Hydrological simulation using the SWAT model in a semi-arid region in the southern part of Zacatecas, Mexico

Today, hydrological models are important tools for successfully managing water resources and understanding the hydrologic response of watersheds, especially in arid regions with limited water resources. The purpose of this study is to evaluate the potential of the Soil Water Assessment Tools model (SWAT) for predicting the hydrological regime of El Chique Dam watershed, in a semi-arid region of the state of Zacatecas, Mexico. The model was calibrated and validated on a daily and a monthly scale using the sequential uncertainty fitting algorithm (SUFI-2) within the Calibration and Uncertainty Program SWAT-CUP. The performance of the SWAT model was evaluated by means of four statistical measures: the Nash–Sutcliffe coefficient (NSE), the determination coefficient (R 2 ), the Percentage of Bias (PBIAS) and the mean square error (RSR). The results showed a better performance of the model in the simulation on a monthly scale, obtaining values of NSE, R 2 , PBIAS and RSR of 0.78, 0.79, -14.00 and 0.47 during the calibration period, and 0.60, 0.61, -18.70 and 0.63 during the validation period, respectively. The most sensitive parameters that affect the behavior of runoff in the watershed are: ALPHA_BF, CH_K2, CN2, SOL_BD and OV_N. The overall hydrological water balance analysis showed that more than 90% of precipitation is lost by evapotranspiration, while 8.6% of precipitation contributed to streamflow. Thus, the SWAT model can predict runoff on a monthly scale very well and therefore, these results are expected to assist in parameter selection for calibration processes in other semi-arid regions, as well as in decision-making related to water resource management and planning in the watershed under study.