This study presents an event-based rainfall–runoff simulation of the 2019 flood in the Vishwamitri River basin, Gujarat, India, using HEC-HMS 4.14. The basin was represented by three subbasins with a combined drainage area of 516.33 km², connected through junctions and reaches to Sink-1, the principal basin outlet which was defined at location on Vishwamitri river at a point before entering the city of Vadodara. Rainfall observations at 15-minute intervals were obtained from Vadodara Municipal Corporation (VMC), while observed discharge at the Pilol gauging station was obtained from the State Water Data Centre (SWDC), Gandhinagar. Data preparation covered a two-year record; however, the simulation and optimization presented in this study were restricted to 2019. The Initial and Constant loss method was coupled with the Clark Unit Hydrograph transform method. Model parameters were optimized against the discharge series at JH2, representing the Pilol comparison location. The optimized trial yielded R² = 0.57, Nash–Sutcliffe efficiency (NSE) = 0.437, normalized NSE = 0.64. The optimized simulation produced a peak discharge of 944.4 m³/s at the principal basin outlet. The results demonstrate the applicability of an optimized HEC-HMS configuration for reconstructing the 2019 basin response and generating a principal-outlet hydrograph for subsequent flood studies.
Introduction
The study focuses on modelling the 2019 flood response of the Vishwamitri River basin in Vadodara, Gujarat, using the HEC-HMS 4.14 rainfall–runoff modelling software. Flooding in monsoon regions is influenced by intense rainfall, soil moisture, catchment characteristics, urbanization, and river channel capacity. Accurate rainfall–runoff modelling is therefore essential for estimating discharge, understanding flood behaviour, and supporting flood management and hydraulic studies.
The research uses an event-based hydrological modelling approach to examine how rainfall is converted into runoff. The model consists of three subbasins with a combined drainage area of approximately 516.33 km². Rainfall data recorded at 15-minute intervals were obtained from the Vadodara Municipal Corporation (VMC), while observed discharge data at Pilol were collected from the State Water Data Centre (SWDC), Gandhinagar.
The Initial and Constant Loss Method was used to estimate precipitation losses, while the Clark Unit Hydrograph Method was applied to transform excess rainfall into direct runoff. Model parameters were optimized by comparing simulated discharge with observed discharge at JH2, representing the Pilol gauging station. Model performance was evaluated using statistical indicators such as the coefficient of determination (R²) and Nash–Sutcliffe Efficiency (NSE).
Previous studies on the Vishwamitri River have used different modelling approaches, including MIKE-NAM, MIKE-FLOOD, HEC-RAS, GIS-based analysis, and SWAT. These studies have examined flood hazards, watershed characteristics, flood inundation, and dam-break scenarios. In comparison, the present research provides an independent HEC-HMS-based assessment of the August 2019 flood event using high-resolution rainfall and observed discharge data.
The main objectives are to develop a three-subbasin HEC-HMS model, incorporate rainfall and discharge data, calibrate and optimize model parameters, evaluate simulation accuracy, and estimate the peak discharge and its timing at Sink-1, the main outlet located upstream of Vadodara city.
Conclusion
An event-based rainfall–runoff model was developed in HEC-HMS 4.14 for the 2019 flood response of the Vishwamitri River basin. The basin model comprised three subbasins with areas of 245.29, 204.25 and 66.786 km², giving a combined drainage area of approximately 516.33 km². Rainfall observations at 15-minute intervals were obtained from Vadodara Municipal Corporation (VMC), and observed discharge at the Pilol gauging station was obtained from SWDC, Gandhinagar. Although data preparation covered a two-year record, the simulation and optimization reported in this study were restricted to 2019.
The Initial and Constant loss method and Clark Unit Hydrograph transform method were used. For subbasin SB1, optimization changed the Clark storage coefficient from 19.05 to 47.191 h and the time of concentration from 10.26 to 31.426 h. The optimized constant loss rate changed from 5.0 to 4.7666 mm/h, and initial loss changed from 3.0 to 15.673 mm.
At JH2, corresponding to the Pilol comparison location, the optimized model yielded R² = 0.57, NSE = 0.437 and normalized NSE = 0.64. At the principal outlet Sink-1, the simulated peak discharge was 944.4 m³/s at 03:15 on 1 August 2019.
The results demonstrate the use of 15-minute rainfall observations and HEC-HMS parameter optimization to represent the 2019 rainfall–runoff response of the Vishwamitri River basin. The optimized performance statistics at JH2 and the simulated peak at Sink-1 provide complementary information for hydrologic assessment and subsequent hydraulic flood modelling. The model outputs should be interpreted according to their respective locations within the basin network.
References
[1] National Institute of Hydrology, “Hydrodynamic Modelling of River Vishwamitri for Preparation of Flood Mitigation Plan for Vadodara City,” in All Studies/Publications, 40th Foundation Day publication. Available:https://www.yumpu.com/en/document/read/62297189/all-studies-publications-40thfd-nih/11
[2] U.S. Army Corps of Engineers, Hydrologic Engineering Center, HEC-HMS Technical Reference Manual. Davis, CA, USA. Available: https://www.hec.usace.army.mil/confluence/hmsdocs/hmsum/latest
[3] J. E. Nash and J. V. Sutcliffe, “River flow forecasting through conceptual models: Part I—A discussion of principles,” Journal of Hydrology, vol. 10, no. 3, pp. 282–290, 1970, doi: 10.1016/0022-1694(70)90255-6.
[4] U.S. Army Corps of Engineers, Hydrologic Engineering Center, “Initial and Constant Loss Model,” HEC-HMS Technical Reference Manual. Available: https://www.hec.usace.army.mil/confluence/hmsdocs/hmstrm/canopy-surface-infiltration-and-runoff-volume/infiltration/initial-and-constant-loss-model.
[5] U.S. Army Corps of Engineers, Hydrologic Engineering Center, “Clark Unit Hydrograph Model,” HEC-HMS Technical Reference Manual. Available: https://www.hec.usace.army.mil/confluence/hmsdocs/hmstrm/transform/clark-unit-hydrograph-model.
[6] D. N. Moriasi, J. G. Arnold, M. W. Van Liew, R. L. Bingner, R. D. Harmel, and T. L. Veith, “Model evaluation guidelines for systematic quantification of accuracy in watershed simulations,” Transactions of the ASABE, vol. 50, no. 3, pp. 885–900, 2007, doi: 10.13031/2013.23153.
[7] Vadodara Municipal Corporation (VMC), Rainfall Data for the Vishwamitri River Basin, 2019, Vadodara, Gujarat, India, unpublished institutional dataset.
[8] State Water Data Centre (SWDC), Observed Discharge Data at Pilol Gauging Station, 2019, Gandhinagar, Gujarat, India, unpublished institutional dataset.
[9] Japan Aerospace Exploration Agency (JAXA), digital elevation model used for terrain representation and basin delineation. Product metadata may be supplied from the original GIS processing records.
[10] V. K. Rana and T. M. V. Suryanarayana, “Estimation of flood influencing characteristics of watershed and their impact on flooding in data-scarce region,” Annals of GIS, vol. 27, no. 4, pp. 397–418, 2021, doi: 10.1080/19475683.2021.1960603.
[11] Z. Shah, A. Saraswat, D. R. Samal, and D. Patel, “A single interface for rainfall-runoff simulation and flood assessment—a case of new capability of HEC-RAS for flood assessment and management,” Arabian Journal of Geosciences, vol. 15, article 1526, 2022, doi: 10.1007/s12517-022-10721-2.
[12] N. Kumar, “Fluvial flood risk in contemporary settlements: A case of Vadodara City in the Vishwamitri Watershed,” Engineering Proceedings, vol. 56, no. 1, article 70, 2023, doi: 10.3390/ASEC2023-15270.
[13] P. Pandya and R. Jain, “Dam break analysis of Ajwa dam during an extreme flood event in Vishwamitri basin,” ISH Journal of Hydraulic Engineering, vol. 30, no. 4, pp. 522–533, 2024, doi: 10.1080/09715010.2024.2372853.