Analyzing the fixed or variable effect of considering hydrological loss functions on the runoff coefficient in continuous modeling

Abstract

Hydrological losses play a significant role in determining the runoff coefficient, influencing the amount of precipitation that ultimately contributes to surface runoff. These losses are influenced by various characteristics, including rainfall, the physical and geographical attributes of the watershed (such as slope and land use), and the soil moisture content within the watershed. Here, we evaluate the effect of the variability of the loss function on the amount of simulated runoff and the runoff coefficient in specific watersheds in Iran. The investigation entails the assessment of two distinct conditions. First, the runoff coefficient is calculated under the assumption of a constant loss, utilizing the φ index. Second, a variable loss function, derived from a soil moisture algorithm, is employed to determine the runoff coefficient. Our analysis shows that the assumption of a variable loss function yields more realistic results. When the variable losses are considered, the simulated runoff coefficient is closer to the observed values and determines the runoff coefficient for all months, including those characterized by low rainfall. The constant loss φ index exhibits two significant practical limitations: the overestimation of runoff coefficient values, and an inability to estimate runoff coefficient during months with low rainfall. The study emphasizes the need for a variable loss function to provide more realistic results. Our findings suggest that utilizing the variable loss function within the soil moisture algorithm produces more accurate results. Thus, the application for improved forecasting of rainfall and runoff processes is recommended.