National Weather Service

National Weather Service and Severe Weather 









JUNE 29, 2005


Mr. Chairman and Members of the Subcommittee, thank you for the opportunity to speak with you on behalf of the U.S. Geological Survey (USGS) on inland flooding and coastal-change impacts of extreme storms.  Each year, natural hazards in the United States such as earthquakes, fires, floods, hurricanes, landslides, and volcanoes result in hundreds of lives lost and cost billions of dollars in disaster aid, disrupted commerce and destroyed public and private properties.  At USGS, it is our goal to provide scientific research and analysis to help citizens, emergency managers, and policy makers decide how to react to each hazard and how to safeguard society.  By collecting long-term data and information assessing past and present hazards events and by providing continuous monitoring and data collection, we hope to arrive at the place where we are able to predict these natural events and mitigate their potential impacts, providing precious time to save lives and property.  By conducting research on coastal change that occurs during extreme storms, and by improving understanding of erosion and deposition that can destroy infrastructure and permanently change the coastal landscape, USGS will assist in efforts to reduce the impact these severe storms have on lives and communities.

There are two major objectives of this USGS research effort. The first is to improve predictive capabilities so that, as a hurricane approaches the United States, assessments can be made of impacts to the threatened coastal setting prior to landfall. The second major objective is to provide the information and knowledge required to assess the changing vulnerability of our coastline to hurricanes for longer-term hazard planning and mitigation so that new buildings and infrastructure, particularly those being rebuilt following a storm disaster, can be sited away from hazardous areas.

The 2004 Atlantic hurricane season was one of the busiest and most destructive in history.  For example, Hurricane Ivan caused severe beach and dune erosion that undermined five-story oceanfront condominium towers, some of the largest buildings to fail during a hurricane in United States history.  Today, after giving an overview of the USGS research program on severe storms, I will focus on lessons learned from the coastal change impacts observed last year.  

Research program on extreme storms

As part of USGS National Assessment of Coastal Change Hazards, impacts of extreme storms have been intensively investigated since the 1997-98 El Nino when severe winter extratropical storms ravaged much of the U.S. west coast, causing extensive erosion of beaches and sea cliffs and resulting in loss of property. The USGS worked cooperatively with National Aeronautic and Space Administration (NASA) and National Oceanic and Atmospheric Administration (NOAA) to acquire airborne lidar surveys of the coast both before and after the El Nino. These data were used to test models of the interaction between storms and coasts.  Since the 1997-98 El Nino, USGS has continued to work with NASA, focusing primarily on hurricane impacts in the southeast U.S., again using airborne lidar to survey the coast before and after storm impact.

Airborne lidar survey systems utilize the Global Positioning System (GPS) and a laser mounted in an aircraft to measure ground topography. If the water is clear enough, some lidar systems can penetrate the ocean and measure shallow seafloor bathymetry. The before- and after-storm surveys gathered as part of USGS research are compared to detect changes in the elevation and configuration of the ground, changes that occur during a storm due to erosion and deposition.

These data are used to test and validate predictive models that can forecast coastal change prior to hurricane landfall. The data are also used to develop a quantitative means to assess the vulnerability of U.S. coasts to future extreme storms.  Currently, USGS is developing the means to assess:

  • The location of potential breaches that sever barrier islands and evacuation routes during hurricanes.  Most of the East and Gulf of Mexico mainland coasts of the United States are protected from the open ocean by a nearly continuous string of barrier islands. These long, thin strips of sand are, in places, low-lying (less than 9 feet in elevation) and subject to being inundated and cut during extreme storms. In fact, most of the present inlets through barrier islands in the southeast United States, which allow boats and ships to transit between ocean and mainland ports, were cut naturally during hurricanes. Most recently, breaches severed barrier islands during Hurricane Isabel on the North Carolina coast in 2003, on the southwest coast of Florida during Hurricane Charley in 2004, and during Hurricane Ivan on the Alabama and Florida panhandle coasts in 2004. Results of USGS research indicate that these catastrophic island breaching events occur where storm processes intersect with low-lying topography. USGS research also suggests that the underlying geology may contribute to the vulnerability of barrier islands to inlet formation.
  • Extreme beach and dune erosion that lowers the elevation of barrier islands, making the islands, and the back bays they shelter, more suceptible to inundation by storm surge. During extreme storms, wind can push water against the coast, raising sea level in a storm surge.  This allows waves to attack beaches and dunes that are normally beyond their reach. During Hurricane Ivan, Santa Rosa Island, offshore of Pensacola, Florida, was reduced in elevation an average of approximately 3 feet; however, in places, the reduction was as much as eight feet where new breaches opened.  This reduction in elevation allows more water to be driven across the island during a severe storm, raising the storm surge in the back bays higher than would have been possible had the dunes remained intact. Thus, up-to-date and accurate information of coastal elevation, and understanding of the coastal response to storm processes, is critical to providing accurate forecasts of hurricane impacts.

The 2004 Hurricanes: Charley, Frances, Ivan and Jeanne

In a cooperative effort between USGS, NASA, and U.S. Army Corps of Engineers, the impact zones of the four Atlantic hurricanes that made landfall in the United States in 2004 were surveyed with airborne lidar and photography both before and after landfall of each storm. Initial results for each hurricane can be found on the USGS World Wide Web site ( ). Pre-storm surveys were combined with models of storm processes and coastal response to assess vulnerability of the threatened coast prior to landfall. After landfall, pre- and post-storm surveys were compared to quantify change and showed that coastal response was unique for each storm, depending on characteristics of both the storm and the shoreline setting impacted.

For example, the swath of hurricane-force winds associated with Hurricane Charley was narrow.  Major coastal-change impacts were limited to several tens of miles of shoreline near landfall, where a breach, 1,500 feet wide, opened through North Captiva Island, Florida. In contrast, Hurricane Frances was a larger, weaker storm that caused moderate coastal erosion extending for nearly 100 miles along the Florida south-central east coast. However, Hurricane Frances’ greatest legacy may have been in making the coastline more vulnerable to erosion from Hurricane Jeanne, which followed the same storm track several weeks later.  Surviving structures left exposed on the brink of eroded dunes following Hurricane Frances in Vero Beach and Floraton, Florida, were later destroyed during Hurricane Jeanne.

The most extensive coastal change observed during the 2004 Atlantic hurricane season occurred during Hurricane Ivan on the Alabama and Florida Panhandle coasts.  On average, the shoreline retreated 40 feet during the storm. In Gulf Shores, Alabama, where the storm’s strongest winds made landfall, the relatively low-lying barrier islands were completely inundated by storm surge. The sea-level difference between the Gulf of Mexico and back bays drove a strong landward current that transported sand across the island and opened a new inlet. In contrast, several miles to the east in Orange Beach, Alabama, where land elevations were higher, the response was dune erosion. In places, the vertical scour associated with dune retreat approached nine feet and undermined structures including several five-story condominium towers that had been built on top of the dunes. These are some of the largest buildings to be destroyed by  hurricane impact in United States history.

Assessments of storm impacts prior to hurricane landfall

Forty-eight hours prior to Hurricane Ivan’s landfall, the USGS posted on its extreme-storm web site an experimental product that showed the vulnerability of the threatened coast to change. This assessment was based on the difference between worst-case storm-surge elevations, calculated by NOAA using computer models, and high-resolution coastal elevations, measured with airborne laser mapping.  For each location along the coast, the posted maps showed where Ivan’s worst-case storm surge would exceed coastal elevations and submerge barrier islands as if they were shoals. At these locations, water level differences would drive strong currents across the islands, changing their form and undermining buildings and infrastructure. The coastal change during Hurricane Ivan measured with airborne lidar was later found to be consistent with USGS assessments of coastal vulnerability made prior to the storm’s landfall.

The future

The unusual failures of large, oceanfront buildings during Hurricane Ivan may be because southeast U.S. coastal communities have not been severely tested by hurricane-induced erosion until recently. Between 1966 and 1990, when southeast coastal developments grew dense, only two major hurricanes made landfall along the east coast or the peninsula of Florida - most developments survived unscathed. However, recent research on decadal scale changes in hurricane activity suggests that the Atlantic Basin has re-entered an active hurricane period similar to that of the period 1941 - 1965 when seventeen major hurricanes made U.S. landfall.  It is likely that this active period will persist for decades.  Hence, the loss of multi-story buildings during Hurricane Ivan may be a warning of what is to come along our hurricane threatened coasts.

The USGS, working with our partners, will continue to develop extreme storm vulnerability assessment methodologies and provide these assessments of coastal change to user agencies. Several weeks ago when Tropical Storm Arlene threatened the Alabama and Florida panhandle coasts - the same area where Hurricane Ivan made landfall nine months before - USGS provided NOAA storm surge modelers with assessments of dune erosion within the forecast impact zone. The modelers were concerned that barrier island elevations had been lowered during Hurricane Ivan, which would allow more water to be driven across the islands, resulting in higher surge in estuaries than their models would account for. The USGS provided dune erosion data and assessments that were incorporated into NOAA storm-surge models and were used to help forecast potential flooding from Tropical Storm Arlene.

Ongoing data collection efforts, combined with existing models, provide the basis for a collaborative effort with other Federal partners, such as National Weather Service (NWS), to assess the likely impacts of coastal storms. Both pre-storm assessments of dune and beach erosion and post-storm damage assessments, provided in a timely manner, support the efforts of Federal and local emergency planners and responders. These activities are also an integral part of persistent research efforts to better understand and assess the vulnerability of U.S. shorelines to coastal change impacts from extreme storms. Integration of scientific information and coastal change models developed by USGS with the meteorological models of impending storm processes from NWS will support more timely and accurate forecasts of the location and type of coastal response to severe storm events.

Inland flooding from excessive rainfall

As population and development continue to increase in coastal areas, more people and property are vulnerable to hurricane threat.  However, coastal residents and visitors are not the only ones vulnerable to the ravages of hurricanes and extreme storms.  Hurricane winds and waves impacting the coastal zone are often accompanied by extreme rainfall that can contribute to local and regional flooding of coastal and inland areas.  Flooding is the most frequent natural disaster.  During the 20th century, floods arising from extreme storms, both tropical and extra-tropical, were the worst natural disaster in the United States in terms of number of lives lost and property damaged. Flooding from extreme storms can occur at any time of the year, in any part of the country, and at any time of the day.  Property damage, including inundation by sediment-laden water, demolished buildings, and erosion that undermines bridge foundations and footings leading to the collapse of structures, results in approximately $5B in losses per year.

Hurricanes and tropical storms can be especially dangerous and destructive as they move inland from coastal areas. For example, floods from remnants of Hurricane Camille in 1969 killed hundreds of people throughout Appalachia.  In 1999, eastern North Carolina endured record rainfall and two months of continuous flooding from Hurricanes Dennis, Floyd, and Irene.  Notable, the 2004 Atlantic hurricane season was the most costly on record -- $42B.  Widespread rainfall amounts over 6 inches caused extensive flooding.  In Florida, USGS field crews obtained some of the highest flow measurements ever recorded.  This flooding was compounded by the remnants of Hurricane Ivan less than 2 weeks later.

The USGS, in cooperation with NWS River Forecast Centers and others, is making significant progress in development of new tools and techniques to address flood risk.  The following are examples of USGS research and modeling activities relative to inland flooding:

  • Prioritizing Streamgaging Network investments and Improved Streamflow Information Delivery.  The USGS managed streamgage network includes 3200 gages that support NWS streamflow forecasts and flood predictions to calibrate their streamflow forecast models and make flood predictions.  The USGS is working to improve delivery of streamgage information to meet this and other national needs for streamflow information.  As part of that effort, USGS is installing new high data rate transmitters to improve real-time data access, flood-hardening streamgages critical to the National Weather Service for flood predictions, and building a robust data storage and processing system to ensure reliable and timely streamflow information delivery to users of the information.
  • Development of a real-time flood inundation mapping capability using forecasts from the NWS River Forecast Centers.  Emergency managers need to know what is (or shortly will be) under water when a flood is occurring.  Inundation maps help emergency managers plan evacuation routes, deploy critical resources, understand the magnitude of events, and, in general, respond quickly to save lives and property.  In creating real-time inundation maps, forecast flood hydrographs are routed through lidar-derived elevation models of reaches of a river with multi-dimensional flow models that allow predictions of the timing, depth, velocity, and impact of flood waters for any location in the mapped floodplain. These inundation forecast maps can be posted on the worldwide web hours to days prior to the arrival of the flood. Near-real-time simulation and internet-based delivery of forecast-flood inundation maps using two-dimensional hydraulic modeling has been developed through a pilot study of the Snoqualmie River, Washington (see USGS Water-Resources Investigations Report 02-4251, 36p.)
  • Development of a map-based Web application, “StreamStats,” to obtain streamflow and flood statistics. “StreamStats” provides streamflow information for all locations in the Nation, and specifically for ungaged sites, by using statistical models and established hydrological relationships.  This application results in major cost savings by reducing the time needed to obtain streamflow estimates for a site from an average of about a day to only a few minutes.  “StreamStats” is currently available for 6 states.  By the end of fiscal year 2005, information from 12 states will be included in “StreamStats.”
  • Development of new technologies to measure flood water levels that heretofore were too dangerous or practically impossible to measure. Accurate determination of the magnitude of floods is essential for establishment of flood-frequency relationships, required for long-term hazard assessment and design of critical infrastructure. These technologies include hydroacoustic current profilers and totally non-contact methods to measure river discharge from the ground or the air (see  These technologies keep personnel out of high flowing streams and increase the margin of safety when taking streamflow measurements in hazardous conditions.

USGS will continue to work with partners at the Federal, State, and local level to assist in efforts to reduce the impact that severe storms have on lives and communities.  Natural hazards, such as hurricanes and inland flooding, will always be with us and may be difficult to predict.  With USGS science, however, we are striving to prevent these natural hazards from becoming natural disasters.  Our efforts in hazards monitoring and long-term data and information collection from past and present hazard events is not simply a scientific research endeavor - - it is a matter of public safety.

Mr. Chairman, thank you for the opportunity to appear before you today.  I am happy to answer any questions that you and Members of the Subcommittee may have.

Was this page helpful?

Please provide a comment