Towards a Robust Rainwater Harvesting Site Selection: Integrating Hydrological Dynamics into GIS-Based Decision Models

Water scarcity is recognized as a critical global challenge, particularly affecting arid and semi-arid regions where climate change and recurrent droughts reduce water availability while demand continues to increase. This growing imbalance threatens the achievement of key Sustainable Development Goals, especially universal access to safe water and sanitation by 2030. In this context, Rainwater Harvesting (RWH) has emerged as a viable strategy to mitigate water scarcity and improve local water resilience. The effectiveness of RWH systems is strongly influenced by hydrological and geomorphological conditions such as rainfall patterns, topography, and runoff behavior. Factors like slope, drainage characteristics, and land use determine how water flows, accumulates, and can potentially be captured. Therefore, selecting suitable locations for RWH structures requires an integrated assessment of these physical characteristics. Recent advances in spatial analysis have led to the development of GIS-based Multi-Criteria Decision Analysis (MCDA) frameworks, which combine various environmental and hydrological criteria into suitability maps. These approaches typically rely on composite indices that integrate parameters such as slope, drainage density, land use, and runoff coefficients. While effective for large-scale screening, conventional approaches are often limited because they rely on local runoff indicators and neglect the cumulative processes of water movement across a catchment. This dissertation proposes an enhanced methodology by introducing a new criterion called Aggregated Runoff (AR), which accounts for upstream runoff contributions through flow accumulation. Unlike traditional pixel-based approaches, AR represents the cumulative nature of runoff, providing a more physically consistent representation of how water is distributed across a basin. This criterion is integrated into an AHP/GIS framework alongside geomorphological parameters, enabling the generation of improved suitability maps for RWH applications. A further innovation is the use of a quantile-based classification system, which defines suitability classes based on the spatial distribution of runoff values rather than fixed thresholds. This approach improves robustness and adaptability by reducing sensitivity to variations in rainfall and basin size, allowing the methodology to be applied across different climatic and geomorphic contexts. The study evaluates the performance of the proposed framework over a 72-year period (1951–2022), enabling a detailed analysis of spatial and temporal variability. The results demonstrate that conventional runoff coefficient-based methods show significant fluctuations in suitability patterns under changing rainfall conditions. In contrast, the AR-based index exhibits greater temporal stability and maintains consistent identification of areas with higher runoff accumulation potential. Finally, the methodology is tested in a new basin with different climatic characteristics to assess its transferability. Despite strong seasonal variability, the approach performs consistently, demonstrating its applicability across diverse conditions. Overall, the integration of aggregated runoff and quantile-based classification enhances the reliability and scalability of RWH site selection, supporting improved decision-making for sustainable water resource management under climate variability.