Improving Simulations of the Sugarcane Crop Based on a Combined Agro-Hydrological Model.

Sugarcane has emerged as the second largest source of biofuel and the main source of sugar in the world. It is a major crop of social, economic and environmental importance in many tropical countries, including Brazil, currently the largest producer (50% of global production). The optimization of crop inputs (fertilizer and water) and climatic risks analysis can and has been assessed by process based crop simulation models. In contrast to other crops, however, sugarcane has only two simulation models available for end users. Moreover, modifications of these models are required to better assess and support sustainable sugarcane production in Brazil, notably in terms of water limited since more than 90% of Brazilian sugarcane fields are rainfed. Therefore, this paper aimed to couple a new sugarcane PBM to the agrohydrological SWAP model in order to reduce the prediction uncertainty and provide a better tool for crop water consumption and hydrological analysis. The SWAP-Sugarcane model exhibited slight better performance to simulate stalk fresh and dry mass (3 and 7 % lower RMSE) and notably better leaf area index simulations performance (38% lower RMSE) than the standalone version. Soil water balance simulations were improved (32% lower RMSE) compared to the initial water balance routine, enabling more robust and less uncertain simulations of crop water consumption.