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

W B M A C Bandara,Preecha Kapetch,Tamotsu Nakandakari,Shinya Nakamura,Mitsumasa Anan,Kazuhito Sakai,R H K Rathnappriya,Hideki Setouchi

doi:10.3390/agronomy11071379

Abstract

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

Highlights

Sugarcane is an important crop for sugar and bioenergy worldwide
Many crop models are available for sugarcane, including Agricultural Production Systems sIMulator (APSIM)Sugarcane [9], DSSAT-Canegro [10], MOSICAS [11], STICS-Sugarcane [12], QCane [13]
The current study was focused to use a differential evolution (DE) algorithm for global optimization of cultivar trait parameters implemented in the APSIM-Sugarcane model

Summary

Introduction

Sugarcane is an important crop for sugar and bioenergy worldwide. As the fourthlargest sugar producer and the second-largest sugar exporter in the world, sugarcane has become the most important crop in Thailand’s agriculture [1]. recent evidence indicates that Thailand’s sugarcane production is highly affected by climate change. Identification of suitable management strategies to cope with such temporal and spatial variability of sugarcane production has become important for the Thai sugarcane industry. In this respect, process-based crop models are advantageous. Since they can be used for assessing climate impacts on sugarcane [3,4,5] evaluating cultivar responses under various environments and management strategies, they can predict yield [6,7] and provide information for economic and policy decision-making [8]. APSIM-Sugarcane is one of the most widely-used platforms for the modelling and simulation of sugarcane production systems [14,15]

Objectives

Methods

Results