Evaluating soybean yield responses to future climate change and irrigation regimes: a DSSAT multi-model assessment

Introduction Climate change is expected to intensify temperature and precipitation variability in Mediterranean regions, creating uncertainty for soybean production. Methods In this study, the DSSAT-CROPGRO-Soybean model was run with three global climate models under two emission pathways (RCP 4.5 and RCP 8.5) across three future time periods: near-future (2016–2040), mid-century (2041–2070), and late-century (2071–2098). Bias correction significantly enhanced the reliability of climate inputs by reducing systematic temperature deviations and improving agreement with observed meteorological conditions. The model was calibrated and validated using observed phenology, leaf area index (LAI), biomass, and yield across three irrigation treatments (I 100 , I 70 , and I 50 ). Results The model showed good correspondence between observed and simulated values. Taylor diagram analysis revealed correlation coefficients generally exceeding 0.95, coefficients of determination (R²) ranging from 0.74 to 0.99, and acceptable RMSD values across treatments. Future projections indicated that yield responses varied across future periods and irrigation conditions. Under late-century conditions, RCP 8.5 produced higher yields than RCP 4.5 by approximately 4–17% under irrigated conditions and 6–20% under rainfed conditions across the considered GCMs. Discussion Elevated CO 2 partly mitigated the effects of warming; however, seasonal soil water availability remained the primary constraint on yield. The results demonstrated that the calibrated DSSAT-CROPGRO-Soybean model provides a reliable basis for predicting the adverse effects of future climatic conditions on soybean production, while multi-GCM climate projections indicated that the magnitude of these effects may vary substantially depending on emission scenario, projection period, and water availability.