Improving CERES-Maize for simulating maize growth and yield under water stress conditions

Abstract

• The CERES-Maize simulations were poor under the water stress conditions, especially with the seedling stage water stress. • The soil water conductivity coefficients were modified to improve maize growth simulations under water stress conditions. • The severe water stress during the seedling stage damaged the photosynthetic membrane and reduced RUE of maize. • A RUE factor was applied to the optimal RUE in the model to reflect the membrane damages on the maize growth. Water stress in different maize growth stages has different effects on maize growth and development. Accurately simulating these effects with a maize model are important to agriculture water management and savings. The objective of this study was to improve the performance of the Crop-Environment Resource Synthesis Maize (CERES-Maize) model for simulating maize (Zea mays L.) growth and yield under water stress conditions. The model was evaluated based on the measurements of phenology date, leaf area index (LAI), aboveground biomass, kernel weight, yield, and soil water content. Those data were collected from the planting to maturity of maize under both water stress and full irrigation conditions in the Guanzhong Plain, China from 2013 to 2015. Our results found that the model simulated those variables realistically under the full irrigation condition, while the simulations were poor under the water stress conditions, especially with the seedling stage water stress treatments. In this study, we included the water stress factor in the phenology calculation and modified the soil water conductivity coefficients to increase root water uptake and improve maize growth and yield simulations under the water stress conditions. In addition, we found the severe water stress during the seedling stage damaged the photosynthetic membrane and reduced the chlorophyll content in the leaf based on observations, significantly affecting the radiation use efficiency (RUE) during the later stages of maize growth without water stress. A linearly regressed RUE factor was applied to the optimal RUE in the CERES-Maize model to reflect such damages on the maize growth and yield. Our simulations of the aboveground biomass and yield were further improved by considering such a factor. Additional data were used to validate our modified model, and the simulated biomass and yield were also improved when compared to those generated by the original model. The results from this study provide strong insights into better maize growth and yield simulations in the water limited regions.