RESULTS OF APPLYING THE HBV MODEL FOR STREAMFLOW MODELING OF RIVERS IN THE ARAL-SYRDARYA BASIN

Abstract

This study evaluates the applicability of the HBV model for mountainous rivers in the Aral-Syrdarya Basin. The runoff of mountain rivers with catchment areas ranging from 88 to 1,481 km² was simulated. For the period 2000...2016, model parameters were calibrated using the GAP optimization algorithm. Model performance was assessed using several criteria: Nash-Sutcliffe Efficiency (NSE), Percent Bias (PBIAS), and the ratio of the root mean square error to the standard deviation of observations (RSR). Based on the selection of optimal parameters, model performance was obtained as follows: NSE ranged from 0.56 to 0.94, PBIAS varied between -8.1 and -28.9%, and RSR ranged from 0.02 to 0.13. The calculated model performance indicated that the dynamics of the simulated runoff during the calibration period were well correlated with observed data. In addition to these criteria, the model’s applicability as a forecasting tool was assessed using the S̅/σ ratio, which indicated that the model could be used for streamflow prediction in the Aksu, Sairam, Shayan, Boldybrek, and Kattabugun rivers. For the period 2019...2022, HBV model parameters for all studied rivers were validated. As a result of calibration and validation, the calibrated parameters of the HBV model for the mountainous rivers of the Aral-Syrdarya Basin are recommended for use in runoff simulation and prediction, particularly for short- and medium-term water flow forecasts, except for the Zhabaglysu and Karashik river basins.