When the Levee Breaks: Part One
(WDAY News Script)
Since the disaster at Grand Forks and East Grand Forks, the National Weather Service has been receiving a lot of criticism for their river forecast...

Marv/News Center 6:
Our meteorologist John Wheeler has been investigating this matter for the last several weeks. Tonight, and the following two nights, we will bring you what John learned about. In part one of "When the Levee Breaks" John looks at what made this year's flood so difficult to forecast.

Sounds Of... Drum Beat, Guitar..... Wind

John Wheeler/News Center 6:
It was during our barrage of blizzards back during November, December, and January, that the engineers and hydrologists of the National Weather Service river forecast center in the Twin Cities make began worrying about the potential for extraordinary spring flooding on the Red River. Located in Chanhassen, Minnesota, the river center produces all the flood predictions for the major rivers in the midwest. Grand Forks weather service hydrologist Wendy Pearson then relays this information to local officials... And answers questions. Their early outlooks are based on the measured water content of the snowpack. With consideration given to average late winter precipitation and average melting conditions. The process is made difficult by the flat terrain of the Red River Valley. Run off tends to be slow. Especially if, like this year, the ditches are full of ice and snow. Weather also complicates the forecast. Precipitation and temperature are rarely average in this part of the country.

Sounds Of... Man Talking

John Wheeler/News Center 6:
Using past floods as guidance, computers translate the volume of water moving into the river into flow rates, which usually correspond to actual river levels. But this year, there was more water in the snowpack than the weather service's computer model had ever dealt with before. There was no reference point to go by. River forecaster Mike Anderson says this year's flood was their greatest challenge ever.

Mike Anderson/ Hydrologist:
We're quite proud of what we did do at the other locations. Our modelling techniques held tight. We just didn't have a real solid handle on everything that came into play; the bridges, the levees, and everything; at that heighth in the city of Grand Forks.

Sounds Of... Helicopter Propeller

John Wheeler/News Center 6:
The river at Grand Forks East Grand Forks behaved differently than elsewhere in the valley, rising at two feet a day right up to the crest. The weather service computer models kept indicating that the rise of the Red would slow, but it never did, until the two town's dikes, built three feet higher than the original crest prediction, either broke down or were overrun. Clearly the forecast process failed. But why?

Sounds Of... Wind Roaring

John Wheeler/News Center 6:
One problem was the weather. Upstream at Fargo Moorhead, freezing temperatures had slowed the field run off, producing a broader and lower crest. But as the crest approached Grand Forks, the weather warmed and field runoff was quicker. The Red River spread out ten miles wide and then was channeled between the dikes, like into a bottleneck.

Mike Anderson/National Weather Service:
In Grand Forks, all the bridges except the Kennedy Bridge had water either on them or over them, and that put more or less a cap on the top of the stream where you had the water pushing up against them, and obviously the volume of water that came into that town at one, key spot. And you have the levees constructed, and they were on top of the levees at that time with more sandbags creating straight walls instead of the channel-shape walls. And you had this on the side and the bridges on top and it was almost a hydrolic pipe flow going through. And there's the hourglass.

Sounds Of... Man Talking

John Wheeler/News Center 6:
Actually, the weather service's predictions of the rate of flow through Grand Forks were correct. But the river level expected to correlate with that flow rate was wrong. Here's another twist, at most locations along the red river, the crest was one to two feet higher than the previous flood of the century, but at Grand Forks, the crest was five feet higher, while just downstream at Oslo, the crest was actually a half a foot lower than the previous record, further indicating that there was something obstructing the flow of the river.

Mike Anderson/National Weather Service:
It's been very emotionally draining on us in the River Center. Two of the forecasters that worked the Red River, I'm originally from Wahpeton. The other forecaster that was with me during the three or four week event, his grandparents grew up in Grand Forks and that house is no longer there. It's very emotional. We tried to do what we thought was a professional engineering job. It was beyond our model at that time and now as we go back we can start to put the pieces back together and see more about what we did.

John Wheeler/News Center 6:
John wheeler... News center six