Calibration and application of DSSAT Model to estimate area yield index insurance premium for common bean at different growth stages and weather perils

CROPGRO-Dry bean model of Decision Support Tool for Agro technology Transfer (DSSAT) was calibrated and validated using observed yield data collected from field experiments conducted in three bean growing areas of Rwanda. These experiments were conducted to identify the most critical bean growth stage to natural bean disease pressure, drought stress and waterlogging and their impact on yield losses. The calibrated DSSAT showed good agreement with the observed bean yields with relative root mean square error (RRMSE) of around 0.5 for both drought and waterlogging and around 1.5 for no pesticide application. The findings revealed that GROPGRO-Dry bean model was an effective tool for predicting bean yield under the investigated crop growing conditions. The estimated yield losses due to investigated treatments were further used to estimate corresponding insurance premium rate of an area yield index insurance model. Drought affected most the seed filling stage with bean yield reduction rate estimated at 23% and a corresponding premium rate of 257 kg ha-1. Pod setting stage was the most sensitive to natural bean disease pressure with yield reduction rate of 30% and a corresponding premium rate of 467 kg ha-1. Seed filling was most sensitive to waterlogging with a yield reduction of 28% and a corresponding premium rate of 429 kg ha-1. This study provides scientific knowledge and basis that can inform insurers to set up area yield index insurance sub-products targeting to cover weather related perils with specific focus on crop developmental stages.