A Quantitative Method for Ranking the Geomorphic Controls on Storm Surge Penetration along the Mississippi Coast after Hurricane Katrina

Rochelle Legaspi, Department of Geosciences, University of West Georgia, Carrollton, GA 30118

In partial fulfillment of the GEMS Summer Research Project under the supervision of Dr. David Bush

July 2007

Abstract

The proposed project will statistically assess selected portions of the Hurricane Katrina impacted coast of Mississippi. Previous studies of geomorphic controls on storm damage in Florida and South Carolina indicate that along those coasts, site elevation provided the best protection against property damage, followed by dune height (in front of the site), and beach width. Each setting is different and different factors may exert more control along the Mississippi coast. Ultimately, the new information can be used to produce a series of coastal hazard maps that can be easily accessed via the internet by both public and private entities interested in learning about and managing their coastal hazards.

Introduction:

A project is currently underway in Puerto Rico assessing a completely different geologic setting. The Katrina study will be a major addition to the data base because of the intensity of the storm and its geographic extent. Each hurricane and site is different and different factors may exert more control along the Mississippi coast. Ultimately, the new information can be used to produce a series of county-scale coastal hazard maps for Hancock, Harrison, and Jackson Counties that can be easily accessed via the internet by both public and private entities interested in learning about and managing their coastal hazards. Maps will be posted on the Coastal Hazards Information Clearinghouse web site (http://www.wcu.edu/coastalhazards/).

Methods Employed:

Pre-storm geomorphic characteristics, such as digital elevation model, Q3 FEMA flood data, measurement of slope and a development layer has been entered into a digital data base. This has entailed both hand digitizing of features and inputting data already in digital form into the data base. Each characteristic will constitute a single data layer in a Geographic Information System (GIS). A final map of storm damage has also been digitized and generated from aerial photographs layering the study area of overwash penetration, the vegetation, a no damage layer and total development destruction layer. The ultimate goal of this project is to explain this alongshore pattern. It is believed that the alongshore variability is primarily controlled by various physical and geomorphic features of the shoreline and nearshore. To test this hypothesis, several shoreline characteristics will be quantified at each shoreline station. Most of this work can be performed using digital data within the GIS database. The following coastal attributes will be quantified for each station: bathymetric slope, fetch, average inland slope, shoreline curvature, degree of vegetation cover, pre-storm beach width, and pre-storm dune volume/height. Finally, a statistical logistic regression analysis will be performed using a newly implemented program “R” developed by Chester W. Jackson, in order to determine which of these pre-storm coastal attributes controlled the nature of the storm impact as measured by overwash penetration and/or buildings damaged. It is hoped that a quantitative ranking can be produced of the most important attributes.

Each degree of slope, meter of elevation and different flood zone will be classified with a 1 value and then layered with the storm damage layers to run with the logistic regression analysis. The logistic regression analysis will tell us which pre storm geomorphic parameters correlate most closely with the damage pattern observed after Hurricane Katrina (2005).

Conclusions:

The proposed research will provide an important boost to our scientific understanding of coastal storm processes, and possibly of greater importance, this information will be invaluable to coastal managers seeking to delineate those shoreline areas that are most vulnerable to storm surge driven damage. The results will help delineate a coastal construction or setback line and form a sound geologic and oceanographic framework on which to base reconstruction and future planning efforts. The maps will be made available via the internet and all maps produced will become public domain.

It is further hoped that the methodology will serve as a model for future studies of hurricanes impacting other shorelines, the ranked list of geomorphic attributes which control storm energy can be applied elsewhere along the Gulf and Atlantic Coasts.

References:

http://terraserverusa.com/default.aspx - TerraServer USA

http://datagateway.nrcs.usda.gov/ - United States Department of Agriculture Geospatial Data Gateway

http://www.maris.state.ms.us/ - Mississippi Automated Resource Information System

http://www.ncddc.noaa.gov/Katrina-2005/InteractiveMaps/ - National Oceanic and Atmospheric Administration/Katrina Impact Assessment