How important is the intra-regional soil heterogeneity for the design of future stress-avoidant wheat ideotypes? A modeling study in central France

Abstract Accurate projections of crop adaptation to climate change require accounting for the spatial heterogeneity of soils, which modulates both water availability and the effectiveness of genetic adaptation. Using the process-based crop model Sirius, we investigated how intra-regional variability in soil available water capacity (AWC) influences wheat yields and the adaptive value of stress-avoidant ideotypes under future climates in central France (Limagne plain). Detailed soil databases were aggregated across five representative sites and combined with multiple climate projections (CMIP6), two emission pathways (SSP2-4.5 and SSP5-8.5), and three time horizons (2031–2050, 2051-2070 and 2071-2090). Variance decomposition revealed that soil AWC accounted for 23% of the simulated yield variability, significantly exceeding the contribution of local climate contrasts (10%), a pattern consistent across current and future periods. Deep soils (>80 mm AWC) buffered drought effects whereas yields stagnated in shallow soils (<80 mm AWC) where water deficits persisted despite phenology hastening. On average, the reference cultivar showed earlier anthesis by 8–21 days under future climates, leading to higher yields mainly in deep soils. Optimization of flowering timing through stress-avoidant ideotypes provided mean yield gains of +6.33 dt·ha−1 in deep soils, but limited benefits (+1.71 dt.ha-1) in shallow ones, highlighting pedological dependence of breeding efficiency. Advancing anthesis also increased exposure to early-spring frost: frost probability rose from 0.4 when flowering occurred more than 250 °C.days earlier, particularly in the frost-prone part of the study area. Hence, frost risk remains a critical constraint for early ideotypes, even under strong warming. Overall, our results demonstrate that intra-regional soil heterogeneity remains a dominant driver of wheat yield variability and adaptation potential under climate change. Designing stress-avoidant ideotypes without explicit consideration of local soil AWC could lead to maladaptation, especially in regions with shallow soils represent a significant portion of cropped areas. In such situation, breeding for terminal stress avoidance may offer only limited benefit. We advocate that breeding and modeling frameworks integrate high-resolution soil data to refine regional ideotype design, reconcile terminal-stress avoidance with frost tolerance, and better capture the spatial realism required for sustainable crop adaptation strategies. Highlights - Local soil water capacity limits wheat adaptation to climate change. - Deep soils favor earlier, stress-avoidant ideotypes. - Shallow soils restrict the benefits of phenological adjustment for stress avoidance. - Frost exposure remains a key risk when shifting phenology toward earliness. Competing Interest Statement The authors have declared no competing interest.