The advancement on the storm surge and street-level inundation modeling — A case study in New York City during Hurricane Sandy

A large-scale, unstructured grid storm tide model SELFE was used to hindcast water level variation during Hurricane Sandy in the mid-Atlantic potion of the US East coast. The model was forced by 8 tidal constituents at the open boundary condition, 1500 km away from the coast, and the wind and pressure fields from atmospheric model RAMS provided by Weatherflow Inc. From the large-scale modeling results, it obvious that during Hurricane Sandy there are two storm surges converged into the New York City; one from Long Island Sound propagating westward and the other from the Now York Bight into New Jersey Coast. The comparisons of modeled storm tide with the NOAA gauge stations from Montauk, NY, Long Island Sound, encompassing New York Harbor, Atlantic City, NJ to Duck, NC were in good agreements, with overall root mean square error and relative error on the order of 15-20 cm. Furthermore, using large scale model outputs as the boundary conditions, a separate, sub-grid inundation model for the major portion of the New York City was also setup to investigate the detailed inundation process. The sub-grid modeling is a novel technology which incorporates LIDAR bathymetric data within a coarse grid model to further improvement of the model accuracy without the high computational costs of simulation on a fullblown high resolution grid. The model grid designed for the New York City modeling makes use of a 200 m × 200 m square base grid and in each of the base grid, there are 40 by 40 numbers of 5m × 5m sub-grids embedded within each base grid cell. To retain the accuracy, the uniform square grid LIDAR data were imported directly into the sub-grid. When coupled with nonlinear wetting and drying solver, the shoreline is intrinsically resolved in the subgrid, which allow partial wetting and drying within a coarse grid when the water level changes in each time step. The sub-grid inundation results compared favorably with USGS´ Hurricane Sandy Mapper database. in - erms of its timing, local inundation area, and the depth of the flooding water. The maximum extent of horizontal inundation was calculated, which was within 30 m of the data-derived estimate by USGS.