Assessment of Drivers Influencing Surface Water Temperature in Lake Gregory Using the General Lake Model (GLM) and Satellite Observation

Small and shallow lakes in tropical regions often lack reliable ground-based temperature data, making it challenging to fully understand their thermal dynamics, which are crucial as they directly influence water quality, ecosystem functioning, and the lake’s response to climate variability. This study employed the General Lake Model (GLM) to simulate daily Lake Surface Water Temperature (LSWT) from 2017 to 2021, using NASA POWER reanalysis data as meteorological input. Landsat Level 2 Surface Temperature (LST-L2) data were used for calibration and validation, resulting in Root Mean Squared Error (RMSE) values of 2.15°C and 2.17°C, and correlation coefficients of 0.65 and 0.73, respectively. The research further examined the effects of climate variability on LSWT by analysing 625 climate scenarios that incorporated variations in air temperature, solar radiation, wind speed, and rainfall. Multiple independent methods, including linear regression, random forest, and ANOVA, were used in the analysis, consistently demonstrating that air temperature was the primary driver of LSWT, accounting for over 65% of the variance. Wind speed was the second most influential factor, describing over 31% of the variance. Shortwave radiation, although statistically significant, had a relatively minor contribution. Precipitation, although included in all scenarios, showed negligible explanatory power for LSWT under the simulated conditions. These findings suggest that, for Gregory Lake, future LSWT variations will be primarily influenced by atmospheric heating and wind-driven mixing. The study also demonstrates the effectiveness of combining satellite observations and reanalysis datasets with hydrodynamic modelling to estimate LSWT in ungauged lakes with limited climate records. Furthermore, this analytical framework can be applied to other lakes to identify their principal climatic drivers.