Effective factors on flood occurrence using the system dynamics approach (Case study: Eskandari watershed-Isfahan, Iran)

Abstract

This study investigates the key factors influencing flood behavior through system dynamics modeling. Initially, the HEC-HMS model was employed to estimate flood levels across sub-basins due to its well-established capacity to simulate hydrological processes and quantify runoff within watershed areas. Its ability to integrate critical parameters, such as land cover and soil permeability, which influence the curve number (CN) and its relationship with runoff and slope, makes HEC-HMS particularly suitable for flood risk assessment. After identifying these relationships, system dynamics modeling was conducted using Vensim to simulate flood dynamics. Vensim's capability to model complex systems, especially feedback loops and nonlinear interactions among multiple variables, enabled a comprehensive representation of the interdependencies among flood-related factors. Sensitivity analysis highlighted land cover as the most significant variable affecting flood behavior. To evaluate its impact, two rainfall scenarios were analyzed: a 26.75 mm event on January 3, 2011, affecting 30% of the watershed, and an 18 mm event on February 20, 2011, covering 20% of the area. The analysis revealed that precipitation, land cover, watershed size, slope, and soil permeability were the primary drivers of flooding. This study demonstrates the value of combining HEC-HMS and system dynamics modeling to predict floods, particularly in data-scarce watersheds. The integration of these tools facilitates the formulation of mathematical relationships among key hydrological variables, thereby enhancing our understanding and management of flood risks.