Abstract:
Abstract: Crop evapotranspiration (ETC) consists of transpiration and soil evaporation. It is important to know the proportion of transpiration and soil evaporation in field of agricultural water saving. The dual crop coefficient approach can calculate transpiration and soil evaporation by dividing a crop coefficient (Kc) into a basal crop coefficient (Kcb) and a soil evaporation coefficient (Ke). In this study, we used SIMDualKc model, a computer model developed by using the theory of dual crop coefficient approach, to calculate evapotranspiration and soil evaporation of winter wheat under different irrigation schedules in arid region Northwest China. Before modeling, a 2-year water-controlled experiment of winter wheat was conducted to calibrate and validate SIMDualKc model. Evapotranspiration was observed by a large-scale weighing lysimeter or calculated by water balance approach, and soil evaporation in 2013-2014 was measured by a micro-lysimeter. We compared the observed evapotranspiration and soil evaporation with the simulated ones. The regression coefficient (b), determination coefficient (R2), root mean square error (RMSE), average absolute error (AAE), Nash-Sutcliffe efficiency (NSE), RMSE-observations standard deviation ratio (RSR) and index of agreement were used to evaluate the errors between observed and simulated evapotranspiration and soil evaporation. The results showed that the SIMDualKc model could accurately simulate evapotranspiration and soil evaporation of winter wheat under different irrigation schedules. Modeling result showed that the key water requirement period of winter wheat was from jointing to grain filling stage, and the soil evaporation ratio in the whole winter wheat growth stage was highest in the mid season stage, followed by the late season stage, crop development stage, and initial stage. On the basis of the simulation, we investigated the relationship between soil evaporation ratio and meteorological factors (minimum temperature, maximum temperature, average relative humidity, wind speed at 2 m above ground surface, solar radiation), and crop factor (ground surface coverage) by using the path analysis method. The result showed that the irrigation only affected soil evaporation in a short period, while ground surface coverage affected it most in a long period with a total indirect influential coefficient of -0.857. A regression model of soil evaporation ratio and ground surface coverage was developed by using the soil evaporation ratio simulated by SIMDualKc model and the ground surface observed in the field experiment. The regression model could accurately calculate soil evaporation of winter wheat under different water conditions with determination coefficients 0.721-0.902 and it could be used as a simplified method to calculate soil evaporation. These findings confirmed that SIMDualKc model was a useful tool to study the change of evapotranspiration and soil evaporation under different irrigation schedules. The further study should focus on the integration of SIMDualKc model and other crop growth models.