Spring floods on the Red River

Guest post by Phil Gerla, associate professor of geology and geological engineering

Since 1904, the U.S. Geological Survey has collected daily flow data for the Red River at Grand Forks. Chris Laveau (UND M.S. 2005) currently leads the agency’s Grand Forks office and his hard working team continues to provide valuable long-term water-resource data, which are crucial to flood prediction and water management.

Not only has the size of the Red’s spring floods varied greatly through the years, but the timing of peak flow can occur within a wide range of dates. The bar graph shows the number of years during the 115-year record that the river first crested on a particular day in the spring. Some of the bars are labeled to show recent records. On average, the Red’s spring flood peaks most often on or around April 6. Remarkably, both the earliest (February 28, 2017) and the latest (April 30, 2013) peak in this long record happened just recently and only four years apart!

The river flow — its discharge — is not shown in the plot, but examination of the USGS records shows that 1981 was an unusual year with the lowest and most protracted spring peak, reaching a maximum flow of only 1,470 cubic feet per second (cfs) on March 28. In contrast, the large and destructive flood of 1997 crested on April 18 when it reached an estimated 137,000 cfs, almost 100 times more than the 1981 peak flow!

Many factors contribute to the timing of the spring peak flow, including snow cover, depth of frozen ground, fall soil moisture, location of deepest snowpack in the basin, prevailing global weather patterns, among others. Since 1904, five of the ten largest spring floods have occurred in the last 20 years, with the other five happening between 1978 and 2000. None of the ten smallest floods have occurred since 2000. The flow of the Red River has increased during the last century. Interestingly, spring floods of the last two decades have contributed significantly to the early scatter of peaks in mid-March, which may reflect a changing climate in our region.

The bar graph has a typical “bell shape” centered on the second week of April. But there are two notable features: first is the scattering of years when the peak was unusually early (falling on a day prior to March 21). The second is a surprising gap in the middle of the bell curve. In 115 years of records, no flood peak has been recorded on April 9, and only a couple of years when the peak happened on April 7 and 8. I can only attribute the paucity of flood peaks on those particular days to coincidence.

It looks like this year’s peak will not arrive until mid- or late April.

All of the data for the Grand Forks river gauge can be accessed here.