Memoire Online - An Assessment of plankton diversity as an water quality indicator in small man-made reservoirs in the Mzingwane catchment, Limpopo basin, Zimbabwe

An assessment of plankton diversity as a water
quality indicator in small man-made reservoirs in
the Mzingwane Catchment, Limpopo basin,
Zimbabwe

A thesis submitted in partial fulfillment of the requirements for the degree of Masters in Integrated Water Resources Management

AN ASSESSMENT OF PLANKTON DIVERSITY AS A WATER QUALITY INDICATOR IN SMALL MAN-MADE RESERVOIRS IN THE MZINGWANE CATCHMENT, LIMPOPO BASIN, ZIMBABWE

By
Lefranc BUSANE Basima
A thesis submitted in partial fulfillment of the requirements for the
degree of Masters in Integrated Water Resources Management

ABSTRACT

Land use changes are believed to have considerable impacts on water quality of reservoirs, which is at present a global issue considering its implications to humanity. This thesis reports on a study carried out in the southern part of Zimbabwe in the Mzingwane catchment, Limpopo basin to investigate the impacts of land and water use on the water quality and ecosystem health of eight small man-made reservoirs. Four reservoirs were located in communal lands while the remaining four were located in a National Park (Matopos) considered pristine. Plankton community structure was identified in terms of abundance and diversity as an indirect assessment of water quality and ecosystem health. In addition, phosphorus, nitrogen, pH, transparency, electroconductivity and hardness were analysed. The results obtained indicate that the communal lands' areas have not gone through major land and water use changes that impact on the quality of reservoirs since no significant difference was obtained between communal lands and the National Park in terms of plankton community (P>0.05). Though the highest phytoplankton abundance was observed in April, February showed the highest number of taxa (highest diversity). Chlorophytes was the major group in both periods with 29 genera in February and 20 in April followed by Diatoms with 17 genera in February and 12 in April. The zooplankton community was less diverse and less abundant and did not show any seasonality pattern. Normal levels of phosphorus (0.022#177;0.037 mg/l) and nitrogen (0.101#177;0.027 mg/l) were obtained and comply with the natural levels in fresh water and WHO guidelines for drinking water. The transparency of water was very low (ca. 27 cm secchi depth) in 75% of the reservoirs with some having a whitish (milky) colour with a likely effect of reducing light penetration and therefore photosynthesis in addition to low nutrient levels. The thesis concludes by acknowledging an acceptable status of the communal lands vis-à-vis the water quality and ecosystem health of reservoirs but urges water managers to continuously monitor these multipurpose reservoirs in order to be assured of their quality as well as to prevent possible detrimental land and water uses.

DECLARATION

I, Lefranc Busane Basima, declare to the senate of the University of Zimbabwe that this thesis is my original work and all other sources of material used are duly acknowledged. This work has not been submitted to any other university for any academic award.

Lefranc Busane Basima University of Zimbabwe Dept of Civil Engineering Zimbabwe

DEDICATION

To my father Busane Chentwali Lambert who guided my first steps in assiduity at work;

To my mother Buhendwa Brigini Nabintu who showed me, through perseverance and abnegation, the way of salvation and of hard work;

To Prof. Muhigwa Bahananga, J-B who taught me to think and dream big and who opened the «English world» to me;

To all those who have endured sacrifices in one way or another to make me complete this work. May you learn that «sacrifice is the great work of joy, the filial act par excellence, the act through which a profane thing becomes sacred, a lost being can find himself again, a temporal thing becomes eternal, and something soiled is consecrated».

«Take away from me the noise of your songs... but let justice roll down like waters, and righteousness like an overflowing stream». Amos 5,21-24

This work has been the fruit of the efforts of many people to whom we feel greatly indebted.

A sincere acknowledgment is made to Prof. B. Marshall from the Biology Dept. who kindly accepted to supervise this work and to generously provide equipments, manuals and other facilities in addition to his valuable guidance.

I am greatly indebted to the co-supervisor Dr A. Senzanje who made everything
possible. His great sense of humanity and understanding, availability and
disposition to assist, and his allocation of working facilities will never be forgotten.

Chhatra Mani Sharma (PhD candidate at the NORAGRIC/Sweden) and Kate Shick (Montana/USA) kindly revised the manuscripts. Prof. Muhigwa Bahananga Jean-Berckmans (from the University of Kisangani-CUB) assisted in the statistical analysis of our data. Ass. Mulongaibalu Mbalassa (MAKERERE University) helped us in the identification of macrophytes.

The fieldwork was made possible by the unconditional assistance of Lennart Woltering (TU Delft/Netherlands), Terry Marima (SSAE-UZ) and Sawunyama Tendai.

The kind assistance of David Love with accommodation at ICRISAT-Bulawayo, Mrs Elisabeth Munyoro with books and lab materials, Chief technician T. Tendayi for the physico-chemical analysis and The National Parks offices in Harare and Bulawayo in allowing us to sample in the Matopos National Park was greatly appreciated.

Many thanks to the support in any kind of Prof Muhigwa (UNIKIS-CUB), Papa Lunanga (UNIKIS-CUB), Dr Trinto Mugangu, Prof Bashwira S., Jean de Dieu Boroto (GWP-SA), Dr Kim Prochazka (IOSA-SA), Anton Earle (IWMI-SA), Ian Ford (UWC), Me Wenceslas Busane Ruhanamirindi, Bellarmin B. Ntaboba, Emmanuel B. Aganze, Sr Bertille B. Bintu, Flle Lambertine B. Lola, Flle Jacques Birugu and Solange Bashwira.

Last but not least, I am very grateful to Prof. Simbi (Dean of Faculty of Engineering), without whom this work could have never been started. His great paternal sense and understanding, allowed me to become what I am today. May you feel rewarded by this humble acknowledgement!

The CPWF Small Reservoirs Project (SRP) PN46 sponsored this thesis fieldwork while WATERnet sponsored our studies in IWRM Masters program in collaboration with UNESCO-IHE/Netherlands.

4.2 Plankton commun ity composition, divers ity and abundance in relation to
land and water use 40

Table 2.2. Some characteristics of small reservoir the studied reservoirs in the communal

lands, Insiza district, Zimbabwe. 13
Table 2.3. Rainfall pattern in Filabusi district and Sibasa village area during 2004/2005

(Insiza District) 14
Table 3.1. Littoral soil samples Color, pH and Electro conductivity (uS/cm), water colour

Table 3.4. Composition and density (ind. l^-1) of the phytoplankton in eight reservoirs

Table 3.5. Phytoplankton diversity: number of taxa recorded in the study area 25

Table 3.6. Simpson's diversity index calculated on February and April samples 26

Table 3.9. Abundance of zooplankton (no.l^-1) for the samples of February and April 29

Fig. 3.3. Zooplankton species distribution and abundance (numbers of zooplankton per litre in the right) among the 8 reservoirs/ February 30
Fig. 3.4. Zooplankton species distribution and abundance (numbers of zooplankton per

Fig. 3.9 Clustering of stations according to the overall plankton abundance/ February 34

Fig. 3.10 Clustering of stations according to the overall plankton abundance/ April 35

Fig. 3.11 Relationship between chemistry and zooplankton in the studied reservoirs 36

Fig. 3.12 Relationship between chemistry and phytoplankton in the studied reservoirs 37

Annex 1. Fig. A to F: A. Whitish color of Sibasa reservoir waters; B. Algal bloom in Chitampa reservoir; C. Sibasa reservoir; D. Mpopoma reservoir; E. Maleme reservoir dam wall; F. Dewa reservoir Error! Bookmark not defined.

Annex 2. Fig. G-H-I-J: Littoral and floating dominant vegetation in Mezilume reservoir. G: Polygonum sp., H. Numpheae sp, Cyperus sp. and Ceratophyllum sp.; I: Myriophyllum sp. and J. Numphaea sp. Error! Bookmark not defined.

Annex 3. Fig. K-P. Examples of phytoplankton taxa identified. K: Rhizosolenia; L: Anabaena; M: Ceratium; N: Pediastrum; O: Phacus; P: Micrasterias sp Error! Bookmark not defined.

Annex 4. Fig. Q-V. Examples of Zooplankton taxa identified. Q: the rotifer Brachionus sp.; R: the cladoceran Bosmina sp.; S: a cladoceran (Daphnia???); T: the rotifer Keratella sp.; U: the copepod Cyclops sp.; V: the rotifer Keratella sp. Error! Bookmark not defined.

Annex 5. Water colors influence. Analysis of variance (combiapr.sta) Marked effects are

significant at p< 0.05) Error! Bookmark not defined.
Annex 6. Influence of soil classes. Analysis of variance (combiapr.sta) Marked effects

I. Introduction

1.1 Background

Water quality is at present a global issue, especially when considering its implications to humanity in terms of water borne diseases (Ongley, 1996). The deterioration of water quality has led to the destruction of ecosystem balance, contamination and pollution of ground and surface water resources (Ongley, 1996). Very soon quality will become the principal limitation for sustainable development in many countries. The crisis of freshwater quality is predicted to have the following global dimensions according to Ongley (1996): decline in sustainable food resources due to pollution; cumulative effect of poor water resources management decisions because of inadequate water quality data in many countries; many countries can no longer manage pollution, leading to high level of aquatic pollution, escalating cost of remediation and potential loss of «creditworthiness». It is widely accepted nowadays that land use changes have a huge impact on the water quality of reservoirs. In the Great Lakes of America for example, rapid population growth, intensive industrial and agricultural activities, and sprawling urban development have resulted in significant stress to the near-shore ecosystem (Thorp et al., 1996). Anthropogenic development of Lake Kivu's shoreline in the Bukavu basin has resulted in a change of littoral aquatic biota of the area, many of the species having disappeared (Basima et al., 2005).

Water quality can be defined as an ensemble of physical, chemical and biological (including bacteriological) characteristics of the given water (Straskraba and Tundisi, 1999). Which characteristics are considered important depend on the intended use of the corresponding water; safe drinking water has to fulfil many restrictions, for example. Water quality investigations are carried out to provide information on the health of water bodies and for developing strategies that help in better management of catchment and water resources. In particular, they may assist in preparing an impact assessment, forecasting `what if' scenarios, and assessing the state of the ambient water environment and trends. The investigations may be a single study to tackle a particular water quality

related issue, or they may be an ongoing program to monitor water quality and understand long-term impacts of land uses and other activities in the catchment (Brainwood et al. 2004).

A number of techniques are available, which use the response of different biological species to assess water quality. These techniques include the South African Scoring System (SASS) developed by Dr Mark Chutter (Davis and Day, 2002), The British River Invertebrate Prediction Classification (RIVPACS), The Australian National River Health Programme (AusRivAS) (Muthimkulu, 2004) and The Nepalese Biotic Score (NEPBIOS) (Sharma, 2003). The biological assessment of surface waters started more than a century ago and consisted mainly of the analysis of the differences of organisms living in clean waters from organisms living in polluted waters (Rosenberg and Resh, 1993, Sharma and Moog, 1996 in Sharma, 2003). It is known that ecosystem functioning reflects the collective life of plants, animals, and microbes and the effect their activities-, such as feeding, growing, moving, and excreting waste, -have on the physical and chemical conditions of their environment (Naeem et al. 1999).

Plankton constitute the foundation of the food web in aquatic ecosystems and represent one of the most direct and profound responses to pollution entering reservoirs¹. Plankton is regarded as the community of plants and animals adapted to suspension in the sea or in freshwater, which is liable to passive movement by wind and current (Reynolds, 1984). These microscopic plants and animals are conveniently segregated into the terms «phytoplankton» and «zooplankton» respectively, though there are differences in opinion where the dividing line is drawn (Cander-Lund and Lund, 1995). The plankton proliferation is greatly affected by the water quality (Schindler, 1978) and the predatorprey relationships in the reservoir (Arcifa et al., 1986). Water quality is in turn affected by land use and water sources. Brainwood et al. (2004) presented evidence to suggest that within reservoirs, chemical trends are strongly linked with the differing water sources. Trends were evidenced by quite distinct patterns of water chemistry (Brainwood et al. 2004), which are related to nutrient inputs.

The great majority of Zimbabwean rural population lives in areas where the mean annual rainfall is below 80 mm and extremely erratic. Soils in this region tend to be infertile sands to sandy loams (Moyo, 1995; Grant, 1981). This unfortunate status makes small, man-made water reservoirs important in Zimbabwe. Their importance is indicated by the fact that 7,000 such reservoirs have been constructed since independence in response to recurrent droughts (Senzanje and Chimbari, 2002). Reservoirs are classified in Zimbabwe as small, medium, large or major in consideration of the maximum height above cleared foundation level and gross capacity (Kabell, 1986). Kabell (1986) defines small reservoirs as one below 106 m³ of capacity and a dam wall height below 8 m. Straskraba and Tundisi (1999) classify reservoirs according to their size capacities. A reservoir was considered small if it had an area of 1 to 100 km² and a capacity of 1 million to 100 million m³ (10⁶-10⁸ m³). Straskraba and Tundisi (1999) call very small reservoirs those of less than one million m³ of capacity and an area of less than 1 km². Small reservoirs in Zimbabwe are often neglected though they support a number of activities in many parts of the country. This is shown by the fact that very little information exist on the capacities, year of construction, water storage capacity, depth of water, sedimentation and most of all about their ecosystems health. Water quality is not monitored and it is difficult to state whether the quality is acceptable or not with regard to the intended uses.

The small reservoirs that are the object in this study are of a size less than three million cubic meters (3 x 106 m³⁾ of capacity and a maximum height of eight (8) m. These reservoirs are multipurpose in use and the main functions include: water supply for rural people and their livestock, brick making; small-scale irrigation, gardening, recreation and tourism, focal point for rural community, small-scale fisheries, wildlife, aquatic life, and microclimate control (Senzanje and Chimbari, 2002). The long-term effects of land use dynamics on water quality, and their potential harm to ecosystem integrity has often been neglected during the planning of small reservoirs. There is also limited information on the small reservoirs in terms of biological diversity and responses of organisms to land and water use factors. Nhiwatiwa (2004) concluded there is a historical bias in limnological investigations towards larger water bodies like Lake Kariba that resulted in small reservoirs not being studied. Studies on these water bodies in Zimbabwe are mainly

focused on improving water availability for sustainable food production and rural development and are mainly concerned by sedimentation and siltation problems (Lawrence and Hasnip, 2004; Lawrence and Lo Cascio, 2004a, Lawrence and Lo Cascio, 2004b, Lawrence and Lo Cascio, 2004c, HR Wallingford, 2004), and increasing productive water use (Mugabe et al. 2003). An inventory of small reservoirs in Zimbabwe has been done by Senzanje and Chimbari (2002) and was mainly concerned with ways to rehabilitate or build new reservoirs, or to assist communities in developing sustainable management plans/strategies for the catchment area of their reservoirs. However, the limnology and ecology of these small reservoirs was not addressed. Historical profiles of small reservoirs in communal areas have been presented by Zirebwa and Twomlow (1999). A comparison of organochlorine pesticide residues in Upper Ncema and Lower Umguza Dams (Zimbabwe) was done by Siwela et al. (1996). DDT and other pesticides originating from farms were found in fish species. These pesticides might have detrimental consequences to aquatic organisms and to a certain extent to humans who feed on them. On the few works done on plankton in Zimbabwe, Thornton and cotterill (1978) provided a list of phytoplankton and zooplankton species occurring in five small reservoirs in the Eastern Highlands, while Green (1990) made a list of zooplankton in eighteen small reservoirs in Marondera and Nyanga. Nhiwatiwa (2004) is among the few authors who studied the limnology and ecology of two small reservoirs in Zimbabwe. The two small reservoirs were built in series on the Munyahuku River. He concluded that very little is known about the ecology of such small water bodies, and that further investigation into this topic is warranted.

Because of this lack of basic ecological knowledge, and the potential for water quality to change as a result of anthropogenic activities, it is important to ecologically assess small reservoirs that exist in rural Zimbabwe. Certain knowledge of the responses of biota to changes in water quality could constitute an important tool to be used by water managers in Zimbabwe to continually and rapidly asses s the quality of waters that they are managing. The present work will focus on an assessment of zooplankton and phytoplankton diversity as a well as some water quality parameters in relation to land and water use on eight selected small reservoirs in the semi arid region of Zimbabwe.

The variability in consideration of reservoirs is highly dependent on the many purposes for which they have been built, as there is some correspondence between their features and their uses (Straskraba and Tundisi, 1999). Small reservoirs differ from large ones in that there are usually no strict rules and regulations guiding access and usage of the water resource. Also they have a semi permanent character in terms of water content and tend to dry out during part of the year (Wotro, 2004).

The importance of the ecological status of surface water systems have been recognized and have contributed to our understanding of the health of our systems². The biological state of health of small reservoirs is to a large extent a function of their use and the land use in the reservoir catchment area. Small reservoirs are known to cause some negative environmental and health impacts; they create or enhance ecological environments that are favourable to the proliferation of vector borne diseases like malaria and schistosomiasis (Hunter et al., 1982; Oomen et al., 1994; Bolton, 1994) and water related diseases. It is believed that different nutrient inputs are flowing to small reservoirs in such areas. Since nutrients are the major support of the proliferation of the phytoplankton community, it is interesting to know how the discharges from different activities affect the plankton community structure. National Parks are considered to have conditions near to pristine and have been selected as control sites while doing this investigation. This study may have important implications on the integrity of ecosystems and therefore on the integrity of the catchment as a whole. The major activities that take place in the study area will be identified for the characterization of the reservoirs and for effective comparisons. The results of the study may show some water resources management implications and might provide guidance for better management of water resources with due respect to land use activities. Findings of this investigation will enable policy and decision makers to find a way forward in the management plans and actions in areas that are having similar characteristics.

Water quality management is important to water managers and decision makers who
often need an understanding of the interactions in small water bodies ecosystems as well

as short term corrective solutions. The short-term solutions aim at diminishing difficulties and long-term solutions that aim to prevent problem creation. The most important management items are summarized in Straskraba and Tundisi (1999). These management items include: - negligence of interrelations of problems might create new unexpected ones; - planning ahead is a step toward success; - responsibility to our children and grand-children mandate fundamental attention to sustainable development; - consideration of biogeophysical, economic and social aspects is fundamental; - environmental impact assessment procedures are useful decision-making tools; - monitoring is an important decision-making tool; - most reservoirs are or eventually become multipurpose that is the basis for conflicts, resolution of these conflicts is enabled by joint participation of the respective parties. The reservoir water quality management publication (Straskraba and Tundisi, 1999) shows managers the importance of a whole system approach and respect to water quality problems. It further shows water quality engineers the need for more advanced theories and inclusion of biological considerations in management decisions. More, it shows limnologists the strength of the whole system approach and the need to include catchments and considerations of human activities including socio-economic and political aspects.

It is important to note that very few investigations have been done on small reservoirs in Zimbabwe, particularly on biological assessment of the integrity of these fragile ecosystems (Twomlow, pers.comm.). This investigation will be therefore a contribution to the understanding of the biology and health of small reservoirs as well and will provide a starting point to future researchers in their management.

1.3 Hypotheses

The main hypothesis assumed is the following: Land and water use activities have negative influence on the water quality of small man made reservoirs in Zimbabwe.

column on communal lands and a less turbid and nutrient rich water column in National Parks).

2. Differences in the physical and chemical characteristics of reservoirs between communal lands and National Parks result in ecological differences in plankton communities (plankton found in reservoirs on communal lands is expected to be more abundant but less diverse, while plankton in reservoirs in National Parks should be more diverse).

3. There are no seasonal changes in the plankton community structure of the small reservoirs.

1.4 Objectives

The main objective of this study is to investigate the impacts of land and water use activities on the water quality of small man-made reservoirs. This will be achieved by studying the plankton community structure as an indirect assessment of the water quality. Therefore, the present study focuses on the following specific objectives:

II. Materials and Methods

2.1 Study area

This study was carried out in the southern part of Zimbabwe (Fig. 2.2) in the Mzingwane catchment, within the Limpopo basin. The Limpopo basin (Fig.2.1) covers up to 1.3 % of the African continent. The Mzingwane catchment is one of the seven catchment areas that have been demarcated by the new water legislation of 1998. The six other catchments are Gwayi, Sanyati, Manyame, Mazoe, Save and Runde. The Mzingwane catchment is located in a semi arid region that is characterized by highly variable rainfall concentrated in one rainy season, separated by relatively long dry seasons (Lawrence and Hasnip, 2004). The Mzingwane catchment is divided into four sub catchments, which are Upper Mzingwane, Lower Mzingwane, Mwenezi and Shashe (Fig. 2.3). The annual rainfall in the Mzingwane catchment ranges from 300mm in the south to 635mm in the north (Table 2.1). Temporal distribution of rainfall follows the general pattern of southern African region. The coldest month (July) when minimum temperature goes below 0^oC as well as the hottest month (October) are both comprised in the dry months that go from April to October. The hottest areas are found around Beitbridge (Conley, 1996 in Mwenge, 2004). The Mzingwane catchment has three major rivers: the Shashe, Umzingwane and Mwenezi. These rivers flow in the southeastern direction into the Limpopo River carrying and depositing sediments along their courses. There is almost no flow in the dry season and riverbeds are sandy alluvial of considerable thickness providing big storage of water³. One of the major features of aridity of the catchment is that the mean annual potential evapotranspiration rates are higher than the mean annual rainfall (1800mm as compared to 465mm). This status indicates low runoff generation and aquifer recharge (Mwenge, 2004).

The Limpopo River in southeastern Africa, rising as the Krokodil River in the mountainous Witwatersrand region of northern South Africa, is a trans-boundary river shared by Botswana, South Africa, Mozambique and Zimbabwe. The Krokodil River then becomes the Limpopo and continues northeast along the border between Botswana and South Africa, then east along the border between Zimbabwe and South Africa. It flows southeast through southern Mozambique before entering the Indian Ocean near Xai-Xai. The Limpopo river is one of the most anthropogenically impacted Rivers in the Southern African region and has vast tracks of its length covered by sand and silt and seasonal flow only during the rainy season which spans from November to April.

The eight selected small reservoirs are located in the Mzingwane catchment, with four in the Matobo (Matopos) National Park and four in the Insiza district around Avoca Business centre (Fig.2.4). The National Park is located 34 km south of Bulawayo, at 28^o Longitude East and 20^o Latitude South. It represents one of the major tourist attractions in Zimbabwe. It contains majestic granite scenery; the landscape has been carved out from an almost flat surface of granite by millions of years of weathering, resulting in great «whalebacks» and domes, and castle-like formations (Tregold, 1996). The vegetation in the park is quite different from that of the surrounding countryside, and supports many species of animals and prolific bird life. The park is comprised of a game reserve and a recreational park, both of which contain a number of small reservoirs. The recreational park (western part) is separated from the game reserve (eastern part) by the main road Bulawayo-Kezi. Mpopoma and Chitampa reservoirs, in the game reserve, are respectively located at 7 and 2 km from the entrance gate. In the recreational park, Mezilume reservoir is located at approximately 5 km from the game park entrance gate while Maleme is at 7 km. Mezilume and Maleme are located in the so-called Central Wild Area while Chitampa and Mpopoma are located in the Whovi Wild Area (Game Park). Fish are abundant in most of these reservoirs. Avoca Business Centre (in Insiza district) is located at 127 km east of Bulawayo and its geographic coordinates are 20^o 47' latitude south and 29^o 31'longitude east. Fig 2.3 and 2.4 show the location of selected small reservoirs.

A summary of the studied reservoirs located in communal lands is presented in Table 2.2. The main users of the reservoir are farmers located mainly downstream of the dam. Sibasa is an old perennial reservoir known to have survived the drought period being recharged by groundwater.

Table 2.2. Some characteristics of small reservoir the studied reservoirs in the communal lands, Insiza district, Zimbabwe.

2.1.1 Land and water use in the communal Lands

Makoshe reservoir is located in a relatively flat area and its water occupies a large surface. Many standing trees are still present in the reservoir showing that the area was vegetated before the construction of the reservoir. The substrate is mainly composed of dark grayish sand (Table 3.2) (expansive and slippery since mixed with clay) and the littoral land fairly covered by vegetation. The area may be subject to erosion in time of heavy rainfalls. Big rocks are located mainly along the dam wall and some are scattered in the reservoir. Water from the reservoir is allocated to the following uses: irrigation, domestic, livestock, fishing, garden and drip irrigation. Drip irrigation is pumping water for 18 ha plot for some beneficiaries involved in the construction of the reservoir. Approximately 200 households that belong to two villages, Bambanani-Makoshe and Mbawulo, utilize Makoshe reservoir's water. There are no farms upstream of the reservoir, all the gardens and farms being located downstream of the reservoir. A number of birds were observed belonging to the genus Phalacrocorax, Ceryle, Ceyx, Hirundo and Nectarinia.

The shoreline vegetation of Sibasa reservoir is dominated by Cyperus spp and the Water Clover Marsilea vestita. The water is having a whitish (milky) colour. The reservoir is located at 200 m to the first homestead of Sibasa village that is slightly elevated compared to the reservoir. The reservoir is mainly allocated to livestock use, though it is used as well for domestic purposes and recreation (fishing). No farm is located in the close vicinity of the reservoir. A white granite hill is located upstream of the reservoir. Some trees are found in the surroundings of the reservoir as well as some rocks, which are also part of the substrates in the reservoir. The area has suffered from a shortage of rainfall the whole period except January where 211 mm of rain where recorded (Table 2.3). Few birds belonging to Phalacrocorax sp., and Ceryle rudis were observed actively fishing in the reservoir.

Table 2.3. Rainfall pattern in Filabusi district and Sibasa village area during 2004/2005 (Insiza District)

Source: Filabusi District Administration and Meteorological Office in Harare, Zimbabwe

Denje reservoir is mainly located in a rocky and slightly slanted area with a low vegetation cover in its littoral zone (score 2 in Table 3.1). The littoral vegetation includes Cyperus spp, Typha spp. and Phragmites spp. A communal garden is situated about 5 meters away from the reservoir. The garden uses manure as a fertilizer. The littoral soils have a yellowish brown colour and the reservoir water has a whitish brown colour. The highest soil pH value (7.8) of the communal reservoirs studied was recorded at Denje. A small business centre is located upstream of the reservoir at approximately 200 m. This Denje business centre as well as the population of Denje village gets water from the reservoir.

Cyperus spp. is the main constituent of the littoral vegetation at Dewa reservoir with
some areas also covered by Polygonum spp. From previously green during the first
sampling (February), the vegetation at the shore and areas surrounding the reservoir has

decayed. The colour of the waters is gray whitish brown while the colour of dry soils is grayish brown (Table 3.1). The littoral zone is full of dung making reference to the considerable number of cattle (10 heads per 10 m² of littoral area) that is ever grazing in the area. The cattle observed were composed of cows, donkeys, goats and sheep. A small communal garden is located in the littoral zone area at approximately 5m from the water level. Manure is the main, if not the sole, type of fertilizer used in the area. The reservoir' s water is mainly used for cattle and domestic purposes. Some points of cloth washing are also located in the same area. Few fishermen were observed on the reservoir. No bird was observed during the two sampling visits. No human settlement is present upstream of the reservoir that is mostly dominated by rocky hills.

2.1.2 Land and water use in the National Park

Most of the reservoirs' shoreline in the National Park is very well covered by vegetation and the reservoirs are surrounded by granite hills. The rainfall pattern in the Matopos area during 2004-05 is presented in Table 3.3.

Chitampa is the only reservoir that was not very well covered by vegetation in addition to its location in a sloppy area. The main vegetation around this reservoir is composed of Phragmites sp., and Typha sp. The waters have a very brownish colour; very similar to the colour of the dry soils that is light brownish gray (Table 3.1). The pH of the soils is 6.6 and the vegetation score is 2 as shown in Table 3.1. An algal bloom was observed close to the shore (see Fig. B, annex 1). Two hippopotamus were also observed in the reservoir.

Maleme reservoir has a littoral zone covered by Cyperus spp. and short vegetation like Polygonum spp. that occupies areas where the levels of water have receded. It has a concrete dam wall in a rocky area (Fig. E, annex 1). The soils of the shoreline have a pH of 5 and grayish brown colour. The reservoir waters were clear like normal drinking water.

Mezilume reservoir is very thickly covered by vegetation dominated by Cyperus spp,
Phragmites sp., Typha sp., Numphaea sp., Myriophyllum sp., Polygonum spp. Some of

the littoral vegetation is shown in the annexes (Fig G to J, Annex 2). The reservoir is surrounded by four rocky granite hills. The reservoir water has a coffee colour. The water level had not dropped since the first sampling trip, as it was the case at Maleme.

Mpopoma reservoir has a large surface area with clear water and very good and thick vegetation covers its shoreline. Litter is easily observed in the littoral areas from the falling vegetation. Many granite rocks are found in and around reservoirs.

2.2 Data collection

2.2.1 Plankton samples

Zooplankton samples were collected with a zooplankton net of 40 cm diameter and 62um mesh size while phytoplankton samples were collected using similar nets of 20um mesh size. The samples were collected using a standardized method presented in Edmondson and Winberg (1971). The concentrated samples were collected in small 130 ml bottles that were labelled. Four samples were collected on each reservoir at a horizontal line situated at 10 to 20 m facing the dam wall for zooplankton and four for phytoplankton. A preservation solution of 4% formalin was added to the sample bottles of zooplankton and Lugol solution was added to the bottles containing phytoplankton for fixing purposes. The samples were then taken to the fish laboratory of the Biological Sciences Department of the University of Zimbabwe. Taxa were then identified and counted under an inverted microscope OLYMPUS CK40 and species pictures were taken using a digital camera NIKON model E995 mounted on the inverted microscope. The identification of taxa was done using a dichotomic identification keys presented in Durand and Lévêque (1980) supplemented by Fernando (2002) and Cander-Lund and Lund (1995).

Transparency was measured on all the selected reservoirs using a secchi disk.
Transparency of waters is linked to light attenuation in reservoirs, which impacts the
photosynthetic potential of primary producers and consequently impact the whole biotic

composition of a reservoir (Hart, 1990). The depth of a reservoir influences its water quality. Of particular importance is the depth relative to the surface area and wind intensity because these factors effect the intensity of mixing in the reservoir (Straskraba and Tundisi, 1999). Water quality is therefore related to reservoir depth, size and basin morphology. Therefore, the depth and morphological characteristics of the reservoirs were described. A stadia rod was used to measure the depth. The slope of the dam was observed and recorded as steep or gentle.

Soil samples of the substrate of the reservoirs were collected and brought to the Soil Science and Agricultural Engineering laboratory of the University of Zimbabwe for analysis of pH, electroconductivity, texture and colour. A Jenway pH meter Model 3510 (ESSEX) was used for pH analysis of soils. A 1:1 soil solution (which is deionised water) ratio was used. A conductivity meter (Ecoscan Con5) was used for electroconductivity analysis. The texture of soils was analysed by the Boyocous hydrometer method. Rainfall data for Filabusi and Sibasa areas was collected at Filabusi District Administration meteorology office while the rainfall data for the National Park was collected from the Meteorology office in Harare. The colour of soil samples was analysed using the Munsell soil colour charts (Munsell, 1975).

The vegetation cover surrounding reservoirs was observed and estimated as abundance scores of 1 to 4 (1 means no vegetation coverage; 2: poor coverage; 3: good coverage and 4: very good coverage). Presence or absence of farms upstream of reservoirs was noted as well as the proximity of homestead upstream and downstream the reservoirs. The activities (anthropogenic) taking place in the vicinity of the reservoirs were also investigated. A digital camera CAMEDIA C120 was used to take pictures of the activities in the study area as well as of the main vegetation cover and soils to facilitate their identification.

2.2.3 Water samples

Water samples were collected using a Ruttner' s bottle at 0.5 m depth from the surface at the four sampling sites located in face of the dams' wall. These samples were immediately placed into a cooler box and kept at low temperatures using ice blocks pending their transportation to a deep freezer in the laboratory. The following water quality parameters were analysed after the samples had been brought to the laboratory: Nitrogen, Phosphorus, pH, electric conductivity (EC) and total hardness. The MuphyRiley Method (ascorbic acid method) was used for total phosphorus analysis using UV visible spectrophotometer Spectronic 21 Bausch and Lomb. Total nitrogen was analysed using the titrimetric method using 0.01 M HCl (Hydrogen chloride). The solution was made alkaline by MgO (Magnesium oxide) and Dervada alloy. Electroconductivity and pH were analysed using the equipment described in section 2.2.2.

Electroconductivity and total hardness are considered because they might be related to soil composition and exchanges between soil and small reservoir waters. There is abundant literature that stresses the importance of phosphorus and nitrogen in the shaping of the structure and abundance of phytoplankton in reservoirs (Crawley, 2000; Talling & Lemoalle, 1998; Lemoalle et al. 1981; Pinel-Alloul et al. 1995; Schindler, 1978; Drenner, 1989). Nitrogen and phosphorus are considered because they are major factors that limit primary production of phytoplankton in reservoirs (Straskraba and Tundisi, 1999). Nutrient-rich animal excrement deposited along and within reservoirs constitute a major input of nitrogen and phosphorus in these areas. Hippos are present in Mpopoma and Chitampa reservoirs. The cattle found around Denje and Dewa reservoirs might have similar effects, though dung is mainly in the shoreline of reservoirs and might be transported to the waters indirectly by rains.

The supply of nitrogen is known to be a key factor controlling the nature and diversity of plant life, the population of both grazing animals and their predators, and vital ecological processes such as plant productivity and the cycling of carbon and soil minerals (Vitousek et al., 1997).

2.3 Data analysis

Clusters were used for the statistical analysis of the data. The eight reservoirs were grouped in terms of species composition and abundance. Zooplankton and phytoplankton taxa were also grouped as they co-occurred or not in the studied reservoirs. Correlations between all means including the abiotic data were computed. Graphs were drawn and analyses were done using SigmaPlot, SPSS and Statistica softwares.

A Simpson's index of diversity (a measure of diversity which takes account of species richness and evenness⁴, which is 1-D) was calculated using the following formula for D: D = ~ (n/N)² where: n = the total number of organisms of a particular species; N = the total number of organisms of all species. The value of this index ranges between 0 and 1. The greater the value, the greater the sample diversity. Spearman' s rank correlations were calculated.

III. Results

3.1 Land and water use in the area

The characteristics of the littoral and surrounding areas of reservoirs are presented in the Table 3.1. These characteristics are mostly grouped in terms of littoral soil quality (pH, electroconductivity, hardness) and colour. Electroconductivity of the soils is of an average value of 420 jiS/cm and is evenly distributed in the whole study area. On average electroconductivity had a value of 394 jiS/cm in the communal lands as compared to 450 jiS/cm in the National Park. However a high value of 730 jiS/cm was noticed in Mpopoma. The pH of the soils in the National Park is slightly acidic 5-6.6 with an average of 5.5 while communal lands are comprised of alkaline soils ranging from 7.5 to 7.8 with an average of 7.6. Vegetation scores ranged from 2 to 4, and tended to be higher at the reservoirs in the National Park as compared to communal lands. Water colour is quite different from a reservoir to another with a predominance of a whitish colour in the communal lands. This whitish colour was not observed in the National Park. Mpopoma had a quite distinct brownish colour.

Table 3.1. Littoral soil samples Color, pH and Electro conductivity (uS/cm), water colour and littoral vegetation score (estimated)

*For the estimation of vegetation abundance, 1 means no vegetation coverage; 2: poor coverage; 3: good coverage and 4: very good coverage.

The classification of the soils using their texture is presented in Table 3.2. Maleme and Mezilume have loamy sand while Mpopoma, Sibasa and Makoshe have loamy sand and sand is predominating around Dewa, Denje and Chitampa. The soil texture is similar in the communal lands and National Park with Mpopoma, Sibasa and Makoshe having loamy sandy soils; Chitampa, Dewa and Denje having sandy soils.

3.2 Water quality

The overall pH was slightly alkaline (7.6-8.5), the electrical conductivity was lower than 200 jiS/cm while the hardness ranged from 23 to 104 mg/l and total nitrogen as well as total phosphorus had values respectively ca. 0.1 mg/l and 0.01 mg/l (Table 3.3). pH readings seem to vary seasonally in communal lands' reservoirs more than in those located in the National Park. Conductivity looks also generally higher in the communal lands. The transparency of water measured by the secchi depth was higher in the National Park as compared to the communal lands. However, the lowest transparency was also recorded in Chitampa reservoir in the National Park.

Parameter	Communal lands				National Park
Parameter	Sibasa	Dewa	Denje	Makoshe	Maleme	Mezilume Mpopoma		Chitampa
pH FEB	7.6	8.1	7.7	7.7	8.5	7.9	8.0	7.6
APR	8.2	8.5	8.4	8.4	8.4	7.6	8.3	7.7
EC (jiS/cm) FEB	107	200	200	157	147	76	109	93
APR	111	250	183	198	138	102	110	109
Secchi depth (m)	0.2	0.3	0.2	0.3	0.5	1.7	2	0.1
Total Hardness
(jig/L)	42	103.8	58.4	83.5	46.8	22.5	36	35.5
Total Nitrogen
(jig/L)	94	87	84	99	8	12	11	14
TP (jig/L) FEB	7	11	6	6	4	3	3	7
APR	8	17	38	5	4	59	2	34

Chemical parameters' abbreviations: Total phosphorus ( TP), electric conductivity (EC).

3.3 Plankton community composition

3.3.1 Phytoplankton

The flora identified and counted in February and April 2005 is presented in Table 3.4. The taxon that got the highest abundance was Hydrodictyon spp., which accounted for an average of 30% of the overall phytoplankton sampled in April. Hydrodictyon was followed by Anabaena (19.9%), Peridinium (15.7%), and Melosira (11.7%), all sampled in April. Hydrodictyon was very rare during the first sampling (February) period accounting for around 0.1% of the phytoplankton sampled in April. The phytoplankton sampled in February showed an abundance of Melosira (18.7%) followed by Ceratium hirundinella (17.3%) and Pinnularia (11.9%). It can be noticed as well from Table 3.4 that the taxa collected in April were much more abundant than that of February with a percentage of 84 in April against 15.8 in April. Though the highest abundance was observed in April, February showed the highest number of taxa (highest diversity). Chlorophytes was the major group in both periods with 29 genera in February and 20 in April. Chlorophytes was followed by bacillariophytes (diatoms) with 17 genera observed in February against 12 in April. Cyanophytes (5 genera), euglenophytes (4 genera), Fungi (3 genera), dinophytes (2 genera), xanthophytes (1 genra) and canophytes (2 genera) were also observed in February. Cyanophytes (4 genera), euglenophytes (2 genera) and Fungi were observed in addition to chlorophytes and bacillariophytes.

Table 3.4. Composition and density (ind. l^-1) of the phytoplankton in eight reservoirs located on communal lands and National Park in rural Zimbabwe.

Class	Taxa		February			April
Class	Taxa	Communal lands	National Park	Abundance	Communal lands	National Park	Abundance
Chlorophyta	Volvox	128	0	128	0	0	0
	Amscottia	304	616	921	21	0	21
	Sphaerocystis	166	154	320	176	122	298
	Dictyosphaerium	47	0	47	0	0	0
	Micractinium	54	26	80	0	0	0
	Scenedesmus	14	31	46	98	5	104
	Staurodesmus	13	173	186	5	248	254
	Ankistrodesmus	5	0	5	10	78	88
	Pediastrum	31	73	105	2562	28	2590
	Sorastrum	2	0	2	0	0	0
	Haematococcus	0	647	647	0	0	0
	cladophora	4	12	16	0	0	0
	Staurastrum	10	1252	1262	1371	2264	3636
	Unidentified2	0	1161	1161	0	0	0
	Unidentified1	0	864	864	0	0	0
	Unidentitfied3	0	263	263	0	0	0
	Cosmarium	8	441	449	13	186	199
	Euastrum	21	94	116	0	0	0
	Sphaerozoma	0	78	78	0	93	93
	Xanthidium	0	93	93	0	0	0
	Actinastrum	0	462	462	0	0	0
	Arthrodesmus	0	38	38	0	26	26
	Hydrodictyon	0	27	27	0	27221	27221
	Closterium	4	2	5	132	23	155
	Selenastrum	0	13	13	0	0	0
	Spondylosium	0	73	73	0	0	0
	Onynchonema	0	26	26	0	0	0
	Pleurotaenium	0	28	28	0	1082	1082
	Micrasterias	32	96	128	0	52	52
	Indet1	0	0	0	0	233	233
	Onynchonema	0	0	0	0	62	62
	Penium	0	0	0	0	16	16
	Zygnema	0	0	0	0	383	383
	Spirogyra	0	0	0	0	16	16
	Cylindrocystis	0	0	0	0	10	10
Bacillariophyta	Navicula	142	825	968	574	717	1291
	Surirella	53	5	58	47	0	47
	Melosira	3055	2310	5365	7923	2585	10508
	Achnantes	0	61	61	0	127	127
	Asterionella	0	3	3	0	0	0
	Cymatopleura	0	53	53	0	0	0
	Rhopalodia	0	60	60	0	0	0
	Oscillatoria	35	28	62	0	0	0
	Gomphosphaerium	0	28	28	0	0	0
	Gomphonema	6	0	6	0	0	0
	Fragilaria	4	0	4	0	31	31
	Rhizosolenia	38	125	162	147	414	562
	Synedra	0	1430	1430	78	72	150
	Pinnularia	0	3397	3397	5	427	432
	Stephanodiscus	0	2	2	0	0	0
	Amphiprora	0	4	4	0	0	0
	Cymbella	2	0	2	0	194	194
	Gyrosigma	0	0	0	18	0	18
Cyanophyta	Coelosphaerium	800	306	1106	0	0	0
	Microcystis	153	101	254	0	0	0
	Microchaete	4	0	4	0	0	0
	Merismopedia	4	0	4	0	0	0
	Anabaena	48	33	80	17916	0	17916
Canophyta	Nostoc	93	0	93	0	0	0
Euglenophyta	Trachelomonas	0	21	21	0	0	0
	Phacus	116	427	543	536	585	1120
	Euglena	115	7	122	163	18	181
	Astasia	150	0	150	0	0	0
Fungi	Rhizosiphon	0	2	2	0	0	0
	Chytridium	24	96	120	0	0	0
	Sporangium	441	308	749	16	0	16
Dinophyta	Ceratium	4945	0	4945	6151	18	6169
	Peridinium	1202	2	1204	14084	5	14089
Xanthophyta	Ophiocytium	3	0	3	0	0	0
Cryptophyta	Cryptomonas	0	11	11	0	0	0
Chrysophyta	Dinobryon	0	0	0	0	471	471
Total		12276	16386	28662	52046	37812	89859

There was no significant difference in phytoplankton species composition in February and April (Spearman's rank correlation coefficient r_s = 0.203, N=71). The Student's t-test did not show any significant difference between February and April' s phytoplankton species (t = -1.71; P= 0.087; N = 71).

Reservoirs in the National Park were more diversified in taxa compared to those in the communal lands, with 49 taxa against 38 sampled in February and 32 against 22 taxa identified in the samples of April (Table 3.5). This difference of diversity in the National Park compared to the communal lands is confirmed by the significant difference obtained using the paired samples Student' s t-test (t = 21.0; df = 1; P = 0.03).

Table 3.5 shows, as well, that phytoplankton communities were more diverse in February on both communal lands and within the National Park. A significant difference in species diversity was found between the two study periods, using the Student' s t-test (t = 33.0; df=1; P = 0.019).

A significant correlation was found between the phytoplankton diversity recorded in communal lands and in the National Park, and between February and April, with Spearman's rank correlation coefficient (r_s =1.000**, n=2, with significance at 0.01 level (2-tailed)).

The Simpson's index of diversity is presented in Table 3.6. The highest as well as the lowest indices were recorded in the National Park respectively in February and in April. The diversity in communal lands was consistently high.

Though the chlorophytes division was more diverse in February, it only constituted 7 % of taxa abundance in communal lands (Table 3.6). However it constituted 41% in the National Park.

Table 3.7, Table 3.8, Fig. 3.1 and Fig 3.2 highlight the difference in the phytoplankton composition and abundance in communal lands and in the National Park and for February and April samples. Phytoplankton abundance was dominated in February by dinophytes (50 %) followed by bacillariophytes or Diatoms (27%) in communal lands. In the National Park bacillariophytes were more abundant (5 1%) (Table 3.7, Fig.3.1). Sibasa reservoir dominated the February abundance of taxa in communal lands with 54% followed by Denje reservoir (3 1.5%) while Mpopoma dominated in the National Park with 53% followed by Mezilume and Maleme with respectively 24% and 17% of the total abundance (Table 3.7).

Class		Reservoirs			Communal Lands total %		Reservoirs			National Park total %
Class	Denj Sibasa e		Dewa	Makos he	Communal Lands total %	Male me	Mezilu Chita me mpa		Mpop o ma	National Park total %
Chlorophyta	82	548	25	189	845 6.9	179	3224	495	2847	6744 41.2
Bacillariophyta	34	2810	180	311	3334 27.2	2179	548	48	5553	8329 50.8
Cyanophyta	10	152	689	157	1008 8.2	0	111	195	134	440 2.7
Canophyta	0	53	0	40	93 0.8	0	0	0	0	0 0.0
Euglenophyta	342	20	3	15	381 3.1	330	58	45	22	454 2.8
Fungi	74	221	132	37	465 3.8	162	20	91	133	406 2.5
Dinophyta	6079	63	0	5	6147 50.1	2	0	0	0	2 0.0
Xanthophyta	3	0	0	0	3 0.0	0	0	0	0	0 0.0
Cryptophyta	0	0	0	0	0 0.0	0	0	0	11	11 0.1
Total	6624	3868	1030	754	12277 100.0	2852	3961	874	8700	16386 100
%	54.0	31.5	8.4	6.1	100	17.4	24.2	5.3	53.1	100

April samples were dominated by dinophytes (38.9%) and cyanophytes (34.4%) in the communal lands and by chlorophytes (85%) in the National Park (Table 3.8). Dinophytes and cyanophytes were rare in the National Park. The second more abundant group in the National Park was bacillariophytes (Diatoms) with 12 % of the total abundance. As for February, Sibasa reservoir recorded the second highest abundance of taxa 41% of total abundance in the communal lands after Makoshe reservoir (43 %). Mezilume (83.61%) dominated the overall abundance in the National Park followed by Mpopoma (13.2%). It is important to note that In February Mpopoma dominated over Mezilume.

A positive Spearman's rank correlation coefficient was found between the phytoplankton major groups in communal lands and the National Park (r_s=0.477; P=0.0194; N=9. The paired t-test gave a non-significant difference (t=-0.39; P=0.70; df=8).

The major groups in February obtained a Simpson's index of diversity of 0.77 in the communal lands and 0.57 in the National Park. The Simpson's diversity index for April samples showed high diversity in the communal lands (0.3 1) while the National Park obtained low diversity (0.74).

The Spearman' s rank correlation between communal lands and National Park was not significant at 5 % level (P=0.05) with r_s=0.072 and N=7. No significant difference in the phytoplankton major groups was found between communal lands and the National Park in April samples using t-test: t=0.34; P=0.74; df=6.

There was a significant difference in abundance between February and April phytoplankton samples in the communal lands using the t-test (t=-2.06; P=0.05; df=14) but no significant difference was found in the National Park (t=-0.86; P=0.39; df=14). However, there was high significant difference in phytoplankton abundance between the communal lands and the National park using a chi-square test (P<0.01).

3.1.2 Zooplankton

The zooplankton community investigated was represented by the freshwater common groups, the crustacean cladocerans and copepods (Cyclops and Calanoids) and rotifers. A few individuals belonging to the Ostracods group were also recorded. The zooplankton community was dominated in February by copepods (Table 3.9); Cyclopes having 28.6 % followed by their youngsters (Nauplii) with 15.2%, the rotifer Keratella (14.2%) and copepod Calanoids nauplii (13%). Communal lands had the highest zooplankton abundance in both February and April samples with respectively 63% and 57%. The Cyclopes were again dominant in April with 27 % of the total abundance followed their nauplii, a cladoceran species, the rotifers Keratella and Brachionus all getting ca. 10%.

Table 3.9. Abundance of zooplankton (no.l^-1) for the samples of February and April

Taxa (February)	Communal lands National Park		Abundance	%
Daphnia	109	18	127	6.3
Moina	68	67	134	6.6
Ceriodaphnia	14	0	14	0.7
Simocephalus	2	0	2	0.1
Bosmina	11	2	13	0.6
Chydorus	3	24	27	1.3
Cyclops	447	133	580	28.6
calanoida	59	5	64	3.1
nauplius calanoida	157	105	262	12.9
nauplius cyclopoida	123	185	309	15.2
Brachionus	32	44	76	3.7
Keratella	235	50	286	14.1
A (long avec queue)	26	84	110	5.4
X (Rotifera)	0	26	26	1.3
Leptodora	0	2	2	0.1
Total	1286	745	2031
%	63.3	36.7
Taxa (April)
Daphnia	34	2	36	5.3
Moina	32	2	34	5.0
Cladocera indet	36	43	80	11.8
Brachionus	36	25	61	8.9
Microdides	22	28	50	7.3
Keratella	48	20	68	10.0
Rotifera indet	3	5	8	1.1
Cyclops	119	63	182	26.8
calanoida	9	8	18	2.6
nauplius calanoida	22	34	56	8.2
nauplius cyclops	24	56	80	11.8
Macrothrix	1	0	1	0.2
Ceriodaphnia	1	0	1	0.2
Ostracoda indet	0	3	3	0.5
Lepadella	0	2	2	0.3
Total	386	292	678
%	56.9	43.1

There was no significant difference in species composition for the student's t-test between communal lands and National Park in February (t=1.41; P=0.17; df=14) and in April samples (t = 1.13; P = 0.27; df = 14).

Seasonal differences in zooplankton distribution and abundance were seen in both the
communal lands and National Park (Figure 3.3, Figure 3.4, Table 3.10 and Table 3.11).
Copepods (composed of Cyclopoids and Calanoids) dominated the samples, comprising

60% of total zooplankton abundance in February and 49 % in April. Ostracods were not found in February samples. Few Ostracods were recorded in April in Chitampa reservoir and Maleme reservoir, both located in the Matopos National Park (Table 3.9).

Fig. 3.3. Zooplankton species distribution and abundance (numbers of zooplankton per litre in the right) among the 8 reservoirs/ February

Fig. 3.4. Zooplankton species distribution and abundance (numbers of zooplankton per litre in right) among the reservoirs/ April

Sibasa reservoir recorded the highest abundance of zooplankton in February (30%) followed by Maleme (20%) and Dewa (17%) while the highest abundance in April was found in Mpopoma reservoir (21%) followed by Denje, Sibasa and Makoshe. Fig. 3.5 and Fig. 3.6 highlight the differences in abundance between February and April and among reservoirs. It is worth note that Mpopoma had the lowest abundance of taxa in February (2%) and Mezilume had the lowest abundance in April (4%). Fig. 3.5 also highlights the predominance of Cyclopoids in February samples (in Dewa reservoir) followed by Rotifers.

Taxa abbreviations are: Cladocera (clado), Cyclopoids (Cyclo), Calanoida (calan), Rotifer (rotif), Ostracoda (ostrac).

Reservoir abbreviations are: Sibasa (Sib), Makoshe (Makosh), Mpopoma (Mpop), Chitampa (Chita), Mezilume (Mezil), Maleme (Malem)

3.1.3 Overall plankton abundance

The study sites exhibited seasonal variation in the abundance of phytoplankton and zooplankton (Fig 3.7 and Fig. 3.8). The highest abundance of zooplankton was recorded in February while the highest abundance of phytoplankton was recorded in April.

A clustering of stations according to plankton abundance using complete linkage is presented in Fig 3.9 and Fig.3.10. These figures show Sibasa and Mpopoma reservoirs' abundance being very dissimilar to all other reservoirs in February while Sibasa, Makoshe and Mezilume are very dissimilar to others in April. No similarity is found between reservoirs according to their presence or not in the National Park or in the communal lands. A first level similarity is found in February between Chitampa (National Park), Makoshe and Dewa (Communal Lands) similarity in abundance of stations according to their occurrence in the National Park or in the communal lands is depicted. However, these Figures show how close the studied reservoirs were in terms of abundance. Different linkages between the two sampling periods can be found as well when comparing these clusters.

Fig. 3.9 Clustering of stations according to the overall plankton abundance/ February

Fig. 3.10 Clustering of stations according to the overall plankton abundance/ April

3.1.4 Relationships between physico-chemical parameters and plankton species in the studied reservoirs

A comparison of the National Park and communal lands for a number of parameters and species relationships showed significant differences (annexes 5 and 6). The following elements showed a significant difference taking Mezilume (National Park) and Makoshe (Communal lands) as an example: pH of water (P<0.01), electroconductivity of water (P<0.001), total nitrogen (P <0.05), the cladoceran Moina (P <0.05), the cyanophyte Anabaena (P <0.05), and the dinophyte Ceratium (P <0.05).

The colour of water may have had an influence on the following elements using ANOVA: pH of water (P<0.01), electroconductivity of water (P<0.001), total nitrogen (P<0.0 1), hardness of water (P<0.0 1), soil' s electroconductivity (P<0.0 1), the species Daphnia (P<0.001), Brachionus (P<0.01), Microdides (P <0.05), nauplii cyclopoids (P <0.05), nauplii calanoids (P <0.05), Anabaena (P<0.01), Peridinium (P<0.001), Melosira (P <0.05), Scenedesmus (P <0.05), Pediastrum (P<0.01), Hydrodictyon (P<0.01), Dinobryon (P <0.05), Pleurotaenium (P<0.01), Arthrodesmus (P <0.05), Fragilaria (P <0.05) and Zygnema (P <0.05). The soil type also showed a significant influence on some phytoplankton species.

The clustering of all the components of the main biota in relation to the water quality is presented on Fig. 3.11 and 3.12. These figures highlight the relationships of species found in the study area with its chemistry. Total phosphorus, total nitrogen in the small reservoirs and electroconductivity of soils had a relationship with zooplankton (Daphnia, Rotifera, Brachionus and Ostracoda) and phytoplankton taxa (Navicula, Phacus, Lepadella, Sphaerocystis, Staurodesmus, Cosmarium and Sporangium). The pH of water and soils, water hardness and electroconductivity also influenced some species (Fig. 3.11 and 3.12).

Factor Loadings, Factor 1 vs. Factor 2
Rotation: Unrotated
Extraction: Principal axis factoring

Fig. 3.11 Relationship between chemistry and zooplankton in the studied reservoirs

Taxa abbreviations: Nauplii cyclopoids (NAUPLCYC), Nauplii calanoids (NAUPLCA), Microdides (MICRODID), Cladocera (CLADOCER),

Abiotic elements abbreviations: Total Nitrogen (TNWATER), Total Phosphorus in water (TPWATER), electroconductivity in soils (ECSOIL), water hardness (HARDWAT), water' s electroconductivity (ECWAT)

Factor Loadings, Factor 1 vs. Factor 2
Rotation: Unrotated
Extraction: Principal axis factoring

Fig. 3.12 Relationship between chemistry and phytoplankton in the studied reservoirs.

Taxa abbreviations: Gyrosigma (GYROSIGM), Lepadella (LEPADELL), Staurodesmus (STAURODE), Cosmarium (COSMARIO), Rhizosolenia (RHIZOSOL), Sphaerocystis (SPHAEROC), Sporangium (SPORANGI), Closterium (CLOSTERI).

Abiotic elements abbreviations: Total Nitrogen (TNWATER), Total Phosphorus in water (TPWATER), electroconductivity in soils (ECSOIL), water hardness (HARDWAT), water' s electroconductivity (ECWAT)

IV. Discussion

4.1 Water quality aspects, land and water use

The water quality parameters that were analysed in the studied reservoirs indicate a general trend that is acceptable in comparison with the WHO guidelines for drinking water (Chapman, 1992), natural levels in freshwater (Sinkala et al. 2002) and the Zimbabwean effluent standards (ZINWA, undated) (Table 3.3). In the whole study area, pH was more or less alkaline, ranging from 7.6 to 8.5- values that are in agreement with the pH of most natural waters that ranges between 6.0 and 8.5 (Chapman, 1992). Total nitrogen, total phosphorus, total hardness and electroconductivity were not significantly different in the National Park as compared to communal lands.

The similarity in the water quality parameters analysed in the National Park and in the communal lands suggests that ecosystem health, as defined by water quality, is currently not under serious threat due the land and water use in the surrounding communal lands. This result was not expected, based on findings in other areas and the documented impact humans can have on water quality (Brainwood et al. 2004; Vitousek et al. 1997; Siwela et al. 1996, Sharma, 2003). However, in the communal areas of this study, the high quality of water of the reservoirs may be explained by the fact that few or no significant land and water uses are taking place upstream, of the reservoirs; all of the farms in the study area were located downstream of the reservoirs and the human settlement appeared to be located far enough away from the reservoirs to not constitute a threat. The homesteads that were found close to reservoir might have not had significant influence, probably due to their low-density status.

It is widely accepted nowadays that excess fertilization and manure production create phosphorus surplus that accumulates in soils. Some of this surplus is transported in soil runoff to aquatic ecosystems (Carpenter et al., 1998) with respect to rainfall levels. Rainfall levels contribute to water quality in the sense that little or no runoff result in little rainfall. This enrichment of water bodies can only happen when farms and activities

are located upstream of reservoirs. Though the study area under investigation in this study currently lacks upstream influences, it would be expected that increased human settlement or any development upstream of the reservoirs could potentially result in a decrease in water quality in reservoirs near communal lands.

The study may also document the role vegetative cover can play in mitigating the impacts of local soil conditions, caused by natural or anthropogenic phenomena, on water quality in the reservoir. Conductivity measurements in the soils in both National Park and communal lands were three to four times higher than measurements in the water (Table 3.1).

The pH was far more acidic in the National Park soils (pH= 5.3) as compared to the waters in the same area that is slightly alkaline (pH=8). The same trend is found in the communal lands where the soils have a pH around 7.5 and waters a pH=8.2. This contrast in values suggests that the surrounding soils have little influence on the water quality of the reservoir waters that might be attributed to the presence of good vegetation cover around reservoirs, which constitute a buffer to large transfers of elements. The presence of riparian vegetation is crucial in retaining some nutrients. This is confirmed by Carpenter et al.(1998) who state that the maintenance of vegetated riparian zones or buffer strips may reduce the transport of phosphorus and nitrogen to reservoirs. It might be suspected that the degradation of vegetation cover due to human activities on the communal lands could increase the transport of nutrients to the reservoirs, and alter water quality in the future.

While it appears that water quality was not directly impacted by the surrounding soils (values of surrounding soil quality being different from those of the water quality), it is interesting to note that water colour may have been influenced by local conditions. Though water quality was found to be acceptable in the communal lands, a whitish colour of water was present in almost all of the reservoirs. This colour is very close to a white granite rock located 250m upstream of Sibasa reservoir and might be the origin of the colour. Such a strong whitish colour might have an effect on light penetration in the reservoir and compromise the primary productivity within the water column. Thus, it may be impacting biota within the reservoirs. This has been shown by the possible influence

discovered on the quality parameters (pH, electroconductivity, total nitrogen and hardness) as well as some plankton species (Annex 5). Sibasa, having the strong whitish colour, might have got a very high abundance.

4.2 Plankton community composition, diversity and abundance in relation to land and water use

4.2.1 Seasonal variation in plankton diversity and abundance

This study found that the diversity and abundance of plankton species varied seasonally. While this study failed to conclusively support this variation with statistical significance, it is believed that rainfall patterns were responsible for the noted seasonal variation. The two sampling periods fell within the span of the normal rainy season that extends from November to April. More conclusive evidence for seasonal variation in plankton diversity and abundance may be reached if future sampling sessions took place both within the normal rainy and dry seasons.

The slight differences in plankton species distribution amongst the communal land sites (Fig 3.3) are probably due to documented rainfall patterns. February sampling was conducted just after the short rains of January and therefore could probably show a sign of seasonality in April since March and April did not have rains. Differences in species composition observed in the figures from communal lands and National Parks might also be explained by the fact that the Matopos National Park received more rains from November to January as compared to the communal lands that received rains only in January. Thus, the abundance and diversity found in the National Park varied more from February to April. It would be expected even greater seasonal differences in plankton community structure should the samples be taken in the period of November to April for the rainy season and the dry season that spans from May to October.

The results of this study suggest that rainfall variability can significantly impact the
diversity and abundance of plankton communities. This is an important concept, as
rainfall patterns in this area have been erratic in the past 10-15 years, and have the

potential to continue following non-normal trends due to climate variability or climate change in the future. Rainfall patterns are considered here in the sense that their intensity might induce the transport of sediments from upstream of reservoirs to the reservoirs. These sediments, if transported are likely to affect the reservoirs water quality, their biotic composition and ecosystems health. This affirmation is supported by the findings of a relation between the water quality parameters and the species composition. These findings are presented in terms of factor analysis in Fig. 3.11 and Fig. 3.12.

Naz and Turkmen (2005) and Reynolds (1984) acknowledge the fact that seasonal variations in plankton species composition and abundance are believed to depend on interactions between physical and chemical factors that are in turn influenced by climatic factors. The informal climatic sub-seasons noticed during the sampling period were, therefore, likely to alter, though not necessarily with significance, the plankton community structure.

Apart from this seasonal variation due probably to rainfall patterns, no difference was noted between the communal lands and the National Park related to land and water use. The reason, therefore, for a tendency to acknowledge the good status of the health of the environment in the communal lands. This shows similarities between reservoirs in communal lands to the pristine-considered reservoirs in the National Park. However, care should be taken to applaud this finding since a more detailed investigation including all the seasons need to be done in order to be sure of the behaviour of these ecosystems. As Cander-Lund and Lund (1995) confirm that like humans need a regular check-up at the hospital, the health of aquatic ecosystems need to be monitored through the observation of their plankton composition- a regular monitoring might also be interesting to get on track with the evolution of the status of these ecosystems.

4.2.2 Plankton composition

There significant difference found in phytoplankton abundance samples using the chi- square method might be due to natural differences between the National Park and the communal lands. The most abundant phytoplankton taxon found during this investigation was Hydrodictyon spp. in April samples. Its abundance might be due to the fact that the species is known to break into pieces (Cander-Lund and Lund, 1995) and the counting of those small pieces might mislead to an acknowledgement of abundance. It was found, however, in April samples that the taxa was widely distributed. Hydrodictyon was rarely observed in February samples. Anabaena sp., a blue-green algae (Cyanophyta), was the second most abundant species observed in April though it was rare in the February samples. Though Anabaena sp. is always associated with algal blooms, its abundance was not high enough to create an algal bloom. Literature shows, in fact, that Anabaena sp. can be found in non-polluted waters (Cander-Lund and Lund, 1995). However, the presence of this species, and others that prefer similar ecological conditions, in areas where they are not expected to normally occur might be a sign of the enrichment of waters, a term referred to as eutrophication. The current aquatic community structure would likely change with the onset of eutrophication, perhaps altering water quality and rendering the reservoirs unsuitable habitat for a variety of plankton species and unsuitable for human uses as they currently stand. One particular risk of the cyanophytes group is the fact that most of the species (including Anabaena sp.) contain toxic substances that can lead to fish kills wherever their blooms occurs, especially in hyper-eutrophic ecosystems. They have Nitrogen-fixing sites (heterocysts) on their organisms and are therefore able to fix nitrogen; which means that they can proliferate rapidly. Anabaena is, particularly, known to produce neurotoxins that affect the human central nervous system and hepatotoxins that affect human liver (Chipfunde, person.comm.).

Ceratium (fig. M, annex 3), a dinophytes that is likewise known to produce toxic substances and red water blooms, was found in both samples of February and April. This taxon belongs to the major group of dinophytes and is common in the plankton of lakes. Some species are rich in plants nutrients such as phosphates and nitrates (Cander-Lund

and Lund, 1995). Cander-Lund and Lund (1995) states that even in such lakes it is often accompanied by cyanophytes. The results obtained in this work have shown the presence of Ceratium as well as cyanophytes, though the water bodies were nutrient-poor (oligotrophic). Ceratium and Peridinium (another dinophytes) increased in abundance in April samples due probably to favourable conditions to their proliferations such as an increase in total nitrogen levels. This cannot be confirmed since total nitrogen was not analysed for February samples. An increase in nutrient levels in the study area would enhance a high productivity level of dinophytes and cyanophytes, leading to algal blooms, which would compromise health of the ecosystems as they currently stand. It is therefore crucial to keep the water bodies under observation.

The species that could cause algal blooms like Anabaena were mostly present in the communal lands (Table 3.4). Ceratium and Peridinium have also been found in high abundance in the communal lands as compared to the National Park. These taxa are known to be proliferating in nutrient rich waters (Cander-Lund and Lund, 1995); Ceratium being able to exploit organic and inorganic nutrients and gain competitive advantage over purely photosynthetic species (Smalley and Coat, 2002). Because these nutrient enrichment indicative species are abundant in the communal lands, an argument would be made that communal land sites should be monitored for an influx of nutrients that could spur them into an algal bloom.

The zooplankton community was less diverse and less abundant as compared to ponds and reservoirs (from other studies) though the most common zooplankton groups were represented. The lower diversity and abundance found in this study might be explained by the presence of planktivorous fishes and most probably low light penetration (low transparency, especially in the communal lands). Though fish abundance was not part of this study, it was noted that fishes were present in all of the reservoirs. Humans were observed actively fishing on the reservoirs. Planktivorous organisms have preferences for specific food items (Wetzel, 1983). Large planktons are the preferred food item, as they contain the most energetic reward to balance the energy loss the fish has most incurred when hunting. In the case of the studied reservoirs, large planktons were composed of big cladocerans like some species of Daphnia and Calanoids. The idea of

active hunting on large zooplankton can easily be depicted from Table 3.9 where large Calanoids are scarce while their juveniles are abundant. Large Cyclopoids and cladocerans were also rarely found in the samples and most of the time when they were found they were only carcasses. So there seems to be a good zooplankton productivity, which is very well regulated by high predation by fish. This is in accord with Hrbàéek et al. (1958) in Wetzel (1983) who shows that the size of the zooplankton community is regulated by the presence of fish predators. The zooplankton community structure found is also in agreement with Arcifa et al. (1986) who concluded that plankton proliferation is greatly affected by the predator-prey relationships in reservoirs.

The correlation found in zooplankton and phytoplankton species might be explained by the availability and preference of food. The availability of a certain phytoplankton species that constitute a preferred source of food to a zooplankton counterpart will allow it to grow easily, following the same abundance curve. This tends to confirm that the availability of nutrients and necessary conditions for phytoplankton growth has a pulling effect on zooplankton species.

4.2.3 Study limitations

The samples were taken during a drought period (Table 3.3). It is therefore necessary to be aware that the results might have been different if the rainy period (November-April) were normal. This is because low rainfall mobilises less nutrient laden sediments than high intensity (Basnyat et al., 2000 in Brainwood et al., 2004). It is, however, important to remember that the studied reservoirs are located in a semi arid region characterised by low and erratic rainfall as shown by Zirebwa and Twomlow (1999). Since heavy rains should not be expected in the area, the water managers can consider that the reservoirs are protected and environmentally healthier. Water managers should therefore note that water quality investigations are carried out, as stated by Brainwood et al. (2004), to provide information on the health of the water bodies and allow them to develop strategies for better management of catchment and water resources.

4.2.4 Management implications

The results of this study can be used to guide future management of these and similar man-made reservoirs in rural Zimbabwe. Reservoirs on communal lands had similar water quality as found in the National Park, which is attributed to the lack of upstream development surrounding these particular reservoirs. However, it is believed that if human populations alter the current use of these water bodies and develop upstream areas, water quality will suffer. Therefore, it would be stressed that upstream development, particularly development that would result in an influx of nitrogen and phosphorus, be limited in these areas. This is particularly important as the reservoirs contained phytoplankton that would proliferate into toxic algal blooms with the influx of those particular nutrients. Such blooms would compromise the quality of water for both human use and the health of the current aquatic community.

Secondly, this study found that local soil conditions were very different from water conditions. This result can be attributed to the presence of a healthy vegetative cover layer surrounding the reservoirs. Such a vegetative layer acts as a buffer to influxes of elements, and helps to maintain stable and healthy water conditions. Reservoir managers should maintain a healthy vegetative buffer around the water body to assist in mitigating any future changes in local conditions.

V. Conclusion

The results of an assessment of plankton diversity as a water quality indicator of small reservoirs in communal lands and in the National Park (Matopos) were presented. The study hypothesized that water in small reservoirs in communal lands might be turbid but richer in nutrients as an effect of land and water use. The expected nutrient richness would in turn influence phytoplankton abundance, which would therefore be expected to be greater and less diverse. And a second hypothesis followed by expecting reservoirs in the National Park to have less nutrients but more diverse plankton community than those in communal lands.

This study found that water quality throughout the study area was at acceptable levels, and did not significantly differ between National Park and communal lands. This finding was contrary to expectations, and indicates that water conditions may be better in areas of human influence than currently thought. This pattern is attributed to limited upstream development, a condition that should be maintained to ensure the integrity of the aquatic ecosystems. High levels of vegetative cover were also thought to mitigate the impacts of local conditions on water bodies, and should be preserved to protect these systems from future change.

A third hypothesis expected no seasonal changes in plankton community structure of the studied small reservoirs. Contrary to this expectation, the diversity and abundance of plankton communities in the study area were influenced by an informal sub-seasonal rainfall patterns. Species composition of both phyto- and zooplankton were similar to expectations based on other ponds and lakes. However, species of phytoplankton were found that could potentially develop into toxic algal blooms with changes in water quality. Zooplankton species abundance was at a lower level than expected, possibly due to the presence of planktivorous fish in the reservoirs.

VI. References

Arcifa, M.S, Northcote, T.G. and Froehlich, O. 1986. Fish-zooplankton interactions and their effects on water quality of a tropical Brazilian reservoir. Hydrobiologia. 139:49-58

Bolton, P. 1994. Assessing irrigation impact on the environment. Bulletin no 29, HR Wallingford, ODU. UK. 29:4-6

Brainwood, M.A., Burgin S. and Maheshwari, B. 2004. Temporal variations in water quality of farm dams: impacts of land use and water sources. Agric. Water Manage. 70: 151-175

Basima, B., Mbalassa, M., Muhigwa, B. and Nshombo, M. (in press). Anthropogenic influences of the biota of the littoral zone of lake Kivu, Bukavu basin, DRCongo. SILCongress2004 Proceedings. Paper presented for the SIL Congres s 2004. Lahti, Finland.

Cander-Lund, H. and Lund, J.W.G. 1995. Freshwater algae. Their microscopic world explored. Biopress Ltd. England. UK. 360 pp

Carpenter, S., Caraco, N.F., Corell, D.L., Howarth, R.W., Sharpley, A.N. and Smith, V.H. 1998. Nonpoint Pollution of surface waters with Phosphorus and Nitrogen. Ecol. Appl. 8: 559-568

Chapman D. (éd.) 1992. Water Quality Assessments-A guide to the Use of Biota, Sediments and Water in Environmental Monitoring- Second Edition. UNES CO/WHO/UNEP. Chapman and Hall publishers.

Crawley, M.J., Little, C., Southwood, T.R.E. and Ulfstrand, S. 2000. The biology of lakes and ponds. Biology of habitats. Oxford University Press. Pp 29-50.

Davies, B. and Day, J. 1998. Vanishing waters. UCT Press. Cape Town. South Africa. 487 pp.

Drenner, R.W., Threlkeld, T.S., Smith, D.J., Mummert, J.R. and Cantrell, P.A. 1989. Interdepedence of phosphorus, fish and site effects on phytoplankton biomass and zooplankton. Limnol. Oceangr. 34: 1315-1321

Durand, J.-R. and Lévêque, C. 1980. Flore et faune aquatiques de l'Afrique Sahélosoudanienne. Éditions de l'Office de la Recherche Scientifique et Technique Outre- Mer Collection Initiations-Documentations Techniques no 44. Paris. France.

Edmondson, W.T. and Winberg, G.G. 1971. A manual on methods for the assessment of secondary productivity in fresh waters. IBP Handbook No 17. Blackwell Scientific Publications, Oxford.

FAO, 1997. Irrigation potential in Africa: A basin approach. FAO-Land & Water Dvpt. Div. Rome. Italy.

Fernando, C. H. (éd.) 2002. A guide to Tropical Freshwater Zooplankton. Identification, Ecology and Impact on Fisheries. Backhuys Publishers, Leiden.

Grant, P.M.1981. The fertilisation of sandy soils in peasant agriculture. Zimbabwe Agricultural Journal. 81: 97-102.

Green, J. 1990. Zooplankton associations in Zimbabwe. J. Zool., Lond. 222: 259-283.

Hart, R.C. 1990. Zooplankton distribution in relation to turbidity and related environmental gradients in a large subtropical reservoir: patterns and implications. Freshwater Biology 24: 241-263

HR Wallingford 2004. Guidelines for Predicting and Minimising Sedimentation in small dams. HR Wallingford/DFID. Report OD 152

Hunter, J.M., Rey, L. and Scott D. 1982. Man-made lakes and man-made diseases: Towards a policy revolution. Social Sces & Med. 16: 1127-1145

Kabell, T. 1986: Assessment of Design Flood Hydrographs. The Zimb. Eng. 24: 573 - 578

Lawrence, P. and Hasnip, N. 2004. Sedimentation in small dams. Impacts on the income of poor rural communities. HR Wallingford, DFID

Lawrence, P. and Lo Cascio, A. 2004a. Sedimentation in small dams. Hydrology and drawdown computations. HR Wallingford, DFID

Lawrence, P. and Lo Cascio, A. 2004b. Sedimentation in small dams. Estimating the impact of catchment conservation, check dams and sediment bypassing in reducing dam siltation. HR Wallingford, DFID

Lawrence, P., and Lo Cascio, A. 2004c. Sedimentation in small dams. Development of a catchment characteristisation and sediment yield prediction procedure. HR Wallingford, DFID

Lemoalle, J., Adeniji, A., Compère, P., Ganf, G.G., Melack, J. and Talling, J. 1981. The ecology and utilization of African Inland Waters. UNEP Reports & Proceedings. Série 1. Symoens J.J., Burgis M., Gaudet J.J. (éds). Nairobi. Kenya.

Mugabe, F., Hodnett, M. and Senzanje, A. 2003. Opportunities for increasing productive water use from dam water-A case study from semi-arid Zimbabwe. Agricultural Water Management, Vol. 62 (2): 149-163

Muthimkulu, S.N. 2004. Biological assessment of the state of the water quality using the South African Scoring System. A case of the Mbuluzi River, Swaziland. Masters thesis. Department of Civil Engineering, University of Zimbabwe. Zimbabwe.

Mwenge, K.J-M. 2004. Water productivity and yield gap analysis of water harvesting systems in the semi-arid Mzingwane catchment, Zimbabwe. MSc Thesis, Department of Civil Engineering, University Zimbabwe. Zimbabwe.

Naeem, S., Chapin III, F.S., Costanza, R., Ehlrich, P.R., Golley, F.B., Hooper, U.D., Lawton, J.H., O'Neill, R.V., Mooney, H.A., Sala, O.E., Symstad, A.J. and Tilman, D. 1999. Biodiversity and Ecosystem functioning: Maintaining Natural Life Support Processes. No 4. Issues in Ecol.

Naz, M. and Turkmen, M. 2005. Phytoplankton Biomass and Species Composition of Lake Gölbasi (Hatay-Turkey). Turk J Biol 29:49-56

Nhapi, I., Hoko, Z., Siebel, M.A. and Gijzen, H.J. 2002. Assessment of the major water and nutrient flows in the Chivero catchment area, Zimbabwe. Physics & Chem. Earth 27: 783-792

Nhiwatiwa, T. 2004. The limnology and ecology of two small man-made reservoirs in Zimbabwe. Mphil thesis, Dept of Biological Sciences, University of Zimbabwe, Harare, Zimbabwe.

Ongley, D. 1996. Control of water pollution from agriculture-FAO irrigation and drainage paper 5. GEMS/Water Collaborating Centre. Canada Centre for Inland waters. Burlington, Canada.

Oomen, M.V., de Wolf, J. and Jobin, W.R. 1994. Health and Irrigation. Volume I and II. ILRI publication 45 Wageningen, The Netherlands

Pinel-Alloul, B., Niyosenga, T. and Legendre, P. 1995. Spatial and environmental components of freshwater zooplankton structure. Ecoscience. 2(1): 1-19

Reynolds, C.S.1984. The ecology of freshwater phytoplankton. Cambridge University Press.

Scheffer, M., Taddese, G., Boelee, E., Senzanje, A., Yohannes, M., Laamrani, H. 2004. Understanding effects of micro-dams on productivity and health: towards strategies to the reduce environmental disease risk in Morocco, Ethiopia and South Africa/Zimbabwe. Multidisciplinary programme grant. Preliminary proposal. WOTRO.

Schindler, D.W. 1978. Factors regulating phytoplankton production and standing crop in the world's freshwaters. Limnol. Oceanogr. 23, 478-486.

Senzanje, A. and Chimbari, M.J. 2002. Inventory of small dams in Africa-A case study of Zimbabwe. Report for International Water Management Institute (IWMI), Colombo, Sri Lanka.

Sharma, C. 2003. Biological impacts and local perceptions of Tinau River Dam, Nepal. Noragric MSc thesis. Agricultural University of Norway.

Sinkala, T., Mwase, E.T and Mwala, M. 2002. Control of aquatic weeds through pollutant reduction and weed utilization: a weed management approach in the lower Kafue River of Zambia. Physics and Chemistry of Earth 27: 983-99 1

Siwela, A.H., Marufu, G. and Mhlanga, A.T. 1996. A comparison of organochlorine pesticide residues in Upper Ncema and Lower Umguza Dams, Zimbabwe. Journal of Applied Science in Southern Africa. 23-36.

Smalley, G.W. and Coats, D.W. 2002. Ecology of the red-tide dinoflagellate Ceratium furca: distribution, mixotrophy, and grazing impact on ciliate populations of Chesapeake Bay. Eukaryot Microbiol. 49: 63-73.

Straskraba, M. and Tundisi, J.G. 1999. Reservoir Water Quality Management. Guidelines of Lake Management. Volume 9. International Lake Environment Committee (ILEC) Shiga, Japan.

Sugunan, V. V. 1997. Fisheries Management of small water bodies in seven Countries in Africa, Asia and Latin America. FAO Fisheries Circular. No. 933. FAO, Rome. 149pp.

Talling, J.F. and Lemoalle J., 1998. Ecological dynamics of Tropical Inland Waters. Cambridge University Press. 261pp.

Thornton, J.A. and Coterill, N.G. 1978. Some hydrobiological observations on five tropical African montane impoundments. Trans. Rhod.Sci.Ass. 59:22-29.

Thorp, J.H., Black, A.R., Jack J.D. and Casper, A.F. 1996. Pelagic enclosures - modification and use for experi-mental study of riverine plankton. Archive fUr Hydro-biologie (Suppl. 113). Large Rivers, 10, 583-589

Tredgold, R. 1956. The Matopos. Federal Department of Printing and Stationery, Salisbury.

Vitousek, P.M., Aber, J., Howarth, R.W., Likens, G.E., Matson, P.A., Schindler, D.W., Schlesinger, W.H. and Tilman, G.D. 1997. Human alterations of the Global Nitrogen Cycle: Causes and Consequences. Issues in Ecol. No 1.

Ward B. H. and Whipple C.G. 1959. Freshwater biology 2^nd edition. Edmondson W.T. (éd.). USA

ZINWA (undated). Operational Guidelines for the Control of Water Pollution in Zimbabwe.

Zirebwa J. and Twomlow, S. 1999. Historical profiles of selected small dams in communal areas of Zimbabwe. Paper presented at the Engineering Technology for Increased Agricultural Productivity Conference. HICC, Harare, Zimbabwe. 8-10 September.

An Assessment of plankton diversity as an water quality indicator in small man-made reservoirs in the Mzingwane catchment, Limpopo basin, Zimbabwe