Towards High-Resolution River Forecasting over Narmada Using Surface Water Elevation Data Assimilation in Triton

Accurate representation of river water levels is essential for flood forecasting in Narmada river basin, where complex river networks and limited observations cause significant challenges. In this study, we present data assimilation framework to assimilate surface water elevation observations into the 2D hydrodynamic model Triton. We will use ensemble Kalman filter (EnKF) data assimilation techniques with grid-to-grid along the stream localization by leveraging both upstream and downstream network information to account for hydrodynamic uncertainties. We will assimilate the surface water elevation from the central water commission (CWC) of India and HydroWEB. The proposed approach is expected to improve the simulation of flood propagation, river depth, and inundation dynamics over the Narmada river basin. By integrating observational data directly into Triton, we anticipate enhanced accuracy in peak water levels and flood timing. This study demonstrates the potential of combining hydrodynamic modeling with real-time data assimilation to provide actionable insights for flood risk