Simulation of storm surge, wave, and coastal inundation in the Northeastern Gulf of Mexico region during Hurricane Ivan in 2004

Abstract

Hurricane-induced storm surge, waves, and coastal inundation in the northeastern Gulf of Mexico region during Hurricane Ivan in 2004 are simulated using a fine grid coastal surge model CH3D (Curvilinear-grid Hydrodynamics in 3D) coupled to a coastal wave model SWAN, with open boundary conditions provided by a basin-scale surge model ADCIRC (Advanced CIRCulation) and a basin-scale wave model WW3 (WaveWatch-III). The H∗wind, a reanalysis 10-m wind produced by the NOAA/AOML Hurricane Research Division (HRD), and a relatively simple analytical wind model are used, incorporating the effect of land dissipation on hurricane wind. Detailed comparison shows good agreement between the simulated and measured wind, waves, surge, and high water marks. Coastal storm surge along the coast is around 2–3 m, while peak surge on the order of 3.5 m is found near Pensacola, which is slightly to the east of the landfall location on Dauphin Island. Wind waves reach 20 m at the Mobile South station (National Data Buoy Center buoy 42040) on the shelf and 2 m inside the Pensacola/Escambia Bay. Model results show that wave-induced surge (total surge subtracted by the meteorologically-induced surge due to wind and pressure) accounts for 20–30% of the peak surge, while errors of the simulated surge and waves are generally within 10% of measured data. The extent of the simulated inundation region is increased when the effects of waves are included. Surge elevations simulated by the 3D model are generally up to 15% higher than that by the 2D model, and the effects of waves are more pronounced in the 3D results. The 3D model results inside the Pensacola/Escambia Bay show significant vertical variation in the horizontal currents. While the estuary has little impact on the surge elevation along the open coastal water, surge at the head of Escambia Bay is more than 50% higher than that at the open coast with 1.5 h delay.