Optimizing cereal yields in Morocco’s Favorable agro-ecological zone: Integrating dynamic vegetation modeling, machine learning, and climate projections for adaptive agricultural strategies

Climate change poses significant challenges to cereal crop productivity, highlighting the urgent need for effective adaptation strategies to ensure agricultural sustainability. This study investigates the impact of sowing dates on the yields of soft wheat, durum wheat, and barley in Morocco's Favorable agro-ecological zone, a region of critical national food security importance, under changing climate conditions from 2024 to 2099. Utilizing the CARAIB dynamic vegetation model, Random Forest machine learning, and high-resolution regional climate projections from the Euro-CORDEX initiative, this study provides a robust framework for evidence-based adaptation planning. December sowing dates were identified as the most favorable, producing consistent yield increases of up to 15% compared to the 30 October baseline, while late January and early September sowing dates resulted in significant yield reductions of up to 32%. The study also highlights a shift in optimum sowing dates from November to late December over the century, reflecting the need to align planting schedules with changing climatic conditions. Additionally, average growth cycles for wheat and barley are projected to shorten by 2099, suggesting accelerated phenological development driven by rising temperatures, thereby emphasizing the need for the development of heat- and drought-tolerant varieties with adapted growth durations. Despite this shortening, yield increases under optimal sowing dates result from improved synchronization of critical growth stages with periods of higher precipitation and lower thermal and water stress, particularly during grain filling. These findings provide a scientifically grounded basis for informing adaptation strategies and strengthening the resilience of cereal production systems under accelerating climate change. • Using CARAIB model, machine learning, and RCMs to assess the impact of sowing dates. • December emerges as the optimal sowing period for cereal crops. • Climate change shifts optimal sowing dates from November to December. • Precipitation and water stress are also major drivers of yield variability. • Length of growing periods will decrease, impacting productivity of current varieties.