Evaluation index comparison of different irrigation water use efficiency based on DSSAT simulation

Abstract: To properly assess irrigation water use efficiency, the definite and comparable indicators are urgently needed. Many evaluation indexes of irrigation water use efficiency are used, which often cause confusion. In order to quantify their differences, correlation, influence factors and application conditions, a comparative study was carried out in this paper. The winter wheat, the most important staple crop in China, was chosen as an example. The Decision support system for agrotechnolgy transfer (DSSAT) CERES-Wheat model, which was driven by daily weather data, soil data and crop management data, was applied to construct two scenarios, namely, irrigated and rainfed of winter wheat in Xinxiang area of Henan province, using data from 1963 to 2013. And the rainfed yield, rainfall of the growth period, required irrigation water, evaporation, irrigation yield in irrigation situation were simulated by DSSAT- CERES-Wheat model. Thus, five evaluation indexes of irrigation water use efficiency, namely, farmland total water use efficiency(WUEg), generalized water resources use efficiency(WUEu), irrigation water use efficiency(WUEi) and crop water productivity(WP) in irrigation situation, real irrigation water use efficiency(WUEti) by rainfed and irrigation situation synthetically, were obtained. The results showed that WUEg and WUEu, WUEg and WP, and WUEu and WP had strong positive linear correlation, and correlation coefficients were all greater than 0.8. WUEi and WUEti were negatively correlated. The average value of WUEg, WUEu and WP for winter wheat in Xinxiang area during 1963-2013 were 1.43, 1.48 and 1.63 kg/m3, respectively. Three results had the consistent change trend, and were on the rise over time. The value of WUEi was larger than WUEti. The mean values of two indicators were 2.46 and 1.49 kg/m3, respectively. And the change of interannual evaluation value for the two indexes was inconsistent. Rainfall of growth period had influence on these indexes. WUEg and WP had a downward trend with the increase of rainfall, and the two indexes remained consistent. With the increase of rainfall in growth period, WUEi showed a trend of increase while WUEti only fluctuated, and the differences between these two indexes increased significantly. The application of WUEi was confined to the influence of rainfall. Compared to WUEi, WUEti had more reasonable definition, but the increased yield in irrigation situation, in contrast to rainfed situation, was difficultly acquired. So WUEi and WUEti are not practical for irrigation water use efficiency evaluation. The values of WUEg and WUEu were both about 1.45 under different hydrological years. For all the five indexes, there was no significant change in different hydrological year. These conclusions provide information for evaluation of agricultural water resource.