Reconstruction of extreme flood events by performing integrated real-time and probabilistic flood modeling in the Periyar river basin, Southern India

The main purpose of this study is to mitigate the flood impact which may be caused by uncertain future extreme high rainfall-runoff events. To accomplish the objective, an integrated hydrologic–hydrodynamic flood modeling has been performed in the Periyar river basin by utilizing the HEC-RTS framework including reservoir simulation, optimization and calibration of modeling parameters. For the reconstruction of historical floods, seven historical extreme high rainfall-runoff events have been identified (2002 to 2018) and the event-based flood modeling and continuous flood modeling were performed. Simulated floods showed a good agreement with reference to observed floods (at Neeleswaram gauge) (where R2 is recorded as 0.80 and 0.67 for flood event 2018 and continuous simulation, respectively). The historical flood event (i.e., July–August 2018) was reconstructed using HEC-RESSIM and the operating rules were modified for Mullaperiyar and Idukki reservoirs. The regulated flood flows were generated for all seven historical extreme events utilizing the daily meteorological time series datasets. The flood frequency analysis (FFA) has been performed to generate the return period floods and probabilistic flood maps. For hydrodynamic flood simulation, one-dimensional (1D) flood routing was performed in the downstream sections. For the accuracy assessment, the computed flood inundation area and depths were compared with the existing reference datasets like (i) Central Water Commission (CWC) flood data, (ii) real-time flood inundation captured in Sentinel satellite data and (iii) flood maps prepared by DARTMOUTH Flood Agency USA, which showed a reliable agreement.