FACT SHEET
Monitoring pollution in 4 biodiversity hotspots in Lake
Victoria (Nyando, Awach, Nzoia and Yala river mouths
for informed decisions
May 2020
KMF/RS/2020/C1.20
KENYA MARINE AND FISHERIES RESEARCH INSTITUTE
FRESH WATER SYSTEMS
KMFRI Headquarters P.O. Box 81651 – 80100, Mombasa Tel: +254 (041475151/4) Email: [email protected]
KMFRI Kisumu P.O. Box 1881 – 40100, Kisumu
Tel: +254 (0) 711 233774 Email: [email protected]
AUTHORS
Mwamburi, J.,
Miruka J. B.,
Wawiye, P., Guya,
F., Aura C.M.
Nyamweya C.S.
Njiru J.M.
Major rivers of Lake Victoria
Kenya basin, are considered as
important hotspots of both
aquatic and terrestrial
biodiversity, besides the
numerous ecosystem services
to the surrounding lake basin
communities. Rivers
discharge contribute about
38% of the total water inputs
into the main lake. River water
also supports pond
aquaculture, irrigation
agriculture, industrial,
domestic and livestock water
uses, and energy generation,.
Many native and rare fish
species spawn in river mouth
areas. However, habitat loss
from pollution is of concerns
and a threat to sustainability of
both lotic and lentic
ecosystems services.
Background information
The Kenyan catchment area covers about 42,460 Km2 (about 22% of the whole Lake Victoria basin
catchment area). This covers areas in Busia, Siaya, Kisumu, Bungoma, Kakamega, Vihiga, Uasin
Gishu, Nandi, Kericho, Bomet, Kisii, Homa bay and Migori counties. Major river drainage systems
(Nyando, Sondu-Miriu, Awach Kibuon, Oluch, Kibos) discharge directly into the Winam Gulf,
whereas rivers Nzoia, Yala, Sio and Kuja are connected directly to the main lake. River discharge
contributes over 18% of the water inputs into Lake Victoria. In Kenya, this amounts to a mean
discharge of 292.1 m3s-1 representing 38%. The Nyando, Yala, Nzoia Sondu - Miriu, Awach and
Kuja drainage basins form key ecosystems supporting a diverse aquatic and terrestrial biodiversity,
and the river mouth areas are recognized as important critical fish habitats. These areas are hotspots
of biodiversity, and are demarcated as key fish breeding and nursery grounds.
Riverine ecosystems form extensive conduits for supply of particulate and dissolved chemical and
nutrient elements into recipient downstream lakes. A comparison of Landsat image from 1973 and
2001 shows a significant increase in suspended sediment in the gulf (UNEP 2014). Recent surveys
in Winam gulf showed high fluctuations in surface water coverage by water hyacinth. (Ongore et
al., 2018), with the highest peak infestations in September and November 2016 and river mouth
and sheltered bays as hotspots of the aquatic weed. Fish based Index of Biotic Integrity (FIBI)
provided a tool form monitoring of rivers within the Lake Victoria basin (Raburu and Masese,
2012; Raburu et al., 2009). Some fish species migrate up-to 80 Km upstream in major rivers
(Whitehead 1959; Cadwalladr, 1965).
Significant changes in landuse as a result of expanding human population, urbanization and
settlements, encroachment of forested lands and wetlands, has been accompanied by varying scales
of upstream land degradation and loss of ecosystem services, in different drainage basins. Notable
ecological changes within L. Victoria ecosystem with potential influences on aquatic biodiversity
includes the Nile perch ( now a major commercially important fish species) fish introductions of
1950’s and 1960’s, which preyed on the native haplochromine species.
Uncontrolled and poor solid waste disposal and management; untreated waste water discharge into
rivers; leaching of residual agricultural chemicals; diffuse surface runoffs and storm waters) from
anthropogenic activities and increasing river water demands are a threat to riverine ecosystems and
fish habitats.
Declining quality of water sources can lead to conflicts among different upstream and downstream
users and stakeholders, human – wildlife, irrigation farmers and water resources managers among
others. Expanding pond aquaculture will be sustained by protecting available water sources from
rivers and their recharge areas. Habitat loss can result from river channel modification,
sedimentation, poor water quality and other invasive species, and hence a cause of concern to the
aquatic biodiversity conservation and sustainability of the fisheries.
A
visual comparison of the colour of surface water in rivers Awach, Nyando, and Yala.
Monitoring of the pollution, key ecological components and indicators of water quality and
changes in environmental conditions in riverine habitats will ensure creation of more
awareness for protection of their ecological health and integrity, to the immediate
community users and other stakeholders, and for sustainable ecosystem services. Thirteen
of the most accessible sites representing the downstream and river mouth areas in each
river of the 4 river channels (Nyando, Yala, Awach and Nzoia) were studied and compared
to previously documented ecological and environmental conditions, for more informed
decisions in their development and overall water resources management. The field surveys
were conducted in November 2019 and March 2020 during the wet and dry seasons
respectively. Challenges often encountered in sampling of rivers were experienced, especially
during high flows and unaccessibility of some river mouth areas.
Clean water abstraction (left) and fish traps (right) in relatively un-degraded habitats.
Monitoring of pollution was undertaken through determination of the spatio - temporal
changes in the physico - chemical conditions; and their influence on the distribution of select indicator
taxa in the four-biodiversity hotspots. The composition, abundance and distribution of fish species,
zooplankton, phytoplankton, macroinvertebrates and other associated terrestrial and aquatic species
diversity; to try and assess how the taxa are affected by environmental and habitat changes.
The immediate communities perceptions of the on the historical changes in hydrological conditions,
potential pollution and ecological impacts affecting the riverine ecosystems within the lake Victoria
Kenya basin were also collected. This will help in understanding the challenges facing the immediate
communities, and other stakeholders around the hotspot areas; in protection of the biodiversity for a
sustainable fisheries and future development and management planning of the water resources within
the drainage basins.
Results
The results show a longitudinal variation the mean values of some of the physic-chemical parameters
The mean water pH lies within the acceptable range of 6.5 to 8.5. The surface water temperature
decreased upstream with a mean range of 23.7±1.6 to 25.6±1.1 °C in all rivers (March 2020) and
21.1±0.5 to 24.5±1.1 (November 2019). In both seasons, the lowest surface water temperatures were
21.7°C and 20.7 °C (R. Awach at Kokwanyo upstream); whereas the highest was recorded at the mouth
of rivers Yala (25.9 °C) and Nyando (27.5 °C).
High variations in D.O were evident. The river surface water was well oxygenated with mean
concentrations above 7.12 mgL-1, except for the lower concentrations at Goye, R. Yala, (4.18 mgL-1).
During November 2019, water conductivity and TDS decreased upstream, with rivers Nyando and
Awach recording the highest and lowest conductivity and TDS concentrations. Less defined trends in
R. Nyando were observed in March 2020. Water salinity showed a wide variation from 0.02 ‰ – 0.15
‰, with the lowest and highest concentrations in R. Awach and R. Nyando (Nyando at Chemelil)
respectively.
In general, all the water quality parameters determined were found to be lower than the drinking water
guideline values recommended by NEMA, except for the high concentration of suspended solids and
turbidity. Much higher total phosphorus concentrations, and which reflect eutrophic conditions were
observed in all rivers, but with lower river Yala (57.57±21.99 µgL-1) during November 2019. Lower
dissolved silica-Si and P were observed in March 2020 compared to the wet period.
The variations in total and soluble phosphorus in river water during Nov 2019 and Mar 2020
Seasonal and anthropogenic influences on the water quality were observed in all the river channels,
which is likely to impact resident species to some extent. River drainage basins are extensive and under
heavy agricultural and human activities with reduced soil cover which contributes to increasing loading
of eroded soils and materials which imparting a heavy visually evident brownish colouration in most
of the surface water. In contrast, clear waters of the Yala river mouth is an indication of reduced direct
surface transport of such materials into the lake waters, as it passes through the swampy areas.
Among the algal families, Diatoms, Chrophytes, Euglenophytes and Cynophyceaen were the most
dominant. 75 different species of algae identified. Of the 12 different species of cyanophytes
encountered, Chroococcus sp., Cylindrospermopsis, Planktolyngbya spp. and anabaena spp. were the
most common genera. Only one species of dinoflagellates (Glenoridinium spp.) was encountered in
Koru and Bondo water works. There were 29 different species of diatoms constituted mainly by of
Aulacoisera, Cyclotella, Chodatella spp., Fragillaria spp., Navicula spp., Surillella spp.
The most significant driving forces that cause predictable variations in phytoplankton composition is
due to algal species occupying different ecological niches within the sampling stations. The high
abundance of diatoms and chrophytes as dominant phytoplankton families, is an indication of cultural
eutrophication. High nutrients enrichment seems to enhance growth and increased algal density, and
more especially blue green algae which are proportionately high in most of the stations. Turbidly
favours the dominance of diatoms especially centric diatoms which are able to attach on detritus.
The percentage composition of phytoplankton groups during November 2019 and March 2020
There was a low zooplankton species diversity which is attributed to sampling during wet season.
However, the low diversity may also be attributed predation pressure from zooplanktivorous fish as
well as the carnivorous zooplankton which feeds on algae especially Merismopedia sp. The dominant
algae are not easily digested by zooplankton due to its fibrous nature and colonial formations. This
impacts the zooplankton community species especially rotifers found at Yala river mouth. Branchionus
spp. conditions only favour predation species on the large bodied zooplankters. The genus, Cyclopoida
genus is known to inhabit areas which are rocky, and could have a risen from sweeping effect of water
through adjacent macrophytes. The latter is a new finding in the river ecosystem health, there is need
to focus on the same in future sampling activities for temporal and spatial scales of the physico-
chemical environment that determine levels of primary production and plankton dynamics in aquatic
ecosystems hence affect the quantity and quality of food items available for invertebrates as well as
fish.
The Cyanophytes cell abundances were positively correlated with temperature (r = 0.701, p < 0.01)
and D.O (r = 0.711, p < 0.01), whereas diatoms were negatively and significantly associated with
water temperature (r= -0.740, p < 0.01).
Total zooplankton and copepods abundance of was positively and significantly correlated (p< 0.01)
with surface water temperature and D.O concentration. Abundances of zooplankton are often
regulated by temperature, food availability (phytoplankton) and predation. Temperature influences
the egg development time, growth rate, brood size and mortality of zooplankton (Herzig 1994).
The dissolved P, nitrate –N, and Nitrite N were all negatively associated with zooplankton
abundances. Among the nutrient elements, only ammonium –N was positively correlated with
Dinoflagellates. This may mean food availability was more important for regulating zooplankton,
than eutrophication, as reflected by the low contribution of rotifer (prefer more eutrophic
conditions) in the stations.
River channels are relatively of relatively shallow depths and with high water mixing rates due to
constant turbulence upstream, which decreases downstream. Most of the channels are exposed to
several diverse non-point (NP) anthropogenic pollution sources along the sub-basins. Known
industrial discharges (PS - point source pollution) are upstream rivers Nzoia and Nyando.
Normally, a combination of several abiotic and biotic factors influence algal production in water.
Light penetration, nutrients, zooplankton and fish grazers are factors influencing algae production.
Micro-invertebrates also prey on zooplankton. Increased presence of particulate mineral matter
reduce light availability for algae. However, the positive correlations with water temperature may
arise due to the increased heating effect from the high presence of suspended mineral matter which
reflects solar radiation within water. High river flows, high turbidity and inadequate sampling of
inaccessible river channels can contribute to the differences numerical abundances of zooplankton
observed. High turbidity may also reduce predation of smaller zooplankton by fish and micro-
invertebrates, resulting in variations numerical abundances and low diversity. In comparison, all
the peak abundances in the Kenyan portion of L. Victoria (lowest 41 Ind.L-1 and highest 474 ind.
L-1 at Kisumu pier), were reported at river mouths and bays (KMFRI 2013), with a similar
dominance by Copepoda.
Abundance of zooplankton du ring Nove 2019 and March 2020
Abundance of zooplankton in rivers during November 2019 and March 2020
The molar ratios of N:P were varied between March 2020 (dry) and November 2019 (wet) with a range
of 0.18 to 36.92, and suggested N –limitation in most of the rivers except for rivers Awach and Nzoia
in November 2019, where P limitation was observed.This may imply less variations in algal
compositions between the rivers. In total there were 75 different species of algae identified from rivers
during the November 2019 and 54 in March 2020.
A total number of 22 macroinveterbrate families representing 13 orders; the orders; Coleoptera,
Decapoda, Diptera, Ephemeroptera, Hemiptera, Hirudineae, Odonata, Plecoptera, Pulmonata,
Trichoptera, Tubificida, Unionoida and Veneroida.were recorded in November.
The percentage of tolerant and intolerant macroinvertebrates, Shannon’s and Simpson’s diversity
indices during March 2020
The percentage of the macroinvertebrate functional groups during March 2020
A total of 9 fish families were found in the study sites, which consisted of 19 fish species (wet season)
and 18 species (dry) from 8 families. The Cyprinidae formed the most dominant species, with the
Cichlids which were represented by O. niloticus and Haplochrois only encountered in rivers Nyando
and Yala. In both dry and wet season, the weight and numbers of fish caught was dominated by E.
nyanzae, E, cercops, L. altianalis, L. victorianus, Haplochromis and B. jacksonii fish species.
However, the fish are exposed to increasing degradation of the natural riverine habitats, and loading of
land derived materials, causing increasing turbidity, which affects aquatic productivity.
There exists a huge potential for conservation of non-commercial native fish species within the riverine
habitats, as indicated by the high diversity and abundance of some of the taxa. It is not possible to
effectively sample the riverine sites due to observed challenges of effective gears and accessibility.
Therefore in order to improve sample collection and representation of the extensive and diverse habitats
in rivers, more techniques and mixed gears need to be employed, with focus on the smaller streams and
tributaries. This should be accompanied by updated freshwater fish biodiversity hotspots guides for
specific river drainage basins to understanding effects of increasing ecological changes due to
degradation of riverine ecosystems. The riverine biodiversity hotspots should be clearly identified in
water resource management databases.
Most of the inflowing rivers serve as important sources of water into fish culture ponds, and receive
effluents from such ponds. Oreochromis niloticus and Clarias gariepinus are fish species widely
cultured in fish ponds within the lake basin. However, there are no reports of new species in the studied
rivers.
Previous biodiversity studies recorded 37, 28 and 25 fish species belonging to 11, 10 and 5 families in
Lake Victoria (Kenya), satellite lakes and dams (Masai et al., 2005) within the Kenyan catchment
respectively. Anabatidae and Propteridae, were the only families not represented in this riverine study.
Fish species composition (%) during November 2019mand March 2020.
Socio-economic aspects
Majority of the respondents were males (79%, n = 58) who are mostly involved in fishing, fish trade
and farming activities. The dominant age bracket indicates that the respondents were knowledgeable
on the historical trends within each river system. Almost over 80% were educated past primary level
of education, however, the income levels were low. Majority of them have adequate knowledge on the
river drainage basin (above 70% in all rivers).
The respondent’s perception on historical trends during the past 10 years revealed that floods are
increasing in rivers Yala, Nzoia and Awach. They felt that inter-seasonal variations were more less
stable (except for R. Awach Kibuon), but water levels tended to decrease during the drought. Changes
in amount of rainfall and water table rise between seasons has an impact of the variations in river water
flows.
Respondents occupation (%) Age of respondents (%)
In all the rivers, respondents felt the water sources are still very useful as there are still relied for most
of the historical domestic uses, fishing and livestock watering, with more potential demands of water
for irrigation and agriculture.
Silt, domestic washing and human and animal refuse were ranked highly as important pollutant sources
in the four inflowing rivers, except litter and obnoxious weeds.
0 10 20 30 40 50
<20
20 - 30
30 - 40
40 - 50
> 50
% of respondents
Age
in y
ears
The proportion of respondent’s perception on whether rivers are polluted.
Respondents in each river basin did not identify resource use conflicts as an impact of concerns.
However, poor river water quality was more perceived as the main impact from pollution. All pollution
impacts (poor water quality, unsafe water, high costs of purification) were of concerns in utilization of
water of river Awach, which reflects its overall low quality in comparison.
Conclusion
Seasonal and anthropogenic influences on the water quality were observed in all the river channels,
which is likely to impact resident species to some extent. River drainage basins are extensive and under
heavy agricultural and human activities with reduced soil cover which contributes to increasing loading
of eroded soils and materials which imparting a heavy visually evident brownish colouration in most
of the surface water. In contrast, clear waters of the Yala river mouth is an indication of reduced direct
surface transport of such materials into the lake waters, as it passes through the swampy areas.
River sites show a high concentrations of algal families dominated by Diatoms, Chrophytes,
Euglenophytes and Cynophyceae. The high abundance of diatoms and chrophytes as dominant
phytoplankton families, is an indication of cultural eutrophication.
High nutrients enrichment seems to enhance growth and increased algal density, and more especially
blue green algae which are proportionately high in most of the stations.
Turbidly favours the dominance of diatoms especially centric diatoms which are able to attach on
detritus.
A new occurrence of the Cyclopoida genus, known to inhabit areas which are rocky could be due to
the sweeping effect of water through adjacent macrophytes reported in this survey.
The high temporal variations in flow regimes in tropical areas necessitates the development of more
information on environmental flows in the rivers which is lacking for may basins and sub-basins.
Recommendations
Improved management of catchment area through improved agricultural practices and protection
existing forest areas with increased maintenance of vegetation cover through planting of perennial
crops and trees will reduce soil erosion and loading in rivers. Planned harvesting of sand minimizes
destruction of riverine aquatic habitats,, and also maintenance of the riparian zone vegetation traps
sediment and reduces suspended sediment loadinsg.-- This should be coupled with better
management of both solid and liquid waste discharges, and treatment of effluent discharged into
recipient river channels.
To address the increasing degradation of riverine ecosystems, and to reduce resource use conflicts,
which undermine biodiversity protection; there is need to develop, education, awareness and
sensitization programes, involving the communities and different stakeholders in improved
pollution control measures; adoption of cleaner production technologies and good agricultural
practices; sustainable use and management of specific river basins.
KMFRI to continue monitoring activities on impacts of pollution on riverine biodiversity (with a
wider focus to detail in smaller streams and tributaries) since non-point pollution sources are
diverse within the drainage basins.
A river monitoring program (gauged stations with long term data) for development of long-term
trends data on nutrients and for management of nutrient loadings and lake eutrophication is
required for the whole Lake Victoria Kenya basin, as some of the rivers lack gauge stations for
involvement of communities in river water quality monitoring activities..
Longterm flood management strategies are urgently required as part of development planning to
reduce associated risks.