5)} transformed.
1 & 2 = Alley cropping of Calliandra calothyrsus and Leucaena leucocephala
respectively with their respective prunings incorporated and no fertilizer applied.
3 & 4 = Alley cropping of Calliandra calothyrsus and Leucaena leucocephala
respectively with no prunings incorporated and no fertilizer applied.
5 & 6 = Maize only, no alley cropping, prunings of Calliandra calothyrsus and Leucaena
leucocephala respectively with their respective prunings incorporated from
outside and no fertilizer applied.
7 & 8 = Maize only, no alley cropping, prunings of Calliandra calothyrsus and Leucaena
leucocephala respectively with their respective prunings incorporated from
outside and fertilizer applied
9 &10 = Maize only, no alley cropping with and without fertilizer applied respectively
fertilizer applied
Table 32.7: Average monthly temperature and total rainfall during the long (first season)
and the short rain (second season) seasons of the year 2000 and 2000/2001 respectively
in Embu, Kenya
Season
Temperature °C
Rainfall mm
1
2
18.7
17.5
262.0
340.9
Total
18.3
602.9
462
Mwangi, M. et al
The higher concentrations of polyphenols in C. calothyrsus than in
L. leucocephala could have caused immobilization of nutrients. Lignin
and polyphenolics are known to be highly resistant to microbial
decomposition according to studies by Melillo et al. (1982), Chesson,
(1997) and Mafongoya et al. (2000). Therefore, based on these varied
nutrients concentrations L. leucocephala is of higher quality than C.
calothyrsus and this may explain why the former decomposed and
released nutrients at a faster rate than the latter. Mafongoya et al. (1998)
and Handayanto et al. (1997), made similar observations that the
potential of the organic inputs from agroforestry species to supply
nutrients depends on their quality and different tree species could be
having varied chemical constituents. The results are in agreement with
Hamada et al. (2000) that litter decomposition is affected by lignin content
and that the relationship between lignin content and the decomposition
rate is inverse. Supply of nutrients by an organic input is largely
determined by the rate at which such organic decomposes and therefore,
L. leucocephala may be of better quality than C. calothyrsus.
Generally, the decomposition rates were faster in the second season
(which received some rains) as opposed to the first season (where rains
failed). This could have been due to increase in moisture and faunal
activity, which promoted the release of nutrients hence rate of
decomposition. Decomposition and release of nutrients contained in
Leucaena leucocephala and Calliandra calothyrsus is thus determined
by their respective quality and/or the environment and the decomposer
organisms present.
Conclusions and Recommendations
Results from this study depicted that different management practice
and/or land use affect soil macrofauna in varied manner. Soil
invertebrate macrofauna populations are high in management practices
that entail incorporation of organic material into the soil as opposed to
those that do not. Agroforestry system enhances the fauna population
unlike forested ecosystems.
Soil invertebrate macrofauna enhanced the rate of decomposition of
C. calothyrsus and L. leucocephala litter. Leucaena leucocephala
decomposed faster than C. calothyrsus and the former had lower lignin
and/or polyphenols to nitrogen ratio than the latter. Therefore, the rate of
litter decomposition and nutrients release is related to tree species hence
resource quality. Leucaena leucocephala could be more suitable for use to
improve maize yields in alley cropping compared to C. calothyrsus.
Farmers should therefore be encouraged to use L. leucocephala as a
source of nutrients for agricultural crops. Calliandra calothyrsus may
Soil Invertebrate Macrofauna Composition Within Agroforestry and Forested
Ecosystems and Their Role in Litter Decomposition in Embu, Kenya
463
be applied at some predetermined time before sowing, to make the
nutrients released from the same be utilized by the growing crop. There
is also a need to investigate litter decomposition trends in an agroforestry
setting in arid and semi-arid lands, which forms the highest percentage
of the Kenyan land.
References
Anderson, J.M. and Ingram, J.S.I. (1993) Tropical Soil Biology and Fertility: A
Handbook of Methods. (2nd edn). Wallingford, UK: CAB International.
Beck, L. (2000) Litter decomposition and soil fauna in temperate and tropical
forests. In Carolinea 58: 243-256.
Borlaug, N. (2000) Meeting Africa’s food production challenges in the 21st
century. Paper presented on June 4th, 2000 by the President, Sasakawa
Africa Association at ICRAF Nairobi, Kenya.
Bruyn, L.A. (1997) The status of Soil Microfauna as indicators of soil health to
monitor the sustainability of Australian Soils. Ecological Economics
Amsterdam 23:167-178.
Bubb, K. A. Xu, Z. H. Simpson, J. A and Saffigna, P.G. (1998) Some nutrient
dynamics associated with litterfall and litter decomposition in hoop pine
plantations of southeast Queensland, Australia. Forest Ecology and
Management 110: 343-352.
Chesson, A.(1997) Plant degradation by ruminants: Parrallels with litter
decomposition in soils. In Cadisch G and Giller (eds). Driven by nature:
Plant litter quality and decomposition. CAB International, Wallingford, UK.
Christopher, M. (1994) Diversity and Ecology of Termites in Amazonian forests.
Pedobiologia 38: 407-428.
Doube, B.M. (1997) Can the Abundance or Activity of Soil fauna be used to
indicate the Biological health of soils? Biological indicators of soil health.
CAB International; Wallingfors; UK. Pp. 265-295.
F.A.O. (1989) Forestry and food security. FAO forestry paper, FAO (UN), 90:1-2.
Giashuddin, M. M., Garrity, D.P. and Aragon, M.L. (1993) Weight loss, nitrogen
content changes, and nitrogen release during decomposition of legume tree
leaves on and in the soil. Nitrogen Fixing Tree Research Reports 11:43-50.
Gupta, P.K and Oli, B.P. (1998) Consumption and assimilation of evergreen oak
litter by the slug Anadenus altivagus in Kumaon Himalayan Forests, India.
Ecoscience 5: 494-501.
Hamada, Z., Steinberger, Y., Kutiel, P., Lavee, H. and Barness, G. (2000)
Decomposition of Avena sterilis litter under arid conditions. Journal of Arid
Environments 46: 281-293.
Handayanto, E. Giller, K. E. and Cadisch, G. (1997) Regulating nitrogen release
from legume tree prunings by mixing residues of different quality. In Soil
Biology and Chemistry. Vol. 29(9-10). Pp 1417-1426.
464
Mwangi, M. et al
Heinemann, A. M., Mengich, E. K., Olang, A. D., and Otieno H. J. O. (1990)
Agroforestry Research Network for Africa Report no. 27. Maseno progress
report for the period January 1988 to January 1990. ICRAF, Nairobi, Kenya.
ICIPE, (1997) International Centre for Insect Physiology and Ecology. Annual
Report Highlights. pp 2, Nairobi, Kenya.
ICRAF, (2000) International Centre for Research on Agroforestry. Laboratory
methods for soil and plant analysis, Nairobi, Kenya.
Jaetzold, R. and Schmidt, H. (1983) Farm handbook of Kenya: National conditions
and farm management information of East Kenya (Eastern and Coast
provinces), Volume II A: 37-38. Typo-druck, Rossdorf, West Germany.
Kochy, M. and Wilson, S.D. (1997) Litter decomposition and nitrogen dynamics
in Aspen forest and mixed-grass prairie. Ecology. Washinghton-D-C.
l78: 732-739.
Kwabiah, A.B. Voroney, R.P. Palm, C.A. and Stoskopf, N.C. (1999) Inorganic
fertiliser enrichment of soil. Effect on decopmsition of plant litter under
subhumid tropical conditions. Biology and Fertility of Soils 30: 224-231
Lavelle, P., Dangerfield, M., Fragoso, C., Eschnebrenner, V., Hernadez, D.L.,
Pashanasi, B. and Brussard, L. (1994a) The relationship between soil
macrofauna and tropical soil fertility. In: Woomer, P.L. and Swift, M.J. (eds).
The Biological Management of Tropical Soil Fertility. Pp 136-169. A WileySayce publication.
Lavelle, P., Gilot, C., Fragoso, C. and Pashanasi, B. (1994b) Soil fauna and
sustainable land use in the humid tropics. In: Greenland, G.A. and Szalbolcs,
I. (eds), Soil Resilience and Sustainable Land use 18:291-307. CAB
International, Wallingford.
Mafongoya, P.L., Barak, P. and Reed, J.D. (2000) Carbon, Nitrogen and
phosphorus mineralisation from tree leaves and manure. Biology and Fertility
of soils 30: 298-305
Mafongoya, P.L., Nair, P.K.R. and Dzowela, B.H. (1997) Multipurpose tree
prunings as a source of nitrogen to maize under semiarid conditions in
Zimbabwe. 3. Interactions of pruning quality and time and method of
application on nitrogen recovery by maize in two soils types. Agroforestry
systems 35: 57:70.
Mafongoya, P.L. Giller, K.E. and Palm, C.A. (1998) Decomposition and nitrogen
release patterns of tree prunings and litter. Agroforestry Systems .
38: 77-97.
Melillo, J.M. Aber, J.D. and Muratore, J.F. (1982) Nitrogen and lignin control of
hardwood leaf litter decomposition dynamics. Ecology 63: 621-626.
Mugendi, D.N. (1997) Tree-Biomass Decomposition, Nitrogen Dynamics, and
Maize Growth under Agroforestry Conditions in the subhumid Highlands
of Kenya. In, thesis submitted to the school of Forest Resources and
Conservation, University of Florida.
Mugendi, D.N. Mochoge, B.O. Coulson, C.L. Stigter, C.J. and Arap Sang, F.K.
1994. Decomposition of Cassia siamea loppings in semiarid Machakos,
Kenya. Arid Soil Research and Rehabilitation 8: 363-372.
Rao, M.R., Niang, A., Kwesiga, F.R., Duguma, B., Franzel, S., Jama, B. and
Soil Invertebrate Macrofauna Composition Within Agroforestry and Forested
Ecosystems and Their Role in Litter Decomposition in Embu, Kenya
465
Buresh, R. (1998) Soil fertility replenishment in sub-Saharan Africa: New
techniques and the spread of their use on farms. A groforestry Today
10: 3-8.
Rowland, A..P. and Roberts, J.D. (1994) Lignin and cellulose fraction in
decomposition studies using acid-detergent fibre methods. Commum. Soil
Sci. Plant 25: 269-277.
Rusek, J. (1998) Biodiversity of collembola and their functional role in the
ecosystem. Biodiversity and Conservation 7: 1207-1219.
Thomas, K.D. and Prescott, C.E. (2000) Nitrogen availability in forest floors of
three tree species on the same site: The role of litter quality. Canadian
Journal of Forestry Research 30: 1698-1706.
Thomas, R.J., and Asakawa, N.M. (1993) Decomposition of leaf litter from tropical
forage grasses and legumes. Soil Biology and Biochemistry 25: 1351-1361.
Tian, G., Kang, B.T., Brussard, L. and Swift, M.J. (1997) Soil fauna mediated
decomposition of plant residues under constrained environmental and
residue quality conditions. In: Cadisch, G. and Giller, K.E. (eds), Driven by
nature: Plant litter quality and decomposition, pp 125-134, CAB International
Wallingford, U.K.
Upadhyay, V. P. and Singh. (1989) Patterns of nutrients immobilization and
release in decomposing forest litter in central Himalaya, India. Journal of
Ecology 77: 127-146.
Vanlauwe, B., Dendooven, L., Merckx, R., Vanlangenhove, G., Vlassak, K. and
Sanginga, N. (1996) Residue quality and decomposition under controlled
and field conditions. International Institute of Tropical Agriculture (IITA)
Research Highlights and Annual Report. 12: 1-6.
Vanlauwe, B., Sanginga, N., and Merckz, R. (1997) Decomposition of four
Leucaena and Senna prunings in alley cropping systems under subhumid
tropical conditions: the process and its modifiers. Soil Biology & Biochemistry
29: 131-137.
Vityakon. P. Meepech, S. Cadisch, G. and Toomsan, B. (2000) Soil organic matter
and nitrogen transformation mediated by plant residues of different qualities
in sandy acid upland and paddy soils. Netherlands Journal of Agricultural
Science. 48: 75-90.
Wieder, R. K., and Lang, G. E. (1982) A critique of the analytical methods used
in examining decomposition data obtained from litterbags. Ecology
63: 1636-1642.
466
Mwangi, M. et al
Selection of Arbuscular Mycorrhizal Fungi for Inoculating Maize and Sorghum Grown in
Oxisols/Ultisol and Vertisols in Cameroon
Selection of Arbuscular
Mycorrhizal Fungi for
Inoculating Maize and
Sorghum Grown in Oxisol/
Ultisol and Vertisol in
Cameroon
467
33
Nwaga, D.1*, The, C.2, Ambassa-Kiki,
R.2, Ngonkeu Mangaptché, E.L.1 and
Tchiegang-Megueni, C.3
1
Biotechnology Centre & Plant Biology Department,
University of Yaoundé I, Cameroon
P.O. Box 812, Yaoundé, Cameroon; E-mail:
dnwaga@uycdc.uninet.cm.
2
Institute of Research for Agriculture & Development, P.O.
Box 2123, Yaoundé-Messa, Cameroon
3
Department of Biological Sciences, University of
Ngaoundéré, P.O. Box 455, Ngaoundéré, Cameroon
*Corresponding author
Abstract
Mycorrhizal plants have been shown to be more adapted to
environmental stresses such as soil acidity, drought and low
fertility in O xisol/Ultisol or Vertisol. Three group of
experiments were carried out to identify mycorrhizal isolates
that showed acidity tolerance, that improve plant P uptake
using pearl millet and cowpea as test crops. Field studies
Nwaga, D. et al
468
were carried out to evaluate selected inoculants using maize
in humid forest zone (HFZ) on Ultisol, to compare chemical
fertilisers to bio-fertiliser using maize on HFZ in Oxisol and
three inoculation methods were tested using muskwari, a
dry season sorghum on Vertisol in savannah zones (SZ). Some
local isolates from our microbial resource bank were identified
and tested. Most of the mycorrhizal isolates were able to
germinate at pH 3.8, but only some were significantly affected
by acidity for their growth. Two isolates (Glomus clarum,
GCDM and Gigaspora margarita, GiMV) were selected out of
eleven for hyphal growth, plant dry weight and P uptake
improvement. An improvement of 75 % and 86 % on efficiency
were recorded over the control on dry weight basis for the
best isolates. Field experiment showing significant
improvement on biomass yield, grain yield and nutrient
content of inoculated maize and sorghum were observed. A
grain yield increase of 52 % to 66 % was recorded on maize
after mycorrhiza inoculation in HFZ on Oxisol/Ultisol.
Mycorrhiza inoculants produced a yield increase ranging from
7 to 48 % in sorghum on Vertisol in SZ according to the
method of inoculation. These results support the idea that
mycorrhiza inoculants may help improve on nutrient uptake,
especially phosphorus or nitrogen which are limiting factors
for cereal productivity in Oxisol/Ultisol, but also water uptake
and yield in Vertisol. What are the limitations for the use of
this technology in the tropics ? What can be done to reduce
these limitations for extension ?
Keywords: Maize, mycorrhiza inoculation, Oxisol, sorghum, Ultisol,
Vertisol
Introduction
To improve crop production in infertile African soils, chemical fertilisers
have been intensively used, organic matter have been used in many
East African countries and some soil management technologies such
as fallow or legumes cultivation have also been used.
Recently, ecologically sound technologies were recommended for
sustainable management of tropical soil. Reliance should be on biological
processes by adapting germ-plasm to advance soil conditions, enhance
soil biological activity and optimise nutrient cycling to minimise external
inputs and maximise the efficiency of their use” (Sanchez, 1994). This
approach has been developed for soil biota management for the tropics
Selection of Arbuscular Mycorrhizal Fungi for Inoculating Maize and Sorghum Grown in
Oxisols/Ultisol and Vertisols in Cameroon
469
using key functional groups approach, such as earthworm and microsymbionts (Woomer and Swift 1994; Swift, 1998).
These soil organisms may represent more than 90 % of soil biological
activity and thus contribute to nutrient cycling, soil fertility and symbiotic
processes in the rhizosphere.
Soil fungal diversity and activity have not been actively studied and
understood (Hawksworth, 1991). Mycorrhiza represent an important
group because they have a wide distribution; may contribute significantly
to microbial biomass and to soil nutrient cycling processes in plants
(Harley and Smith, 1983). Mycorrhizal associations are beneficial to
plants and thus crop productivity for sustainable agriculture (GianinazziPearson et al. 1982; Bethlenfalvay, 1992). They improve on nutrient
uptake, especially phosphorus, and also micronutrients such as zinc
or cooper, they stimulate growth substances and may reduce stresses,
diseases or pest attack (Sylvia and William, 1992; Davet, 1996; Smith
and Read, 1997).
For an appropriate use of this technology, it is necessary to select
the best inocula adapted to specific limiting environmental factors for
crop productivity. Oxisol/Ultisol are acidic soils, low in nutrient
availability, low in available phosphorus and P deficiency is common
and available P is generally less than 5-10 ppm. They have high
aluminium concentration which is highly toxic and may inhibit fungal
spore germination and symbiosis, root growth, hence plant development
and yield (Sanchez and Salinas, 1981; Janos, 1987; Robert, 1992).
Vertisol are rather alkaline rich soils with a montmorillonite swelling
clay, high cation exchange capacity (CEC) and water holding capacity
but the main problem is water availability during dry season (Barrault
et al., 1972; Robert, 1992).
Maize and sorghum are the source of carbohydrates in the humid
forest (HFZ) and the savannah zones (SZ) of Africa. In the HFZ, soil
acidity couple with aluminium and manganese toxicity, limiting maize
productivity. In Cameroon, acid soil cover 75 % of arable land and Al
toxicity account for 40-80 % reduction or 1-2 t ha -1 loss in maize yield
(Bindzi Tsala, 1987; The, 2000). In the SZ, drought and low soil fertility
due to continuous cropping in the same field for a long time, are the
main causes of low maize yields. Sorghum yield in Africa are very low,
less than 0.7 t ha-1 compared to Asia or America, 1.2.-3.7 t ha-1 (Anonyme,
1991)
According to Hayman (1987), maize and sorghum form effective
symbiotic association with indigenous arbuscular mycorrhizal (AM) fungi.
Pot experiments using maize showed that shoot dry weight increase of
biomass weight varying from 3%, 56% to 101%. (Harley and Smith,
1983). It is hypothesized that AM fungi may enhance maize tolerance to
acidity and drought in tropical soils. There is limited available data on
mycorrhizal diversity and their activity on diverse African land use
Nwaga, D. et al
470
systems, but according to Nwaga and Ngonkeu (1998), slash and burn
agriculture may seriously reduce mycorrhiza inoculum potential in
farming systems.
The main objective of this study was to isolate and select mycorrhiza
strains that are tolerant to acidic environment, enhance P uptake leading
to improved crop yield in soils of the humid forest and savannah zones
of Cameroon.
Materials and Methods
Site description and soil characteristics
Site and soil characteristics of the experimental field are summarised
in Tables 33.1 and 33.2.
Table 33.1: General characteristics of the experimental sites in Cameroon
Site
(Village)
Altitude
(m)
Ebolowa
(Nkoemvone)
500-700
Maroua
(Moutourwa)
200-500
Climate
Rainfall
(mm)
Temp.
mean
(°C)
Dominant
vegetation
(Land use)
Sub- 1500-2000
equatorial
24
Forest (Hevea
plantation)
Sahelo- 600-900
soudanian
29
Steppe with
spikes (Sorghum)
23
Forest (Fallow)
Yaoundé
500-700
Sub- 1400-1600
(Minkoameyos)
equatorial
Longitude Latitude
11°20 'E
2°90'N
14°27'E 10°11'N
11°27'E
3°51'N
Table 33.2: Physico-chemical parameters of the soil types used for experimentation in Cameroon
Site
Soil class
(% Clay
content)
pH
O.M.
CEC
(%) (cmol(+)kg -1)
N
(%)
Ebolowa 1
(Nkoemvone)
Oxisol (52)
3.9
3.03
14.4
0.27
7
2.6
Maroua
(Moutourwa)
Vertisol (45)
7.2
1.27
35.6
0.68
23
0.0
Yaoundé
Ultisol (23)
(Minkoameyos)
4.9
2.75
14.6
0.17
6
0.0
1
This site is one with the highest aluminium toxicity in the country
Avail. P
(mg kg -1)
Al
(cmol(+)kg -1)
Selection of Arbuscular Mycorrhizal Fungi for Inoculating Maize and Sorghum Grown in
Oxisols/Ultisol and Vertisols in Cameroon
471
These rich clay soils are also high phosphates fixing from 1600 to
2000 ppm. (Barrault et al., 1972; Robert, 1992). Vertisol represents
more than 1,200,000 ha in North Cameroon, their clay content is up
from 35 %, to 70 %, the CEC is up to 35 meq/100 g soil and available P
is generally more than 20 ppm (Seiny Boukar, 1980).
Source of mycorrhizal strains (isolates)
Arbuscular mycorrhizal spores used were isolated from soils obtained
from four agroecological regions of Cameroon (Bafia, Ngaoundere, Ndupe
and Ekona). Physical and chemical characteristics of these soils are
given on Table 33.3.
After culture were trapped on millet (Pennisetum americanum) and
cowpea (Vigna unguiculata), spores were extracted by wet sieving and
decanting methods (Schenk, 1982; Sieverding, 1991; Brundrett et al.,
1996). Pure cultures were obtained and multiplied on Brachiaria
decumbens plants in pot containing sterilised soil and sand mixture
(2:1), allowed to produce each inoculum with high number of similar
spores per species (more than 10 000). The original strains are: GGMN1
(Ndupe ), GMDE1 ( Ekona), GACB1 ( Bafia ), SGMN1 and GMVN1
(Ngaoundere) (Table 33.3). A standard pure culture of GCXH were from
the University of Hawaii (USA). The soil-based-inocula from our micro
organisms bank were air dried and packaged for longterm storage and
tests (Nwaga et al., 2000). Three experiments were carried out to select
AM fungi and assess their potential on cereal crops.
Experiment I. In vitro selection of acid tolerant AM fungi
strains
A pH ranging between 3.8 to 6.7 were obtained by using pH indicators
mixed to the culture media. These range of pH correspond to most
Cameroonian soils. Spores were sterilised in 2% chloramine T (Sigma)
and 0.025 % of sulphate streptomycin for 20 min for each antibiotic
(Schenck, 1982). Antibiotic solutions were prepared during the day of
experiment and sterilised by millipore filtration. (0.22 µm). Then the
spores rinsed 3 times in distilled sterilised water and placed to germinate
on agar medium (Difco) at 0.7 % in Petri dish over filter paper square (9
mm x 9 mm), sterilised at 120°C for 20 min. Coloured indicators (at
0.02 % concentration) were mixed with agar medium at different pH:
bromophenol blue green in colour at pH 3.8, the methyl red orange in
colour at pH 5.3 and bromocresol purple of reddish colour at pH 6.0.
Agar media at pH 4.5 and 6.7 did not receive coloured indicator. For
each pH value, spores of 5 species were tested.
Nwaga, D. et al
472
Table 33.3: Soil pH, phosphorus, origin and species of arbuscular mycorrhizal fungi tested
Strains code 1 Species
Locality
(host plant)
Soil
pH
GABC1
Glomus
albidum
Bafia
(Crotalaria)
6.3
2.5
55
Yellow,
80-90 mm,
A(EL)B(L)
GMVN1
Gigaspora
margarita
Ngaoundéré
(Vigna)
5.0
5.0
75
White, 300-410
mm, A(L)
SGMN1
Scutellospora
gregaria
Ngaoundéré
(Manihot)
5.0
3.5
62
Black,
310-420 mm,
A(UL)B(L)C(M)
GGRN1
Glomus
geosporum
Ndupe
(Rinorea)
4.2
9.0
20
Brown,
120-200 mm,
A(ELM)
GiMDE
Gigaspora
margarita
Ekona
(Dioscorea)
5.8
18.0
63
White,
180-210 mm,
A(L)
GCHX
Glomus
clarum
Hawaï
University
-
-
60
Yellow to brown,
80-150 mm,
A(EL)B(UM)
Soa
(Zea mays)
6.3
3.5
78
Brown,
45-120 mm,
A(EL)
GISM (or GIYM) Glomus
intraradices
Available P Mycorrhiza
(ppm) colonisation
(%)
Observations
(spore type)
GABC2
Glomus
albidum
Bafia
6.3
2.5
58
Yellow or white,
80-120 mm,
A(EL)B(L)
GCDM
Glomus
clarum
Douala
(Manihot)
4.9
5.8
90
45-100 mm,
A(EL)B(UM)
GAMN1
Glomus
aggregatum
Nkolbisson
(Manihot)
4.3
10.0
52
Yellow,
45-100 mm,
µm, A(EL)B(L)
GiMNV
Gigaspora
margarita
Ngaoundéré
(Vigna)
4.9
2.5
98
Whitish,
300-410 mm,
A(L)
GiME13
margarita
Gigaspora
(Dioscorea)
Ekona
5.8
18.0
76
Hyaline,
180-320 mm,
µm, A(L)
GiXSC
(or GiXYC)
Gigaspora
sp
Soa
(Crotalaria)
6.3
3.5
65
-
GVAM
Glomus
versiforme
Akokas
(Manihot)
5.6
11
32
Yellow,
45-80 mm
1- Internal code of the Applied Microbiology & Biofertiliser Unit (UMAB), University of Yaounde I
2- Using pearl millet as host-plant (Pennisetum americanum ) and acid fuchsin dye for root staining
(Kormanick and Mc Graw, 1982)
3- Spores characteristics were used for species identification (color, shape, size, number nature of
membranes or muronymes)
Selection of Arbuscular Mycorrhizal Fungi for Inoculating Maize and Sorghum Grown in
Oxisols/Ultisol and Vertisols in Cameroon
473
For each AM fungi species, 10 spores were used per replicate
and three replicates per treatment (30 spores /treatment). The Petri
dishes were incubated at 30°C in the dark. Every three days,
observations were made on the rate of germination and this lasted
for ten days. The length of hyphal growth were also evaluated. The
modifications of the colour of the media containing colour indicators
were observed to evaluate acidity, neutrality or alkalinity of the
strain. A spore was considered as germinated if hyphae length
reached at least 100 µm. Analysis of variance are done and the
means are compared according to Student Fischer test to know if
the results were meaningful.
Experiment II. Identification of AM isolates that enhance P
uptake using peal millet and cowpea
The spores and the roots of five species were screened in greenhouse on
pearl millet and cowpea to evaluate their agronomic performances. The
growth media was sterile soil and sand at a ratio of 3:1 (v/v). Ten
kilogramme of the mixture was placed in polyethylene bags. The seeds
of millet and cowpea were surface sterilised and pre-germinated in a
growth chamber. The inoculum contained roots and spores of each of
the six isolates (cultures) (GACB1, GCXH, GMVN1, SGMN1, GMDE1
and GGRN1). Each inoculum 50 g soil was diluted to 500 g of sterilised
sand. The seeds were germinated on this inoculum containing 10 spores
g-1. After three weeks, plants were transferred in greenhouse in plastic
bags containing 10 kg of substrate with five plants per bag. The
experiment was laid out in a 6 x 2 factorial design replicated five times.
A rotation for all bags was made once per week. Rorison’s solution was
used for watering plants twice a week.
Cowpea and pearl millet were harvested after 45 days and 90 days
respectively. Shoot dry weight were obtained after drying the shoots at
80°C for 48 hours. The fine roots were collected, stained and mycorrhizal
colonisation rate assessed according to Kormanik and Mc Graw (1982).
Shoot phosphorus content was evaluated after mineralisation according
to Anderson and Ingram (1993). Efficiency was obtained using the
formula according to Plenchette et al. (1983):
(Mycorrhizal dry weight − control dry weight)
×100
Mycorrhizal dry weight
Data were analysed for a comparison of means with GENSTAT
software.
474
Nwaga, D. et al
Experiment III. Evaluation of mycorrhiza bio-fertilisers
inoculation in the field
The experimental design and methodologies are similar for the three
tests. Soils physico-chemical characteristics of the sites were
summarised on Table 33.1 and 33.2.
Selecting mycorrhizal fungi isolates for maize growth on Ultisol
The experiment was done in Minkoameyos-Yaoundé. The treatments
applied were: control (without inoculum) and inoculated treatments:
myco 1 (GCXH), myco 2 (GCXH+GiMVN1), and myco 3 (GCXH + GiMVN1
+ GCDM + GGRN1). The total surface of experiment site was 384 m 2.
The elementary plots were 6 x 4 m size. The experimental design was a
randomised complete block with 4 treatments and 4 replications. The
sowing density was 0.80 m x 0.25 m at the rate of 2 seeds per planting
hole. Maize population was 50,000 plants per hectare. Before sowing,
10 g of inoculum were introduced in each planting hole, this correspond
to 100 spores per plant. The stage of 50 % male flowering was evaluated
by counting between 49th and the 59th days after sowing. The stage of
50 % flowering female was between the 55 th and the 58 th days after
sowing. Root colonisation was evaluated (Giovannetti and Mosse, 1980)
after root staining (Kormanik and Mc Graw, 1982). Plant were harvested
174 days after sowing on 20 plants per replicate. After drying the plants,
their dry weight was determined. Dry seed weight was determined by
the same process. Data analysis done using Student-Fisher and
Duncan’s tests.
Testing mycorrhizal fungi inoculation in two maize varieties for
acid tolerance
The Orisol experiment was done in Nkoemvone-Ebolowa site. Height
treatments were applied. Two varieties of maize: a tolerant variety ATP
and a sensitive variety CMS. Three levels of fertiliser (F0: no fertiliser,
F1: N-37, P-27, K-14; F2: N-37, P-87, K-14), and two levels of AM fungi
(M0: no AM fungi and M+: 10 g hole-1 of myco 4 AM fungi mixture) were
used in a combination of treatments. The mixture of strains Glomus
clarum, Gigaspora margarita, Glomus albidum, Glomus geosporum,
Glomus aggregatum and Glomus occultum were produced on a Vetiver
grass at a concentration of 200 infectives propagules g-1 (MPN). The total
surface of experiment site was about 1400 m2. The elementary plot was
4 x 6 m size.
Evaluating inoculation methods using Sorghum on Vertisol
Four treatments were applied for this experiment. Nursery was raised
for Sorghum bicolor (Muskwari variety). Half of the nursery surface was
inoculated using 2 kg of inoculant (Glomus clarum, GCDM1 and Glomus
Selection of Arbuscular Mycorrhizal Fungi for Inoculating Maize and Sorghum Grown in
Oxisols/Ultisol and Vertisols in Cameroon
475
aggregatum, GACB). The other half was non-inoculated. The three
different treatments include: INS treatment (Inoculation in nursery at
sowing) achieved while sowing seeds in nursery bed; IFS treatment
(Inoculation in farm by steeping roots) done by soaking plant roots in
inoculum solution (2 kg litre -1) and IFP treatment (inoculation in farm
on planting hole) achieved by introducing 5 g of inoculum in planting
hole before. The control plants were not inoculated. A complete
randomised block design with four replicates was used and 1m spacing
between plants. The elementary plot had a surface of 76 m 2 and
contained 200 plants (2 plants per pocket). The first harvest was done
at the date of 50 % flowering and mycorrhizal colonisation was evaluated
according to Giovannetti and Mosse (1980) and staining according to
Kormanik and Mc Graw (1982). The efficiency of mycorrhiza was analysed
using phosphorus uptake (Rodier, 1978), nitrogen content (Devani et
al., 1989). The second harvest was done at grain maturity and plant dry
weight, seed yield were evaluated. Data analysis was done using
STATITCF software. ANOVA and Newman-Keuls test was used to
compare the treatments.
Results
In vitro selection of acid tolerant AM isolates based on
germination and hyphal growth
After 3 days GMVN1 and GCXH from our microbial resource bank had
germinated at acidic pH of 3.8 and by the 6 th day all the spores had
germinated (Table 33.4).
Table 33.4: Effect media acidity on in vitro spore germination (%) of 6 arbuscular
mycorrhizal fungi after 6 days
Strains code1
Media acidity (pH)
3.8
GACB1
GMVN1
SGMN1
GGRN
GMDE1
GCXH
Mean
4.5
a
20.0
76.7b
45.0a
03.3a
50.0a
70.0c
44.0
5.3
b
30.0
73.0b
50.0a
13.3a
46.7a
63.0ab
46.0
6.0
b
26.7
90.0c
96.7b
13.3a
60.0b
86.7d
62.0
6.7
b
26.7
86.7c
90.0b
16.7b
83.3c
60.0a
61.0
40.0c
56.7a
46.7a
20.0c
50.0a
66.7bc
47.0
Means followed by the same letter in the same line are not significantly different at 5 %
level of probability Duncan’s new multiple range test. 1: Internal code of the Applied
Microbiology & Biofertiliser Unit (UMAB), University of Yaounde I
Nwaga, D. et al
476
However, the rate of germination varied with strains and pH of the
media. An average of 44 % to 62 % spores germinated in the range of pH
3.8-6.7 but the optimum was around pH 5 and 6. Glomus geosporum
(GGRN), a low germinating strain show no variation on germination
accross the pH range tested. For hyphal growth, media acidity has a
marked effect on some strains (Table 33.5).
Table 33.5: Effect of media acidity on in vitro hyphal growth (µm) of 6 mycorrhizal fungi
after 6 days
Strains code1
Media acidity (pH)
3.8
GACB1
GMVN1
SGMN1
GGRN
GMDE1
GCXH
Mean
b
140
650a
1000b
100a
100a
450c
407
4.5
b
140
800b
1000b
101a
100a
400b
424
5.3
6.0
b
136
2700d
4000d
101a
123b
400b
1243
6.7
a
109
2000c
3000c
130b
135c
400b
962
104a
800b
800a
103a
120b
350a
380
Means followed by the same letter in the same line are not significantly different at 5 %
level of probability Duncan’s new multiple range test. 1: Internal code of the Applied
Microbiology & Biofertiliser Unit (UMAB), University of Yaounde I
In some slow growing strains (GACB1, GGRN, GMDE1 and GCXH),
the effect was at least 4 times lower at the optimum pH of 5.3 compared
to pH 3.8. The optimum pH for hyphal growth for most strains is around
5 to 6. These six strains are adapted to a range of 3.8 to 6.7 pH, but pH
up to 6 are inhibitory for hyphal growth. The results shows that in vitro
pH 5 to 6 are optimum for spores germination and hyphal growth of
these 6 isolates of AM fungi. Average spores germination at these pH
are 61-62 %; hyphal growth are 1243-962 µm after six days respectively.
The strains GiMVN1 and GCXH adapted to very high pH, also produce
much more acidity in the culture media compared to other strains as
indicated by the changes in colour indicators.
Selection of AM fungi for plant biomass, P uptake and
efficiency under nursery
Eleven locally isolated AM isolates were tested using sterile soil in nursery
with pearl millet and cowpea as test plants. The plants were harvested
before flowering. Non-inoculated pearl millet show chlorotic symptoms,
while inoculated ones were healthy and dark green in colour. Dry matter
yield of cowpea and pearl millet ranged from 1.4-2.6 to 1.4-7.2
respectively (Table 33.6).
Selection of Arbuscular Mycorrhizal Fungi for Inoculating Maize and Sorghum Grown in
Oxisols/Ultisol and Vertisols in Cameroon
477
Table 33.6: Performance of arbuscular mycorrhizal inoculation on cowpea ( Vigna
unguiculata) and millet (Pennisetum americanum) under controlled conditions
Mycorrhizal Mycorrhizal
strains 1
colonisation
(%)
Cowpea Millet
GCXH
GISM
GABC2
GCDM
GAMN1
GiMV
GiME13
GiXYC
GVAM
GGNR
GI
SM +
GABC2
Control
Treated/
Control 2
LSD (5 %)
Dry weight (g)
P content (%)
Cowpea
Millet
Cowpea
Efficiency 2 (%)
Millet Cowpea Millet
65
72
46
53
62
89
85
85
58
23
71
78
58
90
52
98
76
65
32
19
2.34ab
3.80bc
2.54a
2.50a
3.02b
3.78bc
3.26b
4.60c
3.40b
2.60a
6.86ab
9.08bc
10.58cd
17.18e
13.86de
30.48f
6.32ab
5.86ab
7.50abc
10.88cd
0.47de
0.50def
0.52ef
0.62g
0.39c
0.70g
0.45d
0.30b
0.55f
0.52ef
0.52d
0.42c
0.23ab
0.68e
0.27b
0.78e
0.44c
0.29b
0.29b
0.28b
24a
53ef
30bc
29abc
41d
53ef
45d
61g
48de
32c
38b
53c
60d
75f
69a
86g
33a
27a
43b
61d
69
0
70
0
4.90c
1.78 a
10.80cd
4.26a
0.79h
0.21a
0.44c
0.16a
64g
0
61d
0
-
-
1.4-2.6
1.12
1.4-7.2
3.51
1.4-3.8
0.38
1.4-4.9
0.46
5.3
5.4
Means followed by the same letter are not significantly different at 5 % level of probability
Duncan’s new multiple range test. 1: Internal code of the Applied Microbiology &
Biofertiliser Unit (UMAB), University of Yaounde I. 2: Efficiency obtained using the formula
according to Plenchette et al. (1983). 2 extreme values between treated/ uninoculated
control
For pearl millet, the best strains were GiMVN1 and GCDM while for
cowpea, the highest shoot weight were GiXYC, GISM and GiMVN1.
Cowpea and pearl millet phosphorus content, ranged from 0.21-0.16 %
in the uninoculated control. The inoculated cowpea plant contained
0.28% to 0.79 % phosphorus in the tissue. Mycorrhiza inoculation
increased P biomass content from 40% to 280 % for cowpea and 40% to
390 % for pearl millet over the control. From these results, Gigaspora
sp (GiXY strain) was selected for cowpea and Gigaspora margarita
(GiMVN1 strain) and Glomus clarum (GCDM strain) for pearl millet
production for inoculant use in field experiments.
Some other strains such as Glomus intraradices (GISM) were good
candidates because their efficiency was stable on both cowpea and millet
(53 %). For most other strains, there seemed to be a preference for a
given plant. The mixture of strains (GISM + GABC2) also produce a
Nwaga, D. et al
478
good efficiency on both plants (61-64 %). Mycorrhiza root colonisation
was generally high on both plants except for Glomus geosporum (GGNR),
because only 23% and 19 % were recorded on cowpea and pearl millet
respectively. For millet dry weight and P content, the most efficient strain
is Gigaspora margarita (GiMVN1), the second one is Glomus clarum
(GCDM), because their P content are 0.7-0.8 to 0.6-0.7 respectively and
provide the highest biomass. The least efficient strains provided a shoot
weight and P content increase of 40 % over the uninoculated control.
Maize response to AM inoculation in an Oxisol/Ultisol under
field conditions
AM inoculant selection under Ultisol
In order to assess strains effect on maize yield under field conditions in
oxisol this experiment were carried out. The number of days to 50 %
flowering varied from 54 to 58 days (Table 33.7).
Table 33.7: Response of maize to 3 strains of mycorrhiza inoculation under farm
conditions on Ultisol at Yaoundé-Minkoameyos (Centre, Cameroon)
Treatment
50 %
strains
flowering
(days)
Control
Myco 1
Myco 2
Myco 3
58
56
56
54
Colonisation
(%)
Biomass
yield
(t ha-1)
Grain
yield
(t ha-1)
Yield
increase
(%)
5
34
32
47
8.80b
10.53a
10.28a
11.30a
4.5b
7.0a
6.0a
7.5a
55
33
66
Myco 1: Glomus clarum, Myco 2: Glomus clarum + Gigaspora margarita, Myco 3: Mixture
of strains (Glomus clarum+ Gigaspora, and others)
Means followed by the same letter are not significantly different at 5 % level of probability
Duncan’s new multiple range test.
Maize inoculation by all the strains produced a significant increase
on plant biomass. Seed yield varied from 4.5-7.5 t ha-1 and all the AM
fungi strains significantly increased productivity. The yield increase by
mycorrhiza inoculation was between 33-66 %, representing 1.5-3.0 t ha-1.
This effect was correlated with the level of root colonisation which is
very low on non – inoculated (5 %) compared to inoculated treatments
(32-47 %).
Maize variety compared to chemical fertiliser under Oxisol
field
The experiment was carried out in Ebolowa-Nkoemvone, which is the
site with one of the highest aluminium toxicity in humid forest zone of
Selection of Arbuscular Mycorrhizal Fungi for Inoculating Maize and Sorghum Grown in
Oxisols/Ultisol and Vertisols in Cameroon
479
Cameroon. The number of days to anthesis (63-66), was not significantly
different between the treatments for the susceptible variety CMS, while
for the tolerant one it was lower (62-68) at the highest fertilisers and
inoculated treatment, F2 M+ (Table 33.8).
Table 33.8: Response of two maize varieties to AM inoculation compared to chemical
fertilizers (N-P-K) on station conditions on Oxisol at Ebolowa-Nkoemvone (South,
Cameroon)
Treatment
Variety CMS
(acid susceptible)
F0 MF0 M+
F1 MF1 M+
F2 MF2 M+
Variety ATP
(acid tolerant)
F0 MF0 M+
F1 MF1 M+
F2 MF2 M+
Days to
anthesis
Mycorrhiza
colonisation (%)
Grain yield
(t ha-1)
Yield
increase (%)
62.50a
63.50a
63.50a
63.50a
65.00a
66.00a
32.5
31.8
08.0
35.5
12.3
60.0
2.29a
3.60bc
3.09b
2.97b
3.04b
2.96b
58.8
34.7
29.6
32.3
29.2
68.00 a
66.50a
65.25a
64.25a
64.00a
61.50b
15.5
77.5
32.5
92.5
0.00
57.5
2.52a
3.84c
3.28b
2.95b
2.62a
2.95b
52.3
30.1
17.1
4.0
29.5
Means followed by the same letter are not significantly different at 5 % level of probability
Duncan’s new multiple range test.
1 This site is one with the highest aluminium toxicity in the country
F0: no fertiliser (control), F1: NPK + Urea fertiliser: 37-24-14, F2: NPK + Urea + DAP
fertiliser: 37-84-14; Urea (46-0-0), DAP(18-45-0).
M-: non-inoculated ; M+: mycorrhiza inoculation by Glomus clarum and Gigaspora
margarita
The days to anthesis is reduced by 3 days in the acid susceptible
variety CMS and significantly increased in the acid tolerant one ATP by
6 days. Mycorrhiza inoculation produced a significantly higher yield
compared to the control on both varieties. For the variety CMS, yield of
58.8 % were recorded, corresponding to 1.3 t ha -1 after inoculation
compared to the control. When fertiliser was applied alone yield increase
was 34.7 % (0.8 t ha-1). For the acid tolerant variety (ATP) AM inoculation
improved grain yield by 52.3 % compared to the control, while fertiliser
application yielded 30.1 % only. When mycorrhiza is assessed for the
acid susceptible variety CMS, inoculation was able to increase root
Nwaga, D. et al
480
colonisation of AM fungi only for one treatment (F2 M+), from 32.5 % to
60 %; fertiliser generally reduce it for the non-inoculated fertilised
treatments (F1 M-, F2 M-). The native mycorrhiza fungi was able to
colonise maize roots as high as 35.5 % for the control (Fo M-). For the
tolerant variety ATP, mycorrhiza inoculation increased root colonisation
from 16 % to 78%, 93 and 58 % respectively for the uninoculated
(F0 M-), and inoculated (F0 M+, F1 M+ and F2 M+) treatments. Mycorrhiza
colonization in acid tolerant variety ATP are much more higher than in
the acid susceptible variety CMS without any added fertiliser, 78 and
32 % respectively.
Influence of the inoculation method on sorghum
performances under Vertisol field
The aim of this experiment was to evaluate three inoculation methods
on a specific sorghum variety, muskwari on savannah zone of Cameroon.
Compared to the uninoculated control, there was a reduction of 6 days
by the method at sowing (INS) on the date to 50 % flowering (Table 33.9).
Table 33.9: Evaluation of AM inoculation methods on Sorghum bicolor performances
under farm conditions on vertisol at Maroua-Moutourwa (Far-North, Cameroon)
Treatment
50 %
Root Biomass
flowering Colonisa- yield
(days)
tion(%) (t ha-1)
Control
(uninoculated)
At sowing
(INS)
On planting
hole (IFP)
By steeping
roots (IFD)
Grain
Yield
Grain N N response
yield increase uptake
(%)
(t ha-1)
(%)
(kg ha-1)
62
20
3.07
1.54 b
–
28.34
–
56
38
3.50
1.64 ab
6.5
22.96
- 19
64
43
4.04
2.00 ab
29.9
40.00
+ 41
63
65
4.24
2.28 a
48.1
43.55
+ 54
Means followed by the same letter are not significantly different at 5 % level of probability
Duncan’s new multiple range test.
INS: mycorrhiza inoculation on nursery at sowing (inoculant-soil mixture: 2 kg /6 m 2);
IFP : mycorrhiza inoculation on farm at planting on pocket (5 g hole-1); IFD : mycorrhiza
inoculation on farm at planting by steeping roots (inoculant-water mixture: 2 kg liter-1).
All the mycorrhiza inoculation methods increased sorghum biomass
from 0.43 t to 1.17 t ha-1 (14% to 38 %). The inoculation by steeping the
roots on mycorrhiza inoculum (IFD) gave the best results, significant
yield increase of 48 % were recorded when compared to the uninoculated
Selection of Arbuscular Mycorrhizal Fungi for Inoculating Maize and Sorghum Grown in
Oxisols/Ultisol and Vertisols in Cameroon
481
control (Table 33.9). Mixing mycorrhiza with soil at sowing did not give
high yield (7 %) probably because of the low inoculum density in this
treatment. Seed N uptake evaluation shows that, when compaerd to
the uninoculated treatment, there were a deficiency of -19 % nitrogen
in the seeds of the treatment on nursery at sowing (INS), while at planting
on the field (IFP, IFD) an important increase of 41% to 54 % were
recorded.
Discussion
Mycorrhizal strains selection
These isolates of AM fungi may be differentiated into two groups: some
strains are not influenced by acidity (Glomus albidium, G. geosporum,
Gigaspora margarita and Scutellospora gregaria). This is in accordance
with some authors (Hepper, 1984; Sieverding, 1991) who reported that
the suborder Gigasporineae are better adapted to soil acidity than the
suborder Glomineae (Glomus type). Theses strains (GMDE, GMV1 and
SGMN 1) tend to have an acidifying culture media during their in vitro
growth. The size of spore of most Gigasporineae (151-400 mm) are
more larger than that of Glomineae (46-150 mm) of our collection.
This may indicate that large reserves in these spores are important
factor in maintaining AM fungi infectivity during plant root colonisation
(Smith and Read, 1997). The mixture of strains (GISM + GABC2) are
more performant than the individual strains, this confirm observations
by Sieverding (1991). Criteria for AM fungi selection are: spore
germination and hyphal growth, root colonisation, plant biomass and
yield, nutrient uptake and adaptation to environmental constraint such
as soil acidity and competition (Sieverding, 1991). These criteria have
been used in this work to select few strains of our microbial bank
Gigaspora margarita (Strains GiMVN1 and GiXYC) and Glomus clarum
(GCDM) for acidity tolerance, plant biomass and P uptake performances
(Nwaga et al., 2000). Mycorrhizal isolates from three agroecological
zones of Cameroon have shown an important diversity for symbiotic
efficiency and P uptake of crops (Ngonkeu Mangaptche and Nwaga,
1998).
Field inoculation of maize under Oxisol/Ultisol
According to Harley and Smith (1983), pot experiment using maize
provided 28-33 % mycorrhiza colonisation after inoculation on sterilised
soil. These results indicate colonisation of 19 % to 98 % by eleven different
inocula of AM fungi under controlled conditions using pearl millet. Under
482
Nwaga, D. et al
field conditions, maize inoculation resulted 32% to 47% colonisation at
Yaounde-Minkoameyos while it was 32% to 93% at Ebolowa-Nkoemvone.
These results can be compared with observations by Sylvia et al ,
(1993):10-61% and Sanginga et al . (1999): 17-33%) both working on
maize after inoculation with AM fungi.
The grain yield increase obtained from maize crop at YaoundeMinkoameyos was 1.5-3.0 t ha-1 with mycorrhiza inoculation as compared
to 0.7 t ha -1 grain reported by Islam and Ayanaba (1981) on sterilised
soil or 1.0-1.2 t ha-1 cobs by Baltruschat (1987) both on maize. Maize
yield increase of 52-59 % after inoculation on Ebolowa-Nkoemvone site
and the reduction of maize response to inoculation at higher fertiliser
application are consistent with observations of Smith and Read (1997),
according to which P concentration generally reduce root colonisation
and thus response to inoculation. For example, on barley, a grain yield
increase of 11-37 % were obtained by AM fungi inoculation, 60 kg P ha1
application lead to 3-16 % decline in barley yield (Hayman, 1987). The
acid tolerant maize cultivar (ATP) seemed to be less responsive to high
fertiliser application compared to the susceptible cultivar (CMR).
Conventional agriculture may significantly reduce spores number (100
times) and mycorrhiza colonization (2.5-10 times) when compared to
low input agriculture (Douds et al., 1993), but this situation is likely to
happen only in tropical industrial plantations.
Field inoculation of Sorghum under Vertisol
Inoculation of finger millet with selected AM strain improved grain yield
by 18 % compared to fertiliser (Powell, 1987). Sorghum yield increased
from an average from 0.10 to 0.75 t ha -1 from either of the three
inoculation methods which are commensurable with result obtained by
Miranda (1982) from sorghum. According to observations by Simpson
and Daft (1990), development of mycorrhiza inoculation have shown
not to have interactions with water stress on maize and sorghum, but
Douds et al. (1993) showed that the proportional plant response to
inoculation increase with drought stress. The yield increase of muskwari
sorghum observed could be attributed not only to a better nutrient uptake
but also to a better water stress tolerance by mycorrhiza inoculation
compared to indigenous populations on this vertisol. These results could
be compared to those described by Powell (1987), on finger millet where
a selected AM strain gave 18 % better grain yield than uninoculated
plants over three fertiliser rates. The results in this study showed a
yield increase of 14 % to 15 % by the inoculated maize over the best
uninoculated plants for the two fertiliser rates in this aluminium toxicity
site. Under field conditions, a preliminary result obtained on Mucuna
veracruz inoculation by mycorrhiza produced a significant increase of
Selection of Arbuscular Mycorrhizal Fungi for Inoculating Maize and Sorghum Grown in
Oxisols/Ultisol and Vertisols in Cameroon
483
43 % on litter, 88 % on shoot biomass, 100 % on plant P content and for
P use efficiency, from 13 % for the control to 39 % (Nwaga et al , 1999).
The development of an integrated, ecosystem level, approach to soil
management, including inoculation with micro-symbionts should be
encouraged (Swift, 1998; TSBF, 2000).
Conclusion
It is not yet possible to produce pure AM fungi inoculant, since the
fungi does not grow in artificial media which is a major drawback.
Field responses to inoculation are sometimes low and erratic since
there may be effective high populations of indigenous AM fungi for
most common crops. In order to develop mycorrhiza technology for
field application, many problems will have to be solved for example, to
standardise large scale inoculation production (experimental
conditions, spore numbers to increase their infectivity, reduce
contamination with deleterious organisms), to develop assessments
on the best inoculation methods for a specific crop. A preliminary
screening of some AM fungal isolates for acidity tolerance was carried
out and three of them were selected. It will be more effective to assess
or select for their tolerance to aluminium or manganese toxicity which
are more toxic than hydronium acidity. Using a mixture of selected
AM fungi, many field assessments were done on the potential of
inoculation of maize on Oxisol/Ultisol. The results obtained shows
that significant yield were enhanced by inoculation under diverse poor
soil. Also, sorghum produced a good yield using few quantity of
inoculum under Vertisol which are known not to respond to chemical
fertiliser for muskwari sorghum variety. In this study, it is clear that
strain efficiency test may be beneficial as well as environmental factors
such as fertiliser or variety effect.
Do we need to put efforts into this biotechnology ? Yes, because
slash and burn agriculture may reduce indigenous populations of AM
fungi. Most field evaluations done in Cameroon have shown that
indigenous populations are low and poorly efficients; also because a
minimum yield of twenty percent is good and the returns are much
more higher in legume crops or fruit trees than cereal crops; the
technology may help save currencies at farmer or country level. In order
to minimise mycorrhiza inoculation technology cost for cereal, to quickly
increase soil organic matter build up and fertility through biological
means, we suggest an inoculation strategy of a legume (grain or cover
crop) by selected and competitive rhizobial and mycorrhizal isolates.
The development of a low input technology for farmers such as integrated
biological management of soil fertility to enhance crop productivity in
the tropics is urgently needed.
484
Nwaga, D. et al
Acknowledgements
The authors wish to acknowledge the followings persons for their
technical assistance, Mouliom A., Oloumane Nloubou, J, Oneya S.,
University of Yaounde I, Adamou S. University of Ngaoundere and
Ndoumbe-Nkeng, Institute of Agricultural Research for Development
(IRAD) Yaounde for data analysis.
References
Ambassa-Kiki, R. (1990) Un site d’expérimentation du réseau IBSRAM à
Minkoameyos, Yaoundé. IBSRAM Proceedings. 10: 425-440.
Anderson, J.M. and Ingram, J.S.I. (1993) Tropical Soil Biology and Fertility. A
handbook of methods. CAB International, Walingford.
Anonyme, (1991) Mémento de l’agronome. Ministère de la Coopération, Paris.
Balthruschat, H. (1987) Evaluation of the suitability of expanded clay as carrier
material for VA mycorrhiza spores in field inoculation of maize. Angew.
Botanik 61: 163-169.
Barrault, J. Erkebil, J.P. et Vaille, J. (1972) Point sur les travaux de l’IRAD
sur les sorghos repiqués du Nord-Cameroun. L’agronomie Tropicale
27(8): 791-814.
Bethlenfalvay, G. J. (1992) Mycorrhizae and crop productivity. In : Bethlenfalvay
G. J. Linderman R.G (eds), Mycorrhizae in sustainable agriculture. ASA
Special Publication N° 54, Madison, USA, pp 1-27.
Bindzi Tsala, J. (1987) Les sols rouges du Cameroun. 8e réunion du souscomité ouest et centre Africain de correlation des sols pour la mise en valeur
des terres. MINREST-FAO. Yaoundé, Cameroun.
Brundrett, M.C., Bougher, N., Dell, B., Grove, T. and Malajczuk, N. (1996)
Working with mycorrhiza in forest and agriculture. ACIAR Monograph 32:
374 p.
Davet, P. (1996) Vie microbienne du sol et production végétale. INRA, Paris,
Devani, M.B., Shihoo, J.C., Shal, S.A. and Suhagia, B.N. (1989).
Spectrophotometrical method for micro determination of nitrogen in Kjeldahl
digest. J. Assoc. Off. Anal. Chem. 72(6): 953-956.
Douds, D.D. Jr, Janke, R.R. and Peters, S.E. (1993) VAM fungus spore
populations and colonisation of roots of maize and soybeans under
conventional and low-input sustainable agriculture. Agric. Ecosyst. Environ.
43(3-4): 325-335.
Gianinazzi-Pearson, V. and Diem, H.G. (1982) Endomycorrhizae in the tropics.
In: Microbiology of tropical soil and plant productivity. Dommergues Y.R.,
Diem H.G. (Eds), Martinus Nijhoff /Dr Junk W. Publishers, The Hague.
Giovannetti, M. and Mosse, B. (1980) An evaluation of techniques for measuring
vesicular arbuscular mycorrhizal infection in roots. New Phytol.
84: 489-500.
Selection of Arbuscular Mycorrhizal Fungi for Inoculating Maize and Sorghum Grown in
Oxisols/Ultisol and Vertisols in Cameroon
485
Harley, J.L. and Smith, S.E. (1983) Mycorrhizal symbiosis. Academic Press,
London.
Hawksworth, D. L. (1991) The fungal dimension of biodiversity: magnitude
significance and conservation. Mycological Research 95: 641-655.
Hayman, D.S. (1987) Mycorrhiza in field crop systems. In: Ecophysiology of VA
mycorrhizal plants. Safir G.R. (ed), CRC Press Inc, Boca Raton, Florida. pp.
171-192.
Hepper, C.M. (1984) Isolation and culture of VA Mycorrhizal (VAM) Fungi. In:
VA Mycorrhiza. Conway, L.I. P. and Bagyaraj, D. J. (Eds), CRC Press, Florida,
pp. 95-112.
Islam, R. and Ayanaba, A. (1981) Growth and yield response of cowpea and
maize to inoculation with Glomus mossae in sterilised soil under field
conditions. Plant and Soil 63: 505-509.
Janos, D.P. (1987) VA mycorrhiza in humid tropical ecosystems. In: Ecophysiology
of VA mycorrhizal plants . Safir G.R. (Ed), CRC Press Inc, Boca Raton, Florida
pp. 107-134.
Kormanik, P. P. and Mc Graw, A. C. (1982) Quantification of vesicular- arbuscular
mycorrhizae in plant roots. In : Methods and principles of mycorrhizal
research, N. C. Schenck (Ed), The American Phythological Society, St. Paul,
Minnessota. pp. 37-45.
Miranda, J.C.C. de (1982) Influencia de fungos endomicorrizicos inoculados a
campo, na cultura de sorgo e soja em um solo sob cerrado. R. Bras. Ci. Solo
8, 31-36.
Ngonkeu Mangaptche, E.L. and Nwaga, D. (1998) Diversité, potentiel infectieux
des mycorhizes à arbuscules de quelques sols du Cameroun, et réponse du
niébé (Vigna unguiculata) à l’inoculation. Cam J. Biol. Biochem. Sci.
8(1): 56-67.
Nwaga, D. and Ngonkeu, M.E.L. (1998) Diversité des mycorhizes à arbuscules
et biomasse microbienne du sol des écosystèmes du sud-Cameroun.
Biosciences Procedings. 5:76-84.
Nwaga, D., Nolte, C., and Jemo, M. (1999) Field response of Mucuna veracruz, a
leguminous cover crop to rhizobia and mycorrhiza inoculation compared to
chemical fertilisers in humid forest zone of Cameroon. Paper presented for
the 7th AfNet-TSBF Workshop, 11-15 September 1999, Kampala, Uganda.
Nwaga, D., Ngokeu M.E.L., Ogong M.M., Ngakou A., Abelong M.P. and Foko J.
(2000) Soil beneficial microorganisms and sustainable agricultural
production in cameroon: current research and perspectives. In: The biology
and fertility of tropical soils. TSBF report 1997-1998. pp. 62 -65.
Plenchette, C., Fortin, J. A. and Furlan, V., (1983) Growth responses of several
plant species to mycorrhizae in a soil of moderate P fertility. Plant and Soil
70: 199-209.
Powell, C.L. (1987) Field inoculation with mycorrhizal fungi In: VA mycorrhiza,
Powell C.L., Bagyaraj D.J. (eds), CRC Press, Inc Boca Raton, Florida, pp.
205-222.
Robert, M. (1992) Le sol, ressource naturelle à préserver pour la production et
l’environnement. Cahiers Agricultures 1: 20-34.
Rodier, J. (1978) L'analyse de l'eau: chimie, physico-chimie, bactériologie, biologie.
6e édition. Dunod Technique, Paris.
486
Nwaga, D. et al
Sanchez, P.A. (1994) Properties management of soils in the tropics. WileyInterscience Publication, New-York.
Sanchez, P.A. and Salinas J.G. (1981) Low-imput technology for managing oxisols
and ultisols in tropical America. Adv. in Agron. 34: 279-406.
Sanginga N. Carsky, R.J. and Dashiell, K. (1999) Arbuscular mycorrhizal fungi
respond to rhizobial inoculation and cropping systems in farmers fields in
the Guinea savanna. Biol. Fert Soils 30 (3): 179-186.
Schenck, N. C. (1982) Methods and principles of mycorrhizal research. American
Phytopathological Society, St. Paul, Minnesota.
Seiny Boukar, L. (1980) Etude du niveau de fertilité de quelques sols de la
région de Kolofata sous culture de mil de saison sèche. Mémoire de Diplôme
d’Ingénieur Agronome, ENSA, Dschang, Cameroun.
Sieverding, E. (1991) Vesicular-arbuscular mycorrhiza management in tropical
agrosystems, GTZ, Eschborn, Germany.
Simpson, D. and Daft M.J. (1990) Interactions between water-stress and different
mycorrhizal inocula on plant growth and mycorrhizal development in maize
and sorghum. Plant and Soil. 121: 179-186.
Smith, S.E. and Read, D.J. (1997) Mycorrhizal symbiosis. Academic Press,
London.
Swift, M.J. (1998) Toward the second paradigm: integrated biological management
of soil. Paper presented for the FERTBIO Conference, Brazil, September
1998.
Sylvia, D. M. and Williams, S.E. (1992) Mycorrhizae and environmental stresses.
In : Bethlenfalvay G. J. Linderman R.G (eds), Mycorrhizae in sustainable
agriculture. Madison. ASA Special Publication 54: 101-124.
Sylvia, D.M., Hammond, L.C., Bennet, J.M. Haes, J.H. and Linda, S.B. (1993)
Field response of maize to a VAM fungus and water management. Agron. J.
85(2): 193-198.
The, C. (2000) Identification of heterotic groups for acids soil on some maize
varieties in Cameroon. INCO 1 & 2 Meeting, June 2000, Yaoundé, Cameroon.
Woomer, P.L. and Swift, M.J. (1994) The biological management of tropical soil
fertility. John Wiley and Sons, Chichester, TSBF & Sayce Publ., United
Kingdom.
Macrofaunal Abundance and Diversity in Selected Farmer Perceived Soil Fertility
Niches in Western Kenya
Macrofaunal Abundance and
Diversity in Selected Farmer
Perceived Soil Fertility
Niches in Western Kenya
487
34
Tabu, I.M.1*, Obura, R.K.1 and Swift, M.J.2
Department of Agronomy, Egerton Univesity, Box 536
Njoro, Kenya.
2
TSBF -CIAT, Box 30677, Nairobi, Kenya.
1
*Corresponding author: E-mail: immtabu@yahoo.com
Abstract
Most farmers in Kenya are resource poor and rely on the
soil biological processes (nutrient cycling, decomposition
and mineralization) for subsistence. Soil macrofauna is an
integral part of such systems and their abundance and
diversity has been suggested as an indicator of functional
status. Most soil macrofaunal studies have however involved
natural or controlled ecosystems but little of the farmers’
fields. A study was therefore carried out between 1999 and
2000 to determine macrofauna abundance and diversity in
farmer perceived soil fertility niches of Kabras Division,
Western. Macrofauna were determined in sixteen niches
within four farms that had previously been ranked as having
similar soil fertility management. The soils within the farms
were humic Acrisols and mollic Gleysols. Sampling was done
Tabu, I.M. et al
488
using a monolith 20 cmx20 cmx30 cm in layers of 0-10cm,
10-20cm and 20-30cm. The organisms were hand sorted,
preserved in 70% alcohol and taken to the laboratory for
identification and determination of biomass. The physical
and chemical soil characteristics of the niches were
determined. Termites, earthworm and ants were the most
dominant organisms. The niches varied significantly in
abundance and diversity of these organisms. The diversity
(Shannon –Werner) index was low compared to that of the
natural ecosystem. This was attributed to the predatory
macrofauna, microclimate and differences in food resources.
It was concluded that abundance varied with management
and environment. Diversity was however not different in
niches within the agro-ecosystems.
Key words: Macrofauna, Termites (Isoptera), Earthworm (Lumbricus),
Ants (Formicidae), farmers fields, soil fertility management
Introduction
Only about 30% of Kenya is medium to high potential agricultural land.
With the increase in population (about 28 million people in 2000), farmers
are forced to abandon their local farming practices (shifting cultivation)
for the high external input ones like continuous cultivation and use of
pesticides. That most small-scale farmers are low resource and cannot
afford inorganic fertilizers generally leads to the assumption that farmer
practices impact negatively to soil fertility. Sanchez et al. (1997) observed
that soil fertility is one of the most important factors limiting crop
production in Kenya.
The advent of participatory approaches led to realization that farmers
use alternative methods to counter resource limitation. Farmers use
variation in soil fertility and crop management to maximize production,
get product variety and avoid risks of crop failure. As an aggregate, a
combination of practices such as fallowing, continuous cultivation and
manuring within the farm perspective. This management can be
understood better by disaggregating farms into management niches.
This also forms a framework for integrated nutrient management (judicial
manipulation of inputs and outputs) approach. Biological functioning
of the soil is important in this framework.
Lavelle et al. (1997) defined macrofauna as organisms that are more
than 10mm is size and manipulate the soil by making biogenic structures
and galleries that sometimes persist longer than the organisms
themselves. Through their activities, macrofauna influence the physical,
Macrofaunal Abundance and Diversity in Selected Farmer Perceived Soil Fertility
Niches in Western Kenya
489
chemical and biological status of the soil. They also communite organic
resources and determine diversity and abundance of other organisms
through availability of nutrients. They have thus been suggested as
early ideal warning indicators of soil fertility decline because of their
large body size, fast turnover rate and role in soil fertility management
(Linder et al., 1994).
Sometimes the farmer’ practices like continuous cultivation and use
of pesticides result in decline of soil organic matter, pH and pollution
respectively. Diverse macrofauna interact to cushion the soil against
such stress. Characterizing and understanding the way farmers’
practices affect the macrofaunal level and soil processes is important
for sustainable production.
In addition to their role as soil fertility indicators, macrofauna are
also important resources which farmers can manipulate for subsistence.
Termites with their mutual relationship with fungi (fungal gardens) and
bacteria make the mounds rich in soil organic matter and nutrients
(Lavelle et al., 1997). In Zimbabwe, termite mounds (termitaria) is used
as fertilizer (Murwira and Carter, 1994; Swift et al., 1998) while in the
Sahel, farmers add vegetation to the soil to enhance termite activity
(Lamers et al., 1994). Woomer et al. (1999); Murage et al. (1999); Murwira
and Carter (1994) in Kenya and Zimbabwe, observed that farmers
concentrated organic resources in homegardens where priority crops
are grown. Apart from using macrofauna to compare natural eco-systems
(forests) and cultivated land or in controlled experiments, (Okwakol et
al., 1994; Ayuke et al., 1999; Lavelle et al., 1997) macrofauna have
rarely been used to characterize the farmers practices.
Objectives
i)
Determine the faunal abundance and diversity in farmer perceived
niches of western Kenya.
ii) Relate faunal abundance and diversity to the soil fertility and crop
management in the farmer perceived niches.
Material and Methods
i) Site decription
The study was carried out in four farms on humic Acrisol (uplands) and
mollic Gleysols (valley bottomlands) in LUZ 8 (forest peripheral) of Kabras
division, western Kenya. The area is densely populated with a maize/
sugarcane cropping system and situated between longitudes 340 20’
Tabu, I.M. et al
490
and 350E, latitude 000 15’ and 100N and altitude 1300-1900 metres
above sea level. It receives bimodal rainfall of between 1000-2000 mm
per annum (Figure 34.1) and mean minimum and maximum
temperatures of 80C and 250C respectively.
Participatory mapping of the on-farm niches was carried out and
scientific soil survey and classification used for truth proofing.
Figure 34.1: Rainfall distribution in Kabras Division, Western Kenya (1998-2000)
Rainfall quantity (mm)
400
350
300
1998
250
1999
2000
200
150
100
50
0
Jan.
Feb. March April
May
Jun.
Jul.
Aug. Sep.
Oct.
Nov.
Dec.
Month
ii) Niche Characteristics
Since farmers cannot afford the inorganic fertilizers, they use a variety
of soil fertility and crop management practices based on the holistic
farm perception to optimize production and avoid the risk of crop failure.
They organize their farms into spatial distributed soil fertility niches as
follows:
Oboma/Homegarden – Old boma site i.e. a former kraal site which
farmers used for maize and vegetables production. The niche represented
a high soil fertility and crop management niche. As a result of the high
soil fertility, this niche also had high cultivation intensity i.e. vegetables
were planted immediately after maize.
Npasture – This niche was under fallow (natural pasture) for more than
five years. It represented the farmers’ common practice of leaving land
fallow whenever soil fertility declined. The niche was overgrazed and
soil was compacted.
Cmaize – This niche was continuously under maize cultivation. Maize
was planted in March/April and harvested between October/ November.
Macrofaunal Abundance and Diversity in Selected Farmer Perceived Soil Fertility
Niches in Western Kenya
491
The niche was depleted mainly through nutrient transfer i.e. maize
market external to the farm and stover fed to livestock and the resulting
FYM used elsewhere (homegardes and Oboma). Leaching and erosion
was also high because of low soil organic matter. It therefore represented
infertile (depleted) niches
Vbottom – High soil moisture and nutrient recharge from erosion
characterized this niche. Although generally perceived as fertile,
waterlogging was the main constraint. Farmers therefore used it for
maize production between December and April. Between June and
October the niche was usually waterlogged thus not used for crops.
Forest – represented the natural ecosystem. Data by Brown et al. (1996)
from the nearby Malava forest was used as a reference point.
iii) Macrofauna sampling
Soil macrofauna characterization and identification was based on the
body size as described in Blair et al. (1996) and Anderson and Ingram,
(1993) and the existing species. Sampling was done using a monolith
of size 25 cm x 30 cm x 30 cm to quantify the macrofaunal groups.
The monolith was placed at randomly selected points within the niche
and driven into the soil using a metal mallet. The soil from the monolith
was removed by hand in order of the successive depths (0-10 cm, 1020 cm and 20-30 cm) that were placed into different plastic buckets.
The soil was later placed on plastic trays of size 20cm x 30 cm and
gently sorted out to locate the animals. The organisms were hand sorted
using a pointer put in 70% alcohol for preservation and taken to the
laboratory for determination and identification of taxonomic groups
and abundance. The method was preferred because it was easy to
handle, different stages of macrofauna (sedentary and mobile) could
be extracted, and did not depend on macrofaunal behavior or presence
of substrate.
Species diversity based on the Shannon and Werner diversity index
was used to assess changes in soil fauna across on-farm niches. The
index was calculated using the following equation:
H = - (PilnPi ), where:
H is the Shannon index,
Pi is the proportion of individuals found in the i species and estimated
as ni/ N; where ni is the number of individuals of the ith species and
N the total number of individuals within the sample.
The diversity index assumes that individuals are randomly sampled
from a large population and that all species are represented in the
492
Tabu, I.M. et al
sample. The index combines species richness (total number of species
present) and evenness (relative abundance).
iv) Soil analysis
For soil fertility analysis, samples were taken randomly from four spots
within each niche and bulked to form a composite sample. A sub-sample
of about 500gm was then taken, air-dried and ground to pass through
a 2 mm sieve and analyzed in the laboratory for soil texture, total N,
available P, pH, and CEC. Soil texture was determined by the hydrometer
method, pH by the soil: water ratio of 1:2.5, organic carbon by oxidizing
the soil with potassium dichromate and concentrated Sulphuric acid
and the remaining concentration of dichromate and ferrous ions
determined by titration (Okalebo et al., 1989). Total nitrogen (N) was
determined by semi micro-kjedhal digestion with sulphuric acid and
selenium as a catalyst and copper sulphate to raise the boiling point
(Bremner and Mulvaney, 1982). Phosphorus (P) was extracted using
double acid (0.01M hydrochloric acid and 0.0025N sulphuric acid) and
Ammonium Vanadate/Ammonium molybdate used to develop the colour
whose intensity was measured by a spectrophotometer at wavelength
430nm (colorimetric method).
Cations exchange capacity (Ca, Mg, K, Na) was determined by the
summation method. Five grams of air dried soil ground to pass through
a 2mm sieve was extracted using the double acid (0.01 M hydrochloric
acid and 0.0025 N sulphuric acid) and filtered using a Whatman No. 1
size 15cm filter paper (Anderson and Ingram, 1989; Okalebo et al., 1992).
The Mehlich extractable Ca and Mg were deter mined by
spectrophotometer while Na and K were determined by flame photometer.
Exchangeable acidity (Hp) was determined on soils with pH (H2O) less
than 5.5 and entailed extracting (Al +H) using 125ml 1M KCl followed
by titrating the filtrate against 0.05M NaOH with phenolphthalein
indicator (Anderson and Ingram, 1989).
Results and Discusion
i) Macrofaunal abundance
The fauna sampled within the farm included earthworms, termites, ants,
beetles, centipedes, crickets and grubs (Appendix 1). In 1999,
earthworms were the most abundant (46%), followed by termites (24%)
and ants (19%). The remaining macrofauna species that were mainly
litter dwelling (beetles etc) constituted about 10%. In 2000, ants
(Formicidae) were dominant (81%) followed by termites (9%) and
Macrofaunal Abundance and Diversity in Selected Farmer Perceived Soil Fertility
Niches in Western Kenya
493
earthworms 8% with remaining macrofauna species accounting for 2%.
The lower earthworm and termite population in 2000 compared to 1999
could be attributed to the high population of predatory ants (Formicidae
spp). Lavelle et al. (1994) and Brown et al. (1996) also observed that
termites and earthworms (ecosystem engineers and litter transformers)
were the dominant macrofauna in agro-ecosystems.
Niche types also affected the abundance and composition of
macrofauna. Termites (41.5%), earthworms (26%), and ants (24%)
dominated the Oboma while beetles, spiders and centipedes together
were the minority (less than 10%) (Figures 34.2 and 34.3). In the year
2000, ants increased to 89% at the expense of earthworms (4%) and
termites (5%). The high percent population of termites and ants in
"Oboma" in 1999 could be attributed to the large amounts of food resources
present (Figures 34.2 and 34.3). Frequent cultivation that was common
in Oboma may have contributed to the low population of earthworms
through abrasion and overgrazing and trampling (Castilla, 1992).
In the "Vbottom", earthworms contributed (86%) while beetles and
grubs accounted for 11% and 3% of the macrofauna respectively in
1999. Termites were absent from this niche in 1999. In 2000, earthworm
population decreased to 64%, termites 10%, ants 22% and the remaining
macrofauna 3%. The high earthworm and low termite population in the
"Vbottom" was not a surprise because the former prefer moist conditions
while the later are adapted to dry environments (Lavelle et al. 1997). As
a result of the El Nino effect, 1999 was generally wet (Figure 34.1). This
may have further occluded the termites from the "Vbottom".
In the "Cmaize" niche macrofaunal termites and earthworms
consisted of about (47%) and (43%) respectively in 1999. In 2000,
termites and earthworms constituted about 44% each. Ants and other
macrofauna constituted about 10%. Maize stover in the "Cmaize" niche
provided substrate (food resource) and a comfortable microclimate that
may have contributed to the high population of termites and earthworms.
"Oboma" with higher nutrient level (Table 34.1) was expected to have
higher macrofaunal population compared to "Cmaize" (soil fertility
depleted niche). The higher intensity of cultivation (maize/vegetable in
a year) compared to maize alone in "Cmaize" may have caused abrasion
thus the lower density of earthworms and termites. The anisosymbiotic
association of termites with fungi enables them to digest the low quality
(lignin and tannin-protein complexes rich) crop residues (Wardle and
Lavelle, 1997). Tian et al. (1993) and Ayuke (1999) also observed higher
termite population in niches where mulch of low quality was applied.
Tian et al. (1997) observed that soil fauna contributed more to the
decomposition of low quality residue than high quality because they
stimulated microbial activity. This may be the reason why "Cmaize" had
a higher population of earthworms and termites than in the "Oboma".
The strong and longer lasting mulching effect (reduced moisture loss
Tabu, I.M. et al
494
and soil temperature) of the maize stover coupled with the substrate
effect may have led to the high earthworm population in the "Cmaize".
Table 34.1: Chemical characteristics of the farmer perceived niches
Niche
Ca
5.41a 2.91a
(0.09) (0.31)
1.09a
(0.07)
1.61a
(0.14)
2.68a
(0.38)
6.8a
(0.51)
0.95a 0.08a 2.27a
(0.33) (0.01) (0.09)
Npasture 5.5 a 2.55a
S.E.
(0.28) (0.1)
1.28a
(0.22)
1.79a
(0.44)
2.40a
(1.21)
5.95
(1.63)
0.76a 0.08a 2.15a
(1.06) (0.04) (0.29)
5.11a 3.43a
(0.21) (0.73)
0.59b
(0.16)
1.41a
(0.33)
2.14a
(0.90)
5.94a
(1.2)
1.77a 0.15b 2.40a
(0.78) (0.03) (0.22)
0.84
3.74
5.00
12.20
0.65b
(0.08)
1.31a
(0.15)
2.01a
(0.42)
5.67a
(0.56)
Vbottom
S.E.
Forest
(Malava)
Outfield
S.E.
5.35
P
ppm
2.61
4.99 a 2.51b
(0.09) (0.34)
Mg
ECEC
Cmolkg-1
Ex.
Acidity
1.2
N
Organic
Carbon
%
K
Oboma
S.E.
pH
0.21
4.97
1.54a 0.07a 1.90b
(0.37) (0.56) (0.10)
*Standard error in parenthesis
**Values followed by the same letter in a column are not significantly different
In the "Npasture", ants constituted 49%, earthworms 36% while
termites, grubs and beetles constituted less than 5% each. Termite
population in the "Npasture" was reduced probably because of the
predating ants (Formicidae spp) and lack of food resource (litter) attributed
to overgrazing. Endogenous earthworms dominated the "Npasture"
compared to other surface feeding macrofauna probably soil compaction
at the surface could allow only those macrofauna that could be sheltered.
The earthworm population was lower in "Npastures" and "Oboma"
compared to other niches. The most abundant macrofauna were mainly
termites and earthworms, which had little functional diversity i.e., focus
mainly on decomposition and soil structural changes. Generally, the
population of litter dwelling and surface feeding macrofauna was low in
all niches probably because of the low amounts of litter.
Farmer perception of macrofauna was varied. While in the two
villages farmers did not appreciate the influence of termites on the
soil fertility, their counterparts in the neighbouring village observed
better performance of crops around the termite mounds. The
contrasting observation was probably because soils in these villages
were different and the influence of termite activity (transport and
comminution of litter and supply the limiting nutrients (N, P and K)
also varied. Through gallery and mound construction, termites turn
the soil particles and increase the clay content thus affecting the soil
Macrofaunal Abundance and Diversity in Selected Farmer Perceived Soil Fertility
Niches in Western Kenya
495
physical conditions. In Cameroon, Hugulle and Ndi (1993) observed
that termite mounds had more clay compared to the neighbouring site
that did not have mounds.
Figure 34.2: Distribution of macrofauna in farmer perceived niches (1999)
25
Macrofaunal number/m2
20
15
Oboma
Npasture
Cmaize
10
Vbottom
5
0
earthworms
E/worms
Ants
centipedes
termites
Beetles
millipedes
M/pedes
Grubs
Macrofaunal types
This may be responsible for the higher crop yields on termite mound
located on sandy soils. Although farmers noted termite mounds as
sources of variation in soil fertility, they ploughed over and did not
flatten them thus preserving spatial variation in soil fertility. The cost of
flattening these termite mounds may also have overshadowed the
potential benefits. Most research about influence of termite mounds on
soil fertility focuses on comparing the chemical properties of the termite
mounds with those of the adjacent sites.
Conversion of forests into annual crops changed the food structure
and may have eliminated a majority of macrofaunal species that rely on
wood or leaf litter or required specialized microclimate. Okwakol (2000)
also observed that cultivation of forests reduced variety in food types
and ultimately the termite species. Efforts have been made towards
manipulation of macrofauna seeding in earthworms and relating them
to crop yield. Much more has also been done to relate them to soil
structure. Although the termites’ role tends more towards pests, farmers
in a nearby village recognized its influence in soil fertility.
Tabu, I.M. et al
496
Figure 34.3: Distribution of macrofauna in farmer perceived niches (2000)
180
160
Macrofaunal number/m2
140
120
100
Oboma
Npasture
Cmaize
Vbottom
80
60
40
20
0
Earthworms
Centipedes
Ants
Termites
Beetles
Millipedes
Macrofaunal types
ii) Macrofaunal diversity
Niche management significantly affected the macrofaunal diversity
(Figure 34.4). Agro-ecosystems had significantly lower diversity compared
to the natural ecosystem’s (forest) diversity of 2.31. The number of soil
fauna was also generally lower within most cultivated niches. In 1999,
"Oboma" had the highest diversity (0.59) followed by natural pasture
(0.51), continuous maize plot (0.44) and lastly valley bottomland (0.21).
In 2000, "Cmaize" had the highest diversity (0.42) followed by Vbottom
(0.42) , "Npasture" (0.23) and lastly "Oboma" (0.2). The decrease in
diversity in "Oboma" and natural pasture could be attributed to the
high population of predatory ants that scavenged on other organisms.
Brown et al. (1996) also observed comparable diversity in western Kenya.
The reduction in diversity could be attributed to the reduction in
amount, range and diversity of food resources. Micro-climatic conditions
(waterlogging and low temperatures) may have contributed to the low
diversity in the valley bottomlands. Although "Oboma" was managed
highly (with respect to nutrient availability), macrofaunal diversity was
still low compared to that of the forest (natural ecosystem). This implies
that the process in the farmer perceived soil fertility niches may be
Macrofaunal Abundance and Diversity in Selected Farmer Perceived Soil Fertility
Niches in Western Kenya
497
different. A shift from natural ecosystems (forest) to cultivated ecosystems
therefore reduced macrofauna diversity. Apart from the periodic
waterlogging, valley bottomlands are usually perceived as fertile (with
respect to nutrient level). The macrofaunal abundance and diversity
also appeared to respond to the waterlogging and temperature
constraints.
Figure 34.4: Macrofauna diversity as indicated by Shannon-Wiener index under different
management practices at Kabras, Malava, western Kenya
2.5
Macrofaunal diversity index
SED
2
1.5
1
0.5
0
Forest
Cattle boma
Natural pasture
Valley bottom
Management practices
In comparison, even the highly managed niche ("Oboma") did not
measure up to the forest in terms of macrofauna abundance and
diversity. Giller et al. (1997) observed that it was difficult to revert to the
original (forest) macrofaunal status probably because some of the
organisms became redundant and subsequently extinct and others took
up their functions. Thus comparing high and low management status
in the agro-ecosystem may be a better indicator. Variations in
macrofauna population imply that their effect on nutrient cycling was
important.
Forests normally have mixed litter, a high soil organic matter and
microclimate that support a wide range of fauna (high diversity).
Cultivation however removes vegetation that buffers macrofauna against
fluctuation in microclimate. It also alters soil structure, aeration and
physical quantity of litter for decomposition, which subsequently lowers
498
Tabu, I.M. et al
the abundance, and diversity of the decomposer community and
physically damages the macrofauna. In harsh agro-ecosystem
environment, only macrofauna that are buffered (in a nest or stay deep
in the soil) will not immediately be adversely affected by cultivation.
Brown et al. (1996) also observed lower diversity indices in cultivated
sites than natural (forest) sites and associated it to the negative impact
of cultivation on the ecosystem functions (comminution, decomposition)
mediated by macrofauna.
Management practices such as continuous tillage alter the population
structure, eliminate/reduce key groups and species of soil fauna (Beare
et al., 1997). Warren et al. (1987) observed that microclimate, food
resources and land use were major factors affecting diversity and
abundance of soil fauna communities. Temporal activities also affected
macrofauna population. In the "Vbottom" lands, higher diversity could
be attributed to the diversity in food sources and improved aeration
during 2000.
iii) Soil fertility of the niches
Generally, "Oboma" was more fertile (had higher soil organic matter, pH
and ECEC) compared to "Cmaize" and "Npasture". The nutrient level
was even comparable to those available in the forest (Table 34.1). That
the macrofaunal status did not follow the same pattern implies that
other parameters may have contributed to the macrofaunal performance.
Conclusion
Earthworms, ants and termites were the dominant macrofauna in the
agro-ecosystem. The abundance and diversity was however lower than
that in the forests. That the niches had different diversity indices
confirms further that farmers’ spatial-temporal management was
translated into variations in biological status of the niches. Macrofauna
abundance and diversity was related to both soil fertility management
and the agro-ecological conditions in different niches. The use of
Macrofauna as bio-indicators thus need further work i.e. relating with
keystone species and associated processes. Temporal dynamics of this
macrofauna should also be considered before fully using them to
characterize niches.
References
Anderson, J.M. and Ingram, J.S.I. (1993) Tropical Soil Biology and fertility. A
handbook of methods (2nd edn) C.A.B. International, Wallingford United
Kingdom. 221pp.
Macrofaunal Abundance and Diversity in Selected Farmer Perceived Soil Fertility
Niches in Western Kenya
499
Ayuke, O.F. (2000) Diversity, abundance and function of soil invertebrate fauna
in relation to quality of organic residues. M.Phil. Thesis, Moi University,
Kenya.
Carter S. E. and Murwira, H. K. (1995) Spatial variability in soil fertility
management and crop response in Mutoko Communal Area Zimbabwe.
Ambio 24: 77-89.
Beare, M.H., Reddy, V.M., Tian, G. and Srivastava, S.C. (1997) Agricultural
intensification, soil biodiversity and agricultural function in the tropics:
the role of decomposer biota. App. Soil Ecology 6: (1): 87-108.
Blair J.M., Bohlen, P.J. and Freckman, D.W. (1996) Soil invertebrates as
indicators of soil quality pp 273-289. In Doran J. W. and A. J. Jones (eds)
Methods for assessing soil quality. SSSA Special publication No. 49.
Brown G.G., Moreno, A.G. and Lavelle, P. (1996) Soil fauna in Agricultural and
Natural Ecosystems in Western Kenya and Central Tanzania. In Biological
management of soil fertility in small scale farming systems in tropical Africa.
2nd Annual report, 1996.
Giller K.E., Beare, M.H., Lavelle, P., Izac, A.M.N. and Swift, M.J. (1997)
Agricultural intensification soil biodiversity and agro-ecosystem function.
App. soil Ecol. 6: 3-16.
Hullugalle N.J.R. and Ndi, J.N. (1993) Soil properties of termite moulds under
different land uses in a Typic Kandult of Southern Cameroon. Agric. Ecosy.
and Env. 43: 69-78.
Lavelle, P. Bignell, D. and Lepage, M. (1997) Soil function in a changing world.
The role of invertebrate ecosystem engineers. Eur. J. of soil Biology 33:159193.
Lavelle, P.M. Dangerfield, C. Fragoso V. Eschnebrenner, D.L. Hernadez, B.
Pashanari, and Brussard, L. (1994) The relationship between soil
macrofauna and tropical soil fertility. In Woomer P.L. and Swift M.J. (eds).
The biological management of tropical soil fertility. pp.136-169. A wileySayce publication.
Lamers, J.P.A. and Feil, P. R. (1995) Farmers knowledge and management of
spatial soil and crop growth variability in Niger, West Africa. Neth. J. Agric.
Sci. 43: 375-380.
Okalebo, J.R., Gathua, K.J.W. and Woomer, P.L. (1992) Laboratory methods of
soil and plant analysis. A working manual. TSBF, Nairobi.
Okwakol, M.J.N. (1994) The effect of change in land use on soil macrofauna
communities in Mabira forest, Uganda. Africa J. Ecol. 32: 273-282.
Okwakol, M.J.N. (2000) Changes in termite (Isoptera) communities due to the
clearance and cultivation of tropical forest in Uganda. Africa J. Ecol. 38: 17.
Sanchez, P.A., Shepherd, K.D., Soale, M.J., Place, F.M., Buesh, R.J. and Izac,
A.M.N. (1997) Soil fertility replenishment in Africa: An investment in natural
resources capital. In: Replenishing soil fertility in Africa. SSSA special
publication No.51.
Tian, G., Kang, B.J.T., Brussand, L and Swift, M.J. (1997) Soil fauna mediated
decompositions of plant residues under custaed environmental and residue
Tabu, I.M. et al
500
quality and conditions. In cadish G. and Giller, K.E. (eds) Driven by Nature:
Plant lither quality and decomposition PP 125-134, CAB.
Wardle D.A. and Lavelle, P. (1994) Linkages between soil biota, plant litter and
decompositions. In Cadisch G. and K.E. Giller. Driven by Nature: Plant
litter Quality and decomposition PP 125-134, CAB.
Woomer, P.L., Karanja, N.K. and Okalebo, J.R. (1999) Opportunities for improving
integrated management by smallhold farmers in the central highlands of
Kenya. African Crop Sc. J. 7(4): 441-454.
Appendix 1: Macrofauna species sampled from the different farmer perceived niches in
Kabras Division, western Kenya
Order/Group
Family/sub-family
Genera; species
Authority
Oligochaeta
(earthworms)
Lumbricidae
Lumbricus sp.
Linnaenus
• Termitinae/
Macrotermitinae
• Termitinae/
Macrotermitinae
• Termitinae/
Macrotermitinae
Pseudocanthotermes
spp.
Microtermes spp.
Wasmus
• Staphylinidae
Philonthus sp.
Leptocinus
fuscipennsi com
Hypothenemus sp.
Isoptera
(Termites)
Coleoptera
(Beetles)
Microtermes
pusillus.
• Scolytidae
Chilopoda
(Centipedes)
Dermaptera
(Earwigs)
Oligochaeta
(Enchytraeid
worm)
–
–
Forficuldae/
Forciculoidea
Kaischella sp.
Enchytraeidae
–
As special identification keys were not available for some fauna, identification was
therefore based on already available collections. Identification at species level was not
possible in the case of many samples.
Understanding Soil in its Social Context: Integrating Social and Natural Science
Research within AfNet
Understanding Soil in its
Social Context:
Integrating Social and Natural
Science Research within
AfNet
501
35
Ramisch, J.J.
Social Science Officer, TSBF, P.O. Box 30677 Nairobi, Kenya
(j.ramisch@cgiar.org)
Abstract
Continuing dialogue between the natural and social sciences
means that the conception of integrated natural resource
management (INRM) is evolving from largely discipline-based
approaches to more integrative, holistic ones. This paper
presents examples of opportunities for integrating natural
and social sciences including understanding the social
forces driving soil fertility changes, identifying the clients
for new technologies, and improving the sharing of
knowledge and information between farmers and
researchers. It also outlines theoretical and methodological
[This is a substantially modified version of the paper by Ramisch, Misiko, and
Carter entitled “Finding common ground for social and natural sciences in an
interdisciplinary research organisation – the TSBF experience”, presented at
the Social Research conference Looking back, looking forward: Social Research
in CGIAR System, hosted by CIAT, 11-13 September, 2002 in Cali, Colombia].
502
Ramisch, J.J.
approaches for integrating social science into TSBF’s
research activities, and identifies strategic lessons from the
past decade’s research that would be relevant for TSBF’s
partners within the Africa Network for Soil Biology and
Fertility (AfNet).
While individual disciplines still retain preferred modes
of conducting fieldwork (i.e.: participant observation and
community-based learning for “social” research, replicated
trial plots for the “biological” research) a more “balanced”
integration of these modes is evolving around activities of
mutual interest and importance, such as those relating to
understanding on-farm variability and providing decision
support for farmers. Since TSBF works through
partnerships with national research and extension services,
it has an important role in stimulating the growth of common
bodies of knowledge and practice at the interface between
research, extension, and farming. To do so requires strong
champions for interdisciplinary, collaborative learning from
both natural and social science backgrounds, the
commitment of time and resources, and patience.
Introduction
The Tropical Soil Biology and Fertility (TSBF) Programme (now Institute)
was created in 1984 under the patronage of the Man and Biosphere
programme of UNESCO and recently incorporated into the Future
Harvest system of food and environment research centres as a research
Institute of the Centro Internacional de Agricultura Tropical (CIAT). As
an international research body, the underlying justification of TSBF’s
work has been that “the fertility of tropical soils is controlled by biological
processes and can be managed by the manipulation of these processes”
(Woomer and Swift, 1994).
Being an organisation with an explicitly biological and ecological
mandate and origin, TSBF has nonetheless sought social science input
into its research program. However, since TSBF has always been a small
team (never more than six internationally recruited scientists), much of
TSBF’s considerable output has therefore been generated through
collaboration with partner organisations (both national and
international), with a special focus on sub-Saharan Africa through the
African Network for Soil Biology and Fertility (Afnet). The decision to
develop and maintain a core competency at the interface of social and
natural sciences at TSBF has also provided a helpful nucleus for building
social science competency with partners.
Understanding Soil in its Social Context: Integrating Social and Natural Science
Research within AfNet
503
This paper explores the need for greater integration of social and
natural science methods in dealing with soil biology and fertility
management, and the potential for doing so within AfNet or other African
organisations. It presents key programmatic areas where the potential
for synergy is high, and suggests ways of building familiarity and
competency with interdisciplinary methods and approaches.
The second half of the paper examines the historical record of TSBF
and AfNet as “laboratories” for developing meaningful interdisciplinary
dialogue and collaboration, and asks whether what has emerged so far
has been “social soil science” or merely “soiled social science”. Examples
of theoretical and methodological evolution are drawn from “grey” project
literature, personal commentary, and publications. The strategic lessons
from these examples reflect in microcosm the much broader debates
about the potential for “rigorous” science under competing disciplinary
approaches to integrated natural resource management (INRM). They
also address the all too common assumption that the responsibility for
developing a common institutional culture and language within INRM
falls more to social scientist “newcomers” than to biological or natural
scientists.
Relating Natural and Social Sciences
Agriculture is a human endeavour, manipulating plants and soils in a
complicated environment to sustain life and support economic and social
livelihoods. As such, the management of soils always occurs in a social
context and improvements to soil fertility management strategies will
only come about if they satisfy the social and economic needs of farmers.
Traditional agronomic (or soil) research has tended to neglect these
social components as “externalities” that merely impinge on the studied
processes. However, there is great potential for synergy by understanding
the social context of soil management, as the examples provided in the
following sections will show. These examples are grouped around three
key topics that can integrate natural and social sciences in integrated
soil fertility management (ISFM) research:
a) Identifying the social forces shaping soil fertility change, including
economic and demographic drivers, cultural factors, and policy
environments.
b) Identifying the clients for new or improved technologies, the uses
they will have for ISFM, and the decision-making criteria they use
for evaluating both current and improved options.
c) Improving the sharing of soil fertility expertise, by better
understanding existing knowledge systems and improving
communication and dissemination strategies to strengthen them.
504
Ramisch, J.J.
The social context for fertility changes
Because the rural landscape is full of farms, and people living and
working on them, it is often natural to conclude that rural people are
therefore “farmers”. Of course farming (whether for subsistence or
market-oriented production) is only one of multiple rural livelihoods,
such as artisanal work, petty trading, labour exchange, or seasonal
migration.
Even people who indeed consider themselves “farmers” are not just
soil managers, and management ability is a function of knowledge, and
access to key resources (such as land, labour, germplasm, finances,
and inputs). Beyond the farmer’s management ability, the “fertility” of a
soil is also function of inherent bio-physical properties, nutrient balances,
and broader social contexts (Figure 35.1). As a result, decisions to
manage (or to ignore managing) the soil resource are part of a trade-off
analysis that considers the soil within a wider economic or livelihood
sustainability framework. Research can play an important role here in
understanding the conditions under which different interventions are
likely to be profitable or attractive to farmers (cf. papers in this volume
by Kaliba and Rabele, Kipsat et al., and Mutiro and Murwira).
Figure 35.1: Soil fertility as an interaction of socio-economic and bio-physico-chemical
properties.
•
•
•
•
Soil fertility changes, therefore, have their origins in many humanmediated processes that influence the rate and nature of the key
biological and pedological processes (i.e.: erosion / sedimentation,
organic matter decomposition / accumulation, etc.). Social differences
between farmers (in terms of capital assets like land, labour, cash, and
knowledge) and their institutional context will in turn systematically
influence the types of soil management options available and the ultimate
soil fertility status outcomes. Furthermore, social and soil fertility
Understanding Soil in its Social Context: Integrating Social and Natural Science
Research within AfNet
505
changes interact with each other over the long term – strong soil
managers are likely to improve their economic and social well-being
while weaker ones may become trapped in declining or vulnerable
livelihoods.
For example, Vihiga district in western Kenya is one of the most
densely populated regions in sub-Saharan Africa, with between 1200
and 1400 people per km2. Although endowed with a high potential climate
and inherently fertile soils, the region’s political history has left longdistance markets and infrastructure poorly-developed. Out-migration
(particularly by young men) is extremely high, which serves both to give
households access to off-farm income, and also as a means to reduce
pressure on households to sub-divide their land amongst sons. Wealth
ranking conducted in this area (Table 35.1) shows that the “wealthiest”
households have better access to off-farm income, which can be used
to pay hired labour and to support more intensive soil management.
Less intensive soil fertility management strategies are associated with
the middle and “poor” households, which often have to sell household
labour to others.
Table 35.1: Relationship between ISFM practices and wealth class in Vihiga District, Kenya
Agricultural labour
Off-farm income
Use fallowing
Rotate crops
Make compost
Regularly use manure
Ever used inorganic fertiliser
Land has SWC terraces
Wealthiest farmers
(n=34)
Poorest farmers
(n=59)
Hired + family
Family only
56 %
12 %
32 %
53 %
91 %
68 %
91 %
20 %
0%
22 %
42 %
59 %
42 %
39 %
(Carter and Crowley . Unpublished data)
It is apparent from these data that the soil fertility problems of the
“wealthy” farms would therefore differ significantly from those of the
other households, and that soil fertility changes (either improvement or
decline) are strongly related to the socio-economic and political dynamics
of households’ access to resources. The notion of establishing farmer
typologies that relate household characteristics to land use behaviours
and soil fertility outcomes has therefore taken root as one of the most
useful interactions between social and natural sciences within ISFM
research. Not only do these typologies improve the ability to explain
existing patterns of soil fertility, but they facilitate better targeting of
recommendations and decision support advice.
506
Ramisch, J.J.
Technologies for whom?
It is widely recognised that the adoption of new soil fertility management
technologies is uneven. Since not all farmers have similar needs or
constraints, many studies attempt to determine the adoption potential
for new technologies based on farmer characteristics (cf. Kaliba and
Rabele, this volume). Socio-economic differences (as discussed in the
previous section) may indeed help explain why some farmers are better
able to afford the land, capital, or labour needed to experiment with or
to use a new technology.
However, equally important for understanding the relationship
between farmer difference and the acceptability of technologies is the
notion of a farming system’s “precision” (Reese and Sumberg, 2003).
For a variety of reasons, farmers in rural Africa are often not in a position
to act on or implement their decisions or plans in the precise manner,
or at the precise times, that they might wish. Richards (1989)
demonstrates that “how people actually farm” often contrasts sharply
with how they might “ideally like to farm”. The intervening reasons might
be climatic (the rains may be early or late, too short or too heavy),
institutional (the desired inputs such as seeds or fertilisers may not be
available when required or at a reasonable price), or related to the
household itself (labour appropriate to a specific task might not be
available, because of competing demands, illness, or indeed simple “bad
luck”).
Farming systems where farmers exercise relatively little control over
key components of their environment (low precision systems) differ
markedly from those where they exercise more control (high precision
systems). For example, maize in sub-Saharan Africa is often planted
later than the ideal date because of labour constraints, risk
considerations, and crop rotations, with consequent yield reductions of
up to 75% compared to the optimal planting date (Byerlee and Heisey,
1996, using Zimbabwe as example).
The argument of Reece and Sumberg (2003) is that agricultural
research has historically tended to neglect differences in farming system
“precision”, even while working to ensure that technologies continue to
give acceptable yields across a range of environmental conditions. While
plant breeding prioritises the yield “stability” of a crop variety over an
environmental gradient (subject to minor location-specific adjustments)
farmers who are unable to provide the precise management anticipated
by the researcher may suffer significant yield losses. Clearly, research
that is developing technologies for use in “low-precision” farming systems
must acknowledge that farmers’ management practices will vary, making
questions of management adaptability as important as those of
environmental adaptability.
Understanding Soil in its Social Context: Integrating Social and Natural Science
Research within AfNet
507
Related to the concept of a farming system’s “precision” is that of a
technology’s “solution space”: the “area around an optimal set of
operator-influenced conditions within which a technology will still yield
‘positive’ results” (Reece and Sumberg, 2003: 416). In Figure 35.2, the
yield response of the crop in (A) is highly sensitive to the date of input
application, whereas the option shown in (B) obtains a lower maximum
but sustains favourable yields over a wider range of application dates.
The area under the two curves represents the “solution space” of the
two different technologies. The narrower solution space of technology
(A) would be appropriate for a farmer who can control the management
variable (in this case application date) with the needed precision. The
second option (B) has a broader solution space and so would be more
suited to a lower precision farming system.
Figure 35.2: Comparing the response of two technologies with different “solution spaces”.
To better match the precision of farming systems to technologies,
technology design must involve the intended beneficiaries earlier. Figure
35.3 shows an idealised flowchart of a participatory technology design
process, with control of the design moving from “formal” to “farmer-led”
as soon as the comparative advantage shifts from researcher to farmer.
The actual solution space of a given technology will become apparent as
technology development takes place, defined as a direct result of the
choices made (and options excluded) during this process by the people
who will ultimately decide to use the technology. Thus the solution space
that is defined for a given technology being perfected by its users will
inevitably come to correspond to the precision of their farming system,
and that space will also be smaller than the range of possibilities that
had been open at the initial stages.
508
Ramisch, J.J.
Figure 35.3: Participatory technology design process.
The role of research in this process is therefore to identify “prototype”
technologies of interest to farmers, and then immediately involving the
group(s) most likely to benefit in the next steps of designing and refining
the technologies (cf. papers in this volume by Odendo et al., and Miiro
et al.). After all, most agricultural technologies in use today were designed
by farmers. Such a collaborative research strategy is attractive not only
because it empowers farmers to seek new options more confidently, but
also because it reduces the likelihood of investment in “dead end” or
non-adopted technologies, thereby ultimately reducing research costs.
Participatory research strategies, however, are only slowly taking
root in TSBF and AfNet. It should be acknowledged, though, that the
“over-designing” of technologies before involving farmers in their
development is a natural consequence of scientists failing to a) trust in
the innovative capacity of farmers or b) know how to apply farmers’
knowledge and innovation as contributions to “formal” scientific activity.
It limits farmers’ role to relatively passive activities, such as selecting
niches or adapting application rates to local circumstances, which
ultimately discourages any sense of ownership of the technology
development process. However, to recognise certain behaviour as an
“innovation” requires channels of communication and trust to exist
between farmer and scientist (see the next section), and a willingness to
see all modifications of practice (including abandonment and complete
reversals) as potentially useful.
Understanding Soil in its Social Context: Integrating Social and Natural Science
Research within AfNet
509
Observations of innovative farmer practice can feed into researchable
topics, such as the use of Tithonia as a nutrient-rich mulch (now a staple
“technology” promoted by TSBF and others in East and Southern Africa).
When translating the Tithonia biomass transfer technology to other farms,
a commonly heard comment is that the cut-and-carry system is “labour
intensive”. Harvesting biomass from hedgerows all at once before planting
one’s crops is indeed a large, and previously non-existent task, even if
pruning hedgerows or applying plant material on cropland are familiar
activities already in the household calendar. As a result, many farmers
have begun harvesting their Tithonia sporadically (as part of normal hedge
maintenance) and transferring it to their compost pile (another familiar
task). Clearly the decision not to continue with the cut-and-carry operation
and instead supplement the compost pile with Tithonia should be seen
as an “innovation” or indeed as a logical supplementation of existing
practices. However, while Tithonia had been identified as a “best bet” for
direct application to fields, it may not be the “best” option for materials to
be added to compost piles. A natural entry point for truly interdisciplinary
research would be experimentation based on farmers’ own practices (many
report that Tithonia speeds the “cooking” of compost piles making it ready
for use sooner) to validate the use of Tithonia or other materials as part
of the composting process.
Sharing experience better: local knowledge and decision
support
If farmers’ experience of soil fertility change is a function of their different
socio-economic constraints and opportunities then there are also clear
implications for dissemination and technology adaptation. One is that
local knowledge of soil ecology (“Folk Ecology”) is itself an important
entry point for scientists wishing to understand and build on local
practices. Building a shared language then facilitates the translation of
strategic research principles into applied tools, such as those that can
assist farmers making land use decisions. The second implication is
that dissemination and adaptive learning strategies must acknowledge
that not everyone will be reached by the same methods. This suggests
that methods must be targeted towards the known potential users, and
also that if a diversity of potential users is identified, multiple strategies
might need to be employed to avoid favouring some groups over others.
The original starting point for scientists and farmers trying to build
a common understanding of soil has been local taxonomies. Local names
and descriptions of soils have the longest history of use by soil scientists,
who recognised that the subtleties of farmers’ terminologies reflect the
intimacy of frequent interactions and reliance on the land around them.
However, it is also important to understand how local people recognise
and monitor changes in the soils that support their livelihoods. Many
510
Ramisch, J.J.
local concepts of soil fertility mirror the terminology used to describe
human health (this is also true of some scientist versions of these
concepts). Farmers will refer to soils that are “tired”, “sick”, or “thirsty”,
and also to soils that have become “addicted” to chemical fertilisers.
However, many local knowledge systems treat soil in a much more holistic
fashion, seeing its well-being as embedded within broader systems, for
example recognising that crop-soil health is itself strongly influenced by
pest dynamics and climate variation. The problems of soils may also be
attributed to supernatural origins, such as the neglect of traditions, taboos,
or rituals that would have renewed the soil’s fertility (cf. Richards, 1989).
These latter “cosmological” aspects of local knowledge are often the
ones most criticised by scientists when minimising the importance of
dialogue between local and scientific traditions. Nonetheless, it is not
possible to ignore this local knowledge base, since local people will
continue to make land use decisions based on its assumptions. Initiating
a dialogue that will build on the strengths of local knowledge can also
facilitate the process of filling the “knowledge gaps” that are also present,
and modifying or replacing negative practices. The very fact that local
knowledge often varies between individuals (as a function of gender,
age, geography, ethnicity, or livelihood), and indeed that it is not
necessarily organised as systematically, coherently, or comprehensively
as more “formal” knowledge means that it is essential to find ways to
bring local and new knowledge systems together.
Dissemination and decision support strategies must therefore
confront this diversity. Materials and methods must recognise that
farmers have greatly varying abilities, knowledge, and assets, and that
“one size” will not “fit all”. As shown in Figure 35.3 above, any given
project or technology can conceivably result in multiple finished
knowledge outputs, depending on how well the initial ideas have been
used and modified by the people who are likely to be interested in or
able to benefit from that knowledge. The decision support guides to
support improved knowledge should therefore reflect the production
and livelihood goals of those clients / co-researchers, as well as their
biophysical and socio-economic assets. These are daunting challenges,
but with the help of better understanding of local conditions, local
knowledge, and the use of better simulation and modelling tools TSBF
and AfNet are helping to meet them (cf. Amede, this volume).
Evolution of Theories and Methods within TSBF
and AfNet
The development of a TSBF research agenda that looked beyond the
soil to the people cultivating it has moved from descriptive,
characterisations of farming systems to more strategic study of social
Understanding Soil in its Social Context: Integrating Social and Natural Science
Research within AfNet
511
differentiation, power, and networks as they relate to soil fertility
management innovation. An interest in dissemination has broadened
into investigation of social dynamics, knowledge, and farm-level decisionmaking. There has also been a tradition of self-reflection, examining
the consistency and coherence of TSBF’s stated goals, methods, and
actual practice, as well as the extent to which grassroots action conforms
to its depiction to outsiders. As such, social science practice has
developed quite healthily over the last ten years 1992-2002, driven
significantly by the following factors:
• The disciplinary background of the Social Science Officer (and to
a lesser extent, that of field staff). Three people have held this position
– Simon Carter (1992-1997, Geographer), Patrick Sikana (1998-2000,
Anthropologist), Joshua Ramisch (2001-present, Human Ecologist)
– and each has had preferred research topics and interests. In
addition, Eve Crowley (1994-1996, Anthropologist) worked with TSBF
on a Rockefeller Social Sciences Fellowship; a position shared half
time with ICRAF.
• The demand for “socio-economic” understanding of processes
being studied by other TSBF staff and collaborators.
• The natural evolution of projects from inception to later stages.
This organic growth has typically moved from characterisation using
very descriptive studies to more explanatory work building on existing
practices through to development of longer-term interactive learning
activities.
• Evolving social science debates concerning knowledge, power, and
participation. The co-supervision of MSc and MA students has been
an especially useful vehicle for maintaining contact with these
debates.
• Responding to donor agendas, including but not limited to perceived
needs for research results readily useful to farmers, a clearer
understanding of agrarian change and its links to changes in soil
fertility, livelihoods analysis, impact assessment, and identifying the
most effective ways of “scaling up” organisational successes.
Demand driven – but by whom?
There has always been a tension between the research agendas
demanded from within TSBF by social scientists (i.e.: disciplinary
interests, evolving projects and debates) and those expected from
outside (i.e.: from other TSBF staff, partners, donors). This tension
results from different research paradigms and differing ideas about
the role of research in relation to social change. From the natural
science perspective, the key contribution of social science to INRM
often appears to be identifying and understanding the social factors
that limit “adoption” or the “appropriateness” of given technologies.
512
Ramisch, J.J.
Other socio-cultural phenomena, such as “policy” might be
acknowledged as important to the fate of different innovations, but
most teams (even multi-disciplinary ones) lack the capacity to generate
relevant policy-related questions, experiments or interventions. In other
words, when the organisation is researching natural resource problems,
the natural-social science dialogue has most often begun with
identifying “black boxes” of external, social forces that need
illumination, rather than defining truly interdisciplinary questions
about how research (including technical research) can support positive
change in rural societies.
This tension is reflected clearest in the history of the social science
position itself (for fuller discussion, cf. Ramisch et al., 2002). Created in
1992, the post was originally charged with “Resource Integration”. This
step was perceived as a natural evolution for TSBF, which always held
an ecological, systems-oriented approach to thinking. Although TSBF’s
strength remained at the plot level, the diversity of forces impinging on
the plot draws attention naturally towards a broader, systemic analysis
(Scholes et al., 1994).
The Resource Integration Officer was therefore initially charged with
“developing a model for integrating biophysical and socio-economic
determinants of soil fertility for small-scale farms” (Swift et al., 1994).
Under this rubric, social factors were expected to be integrated into
holistic models as additional explanatory variables. Once key and
perhaps universal variables were identified, these could then be added
to a “minimum set” of characterisation data collected for TSBF sites (cf.
Anderson and Ingram, 1993). However, the main contributions to the
TSBF programme remained in terms of site selection, selection of themes
for process research, and client group selection, with much less emphasis
on experimentation, or monitoring and evaluation (Crowley, 1995). This
can be seen in the earliest social science work of AfNet, which included
developing simple GIS databases for East Africa, a more detailed one
for western Kenya, detailed formal survey work in western Kenya,
participatory characterisation of farmers’ recognition and management
of farm and landscape-level management of soil variability in Kenya
and Zimbabwe.
The most fundamental methodological evolution over the last decade
has been from largely descriptive, empirical work towards developing
more theory-driven, strategic research and the broader use of
participatory approaches. At the same time, there has been a search for
the optimal degrees of participation relating to the “fieldwork” aspects –
which actors, doing which tasks, using which methods. This search
has highlighted some of the still extant divides between the rhetoric of
research aims and the realities of operational daily practice, as well as
tensions that exist between different models of the role of research in
stimulating change.
Understanding Soil in its Social Context: Integrating Social and Natural Science
Research within AfNet
513
Social science within AfNet
The AfNet membership is still overwhelmingly natural scientists (over
150 soil scientists, biologists, agronomists) with social science
represented in 2002 only by six (socio)economists. While there is a
general appreciation that “social science” is important to the network,
there is still great unfamiliarity with what can really be offered or
understood. The emphasis remains on economic information about the
“profitability” or “adoptability” of known technologies, with no expertise
or experience in applying strategic, interdisciplinary research questions
at the interface of human-environment interactions to soil fertility
management. AfNet could have made it a higher priority to try to attract
more social scientists, but soil and agricultural scientists need to be
trained to recognise where social science can make their lives easier.
This has to happen at university and in special training courses, and
(rather like gender mainstreaming) has to have soil and agricultural
scientists as its champions, not just the social scientists. Host institutions
have also to provide the space for scientists to engage in interdisciplinary
research. Unfortunately, while recognised by the various AfNet
coordinators, this has tended to be subsumed, and therefore obscured,
within the larger problem (true within AfNet as within the CG system
more generally) of declining numbers of soil scientists faced with
increasing obligations and expectations.
There has been significant turnover of personnel at TSBF since 1992,
most notably the tragic loss of Patrick Sikana in the 2000 crash of
Kenya Airways flight KQ431. Problems with the continuity of personnel
at TSBF and within AfNet have had major impacts on developing an
interdisciplinary and social science research agenda that is based on
institutional memory and a coherent agenda. Within partner
organisations, the retrenchment of public sector employees (as part of
structural adjustment or other “reform” programmes) has gutted national
research bodies and extension services. The relatively low numbers of
social scientists present in national systems must also be seen in the
light of the stark fact that they tend to be much more attractive to
donors and thus more likely to move on from low paid national positions.
Social scientists trained in participatory methods are also much less
likely to return to agricultural research jobs when conservation and
health present opportunities in more prominent and well-funded fields.
Finally, staff turnover in African organisations has tragically been
exacerbated by sudden deaths like Patrick’s, attributable to accidents,
disease, and general insecurity.
The 8th AfNet meeting held in Arusha in May 2001, also clearly
demonstrated that amongst partners TSBF is still perceived essentially
as a biology-based organisation with minimal social science input.
Active recruiting of social scientists has begun through networking
514
Ramisch, J.J.
and proposal development, but has been complicated by the rapid
expansion of AfNet in the past two years. The massive influx of new
members and the expansion of activity into West Africa have
simultaneously increased the potential demand for ISFM input and
diluted the few interdisciplinary voices present within the network.
The AfNet mandate of increasing the use of “integrated” approaches
frequently takes a back seat to its more “traditional” and familiar
mandate of increasing support of biological approaches to partner
institutes through curriculum development and networked
experiments. The role of social science within AfNet remains an
unresolved problem, acknowledged as important (for “integrated”
resource management, for greater “adoption”, and ultimately donor
approval of soil fertility management topics) but not backed by
resources or strong champions within the network.
The lack of “champions” for social science research within TSBF
can also be seen in the example of Ritu Verma, an IDRC-funded MA
student who worked with TSBF in western Kenya from October 1997 to
April 1998. Her research comprehensively examined gender and
agricultural practice but without a strong link to the core of TSBF was
never meaningfully integrated into other projects. Ironically, her book
Gender, Land, and Livelihoods in East Africa: Through Farmers’ Eyes
(Verma, 2001) is the most extensive TSBF text produced by social science
research but presents its arguments in such detail that it has been
difficult to absorb or disseminate, making it a testimony to missed
opportunities.
A final point to note is that all of the social scientists who have
worked at TSBF have been relatively young and in the early stages of
their careers, whereas the biological scientists have generally been more
senior. The onus has been on the social scientists to communicate novel
ideas in terms their colleagues could understand or accept; this was
relatively easy with concepts such as spatial variability, but much harder
with feminist political ecology. Furthermore, in the past, strong
personalities or opinions have tended to block communication between
individuals and to limit interactions within the team. The new team
that came together in early 2001 has begun to overcome some of these
historical difficulties, further stimulated by meetings held in conjunction
with the union with CIAT and the formation of the strategic Alliance for
ISFM between CIAT, TSBF, and ICRAF. However, without a more senior
social scientist or generalist present to mentor or to mediate
communication, interdisciplinarity will always be a challenge.
“Research” or “action research”?
The development of social science at TSBF has been implicitly predicated
on two very different models of how change is brought about in rural
Understanding Soil in its Social Context: Integrating Social and Natural Science
Research within AfNet
515
communities and what role outsiders and scientists can play in that
process:
• The more conventional approach suggests that a “good technology
sells itself” and that working with communities merely requires that
the “best bet options” are made available to the “categories of farmers”
who are likely to benefit from them. In this model, which is still
widely held by many natural scientists including TSBF partners, a
“research” organisation has too few resources and no comparative
advantage in doing dissemination, and is better placed to research
and evaluate the dissemination and technology promotion activities
carried out by partners (local NGO’s or national agricultural bodies).
• The alternate approach argues that understanding local processes
of innovation, resource distribution, resource allocation decisions,
and information transfer is essential to developing technologies
relevant to their users’ conditions. Integral to this second approach
is the development of meaningful communication and learning across
disciplinary boundaries – something that TSBF has attempted to do
repeatedly, but which still remains problematic.
As TSBF and its partners have become more versed in participatory
methods, tension has developed between these models. The desire for
more “development” oriented activity has been highlighted in the
redesigning of the “Resource Integration” theme of TSBF in 2000 into
the new Focus 1, demonstratively titled “Empowering Farmers”, into
which all the other bio-physical Foci’s arrows flow. It may also have
been further accentuated by the recruitment in the late 1990s of TSBF
field staff for Kenya with NGO backgrounds in action research. The
argument has been that without actively engaging in dissemination and
community organisation the phenomena of interest to research
(knowledge flows, further innovation and adaptation, etc.) will be too
scarce to be viable or observable. Indeed, these staff members have
found it difficult to define or implement “research” as an independent
activity, devoid of extension or development components.
In reality, most partner organisations have lacked the resources
(personnel, transport, and operating funds) to carry out such work,
and indeed have often turned to TSBF for material or logistical support.
The decision to devolve more of the research, experimentation, and
dissemination activities to the host communities, therefore, is not so
much ideologically driven as pragmatic. The increasing use of farmerdesigned and farmer-run experiments, farmer-to-farmer training, and
group-based activities has effectively begun to address the desire for
more “action” oriented work while providing social processes worthy of
investigation. What has emerged in the project areas of western Kenya
(where TSBF and local groups have had a reasonably long, 5-8 year
history of contact) are prolonged, one-to-one relationships between
516
Ramisch, J.J.
scientists and farmers. Interactive, two-way learning, through
community-based interactive sessions and farmer-based
demonstrations, has been enhanced by researchers, and is widely
conducted in local dialects. The ongoing challenge, however, has been
finding optimal roles for researcher, extensionist, and farmer
participation under these continuing conditions of resource constraint.
Collaboration and “participation”
Under the prevailing orthodoxy of participation, it is difficult to find
projects that do not describe themselves as using and embracing
“participatory” methods, to the extent that the term invites dismissal or
covert cynicism (cf. Cooke and Kothari, 2001). These methods are usually
assumed to apply only to relationships between researcher / extensionist
and “client”, where they are used to “level” the power relationships
between actors. Yet in the TSBF context, where planning and
implementation of activities is explicitly done in partnership with national
research and extension institutions, participatory methods of
collaboration have had to evolve. If cross-disciplinary learning has been
difficult within TSBF, it has been even more so between TSBF and its
partners, a fact which must be acknowledged before looking at the
effectiveness of “participation” in the dealings of “researchers” with
farmers.
This point needs to be based on what might be called “realistic
expectations” of change. True collaboration must recognise (however
reluctantly) that working with the human resources that are on hand
within networks means starting from the perceptions and skills of those
partners and moving at the best pace possible. It would have been easy
to “cook” fancy results about participation if the social scientists had
simply gone it alone. Working in partnership through AfNet, however,
has forced TSBF to confront the realities of public funded research in
Africa, the conservatism and logistical difficulties of which demand
considerable patience. It is relatively easy for partners to influence each
other’s rhetoric, harder to alter each other’s conceptualisations of
problems, and harder still to make lasting changes in the way each
carries out research tasks. “Participation” is not an approach whose
benefits are learned or appreciated quickly and the socialisation of
knowledge backwards and forwards between scientists and farmers
depends fundamentally on the generation of experience.
The progress of AfNet towards “internalising” the rhetoric of farmer
participatory research may seem slow even if it is one of the more
advanced scientific networks (cf. review of on-farm research in the EUfunded project, Carter et al., 1998). As mentioned above, the scarcity of
AfNet members trained in participatory methods able to act as
“champions”, and the lack of continuity in many institutions facing
Understanding Soil in its Social Context: Integrating Social and Natural Science
Research within AfNet
517
financial crisis, hinder the development of a more interdisciplinary
research culture.
However, progress is being made in learning new attitudes and
unlearning old ones. For example, in the EU-funded project, the Zambian
team decided to work on the local fundikila mound systems and to replicate
the farmers’ practices on-station to validate the system in full view of
their peers. Among other AfNet partners, research teams in Zimbabwe
and Kenya now acknowledge the various micro-niches that farmers
recognise and manage and have incorporated these into various research
designs. Increasingly sophisticated understanding of wealth and gender
differences as they relate to soil fertility management have also been
incorporated into more recent project designs. Finally, previously distinct
elements of process and on-farm research have been combined in activities
where complex soil-crop scenario modelling has been fed back into
negotiation or decision support work conducted with farmers.
The politics of community-based research
It is, of course, never easy to surrender control of research agendas,
even where the research is ostensibly for the benefit of the rural poor
(i.e.: TSBF’s Theme 1 is “Empowerment of Farmers” with new
technologies). If TSBF has truly embraced the devolution to farmers (or
other stakeholders) the major responsibility for adaptive testing and
sharing of accountability for quality control over research, what have
the political implications of this move been?
As TSBF placed more attention on building capacity in its partners
for farmer participatory research, it also shifted to working with local
farmers as groups and individuals. In the earlier 1990s, on-farm trials
were based on individuals' farms. In such arrangements, host farmers
were expected to define and explain experiments to other local and
visiting farmers. While we do not know the exact accomplishment
through this arrangement, there are indications in Kabras and Vihiga
that selecting “model” farmers to work with disaffects them from many
other farmers.
Down the road, focus shifted to the group approach. Initially, it
seemed obvious that involving many farmers would have a multiplier
effect. However, it soon became apparent that the manner in which
TSBF talks (and to whom) is more important than mere numbers. Groups
are frequently unstable and many are not especially open to new
membership. When researchers request farmers to work with them
collectively, “new” groups emerge. But these “new” groups usually
comprise members of a previous, defunct group. This means that one
has to deliberately seek the inclusion of all types of farmers (within and
outside groups) in research and dissemination. This role of a local unifier
is tricky and can even appear comical before local farmers.
518
Ramisch, J.J.
Intervening research on the nature of social capital and the role of
local groups and networks in passing agricultural information (Misiko,
2001) has shown that there is still a tendency for some groups or
individuals to view their participation in TSBF as “secret knowledge”
that is not to be shared with others. Likewise, non-participants are
often wary of inquiring about project activities, assuming that they are
not welcome or need to be invited by some patron. This attitude has
persisted for multiple reasons, and in spite of the considerable efforts of
TSBF and other research bodies to present their work as “open to all”
by actively seeking to include marginalized groups. Because local politics
takes precedence even over the “good intentions” of outsiders, the vast
exposure that many farmers have had to project work in western Kenya
does not, therefore, translate into widespread use or understanding of
ISFM.
The initial willingness of TSBF to accept “groups” as representatives
of community interests has led to numerous problems. After all, groups
exist and persist when they have strong roles and identities, histories of
their own which often only become known with time. For example, the
most vocal members of groups have frequently been people who are
either not well respected by others locally, or possessed of agendas that
run far beyond ISFM. This later group tends to see the research project
as a vehicle for access to new resources and political leverage than as
an opportunity for new learning (Sikana, 1995), although it may take
project staff a long time to appreciate this reality. Since much of TSBF’s
on-farm work has been initiated in the context of structural adjustment
programmes and the cessation of donor funding for major local
development projects, it is natural that farmer concerns about water,
health, poor infrastructure, or education would be mapped onto the
“research” activities if TSBF was the only “development” agency working
in their area. Beyond such explicit “hijacking” of groups, there are
frequently tensions between participants over the definitions of goals,
membership, and indeed the “success” of the group’s activities.
Nevertheless, working through groups provides an opportunity to
diffuse risk and broaden responsibility and ownership of activities.
Groups should be seen neither as a panacea for community-based
management’s difficulties, nor as a replacement for effective
dissemination strategies. When setting up experiments or
demonstrations at the local level, having wider input about where in
the landscape, whose land, or which soils are suited to which types of
research activity has proven invaluable. With our broadened knowledge
of the diversity of local soil types, requests by farmers to have activities
replicated on different soils become logical and understandable, when
previously they might have been dismissed as unjustified demands for
a share of a perceived research “pie”. In the end, such replication turns
out to be both good science and good politics.
Understanding Soil in its Social Context: Integrating Social and Natural Science
Research within AfNet
519
Strategic Lessons: Finding Common Ground
Building on the easiest topics
The challenges that TSBF has tried to address are highly complex in
both biophysical and social terms. As such, interdisciplinary
collaboration depends on developing a better understanding of what
changes are taking place, and of developing a modus operandi that can
generate useful knowledge as part of an on-going dialogue between
scientists and farmers.
The parallel dialogue that must take place, between social and
natural scientists, has been easiest around themes that integrate
themselves readily into natural science work, including spatial variability,
wealth ranking or farmer typologies as they relate to ISFM practice, and
understanding the strengths and weaknesses of existing local knowledge.
It has been considerably harder to incorporate elements that relate to
the political nature of “research”, such as using livelihoods analysis or
feminist political ecology to find the place of ISFM and research
interventions within local practice.
Championing workable models
If AfNet, collaborators have been slow to adopt interdisciplinary and
participatory approaches. It is due in part to the relative lack of
successful, convincing models of how such approaches pay short or
long-term benefits to ISFM research. Further constraints have been
staff turnover (which leads to fragmented agendas and loss of
institutional memory), scarcity of time and resources, and a shortage of
generalists or social scientists within partner organisations. The rhetoric
of interdisciplinarity and participation have rapidly infiltrated research
bodies because they are relatively cost free and often there is the
perception that donor funding is linked to such language. Simplified
versions of interdisciplinary activities, linking ISFM with participatory
wealth ranking, or moving from local soil taxonomies to broader
understanding of how soil fertility is managed locally, have also begun
to take hold within local practice. While some natural scientists are
“afraid of having to become social scientists”, there is a slowly growing
constituency within AfNet that sees advantages for interdisciplinary
collaboration. Nevertheless, without relatively senior “champions” for
interdisciplinary or socially oriented approaches within TSBF, new
methods and approaches are at a disadvantage compared with the more
familiar status quo.
520
Ramisch, J.J.
Negotiating the role and nature of “research”
Given the variables of donor climate, institutional and personnel changes,
and socio-political change on the ground, truly interdisciplinary ISFM
research will need to develop a common language and common priorities
that can form a core identity in dealing with outside forces. This requires
an iterative process of negotiating the role of “research” in the
development of local communities. If donors, researchers, and
extensionists feel the need to “scale up” local successes and achievements
to broader communities, it must be reconciled with the desires of the
initial community members for taking research accomplishments to
greater depth. If moving towards group-based research methods means
shifting the burden of implementation to national partners, a common
path for “participation” will need to be negotiated. In particular, the
skills and attitudes necessary to support more decentralised forms of
research need to be cultivated by the scientists, agents, and farmers
involved.
Despite the rhetoric of interdisciplinary collaboration, crossdisciplinary learning and communication remain complicated by the
divergent ideas of what role “research” can and should play in bringing
about change in rural communities. Resolving these divergences often
falls to social scientists, since their disciplinary orientation predisposes
them to thinking about such issues and their colleagues are more likely
to see these issues as somehow separate from their daily activities of
research. However, building common bodies of knowledge and practice
can only happen with the full participation of all disciplines involved in
ISFM.
References
Anderson, J.M. and Ingram, J.S.I. (1993) Tropical Soil Biology and Fertility: A
Handbook of Methods. 2nd edition. Wallingford, UK: CABI.
Byerlee, D. and Heisey, P.W. (1996) Past and potential impacts of maize research
in sub-Saharan Africa: a critical assessment. Food Policy. 21(3): 255-277.
Cooke, B. and Kothari, U. (2001) Participation: The New Tyranny? London: Zed
Books.
Carter, S.E., Chuma, E., Goma, H.C., Hagmann, J., Mapfumo, P., Ojiem, J.,
Odendo, M., Riley, J. and Sokotela, S.K. (1998) On-farm research in the
TSBF Programme: Experiences in smallholder systems of tropical Africa.
Chapter XI in, Carter, S.E., Riley, J. (eds.) Final Report: Biological
Management of Soil Fertility in Small-scale Farming Systems in Tropical Africa
(EU Project ERBTS3*CT940337). Pp 189-206.
Crowley, E.L. (1995) Some methods for characterising social environments in
soil management research. In, Carter, S.E. (ed.) Proceedings of the 1st project
Understanding Soil in its Social Context: Integrating Social and Natural Science
Research within AfNet
521
workshop (Annual Report to the EU for 1995), held at Sokoine University of
Agriculture, Morogoro Tanzania, April 19-24, 1995. Project
ERBTS3*CT940337: Biological management of soil fertility in small-scale
farming systems of tropical Africa. Pp. 99-110.
Crowley, E.L. and Carter, S.E. (2000) Agrarian change and the changing
relationship between toil and soil in Kakamega, Western Kenya, 1900-1994.
Human Ecology. 28(3): 383-414.
Misiko, M.T. (2001) The Potential of Community Institutions in Dissemination
and Adoption of Agricultural Technologies in Emuhaya, Western Kenya. MA
Thesis. Nairobi: Institute of African Studies, University of Nairobi.
Ramisch, J.J., Misiko, M.T. and Carter, S.E. (2002) Finding common ground for
social and natural sciences in an interdisciplinary research organisation –
the TSBF experience. Looking back, looking forward: Social Research in CGIAR
System, CGIAR conference hosted by CIAT, 11-13 September, 2002. Cali,
Colombia.
Reece, J.D. and Sumberg, J.E. (2003) More clients, less resources: towards a
new conceptual framework for agricultural research in marginal areas.
Technovation. 23: 409-421.
Richards, P. (1989) Agriculture as a performance. In, Chambers, R., Pacey, A.,
Thrupp, L.A. (eds.) Farmer First: Farmer Innovation and Agricultural Research.
London: Intermediate Technology (IT) Publications. Pp 185-195.
Sikana, P.M. (1995) “Who is fooling who? Participation, power, and interest in
rural development” Paper presented by special invitation at the International
Development Research Centre (IDRC), June, 1995. Ottawa, Canada: IDRC
(unpublished).
Scholes, M.C., Swift, M.J., Heal, O.W., Sanchez, P.A., Ingram, S.J.I. and Dalal,
R. (1994) Soil fertility research in response to the demand for sustainability.
In, Woomer, P.L. and Swift, M.J. (eds.) The Biological Management of Tropical
Soil Fertility. Chichester, UK: John Wiley-Sayce. Pp 1- 14.
Swift, M.J., Bohren, L., Carter, S.E., Izac, A.M. and Woomer, P.L. (1994) Biological
management of tropical soils: Integrating process research and farm practice.
In, Woomer, P.L. and Swift, M.J. (eds.) The Biological Management of Tropical
Soil Fertility. Chichester, UK: John Wiley-Sayce. Pp 209-228.
Verma, R. (2001) Gender, Land, and Livelihoods in East Africa: Through Farmers’
Eyes. Ottawa: IDRC.
Woomer, P.L. and Swift, M.J. (eds.) (1994) The Biological Management of Tropical
Soil Fertility. Chichester, UK: John Wiley-Sayce.
522
Ramisch, J.J.
Linking Research Results with Rural Development Projects: Experiences from
Southern Africa
Linking Research Results
with Rural Development
Projects: Experiences from
Southern Africa
523
36
Murwira, H.K.
TSBF, c/o Faculty of Agriculture, University of Zimbabwe,
P.O. Box MP228 Mt.Pleasant, Harare, Zimbabwe
Abstract
This paper presents an approach that has been used to
translate research results into farm practice within the
context of rural development projects in Malawi, Zambia
and Zimbabwe. It is an impact oriented approach designed
to ensure that research that is conducted with a
development focus should take account of beneficiary
interests and be able to address problems in a more holistic
fashion. Examples are drawn from the work on the croplivestock systems that are predominant in much of southern
Africa. The main conclusions drawn are that soil fertility
management is context specific and requiring adaptive
responses that consider local knowledge of the farmer as a
starting point in addressing problems. Research for
development is not just about technologies, it is also about
the people and enhancing their decision-making processes.
To achieve greater impact of integrated soil fertility
management research requires interdisciplinary teamwork,
inter-institutional partnerships, stakeholder involvement,
participatory approaches and systems thinking.
524
Murwira, H.K.
Introduction
The Context of Rural Development Projects
These are most often investment projects co-financed by governments
and development agencies in this case, the International Fund for
Agricultural Development (IFAD). These investment projects are varied
in their nature but are often aimed at infrastructure development, food
security, irrigation and market development. The soil fertility constraint
is often ill defined within these projects even for those focussing on food
security, yet it is a pervasive issue contributing directly to poor land
quality and low productivity (Figure 36.1). The result is that most of
these projects do not give soil fertility issues the prominence required
nor is there sufficient involvement of relevant expertise. The challenge
of linking soil fertility research results to these development projects is
therefore great and entails recognizing that needs between projects vary,
and that there is a complexity of problems addressed. Relevance of results
is dependent on how the research addresses the hierarchy of needs and
the multiplicity of objectives of target clients. This is the essence of
research for development. This paper is an attempt to show an approach
for linking soil fertility research results with development projects using
examples from southern Africa.
Research for development is research carried out in response to the
needs of the beneficiary communities. It is impact oriented and by design
involves participatory evaluation of options. The overall framework
includes the whole research continuum from process research to
adaptive research and dissemination though with a bias towards the
latter two (Figure 36.2). At the process level, the research is designed to
generate an understanding of the regulation of nutrient supply, and of
local knowledge about farmers’ priorities, access and management of
resources and how they are socially differentiated. The key challenge is
to use knowledge of social and biophysical processes to facilitate the
process of change to achieve more impact on livelihoods and on the way
that resources are managed. There are different methodologies that can
be used to facilitate change in farm practice. They all entail bridging the
gap between researchers and farmers through approaches that
encourage participatory diagnosis and evaluation of problems and
solutions. This whole process is complemented by the use of decision
support tools that aid both the farmers and researchers in making
decisions.
Poor land quality
Inadequate water
Poor water quality
Lack of
Poor extension
knowledge
approach
Lack of capital
Labour shortage
Lack of catchment based
Siltation
approach to water
Overgrazing
management
Market constraints
Land degradation
Poor land management
Inadequate inputs
Inappropriate
technology
Inadequate farmer
involvement
Poor service
provision
Overuse by livestock
Poor markets
Cultural Economic
Poor targetting of
technology
Poor tillage
systems
Insufficient supply from
Poverty
farms
Lack of alternatives
Poor policy
framework
Encroachment into grazing
areas
Poor internalization of values
High cost of participation
Distant
markets
Land tenure
No collateral
Little credit
Poor policy environment
Insufficient inputs
High input
costs
525
Limited facilitation
Deforestation
Low
adoption
Socioeconomic
constraints
Linking Research Results with Rural Development Projects: Experiences from
Southern Africa
Inadequate conservation methods Inadequate irrigation development
Figure 36.1: Constraint diagnosis of rural development projects
Low Productivity
Murwira, H.K.
526
Figure 36.2: A framework for linking integrated soil fertility management research with
farm practice
INTEGRATED SOIL FERTILITY MANAGEMENT RESEARCH FRAMEWORK
Process Research
Adaptive Research
Farmer knowledge
FACILITATING CHANGE IN PRACTICE
• Nutrient resource
• Nutrient access
• Farmer priorities
• Wealth, labour, etc/
Joint evaluation
Strengthen
farmer research
and extension
Linkages
Researcher knowledge
• Organic resource quality
• Fertilizer equivalencies
• Organic + inorganics
• Residual effects
Methods
Tools
Dissemination
PRA, trials
Databases
Participatory evaluation
Decision support tools
• Practicability
• Land, capital, labour
Demonstration, Field days, plots
Successful practices can be scaled out with partners within rural
development projects but recognizing that the project sites and conditions
are not amorphous (Figure 36.3). Soil fertility management is context
specific and requiring adaptive responses that consider local knowledge
of the farmer as a starting point in addressing problems.
Figure 36.3: Linking research results into the development process
SCALING UP /OUT
Farmers’
Objectives
Agricultural development
project
(diffusion of successful
interventions to a larger
geographical area)
Farmer evaluation
Benefits and costs
of organic matter
management
Adapting/testing
Resource
Management
Manipulation,
production potential
Fertilizers
Organic inputs
Soils
Nutrient
availability
Soil organic matter
Soil biota
Resource
availability,
access, soil
based
constraints
Process
Linking Research Results with Rural Development Projects: Experiences from
Southern Africa
527
The case of crop-livestock systems
Crop-livestock systems pre-dominate in much of southern Africa and
several investment projects are being implemented in the region (eg
Southeastern Dry Areas Project and Smallholder Dry Areas Resources
Management Project in Zimbabwe Southern Province Household Food
security Project in Zambia) to address food security concerns. It is clear
from an analysis of the literature that livestock provide an immense
contribution to livelihoods and that crop production is intricately linked
to the herd size that a household might have (Table 36.1). This is related
to both the capacities to produce manure and the provision of draft
power from livestock. The sources of manure and the management
strategies that farmers use are very diverse (Table 36.2) making
prescription of best manure utilization practices difficult. The manure
produced is often of poor quality hence options are needed to improve
on efficacy of this key resource (Murwira et al. 1995). This is not only
true from the research perspective, but also from numerous discussions
with farmers on their perceptions on how effective communal area
manures are (Nzuma et al. 1998).
Table 36.1: Grain yield per household and production in relation to the size of the cattle
herd in five communal areas of Zimbabwe, 1986
Size of cattle herd
per household
Maize grain yield
(kg ha-1)
Maize grain production
per household/year (kg)
0
1-4
5-8
9-12
> 12
669
903
1148
1249
1831
629
876
1366
1599
2362
Source: Adapted from GFA, 1987 (report covering Chivi, Makoni, Nswazi, Chirumanzi
and Merengwa communal areas)
One simple approach taken in the study areas was to look at ways
in which farmers could manipulate biological processes to enhance
quality of the manures. Anaerobic composting of manure in pits, an
innovation on the conventional practice of curing manures in heaps,
was proved to be a more efficient process that resulted in higher N
contents in the manure. The pitted manure produced higher maize yields
in the first year of application than heaped manure at the equivalent N
application rate of 100kg ha-1.
Manure Management and Use
Farmer 2
Method of
application
Manure
application rotations
Type of Fertilizer
Rate of Application
4 t ha-1
Broadcasted
Ammonium nitrate (AN) applied
17th t ha-1
Banded
2 year
rotation
maize-g/nutmaize
Uses Coca-cola bottle
top for pit stored
manure and cup number
2.5 for heap stored manure
3 year
rotation
maizemaize g/nut
No mineral fertilizer
for pit stored manure
Used to apply AN when
using heap stored manure
Could not give the rate
of application of AN
when heap stored
manure was use
3 year
rotation
maizemaize-g/nut
No amonium
nitrate applied
Seems application of
AN is targeted or
done only when crop
deserves it
Farmer 4
2 year
rotation
maizeg/nut-maize
No amonium
nitrate applied
Seems application of
AN is targeted or
done only when crop
deserves it
21 t ha-1
first season
16th t ha-1
second season
3 t ha-1
banded and
additional 4 t/ha
broadcasted
Banded at
2 cm depth
in ridge
Broadcasted
Banding
and
broadcasting
Banding
Note: Application rates were converted from scotch carts to tonnes per hectare and each scotch cart can carry approximately 400 kg of manure.
Murwira, H.K.
Farmer 3
Mineral Fertilizer Use
Rare of
application
Farmer 1
528
Table 36.2: Diversity in manure management and fertilizer use strategies for different farmers in Shurugwi District, Zimbabwe
Linking Research Results with Rural Development Projects: Experiences from
Southern Africa
529
Residual yields were however lower in the second and third years in
the pitted manure but overall yields after 3 years (including the 1st
year) were higher (Figure 36.4). This demonstrates that farmers can
benefit from putting science into practice and from the choices provided
on how they can maximize on immediate returns or alternatively build
on soil fertility and lose on the short term benefits (Table 36.3) (Refer to
Mutiro and Murwira in this volume). These results need to be interpreted
in terms of the social discount rates that poor people use and the impact
it will have on the soil fertility investment strategies (Figure 36.5).
Discount rates are quite often higher for poorer households, which might
negate on investments with a lower immediate return.
Table 36.3: Overall benefits over 3 years of using pit and heap stored manure
Factor
Control
Total harvest (tonnes)
Pit
Heap
1.83
9.82
8.79
Total Gross Benefit (Z$)
552.24
2835.35
2885.87
Total Variable Cost (Z$)
1748.22
1814.25
1818.32
Total Financial Benefit (Z$)
-1195.98
1021.10
1067.55
Net Present Values (NPV)
-801.46
767.04
497.64
Noe: 1 US$ = Z$ 55
Figure 36.4: Effect of different manure storage methods on maize yield
12
Control
10
Heap
Pit
8
6
4
2
0
1997/98
1998/99
1999/00
Overall 3 years
Murwira, H.K.
530
Figure 36.5: Discount rates and their implications on soil fertility management investment
strategies
Future benefits high
• More short term
investment
Limited value of future benefits
• More short term investment
• Technologies yielding most in the
short term preverred
• Social discount rates higher for poor
people hence use of exhaustive resource
utilization strategies
Future
benefits
Threshold of 10%
for heaped manure
A key lesson from these results is that farmers need to be engaged
in a dialogue on how they can arrive at solutions that suit their
requirements and circumstances. Developing a framework for such a
learning process can be very fruitful but demanding. Attempts have
been made to come up with a framework for manure decision making in
Zimbabwe (Figure 36.6). The framework looks complex but has been
widely tested on its usefulness and it has been demonstrated that it
can stimulate discussions on various aspects of manure management
and the decisions that farmers take before and after application of
manure to soil. It is important to emphasize that the decision tree is
more of a conceptual framework for social learning rather than a clear
guide for decisions.
The arguments above point to the fact that translating research
results into farm practice is not just about technologies, its about people
and reinforcing their decision making and their capacity to analyze tradeoffs and options. This has to be firmly grounded in their livelihood
objectives and aspirations. Livelihood income is diversified and
dependent on the opportunities presented to farming households by
proximity to markets, crop/livestock productivity and other off-farm
activities (Figure 36.7). The contributions (potential and actual) of each
of the enterprises needs to be known in order to set priorities for
technology testing. It is no point over-emphasizing on manure use and
cropping in an environment where farmers can get little recompense
from these activities. However opportunities can also be identified where
livelihood strategies can be reinforced.
Linking Research Results with Rural Development Projects: Experiences from
Southern Africa
531
Figure 36.6: Farmer manure decision guide developed using a spidogram analysis
Leaf litter
Household
compost
Lime
SOIL FERTILITY
IMPROVEMENT
Anthill
Inorganic fertilizers
Rotations
Cattle manure
Add cactus
gavakava
Quantity
Add anthhill
Add nothing
Add residual maize groundnut, leaf litter, grass
STORAGE
Quality indications
(see notes)
1. Colour
2. Weight
3. Moulds
4. Components
5. Temperature
Heap
Pit
Deep stall
Quality and effectiveness
High quality
With anthill
Frequency/
Residual effects
Method of
application
(see notes)
Supplementation
(see notes)
Pit stored
Medium quality
Low quality
With residues
Nothing added
Supplementation
(1) Assess crop performance
after germination
(2) If performance is poor use
compound X(a combination of
D & AN)
(3) Check soil pH
* Generally top dressing is
required - cash is usually the
limiting factor
Heaped
,,
3 years
1 year
2 years
1 year
1 year
Broadcasting
Banding
Broadcasting
Broadcasting
Broadcasting
No top dress
but (1)
Top dress (1)
and (2)
Top dress (1)
and (2)
Top dess (1)
Use in gardens
Top dress (1)
and (2)
Use for field crops
Notes on rates of manure application
• It depends on the quantities of manure available
• Rotation of manure application depend on the plot/farm size
• Manure is usually targetted for high potential fields (for food security)
Considerations
• Soil type
• Presence of witch weed-striga
Low rates or no manure applied to low potencial fields
Figure 36.7: Livelihood incomes of smallholder farmers in Chivi, Zimbabwe
Percentage
100
90
80
70
60
50
40
30
20
10
0
Remittances
Wages/home industry
Woodland sales
Woodland subsistence
Livestock subs
Livestock sales
Garden sles
Garden subsistence
Dryland ag sales
Very poor Medium “Wealthy”
Medium poor “Wealthy”
Dryland ag subs
Wealth class
Wealth class
From Research Results to Creating Partnerships for
Development
Successful interventions in any project area require the participation of
key stakeholders (Figure 36.8). This derives from possible institutional
synergies that can obtain, and from the need to analyze diffusion
Murwira, H.K.
532
pathways to increase impact beyond the plot level and pilot villages. In
implementation of work in rural development, research and innovation,
institutions are more and more confronted with issues that are too
complex to be resolved by a single organization on its own. Nowadays,
rural development has to meet many objectives such as improving the
livelihoods of poor people, promoting sustainable use of natural resources
and biodiversity, linking small-scale farmers to markets and enhancing
food security and safety simultaneously. A single institution can no
longer make isolated contributions to rural development in their
specialized field, but need to ensure that their products and services,
jointly with those of other organizations, contribute to these broader
objectives. For this to happen, organizations need to combine different
kinds of expertise and to work in partnership with other rural
development and research organizations. Together, these partners need
to work closely with the beneficiaries of rural development activities.
They also need to involve and collaborate with other groups that have a
role to play in tackling the issues and achieving the broader development
objectives such as the private sector (agro-dealers), policy makers and
other interest groups (Figure 36.8).
Figure 36.8: Creating partnerships for effective diffusion of integrated soil fertility
management research (Murwira and Wopereis, 2003)
Marketing
ISFM
Farmers’
organizations
Outputs
Inputs
Input
dealers
ISFM
Public
awareness,
information
Input dealers,
credit
providers
Pilot village
Learning plot
Diffusion plot
Policy and
decision
makers
NARES,
NGOs, IFDC,
TSBF
Linking Research Results with Rural Development Projects: Experiences from
Southern Africa
533
The major challenge observed to date in linking research results
with rural development projects in southern Africa has been in bringing
together the critical mass of expertise required to effect a coherent
participatory research and development program. There has been huge
staff turn-overs in most of the key national agricultural research systems
(e.g. Zambia and Zimbabwe) or the personnel are simply not there (e.g.
Malawi).
Conclusion
The main lessons from the work in southern Africa are that translating
research results into farm practice is not just about technologies, its
about people and reinforcing their decision making and their capacity
to analyze trade-offs and options. All this calls for a new approach to
doing business in rural development and research. This new way should
put emphasis on interdisciplinary teamwork, inter -institutional
partnerships, stakeholder involvement, participatory approaches and
systems thinking. It sees rural development and innovation and the
knowledge needed for it as the result of collective learning to which all
these actors contribute, not as the result of the transfer of knowledge
generated by a single organization. Most of the work reported is still ongoing but it is hoped that the approaches expounded in this paper could
lead to more tangible benefits at the farm level.
Acknowledgements
The support of the International Fund for Agricultural Development is
gratefully acknowledged.
References
Murwira, H.K., Swift, M.J. and Frost, P.G.H. (1995) Manure as a key resource
in sustainable agriculture: a case study of communal area farming systems
in Zimbabwe.p131-148. A keynote address. In Livestock and Sustainable
Nutrient Cycling in Farming Systems of Sub-Saharan Africa, Vol.II ILCA,
Addis Ababa, Ethiopia, 22-26 Nov, 1993. Ed. J.M. Powell, S. FernandezRivera, T.O. Williams and C. Renard.
Murwira, H.K. and Wopereis, M. (2003) Collaboration with rural development
projects. Proceedings of the IFDC/TSBF-CIAT project results and planning
workshop held in Nairobi, Kenya, 28-30 April, 2003. Unpublished.
534
Murwira, H.K.
Nzuma, J.K., Murwira, H.K. and Mpepereki, S. (1998) Cattle manure
management options for reducing nutrient losses: farmer perceptions and
solutions in Mangwende, Zimbabwe. P183-190. In: S.R.Waddington, H.K.
Murwira, J. Kumwenda, D.Hikwa and F. Tagwira (eds). Soil fertility Research
for Maize-based farming systems in Malawi and Zimbabwe. Proceedings of
the Soil Fert Net Results and Planning Workshop held from 7-11 July at
Africa University, Mutare, Zimbabwe. 312pp.
Economic Analysis of Non-Conventional Fertilizers in Vihiga District, Western Kenya
Economic Analysis of NonConventional Fertilizers in
Vihiga District, Western
Kenya
535
37
Kipsat, M.J.1, Maritim, H.K.1, and
Okalebo, J.R.2
1
Department of Agricultural Resource Economics
Department of Soil Science
Moi University, P.O. Box 1125, Eldoret, Kenya
2
Abstract
Most farmers in Vihiga district are faced with the problem
of low income. Population pressure is high, land sizes are
small and the cost of hired labour is high. With the onset of
market liberalisation, prices of conventional fertilizers have
been rising faster than farm produce prices. There are many
available soil fertility technology options but their adoption
is subject to farmers’ perception of benefits and limitations
to their use. The overall objective of this study was to carry
out an economic analysis of some non-conventional fertilizer
materials used to improve food production in Vihiga district.
A random sample of 150 farmers was selected from three of
the six divisions of Vihiga district. Primary and secondary
data were used. Gross-margins and cost to benefit ratios
were used as the main tools in data analysis. Evaluation of
the use of the organic matter technologies on maize/ bean
production indicated that there were significant profitability
536
Kipsat, M.J. et al
differences between them at 95 % confidence level. Among
the organic matter technologies considered, use of
agroforestry shrubs in food (maize and bean) production
gave the highest profitability. Increasing availability of seeds
of agroforestry shrubs to farmers can therefore improve food
production in the study area.
Introduction
Vihiga is one of the eight districts of western province. It has an altitude
range of between 1750 and 2000 meters above sea level. It is
characterized by undulating hills and valleys with vast network of
streams and brooks that are tributaries of rivers Esalwa and Yala. Its
bimodal, reliable, adequate and well distributed rainfall of between 1800
and 2200 mm per annum peaks in April and June for long rains and
September and November for short rains. Ninenty five percent (95%) of
the District is in the Upper Midland Zone (UM1) while the rest (5%) is in
the Lower Midland Zone (LM1).
The district’s warm and humid climate supports growing of many
crops. However, soils are of low fertility, limited water-holding capacity
and are prone to erosion due to their sandy texture, high land use intensity
and heavy rainstorms. Widespread N and P deficiencies in soils due to
continuous cropping (Niang’ et al., 1996) and inability of smallholder
farmers to invest on fertilizers to replace the lost nutrients (Okalebo et
al., 1996) has led to low agricultural productivity in the district. The high
population densities of 1100 people or 294 households (of on average 8
persons) / km2 has resulted in land subdivision into small units, further
lowering agricultural productivity of the area. Vihiga was expected to
have 80,000 households in the year 2001. The average land holding is
0.6 ha and this is considered to be below the FAO recommendation for
subsistence food purposes of 1.4 ha / household (FAO, 1999).
The problem of persistently low agricultural productivity in Vihiga
district has resulted in a vicious cycle of soil degradation and food
insecurity. Crop yields have continued to decline despite the existence
of a wealth of already developed technologies that farmers could use to
improve soil fertility. In 1982 when the Ministry of Agriculture conducted
fertilizer trials in various districts of western Kenya, maize yields in
Vihiga increased from 3800 kg ha -1 (without addition of fertilizer) to
6100 kg ha-1 with addition of (60-60-0) NPK fertilizer. The highest maize
yields of 14220 kg ha-1 with addition of 178 kg N and 104 kg P ha-1 was
also recorded. This high yield increase realized in the 1980s contrasts
greatly with 1990’s report of maize yields of on average 122 kg ha -1,
(Aritho, 1994). This shows a drastic decline in land productivity in Vihiga
over the years.
Economic Analysis of Non-Conventional Fertilizers in Vihiga District, Western Kenya
537
Adoption rates of chemical fertilizers in western Kenya are low despite
farmers knowing the benefits to their use (Hoekstra and Gorbett, 1995).
Use of commercial fertilizers on subsistence food crops such as maize
and beans in the area has been restricted to only a few farms with high
endowment of resources such as cattle and land (Shepherd and Soule,
1998). Fertilizer recommendations as given in the MoA Bulletin are
infeasible in most districts in western Kenya (Okello, 1997). This is
mainly due to increased prices after SAPs, unavailability of cash, or
lack of access to appropriate fertilizer materials that can be reached
easily and at the right time (Nandwa et al., 1997). Research has shown
that non-conventional fertilizers are major resources available to farmers
to manage soil fertility. They are environmentally friendly and provide
longer term beneficial effects to the soil than chemical fertilizers. Nonconventional fertilizers refer to soil fertility management technologies
other than the exclusive use of chemical fertilizers. They include organic
matter of plant and animal origin used alone or fortified with inorganic
materials. Non-conventional fertilizers in this study also include
Phosphate Rock (PR), a source of P that naturally occurs in some parts
of Africa.
There is a high potential for using non-conventional fertilizers in
Vihiga because farmers are keen on improving the fertility of their soils.
They have a long history of using traditional soil fertility improvement
strategies such as fallowing and farmyard manure in their fields. Nearly
ten years of collaborative research by government institutions and Non
Governmental Organizations (NGOs) in western Kenya has addressed
soil fertility improvement using non-conventional fertilizers. The crop
response trials with the most commonly used non-conventional fertilizers
have produced technically good yield responses. It also comes out so
clearly from research publications that technologies have been studied
for potential yields but comparative economic analysis has not been
part of it. Economically speaking, however, output (maize and bean
yields) alone does not reflect much about efficiency of production.
Research scientists in the past laid more emphasis on the ability of
technologies to achieve high crop yield responses than on the
performance of the technologies based on economic considerations. This
explains why some technologies that appear superior in improving crop
yields under research conditions are not always the most adopted in
farmers’ fields. This has resulted in problems of determining the
superiority of the existing non-conventional fertilizer technologies based
on efficiency of use of the productive resources. Health conscious
consumer groups are also lobbying for the use of organic materials but
the viability of the option has not been assessed economically.
Non-conventional fertilizers when used in the right amounts have
as high yield responses as those of chemical fertilizers used at required
levels. The overall objective in this study was to economically evaluate
538
Kipsat, M.J. et al
some of the commonly used non-conventional fertilizer materials,
namely, agroforestry shrubs, farmyard and compost manure and
Tithonia diversifolia on improved and sustained maize and bean
production in the populous Vihiga district. The hypothesis that was
held in this study was that there is a significant profitability difference
in maize-bean intercrop when the selected organic matter technologies
are used. This implies that farmers need to consider profitability of use
of the technologies in adoption. The technologies have comparable yield
responses but the cost of adoption vary from technology to technology
resulting in profitability differences.
Methodology
The study was based on three out of the six divisions of Vihiga district,
namely Sabatia, Emuhaya and Luanda. According to Vihiga District
Development plan 1997-2001, Emuhaya has a total area of 75 km2, of
which 60 km2 is arable land. It has a population estimate of 89,952
persons and 11,244 households, farm sizes of on average 0.4 ha and a
population density of 1,199 persons km-2. Sabatia has an area of 115
km2 with 101 km2 of arable land expected to be supporting 150,000
people. It has 21,428 households with average farm sizes of 0.5 ha.
Luanda has a population estimate of 114,936 people and 14370
households in an area of 104 km2, of which 68.5 km2 is arable. Like the
rest of the district, Luanda, Emuhaya and Sabatia divisions have
generally good climate for production of most crops but soils are depleted
of N and P.
Primary and secondary data was used in this study. Primary data
was collected through administration of structured questionnaires in
some purposively selected households. Secondary data was mainly
from research institution and government publication such as
agricultural annual reports. Information collected included those on
operation costs of use of the selected technologies in maize and bean
intercrop, input and output prices, application rates of the specified
non-conventional fertilizers by farmers and the associated average
maize and bean intercrop yields. The average output and input prices
were obtained from time series data in the study area divisional
Agriculture Offices. Operational costs considered included costs such
as those of crop management (planting, pest control, harvesting,
shelling and collecting, preparing, carrying and application of organic
materials) and marketing of produce. The opportunity cost of second
season maize and bean crop was considered as a cost in case of
improved fallows because no crop was planted in the second rainy
Economic Analysis of Non-Conventional Fertilizers in Vihiga District, Western Kenya
539
season as agroforestry shrubs were left growing in the crop field during
the season. The value of labour used on various operations in the
production was based on survey of farmers estimation, while the cost
of land included the land rent in the area. Costs were arrived at after
grants and subsidies on agricultural products have been excluded.
The benefits included increased maize and bean yields and wood fuel
(in case of agroforestry shrubs). Information on use levels of fortified
or unfortified organic materials and associated crop yields were
obtained from some selected farmers, who were known to enumerators
employed in this study as using the organic matter technologies
considered in this study on maize and bean intercrop. Interview of
farmers who did not use fertilizer in subsistence food (maize and bean)
production provided data that was used as control.
The population was divided into three sampling units represented
by three selected divisions (Sabatia, Emuhaya, and Luanda). The
selection was based on agro-ecological zones and prevalence of organic
matter technologies under consideration. The three divisions provided
the survey sites for the study. From the selected divisions, households
that were known to the enumerators as using the selected organic matter
technologies were selected randomly from each location. The exact
number of farmers selected in each location depended on prevalence of
organic matter technologies that the study focussed on in the location.
At least two farms were selected from each sub-location and this study
collected data from 20 sub-locations in which a total of 150 households
were interviewed.
Data collection exercise was done between August and December
2000. Ten enumerators were appointed, trained for the enumeration
exercise and provided with questionnaires. Single visit formal surveys
that were conducted using structured questionnaires were orally
administered to farmers with the help of the enumerators who knew
and were conversant with the farmers’ local language and customs.
During the survey the enumerators made arrangements to meet the
sample farmers in farmers’ fields.
The economic analysis of the technologies involved the Net Present
Value (NPV) or the net worth and Benefit- Cost Ratios (BCR). NPV is
defined as the present worth of the benefits less the present cost of a
project. In this study each of the four organic matter technologies is
taken as a project. BCR is a discounted measure of project worth. It is
given as the present worth of the benefit stream divided by present
worth of the cost stream. The NPV and B/ C analyses were used in
economic evaluation of non-conventional fertilizer technologies in this
study to ensure that the residual effects of use of organic matter
technologies are captured. Mathematically NPV is given as:
540
t = n
Kipsat, M.J. et al
−t
∑
Bt − C t (1 + i )
t =1
(
)
Bt = benefits in year t,
Ct= cost in year t,
t =1,2,….n, time in years
n = year n/last year under consideration
i = interest/compounding rate, taken as the interest rate in commercial
banks.
Results and Discussions
Evaluation of soil fertility technologies in terms of maize
and bean yields
Farmers interviewed in this study fortified organic materials with half
the recommended levels of inorganic materials mainly DAP and CAN.
Very few farmers used Phosphate Rock (PR) due to its limited availability
in retail shops in the area. Table 37.1 shows the three-year (1998-2000)
average maize and bean yields among the farmers interviewed in Vihiga
district when the specified soil fertility technologies are used. The yields
considered in this study refer to averages obtained when maize and beans
were intercropped. Control is taken as the current crop yields obtained
in a maize-bean intercrop in the study area when no fertilizer is applied.
Table 37.1: Average yield responses to soil fertility management technologies
Soil Fertility Management
Technology
Fertilizer use level/
Seed Rate
Maize Yield
kg ha -1
Bean Yield
k g ha-1
Zero fertilizers added (control)
Tithonia diversifolia alone
Fortified Farmyard manure
Fortified Compost manure
Use of Tephrosia vogelii
Fortified Tithonia diversifolia
Use of Crotalaria grahamiana
Inorganic fertilizer (DAP and CAN)
5 t ha-1
2.5 t ha-1
2.5 t ha-1
2.5 t ha-1
2.5 t ha -1
2.5 t ha-1
DAP 2.5 Bags
CAN 2.5 bags
1 bag = 50 kg
970
1674
2025
2109
2174
2270
2490
2700
100
165
182
180
185.5
193.7
212.5
225
Table 37.1 shows that yields of both maize and beans were higher
with application of conventional fertilizers than with non-conventional
fertilizers. Using fortified organic materials however, also improves crop
yields. Fortifying tithonia for example increased maize yields by 36%
relative to tithonia applied on its own and 134% compared to maize
produced under conditions of no fertilizer. Table 37.2 shows the % change
in crop yields arising from the use of the specified soil fertility technologies
in the study area compared to the control of no fertilizer use. Applying
Economic Analysis of Non-Conventional Fertilizers in Vihiga District, Western Kenya
541
organic materials results in substantial crop increases of between 72
and 157 %. Fortifying organic materials is thus recommended because
of the low levels of N and P in organic materials. Table 37.2 indicates
the % crop yields increases arising from use of fortified organic materials.
Table 37.2: Yield Increases with Use of Fortified Organic Materials in Vihiga District
Soil Fertility Management Technology
Zero fertilizers added (control)
Use of Tithonia diversifolia alone
Use of fortified farmyard manure
Use of fortified Compost manure
Use of fortified Tephrosia vogelii
Use of fortified Tithonia diversifolia
Use of fortified Crotalaria grahamiana
Percent (%) improvement in crop yield
Maize
Beans
0
72
109
117
124
134
157
0
65
82
80
85.5
93.7
112
Table 37.2 shows that Crotalaria is the best of the organic matter
technologies in improving yields of both maize and beans in Vihiga by
on average 157 and 112 % respectively. This means that based on crop
yield responses, organic materials are the best alternatives to inorganic
materials in crop production in Vihiga district. A single factor ANOVA
carried out to make comparative yield analyses indicate significant maize
and bean yield differences arising from the use of the organic matter
technologies at 5% level.
Economic evaluation of soil fertility technologies in vihiga
Cost of labour form a major part of the total cost in the use of organic
materials in western Kenya, particularly in Vihiga district (Kipsat, 2001).
Table 37.3 gives a comparative analysis of the total variable and labour
costs of using the reviewed technologies in maize and bean intercrop
production in the study area. The table indicates that labour form more
than half of the total variable cost of production when the organic matter
technologies are used. This is because use of organic materials is labour
intensive.
The figures in parentheses in Table 37.3 indicate the proportion
that labour costs make of the total variable costs of using the specified
technologies in maize/ bean production in Vihiga. The cost of labour
forms over 60% of the total variable costs in all cases. Labour contributes
less to total variable costs (60.74%) when inorganic fertilizers are used
than it does in use of organic fertilizers, where it contributes between
542
Kipsat, M.J. et al
65.71 and 74.76% of the total costs in maize and bean production. This
is explained by the differences in labour requirements for using organic
and inorganic materials in crop production.
Table 37.3: Costs associated with soil fertility technologies in Vihiga District
Soil Management Practice
Total costs (Kshs ha-1)
of Maize and Bean
Inter-crop
Cost of labour (Kshs ha -1)
of Maize and Bean
Inter-crop
Inorganic Fertilizer
Fortified Crotalaria
Fortified Tephrosia vogelii
Fortified T. diversifolia
Fortified Compost Manure
Fortified Farmyard Manure
Unfortified T. diversifolia
Zero fertilizer (control)
47031.10
43126.00
42304.55
46231.70
44220.80
42996.35
44995.80
35560.70
28565.00
28337.80
27972.50
30861.10
29897.50
28897.50
33639.10
26630.50
(60.74 %)
(65.71%)
(66.12%)
(66.75%)
(67.61%)
(67.21%)
(74.76%)
(74.88%)
In economics the goal or standard for return to labour and
management should be an amount at least as great as the opportunity
cost of owner’s labour and management in a non-farm occupation. The
minimum standard or goal for return to capital is that the rate of capital
return should approximate the interest rate of borrowed capital. The
above goals though desirable are not achievable in farmers’ production
conditions as seen in Vihiga district. The aim should therefore be that
farmers select technologies that are more economically efficient than
others in the use of resources. Table 37.4 shows the economic evaluation
of organic matter technologies.
Table 37.4: Results of economic evaluation of organic matter technologies
Soil Fertility Management
Technology
Net Present Value
NPV ha -1 in (Ksh ha -1)
Benefit to
Cost Ratio
Rank based
on NPV
Fortified Crotalaria
Fortified Tephrosia
Fortified Tithonia
Fortified Compost Manure
Fortified Farmyard Manure
Tithonia alone
No fertilizer (control)
33 568.20
13745.90
11047.20
6020
4592.10
-2130
-11719
1.27:1
1.13:1
1.08:1
1.05:1
1.036:1
0.87:1
0.61:1
1
2
3
4
5
6
7
Table 37.4 indicates that the use of agroforestry shrubs (Crotalaria
and Tephrosia) on maize and bean production gives the best profitability
in relation to the other non-conventional fertilizers considered. Although
Economic Analysis of Non-Conventional Fertilizers in Vihiga District, Western Kenya
543
tithonia is associated with high crop yield, the net worth is lowered by
the high labour demands that translate into high cost of using it. Fortified
farmyard and compost manures have relatively low crop yields and high
labour costs lowering the returns to their use. The NPV of using unfortified
tithonia and production under conditions of no fertilizers are negative.
This means that those farmers who produce under the two systems are
incurring losses in maize and bean production and should be advised to
find alternative use of the invested labour, land and capital. The two
production systems result in very low BCR while the rest of the soil fertility
management technologies under consideration have favourable (greater
than one) BCR values. This means that apart from using tithonia, a farmer
can make positive returns by using the rest of the Non-Conventional
fertilizer technologies in maize bean production although the relative
returns vary from technology to technology.
To test the null hypothesis that there are no significant profitability
differences arising from the use of the four organic matter technologies
on maize and bean production, a single factor ANOVA was carried out.
The results indicate that there are significant profitability differences
between the Non-Conventional fertilizers at 95 % confidence level and
therefore agricultural extension agents should consider profitability
differences in choice of technologies to promote in Vihiga district.
Conclusion and Recommendations
Resource limitations are a major hindrance to adoption of soil fertility
improvement technologies in Vihiga district. The district is characterized
by high cost of land, labour and capital. To improve food security in the
district, policy makers should focus more on measures to improve the
resource base of the resource poor farmers than on aspects of generating
more technologies in the area.
Promotion of agroforestry can improve food production in the study
area. From the survey, it was realized that farmers have a problem of
accessing agroforestry shrubs’ seeds to use in their fields. To promote
agroforestry as a technology, therefore, involves providing agroforestry
shrubs’ seed to the farmers and teaching them on how to propagate
and manage the plants for soil fertility improvement. The economic
analysis of other soil fertility management technologies not covered in
this study should be made and information availed to the farmers to
enable them make informed decisions. The benefits of using organic
materials can be improved by fortifying it with PR that is a cheaper
source of P than inorganic P fertilizer. The availability of PR in retail
shops should thus be improved. From the research, it was realized that
farmers were aware of the benefits in the use of PR but complain of the
material not being available.
544
Kipsat, M.J. et al
Acknowledgements
The first author would like to appreciate the Ministry of Environment
and Natural Resources (LVEMP), Kenya and Tropical Soil Biology and
Fertility programme (TSBF). LVEMP for providing funds to carry out the
research and TSBF for facilitating the presentation of this paper in the
TSBF 8th African Network for Soil Biology and Fertility (AfNet) workshop
in Arusha, Tanzania.
References
Aritho, R. (1994) The Effect of Household Income and Seasonal Price Changes on
Household Food Expenditure Patterns. Fertility and Nutrition Studies Project,
Egerton University.
FAO, (1999) FAO Fertilizer yearbook (1998) Vol. 44. Food and Agriculture
Organization of the United Nations, Rome.
Hoekstra and Gorbett (1995) Sustainable Agricultural Growth for the Highlands
of East and Central Africa: Prospects to 2020. International Food Policy
Research Institute, Washington D.C.
Kipsat, M.J. (2001) Economics of Non-Conventional Fertilizers in Vihiga District,
Western Kenya. M.Phil. Thesis Department of Agricultural Resource
Economics, Moi University. Kenya.
Nandwa, S.M., Kanyanjua, S.M., and Thuraniva, E.G. (1997) Soil Phosphorus
Recapitalization of Western Kenya Soils with Minjingu Rock Phosphate;
Some Evidence of Limited Soil P Availability and Maize Response to Fertilizer
Application.
Niang’, A.I., Amadolo, B.A., Gathumbi, S.M., Otieno, J.H., Obonyo, C.O. and
Obonyo, E. (1996) KEFRI/ KARI/ ICRAF – AFRENA, Maseno Project Report
no.10.
Okalebo, J.R., Woomer, P.L., Maritim, H.K., Kapkiyai J. and Mwakuko, P.E.
(1996) “Phosphorous Research Exploratory Project for Western Kenya (PREP)”.
Project Proposal Presented to “Rockefeller foundation, Forum on Agricultural
Resources Husbandry.”
Okello, D. (1997) Improving Food Production through Improving Soil Fertility in
Siaya District. A paper presented at PREP workshop Chepkoilel campus
12th–15th May 1997.
Shepherd, K.D. and Soule M.J. (1998) Assessment of the Economic and
Ecological Impacts of Agroforestry and other Soil Management Options on
Western Kenya Farms Using A Dynamic Simulation Model. Agricultural
Ecosystem Environment (in Press).
Early Farmer Evaluation of Integrated Nutrient Management Technologies in Eastern Uganda
Early Farmer Evaluation of
Integrated Nutrient
Management Technologies in
Eastern Uganda
545
38
Miiro, R.1, Kabuye, F.2, Jama, B.A.3,
Musenero, E.4, Zake, J.Y.K.5, Nkwiine,
C.5, Kakinda, M.J.2, Onyango, O.6,
and Delve, R.J.7
1
Department of Agricultural Extension/Education, Faculty of
Agriculture, Makerere University, P.O. Box 7062, Kampala,
Uganda, rfmiiro@yahoo.com; rfmiiro@agric.mak.ac.ug
2
Africa 2000 Network, P.O. Box 7184, Kampala, Uganda,
anetwork@imul.com
3
International Centre for Research in Agroforestry (ICRAF),
Nairobi, Kenya, b.jama@cgiar.org
4
Department of Production, Tororo District, Uganda
5
Makerere University, Department of Soil Science, Faculty of
Agriculture, P.O. Box 7062, Kampala, Uganda,
acss@starcom.co.ug or plectumu@imul.com
6
Tororo District Local Government, Tororo, Uganda
7
Tropical Soil Biology and Fertility Institute (TSBF)/International
Centre for Research in Tropical Agriculture (CIAT), P.O. Box 6247,
Kampala, Uganda, r.delve@cgiar.org
Abstract
There has been widespread recognition of the need to
rejuvenate the fertility of soils for sustainable agricultural
productivity, food security, household income, and
546
Miiro, R. et al
poverty alleviation in sub-Saharan Africa. Since 1999
different Integrated Nutrient Management (INM),
technologies have been under on-farm adaptive/
dissemination trials in Tororo district, eastern Uganda.
The area is well known for its highly unproductive sandy
ferralsols. Options promoted included the use of
leguminous trees/shrubs and cover crops such as,
Mucuna, Canavalia, Tephrosia, and Crotalaria species.
Also promoted were Tithonia biomass transfer and
Rhizobia inoculation. For the P deficient soils, various P
fertilizers were evaluated including Busumbu Phosphate
Rock from nearby deposits, TSP and Minjingu Phosphate
rock from Tanzania. Initial assessments by farmers
indicate wide-scale testing in the pilot areas and farmer
adaptation and innovation of the options promoted. Eighty
eight percent (88%) of the farmers who tested the options
indicated willingness to expand 56% of these to more than
one acre of land. To scale up these efforts to the entire
district and to address constraints like seed availability,
awareness creation and training, a consortium of R&D
partners through a project called Integrated Soil
Productivity Initiative through Research and Education
(INSPIRE) steered by the District administration has been
initiated. Farmer to farmer extension, participatory crop
management training, fertiliser use, green manure crop/
shrub husbandry frame the way forward of the project.
Key words: Integrated Nutrient Management, Green manure legume
cover crops, Participatory evaluation, Dissemination.
Introduction
Soil fertility depletion in smallholder farms is recognised as a major
biophysical root cause of the declining per-capita food production in
most of sub-Saharan Africa (Sanchez et al. , 1997). Soil fertility
rejuvenation for increased productivity, food security and income has
been recognised by the Government of Uganda as a strategy towards
poverty alleviation (Soil Fertility Initiative –SFI, 1999). The widespread
belief by many, including politicians that Ugandan soils are fertile has
also been corrected to a recognition that there has been a lot of soil
degradation of all forms leading to poorer soils (PMA, 2000).
Early Farmer Evaluation of Integrated Nutrient Management Technologies in Eastern Uganda
547
The government has set up a strategy to combat soil degradation
under the Plan for modernisation of agriculture through the Soil Fertility
Initiative (SFI).
The SFI in Uganda aim at correcting the negative nutrient balance
in smallholder farming. Households needs to move beyond the two
extremes of high external input agriculture and the low external input
agriculture to integrated forms of agriculture such as integrated nutrient
management. The advantage of the integrated approaches to agriculture
is the synergism between locally known practices and introduced or
research inputs and practices that are developed either through a
participatory technology development (PTD) process or participatory
learning and action research (PLAR) process. These processes are part
of development thinking for poverty alleviation that promotes
empowerment of the beneficiaries through partnerships and pluralism
(RoU, 1997; Ashley & Carney, 1999).
In Tororo District since 1997, various efforts by the National
Agricultural Research Systems (NARS) and an international NGO, (Africa
2000 Network), have been in place to fight poverty of farm households
by increasing food security through the promotion of integrated nutrient
management technologies. This has been through a consortium involving
civil society, NGOs, national agricultural research systems, international
agricultural research systems and government, called the Integrated
Soil Productivity Initiative through Research and Education (INSPIRE)
project. The efforts of the consortia have been directed to Tororo District,
because of its dense population of over 280 persons per square kilometre,
poorly endowed natural resources, acidic and sandy soils, and a reversion
to mainly annual crop system. About 82 % of the district land is farmed
making this area a high incidence poverty area (RoU, 1991; Zake et al.,
1998; World Bank, 1993).
The INSPIRE Project
The INSPIRE project is a broad based consortium of district based
development organisations, the NARS and district local government,
formed to address increasing poverty and food insecurity by promoting
appropriate soil management technologies for increased agricultural
productivity. The partners include: the district government’s
department of production, Africa 2000 Network (A2N), Sasakawa Global
2000 (SG2000), Tororo District Farmers Association (TODIFA), the Food
Security and Marketing organisation (FOSEM) and appropriate
Technology (AT) Uganda; a local branch of Enterprise Works Worldwide.
Others are Plan International, the International Centre for Research
in Agroforestry (ICRAF) of western Kenya and Uganda, Tropical Soil
Biology and Fertility (TSBF), The International Centre for Tropical
548
Miiro, R. et al
Agriculture (CIAT), Makerere University, Faculty of Agriculture and
the National Agricultural Research Organisation (NARO). The
consortium is overseen by a steering committee led by the District
Production Officer and the project activities are co-ordinated by Africa
2000 Network.
A five-year activity plan has been made to address the soil needs of
the district using an integrated approach. The project goal is the
improvement of livelihoods of farmers in eastern Uganda by empowering
them to overcome food insecurity and poverty. The project purpose is to
improve the soil fertility productivity in Tororo District in a sustainable
manner. Four outputs are expected:
1. Establish through baseline studies the current soil fertility levels
and farmer management practices, and develop, test and verify with
farmers the appropriate and sustainable soil management
technologies,
2. Disseminate verified soil management technologies and practices,
3. Improve access to soil fertility enhancing agricultural inputs in the
district such as fertilizers, cover crops, seeds, Rhizobia, compost
inputs, tree seedlings, and mulching materials and
4. Enhance capacity of the stakeholders to implement sustainable soil
fertility management technologies.
To date, the technologies tested on-farm include the use of
leguminous trees/shrubs and cover crops such as Mucuna, Canavalia,
Tephrosia and Crotalaria species. Also promoted is Tithonia biomass
transfer and the use of rhizobium inoculum for legume crops. Since the
soils in the district are highly P deficient, P fertilisers including Busumbu
Phosphate rock from the nearby Tororo hills have been evaluated in
comparison with other sources such as Triple Super Phosphate (TSP)
and Minjingu Phosphate rock from Tanzania. The demonstration/
dissemination strategy used several farmers from a farmer research
group, who hosted the trials on behalf of the group. Training, field days
and farmer evaluation of these trials are conducted throughout the
season. Exchange visits between farmer groups were conducted to allow
different farmer groups in the District and from outside the District to
visit and learn from each other.
This paper that provides an overview of the INSPIRE project, does
share results of an early qualitative evaluation of the soil fertility
enhancing adaptive trials. The study established the performance of
the technological options tested, the initial benefits and constraints,
adaptations, relevance of pre-trial training, on farm expansion, continuity
and dissemination/diffusion. The future direction of the consortium’s
planned activities is outlined.
549
Early Farmer Evaluation of Integrated Nutrient Management Technologies in Eastern Uganda
Methodology
The study was conducted in the district of Tororo in Osukuru and Kisoko
sub-counties. Tororo district is found in eastern Uganda bordering
Kenya. Most of the district is flat, lying at an altitude of 1,097 to 1,219
m above sea level and a temperature range of 15.7° to 30.6° C. The
annual rainfall is more than 1,200 mm per year. It has a population
density of about 280 persons per sq. km, with over 82% of the land
under agriculture. Soil type in the area is sandy loam often acidic and K
deficient. Soil erodibility and erosivity is moderate (Wortmann and Eledu,
1999).
There were two stages to the qualitative analysis:
1. Data were collected using a structured interview schedule, developed
by the NARS partners of the INSPIRE project. The sample size was
25 out of the 92 farmers who conducted demonstration trials between
1998 and 2000 (Table 38.1). Data were collected in June-July 2000
by the extension workers of A2N who had been trained on the data
collection processes.
2. A qualitative group evaluation of the legume cover crops to capture
their performance was conducted in February 2001. A pair-wise
ranking of the legume cover crop preferences was done with the
farmers. Farmers hosting the green manure trials were invited along
with those who were interested in trying out the technologies. Sixty
farmers were in attendance in the Kisoko meeting and more than
70 from Osukuru. Only data from Kisoko will be presented in this
paper.
Table 38.1: Numbers of farmers who hosted the trials
Trial description
No. of farmer
trials
1. Determining the effectiveness of Busumbu rock phosphate
as a source of P using maize as a test crop
2. Determining the effectiveness of Busumbu rock phosphate
as a source of P using groundnuts as a test crop
3. Determine the contribution of Nitrogen by inoculated
groundnuts
4. Compare the integration of Tithonia biomass with TSP
to the sole application of inorganic fertilizer at an
equivalent rate of NPK
5. To determine the effectiveness of using legume cover
crops (LCCs) as an organic source of fertilizers on maize
Number of
evaluated
farmers
20
14
20
7
16
10
10
5
10
6
550
Miiro, R. et al
Results and Discussion
Farmer evaluations of phosphate rock, biomass transfer
and rhizobia innoculum
Farmers evaluating the technologies were asked about what they had
feared before hand and what had actually been constraints or problems
during the trials. They also explained what they had learned from the
trials, and how the technologies compared with each other. Nearly half
of the farmers (48%) indicated that their main fear prior to testing the
different options was related to their experience with inorganic fertilizers
(Table 38.2). In particular, Busumbu phosphate rock was unfamiliar to
nearly a third of the farmers (32%). In contrast, few farmers were
unfamiliar and worried about the use of the organic fertilizers. Worries
about lack of funds to buy inorganic inputs or hybrid maize seed were
also common (28%). Lack of funds for inputs was not an issue for the
use of Tithonia, compost and Rhizobia, because these relied more on
labour than capital inputs, the former being cheap.
Table 38.2: Fears held by farmers before the trials were hosted (n=25)
Impro- Busu- Busu- Minjived mbu mbu ngu
maize blend PR
PR
TSP
Urea
P and TithoK
nia
Compost
Rhiz
obia
Had never
used it before
Lack of funds to
buy inputs
Not known as
a fertilizer
_
40% 40% 40%
40% 40%
40%
_
_
_
28% 28% 28% 28%
28% 28%
28%
8%
4%
4%
_
_
_
_
28%
_
_
_
_
32% 32%
_
_
_
Source not known
or not available
16%
_
_
28%
28% 28%
Amongst the things learned from the trials (Table 38.3), the most
indicated was the use of inorganic fertilisers (69%). It is also interesting
to note that most of the new knowledge farmers indicated to have learned
addressed general techniques of crop management, such as using
fertilisers, line planting, thinning, weeding and top dressing, rather than
specific organic technologies such as the legume cover crops (12%) or
applying tithonia biomass (4%).
Early Farmer Evaluation of Integrated Nutrient Management Technologies in Eastern Uganda
551
Table 38.3: Things farmers learnt from the trials (n=25)
Things farmers learnt
Frequency
Percentage
18
11
4
3
3
3
2
1
1
69
42
15
12
12
12
8
4
4
Use of fertilizers
Line planting/spacing
Inoculation with Rhizobia
Thinning
Top dressing
Planting legume cover crops
Frequent weeding
Planting tithonia
Measuring of harvest area
Table 38.4: Current impressions of the soil fertility enhancing options (n=25)
Busumbu
Busumbu
blend
Minjingu
PR only
TSP
Urea P and
K
Tithonia
Compost
Rhiz
obia
– – – – – – – – – – – – – – – – – – – Percentage (%) – – – – – – – – – – – – – – – – – –
Gives higher yields
Not well known
Poor performance
Not aware of option
Increased number
of groundnut pods
48
28
4
4
48
28
4
4
48
28
_
4
48
28
_
4
48
28
_
4
48
28
_
4
16
_
_
_
_
_
_
_
16
_
_
4
_
_
_
_
_
_
_
_
16
In general, nearly half of the farmers (48%) felt that yields had
improved from the use of inorganic fertilisers and only 4% felt that the
Busumbu PR or Busumbu blend had performed poorly (Table 38.4).
The use of organic materials appeared less compelling – 16% said Tithonia
had increased yields, while a similar proportion felt that the use of
Rhizobia had increased yields or the number of groundnut pods. When
asked about whether they wanted to expand the areas dedicated to the
trials, 83% were prepared to do so. Over half (56%), would increase the
area to between one and ten acres, 28% would dedicate at least half an
acre , and 8% would allocate a quarter acre.
Farmers' own experiments and dissemination efforts
Farmers learnt how to conduct own experiments following contact with
the INSPIRE project activities. The own experiments given as provided
by the farmers included planting:
1) maize with compost,
2) maize with FYM manure,
3) groundnuts with compost,
552
4)
5)
6)
7)
8)
Miiro, R. et al
groundnuts and spraying with urine to control pests and diseases,
pineapples with and without compost and ash,
beans with compost,
sorghum with compost and
maize with and without Di-Ammonium Phosphate (DAP).
Farmers indicated to have had regular discussions about the trials
and the newly learned techniques with their neighbours. Over three
quarters of the farmers (76%) had discussed the trials specifically, while
92% said they had discussed related topics. These topics included: use
of both organic and inorganic fertilisers to increase yields, spacing and
line planting of maize and groundnuts, timely planting as well as group
formation and working together. In the process, a total of 179 other
farmers were talked to (Table 38.5). On average each farmer had talked
to 9 farmers. These discussions resulted in neighbouring farmers
wanting to participate in the evaluation of some of the technologies,
line planting and spacing (48%), planting maize with fertilizers/Tithonia
(20%) and use of compost (4%).
Table 38.5: Number of neighbours the farmers told about the technologies
Number of other
farmers contacted
Frequency
Total
20
15
10
6
5
3
1
None
4
1
4
3
2
5
1
5
80
15
40
18
10
15
1
25
179
Participatory group evaluations of the legume cover crops/
shrubs
Six farmers who hosted the legume cover crops/improved fallow
experiments for two seasons were involved in the participatory evaluation
of the cover crops. In addition over 60 farmers who included those who
had grown the fallows and were about to incorporate the fallows and
those who wanted to start the on-farm test trials, were present at the
evaluation. A number of issues arose from the evaluation of the
technologies and the highlights based on specific criteria/aspects are
given in Table 38.6.
Early Farmer Evaluation of Integrated Nutrient Management Technologies in Eastern Uganda
553
Table 38.6: Qualitative evaluations of the legume cover crops/shrubs
Legume cover crops/shrubs
Crotalaria Tephrosia
grahamiana
Mucuna
Legume shrubs/trees
Canavalia Crotalaria Sesbania
pancilla
Germination Took a week to germinate
Took 4
days if
seed is
soaked in
hot water
Vegetative
growth
Good
Pests
Attacked by
caterpillar
when its
moist
Solutions
to pests
Spray with
pesticides
or use ash
Drought
resistance
Good if
planted
closely
The Best
Labour
requirements
High for
weeding
High for
weeding
Harvesting
seed
Seed harvesting need a lot of labour
Family
labour
contributor
All participate but the men do more work on the crops
Seeding
Seed
Seed
needed
Very
good
Easily
attacked
by caterpillars,
Calliandra Leucena
Good
Fair
Good if
planted
closely
Gives
higher
seed
yields
Farmers have their own seed stands
Want more seed
Easily
eaten by
caterpillars
Most
scarce
Farmers have their own seed
stands
Want
More
seed
Initial benefits of the green manure cover crops
The farmers indicated the initial benefits of growing the green manure
cover crops. With germination, mucuna was considered the best, followed
by canavalia and Crotalaria grahamiana. Mucuna was indicated to
rapidly produce thick vegetation on the land that smothered weeds.
554
Miiro, R. et al
Maize planted with mucuna grew very fast and looked healthy. It was
rated the best in terms of drought resistance, while C. grahamania, and
C. pancilla were good at drought resistance if planted closely. Mucuna
also gave the highest seed yields. It was used as a livestock feed, and its
boiled seeds were edible to humans.
Tephrosia rejuvenated soil fertility, killed mole rats and was quick
maturing. It could also be used for harvesting fish.
Sesbania improved soil fertility and could eradicate the striga weed
if planted as a fallow. It also provided firewood and its poles were used
to make fences. Canavalia controlled weeds when planted in banana
plantations, and its seed were edible after boiling. Tithonia could be
used as medicine against cough, and stomach pains. As a measure to
ensure seed sufficiency, farmers had established own seed stands for
most of the legume cover crops/shrubs except for C. pancilla. However,
they wanted more seeds for those crops with seeding difficulties (C.
grahamania, and C. pancilla). Apparently men worked more on the green
manure crops than women.
Difficulties with the fallows planted
Farmers indicated the difficulties they encountered with the fallow crops.
Crotalaria species had small seed and were easier to broadcast than to
plant in lines. Weeding crotalaria before it establishes was a laborious
exercise and it needed spraying or use of ash when infested by the
caterpillars. All legume cover crops and shrubs were indicated to have
difficulties in getting seeds. Sesbania had beetles that ate the leaves
and young shoots, which tended to kill the plants. Sesbania pods were
usually sharp pointed at the end and pierced during transportation.
Mucuna which has climbing characteristics increased the labour
demands to have it removed from the crops it was intercropped with
such as maize. Canavalia pods were hard to split during threshing,
while tephrosia was difficult to weed and produced itching dust when
threshing. Tithonia was observed to leave a bitter taste in ones hands
after working with it. A number of these findings concur with those in
Miiro et al. (in press) who reported on the integration of green manure
cover crops in the farming systems of small scale farmers in Iganga
district. They particularly point out labour difficulties associated with
mucuna intercrops, and harvesting crotalaria.
Table 38.7 shows the pairwise ranking that was done for six legume
crops including mucuna, canavalia, Crotalaria pancilla, Crotalaria
grahamiana, sesbania, and tephrosia. The criteria for evaluating the
crops included their ease to germinate, vegetative production, and ease
to manage. Results show that farmers ranked mucuna first followed by
sesbania then Crotalaria pancilla, C. grahamiana, tephrosia and
Early Farmer Evaluation of Integrated Nutrient Management Technologies in Eastern Uganda
555
canavalia. In Iganga district, small holder farmers integrated mucuna
more than the other species because of its fast growing nature, and
ability to improve soil fertility (Miiro et al. in press)
Table 38.7: Pairwise ranking of the five important legume crops/shrubs in Kisoko
Mucuna Canavalia Crotalaria Crotalaria Sesbania Tephrosia Score
pancilla
grahamiana
M
Rank
M
M
M
M
5
1
Mucuna
P
G
S
T
0
6
Canavalia
P
S
P
3
3
C. pancilla
S
G
2
4
C. grahamiana
S
4
2
Sesbania
1
5
Tephrosia
Conclusions and implications for the INSPIRE project
The testing on-farm of the various inorganic and organic sources of
nutrients by farmers in eastern Uganda, seems to yield useful learning
experiences to both the farmers and the INSPIRE members. Initial worries
about the experiments were hinged around the use of inorganic fertilisers
and their being expensive. This however reveals a knowledge gap and
need to expose farmers to the various soil fertility options including their
crop yield and cost implications. Further sustainable promotion of
integrated nutrient management system should take care of this. The
choice of fallow species depends very much on the production constraints
that the farmer wants to address; whether smothering invasive weeds
like couch grass, or introducing a multi-purpose legume for livestock
feed or fuelwood not just soil fertility. Not surprisingly, due to financial
constraints farmers are more inclined towards technologies that do not
require a large capital investment and would rather allocate their time to
production and management of legume cover crops and improved fallows.
Farmer to farmer communication was seen as a very successful
way of disseminating new technologies to a wider audience. This should
be reinforced with farmer exchange visits, field days, and training farmer
extensionists or farmer to farmer training. There was willingness among
the farmers of Tororo district to test new technologies. Interestingly,
after many years of extension services on crop production, one of the
main results from the technology testing was the increased knowledge
on crop management in terms of planting, thinning and weed
management. Understanding of farmer priorities and constraints,
decision making and how farmers trade-off production technologies and
their time will be increasingly important in targeting interventions/
556
Miiro, R. et al
technologies to smallholder farmers. The INSPIRE project needs to refine
these technologies through more participatory technology testing
approaches to foster adaptability to farmer conditions and empower
farmers with a sustainable green manure husbandry system that
includes a strategy to ensure seed sufficiency.
Acknowledgements
The authors acknowledge the following for their role in data collection,
Jacinta Namubiru, Ali Mawanda, John Obwego of Africa 2000 Network
Tororo and Tom Ochinga of ICRAF-Maseno. They also acknowledge TSBF
for funding the conference fees towards the lead author's attending of
AfNet 8.
References
Ashley, C. and Carney, D. (1999) Sustainable Rural Livelihoods: Lessons from
early experience. Department for International Development (DFID) Russel
Press Ltd. Nottingham United Kingdom.
PMA, (2000) "Plan for the Modernisation of Agriculture: Eradicating Poverty in
Uganda” Government Strategy and Operational Framework. Ministry of
Agriculture, Animal Industry and Fisheries (MAAIF) and the Ministry of
Finance. Planning and Economic Development (MFPED). Government of
Uganda.
Miiro, R., A. Esilaba and S. David (in press). Integration and dissemination of
green manure cover crops in African small scale farming systems: Successes
and constraints in Eastern Uganda. Paper accepted by the Makerere
University Agricultural Research Institute Kabanyolo Bulletin
Republic of Uganda (1991) Housing and Population Census.
Republic of Uganda (1997) The Local governments Act 1997 and the Local
Governments (Amendments) Act, 1997.
Sanchez, P., Izac, A.M., Buresh, R., Shepherd, K., Soule, M., Mokwunye, U.,
Palm, C., Woomer, P. and Nderitu, C. (1997) Soil Fertility Replenishment in
Africa as an Investment in Natural Resource Capital. In: Buresh, R., Sanchez,
P. (eds.), Replenishing Soil Fertility in Africa. SSSA Special Publication 51
Madison WI: Soil Science Society of America (SSSA).
SFI (1999) Uganda. Soil Fertility Initiative. Concept Paper. Ministry of Agriculture,
Animal Industry and Fisheries (MAAIF), National Agricultural Research
Organisation (NARO), Food and Agricultural Organisation (FAO) and
Investment Centre Division FAO/WORLD BANK Co-operative Programme.
The Republic of Uganda.
Zake, J.Y.K., Nkwiine, C., Miiro, R. (1998) A baseline study of the Farming Systems
of Client Farmers of Africa 2000 Network in Tororo District, Eastern Uganda.
Wortmann, C.S., Eledu, C.A.. (1999) Uganda’s Agro-ecological Zones: A guide
for planners and policy makers. Kampala Uganda: Centro Internacional de
Agricultura Tropical (CIAT).
World Bank (1993) Uganda Growing out of Poverty. World Bank Country Report.
Potential for Adoption of Legume Green Manure on Smallholder Farms in Western Kenya
Potential for Adoption of
Legume Green Manure on
Smallholder Farms in
Western Kenya
557
39
Odendo, M., Ojiem, J. and Okwosa, E
Kenya Agricultural Research Institute, Regional Research
Centre, P.O. Box 169, Kakamega, Kenya
Abstract
Low soil fertility is a major constraint to increased
agricultural productivity amongst smallholder farmers in
Kenya. The use of inorganic fertilizer to alleviate the
constraint is limited mainly by its high cost, untimely
availability and low producer prices. Under the Legume
Research Network Project (LRNP), various legume species
that can be utilized as green manure to supplement the
use of inorganic fertilizer were screened and the species
well adapted to different agro-ecological zones of Kenya were
selected. This study was conducted when the species
selected for western Kenya were being tested on the farmers’
fields to assess their effect on soil fertility improvement and
crop yields. The legumes, as a component of integrated
nutrient management (INM), were grown in rotation with
maize. Seven treatments composed of inorganic fertilizers,
farmyard manure, green manure and combinations of these
compounds were evaluated. A significant maize grain yield
response to treatments was observed during
experimentation.
558
Odendo, M. et al
The objective of this study was to assess diffusion and
potential for farmers’ uptake of the various components of
the green manure technology introduced to them and
determine technological and socio-economic predictors to
adoption. A survey was conducted in Kabras cluster site,
Kakamega District, one of the six LRNP cluster sites in Kenya.
The study involved a survey of key informants and farm
households, comprising 11 farmers who hosted the green
manure trials and 34 randomly selected households that
were non-participating in the trials. Results show that most
farmers were not aware of performance and mechanisms of
carrying out the various components of the technology.
Although none of the respondents had adopted legume green
manure, all the experimenting farmers expressed willingness
to adopt. High labour demand, especially for establishment
and timely incorporation of the manure and inadequate
availability of legume seed were the most important
constraints envisaged by the farmers. More farmers should
be involved in testing of the technology and efforts should be
made to sensitize farmers on the potential of the technology.
There is need to synchronize the recommended activities for
management of the manure in order to combine certain
operations to minimize labour demand.
Key words: Adoption, green manure, legume, smallholder, soil fertility
Introduction
Smallholder farming in many parts of Kenya is mainly constrained by
declining soil fertility. Recent studies in western Kenya have identified
soil fertility decline as one of the most important biophysical constraints
to increased agricultural productivity (KARI, 1994; Odendo and Ojiem,
1995; Ojiem and Odendo, 1996). The use of inorganic fertilizers to
mitigate soil fertility decline is limited by mainly high costs, low producer
price of most food crops and erratic availability of the fertilizers. A few
farmers that use inorganic fertilizers cannot afford recommended rates.
Under such conditions, one of the alternatives is to supplement inorganic
fertilizers with other sources of plant nutrients such as green manure.
Integration of legume green manure into the farming systems can be a
cheaper alternative of alleviating low soil fertility and erosion problems
(KARI, 1997; Kanyanjua et al., 2000). To enable introduction and
integration of legume green manure into farming systems in Kenya,
legume screening under the auspices of Legume Research Network
Project (LRNP) was set up to screen legume species and select those
Potential for Adoption of Legume Green Manure on Smallholder Farms in Western Kenya
559
adapted in various regions of Kenya. After two years of on-station
screening in the regions, the well adapted legume species were selected
(Ojiem and Okwosa, 1999; Kanyanjua et al., 2000). About 8 legumes
were selected for western Kenya, of which two; velvet beans (Mucuna
pruriens) and sunnhemp (Crotolaria ochroleuca) were introduced in
Kabras Division (Ojiem and Okwosa, 1999; Ojiem et al., 2000). Seven
treatments comprising recommended rate of inorganic nitrogen (60 kg
N ha -1) (full N), recommended rate Farm Yard Manure (5 t ha-1 FYM),
legume green manure biomass (GM) alone, GM + half N, GM + half
FYM, half N+ half FYM, and GM+ half N + half FYM were evaluated
against the farmers’ practices in Kabras division in 1998 and 1999.
Phosphorus was applied at a blanket rate of 30 kg P205 ha -1 as Triple
super phosphate. Half the farms were planted with Crotalaria ochroleuca
and the other half with Mucuna pruriens. Crotalaria ochroleuca and
Mucuna pruriens, as components of integrated nutrient management
(INM), were grown in rotation with maize. The legumes were established
in the short rain season, incorporated into the soil in the long rain
season and then maize (H512), was grown in the fields as a test crop.
Significant (p=0.05) maize grain yield response to treatments were
observed in 1998 and 1999. Over the two years, maize grain yield was
3.98 t ha-1 for full N, 3.67 t ha-1 for GM+ half N, and 3.67 t ha-1 for GM
+ half N + half FYM. The maize grain yield of GM + half FYM was 2.24
t ha -1 and was slightly lower than the farmers’ practices (2.5 t ha -1).
These results indicate that inorganic N requirements can be reduced by
50%, but high maize grain yields maintained by supplementing with N
from GM and FYM (Ojiem et al., 2000).
There is often time lag between technology development and adoption
(Mills et al., 1998). In order to monitor adoption, it is important to have
a close working relationship between farmers and researchers, as a
technology is being developed and tested. This interaction provides early
indication of whether or not a new technology is acceptable.
Experimenting farmers are very important in assessing acceptability of
a technology because such farmers provide insights about potential
adoption of a new technology such as green manure. These farmers
have relatively more experience with the components and performance
of the technology than the rest of the farmers. A technology may be
changed or modified by a user in the process of diffusion and adoption.
As Rogers (1983) observes, potential users may play an important role
in the process of technology generation by being involved in the
generation process, rather than being merely passive recipients of
innovation once it has been generated.
A review of adoption studies in developing countries reveals that no
study had analysed the direct effects of farmers’ subjective assessment
of agricultural technology on adoption decisions (Feder et al., 1985).
Most adoption studies analyse the reasons for adoption or non-adoption
560
Odendo, M. et al
at a point in time, principally in terms of socio-economic characteristics
of adopters and non-adopters. Recent studies (e.g. Adesina and Zinnah,
1993; Adesina and Baidu-Forson, 1995), however, demonstrate that
farmers’ perceptions of the characteristics of the technology significantly
affect adoption decisions. In view of the recent studies, this study
considers both characteristics of potential users and the technology
attributes as important explanatory variables for farmers’ decisions to
adopt legume green manure.
Measuring adoption of soil fertility improvement technologies is a
major challenge facing researchers who attempt to model soil fertility
management decision processes (Ervin and Ervin, 1982; Lynne et al.,
1988, Purvis et al., 1989). Some of the measures that have been used
include willingness to adopt, actual adoption decision and the extent of
adoption. Purvis et al.(1989), for instance, measured the willingness of
Michigan farmers in the USA to accept yearly payments for participating
in filter strips program using contingent valuation method (CVM). The
CVM involves eliciting values; people are asked directly to state or reveal
what they are willing to pay for some change in provision of a good or
service or what they are willing to accept to forego a change or tolerate
a change (Mitchel and Carson, 1989). For this study, since there was a
short duration between technology testing, diffusion and assessment of
adoption, farmers’ willingness to adopt legume green manure was a
plausible method of assessing adoption and CVM was applied. However,
because the actual values (amounts) the farmers were willing to pay in
order to adopt the technology was not elicited, only some principles of
CVM were adapted.
The objective of this study was to assess diffusion and potential for
farmers’ uptake of the various components of legume green manure
technology introduced to them and determine technological and socioeconomic predictors to adoption. The knowledge gained from the study
could be used by both researchers and farmers to fine-tune the
technology with a view to enhance relevance and likelihood of adoption.
Methodology
The study area
This study was conducted in the LRNP cluster site in Kabras Division of
Kakamega District in western Kenya. Kabras Division covers an area of
about 429 km2, of which 367 km2 is arable. With population of about
169,000 persons and density of 386 persons per km2, Kabras division
is relatively sparsely populated in relation to other Divisions of Kakamega
district. The average land holding per household is about 1.4 ha (MoRD,
2000; Ministry of Agriculture staff, Kabras, pers. comm.).
Potential for Adoption of Legume Green Manure on Smallholder Farms in Western Kenya
561
The Division falls within a relatively high agricultural potential area,
with a third of its landmass falling within the high potential Upper
Midland and Lower Midland agro-ecological zones (Jaetzold and Schmidt,
1983). It falls at an altitude of 1300-1500 m above sea level. The mean
annual temperature is 22-29°C, which is one of the highest in the District.
Annual average rainfall is 1500-1800 mm, with peaks in April and
August. The rainfall pattern is bimodal, providing for two cropping
seasons per annum. The long rain season (March-July) receives 650750 mm and the short rain season (August-November) 500-600 mm of
rainfall, whilst the mean annual pan evaporation ranges between 1600
and 1800 mm. There is a significant variability of soils; the major soil
types are Nito-rhodic Ferralsols, Acrisols, Combisols and Gleysols.
However, the predominant is Nito-rhodic ferralsols. These are mainly
developed from granites and are well drained, very dark, dusky red to
yellowish red and in some places friable clay loams with acid humic
topsoil (Jaetzold and Schmidt, 1983; Ministry of Agriculture staff, Kabras,
pers. comm).
Agriculture is not only the main economic activity, but also has a
social function as it is involved in food security (Ministry of Agriculture
staff, Kabras, pers. comm). The farming system incorporates crops and
livestock. The main food crops grown include maize/beans (the staple
food), sweet potato, sorghum, groundnuts, pineapples, kales, tomatoes
and cabbages. Major cash crops are sugar cane and maize. Most of the
sugar cane is grown in south and west Kabras locations, which are
situated near the sugarcane-processing factory. Livestock comprise
mainly of zebu cattle, though some crosses and grade cows are found in
areas bordering Nandi District. Milk is the main livestock product. The
rearing system is commonly free range coupled with tethering. Other
livestock types include sheep, poultry and rabbits. Cattle rearing has
been ranked third after maize and sugarcane in priority ranking
exercises. The ranking exercises considered the enterprise distribution
in scale (number of households), significance in meeting subsistence
needs and the level of income generated from the enterprises.
Low maize harvests and low milk yields are identified as the major
problems affecting the farming households in the area. Potential
solutions to these problems include improvement in quality and quantity
of feeds for livestock and improvement of soil fertility for increased crop
yields, especially maize. A cursory observation of the two problems and
solutions suggest that an integration of crop and livestock enterprises
would address the problem to some extent. Thus, the manure from
livestock, especially cattle, would be used to improve soil fertility, more
so when the livestock is fed on high quality feed such as legumes, whilst
crop residues from increased crop yields would be used to increase
livestock productivity. Integration of legume green manure in the farming
system would contribute a great deal in improving soil fertility for
562
Odendo, M. et al
increased crop production and improved livestock productivity through
provision of high quality legume feeds.
Sampling procedure
The sites for on-farm green manure experimentation and for this study
were distributed in three administrative Locations; Chemche, Mahira
and Lukume, to represent biophysical and socio-economic variability.
A list of all households in the study clusters, which formed the sampling
frame, was obtained from the local elders (liguru). A sample of 11 out of
the 20 households that hosted LRNP experiments and a sample of 34
non-participating households from within the villages where the onfarm trials were being conducted and contiguous villages, were sampled
from the household lists.
Data collection and analysis
A structured questionnaire composed of both open and closed ended
questions was designed by socio-economists in collaboration with LRNP
implementing scientists, to help collect relevant primary data. In addition
to the questionnaire, personal field observations and interview of key
informants such as extension officers, KARI researchers, local leaders
and farmers were conducted using a checklist to supplement the
questionnaire. A draft questionnaire was pre-tested on five farm
households in the vicinity of the study clusters and two trial participating
farmers. The result of the pre-testing helped in final restructuring of
the questionnaire by incorporating missing variables or information,
omitting irrelevant questions and paraphrasing questions that appeared
ambiguous to the respondents. The final questionnaire had three
sections. The first section consisted of general questions, including
demographic and socio-economic characteristics of the respondents and
their households, as well as farming systems under which they operate.
Questions in section two focused on technological components of legume
green manure that respondents were aware of, were willing to uptake
or had adopted and any modifications they had made, and factors
affecting uptake of the components.
Prior to the interviews, the implementing team was trained on a
number of relevant aspects of green manure. These included elements
of legume green manure as a component of integrated nutrient
management, general information about the legumes (growth habits),
methods of establishment, weed control and biomass incorporation.
Others were management, storage and application of farmyard manure,
Potential for Adoption of Legume Green Manure on Smallholder Farms in Western Kenya
563
time and rate of application of inorganic nitrogen and agronomy of maize.
The authors and research technicians as well as extension officers
conducted personal interviews. Heads of the households were interviewed
however in their absence, a household member conversant with farm
activities was interviewed. For the LRNP participating farmers, the
interviews focused on application of technological components on fields
or plots other than the experimental ones.
Respondent farmers were asked to state their willingness to adopt
legume green manure by exposing them to costs and benefits, not
necessarily in monetary terms, involved in utilisation of the technology.
Willingness to adopt reflects individuals’ preference for a good (or service)
in question such as a component of soil fertility management technology.
We presented the non-participating farmers with a hypothetical
opportunity on use of legume green manure, especially Mucuna pruriens
and Crotolaria ochroleuca as components of integrated nutrient
management. Two types of information were provided to respondents.
First, was a detailed description of components of legume green manure,
their purpose and expected outcomes of their usage in form of a scenario.
Immediately following this scenario, respondents were asked whether
or not they were willing to uptake the manure. In addition, they were
asked to give reasons for their willingness to adopt using an open ended
format. Such follow-up questions were essential in order to probe
respondents’ perceptions and their reasoning behind the responses.
The data were analysed by descriptive statistics using Statistical Package
for Social Sciences (SPSS) software.
Results and Discussion
Socioeconomic and demographic characteristics of
sampled households
The selected socioeconomic and demographic profiles of the sampled
households are presented in Table 39.1. Over 90% of the households
were male-headed. The mean age of the household heads was 42.5 years.
The highest education level for a majority heads of households was
primary, whilst the mean number of persons per household was nine.
A paltry 31.1% of the farmers received some credit in the last 5 years,
while 35.6% received some off-farm incomes. The mean land size was
6.5 acres. Majority of the households owned cattle, mostly local zebus
at an average of four cattle per household.
564
Odendo, M. et al
Table 39.1: Socioeconomic and demographic profiles of sample households
Characteristic
n=45
Mean (S.D.)
Percent
6.5 (2.0)
Farm Size (Acres)
Household head:
Male
Female
–
–
91.1
8.9
42.5
–
Highest Education:
None
–
18
Primary
–
45.5
Secondary
College
–
–
34.4
2.1
Household size
Number working on farm full-time
8.7 (3.9)
2.6 (1.7)
–
–
Number working on farm part-time
3.0 (2.0)
–
–
–
31.1
35.6
4.3 (3.8)
–
Age (Years)
Received credit in last 5 years
Received off-farm income
Number of cattle/household
Notes:
1) The sample (n) includes 11 households that hosted the green manure experiments.
2) Figures in parentheses are standard deviations (S.D)
Farmers’ awareness and uptake of components of
integrated nutrient management
Awareness of the potential benefits of a new technology is a commonly
acknowledged prerequisite for farmers to decide whether to or not adopt
a technology. Farmers’ awareness and the rate of adoption of some
recommended components of integrated nutrient management (INM)
are shown in Table 39.2.
Potential for Adoption of Legume Green Manure on Smallholder Farms in Western Kenya
565
Table 39.2: Percentage of farmers who are aware and/or are practicing some INM
components
Technology components
Inorganic fertilizer
Organic fertilizer
Green manure
Maize spacing
Percentage of farmers
n=45
Aware
Practicing
100.0
100.0
28.9
68.8
35.5
88.9
0.0
33.3
Note: Percentages are computed independently, hence do not add up to 100
All respondents knew some Crotalaria sp, which they grow and
utilize mainly as a vegetable in the area and most of them were not
aware whether the species or other species could be used as green
manure. About 29% of the respondents were aware that legume green
manure, especially Mucuna pruriens and Crotolaria ochroleuca, could
be utilized to improve soil fertility. These are the farmers who were
either hosting legume green manure trials or had interacted with
experimenting farmers, researchers or extension agents.
Farmers were utilizing certain components of INM technology on
maize crop. About 36% had adopted inorganic fertilizers, particularly
Calcium Ammonium Phosphate (CAN) and Di-Ammonium Phosphate
(DAP), whilst nearly 90% of the study households applied some organic
manure, mainly on maize. The doses of organic and inorganic fertilizers
were lower than the recommended rates. For example, the mean rate of
application of DAP, which the most popularly used basal fertilizer in
Kakamega district, was 37.6 kg acre-1 compared to recommended 50 kg
acre-1 . Only about 18% of the respondents top dressed their crops,
especially using CAN at the mean rate of 26.7 kg acre-1 , instead of the
recommended 50 kg acre-1. This was attributed to high costs of the
fertilizers, low cash incomes of the households and lack of adequate
knowledge about the recommended types and rates of inorganic
fertilizers. The low doses of farm yard manure were mainly associated
with the small number of cattle kept which could not produce enough
manure in a season, and free range rearing system that does not allow
efficient collection of manure. The recommended maize spacing of 75cm
× 30cm was adopted by 33% of the households. In most cases wider
inter-row spacing, ranging between 80cm to 100cm, were used. This
was meant to create more space for inter-cropping with other crops,
especially beans.
None of the respondents had actually adopted legume green manure
for soil fertility management. The low adoption was associated with lack
of exposure of the technology to the farmers. Even the experimenting
farmers indicated that they wanted to have more experience about the
566
Odendo, M. et al
use and performance of the technology via experimental plots, before
trying on their own plots. This is in line with what is fairly general
agreement that most people, including farmers are risk averse (Upton,
1987) . This means they are willing to forego some income or face extra
costs in order to avoid risk, and are hence cautious in their decisionmaking.
Although 63.4% of the experimenting farmers indicated that
application of green manure technology in the experimental plots did
not require much labour, they foresaw that more labour would be needed
on their own larger fields since legumes for green manure require
establishment at onset of rains when they are busy preparing land and
planting other crops. High labour demand was also perceived to be
required for incorporation—chopping before incorporation and timely
incorporation. Mucuna was perceived to have highest labour requirement
because of its high biomass. Although the mean number of persons per
household was large (Table 39.1), the proportion that works on-farm on
full-time basis was only about one-third, indicating labour shortage.
Farmers also lacked access to the green manure legume seed. These
notwithstanding, all the experimenting farmers expressed willingness
to adopt green manure legumes for soil management. However, it was
not easy to assess whether this was the actual farmers’ expressed
preference. As Swinkles and Franzel (1997) and Franzel et al. (1999)
observe, farmers often state that they like a technology, even if they do
not, because they hope to obtain material or social benefits from
interacting with technology facilitators or because of cultural taboos
against criticism or they simply try to please researchers.
In the on- farm experiments, the legumes for green manure were
grown in rotation with maize in the first season. The farmers initially
proposed the rotation system whereby the green manures were
established in the short rain season and incorporated in the long rain
season, expecting dramatic yield increment in the long rains. On the
basis of the yields obtained from the trial plots, farmers expressed
preference for either intercropping or relay cropping the legumes in
maize. It was also noted that although 84% of the farmers practiced
some fallowing, especially in the short rain season, the fallows were
mostly utilized as grazing land. Furthermore, the farmers preferred some
yield in the season, rather that having no crop at all in the fallows.
Therefore, rotation system of green manure application tested in Kabras
Division conflicted with grazing of livestock and other farmers’
preferences. Similar observations were reported by Franzel et al. (1999)
in a study conducted in western Zambia. Whereas farmers recognized
that intercropping of improved fallows such as sesbania and tephrosia
appeared to reduce maize yields and tree growth during the year of
establishment, many farmers preferred inter-cropping because it
economizes on land and labour use relative to planting pure tree stands.
Potential for Adoption of Legume Green Manure on Smallholder Farms in Western Kenya
567
This finding is in line with high discount rate that is associated with
most resource poor farmers. They prefer short-term benefits to satisfy
their basic needs.
Technology characteristics affecting potential adoption
Farmers considered certain characteristics of the various legume species
in deciding to adopt green manure (Table 39.3). High biomass was cited
as both a constraint and an opportunity. Mucuna pruriens was preferred
because of its high biomass production. The legume was perceived as
being able to improve soil fertility within a short time because of its
high biomass. Again, high biomass was considered a constraint because
it requires more labour to manage and does not allow for intercropping
with other crops, especially maize because of its climbing growth habit
and high ground cover. Crotolaria, on the other hand, was highly
preferred because of its potential multiple uses as a vegetable and for
soil fertility improvement. It was also considered to have medicinal value.
The farmers had, however, not explored other alternative uses of Mucuna
since it was a relatively new crop in the farming system.
Table 39.3: Major farmer desired legume characteristics for green manure use in western Kenya
Legume characteristics
High biomass
Suitable for intercropping with maize
Multiple uses
% farmers citing preference
for the characteristic
67.0
91.0
100.0
Note: percentages were computed independently and do not add up to 100.
Conclusions and Recommendations
Assessing potential adoption of a new technology such as green manure
legumes only a few years of technology testing is a difficult task. Except
for farmers hosting the experiments, most farmers were unable to
conceptualize how the technology is applied, how it works and its
performance. Indeed, even most experimenting farmers were not able
to confidently perceive the performance of the technology because of
the short duration experience they had had with the technology. The
technology takes a relatively long time to realize full benefits. More time
is required for testing and dissemination of the technology to many
farmers so as to create awareness and enable farmers make informed
decisions on whether they are willing to adopt or not. A larger number
of farmers should be involved in the technology testing and field days
568
Odendo, M. et al
should be held to disseminate the technology. It is only after this that
actual adoption can be meaningfully assessed.
The key socio-economic factors that constrained the technology
uptake across the sites was high labour demand at the time of planting
and incorporation of the green manure legumes and unavailability of
adequate amount of the legume seed. To minimize labour demand for
establishment and incorporation of the legumes, research should devise
ways of saving labour such as relay cropping the legumes in maize in
the first season and synchronize incorporation the legumes so that it
coincides with land preparation in the second season, whilst considering
the trade-offs and synergies.
To fit the green manure into the existing farming system, emphasis
should be on inter-cropping or relaying of the green manures rather
than establishing them in a rotation. Although most households
practiced some fallowing in the second season, most of the fallows were
utilized as grazing land. Seed for legume establishment was one of the
main constraints. There is need to develop a local farmer-managed
legume seed production system to enable farmers have access to seed.
Further research is required to investigate opportunity costs of the
alternative fallowing systems such as Mucuna or Crotolaria in relay and
rotation systems. This will help in assessing farmers’ willingness or
unwillingness to forego one season’s crop for the future expected benefits
from green manure legume rotation system. A more in-depth
investigation is required to assess competitive demand for legume species
with multiple uses.
Acknowledgments
This research was made possible by financial support of the Rockefeller
Foundation. We particularly acknowledge, with thanks, Drs. Joseph
Mureithi and John Lynam for their support and keen interest in the
study. We also thank the Centre Director for RRC-Kakamega, Dr. Apollo
Orodho, for his support. Special thanks to our colleagues in KARI and
Ministry of Agriculture for their help in data collection and the farmers
for sacrificing their time to discuss with us. Responsibility for errors
remains ours.
References
Adesina, A. A. and Zinnah, M.M. (1993) Technology Characteristics, Farmer
Perceptions and Adoption Decisions: A Tobit Model Application in Sierra
Leone. Agricultural Economics 9:297-311.
Potential for Adoption of Legume Green Manure on Smallholder Farms in Western Kenya
569
Adesina, A.A. and Baidu-Forson, J. (1995) Farmers’ Perceptions and Adoption
of New Agricultural Technology: Evidence from Analysis in Burkina Faso
and Guinea, West Africa. Agricultural Economics, 13:1-9
Ervin, C.A. and Ervin, D.E., (1982). Factors Affecting the Use of Soil Conservation
Practices: Hypotheses, Evidence and Policy Implications. Land Economics,
58:277-292.
Feder, G., Just, R.E. and Zilberman, D. (1985) Adoption of Agricultural
Innovations in Developing Countries: A survey. Economic Development and
Cultural change 33:255-297.
Franzel, S., Phiri, D. and Kwesiga, F. (1999) Assessing the Adoption Potential of
Improved Fallows in Eastern Zambia. AFRENA Report No. 124. Nairobi:
ICRAF.
Lynne, G.D. , Shonkwiler, J.S. and Rola, L.R. (1988) Attitudes and Farmers’
Conservation Behaviour. American Journal of Agricultural Economics 70(1):
12-19.
Jaetzold, R. and Schimdt, H. (1983) Farm Management Handbook: Natural and
Farm Management Information Vol.II/B. Nairobi: Ministry of Agriculture,
Livestock Development and Marketing.
Kanyanjua, S.M.; Mureithi, J.G.; Gachee, C.K.K. and Saha, H.M. (2000). Soil
Fertility Management Handbook for Extension Staff and Farmers in Kenya.
KARI Technical Note Series No. 6. Nairobi: KARI and LRNP.
KARI. (1994) Participatory Rural Appraisals (PRAs) for Inception Phase of the
Netherlands Funded Regional Research Programme at KARI-Kakamega
Mandate Region. Kakamega: Kenya Agricultural Research Institute (KARI).
KARI. (1997) Legume Screening Network Research Project Proposal. Nairobi:
KARI - National Agricultural Research Laboratories, Nairobi.
Mills, B., Hassan, R.M and Mwangi, P. (1998) Estimating Potential Benefits
from Research and Setting Research priorities for maize in Kenya. In: Hassan,
R.M. Maize Technology Development and Transfer: A GIS Application for
Research Planning in Kenya. Cambridge University Press
Mitchel, R.C. and Carson, R.T. (1989) Using Surveys to Value Public Goods:
The Contingent Valuation Method: Resources for the Future. Washington
DC.
MoRD. (2000) Annual Report, Kabras Division, Kakamega District. Kakamega:
Ministry of Agriculture and Rural Development (MoRD) Kakamega.
Odendo, M. and Ojiem, J. (1995) Socio-Economic Characterisation of Farmers:
A Case Study of Kabras Division, western Kenya. Paper presented to FARMAfrica/IIED Nutrient Cycling Workshop, Soddo, Ethiopia, November 27 December 2, 1995.
Ojiem, J. and Odendo, M. (1996) Biological Management of Soil Fertility in Kabras
and Vihiga Divisions of Western Kenya. Second Annual Report to the
European Union. In: Carter, S.E.(ed.) Biological Management of Soil Fertility
in Small-Scale Farming Systems in Tropical Africa. Nairobi: TSBF.
Ojiem, J. and Okwosa, E.A. (1999) Role of Legumes in Small-Scale Productivity
Improvement. A News Bulletin. Kakamega: Kenya Agricultural Research
Institute.
570
Odendo, M. et al
Ojiem, J.O. Mureithi, J.G. and Okwuosa, E. A. (2000) On-farm Evaluation of
Promising Organic-inorganic Soil Fertility Management Options for
Smallholders in western Kenya. Paper presented at the 2nd Conference of
the Soil Management Project and Legume Research Network Project (LRNP),
Continental Hotel, Mombasa; June 26 - 30, 2000.
Purvis, A., Hoehn, J.P., Sorenson, V.L. and Pierce, F.J. (1989) Farmers’ Response
to a Filter Strip program: Results from a Contingent Valuation Survey.
Journal of Soil and Water Conservation 44(5): 501-504.
Rogers, E. (1983) Communication of Innovations. New York: Free Press.
Swinkles, R. and Franzel, S. (1997). Adoption Potential of Hedgerow Inter
cropping in Maize - based Cropping Systems in the Highlands of Western
Kenya 2. Economic and Farmers’ Evaluation. Experimental Agriculture
33:211-223.
Upton, M. (1987). African Farm Management. New York: Cambridge University
Press.
571
The Profitability of Manure Use on Maize in the Small-holder Sector of Zimbabwe
The Profitability of Manure
Use on Maize in the Smallholder Sector of Zimbabwe
40
Mutiro, K. and Murwira, H.K.
Tropical Soil Biology and Fertility, c/o University of
Zimbabwe, Faculty of Agriculture, Department of Soil
Science and Agricultural Engineering, P.O. Box MP 228
Mount Pleasand, Harare, Zimbabwe
Abstract
Several manure use options were analysed for profitability
using results from farmer participatory trials conducted in
the small holder farming sector of Zimbabwe. The options
analysed were
a) not using any manure
b) using aerobically composted (heap stored) manure,
c) using manure improved through anaerobic storage (pit
stored),
d) different manure application methods such as banding,
broadcasting and station placement, and
e) supplementing manure with mineral fertilizer.
The use of manure provided a marginal rate of return
(MRR) of at least 215% compared to not using manure. The
MRR on manure use was increased significantly by
composting manure in pits. Financial benefits obtained from
pit stored manure were much higher in the first year of
manure application compared to those of heap stored
manure. Higher returns from heap stored manure were
obtained in the second and third seasons after manure
application. Overall undiscounted financial benefits for the
572
Mutiro, K. and Murwira, H.K.
three years were marginally higher for pit stored manure.
Higher financial benefits were obtained from supplementing
manure with mineral fertilizer compared to using manure
alone. Banding and placing manure on-station increased
returns from using both pit and heap stored manure. The
conventional practice of broadcasting manure was found
not to be profitable.
Key words: Economic Returns, Profitability, Manure, Mineral Fertilizer,
Smallholder, Zimbabwe
Introduction
Soil fertility depletion in the smallholder farms is the fundamental cause
for declining per capita food production in sub-Saharan Africa (Sanchez
et al., 1997). Increased food insecurity, reduced farm incomes, limited
returns from agricultural investment and rural poverty are some of the
consequences of declining soil fertility. Input purchases by smallholder
farmers have not been enough to cover for the nutrient outflows (Smaling
et al., 1997). Mineral fertilizers have become unaffordable for most
smallholder farmers since the removal of subsidies in 1991 when the
Zimbabwe government embarked on the Economic Structural
Adjustment Program (ESAP). Most smallholder farmers in Zimbabwe
have turned to the use of manure as a low cost option. However the
effective utilisation of manure is constrained by severely limited
quantities available, poor quality of the manure with most manures
having less than 1 % nitrogen and a high sand content (Mugwira, 1985;
Tanner and Mugwira, 1984). Improving manure management and
storage is an important option for improving yields and returns to
investment in crop production.
This paper is an economic appraisal of the promising manure related
soil fertility management options that were tested in the smallholder
farming environment.
Materials and Methods
Three sets of different farmer participatory trials were conducted in the
smallholder farming area of Murewa. The trials were on (1) the effect of
supplementing 5 tonnes of manure with varying levels of fertilizer N (0,
20, 40, 80, 100 kg N ha -1) on maize yield, (2) the effect of differently
cured manure (aerobic and anaerobic) on maize yield and residual effects
in subsequent seasons and (3) the effect of different manure placement
methods on maize yield. All the trials were implemented in Murewa,
The Profitability of Manure Use on Maize in the Small-holder Sector of Zimbabwe
573
north east of Harare. The trials were conducted on sandy soils, typical
of most smallholder farming areas of Zimbabwe.
In the trial on supplementing manure with mineral fertilizer, the
nitrogen was applied twice, at 6 and 10 weeks after planting and this
conforms to the normal farmer practice. Manure was broadcasted at
planting. The trial was conducted for two seasons, 1997/98 and 1998/
99.
Most smallholder farmers store their manure for at least three
months before application in the field. The conventional way involves
digging manure out of the kraal and heaping it beside the kraal for 3
months. This was compared to the new innovation of digging a pit beside
the kraal and then putting manure in the pit that is then covered to
ensure anaerobic decomposition. The manure is kept in the pit for three
months. The manure was banded and applied at a rate of 10kg N ha-1.
The other option was to test different manure placement methods;
broadcasting, banding and spot application. Broadcasting is the
conventional method of applying manure used by most smallholder
farmers in Zimbabwe. For the comparison of different application
methods an application rate of 100kg N ha-1 equivalent was used basing
upon the total N concentrations of the manures.
A financial analysis was conducted to appraise the different options
for private profitability. A cost benefit analysis was conducted for the
trial on different manure storage systems and their residual effects. A
full budget for the maize enterprise under the different treatments or
trials was prepared based on marketable output. The trial yields were
adjusted by 10% to cater for field losses. Factory gate maize price was
used in the analysis. The cost of transporting the maize to nearest depot
was included in the analysis. Farm gate prices were used for all the
inputs namely mineral fertilizer, seed and insecticides. The prices of
inputs were collected from the nearest rural service centre offering such
inputs.
One of the major costs in the utilization of manure is the labour
used in the digging, curing, transportation and application of the manure.
A survey was undertaken in two communal areas, Murewa and
Tsholotsho, to collect information on labour. Farmers were asked to
state the time they take and the cost of the labour used in the
management and utilization of manure. Information was collected for
each manure related operation, digging manure from the kraal, heaping
or placing manure into the pit, transporting manure to the field and
application of manure in the field. Discussions were also held with
farmers to confirm survey findings.
The survey results and discussions with farmers revealed that pitting
of manure require an additional 5 person days compared with curing
manure on the heap. Farmers also indicated that heap stored manure
has a lot of weed seed compared to pit stored manure and farmers
574
Mutiro, K. and Murwira, H.K.
allocate more labour days on weeding fields where heap stored manure
is applied compared to where pit stored manure is applied.
Financial returns from the technologies are a function of the maize
yield obtained from the technology, cost of implementing the technology
and improvement in the fertility status of the soil (residual effects) over
time.
The financial returns are given by the function:
∆R
∆Q ∆C o ∆SF
= f o ,
,
∆T
∆T
∆T
∆T
where
∆R
= returns from the technology
∆T
∆Qo
= yield
∆T
∆Co
= total cost
∆T
∆SF
= residual benefits
∆T
Residual benefits are a function of the change in the fertility status
of the soil over time and are given by the function:
∆Qt + i
∆SFt + i
= f
∆T
∆T o
where
∆Qt + i
= yield obtained the following year
∆T∆
∆SFt + i
= residual soil fertility
∆To
Residual soil fertility is a result of the improvement of the soil due to
additions of manure. Economic quantification of other benefits related
to residual soil fertility other than the yield obtained is beyond the scope
of this paper.
Net Present Values (NPV), present value of expected future earnings
or benefits (Gittinger, 1982), were calculated for the future stream of
benefits from residual soil fertility. Normally the going interest rate is
used as the discount rate. Most smallholder farmers obtain farming
loans from the Government owned commercial bank, Agribank at 20%
interest rate. A 20% discount rate was therefore used to discount future
benefits into today’s values.
The Profitability of Manure Use on Maize in the Small-holder Sector of Zimbabwe
575
The Benefit Cost Ratio (BCR) was also calculated with the provision
that if the BCR ≥1 then it is profitable to adopt the technology when the
cost and benefit streams are discounted at the opportunity cost of capital
(Gittinger, 1995).
Results
Supplementing manure with mineral fertilizer
Greater benefits were obtained when manure was supplemented with
some mineral fertilizers. Using 5 t ha -1of manure produced a yield
advantage of 84% compared to not using any fertility inputs. The addition
of 20 kg N provided a further 45% yield gain compared to using manure
only (Table 40.1). The maize yield increased at a decreasing rate with
successive additions of mineral fertilizer.
Table 40.1: Marginal rates of return for manure and mineral fertilizer combinations
Variables
0 Manure+
0 fertilizer
Yield (t ha-1)
1.18
Adjusted yield (10%)
1.06
Selling Price(Z$ ton-1)
8500.00
Gross Benefit(Z$)
9027.00
Total Variable Costs (Z$ ha-1)
7877.40
Net Benefit (Z$ ha-1)
1149.60
Rate of Return (%)
15
Marginal Net Benefit(Z$)
NA
Marginal Variable Cost (Z$)
NA
Marginal Rate of Return (MRR) % NA
5 t/ha manure +
0 fertilizer
2.17
1.95
8500.00
16600.50
8745.90
7854.60
90
6705.00
868.60
772%
20kgN ha-1
40kg N ha-1
80kg N ha-1 100kg N ha-1
3.14
3.96
4.59
2.83
3.564
4.13
8500.00 8500.00 8500.00
24021.00 30294.00 35113.50
10555.80 12365.70 17795.40
13465.20 17928.30 17318.20
122
134
105
5322.10 4174.60
622.70
2098.40 2098.40 4196.80
254%
199%
15%
5.03
4.53
8500.00
38479.50
19605.20
18874.30
100
1267.60
2098.40
60%
Note: 1US$ = Z$55
All treatments produced positive net financial benefits including the
no fertility inputs option (Table 40.1). Supplementing 5 t ha-1 of manure
with 40 kg N ha-1 produced the highest rate of return (134% $-1) invested
(Table 40.1). Higher levels of N offered lower returns per dollar invested
compared to 20 and 40 kg N ha-1. The practice of not using any manure
and mineral fertilizer only offered a 15% return on investment. The use
of 5 t ha-1 of manure without any mineral fertilizer increased the rate of
return 6 fold to 90% (Table 40.1), compared to not using any manure
and mineral fertilizer. Use of manure alone offered a marginal rate of
576
Mutiro, K. and Murwira, H.K.
return of more than 300% compared to not using manure. The marginal
rate of return increased to more than 400% by supplementing the 5
t ha-1of manure with 20kg N ha-1. The marginal rate of return declined
with higher levels of mineral fertilizer (Table 40.1).
Marginal net benefit first increased with the first 20 kg of N but
declined with successive additions of N. On the other hand the marginal
variable cost, which is the extra cost incurred by using an additional
bag of mineral fertilizer, remained constant, since it is the price of each
additional bag of fertilizer. On the basis of the Marginal Approach in
evaluating profitable level of input use, the most profitable level of
fertilizer is given where the marginal benefit will be equal to the marginal
cost (Hill, 1990). The most optimum level of N to apply per hectare was
found to be 43 kg N ha-1 .
Comparisons of the effectiveness of pit and heap stored
manure
In the first year of manure application, manure stored anaerobically in
pits produced a 104% yield gain compared to that aerobically stored on
a heap (Figure 40.1). Heap stored manure offered 11 and 88 % yield
gain in the second and third seasons respectively and offered higher
residual fertility compared to storing manure in pits.
Figure 40.1. Residual effects of pit and heap stored manure on maize yield on a sandy
soil in Murewa, 1997/98 to 1999/00 season
12
10
8
6
4
2
0
1997/98
1998/99
control
1999/00
heap
overall 3 year yield
pit
The Profitability of Manure Use on Maize in the Small-holder Sector of Zimbabwe
577
The two manure storage methods all produced positive net financial
benefits in all the three seasons. (Table 40.2). Negative financial returns
were obtained in all the three seasons for the control. In the first season
of application, a rate of return of 219% was obtained from pit stored
manure compared to 70% from heap stored manure. The trend was
reversed in the second season in which manure stored on a heap offered
a 244% return compared to 210% from pit stored manure (Table 40.2).
Investment in the use of heap stored manure provided a more than
200% MRR compared to the control in the first year of application. A
MRR of more than 1900% was obtained from pit stored manure in the
first year of application compared to heap stored manure (Table 40.2).
Table 40.2: Analysis of the profitability of using pit and heap stored manure and residual effects over 3 years at
prices deflated for inflation
Variables
1997/98 Season
Control
Yield (t ha )
0.94
Adjusted yield
(10%)
0.84
Selling Price
(Z$ ton-1)
8500.00
Gross Benefit
(Z$)
7191.00
Total Variable
Costs (Z$ ha-1) 8247.40
Net Benefit
(Z$ ha-1)
-1056.40
Rate of Return
(%)
-13
Marginal Net
Benefit
NA
Marginal Variable
Cost
NA
Marginal Rate of
Return (MRR)
NA
Net Present
Values (NPV) -1056.40
-1
1998/99 Season
1999/2000 Season
Heap
Pit
Control
Heap
Pit
Control
Heap
Pit
2.89
5.88
0.69
3.71
3.34
0.41
3.17
1.69
2.60
5.29
0.62
3.34
3.01
0.37
2.85
1.52
8500.00
8500.00 8500.00
8500.00 8500.00 8500.00 8500.00 8500.00
22108.50 44982.00 5278.50
28381.50 25551.00 3136.50 24250.5012928.50
12982.70 14096.60 8247.40
8247.40 8247.40 8247.40 8247.40 8247.40
9125.80 30885.50 -2968.90 20134.10 17303.60 -5110.90 16003.10 4681.10
70
219
10182.20
585.30
4735.40
5.60
215%
1954%
-36
244
210
-62
194
57
9125.80 30885.50 -2474.00 16778.50 14419.70 -4259.00 13336.00 3901.00
Note: 1US$ = Z$55
Undiscounted overall three year net financial benefits were 17% higher
for pit than for heap stored manure and the benefits increased to 25%
when discounted using a 20% discount rate (Table 40.3). Sensitivity
analysis revealed that the higher the discount rate, the higher the benefits
from pitting manure compared to heap stored manure. Pit stored manure
had a higher BCR, 1.72 compared to 1.49 from heap stored manure.
This further confirms the profitability of pit stored manure compared to
578
Mutiro, K. and Murwira, H.K.
heaping. A 47% increase in all costs will render heap storage unprofitable
with a BCR of less than 1 whereas it will take more than a 77% increase
in overall costs to make pit storage unprofitable. Another sensitivity
analysis on labour revealed that a 200% increase in the price of labour
would make storing manure on heaps unprofitable. It would take a 600%
increase on the current labour prices to make pit storing manure
unprofitable. The break even grain price for heap and pit stored manure
is $3400 and $ 3100 respectively all being equal.
Table 40.3: Overall benefits over 3 years of using pit and heap stored manure on sandy
soils in Murewa
Factor
Total harvest (tonnes)
Total Net Financial Benefit (Z$)
Net Present Values (NPV)
Control
Pit
Heap
1.83
-9136.10
-7789.50
9.82
52870.20
49206.10
8.79
45263.10
39240.20
Note: 1US$ = Z$55
Effect of Manure Placement Methods on Maize Yield
Banding and placing manure on-station produced higher yields for both
pit and heap stored manure compared to the farmer practice of
broadcasting manure (Table 40.4). Pit stored manure yielded more than
heap stored manure in all the different placement methods, banding,
broadcasting and station placement. Banding heap stored manure
yielded more compared to placing it on-station or broadcasting it. Onstation application of pit stored manure marginally out-yielded banding
though it was not statistically significant. Broadcasting manure gave
the least yield compared to the other two methods, (banding and station
placement).
The rate of return for heap stored manure was negative for
broadcasting and station placement. A 6% rate of return was obtained
for banded heap stored manure (Table 40.4). Net financial benefits for
pit stored manure were positive for all the three different placement
methods. Banding and station placement produced more than a 70 %
rate of return while broadcasting offered a 50% return to investment
(Table 40.4). A sensitivity analysis on labour rates revealed that a 50%
increase in labour costs made heap stored manure unprofitable.
Increases in labour rates of more than 100% reduced rates of return for
pit stored manure to less than 20%.
The Profitability of Manure Use on Maize in the Small-holder Sector of Zimbabwe
579
Table 40.4: Marginal rates of return for pit and heap stored manure using different
application methods (banding, on-station and broadcasting)
Variables
Pit, Banded
Heap, BroadcastedHeap, BandedHeap, On-stationPit Broadcasted
Pit On-station
Yield(t ha-1)
Adjusted yield (10%)
Selling Price(Z$ ton-1)
Gross Benefit(Z$)
Total Variable Costs
(Z$ ha-1)
Net Benefit (Z$ ha-1)
Rate of Return (%)
Marginal Net Benefit
(Z$)
Marginal Variable
Cost (Z$)
Marginal Rate of
Return (Z$)
1.01
1.79
1.39
2.76
3.22
3.30
0.91
1.61
1.25
2.48
2.90
2.97
8500.00 8500.00 8500.00 8500.00 8500.00 8500.00
7726.50 13693.50 10633.50 21114.00 24633.00 25245.00
12804.30 12891.20 13291.80 13964.00 14050.80 14451.40
-5077.80 802.30 -2658.30 7150.00 10582.20 10793.60
-40
6
-20
51
75
75
NA
-1574.10
-1564.70
1861.70
1516.10
2031.50
NA
-56.70
85.10
28.40
28.40
28.40
NA
2776%
-1840%
6567%
5348%
7176%
Note: 1US$ = Z$55
Discussion
The use of manure with smaller quantities of mineral fertilizers offers
much larger productivity gains compared to using mineral fertilizer alone
or manure alone. Combinations of mineral fertilizers and manure
generally yield better though responses are very variable across sites
because of the variability of the manure quality and site characteristics
(Murwira et al., 1998). From this study the most optimum level of
supplementing 5 tonnes of manure was 43 kg N ha-1. Sensitivity analysis
revealed that an increase of more than 50% in the price of mineral
fertilizer would make higher rates of supplementation less favourable
with MRR. The variation in responses across different sites makes
blanket recommendations impractical. Recommendations on
supplementing organic materials with mineral fertilizers should be area
specific. Application of inorganic fertilizer with manure can reduce the
risks of economic losses and increase the probability of higher financial
returns. Results from the study indicate that supplementing manure
with mineral fertilizers can significantly increase net financial returns.
The improvement of manure quality through pit storage on the farm
provides a realistic option for improving productivity in the smallholder
sector. For resource constrained households that cannot raise enough
cash to buy mineral fertilizers, pit storage is a technology which makes
it possible for farmers to substitute cash requirements for soil fertility
580
Mutiro, K. and Murwira, H.K.
management with their labour. Despite the evident benefits of storing
manure in pits, more than 65% of smallholder farmers who use manure
in Murewa store their manure on a heap. Extension is silent on how
farmers can improve their manure for effective utilization despite the
research evidence that manure from the smallholder farming sector is
of very poor quality. Most farmers are likely to adopt this technology
given that technologies that offer larger benefits in the first season of
adoption are likely to be adopted than those which yield benefits later
in the project cycle like heaping manure (Gittinger, 1995).
The yield benefits from manure application can be increased for
both heap and pit stored manure if the appropriate method of application
is used. Pit stored manure produced higher rates of returns on all the
different application methods compared with heap stored manure.
Studies done in Zimbabwe have identified banding and station placement
as the most rewarding placement methods (Mubonderi et al., 1999).
Though most farmers in Zimbabwe broadcast their manure, results
from this analysis show that this is not a profitable option especially for
heap stored manure. Farmers can realise greater returns by either
banding or station placement of the manure. However labour
requirements for banding and station placement make these options
unattainable for labour constrained households. Manure application
methods are becoming more targeted in reaction to reduced livestock
numbers and increase in the prices of mineral fertilizers (Snapp et al.,
1997; Ahmed et al., 1997).
Smallholder farmers in Zimbabwe supplement manure with varying
levels of mineral fertilizers but yields are still below potential due to
inadequate amounts, poor quality of organic materials and inefficient
combinations (Murwira and Palm, 1998). To these farmers what is more
critical is how much of mineral fertilizer should supplement the manure
for maximum benefit. A range of combinations that are profitable should
be recommended to farmers given their different resource endowments
and variability of the quality of manure from 0.1 to 1.9% N content
depending on the management and handling of the manure (Nzuma
and Murwira, 2000). Specific decision guides could be developed to
provide farmers with guidelines for supplementing different quality
manures with mineral fertilizers (Figure 40.1). These decision guides
need to be farmer-friendly and take account of available quantities of
manure and fertilizer N, farmer perceptions of quality and other
management factors like soil type and methods of application.
Conclusion
Economic returns are an important determinant of technology use
(CIMMYT, 1988). The options that offer higher economic returns have
been identified in this study and merit further testing with farmers.
The Profitability of Manure Use on Maize in the Small-holder Sector of Zimbabwe
581
Extension can play a major role in expanding further testing and adoption
of these options by farmers. Information on how farmers can realise
maximum returns by supplementing organic materials with mineral
fertilizer seems not available or if available it is scanty yet this option
provides substantial opportunities for increasing productivity in the
smallholder farming sector.
Most farmers find it difficult to raise the capital required for
investments in mineral fertilizer and find it cheaper to invest their labour
than capital. Despite the additional labour requirements of pit storing,
manure farmers can invest their labour and be able to realise returns
of more than 100% from utilizing pit stored manure. Concerns have
been raised about labour availability in the smallholder farming sector
especially considering the high incidences of HIV. Labour shortages are
likely to discourage adoption of pit storing but an in-depth study is
required to ascertain labour availability and its impact on adoption of
labour intensive technologies.
Acknowledgements
We would like to thank IFAD and the Rockefeller Foundation for
sponsoring this work. Acknowledgements also go to N. Nhamo and J.
Nzuma who provided data from their farmer participatory trials on
manure and mineral fertilizers.
References
Ahmed, M.M., Rohrbach, D.D., Gono, L.T., Mazhangara, E.P., Mugwira, L.,
Masendeke, D. and Alibaba, S. (1997) Soil fertility management in communal
areas of Zimbabwe: current practices, constraints and opportunities for
change. Results of a diagnostic survey. Southern Eastern Africa Region
Working Paper no. 6. International Crops Research Institute for the SemiArid Tropics (ICRISAT).
CIMMYT, (1988) From Agronomic data to farmer recommendations: An economic
Training manual. Completely revised edition. CIMMYT, Mexico, D.F.
Gittinger, J.P. (1995) Economic analysis of agricultural projects. The John Hopkins
University Press, Baltimore, Maryland 21218, U.S.A.
Hill, B. (1990) An introduction to economics for students of Agriculture. Pergamon
Press plc, Headington Hill Hall, Oxford, England.
Mubonderi, T.H., Mariga, I.K., Mugwira, L.M., and Chivinge, O.A. (1999) Maize
response to method and rate of manure application. African Crop Science
Journal Vol. 7. No. 4. pp. 407-413.
Mugwira, L. M. (1985) Effects of supplementing communal area manure with
lime and fertilizer on plant growth and nutrient uptake. Zimbabwe
Agricultural Journal 82: 153-159.
582
Mutiro, K. and Murwira, H.K.
Murwira, H.K. and Palm, C.A. (1998) Developing multiple fertilizer use strategies
for smallholder farmers in Southern Africa in CIMMYT and EARO.1999.
Maize Production Technology for the future: challenges and opportunities:
Proceedings of the Sixth Eastern and Southern Africa regional maize
conference, 21-25 September, 1998, Addis Ababa, Ethiopia:
CIMMYT(International Maize and Wheat Improvement Center) and EARO
(Ethiopian Agricultural Research Organisation). Pp 205-209.
Murwira, H.K., Tagwira, F., Chikowo, R and Waddington S.R. (1998) An
evaluation of the agronomic effectiveness of low rates of cattle manure and
combinations of inorganic N in Zimbabwe in CIMMYT and EARO.1999. Maize
Production Technology for the future: challenges and opportunities:
Proceedings of the Sixth Eastern and Southern Africa regional maize
conference, 21-25 September, 1998, Addis Ababa, Ethiopia:
CIMMYT(International Maize and Wheat Improvement Center) and EARO
(Ethiopian Agricultural Research Organisation). pp 179-182.
Nzuma, J.K., and Murwira, H.K. (2000) Improving the management of manure in
Zimbabwe in Managing Africa’s Soils No. 15. Russel Press, Nottingham.
Sanchez, P.A. (1997) Soil fertility replenishment in Africa: An investment in
natural resource capital. In R.J. Buresh et al. (ed), Replenishing Soil Fertility
in Africa. SSSA Special Publication no. 51. SSSA. Madison, USA. pp 1-46.
Palm, C.A; Murwira, H. K and Carter S., E. 1998 Organic matter management:
from science to practice. In: Waddington, S. R., H.K. Murwira, J. D. T.
Kumwenda, D. Hikwa and F. Tagwira (eds) 1998. Soil Fertility Research for
Maize-Based Farming Systems in Malawi and Zimbabwe. Proceedings of
the Soil Fert Net Results and Planning Workshop held from 7 to 11 July
1997 at Africa University, Mutare, Zimbabwe. Soil Fert Net and CIMMYTZimbabwe, Harare, Zimbabwe.pp 21-28.
Smaling, E. M. A., Nandwa, S. M and Janssen. (1997) Soil fertility in Africa is at
stake. In: R.J. Buresh et al. (ed) Replenishing Soil Fertility in Africa. SSSA
Special Publication no. 51. SSSA. Madison, USA. pp 47-62.
Snapp, S., Phiri, R and Moyo, A. (1997) Soil fertility experimentation and
recommendations for drought-prone regions of Zimbabwe and Malawi. In
CIMMYT. 1999. Risk management for maize farmers in drought prone areas
of Southern Africa: Proceedings of a workshop held at Kadoma range,
Zimbabwe, 1 –3 October 1997. Mexico, D. F. pp13-24.
Tanner, P.D and Mugwira, L.M. (1984) Effectiveness of communal area manure
as sources of nutrients for young maize plants. Zim. Agric. J.81:31-35.
Improved Food Production by Use of Soil Fertility Amendment Strategies in the Central
Highlands of Kenya
Improved Food Production
by Use of Soil Fertility
Amendment Strategies in the
Central Highlands of Kenya
583
41
Mucheru, M.1, Mugendi, D.1, Micheni,
A.2, Mugwe, J.3, Kung'u, J.1, Otor, S.1
and Gitari, J.2
1
Kenyatta University, Faculty of Environmental Studies, P.O.
Box 43844, Nairobi, Kenya
2
Kenya Agricultural Research Institute, P.O. Box 27, Embu,
Kenya
3
Kenya Forestry Research Institute, P.O. Box 20412, Nairobi,
Kenya
Abstract
Declining soil and crop productivity is a major problem
facing smallholder farmers in eastern and central highlands
of Kenya. This is caused by continuous cropping without
addition of adequate external soil fertility inputs. A
multidisciplinary and farmers participatory trial is being
implemented in the main maize growing areas of the central
highlands of Kenya to address the above problem. The trial
is farmer-researcher managed with a general expected
output of offering small-scale resource poor farmers feasible
584
Mucheru, M. et al
soil management techniques for combating soil nutrient
depletion. Results for the two seasons reported here indicate
that the general maize performance may be improved by
combining fast decomposing plant biomass (e.g. Tithonia
diversifolia) and half the recommended rate of nitrogen
fertilizer.
Key words: biomass transfer, nitrogen replenishment, N leaching, maize
yield
Introduction
One of the challenges facing Kenya today is the production of adequate
food to feed the rapidly growing population and in particular, the
inhabitants of the densely populated highlands of central Kenya with
over 500 persons km-2 (Government of Kenya, 1994). The soils in this
area are deep, well drained, weathered humic nitisols with moderate to
high inherent fertility (Jaetzold and schmidt, 1983). Over time, the soil
fertility has declined due to continuous cropping with little nutrient
replenishment (Ikombo, 1984) and crop yield decline has been a major
problem facing smallholder farmers in the area. Though high yielding
maize varieties have been developed with yield potentials of 7-12 Mg ha-1,
maize yields at the farm level hardly exceed 1.5 Mg ha-1 (Wokabi, 1994).
The use of inorganic fertilizers is generally low, less than 20 kg N and
10 kg P ha-1 (Muriithi et al., 1994). The amount is inadequate to meet
the crop nutritional requirements for optimum crop yields at the farm
level. Due to the high cost of inorganic fertilizers and low prices of farm
produce, over 80% of the farmers use farmyard manure (FYM) to improve
soil fertility and crop productivity (Maize Data Base Project, 1993). The
usefulness of FYM is limited mainly due to its variability and often-low
nutrient contents and also the large quantities (5–10 Mg ha-1) needed to
supply adequate nutrients (Kihanda, 1996; Nzuma et al., 1998).
Surveys carried out in the area indicate that farmers are fully aware
of the declining soil fertility (as expressed by declining crop yields), but
in most cases they do not have readily available resources to replenish
it (Muriithi et al., 1994). Research work by Mugendi et al. (1999); Mutuo
et al. (1998) and Nziguheba et al. (1998) reported positive results from
use of tithonia, calliandra and leucaena biomass for soil fertility
improvement. These biomass are therefore supplementary components
in soil fertility improvement and needs to be evaluated on-farm by
farmers and other stakeholders in agricultural production processes.
The information reported herein is from a participatory on-farm trial
conducted in the predominantly maize growing zones (UM2– UM3) of
Meru South District. The aim of the evaluation is to provide a menu
Improved Food Production by Use of Soil Fertility Amendment Strategies in the Central
Highlands of Kenya
585
(demonstration) on integrated soil fertility management strategies for
increased agricultural production by smallholder farmers in Central
highlands of Kenya. This project sought to i) integrate nutrient
management practices that will arrest the current nutrient depletion
and increase food production, and ii) encourage farmers to adopt
improved nutrient management practices.
Materials and Methods
Experimental site
The study was carried in Meru South District. According to Jaetzold et al.
(1983), the area is in upper midlands 2 and 3 (UM2-UM3). Coffee and
dairy are the main Land Use Systems (LUS) with an altitude of approximately
1500 m above sea level, annual mean temperature of about 200 C and
annual rainfall of about 1200 mm. The rainfall is bimodal, falling in two
seasons, the long rains (LR) lasting from March through June and short
rains (SR) from October through December. About 65% of the rains come
during the long rainy season. The main food crop is maize.
Experimental layout
The experiment was established in March 2000 on a farm with poor
and impoverished soils and laid out as a randomized complete block
design (RCBD) with 3 replicates. The plot sizes were 6 m x 4.5 m. The
test crop, maize, (Zea mays L, var. H513) was planted at a spacing of
0.75 m and 0.25 m inter- and intra-row, respectively. Three (3) seeds
were sown and thinned at four weeks later to 2 plants per hole to give
an approximate population of 53,300 plants ha -1. Nine external soil
fertility amendment inputs (Table 41.1) were applied to give an equivalent
of 60 kg N ha-1. The tenth was an absolute control (no external input)
representing farmers on the lower end of resource endowment.
Organic materials were applied and incorporated into the soil to a
depth of 15 cm during land preparation just before the onset of the
rains. Nutrients compositions of the applied organic inputs are
represented in Table 41.2. The inorganic source of N and P was the
compound fertilizer (23:23:0) that was applied during maize sowing. All
the agronomic procedures for maize production were appropriately
followed after planting. The plots were hand weeded twice, four weeks
after planting and at maize flowering. Stalk borers were controlled by
use of borericide (Buldock dust) four weeks after the crop emergence.
During the first season a general P deficiency was noted, thus a uniform
top dressing for P, as TSP, was carried out in the second season.
586
Mucheru, M. et al
Table 41.1: Experimental treatments indicating the different soil applied fertility
amendment inputs at Chuka, Kenya
Treatment. No.
Treatment
1
2
3
4
5
6
7
8
9
10
Cattle manure
Tithonia diversifolia
Calliandra calothyrsus
Leucaena leucocephala
Cattle manure + 30 kg N and 30 kg P ha -1
Tithonia + 30 kg N and 30 kg P ha -1
Calliandra +30 kg N and 30 kg P ha -1
Leucaena + 30 kg N and 30 kg P ha -1
60 kg N and 60 kg P ha -1
Absolute control (no inputs)
Table 41.2: Nutrient composition (%) of organic materials inputs applied in the Soil at
Chuka, during season 1 and 2
Treatment
Cattle manure
Tithonia
Calliandra
Leucaena
N
P
Ca
Mg
K
Ash
1.4
3.0
3.3
3.8
0.2
0.2
0.2
0.2
1.0
2.2
0.9
1.4
0.4
0.6
0.4
0.4
1.8
2.9
1.1
1.8
46.1
13.2
5.8
8.7
Sampling and analyses
Maize was harvested at maturity from a net area of 21.0 m2. This was
after leaving out one row on each side of the plot and the first and the
last plants of each row in order to minimize the edge effect. At the end of
the second season soil samples were taken with an Alderman auger.
The soils were sampled from three different depths: 0-30, 30-100 and
100-150 cm. One sub-sample of each soil sample was dried at 1050 C
for 48 hours in order to determine gravimetric water content.
For determination of ammonium and nitrate, about 20 g of field
moist soil was extracted with 100 ml 2 M KCl by shaking for one hour at
150 reciprocation per minute and subsequent gravity filtering with
prewashed whatman paper. Soil water content was determined on the
stored field moist soil at the time of extraction in order to calculate the
dry weight of extracted soil. Ammonium in the extract was determined
by a calorimetric method (Anderson and Ingram, 1993) and nitrate was
determined by cadmium reduction (ICRAF, 1995). Biophysical data was
statistically analyzed using Genstat program (1995).
Improved Food Production by Use of Soil Fertility Amendment Strategies in the Central
Highlands of Kenya
587
Results and Discussions
Maize grain yield
The average maize grain yield across the treatments was 0.9 Mg ha -1
(Table 41.3) during the first season. Application of recommended
inorganic fertilizer (60 kg N and P ha -1) gave the highest maize grain
yield with an average of 1.6 Mg ha -1. Average maize grain yield from
calliandra was lowest (0.2 Mg ha-1) and was worse than the control. The
maize yields in this season (long rains) were not significantly different.
The average maize grain yield was against the expected grain yield of
greater than 6 Mg ha -1 (Var. H513) for the area. The low maize grain
yield in calliandra treatment could be attributed to the lower rate of
decomposition and mineralization due to the high polyphenol and lignin
content of calliandra, which could have resulted to net immobilization
of nutrients. The low soil moisture content resulting from the low rainfall
(126 mm received in the first 20 days of the season) during this season
could have exacerbated the situation. The higher maize grain yield with
the inorganic fertilizer could be due to the readily available N compared
to the N from organic inputs which must first decompose and mineralize
before the N is available to the plant. Rains that stopped very early in
the season could have meant that the organics did not have sufficient
water (moisture) to decompose and mineralize and even if they did,
water was not available for the mineralized nutrients to be taken up by
the plants.
Table 41.3: Maize yields under different soil fertility amendment inputs in Chuka during
season 1 and 2
Treatment
1st season
Grain wt (Mg ha -1)
2nd season
Grain wt (Mg ha -1)
Cattle manure
Tithonia
Calliandra
Leucaena
Cattle manure + 30 kg N & P ha -1
Tithonia + 30 kg N & P ha-1
Calliandra + 30 kg N & P ha -1
Leucaena + 30 kg N & P ha -1
60 kg N & P ha-1
Control
0.7
0.8
0.2
0.9
1.4
1.5
1.1
1.0
1.6
0.3
5.0
5.9
4.0
5.1
5.7
6.2
4.7
5.1
5.4
3.1
Mean
SED
0.9
0.7
5.0
1.2
588
Mucheru, M. et al
The average maize grain yield across all the treatments during the
second season was 5.0 Mg ha-1. During this season tithonia with 30 kg
N and P ha-1gave the highest maize grain yield of 6.2 Mg ha-1. The control
had the lowest maize grain yield (3.1 Mg ha-1). The maize grain yields in
the second season were significantly different (P<0.05) between the
treatments. The better performance of tithonia during this season could
be attributed to the faster release of N and P from the leaf biomass
(Gachengo et al., 1999). The integration of tithonia and mineral fertilizer
had higher maize grain yields than the recommended rate of mineral
fertilizer; this could have been as a result of the provision of additional
benefits (besides N and P) by the tithonia (organic).
The integration of organic and inorganic nutrient sources of N gave
higher maize grain yields as compared to the sole application of organic
materials in both seasons. These results concur with results by Gachengo
(1996), Kihanda (1996), Mutuo et al (1998), Nziguheba et al. (1998) and
Mugendi et al. (1999) on the integration of organic and inorganic soil
fertility inputs. Integration of inorganic and organic nutrient inputs can
be considered as a better option in increasing fertilizer use efficiency
and providing a more balanced supply of nutrients (Donovan and Casey,
1998). Kapkiyai et al. (1998) reported that combination of organic and
inorganic nutrient sources has been shown to result into synergy and
improved synchronization of nutrient release and uptake by plants
(leading to higher yields).
The low maize grain yields in the first season could be associated
with the very low precipitation (average 126 mm) with most of it being
recorded within the first three weeks of the season. This low precipitation
could have reduced the availability of nutrients to the maize plants.
However, the second season was characterized by high precipitation
(average 698 mm) occurring throughout the season and this could have
led to the higher maize grain yield.
Residual Mineral N
Leucaena with 30 kg N and P ha -1 had the highest (97.7 kg N ha -1)
residual mineral N while recommended level of inorganic fertilizer had
the lowest (51.4 kg N ha-1) residual mineral N at 0-30 cm depth. There
was a significant difference (P<0.05) of mineral residual N between
treatments at the end of the second season (Table 41.4). All treatments
were higher in residual mineral N than the control with the exception of
the recommended level of inorganic fertilizer. This trend could be
attributed to the beneficial ability of the soil incorporated biomass to
release N gradually (especially in the present scenario where they were
soil incorporated when dry) unlike inorganic fertilizers which release N
drastically after application making them have a low residual effect. No
Improved Food Production by Use of Soil Fertility Amendment Strategies in the Central
Highlands of Kenya
589
significant differences were observed for ammonium between the
treatments at 0-30 cm depth but significant differences were observed
in the other depths. Nitrate concentration was significantly different
(P<0.05) between the treatments as well as depth. Lower levels of
ammonia-N were noted in comparison with nitrate-N in all the
treatments. This could be as a result of the rapid conversion of ammonia
to nitrate following mineralization of inputs in the soil.
Table 41.4: Treatment effects on soil residual mineral N (kg ha -1) at various soil depths
at Chuka at the end of the second season
Treatment
0-30 cm
30-100 cm
100-150 cm
Cattle manure
Tithonia
Calliandra
Leucaena
Cattle manure + 30 kg N & P ha -1
Tithonia + 30 kg N & P ha-1
Calliandra + 30 kg N & P ha -1
Leucaena + 30 kg N & P ha -1
60 kg N & P ha-1
Control
77.7
67.8
64.1
77.5
64.7
77.7
87.1
97.7
51.4
51.7
143.4
282.8
104.4
70.6
131.9
207.2
56.4
161.5
38.2
94.2
43.2
44.1
84.2
87
49.8
95.8
Mean
SED
70.8
10.6
144.8
28.5
67.1
15.6
The average concentration of mineral N was highest (144.8 kg N
ha-1) in the 30-100 cm soil depth and lowest (67.1 kg N ha-1) in the 100150 cm soil depth. Mineral N concentration was lower in the 100-150
cm depth in all the treatments. Cattle manure had the least concentration
(38.2 kg N ha -1 ) in the 100-150 cm depth (Table 41.4) while the
recommended level of fertilizer had the least concentration in both the 0-30
cm and 30-100 cm depth with 51.4 kg N ha-1 and 70.6 kg N ha-1 respectively.
The mineral N in the 100-150 cm soil depth is below the rooting zone of
maize plants and may not be available to the maize plants (Mugendi et
al., 2000). It may also not be readily transformed (denitrified or
assimilated) because of the limited microbial population and available
C at this depth (Paramasivam et al., 1999). This mineral N is therefore
prone to leaching into ground water therefore careful management of
soil fertility inputs like timely application and split fertilizer application
that have been reported to reduce N leaching should be encouraged
(Paramasivam et al., 2001).
A bulge in nitrate occurred at 30-100 cm in all the treatments. This
concurs with Kindu et al. (1997) who reported a bulge in nitrate at 0.3 to
590
Mucheru, M. et al
1.5 m soil depth in the maize land use system in Western Kenya. The
bulge (accumulation) in nitrate at this depth could be attributed to greater
N mineralization compared to plant uptake of top soil N immediately
after the onset of the rainy season, subsequent nitrate leaching and then
adsorption of nitrate on positively charged soil surfaces. Hartemink et al.
(1996), also working in the land use systems of western Kenya reported
that about 60% of nitrate at 1 to 2 m depth was sorbed on soil surfaces;
this sorption of nitrate is known to delay its downward movement resulting
in nitrate accumulation in the subsoil (Wong et al., 1987).
Farmers’ perception
Two farmers’ field days were held during the two seasons. The farmers
were impressed with what they saw in the field and they were willing to
experiment with these technologies especially with tithonia, which grows
locally along the boundaries and roadsides.
Conclusion
The high maize grain yields in the second season demonstrate the positive
impact of these technologies in the area. Tithonia with half recommended
rate of inorganic fertilizer gave impressive yields and hence the farmers
are encouraged to adopt it on their farms to improve their food security.
The bulge in nitrate at 30-100 cm depth indicates that there is nitrogen
leaching in the soil and this calls for action.
Acknowledgements
The authors wish to thank the Rockefeller Foundation for providing
financial support for the field experimentation. They also appreciate
the contribution and collaborative efforts by the Kenya Agricultural
Research Institute (KARI), Kenya Forestry Research Institute (KEFRI)
and Kenyatta University (Faculty of Environmental Studies) in
administering field activities. Lastly, the cooperation and contributions
by researchers and Kirege community members (farmers, administration
and agricultural extension staff) are acknowledged.
References
Anderson, J.M. and Ingram, J.S.L. (1993) Tropical Soil Biology and Fertility: A
Handbook of Methods. CAB International, Wallingford, UK.
Donovan, G. and Casey (1998) Soil fertility management in Sub- Saharan Africa.
World Bank. Technical Paper. 408 Pp
Improved Food Production by Use of Soil Fertility Amendment Strategies in the Central
Highlands of Kenya
591
Gachengo, C. (1996) Phosphorus release and availability on addition of organic
materials to Phosphorus fixing soils. MPhil thesis, Moi University, Eldoret,
Kenya.
Gachengo, C.N., Palm, C.A., Jama, B. and Otieno, C. (1999) Tithonia and senna
green Manures and inorganic fertilizers as phosphorus sources for maize
in Western Kenya. Agrof. Syst. 44:21-26.
Genstat (1995) Genstat Release 3.2 for Windows 95. Lawes Agricultural Trust,
Rothamstead Experimental Station, UK.
Government of Kenya (1994) The 1989 Kenya National Census results. Ministry
of Home Affairs, Nairobi, Kenya.
Hartemink, A.E., Buresh, R.J., Bashir, J. and Janssen, B.H. (1996) Soil nitrate
and water dynamics in sesbania fallow, weed fallow, and maize. Soil Sci.
Soc. Am. J. 60: 568-574.
ICRAF (1995) Laboratory Methods for Soil and Plant Analysis: Version 1.1 Nairobi.
Ikombo, B.M. (1984) Effects of farmyard manure and fertilizers on maize in
semi-arid areas of Eastern Kenya. East. Afr. Agric. for. J. 44:266-274.
Jaetzold, R. and Schemidt, H. (1983) Farm management handbook of Kenya.
Natural conditions and farm information. Vol.11/ C. East Kenya. Ministry
of Agriculture, Kenya.
Kapkiyai, J.J., Karanja, N.K., Woomer, P., and Qureshi, J.N. (1998) Soil organic
carbon fractions in a long-term experiment and the potential for their use
as a diagnostic assays in highland farming systems of central Kenya. African
Crop Science Journal 6: 19-28.
Kihanda, F.M. (1996) The role of farmyard manure in improving maize production
in the sub-humid highlands of central Kenya. PhD Thesis. The University
of Reading. UK.
Kindu, M., Buresh, R.J. and Bashir, J. (1997) Root and inorganic nitrogen
distributions in sesbania fallow, natural fallow and maize fields. Plant and
Soil 188: 319-327.
Maize data base project (1993) Preliminary results for the Embu area presented
in workshop held in Nairobi, Kenya, September 23-24, 1993.
Mugendi, D.N., Nair, P.K.R., Mugwe, J.N., O’Neill, M.K. and Woomer, P.L. (1999)
Calliandra and Leucaena alley cropped with maize. Part 1: Soil fertility
changes and maize production in the sub humid highlands of Kenya. Agrof.
Syst. 46: 39-50.
Mugendi, D.N., Kanyi, M.K., Mugwe, J.N., Kung’u, J.B. and Wamicha, W. (2000)
Mineral N movement and management in the Agroforestry systems in the
central highlands of Kenya. Paper submitted to the 18th conference of the
Soil Science Society of East Africa at Mombasa Kenya.
Muriithi, F.M., Thijssen, H.J.C., Mugendi, D.N., Mwangi, J.N., O’Neill, M.K. and
Nyaata, O.Z. (1994) Report of a survey on agroforestry technologies used
for fodder production and soil fertility improvement in Meru district, Kenya.
National Agroforestry Research Project, Regional Research Centre, Embu,
Kenya
Mutuo, P.K., Mukalama, J.P. and Agunda, J. (1998) On-farm testing of organic
and inorganic phosphorous source on maize in Western Kenya. In: The
Biology and Fertility of Tropical Soils: TSBF Report, pp22.
592
Mucheru, M. et al
Nziguheba, G. and Mutuo, P.K. (1998) Integration of Tithonia diversifolia and
inorganic fertilizer for maize Production. In: The Biology and Fertility of
Tropical Soils: TSBF Report, pp 23.
Nzuma, J.K., Murwira, H.K. and Mpepereki, S. (1998) Cattle manure
management options for reduction nutrient losses. Farmer perceptions and
solutions in Mangwada, Zimbabwe. In: Soil Fertility Research for maize
based system in Malawi and Zimbabwe. CIMMYT, Soil Fertility Network.
Paramasivam, S., Alva, A.K., Prakash, O. and Cui, S.L. (1999) Denitrification in
the vadose zone and in surficial groundwater of a sandy entisol with citrus
production. Plant Soil 208: 307-319.
Paramasivam, S., Alva, A.K., Fares, A. and Sanjwan, K.S. (2001) Estimation of
Nitrate Leaching in an Entisol under Optimum Citrus Production. Soil Sci.
Soc Am. J. 65: 914-921.
Wokabi, S.M. (1994) Quantified land evaluation for maize yield gap analysis at
three sites in the Eastern slopes of Mt. Kenya. International Institute for
Aerospace Survey and Earth Science (ITC), The Netherlands.
Wong, M.T.F., Wild, A., and Juo, A.S.R. (1987) Retarded leaching of nitrate
measured in Monolith lysimeters in southeast Nigeria. J. Soil Sci. 38: 511518.
593
Impact of Adopting Soil Conservation Practices on Wheat Yield in Lesotho
Impact of Adopting Soil
Conservation Practices on
Wheat Yield in Lesotho
42
Kaliba, A.R.M.1 and Rabele, T.2
1
Aquaculture/Fisheries Center, University of Arkansas at
Pine Bluff, 1200 North University, Pine Bluff AR, USA
2
Department of Economics, National University of Lesotho,
P.O. Roma, 180, Lesotho, Tel: 266 12 3593; Fax: 266 12
340000, E-mail: akaliba@uaex.edu and trabele@yahoo.com
Abstract
This study assesses the impact of adopting soil conservation
practices on wheat yield in Lesotho. The study uses inputoutput data collected from 50 smallholder farmers in
Mafeteng and Maseru districts, the major wheat growing
areas in the country. A system of equations was used to
estimate factors affecting adoption of soil conservation
measures, and impact of soil conservation measures on
wheat yield. For each farmer’s wheat field, based on the
soil conservation practices adopted by the farmer, two soil
conservation variables related to farmer’s soils erosion
control methods were constructed. Two Tobit models and a
modified Cobb-Douglas production function were used to
model adoption, and impact of, soil conservation measures
respectively. The adoption of two soil conservation measures
was modeled as function of household’s demographic
594
Kaliba, A.R.M. and Rabele, T.
characteristics and availability of extension services. The
yield equation was modeled as a function of inputs used in
production and soil conservation efforts. The results indicate
that soil conservation efforts were superior to inorganic
fertilizer application in terms of increasing wheat yield.
Increase in soil conservation efforts, coupled with low
inorganic fertilizer use has a potential of increasing wheat
production among smallholder farmers in the area.
Key words: Adoption, Impact, Improved Wheat Varieties, Lesotho, Soil
Conservation
Introduction
Information on the causes and effects of soil erosion on Lesotho’s
agricultural productivity is quite limited, despite considerable erosion
being visible in many parts of the country. It is evident that soil loss
and land degradation have escalated in recent years with a consequence
of decreasing agricultural productivity in the country. Coupled with
Lesotho’s topographical and climatic variation, soil erosion is severe in
most parts of the country. Only 13% of land is arable for growing crops
and in recent years, soil erosion has reduced this to 9% (LMRG, 1996).
Because of high human population in the lowlands, even with a moderate
livestock population, extensive overgrazing, soil erosion and rapid
deterioration of water and soil resources occur at an alarming rate.
Everywhere, the plateau and hill slopes are marked by signs of heavy
erosion by water and wind, leaving behind bare bedrock, laterite hard
pans and stony soils lacking in organic matter.
According to the Kingdom of Lesotho’s report to the United Nation
conference on environment and development (1980), about 54% of
cropland and 28% of those in the mountains are subject to severe sheet
erosion (KoL, 1980). Further, about 40% of the cultivated area should
ideally be under fodder or pasture. Soils are low in organic matter; yields
are low and decreasing, and cultivated area is diminishing. About 5060% of rangelands show severe soil erosion and degradation.
Significantly, in quantitative terms, soil losses per year amount to 15
million tonnes from croplands and 23 million from rangelands, and 1
million tonnes from gully erosion (LMRG, 1996).
To increase agricultural productivity, soil conservation is of
paramount importance to farmers involved in crop production.
Replenishment of soil fertility by artificial fertilizer would be very
expensive. Even then, the use of inorganic fertilizers does not
compensate for the loss of soil organic matter. This study assesses the
potential impact of adopting soil conservation practices on wheat yield.
Impact of Adopting Soil Conservation Practices on Wheat Yield in Lesotho
595
The focus is on conservation measures advocated by the extension
department of the ministry of agriculture co-operatives and land
reclamation. These farming practices include contour farming, crop
residue management for the improvement of infiltration and planting
of cover crops like fodder to increase land productivity. Structural
methods include cut-off drains or diversion to cater for excessive runoff, terracing within cropland for reduction of steepness and length of
slope, waterways to improve water disposal from the terraces and cutoff drains, and gully treatment by vegetative methods and the
construction of silt traps to retard water flow (especially on the
catchments area) (MoA, 1988).
A system of equations is used to estimate factors affecting adoption
of short-term and long-term soil conservation measures and impact of
adopted soil conservation measure on wheat yield. This study deviate
from the past studies that assess the impact of soil conservation
measures on crop yield in two folds. First, the methodology allows using
the whole samples. Past studies that used systems of equations to
estimate adoption, and the impact of soil conservation measure, dropped
out non-adopters in the yield equation to take into account the sample
selection bias. Second, the model takes into account the zero values in
the explanatory variables by using the Battese’s modified Cobb-Douglas
hence BMCD production function. The basic assumption is that farmers
who did not use some of the inputs should have different level of yield
(thus intercept) from those farmers who used the corresponding inputs.
This is important when estimating a system of equations that involve
adopters and non-adopters. The modified production function
incorporates both intercept and divergent shifts in the model.
Literature Review
Feder, Just and Zilberman (1985) define adoption as the degree of use of
a new technology in long run equilibrium when a farmer has full
information about the new technology and its potential. Therefore,
adoption at the farm level describes the realization of farmer’s decision to
apply a new technology in the production process. On the other hand,
aggregate adoption is defined as the process of spread or diffusion of a
new technology within the region. Therefore, a distinction exists between
adoption at the individual farm level and aggregate adoption within the
targeted region. Adoption at the farm level is often quantified or represented
by a binary variable (adoption = 1, non-adoption = 0). In the case of a
divisible technology, a continuous variable describing the intensity of
adoption (e.g., hectares devoted to a new technology or number of livestock
under a new treatment) or extent of adoption (e.g., share of land devoted
to a new technology) are used. Other researchers suggest other variables
which include the earliness of adoption; the time the technology was first
596
Kaliba, A.R.M. and Rabele, T.
used by the farmer; the thoroughness of adoption; the number of technical
components adopted by farmer from the recommended package and an
index of innovativeness which aggregate the adoption dimensions
mentioned above. In most cases, the adoption response depends on the
problem at hand, study objective, available data, and sometimes the
available computer package (Feder, Just and Zilberman, 1985).
A particular technology is adopted when the anticipated utility from
it exceeds that of non-adoption (Rahm and Huffman, 1984). Since utility
is not observable, change in utility can be inferred from farmers’ decision
of adopting or not adopting a technology (incidence of adoption) or
adopting some continuous choice over a predefined interval (Kazianga
and Masters, 2001). When assessing the incidence of adoption implies
using Probit models (Maddala, 1992). To consider intensity or extent of
adoption involves using Tobit models as in Baidu-Forson (1995).
Sometimes, a two stage procedure is used to model adoption when it is
observed that adoption of one innovation leads to adoption of another
complementary farming techniques (Nkonya, et al., 1997; Kaliba, et al.,
1999).
Assessing adoption of soil conservation measures, involves defining
the soil conservation variable or index, which varies from researcher to
researcher. The tendency is to choose one specific soil conservation
measure as an indicator of adoption. For example, see Shively (1999)
on adoption and impact of contour hedgerows in the Philippines for the
case of Probit model, and Kazianga and Masters (2001) on adoption
and impact of field bunds and micro-catchments in Burkina Faso for
the case of Tobit Model. The difficulty arise when there are several soil
conservation measures or a package of soil conservation innovations.
Nowak (1987) used the ratio of adopted practices related to total numbers
available in the package to define the soil conservation variable for each
farmer. This approach is simple because it involves just observing soil
conservation practices adopted by the farmer. Kastens and Dhuyvetter
(1999) used the ratio of total farm costs relative to the costs of adopted
soil conservation measures. This procedure evaluates all inputs used
in production at market prices and requires proper record keeping. This
study adopted the former approach due to its simplicity and
unavailability of price data on all input used in production.
From the literature, factors influencing adoption of new agricultural
innovation can be divided into three major categories: farm and farmers’
associated attributes, attributes associated with the technology (Adesina
et al., 1992) and the farming objective (CIMMYT, 1988; Ockwell, 1991).
In the first category, factors discussed in the literature include human
capital represented by the level of education of the farmer (Rahm and
Huffman, 1985; Goodwin and Schroder, 1994), the risk and risk
management strategies (Saha and Love, 1994), the institutional support
system, such as marketing facilities, research and extension services,
Impact of Adopting Soil Conservation Practices on Wheat Yield in Lesotho
597
transportation etc (Feder, Just and Zilberman, 1985), availability of
production factors and factor endowment such as farm size, number of
livestock owned (Rahm and Huffman, 1984), and the level of off-farm
income and income sources (Kimhi, 1994). The second and third categories
depend on the type of technology and are important when farmers have
access to different types of technical innovation e.g., different type of
crop varieties with differences in production characteristics and
performance, or when dealing with heterogeneous farmers with different
farming objectives e.g., small and larger farmers, subsistence and market
oriented farmers. Therefore, the influence of each exogenous variable on
adoption responses is unique and specific to the study area. These
characteristics make adoption studies site specific and often incomparable.
Two techniques are commonly used to study the impact of soil
conservation practices on crop productivity. The first approach is the
crop modeling system, which uses several ecosystem factors to evaluate
the dynamics of crops and soils processes that include soil conservation
measures. The technique requires a lot of data and experience in system
modeling (Cox, Hammer and Robertson, 2001). The second approach
uses abstract models such as a production function based on relatively
few variables to relate production to soil conservation activities (Kazianga
and Masters, 2001). This technique concentrates on modeling the response
of crop growth relative to several exogenous variables that ensure the
survival and growth of the crop. The commonly used production functions
are the Cobb-Douglas and the translog. The unrestricted translog
production function is sometimes preferred because it is general and
flexible and allows analysis of interaction of variables (Byiringiro and
Reardon, 1996). The Cobb-Douglas is a special case of a translog function,
when the interaction terms have zero coefficients (Gujarti, 1995 ). Unlike
the Cobb-Douglas, the translog function does not always generate
elasticities of substitution of one, and the isoquant and marginal products
derived from the translog depend on the coefficients on the interaction
terms. However, under low-input agriculture, most smallholder farmers
produce on the increasing side of the production function, and the translog
production function may not represent an actual data generating process.
Whereas the translog functions are superior when the objective is to
calculate the optimal mix of inputs, Cobb-Douglas function often behaves
better when the objective is tracing the production frontier under low
input agriculture (Shively, 1998).
Empirical Model and Estimation Procedure
To rationalize the model, consider a farmer who chose to adopt some or
all soil conservation measures as advocated by the Extension Department
in Leshoto. The farmer has two choices: to adopt innovations from shortterm soil conservation package (i.e., contour ploughing, crop residue
598
Kaliba, A.R.M. and Rabele, T.
management and cover crops); and /or from long term soil conservation
package (i.e., structural methods that include cut-off drains, terracing,
gully treatment, and slit traps). Let Aij represent percent of innovations
adopted from any package (extent of adoption) such that Aij=0 for nonadopter, and (0<Aij100) for adopters. Also, a binary variable dij (incidence
of adoption) can be created such that dij=0 for non-adopter, and dij=1
for adopter. Formally, this relationship can be presented as follows:
2
∑A
j =1
*
ij
= ΒTj Zij + εij ,
1
dij =
0
if
d ij*
if
d ij*
2
∑d
j= 1
*
ij
= θ Tj X ij + ν ij ,
(1)
> 0
,
= 0
Aij = d ij Aij*
(i = 1, 2 ,....., T; j = 1, 2).
In the equations, A* ij is the latent variable representing extent of
adoption and is generated by the classical linear regression, d*ij is the
latent variable representing incidence of adoption and is generated by
the classical Probit regression, j and θ j are parameters of the models,
superscript T is the transpose function, matrices Z ij and X ij contains
variables associated with adoption such that matrix X is contained in
matrix Z, and eij and vij are random errors. The basic assumption is that
the farmer takes a two-step decision process. First, the farmer decides
either to adopt or not to adopt any soil conservation innovation. Second,
if the farmer decides to adopt, a decision is also made on the number of
innovations from the advocated technical innovations from the package.
The system represents simultaneous double-bounded Tobit equations,
were the lower limit is zero and the upper limit is 100. In the system
(j=1) represent extent of adopting short-term soil conservation measures,
and (j=2) represent the extent of adopting long-term soil conservation
measures. As shown by Shonkwiler and Yen (1999), the unconditional
mean of Aij in Equation (1) can be also represented as:
Aij = Φ(θ Tj X ij )(β Tj Zij ) + δ ijφ (θ Tj X ij ) + ηij ,
(η ij ~ N (0, σ u j )),
(2)
Such that :
E ( Aij | Z ij , X ij ) = Φ(θ j X ij )( β j Zij ) + δ ijφ (θ j X ij ),
T
T
T
were Φ (.) and φ (.)are cumulative distribution function and univariate
standard normal probability density function, E(.) is the expectation
operator and η ij is the identically normally distributed error term. Notice
that in the adoption equations, previous decision to adopt some of the
long-term soil conservation measures may induce a farmer to adopt
specific measures from the short-term soil conservation package and
vise versa.
Impact of Adopting Soil Conservation Practices on Wheat Yield in Lesotho
599
To evaluate the impact of adopting soil conservation measures on
wheat yield, consider a Cobb-Douglas production function that relates
agricultural input and soil conservation measures to yield. The function
accounts for the fact that expected yield depend on inputs used in
production and current or past decisions to adopt soil conservation
measures. If Yi represents the yield observed on plot i, the corresponding
Cobb-Douglas production function is:
Yi = α 0 M i Aiα1 21 Aiα222 µ ,
µ
~
N (0, σ µ ).
( 3)
In Equation (3), α ‘s are parameters to be estimated, M is the matrix
of production inputs used to produce wheat on plot i, and other variables
are as explained before. Using information contained in Equations (1)
to (3), the adoption and yield equations can be formulated as follows:
Ai1 = Φ(θ1T X i )[β 10 + β 2 Ai1 + β 11z1 + β 21 z2 + β 31 z3 + β 41z4 + β 51 z5 + β 61z6 + β 71 z7 ] + δ1φ (θiT X i ) + η i1
(4)
Ai 2 = Φ (θ 2T X i )[ β 20 + β 1 Ai2 + β 12 z1 + β 22 z2 + β 32 z3 + β 42 z4 + β 52 z5 + β 62 z6 + β 72 z7 ] + δ 2φ (θ iT X i ) + ηi 2
LY1 = α 0 + α 1 LnLi + α 2 LnFi + α 3 LnAi1 + α 4 LnAi2 + α Hi + (α 0 − λ ) DFi + µi
In the adoption equations, z 1 is sex of the household head (z1=1, if
respondent is male; z2=0, otherwise), z2 is age of respondent in years, z3
is education of the respondent in years, z4 is the number of adults in
the households, and z5 is the estimated monthly income of the respondent
in Lesotho’s Maluti. Other variables were defined as: z6 the experience
of the farmer measured as years in growing wheat, and z7 is a variable
representing availability of extension services to the farmer. The last
but one item, φ (.), is known as the correction factor,, and η ij are random
errors. The extension service variable was constructed as in Kaliba et
al. (2000).
In the yield equation, Ln is the natural logarithm function, Yi is yield
for plot in bags/acre (one bag is 90 kg), Li is labor used in production in
mandays equivalents (family and hired labor was combined together
because few farmers used hired labor). Other variables are: F the quantity
of NPK (3:2:1) fertilizer used per plot in 25kg bags; Ai1 and Ai2 are extent
of adopting short and long terms soil conservation measures; Hi the
dummy variable representing hybrid wheat varieties (H=1 if used hybrid
seeds, H=0 otherwise); and DFk is dummy variable introduced to capture
the influence of non use of fertilizer as suggested by Battese (1997), and
ìi is the identically normally distributed error term. The dummy variable
is such that: DFk=1 if the farmer did not use any fertilizer, DFk=0 if the
farmers reported the use of fertilizer. However, the zeros (non-use of
fertilizer) in the fertilizer variable (Fi) are replaced by ones for the model to
be identified. The important assumption of the MCD model is that farmers
who did not use any fertilizer have different intercept from those who
used fertilizer. This assumption is true if the parameter λ is statistically
different from zero. The use of manure and other organic fertilizer are
very limited as animals usually stay away from the cropland.
600
Kaliba, A.R.M. and Rabele, T.
The inclusion of adoption variables as independent variables
introduces endogeneity and contemporaneous correlation problem in
the model (i.e., cov( η ij, µi) ~ 0). Zero observations in the fertilizer variable
imply that users and non-users have different intercepts. In order to
increase the efficiency of the estimated parameters and to correct for
correlation between errors, Equation 4 was estimated in a two-step
procedure. First, the estimates of θ j were obtained using maximum
likelihood probit (Maddala, 1992) where the dependent variables were
the binary outcome of dij=1 and dij=0 for each j but without including
corresponding Aji as independent variable in each adoption equation.
Second, the results were used to estimate Φ (θ jXij) and (θ jXij) in Equation
(2). The estimated of β , α and δ in Equation (4) were estimated using
nonlinear seemingly unrelated regression (SHAZAM, 1997) as suggested
by (Shonkwiler and Yen, 1999). Table 42.1 list the variables included in
the model, expected signs and reasons.
Source of data
This study uses cross-sectional data collected through a survey using a
structured questionnaire. The survey covered a sample of 50 smallholder
farmers selected randomly from Maseru (25) and Mafeteng (25) districts.
The districts are the main wheat growing regions and are easily accessible
from the National University of Lesotho. The sample size took into
consideration the budget constraint. The data collected were on inputs
used in production, wheat varieties grown and the demographic
characteristics of the respondent. Soil conservation measures for each
farm field were determined by observation. A soil conservation variable
was then developed based on the number of soil conservation practices
adopted by the farmer out of the soil conservation package as advocated
by extension agents working within the area. The major respondent
was the household head.
Results and Discussion
Summary statistics of the variables
Table 42.2 presents summary statistics of the variables used in the
model. On average, every respondent farmed nearly five acres of wheat
with a standard deviation of about 4.91. The total harvest was roughly
33 bags of wheat per plot (6.6 bags/acre). The respondents used about
9.5 man-days equivalent (about 77.5 hours) to complete all field activities
601
Impact of Adopting Soil Conservation Practices on Wheat Yield in Lesotho
involving wheat production. This included family and hired labor. About
6 bags of NPK fertilizer were applied in the five acres plot. About 48% of
respondents were growing hybrid varieties and about 16% of respondents
did not use inorganic fertilizer in their wheat field. On average, the
farmers adopted four measures of soil conservation out of the package
with nine recommendations (see also Table 42.3). Whereas few
respondents have no formal education, about 38% of respondents
indicated that an extension agent to discuss wheat production has never
visited them. On average, a respondent had attended a two-day formal
training on wheat production. These formal training included seminars,
workshops and attending field days organized by the extension services
department, or any other non-governmental organization involved in
agricultural development.
Table 42.1: Exogenous variables included in the model, expected signs and justification
Variable
Expected
sign
Justification
Sex
+
Male headed households have more resources
and are more likely to adopt new innovations than
female headed households
Age
+
Older farmers have more resource than younger
farmer and are more likely to adopt new
innovations
Education
+
Educated farmers are best farmers as they know
the benefits of soil conservation
Number of adults in
households
+
Availability of adult labor increase the ability to
adhere to all important agronomic practices
Income
+
High income avails necessary inputs for better
farming methods such as soil conservation
Experience in farming
+
Farmer’s experience increase the likelihood of
understanding the benefits of soil conservation
Extension services
+
Extension services
performance
+
Availability of labor improve crop management
+
Fertilizers increase soil nutrient and crop growth
Soil conservation
measures
+
Conservation improves soil structure and texture
and thus yields
Hybrids
+
Hybrid are high yielding than local varieties
Adoption Equations
Yield Equation
Labor (mandays
equivalent)
Quantity of fertilizer
increase
productive
602
Kaliba, A.R.M. and Rabele, T.
Table 42.2: Summary statistics of variables used in the regression analysis
Variable
Unit
Mean/
Percent
STD
Deviation
Total production per plot
Size of the plot
Total Labor used per plot
Fertilizers (NPK:3:2:1)
Age of household head
Experience of the farmers
Education of household head
Monthly income
Training in wheat production
Extent of short-term soil conservation
measures adopted
Extent of long-term soil conservation
measures adopted
Sex of household head: (male)
Farmers growing hybrid varieties
Extension visits: Always
Sometime
None
Bags/kg
Acres
Man days
25 kg bag
Years
Years
Years
Maluti
Days
32.62
4.91
9.68
6.28
39.96
15.78
4.75
670.76
2.34
26.36
3.51
7.03
4.74
5.59
6.89
10.56
890.54
4.79
%
0.40
0.18
%
%
%
%
%
%
0.40
76.00
48.00
30.00
32.00
38.00
0.18
Table 42.3: Percentage of farmers adopting soil conservation practices
Soil conservation variables
Long-term Soil Conservation Measures
Terraces
Silt traps
Water ways
Sandbags
Short-term Soil Conservation Measures
Crop rotation
Inter-planting
Fallowing
Contour ploughing
Vegetation cover
%
11
4
15
2
24
2
16
12
13
Table 42.3 indicates the types of soil conservation variables adopted
by different respondents in their wheat fields. From this table, the most
common measure adopted by the farmers is crop rotation, (24% of
respondents). Other popular soil conservation measures were fallowing
(16%), construction of waterways (15%), vegetable cover (13%) and
contour farming (12%). Sandbag construction and interplanting were
the least common among the respondents. However, all respondents
have adopted at least one soil conservation measure in their wheat fields.
603
Impact of Adopting Soil Conservation Practices on Wheat Yield in Lesotho
Regression Results
Table 42.4 presents the results on factors affecting extent of adopting
short-term soil conservation measures. During the analysis, the lower
limit for the Tobit mode was set at 0 and the upper limit 100. The
likelihood ratio statistics for the null hypothesis that all parameters in
the model are zero was rejected at 1% probability level, meaning that
variables included in the model explain some of the variation in extent
of adopting short-term soil conservation measures. Positive and
negative signs on the exogenous variable indicate that the variable’s
marginal effect on short-term soil conservation measures were positive
(increasing extent of adoption) or negative (decreasing extent of
adoption).
Table 42.4: Factors affecting adoption of short-term soil conservation measures
Variable name
Estimated
coefficient
Asymptotic
T-Ratio
Constant
Long-term soil conservation measures index
Sex of household head (Male=1,0 otherwise)
Age of household head in years
Education of household head in years
Number of adults in the households (> 18 years)
Household monthly income in maruti
Experience of growing wheat in years
Availability of extension services variable
Correction factor ( 1)
δ
R-square (%)
Log of Likelihood ratio test
-0.2478
1.4123
-0.1081
0.0418
0.1078
-0.0303
-0.0750
0.0012
-0.0116
1.5487
59.4**
88.0313**
-0.6992
2.2450**
-0.8986
1.0119
2.1568**
-2.3481**
-2.2271**
-0.4000
-1.1016
2.6655**
Double and single asterisks denote statistically significance at 5% and 10% level
The statistically significant variables and variables with positive
influence were adoption of long-term soil conservation measures,
education of household head, and the correction factor. Sex, age, and
experience of the household head and availability of extension services
have no influence on the extent of adopting short-term soil conservation
measures as anticipated. Other statistically significant variables but
with unexpected negative influence were number of adults in the
households and household monthly income.
The results of Tobit models that examine factors affecting the adoption
of long-term soil conservation measures are presented in Table 42.5.
604
Kaliba, A.R.M. and Rabele, T.
Again, positive and negative signs on the exogenous variables indicate
that higher values of the variables will increase or decrease adoption of
long-term soil conservation measures. The likelihood ratio test statistics
was significant at 1% probability level. The statistically significant variables
included extent of adopted short-term soil conservation measures, number
of adults in the households and household monthly income.
Table 42.5: Factors affecting adoption of Long-term soil conservation measures
Variable names
Estimated
coefficient
Asymptotic
T-Ratio
Constant
Short-term soil conservation measures index
Sex of household head (Male=1,0 otherwise)
Age of household head in years
Education of household head in years
Number of adults in the households (> 18 years)
Household monthly income in Lesotho Maruti
Experience of growing wheat in years
Availability of extension services variable
-0.3619
1.1739
0.1144
-0.0413
-0.0976
0.0296
0.0573
0.0018
0.0103
-1.1753
2.2780**
1.0930
-1.1921
-2.0435**
2.7527**
1.7968*
-0.0660
1.1280
Correction factor ( δ 2)
R-square
Log of Likelihood ratio test statistic
0.7465
44.62
94.25**
1.6193
Double and single asterisks denote statistically significance at 5% and 10% level
Table 42.6: Multiple Regression Results on Wheat Production per Plot
Variable
Constant
Log of total labor (mandays)
Log of quantity of fertilizer (25 NPK bags)
Log of short-term soil conservation variable
Log of long-term soil conservation variable
If used Hybrid (yes=1, No=0)
Dummy for quantity of fertilizer (Dk)
R-squares (%)
F-statistics (zero slopes)
Estimated
coefficients
Asymptotic
T-Ratio
1.9285
0.0308
0.0887
0.0892
0.1141
0.1256
-1.4995
77.5600
85.11**
4.7994
0.2023
2.2553**
2.1995**
1.9452*
0.8537
-3.4716**
Double and single asterisks denote statistically significance at 5% and 10% level
Impact of Adopting Soil Conservation Practices on Wheat Yield in Lesotho
605
The results of both models suggest the followings. First, a decision to
adopt long-term soil conservation measures has a great influence on the
adoption of short-term soil conservation measure and vise versa. However,
once adopted, long-term soil conservation measures stay in the field and
will always acts as benchmark for adopting short-term soil conservation
measures. Extension efforts that promote soil conservation should
therefore be directed more to long-term soil conservation measures in
order to stimulate the adoption of short-term soil conservation measures.
Second, the signs on the estimated parameters indicate that the two
technologies are considered to be substitute to each other rather than
complimentary. For example, households with more available labor
(number of adults) will tend to focus more on adopting long-term soil
conservation measures than both technologies. Relatively educated
farmers will tend to adopt short-term rather that long-term soil
conservation measures. Demonstration plots that show the benefit of
adopting both technologies is highly recommended. Third, the indication
that availability of extension services has no influence on adoption, surpass
all logic. Nevertheless, this may be a sign of weak extension services in
the country, meaning that the current extension services delivery system
is too weak to influence any technological adoption
For the yield equation, the estimated coefficient of determination
(R2) was about 78%, indicating that the model explains at least 78% of
the variation in wheat production as reported by sample respondents.
The likelihood ratio test of the null hypothesis that all variables included
in the model have zero slopes was rejected at 1% level of significance.
All signs were as expected. Statistically significant variables were quantity
of fertilizer, and both adoption of short-term and long-term soil
conservation measures. The dummy variable for non-use of fertilizer
was statistically significant and negative as expected, indicating that
the yields of farmers not using fertilizers was more likely to be less than
those of farmers using fertilizer.
Because the variables used in the model are in the logarithm form,
the estimated coefficients for continuous variables are elasticities and
for dummy variables, the coefficients are intercept shifters. Therefore,
the average marginal product (AMP) of an input is the product of the
estimated elasticity times the output-input ratio (i.e., AMP=ÄY/ÄX=áY/
X). At the sample mean, the calculated marginal product of labor was
0.021, implying that increase in labor by one unit will increase yield by
0.02 bags/acre (0.3%). The marginal products of the fertilizer and shortterm and long-term soil conservation efforts were respectively 0.024,
1.18 and 1.80, implying that a unit increase in the use of fertilizer,
therefore, increases yields by 0.024bag/acre (0.3%). Adoption of
additional one unit of short-term or long-term soil conservation measure,
however, has a much greater impact, increasing yield by 1.18bags/acre
(17.9%) and 1.8bags/acre (27.3%) respectively.
606
Kaliba, A.R.M. and Rabele, T.
Conclusion and Policy Implication
The need to address the problem of soil erosion in Lesotho is widely
known. The solution to this problem lies in reducing the negative impact
of soil erosion on crop yields through soil conservation measures and
improved land management practices. As indicated by the regression
results, wheat farmers stand to gain more through increased soil
conservation efforts than use of inorganic fertilizers alone. Given the
limited availability of land, increase in acreage is not a viable solution.
Agricultural intensification through adopting soil conservation measures
may be best option for most farmers.
A thing to note is the limited influence of extension services on
adoption of soil conservation measures. Extension services are important
in enhancing the adoption of any new farming practices. Adequately
trained, well-supported extension services can effectively induce the
adoption of soil conservation practices. Field trials and demonstrations
successfully create awareness on returns associated with soil
conservation measures to farmers. It is therefore imperative that
extension services need to be strengthened in order to enhance the
adoption of soil conservation measures. Moreover, farmers will adopt
the practices that give high returns. Farm management studies aimed
at establishing soil conservation mixes that optimize returns to the
farmers are highly recommended.
References
Adesina, A. and M.M. Zinnah (1992) Adoption, diffusion, and economic impacts
of modern mangrove rice varieties in Western Africa: further results from
Guinea and Sierra Leone. In: Towards a New Paradigm for Farming System
Research/Extension. Working Paper for the 12th Annual FSR Symposium
1992. Michigan State University, 1992: 443-466.
Anderson, D. and Grove, R. (1987) Conservation in Africa, People, Policies and
Practices, Cambridge University Press: Great Britain.
Battese, G.E. (1997) A note on the Estimation of Cobb-Douglas Production
Functions when Some Explanatory Variables have Zero Values, Journal of
Agricultural Economics 48: 250-252.
Byiringiro, F. and T. Reardon (1996) Farm Productivity in Rwanda: Effects of
farm size, erosion, and soil conservation investments. Agricultural economics
15:127-136.
CIMMYT (1993) The adoption of agricultural technologies: a guide for survey
design. International Wheat and Improvement Center, Londres: Mexico.
Cox, H.W., Hammer, G.L., and Robertson, M.J. (2001) Opportunities for crop
modeling in barley. Proceedings of the 10th Australian Barley Technical
Symposium. The Regional Institute Ltd:Australia.
Impact of Adopting Soil Conservation Practices on Wheat Yield in Lesotho
607
Feder, G., Just, R.E., and Zilberman, D. (1985) Adoption of agricultural
innovations in developing countries: a survey. Economic Development and
Cultural Change 33 (1985): 255-297.
Goodwin, B.K. and Schroeder, T.C. (1994) Human capital, producer education,
and adoption of forward-pricing methods. American Journal of Agricultural
Economics 76: 936-947.
Greene, W.H. (1997) Econometric Analysis, Prentice, Hall, Upper Saddle River:
New Jersey.
Gujarati, D.N. (1995) Basic Econometrics. McGraw-Hill International,Singapore.
Hall, B.H. (1994) Time Series Processor User’s Guide. TSP International: Palo
Alto, CA.
Kaliba, A.R.M., Verkuijl, H.J.M. and Mwangi, W. (2000) Adoption of Maize
Production Technologies in the Intermediate and Lowlands of Tanzania.
Journal of Agricultural and Applied Economics 32(1): 35-47.
Kazianga, H. and Masters, W. (2002) Investing in Soils. Field bands and Microcatchments in Burkina Faso. Environmental and Development Economics,
7(3):571-592.
Kimhi, A. (1994) Quasi maximum likelihood estimation of multivariate models:
farm couples’ labor participation. American Journal of Agricultural Economics
74(4): 828-835.
Kingdom of Lesotho (KoL). (1980) National Papers on Environment and
Development in Lesotho.Government of Lesotho for the United Nation’s
Conference on Environment and Development, Maseru: Lesotho.
Kastens, T.L. and Dhuyvetter, K.C. (1999) Management Factors: What is
Important? Prices,Yield,Costs, or Technology Adoption? Staff paper,
Agricultural Economics Department, Kansas State University: Manhattan,
Kansas.
Lesotho Mountain Research Group (LMRG). (1996) Towards an Integrated
perspective on Land Degradation in the Mountains of Lesotho. A report on
the Inaugural Meeting held at Malealea Lodge, Ministry of Environment,
Maseru, Lesotho.
Maddala, G.S., 1992. Introduction to Econometrics . Macmillan Publishing
Company: New York
Ministry of Agriculture (MOA). (1988) Soil Conservation Handbook for Lesotho,
Conservation Division: Maseru: Lesotho.
Nkonya, E., Schroeder, T. and Norman, D. (1997) Factors affecting adoption of
improved maize seed and fertilizer in northern Tanzania, Journal of
Agricultural Economics 48:1-12.
Nowak, P.J. (1987) The Adoption of Agricultural Conservation Techniques:
Economic and Diffusion Explanations. Rural Sociology 52(2): 208-210.
Ockwell, A.P. (1991) Characteristics of improved technologies that affect their
adoption in semi-arid tropics of eastern Kenya. Journal of Farming System
Research and Extension 2:6-18.
Rahm, M.R., and Huffman, W.E. (1984) The adoption of reduced tillage: the role
of human capital and other variables. American Journal of Agricultural
Economics. 66: 405-413.
608
Kaliba, A.R.M. and Rabele, T.
Saha, A., Love, H.A. and Schwart, R. (1994) Adoption of emerging technologies
under output uncertainty. American. Journal of Agricultural Economics 76:
836-846.
Shively, G.E. (1998) Modeling Impacts of Soil Conservation on Productivity and
Yield Variability: Evidence from a Heteroskedastic Switching Regression.
Selected paper at Annual meeting of the American Agricultural Economics
Association, 2-5 August 1998, Salt Lake City, Utah.
White, H. (1980) A Heteroskedasticity-Consistent Covariance Matrix and a Direct
Test for Heteroskedasticity. Econometrica 48:721-746.
Patrick Mutuo