Integration of geospatial-based algorithms for groundwater potential characterization in Keiskamma Catchment of South Africa

Abstract

Groundwater supports over 2.4 billion people across the globe and is critical to food security. The spatial dynamics of groundwater vary from place to place. The irregularity of groundwater resource exploitation is recognized in drought-prone areas, putting pressure on the resource. Hence, accurate groundwater potential characterization is critical for sustainable development and management of groundwater, particularly in drought-prone environments. Therefore, this study aimed at utilizing remote sensing satellite data and geospatial-based (analytical hierarchy process (AHP) and frequency ratio (FR)) algorithms to characterize groundwater potential zones (GWPZs) in the Keiskamma Catchment of South Africa. Seven (7) selected factors, including geology, soil type, slope, rainfall, drainage density, lineament density, and land use land cover, were assigned weights based on the AHP and FR algorithms. The validation results showed that the FR model performed better than the AHP, with the area under curve (AUC) accuracies of 62% and 50%, respectively. Based on the findings of this study, we infer that FR is more reliable than AHP when characterizing GWPZ. Lastly, GWPZ maps produced will be beneficial for improving efficient planning, management strategies, and decision-making.