Detecting land surface water changes in the Upper Mzingwane sub-catchment using remotely sensed data

Globally, water is acknowledged as indispensable. It is essential for both human life and environmental needs. However, surface water resources are threatened by human and climatic influences, which may result in changes in size and density. This study aimed to evaluate the effectiveness of the normalised difference water index (NDWI), modified normalised difference water index (MNDWI) and automated water extraction index (AWEI) in detecting land surface water changes using Landsat satellite data. The results showed that the AWEI performed considerably better than the MNDWI and NDWI for extracting water surface area in the Upper Mzingwane sub-catchment, with an overall accuracy of 0.93 and a kappa coefficient of 0.82. The MNDWI and NDWI had overall accuracy/kappa values of 0.88/0.74 and 0.89/0.73, respectively. The AWEI can enhance surface water features while effectively suppressing or eliminating pollution and noise from surrounding vegetation and muddy soil. NDWI/MDWI water information is often mixed with pollution noise, vegetation and muddy soil, overestimating the area of water. All the applied indices indicate a progressive loss in the surface area of the water bodies in the sub-catchment. The decrease in water surface area could be a result of degradation, as it coincided with a decrease in vegetation cover and an increase in degraded areas. Future research needs to investigate the hydrological response of the sub-catchment to the potential influence of climate, variability, change, and land use land cover (LULC) changes.

Automated summary

This study evaluated the effectiveness of different indices in detecting land surface water changes, with AWEI performing the best in extracting water surface area and effectively eliminating pollution and noise.