Assessment of SWAT Hydrological Model in Catchments' Water Balance Simulation Located in Semi-Arid Regions (Case Study: Zayandeh-Rud River Basin)

Understanding the concept of water balance is one of the most important prerequisites for sustainable management of water resources in the watersheds. Therefore, the components of water resources in a catchment system should be compared at different time periods, and also the effect of each of them should be identified on varied hydraulic components of the hydrological systems. The SWAT model is an example of a physically based hydrologic model which can be used for large-scale simulating and monitoring of water cycle processes based on the characteristics of the catchment area and its climatic conditions. The main object of this study is the hydrologic simulation and water balance estimation for the period 2000-2009 in the Zayandeh-Rud River Basin. The Zayandeh-Rud River Basin is located in the arid and semi-arid central region of Iran. This area is very variable in terms of rainfall. As well as the state of water resources and water consumption is very complicated in this catchment. In the present study, the soil and water assessment tool (SWAT) used to simulate water balance in the Zayandeh-Rud River Basin. The input required data included digital elevation model, land use map, soil texture map and meteorological information including daily rainfall data and minimum and maximum temperature data were introduced to the model and the model was implemented with these data. The sensitivity of the flow-effective parameters was determined using the p-value and t-state criteria by the SUFI2 algorithm in the SWAT-CUP program. The model was calibrated monthly and validated with the selected parameters in the sensitivity analysis using the Nash-Sutcliff criteria and the coefficient of determination by the application of the data of six stations including. Calibration of the model was conducted for 2000-2006 and validation of the model for the years 2007-2009. The results of sensitivity analysis showed that considering the characteristics of the study area, the SWAT model is more sensitive to the 17 effective parameters on runoff. The selected parameters also confirm the results of previous research carried out in the region. The sensitive parameters selected in the sensitivity analysis step were used to calibrate the model. In the next step, the parameters of SWAT-CUP software were entered. After that, these parameters were repeated 1000 times with the SUFI2 algorithm, and the optimal value for each parameter was determined. The Nash-Sutcliff coefficient and the coefficient of determination in the six hydrometric stations are greater than 0.56 and 0.69 in calibration and verification periods respectively, which indicates that the model has a satisfactory ability to run in runoff simulation. The contribution of the components of the water balance including evapotranspiration, surface runoff, lateral flow, groundwater flow, and deep aquifer recharge was calculated from annual basin precipitation. The amount of extracted water from the hydrological components indicated that the largest share of the water balance was related to actual evapotranspiration, the range of variations in the type of precipitation in the study area ranged from 60.1% (2000) to 92.7 % (2007). After evapotranspiration, surface runoff with a change of 22.2% (2005) to 8.6% (2009) and groundwater flow with a change of 14.2% (2000) to 20.5% (the year 2007) had relatively high fluctuations and a large share in the basin balance. These results indicate that the lateral flow with a range of 3.1 to 1.9% had no significant change in these years. Also, the deep aquifer recharge with the range of 1.2 to1.5% was the lowest in 2003 and 2009, respectively. The results showed that the calibrated model for the Zayandeh-Rud River Basin had a desirable performance for both calibration and validation periods. Therefore, the SWAT model has acceptable performance for simulating the water balance of the area. In addition, the results of this study showed that 65.98% of the total annual precipitation in the basin is in form of evapotranspiration, which compares to the other water balance components has the highest part. As well as surface runoff with 15%, groundwater flow with 13.7%, lateral flow with 1.5%, and deep aquifer recharge with 0.8% have other parts of the water balance components in Zayandeh-Rud River Basin. The results also indicate that the highest water losses in the soil and groundwater resources of the basin are due to evapotranspiration. Therefore, serious measures to prevent the loss of water through evapotranspiration in the region to be necessary. The results of this research can be used to predict the effects of climate change and the applicable management practices in the region, which are presented in scenarios to the model.