Assessing Groundwater Potential in Semi-Arid Areas of Uganda: Case study of Agago District.

Abstract

Increasing demand for groundwater resources necessitates effective mapping and categorization of groundwater potential zones to ensure long-term water security, especially in semi-arid regions like Agago District, Uganda. Groundwater serves as the primary water source in the district, yet access is often unreliable due to seasonal variability and limited surface water options. Conventional groundwater exploration methods are costly and provide localized data, creating a need for more efficient and comprehensive approaches that support multiple factor assessment. This study seeks to evaluate groundwater potential in the semi-arid regions of Agago District, Uganda, by determining key factors that impact groundwater availability and modeling groundwater potential zones, aiming to categorize these zones effectively. This study utilized the Delphi technique to attain expert consensus on eight pivotal factors affecting groundwater potential: Rainfall, Geology, Slope, Soil, Land Cover, Drainage Density, Lineament Density, and Distance from Rivers. Esteemed for its efficacy in interdisciplinary domains, the Delphi method transcends conventional groundwater exploration techniques by promoting expert collaboration through its principles of anonymity, iterative rounds, controlled feedback, and statistical response aggregation. Thematic maps were tested for multicollinearity using the Variance Inflation Factor (VIF), indicating factor independence. Factor weights were determined through the AHP, producing a consistency ratio below 0.1, confirming reliability. A weighted overlay analysis was then conducted to map groundwater potential zones. It was found that they could be categorized as poor (2%), low (13%), moderate (51%), high (30%), and very high (4%). Validation with borehole data by comparison with categorized zones indicated a 73% accuracy rate and a ROC analysis yielded an AUC score of 0.724, demonstrating reliable model performance. The highest groundwater potential was identified in the southern district near the Agago River, where alluvial and lacustrine deposits suggest favorable conditions for productive aquifers. Sensitivity analysis highlighted geology, soil, and rainfall as the primary factors influencing model accuracy. This study suggests that areas with high and very high groundwater potential, especially in the southern region, should be prioritized for exploration. Planning authorities should employ groundwater potential maps to inform infrastructure siting, thereby safeguarding infiltration paths. Consistent groundwater mapping is recommended to mitigate climate effects and promote sustainable resource management.