عليرضا نوين پري زاد

The topic of hydrological modeling for the Hirmand watershed holds special significance. Accurate estimates of the water inflow to the Hamoun wetlands are essential for proper planning regarding these wetlands. This information helps to understand the current state of the wetlands and to make necessary plans. Additionally, a treaty between Iran and Afghanistan on managing the Hirmand River, which lies upstream of the Kajaki Dam, necessitates precise monitoring of this river’s status. In negotiations between the two countries, it is crucial to present documented data on the Hirmand River’s status each year and even to provide future predictions. Achieving these goals requires a precise hydrological model that can be calibrated using reanalysis and available ground data, even with limited terrestrial data. In this research, various reanalysis datasets, including ERA5, GLDAS, and CRU, have been used. These datasets have been eva‎luated, modified, and integrated in previous studies. A combined precipitation and temperature dataset based on global ERA5 and GLDAS data has been used for hydrological modeling. The calibration approach in this study is based on two variables: runoff and snow cover, using river discharge data from Afghanistan (1960-1980) and MODIS satellite snow cover data (2000-2022). Snow cover significantly influences the natural flow regime of the Hirmand River. The SWAT model has been used for hydrological modeling, and single-variable and dual-variable calibration scenarios are compared. Dual-variable calibration was performed upstream of the Kajaki Dam, while single-variable calibration based on river discharge observations was applied to the entire Hirmand basin. The SUFI-2 algorithm was used for calibration. The results for single-variable calibration show that the NS coefficient values for stations such as Dehrawud (upstream of Kajaki Dam), Deh La (Arghandab), Delaram (Khash), Chahar Burjak (Lower Hirmand), Ardeskan (Harut), Farah (Farah), and Qala Bist (Arghestan) range from [0.0 to 0.95]. Therefore, the NS index in all regions reflects a favorable state. Single-variable calibration based on river discharge observations in the upstream basin of Kajaki Dam shows that the NS values for runoff and snow cover are 0.64 and 0.4, respectively. In single-variable calibration based on snow cover observations, the NS values for runoff and snow cover are -343.13 and 0.77, respectively. In contrast, dual-variable calibration (snow and runoff) improves the NS criteria for runoff and snow cover to 0.8 and 0.79, respectively, yielding more accurate estimates. In a study by Tarbiat Modares University for simulating the lower Hirmand basin, the NS index for the validation period is 0.63, and for the calibration period, it is calculated as 0.67. In the present study, the NS index for the calibration period is 0.95 and 0.82 for the validation period. Also, in the Tarbiat Modares University study, the NS criterion for the Dehrawud station during the calibration period was 0.72 and 0.80 for the validation period. In this study, these criteria for the calibration and validation periods are 0.80 and 0.86, respectively, indicating higher accuracy compared to previous research.

مدل‌سازي , مدل SWAT , كاليبراسيون دو متغيره , روش SUFI-2

Modeling , SWAT model , dual-variable calibration , SUFI-2 method

Alireza Novinparizad

Dr. Alizadeh