Simulating the Influences of Soil Water Stress on Leaf Expansion and Senescence of Winter Wheat

Abstract

Current crop models usually need a lot of input information to simulate crop leaf area index and have relatively large simulation errors under soil water stress, which need to be improved. In this study, based on experiments conducted in glass soil columns and field under rainout shelters in four years (2012-2016) in Yangling, Shaanxi Province in China, we attempted to establish a dynamic model for simulating leaf expansion and senescence of winter wheat under soil water stress. First, a temperature response function was established with four cardinal temperatures (base temperature, lower optimum temperature, higher optimum temperature, and maximum temperature). Then two soil water stress functions were established to quantify the effect of soil water stress on the processes of leaf expansion and leaf senescence of main stem per plant. The first order derivative of a logistic function was then modified with the temperature and soil water stress response function and was used to simulate the daily rate of leaf area expansion and senescence of main stem. The parameters of the new model were estimated using the Solver add-in in MS Excel and then validated based on the data of soil column experiments in 2014-2015 growing season. Then the new model was evaluated with another dataset of column experiments conducted in 2015-2016. Furthermore, the influence of soil water stress on wheat tillering was included in the new model based on open field experiments in 2012-2013 growing season. Then, the new leaf area index (LAI) simulation model was further verified with the data of open field experiments under rainout shelter (2013-2014) and rainfed condition (2004-2005, 2005-2006, and 2008-2009) in different sites. A comparison was conducted between new LAI model and the original LAI module of CERES-Wheat. The results showed that the leaf expansion rate was not affected when the relative soil water availability was greater than 0.7; water stress inhibited the leaf expansion when relative soil water availability was between 0.2 and 0.7; leaves withered and yellowed when relative soil water availability was less than 0.2. The overall average root mean square error (RMSE) and residual accumulation coefficient (CRM) were 9.78 cm2 plant−1, -0.03, and 6.51 cm2 plant−1; 0.05 for model calibration and validation, respectively. The RMSE of LAI decreased by an average of 47.89% compared with the original LAI module of CERES-Wheat. The results of this study can help to improve the current CERES-Wheat model for model applications in the arid and semi-arid regions of China.