Improvement and testing of SWAT for multi-source irrigation systems with paddy rice

Abstract

According to the hydrological characteristics of a multi-source irrigation system with paddy rice, the Soil and Water Assessment Tool (SWAT) is improved to develop a distributed hydro-agronomic model. A technique with three critical depths is incorporated to control the irrigation and drainage of paddy fields. In addition, the evapotranspiration (ET) of a paddy field is estimated based on two types of water storage conditions within SWAT. Specifically, we propose a multi-source auto-irrigation framework for SWAT to estimate agricultural irrigation water consumptions (AIWCs) from different water sources. Furthermore, we apply the improved SWAT to the Yangshudang (YSD) basin, located in the Zhanghe Irrigation System (ZIS) of southern China, and the simulation results are compared with the observed data as well as with those of the original SWAT. The results indicate that compared with the original SWAT, the Nash-Sutcliffe efficiency (NSE) of daily discharge increases from 0.48 to 0.80 in the calibration period and from 0.68 to 0.84 in the validation period; in addition, the NSE of the daily ET increases from −0.09 to 0.90, and other evaluation metrics also increase. Moreover, the percent bias (PBIAS) between the total AIWC simulated by the improved SWAT and the observed value is 2.75%, and that is −48.59% for the original SWAT. In particular, the improved SWAT estimates AIWCs from local water sources (ponds and rivers) and the main Zhanghe Reservoir, and the simulated AIWCs approximate the observed values. Therefore, the improved SWAT is more effective and suitable for the simulation of hydrological processes than the original SWAT in the multi-source irrigation systems with paddy rice, and it can estimate the AIWCs from different water sources, which can provide irrigation management decision support for irrigation managers.