Final Environmental and Social Impact Assessment Volume 2: Technical Reports

Transcription

1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Government of the Republic of Malawi Ministry of Water Development and Irrigation National Water Development Programme Independent Environmental Impact Assessment for the Upgraded Kamuzu Barrage Final Environmental and Social Impact Assessment Volume 2: Technical Reports December 2013

2 Independent Environmental Impact Assessment for the Upgraded Kamuzu Barrage Final Environmental and Social Impact Assessment Volume 2: Technical Reports Table of Contents 1. Fisheries Report 2. Assessment of Wildlife Impacts 3. Vegetation Assessment 4. Water Quality Assessment Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol. 2, December 2013

3 Government of the Republic of Malawi Ministry of Water Development and Irrigation National Water Development Programme Independent Environmental Impact Assessment for the Upgraded Kamuzu Barrage Final Environmental and Social Impact Assessment Volume 2: Fisheries December 2013

4 Independent Environmental Impact Assessment for the Upgraded Kamuzu Barrage Final ESIA Volume 2: Fisheries Table of Contents... i Abbreviations... iv Executive Summary... v 1 Background EIA Objectives Methodology Scope of the Fisheries Study Area of Study EIA Approach Desk Study Field Surveys Collection of Fish Samples Identification of Fish Species and Data Collection Determination of Environmentally Sensitive Hot Spots and Endangered Fish Species in the Shire River System Statistical Data Analysis Major Findings The Shire River Biophysical Environment The Upper Shire River Lake Malombe Middle Shire River The Lower Shire River Fish Species Composition in the Shire River Breeding Seasonality and Migration of Fish in the Shire River State of Fish Stocks and Fisheries Management in the Shire River Malawi Government Policy on Fisheries Environmentally Sensitive Aquatic Areas/Habitats which may be Impacted by the Upgraded Kamuzu Barrage Project Endangered Fish Species that are Potentially Impacted by the Proposed Kamuzu Barrage Upgrading Fishing Catch Biomass by Species and Gear Gear selectivity Fish Species Composition and Abundance Size Distribution by Species Distribution and Size at Capture of Major Fish Groups Discussion Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December 2013 i

5 4.1 The Shire River Aquatic Environment Fish and Fisheries in the Shire River System Possible Impacts of the Upgraded Kamuzu Barrage at Liwonde on Fish and Fisheries in Lakes Malawi and Malombe and the Shire River Impediment to Fish Migration Destruction of Fish Breeding Aquatic Habitats through Flooding and Siltation Changes in Water Flow Downstream Impacts on Fish of Lake Malawi Particularly the Rock Dwelling Cichlids (Mbuna) Impacts of Construction Activities on Fish and Fisheries Resources Mitigation of Adverse Impacts Conclusion and Recommendation Reference Material Appendix 1: Summary of Key Questions Table of Figures Figure 1. Kamuzu Barrage (Photo: F. Kapute)... 1 Figure 2. Map of the south east Arm of Lake Malawi and Lake Malombe... 2 Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. Fishers operate the beach seine at Namiasi (Area A of south east arm of Lake Malawi, Photo: F. Kapute)... 3 Mouth of the Shire River from the south east arm of Lake Malawi near Mpondasi in Mangochi (Photo: F. Kapute)... 4 Bakili Muluzi Bridge (first and new bridge on the Shire River from mouth) RIGHT: part of the old bridge (Photo: F. Kapute)... 4 Illegal fishing (fishers in a dugout canoe shown by arrow) in the protected area of the Shire River at Mvera (Photo: F. Kapute)... 5 Scoop nets are one of the most dynamic and active gears heavily used at Kamuzu Barrage when the gates are open (Photo: F. Kapute)... 5 Figure 8. Cast nets are also used to catch fish in the Shire River around Kamuzu Barrage (Photo: F. Kapute) Figure 9. The narrow and meandering Middle Shire River near Nkula Falls (Photo: F. Kapute)... 6 Figure 10. Figure 11. ESCOM intake at Nkula Falls end point of the Shire River before rapids (Photo: F. Kapute)6 The mouth of the Shire River (far back) from the south east arm of Lake Malawi (Photo taken from Muluzi Bridge at Mangochi by F. Kapute)... 8 Figure 12. Hippopotamus in the Shire River at Liwonde National Park... 9 Figure 13. Lake Malombe has a marshy and muddy shoreline (Photo: F. Kapute) Figure 14. An aerial view of the vast Lower Shire River ( 11 Figure 15. Water hyacinth plants in the Shire River (Photo: F. Kapute) Figure 16. Fish species in the Shire River (Photo taken at Kamuzu Barrage by F. Kapute) Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December 2013 ii

6 Figure 17. Fish traps, cast nets and scoop nets are heavily employed by fishers to catch fish in the Shire River (Photo: F. Kapute) Figure 18. Lake Malombe fish catch composition (Source: Malawi Department of Fisheries) Figure 19. Lower Shire catch statistics from (Source: Malawi Department of Fisheries) Figure 20. Figure 21. Figure 22. Figure 23. Figure 24. Receded water levels in Area A of the south-east arm of Lake Malawi (near MALDECO) at Mangochi (Photo: F. Kapute) The Upper Shire River at Mangochi (Left = upstream and Right = downstream) on its way to Lake Malombe (Photo: F. Kapute) The Shire River at Liwonde see the existing floating weed barrier across the river (Photo: F. Kapute) Species catch biomass for the Upper Shire River caught using Hand line and larger cichlids caught using Hand line from the Upper Shire (Photo: F. Kapute) Species catch biomass for the Upper Shire River caught using Hand line and larger cichlids caught using Hand line from the Upper Shire Figure 25. Species catch biomass for Lake Malombe caught using Scoop net (Photo: F. Kapute) Figure 26. Figure 27. Figure 28. Species catch biomass for Nkula Falls ESCOM Intake Dam caught using Hand line (Photo: G. Steeves Size distribution, logistic selection ogive and size at capture of Cichlids caught in Beach seines Size distribution, logistic selection ogive and size at capture of Cichlids caught in Mosquito nets Figure 29. Distribution of fish families by % weight sampled during the survey Figure 30. Most abundant (L) and least abundant (R) fish genera sampled during the survey Figure 31. Most abundant (L) and least abundant (R) fish species sampled during the survey Figure 32. Small cichlid (kambuzi) catch from south east arm of Lake Malawi caught using a small meshed beach seine (Photo: F. Kapute) Figure 33. Length frequency histograms of different species Figure 34. Distribution and size at capture of cichlids caught using scoop nets and hand line respectively Figure 35. Shire River (L) and Lake Malawi (R) marshy and sandy shorelines respectively (Photos: F. Kapute) Figure 36. Beautifully coloured rock dwelling Cichlids (Mbuna) Table of Tables Table 1. Fish catch composition by weight for the Lower Shire from 2000 to Table 2. Fish families and genera sampled during the survey Table 3. Shire River fish species composition Table 4. Summary of impacts and suggested mitigation measures/plans Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December 2013 iii

7 Abbreviations CITES EIA ESCOM FAO HEP IUCN km m masl NEP RAMSAR TL Convention on International Trade in Endangered Species of Wild Fauna and Flora Environmental Impact Assessment Electricity Supply Corporation of Malawi (Ltd) Food and Agriculture Organization (of the United Nations) Hydro Electric Power International Union for Conservation of Nature kilometre metres metres above sea level National Environmental Policy The Convention on Wetlands of International Importance Total Length Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December 2013 iv

8 Executive Summary The Kamuzu Barrage at Liwonde across the Shire River in southern Malawi which was commissioned in 1965 is said to have outlived its design life span. The Government of Malawi is therefore planning to upgrade the Barrage to a modern structure. The upgrading includes raising the Maximum Highest Regulating Water Level of the Barrage up to 40 cm corresponding to a Malawi Lake level of masl. In anticipation of a likelihood that this may cause environmental and social impacts in the Shire River and Lakes Malawi and Malombe and the communities that directly or indirectly obtain sustenance from the River and Lakes, an environmental impact assessment (EIA) was undertaken. The objectives of the EIA were; to identify environmentally sensitive aquatic areas/habitats which may be impacted by the proposed project; identify any locally, nationally and/or regionally rare and/or endangered species of fish that might be impacted by the proposed project; and finally to assist in the identification of potential environmental impacts and suggest mitigation measures including any potential risks to Lake Malawi's aquatic biodiversity. The study established that the Liwonde National Park is an important refuge and breeding area for fish in the Shire River and Lake Malombe while the surrounding areas of Middle Shire River and Lake Malombe are subjected to intensive and uncontrolled fishing. After breeding, fish migrate to these areas where they are caught by fishers. The Kamuzu Barrage which is located downstream of Liwonde National Park, therefore, does not interfere with any fish migrations in the Shire River between the Park and Lake Malombe. The proposed upgraded Kamuzu Barrage thus poses essentially the same type of physical barrier to fish movements as the existing one; hence no change from the status quo. Also, downstream migration of fish will still occur as the gates are often fully open under low flow and high flow conditions. Upstream migration may also be possible during these times since by their nature, fish prefer to swim against the water current. One important way of aiding upstream migration of the fish along the Barrage is the construction of fish passes. However, heavy and uncontrolled fishing currently prevailing at the Barrage defeats the whole idea as fish would be subjected to easy and massive exploitation by fishers. It is suggested that sound management of the fisheries resource in the Park especially preventing fishing, would help maintain the fish stock in the Shire River and probably Lake Malombe. Based on a careful analysis of the impacts and their proposed mitigation measures, this report concludes that the project of refurbishing the Kamuzu Barrage on the Shire River does not pose significant threats to the fisheries, fish resources and the aquatic habitat of the Shire River. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December 2013 v

9 1 Background The Kamuzu Barrage (also referred to as Liwonde Barrage), built across the Shire River (Figure 1), at Liwonde, was commissioned in The Barrage serves as a water regulation device for the generation of electricity about 80 kilometres down the river Shire at Nkula Falls Hydro Electric Power Station. Other water users to benefit from the operation of the Barrage include downstream irrigators, the Sugar Corporation of Malawi (SUCOMA) Plantation and town water supplies such as Blantyre city. At present, the Barrage is no longer fully operational and requires extensive maintenance. Figure 1. Kamuzu Barrage (Photo: F. Kapute) The government of Malawi through funding from the World Bank is therefore planning to refurbish/upgrade this Barrage to a modern structure. The upgraded Barrage includes raising the highest regulated water level by up to 40 cm to masl. This calls for an extensive and detailed environmental impact assessment (EIA) as the change in water levels upstream and change in regulated flow downstream will result in some negative and positive outcomes whether environmental, socio-economic etc. 1.1 EIA Objectives The purpose for this study was therefore to conduct a detailed environmental impact assessment on what could be the likely impacts of the changed water levels upstream of the Barrage and the changed flow regime downstream on aquatic habitat, fisheries and fish resources and consequently how this would eventually affect the socio-economics of the people especially those around the Shire River ecosystem who rely on fishing for their livelihood. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

10 2 Methodology 2.1 Scope of the Fisheries Study The scope of the Fisheries Study is as follows: 1) To identify environmentally sensitive aquatic areas/habitats which may be impacted by the proposed project; 2) To identify any locally, nationally and/or regionally rare and/or endangered species of fish that might potentially be impacted by the proposed project; 3) To assist in the identification of potential environmental impacts and mitigation measures including any potential risks to Lake Malawi's aquatic biodiversity, such as the numerous endemic fish species that live only within the shallow waters of the lake. 2.2 Area of Study The area under study stretches from Area A (Figure 2) of the south east arm of Lake Malawi where the Shire River starts to Nkula Falls and further downstream to the lower Shire valley. Figure 2. Map of the south east Arm of Lake Malawi and Lake Malombe The justification is that according to the biophysical environment of the Shire River, there is no upstream migration of fish from the Lower Shire beyond the falls and rapids which start at Nkula. As such, the Upper and Middle Shire including Lake Malombe appear to be the major focus as far as changes in water levels in the Shire River are concerned. 2.3 EIA Approach The study was under taken in two phases including a desk study and field surveys. The desk study provided the basis for the design of field studies as well as reference for validating findings at local, national and regional level. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

11 2.3.1 Desk Study The desk study was conducted by reviewing literature which includes Government Frame Survey Reports, academic research and publications as well as surfing the web for relevant comparative studies and reference material. Documentation pertaining to the construction and review of impacts of the existing Barrage at Liwonde were also reviewed Field Surveys Collection of Fish Samples The focus of this work concentrated on the section of the Shire River from its mouth at the south east arm of Lake Malawi to Nkula Falls where the river flows into a series of rapids then into a wide valley (Lower Shire). The justification here is that as far as fisheries resources are concerned, changes in water levels as a result of the upgraded Barrage at Liwonde would only affect (if any at all) fish and fishing activities up to the escarpments (falls) at Nkula. Beyond Nkula Falls (which is the first fall in a series from the mouth of the Shire River), no migration of fish is possible from the Lower Shire upstream due to the rapids and the high and rocky waterfalls. Samples of fish were collected from randomly selected areas in the Shire River in early October Samples were collected from the south east arm of Lake Malawi at Namiasi from fishers using beach seine (Figure 3) to catch the fish. Figure 3. Fishers operate the beach seine at Namiasi (Area A of south east arm of Lake Malawi, Photo: F. Kapute) From the south east arm of Lake Malawi, the next lot of fish samples was collected from the Upper Shire River section between the mouth of the Shire River (Figure 4) and Bakili Muluzi Bridge at Mangochi Boma (Figure 5). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

12 Figure 4. Mouth of the Shire River from the south east arm of Lake Malawi near Mpondasi in Mangochi (Photo: F. Kapute) Samples from Lake Malombe were purchased from fishers at Mpinganjira (a fish selling point near one of the beaches of Upper Lake Malombe). One problem faced during collection of fish samples from Lake Malombe was that the lake is closed to fishing in October every year by the Malawi Department of Fisheries to allow the fish to breed. Normal fishing is therefore suspended during this time. However, we realized through well planned sporadic interviews with the fish sellers and buyers that the fish that we purchased at this market were actually illegally caught from Lake Malombe despite the closed season. This highlighted the difficulties faced by the Malawi Department of Fisheries in enforcing fisheries regulations. Figure 5. Bakili Muluzi Bridge (first and new bridge on the Shire River from mouth) RIGHT: part of the old bridge (Photo: F. Kapute) Collecting fish samples from the section of the Shire River between Lake Malombe and Kamuzu Barrage was very difficult also because this stretch of the river forms part of the Liwonde National Park which is a protected/restricted area from fishing. During our preliminary visit to Liwonde National Park prior to the inception of this work, we had firsthand experience of illegal fishing activity continuing in the Shire River section of the National Park (Figure 6). However, as was the case with Lake Malombe, we managed to buy fish samples at Kamuzu Barrage which after detailed scrutiny, revealed that most of the fish that is sold at the Barrage is predominantly the Shire strain of the tilapia (Oreochromis shiranus) and is actually poached from the National Park. It is clear therefore that the Park acts as a reservoir of fish stock which upon migration is caught by fishers in the peripherals of the Park and sometimes even within the Park. At the Kamuzu Barrage (Figure 1), fish samples were sourced from fishers using scoop nets and cast nets (Figures 7 and 8 respectively). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

13 Figure 6. Illegal fishing (fishers in a dugout canoe shown by arrow) in the protected area of the Shire River at Mvera (Photo: F. Kapute) Fishers here usually use the scoop (Figure 7) and cast nets (Figure 8) to catch riverine fish species moving downstream with the strong water currents at the Barrage when the doors have been opened to allow water downstream to Nkula Falls Hydro Electricity Power Station. Fishing using scoop nets is thus only active when the Barrage gates are opened. Figure 7. Scoop nets are one of the most dynamic and active gears heavily used at Kamuzu Barrage when the gates are open (Photo: F. Kapute) No active fishing is documented for the section of the Shire River between Kamuzu Barrage and Nkula Falls Hydro Electricity Power Station as evidenced by the Malawi Department of Fisheries which did not have fisheries data available for the area. Figure 8. Cast nets are also used to catch fish in the Shire River around Kamuzu Barrage (Photo: F. Kapute). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

14 The Shire River in this area meanders (Figure 9) through a remote area where fishing is highly uncoordinated and fish catch data is not documented (personal communication, Malawi Department of Fisheries). Figure 9. The narrow and meandering Middle Shire River near Nkula Falls (Photo: F. Kapute) We managed to collect fish samples at Nkula Falls Electricity Supply Corporation of Malawi (ESCOM) dam (water intake, Figure 10) where water from the Shire River is kept before being channelled into a tunnel to the hydro electricity generation power houses. The commonly used gear here was hand lining and sometimes use of fish traps. Figure 10. F. Kapute) ESCOM intake at Nkula Falls end point of the Shire River before rapids (Photo: Thus, in summary, fish samples were collected from Area A of the south east arm of Lake Malawi, the Upper Shire (between the mouth and Lake Malombe), Lake Malombe, section of the Shire River in the Liwonde National Park, Kamuzu Barrage and finally at Nkula Falls where the terrain of the river drops into rapids before going into the Lower Shire. Fish catches were collected/purchased fresh sometimes live from fishermen using a variety of fishing gears ranging from beach seines, gill nets, scoop nets, hand lines, cast nets and mosquito nets Identification of Fish Species and Data Collection The catch from all fishing gears sampled was divided where necessary into three categories, small fish, catfishes and all big cichlid and non-cichlid fishes. Fish in each category were sorted and identified to species level. Standard length (mm) and whole weight was recorded (g) using a calibrated measuring board and a battery operated electronic weighing scale respectively. Length frequencies and catch data were raised to the total catch where sub-sampling was done. The type of fishing gear and areas where fish were caught were also documented. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

15 Determination of Environmentally Sensitive Hot Spots and Endangered Fish Species in the Shire River System During the field survey, environmentally vulnerable hot spots/areas were examined i.e. the aquatic habitat of the Shire River and species of fish likely to be impacted by the project. This was done by referring to literature and soliciting views from fishers and other stakeholders about past and current trends in fish catches and species composition as well as issues of environmental degradation Statistical Data Analysis The data were entered into computer Microsoft Excel spreadsheet for analysis. The data were analysed for species composition (diversity), gear selectivity, size distribution by species, weight by species and total length by gear by way of plotting length frequency histograms. Size-specific selectivity of Mosquito nets and Kambuzi beach seine fishing gear was modelled using the logistic model (Butterworth et al. 1989; Booth and Punt 1998) described as: L L 1 50 / P 1 exp where P is the selectivity of the gear on a fish of size L, L 50 is the size-at-50%-selectivity or the mean length at which 50% of fish are retained in the net (mean length at first capture), and is a parameter related to the size range over which the selectivity changes from values near 0 to values near 1. As tends to zero, this function approaches knife-edged selection (Butterworth et al. 1989). The size-specific selectivity of Scoop nets and Hand line fishing gears was modelled using the Gamma distribution function to compensate for the skewed distribution of fish catch data from the two fishing gears. The gamma probability density function (Malcom, 2001) is described as: x is the value at which the distribution is evaluated, α is a parameter to the distribution and β is a parameter to the distribution. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

16 3 Major Findings 3.1 The Shire River Biophysical Environment The Shire River, the only outlet of Lake Malawi located at 35 15' E and 14 25' S is the largest and longest river in Malawi (Tweddle et al. 1979; FAO, 1993; Ribbink, 2001). It stretches over a distance of about 410 km from its mouth at the south east arm of Lake Malawi near Mangochi to the Zambezi River (35 20' E and 17 50' S) (Osborne, 2000). The flow of water in the Shire River is directly affected by the highly variable flow rate in Lake Malawi (the source) creating fears that during low flow extremes, there may be a no-flow scenario in the Shire River or, that extreme high flows result in floods that could cause damage to agricultural lands, infrastructure and loss of life (World Bank, 2007). It is reported that the River outflow from Lake Malawi was once cut off because of a low Lake level between 1908 and 1935 (Parry et al., 2009). It is also reported that a Lake water level of masl in the early 1980s triggered flooding around the shoreline of the River (Parry et al., 2009). The Lake Malawi - Shire River hydrological system is probably Malawi s single most important natural resource system which supports tremendous livelihoods as well as supporting important environmental functions. The River supplies over 96% of the country's power generating capacity. It is also a source of water to the major urban centres of the southern region of Malawi such as Blantyre and Limbe and the rural water users along the length of the River. Many irrigation projects have also flourished along the River (World Bank, 2007). The Shire River is demarcated into three sections based on the geographical areas through which it flows from Lake Malawi to the confluence at Zambezi River in Mozambique, namely; the Upper, Middle and Lower Shire sections. 3.2 The Upper Shire River The Upper Shire (Figure 11) is the stretch of about 130 km between its outlet at the south-east arm of Lake Malawi (Samama) to Matope with a total bed drop of 15 m including a 1.5 m drop from Mangochi to Liwonde (Osborne 2000). Other earlier literature (Pike, 1972) indicates the distance as from Lake Malawi mouth to Lake Malombe. The latter description appears to be realistically based on the physiography that this is a relatively flat area but also with nearly similar characteristics such as shoreline vegetation etc. Figure 11. The mouth of the Shire River (far back) from the south east arm of Lake Malawi (Photo taken from Muluzi Bridge at Mangochi by F. Kapute) In this area, the River flows across a submerged sand bar about 4 km to the north of Mangochi town at an altitude of 468 m above sea level (Chimatiro, 2004) resulting in the mouth of the River from Lake Malawi being low and thus has been an issue of concern about reduced water levels in the Shire River during times of low water levels in Lake Malawi. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

17 Notably important aquatic habitats/structures within this stretch of the river are the Liwonde National Park and the Kamuzu Barrage. The Liwonde National Park is home to many riverine fish species e.g. Cyprinids such as Opsaridium microlepis (Mpasa) including the Shire Tilapia Oreochromis shiranus and a diversity of aquatic life (Figure 12). The park has diverse vegetation ranging from the swamps, lagoons and reed-beds along the river edge, grassland in the drained areas where palm, baobab and many trees grow ( From Lake Malombe mouth to Kamuzu Barrage, the river stretches over a distance of about 30 km (Pike, 1972). The Barrage at Liwonde (Figure 1) is a concrete structure, 156 m long with 14 radial steel gates and a 6.6 m carriage way providing a bridge across the River (Liabunya, 2007). It was constructed as part of the hydroelectric scheme and was commissioned in 1965 to regulate the flow of the Shire River so that water entry into the Hydro Electric Plant at Nkula Falls is equally controlled (Liabunya, 2007; Chapalapata, 2007). Figure 12. Hippopotamus in the Shire River at Liwonde National Park When there isn t enough water at the power station, the gates at the Barrage at Liwonde are opened and vice versa when too much water is not required at the power station. The Barrage also helps to prevent weeds from finding their way into the power stations downstream. 3.3 Lake Malombe Approximately 15 km downstream of Lake Malawi, the River enters into Lake Malombe (Figure 13), a shallow lake (30 km by 15 km) with an area of about 450 km 2 and depth range of 5 to 7 m located at S E (Bell 1998). Lake Malombe is perceived to be an ox-bow lake formed due to the swelling of the Shire River. The lake therefore has similar characteristics to those of the southern part of Lake Malawi s aquatic ecology and also a high level of fish biodiversity and genetic variation. The lake is shallow, turbid and nutrient-rich, with vegetated beaches/shores. The marshy and muddy beaches or shoreline of Lake Malombe make it lack the appeal and tourism potential compared to Lake Malawi. Lake Malombe is a polymictic lake in nature, that is, it experiences continuous and complete mixing due to its shallowness where nutrients are well recycled resulting in the productive zone extending to the bottom. It is reported that rainfall runoff contributes significantly to the productivity of the Lake (Welcomme 1985). However, despite its richness, Lake Malombe contributes about 1.2% of total fish landings in Malawi (FAO 2005). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

18 Figure 13. Lake Malombe has a marshy and muddy shoreline (Photo: F. Kapute) It is believed that after being dry for many years in the early 1900s, Lake Malombe refilled around the 1930s suggesting that the Lake Malombe population of fish is a relatively recent peripheral isolate (Anseeuw, 2011). Furthermore, migration of fish between the Upper Shire River and the south-east arm of Lake Malawi has also been reported (FAO, 1993). 3.4 Middle Shire River The Middle Shire covers a distance of about 80 km starting from Matope to Chikhwawa (Osborne, 2000). Other works (Tweddle & Willoughby, 1979) refer to the Middle Shire as being from Lake Malombe to Kapichila Falls. This is the same stretch which Pike (1972) describes as the third section of the Middle Shire River. The middle Shire is characterised by numerous falls and rapids and has total fall of 370 m (Osborne 2000; Tweddle & Willoughby, 1979) thus creating the potential for hydro-electric power production. The Kapichila Falls in Chikhwawa district and the Middle Shire rapids are physical and ecological barriers to the upstream migration of the Lower Zambezi fauna (Tweddle & Willoughby, 1979). Thus, fish cannot migrate from the Lower Shire to the Middle Shire because of the gorges and waterfalls. That is why species from the Lower Shire such as Oreochromis mossambicus are endemic to the Lower Shire and the Zambezi River. Whilst the Upper Shire is relatively shallow and flow is slow, the water gradually increases from the middle to the Lower Shire with an annual flow of 405m 3 /s at Matope (Osborne 2000). 3.5 The Lower Shire River The Lower Shire (Figure 14) starts from the Kapichira Falls (Chikhwawa) to the end of Ndindi Marsh on the border with Mozambique (34 50'-35 17'S). It is one of the seventeen major floodplains in Africa, covering an area of more than 820 km 2 of marshes at peak floods. The marshy area of the Lower Shire valley covers an area of about 650 km 2 (Chimatiro 2004). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

19 Figure 14. An aerial view of the vast Lower Shire River ( The Lower Shire River sustains an important river-floodplain fishery (where fish production is highly dependent on the quantity and quality of annual flooding regime) contributing about 11% of total fish landings in the country (Donda and Njaya 1997). The fishery in the Lower Shire which is mainly subsistence in nature with small-scale commercial operations provides livelihoods for about 4,000 people as gear owners or fishing crew members (Fisheries Department, 2003). The fisheries in the Lower Shire like in Lake Malawi are multi-species and multi-gear and dominant fish species in the catches include Mphende (Oreochromis species predominantly Oreochromis mossambicus) representing more than 50% of the catch and Mlamba (Clarias spp.) contributing about 40% of the total catch). The Lower Shire valley is categorised into three ecosystems namely; flood plain and swamp, lagoon and riverine areas. Vegetation in the Lower Shire valley is subject to tremendous seasonal variation due to i) the local rains which occur between December and April, ii) the Barrage at Liwonde which regulates the flow of the Shire River (Liabunya 2007; Chapalapata, 2007). Marshes are usually dry between August and November, and fill during December (Chimatiro, 2004). Water hyacinth (Eichhornia crassipes, Figure 15) is a weed causing considerable problems in the Shire River. Figure 15. Water hyacinth plants in the Shire River (Photo: F. Kapute) Water hyacinth and other aquatic weeds continue to cause serious problems to power generation at Nkula Hydro Electric Power Station (the first HEP Station downstream of Kamuzu Barrage) by blocking the screens at the inlet to the turbines (Mzuza, 2010; Liabunya, 2007; Chapalapata, 2007). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

20 3.6 Fish Species Composition in the Shire River Although the Upper Shire experiences heavy fishing, it appears that the fish catch data is scanty (Fisheries Department, personal observation) probably due to poor recording of fish catches and unregulated fishing. According to the FAO, the Upper Shire River contributes about 1% to total fish landings in the country (FAO, 2005). A section of the River that passes through Liwonde National Park is a protected area hence closed to fishing. This contributes at least as much to the overall fish catch but also conserving many fish species whose populations are severely reduced in other areas (Figure 16). Figure 16. Fish species in the Shire River (Photo taken at Kamuzu Barrage by F. Kapute) In fact, it appears that most of the fish that are caught around the Kamuzu Barrage by fishers using scoop nets (Figure 7), cast nets (Figure 8) and bamboo woven fish traps (Figure 17) probably originate from the Park suggesting that the Park acts as a breeding area for the fish. Figure 17. Fish traps, cast nets and scoop nets are heavily employed by fishers to catch fish in the Shire River (Photo: F. Kapute) The species composition for Lake Malombe like the entire Upper Shire and the south-east arm of Lake Malawi is diverse i.e. multi-species (Figures 16 and 18). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

21 Figure 18. Lake Malombe fish catch composition (Source: Malawi Department of Fisheries) Most of the species in the Upper Shire are also found in the south east arm of Lake Malawi and Lake Malombe probably due to migration. Though Lake Malombe supports a highly productive fishery because of its high primary productivity, uncontrolled fishing has overexploited the fish stock and what remains are small cichlids locally known as kambuzi (Weyl et al., 2004a & b; Banda et al., 2005; Kapute et al., 2008). The tilapia (Chambo) fishery which used to be vibrant in the lake has almost collapsed (Weyl et al., 2004a & b). Figure 18 shows that Copadichromis chrysonotus species (Utaka) also dominate the kambuzi (small cichlids) catch (Turner, 1995). Oreochromis shiranus (Shire River Tilapia) appears to be the most commercially abundant species which is usually sold around the Barrage either fresh or in smaller quantities processed (smoked) (personal observation). In the Lower Shire, the fish fauna is essentially of Zambezi River Basin origin as more than 60 species are caught in this fishery, but only three namely, Mlamba (Clarias gariepinus), Chikano (Clarias ngamensis) and Mphende (Oreochromis mossambicus) are of commercial importance (Willoughby and Tweddle 1978). These three contribute 90% to the total fish catch. In general, the Shire River has a multi-species fishery especially the Upper and Lower Shire sections. 3.7 Breeding Seasonality and Migration of Fish in the Shire River The riverine species usually breed during the rainy season and also when the river is flooding. The diverse ecosystems of the Shire River ranging from falls, rapids and swamps to floodplains may contributed to the varied breeding pattern of fish in the River. This is due to the fact that spawning periodicity in fish is to a greater extent influenced by abiotic factors such as hydrological regime of the river, climatic seasonality and habitat characteristics (Welcomme, 1995). For example, most of the life cycle of the fish in the Lower Shire Valley is subject to the seasonal drying of the marshes while breeding occurs during high water when feeding is also at its peak (Willoughby and Tweddle, 1978). In the Upper Shire River, fish seasonally migrate to and from the south-east arm of Lake Malawi via the Shire River to Lake Malombe (FAO 2003). Juveniles are found in the river from May to July moving towards Lake Malombe (Seisay, 1992). This suggests why management of fish in one system leaving the other as is the case for areas around the protected Liwonde National Park is counter-productive. In the Lower Shire, the general observation is that most of the fish species breed intensively during the rainy season although some species breed throughout the year. 3.8 State of Fish Stocks and Fisheries Management in the Shire River Catches have tremendously declined in all natural water bodies in Malawi including the Shire River due to uncontrolled fishing effort (FAO, 1993). Many attempts to manage the fishery by the government through the Department of Fisheries such as co-management have failed resulting in exploitation of the fish resources outside management restrictions. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

22 In Lake Malombe and the Upper Shire River the fish stocks are exploited only by the artisanal sector which employs gillnets, chambo seines, kambuzi seines and nkacha nets. In Lake Malawi, in addition to the artisanal sector, the stocks of the south-east arm are exploited by the semi-industrial and the industrial fisheries. It is reported that total annual catches from the Upper Shire River have declined from more than 1,200 tons in the early 1980s to the recent levels of around 500 tons (FAO, 1993). In the Lower Shire, the species diversity is much less than that of the middle Shire, Lake Malombe and the Upper Shire River. Dominant and commercially exploited species by fishers which makes up about 60% of the catch (Donda and Njaya, 1997) include Mphende (Tilapia species), Mlamba (Clarias species) and Chikano (Table 1). Fisheries in the Lower Shire are managed through five areas namely: Elephant Marsh (518 km 2 ), Ndinde Marsh (155 km 2 ), Lagoons around Chikhwawa (3.9 km 2 ), Bangula Lagoon (18 km 2 ) and Shire River (337 km 2 ) (Ratcliffe, 1971). The Lower Shire and associated marshes support a Lower Zambezi fish fauna, separated physically and ecologically from the Lake Malawi fauna by a series of cataracts, rapids and falls between Matope and Chikhwawa. Current catch results for catches from the Lower Shire suggest that the fishery in the Lower Shire is picking up compared to other sections of the River. TABLE 1. FISH CATCH COMPOSITION BY WEIGHT FOR THE LOWER SHIRE FROM 2000 TO 2010 Year Species Mphende Others Mlamba Chikano Year total , , , , , , , , , , , , , , , , , , , , , , , , , Total 22, , , , , Source: Malawi Department of Fisheries; Weight in tons Previously, the Lower Shire Valley used to contribute between 10 and 15% of the total national landings (Malawi Department of Fisheries 1989; Bulirani et al., 1999) and catches fluctuate (Figure 19). The system is highly productive and efficiently utilized by fishermen suggesting why catches have increased recently. The fishery is small scale and subsistence in nature. It is pursued almost exclusively using dugout canoes from numerous permanent and temporary traditional fishing villages. The main fishing methods in the Lower Shire include seine nets, gill nets, fish traps, scoop nets, cast nets and encircling fish fence and dug-out canoes and plank boats without engines are the main fishing crafts employed. Gill nets are however, the commonest fishing gear used. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

23 Figure 19. Fisheries) Lower Shire catch statistics from (Source: Malawi Department of It is reported (Chimatiro, 2004) that seasonal fluctuation of fish catches are due to the flooding pattern which seems to be influenced in part by the operations of the Kamuzu Barrage. Water Hyacinth is another current major problem in the area adversely affecting the fishery. The challenge in managing fishery resources in Malawi is that the artisanal fishery which includes the Shire River is open-access in character. Due to poor enforcement by the Malawi Department of Fisheries, many fishers continue to fish without any access restrictions. In fact, as opposed to Lakes Malawi and Malombe, the fishery in the Shire River appears to be unregulated. This could explain why catch data for most parts of the Shire River is not available. For the Shire River fishery, the only and well protected area is around Liwonde National Park. Evidence shows that fish catches in areas around the Park such as the southern tip of Lake Malombe (near the Park area) are better than the areas far from the park (FAO, 2005). 3.9 Malawi Government Policy on Fisheries Regarding riverine and floodplain fisheries, the Malawi Government Fisheries Policy (GoM, 2001) aims at sustainable and appropriate utilisation of rivers to facilitate migration and spawning of fish and other aquatic vertebrates and the provision of nursery areas to maintain biodiversity. The revised National Environmental Policy (NEP) places an emphasis on conservation of biological diversity ( 2011) by seeking to manage, conserve and utilise biological diversity (ecosystems, genetic resources and species) for the preservation of national heritage. The NEP is also supported by the Environmental Management Act of Malawi in providing general environmental protection through Sectoral policies (e.g. policies on land, water, fisheries, waste, and forestry) to maintain consistency. Malawi is a signatory to a number of international conventions supporting the conservation of biodiversity such as the African Convention on the Conservation of Nature and Natural Resources, Convention on Wetlands of International Importance (RAMSAR), Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES), Convention on the Conservation of Migratory Species on Wild Animals and the Convention on Biological Diversity ( 2011). All fisheries regulations governing the fisheries in Malawi are enshrined in the Fisheries Conservation and Management Regulations (2000). These regulations are formulated to meet the needs of the various functions of fisheries management, such as protection of the breeding stock, the juvenile fish and fishing effort. Broad categorization of these fisheries management regulations are: closed seasons, closed areas, mesh size restrictions, minimum take-able size of fish, fishing net maximum headline length, and fishing licenses. With these regulations in place, it appears that the challenge to managing fisheries resources in Malawi is the lack of enforcement rather than good policies. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

24 3.10 Environmentally Sensitive Aquatic Areas/Habitats which may be Impacted by the Upgraded Kamuzu Barrage Project At the time of the field survey, it was notably clear that the water level in Lake Malawi at the south-east arm had receded tremendously as evidenced by wide sandy beaches which are usually inundated during periods of more water in the Lake (Figure 20). During peak water levels, the vast sandy beach stretches along the Lake are submerged under water creating breeding areas for cichlids. Unfortunately, in the south east arm of Lake Malawi (Mangochi) many of these fish breeding sandy beaches have been cleared for holiday resorts which have flourished in recent years. This suggests that irrespective of whether it is peak or low water level in the Lake, breeding of fish is compromised by the fact that these sandy beaches are frequently used by swimmers, for boat landings and also for beach seine netting. Figure 20. Receded water levels in Area A of the south-east arm of Lake Malawi (near MALDECO) at Mangochi (Photo: F. Kapute) The area at the outflow from Lake Malawi is marshy (Figure 4) and relatively shallow due to a sand bar. Common fishing gear in this area are gill nets; the muddy and marshy beaches do not favour operation of beach seines. Fishing in this area is very active especially Chambo (tilapia) fish caught with gill nets. The river then becomes wide (Figure 21) up to Lake Malombe and beyond to the Kamuzu Barrage. As explained earlier, the Upper Shire is a very active section for fishing activities. Many types of gear are employed ranging from beach seines, gill nets and the recent destructive light fishing using mosquito nets. Lake Malombe is the most productive lake in Malawi with a marshy shoreline and muddy beaches an aquatic environment which provides good habitat for fish breeding. However, Lake Malombe is the most heavily fished lake in Malawi and fish catches especially the chambo (tilapia) have declined nearly to extinction (Kapute et al., 2008). Figure 21. The Upper Shire River at Mangochi (Left = upstream and Right = downstream) on its way to Lake Malombe (Photo: F. Kapute) Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

25 The section of the Shire River after Lake Malombe up to a few kilometres from Kamuzu Barrage is a National Park hence a protected area from fishing. The area thus acts as a breeding site for fish which later migrate to other parts of the River and into Lake Malombe. Near the Barrage at Liwonde, huge moving masses of aquatic vegetation are seen upstream (Figure 22) descending towards the Barrage. These have been a big problem for ESCOM which spends a large effort removing weeds at the Barrage to prevent their movement into the hydro electricity power station at Nkula Falls some 80 km downstream. The shoreline of the section of the Shire River from Lake Malombe down to Nkula Falls does not have much infrastructure (buildings) or an active agricultural industry so that much of the shoreline has not been disturbed. This suggests that this is a favourable area for fish spawning especially riverine species which are chiefly Cyprinids such as Opsaridium microlepis (Mpasa), Opsaridium microcephalus (Sanjika), Labeo cylindricus (Ntchira) and several Barbus species. Figure 22. The Shire River at Liwonde see the existing floating weed barrier across the river (Photo: F. Kapute) As described earlier (Figure 10), the Shire River between Liwonde and Nkula is relatively narrow with many meanders and a steeper gradient. As opposed to the flat terrain between Lake Malawi and Liwonde, in this area, the topography is hilly hence the steeper gradient. Liwonde National Park (about 580 km 2 ) lies on the banks of the Upper Shire River, bordering Lake Malombe to the north and encompassing a large area east of the river. It is the best described of the nine national parks and wildlife reserves in Malawi. The vegetation of the park along the Shire River is diverse ranging from the riverine swamps, lagoons and reed-beds. During the field study, it became apparent that the National Park is by far the most environmentally important area in relation to the upgrading of the Barrage at Liwonde. The Park as mentioned earlier is home to many fish species and acts as an important breeding area for the fish. Elsewhere in the Shire River System, fish hardly have a chance to breed because of the uncontrolled heavy fishing pressure. The existing Kamuzu Barrage is capable of regulating upstream water levels in Lake Malawi, Lake Malombe, and the upper Shire River within Liwonde National Park, when they are within the range of to metres above sea level (masl), as measured at the Lake. At the time of construction of the Barrage this water level increase was considered to be able to minimise the impacts of flooding upstream of the Barrage in the Shire River, Lake Malombe and Lake Malawi while still providing reliable regulated flows for hydropower generation downstream. A Feasibility Report prepared in 2003 (NORPLAN, 2011; Norconsult and Associates, 2003) considered a 20 cm water level increase with the refurbished Kamuzu Barrage, which would lead to a regulation range of to masl. Historical records indicate that Lake Malawi levels have varied from a low of masl to a high of masl (Parry et al., 2009) a maximum difference of 7.31 m well beyond the Barrage s influence. Also, the water level will not be permanently raised by 40 cm to masl. As such, the Barrage Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

26 will result in a slight increase in the duration (length of time) at which the water level will correspond to the masl. This level is much lower than the natural fluctuations in Lake levels under which the fish species have evolved. Therefore, impact on fisheries and fish resources in Lake Malawi, and to a lesser extent, Lake Malombe and the Upper Shire River is expected to be limited Endangered Fish Species that are Potentially Impacted by the Proposed Kamuzu Barrage Upgrading Apart from the Chambo (tilapia) fish stock which have tremendously declined in recent years, Opsaridium microlepis (Mpasa or Lake salmon), a cyprinid and a lacustrine fish species, is the most commercially sought after riverine species in Malawi. Mpasa is at the verge of extinction in Malawian natural waters due to multiple factors such as use of bad fishing methods e.g. river/stream blocking when fish are migrating for breeding; degradation of the riverine environment due to catchment destruction e.g. by farming and use of chemicals; siltation and in extreme cases drying of rivers and streams under natural climatic conditions. For these reasons, the fish was included in the IUCN Red List of threatened species in 2006 (IUCN, 2006). Although Mpasa has been fished out in nearly all rivers and streams in Malawi, there appears to be some remnants in Liwonde National Park where it is still found in abundance Fishing As explained earlier, fishing in the Shire River, with an exception of the Upper and Lower Shire, is highly uncoordinated especially by the fact that some parts of the middle Shire lie within the protected zone of the Liwonde National Park. Thus, although fishing is still carried out in this area, there is no recorded catch data because it is done illegally and consequently fishers usually do not co-operate with fisheries personnel. The other part of the middle Shire up to Nkula Falls is very remote and not active as far as commercial fishing is concerned. Fishing gear, including: gill nets, long lines, hand lines, mosquito nets, cast nets, scoop nets and fish traps (with or without weirs) are operated in the Shire River System. Dugout canoes and rafts were the common fishing crafts spotted during the current survey. Fishing gear is generally selective in catching fish but its selectivity varies according to the design and behaviour of susceptible fish species. Two mathematical models, Gamma and Logistic were used to determine selectivity and to estimate size-at-first capture for major types of fishing gear in the Shire River System. The gear assessed included: mosquito net, kambuzi beach sine, scoop net and hand line. The major target fish groups were cichlids which comprised 61% of total landings. Sizes at capture ranged from 75 mm Total Length (TL) to 101 mm TL. Mosquito nets and kambuzi beach seines caught cichlids at smaller sizes at capture of 75 mm and 87.5 mm TL, respectively while hand lines and scoop nets landed cichlids at slightly bigger sizes of 93.4 mm and 101 mm TL, respectively Catch Biomass by Species and Gear Figures 23 to 26 show species catch biomass according to the fishing gear that was used. As a multi-species fishery, a variety of gear is deliberately employed to catch different species and sizes of fish. Figure 30 shows that a majority of the fish species that are caught in beach seines are usually the small cichlids also known as kambuzi. Placidochromis, Protomelas and Pseudotropheus species are all categorized as kambuzi i.e. small cichlids shown in Figure 23. An exception is Oropharynx argyrosoma, which is a relatively small species often classified under the kambuzi fishery, the rest are generally bigger species. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

27 Figure 23. Species catch biomass for the Upper Shire River caught using Hand line and larger cichlids caught using Hand line from the Upper Shire (Photo: F. Kapute) This agrees with the fact that a hand line is highly selective. Also the small cichlid (kambuzi) fishery is not well pronounced in the Upper Shire in the south east arm of Lake Malawi as is the case in Lake Malombe. Figure 24 presents a very good parallel attribute of two types of gear selective (gill net) and non-selective (mosquito net). In the mosquito net, there is a combination of both large and small cichlids while catches in gill nets are large cichlids and other large non-cichlid species such as Bagrus, Mormerids and Clarids. At the Barrage in Liwonde, scoop nets are extensively and intensively employed to catch different species of fish. Though partly selective, scoop nets usually catch bigger fish that swim with the strong water current when the Barrage gates have been opened. Figure 24. Species catch biomass for the Upper Shire River caught using Hand line and larger cichlids caught using Hand line from the Upper Shire Figure 25. Species catch biomass for Lake Malombe caught using Scoop net (Photo: F. Kapute) Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

28 Apart from the cichlids, many riverine species such as Opsaridium microlepsi (Mpasa), Opsaridium microcephalus (Sanjika), and Barbus litamba (Thamba) are caught by scoop netters at the Barrage. This is the only site where we managed to sample many cyprinid species suggesting that indeed most of the fish caught around the Barrage actually migrate from the protected Liwonde National Park. Hand line catches at the ESCOM Hydro Electricity Power generation intake at Nkula Falls generally comprised of A. calliptera (Figure 26), a beautiful "haplochromine" species found amongst vegetated areas in shallow waters ( Other species included Oreochromis placidus which is actually a Lower Shire River tilapia ( Tilapia rendalli a Red breasted Tilapia which prefer well-vegetated water along river littorals or backwaters, floodplains and swamps ( and O. Microcephalus a riverine species. Figure 26. Species catch biomass for Nkula Falls ESCOM Intake Dam caught using Hand line (Photo: G. Steeves - In summary, most of the fish catches are from gill nets followed by mosquito nets then beach seines Gear selectivity Size distribution and size at capture of cichlids caught in kambuzi beach seines is shown in Figure 27. Size ranged from 30 mm to 180 mm TL with an average of 91 mm ± 40 mm TL. The size-at-capture was estimated at 87.5 mm TL using the logistic selection ogive. Figure 27. Beach seines Size distribution, logistic selection ogive and size at capture of Cichlids caught in A logistic selection ogive or curve is a graphical estimation of the size (length) at which 50% of that fish stock is subjected to fishing. Size distribution and size at capture of cichlids caught in mosquito nets is shown in Figure 28. Size ranged from 30 mm to 120 mm TL with an average of 75 mm ± 24 mm TL. The size-at-capture was estimated at 75 mm TL using the logistic selection ogive. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

29 Figure 28. Mosquito nets Size distribution, logistic selection ogive and size at capture of Cichlids caught in A small difference in the size distribution ranges and estimated size at capture for the cichlids caught in both beach seines and mosquito nets was observed supporting the fact that even the beach seines use a very small meshed net that scoops almost everything within its coverage area Fish Species Composition and Abundance Like Lake Malawi, the fishery in the Shire River system which includes Lake Malombe is multi-species (Table 2). In this study, fish samples were collected from a variety of fishing gear including mosquito nets, gill nets, scoop nets, cast nets and hand lines. TABLE 2. FISH FAMILIES AND GENERA SAMPLED DURING THE SURVEY Family Cichlidae Bagridae Clariidae Cyprinidae Mochokidae Mormyridae Genus Astatotilapia Bagrus Clarias Barbus Synodontis Mormyrus Aulonocara Engraulicypris Buccochromis Labeo Chilotilapia Opsaridium Copadichromis Corematodus Ctenopharynx Dimidiochromis Hemitaeniochromis Lethrinops Mylochromis Oreochromis Otopharynx Placidochromis Protomelas Pseudotropheus Stigmatochromis Tilapia Trematocranus Over 40 species belonging to 27 genera and six families were sampled in Lake Malombe, Upper and Middle Shire River (Table 2, Figures 29-31). The six families included Cichlidae, Clariidae, Cyprinidae, Mochokidae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

30 and Mormyridae. The most important in terms of distribution by weight were the families Cichlidae (61%), Clariidae (17%) and Mormyridae (10%) while the least important were the families Bagridae and Mochokidae with 4% and 1%, respectively (Figure 29). The family Cichlidae was the most diverse with 19 genera followed by the family Cyprinidae with four genera. All other families were represented with one genus each. It is clear from Table 2 and Figures 29 to 31 that most of the catch in the Shire River system as is the case with Lake Malawi is dominated by cichlids (i.e. fish that brood their eggs in the mouth also known as mouth brooders) comprising about 61% of the total catch. In the riverine environment, cyprinids also make a good composition in the catch. Figure 29. Distribution of fish families by % weight sampled during the survey At genus level, the most important in terms of weight were Clarias (17%), Oreochromis (15%), Mormyrus (10%), Protomelas (9.8%), Otopharynx (9%) and Astatotilapia (6%) while the least important were Corematodus (0.1%), Aulonocara (0.2%), Engraulicypris (0.9%) among others. Some of the least commercially important genera like Corematodus and Stigmatochromis are also quite rare naturally. Figure 30. Most abundant (L) and least abundant (R) fish genera sampled during the survey It should be emphasised that in Figure 30, the Oreochromis species comprise the Shire River tilapia (Oreochromis shiranus) and the three variants of tilapia Oreochromis karongae, Oreochromis squamipinnis and Oreochromis lidole found both in Lakes Malawi and Malombe as well as some parts of the Upper and Middle Shire River. In terms of fishery importance, the majority of the catch in the Shire River System is landed from species like Clarius gariepinus, Oreochromis spp., Mormyrus longirostris and small cichlids such as Otopharynx argyrosoma, Protomelas spp. and Astatotilapia calliptera (Table 2). Some species, like those listed in Table 3, were sampled in low numbers in the River System but their contribution to total landings in Lake Malawi is quite enormous. Species like Engraulicypris sardella, Copadichromis spp., Buccochromis lepturus, Synodontis njassae and Lethrinops species dominate fish landings in both the large and small scale fisheries of Lake Malawi. After the collapse of the Chambo (Tilapia) fishery in Lake Malombe and the south-east arm of Lake Malawi due to heavy fishing pressure, the catch is mainly dominated by the small cichlids which are known as kambuzi (Figure 32) hence the versatile kambuzi fishery of Lake Malombe. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

31 Figure 31. Most abundant (L) and least abundant (R) fish species sampled during the survey Figure 32. Small cichlid (kambuzi) catch from south east arm of Lake Malawi caught using a small meshed beach seine (Photo: F. Kapute) The kambuzi fishery is characterised by use of destructive small meshed gear such as beach seines and mosquito net patched purse seines (chilimira and nkacha) to increase the catch effort Size Distribution by Species An analysis of the figures and length data in Figure 33 demonstrates the fact that nearly all of the cichlids caught in the Shire River system are small sized probably for two reasons: firstly, being an indication of overfishing where small meshed gear such as mosquito nets dominate, resulting in the removal of larger sizes; secondly, is the fact that the species are inherently small sized such as Pseudotropheus elegans and Allistotilapia calliptera, Oropharynx argyrosoma, Placidochromis subocularis and many others. Astatotilapia calliptera Size range: mm TL Average 80.6±14.0 mm TL Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

32 Copadichromis chrysonotus Size range: mm TL Average: 68.3±17.6 mm TL Copadichromis species Size range: mm TL Average: 83.7±11.2 mm TL Ctenopharynx intermedius Size range: mm TL Average: 73.8±20.6 mm TL Ctenopharynx nitidus Size range: mm TL Average 82.2±63.2 mm TL Engraulicypris sardella Size range: mm TL Average: 68.4±11.0 mm TL Hemitaeniochromis oxyrhynchus Size range: mm TL Average: 106.2±6.8 mm TL Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

33 Opsaridium microcephalus Size range: mm TL Average: 109.5±26.6 mm TL Oreochromis species Size range: mm TL Average: 239.2±69.2 mm TL Otopharynx argyrosoma Size range: mm TL Average: 78.3±26.7 mm TL Placidochromis subocularis Size range: mm TL Average: 83.4±31.6 mm TL Protomelas similis Size range: mm TL Average: 96.8±24.6 mm TL Protomelas species Size range: mm TL Average: 120.9±31.3 mm TL Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

34 Pseudotropheus elegans Size range: mm TL Average: 55.9±7.7 mm TL Figure 33. Length frequency histograms of different species 3.17 Distribution and Size at Capture of Major Fish Groups Although six fish families were sampled during the study, only cichlids occurred in sufficient numbers to warrant analysis of gear selectivity. Therefore the analysis that follows focused on cichlid fishes only. The size distribution and selectivity pattern of Cichlids are shown in Figure 34. Size distribution for cichlids caught in Scoop nets ranged from 60 mm to 315 mm TL with an average of 114 mm ± 57mm TL. The size-atcapture was estimated at 101 mm ± 57mm TL using the gamma distribution function of selectivity (Figure 34). Figure 34. respectively Distribution and size at capture of cichlids caught using scoop nets and hand line Size distribution of cichlids caught in Hand lines ranged from 60 mm to 230 mm TL with an average of 105 mm ± 36 mm TL. The size-at-capture was estimated at 93.4 mm ± 28 mm TL using the gamma distribution function of selectivity (Figure 34). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

35 4 Discussion 4.1 The Shire River Aquatic Environment The shoreline for the Shire River system as observed during the field survey is generally covered by vegetation all the way from the outflow point of Lake Malawi (Figure 35L). This is different from Lake Malawi, where the shoreline has clear and sometimes wide sandy beaches (Figure 35R). Breeding of the cichlid fishes that make nests and spawn their eggs along the sandy beaches is therefore more pronounced in Lake Malawi than in the Shire River. The latter favours riverine species more than all cichlids in general. Figure 35. Shire River (L) and Lake Malawi (R) marshy and sandy shorelines respectively (Photos: F. Kapute) It should also be emphasised that the sand bar at the start of the Shire River at the south east arm of Lake Malawi plays a key role on how much water moves into the River from the Lake. There have always been fears of the Shire River and Lake Malombe drying up due to the cessation of flows from Lake Malawi in an event of extreme low water levels in the lake as occurred from 1908 to 1935 when outflow was near zero for about 27 years (Parry et al., 2009). The abundant presence of vegetation along the banks of the Shire River influences fishing in that it limits the operation of beach seines commonly used in clear sandy beaches of Lake Malawi and Lake Malombe. Gill nets and other smaller traditional fishing gear such as hand line, scoop nets etc. are therefore used extensively in the Shire River. 4.2 Fish and Fisheries in the Shire River System The fish community of the Shire River including Lake Malombe is typical of that of Lake Malawi comprising key genera such as Lethrinops, Otopharynx, Copadichromis, Oreochromis, Engraulicypris, Clarias, Synodontis, Opsaridium, Barbus among many others (Table 3). The list in Table 3 is comprehensive and covers the entire Shire River system fish species above and/or below the Kapichira and other falls that serve as a bio-geographic divide between the Lake Malawi and the Zambezi River system fish fauna. Most of the fish species in Table 3 are demersal (bottom feeders) and feeding high in the trophic level - a niche that is unlikely to be significantly affected by water levels. Also, most of the species are Cyprinids hence increasing water levels may not affect breeding. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

36 TABLE 3. SHIRE RIVER FISH SPECIES COMPOSITION Species Family Local name Habitat Status Amphilius platychir Amphiliidae? Demersal Native Anguilla nebulosa Anguillidae Mkunga Demersal Native Aplocheilichthys johnstoni Poeciliidae? Benthopelagic Native Aplocheilichthys katangae Poeciliidae? Benthopelagic Native Astatotilapia calliptera Cichlidae? Benthopelagic Native Barbus afrohamiltoni Cyprinidae Matemba Benthopelagic Native Barbus arcislongae Cyprinidae Matemba Benthopelagic Native Barbus atkinsoni Cyprinidae Matemba Benthopelagic Native Barbus johnstonii Cyprinidae Ngumbo Benthopelagic Native Barbus kerstenii kerstenii Cyprinidae Matemba Benthopelagic Native Barbus lineomaculatus Cyprinidae Matemba Benthopelagic Native Barbus macrotaenia Cyprinidae Matemba Benthopelagic Native Barbus microcephalus Cyprinidae Matemba Benthopelagic Native Barbus paludinosus Cyprinidae Matemba Benthopelagic Native Barbus radiatus aurantiacus Cyprinidae Matemba Benthopelagic Native Barbus trimaculatus Cyprinidae Matemba Benthopelagic Native Barbus viviparus Cyprinidae Matemba Benthopelagic Native Brycinus imberi Alestiidae Nkhalala Demersal Native Chiloglanis neumanni Mochokidae? Benthopelagic Native Chilotilapia rhoadesii Cichlidae Gundamwala Benthopelagic Native Clarias gariepinus Clariidae Mlamba Benthopelagic Native Clarias ngamensis Clariidae Mlamba Demersal Native Clarias theodorae Clariidae Mlamba Demersal Native Cyathochromis obliquidens Cichlidae? Benthopelagic Native Dimidiochromis kiwinge Cichlidae Mbaba Benthopelagic Native Dimidiochromis strigatus Cichlidae Mbaba Benthopelagic Native Hemigrammopetersius barnardi Alestiidae? Pelagic Native Hemitilapia oxyrhyncha Cichlidae Mbaba Benthopelagic Endemic Hippopotamyrus discorhynchus Mormyridae Mphuta Demersal Native Hydrocynus vittatus Alestiidae? Demersal Native Irvineia orientalis Schilbeidae? Demersal Native Labeo altivelis Cyprinidae? Benthopelagic Native Labeo cylindricus Cyprinidae Ningwi Benthopelagic Native Leptoglanis rotundiceps Amphiliidae? Demersal Native Malapterurus shirensis Malapteruridae? Benthopelagic Native Marcusenius macrolepidotus Mormyridae? Demersal Native Marcusenius nyasensis Mormyridae? Demersal Native Micralestes acutidens Alestiidae? Pelagic Native Mormyrops anguilloides Mormyridae? Demersal Native Mormyrus longirostris Mormyridae Samwamowa Demersal Native Opsaridium ubangiensis Cyprinidae? Benthopelagic Native Opsaridium zambezense Cyprinidae? Benthopelagic Native Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

37 Species Family Local name Habitat Status Oreochromis mossambicus Cichlidae Makakana Benthopelagic Native Oreochromis placidus placidus Cichlidae Chambo Benthopelagic Native Oreochromis shiranus chilwae Cichlidae Chambo Benthopelagic Native Oreochromis shiranus shiranus Cichlidae Makumba Benthopelagic Native Otopharynx tetraspilus Cichlidae? Demersal Native Otopharynx tetrastigma Cichlidae Kambuzi Demersal Native Protomelas kirkii Cichlidae Kambuzi Benthopelagic Native Protomelas triaenodon Cichlidae Kambuzi Benthopelagic Native Protopterus annectensbrieni Protopteridae? Demersal Native Pseudocrenilabrus philander Cichlidae? Benthopelagic Native dispersus Serranochromis robustus jallae Cichlidae Tsungwa Demersal Native Synodontis nebulosus Mochokidae? Benthopelagic Native Synodontis njassae Mochokidae Nkholokolo Benthopelagic Native Tilapia rendalli Cichlidae Chilunguni Benthopelagic Native Trematocranus placodon Cichlidae Mbaba Demersal Native Source: Most of the Cyprinid fishes are free spawners, that discharge eggs into the open water rather than building nests as is the case with Cichlids. Later the eggs will adhere to whatever substrate they happen to land on. Therefore, though higher water levels upstream of Kamuzu Barrage would have an impact of inundation of riparian vegetation and river sandbars within Liwonde National Park, this will not impact on the breeding of most fish species in the Park. This is not surprising considering that the Shire River is the only outlet of Lake Malawi and it is therefore essentially one system implying that the fish community is also expected to be homogenous. Information gathered through literature search and results from the field survey support the fact that the fishery in the Shire River aquatic system like Lake Malawi is multi-species. While literature suggests that more than 50 fish species are found in the Shire River system ( about 40 species were recorded during the field survey, a majority of these being cichlids (mouth brooders). Conspicuously missing were riverine species (Cyprinids) save a few specimens of Opsaridium microcephalus, Barbus paludinosus and Barbus litamba identified at the Barrage from fishers and believed to be illegally caught from the protected Liwonde National Park. No specimen of Opsaridium microlepis (Mpasa) was recorded from the Shire River explaining the ever increasing scarcity of this commercial IUCN Red Listed species. However it is likely that some fish species were under-sampled or entirely missed out because of limitations in efficiency and selectivity inherent in fishing gears from which fish were sampled. The Shire River s beaches are generally covered with more vegetation from Lake Malombe where more cichlids than riverine species are found. While riverine species favour such aquatic environment, presence of more cichlids is due to fish migration from the south east arm of Lake Malawi to Lake Malombe (which is part of the Shire River system) through the Upper Shire River. Catches from the southern part of the overexploited Lake Malombe appear better than the rest of the areas of the Lake suggesting yet more migration of fish from the Liwonde National Park that borders the southern part of Lake Malombe (FAO, 2005). It was clear from the findings that enforcement of fisheries management by the Malawi Department of Fisheries is facing challenges because, despite there being a closed season for Lake Malombe, illegal fishing was still commonly observed. The study concluded that the Liwonde National Park is an important breeding area for fish that migrate and are caught by fishers in the southern part of Lake Malombe around Mvera area (also as reported earlier by FAO, 2005) and the stretch of the Shire River down to the Barrage at Liwonde. Protecting the fishery in the Park would therefore have far reaching advantages to the fisheries especially in the Middle Shire River and southern part of Lake Malombe. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

38 4.3 Possible Impacts of the Upgraded Kamuzu Barrage at Liwonde on Fish and Fisheries in Lakes Malawi and Malombe and the Shire River Impediment to Fish Migration Riverine fish species are generally migratory species (Welcomme 1985) mainly for breeding purposes. In the Shire River, important and commercially valuable migratory species include Opsaridium microlepis (Mpasa) and several Barbus species found in the Upper and Middle Shire ( Mpasa has been included under the IUCN Red List of endangered species. Physical barriers eliminate diadromous or obligate migrating fish species by preventing movement to upstream breeding sites by adults but also slowing downstream movements of juveniles. The fish fauna of the Upper and Lower Shire River are separated by natural barriers (waterfalls) downstream of the existing Barrage at Liwonde. As a result, fish species are very distinct and endemic to each geographical area. Most of the species in the Upper Shire are also found in the south east arm of Lake Malawi and Lake Malombe probably due to migration. In the Lower Shire, the fish fauna is essentially of Zambezi River Basin eco-region as more than 60 species are caught in this fishery. Out of the 60 species, only three species are of commercial importance namely, Mlamba (Clarias gariepinus), Chikano (Clarias ngamensis) and Mphende (Oreochromis mossambicus) (Willoughby and Tweddle, 1978). From the outflow of Lake Malawi near Mangochi downstream to Lake Malombe, the Shire River flows through a relatively flat area without any obstruction. The natural barriers to fish migration in the River are the water falls at Nkula downstream to Kapichira Falls. The distance between the Barrage (Liwonde) to the first natural barrier (falls) at Nkula is about 80km. The gates at the Barrage at Liwonde when fully open are not an impediment to migration of fish. That is why fishermen catch a lot of migrating fish at the Barrage using Scoop and Cast nets when the gates are open. As explained earlier, the Liwonde National Park is an important refuge and breeding area for fish in the Shire River and Lake Malombe since the rest of the areas are subjected to intensive and uncontrolled fishing. Since the Kamuzu Barrage is located downstream of Liwonde National Park, it does not interfere with any fish migrations in the Shire River between the Park and Lake Malawi. Further, the upgraded Kamuzu Barrage would pose essentially the same type of physical barrier to fish movements as the existing one; hence no change from the status quo is expected with respect to fish migrations. The existing natural impediment (falls) at Nkula and Kapichira downstream will continue to block migration of fish from the Lower Shire to the Upper Shire. It is important to note nevertheless that downstream migration of the fish will still occur as the gates are often fully open under low flow and high flow conditions. Upstream migration may also be possible during these times since by their nature, fish prefer to swim against the water current. One way of facilitating migration of fish in the Shire River past the Barrage is construction of fish passes also known as fish ladders. However, any benefit from providing fish passage is limited to the 80km stretch of the Shire River between the Barrage and Nkula falls. The following section provides a discussion on existing fish passage experience in Malawi. Fish passes in Malawi Unpublished information sourced from the Malawi Department of Fisheries indicates that fish passes in Malawi have previously been constructed in several places though only limited data is available. In Bua River (Central Malawi), fish passes were constructed within the context of an environmental impact assessment (EIA) requirement for rice cultivation along the lower reaches of the River. The fish of interest here was the Lake Salmon Opsaridium microlepis (Mpasa). Other fish passes have been constructed at Mulunguzi Dam (on Zomba Mountain) for trout species. It is also reported that fish passes were constructed at Kapichira Falls down the Shire River by ESCOM. However, in all these reported cases there was not a detailed assessment of the effectiveness of the fish passes and information remains scanty (personal communication from the Malawi Department of Fisheries). Also one of the major disadvantages of existing fish passes is that they expose fish to being easily captured. The experience of uncontrolled blocking of rivers to catch breeding riverine fish in Malawi which has resulted in the elimination particularly of Mpasa (Opsaridium microlepis Lake Salmon), presents a great challenge of the effectiveness of fish passes. Linthipe River in the Central Region of Malawi which used to support large stocks of Mpasa is a good example. Fishers have been uncontrollably blocking the river catching migrating breeding Mpasa resulting in collapse of the Mpasa fishery there. Therefore, unless followed by a strong Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

39 fisheries management and enforcement mechanism, fish passes would create a more conducive environment for complete exploitation of already declining riverine fish stocks and mostly the IUCN Red Listed Opsaridium microlepis (Mpasa). The Food and Agriculture Organization (ftp://ftp.fao.org/docrep/fao/010/y4454e/y4454e01.pdf) has previously demonstrated use of the following fish passes outside of Malawi. i) Bottom ramp and slope A sill having a rough surface and extending over the entire river width with as shallow a slope as possible, to overcome a level difference of the river bottom. This category also includes stabilizing structures (e.g. stabilizing weirs), if the body of the weir has a shallow slope similar to the slope of a ramp or slide and is of loose construction ii) Bypass channel A fish pass with features similar to those of a natural stream, bypassing a dam. The whole impounded section of the river can thus be bypassed. iii) Fish ramp A construction that is integrated into the weir and covers only a part of the river width, with as gentle a slope as possible to ensure that fish can ascend. Independently of their slope, they are all called ramps; in general the incorporation of perturbation boulders or boulder sills is required to reduce flow velocity Destruction of Fish Breeding Aquatic Habitats through Flooding and Siltation Physical infrastructure such as barrages have been reported to cause habitat loss (Welcomme, 1985). Turbidity and siltation of the water increases after the water passes through the barrage gates ( 2011). Sedimentation could entail burying already made nests for spawning of the cichlids such as the Tilapine family found in the river course. Increase in silt load also accelerates the natural evolutionary processes of the river system resulting in choking of the substrate rendering it unsuitable for spawning by species requiring well aerated flows and clear pebble or gravel bottoms (Welcomme 1985). Opsaridium microlepis (Mpasa) appears to be one of such species ( 2011). Heavy siltation could also affect the breeding of fish by destroying breeding nests for the spawning cichlids such as Oreochromis shiranus along the shoreline of the Shire River. Siltation is also a problem to reproductive success of fish species such as O. microlepis which requires silt free water to reproduce ( However, water released from a dam or weir is usually low in sediment because sedimentation in the reservoir is dependent on residence time and flow rates within the reservoir. Therefore, the problem of destruction of fish breeding aquatic habitats through flooding and siltation due to the upgraded Barrage is unlikely. Further, riverine fishes generally migrate upstream to breed during the rainy season when water levels are already high. This is a time when the Barrage s influence at Liwonde on the water flow is least. Also, most fish species in the Shire River (Table 2 and 3) are riverine which do not build breeding nests but eggs that attach to substrates such as grass, rocks etc. Inundation of breeding areas as a result of the Barrage is therefore unlikely to be an issue. The upgraded Barrage will also not release higher regulated flows than the existing situation. In any case, the Barrage has no influence over natural flood flows in the Shire River system they are beyond the operating level of the Barrage. Also any impact such as siltation will be short-lived and minor due to the potential dilution properties of the Shire River Changes in Water Flow Downstream Operation of the Barrage gates at Liwonde has a direct effect on the flow of water in the Shire River downstream. In the Lower Shire, the River has numerous meanders hence reduced water flow may result in the progressive restriction of the stream to a smaller bed within the original channel leading to loss of habitats for fish. Riverine aquatic organisms are usually adapted to particular water flow patterns (Welcomme, 1985) explaining why riverine fish species usually breed during flooding regimes which are also associated with more food availability and when most of the fish become sexually ripe. Sudden or periodic increased flow of Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

40 water as a result of opening of the gates at the upgraded Barrage may trigger sexual maturity and consequently spawning of fish downstream especially the Lower Shire. This may eventually change the breeding pattern (timing) for other species of fish. Water release patterns from the upgraded Kamuzu Barrage that would be generally favourable to downstream fish and fisheries include: i) allowing some natural flooding to occur during the wet season, to stimulate natural fish reproduction and to maintain the vitality of the Elephant Marsh and other areas with important fish populations; ii) avoiding significant daily fluctuations in flow rates, such as for peaking power (which can upset the natural cycles of fish and other aquatic life); and iii) avoiding very sudden decreases in flows due to rapid gate closures at the Barrage since such sudden drops can cause large-scale fish stranding. However, as explained earlier, the upgraded Barrage will not result in a reduction of flow volume downstream but only a change in the flow pattern. Also, likely downstream impacts of the upgraded Kamuzu Barrage will depend on how water releases at the Barrage are managed, and whether there will be any appreciable difference from the status quo water release regime for the existing Barrage. The Barrage also acts as a facility to regulate flow in the Shire River so that there is no change in the total volume of water but a change in the timing of flows, i.e. there will be slightly less flow during the high flow season as more water is stored behind the Barrage and then slightly more flow during the low flow season as more water is released from the Barrage. There will therefore be no change in the total volume of water released downstream by the Barrage, only a change in the timing and magnitude of the releases. Similarly, fishing activities will also to that effect not be affected in the Lower Shire River since water levels will not be significantly affected Impacts on Fish of Lake Malawi Particularly the Rock Dwelling Cichlids (Mbuna) In Lake Malawi, Mbuna (beautifully coloured aquarium) fish species (Figure 36) are very common in Cape MaClear near Monkey Bay, the area between the south-east and south-west arms of Lake Malawi (Figure 2). Figure 36. Beautifully coloured rock dwelling Cichlids (Mbuna) ( This area of the Lake is part of the protected Lake Malawi National Park. The Mbuna are more dominant in Cape MaClear where the Lake is deep and has a rocky bottom which favour Mbuna s feeding on the rocky substrates. Though Mbuna have varied depth differences extending from the surface to more than 40 m (Lewis, Reinthal and Trendall, 1986), its population in the south east arm of Lake Malawi is insignificant. During the field survey, not even one was found (Table 2 and figure 33) in south east arm of Lake Malawi, Lake Malombe and the Shire River. Therefore, as far as the south east arm of Lake Malawi (Area A) is concerned, the threat of the Mbuna species being impacted by the upgraded Barrage is not valid Impacts of Construction Activities on Fish and Fisheries Resources Clearing of vegetation and earth moving works during the upgrading of the Barrage is likely to cause accelerated erosion in the project area with sediments contaminating water in the Shire River and silting up spawning grounds on the river bed. Solid wastes and oils from construction plant and sites may spill into the Shire River thereby contaminating its waters and likely kill some aquatic organisms including fish. These impacts are expected to be short term and minor given the potential dilution properties of the Shire River. Thus such water pollution in the River will be localised and not affect the entire Shire River system. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

41 Table 4 presents a summary of the potential environmental impacts, their sources, areas of impact, potential mitigation measures and brief description of the occurrence of the impact and mitigation. This summary is a synthesis of the results from both the desk study and the field surveys (Appendix 1). TABLE 4. Impact Impediment/blocking of fish migration Destruction of fish breeding aquatic habitats through siltation SUMMARY OF IMPACTS AND SUGGESTED MITIGATION MEASURES/PLANS Type of Impact Negative Negative Source of Impact Impact Area Mitigation Measure / Comments The Barrage and the action of closing the gates at the Barrage could eliminate diadromous or obligate migrants by preventing movement to upstream breeding sites by adults and slowing down- stream movements of juveniles. However, the gates when fully open are not an impediment to migration of fish. That is why fishermen catch a lot of fish at the Barrage using Cast nets when the gates are open. Increase in turbidity and siltation of the water after passing through the Barrage gates. Sedimentation thus means burying already made nests for spawning of the cichlids. Increase in silt load accelerates the natural evolutionary processes of the river system resulting into choking of the substrate rendering it unsuitable for spawning by species requiring well aerated flows and clear gravel bottoms. Siltation is also a problem to reproductive success of fish such as O. Microlepis (Mpasa) which requires silt free water to reproduce. *At the Barrage, *Upstream breeding areas, *Immediate downstream as far as Nkula (80km) where there is a natural barrier to fish movement upstream Upstream fish breeding areas Downstream potential fish breeding areas Constructing fish passes Fish passes or ladders have satisfactorily worked for salmon in northern temperate rivers. In the tropics, fish passes have successfully demonstrated to be useful. Fish ways at dams on the Tigris and Euphrates rivers have also been used successfully by many of the migratory cyprinid species in these rivers. Information about fish passes in Malawi is scanty, However, fish passes expose fish to easy way of being captured. The experience of uncontrolled blocking of rivers to catch breeding fish in Malawi which has resulted in wiping out of riverine species presents a great challenge of the effectiveness of the fish passes. Riverine fishes generally migrate upstream to breed during rainy season when water levels are already high. However, this is when the Barrage has the least influence on water flow. Thus, increased water levels cannot affect breeding. But fish are sensitive to habitat changes, therefore proper gate control and enforcement of fishing regulations especially for breeding sanctuaries (e.g. Liwonde National Park) would to protect the Red listed species. The section of the Shire River at Liwonde does not have sandy beaches that favour breeding of cichlids. Thus, the area does not naturally favour the breeding of cichlids suggesting why the Chambo (tilapia) fishery is not vibrant in the Shire Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

42 Impact Changes in flow of water downstream Disappearance of fish spawning habitats through flooding Type of Impact Negative / Positive Negative / Positive Source of Impact Impact Area Mitigation Measure / Comments River. Another mitigation measure (though it is high cost) is improvement of the Spawning areas such as making artificial breeding beaches for fish. This is improvement of bottom texture for fish which spawn over gravel, particularly where siltation has resulted in the loss of the normal substrate. However, most fish species are riverine which do not build breeding nests but eggs attach to substrates such as grass, rocks etc. hence the issue of inundation of breeding areas is not valid. There will be no change in the total volume of water released downstream by the Barrage, only a change in the timing and magnitude of the releases The upgraded Barrage will not release higher flows than the existing situation. In any case, the Barrage has no influence over natural flood flows in the Shire River system they are beyond the operating level of the Barrage Barrage downstream and the Lower Shire Zone Both downstream and upstream Operation of the gates at Kamuzu Barrage to maintain downstream aquatic environment. Closing and opening of gates should be monitored and balanced i.e. releasing sufficient water downstream to meet hydropower, town water supply and irrigation needs as well as meeting environmental requirements. This could be a positive impact in a way that less or too much water tends to favour different species of fish. In fact, flooding may trigger spawning among some riverine species including catfish (Clarias spp.) Most fish species are riverine which do not build breeding nests but eggs attach to substrates such as grass, rocks etc. hence the issue of inundation of breeding areas is not valid. Flooding would create new breeding areas for the fish hence may also be a positive impact. Only siltation is a problem not flooding as far as breeding of riverine species and cichlids is concerned. Increased water levels would actually make Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

43 Impact Loss of fish landing beaches and fish drying areas upstream due increased water levels Reduced fish activities downstream due to low water levels Changes in water quality Type of Impact Negative Source of Impact Impact Area Mitigation Measure / Comments breeding of riverine species much easier i.e. by facilitating migration up stream. Some beaches may be inundated for a slightly longer time because of the Barrage. Barrage upstream and Upper Shire Lake Malawi water levels fluctuate naturally much greater than the operating level of the Barrage and the proposed 40cm increase in maximum regulated height. Sandy beaches are inundated by natural flows irrespective of Barrage operation. Negative Barrage downstream The total amount of water going down the system will not change as a result of upgrading the Barrage. Negative Downstream flooding, upstream water retention, construction may affect the physical and chemical properties of water e.g. ph, conductivity etc which may eventually affect the physiology of fishes Both upstream and downstream Constant flushing of the system i.e. the opening and closing of Barrage gates. As regards construction, this is not a long term problem as the system would stabilise sometime after construction is completed. 4.4 Mitigation of Adverse Impacts Analysis of these findings suggests that potential impacts will be impediment of migration, siltation, destruction of aquatic habitats-reproductive and schooling habitats, destruction of fish processing/drying areas upstream, reduced catches especially of riverine and Red-listed species such as the Mpasa especially downstream due to reduced water levels, flooding and water quality deterioration. However, these potential impacts from the upgraded Kamuzu Barrage at Liwonde are not anticipated to occur, or if they do occur will only be minor in nature for the reasons listed below: i) The upgraded Kamuzu Barrage would pose essentially the same type of physical impediment to fish movements as the existing one; hence no change from the status quo is expected with respect to fish migrations. ii) Downstream migration of the fish will still occur as the gates are fully open under extreme low flow and high flow conditions. iii) Upstream migration may also be possible during opening of gates since by their nature, fish prefer to swim against the water current. iv) One suggested way of facilitating migration of fish in the Shire River passed the Barrage apart from the opening of the gates would be construction of a fish pass. However, success of a fish passage facility at the Barrage is questionable because fish passes expose fish to easily being captured. Its effectiveness can only be guaranteed with strict fisheries management and enforcement. Also, riverine fishes generally migrate upstream to breed during rainy season when water levels are already high. This is a time when the Barrage s influence at Liwonde on the water flow is least. v) On destruction and inundation of fish breeding grounds: Most fish species in the Shire River are riverine which do not build breeding nests but eggs that attach to substrates such as grass, rocks etc. The issue of inundation of breeding areas is therefore not valid. vi) Any impact from increased siltation arising from construction activities will be short-lived and minor due to the potential dilution properties of the Shire River. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

44 vii) On changes in water flow in the Shire River: The upgraded Barrage will also not release higher flows than the existing situation. The Barrage has no influence over natural flood flows in the Shire River system they are beyond the operating level of the Barrage. viii) Water release patterns from the upgraded Kamuzu Barrage should be managed so they are generally favourable to downstream fish and fisheries such as: (i) allowing some natural flooding to occur during the wet season, to stimulate natural fish reproduction and to maintain the vitality of the Elephant Marsh and other areas with important fish populations; (ii) avoiding significant daily fluctuations in flow rates which can upset the natural cycles of fish and other aquatic life; and (iii) avoiding very sudden decreases in flows due to rapid gate closures at the Barrage. Such sudden drops can cause large-scale fish stranding. ix) The Barrage acts as a facility to regulate flow in the Shire River so that there is no change in the total volume of water but a change in the timing of flows, i.e. there will be slightly less flow during the high flow season as more water is stored behind the Barrage and then slightly more flow during the low flow season as more water is released from the Barrage. There will therefore be no change in the total volume of water released downstream by the Barrage, only a change in the timing and magnitude of the releases. x) Similarly, fishing activities will also to that effect not be affected in the Lower Shire River since water levels will not be significantly affected. xi) This area of the Lake is part of the protected Lake Malawi National Park. The Mbuna are more dominant in Cape MaClear where the Lake is deep and has a rocky bottom which favour Mbuna s feeding on the rocky substrates. Though Mbuna have varied depth differences extending from the surface to more than 40 m, its population in the south east arm of Lake Malawi is insignificant. During the field survey, not even one was found in south east arm of Lake Malawi, Lake Malombe and the Shire River. Therefore, as far as the south east arm of Lake Malawi (Area A) is concerned, the threat of the Mbuna species being impacted by the upgraded Barrage is not valid. xii) Construction impacts such as clearing of the area, water pollution and siltation are expected to be short term and minor given the potential dilution properties of the Shire River. Thus such water pollution in the River will be localized and not affect the entire Shire River system. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

45 5 Conclusion and Recommendation It is concluded, based on careful analysis of findings from this Report, that the project of upgrading the Kamuzu Barrage which may likely result in rising water levels upstream (but within the natural fluctuating water levels) does not pose significant danger to the fish and fisheries of the Shire River System if mitigation measures which are proposed in this report are adopted and implemented. Key observations and recommendations made are as follows: The upstream area between the Barrage, Lake Malombe and the Liwonde National Park, where fish migration takes place is sensitive to water level rise likely to be affected by the upgrading of the Kamuzu Barrage. However, the upgraded Barrage will not cause any significant changes in water levels but rather have the same operation as the present Barrage. Lowering water levels downstream may not prohibit fish breeding but rather reduce the breeding area and hence limit fish catches. On the other hand, increased water levels will simply trigger breeding which is an advantage. Most of the fish caught from areas of the Shire River around the Liwonde National Park appear to migrate from the Park which appears to be an important breeding area. Most fish found in the Shire River especially upstream from the Barrage breed in the Liwonde National Park which is a protected area hence reduced threat to breeding of fish. As long as migration of fish between Lake Malawi and Lake Malombe through the 15 km Upper Shire stretch continues, threats to breeding of the fish upstream will be minimal. Except for the part of the Shire River within Liwonde National Park, elsewhere, management of fish exploitation is very minimal if not non-existent at all. The fisheries for the Shire River except for the Lower Shire and small parts of the Upper Shire appear to be uncoordinated requiring reinforcement of varied participatory monitoring and data collection approaches, owing to its multi-species nature. This will ensure sustenance of fisheries livelihoods while mitigating impacts of the Barrage. As an important aquatic habitat for breeding of fish that migrate to the main part of the Shire River and Lake Malombe, there is need to institute sound fisheries management strategies in Liwonde National Park to monitor recruitment of fish stocks. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

46 6 Reference Material Anseeuw, D., Raeymaekers, J.A.M., Busselen, P., Verheyen, E. and Snoeks, J. (2011). Low Genetic and Morphometric Intraspecific Divergence in Peripheral Copadichromis Populations (Perciformes: Cichlidae) in the Lake Malawi Basin. International Journal of Evolutionary Biology. doi: /2011/ Research. Banda, M., Jamu, D, Njaya, F, Makuwila, M. and Maluwa, A. (eds) (2005). The Chambo Restoration Strategic Plan. WorldFish Center Conference 71 Proceedings, Penang, Malaysia, 112 pp. Bell, R. (1998). The World Bank/WBI s CBNRM Initiative. Booth, A.J. and Punt, A.E. (1998).Evidence for rebuilding in the panga stock on the Agulhas Bank, South Africa. Fish. Res. 34: Bulirani, A. E., M.C. Banda, Ó.K. Pálsson, O.L.F. Weyl, G.Z. Kanyerere, M.M. Manase, and R.D. Sipawe (1999). Fish stock and fisheries of Malawian waters. Resource Report Fisheries Research Unit, Department of Fisheries, Monkey Bay, 55 pp. Butterworth, D.S., Punt, A.E., Borchers, D.L., Pugh, J.G., and Hughes, G.S. (1989).A manual of Mathematical techniques for line fish assessment. S. Afr. Natl. Sci. Pro. Rep., pp 160: 89. Chapalapata, D.A.J. (2007). Problems of Aquatic Weeds and Power Generation in ESCOM. A paper presented at the 17th Africa Hydro Symposium. Chimatiro, S.K. (2004). The Biophysical Dynamics of the Lower Shire River Floodplain Fisheries in Malawi. PhD Thesis. Rhodes University, RSA. Donda, S. and Njaya, F. (1997).Food Security and Poverty Alleviation through Improved Valuation and Governance of River Fisheries in Africa. Fisheries Co-Management in Malawi: An Analysis of the Underlying Policy Process. World Fish Center, Cairo, Egypt. FAO (2005). FAO Fishery Country Profile: The Republic of Malawi. FID/CP/MLW FAO (1993). Fisheries Management in the South-east arm of Lake Malawi, the Upper Shire River and Lake Malombe, with particular reference to the fisheries on chambo (Oreochromis spp.). GOM/FAO/UNDP Chambo Fisheries Research Project. CIFA Technical Paper. No. 21. Rome, FAO p. Fisheries Department (2003). Annual fish production, Lilongwe. GoM (2001). Fisheries and Aquaculture Policy. Ministry of Natural Resources and Environmental Affairs, Department of Fisheries, Lilongwe, Malawi. IUCN (2006) IUCN Red List of Threatened Species. Kapute, F., Kaunda, E., Banda, M. and Morioka, S. (2008). Maturity, age and growth of Oreochromis karongae (Teleostei: Cichlidae) in Lakes Malawi and Malombe, Malawi. African Journal of Aquatic Science 2008, 33(1): 69 76, doi: /AJAS Lewis, D.S.C., Reinthal, P. and Trendall, J. (1986). A Guide to the Fishes of Lake Malawi National Park. WWF, Gland, Switzerland. Liabunya, W.W. (2007). Malawi Aquatic Weeds Management at Hydro Power Plants. International Conference on Small Hydropower - Hydro Sri Lanka, October Malcolm, H. (2001). Modelling and Quantitative Methods in Fisheries. Chapman and Hall, Boca Raton, 406 pp. Mzuza, M.K. (2010). Assessment of the effects of aquatic weeds on production of electricity in Malawi: A case of Nkula Hydro Electric Power Station. BSc Dissertation, Catholic University of Malawi (CUNIMA), Malawi. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

47 NORPLAN (2011). Environmental and Social Assessment Report (ESIA) for detailed design of the Upgraded Kamuzu Barrage. Government of the Republic of Malawi, Ministry of Irrigation and Water Development National Water Development Project II. Norconsult and Associates (2003). The Integrated Water Resources Development Plan for Lake Malawi and Shire River System Lake Malawi Level Control Stage 2, Final Feasibility Report, Vol. III Baseline and Study Process Papers, Part A Environment and Water Resources. Report to Ministry of Water Development by Norconsult and Associates, December Osborne, O.N. (2000). Naturalization of Lake Malawi levels and Shire River flows: Challenges of Water Resources Research and Sustainable Utilization of the Lake Malawi Shire River System. Interconsult (pvt) Limited. 1st WARFSA/Water Net Symposium: Sustainable Use of Water Resources, Maputo, 1-2 November Parry, D., Ecosury and Burton, K. (2009). Impact assessment case studies from southern Africa. Lake Malawi level control Integrated Water Resources Development Plan for Lake Malawi and Shire River System. Southern African Institute of Environmental Assessment (SAIEA). Norconsult, Ministry of Water Development, Malawi. Pike, J.G. (1972). Hydrology. In S. Agnew and M. Stubbs (eds.) Malawi in Maps. University of London Press LTD, pp Ratcliffe, C. (1971). The Fishery of the Lower Shire Area Malawi. Fisheries Bulletin No. 3. Government of Malawi, Ministry of Agriculture, Fisheries Department, Zomba. Ribbink, A.J. (2001) "Lake Malawi/Niassa/Nyasa Ecogion-based Conservation Programme: Biophysical reconnaissance". Harare, Zimbabwe. WWF Southern African Regional Programme Office. Seisay, M. (1992). Population Dynamics and Stock Assessment of Chambo in the South East Arm of Lake Malawi and Lake Malombe. FI:DP/MLW/86/013, Field document 19 in prep. Turner, G.F. (1995). "Management, conservation and species changes of exploited fish stocks in Lake Malawi" In Pitcher, T.J.; Hart, P.J.B. (Ed.). The impact of species changes in African Lakes. (pp ) London, UK: Chapman and Hall. Tweddle, D. and N.G. Willoughby (1979). An annotated checklist of the fish fauna of the river Shire South of Kapichila Falls, JLB Smith Inst. Ichthyol. Bull. 39: Welcomme, R.L. (1995). Relationships between fisheries and the integrity of river systems. Regulated Rivers: Research and management, 11, Welcomme, R.L. (1985). River fisheries.fao Fisheries Technical Paper 262: 330 p. Weyl, O.L.F., Kazembe, J., Booth, A.J. and Mandere, D.S. (2004a). An assessment of the light attraction fishery in southern Lake Malawi. African Journal of Aquatic Science 29: Weyl, O.L.F., Mwakiyongo, K.R. and Mandere, D.S. (2004b). An assessment of the nkacha net fishery of Lake Malombe, Malawi. African Journal of Aquatic Science 29: Willoughby, N.G. and Tweddle, D The ecology of the catfish Clarias gariepinus and Clarias ngamensis in the Shire Valley, Malawi. J. Zool. 186: Willoughby, N.G. & D. Tweddle (1978). The ecology of commercially important species in the Shire valley fisheries, Malawi. In R.L. Welcomme (eds.) Symposium on river and floodplain fisheries in Africa. CIFA Tech. Pap./Doc. Tech. CPCA, 5: World Bank (2007). Second National Water Development Project Appraisal Document, Report No: MW, Malawi. Web Pages (Accessed 23 rd October 2011) Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

48 ftp://ftp.fao.org/docrep/fao/010/y4454e/y4454e01.pdf (Accessed 23 rd October 2011) (Accessed 23 rd October 2011) (Accessed 24 th October 2011) (Accessed 24 th October 2011) Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

49 Appendix 1: Summary of Key Questions 1. Does the project affect species composition and the fishery of the Shire River? YES and/or NO a. Any suggestions/comments proposed to mitigate this impact? YES. b. Some fish species would flourish better in high waters (rheophilic) while other wouldn t (limnophilic species) c. Also migration would stabilize and/or compensate for any changes or species loss that may occur as a result of increased water levels 2. May the project likely affect fishing in the Shire River? NOT REALLY a. If yes, what measures have been put in place, and if no, why? b. Most of the fishing gears that are extensively employed in the Shire River fisheries are cast nets, scoop nets, fish traps and gill nets. Beach seines are common in Area A of the south east arm of Lake Malawi). Operation of these gears cannot be hindered by increased water levels. 3. Does the project have any effect in impeding/blocking migration of fish? YES a. Has this work provided solutions or suggestions to mitigate this impact? YES 4. Is there likelihood that increasing and/or low water levels upstream and downstream respectively would result into destruction of fish breeding habitats? YES a. Are there any mitigation measures for this impact provided by this work? YES 5. Does the potential for flooding downstream present a danger to destruction of fish breeding habitats downstream? YES a. Has this work suggested proper mitigation measures for the same? YES 6. Are there any nationally endangered fish species that could be affected as a result of the upgraded Barrage? YES a. Has this EIA made any suggestions to mitigating this impact? YES 7. Does the upgraded Barrage pose any challenge to destruction of fish landing beaches and fish drying areas? YES, but for isolated areas only especially Lake Malombe a. Are there suggestions for mitigating this impact? Not really, because this is more of a social issue hence to be tackled by other expertise. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Fisheries, December

50 Government of the Republic of Malawi Ministry of Water Development and Irrigation National Water Development Programme Independent Environmental Impact Assessment for the Upgraded Kamuzu Barrage Final Environmental and Social Impact Assessment Volume 2: Wildlife December 2013

51 Independent Environmental Impact Assessment for the Upgraded Kamuzu Barrage Final ESIA Volume 2: Wildlife Table of Contents... i Abbreviations... iii Acknowledgements... iv Executive Summary... v 1 Introduction Objectives of the ESIA Methods Description of the Existing Environment Wildlife Habitats Critical Habitats Sand Bars Sand Beaches Emergent Boulders Fauna Biodiversity (species diversity/richness, distribution, movements) Species of Conservation Interest Environmental Impacts Identification of Key Impacts to Wildlife Environmental Mitigation Measures Conclusion Reference Material Appendix 1: People Met During Field Work in the Project Area (October 2011) Appendix 2: Description of Field Activities in the Project Area by the Wildlife Expert in October Table of Figures Figure 1. National Parks and Wildlife Reserves within the Project area... 3 Figure 2. Sand bars in the middle of the Shire River and floodplain in Liwonde National Park close to Mvuu Camp Figure 3. Figure 4. Sand beach at Namakazo area of Mpiganjila village on the western shores of Lake Malombe. Note water birds utilising the sand beach A beach at Mwalija or Chimwala village is used by residents for domestic chores (washing, bathing, fetching water) and as fish landing site Figure 5. Sand beach at the Elephant Marsh Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December 2013 i

52 Figure 6. Figure 7. Figure 8. Figure 9. Shire River at Chipeta village looking across the river. Open water habitat is one of the critical habitats for a number of fauna species including Hippopotamuses ( ) Namakazo beach at Mpiganjila village. Note goats grazing on grass on the lake beach. Flooding of the beach will also affect livestock grazing areas. Beaches like this are important for crocodile and terrapin nesting ( ) Impala and waterbuck grazing on the floodplain in Liwonde National Park. Flooding from the Barrage could temporarily inundate some low lying grassland areas and reduce the wildlife grazing area. Note the Shire River on the horizon ( ) Mopane woodland in Liwonde National Park. Lillian s Lovebird is reported to use crevices of the mopane trees for breeding. Such woodlands are critical for the survival of this threatened species. Over 70% of Liwonde National Park is covered by mopane and most is not at risk of being flooded by Barrage operation ( ) Table of Tables Table 1. Places visited by the wildlife expert in the project area in October Table 2. Critical habitats in the project area... 4 Table 3. Summary of wildlife population trends in Liwonde NP and surrounding from 2004 to Table 4. Potential impacts on key wildlife habitats... 8 Table 5. Wildlife Impact Correlation Matrix for the proposed upgrading of Kamuzu Barrage... 9 Table 6. Summary of proposed mitigation measures for the adverse effects arising from upgrading of Kamuzu Barrage Table 7. Wildlife monitoring in upgraded Kamuzu Barrage project area Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December 2013 ii

53 Abbreviations AEWA CBD CITES CMS EIA ESCOM ESIA ESMP HRWL IFC IUCN km LNP m masl MNREE MoIWD NGO NP NT PPP SMEC USD UTM WESM Agreement on the Conservation of African-Eurasian Migratory Waterbirds Convention on Biological Diversity Convention on International Trade in Endangered Species of Wild Fauna and Flora Convention on Migratory Species Environmental Impact Assessment Electricity Supply Corporation of Malawi (Ltd) Environmental and Social Impact Assessment Environmental and Social Management Plan Highest Regulated Water Level International Finance Corporation International Union for Conservation of Nature kilometre Liwonde National Park metres metres above sea level Ministry of Natural Resources, Energy and Environment Ministry of Water Development and Irrigation Non-Governmental Organisation National Park Near Threatened Public Private Partnership Snowy Mountains Engineering Corporation Ltd. United States Dollar Universal Transverse Mercator Wildlife and Environment Society of Malawi Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December 2013 iii

54 Acknowledgements I am grateful to SMEC for entrusting me to participate in this ESIA exercise as a wildlife expert. Special thanks are extended to the Team Leader Michael Holics, SMEC secretary at Lilongwe Evelyn Mwase and other staff at Lilongwe office were very helpful in logistics during my field visit to Malawi. Mr Felix Kalowekamo accompanied me to the field and was extremely helpful in a number of ways. The Chiefs and villagers in places we visited were very cooperative and Liwonde National Park senior management, Mvuu Camp staff and Majete Wildlife Reserve devoted their time for discussions and provided useful information on the wildlife of Malawi. Government officials in particular the Ministry of Natural Resources, Environment and Energy, Department of Wildlife and National Parks and other Departments availed vital information on the project in addition to their opinions of the potential impacts. I enjoyed my stay in Malawi and the people were very kind and it demonstrated why Malawi is the Warm Heart of Africa. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December 2013 iv

55 Executive Summary The Government of Malawi through the Ministry of Water Development and Irrigation is implementing an integrated water development plan for the Shire River Basin. One of the major activities at this stage is the upgrading of the Kamuzu Barrage at Liwonde along the Shire River. In this second phase the rehabilitation is expected to increase the regulating capacity of the Barrage by raising the highest regulated water level (HRWL) by 20 40cm compared to the existing HRWL of masl. However, the Barrage control capacity does not include situations of severe floods or drought. The Government of Malawi contracted SMEC to undertake an ESIA which involved interdisciplinary fields including the wildlife component. The specific objectives for the wildlife component in the ESIA included: Collecting wildlife baseline data for the area of project impact. Identifying environmentally sensitive areas/habitats (e.g. wetlands, ecologically important areas) which may be impacted by the Project. Identifying any locally/regionally/nationally rare/endangered species of wildlife potentially impacted by the Project. Assisting in the identification of potential environmental impacts and mitigation measures. Working with other Team members in the identification and assessment of Project alternatives to minimise adverse environmental impacts. Providing technical support to Team members on wildlife aspects Providing written input into the ESIA and ESMP reports including sections/chapters on methodology, description of wildlife, impacts, mitigation and monitoring measures A combination of techniques including secondary sources and primary data were used to collect wildlife information used in assessing potential impacts of the project to wildlife. Primary data was collected from a site visit and included direct observation, transects, interviews and discussions. During the field visit all wildlife encountered was recorded. The field survey also involved interviews and discussions. Risk analysis suggested possible negative impacts to wildlife arising from flooding and changes in water flow and presence of barriers to wildlife movement and migration. Modelling of flooding regimes from the planned upgraded Barrage suggests that there will be little or insignificant deviation in areas to be covered by water as a result of enhanced impoundment. While species like crocodiles, the African Skimmer and terrapins may experience limited effects from the temporary flooding of sand bars and beaches, other species of conservation importance such as hippopotamus, Lillian s Lovebird, migratory birds, rhinoceros and elephants will not be affected. Also the upgraded Barrage is anticipated to have little effect to the critical habitats in the project area. The Barrage will continue to be a physical barrier for hippopotamuses and crocodiles to cross from upstream to downstream. On the positive side, the Barrage will continue to be a reliable source of water for the Shire River through regulation of flows and will ensure there is constant water availability for wildlife and other aquatic organisms. This is important given the unreliable water flow due to the erratic weather of Lake Malawi and the Shire River Basin. However, the anticipated effect of the upgraded Barrage to wildlife will also very much depend on how the Barrage gates will be opened to regulate water flow downstream. Monitoring of species diversity, recruitment, abundance, distribution and movement of species of interest and critical habitats is recommended to determine if there will be any significant impacts from the changed flooding regimes. Human wildlife interactions will also need to be monitored to assess whether the flooding regimes will result in any new or increased occurrence of human wildlife conflicts. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December 2013 v

56 1 Introduction This report covers the activities of the wildlife expert following a field visit to Malawi from 10 th to 23 rd October 2011 and evaluation of the wildlife data and information collected. 1.1 Objectives of the ESIA The objectives are as stipulated in the terms of employment covering activities to be undertaken by the expert and advisory role on wildlife matters to other members of the ESIA team. The specific objectives for the wildlife component in the ESIA included: 1.2 Methods Collecting wildlife baseline data for the area of project impact. Identifying environmentally sensitive areas/habitats (e.g. wetlands, ecologically important areas) which may be impacted by the Project. Identifying any locally/regionally/nationally rare/endangered species of wildlife potentially impacted by the Project. Assisting in the identification of potential environmental impacts and mitigation measures. Working with other Team members in the identification and assessment of Project alternatives to minimise adverse environmental impacts. Providing technical support to Team members on wildlife aspects Providing written input into the ESIA and ESMP reports including sections/chapters on methodology, description of wildlife, impacts, mitigation and monitoring measures A combination of techniques including secondary sources and primary data were used to collect wildlife information used in assessing potential impacts of the project on wildlife. Considerable secondary information on wildlife was available from Liwonde National Park, Department of National Parks and Wildlife, SMEC office in Lilongwe, Majete Wildlife Reserve, published/unpublished literature and Malawi Government website. Primary data was collected from site visit and included direct observation, transects, interviews and discussions. During the field visit all wildlife encountered was recorded. For the interviews and discussions some guiding questions were prepared to guide the conversation and included: Which animals are found in the area. What animals are rare in the area or have unique significance (of any kind)? Are there any human wildlife conflicts in the area? Are there any wildlife movements experienced here? Are there places where some animals breed in this area? What is the effect of the existing Kamuzu Barrage to wildlife? Are you aware that there are plans to upgrade the Kamuzu Barrage at Liwonde? What will happen if water levels rise due to the upgrading of the Barrage? What will be the effect to wildlife? Do you have any suggestions on what could be done? For some people and institutions there were additional questions that included: What was the situation before the Barrage was commissioned in 1965? What was the effect to wildlife soon after the Barrage was commissioned? Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

57 What is the institutional position on the upgrading of the Barrage in respect to environmental issues? Most of the key areas that cover the project area were visited (Table 1). Starting from the point where the mouth of the Shire leaves Lake Malawi we followed the river down to the Elephant Marsh to the south. Some of the areas surveyed were Mtongole village located at the mouth of Shire River as it leaves Lake Malawi, Chipeta, Mpanganjila, Chimwala, Kausi all located to the west of Lake Malombe and Kadewele village to the east of Lake Malombe. The survey also included the Shire River at Mangochi, Kamuzu Barrage at Liwonde, Liwonde National Park and Mvuu Camp. Further south we visited the Shire River at Chikhwawa, the Elephant Marsh, Majete Wildlife Reserve and Kapichira Power Station. TABLE 1. Mangochi Machinga PLACES VISITED BY THE WILDLIFE EXPERT IN THE PROJECT AREA IN OCTOBER 2011 District Village/Area Remarks Mtongole, Chipeta, Mpinganjila, Kadewele (Mpondas), Chimwala, Kausi Liwonde National Park, Mvuu Camp These villages except Kadewele appear to the west of Lake Malombe. Kadewele is on the eastern shores of Lake Malombe These include some of the prime wildlife areas, key wildlife habitats and tourism Balaka/Machinga Kamuzu Barrage at Liwonde This is the actual site of the project and Shire River is the boundary between the two districts Chikhwawa Slopes of Shire plateau overlooking the rift valley near Chikhwawa This is a catchment area that has been seriously affected by destruction of vegetation and wildlife habitat Chikhwawa Kamuzu Bridge at Chikhwawa Borders SUCOMA sugar cane plantation Chikhwawa/Nsanje Elephant Marsh A vast wilderness area currently threatened by human encroachment and cultivation Chikhwawa Majete Wildlife Reserve and Kapichira Power Station This is a reserve that is rapidly improving in conservation and tourism. The power station is located inside the reserve. At least 56 people were interviewed and provided information on wildlife related to the project area. They include villagers, chiefs, village headmen, fishers, wildlife managers, NGOs and government officials (Appendix 1). Potential impacts to wildlife anticipated from the changed flow conditions downstream of the Barrage and changed water levels upstream have been predicted based on the ecology and sensitivity of the individual species found in the impacted areas. It involved the assessment of species habitat utilisation of the flooded area and potential risks when the species is denied use of the habitat due to inundation. For other groups including migratory birds and animals a community ecology approach was used based on the collective ecological requirements of the species utilising the project area particularly the wetlands as a refuge area. The impacts were therefore ranked depending on the possible severity of the impact if the species or community is denied access to the habitat or the habitat is destroyed or degraded by inundation due to the refurbished Barrage or from floods. Experiences gained from what happened to the wildlife from the impacts of the Barrage commissioning in 1965 have been taken into account. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

58 2 Description of the Existing Environment 2.1 Wildlife Habitats The project area spans over a variety of habitats from open water, floating meadows, lagoons and reeds, floodplain grassland, woodland and mixed woodlands. The immediate (riparian) habitats along the Shire River and Lake Malombe include wetlands, open grasslands and mixed woodlands. Wetlands and open short grasslands occupy most of the floodplains. The extent of coverage varies from place to place but the project area being generally flat allows water stagnation during flooding and rain. There are also a number of protected areas within the project area including Liwonde National Park and Majete Wildlife Reserve (Figure 1). Figure 1. (Source: Department of National Parks and Wildlife) National Parks and Wildlife Reserves within the Project area 2.2 Critical Habitats In the World Bank s Natural Habitats Policy (OP 4.04), critical natural habitats are defined to include most types of protected areas, as well as other areas of known high conservation value. Critical habitat can also be construed as habitat that is essential for maintaining the integrity of an ecosystem, species or assemblages of species. Evaluation of the habitats in the project area with the associated species which utilise the habitats and taking into consideration the ecology of the species and community ecology the habitats shown in Table 2 can be considered as critical habitats responsible for the survival of a number of species in the project area. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

59 TABLE 2. Open water CRITICAL HABITATS IN THE PROJECT AREA Type Significance Remarks Sand banks (beaches) along Lakes Malawi and Malombe and along Shire River River sandbars Emergent boulders in middle Shire River channel Reeds Riparian short grasslands on river or lake banks Mopane woodland Riverine vegetation including Brachystegia wooded grassland in Majete Wildlife Reserve Important habitat for hippopotamus, crocodiles, terrapins, amphibians, fresh water invertebrates and other species Nesting areas for crocodiles and terrapins Serve as roosting and nesting areas for African Skimmers and other birds including waterfowl Habitat for the highly localised Rock Pratincole (Glareola nuchalis) Breeding areas for some birds including weavers Important grazing area for many herbivores especially during the dry season when most other areas are dry and void of grass and water. Prime habitat and breeding area for the threatened Lillian s Lovebird and also important habitat for many herbivores Habitat for the reintroduced black rhinoceros in the reserve. Open water habitat will increase upstream of the Barrage so these species will benefit from predicted changes Lack of these habitats or being inundated will prevent these species from breeding affecting their recruitment and population growth Inundation of river sandbars from regulated flows downstream and from higher water levels upstream will reduce the habitat for African Skimmers Inundation of boulders from higher regulated flows will reduce habitat for the Rock Pratincole Loss or degradation of these habitats through destruction or inundation will affect recruitment of a number avian species Inundation of short grasslands along river and/or lake banks will negatively impact many herbivore species in the project area Destruction of mopane woodland will affect the lovebird and other species. However, 70% of Liwonde National Park is covered by mopane woodland and inundation may likely affect a very small portion of the woodland. As such flooding is not really an immediate threat to the survival of Lillian s Lovebird. This is a critically endangered species which has been recently reintroduced into the reserve and is progressing well. The riverine habitat is affected by flooding from the Shire River which will impact on the rhinos. The Barrage does not exacerbate natural flooding in the Shire River. Marshes and wetlands Critical habitat for migratory birds Palaearctic and intra African migratory birds utilise the marshes and wetlands in the project area as winter refuge and breeding areas. The marshes and mud flats harbour invertebrates and other small vertebrates that are food for the birds. Wetland vegetation is susceptible to a change in flood regime which may result in a change in the habitat and ecology of the affected area. The Shire River system which includes the project area from Lake Malawi down to the Zambezi River comprises of various habit types described earlier including sand bars, sand beaches and emergent boulders. The occurrence of these habitats in the system depends on various factors such as geomorphology, nature of parent rock and soils, terrain and volume of flowing water. Sand bars, beaches and emergent boulders are important habitats for various fauna species depending on where the habitat is located, human influence and season of the year Sand Bars A sand bar may be described as a long mass or low ridge of submerged or partially exposed sand built up in the water along a shore or beach by the action of waves or currents. Usually these are found in mature river Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

60 systems or beaches with gentle or flat gradient and where the water flow is not very fast. Most are formed from deposition of sand and alluvium or may be formed following water recession due to reduced water flow. Sand bar presence depends on the amount of water flowing and may be covered under water or even disappear during floods. In the project area sand bars were observed in a number of places visited. They appear from where the Shire River leaves Lake Malawi to Lake Malombe and downstream to locations such as Mvuu Camp in Liwonde National Park (Figure 2) and Elephant Marsh. No attempt was made to document all sand bars in the project area but those encountered during the field work were noted. Nevertheless local people report of presence of sand bars in many parts of the Shire River. In some locations such as Kausi village some vegetation patches were observed in the middle of the river which later due to water currents and erosion the vegetation may die away and remain with emerged sand bars. African Skimmers (Rynchops flavirostris), waders and other water birds are some of the major users of the sand bars along the Shire River. AEWA (1998) reports that the skimmer is an intra-african or Afrotropical migrant associated with the Zambezi river system and it appears in both the Bonn Convention and AEWA appendices of migratory species. It is usually present in the area between April and December. The species migrates up and down larger rivers, and to and from inland lakes, dispersing widely after the breeding season. Migration is driven by the need for calm weather. African skimmers require a predictable river level that neither floods out of season nor drops so much that the birds are exposed to terrestrial predators. AEWA (1998) further reports that the species has an unfavourable conservation status due to the dam activities in the Zambezi basin, and the need for unseasonal river flow for the production of hydroelectricity. This makes the sandbars no longer safe from floods, and rocks emerge too far out of the river and banks are undercut by unseasonable flow (and also by speed-boat activity) all with a potential risk to the survival of the species. The African Skimmer is ecologically dependent on wetlands specifically rivers. In addition, the Skimmer is very vulnerable to tourists activities. According to the IUCN red list (2012) the African Skimmer is near threatened (NT) and it requires expanses of calm water for feeding. Breeding takes place during the dry season when rivers are at their lowest and sandbars most exposed. This generally occurs from July to November south of the equator. Breeding takes place along broad rivers on large, dry sandbars that are largely free from vegetation (IUCN red list 2012). In Malawi the Shire River is the main habitat and breeding area for the African Skimmer. Pairs nest in loose colonies and lay clutches of one to four eggs in a large usually unlined scrape in the sand. Scrapes occur within 30 m of water but receding water levels may increase the distance. The incubation period is about 21 days Sand Beaches A sand beach may be described as a gently sloping zone where deposits of unconsolidated sediments are subject to wave action at the shore of an ocean, lake or river. Most of the sediment making up a beach is supplied by rivers or by the erosion of highlands adjacent to the coast. Beaches extend from a low waterline landward to a definite change in material or physiographic form, such as the presence of a cliff or dune complex marking a clear demarcation of the edge of a coast or beach. In the project area sand beaches occur along parts of the Shire River and Lake Malombe (Figure 3). Beaches are only found in places where there is very gentle or almost flat gradient to the water level. Rocky banks or cliffs do not allow formation of beaches. Sand beaches are important habitats for crocodiles and in most cases crocodile conservation is in direct competition with other forms of land use such as agriculture and fisheries (Figure 4). Crocodiles need sand banks as nesting sites but the same sites are required as boat yards by fishers and as recreation areas. In marshy areas crocodiles need the islands as nesting sites but these are required for dimba cultivation (dry season cropping). For example in the Lower Shire Valley dimba land supplies a considerable proportion of total food production and it appears that dimba production is increasing in importance (Mphande, 1987). On the other hand young crocodiles tend to hide in the reeds and bask in isolated areas which are not easily accessible making these habitats also vital for the recruitment and survival of crocodile populations. One of the threats reported previously to the crocodile population has been the closing of Kamuzu Barrage gates on the nesting crocodiles in Liwonde National Park. The Barrage is opened and closed to regulate the flow of water in the middle Shire for hydroelectric power generation. The closure of the Barrage causes a rise in the water level in the Liwonde National Park. The Barrage is normally closed after the crocodiles have laid their eggs (i.e. October) hence submerging the eggs. The Barrage in such years is a major mortality factor for Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

61 the crocodiles. Ways and means have to be found to alleviate this problem. A compromise solution between the Ministry of Forestry and Natural Resources and ESCOM was proposed to be worked out and if possible the Barrage should be closed before crocodiles start nesting i.e. August (Mphande 1987). The average incubation period for crocodiles is about 80 days and by laying their eggs in October hatching would take place at the beginning of the long rainy season which starts in November and runs up to April. Terrapins (turtles living in fresh or brackish water) are reported to occur in the project area occupying logs, rocks and mud banks while in the rainy season they are common in isolated pans and waterholes. Nesting takes place in October November at the beginning of the rainy season and hatchlings emerge in November to January. Nests are vulnerable to predation by water monitors and mongoose (Broadley and Boycott 2009) Emergent Boulders These are rock outcrops found along river valleys above water level. They are formed in areas where the geomorphology comprise of hard rock not easily eroded by water current or other erosion agents. Rock outcrops are common in the middle Shire for about 80 km from Matope to Chikhwawa and south wards to the Majete Wildlife Reserve. On this stretch the river is characterised by rock bars and outcrops. Some areas of the Shire River, like parts of Mvuu Camp, are characterised by steep rocky banks. Series of rapids were observed along the Shire River at Kapichira power station located within Majete Wildlife Reserve but photos could not be taken due to security restrictions covering the area. The rocky islands of the Shire River in the Majete Wildlife Reserve provide habitat for breeding populations of the Rock Pratincole (Glareola nuchalis). The Rock Pratincole is another intra-african or Afrotropical migrant in the Zambezi River system, appearing in July and leaving sometime in December or January though it is reported to be endemic in the Majete area. It also requires a predictable river level that neither floods out of season nor drops so much that the birds are exposed to terrestrial predators. This little bird nests on the emergent rocks in white water areas, often near waterfalls. It is classified as Least Vulnerable on the IUCN Red Data List, the Rock Pratincole is widely distributed in Africa. 2.3 Fauna Biodiversity (species diversity/richness, distribution, movements) The project area is rich in biodiversity particularly the Liwonde National park (Table 3). Wildlife is found in almost all villages surrounding the project. However the eastern banks of Shire River record more species than the western banks possibly due to having greater wilderness areas and easy mobility of wildlife in the areas. The western parts are more populated with increased human activities that include farming, settlements and infrastructure development. TABLE 3. SUMMARY OF WILDLIFE POPULATION TRENDS IN LIWONDE NP AND SURROUNDING FROM 2004 TO 2009 Species Population Estimate Elephant Waterbuck 1,522 2,524 2,451 2,517 2,566 3,539 Impala 983 3,656 2,356 3,187 2,652 4,163 Sable Warthog 1,323 1,246 1,655 1,618 1,547 3,156 Bushbuck Kudu Common duiker Bush pig NR NR NR NR Reedbuck Klipspringer 6 8 NR NR NR NR Grysbok NR NR NR NR Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

62 Species Population Estimate Suni 4 2 NR NR NR NR Buffalo Eland Zebra Hartebeest Roan Black rhino NR Oribi NR 4 NR NR NR NR Hippopotami NR 926 NR NR 1,136 NR Crocodile NR 534 NR NR 3,105 NR (Source: Liwonde National Park 2004, 2005, 2006, 2007, 2008, 2009a, 2009b); NR: No records available 2.4 Species of Conservation Interest There are many species of conservation interest. Records show that a number of protected areas in the project area now harbour reintroduced species after the previous populations became locally extinct mainly due to hunting or habitat loss. Of key interest are rhinos, roan antelopes, leopards, buffaloes, eland and zebra. There are also species translocated to Liwonde National Park from Kasungu National Park in the north in order save them from local extinction. However, wildlife population trends from the annual census conducted by Liwonde National Park suggest that the populations of most large game species are stable. There also other species of birds, reptiles, amphibians and selected invertebrates of conservation interest. For example, the globally near-threatened African Skimmer is dependent on river sandbars (surrounded by water) for roosting and nesting, so it only occurs along the Shire River when water levels are low enough for the sandbars to be exposed. Three turtles classified as Least Concern (LC) on the IUCN Red List occur within Malawi. The Yellow-bellied Turtle (Pelusios castanoides) is the most common while the East African Black Mud Turtle (Pelusios subniger) and African Helmeted Turtle (Pelomedusa subrufa) are rarely mentioned though their range of distribution includes Malawi. Yellow-bellied Turtles are semi-aquatic and are found in rivers, marshes and swamps. This is the turtle which is commonly reported in Malawi and appears among the fairly common reptiles found in the marshes and swamps. It feeds mainly on large pulmonate snails and floating water lettuce. The Yellow-bellied Mud Turtles in captivity from Malawi are reported to lay about 25 eggs towards the end of September. It is also reported that during the dry season, this species may aestivate in the mud. The East African Black Mud Turtle eats insects, worms, snails, small fish, amphibians, crabs and aquatic plants that grow in water such as water grasses. These turtles are generally nocturnal but will bask in the sun or wander about on land during the rainy season. East African Mud Turtles tend to like a watery habitat and if their river or pond dries up during the dry season they may aestivate in underground burrows when the temperatures become too warm, too cold, or conditions become too dry, re-emerging again when conditions return to a suitable level when water returns. In captivity, the East African Black Mud Turtle has been recorded laying its eggs in February and March, with hatchlings emerging around 58 days later (when incubated at 28 to 30 C), although this may vary with location. Three to twelve eggs are produced per clutch by the female and buried in a flask-shaped nest cavity. African Helmeted Turtle also known as the Marsh Terrapin or Crocodile Turtle are semi-aquatic, living in rivers, lakes, and marshes, and they also like rain pools and places that are fertilised. The species is omnivorous and will eat almost anything including insects, small crustaceans, fish, earthworms, and snails. They may also feed on carrion. The fine claws on its feet help it tear its prey apart. Groups of these turtles have been observed capturing and drowning larger prey such as doves when they come to drink, the commotion caused by these group attacks is often mistaken for crocodiles. The female will lay 2 to 10 eggs on average, normally during late spring and early summer. The eggs are placed in a flask shaped nest that is about cm deep. The eggs hatch in days. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

63 3 Environmental Impacts 3.1 Identification of Key Impacts to Wildlife Rapid assessment of the situation suggest that some of the immediate impacts that may arise from changed flow regime downstream and changed water levels upstream of the Barrage include loss of habitat for some species due to inundation, loss of breeding areas for crocodiles and terrapins and loss of habitat for migratory birds. A summary of impacts is provided in Tables 4 and 5. Second level impacts may involve increased competition for resources between species and human wildlife conflicts. However flooding will have a dual effect on some wildlife species depending on the ecology of the species. One of the key species of conservation interest the Lillian s Lovebird is not greatly threatened by the rising water levels because most of its habitat is located upslope from the river channel and floodplains. There may be some impacts to tourism again depending on the effect of rising water levels on species as well as tourism infrastructure and navigation systems. TABLE 4. POTENTIAL IMPACTS ON KEY WILDLIFE HABITATS Type Significance Potential Impacts Open water Important habitat for hippopotamus, crocodiles, terrapins, amphibians, fresh Open water habitat will increase upstream of the Barrage so these species will benefit from predicted changes water invertebrates and other species Sand banks Nesting areas for crocodiles and terrapins Loss of these habitats will prevent these species from breeding affecting their recruitment and population growth Sand bars Reeds Seasonal Floodplains Mopane woodland Emergent boulders in Shire River channel Riverine vegetation in Majete Wildlife Reserve Marshes and wetlands Essential roosting and nesting habitat of the globally near-threatened African Skimmer. Breeding areas for some birds including weavers Important grazing area for many herbivores especially during the dry season when most other areas are dry and void of grass and water. Prime habitat and breeding area for the threatened Lillian s Lovebird and also important habitat for many herbivores Habitat for a localised bird species rare in Malawi, the Rock Pratincole Habitat for the reintroduced black rhinoceros in the reserve. Critical habitat for migratory birds Loss of these habitats will prevent these species from breeding affecting their recruitment and population growth Loss or degradation of these habitats will affect recruitment of a number of avian species Increased inundation of short grasslands will negatively impact many herbivore species in the project area Loss of Mopane woodland would affect the lovebird and other species. However, 70% of Liwonde National Park is covered by Mopane woodland and inundation is likely to affect only a very small portion of the woodland. As such flooding is not really a threat to the survival of Lillian s Lovebird This is a riverine habitat found in Majete Wildlife Reserve and can be significantly affected by flooding in the Lower Shire River. The Rock Pratincole is confined to this type of habitat and is rare in Malawi. This is a critically endangered species which has been recently reintroduced into the reserve and is progressing well. The riverine habitat can be affected by flooding and hence threaten the survival of the rhinos. Palaearctic and intra African migratory birds utilize the marshes and wetlands in the project area as winter refuge and breeding areas. The marshes and mud flats harbour invertebrates and other small vertebrates that are food to the birds. Some wetland vegetation cannot withstand long term/permanent inundation and would immediately rot changing the habitat and ecology of the affected area Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

64 TABLE 5. WILDLIFE IMPACT CORRELATION MATRIX FOR THE PROPOSED UPGRADING OF KAMUZU BARRAGE Environmental parameters/impacts Mobilization phase Activities Construction phase Operation phase Remarks Species diversity Large mammals Human activities often have a tendency of displacing large mammals even if the activities do not directly affect them but the associated disturbances do. Birds Human disturbances will displace some bird species Reptiles Construction activities may keep away some reptiles but would return after construction. Amphibians Construction activities will chase away some amphibians. Invertebrates Many invertebrates including insects will possibly be affected at all stages of the project construction and operation. All groups downstream Ensuring minimum adequate regular water flow downstream will enhance survival of a number of wildlife species. Species recruitment Crocodiles Inundated sand banks with nests will affect recruitment. Terrapins African Skimmer Inundated sand banks with nests will affect recruitment. Lillian s Lovebird Most of the Mopane woodland which is the breeding ground for Lillian s Lovebird is found in uplands and will not be affected by flooding. Rock Pratincole This bird breeds and roosts on emergent boulders in the Shire River in the Majete Wildlife Reserve. Flooding has already been reported to reduce its habitat. Other birds Many birds especially weavers breed in the reeds and other vegetation in marshes/wetlands which rot easily when submerged for long periods of time. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

65 Environmental parameters/impacts Mobilization phase Activities Construction phase Operation phase Remarks Wildlife movement Hippos Barrage is a partial barrier for hippos to move upstream or downstream Crocodiles Barrage is a partial barrier for crocodiles to move upstream or downstream Seasonal Floodplains Critical habitat for herbivores during dry season Riparian wooded habitats Important habitats for black rhino and elephants. The refurbished Barrage will only have a minor impact on riparian woodlands. Water availability for wildlife Barrage has ensured water is available to wildlife both upstream and downstream Key: -1=Mild Adverse; -2=Adverse; -3=Highly Adverse; +1=Mild Beneficial; +2=Beneficial; +3=Highly Beneficial; 0=No impact. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

66 4 Environmental Mitigation Measures Table 6 provides some mitigation measures proposed to be included in the environmental management plan for the Project for the purpose of minimising adverse impacts arising from Project activities. These mitigation measures are supported by the suggested monitoring outlined in Table 7. Mitigation measures for restriction of wildlife movement (crocodiles and hippopotamus) are considered unnecessary since the Barrage acts as a partial barrier only and habitat immediately downstream is not favourable to these animals. Barrage gates are normally fully open during extreme low and high flows in the Shire River allowing some movement of animals. The Middle Shire River system is characterised by a narrow rocky channel and fast flowing water with the presence of several rapids and waterfalls (e.g. Kapichira Falls) which are not favoured habitat for crocodiles or hippopotamus. Crocodiles and hippopotamus are also found in abundance in the Lower Shire, particularly in the floodplain wetlands of the Elephant Marsh suggesting wildlife movement is not an issue since the Barrage has been operating since Construction works and associated site clearing and vehicle movements may cause minor environmental impacts which can be mitigated by avoiding construction works in the rainy season as well as keeping vegetation clearance to a minimum. Construction teams should be provided with adequate food rations to avoid poaching of wild animals as well as coking fuel to avoid cutting of trees or use of charcoal. Following the approval and implementation of the project there is a demand for continuous improvement and follow up of the environmental performance within the project system. Environmental monitoring is a process which, when being carried out in a participatory way, can constitute a powerful instrument to raise awareness, to train people and to improve insights in ecological processes and environmental impacts (Kessler 1998). Table 7 provides some of the initial monitoring attributes, indicators and protocols for identified issues in the current project. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

67 TABLE 6. SUMMARY OF PROPOSED MITIGATION MEASURES FOR THE ADVERSE EFFECTS ARISING FROM UPGRADING OF KAMUZU BARRAGE Impact Cause Affected Fauna Effect Status Proposed Mitigation Species diversity Change in inundation patterns upstream of Barrage Large mammals and other species Flooded areas scare animals away Short and long term Controlled flooding upstream is practised to ensure only the necessary water rise is effected to allow limited flooding of wildlife habitat Habitat destruction, degradation Flooding of riparian short grasslands Flooding of reeds, marshes and other wetland vegetation Impaired breeding of wildlife species Flooding of riparian vegetation in Majete Change in inundation patterns upstream of Barrage Change in flow pattern downstream of Barrage Inundation of sand banks and bars used as nesting areas Rock Pratincole Herbivores including hippopotamus, impala, waterbucks, warthogs and many other species Migratory bird communities Crocodiles and terrapins Habitat destroyed by flooding Loss of existing Grassland habitat Habitat affected by changing water levels due to new regulated flow regime Destruction of the nests and eggs Note: National Parks staff should be involved in the institutional structures for managing Barrage operation. Short and long term Short term Short term Short to long term Barrage operation to regulate water flow downstream to avoid/minimise inundation of river boulders during the breeding season. This is an important dry season grazing refuge for herbivores in the project area. Limiting water rise will help to reduce the grassland area that will be covered by water Regulate water levels to maintain the marshes and mud flats supporting migrating birds Barrage operation to limit water rise upstream during critical breeding periods Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

68 TABLE 7. WILDLIFE MONITORING IN UPGRADED KAMUZU BARRAGE PROJECT AREA Monitoring attribute Monitoring indicators Protocols Cost (USD) Species diversity Species performance Number of species in areas affected by flooding as a result of Kamuzu Barrage operation Performance of Rock Pratincole (in Majete), elephants and Lillian s Lovebird Estimating species diversity/richness on annual basis 30,000 Population abundance, structure and distribution evaluated periodically Recruitment of species of interest Population growth of crocodiles and terrapins Census of crocodiles and terrapins and studying population structure conducted periodically Migratory bird communities Riparian short grasslands within Liwonde National Park and Majete Wildlife Reserve Presence and number of migratory birds in marshes and wetlands in the project affected area Expanse and utilisation Observation of migrating birds based on number of species/individuals, timing, duration of stay on annual basis Measuring the size and spatial distribution of the grasslands and level of utilisation including number of species utilising the grasslands conducted periodically Note: Coverage of studies is considered to include the whole Shire River catchment from the source at Lake Malawi to the point it enters the Zambezi River. National Parks staff should play an important part in monitoring programs especially within Liwonde National Park and Majete Wildlife Reserve. 20,000 30,000 15,000 50,000 Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

69 5 Conclusion The construction works and associated vegetation clearance (including quarry site and sand/gravel borrow areas) cover a small area and the impacts will be minor and temporary. The proposed monitoring to collect data on the health and distribution of riparian habitats within Liwonde National Park and Majete National Park, as well as monitoring of specific wildlife species (elephant, hippopotamus, crocodile, African Skimmer) will lead to better operation of the Barrage. This in turn will result in the long term preservation of habitats and improved biodiversity. Careful adherence to mitigation measures such as those outlined in Table 6 will be taken to avoid/minimise negative impacts from the Project. Consequently, increasing the regulating capacity of Kamuzu Barrage by raising the highest regulated water level by cm by upgrading the Barrage is not expected to have significant adverse effects on wildlife populations in the project area. The combined ecological impacts from the project are expected to result in a minor negative overall balance. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

70 6 Reference Material References AEWA (1998). A Waterbird Oversight. African-Eurasian Waterbird Agreement. AEWA Newsletter. Broadley, D. G. and Boycott, R. C. (2009). Pelusios sinuatus (Smith 1838) - Serrated Hinged Terrapin. In Rhodin, A. G. J., Pritchard, P. C. H., van Dijk, P. P., Saumure, R. A. Buhlmann, K. A., Iverson, J. B. and Mittermeier, R. A. (Eds). Conservation Biology of Freshwater Turtles and Tortoises: A Compilation Project of the IUCN/SSC Tortoise and Freshwater Turtle Specialist Group. Chelonian Research Monographs No. 35 pp , doi: 10385/crm sinuatus.vl.2009, Department of National Parks and Wildlife (undated). Environmental and Social Management Framework, Final Report. Effective Management of Nkhotakota Wildlife Reserve Project, GEF Implementing Agency: Wildlife and Environmental Society of Malawi, GEF Project Number: P Mphande, J. N. B. (1987). Status of the Nile Crocodile in Malawi. In Maintenance in Appendix II of the Malawian population of Crocodylus Niloticus Proposal. Republic of Malawi 56pp. Websites (IUCN 2012) (December 2012) Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

71 Figure 2. Sand bars in the middle of the Shire River and floodplain in Liwonde National Park close to Mvuu Camp. Figure 3. Sand beach at Namakazo area of Mpiganjila village on the western shores of Lake Malombe. Note water birds utilising the sand beach. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

72 Figure 4. A beach at Mwalija or Chimwala village is used by residents for domestic chores (washing, bathing, fetching water) and as fish landing site. Figure 5. Sand beach at the Elephant Marsh Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

73 Figure 6. Shire River at Chipeta village looking across the river. Open water habitat is one of the critical habitats for a number of fauna species including Hippopotamuses ( ). Figure 7. Namakazo beach at Mpiganjila village. Note goats grazing on grass on the lake beach. Flooding of the beach will also affect livestock grazing areas. Beaches like this are important for crocodile and terrapin nesting ( ). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

74 Figure 8. Impala and waterbuck grazing on the floodplain in Liwonde National Park. Flooding from the Barrage could temporarily inundate some low lying grassland areas and reduce the wildlife grazing area. Note the Shire River on the horizon ( ). Figure 9. Mopane woodland in Liwonde National Park. Lillian s Lovebird is reported to use crevices of the mopane trees for breeding. Such woodlands are critical for the survival of this threatened species. Over 70% of Liwonde National Park is covered by mopane and most is not at risk of being flooded by Barrage operation ( ). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

75 Appendix 1: People Met During Field Work in the Project Area (October 2011) SN Name Title Affiliation Contacts 1 Felix Kalowekamo Lecturer Bunda College of Agriculture 2 Mariam Binali Ag. Chief Mtongole Village 2 Daniel Wayson Villager Mtongole Village 3 Emily Chipeta Village Headman Chipeta Village 4 Kawile Susa Villager Chipeta Village 5 Metrina Khan Villager Mpiganjila Village Kondani Sekambwa Fisherman Mpiganjila Village 7 Emmanuel Chilungamo Fisherman Mpiganjila Village 8 Rajabu Samuel Fisherman Mpiganjila Village 9 Paul John Namakasu Village Headman Mpiganjila Village 10 Rashid Moses Khan Villager Mpiganjila Village Margreth Anton Villager Chimwala/Mwalija Village 12 Idrissa Ali Abdul Fisherman Chimwala/Mwalija Village 13 Augustine Michael Fisherman Chimwala/Mwalija Village 14 Biriad Bwanali Fisherman Chimwala/Mwalija Village 15 Frank Chalo Fisherman Chimwala/Mwalija Village 16 Sifa Denson Fisherman Chimwala/Mwalija Village 17 Haji Samson Fisherman Chimwala/Mwalija Village 18 Yusuph Sunje Fisherman Chimwala/Mwalija Village 19 Omaa Augustine Fisherman Chimwala/Mwalija Village 20 Jaffari Dickson Fisherman Chimwala/Mwalija Village 21 Kassin Dayton Fisherman Chimwala/Mwalija Village 22 Fatina Augustine Fisherman Chimwala/Mwalija Village 23 Tole Manyozo Fisherman Chimwala/Mwalija Village 24 Askot Danken Fisherman Chimwala/Mwalija Village 25 Maumbo Mmadi Fisherman Chimwala/Mwalija Village 26 Michael Thomas Fisherman Chimwala/Mwalija Village 27 Michael Thomas Fisherman Chimwala/Mwalija Village 28 Rome Sindi Fisherman Chimwala/Mwalija Village 29 Hamilton Wailess Fisherman Chimwala/Mwalija Village 30 Lucy Kaluwa Villager Kadewele Village 31 Loveness Kadewele Villager Kadewele Village 32 Loveness Masanga Villager Kadewele Village 33 Lucy Bula Villager Kadewele Village 34 Wetu Edward Villager Kadewele Village 35 Leonard N gisi Villager Kadewele Village 36 Nsaku Kawina Village Headman Kadewele Village 37 Machinga Dayton Chairman, Village Beach Committee Kadewele Village 38 Manyozo John Villager Kadewele Village Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

76 SN Name Title Affiliation Contacts 39 Daton Mustapha Villager Kadewele Village 40 Wilson German Villager Kadewele Village 41 Hauvi James Villager Kadewele Village 42 Musa Lesta Chief Kausi Village 43 Yunus Yusuph Villager Kausi Village 44 Eneress Mustapha Villager Kausi Village 45 Esnath Jasten Villager Kausi Village 46 Mariam Akim Villager Kausi Village 47 Fatma Alaton Villager Kausi Village 48 Angel Gondwe Safari Guide Mvuu Lodge 49 Dinara Kankhuni Camp Manager Mvuu Lodge 50 Charlington Banda Parks Assistant Liwonde National Park 51 Benson Kandyero Wildlife Scout Liwonde National Park 52 Sam Nyanyale Divisional Manager Liwonde National Park McPhillip Mwithokona Head, Wildlife Management Unit Liwonde National Park Blessings Msikuwanga Research Officer Liwonde National Park Mavuto Kulinji Parks Assistant Liwonde National Park 56 Michael Holics Team Leader SMEC, Lilongwe Dr. Mrityunjay M. Jha Team member SMEC, Dar es Salaam Murphy Kajumi Sociologist SMEC, Lilongwe Sulemani Shuvah Boatman (Elephant Marsh) Tambo Village 60 Nicholaus Finta Boatman (Elephant Marsh) Nyang a Village 61 Ronick Hale Fisher (Elephant Marsh) Sabora Village 62 Patricio Ndadzela Project Manager Majete Wildlife Reserve Shadrack Chado 64 Clement Mbota Senior Operations & Electrical Maintenance Engineer Deputy Director (Wildlife Management) 65 Chiza Manda Assistant Director (CS) 66 Vicent Kaitano Manager 67 Wilfred Ndovi Project Coordinator (Sustainable Management of Wildlife through CBRM) Kapichira Power Station Department of National Parks and Wildlife - Lilongwe Department of National Parks and Wildlife - Lilongwe Wildlife and Environmental Society of Malawi - Lilongwe Department of National Parks and Wildlife - Lilongwe 68 Sosten Lingwalanya Deputy Director Department of Tourism 69 Kasizo Chirambo Assistant Director Department of Forestry / Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

77 Appendix 2: Description of Field Activities in the Project Area by the Wildlife Expert in October 2011 Date Day Description of duties 8 Saturday Review of inception project report and other documents submitted by SMEC. Literature search on Liwonde National Park, Species of conservation importance at the project area including Hippopotamus, Elephants, Crocodiles and Lillian s Lovebird. 10 Monday Travel from Dar es Salaam to Lilongwe. Arrive at SMEC country office at Lilongwe meet with Team Leader Michael Holics, SMEC Country Manager Stayford Phiri, another team member Dr Mrityunjay Jha and local counterpart Mr Felix Kalowekamo. Received project briefing from Team Leader, had orientation of the SMEC offices and started project document review. 11 Tuesday Working meeting with local counterpart Mr Felix Kalowekamo, held discussions of the wildlife issues in the project area and the existing field situation. In collaboration with the Team Leader Mr Michael Holics we planned the field visit to Liwonde and surrounding areas, logistics and key areas and people to meet in the field. Continued with document review at SMEC offices in Lilongwe. 12 Wednesday Worked from the SMEC offices in Lilongwe. Continued with documents review of previous studies in the project area. Reviewing the field visit plan with team leader Michael Holics and Local counterpart Mr Felix Kalowekamo following fuel shortage at Lilongwe. 13 Thursday Worked from SMEC offices in Lilongwe. Continued with documents review and planned for a revised field visit itinerary with the team leader, local counterpart Mr Felix and secretary Ms. Evelyn. Finalised transport hire for the field trip and other logistical arrangements. Confirmed to leave for field work on the next day i.e. Friday 14 th October Friday Travelled from Bridge View Hotel in Lilongwe with local counterpart Mr Felix Kalowekamo to the project area in Mangochi District. Heading southwards from Lilongwe we branched eastwards at Masasa junction and descended into the rift valley. First we visited Mtongole village in the afternoon where the source of Shire River is located. We met the Acting Chief Mariam Binali and other villagers. We visited the source of Shire river from Lake Malawi and discussed with the Chief about the wildlife issues and possible impact of the improved Kamuzu Barrage to the area. We later moved to Chipeta village where we met the Village Headman Emily Chipeta. We discussed with her again the wildlife issues and later visited the Shire river at the village. At his point we could see across the river the marshes to the south east end of Lake Malawi which are believed could be inundated during floods or by water flow control at the barrage. The marshes are reported to have sand bars where crocodiles put their nests. Later we proceeded to Mangochi where we visited again the Shire River at Mangochi and the Bakili Muluzi Bridge. We spent the night at Mangochi. 15 Saturday Started off at around 8am in the morning and managed to visit four villages surrounding Lake Malombe, one village on the eastern shores and three villages on the western shores. At Kadewele village on the eastern side of Lake Malombe we managed to meet the Village Headman Kadewele, Beach Village Committee Chairman Machinga Dayton and other villagers and fishermen. We also visited the Chantoji beach and Nkuju crocodile nesting site. At Mpinganjlila village we met the village Headman Paul John Namakasu and other villagers and fishermen. We visited Namakazo area a fish landing site and sand bars. At Chimwala village we managed to speak to a number of fishermen and visit the Mwalija beach, also a fish landing area. The last village to visit was Kausi where we met the Chief Mussa Lesta and other villagers. The Chief took the liberty of taking the team to Tenesi beach and provided information on wildlife in the village and human wildlife conflicts they experience. As dusk was drawing in we started off to Liwonde and arrived at around 8pm and retired for the day. 16 Sunday Started off at 8am to Liwonde National Park. Met the Scout guard on duty at the gate Mr Benson Kandyero who provided us with an escort scout Mr Charlington Banda who escorted us to Mvuu camp. On the way we managed to see many herds of wildlife suggesting that the Park is relatively rich in wildlife resources. At Mvuu camp we met with Camp Manager Mr Dinara Kankhuni and Tour Guide Mr Angel Gondwe. We discussed with them the issues of upgrading the Kamuzu barrage and its possible consequences on wildlife. The manager offered us a boat ride to see the effect of water currents and waves on the Shire River banks around the lodge, and he commented that rising water levels would exacerbate the problem hence affecting the camp Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

78 infrastructure and consequently tourism. In the afternoon we came back and had a meeting with the Manager of Liwonde National Park Mr Sam Nyanyale. 17 Monday In Malawi it is a public Holiday Mother s Day but the top management of Liwonde National Park agreed to meet us for a discussion. We started off from Hippo View Lodge at about 8.30am and arrived at the LNP Headquarters in the Park. We had a discussion with the Division Manager Mr Sam Nyanyale, Research Officer Mr Blessings Msikuwanga and Head of Wildlife Management Mr McPhillip Mwithokona. After a two and a half hour meeting we proceeded to see the Lillian s Lovebird habitat and breeding site at Namisundu area, located within the Park. On our way back we visited the Kamuzu Barrage at Liwonde township. We returned to Liwonde after midday and spent the afternoon reviewing reports obtained from LNP. 18 Tuesday Set off from Liwonde at around 8.30am and travelled through Zomba, Blantyre, Shire plateau to Kamuzu Bridge at road block Chikhwawa. Observed the Shire River. From the slopes of Shire plateau we could see the Shire meandering as it flows lazily towards the Zambezi and with a number of ox bow lakes and branched streams. We proceeded to the Elephant Marsh at Chilomo where we arrived late afternoon. Observed the marshes through navigating by canoe until late evening. Retuned to Chikhwawa where we spent the night. 19 Wednesday Started off from Chikhwawa at about 8.30am after fixing a flat tyre, we proceeded to Majete Wildlife Reserve. At Majete we met the Project Manager Mr Patricio Ndadzela and discussed with him a number of wildlife issues. We proceeded to Kapichira Power Station and had a discussion with the Senior Operations and Electrical Maintenance Engineer Mr Shadreck Chado. We then started a long journey back to Lilongwe via Blantyre and Kamuzu Bridge at Zalewa. We arrived in Lilongwe at about 1900 hours. 20 Thursday Reported back at the SMEC offices at Lilongwe and briefed the team leader on the field trip to the project area. Then in the company of the local counterpart we visited key people and offices in Lilongwe including National Parks and Wildlife, Lilongwe Sanctuary, Danish Hunters Association and the Environment. At the National Parks and Wildlife we met Mr Clement Mbota the Deputy Director (Wildlife Management), and Mr Chiza Manda, the Assistant Director (CS). At WESM we met the Manager Mr Vicent Kaitano and Project Coordinator (Sustainable Management of Wildlife through CBRM) Mr Wilfred Ndovi. We were not lucky to find any person at the Environment office as all of them had gone for a climate change symposium. We returned to SMEC offices and continued with review of various reports and documents availed to the team related to the ongoing assignment. 21 Friday The day stated by visiting other key personnel who are stakeholders in the project. We managed to visit the Ministry of Tourism, Wildlife and Culture where we met the Deputy Director of Tourism (Marketing) Mr Sosten Lingwalanya and later we went to the Department of Forestry where we were able to meet the Assistant Director Mr Kasizo Z. S. Chirambo. We returned to the office and continued with review of documents and reports and also discussed with team leader issues related to report writing. 22 Saturday Continued with review of documents and other literature and started initial stages of designing the report format and preparations for a return journey back to Tanzania the next day. 23 Sunday Spent the morning hours working on the draft report and later at around 10.30am left for the Kamuzu International Airport in Lilongwe and flew to Dar es Salaam. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Wildlife, December

79 Government of the Republic of Malawi Ministry of Water Development and Irrigation National Water Development Programme Independent Environmental Impact Assessment for the Upgraded Kamuzu Barrage Final Environmental and Social Impact Assessment Volume 2: Vegetation December 2013

80 Independent Environmental Impact Assessment for the Upgraded Kamuzu Barrage Final ESIA Volume 2: Vegetation Table of Contents... i Abbreviations... iii Acknowledgements... iii Executive Summary... iv 1 Introduction Project Background Ecological Setting Lake Malawi Upper Shire River Middle Shire River Lower Shire River Hydrology of Lake Malawi and the Shire River Lake Malawi Lake Outflow and Upper Shire Lake Malombe Middle Shire Lower Shire Biotic Communities Perennial Marshes Seasonal Marshes Seasonal Marshes Cultivated Seasonal Floodplains Riverine Woodlands Methods Field Studies Spectral Analysis Outputs Results Modelling Impacts Upstream of Kamuzu Barrage Area of Inundation Inundation Period Downstream of Kamuzu Barrage Area of Inundation Inundation Period Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December 2013 i

81 4 Discussion Environmental Flow Requirements Implications for Operation of Kamuzu Barrage References Appendix 1: Description of Biotic Communities at Sampling Sites Table of Figures Figure 1. Biotic Communities of Malawi and Project Area... 2 Figure 2. Map of Project area... 3 Figure 3. Lake Malawi... 4 Figure 4. Kamuzu Barrage... 4 Figure 5. Example of a Perennial Marsh... 9 Figure 6. Example of an uncultivated seasonal marsh Figure 7. Example of cultivated seasonal marsh (dambo) Figure 8. Example of seasonal flood plains Figure 9. Example of riverine woodland Figure 10. Inundation map 473 masl Lake Malawi to Kamuzu Barrage Figure 11. Inundation map 473 masl Kamuzu Barrage to Chigaru Figure 12. Inundation map 474 masl Lake Malawi to Kamuzu Barrage Figure 13. Inundation map 474 masl Kamuzu Barrage to Chigaru Figure 14. Inundation map 475 masl Lake Malawi to Kamuzu Barrage Figure 15. Inundation map 475 masl Kamuzu Barrage to Chigaru Figure 16. Inundation map masl Lake Malawi to Kamuzu Barrage Figure 17. Inundation map masl Kamuzu Barrage to Chigaru Figure 18. Inundation map 476 masl Lake Malawi to Kamuzu Barrage Figure 19. Inundation map 476 masl Kamuzu Barrage to Chigaru Figure 20. Habitat inundation above Kamuzu Barrage over a range of water levels Figure 21. Habitat inundation below Kamuzu Barrage over a range of water levels Figure 22. Shire River Hydrograph (World Bank 2010) Table of Tables Table 1. Common Aquatic Macrophytes found in Perennial Marshes of the Shire River... 8 Table 2. Common Emergent Woody Plants... 9 Table 3. Typical Floodplain Plant Species Table 4. Common plants associated with riverine woodlands Table 5. Biotic communities represented in field studies Table 6. Inundation Period Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December 2013 ii

82 cm EIA ESIA ESRI GPS ha HRWL km LRWL m masl mm UNDP UNESCO Abbreviations centimetre Environmental Impact Assessment Environmental and Social Impact Assessment Mapping software Global Positioning System hectare Highest Regulated Water Level kilometre Lowest Regulated Water Level metre metres above sea level millimetre United Nations Development Programme United Nations Educational, Scientific and Cultural Organization Acknowledgements Invaluable assistance and advice was provided to the study team by rangers and staff from the Liwonde National Park; rangers and staff from the Majete National Park; and Dr Montford Mwanyambo, and staff from the National Herbarium of Malawi at Zomba. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December 2013 iii

83 Executive Summary The Ministry of Irrigation and Water Development commissioned the Snowy Mountains Engineering Corporation (SMEC International) to undertake a detailed Environmental Impact Assessment (EIA) for the upgrade (including raising the height) of Kamuzu Barrage in accordance with the requirements of Malawi s EIA process and World Bank Safeguard Policies. This Ecological Technical Report has been prepared as an appendix to the main Environmental and Social Impact Assessment (ESIA) report. The objective the study is to inform the ESIA process by collating baseline data on the ecology of the project area, identifying ecological risks associated with the project, and developing appropriate management measures to minimise or mitigate risks to an acceptable level. The study focuses primarily on the impacts to the biotic communities immediately adjacent to the areas of Lake Malawi and the Shire River that may experience fluctuations in water levels and inundation periods as a result of the proposed barrage upgrade. Impacts to terrestrial and aquatic flora and fauna are addressed in the Wildlife Technical Report and the Aquatic Ecology Technical Report respectively. The study characterised the biotic communities at 30 different sites adjacent to the Shire River, Lake Malombe and Lake Malawi. Each site was within 50 m of the high water level; the location of each was largely determined by its accessibility from the road network. At each site the survey team characterised the biotic community according to a predetermined descriptor of the vegetative communities that were represented. A patch of the dominant vegetation type was captured within a GPS polygon to act as a training site for spectral imagery analysis of the entire project area. Each site was described in terms of topography and distinguishing features and photographed with a GPS enabled camera. In order to quantify the extent of impacts from arising from different inundation models, biotic communities were characterised within the predicted inundation area for a range of water levels. A combination of manual delineation and unsupervised spectral classification was used delineate the boundaries between the biotic communities within the inundation area. Using this technique, the following biotic communities were found to be represented within the inundation area: Perennial Marshes Seasonal Marshes Uncultivated Seasonal Marshes - Under Cultivation (dambos) Seasonal Flood Plains Riverine Woodlands Dry Bush Savannah Cleared Woodland In order to quantify the impact of each inundation model on each biotic community type, the total area within each inundation model was calculated using geometry and summary tools in ArcGIS (ESRI). A separate map showing the results of each calculation was developed for visual comparison. The results of the modelling show that there will be a minimal change in the area of habitat inundated and inundation period associated with a 20 cm increase in Barrage height (corresponding to a new Highest Regulated Water Level (HRWL) of masl). However, a potentially significant increase in the area of inundation and inundation period is possible upstream of the Barrage if the height is raised by 40 cm (corresponding to a new HRWL of masl). This may significantly impact important Perennial Marsh habitat located along the boundary between the Shire River and Liwonde National Park. It is recommended that if the Barrage is raised by 40 cm that a Perennial Marsh habitat monitoring program is implemented within the boundaries of the Park and an adaptive management program is adopted whereby the operation of the Barrage is regulated in response to the findings of the monitoring program. No such monitoring program is deemed necessary if the Barrage is raised by only 20 cm. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December 2013 iv

84 It is also recommended If levels do need to be increased above masl then the Barrage should be operated to mimic the seasonal flooding patterns of system i.e. levels of masl permitted during the wet season only (i.e. Dec April); maintain levels below during the dry season (May Oct). This operational regime will also minimise the incidences of prolonged (i.e. greater than existing levels) inundation of Seasonal Floodplain and Marshland habitats downstream of the Barrage. This is significant as this habitat type may support crocodile (and turtle) nesting during the critical incubation period lasting from October to December. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December 2013 v

85 1 Introduction 1.1 Project Background Over the last century Lake Malawi has experienced severe periodic water level fluctuations which in turn have caused social, economic and environmental changes, mostly of a negative nature. For instance, in the period 1915 to 1937 the water level of the Lake was so low that the outflow from the Lake more or less stopped, leaving the Shire River dry. Different ways of securing and stabilising the water flow have been suggested and studied. The most resent of these studies is the Lake Malawi Level Control Study Phase I and II which analysed different types of interventions to stabilise the flow in the Shire River, including the no-action alternative. The preferred option identified by this study was to upgrade the existing Kamuzu Barrage at Liwonde so that flow in the River could be regulated in response to prevailing environmental conditions. The Ministry of Irrigation and Water Development contracted the Snowy Mountains Engineering Corporation (SMEC International) to undertake a detailed Environmental Impact Assessment (EIA) for the upgrade of Kamuzu Barrage in accordance with the requirements of Malawi s EIA process and World Bank Safeguard Policies. This Ecological Technical Report has been prepared as a technical appendix to the main Environmental and Social Impact Assessment (ESIA) Report. The objective of this technical report is to inform the EIA process by collating baseline data on the ecology of the project area, identify ecological risks associated with the proposed project, and develop appropriate management measures to minimise or mitigate risks to an acceptable level. This Ecological Technical Report is focused primarily on the impacts to the biotic communities immediately adjacent to the areas of Lake Malawi and the Shire River that may experience fluctuations in water levels as a result of the proposed Barrage upgrade (Figure 1). Impacts to terrestrial and aquatic flora and fauna are addressed in the Wildlife Technical Report. 1.2 Ecological Setting The water bodies and catchment area influenced by the project include Lake Malawi and the Shire River water systems (Figure 2). Lake Malawi has a catchment area at the Lake outlet of about 126,500 km 2, consisting of: 64,372 km 2 drainage area in Malawi 26,600 km 2 drainage area in Tanzania 6,768 km 2 drainage area in Mozambique, and 28,740 km 2 of Lake surface The Shire River forms the only outlet of Lake Malawi, running through the southern part of the country. Its lower part below the junction with Ruo River forms the border with Mozambique. A few kilometres further downstream of the southern border of Malawi the Shire River discharges into the Zambezi. The Shire River catchment, below the Lake and excluding the Ruo catchment, covers an area of 18,945 km 2. Although Malawi lies within the tropics, the topography is varied and the range of altitude so great that climatic conditions are a complex mosaic with many gradations between dry and wet and hot and cold. The project area has been defined as the reaches/areas of the Shire River, Lake Malombe and Lake Malawi that may experience permanent or temporary changes in water levels as a result of the proposed upgrade of the Kamuzu Barrage at Liwonde. The project area covers a distance of approximately 100 linear kilometres from the lower reaches of Lake Malawi to Chikhwawa on the Shire River (Figure 2). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

86 Figure 1. Biotic Communities of Malawi and Project Area Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

87 The project area is naturally divided in to four major zones these are: 1. Lake Malawi Southern arms 2. Upper Shire River from Samama to Matope 3. Middle Shire River - from Matope to Kapichira 4. Lower Shire River Kapichira to Mozambique border Kamuzu Barrage Figure 2. Map of Project area Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

88 1.2.1 Lake Malawi Lake Malawi is the southernmost and third largest of the East African great lakes. It is approximately 560 km long and 75 km across at its widest point. Maximum depth is around 700 m. The Lake is located at an altitude of about 475 m and experiences marked seasonal variations in wind, temperature and precipitation. The depth of the Lake falls off quickly from the shore and depths of more than 200 m are found offshore. In the south-eastern area, where the Lake discharges its excess water in to the Shire River, the fall-off in depth is less striking and large shallow sandy areas occur. The Lake, in particular the southern part, supports important fisheries both artisanal and semi-industrial. The Lake has been included on the United Nations Educational, Scientific and Cultural Organization (UNESCO) World Heritage List due to its unique fish biodiversity, represented by the several hundred different mbuna cichlid species. These species are almost exclusively found on stone and rocky bottom and only 2-3 species inhabit sandy substrate. Figure 3. Lake Malawi Upper Shire River The Upper Shire River is defined as the river channel from its outflow from Lake Malawi at Samama down to Matope and includes Kamuzu Barrage (Figure 4). For this 100 km stretch the River flows down an extremely gentle gradient, the water level at Liwonde being only 2 m lower than the Lake exit (the gradient is even less with the Barrage in operation). The flat landscape and reduced stream velocity cause the River to meander creating a network of pools and channels, and to flood adjacent land in the rainy season. Figure 4. Kamuzu Barrage Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

89 The area is relatively dry with mean annual rainfall of around 700 mm (contrasting with over 2,000 mm on the nearby Zomba Mountain). The rainy period is from November to April when more than 90% of the rain falls. May to October constitutes the dry season. Mean annual temperature is around 22 C. The landscape of the Upper Shire takes its character from the Great African Rift Valley, which runs northsouth in a km wide trough. The valley consists of plains and gentle foot slopes, which rise to an altitude of approximately 700 m. To the west the Rift Valley merges into uplands and hills around 1,100 m, while an escarpment delimits the eastern boundary several hundred metres high guarding an extensive area of uplands and hills. The area is underlain by basement complex rocks of pre-cambrian and Cambrian age. Extensive faulting since Mesozoic times has led to the formation of the Rift Valley, which itself is filled with unconsolidated sediments of lacustrine, fluvial and colluvial origin. Soils are generally well drained, yellowish brown to reddish brown, medium to fine textured and are deep if developed in alluvium. In the lowest areas soils have been modified by flooding and sedimentation, giving rise to considerable localised variations. Much of the Upper Shire Valley and the lakeshore plain areas, where soils have reasonable drainage, are dominated by woodland savannahs. Apart from Liwonde National Park, most of the areas of semi-deciduous riverine forest have been cleared for cultivation. Some areas of Mopane (Colophospermum mopane) woodland survive, mostly in association with other trees and shrubs Middle Shire River The Middle Shire, from Matope to Kapichira, drops over 370 m in elevation over a distance of about 90 km. The gradient is gentle but gradually increases. Beginning around Matope, the River falls below the surrounding land and is flanked by alluvial terraces. This section also contains a series of cataracts (Kholombidzo, Nkula, Tedzani, Hamilton and Kapichira). Additional River characteristics include: meanders, rapids flanked by islets, and abandoned river channels. Several tributaries join the main channel, the most important being the perennial Lisungwe and Mkurumadzi Rivers. The Middle Shire Valley is relatively dry with annual rainfall in the order of mm. Mean annual temperature is approximately 24 C. The Middle Shire Valley forms the floor of the Great African Rift Valley and consists of undulating but rugged and densely dissected country. The area is largely underlain by Precambian metamorphic rocks (basement complex) consisting of gneisses and granulite. Soils in general are moderately deep, well drained, brown, and medium textured with weathered rock in the subsoil and gravel and stones throughout the profile. Population pressure in the Middle Shire Valley has resulted in the natural vegetation only appearing in areas deemed unsuitable or very marginal for cultivation. The remnants of the natural vegetation may be classed as woodlands and savannahs Lower Shire River Downstream of Kapichira, the Shire River Valley widens in to a flat alluvial plain. For about 150 km the Shire meanders over this plain. The area is relatively dry, recording rainfall of 700-1,000 mm annually. Dry spells lasting several weeks are not uncommon during the period of the rainy season. Highest mean monthly temperatures fall into the range of C during the rainfed crop growing season. The Lower Shire Valley is a continuation of the Rift Valley, as noted in the preceding sections. The floor of the Rift Valley is flanked by escarpments associated with the major fault lines which follow a south-east to north-west trend. Relief is dominated by the Chikwawa-Thyolo Escarpment to the east, the Majete Escarpment to the north-west, the Nsanje Hills to the south-west and the Lower Shire uplands to the west. Altitude in the Lower Shire Valley ranges from 50 m to 200 m, while to the north-west and south-west the land rises to nearly 1,000m. The hills and uplands surrounding the Lower Shire Valley predominantly consist of basement complex rocks, gneisses and granulite. The Valley itself is characterised by Tertiary unconsolidated sediments. Soils in the Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

90 Valley are deep, medium to fine texture, brown to very dark in colour and have a drainage ability, which varies from good to very poor. The upper reaches of this section host large areas of commercial sugar cane agriculture, whilst the lower reaches are dominated by the seasonally flooded Elephant Marsh with its attendant market garden agriculture and artisanal fisheries. The natural vegetation of the Valley has been largely removed or extensively disturbed by cultivation. Seasonally waterlogged areas adjacent to the Shire River support productive grass vegetation Hydrology of Lake Malawi and the Shire River Lake Malawi The level of Lake Malawi normally exhibits an annual fluctuation of about one metre between the low dry season and the high level at the end of the rainy season. In addition, the Lake has historically exhibited a pattern of temporal cyclical fluctuation. From the evidence of writings of early travellers it is inferred that the Lake level was very low in 1830, very high in , falling in , high in 1882, very low in 1890, but rising in The factors determining the fluctuations in water level are the runoff from the lakes large catchment (half of which originates in the high rainfall areas of southern Tanzania), the rainfall directly on the Lake surface, and loss through evaporation. The residual water, the so called freewater, is either used to change the volume of the lake or is released into the Shire River. The evaporation is the conservative factor, leaving the water level to be heavily dependent on changes in rainfall pattern in particular on the rainfall directly on the Lake. Representative data on Lake rainfall and evaporation are not available. An indicative water balance for the Lake, based on hydrological and meteorological data from the period 1953/54 to 1973/74, has been presented in a Government of Malawi UNDP/WMP project report; calculated into average annual flow figures the balance showed: Inflow from catchment: 595 m 3 /sec Rainfall on the Lake: 1230 m 3 /sec Lake evaporation: 1467m 3 /sec Free water (combined Lake outflow and change in Lake volume); 358 m 3 /sec Lake Malawi is a nutrient poor (oligotrophic) clear lake. The production of phytoplankton which forms the base of the food chain, and consequently influences the total fish production, is limited by the storage of nutrients. The influx of nutrients from the catchment runoff is mostly lost into a stratified anoxic layer originating at a depth of approximately 250 m. Some mixing and mobilisation of nutrients occurs in the cool winter season (May August), due to south-easterly trade winds. In the shallower south-eastern part of the Lake, the entire water column is mixed in the cold season and nutrients sedimented to the bottom are returned to the water column. The existence of internal upwelling under the influence of the strong southerly winds also helps in mixing the water and hence maintaining the productivity of the south east arm of the Lake. This leads to a much higher annual biological production in this part of the Lake compared to the northern segment Lake Outflow and Upper Shire The flow of the Shire River is primarily determined by two components: the outflow from Lake Malawi; and the inflow from the drainage basin downstream of the Lake. Of these, the outflow from the Lake is the dominant component. In the dry season, Lake outflow accounts for virtually all the water entering the Shire River. Evaporation from Lake Malombe accounts for an average annual loss of about 20 m 3 /sec. The Kamuzu Barrage provides some flow regulation capacity and is used to secure the base flow needed for hydropower generation. The long term variations in Lake levels, and thus the outflow, have caused periodic changes in the flow characteristics of the Shire on a historical time-scale. The River ceased to flow altogether in 1915 and only gradually regained its regular flow in when the water level in Lake Malawi rose to the height whereby it was able to break through the sand and sediment bars that the tributaries had built up in the Upper Shire. From the early 1950s until mid 1970s, the River flow at Liwonde was maintained at annual average levels fluctuating between 200 m 3 /sec and 600 m 3 /sec. The Kamuzu Barrage was completed in 1965 for the purposes of storing water in the upper part of the River to maintain a minimum flow of at least 170 m 3 /sec Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

91 downstream. In the period from the mid-1970s and through the 1980s the Lake level increased and the average annual flow was recorded at between m 3 /sec. In the early 1990s, however, the whole of south-eastern Africa experienced serious droughts and a dramatic fall in the Lake level and a reduction in Shire flow resulted. The lowest water level in 40 years was recorded in December 1995, when the lake level reached masl (meters above sea level) with a resulting downstream water flow (measured at Liwonde) of 129 m 3 /sec. This flow was well below that required for maintaining hydropower production at the plants in the Middle Shire. An additional factor that affects the outflow is the status of silt and sand deposits in the river bed of the Upper Shire River. As a result of the modest slope of the whole upper course, sand bars are able to form, preventing some of the water from flowing downstream. A reduced initial outflow from the Lake, due to low water levels, will further exacerbate the problem. A high river flow enables the silt barriers to be broken through and the flushing of sediment downstream. Some of the siltation changes are a result of the dynamics of the main River, but the major problems are caused by the seasonal tributaries and their discharge points in to the Shire. Silt transport by the tributaries has increased in recent times due to poor land-management and the intensification of agriculture in the catchment area. The relatively nutrient rich water exiting the Lake is further enriched as it flows downstream. The erosive force of the River and tributaries add to the concentration of nitrogen and phosphorus available for plant growth. Some of these nutrients will be used by planktonic algae in the River; others will be used to promote the growth of reed and wetland species, including floating or submerged plants. The water hyacinth (Eichhornia crassipes), a noxious floating plant, has recently established itself in the area and is benefiting from the nutrient rich water in some of the more sheltered parts of the Shire River Lake Malombe Lake Malombe is effectively an extension of the Upper Shire, located approximately mid-way between the outflow and Liwonde. Typically, the length of the Lake is about 30 km with a maximum width of about 15 km. The depth averages about 7 m, with a maximum of around 17 m. It is claimed that the depth has decreased in recent years. The existence and dimensions of the Lake are totally dependent on the discharge of the Shire River. Large seasonal fluctuations in Lake area are common and records exist of a total dry-out in the years the Shire ceased to flow in its upper course. Due to its shallow depth, the water in Lake Malombe is comprehensively mixed at all times of the year; the nutrients are efficiently recycled and the productive zone extends right to the bottom of the water column. Additionally, the Lake profits from the inflowing nutrient rich water, and surface runoff entering the Lake from adjacent land further contributes to the productivity levels. Readily observable indicators of the nutrient rich (eutrophic) character of the Lake are high levels of chlorophyll a, and the low transparency of the water Middle Shire Moving downstream the River is fed by a number of tributaries, most of them seasonal. The flow regime of the River here, therefore, exhibits greater variations between peak flows and low flows than the upper part. Typically in normal Lake outflow periods the rainy season flow is about 1.5 to 3 times the magnitude of the dry season flows. Although flood episodes of local origin are important, a significant proportion of peak average flow is due to higher lake levels towards the end of the rainy season. A comparison of Shire River flow discharge at Liwonde and that at Maganga (2.5 km downstream of Kapichira falls) indicates that the Middle Shire tributaries contribute about 30% of the Shire water in the rainy season, and less than 10% in the dry season Lower Shire The Lower Shire is slow flowing over a flat flood plain. During periods of high water level it overflows the area termed Elephant Marsh, creating a large wetland with a multitude of channels and pools. The Elephant Marsh in its downstream portion is also heavily influenced by the Ruo River running from the east at the Malawi Mozambique border. In some years, the Ruo River might even have a greater impact on the water table in the Marsh than the Shire River itself. Downstream of the Ruo River confluence is another extensive wetland area known as the Ndinde Marsh. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

92 The existing water withdrawal for irrigation purposes is at present only 12 m 3 /sec in the peak periods (SUCOMA Sugar Estates). The main water management problem faced by the irrigation interests in the Lower Shire is at present the heavy silt content of the water. Continuous mechanical removal of silt from the intake channel is needed to maintain irrigation operations on the SUCOMA estates Biotic Communities For the purpose of this rapid assessment the biotic communities represented within 100 m of the water edge were included in the assessment; the rationale being that this was the zone of influence of any fluctuation in water levels resulting from the Barrage upgrade. Five distinct biotic communities were identified as being represented within the project area during the field assessment, including: 1. Perennial Marshes 2. Seasonal Marshes Uncultivated 3. Seasonal Marshes - Under Cultivation (dambos) 4. Seasonal Flood Plains 5. Riverine Woodlands A brief description of the characteristics of each biotic community type is provided below Perennial Marshes Perennial Marshes are found along parts of the river and in certain sections of the lakeshore zone on both Lake Malawi and Lake Malombe. The permanent marsh areas are in relatively shallow water, but the soils are almost permanently inundated. The vegetation consists of tall reeds and aquatic grasses, and various smaller floating and/or submerged plants where the river current is either weak or virtually non-existent (Figure 5). Full descriptions of aquatic macrophytes in the Upper, Middle and Lower Shire are given in Blackmore, Dudly & Osborne (1988). Some common species are given in Table 1 to provide the reader with some detail for the aquatic systems that may be affected by this project. TABLE 1. COMMON AQUATIC MACROPHYTES FOUND IN PERENNIAL MARSHES OF THE SHIRE RIVER Reeds, rushes, sedges and grasses Cyperus papyrus Typha capensis Vossia cuspidate Phragmites mauritianus, P australis Thelipterys confluens (Pteridophyta) Submerged plants Ceratophyllum demersum Ottelia spp. (e.g. O. exserta., O. scabra) Potamogeton pectinatus, P. thunbergia Spirodela polyrhiza Water lilies and other floating aquatics Azolla pinnata Commelina benghalensis Ipomoea aquatica Nymphaea caerula, Nymphaea lotus Pycreus mundtii Salvinia hastata Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

93 Figure 5. Example of a Perennial Marsh Seasonal Marshes The lake and river levels rise and fall on average 1 m in a single year. These fluctuations are due to annual rainfall inflow mainly to Lake Malawi. The fluctuations are also affected by the rate of water release at the Kamuzu Barrage, so they are not true seasonal fluctuations, but a combination of natural and managed cycles of water flow. The species found in seasonal marsh areas are similar to those found in perennial marshes, with die-back of aquatic plants when water levels recede, plus the addition of plants that tolerate wet and dry soils (Figure 6). Water tolerant woody plants occur here, as do water tolerant grasses, sedges and reeds (Table 2). The landscape is generally flat with tall reeds and grasses, with scattered, short trees and shrubs that sometimes form small thickets. In addition to the species listed for the permanent swamp areas, there are a few emergent woody plants. These tend to be rooted in silt and humus accumulations formed on the tops of reed banks and papyrus suds. TABLE 2. COMMON EMERGENT WOODY PLANTS Woody Plants Reeds, rushes, sedges and grasses Hibiscus diversifolius Cyperus spp. (numerous) Mimosa pigra Echinochloa haploclada, E. pyramidalis Sesbania bispinosa Hemarthria altissima Syzygium cordatum Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

94 Figure 6. Example of an uncultivated seasonal marsh Seasonal Marshes Cultivated Most of the seasonal marshes around Lake Malombe and in the Upper Shire were under some form of dambo cultivation at the time of the assessment. Brinkman and Blokhuis (1986) define dambos as: areas that have free water at or on the surface for at least the major part of the growing season. The water is sufficiently shallow to allow the growth of a wetland crop or of natural vegetation rooted in the soil. The most commonly planted dambo crop is maize (Figure 7); other common crops are beans, peas, sweet potatoes and sugar cane. Vegetables such as cabbage, rape, pumpkin leaves, champiru or mustard are grown to a lesser extent (Klarer 2008). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

95 Figure 7. Example of cultivated seasonal marsh (dambo) Seasonal Floodplains Seasonal floodplains occur where there are shallow gradients from the lake and river body sides on to dry land. The land is more commonly dry than submerged here, as opposed to seasonal marshlands and/or swamp, which is more commonly submerged, but may dry out from time to time. Here we see scattered tall trees, predominantly palm species, and tree and scrub thickets, most commonly associated with low lying islands and termite mounds. Table 3 provides a list of typical plant species found in seasonal floodplains. The flat-lands areas are covered in tall grasses, reeds and sedges and in the lower lying locations, these merge in to seasonal marshlands (Figure 8). Some seasonal floodplain areas are also under cultivation. TABLE 3. TYPICAL FLOODPLAIN PLANT SPECIES Woody plants - Trees Hyphaene petersiana Phoenix reclinata Kigelia Africana Afzelia quanzensis Acacia nilotica Acacia nigrescens Dalbergia melanoxylon Woody plants - Shrubs Adina spp. Carissa spp. Dichrostachys cinerea Ficus spp. Garcinia spp. Grewia spp. Gymnosporia senegalensis Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

96 Woody plants - Trees Combretum imberbe Lonchocarpus capassa Trichilia capitata Xanthoxercis zambeziaca Ziziphus mucronata Reeds, rushes, sedges and grasses Cyperus spp. (numerous) Kyllinga spp. Mariscus spp. Pycreus spp. Brachiaria brizantha Echinochloa haploclada, E. pyramidalis Hemarthria altissima Oryza spp. Phragmites australis, P. mauritianus Sporobolus festivus, S. robustus Woody plants - Shrubs Maerua triphyllum Ormocarpum kirkii Rhus spp. Non-grasses Asparagus spp. Hibiscus spp. Leonotis spp. Ocimum canum Pluchea spp. Sanseveria aethiopica Sesamum spp. Sesbania spp. Veronia spp. Figure 8. Example of seasonal flood plains Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

97 Riverine Woodlands Riverine Woodlands tend to predominate in the steeper areas of the Middle Shire where the river banks are drier, these woodlands tend to be species rich (Table 4). The trees are relatively tall, varying in height from 12 to 30 m, with secondary shrub and emerging tree understory (Figure 9). The tree canopies are relatively close and vary from interlocking to slightly apart (up to about 5 m). TABLE 4. Woody plants Acacia xanthophloea Albizia brevifolia Balanites aegyptica Cissus quadrangulifolia Colophospermum mopane COMMON PLANTS ASSOCIATED WITH RIVERINE WOODLANDS Combretum mossambicensis Commiphora africana Dalbergia spp. Cont. Dichrostachys cinerea Diospyros spp. Faedherbia albida Strychnos spp. Tamarindus indicus Terminalia sambesiaca Thylachium africanum Xeromphis obovata Figure 9. Example of riverine woodland Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

98 2 Methods This study took the form of a rapid ecological assessment of the project area conducted between 30/10/2011 and 4/11/ Field Studies The study team assessed the biotic communities at 30 different sites adjacent to the Shire River, Lake Malombe and Lake Malawi. Each site was within 100 m of the high water edge; the location of each site was largely determined by its accessibility from the road network. Each site was assessed by a team of ecologists specialising in impact assessment and botany, including: Dr Michael Clarke SMEC Certified Environmental Practitioner in Impact Assessment; and Dr Montford Mwanyambo National Herbarium of Malawi at Zomba. At each site the survey team characterised the biotic community according to the categories described in Section A representative patch of the dominant vegetation type was captured within a GPS polygon to act as a training site for spectral imagery analysis of the project area (see Section 2.2). Each site was photographed with a GPS enabled camera. The botanists then developed a vegetative species list for each site, and collected representative samples of the species present for cataloguing at the national herbarium. All data was entered in to a site specific data sheet; these are attached in Appendix Spectral Analysis In order to quantify the extent of impacts from different inundation models, vegetation classes were defined within the boundary of each inundation model. Vegetation types can be classed using either supervised or unsupervised classification techniques or manual delineation using control points from a ground survey. Supervised classification generally requires the use of multi-spectral imagery to allow for accurate delineation of vegetation classes. No multi-spectral imagery was available for use in the study area, however high resolution aerial photography (taken in 2010) was obtained to assist with the hydraulic modelling. Based on a review of the aerial photography, the number of vegetation classes required and the ability of unsupervised classification techniques to distinguish between vegetation classes, it was determined that a combination of manual delineation and unsupervised classification would provide the most accurate mapping, using the aerial photography as a guide. By using the same imagery used as a backdrop to the hydraulic modelling we also ensured no variation in the boundary of vegetation classes. Based on the field survey the following broad vegetation classes were identified within the study area: Perennial Marshes Seasonal Marshes Uncultivated Seasonal Marshes - Under Cultivation (dambos) Seasonal Flood Plains Riverine Woodlands Dry Bush Savannah Cleared Woodland The distribution of vegetation classes within the study area were initially assessed by reclassing the output of each inundation model (473, 474, 475, and 476). This resulted in a broad boundary (unsupervised classification) for the Water and Permanent inundation classes as well as the Marsh and Floodplain areas. To provide a more accurate vegetation map, the rest of the vegetation classes were manually delineated in ArcGIS (ESRI) and each of the Marsh and Floodplain boundaries were re-inspected at a Scale between 1:100 and 1:4000 and divided where obvious errors were present. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

99 By delineating the vegetation class boundaries for the inundation model that covered the greatest area first (model 476) we were able to use Geo-processing tools ArcGIS (ESRI) to reclassify the boundaries of the other (smaller) inundation models. The output of each inundation model was then divided into two sections, the top half from Mponda to the Kamuzu Barrage and the bottom half from the Kamuzu Barrage to Chigaru (the extent of the available aerial photography). In order to quantify the impact of each inundation model on the vegetation classes, the total area within each inundation model (top and bottom halves) was calculated using geometry and summary tools in ArcGIS (ESRI). A separate map showing the results of each calculation was provided for visual comparisons (see Section 3) Outputs Two types of output were produced (see Section 3): i. Mosaic maps showing the distribution of biotic communities within the project area. ii. The area of each biotic community type within the project area was calculated. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

100 3 Results TABLE 5. BIOTIC COMMUNITIES REPRESENTED IN FIELD STUDIES The number of each type of biotic community represented in the field studies are shown in Table 5. A data sheet for each site is provided in Appendix 1. This information was used to inform the mapping of biotic community mosaic maps in Section 3.1. The outputs from the Design Consultant s hydraulic modelling for the upgraded Barrage were overlaid onto the biotic community mapping. These outputs showed the expected upstream and downstream inundation footprints for a range of water release scenarios. By overlaying a range of inundation footprints on top of the biotic community mapping the areas of each biotic community type that will be impacted at various water levels at the Barrage, was calculated for a range of release scenarios. 3.1 Modelling The existing Kamuzu Barrage has the capacity to control water levels in Lake Malawi within a range from masl (Lowest Regulated Water Level) to masl (Highest Regulated Water Level). The proposed Project to raise the height of the Barrage will increase the HRWL in Lake Malawi by 20-40cm (i.e. from to masl depending on social/environmental impacts; Norplan, 2011). This is still within the range of historical lake levels, which records show vary from a low of masl in 1915 to a high of masl in 1980 (a maximum difference of 7.31 m well beyond the influence of the Barrage; Norconsult 2003). The purpose of this assessment is to provide information on how manipulations of water levels within the operating range of the modified (i.e. raised) Barrage may impact the land and habitats immediately adjacent to the Shire River, Lake Malombe and the southern part of Lake Malawi. It is intended that this information will then be used to inform the development of an operational schedule for the modified Barrage that minimises impacts to fringing habitats that are sensitive to seasonal or temporal modifications in the existing flooding cycles of the system. Hydraulic modelling has considered five scenarios ranging from 473 masl (i.e. below the existing LRWL) to 476 (i.e. above the proposed HRWL for the raised Barrage). The levels included in the models are: 473; 474; 475; and 476 masl. Maps showing the inundation of each habitat / land use type upstream and downstream of the Barrage at the various levels are provided in Figures 10 to 19. The area of each habitat type that will be inundated at each water level immediately upstream and downstream of the Barrage is summarised in Figures 20 and 21. Site Perennial Marshes Seasonal Marshes Uncultivated Seasonal Marshes - Under Cultivation (dambos) Seasonal Flood Plains Riverine Woodlands Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

101 Figure 10. Inundation map 473 masl Lake Malawi to Kamuzu Barrage Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

102 Figure 11. Inundation map 473 masl Kamuzu Barrage to Chigaru Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

103 Figure 12. Inundation map 474 masl Lake Malawi to Kamuzu Barrage Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

104 Figure 13. Inundation map 474 masl Kamuzu Barrage to Chigaru Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

105 Figure 14. Inundation map 475 masl Lake Malawi to Kamuzu Barrage Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

106 Figure 15. Inundation map 475 masl Kamuzu Barrage to Chigaru Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

107 Figure 16. Inundation map masl Lake Malawi to Kamuzu Barrage Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

108 Figure 17. Inundation map masl Kamuzu Barrage to Chigaru Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

109 Figure 18. Inundation map 476 masl Lake Malawi to Kamuzu Barrage Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

110 Figure 19. Inundation map 476 masl Kamuzu Barrage to Chigaru Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

111 Inundation Above Barrage s re e c ta H masl masl 475 masl 474 masl 473 masl 0 Perrenial Marshes (Grade 1) Perrennial Marshes (Grade 2) Seasonal Flood plain Seasonal Floodplain (cu ltivated) Seasonal Marsh Season al Marsh (cultivated) Riverine Woodlands Habitat / Land Use Type Figure 20. Habitat inundation above Kamuzu Barrage over a range of water levels Inundation Below Barrage r e s ta e c H masl masl 475 masl 474 masl 473 masl 0 Perrenial Marshes (Grade 1) Perrennial Marshes (Grade 2) Seasonal Floodplain Seasonal Floodplain (cultivated) Seasonal Marsh Seasonal Marsh (cultivated) Riverine Woodlands Habitat / Land Use Type Figure 21. Habitat inundation below Kamuzu Barrage over a range of water levels Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

112 3.2 Impacts Upstream of Kamuzu Barrage Area of Inundation It can be seen from Figures 10 and 20 that upstream of the Barrage significant inundation of fringing habitats only occurs at water levels of 475 masl and above. As the existing Barrage maintains a mean water level of masl only changes to the structure that result in mean water levels above masl are relevant in terms of the ecological impacts on fringing habitats. At water levels of masl and above the most highly impacted habitat type is Perennial Marsh, with 3, ha and 4, ha of this habitat type being inundated at masl and 476 masl respectively. Most of the Perennial Marsh that is susceptible to inundation lies within the boundaries of Liwonde National Park. Inundation of other habitat types is minimal at 474 masl and moderate at water levels of above 475, with the next highest maximum inundation being that of Seasonal Marsh habitat (i.e. 695 ha at 476 masl) Inundation Period In addition to the area of habitat inundated any changes in the inundation period must also be considered in an ecological context as this can cause a change in species composition. The data presented in Table 6 shows the predicted increases in the amount of time (expressed in both days per year and proportion of the year) that perennial marsh habitat will be inundated above the existing baseline level with a 20 cm and 40 cm increase in Barrage height. The predictions in Table 6 indicate that there is no change in the inundation period of perennial marsh habitat for an increase in regulating height of either 20 cm or 40 cm when the level of Lake Malawi is 474 masl. However, small predicted changes in duration do occur at higher Lake levels and when the Barrage s HRWL is increased by either 20 cm or 40 cm. For example, at a Lake Malawi water level of 475 masl there is negligible difference in inundation period between a 20 cm or 40 cm increase in HRWL of the Barrage. The differences are increased slightly at Lake levels of masl and 476 masl. As perennial marshlands are permanent to semi-permanent wet habitats, inundation is likely to be manifested by an increase in water depth rather than an increase in the time that land is covered with water that would otherwise be dry. In addition, as the projected increases in inundation period are small, they probably lie within the natural annual variation of inundation experienced in the system (however data is not available to substantiate this supposition). Consequently the affect of this impact on available habitat quantity and quality above the Barrage is predicted to be low Downstream of Kamuzu Barrage Area of Inundation Downstream of the Barrage the magnitude of habitat inundation is significantly less than above the structure as the Shire River narrows and becomes more incised until it reaches the low wide floodplain of the Elephant Marsh. Water flows are faster in this section and there are less perennial marsh areas which require relatively slow flowing water in which to become established. In this section significant habitat inundation is only observed at water levels of masl and above with seasonal floodplain and seasonal marsh habitats being the most severely impacted (Figures 17 and 21). Perennial marsh habitat is poorly represented in this section of the River and inundation of this habitat type is approximately 1/5 of that above the Barrage at the highest water level of 476 masl Inundation Period The data presented in Table 6 shows that seasonal flood plain and seasonal marsh habitats are more significantly impacted by increased inundation period downstream of the Barrage than upstream. This could potentially be an issue for egg laying reptiles, such as crocodiles and turtles that inhabit the Shire River below the Barrage. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

113 Female crocodiles will select a sandbank that is above flood water level, with good drainage and cover nearby to build a nest (Compass, 2000). Such conditions are only found in seasonal flood plain and seasonal marsh habitats in this section of the Shire River. Typically, crocodile eggs can only withstand flooding for a period of up to 12 hours (Mazzotti, 2002). If the water regulation regime at the Barrage leads to nesting sites being inundated for longer than this, decreased rates of egg viability may result. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

114 TABLE 6. Lake Malawi Water Level (masl) INUNDATION PERIOD Existing Barrage Height ( masl) 68% (248 days) 40% (146 days) 22% (80 days) 8% (33 days) Duration of Inundation 20 cm Increase in Barrage Height ( masl) 68% (248 days) 44% (161 days) 26% (95 days) 12% (40 days) 40 cm Increase in Barrage Height ( masl) 68% (248 days) 45% (164 days) 30% (110 days) 15% (51 days) Upstream of Kamuzu Barrage to Mponda Area of Perrennial Marsh Inundated (ha) Downstream of Kamuzu Barrage to Chigaru Total Area of Seasonal Floodplain (uncultivated) Inundated (ha) Upstream of Kamuzu Barrage to Mponda Downstream of Kamuzu Barrage to Chigaru Total Area of Seasonal Marsh (uncultivated) Inundated (ha) Upstream of Kamuzu Barrage to Mponda Downstream of Kamuzu Barrage to Chigaru , , , , , , , , , , , ,486.3 Total Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

115 4 Discussion At a water level of 475 masl at Lake Malawi (i.e. the no Barrage condition), the existing Barrage height results in an increase of 0.32 masl to a baseline water level of masl (i.e. the existing baseline condition). A 20 cm increase in the height of the Barrage will increase the water level to masl resulting in the upstream inundation of an additional ha (i.e. a 9.15% increase) of perennial marsh habitat above the baseline condition. A 20 cm increase in Barrage height will extend the inundation period of this ha by 15 days above the baseline condition; whereas a 40 cm increase will extend the inundation period by 30 days. At a water level of 476 masl at Lake Malawi a 20 cm increase in the height of the Barrage will result in the upstream inundation of an additional 1, ha (i.e. a 40.01% increase) of perennial marsh habitat above the baseline condition. A 20 cm increase in Barrage height will extend the inundation period of this 1, ha by 8 days above the baseline condition; whereas a 40 cm increase will extend the inundation period by 18 days. The majority of the additional perennial marsh habitat that will be inundated by any increase in the height of the Barrage lies within the boundaries of Liwonde National Park and has significant ecological value to the wildlife inhabiting the Park. Perennial marshes are essential to the maintenance of the aquatic system in this part of the Shire River as they provide breeding and rearing habitat for economically important species of fish, invertebrates and birds. They also contribute significantly the productivity of the aquatic system in this reach of the river. Perennial marshes and seasonal floodplains are also important foraging areas for elephants and hippopotamus residing within Liwonde National Park (Government of Malawi, 2005). As described in Section 1, the main vegetative types found in Perrenial Marshes are emergent species (e.g. Cyperus papyrus; Typha capensis; Vossia cuspidate; Phragmites mauritianus; P australis; and Thelipterys confluens), floating aquatics (e.g. Azolla pinnata; Commelina benghalensis; Ipomoea aquatica; Nymphaea caerula; Nymphaea lotus; Pycreus mundtii; and Salvinia hastate), and submerged plants (e.g. Ceratophyllum demersum; Ottelia spp; Potamogeton pectinatus; P. thunbergia; and Spirodela polyrhiza). While floating aquatics and submerged plants can tolerate extended periods of submergence, emergent species are less tolerant. Increasing the height of the Barrage by 20 cm results in minor increases in inundation periods above the existing baseline condition (i.e. 15 days and 8 days for the and 476 masl scenarios modelled); this small increase in the period of inundation should be within the tolerance range of emergent aquatic species. However, increasing the Barrage height by 40 cm results in significantly longer inundation periods (i.e. 30 days and 18 days for the and 476 masl scenarios modelled), which may be outside the tolerance range of emergent species and result in a shift in species composition. A 20 cm rise in Barrage height results in minor increases in the area and period of inundation of Perennial Marsh habitat upstream of the Barrage and this is not likely to result in any significant change in species composition or structure of this habitat type. However, a 40 cm rise in height results in the inundation of a much larger area for extended periods of time which is more likely to result in a change in the species composition and structure of this habitat type within Liwonde National Park. Hence, a 40cm rise in the height of the Barrage can only be recommended if accompanied by a monitoring program of the health/extent of Perennial Marshes within Liwonde National Park to track any changes in habitat extent or condition. This should be accompanied by an adaptive management program which regulates the operation of the Barrage in response to the findings of the monitoring program. Downstream of the Barrage water levels above masl primarily result in the inundation of an additional ha (i.e. a 270% increase) of seasonal flood plains and ha (i.e. a 643% increase) of seasonal marshes above the baseline condition. The primary impact from this will be social, as the majority of these seasonally flooded areas have limited ecological value because they have been modified by the local community to grow crops. Increased periods of inundation will effectively reduce the growing season on the fertile River margins resulting in social and economic impacts for the local community. This issue will be addressed in the Environmental and Social Impact Assessment Report. Increased inundation periods of seasonal flood plains and seasonal marshes downstream of the Barrage could be an issue for nesting crocodiles and turtles. For example, the dry-season nesting period of the Nile crocodile results in hatching at the start of the wet season, increased periods of inundation at this time Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

116 resulting from operation of the Barrage may destroy eggs if nests are flooded. In addition, the sandbanks in which they nest are often deposited by wet-season water flows, if Barrage operations result in increased water flows during the early wet-season heavy rains the sandbanks can be severely eroded or washed away, inundating the eggs or washing them out into deep water. 4.1 Environmental Flow Requirements The primary cause of impact to the fringing terrestrial habitats resulting from implementation of the project will be associated with changes in environmental flows and flooding regimes. These impacts will be most pronounced in the low gradient areas of Southern Lake Malawi, Lake Malombe and the Upper Shire River. The Middle Shire River is characterised by a steep gradient, well defined incised river channel and riverine woodlands and escarpments. This reach of the river will be relatively immune to variations in volumes or patterns of water release from the Barrage as the channel has capacity to absorb relatively large fluctuations in water level without flooding the surrounding terrestrial habitats. This will hold true provided a minimal baseline environmental flow is maintained in the Middle Shire Reach. It is important to maintain the seasonal flooding of the marshlands in order to preserve floodplain and seasonal marsh habitat and support the seasonal dambo agriculture that much of the local community relies upon for sustenance during the dry season when seasonal marshes provide temporal fertile land for cropping. Downstream of the Barrage it is important that operations do not result in artificially prolonged inundation of crocodile and turtle nesting sites located on seasonal flood plain and seasonal marshland during the critical nesting season from October to January. 4.2 Implications for Operation of Kamuzu Barrage In order to minimise the impacts of the Project to fringing habitats, the Barrage should be operated to mimic the natural seasonal flooding regime of the Shire River, Lake Malombe and the lower part of Lake Malawi, as closely as possible. This is particularly critical when the Barrage is operating at levels resulting in water levels of 475 masl and above at Lake Malawi. A typical annual hydrograph for the Shire River is shown in Figure 22; peak flows typically occur in February and March while minimum flows are experienced during the dry season in September and October. The proposed Project to raise the height of the Barrage will increase the existing HRWL ( masl) in Lake Malawi by 20-40cm i.e masl (which is within the historical range of Lake levels, with a high of recorded in 1980). Modelling shows that there will be a minimal change in the area of habitat inundated and the period of inundation associated with a 20 cm increase in Barrage height. However, a potentially significant increase in the area of inundation and the inundation period is possible upstream of the Barrage if the height is raised by 40 cm. This may significantly impact important perennial marsh habitat located along the boundary between the Shire River and Liwonde National Park. It is recommended that if the Barrage is raised by 40 cm, a Perennial Marsh habitat monitoring program is implemented within the boundaries of the Park and an adaptive management program is adopted whereby the operation of the Barrage is regulated in response to the findings of the monitoring program. It is also recommended that if levels do need to be increased above masl then the Barrage should be operated to mimic the seasonal flooding patterns of the system i.e. levels of masl permitted during the wet season only (i.e. Dec April); maintain levels below masl during the dry season (May Oct). This operational regime will also minimise the incidences of prolonged (i.e. greater than existing levels) inundation of seasonal floodplain and marshland habitats downstream of the Barrage. This habitat type may support crocodile (and turtle) nesting, during the critical incubation period lasting from October to December (Compass, 2000). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

117 Figure 22. Shire River Hydrograph (World Bank 2010) Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

118 5 References Brinkman, R. and Blokhuis, W.A. (1986). Classification of the soils. In The Wetlands and Rice in Sub- Saharan Africa. Juo, A.S.R. and J.A. Love (eds.). IITA, Nigeria. Compass (2000). Crocodile and Hippopotamus Management in the Lower Shire. Community Partnerships for Sustainable Resource Management in Malawi. Prepared by: Felix Kalowekamo (Consultant) Development Alternatives, Inc Woodmont Ave., Suite 200 Phekani House Bethesda, MD 20814, USA. Government of Malawi (2005). Liwonde Mangochi Protected Area Complex, Malawi. General Management Plan. Kraler, A. (2008). Agriculture in Dambos Around Mzuzu City, Malawi; a Sustainability Assessment. Honors Thesis. Amanda Klarer, International Development Studies and Environmental Science, Dalhousie University, Halifax, Canada. Mazzotti, F. (2002). American Crocodiles (Crocodylus acutus) in Florida. Department of Wildlife Ecology and Conservation, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida. Published: 1999, as SS-WIS-38. Revised: December, 2002 as WEC-QRS Revised: May 2003 as WEC 38. Reviewed May Norconsult (2003). Feasibility Study of Integrated Water Resources Development plan for Lake Malawi 7 Shire River Systems Lake Malawi Level Control Stage 2 Main Report. Republic of Malawi, Ministry of Water Development, pages. Norplan (2011). Phase 1 Report: Optimised Design for Detailed Design of the Upgrade Kamuzu Barrage. Commissioned by the Government of the Republic of Malawi, pages. World Bank (2010). The Zambezi River Basin A Multi-Sector Investment Opportunities Analysis. The International Bank for Reconstruction and Development, World Bank. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

119 Appendix 1: Description of Biotic Communities at Sampling Sites SITE NUMBER: 01 DATE: 29/10/2011 TIME: 06:41 COORDINATES: S / E SITE NAME. Elephant Marsh BIOME CATEGORY: Seasonal Flood Plain DESCRIPTION: Seasonally wet grassland on an alluvium flood plain. Sparsely covered with grasses and some reeds and papyrus in wet areas and at the margins of the river. SPECIES LIST: Species recorded. Family Cynodon dactylon (L.) Pers. Poaceae Potamogeton Potamogetonaceae Parkinsonia aculeata L. Fabaceae- Mimosoideae Acacia (invasive one) Fabaceae- Mimosoideae Eichhornia crassipes (Mart.) Solms Pontederiaceae. Phragmites mauritianus Kunth. Poaceae Eichornia crassipes Pontederiaceae. Echnochloa colona (L.) Link Poacaea Cyperus papyrus L. Cyperaceae Panicum Poaceae Vernonia cinerea Asteraceeae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

120 Mimosa pigra L. Sphaelanthus randii S.Moore var. Randii Azolla nilotica Mett. Helotropium sp. Vossia cuspidata (Roxb.) Griff. Ludwigia erecta (L.) Hara Ipomoea acquatica Forssk. Gomphocarpus fruticosus (L.) Ait. Mimosideae Asteraceeae Azollaceae Boraginaceae Poaceae Onagraceae Convolvulaceae Asclepidoideae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

121 SITE NUMBER: 03 DATE: 29/10/2011 TIME: 08:45 COORDINATES: S / E SITE NAME. Elephant Marsh BIOME CATEGORY: Perennial Marsh DESCRIPTION: Perennial marsh area in the middle of Elephant Marsh. Comprised largely of emergent or floating macrophytes. This area will remain submerged throughout the year. SPECIES LIST: Vossia cuspidata (Roxb.) Griff. Mimosa pigra L. Persicaria serrulata (Lag.) Webb & Moq. Ipomoea Ipomoea aquatica Forssk. Cissampelos mucronata A. Rich. Ziziphus mucronata Willd. Ricinus communis L. Phragmites mauritianus Kunth. Vernonia Calotropis procera (Aiton) W.T. Aiton Ficus Poaceae Fabaceae Mimosoideae Polygonaceae Convolvulaceae Convolvulaceae Menispermaceae Rhamnaceae Euphorbiaceae Poaceae Asteraceae Asclepidoideae Moracea Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

122 Tridax procumbens L. Lactuca Mangifera indica L. Rynchosia Eichhornia crassipes (Mart.) Solms Pistia stratoites L. Zea mays L. Cyperus cyperoides (L.) Kuntze Schoenoplectus corymbosus (Roem. et Schutt) J.Reyn Echinochloa colona (L.) Link Nymphea sp. Ottelia ulvifolia (Plunch.) Walp. Cyperus laxus Lam. subsp. buchholzii (Boeck) K. Lye Acacia xanthophloea Benth. Philenoptera violacea (Klotze) Shrire Persicaria limbata (Meisn.)Hara Asteraceae Asteraceae Anacardiaceae Papilinoideae Pontederiaceae. Araceae Poaceae Cyperaceae Cyperaceae Poaceae Nympeaceae Hydrocharitaceae Cyperaceae Mimosoideae Papilinoideae Polygonaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

123 SITE NUMBER: 04 DATE: 29/10/2011 TIME: 09:32 COORDINATES: S / E SITE NAME. Elephant Marsh BIOME CATEGORY: Perennial Marsh DESCRIPTION: Perennial marsh area in the middle of Elephant Marsh. Comprised largely of emergent or floating macrophytes. This area will remain submerged throughout the year. SPECIES LIST: Vossia cuspidata (Roxb.) Griff. Mimosa pigra L. Persicaria serrulata (Lag.) Webb & Moq. Ipomoea Ipomoea aquatica Forssk. Cissampelos mucronata A. Rich. Ziziphus mucronata Willd. Ricinus communis L. Phragmites mauritianus Kunth. Vernonia Calotropis procera (Aiton) W.T. Aiton Ficus Poaceae Fabaceae Mimosoideae Polygonaceae Convolvulaceae Convolvulaceae Menispermaceae Rhamnaceae Euphorbiaceae Poaceae Asteraceae Asclepidoideae Moracea Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

124 Tridax procumbens L. Lactuca Mangifera indica L. Rynchosia Eichhornia crassipes (Mart.) Solms Pistia stratoites L. Zea mays L. Cyperus cyperoides (L.) Kuntze Echinochloa colona (L.) Link Nymphea sp. Ottelia ulvifolia (Plunch.) Walp. Cyperus laxus Lam. subsp. buchholzii (Boeck) K. Lye Acacia xanthophloea Benth. Persicaria limbata (Meisn.)Hara Asteraceae Asteraceae Anacardiaceae Papilinoideae Pontederiaceae. Araceae Poaceae Cyperaceae Poaceae Nympeaceae Hydrocharitaceae Cyperaceae Mimosoideae Polygonaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

125 SITE NUMBER: 05 DATE: 29/10/2011 TIME: 12:40 COORDINATES: S / E SITE NAME. Elephant Marsh BIOME CATEGORY: Perennial Marsh DESCRIPTION: This site is located within Elephant Marsh; it is a perennial marsh area dominated by Typher spp. SPECIES LIST: Phragmites mauritiana Kunth. Vossia cuspidata (Roxb.) Griff. Eleusine indica (L.) Gaertn. Sida alba L. Mangifera indica L. Nidorella auricullata DC susbsp. auriculata Musa paradisiaca L. Adansonia digitata L. Pennisetum purpureum Schumach. Leucena leucocephala (Lam.) De Wit Zea mays L. Carica papaya L. Ziziphus abyssinica Hochst. ex A.Rich. Poaceae Poaceae Poaceae Malvaceae Anacardiaceae Asteraceae Musaceae Bombacaceae Poaceae Mimosoideae Poaceae Caricaceae Rhamnaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

126 Cyperus mundtii (Nees) Kunth. Echinochloa colona (L.) Link Sphaeranthus pedancularis DC subsp. rogesii Commelina benghalensis L. Ficus capreifolia Delile Solanum esculentum L. Ocimum gratissimum (L.) subsp. gratissimum Solanum sp. Persicaria attenuata (R.Br.) Sojak susbsp. africanum Mimosa pigra L. Abelmoschus esculentus (L.) Moench Sesbania macrantha Welw. ex Phill. & Hutch. Prosopis glandulosa Torrey Cyperaceae Poaceae Asteraceae Commelinaceae Moraceae Solanaceae Lamiaceae Solanaceae Polygonaceae Mimosoideae Malvaceae Papilionoideae Mimosoideae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

127 SITE NUMBER: 06 DATE: 29/10/2011 TIME: 12:52 COORDINATES: S / E SITE NAME. Lower Shire River BIOME CATEGORY: Seasonal Marshes - Under Cultivation (dambos) DESCRIPTION: This site is a seasonally wet grassland which is under dambo cultivation the principal crop being maize. SPECIES LIST: Various crops Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

128 SITE NUMBER: 07 DATE: 29/10/2011 TIME: 14:13 COORDINATES: S / E SITE NAME. Kasinthula Growers Pump Station BIOME CATEGORY: Seasonal Marshes - Under Cultivation (dambos) DESCRIPTION: This is a disturbed seasonally wet grassland area which is under dambo cultivation. Crops include cassava, maize, peas, cassava and tomatoes. Some perennial marsh vegetation exists at the river margin. A large pumping station is located just upstream of this site which extracts a lot of water from the river for irrigation of sugar cane fields. SPECIES LIST: Philenoptera violacea (Klotze) Shrire Trichilia emetica Vahl Phragmites mauritiana Kunth. Senna siamea (L.) Irwin & Bainesby Capparis erythrocarpus Isert Sterculia africana (Lour.) Fiori Diospyros senensis Klotz. Lufa cylindrica (L.) M.Roem Combretum microphyllum (Klotz.) Engl. Cryptolepis oblongifolia (Meisn.) Schltr. Papilinoideae Meliaceae Poaceae Caesalpinoideae Capparaceae Sterculiacaea Ebenaceae Curcubitaceae Combretaceae Asclepidoideae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

129 Azadirachta indica A.Juss. Calotropis procera (Ait.) Ait. f. Jatropha sp. Melanthera scandens (Schum. & Thonn.) Roberty Tabernamontana elegans Stapf. Abelmoschus esculentus (L.) Moench Musa paradisiaca L. Maerua angolensis DC Vetiveria nigritana (Benth.) Stapf. Euphorbia hirta L. Nidorella auriculata DC Cynodon dactylon (L.) Pers Echinochloa colona (L.) Link Sorghum bicolor (L.) Moench Senna alata Panicum maximum L. Phyllanthus reticulata Poir var. reticulata Ficus capreifolia Delile Psidium guajava L. Albizia atunesiana Harv. Solanum sp. Cissampelos mucronata A.Rich. Ageratum conyzoides L. subsp houstonianum (MMI) Sahn. Verbena sp. Lantana camara L. Zea mays L. Vigna unguiculata (L.) Walp. Ipomoea batatas (L.) Lam. Mangifera indica L. Solanum esculentum L. Cajanus cajan (L.) Mill. Amaranthus hybridus Meliaceae Asclepidoideae Euphorbiaceae Asteraceae Apocynaceae Malvaceae Musaceae Poaceae Euphorbiaceae Asteraceae Poaceae Poaceae Poaceae Caesalpinoideae Poaceae Euphorbiaceae Moraceae Myrtaceae Solanaceae Menispermaceae Asteraceae Verbenaceae Poaceae Papilinoideae Convovulacaea Anacardiaceae Solanaceae Papilinoideae Amaranthaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

130 SITE NUMBER:08 DATE: 30/10/2011 TIME: 07:55 COORDINATES: S / E SITE NAME. Majete National Park BIOME CATEGORY: Riverine Woodlands (escarpments) DESCRIPTION: This site is located within Majete National Park just downstream of the Kapichira Hydropower Plant. It is open mixed woodland on an escarpment, one of the few escarpment sites sampled due to lack of accessibility. SPECIES LIST: Sansevieria burdettii Maerua angolensis DC Kirkia acuminata Oliv. Justicia striata (Klotz.) Bullock Asparagus africana Commiphora sp. Dalbergia arbutifolia Bak. Cissus quadrangularis L. Bombax rhodognaphalon K.Schum. Pleurostylia africana Acacia nilotica (L.) Delile Sterculia appendiculata K.Schum. Capparaceae Simaroubaceae Acanthaceae Asparagaceae Bombacacae Mimosoideae Sterculiaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

131 Hibscus debeertsii Wild. & Duv. Acacia nigrescens Oliv. Combretum microphyllum Klotz. Ximenia americanum L. var. microphylla Oliv. Urochloa mossambicensis (Hack.) Dandy Ormocarpum kirkii S.Moore Malvaceae Mimosoideae Combretaceae Poaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

132 SITE NUMBER: 09 DATE: 30/10/2011 TIME: 08:34 COORDINATES: S / E SITE NAME. Majete National Park BIOME CATEGORY: Riverine Woodlands DESCRIPTION: This site is within Majete National Park, located just upstream of a hydropower plant. It is a deciduous riparian woodland. Bank erosion is evident. Wildlife is abundant and Vervet monkeys, baboon, elephants and hippos were observed during the assessment. A number of sandy banks exist in this reach of the Shire which may be crocodile nesting areas (see photo bottom right). SPECIES LIST: Cleistochlamys kirkii (Benth.). Oliv. Xylopia parviflora (A. Rich) Benth. Lantana camara L. Acacia nigrescens Olive Phyllanthus reticulata Poir. Jasminum fluminense Vell Sterculia appindiculata K. Schum Jatropha sp. Croton macrostachys Hochst. Ex. A. Rich Tamarindus indica L. Kiggelaria africana L. Annonaceae Annonaceae Verbanaceae Mimosaceae Euphorbiaceae Oleaceae Sterculiaceae Euphorbiaceae Euphorbiaceae Caesalpinoideae Flacourtiaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

133 Grewia flavescens Juss. Dovyalis macrocalyx (Oliv.) Warb. Zanha golungensis Hiern Cordylia africana Lour. Kigelia africana (Lam.) Benth. Urena lobata L. Sida acuta L. Ageratum conyzoides L. subsp houstonium (Mill.) Sahu Bosia salicifolia Oliv Philenoptera bussei (Harm) Sschrire Ormocarpum kirkii S. Moore Holarrhena pubescens (Buch. Ham.) Wall Tiliaceae Flacourtiaceae Sapindaceae Caesalpinoideae Bignoniaceae Malvaceae Malvaceae Asteraceae Capparaceae Papilionoideae Papilionoideae Apocycenaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

134 SITE NUMBER: 10 DATE: 30/10/2011 TIME: 09:08 COORDINATES: S / E SITE NAME. Hamilton Falls BIOME CATEGORY: Riverine Woodlands DESCRIPTION: Site is within Majete National Park. It is best described as deciduous open woodland, which grows immediately adjacent to the river bank forming a true riparian woodland gallery. Elephants were spotted not far from the site and the Majete Wildlife Ranger who accompanied the assessment team indicated that crocodiles were very common in the area. SPECIES LIST: Tamarindus indica L. Caesalpinoideae Kigelia africana Bignoniaceae Philenoptera violacea (Klotze) Schrire Papilionoideae Maytenus senegalensis (Lam.) Exell. Celastraceae Trichilia emetica L. Meliaceae Taberamontana elegans Apocycenaceae Jasminum fluminense Vell Oleaceae Phragmites mauriana Kunth Poaceae Azanza garckena (F. Hoffim.) Exell & Hilcoat Malvaceae Lantana camara L. Verbanaceae Diospyros senensis Ebenaceae Schoenoplectus corymbosus (Roem el Schutt) J. Reyn Cyperaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

135 Grewia sp. Urena lobata L. Croton macrostachys Hochst. Ex. A. Rich Ficus sp. Cynodon dactylon Hyphane sp. Tiliaceae Malvaceae Euphorbiaceae Moraceae Poaceae Palmae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

136 SITE NUMBER: 11 DATE: 30/10/2011 TIME: 09:42 COORDINATES: S / E BIOME CATEGORY: Riverine Woodlands SITE NAME. Confluence of Mruloumadzi & Shire Rivers DESCRIPTION: Site is within Majete National Park. It is best described as deciduous open woodland, which grows immediately adjacent to the river bank forming a true riparian woodland gallery. Elephants were spotted not far from the site and the Majete Wildlife Ranger who accompanied the assessment team indicated that crocodiles were very common in the area. SPECIES LIST: Acacia nigrescens Olive Piliostigma thonningii (Schumach.) Milne Redh. Taberamontana elegans Trichilia emetica L. Cleistochlamys kirkii (Benth.). Oliv. Tamarindus indica L. Philenoptera violacea (Klotze) Schrire Capparis rosea Sterculia appendiculata K. Schum. Breonardia salicifolia Mimosoidea Caesalpinoideae Apocycenaceae Meliaceae Annonaceae Caesalpinoideae Papilionoideae Capparaceae Sterculiaceae Rubiaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

137 Antidesma venosum Dalbergia sp. Zanha golungensis Hiern Phragmites mauriana Kunth Acacia nigrescens Olive Cynodon dactylon (L.) Pers. Schoenoplectus corymbosus (Roem el Schutt) J. Reyn Maytenus senegalensis (Lam.) Exell. Lantana camara L. Phyllanthus reticulata Poir. Albizia versicolor Euphorbiaceae Papilionoideae Sapindaceae Poaceae Mimosoidea Poaceae Cyperaceae Celastraceae Verbanaceae Euphorbiaceae Mimosoidea Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

138 SITE NUMBER: 14 DATE: 31/10/2011 TIME: 10:15 COORDINATES: S / E SITE NAME. Sifakado Village BIOME CATEGORY: Perennial Marsh DESCRIPTION: Typical Perennial marsh land. Typher spp., Voscia spp. and Phragmites spp. dominate. SPECIES LIST: Typha domingensis Pers. Phragmites mauritianus Kunth Vossia cuspidata (Roxb.) Griff. Faidherbia alba (Del.) A. Chev. Acacia xanthophloea Benth. Cyperus papyrus L. Zea mays L. Musa paradisica Aeschynomene elephraxylon (Guill & Perr) Taub. Ipomea aquatica Forssk. Nidorella auriculata DC Argemone mexicana L. Typhaceae Poaceae Poaceae Mimosoidea Mimosoidea Cyperaceae Poaceae Musaceae Papilionoideae Convolvulaceae Asteraceae Papaveraceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

139 SITE NUMBER: 15 DATE: 31/10/2011 TIME: 10:50 COORDINATES: S / E SITE NAME. Citinounugwa Village BIOME CATEGORY: Perennial Marsh DESCRIPTION: Perennial marsh, slightly disturbed by grazing and domba agriculture. Typical perennial swampy species represented such as Papyrus spp. Phragmites spp. and Typher. SPECIES LIST: Eleusine indica L. Poaceae Phragmites mauriana Kunth Poaceae Echinochloa pyramidalis (Lam.) Hitchc. & Chase Poaceae Sida acuta L. Malvaceae Persicaria attenuata (R. Br) Sojak subsp africana Polygonaceae Argemona mexicana Papaveraceae Cyperus papyrus L. Cyperaceae Cajanus cajana (L.) Millsp. Papilionoideae Ricimus communus L. Euphorbiaceae Typha domingensis Pers. Typhaceae Vossia cuspidata (Roxb.) Griff. Poaceae Aeschynomene elephraxylon (Guill & Perr) Taub. Papilionoideae Launnea cornuta (Hochst. ex Oliv. & Hiern) C. Jeffrey Asteraceae Cynodon dactylon (L.) Pers. Poaceae Ficus sycamorus L. Moraceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

140 SITE NUMBER: 16 DATE: 31/10/2011 TIME: 11:22 COORDINATES: S / E SITE NAME. Nataka Village BIOME CATEGORY: Perennial Marsh DESCRIPTION: Perennial marsh land, flanked by some seasonally wet grassland which is being grazed by cattle. The influence of the river is evident some 200+ m from the waters edge as the land is very boggy (photo lower left). Typical perennial marsh species are common including Phragmites spp. and Typher spp. SPECIES LIST: Vossia cuspidata (Roxb.) Griff. Phragmites mauritianus Kunth Typha domingensis Pers. Sesbania sesban L. Cyperus papyrus L. Ludwigia erecta (L.) Hara Cyperus alopecuroides Rottb. Azolla nilotica Mett. Cyperus sp. Echinochloa pyramidalis (Lam.) Hitchc. & Chase Poaceae Poaceae Typhaceae Papilionoideae Cyperaceae Onagraceae Cyperaceae Azollaceae Cyperaceae Poaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

141 SITE NUMBER: 17 DATE: 31/10/2011 TIME: 14:21 COORDINATES: S / E SITE NAME. Liwonde National Park (east bank) BIOME CATEGORY: Seasonal Floodplain DESCRIPTION: This site is comprised of seasonally wet grasslands on an alluvial flood plain. The flood plain is approximately 150 m wide and will be completely flooded during the wet season (late November late March). This is an important grazing area for many animal species in Liwonde National Park; water buck and baboons were observed at the site during the assessment. SPECIES LIST: Kiggelaria africana (Lam) Benth Phragmites mauritianus Kunth Vossia cuspidata (Roxb.) Griff. Mimosa pigra L. Hyphane sp. Aeschynomene elephraxylon (Guill & Perr) Taub. Eichhornia crassipes (Mart.) Solms Laub Persicaria attenuata (R. Br) Sojak subsp africana Bignoniaceae Poaceae Poaceae Mimosoidea Palmae Papilionoideae Pontederiaceae Polygonaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

142 SITE NUMBER: 18 DATE: 01/11/2011 TIME: 07:00 COORDINATES: S / E SITE NAME. Liwonde National Park (east bank) BIOME CATEGORY: Seasonal Floodplain DESCRIPTION: This is a seasonally wet grassland on an extensive flood plain several hundreds of metres wide. Foraging wildlife is abundant on the site with elephant, hippo, warthog, water buck and impala spotted. Termite mounds common. SPECIES LIST: Cryptolepis oblongifolia (Meisn) Schltr. Maerua angolensis Echinochloa pyramidalis (Lam.) Hitchc. & Chase Phragamite mauritianus Kunth Eichhornia crassipes (Mart.) Solms Laub Taberamontana elegans Philenoptera violacea (Klotze) Schrire Hyphane sp. Acacia xanthophloea Benth. Cissampelose micronata A. Rich Caccia abbreviata Oliv. Apocycenaceae Capparaceae Poaceae Poaceae Pontederiaceae Apocycenaceae Papilionoideae Palmae Mimosoidea Menispermaceae Caesalpinoideae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

143 Schoenoplectus corymbosus (Roem el Schutt) J. Reyn Trichlia emetica Vahl. Flacourtia indica (Burm.f.) Merr. Senna siamea (Lam) Irwin & Bainesby Antidesma venosum E. Mey. ex Tul. Ziziphus abyssinica Hochst ex A. Rich Cyperaceae Meliaceae Flacourtiaceae Caesalpinoideae Euphorbiaceae Rhamnaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

144 SITE NUMBER: 19 DATE: 01/11/2011 TIME: 08:02 COORDINATES: S / E SITE NAME. Liwonde National Park (east bank) BIOME CATEGORY: Seasonal Floodplain DESCRIPTION: This is a seasonal flood plain with mixed Mopanie woodland to the rear; a perennial wetland fringe at the lake margin is absent. Baboons were seen foraging on the grasslands. SPECIES LIST: Ocimum gratissimum L. Subsp. Gratissimum var gratissimum Triumfetta pilosa Roth Philenoptera violacea (Klotze) Schrire Echinochloa pyramidalis (Lam.) Hitchc. & Chase Combretum imberbe Wawra Ziziphus abyssinica Hochst ex A. Rich Pychnostachys urticifolia Hook Cryptolepis oblongifolia (Meisn) Schltr. Sesbania sesban L. Phragmites mauritianus Kunth Eichhornia crassipes (Mart.) Solms Laub Acacia xanthophloea Benth. Lamiaceae Tiliaceae Papilionoideae Poaceae Combretaceae Rhamnaceae Lamiaceae Apocycenaceae Papilionoideae Poaceae Pontederiaceae Mimosoidea Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

145 SITE NUMBER: 20 DATE: 01/11/2011 TIME: 08:25 COORDINATES: S / E SITE NAME. NA BIOME CATEGORY: Riverine Woodlands (fertile Areas) DESCRIPTION: Open Mopane woodland. Tree species grow close to water line, and seasonally wet grassland strip is narrow. This is due to the increased slope[?] in this area and the reduced area of inundation during the wet season (late November late March). The Perennial marsh at the margins of the lake are absent in this area. SPECIES LIST: Euphorbia ingens Boiss. Zanha africana (Radlk.) Exell Combretum imberbe Wawra Albizia amara (Roxb.) Boivin subsp. Colophospermum mopane (Benth.) J. Leonard Maerua angolensis DC. Acacia nilotica (L.) Delile Acacia xanthophloea Benth. Echinochloa pyramidalis (Lam.) Hitchc. & Chase Euphorbiaceae Sapindaceae Combretaceae Mimosoidea Caesalpinoideae Capparaceae Mimosoidea Mimosoidea Poaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

146 SITE NUMBER: 21 DATE: 01/11/2011 TIME: 08:54 COORDINATES: S / E SITE NAME. Mouth of Lake Malombe BIOME CATEGORY: Seasonal Floodplain DESCRIPTION: Mixed woodlands at approximately 50 m from waters edge; there is then a narrow (~50m) seasonally wet grassland on the flood plain merging in to a thin band of perennial marsh land at the lake edge. SPECIES LIST: Schoenoplectus corymbosus (Roem el Schutt) J. Reyn Vossia cuspidata (Roxb.) Griff. Echinocloa pyramidalis(lam.) Hitchc. & Chase Acacia xanthophloea Benth. Cordylia africana Lour. Faidherbia alba (Del.) A. Chev. Diospyros senensis Klotzch Scrocarya birrea (A. Rich.) Hochst. Subsp. Caffra (Sond.) Kokwaro Andropogon gayanus Kunth Philenoptera violacea (Klotze) Schrire Acacia nilotica (L.) Delile Kiggelaria africana (Lam) Benth Cyperaceae Poaceae Poaceae Mimosoidea Celastraceae Mimosoidea Ebenaceae Anarcadiaecea Poaceae Papilionoideae Mimosoidea Bignoniaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

147 Maerua angolensis DC. Euphorbia ingens Boiss. Markhamia zanzibarica (DC) K. Schum. Maytenus senegalensis (Lam.) Exell. Capparaceae Euphorbiaceae Bignoniaceae Celastraceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

148 SITE NUMBER: 22 DATE: 01/11/2011 TIME: 11:37 COORDINATES: S / E SITE NAME. Liwonde National Park (west bank) BIOME CATEGORY: Seasonal Floodplain DESCRIPTION: This site is located in the 1 km wide strip of land within Liwonde National Park located on the west bank of the Shire River. This site is comprised of mixed open woodland in the elevated areas, and seasonally wet grasslands in the lower areas that become flooded during the wet season (late November late March). Perennial marsh land species are found in a narrow band only at the lake margins. The Liwonde National park Ranger who accompanied the assessment team indicated that the few sandy patches at this site were potentially good crocodile nesting sites (see photo lower right). SPECIES LIST: Acacia xanthophloea Benth. Adansonia digitata L. Ziziphus abyssinica Hochst ex A. Rich Phragmites mauritianus Kunth Maerua angolensis DC. Diospyros senensis Klotzch Caccia abbreviata Oliv. Rhus natalensis Bernh. ex Krauss. Eichhornia crassipes (Mart.) Solms Laub Mimosoidea Bombacaceae Rhamnaceae Poaceae Capparaceae Ebenaceae Caesalpinoideae Anarcadiaecea Pontederiaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

149 Echinochloa pyramidalis (Lam.) Hitchc. & Chase Cyperus alopecuroides Rottb. Aeschynomene elephraxylon (Guill & Perr) Taub. Poaceae Cyperaceae Papilionoideae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

150 SITE NUMBER: 23 DATE: 01/11/2011 TIME: 14:36 COORDINATES: S / E SITE NAME. Lake Malombe BIOME CATEGORY: Perennial Marsh DESCRIPTION: Perennial marshes at lake margins with seasonally wet grasslands behind. Some of the seasonally wet grasslands have been cultivated for dombas. Locals report that lake level is very low at the time of the survey and that during the wet season (late November to late March) the levels will rise 20 m above that at the time of the assessment. SPECIES LIST: Phragmites mauritianus Kunth Cyperas papyrus L. Melanthera scandens (Schumach. & Thonn.) Roberty Ipomea rubens Choisy Aeschynomene elaphraxylon (Guill & Perr) Taub. Musa paradisica L. Faidherbia alba (Del.) A. Chev. Persicaria attenuata (R. Br) Sojak subsp africana Schoenoplectus corymbosus (Roem el Schutt) J. Reyn Typha domingensis Pers. Vossia cuspidata (Roxb.) Griff. Poaceae Cyperaceae Asteraceae Convolvulaceae Papilionoideae Musaceae Mimosoidea Polygonaceae Cyperaceae Typhaceae Poaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

151 Aleurites moluccana (L.) Willd Moringa ovalifolia Dinter & A. Berger Carica papaya Euphorbiaceae Moringaceae Caricaeae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

152 SITE NUMBER: 24 DATE: 01/11/2011 TIME: 15:17 COORDINATES: S / E SITE NAME. NA BIOME CATEGORY: Perennial Marsh DESCRIPTION: Lake margins are heavily vegetated with a thin band of perennial marsh species (e.g. Phragmites spp.); this strip is m wide. The Seasonally wet grasslands behind the perennial marsh have been heavily cultivated and a row of mango trees planted. SPECIES LIST: Phragmite mauritianus Kunth Poaceae Senna siamea (Lam) Irwin & Bainesby Caesalpinoideae Ziziphus abyssinica Hochst ex A. Rich Rhamnaceae Ricimus communus L. Euphorbiaceae Calotropes procera L. Apocycenaceae Mangifera indica L Anarcadiaecea Psidium guajava L. Myrtaceae Schoenoplectus corymbosus (Roem el Schutt) J. Reyn Cyperaceae Faidherbia alba (Del.) A. Chev. Mimosoidea Ipomea rubens Choisy Convolvulaceae Sesbania sesban L. Papilionoideae Senna occidentalis (L.) Link Caesalpinoideae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

153 SITE NUMBER: 25 DATE: 02/11/2011 TIME: 07:14 COORDINATES: S / E SITE NAME. Kadewere (E. bank of Lake Malombe) BIOME CATEGORY: Seasonal Marshes - Under Cultivation (dambos) DESCRIPTION: The vegetation fringing the lake universally consists of a thin band of aquatic vegetation at the waters edge. Behind this is seasonally wet grassland which has been cultivated with numerous dambo gardens. Crops observed include: sugar cane; cassava; maize; tomatoes and pumpkin. Locals report that during the wet season (late November to late March) water levels will rise approximately 50 m from the levels at the time of the assessment. SPECIES LIST: Senna occidentalis (L.) Link Melanthera scandens (Schumach. & Thonn.) Roberty Vossia cuspidata (Roxb.) Griff. Phragmite mauritianus Kunth Musa paradisiaca L. Zea mays L. Sesbania sesban L. Luffa cylindlica (L.) Saccharum officinarum L. Caesalpinoideae Asteraceae Poaceae Poaceae Musaceae Poaceae Papilionoideae Cucurbitaceae Poaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

154 Euphorbia Esculenta Senna siamea (Lam) Irwin & Bainesby Ricimus communus L. Persicaria attenuata (R. Br) Sojak subsp africana Persicaria limbata (Meisn.) Hara Ipomea rubens Choisy Cyperus hemsphaevicus Boeck Xanthium sp. Calotropes procera L. Solanum panduriforme E. Mey Sida acuta L. Aeschynomene cristata Vatke Indigofera antunesiana Harms Caesalpinoideae Euphorbiaceae Euphorbiaceae Polygonaceae Polygonaceae Convolvulaceae Cyperaceae Asteraceae Apocycenaceae Solanaceae Malvaceae Papilionoideae Papilionoideae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

155 SITE NUMBER: 26 DATE: 02/11/2011 TIME: 07:50 COORDINATES: S / E SITE NAME. Kwisimba (E. bank of Lake Malombe) BIOME CATEGORY: Seasonal Marshes - Under Cultivation (dambos) DESCRIPTION: Fishing village on the east bank of Lake Malombe. At the edges of the beach permanent marsh land fringes the lake, behind which seasonally wet grasslands have been developed as dambos. Residents indicate that during the wet season (late November late March) the water level is m higher than at the time of the assessment (early November). SPECIES LIST: Faidherbia alba (Del.) A. Chev. Rhynchosia sp. Vossia cuspidata (Roxb.) Griff. Phragmite mauritianus Kunth Sesbania sesban L. Ipomea rubens Choisy Nidorella auriculata DC Sida acuta L. Desmodium tortuosum (SW) DC Euphorbia heterophylla L. Saccharum officinarum L. Mimosoidea Papilionoideae Poaceae Poaceae Papilionoideae Convolvulaceae Asteraceae Malvaceae Papilionoideae Euphorbiaceae Poaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

156 Ricimus communus L. Euphorbia esculenta ( Casava) Cissampelos micronata A. Rich Boerhaavia erecta L. Euphorbiaceae Euphorbiaceae Menispermaceae Nyctaginaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

157 SITE NUMBER: 27 DATE: 02/11/2011 TIME: 09:23 COORDINATES: S / E SITE NAME. Eastern Arm of Lake Malawi BIOME CATEGORY: Seasonal Marshes - Under Cultivation (dambos) DESCRIPTION: This is a fishing village. The vegetation consists of Perennial wetlands at the lake margins with Seasonal wetlands and associated dambos behind. Much of the Typha areas have been cut and used for grazing or agriculture. SPECIES LIST: Phragmites mauritianus Kunth Ziziphus abyssinica Hochst ex A. Rich Faidherbia alba (Del.) A. Chev. Mangifera indica L Sesbania sesban L. Mimosa pigra L. Echinochloa pyramadis (Lam.) Hitchc. & Chase Aeschynomene elaphraxylon (Guill & Perr) Taub. Euphorbia glaziovil Muell. Arg Melia azedarach L. Hibiscus diversifolia Jacq. Poaceae Lamiaceae Mimosoidea Anarcadiaecea Papilionoideae Mimosoidea Poaceae Papilionoideae Euphorbiaceae Meliaceae Malvaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

158 Vossia cuspidata (Roxb.) Griff. Eichhornia crassepes Senna siamea (Lam) Irwin & Bainesby Aleurites moluccana (L.) Willd Ipomoea rubens choisy Crotalaria sp. Poaceae Pontederiaceae Caesalpinoideae Euphorbiaceae Convolvulaceae Papilionoideae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

159 SITE NUMBER: 28 DATE: 03/11/2011 TIME: 10:22 COORDINATES: S / E SITE NAME. NA BIOME CATEGORY: Cleared Woodlands DESCRIPTION: This site is highly modified open woodland, with mature trees growing down to the top of a sandy beach. The area is heavily populated and many of the trees have been felled or coppiced. The slope of the land is relatively steep compared to the flood plain areas, and flooding is minimal, allowing establishment of mature woodland tree species close to the water line. SPECIES LIST: Albizia labbeck (L.) Benth Senna siamea (Lam) Irwin & Bainesby Adansonia digitata L. Moringa ovalifolia Dinter & A. Berger Ziziphus abyssinica Hochst ex A. Rich Catharanthus roea Don. Ficus sp. Delonix regia (Boj. Ex Hook.f. Eichhornia crassipes Typha domingensis Pers. Schoenoplectus corymbosus (Roem el Schutt) J. Reyn Mimosoidea Caesalpinoideae Bombacaceae Moringaceae Rhamnaceae Apocycenaceae Moraceae Caesalpinoideae Pontederiaceae Typhaceae Cyperaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

160 Aeschynomene elaphraxylon (Guill & Perr) Taub. Sesbania sesban L. Ipomoea rubens choisy Combretum imberbe Wawra Senna cristata Vatke Tephrosia sp. Papilionoideae Papilionoideae Convolvulaceae Combretaceae Caesalpinoideae Papilionoideae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

161 SITE NUMBER: 29 DATE: 02/11/2011 TIME: 11:53 COORDINATES: S / E SITE NAME. Lake Malawi Outlet to Shire (west) BIOME CATEGORY: Perennial Marsh DESCRIPTION: Perennial marsh areas at interface with the lake, with seasonal grassland areas behind and associated dambos. There is a large irrigation pumping station at this site which draws water from the Shire River to irrigate crops several kilometres inland. SPECIES LIST: Ziziphus abyssinica Hochst ex A. Rich Faidherbia alba (Del.) A. Chev. Typha domingensis Pers. Sesbania sesban L. Echinochloa pyramadis (Lam.) Hitchc. & Chase Ipomea rubens Choisy Vossia cuspidata (Roxb.) Griff. Tridax procumbens L. Wahheria indica L. Combretum mossambinsis (Klotzch) Engl. Cissampelos micronata A. Rich Rhamnaceae Mimosoidea Typhaceae Papilionoideae Poaceae Convolvulaceae Poaceae Asteraceae Sterculiaceae Combretaceae Menispermaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

162 Azadirachta indica A. Juss Musa paradisiaca L. Hyphane sp. Acacia xanthophloea Benth. Aeschynomene cristata Vatke Nidorella auriculata DC Persicaria attenuata (R. Br) Sojak subsp africana Commelina benghalensis L. Eichhornia crassipes (Mart.) Solms Laub Phragmites mauritianus Kunth Manhot glaziovii Muell. Arg Ocimum gratissimum L. Subsp. gratissimum var gratissimum Crotalaria pallida Aiton Meliaceae Musaceae Palmae Mimosoidea Papilionoideae Asteraceae Polygonaceae Commelinaceae Pontederiaceae Poaceae Euphorbiaceae Lamiaceae Papilionoideae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

163 SITE NUMBER: 30 DATE: 02/11/2011 TIME: 12:49 COORDINATES: S / E SITE NAME. Sunbird Resort BIOME CATEGORY: Dry Bush Savannah DESCRIPTION: Rocky outcrops besides the beach support open canopy woodlands. Behind the beach area is Sunbird Resort with manicured gardens and planted trees. SPECIES LIST: Sesbania sesban L. (Merr.) Trichilia emetica Vahl. subsp. emetica Ficus natalensis Hochst. Ficus sp. Heliotropium indicum L. Eichhormia crassipes (Mart.) Solms Laub Acacia xanthophloea Benth. Hyphane sp. Faidherbia alba (Del.) A. Chev. Boerhaavia erecta L. Poaceae Meliaceae Moraceae Moraceae Boraginaceae Pontederiaceae Mimosoidea Palmae Mimosoidea Nyctaginaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

164 SITE NUMBER: 31 DATE: 02/11/2011 TIME: 14:33 COORDINATES: S / E SITE NAME. Namasso Bay BIOME CATEGORY: Dry Bush Savannah (on adjacent rocks) DESCRIPTION: This is Combretum dominated woodland on the rocky outcrops besides the sandy beach. Due to the steep grade of the rocky outcrops the woodland vegetation extends down to the water edge. Behind the beach are manicured gardens of a residential compound. SPECIES LIST: Hyphane sp. Ficus bubu Terminalia catapa Sterculia africana (Lour.0 Fiori Phyllanthus reticulata Poir. Cissus cucumerifolia Planch. Ziziphus mucronata Willd. Cordylia africana Lour. Dalbergia arbutifolia Baker Cadaba kirkii Oliv. Albizia amara (Roxb.) Boivin subsp. Ximenia americana L. Palmae Moraceae Combretaceae Sterculiaceae Euphorbiaceae Vitaceae Rhamnaceae Caesalpinoideae Papilionoideae Capparaceae Mimosoidea Olaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

165 SITE NUMBER: 32 DATE: 03/11/2011 TIME: 10:12 COORDINATES: S / E SITE NAME. Tiainankumba BIOME CATEGORY: Seasonal Floodplain DESCRIPTION: Seasonally wet grassland with permanent swamp at the interface with the lake. Currently being used for pasture land for cattle, this area will likely flood during the wet season. Ground is perennially wet at the grassland/swamp interface. SPECIES LIST: Hyphane sp. Vossia cuspidata (Roxb.) Griff. Cyperus alopecuroides Rottb. Schoenoplectus corymbosus (Roem el Schutt) J. Reyn Typha domingenonsis Pers. Phragmites mauritianus Kunth Cyperus papyrus L. Aeschynome cristaty Vitke Faidherbia alba (Del.) A. Chev. Mangifera indica L Hibiscus diversifolia Jacq. Palmae Poaceae Cyperaceae Cyperaceae Typhaceae Poaceae Cyperaceae Papilionoideae Mimosoidea Anarcadiaecea Malvaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

166 Eichhornia crassipes (Mart.) solms Laub Ipomoea rubens choisy Blepharis acanthodianss Klotzch Pontederiaceae Convolvulaceae Acanthaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

167 SITE NUMBER: 33 DATE: 03/11/2011 TIME: 06:58 COORDINATES: S / E SITE NAME. NA BIOME CATEGORY: Seasonal Floodplain DESCRIPTION: This site is comprised of a sand berm immediately behind a sandy beach. Behind the berm is a depression which supports seasonally wet grasslands, which is too wet to support the woodland vegetation that proliferates at the margins of the site and on the berm itself. The grasslands were being used for grazing cattle at the time of the assessment but would be flooded during the wet season. SPECIES LIST: Acacia xanthophloea Benth. Faidherbia alba (Del.) A. Chev. Phragmites mauritianus Kunth Senna alata L. Urena lobata L. Hyphaena sp. Ziziphus mucronata Willd. Vossia cuspidata (Roxb.) Griff. Cyperus alopecuroides Rottb. Mimosoidea Mimosoidea Poaceae Papilionoideae Malvaceae Palmae Rhamnaceae Poaceae Cyperaceae Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Vegetation, December

168 Government of the Republic of Malawi Ministry of Water Development and Irrigation National Water Development Programme Independent Environmental Impact Assessment for the Upgraded Kamuzu Barrage Final Environmental and Social Impact Assessment Volume 2: Water Quality December 2013

169 Independent Environmental Impact Assessment for the Upgraded Kamuzu Barrage Final ESIA Volume 2: Water Quality Table of Contents... i Abbreviations / Glossary... iv Acknowledgements... v Executive Summary... vi 1 Background Water Quality Literature review Lake Malawi Water Quality Water Quality Impacts of Barrages Objective of the Study Overall Objectives Specific Objective Materials and Methods Study Area Sampling Points and Design Sampling Point S Sampling Point S Sampling Point S Sampling Point S Sampling Point S Collection and Handling of Water Samples Important Water Quality Parameters Dissolved Oxygen (DO) ph Value Phosphorous Turbidity Water Temperature Nitrate (NO 3- ) Biochemical Oxygen Demand Total Dissolved Solids (TDS) Electrical Conductivity (EC) Chlorophyll a Sulphate Hardness Faecal Coliform Bacteria Water Quality Parameters Determined on Site Water Quality Parameters Determined in the Laboratory... 9 Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec 2013 i

170 2.6.1 Nitrates Soluble Reactive Phosphorous (SRP) Sulphate Biochemical Oxygen Demand (BOD) Faecal Coliform Bacteria (FCB) Chlorophyll a Total Suspended Solids (TSS) Transparency Laboratory Facilities Results and Discussion Selected Water Quality Parameters Turbidity Chlorophyll a Phosphorus Faecal Coliform Bacteria Phosphorous, Nitrogen and Aquatic Weeds Comparison of the Results with Previous Studies Possible Water Quality Impacts due to Heightened Barrage Impacts during Construction Phase Increased Siltation and Turbidity of Water Downstream Increased Levels of Water Upstream Impacts during the Operation Phase Nutrient Enrichment due to Increased Water Retention Potential Contamination by Wastewater from Resorts, Lodges, Markets and Fishing Activities around the Barrage Proposed Mitigation Measures Mitigation Measures during Construction Phase Water Pollution Monitoring Mitigation Measures during Operation Phase Catchment Management Proper Waste Water Management Practices Public Awareness / Stakeholder Involvement Sanitary Measures Water Monitoring System Conclusion References Appendix 1: Standard Guidelines for Flowing Water Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec 2013 ii

171 Appendix 2: Raw Data for the Dry Season Appendix 3: Raw Data for the Wet Season Appendix 4: Pictures of the Shire River Catchment Appendix 5: Persons and Organisations Consulted Table of Figures Figure 1. Map of Malawi showing Shire River and the sampling site... 4 Figure 2. Part of map of Malawi showing Lake Malombe and Kamuzu Barrage and water sampling points along the Shire River Figure 3. Trends of selected parameters in dry and wet seasons at different sampling points Figure 4. Mean concentrations of selected parameters at various sampling points Figure 5. Shire River catchment area at the Kamuzu Barrage, used as a dumping site Figure 6. A researcher getting set for sampling Figure 7. Water sampling and on-site measurements at Sampling Point 4 at the Kamuzu Barrage Figure 8. The Shire River as a water source for domestic purposes Figure 9. Fishing along the Shire River - another source of water pollution Table of Tables Table 1. Mean readings of selected water quality parameters for the dry season Table 2. Mean readings of selected water quality parameters for the wet season Table 3. Mean concentration of selected water quality parameters for the Shire River ( ).. 13 Table 4. Standard guidelines for water quality from UNECE Table 5. Raw data for the dry season Table 6. Raw data for the wet season Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec 2013 iii

172 Abbreviations BOD Biochemical Oxygen Demand C Degrees Celsius C Carbon Ca Calcium CFU Coliform Forming Units Cl Chlorine DO Dissolved Oxygen EC Electrical Conductivity GoM Government of Malawi GPS Global Positioning System FCB Faecal Coliform Bacteria HCl Hydrochloric Acid K Potassium L litre Mg Magnesium N Nitrogen Na Sodium O Oxygen nm nano metres NO 3- NTU P PCA ph PO 4- ppm S μs SMEC SRP TDS TN TP TSS UNECE WHO Nitrate Nephelometric Turbidity Units Phosphorus Principal Component Analysis Measure of acidity Phosphate Parts per Million Sulphur Micro Siemens Snowy Mountains Engineering Corporation Soluble Reactive Phosphate Total Dissolved Solids Total Nitrogen Total Phosphorous Total Suspended Solids United Nations Economic Commission for Europe World Health Organization Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec 2013 iv

173 Glossary BARRAGE EUTROPHIC FAECAL COLIFORM BACTERIA MACROPHYTES MESOTROPHIC MONITORING OLIGOTROPHIC ph PLANKTON SURFACE RUNOFF TDS TSS TURBIDITY ZOOPLANKTON A Special type of dam which consists of a line of large gates that can be opened or closed to regulate the amount of water flowing downstream. The gates are set between flanking piers which are responsible for supporting the water load. They are often used to control and stabilize water flow for various purposes. Water with high primary production due to excessive nutrient enrichment, the water is not clear and has a lot of phytoplankton. A sub-group of total coliform bacteria. They occur in great quantities in the intestines and faeces of people and animals. The presence of faecal coliform bacteria in a drinking water sample often indicates recent faecal contamination, meaning that there is a greater risk that pathogens are present in that water. Are macroscopic aquatic plant that grows in or near water and is emergent, submerged or floating. Water with intermediate level of primary production, greater than oligotrophic but less than eutrophic. Long-term, standardised measurement and observation of the aquatic environment and water quality in order to define status and trends. Water with low primary production due to low nutrient content, often more clear with high drinking quality. The balance of positive hydrogen ions (H+) and negative hydroxide ions (OH-) in water and is a determination of how acidic or basic the water is. Are microscopic drifting organisms (animals, plants, archaea, or bacteria) that inhabit the pelagic zone of a water body. Is the water flow that occurs when the soil is infiltrated to full capacity and excess water from rain, melt water, or other sources flows over the land. Is the total amount of dissolved materials/solids (principally calcium, magnesium, potassium, sodium, bicarbonates, chlorides and sulphates) which become salts when evaporation takes place. Total Suspended Solids (TSS) includes all particles suspended in water which will not pass through a filter. In simple terms, turbidity is a measure of how clear the water is. Are microscopic invertebrate animals that swim or drift in water. They are at the base of the food chain, feeding on microscopic plants and being fed upon by aquatic insects, fish. 23:21:0:+4S Inorganic fertiliser recommended for basal dressing, contains 23% Nitrogen, 21% Phosphorous, 0% Potassium and 4% Sulphur. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec 2013 v

174 Acknowledgements This Report on Consultancy work was prepared by Austin Mtethiwa, Wilson Jere and Elton Nyali. The consultancy team sincerely thank SMEC for instituting and funding this study. The team also sincerely extend its gratitude to Michael Holics and Stayford Phiri of SMEC for their technical guidance. Appreciation is also extended to Bunda College for the equipment and technical expertise provided in the sampling and analysis of water samples. Many thanks to Staff of Liwonde National Park (for provision of boat services, information and space), Fishermen, Community of Businessmen at Liwonde (Kamuzu Barrage) and all those that provided support to this study. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec 2013 vi

175 Executive Summary A water quality study was carried out along the middle Shire River in Malawi in November, 2011 and January, The study was part of the on-going environmental study by Snowy Mountains Engineering Corporation (SMEC) on the Shire River catchment. The environmental study is aimed at determining the environmental impacts of rehabilitating the Kamuzu Barrage which will raise the maximum regulated water level up to 40 centimetres. The aim of this water quality study was to collect baseline information on the Shire River water quality within the Kamuzu Barrage catchment as may be affected by the Barrage rehabilitation. A reconnaissance survey was conducted where five water sampling points were set up followed by actual collection of water samples and onsite water quality measurements. Five sampling points were selected stretching from the outlet of Lake Malombe to an area downstream of the Kamuzu Barrage at Liwonde. The site was selected in consideration that this is the area that will be affected once the work on the Barrage has been accomplished. Temperature, ph, TDS, turbidity and Secchi depth among others, were determined on site. Water samples collected were immediately stored in a cooler box to maintain low temperature and later transferred to Bunda College Laboratory for analysis of various parameters including faecal coliform, BOD, nitrates and phosphorus among others. Results showed that all five sampling points had faecal pollution (118±3.2 CFU/100 ml, mean) but did not have high phosphorus (20.6±0.18 µg/l, mean) and nitrate levels (0.16±0.04 mg/l, mean). However, the rainy season values were higher than those of the dry season, showing the impact of rain and runoff from the catchment to the river system. Also, those points that were closest to the Kamuzu Barrage and at the outlet of Lake Malombe had high concentrations of coliform counts (124±2.2CFU/100 ml, mean) and those sampling points within the stream catchment area and resorts and lodges gave higher concentrations of phosphorus (24±2.6 µg/l, mean) and nitrates (0.22±0.02 mg/l, mean) than the rest, probably due to the nature of pollution emanating from the lodges and resorts. Whilst rehabilitation of the Barrage may bring about low TSS, nitrate nitrogen and phosphorus levels may exceed the recommended limits. The impact of these high level concentrations is that they contribute to overgrowth of aquatic plants and algae. These mats of floating plants can impede fishing boats and nets, block out light for other organisms, clog turbines and provide an excellent habitat for disease vectors such as mosquitoes and the snails which host the schistosomiasis parasite. Furthermore, since the rehabilitation will increase the water level at the Barrage, it is expected that the water levels of Lake Malombe would also increase with time; so too some negligible increase in the level for Lake Malawi. This expectation is made with assumption that the current climatic conditions prevail. Currently the Shire River water (within the area studied) is within the oligotrophic category although there are signs of changes towards the mesotrophic conditions. Therefore, deliberate monitoring and mitigation measures have to be instituted to ensure detection of any imminent pollution and immediate remedial measures. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec 2013 vii

176 1 Background Snowy Mountains Engineering Corporation (SMEC) International of Australia is carrying out an environmental study for the refurbishing of Kamuzu Barrage at Liwonde in which the regulated water level upstream would be increased by a maximum of 40cm. The Project will be implemented under the National Water Development Project Phase II by the Ministry of Water Development and Irrigation. It is expected that the change in the size of the Barrage may have some effects in several areas including food security, fisheries, water quality, traffic, water users as well as tourism. Kamuzu Barrage is located on the Shire River in the Southern Region of Malawi in Liwonde, Balaka District (Figure 1). The existing Barrage constructed in 1965 comprises 14 radial gates and shoulders a highway crossing over the bridge. Kamuzu Barrage partly regulates the water levels of Lake Malawi and the Shire River and provides significant benefits to individuals and the Malawian economy as a whole. The regulated Shire River flows benefit the hydropower generation in the Middle Shire. Historically, the water level of Lake Malawi fluctuated over seven metres. Before the Kamuzu Barrage was constructed, water flows in the Shire River varied strongly and in some years even fell almost dry (GoM, 2012). The Shire River is of great economic importance to Malawi as it generates more than 98% of Malawi s electricity, provides freshwater for irrigation in Malawi s plantations and supports fisheries. It is the only river that flows out from Lake Malawi and joins the Zambezi River before flowing into the Indian Ocean. Wetlands and marshes in the Shire Basin serve as breeding sites for migratory birds, production of crops and are key wildlife conservation areas. It is with expectation of potential impacts on water quality as a result of Barrage refurbishment that work on water quality was instituted to provide a basis of comparison for the future trends and contribute to the required information needed to properly design the refurbishment of the Barrage. 1.1 Water Quality Water quality is the physical, chemical and biological characteristics of water that makes it suitable for a specific use. The quality of water can be determined onsite by using probes or chemically in the laboratory and the results are compared to a set of standards as a yardstick for compliance (Graham, 1990). Water quality affects all water users. Aquatic life, recreational activities such as swimming and boating, agricultural uses such as irrigation are all affected by the physico-chemical, biological and microbiological conditions that exist in water courses. If good water quality is not maintained, it is not just the environment that will suffer; the commercial and recreational value of our water resources will also diminish (Graham, 1990). Many chemical analyses are carried out to ensure that water of appropriate quality is supplied for human consumption. Although the only natural input to any surface water system is precipitation within its watershed, the total quantity of water in that system at any given time is also dependent on many other factors. These factors include storage capacity in lakes, wetlands and artificial reservoirs, the permeability of the soil beneath these storage bodies, the runoff characteristics of the land in the watershed, the timing of the precipitation and local evaporation rates. 1.2 Literature review In its outline, the National Water Resources Policy for Malawi indicates the importance of water of acceptable quality for all the needs in Malawi. In achieving sustainable provision of water supply and sanitation services that are equitably accessible and used by individuals and entrepreneurs for socio-economic development at affordable cost. And in promoting efficient and effective utilisation, conservation and protection of water resources for sustainable agriculture and irrigation, fisheries, navigation, eco-tourism, forestry, hydropower and disaster management and environmental protection among others (GoM, 2003). It is therefore necessary that all activities within water bodies and their catchment adhere to this requirement. Among the water bodies that have to meet this requirement is Lake Malawi from which the Shire River emanates. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

177 1.2.1 Lake Malawi Water Quality Lake Malawi is an oligotrophic lake situated between Malawi, Tanzania and Mozambique in the rift valley of Southeast Africa between 3 and 10 S. With an area of 29,500 km 2, a volume of 8,400 km 3 and a maximum depth of 706m, it is one of the world s largest lakes. Lake Malawi is also the most species rich lake in the world with an estimated more than 1,000 species of cichlids (fish). The Lake is permanently stratified and anoxic below 200 m and has a water cycling time of 750 years due to low outflow volume, with residence time of approximately 114 years. On average, Lake Malawi receives 39 km 3 of precipitation yearly directly on its surface and it has a single rainy season starting in December and ending around May. The climate in Malawi is subtropical and monsoonal. The permanently warm water (23-30 C in surface water, 22.4 C in deep water) makes Lake Malawi different from other lakes. The slow water cycling time of Lake Malawi makes it particularly vulnerable with regards to chemical pollution since dilution will be ineffective in reducing contaminant concentrations. Therefore, it is important to prevent levels of the pollutants from increasing within the Lake Malawi region as well as in other vulnerable ecosystems (Karlsson et al., 2000). Whilst previous research has shown that there is a greater input of nutrients into Lake Malawi by various rivers that empty their water into the Lake from excessive erosion on cultivated lands (Mkanda, 2002), the Shire River outflow only removes nutrients from the surface mixed layer of the Lake. The outflow is minor in comparison to the total inflow of nutrients from other rivers, and is small when compared with other pathways of nutrient removal. For C, N and P, the ratio of particulate to dissolved fractions is greater in inflowing rivers than in the Shire River. Due to its large volume and long hydraulic residence time, Lake Malawi may respond very slowly to increased inputs of certain contaminants. Nutrient concentrations in Lake Malawi are low, and therefore phytoplankton concentration is also low. The surface water concentrations of dissolved nitrogen and phosphorus are below mgl -1 and mgl -1 respectively, and chlorophyll a concentrations (used as an index of phytoplankton abundance) are below 1 μg L -1. The epilithic periphyton of Lake Malawi is dominated by Heterocystous cyanoprokaryotes (blue-green algae/cyanophytes) and diatoms containing endosymbionts capable of nitrogen fixation, though in some locations impacts such as sediment and nutrient loading have shifted the algal community structure towards dominance by diatoms and chlorophytes. In addition to this the Lake is typically slightly alkaline with a ph ranging from 7.7 to 8.6, a carbonate hardness of 107 to 142 mg L -1, and a conductivity of 210 to 285 µs cm Water Quality Impacts of Barrages Connell, Bycroft, Millev, and Lather (1981) reported a reduced Dissolved Oxygen (DO) concentrations and elevated Biochemical Oxygen Demand (BOD) and nutrient concentration downstream of barrages. Reduced downstream flushing as a consequence of construction of a barrage has been suggested as a contributing factor to the poor water quality downstream. In support of this, Turnbull (1974) in Connell et al. (1981) reported an increase in salinity downstream of barrages, suggesting a reduced rate of flushing. Furthermore, Chaman, Tanveera, and Kumar (2008) showed increase of organic pollution and eutrophication as indicated by a greater concentration of NH 4, NO 2, and Total P, as well as lower dissolved oxygen concentration at a barrage. The enhanced concentrations of BOD were in agreement with decreased DO concentrations. Chaman et al. (2008) further attributed high concentrations of Ca, Mg, Cl and Na ions as a result of sewage and industrial effluents within a barrage catchment and Kirby and Retière (2009) showed that oxygen depletion is induced by entraining suspended fine sediment which accumulate nutrients like phosphorous which can be made available to the surface due to changes in aeration and enhance algal blooms. The decline in river water quality due to pollution from sewage and industrial effluent may have a direct detrimental impact on the health of water user groups up and downstream. Incidents of diarrhoea, typhoid, jaundice, cholera and skin diseases are of common occurrence among user groups. Polluted river water can also seriously impair the aquatic ecology and biodiversity along the rivers (Hammer & Linke, 2010). Gray (1992) indicated that the two most significant water quality challenges experienced in a barrage catchment are biological (faecal) and chemical pollution from domestic and industrial effluents which emanate from surrounding residential and industrial sources due to the fact that barrage catchments are often a focus of economic activities. Furthermore, many of the contaminants discharged have an affinity for fine particles and therefore accumulate in the bottom sediments. Episodes of erosion of sediments by storms and changes in oxygen concentration as well as bio-turbation processes may lead to desorption of these contaminants into the water column with potential adverse impacts (Gray, 1992). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

178 Furthermore, changes in water level upstream due to presence of barrages can lead to loss of some macrophytic plants whilst those that are hydrophytic would increase in number and this would affect the biodiversity of macrophytes along the river (Connell et al., 1981; Kirby & Retière, 2009). Water quality monitoring has shown that after heavy rainfall, water quality of barrage reservoirs and other aquatic ecosystems quickly decline. Organic, inorganic and toxic pollutants generated from agricultural and industrial sources are accumulated and get mixed with river water posing a threat to aquatic life and making water unfit for human consumption (Karlsson et al., 2000; Mkanda, 2002). 1.3 Objective of the Study Overall Objectives The overall objective of this study was to describe the water quality in the Project area and assess impacts of the Kamuzu Barrage in Liwonde as part of the environmental study for the refurbishment of the Barrage Specific Objective The following were the specific objectives of the study: To determine the chemical, physical and bacteriological water quality of the Shire River at Kamuzu Barrage. To determine the impacts of refurbishment of the Barrage on water quality. To develop suitable mitigation measures to minimise impacts on water quality during construction and operation of the Barrage. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

179 2 Materials and Methods 2.1 Study Area The study was conducted along the Shire River with five sampling point stretching from the outlet of Lake Malombe to a point downstream of the Kamuzu Barrage. The Kamuzu Barrage is constructed on Shire River at Liwonde Road block in Machinga district in the Southern region of Malawi (Figure 1). The study was conducted in November, 2011 and January, North Lake Malawi Lake Malombe Kamuzu Barrage Figure 1. Map of Malawi showing Shire River and the sampling site (Source: GoM and Mtethiwa, 2012) Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

180 2.2 Sampling Points and Design Five main sampling points were selected, denoted by S1, S2, S3 (S3a, S3b, S3c), S4 (S4a, S4b, S4c) and S5 (S5a, S5b, S5c) in Figure Sampling Point S1 S1 sampling point was located at the outlet of Lake Malombe. The point was selected with the understanding that Lake Malombe is one of the water bodies whose water quality would be affected by the refurbishment and operation of the Barrage which will affect water levels within the Lake. It was therefore important to determine the current levels of concentration of various water quality parameters to be used as a basis for comparison in future. Three water samples were collected and determined for the various water quality parameters Sampling Point S2 S2 sampling point was located adjacent to the Liwonde National Park. This location is one of the busiest areas in terms of wildlife, tourists and boating. The area has a variety of terrestrial as well as aquatic wild animals. Beside this, just above this sampling point, there is a stream that joins the Shire River (Figure 2). As such, it was an important sampling to determine the impact of catchment activities as well as the National Park on water quality Sampling Point S3 The S3 sampling point was located after Liwonde National Park and close to Hippo View Lodge. The point was selected with the background to determine the impact of catchment development activities such as lodges and resorts on the water quality to be used as a basis to project the situation after the Barrage has been refurbished. Here three sampling points were used (denoted S3a, S3b, and S3c) and three water samples were collected from each of the sites at each sampling time Sampling Point S4 This point was about 15 m upstream from the Barrage and about 50 m from the Hippo View Lodge and the ESCOM Weed Trapper. There were three sampling points denoted S4a, S4b, S4c where three water samples were collected at each sampling time. This point was important to determine the impact of the activities around the Barrage on the water quality as well as the weed boom on water quality. Nevertheless, this was also selected to determine if the water retention period, which is more conspicuous within this area, has any significant effect on the concentration of various water quality parameters Sampling Point S5 This point was about 200 m downstream of the Barrage. The point was chosen to determine the water quality after the Barrage, that is, if the Barrage has any impact on the water quality downstream. Just as for Sampling site S4, there were three sampling points here (denoted S5a, S5b, S5c) where three water samples were collected at each sampling time. 2.3 Collection and Handling of Water Samples A Schindler was used to collect the water which was immediately transferred into 500 ml plastic (polythene) bottles which had been previously rinsed with distilled water. The water samples were kept in a cooler box, immediately after being collected, where they were maintained at very low temperatures by using ice blocks placed in the cooler. 2.4 Important Water Quality Parameters Water quality parameters are variables (that are part of the water system) that are used to indicate the status of the water (whether the water is of good or bad quality). The following water quality parameters were used in this study: Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

181 2.4.1 Dissolved Oxygen (DO) This is the amount of Oxygen (O 2 ) dissolved in water. It gets in water either by diffusion (diffusion gradient) from the surrounding air, currents and waves or photosynthesis by aquatic plants and algae. If water is too warm, there may not be enough oxygen in it. When there are too many bacteria, algae or aquatic animal in the area, they may overpopulate hence using DO in great amounts leading to Oxygen deficiency. Oxygen levels also can be reduced through over-fertilisation of water plants by run-off from farm fields containing phosphates and nitrates (the ingredients in fertilisers). Under these conditions, the numbers and size of water plants increase a great deal. Then, if the weather becomes cloudy for several days, respiring plants will use much of the available DO since they cannot photosynthesise. When these plants die, they become food for bacteria, which in turn multiply and use large amounts of oxygen. The amount of DO an aquatic organism needs depends upon its species, its physical state, water temperature, pollutants present, and more. Fish are cold-blooded animals, so they use more oxygen at higher temperatures when their metabolic rate increases. Kamuzu Barrage (Source: GoM and Mtethiwa, 2012) Figure 2. Part of map of Malawi showing Lake Malombe and Kamuzu Barrage and water sampling points along the Shire River. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

182 Boyd (1990) and (Chapman, 1996) suggest that 4-5 ppm of DO is the minimum amount that will support a large, diverse fish population. The DO level in good fishing waters generally averages about 9.0 ppm. When DO levels drop below about 3.0 ppm, even the most DO tolerant fish may die ph Value The balance of positive hydrogen ions (H+) and negative hydroxide ions (OH-) in water determines how acidic or basic the water is. The ph scale ranges from 0 (strongly acidic) to 14 (strongly basic). In pure water, the concentration of positive hydrogen ions is in equilibrium with the concentration of negative hydroxide ions, and the ph measures exactly 7 (Quellette & Rawn, 1997). In a lake or pond, the water s ph is affected by its age and the chemicals discharged by communities and industries. Most lakes are basic (alkaline) when they are first formed and become more acidic with time due to the build-up of organic materials. As organic substances decay, carbon dioxide (CO 2 ) forms and combines with water to produce a weak acid, called carbonic acid. Large amounts of carbonic acid lower water s ph. Most fish can tolerate ph values of about 5.0 to 9.0. The vast majority of Malawian rivers, lakes and streams fall within this range (Bootsma, 2001) Phosphorous The element phosphorus is necessary for plant and animal growth. Nearly all fertilisers contain phosphates (chemical compounds containing the element, phosphorous). When it rains, varying amounts of phosphates wash from farm soils into nearby waterways. Phosphates stimulate the growth of plankton and water plants that provide food for fish. This may increase the fish population and improve the waterway s quality of life. If too much phosphate is present, algae and water weeds such as water hyacinth grow wildly, choke the waterway, and use up large amounts of oxygen. Many fish and aquatic organisms may die Turbidity Is the optical property of a water sample that causes light to be scattered and absorbed rather than transmitted in straight lines through the sample. Simply, turbidity answers the question, How cloudy is the water? Ability of light to pass through water depends on how much suspended material is present. Turbidity may be caused when light is blocked by large amounts of silt, microorganisms, plant fibres, sawdust, wood ashes, chemicals and coal dust. The most frequent causes of turbidity in lakes and rivers are plankton and soil erosion from poor land-use practices in the upper catchment, particularly the clearing of vegetation Water Temperature Is a measure of how hot or cold the water is. The water temperature is results of several factors, among them are: the colour, depth, shoreline vegetation shade, the location (temperate or tropic region), the time of year, the temperature of the water supplying the waterways and the volume of the water. Fish and most aquatic organisms are cold-blooded. Consequently, their metabolism increases as the water warms and decreases as it cools. Each species of aquatic organism has its own optimum (best) water temperature. If the water temperature shifts too far from the optimum, the organism suffers Nitrate (NO3-) Nitrate (NO 3- ) is a form of the element nitrogen, an essential component of life, is found in the cells of all living things. Nitrites (NO 2- ) are relatively short-lived because they are quickly converted to nitrates by bacteria. Nitrites produce a serious illness (brown blood disease) in fish, even though they do not exist for very long in the environment. Nitrates stimulate the growth of plankton and water weeds that provide food for fish. This may increase the fish population. However, if algae grow too abundantly, oxygen levels will be reduced and fish will die. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

183 2.4.7 Biochemical Oxygen Demand Biochemical oxygen demand or BOD is the amount of dissolved oxygen needed by aerobic biological organisms in a body of water to break down organic material present in a given water sample at certain temperature over a specific time period. The term also refers to a chemical procedure for determining this amount. This is not a precise quantitative test, although it is widely used as an indication of the organic quality of water. The BOD value is most commonly expressed in milligrams of oxygen consumed per litre of sample during 5 days of incubation at 20 C. However, it is often used as a robust surrogate of the degree of organic pollution of water (Chapman, 1996; Chiras, 1998) Total Dissolved Solids (TDS) Total dissolved solids (TDS) comprise inorganic salts (principally calcium, magnesium, potassium, sodium, bicarbonates, chlorides and sulphates) and some small amounts of organic matter that are dissolved in a solution (Cech, 2002). TDS is the total amount of dissolved materials/solids which become salts when evaporation takes place and it is an indication of salinity (D. Chapman, 1998). A high TDS load can also increase water hardness rendering laundry soap ineffective; it can also produce an unpleasant taste to drinking water if exceedingly chlorinated (USEPA, 2010). Many crops are also intolerant to high water salinity Electrical Conductivity (EC) EC measures the flow of electric current in a water sample as various literatures report that dissolved solids (ions) conduct electricity. (Patterson, 1999) indicated that EC provides a good measure of total salinity. (USEPA, 2010) and (Fipps, 2009) reported that there is a linear relationship between TDS and EC i.e. the more the total dissolved salts in a solution, the higher the electric current conducted Chlorophyll a D. Chapman (1998) described chlorophyll a as the green pigment in photosynthetic plants, indicative of algal biomass. Algae, being primary producers, support secondary producers and other higher organisms (Jeffrey & Humphrey, 1997). The growth of algae requires nutrients (predominantly phosphates and nitrates) in addition to sunlight and temperature. An increase in phosphates and nitrates in a river/lake will cause a rapid multiplication of the algae. Therefore the degree of chlorophyll a concentration shows the trophic status of that particular river/lake thus an indicator of water quality Sulphate Ceccotti (1996) stated that sulphur is a major constituent of amino acids in all living things and it also enhances the efficient utilisation of N and P. Sulphate occurs in many forms which are highly soluble in water hence if it is mixed with water, acid water may be formed which can damage the environment. Major sources of S include burning fossil fuels as well as decomposition of sulphur-containing organic matter. Bacteria break down the sulphur-containing organic matter into hydrogen sulphide which produces a bad odour. It is indicated that some aquatic organisms which thrive under anaerobic conditions do so by use of sulphate as electron acceptor. Increase in sulphate in drinking water causes gastro-intestinal irritation when magnesium or sodium is present Hardness Hardness in water is defined as concentration of multivalent cations. Multivalent cations are positively charged metal complexes with a charge greater than 1+. They mainly have the charge of 2+. These cations include Ca 2+ and Mg 2+. Hardness can be quantified by instrumental analysis. The total water hardness, including both Ca 2+ and Mg 2+ ions, is reported in parts per million or mass /volume (mg/l) of calcium carbonate (CaCO 3 ) in the water. Although water hardness usually measures only the total concentrations of calcium and magnesium (the two most prevalent bivalent metal ions), iron, aluminium, and manganese can also be present at elevated levels in some locations. The presence of iron characteristically confers a brownish (rustlike) colour to the calcification, instead of white (the colour of most of the other compounds). Because it is the precise mixture of minerals dissolved in the water, together with the water s ph and temperature, that determines the behaviour of the hardness, a single-number scale does not adequately describe hardness. Descriptions of hardness correspond roughly with ranges of mineral concentrations: Soft (0-60ppm), moderate (61-120ppm) and Hard ( ppm), and very hard ( 181ppm). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

184 Faecal Coliform Bacteria The Bacteriological contamination assessment in water is based on measurement of faecal coliform bacteria levels in water. This can be used to determine how much faecal contamination has occurred. Coliform bacteria are a good indicator of the quality of water for drinking or swimming. The higher the coliform count, the more likely the water is to contain some pathogenic agent from faecal contamination. (Chiras, 1998) indicates that rivers in many less developed nations are very heavily polluted with human wastes. Although there will always be coliform bacteria present in natural waters, high levels provide evidence of water pollution from sewage. 2.5 Water Quality Parameters Determined on Site The following water quality parameters: Electrical Conductivity (EC), Total Dissolved Solids (TDS), Turbidity (Turb), Temperature, Dissolved Oxygen (DO), Salinity and ph were measured on site using a Horiba water checker. 2.6 Water Quality Parameters Determined in the Laboratory They included the following parameters Nitrates Determination of Nitrate (NO 3- ) in all water samples was done spectrophotometrically. Absorbance of the solutions was measured on a HACH DR/3000 UV/Visible Spectrophotometer at 410 nm against a blank. The NO 3- concentration was calculated from the calibration curve obtained by using standard solutions for NO 3- (APHA, 2005) Soluble Reactive Phosphorous (SRP) The water samples were filtered immediately through glass fibre filters (Whatman GF/F, 0.6µm pore size. The orthophosphate was then determined colourimetrically using a HACH DR/3000 UV/Visible Spectrophotometer at 880 nm without hydrolysis process. Phosphate concentration was calculated from the calibration curve by using the standard solutions (APHA, 2005), (Wetzel & Likens, 2001) Sulphate Determination of Sulphate was done using the gravimetric method after precipitation by barium chloride in hydrochloric acid. Absorbance was measured on a UV/Visible spectrophotometer, Model HACH DR/3000 at 430 nm (APHA, 2005) Biochemical Oxygen Demand (BOD) Determination of BOD was done using the Winkler (BOD5) method (APHA, 2005). After incubating the sample at 20 C for five days in the dark, BOD was calculated as follows: BOD5 (mg/l) = [(Initial DO-Final DO) x Dilution Factor] Blank Faecal Coliform Bacteria (FCB) Membrane filter method was used where 100 ml of sample water was filtered through a cellulose membrane filter. It allows water to pass through it but it retains coliform on its surfaces. The filter was placed in the media (M endo Broth) and incubated for 24 hours at 35 C. Blue colonies indicated the presence of coliform (APHA, 2005) Chlorophyll a Determination of chlorophyll a was done by using spectrophotometric method as described by (Jeffrey & Humphrey, 1997). The sample was first filtered using a Whatman GF/F glass fibre filter and was then centrifuged. Absorbance of the samples was measured using a U/V spectrophotometer at 750, 664, 647 and 630 nm against a 90% acetone blank. Chlorophyll a concentration was calculated using the formula: Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

185 Chlorophyll a (mg/m 3 ) = (11.85*(E 664 -E 750 ) *(E 647 -E 750 )-0.08(E 630 -E 750 )*Ve/L*Vf (Jeffrey & Humphrey, 1997) Where: L = Cuvette light-path in centimetres, Ve = Extraction volume in millilitres, Vf = Filtered volume in litres Total Suspended Solids (TSS) A well-mixed volume of the sample was filtered through a weighed standard glass-fibre filter and the residue retained on the filter was dried to a constant weight at C. The increase in weight of the filter was taken to represent the Total Suspended Solids in accordance to (APHA, 2005) Transparency This was determined by use of a Secchi disc. A Secchi disc is a circular disc painted black and white with a diameter of about 15 cm. It has a calibrated rope/string tied to it. Transparency was determined by lowering a Secchi disc into the water at a perpendicular position and a reading on the calibrated rope was recorded at a point where it just disappeared and then it was further lowered into the water until it completely disappeared. A second reading of the Secchi disk was obtained at a point where the disc reappeared when it was being lifted from the water. An average of the two readings gave a Secchi depth which determines water transparency. 2.7 Laboratory Facilities Bunda College laboratory and equipment was used to analyse the samples. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

186 3 Results and Discussion Results in Tables 1 and 2 show that of the three selected graphed parameters, turbidity and phosphorus concentrations were higher during the rainy season than the dry season. The same tendency is also portrayed by BOD, TSS and Secchi depth as indicated in Tables 1 and 2 below. Variation in readings of these parameters is based on the seasonality and locality and therefore is largely a sign of effects of catchment activities. 3.1 Selected Water Quality Parameters Turbidity During rainy season, surface runoff carries with it silt and other substances. Once these enter the aquatic ecosystem, they mix with the clear water reducing its transparency and increasing its turbidity (Table 1). The amount of silt in the runoff is exacerbated by poor agronomic practices. This is the scenario around the catchment of Shire River especially the upper area. The GoM (2003) reported that there is widespread dimba (dry season) farming in the Mangochi area (an average of about 13% of farming being dimba farming). However, from observation, the dimba farming in the project area is much higher than the reported 13%. The dimba farming in the area occurs along the Shire River and Lake Malombe and is associated with poor agronomic practices and application of fertilisers which is responsible for release of silts and nutrients into the water that increases turbidity of the water and macrophyte production (GoM, 2003). Also, it was observed that there are a number of buildings along the Shire River especially at the Kamuzu Barrage. These are mostly resorts and lodges. Construction processes often leads to clearing of the land giving rise to bare and loosened soil. Once rains come, soil particles are dislodged and carried away by runoff into the aquatic ecosystem. Turbidity in excess of 50 NTU is disadvantageous to fish because turbid water clogs the gills bringing discomfort and sickness. Besides this, high turbid water reduces visibility in fish. As a result, two disadvantages arise, firstly: the fish will mostly starve because they cannot find food affecting their productivity; and, secondly fish cannot see their predators. As a result most of them are easily caught. The more the fish are caught the lower the recruitment success hence affecting the population and catch in subsequent years. Very high turbid water reduces hatching of eggs since the eggs are covered by sediments (Boyd, 1990). High turbid water also creates ideal conditions for the growth of pathogenic bacteria that affect water users. If turbidity is high, effectiveness of chemicals, such as chlorine, that are used to kill the pathogens in water abstracted for other domestic uses, is poor. TABLE 1. Location Upstream of the weed trapper (S3) Upstream of Barrage (S4) Downstream of Barrage (S5) MEAN READINGS OF SELECTED WATER QUALITY PARAMETERS FOR THE DRY SEASON Chlorophyll a (mg/l) Hardness (mg/l) Phosphorus (µg/l) Secchi depth (cm) Turbidity (NTU) a a a a a b b b b b c c c c c X Means having the same letter in any given column are not significantly different from each other at 0.05% level of significance. Furthermore, high turbid water reduces the amount of sunlight entering the water. This then means that primary production in terms of phytoplankton, which is food for fish, would be heavily reduced. Reduction of the the available food for fish affects fish productivity. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

187 With the fore going levels of turbidity and activities around the catchment, it is predicted that unless other catchment protection measures are taken, there is an impending danger of increased turbidity in the Shire River after the Barrage has been rehabilitated and raised. This is because upstream water levels will increase making the water more readily available for catchment farming. Turbidity levels of above 150 NTU are not recommended for flowing water as they are a source of the above problems and turbidity of more than 5 NTU is not recommended for drinking water Chlorophyll a The findings of this study show that the wet season chlorophyll a concentrations (Table 2) are lower than the dry season values (Table 1). This is attributed to the dilution effects of the rains. This dilution is both directly on the phytoplankton concentration that translates to low chlorophyll a concentration and to the nutrient (nitrogen) that is responsible for proliferation of phytoplankton responsible for this chlorophyll a. Evidence is shown by the NO 3- concentration determined above, there is a decline in the concentration of the nutrient and hence subsequent reduction in chlorophyll a (Tables 1 and 2). The green pigment (chlorophyll) which is present in most photosynthetic organisms provides an indirect measure of algal biomass and an indication of trophic status of a water body. The growth of algae in water is directly related to the concentration of nutrients (principally nitrogen and phosphorous), temperature and light. Water bodies with low levels of nutrients (oligotrophic) have chlorophyll a of less than 2.5 µg/l whereas water with high levels of nutrients (eutrophic) have high chlorophyll a (concentrations of more than 45 µg/l) Excessive concentration of P and N leads to overgrowth of phytoplankton a condition called algal bloom. This is not good for aquatic organisms as it would lead to reduced oxygen concentration and increased carbon dioxide concentration at night. The increase in carbon dioxide leads to formation of weak carbonic acid which may increase the acidity of the water to levels not conducive to aquatic organisms, including fish. The reduced oxygen level at night will lead to a situation where aquatic organisms are stressed and offer reduced productivity. On the other hand, algal blooms will reduce light depth as most of the radiation will be blocked by floating algae. Despite all this, the levels of chlorophyll a concentrations both in the diluted (rainy) and undiluted (dry) state is below the water eutrophication state. This then means that Shire River water, in terms of nutrient enrichment, has not yet reached an alarming state, (Tables 2 and 3 and Appendix 1). It was expected that the increased concentration of phosphorous in the rainy season would trigger an increase in phytoplankton hence chlorophyll a as the two are directly linked. However, the reduced chlorophyll a in the presence of increased phosphorus concentration may signify that at this time dilution had more influence on the chlorophyll a concentration than the nutrients. TABLE 2. Site Shire outlet at Lake Malombe S1) Liwonde National Park (S2) Upstream of weed trapper (S3) Upstream of Barrage (S4) Downstream of Barrage (S5) MEAN READINGS OF SELECTED WATER QUALITY PARAMETERS FOR THE WET SEASON Chlorophyll a (mg/l) Hardness (mg/l) Phosphorous (µg/l) Secchi depth Transparency (cm) Turbidity (NTU) Coliform (CFU/100ml) ± 0.02 c ± ± 0.06 c ± 1.38 a ± ± 1.92 c 1.10 b 1.53 b ± 0.11 b ± ± 0.04 b ± 0.88 a 120.0± ± 1.13 a 1.83 c ± 0.19 b ± ± 0.21 b ± 4.41 b ± ± 1.18 c 0.09 e 5.14 b ± 0.54 a ± ± 0.26 a ± 0.58 bc ± ± 0.75 ab 0.00 d 0.55 b ± 0.22 c ± ± 0.04 b ± 1.12 c ± ± 0.73 d 0.01 c 2.25 a X Means having the same letter in any given column are not significantly different from each other at 0.05% level of significance. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

188 3.1.3 Phosphorus Intense ecological interest in phosphorus stems from its major role in metabolism in the biosphere (Chapman, 1996). In comparison to relative supply of other major nutritional and other structural components of the biota (C, N, O, S), phosphorus is the least abundant and commonly limits biological productivity in aquatic ecosystem (Wetzel & Likens, 2001) Phosphorus will come from the biota as well as the sediments and other colloidal matter. These are either lost by sedimentation or hydrolysed to orthophosphate which is rapidly assimilated by biota leading to low concentrations of phosphorus in the upper layer (trophogenic zone) of the water body, where enough light is present to permit photosynthesis. Because of the role of phosphorus in primary production, an excess amount of it induces phytoplankton and macrophytes production in water bodies. It is therefore imperative to ensure a substantive amount of phosphorus is reduced to ensure oligotrophic conditions (Chapman, 1996). In this study, P concentrations were higher during the rainy season than during the dry season (Tables 1 and 2; Figures 3 and 4). These results support the facts that runoff from rains bring additional input of P into the aquatic ecosystem. As indicated above, the surrounding catchment of the Shire River is heavily cultivated for dimba and field crops and phosphorus fertiliser such 21:23:0:+4S is mostly used. This implies that during the rainy season, a substantial amount of phosphorus is carried into the aquatic ecosystems through runoff. Apart from agricultural areas, runoff also comes from lodges, resorts and other residential areas along the River. These areas can be another source of phosphorus where a lot of detergents are used. There is not much variation in different sites, although higher concentrations were recorded at the Barrage than Lake Malombe (mouth of Shire River). Probably this portrays how the catchment development activities and water residence time contributes to phosphorus input and accumulation into the aquatic system respectively (Mumba and Kaunda, 1999). TABLE 3. PARAMETER Conductivity (µs/cm) MEAN CONCENTRATION OF SELECTED WATER QUALITY PARAMETERS FOR THE SHIRE RIVER ( ) YEAR TDS (mg/l) Nitrate (mg/l) < <0.1 <0.1 <0.1 Sulphate (mg/l) < Hardness (mg/l) Turbidity (NTU) TSS (mg/l) Source: GoM, Faecal Coliform Bacteria Faecal Coliform Count in the form of Coliform Forming Units (CFU) is high (Table 2) showing that the water is contaminated with faecal matter. This may indicate that bacteria have been brought into the river system by rain water which washed away faecal matter from the catchment. In terms of numbers, there is a low count at the mouth of Shire River at Lake Malombe and increased CFU at the Barrage. Since faecal coliform is of animal origin, going by these results means that there is more faecal pollution at Liwonde than at the mouth of Shire River (Table 2). For untreated, chlorinated supplies, such as surface water or shallow or deep well waters, the detection of faecal coliform alone can generally serve as an adequate guide for the presence of pathogenic organisms of faecal origin (WHO, 1996). The faecal matter may have also contributed to the increase in phosphorus concentration during the wet season as shown in Table 2. The source of these could be the fishermen who may defecate into the water whilst fishing, the community which roams around the river bank for plant biomass hunting, livestock grazing along the catchment and effluent discharges from the resorts and lodges within the river catchment. If measures are not taken to ensure sanitary measures are followed, it is expected that when water levels increase, the number of activities along the river would also increase leading to increased faecal pollution. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

189 Figure 3. Trends of selected parameters in dry and wet seasons at different sampling points 3.2 Phosphorous, Nitrogen and Aquatic Weeds A combination of phosphorus and nitrogen is very crucial in relation to aquatic primary production. Excessive concentrations of the two nutrients will lead to a eutrophic state of the water which may be manifested by macrophyte growth (aquatic weeds) rather than in the growth of phytoplankton. The effect of eutrophication can be highly detrimental to water quality and severely limit the uses for which the water is meant. In relation to concentration of the phosphorous and nitrogen, the water of the Shire River is basically within the oligotrophic category. Figure 4. Mean concentrations of selected parameters at various sampling points Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

190 4 Comparison of the Results with Previous Studies Previous studies conducted between (GoM, 2003, Table 3) on selected water quality parameters show higher readings of Total Dissolved Solids (TDS) and Electrical Conductivity (EC) and lower Turbidity (among other parameters) in comparison to the current findings. Increase in turbidity by close to three fold indicates the increase in pollution over the years. However, although there are these differences, the water quality was still in the oligotrophic category in both situations. TDS is a result of inorganic and organic substances with Ca, PO 4-3, NO 3-, Na, K and Cl being the common chemical constituents whilst EC is a result of ion concentration, mobility, oxidation state as well as temperature of the water. TDS is directly related to EC such that TDS=0.5 times conductivity. Also, the environmentally acceptable conductivity of rivers is between µS/cm while less than 200mg/L TDS has no effect on aquatic organisms including fish spawning. Values outside these ranges would negatively affect productivity of the aquatic ecosystem. Therefore using this guideline, the water quality situation is within the oligotrophic category since research has shown that concentrations of less than 200mg/L TDS have no effect on aquatic organisms including fish spawning (Chapman, 1996). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

191 5 Possible Water Quality Impacts due to Heightened Barrage 5.1 Impacts during Construction Phase Increased Siltation and Turbidity of Water Downstream It is anticipated that activities during construction of the Barrage will lead to temporary increased production of silt leading to more turbid water downstream. Apart from this, since water will be impounded, there will be reduced flow of water affecting the nutrient retention of the river and therefore concentration leading to isolated algal blooms and proliferation of other parasitic water borne diseases Increased Levels of Water Upstream It is anticipated that once the water has been impounded due to the increased height of the Barrage gates, the level of the water upstream would increase. Also, nutrients from upstream that have not been assimilated by the aquatic biota will accumulate within the impoundment. An increase in nutrients especially phosphorus and nitrogen may lead to algal blooms as well as growth of aquatic weeds upstream. 5.2 Impacts during the Operation Phase Barrages are likened to damming in rivers and/or weirs in most studies. This is because of the changed flow regime downstream, increased water levels and increased retention time of water in the reservoir upstream. Due to the foregoing outlined factors associated with barrages, there are likely changes in water quality as described below Nutrient Enrichment due to Increased Water Retention There is likely to be sedimentation of nutrients due to reduced flushing of the water within an impoundment. Furthermore, because the depth of the Barrage reservoir is relatively shallow, wind driven events can lead to a substantial turbation (turnover), allowing periodic pulses of nutrients into the surface layer of the water. These nutrient pulses then stimulate algal growth making the water nutrient-enriched (Baldwin et al., 2007). In general, longer water retention times will result in greater settlement of sediments and nutrients within the water storage resulting in lower turbidity. However, during prevailing winds, the bed sediments would be disturbed causing re-suspension of the settled sediments and nutrients and leading to highly enriched water. Therefore whilst TDS is expected to be low, total nitrogen and phosphorus levels may exceed the guidelines. This may cause increased growth of aquatic plants and algae. Nutrient-enriched tropical reservoirs are particularly prone to colonisation by aquatic plants. Mats of floating plants can impede fishing boats and nets, block out light for other organisms, clog turbines and provide an excellent habitat for disease vectors such as mosquitoes and the snails which host the schistosomiasis parasite Potential Contamination by Wastewater from Resorts, Lodges, Markets and Fishing Activities around the Barrage Currently there are a number of resorts and lodges around the Barrage area. Partly it is because the presence of larger volumes of water and its locality. The probability is high that, with time, there would be an increased number of such structures which will have an impact on water quality if proper measures are not adhered to regarding the discharge of both solid and liquid wastes which they will generate. The household wastes are sources for nutrient and pathogen production. Since there is potential increase in primary production, it follows that fish recruitment would increase leading to increased fishing activities around the Barrage. The fishing activities have the potential to reduce water quality in the Barrage reservoir through human waste and food remains entering the water. Under conditions of high river discharge there may be little variation in water quality since nutrients would be flushed out, but under low discharge conditions substantial reductions in dissolved oxygen concentration and increases in nutrient and chlorophyll a concentrations would be noted in the zone immediately upstream and downstream of the Barrage (as noted by Connell et al. 1981). Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

192 6 Proposed Mitigation Measures In order to prevent the anticipated impacts of this Barrage and ensure clean water, the following are proposed mitigation measures. 6.1 Mitigation Measures during Construction Phase Water Pollution Monitoring During construction, it is suggested that ph, nutrient load, chlorophyll a, BOD and turbidity be measured every fortnight to ensure any upsurge is detected for prompt correction measures. 6.2 Mitigation Measures during Operation Phase Catchment Management The river catchment is one of the expected sources of nutrients that may pollute water resources. It is therefore necessary to devise ways to manage the catchment to mitigate the pollution. These may include; Implementation of land conservation practices (catchment management measures) such as agroforestry and recommended land use practices to impede run-off. Protecting the river buffer zone from encroachment by agriculture and settlements through enforcement of existing laws, and imposing heavy fines on offenders. Encouraging the formation of local authorities (committees) in the management of catchment areas Proper Waste Water Management Practices Since waste is one of the expected sources of water pollution in the area the following is proposed: Implementation of proper waste management and treatment procedures; Complying with international standards for solid waste and liquid waste disposal Public Awareness / Stakeholder Involvement Public awareness campaigns should be encouraged on the need to protect water resources from pollution and the dangers of polluted waters. People are encouraged and mobilised to fully participate in any activity if they feel the sense of involvement and belonging Sanitary Measures There is a need to institute proper sanitary measures at the Kamuzu Barrage in Liwonde and in the river catchment to reduce the risk of eutrophication which is a seed for phytoplankton and macrophyte proliferation in aquatic ecosystems Water Monitoring System There is a need to set up at least five monitoring points both upstream and downstream of the Barrage. At these points, ph, N, P, turbidity, BOD chlorophyll a, and faecal coliform bacteria should be measured every three months. This will assist detection of any adverse changes in the water quality and allow corrective measures to be implemented promptly. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

193 7 Conclusion The water quality of the Shire River is within the oligotrophic category although the concentrations of various parameters pose an impending threat of shifting to mesotrophic conditions with time especially due to anthropogenic activities as well as input from runoff from the surrounding communities. The consequences of such activities are already conspicuous on turbidity, faecal coliform bacteria, hardness and nutrients which are increasing. Also it has been shown that catchment development activities significantly contribute to increased pollution of the water. This means that measures have to be taken to ensure that catchment development activities (such as farming and resort developments) do not enrich the Shire River with nutrients and pathogenic bacteria. Furthermore, it has been shown that reduced flow of water also contributes to increased pollution of the river. For example, at the Barrage where water flow is regulated, concentrations of various parameters have been shown to increase in the sediments upstream due to accumulation and downstream due to reduced flushes. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

194 8 References Adams W M, (1985). The downstream impact of dam construction, a case study of Nigeria transaction of British geographers, NS 10: APHA, A. P. H. A. (2005). Standard methods for the Examination of Water and Wastewater, Washngton DC. Baldwin, D. S., Gigney, H., Wilson, J., Boulding, A., Watson, G., & Huzzey, L. (2007). Lake Hume Integrated Water Quality Study, The Murray-Darling Freshwater Research Centre. The Murray-Darling Basin Commission. Bootsma, H. (2001). Lake Basin Management initiative-lake Malawi/Nyasa. Boyd, C. E. (1990). Water quality management for pond fish culture. Elsevier science publishers. New York, USA. Ceccotti, S. P. (1996). Plant nutrient sulphur- a review of nutrient balance, environmental impact and fertilizers. Fertilizer Research, 43, Cech, V. T. (2002). Principles of water resources: History development management Policy, John Wiley and Sons, Inc. London. Chaman, T., Tanveera, T. m., & Kumar, R. (2008). Water Quality of the River Yamuna in the Delhi stretch: Key Determinants and Management Issues. Clean, 36(3), Chapman. (1996). Water Quality Assessments - A Guide to Use of Biota, Sediments and Water in Environmental Monitoring, E&FN Spon, Cambridge, Britain, ISBN (HB) (PB). Chiras, D. D. (1998). Environmental Science: A system approach to sustainable development, Wadworth Publishing Company, New York. Connell, Bycroft, B. M., Millev, G. J., & Lather, P. (1981). Effects of a Barrage on Flushing and Water Quality in the Fitzroy River Estuary, Queensland. Ausi. J Mar Freshwater Res 32 (57-63). Fipps, G. (2009). Irrigation Water Quality Standards and Salinity Management. Texas. GoM (2003). Integrated water resource development for Lake Malawi and Shire River system Lake Malawi level control stage 2- Final Feasibility Study, Ministry of Water Development, Lilongwe. GoM (2012). Project appraisal document on a proposed credit and grant to the republic of Malawi for a Shire River basin management project in the first phase of the Shire River basin management Programme, Ministry of Irrigation and Water Development. Graham, G. (1990). Testing water quality, Mississippi State University, USA. Gray, A. (1992). The Ecological impact of Estuarine Barrages, British Ecological Society (3). Hammer, J., & Linke, R. P. E. (2010). Assessments of the impacts of DAMS on the Dupage River, the Conservation Foundation report, Naperville. Jeffrey, S. W., & Humphrey, G. F. (1997). New Spectrophotometric Equations for Determining Chlorophylls a, b, c1 and c2 in higher plants, algae and natural phytoplankton. Biochem. Physiol. Pflanzen, 167, Karlsson, H., Muir, D., Teixiera, C., Burniston, D., Strachan, W., Hecky, R., Rosenberg, B. (2000). Persistent Chlorinated Pesticides in Air, Water, and Precipitation from the Lake Malawi Area, Southern Africa. Environ. Sci. Technology (34): Kirby, R., & Retie`re, C. (2009). Comparing environmental effects of Rance and Severn Barrages. Proceedings of the Institution of Civil Engineers Maritime, 162, doi: /maen Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

195 Mkanda, F. (2002). Contribution by farmers survival strategies to soil erosion in the Linthipe River Catchment: implications for biodiversity conservation in Lake Malawi/Nyasa. Biodiversity and Conservation, 11, Mumba, P. P., & Kaunda, E. (1999). Chemical Pollution in Selected reservoirs and rivers in Lilongwe. Malawi Journal of Science and Technology, 5. Patterson, D. (1999). Water Flow, Sediment Dynamics and Benthic Biology. Experimental Marine Biology and Ecology, 332, Quellette, R. J., & Rawn, D. (1997). Organic Chemistry- a Brief introduction, Amazon books, Washington. Turnbull, J. D. (1974). Effects of the Fitzroy River barrage on the salinity of the Fitzroy River. ANCOLD, Bull. No. 40, pp UNECE (1994). Standard Statistical Classification of Surface Freshwater Quality of Aquatic Life, in Readings in International Statisitics, UN Economic Commission for Europe, New York and Geneva. USEPA, U. S. E. P. A. ( 2010). USEPA Storage and Retrieval (STORET) System for Water Quality Monitoring in Florida. Washington DC: US EPA, Office of Environmental Information. Wetzel, R. G., & Likens, G. E. (2001). Limnological Analyses, Springer Science Business Media Inc, New York. WHO. (1996). Guidelines for drinking water quality-health criteria and other supporting information, World Health Organization. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

196 Appendix 1: Standard Guidelines for Flowing Water Class I; Oligotrophic, Class II; ά- mesotrophic, Class III; ß-mesotrophic, Class IV; ά-eutrophic TABLE 4. STANDARD GUIDELINES FOR WATER QUALITY FROM UNECE Variable Class I Class II Class III Class IV Class V DO (mg/l) > <3 BOD (mg/l) < >50 Phosphorus (µg/l) < >195 Total Nitrogen (µg/l) < >2500 Chlorophyll a (µg/l) < >165 Hardness (mg/l) < >80 Turbidity (NTU) < >20 Source: (UNECE, 2003) Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

197 Appendix 2: Raw Data for the Dry Season TABLE 5. Sampling Point RAW DATA FOR THE DRY SEASON Temp C ph DO (mg/l) Cond (µs/cm) Turb (NTU) TDS (mg/l) Sal (g/l) Depth (cm) S S S S S S S S S S3=upstream of Barrage; S4=downstream of Barrage; S5=upstream of ESCOM weed trapper NB: Depth denotes the depth at which water samples were taken. Sech (cm) Chl a (mg/l) BOD (mg/l) P (µg/l) Hard (mg/l) Sulph (mg/l) TSS (mg/l) N (mg/l) Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

198 Appendix 3: Raw Data for the Wet Season TABLE 6. RAW DATA FOR THE WET SEASON Sampling Point Temp C ph DO (mg/l) Cond (µs/cm) Turb (NTU) TDS (mg/l) Sal (g/l) Depth (cm) Sechi (cm) Chl a (mg/l) BOD (mg/l) P (µg/l) Hard (mg/l) Sulph (mg/l) TSS (mg/ll) N (mg/l) Coliform (CFU/100ml) S S S S S S S S S S S S S S S NB: Depth denotes the depth at which the water was sampled Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

199 Appendix 4: Pictures of the Shire River Catchment Figure 5. Shire River catchment area at the Kamuzu Barrage, used as a dumping site. Figure 6. A researcher getting set for sampling. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec

200 Figure 7. Barrage. Water sampling and on-site measurements at Sampling Point 4 at the Kamuzu Figure 8. The Shire River as a water source for domestic purposes. Independent EIA for the Upgraded Kamuzu Barrage: Final ESIA Vol 2-Water Quality, Dec