Assessing climate change impacts on Mediterranean durum wheat using DSSAT-CERES-Wheat across contrasting environments

The Mediterranean basin is a climate change hotspot, and this will strongly affect key crops such as durum wheat, a staple for millions and a major commodity in southern Europe. Future productivity remains uncertain, as climate change introduces both limiting and beneficial factors. Understanding crop responses is essential to design effective adaptation strategies and ensure food security.

This study assesses the DSSAT-CERES-Wheat model’s ability to simulate durum wheat growth and yield under climate change across three contrasting Mediterranean environments and support adaptive strategies. The model was calibrated for Tirex, a widely cultivated variety, using long-term field data and Monte Carlo optimization, then evaluated with three independent trials in northern (Rovigo), central (Agugliano), and southern (Genzano) Italy, incorporating leaf area index data from Sentinel-2. Two CMIP5 scenarios (RCP 4.5 and RCP 8.5) and three temporal horizons (baseline 1991–2020, near future 2022–2050, far future 2070–2100) were used to simulate yields, evaluate irrigation as adaptative strategy, and assess water use efficiency (WUE).

Calibration showed strong performance for grain yield (R²=0.98, d-stat=0.98, RMSE=0.3 t/ha), canopy biomass (R²=0.98, d-stat=0.62, RMSE=3 t/ha), and anthesis (R²=0.98, d-stat=0.84, RMSE=7.6 days). Evaluation confirmed good agreement for yield and biomass across sites, while LAI was less accurate.

At Rovigo, climate change reduced yields most under RCP 8.5 near future (-1.8 t/ha, -40%). At Agugliano, responses depended on agronomic management: under enhanced conventional (standard nutrition, supplemental irrigation, and integrated pest management), yields declined most under RCP 4.5 near future (-1.8 t/ha, -28%) but increased under RCP 8.5 far future (+1.0 t/ha, +15%). Under zero-stress (fertigation and chemical pest control), yields increased across all scenarios, reaching the highest gain of +1.5 t/ha (+17%) for RCP 8.5 far future. At Genzano, limited effects were observed, with the largest near-future decline (-0.3 t ha⁻¹, -11%) and a far-future increase (+0.7 t/ha, +30%) under RCP 8.5. The achievement of higher yields in the far future compared to the near future across all scenarios may be due to projected increases in atmospheric CO₂, thereby partially offsetting yield losses caused by changes in temperature and precipitation.

Full irrigation mitigated climate impacts. At Rovigo, it led to +3.9 t ha⁻¹ (+84%) under RCP 8.5 far future and improved WUE from 5 to 10–14 kg/mm. At Agugliano, irrigation increased yields under all scenarios, with the largest gain under RCP 8.5 far future (+3.6 t ha⁻¹, +54%) while maintaining a WUE of 10–15 kg/mm. At Genzano, irrigation produced smaller absolute but higher relative gain (+2.0 t/ha, +90%) under RCP 8.5 far future and WUE slightly increased.

Full irrigation effectively stabilizes and increases wheat yields, but only modestly improves WUE, indicating that higher yields do not necessarily translate into greater water efficiency. Full irrigation should therefore be considered a theoretical upper limit rather than a realistic large-scale strategy due to water availability, cost, and infrastructure constraints. Effective climate adaptation in Mediterranean wheat requires combining agronomic management adjustments and genotype choice, supported by crop modeling to assess trade-offs among productivity, water use, and environmental sustainability.