GIS and Google Earth Engine map Zimbabwe’s groundwater potential nationwide

Researchers have built a national-scale framework for assessing groundwater potential in Zimbabwe, combining geographic information systems, remote sensing, the Analytical Hierarchy Process and Google Earth Engine. The study, published in Earth Science Informatics, addresses a persistent problem in data-scarce crystalline basement terrains, where reliable groundwater assessment remains difficult.

The team combined eleven conditioning factors through an AHP-weighted overlay to delineate five Groundwater Potential Zones across the country. Geology received the largest weight at 24.09 percent, followed by lineament density at 20.46 percent, rainfall at 16.76 percent and drainage density at 16.16 percent. The resulting map classified moderate potential over 46.0 percent of Zimbabwe, high and very high potential over 42.5 percent, and low and very low potential over the remaining 11.5 percent.

To evaluate the map, the researchers compared it against 9,695 geolocated boreholes from the national ZINWA database, using statistics that are robust to skew and clustering. Mean borehole yield increased monotonically across the zones, from 1.33 litres per second in the Very Low potential class to 3.96 litres per second in the Very High class. The share of high-yielding boreholes exceeding 5 litres per second rose from 2.9 percent to 9.9 percent across the same gradient, a trend significant by the Cochran–Armitage test at p < 0.001, though the significance weakened to p = 0.02 after 1 km spatial thinning of the borehole data.

Not every measure supported the zoning equally. Median yields were similar across the potential classes, and point-level agreement between the mapped zones and individual borehole yields was weak, with a Spearman correlation of 0.02 and Cohen’s kappa of 0.02. The authors interpret this pattern as consistent with fracture-controlled yields in basement aquifers, where local fracture networks can dominate productivity at individual sites regardless of the regional classification.

The study also examined storage dynamics using GRACE terrestrial water-storage anomalies and CHIRPS rainfall data for 2002 to 2016. The two showed a positive association with an R² of 0.78 and a one-to-two-month lag, which the researchers interpret as a regional storage response to rainfall. The authors position the framework as a transparent national screening tool that identifies where productive boreholes are more likely to be found, while emphasising that site-specific investigation remains essential for actual borehole siting.

Subject of Research: A national GIS and AHP-based groundwater potential assessment of Zimbabwe validated against borehole yield data

Article Title: Integrating GIS, multi-criteria decision analysis and google earth engine for national-scale groundwater potential assessment in Zimbabwe

Article References: Gumindoga, W., Chikwiramakomo, L., Chisadza, B., Rwasoka, D. T., & Muchaka, F. A. (2026). Integrating GIS, multi-criteria decision analysis and google earth engine for national-scale groundwater potential assessment in Zimbabwe. Earth Science Informatics, 19(12), Article 218. https://doi.org/10.1007/s12145-026-02270-z

Image Credits: AI Generated

DOI: 10.1007/s12145-026-02270-z

Keywords: groundwater, Zimbabwe, GIS, Google Earth Engine, Analytical Hierarchy Process, remote sensing, borehole yield, crystalline basement aquifers, GRACE, CHIRPS, groundwater potential zones, Integrating