Global Optimization of Cultivar Trait Parameters in the Simulation of Sugarcane Phenology Using Gaussian Process Emulation

The global optimization of parameters in process-based crop models is often considered computationally expensive. Gaussian process (GP) emulation a widely used method for reducing the computational burden process. Total above-ground biomass and cane dry weight three Thai sugarcane cultivars (KK3, LK92-11 02-2-058) collected under rainfed irrigated conditions were to optimize cultivar-specific Agricultural Production Systems sIMulator (APSIM)-Sugarcane model through GP emulation. emulators trained validated approximate APSIM-Sugarcane then optimizing differential evolution algorithm. Resulting optimized allowed obtain simulations that quite well approximated observed CDW (validation results between simulated yields: R2 0.93–0.98; normalized root mean squared error: 5–22%; Willmott’s agreement index: 0.87–0.99). best parametrization was obtained lowest water stressed conditions. Based on these results, we suggest can be efficiently implemented parameterization expensive simulators.