Adapting the CROPGRO Legume Model to Simulate Growth of Faba Bean

Abstract

Crop growth models are valuable tools for integrating our understanding of physiological processes and for hypothesizing and evaluating crop management strategies. Our objective was to adapt the CROPGRO model to simulate the growth and yield of faba bean (Vicia faba L.). The CROPGRO model simulates different grain legume species using external parameter files that describe species process sensitivity to environment plus files describing cultivar differences. Specific objectives were to develop a species file and one cultivar file for faba bean based on: (i) values and relationships from the literature and (ii) comparison with observed growth data on faba bean grown in Cordoba (Spain). A systematic procedure was followed to develop relationships and values for the species and cultivar files. Base temperatures for processes of this cool‐season legume are typically between 0 to 2°C for photosynthetic, vegetative, and reproductive processes while corresponding optimum temperatures vary from 22 to 30°C. After adaptation, the model accurately predicted total crop dry matter accumulation, pod mass, and partitioning to plant components. High seed yields of faba bean exceeding 6000 kg ha−1 were predicted in agreement with observed data. Sensitivity analyses on sowing date showed optimum yields from early winter sowing at Cordoba and late winter sowing in northern Europe. Adapting an existing mechanistic model such as CROPGRO had advantages because many processes were similar across species and well simulated. The primary adaptation required was to change the cardinal temperature sensitivities of growth processes, based on faba bean literature or by analogy to other species.