Decomposition and Nutrient Release Patterns of the Green Manure Phytomass in Irrigated Mango Orchard T.C. de A. Gomes, J.A.M. e Silva, E.M.B. Soares and M.S.L. da Silva Embrapa Semi-Árido P.O.Box 23, 56300-970, Petrolina, Pernambuco, Brazil Keywords: mulch, pearl millet, Sudan sorghum, Crotalaria juncea, semi-arid, Brazil Abstract Decomposition rate and N, P, K, Ca and Mg release patterns of biomass of Crotalaria juncea, pearl millet (Pennisetum typhoides) and sudan sorghum in a mango orchard irrigated by micro-sprinkler was studied in the Northeast of Brazil. The litterbag method was used and 20 g of dry weight of plant upper part or fresh leaves of the three species were collected in nylon screen bags. The litterbags were distributed in the soil surface under the mango trees. Three bags of each material were collected after 1, 2, 4, 6, 8, and 12 weeks after distribution. All three species presented a fast decomposition phase from the fourth to the sixth week, followed by a slower decomposing phase. Yearly decomposition constants of leaves and upper part biomass (k) were, respectively, 4.59 and 3.88 for sudan sorghum, 6.34 and 3.71 for pearl millet and 7.01 and 5.30 for C. juncea. Potassium was the fastest releasing element. About 90% of potassium contained in the leaves of all three species and in C. juncea upper part were released during the first two weeks of decomposition. It was observed a strong immobilization of magnesium and calcium during the decomposition of the grasses biomass upper part. INTRODUCTION In the Sub middle of the São Francisco River Valley, Brazilian semi-arid zone, most of the soils cropped with mango trees show low levels of organic matter content, cation-exchange capacity (CEC) and water holding capacity. The "in situ" production of organic matter to improve soil chemical, physical and biological characteristics through mulching constitutes a potential alternative for its preservation. This practice, however, does not guarantee that the nutrients will be liberated in synchronism with plant nutritional demands. Choosing the organic residue type and application time depends on the knowledge on decomposition speed and factors affecting it (Henrot and Brussaard, 1997). When the biomass decomposition process is known, pruning and some other management can be optimized to benefit plants in critical stages of crop cycle (Budelman, 1988). This work aimed to study speed decomposition and nutrient release patterns of upper part biomass and leaves of pearl millet (Pennisetum typhoides cv. IPA-Bulk-1 BF), sudan sorghum cv. AG-2501-C (Sorghum sudanensis x Sorghum bicolor,) and a leguminous species, Crotalaria juncea, in an irrigated mango tree orchard in the Brazilian northeastern semi-arid. MATERIAL AND METHODS The experiment was carried out in a commercial mango (Mangifera indica) orchard, in Petrolina, State of Pernambuco, Brazil. The area was located in the São Francisco River Valley (9°9' South, 40°22' West and 365.5 m altitude), where the annual mean rainfall is 570 mm, the monthly mean temperatures varies from 24.2 to 28.1°C and the air relative humidity ranges from 52% to 70%. ‘Tommy Atkins’ mango trees were spaced 10 m x 5 m in an Entisol Typic Quartzipsamment soil, with 90% sand and 3% clay, under micro-sprinkle irrigation. Leaves and upper part biomass (leaves, stems, inflorescences and branches) decomposition of pearl millet, sudan sorghum and C. juncea, which were sown between mango tree rows, were evaluated after the second growing cycle. Intercropped species Proc. VIIth IS on Mango Eds. A.C.Q. Pinto et al. Acta Hort 645, ISHS 2004 183 were sown in March 2001, during the normal rainfall season, and cut sixty days later. Biomass was left on the soil surface from May to August for decomposition. During this time they were registered the amount of water applied by irrigation and the average daily temperature (Fig. 1), but there was no rainfall record in that time. The experiment was set up in a complete randomized block design with three replications. The litterbag method was used and 20 g of dry weight of plant upper part or fresh leaves of the three species were collected in nylon screen bags (0.3 m x 0.3 m). The litterbags were distributed on the soil surface under the mango trees. Eighteen bags were used for each species (six bags/replicate). In each sampling time (1st, 2nd, 4th, 6th, 8th and 12th weeks after the decomposition process beginning), a bag of each plot was collected and the biomass remaining was weighed and analyzed to determinate mass nutrient losses. Concentrations of nitrogen, phosphorus, potassium, calcium and magnesium were analyzed in the original plant material, as well as in the remaining material in the bags, as described by Malavolta et al. (1997). Data are presented as remaining percent of the original nutrients, leaf mass and upper plant parts. To determine differences in the decomposition and nutrient release patterns of the three species, analysis of variance was performed for each time interval. The decomposition and nutrient loss constants (k) were determined by regression analysis through single exponential model x = e-kt, where x is the proportion of initial biomass or nutrient remaining in each sampling time, t, in years (Wieder and Lang, 1982). The averages for k values were compared by the Scott-Knott grouping test at 5%. RESULTS AND DISCUSSION Nitrogen, potassium, calcium and magnesium concentrations in leaves were generally higher than in the upper part biomass of the studied species (Table 1). Phosphorus concentrations were similar for the three species, in spite of the considered material (leaves or upper part biomass). C. juncea showed the largest values for nitrogen, calcium and magnesium, while the grasses presented larger potassium concentrations. Sudan sorghum and pearl millet showed similar decomposition patterns, mainly when the whole material of the plant upper part was considered, significantly differing from that of C. juncea (Fig. 2). Analysis of variance for each sampling time showed that pearl millet and sudan sorghum had a similar percentage of upper part and leaf remaining, except for fourth and twelfth weeks for leaf remaining. The average values for decomposition and nutrient loss constants (k) ranged from 3.71 to 17.66 yr-1 (Table 2) and were higher than those registered in the literature by Palm and Sanchez (1990) for the humid tropics, which ranged from 1.65 to 8.48 yr-1. The statistical comparison of the constants confirms the observation that C. juncea decompose significantly faster than the other species. Within each specie, the process of leaf decomposition resulted in nutrients release similar to the upper part biomass, except for nitrogen in C. juncea and magnesium in sudan sorghum that were released faster. Nutrient loss rates for both decomposing materials showed the following general trend: potassium > nitrogen, phosphorus, calcium and magnesium. An exception it was the slow release of magnesium from the pearl millet upper part biomass and for sudan sorghum, that was strongly influenced by immobilization problems. During the studied period, the decomposition process of pearl millet, sudan sorghum and C. juncea leaves did not show nitrogen immobilization (increase on nitrogen concentration in the remaining material). However, when the decomposition material was the upper part biomass the immobilization process did happen at the first week, especially for C. juncea. After that time, release of the nitrogen did start. In both studied materials, more than 60% of the initial nitrogen content had already been released until the sixth week after the decomposition process beginning. The decomposition of the C. juncea upper part biomass and leaves clearly evidenced the phosphorus immobilization in the second week of collection (Fig. 2). Potassium was the nutrient liberated quicker, with k values that ranged from 10.49 184 to 17.66 yr-1. In study developed in the humid tropical region of Peru, assessing the decomposition of three arboreal legumes, Palm and Sanchez (1990) also observed the faster release of the potassium. However, the k values found ranged from 4.15 to 9.94 yr-1. Budelman (1988) ponders that the optimum situation for fast decomposition is where high average temperatures are found together with a continuous water supply. The constant irrigation level and observed temperatures during the period of the study (Figure 1), probably made possible, already in the second week, the release of 90% of the existent potassium in the leaves of the three studied species, as well as that present at upper part biomass of C. juncea. For the pearl millet and sudan sorghum, the percentage was reached in the sixth week after the process beginning. Although it has not been observed during leaf decomposition, it should be pointed out the occurrence of calcium immobilization in pearl millet upper part biomass, which release did begin only after the sixth week. On the other hand, for C. juncea and sudan sorghum, the initial immobilization was, respectively, small and null. The phenomenon of net immobilization repeated more deeply with regard to the magnesium of sudan sorghum upper part biomass, with great variation in concentration during the whole study period, resulting in small k values, with low fitness for the single exponential model used. These results resemble those found by Palm and Sanchez (1990) for Inga edulis and Cajanus cajan, and those by Thomas and Asakawa (1993) for Arachis pintoi and Centrosema acutifolium, when they verified long periods of calcium and magnesium immobilization CONCLUSIONS The results obtained during the studied period allowed the following conclusions: 1. The decomposition rate of the upper part biomass and leaves was fast, possibly due to irrigation, with k values ranging from 3.71 to 17.66 yr-1; 2. The upper part biomass of the leaves of C. juncea were decomposed faster than those for pearl millet and sudan sorghum; 3. The decomposition of the leaf material resulted in nutrient release similar to the upper part biomass, except for the nitrogen in C. juncea and magnesium in sudan sorghum which were released faster; 4. The potassium was the nutrient released faster, being around 90% of this nutrient existent in the leaves of the three studied species, as well as that present in the upper part of C. juncea, already released in the second week after the decomposition process beginning; 5. There was nitrogen net immobilization during the upper part biomass decomposition of the three studied species, but especially for C. juncea; 6. The decomposition of the upper part biomass presented phases of calcium and magnesium immobilization, especially in the two grasses species investigated. Literature Cited Budelman, A. 1988. The decomposition of the leaf mulches of Leucaena leucocephala, Gliricidia sepium and Flemingia macrophylla under humid tropical conditions. Agroforestry Systems 7:33-45. Henrot, J. and Brussaard. L. 1997. Determinants of Flemingia congesta e Dactyladenia barter mulch decomposition in alley-cropping systems in the humid tropics. Plant and Soil 191:101-107. Malavolta, E., Vitti, G.C. and Oliveira, S.A. 1997. Avaliação do estado nutricional das plantas: princípios e aplicações. 2.ed. Potafos, Piracicaba. 319p. Palm, C.A. and Sanchez, P.A. 1990. Decomposition and nutrient release patterns of the leaves of three tropical legumes. Biotropica 232(4):330-338. Thomas, R.J. and Asakawa, N.M. 1993. Decomposition of leaf litter from tropical forage grasses and legumes. Soil Biol. Biochem. 25:1351-1361. Wieder, R.K. and Lang, G.E. 1982. A critique of the analytical methods used in examining decomposition data obtained from litter bags. Ecology 63(6):1636-1642. 185 Tables Table 1. Initial nutrient concentrations of leaves (L) and upper part biomass (UP) for pearl millet, sudan sorghum and C. juncea used in the litterbags study. Species 1 N P K Ca Mg UP L UP L UP L UP L UP -1 ------------------------------------------------ g kg --------------------------------Pearl millet 29.00 20.75 2.26 2.40 39.64 27.43 14.10 6.08 3.73 1.95 (1.31) (2.15) (0.34) (1.20) (4.51) (6.55) (4.31) (0.88) (0.84) (0.37) Sudan sorghum 25.17 22.70 2.60 2.61 28.04 23.94 8.83 8.30 3.40 2.78 (4.70) (11.90) (0.40) (0.70) (2.96) (4.79) (0.29) (7.29) (0.20) (0.38) C. juncea 36.13 24.67 2.88 2.26 18.22 18.84 26.8 13.42 4.60 3.20 (4.53) (2.25) (1.12) (0.43) (1.35) (2.21) (0.46) (7.07) (0.10) (1.05) L 1 Number in brackets corresponds to the mean stand deviation. Table 2. Decomposition and nutrient release constants (k) for leaves (L) and upper part biomass (UP) for pearl millet, Sudan sorghum and C. juncea used in the litterbags study. Species 1 Biomass Nitrogen Phosphorus UP CV (%) L UP CV (%) L UP CV (%) Pearl millet 3.71Bb 1.94 7.05 Aa 5.27Aa 20.97 4.50Ba 5.5 Aa 16.25 Sudan sorghum 3.88Ba 16.29 5.08Aa 6.11Aa 26.42 5.65Ba 4.2 Aa 22.62 C. juncea 5.30Aa 10.08 9.46Aa 7.92Ab 4.54 7.32Aa 6.7 Aa 6.57 CV (%) 14.40 10.93 13.68 10.16 16.92 Species Potassium Calcium Magnesium2 L UP CV (%) L UP CV (%) L UP CV (%) Pearl millet 16.46Aa 12.62Aa 7.81 5.55Ba 5.81Ba 21.48 6.32Aa 3.85Aa 27.96 Sudan sorghum 15.15Aa 10.49Aa 18.49 4.74Ba 4.07Ba 19.52 C. juncea 17.66Aa 16.53Aa 18.79 11.69Aa 9.77Aa 6.34 8.96Aa 7.55Aa 10.90 CV (%) 18.81 20.41 9.67 18.92 17.91 12.22 1 Means followed by same letter did not statistically differ from each other by the Scott-Knott Grouping Test at 5%. Upper case letters are for comparing k values among species and lower case letters are for comparing k leaf and upper part biomass values in the same species. 2 The single exponential model did not fit very well for the sudan sorghum magnesium data and the release constants were not presented. L 6.34Ba 4.59Ca 7.01Aa 3.14 186 Figures Irrigation am ount 10 30 M ean day tem perature 9 Irriga tion (m m ) 7 20 6 5 15 4 10 3 2 Te m pe ra ture (°C) 25 8 5 1 0 0 0 1 2 3 4 5 6 7 8 9 10 11 12 Tim e (w e e ks) Fig. 1. Irrigation levels and average daily temperatures during the studied period. 187 Biom ass 120 100 80 Nitroge n 140 P earl m illet leaves S udan sorghum leaves C . juncea leaves P earl m illet upper part S udan sorghum upper part C . juncea upper part 120 100 80 60 60 40 40 20 20 0 0 0 2 4 6 8 10 12 Phosphorus 120 0 4 100 80 80 60 60 40 40 20 20 6 8 10 12 10 12 10 12 Potassium 120 100 % R emaining 2 0 0 0 2 4 6 8 10 Calcium 200 0 12 2 4 6 M agne sium 450 180 400 160 350 140 300 120 8 250 100 200 80 150 60 100 40 50 20 0 0 0 2 4 6 8 10 12 0 2 4 6 8 T im e (weeks) Fig. 2. Dry matter, nitrogen, phosphorus, potassium, calcium and magnesium percent remaining in the leaf and upper part biomass for pearl millet, sudan sorghum and C. juncea during the decomposition time. 188