A data-driven crop model for biomass sorghum growth process simulation.

Accurate simulation of crop growth processes for predicting final yield is critical for optimizing resource management, particularly in regions with variable climates and limited resource availability. This paper proposes a novel data-driven crop model to simulate phenotypic changes during biomass sorghum growth. The model integrates a detailed physiological framework for sorghum development-tracking how phenotypes are determined by genotype, environment, management practices, and their interactions-with data-driven techniques to calibrate genotypic parameters using experimental data. Results demonstrate that the model achieves accurate biomass production predictions and successfully disentangles the effects of environmental and management factors on phenotypic development, even with limited data. This model enhances the accuracy and applicability of biomass sorghum growth and yield prediction models, offering valuable insights for precision agriculture.