Quantifying Diyala River basin rainfall-runoff models for normal and extreme weather events

Abstract

Abstract A variety of methods, including both modeling and non-modeling water balance techniques, are used in this study to estimate water availability allocated for different demands. The area under investigation is the shared Diyala River basin between Iraq and Iran, which is vulnerable to climate change impact and upstream control and aims to enhance watershed management. Two rainfall-runoff models were applied, the Tanh curve model and the modified Tanh curve model for runoff simulation at the Derbendkhan and Hemmrin hydrometric stations. Data from two meteorological stations in Iraq (Khanakin and Sulymania) and one in Iran (Sanandaj) for 20 hydrological years (2000–2020) was used. Some goodness of fit indicators were used to evaluate the reliability of the model outcomes, including the correlation coefficient, the root mean squared error, the mean absolute error, the deviation of runoff volume, and the index of agreement. Statistical analysis shows no statistically significant difference between observed and predicted runoff at all stations in winter, spring, and annual time scales when using the Tanh curve model, or for monthly and extreme wet events when using the modified Tanh curve method. The modified Tanh curve model was more accurate than the Tanh model and applied only to Sulymania and Sanandaj stations, as it required a considerable amount of precipitation in at least semi-humid areas.