Assessment and comparison of crop growth models for estimating wheat production in a semi-arid region of Morocco

Faced with growing food security challenges influenced by global factors such as population growth, climate change, and soil erosion, the need for sustainable agricultural practices is particularly relevant in Africa. In Morocco, wheat is the most dominant crop, but its production is highly dependent on rainfall. In this research, we evaluate several crop growth models, including AquaCrop, among others, focusing on their ability to effectively improve crop production predictions and yield gap analysis in Morocco. This evaluation is essential to develop adaptive agricultural practices that can mitigate the adverse effects of climate change on crop yields. This study employs AquaCrop-OSPy (ACOSP), an open-source Python version of the AquaCrop model, to simulate various indicators of crop growth such as canopy cover (CC), actual evapotranspiration (ETcact), biomass, and grain yield (GY) for wheat under drip irrigation in the semi-arid Chichaoua region of Marrakech in Morocco. The model was first calibrated by using the field data collected over two wheat fields during the 2016/2017 cropping season. Key parameters affecting CC, ETcact, biomass, and GY were calibrated by comparing field measurements with the model outputs. Then, model validation was carried out on the same fields but during the 2017/2018 cropping season. The results demonstrated that ACOSP effectively simulates CC, ETcact, biomass, and GY across two growing seasons. The comparative analysis between observed and simulated parameters yielded the following average values: for CC, R²=95%, RMSE=8.5%, and MSE=1.1%; for ETcact, R²=76%, RMSE=0.61 mm/day, and MSE=0.40 mm/day; and biomass, R²=87%, RMSE=0.22 t/ha, and MSE=0.05 t/ha during the calibration season. GY recorded was 3.87 t/ha. In the validation season, the model achieved similar accuracy for CC R²=95%, RMSE=8.0%, MSE=1.0 %; and biomass R²=91%, RMSE=0.15 t/ha, MSE=0.05 t/ha; with a GY of 3.29 t/ha. These results confirm the model's reliability in simulating key growth parameters of wheat in a semi-arid environment. Two main aspects are addressed through this study: firstly, to provide valuable information for agricultural policy and decision-making in Morocco, and secondly, to enrich the international conversation on sustainable agricultural practices, particularly in arid and semi-arid regions. Leveraging the findings of efficient simulation of wheat growth and production using the ACOSP model, this research provides a solid basis for local, national, and international key actors in developing robust strategies to improve wheat production, thus enhancing the sustainability and resilience of Moroccan agriculture.