Development of Scientific Interpolation Method for Urban Inundation Maps of Arbitrary Return Periods Based on Pre-Simulated Scenarios

Due to recent climate change, urban flood damage is increasing. Current urban flood prediction has been mainly conducted by performing numerical simulations for various return period scenarios and producing inundation maps based on the results. However, this method has disadvantages in that it is difficult to predict arbitrary rainfall events or intermediate frequencies that can occur other than the fixed scenarios, and it is difficult to respond in real-time because the computation time of numerical simulations is significantly long. Therefore, to overcome these challenges, this study developed a 'scientific interpolation method' to estimate urban inundation maps for arbitrary frequencies by leveraging pre-constructed flood scenario data. We utilized simulation results from a self-developed Python-based urban flood model as a benchmark to derive the fundamental governing equations and related parameters. An Inverse Analysis technique was applied to mathematically reconstruct the non-linear relationship between rainfall frequency and inundation depth. Consequently, the inundation depths and extents for arbitrary frequencies interpolated through the derived equations showed a high spatial correlation with the physics-based model results with R²= 0.9. By integrating discrete scenario maps through this interpolation scheme, the proposed method enables rapid flood prediction without the need for repetitive numerical simulations. This approach is expected to significantly enhance the capability for immediate decision-making and response against sudden urban flood disasters.