Coupled Rainfall-Runoff and Hydrodynamic Modeling using MIKE + for Flood Simulation

In water resources planning and management, computer-based models for rainfall-runoff simulation are recognized as powerful tools. The southwest monsoon brings abundant rain to the Krishna basin as the monsoon winds make landfall on the Indian peninsula. With multiple hydrological interventions impacting the basin and causing severe flooding, the need for a reliable flood forecasting model cannot be overstated. Such a model facilitates efficient flood management and enhances community readiness for extreme weather events. In developing countries, integrating rainfall-runoff and hydrodynamic models for flood simulations has been rarely used due to a lack of data availability. Integrating rainfall-runoff and hydrodynamic models leads to better flood estimation, enhancing cohesive flood management strategies. The current study aims to develop an integrated Rainfall-Runoff (RR) and Hydrodynamic model for the Krishna River sub-basin in India. The study utilized the MIKE + RDI model for rainfall-runoff modeling in conjunction with the River network model for hydrodynamic modeling to develop a coupled model. The digital elevation model (DEM), gridded rainfall over the sub-basin, evaporation, discharge, and water level data for 2005 and 2006 were input parameters to develop and calibrate the coupled model and validated for 2009 data. Various performance assessing criteria such as coefficient of determination (R²), mean absolute error (MAE), index of agreement (d), Nash–Sutcliffe efficiency (NSE), and percentage deviation in peak were used to evaluate the model accuracy. The validation results demonstrate a decent agreement between the observed and simulated water levels, with R² of 0.90, MAE of 1.24, d of 0.75, NSE of 0.52, and a percentage deviation of peak at − 2.82%. The coupled model's ability to predict simulated flows and water levels within acceptable accuracy further demonstrated its reliability.