Relationship between lake level fluctuations (WLFs) and water quality parameters
A significant correlation (p < 0.01) was found between nine water quality parameters (conductivity, depth, TP, PO43–, WQI, DO, temperature, turbidity, and NO3) and DLTM (Table II). Only, one parameter (SiO44–) was significantly (p < 0.01) correlated with the yearly amplitude (WLamp) in Lake Baringo indicating the DLTM to be a better predictor of water quality in the lake at an inter-annual scale. Except for the depth and different forms of nitrogen (NO3, NO2 and NH4+) which had negative correlations with DLTM, other parameters demonstrated positive concordance with increasing DLTM over time in the lake. NH4+ that ranged from 21.75 to 48.8 µg L-1, also showed significant negative correlation (r = -0.64, p < 0.05) with DLTM over time in the lake.
Silicate ions (SiO44–) ranged from 18.92 to 27.43 mg L-1 across all years (2008-2021). SiO44– did not demonstrate a concordance with DLTM but showed a positive significant correlation with yearly amplitude of lake levels (r = 0.70, p = 0.008) indicating the ion is tracked by short-term changes in the lake. There was no significant concordance between Chl-a, and TN with any of the WLFs indices over time in the lake. The water quality index (WQI) as a measure of the lake water quality was significantly correlated with the DLTM (r = 0.80,p = 0.001) and not the lake water level amplitude, WLamp (Table II). The Gaussian 4 parameter unimodal model further provided a significant fit (p < 0.05) to the relationship between WQI and DLTM (R² = 0.89) and not with the lake amplitude (Figure 6a and b) indicating that a range of DLTM of ≥ 2 m and in the range of 6.5 to ≥10 m provided a good water quality of the lake as WQI approached the WHO recommended value of WQI ≤ 100, while DTLM values of ≤ 2 m resulted into poor water quality. The Gaussian, 4-parameter model: WQI = 314.68 + 1252.05*exp(-0.5*((x - 3.88)/2.2)²) yielded a significant relationship (p = 0.0024 ) between WQI and DLTM (R² = 0.89) (Figure 6). These results indicated that the water quality index (WQI) was lowest in years closest to the highest long-term mean (DLTM = 9.9 m) and increased with decreasing lake water levels.