Water Stress is a Key Factor Influencing the Parameter Sensitivity of the WOFOST Model in Different Agro-Meteorological Conditions

Abstract

Sensitivity analysis is helpful for improving the efficiency and accuracy of the calibration of crop growth models. However, parameter sensitivity is still not well understood when combined with different meteorological and production conditions, especially adverse conditions such as water stress. This study simulated the production of winter wheat in four ecological areas in Henan Province, China. The Extend Fourier Amplitude Sensitivity Test algorithm (EFAST) was used for analyzing the sensitivity of 43 crop parameters of the WOrld FOod STudies (WOFOST) model to yield, aboveground biomass, and leaf area index (LAI) with or without water-limited conditions. The results demonstrated that yield and biomass were the objective outputs, and the main limiting factors for the model results were assimilation and dry matter conversion efficiency. Under water-limited conditions, the parameter sensitivity of related extinction coefficient, early wheat leaf area, and root growth increased with increased water stress. With the process variable LAI as the target output, the parameter sensitivity varied at different growth stages, whereas the parameter sensitivity was almost the same under different agro-meteorological conditions. Under water-limited conditions, the parameter sensitivity of wheat early extinction coefficient, maximum root depth, and death rate of the leaves also increased with increased water stress. Therefore, water stress is a key factor affecting parameter sensitivity under different agro-meteorological conditions.