Floodwatchers aim to improve forecasting

Snow-covered trees sit in floodwaters in Moorhead
Snow-covered pine trees sit in flood water March 31, 2009 near Moorhead, Minnesota. A snowstorm has slowed recovery efforts as residents of Moorhead and neighboring Fargo, North Dakota try to return to their homes as the Red River begins to slowly recede.
Scott Olson/Getty Images

Six months after flood waters threatened to devastate the southern Red River Valley, federal officials gathered in St. Paul today to talk about why they were wrong in a key area -- the flood forecasts.

They said they're already making adjustments to better prepare residents in the Fargo-Moorhead area next time.

During the prolonged flooding this past spring, the Red River was above flood stage for 61 days -- two months. It took about 18 million sandbags and untold loads of clay and dirt to hold it back.

The U.S. Army Corps of Engineers alone estimates it spent about $30 million to prevent more than $2 billion in damage.

Political Coverage Powered by You

Your gift today creates a more connected Minnesota. MPR News is your trusted resource for election coverage, reporting and breaking news. With your support, MPR News brings accessible, courageous journalism and authentic conversation to everyone - free of paywalls and barriers. Your gift makes a difference.

Three deaths were linked to the flood fight.

But 250 miles from the banks of the Red River, experts gathered during those weeks to talk about what might have been the most important number of all, the shifting crest prediction for the people at the southern end of the Red River.

Forecasters predicted the river would crest as high as 43 feet, which turned out to be nearly 2 feet higher than the actual peak. They warned that a second crest could be even worse, but it never happened. And as bad as the flooding was, the worst fears never came true.

Crests over the years
This is a National Weather Service computer plot showing 57 different possible flood crests on the Red River this month, based on weather since 1949. Back during the last major flood, the National Weather Service didn't have the computer power to run more than a handful of possible flood scenarios. The black line represents only one possible scenario, if precipitation and snowmelt parallel that in 2004. The actual forecast is between 38 and 40 feet.
Graphic courtesy of the National Weather Service

It was an ironic contrast to the unprecedented flooding on the Red River in 1997, when the crest surged past the forecast peak and overwhelmed defenses in Grand Forks.

National Weather Service Hydrologist Steve Buan said the floodwaters were off the forecasters' charts in a different way this time.

"This was the highest, earlier crest in over 120 years of records on the Red River of the North," said Buan. "From the initial point of melt to the peak in the Fargo-Moorhead area, it was the fastest it had ever occurred. And to have our models have to understand that, and then, with that 20 inches of snow that fell, plus the residual moisture out on the landscape, it overwhelmed our capabilities."

Buan said hydrological engineers have been studying the flows and peaks of the flooding and rewriting their models. They're incorporating the river's surprises into the calculations forecasters will use next time.

But it isn't just a matter of mathematics.

Officials from the National Weather Service, the Corps of Engineers and the U.S. Geological Survey are also talking about the human element -- the nature of predictions and flood preparations.

Forecasters have tried to finely weight the risk of flooding against the waste of overpreparing, typically centering on a single crest number.

Chuck Shadie, a hydraulic engineer with the Corps of Engineers in Vicksburg, Miss., said flood watchers are rethinking how they talk about the high water to the public -- and letting the public play a larger role in gauging the risk.

"There's still some uncertainty out there as to how to use the data that they're being supplied. Do we give you one single value, or do you get a range of values?" said Shadie. "I think we're getting some valuable information today that we can take back and we can say, 'OK, we've got the products, but they're not sure what to do with that.' So we need to focus on that some."

This is the second flood summit in as many years for the agencies that monitor the nation's waterways. They gathered last October to talk about the flooding that devastated Cedar Rapids, Iowa, and other parts of the Midwest.

They hope that regular hindsight will someday improve their flood foresight.