Remote sensing application in compliment to in-situ monitoring of water quality: Lower Manyame Sub-catchment, Zimbabwe

Abstract

Lower Manyame Sub-catchment (LMS) is highly threatened by land surface-pollution. This study assessed use of existing empirical algorithms to monitor spatio-temporal variation of surface water quality using Sentinel-2, for catchment protection. Chemical Oxygen Demand (COD), total nitrogen (TN), Biochemical Oxygen Demand (BOD₅), total phosphorous (TP) and Total Suspended Solids (TSS) were determined for 12 sites in LMS. Spearman's correlation coefficient, r was used to assess the effect of river catchment condition on surface water quality. The most significant negative relationship was found between TP and elevation (r = −0.70; p = 0.011) and COD and slope (r = −0.70, p = 0.012) whereas the weakest negative relationship was between BOD₅ and elevation (r = −0.10; p = 0.761). Coefficient of determination was used to assess the prediction capacity of remote sensing algorithms in determining surface water quality. The best R² value of 0.92 (TN) and a fair R² of 0.51 (TSS) were observed during the wet season. The research showed that marginal and inaccessible catchments could be monitored using existing empirical algorithms. This study gave insights to environmental custodians to be able to monitor and determine the spatio-temporal variation of water quality in catchments’ rivers, using LMS as an example.