“…Furthermore, well performance data, particularly yield, are subject to a high degree of uncertainty, limiting the effectiveness of these relationships for well siting and productivity mapping (Chilton & Foster, 1995; Taylor & Howard, 2000). Recent studies addressed the issue of predicting the potential productivity of WBAs by focusing on upscaling and mapping of the hydrogeological properties with various approaches: geostatistical methods (Razack & Lasm, 2006); relationships between geophysical and hydrogeological properties (Chandra et al, 2008; Meju et al, 2001; Shishaye et al, 2019); multicriteria analysis using GIS (Lachassagne et al, 2001); large‐scale fluctuations in hydraulic heads (Dewandel et al, 2017) in combination with satellite data (Dewandel et al, 2017; Macdonald & Edmunds, 2014); the integration of borehole data and geological mapping (Courtois et al, 2010; MacDonald et al, 2012); and probabilistic and machine learning models (Ahmed & Mansor, 2018).…”