A comprehensive evaluation of calibration strategies for flood prediction in a large catchment using HEC-HMS

• Published in: Modeling earth systems and environment Vol. 11; no. 1; p. 32
• Main Authors: Kalakuntla, Nikhil Teja, Umamahesh, N. V.
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.02.2025; Springer Nature B.V
• Subjects: Algorithms; Automation; Benchmarks; Calibration; Catchments; Chemistry and Earth Sciences; Computer Science; Earth and Environmental Science; Earth Sciences; Earth System Sciences; Ecosystems; Environment; Flood forecasting; Flood hydrographs; Flood predictions; Floods; Hydrologic models; Hydrology; Hydrometeorology; Math. Appl. in Environmental Science; Mathematical Applications in the Physical Sciences; Multiple objective analysis; Optimization; Optimization algorithms; Original Article; Outlets; Performance evaluation; Physics; River basins; Search algorithms; Spatial variations; Statistics for Engineering; Stream discharge; Stream flow; Watersheds; Multi‐site calibration; HEC-HMS; Single-site calibration; Single-objective optimization; Multi‐objective optimization; Distributed hydrologic model
• ISSN: 2363-6203; 2363-6211
• DOI: 10.1007/s40808-024-02276-w

Effective calibration is essential for the performance of hydrological models, particularly in flood forecasting. This study aims to explore various calibration strategies for improving the accuracy of an event based hydrological model. The proposed strategies are optimized at the flood forecasting station located at Bamni (Balharsha) in Wardha subbasin of Godavari River Basin, India. The effect of spatial variability of hydro-meteorological processes and catchment characteristics on streamflow for such a large catchment is modelled by employing a distributed hydrologic model (HEC-HMS). Different characteristics of a flood hydrograph for calibration are addressed using single- and multi-objective optimization algorithms. Four calibration strategies are formulated based on single- and multi-site calibration approaches, combined with single- and multi-objective optimization algorithms. These strategies are compared against the default calibration technique used in HEC-HMS, serving as the benchmark. In this event based study, six flood events are chosen for calibration and six for validation. The results of the study seek to identify the optimal calibration strategy for accurate prediction of floods during flood forecasting. While single-site calibration strategies yield similar results, the incorporation of interior streamgage data substantially enhances model performance by improving the hydrological simulations at the outlet. The benchmark calibration strategy outperforms single-site calibration strategies; with the multi-site multi-objective combination demonstrating the best results, albeit slightly better than the benchmark strategy. Additionally, the external automation of HEC-HMS model facilitates the implementation of joint-event calibration and application of global search algorithms, making the optimization process more straightforward and efficient.