Assessing the feasibility of integrating remote sensing and in-situ measurements in monitoring water quality status of Lake Chivero, Zimbabwe

•Water quality in Lake Chivero assessed using remote sensing and in-situ measurements.•MODIS images were used to quantify chlorophyll_a and to predict the pollution level.•High chlorophyll_a levels in the Lake as explained by Harare's expansive nature.•Positive correlation (r2 = 0.89) observed between field based and predicted chlorophyll a.•Results useful for assessing water quality levels in data scarce environments.

This work investigates the likelihood of integrating the cheap and readily-available broadband multispectral MODIS data and in-situ measurements in quantifying and monitoring water quality status of an inland lake within Upper Manyame Catchment in Zimbabwe. Specifically we used MODIS images to quantify inland lake chlorophyll_a concentrations, as a proxy for predicting lake pollution levels. The findings of this study show a high chlorophyll_a concentration of 0.101 ± 0.128 μg/L within the Lake. The results further demonstrated that the chlorophyll_a concentration levels did not significantly vary (p = 0.788) between sites, except among depths (p = 0.05). Further, prediction results based on the relationship between observed and predicted chlorophyll_a produced a high R2 value of 0.89 and a root mean square error (RMSE) value of 0.003 μg/L. Moreover, the derived landuse maps of Upper Manyame Catchment indicated a significant variation in the percentage settlement in 1985, 1994 and 2010 change from 1985 to 2010. For instance, 8% increase in settlement in the period between 1994 and 2010 and over 12% increase from 1985 to 2010 and a decline in percent forest coverage (i.e. 9.8% in 1985 to 2.0% in the year 2010) in the catchment was observed. Overall, the findings of this study highlights the importance of free and readily-available satellite datasets (such as the multispectral MODIS and Landsat) in quantifying and monitoring water quality across inland lakes especially in data-scarce areas like Sub-Saharan Africa.