Gustav bears down on U.S. East Coast: the Hurricane Scenario: Gustav, an intense Cat 4 storm, wreaks supreme havoc upon the Eastern Seaboard, leaving a trail of losses from Key West to Cape Cod

Risk & Insurance, April 15, 2004 by Auguste Boissonnade

Since the devastating hurricane that battered Galveston in 1900 and claimed more than 8,000 lives, major investments have been made in better preparedness, and improved methods of tracking and forecasting hurricanes. As a result there are today far fewer deaths from hurricanes, despite a dramatic increase in the number of people living along the coasts. Between 1900 and 1960, 12 hurricanes caused more than 200 deaths each. Since the 1960s, the average number of people who die every year from hurricanes is 25.

Economic losses due to hurricanes, however, have increased steadily as more people choose to live by the shore.

The highest historical reconstructed losses (modeled with current exposures and values) in the past 100 years would come from a Category Four hurricane that hit Miami in 1926, causing $87 billion in damages and $47 billion in insured losses. The Category Three hurricane that hit New England in 1938 would cause $28 billion in damages and $17 billion in insured losses. The small but intense Category Five, Hurricane Andrew, in 1992 would cause $38 billion in damages and $23 billion in insured losses.

About 80 percent of U.S. coastal residents from Texas to Maine have not experienced a direct hit from a major hurricane since 1900. Risk Management Solutions first developed a catastrophe model for hurricane loss in 1993 and has revised this model over subsequent years. The model simulates tens of thousands of hurricanes that represent the frequency and severity of hurricanes likely to cause significant loss in the United States.

This model shows that there is a 1 percent annual probability that a direct economic loss of $75 billion or more could occur somewhere in the United States from the occurrence of an hurricane--i.e, a $75 billion economic loss has a 100-year "return period."

Results obtained with such a model show that there is a wide range of simulated hurricanes that could give such losses with 1 percent annual probability.

Gustav is a hypothetical scenario. This storm is similar to the "Great Lake Okeechobee" hurricane of September 1928. While the track and the impacts are different, Gustav illustrates the impacts of one of the many intense hurricanes that could strike this region.

Most of the economic losses resulting from Gustav are due to wind and wind-related perils. Surge and rainfall perils contributed to a relatively small portion of the total economic losses. This might not be true for other areas such as New York, for example, where the funneling coastline makes the region more vulnerable to coastal storm surge.

Gustav is by no means a worst-case hurricane scenario. Hurricanes with the same intensity could make a direct hit on a large metropolitan area, such as Miami, which would, in turn, have resulted in larger losses. Intense surge and flooding could have an impact on protective flood structures, such as in New Orleans, leading to widespread urban flooding. It would even be possible for Miami and New Orleans to be hit by the same hurricane in the space of two days. This almost happened in 1992.