Document Type : Research Paper
Authors
1 Department of civil Engineering, Khorramabad Branch, Islamic Azad University, Khoramabad, Iran
2 Department of civil Engineering, Estahban Branch, Islamic Azad University, Estahban, Iran
Abstract
Introduction
Flood estimation in ungauged watersheds is crucial for designing hydraulic structures. The total flood flow in watersheds consists of surface and subsurface flows. In highly permeable watersheds, subsurface flow significantly contributes to total runoff, yet limited studies have addressed this aspect. Runoff mechanisms in watersheds generally follow two models: the Hortonian mechanism, where the soil saturates from the top, and the Dunne mechanism, where saturation occurs from below. This study adopts the Dunne mechanism for runoff generation.
Materials and methods
One method for predicting surface and subsurface hydrographs in ungauged watersheds is the Geomorphologic Instantaneous Unit Hydrograph (GIUH), which utilizes geomorphological data. These data were derived using ArcGIS and hydrological extensions. The GIUH model can separate surface and subsurface flow components. While previous studies have primarily used GIUH for surface flow estimation, the model equations were expanded in this study to estimate the total watershed subsurface hydrograph. The GIUH model was applied to estimate surface and subsurface runoff in two watersheds: Kasillian in Iran and Gagas in India.
Results and discussion
The proposed GIUH model was used to analyze surface and subsurface flows in the Kasillian (Iran) and Gagas (India) watersheds. Simulation results for four rainfall-runoff events in each watershed demonstrated that the model effectively estimated total hydrographs and their components. A comparison of estimated and observed peak discharges in the Kasillian watershed showed that the simulated peak discharge on May 10, 1992, was 10.1 m³/s, whereas the observed value was 11.8 m³/s. The error margin across different events ranged from 3% to 16%, indicating an acceptable model accuracy in runoff estimation. In the Gagas watershed, total peak discharge varied between 44 and 110 m³/s, with subsurface flow contributing approximately 5%–6% of the total flow. In the Kasillian watershed, total discharge ranged from 1.6 to 12 m³/s, while peak subsurface discharge was estimated between 35 and 60 L/s. The relationship between rainfall and subsurface peak discharge revealed that lower rainfall led to reduced subsurface peak discharge. Sensitivity analysis showed that hydraulic conductivity was one of the most influential parameters in subsurface flow simulation. In soils with high hydraulic conductivity, subsurface flow accounted for a larger portion of total flow, and the hydrograph lag time increased. For example, reducing the hydraulic conductivity in the Kasillian watershed from 0.0025 to 0.0009 m/s increased subsurface peak discharge from 0.35 to 1.3 m³/s. Additionally, reducing the Manning’s roughness coefficient from 0.2 to 0.4 resulted in a 31% decrease in flood peak discharge. These findings highlight the importance of hydrological and geomorphological characteristics in accurate runoff estimation and flood control structure design.
Conclusion
This study evaluated the GIUH model for estimating surface and subsurface runoff in the Kasillian (Iran) and Gagas (India) watersheds. Results showed that the model provided accurate peak discharge estimates, with estimation errors ranging from 3% to 16% in Kasillian and 1.6% to 12% in Gagas. The subsurface flow played a significant role in highly permeable watersheds, contributing 5%–6% of total runoff in the Gagas watershed. Sensitivity analysis revealed that increasing hydraulic conductivity led to higher subsurface peak discharge, whereas reducing the Manning’s coefficient increased flood peaks. These findings confirm the importance of geomorphological and hydrological characteristics in runoff modeling. Ultimately, the GIUH model can serve as a useful tool for flood management and watershed hydrological response assessment.
Keywords
)