*Corresponding author
ABSTRACT
The study was conducted in Lake Baringo, a shallow rift valley lake in
Kenya, and determined quantitative relationships between water level
changes, water quality, and fishery production for informed lake basin
management. Long-term (2008 to 2020) data on water level, water quality,
and fisheries yields from Lake Baringo were analyzed using a combination
of statistical methods. Linear and waveform regression analyses were
used to describe patterns of lake level fluctuations over time while,
Pearson’s correlation was used to determine the concordance of lake
level changes with water quality parameters, landings, and condition of
fish species. PCA results grouped the study period into the years (2008
to 2012) marked by higher TP, turbidity levels, and Water Quality Index
(WQI) values demonstrating the periods of poor water quality, and
periods (2015 to 2020) of good water quality characterized by decreased
TP, low turbidity and low WQI values attributed to higher annual lake
water levels. Locally-weighted scatter plot smoother (LOWESS) analysis
showed the annual lake level amplitude to have generally declined over
time with peak values in 1964 (8.6 m) and 2008 (9.4 m). The waveform
regression significantly modeled lake level fluctuations as indexed by
annual deviations from the long-term average (DLTM) and showed a 20-year
oscillation between peak water levels in the lake. The lake water
quality increased during the years when DLTM values were ≥ 2 m. There
were significant positive correlations of Water Level Fluctuations
(WLFs) with water quality variables (conductivity, depth, TP,
PO43–, DO, temperature, turbidity,
NO2–,
NH4+ and
NO3–) and water quality index (WQI)
in Lake Baringo indicating lake water quality deteriorated during lower
water levels and improved during rising water levels. Linear regression
analyses showed a significant concordance (p < 0.05) between
the annual fishery yield and the rising WLFs (r = 0.66). Also, there was
a significant (p < 0.001) relationship between the condition
factor of the native species, oreochromis niloticus baringoensis ,
and the yearly lake level amplitude (r = 0.69) while, catches of the
lungfish, protopterus aethiopicus and Barbus intermedius,showed a differing relationship with WLFs in the lake indicating
species-specific influence of WLFs on catches from the lake. Overall,
the results demonstrate that WLFs of Lake Baringo are a driver of fish
species biomass and physico-chemical properties of the lake. We
recommend the integration of fisheries yields, water quality assessment,
and WLFs modeling at different temporal scales in the management of
Afrotropical lake ecosystems.
Keywords: Fisheries, water quality parameters, waveform and Gaussian
models, and watershed management.
INTRODUCTION
Water level fluctuations in aquatic systems affect their
physico-chemical properties, assemblage structures, ecosystem functions,
and ecological services (Coops and Hosper, 2002). Water level changes in
lakes can be caused by natural factors such as climatic variability
(Bergonzini et al ., 2004; Gownaris et al ., 2015), human
influences such as through damning abstraction (Evtimova and Donohue,
2014) and through climatic forcing leading to extreme variability in
rainfall and drought conditions (Dudgeon et al., 2006). The changes in
the lake water level may intern affect physico-chemical parameters
through volume changes and productivity of the system through
ecohydrological influences (Zalewski et al ., 1997; Coops and
Hosper, 2002). The influence of water level variations on water quality
and fisheries production has been widely studied in most temperate
rivers, lakes, and reservoirs (Welcomme 1970; Hamerlynck et al .,
2011; Kolding et al ., 2016). However, there has been a paucity of
studies in tropical freshwater bodies and especially the Afrotropical
systems. Changes in water level fluctuations (WLFs) tend to affect lake
productivity through nutrient supply variations, influence on breeding
areas, and changes in shallow productive inshore areas (Kolding and van
Zwieten, 2012; Gownaris et al ., 2015; Musinguzi et al .,
2019) in addition to changes in water quality parameters (White et
al ., 2008).
The physico-chemical properties of water bodies play a significant role
in various aspects of their ecohydrobiology (the relationship between
hydrology and ecology) (Zalewski et al ., 1997). The interactions
of physico-chemical properties of water have a significant role in the
composition, distribution, and abundance of aquatic organisms (Hinckleyet al ., 2014). Lake Baringo is a shallow freshwater rift valley
lake in Kenya with a surface area of about 140 km2 and
designated as a Ramsar site (Ramsar, 2002). The lake is facing
human-induced and natural stressors as a result of land and water use
systems as well as climate variability, amongst others (Omondi et
al ., 2014). Additionally, Lake Baringo is characterized by low water
depths (average of 9.5 m, as in 2020, Walumona, personal observation),
the mixing of surface and bottom water induced by the wave actions
together with the clay soils of the catchment contributes to the lake’s
notable high turbidity, reported to affect the primary productivity of
the lake (Hinckley et al ., 2014; Odada et al. , 2006;
Nyakeya et al. , 2020). The lake has a long history of water level
fluctuations (Hickley et al ., 2004; Aura et al ., 2020).
Although there have been reports of the influence of these fluctuations
on its fishery (Kallqvist, 1987; Omondi et al ., 2011), these have
been mostly on an inter-annual or semi-annual basis and have not
accounted for long-term inter-annual variability. Additionally, the lake
water level changes are likely to affect physico-chemical parameter
values, however, despite the periodic fluctuations in the water level of
Lake Baringo, there are no accounts of how these fluctuations have
affected its water quality parameters and ecological functions
including, fisheries yields. In this study, we hypothesized that both
intra- and inter-annual changes in the levels of Lake Baringo will
affect the water quality parameters and fisheries, with cascading
influences on lake functions and livelihoods.
MATERIALS AND METHODS