Simulating and analyzing climate change impacts on crop yields in Morocco

Morocco’s cereal sector, which occupies over 59% of the nation’s cultivated land and constitutes a major pillar of national food security, faces mounting challenges due to climate change. Increased drought frequency, rising temperatures, and erratic rainfall patterns are already destabilizing cereal yields and heightening reliance on imports. Developing robust, data-driven adaptation strategies is thus crucial to sustain productivity, support rural livelihoods, and inform agricultural policy.
This thesis evaluates how Morocco’s cereal production can adapt to accelerating climate risks while maintaining long-term productivity. The research integrates three methodological pillars: (1) calibration and validation of the CARAIB (CARbon Assimilation In the Biosphere) dynamic vegetation model using historical yield data for soft wheat, durum wheat, and barley across all agro-ecological zones; (2) yield simulations under high-resolution regional climate projections; and (3) application of Random Forest machine learning to quantify the impacts of climate variability and agronomic factors.
To explore future vulnerabilities, the study employs climate projections under the RCP 8.5 scenario, which represents a high-emissions “business-as-usual” pathway and reflects the upper bound of climate risk for North Africa. This choice enables a conservative, risk-based assessment and is consistent with recent regional trends in warming and drought.
Key findings reveal that projected yield declines by the 2080s will range from 10% to 35% depending on crop and region, with barley particularly affected. For instance, simulations indicate that the average barley growth cycle will shorten dramatically, from 212 days today to just 163 days by the end of the century. The CO₂ fertilization effect is shown to provide only a temporary buffer: while it may help sustain current yield levels until around 2050, its benefit diminishes substantially thereafter, as drought and heat stress intensify and outweigh any productivity gains.
A major practical contribution of this thesis is the detailed assessment of sowing date optimization in the Favorable agro-ecological zone, which accounts for nearly 40% of national cereal production. By simulating thirteen different sowing windows from early September to late January, the research identifies December sowing as optimal, consistently yielding higher and more stable harvests across both wheat and barley. In contrast, sowing in early September or late January results in yield losses of up to 32%, due to misalignment with rainfall patterns and increased exposure to heat stress during sensitive growth stages. These results highlight the importance of revising sowing calendars as an immediate adaptation lever.
Application of the Random Forest model further clarifies the drivers of yield variability, with precipitation timing, water stress, and temperature extremes during flowering and grain-filling stages emerging as the most influential factors. The study’s results also signal an urgent need to re-evaluate the suitability of current cereal varieties, as shortened growth cycles and shifting climatic conditions could render existing cultivars less viable within the coming decades.
The integrated framework developed here, which combines dynamic vegetation modeling, regional climate projections, machine learning model and advanced statistical tools, offers actionable insights for adaptation planning. Optimizing sowing calendars, accelerating the breeding of short-cycle, drought-resilient varieties, and strengthening seasonal forecasting capacities emerge as central pillars for building a climate-resilient cereal sector in Morocco.
In conclusion, this thesis bridges the gap between climate science and agricultural policy, delivering robust, quantitative recommendations to guide both immediate and long-term adaptation strategies. The approaches and findings are also transferable to other semi-arid regions facing similar threats to food security. By supporting evidence-based decision making, this research contributes to Morocco’s national development goals and aligns with the Sustainable Development Goals for climate adaptation and food security.