MODELLING LAKE LEVEL VARIATIONS, WATER BALANCE, AND FISHERIES OF LAKE BARINGO, KENYA

Abstract

Lake Baringo has the worst eroded catchment area and whose water levels (WLs) have been fluctuating at inter- and intra-annual time scales. These changes have been related to natural events and to human activities in the lake catchment. Changes in Lake Baringo water levels and water quality have affected fisheries production, ecological functioning and livelihoods. However, like for most tropical lakes, little is known about the linkages between WL changes, water quality parameters, fisheries production, and water/nutrient balance components of Lake Baringo. This study aimed to bridge this gap by modelling the linkages between WL variations, fisheries production, water quality, and water balance parameters of the lake. The study used both short and long-term datasets. Shortterm data were generated through monthly sampling of nine stations in the lake that extended from January 2020 to June 2021 while, long-term data (1956-2018) were sourced from grey and published data on the lake. Water samples were analyzed for selected physico-chemical variables using the standard methods for the analysis of water and wastewater at the Kenya Marine and Fisheries Research Institute (KMFRI) Baringo and Kisumu laboratories. Water balance of the lake was modeled through determination of input (river discharges, rainfall) and output (evaporation, abstraction, seepage) components of the balance. Lake level changes were analyzed based on annual and monthly deviations from long-term average (DLTM) and from patterns of lake level amplitude obtained from the annual/monthly maximum and minimum lake levels (WLamp). The fisheries and ecological functioning of Lake Baringo were described using the Ecopath mass-balanced model. Four main inputs required in the Ecopath model were used and included: biomass (B) production/biomass ratio (P/B), consumption/biomass ratio (Q/B), and ecotrophic efficiency (EE). Three annual Ecopath models (1999, 2010, 2020) were generated in order to compare the temporal trend in ecosystem functioning of the lake. Statistical approaches used to analyze the data majorly included; analysis of variance (ANOVA) applications, Locally-Weighted Scatter Plot Smoother (LOWESS), linear and waveform regression analyses, and Principal Component Analysis (PCA). Results indicated that, at inter-annual scale, the lake’s trophic status shifted from eutrophic to mesotrophic status at long-term (2008-2020) and short-term (2020-2021) scales, following the Carlson’s Trophic Status Index (CTSI) values. Water balance modelling indicated the water inputs from the runoff contributed for 75% in 1970-1995 and 71% in 2008-2021 to the lake’s storage while, direct precipitation into the lake contributed 25% in 1970-1995 and 29% in 2008-2021 to the lake’s storage, indicating that runoff is the major input component of Lake Baringo’s water balance. The water losses from the lake were contributed by evaporation (59-42%), abstraction (29-7%) and underground seepage (7-3%) during 1970-1995 and 2008-2021, respectively. Waveform regression significantly modeled DLTM and showed a 20-year oscillation between peak water levels in the lake. There were significant positive correlations between Water Level Fluctuations (WLFs) and both the water quality variables and Water Quality Index (WQI), and between fishery yields and WLFs in Lake Baringo. Three annual Ecopath models (1999, 2010, 2020) confirmed three trophic levels for the lake and suggested a strong bottom-up control in the lake’s food web. Oreochromis niloticus baringoensis is modeled as the keystone species in the lake. The Ecopath network analysis for the three models provided the ecosystem functioning and fisheries indicators whose temporal trends are variable and are described in the thesis. Overall, it is recommended that water resource management policies should guide the uses of water from the lake based on WHO guidelines. Water quality assessment, WLFs, species keystonness and other Ecopath results should be considered in the application of holistic and integrated lake basin management (ILBM) approaches in Lake Baringo and its watershed in order to sustain ecological services and the lake-dependent livelihoods.