• A dual-method multi-metric framework (TOPSIS & Social Choice) evaluates 24 CMIP6 GCMs for a semi-arid basin. • Performance-driven ensemble optimization identifies an 11-member GCM ensemble as optimal. • Projections show earlier peak flows, 8–17% annual runoff decline, and 35–50% summer flow reduction. Quantifying the impacts of climate change on streamflow is crucial for sustainable water resources management, particularly in data-scarce semi-arid basins. To support reliable climate impact assessment in such environments, the robust evaluation and selection of climate models is essential. This study evaluates 24 GCMs for the Karkheh River Basin (KRB) using daily observations and a multi-metric framework. In this framework, each GCM is independently ranked using two multi-criteria ranking methods, namely the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) and the Social Choice method based on the Borda count. To determine an optimal ensemble size, a performance-driven approach was employed, revealing diminishing returns beyond approximately 10–13 models, with 11 GCMs identified as the ideal ensemble. Using the SWAT model calibrated with the SUFI-2 algorithm, monthly streamflow at PayePol station was projected under three Shared Socioeconomic Pathways (SSP1-2.6, SSP2-4.5, and SSP5-8.5) across three periods: 2026–2050, 2051–2075, and 2076–2100. Projections from the selected ensemble indicate a shift in peak flow timing from April–May to March–April, an 8–17% reduction in annual streamflow relative to the historical annual discharge (∼151 m³/s), and a substantial 35–50% decline in summer flows. Uncertainty increases over time and is amplified under higher emissions, particularly in late winter and spring due to increased precipitation variability. Overall, the results consistently indicate earlier peak flows, reduced summer runoff, and increasing uncertainty, offering a robust framework for climate risk assessment and adaptation in data-scarce semi-arid basins. These results support climate-informed and adaptive water resources management in data-scarce semi-arid basins.
Ghasemi et al. (Fri,) studied this question.