U.S. East Coast Storm Surge, Wind Wave and Runup Simulations During the Extratropical Remnants of Hurricane Ida

During 13th -15th of November a powerful extra-tropical storm developed from the remnants of tropical storm Ida and affected the East coast of the USA causing severe flooding in some areas, damaging buildings and eroding beaches. Ida’s low pressure cell was located off the U.S. East coast while a high pressure area was located in New England, resulting in strong winds blowing toward the southwest all along the coasts of Virginia, Delaware and New Jersey resulting in coastal setup and high waves along the coast between Cape Hatteras and New Jersey. From the Carolinas north to New England beach erosion appears to have rivaled that which occurred during Hurricane Isabel in 2003. One of the most critical factors affecting the potential for beach and dune erosion, overwash and breaching is total wave runup. The extent of runup is controlled by a range of factors the most significant being tidal levels, storm surge heights and the height of waves incident to the coast. The present study focuses on hindcasting and analyzing the spatial and temporal distribution of these parameters. We use results derived from a modeling system for wind, waves, and circulation currently applied by the USGS to the U.S. East coast. The system is run here in a hindcast mode with a 5 km resolution grid for both the SWAN wave generation and propagation model and the three dimensional ROMS hydrodynamic model. Model simulations computed a maximum storm surge of approximately 1.1 m (not including wave-induced setup), a value that was also measured by several coastal tidal gauges. Model results also indicate that the most energetic waves with significant wave heights up to 7 m occurred in coastal Virginia, decreasing to 4.5 m at Cape Hatteras. Computed significant wave heights and sea levels are compared with wave data and tidal levels measured along the East US coast. The models performance is analyzed computing the RMSE, the correlation coefficient (R) and the model skill (S) in each considered station. Results showed that there is a high agreement between the measurements and the models outputs, although some discrepancies are observed in specific areas.