Uncertainty-aware MAR planning with spatially explicit data-driven weighting.

Area of Science:

• Hydrogeology and Water Resource Management
• Environmental Science and Engineering
• Geographic Information Systems (GIS) and Spatial Analysis

Background:

• Managed aquifer recharge (MAR) is crucial for augmenting water availability and preserving groundwater quality.
• Traditional MAR suitability assessments often lack robust uncertainty quantification.
• Stakeholder involvement is key in selecting MAR suitability criteria.

Purpose of the Study:

• To develop and apply an uncertainty-aware GIS-based multicriteria decision analysis (MCDA) framework for MAR suitability assessment.
• To reduce subjectivity in criterion weighting and quantify uncertainty in MAR planning.
• To identify robust MAR candidate areas in the Sado and Ribeira do Alentejo River Basins.

Main Methods:

• Integrated GIS-MCDA with stratified stochastic sampling and Moran's I correlogram for weight generation.
• Employed both correlation-based and Principal Component Analysis (PCA)-based weighting approaches.
• Generated suitability-map ensembles to analyze spatial heterogeneity and uncertainty.

Main Results:

• Most of the study area showed low-to-moderate MAR suitability (78% and 65% under PCA- and correlation-based methods, respectively).
• The PCA-based approach exhibited greater variability in criterion importance compared to the correlation-based approach.
• A final classification identified robust MAR candidate zones, conditionally suitable areas, and zones with low suitability or high uncertainty.

Conclusions:

• The proposed uncertainty-aware GIS-MCDA framework offers a transferable methodology for evidence-based MAR planning.
• Explicitly accounting for spatial heterogeneity and autocorrelation improves MAR suitability assessment reliability.
• This approach facilitates informed decision-making and prioritization for MAR projects under uncertainty.