The frequency and severity of hurricane activity has risen sharply over the last few years. Several major hurricanes hit the US Southeast coast in the 2004 and 2005 seasons, with the impacts from Katrina and Wilma being among the costliest in the US history. Typically, the storm surge predictions are focused on local dynamics associated with the wind-induced onshore drift and the wave action. However, the evolution of storm surge also includes an alongshore propagation of the coastally trapped wave (CTW) pulse. Linear theories predict that in such a case the strongest response occurs downstream (in the sense of Kelvin wave propagation) and later in time relative to the location and moment of the landfall. This notion is consistent with the events in New Orleans, when the protective levies were overwhelmed a few hours after the Hurricane Katrina's landfall some distance upstream, on the Mississippi coast.
Unique and detailed observations of sea level and barometric pressure along the Southwest Florida coast collected during Hurricane Wilma's landfall in October 2005 will be analyzed to delineate the coastally trapped wave (CTW) pulse propagation, dispersion and decay as it leaves the forcing region and moves downstream (northward). While the observational array was not designed to capture the complex hydrodynamics of the storm surge, it occupied an optimal position for observing the hurricane-generated CTW pulse.. These data will also be used for tuning-up the primitive equation model in order to conduct a process-oriented study of the hurricane's impact on the wide and shallow shelf. The data set consists of weeklong time series of storm surge (S) and barometric pressure (B) measured by the USGS Florida Integrated Science Center at approximately 30 locations. The survey area spanned more than 100 km alongshore with the instruments deployed both on the coastline and in the inlets/estuaries. These data will be augmented with the existing NOAA wind and sea level measurements.
The atmospheric pressure data will be used in order to determine a temporal and spatial evolution of the wind forcing: the available time series of wind will be extrapolated by applying the structure of atmospheric pressure variations. Detailed barometric pressure measurements will be also applied to accurately adjust the sea level data and thus to detect the CTW pulse evolution along the coastline. The CTW pulse amplitude will be determined by using data from the exposed coastline only, while the phase will be estimated based both on data from the coastline and from the inlets. The latter will be adjusted in time by allowing the signal propagation from the mouth inland at a speed of a long gravity wave. The observations will be reproduced in the model by applying the observed forcing and tuning the bottom stress and horizontal eddy "viscosity" coefficients in order to obtain a realistic propagation speed and amplitude of the storm surge. The shelf topography will be uniform alongshore, except for the upstream boundary, where the shelf width will diminish abruptly, thus mimicking the southern tip of Florida. The role of dispersion, friction and nonlinearity will be systematically studied by conducting realistic model runs with the observed wind forcing and complimentary model runs with the same initial disturbance (induced by the hurricane) but traveling as a free wave pulse with much reduced friction, and also as a pulse of the same spatial-temporal structure but with a smaller amplitude (allowing linear dispersion). In subsequent model experiments, the coastline feature will also be placed downstream, representing the conditions of Hurricane Katrina (i.e., protruding Mississippi delta).
Broader Impacts: The results of data analysis and numerical experiments will reveal the importance of dispersion, nonlinearity, frictional decay and topographic variations in the alongshore evolution of storm surge driven by a hurricane landfall. The results will also improve our understanding and predictability of hurricane-induced flooding with a focus on vulnerability of the downstream areas.
