CHAPEL HILL - Hurricane Gustav, this summer's flooding throughout the Midwest and tropical storm Hanna brewing in the Atlantic should pique our interest in how a large flood would affect North Carolina.

Our research shows that a large flood would substantially damage some areas of North Carolina, while other areas -- having learned from past floods and acted accordingly -- would be far less damaged. Floodplain managers should strictly limit development in flood-prone areas and continue their efforts at moving people and properties out of harm's way.

Floods remain the most costly natural hazard in the United States, with annual averages of 140 lives lost and $6 billion in property damage. Seven to eight million households are exposed to significant flood risk nationally and represent substantial economic liabilities. FEMA estimates that over 175,000 properties are located in flood-prone areas in North Carolina, and that only 60,000 are covered by flood insurance.

Flood policy is based on the 100-year floodplain boundary -- the extent of land that on average has a 1-in-100 (1 percent) chance of flooding in any given year. Similarly, the 500-year floodplain has a 1-in-500 (0.2 percent) chance of flooding in any year.

Despite these small percentages, we should not think that such events are unlikely to happen, or that they will not happen repeatedly. Similar to the Boston Red Sox going 86 years without winning a World Series but then winning two in four years, the Midwest experienced a "500-year flood" this summer -- its second in the past 15 years, a statistically unlikely but devastating event.

Floods of similar magnitude in North Carolina would likely cause even greater damage, since population growth here has far exceeded that in the Midwest.

l l l

THE GOVERNMENT-RUN NATIONAL FLOOD INSURANCE PROGRAM (NFIP) DISCOURAGES NEW DEVELOPMENT in the 100-year floodplain and encourages property owners to purchase flood insurance and adopt more stringent building requirements. The only source of insurance for properties inside the 100-year floodplain is the NFIP, as private industry will not take on such a liability.

Outside the 100-year boundary, there are no federal requirements for flood mitigation or insurance. Properties in this area, where the risk for flooding is only slightly less than the 100-year flood, are often unprepared, uninsured, unprotected and normally rebuilt after being damaged by floods.

We recently completed a study of the effectiveness of floodplain policies for limiting development in flood-prone areas by measuring the change in population in the 100-year and 500-year floodplains in five North Carolina counties between 1990 and 2000. Our study showed that current floodplain policies and implementation have been only marginally successful in limiting development: Orange, Durham, Wake and Buncombe counties all had increased population density and development in the 100-year floodplains.

In contrast, Craven County effectively reduced the number of people residing in the 100-year floodplain. Indeed, many eastern counties have reduced their vulnerability to floods through proactive hazard reduction efforts since Hurricanes Fran (1996) and Floyd (1999), which destroyed over 11,000 homes, the vast majority of which were in the 100-year floodplain.

Since these hurricanes, communities have used hazard mitigation grants to elevate homes or to acquire flood-prone properties. Proactive efforts after Fran reduced the losses during Floyd by almost $6 million in Kinston alone, at a cost of only $2 million.

Unfortunately, directly outside the 100-year floodplain, but within the 500-year floodplain, population density and infrastructure increased in all five counties. The lack of any policies for the 500-year floodplain allows development to proceed unabated, despite the very real flooding threat to these areas.

Reducing flood losses requires proactive planning and action. The success of flood reduction programs in Eastern North Carolina should serve as models for the potential long-term economic and safety benefits these programs could have throughout the state.

The low-hanging fruit for flood damage reduction is the limitation of development in the 100-year floodplain. Next must be a concerted effort to move people and properties out of these vulnerable areas. Over the long term, the state should also consider reducing development on 500-year floodplains.

(Martin Doyle is an associate professor of geography at UNC-Chapel Hill and a GlaxoSmithKline faculty fellow at the Institute for Emerging Issues. Lauren Patterson is a graduate student in geography at the university.)

Get $150+ in coupons in every Sunday N&O. Click here for convenient home delivery.