What we've learned about tornadoes since 1998


I've seen a lot of big storms in Minnesota.

The May 1965 Twin Cities tornado outbreak is my first living memory. I'll never forget the green sky and fist-sized hail pounding the ground.

I've seen a thousand thunderstorms in Minnesota, and I can remember something about nearly every one of them.

Create a More Connected Minnesota

MPR News is your trusted resource for the news you need. With your support, MPR News brings accessible, courageous journalism and authentic conversation to everyone - free of paywalls and barriers. Your gift makes a difference.


That's why when I was living in Arizona on March 29, 1998, and heard the news of the Comfrey and St. Peter tornadoes, I got a sick feeling in my stomach. My mind immediately went back to that May day in 1965 when I was just a boy.

On May 6, 1965, six tornadoes skipped though the western metro, killing 13 and injuring 683 people. This event is still listed as the fifth most important weather event of the 20th century in Minnesota, by the University of Minnesota's Climate Working Group.

The NWS made remarkable advances in severe storm forecasting, radar meteorology and warnings between 1965 and 1998. There are amazing differences between the Twin Cities NWS radar images from May 6, 1965 and March 29, 1998.

It is tragic, but amazing, that only two people were killed in the 1998 Comfrey and St. Peter tornadoes. A big reason for the lower death toll was the work of meteorologist Craig Edwards and the NWS team here in the Twin Cities. Their coverage and warnings that day saved lives.

Tornado bearing down
A resident of St. Peter took this photo of the tornado bearing down on the town on March 29, 1998.
Photo courtesy of Jerry Hawbaker

Even though warnings were issued in 1965, and the civil defense sirens were used for the first time in Twin Cities history, Craig and his team were able to give greater lead time for the 1998 tornado outbreak.

In fact, there was an awareness of the potential that a life-threatening event was on the horizon days in advance.

"Leading up to the tornado outbreak on March 29, there were several days of unseasonably warm weather in southern Minnesota," Edwards recalled. "On Friday, the computer models had the forecasters convinced that we needed to get a strategy in place to handle the probable onslaught of severe thunderstorms on Sunday afternoon."

On Sunday morning, the forecasters on the midnight shift laid out the alert, with highlights of the possibility of tornadoes later in the day. Citizens were put on notice that severe storms were likely," said Edwards.

"The 14 tornadoes seen on March 29, 1998, more than doubled the total number of March tornadoes seen in the state of Minnesota prior to 1998."

As the event began, Edwards recalled some fortunate breaks in where storms developed.

"One thing that played into our favor was the initial storm development igniting just to the north and east of Sioux Falls. Their (Sioux Falls NWS office) communication of vital information on ground truth of storm severity, greatly aided our confidence in warning decisions," said Edwards.

As the event progressed, Edwards says things were uneasily quiet in the Twin Cities NWS office.

"As the storm moved towards St. Peter and the warning was issued for Nicollet County, we had no ground truth of the storm's intensity," said Edwards. "Sirens wailed and residents sought shelter. It wasn't until about 30 minutes after the storm had moved through St. Peter that we were able to begin to gather information on the destruction."

In some areas, the tornado cut off sirens as they sounded to warn of the impending storm. Though we've grown to depend on sirens for tornado warnings, it is important to remember that they are intended primarily as an outdoor warning system.

There were several unusual aspects to the 1998 outbreak. The first and most obvious is occurrence of F2 and F3 tornadoes well outside the bounds of the usual tornado season. The normal peak for tornadoes in Minnesota occurs in mid-June.

Here's an excerpt of the NWS summary of the event.

Tornado track
A map showing the track of the March 29, 1998 tornadoes which struck south central Minnesota.
Image courtesy of the National Weather Service

"Prior to 1998, only five tornadoes with a Fujita rating of F2 or greater had been recorded in the month of March in the entire state of Minnesota --- none of which occurred on the same day.

"In stark contrast to the previous March tornado climatology, the 14 tornadoes seen on March 29, 1998, more than doubled the total number of March tornadoes seen in the state of Minnesota prior to 1998.

"The six F2 or stronger tornadoes also easily eclipsed the total number of strong and/or violent March tornadoes in the state prior to the 1998 outbreak."

The second remarkable aspect is the incredible long track of the Comfrey tornado. The storm was 900 yards wide -- nearly a half mile wide. It was rated an F3 and briefly strengthened to an F4 storm, tearing a path 67 miles long through five counties.

It is the fifth longest tornado track in Minnesota history, according the NWS. If you consider only tornadoes with continuous path length, it may have been the longest.

The March 29, 1998 tornado outbreak was a brutal reminder that tornadoes can occur anytime, anywhere, regardless of the season if meteorological conditions are right.

The fact is that another outbreak like 1965 and 1998 is due to occur again in an urban area such as the Twin Cities. It will happen; it's just a matter of when.

We also learned that advances in severe weather forecasting and technology, while not perfect, have made us safer. We can be thankful that the NWS and the media put so much emphasis into warning for severe storms.

Warnings are great, but the state of the science of meteorology means that not every future tornado will have a warning attached, or a siren wailing. We need to listen for the warnings, but also have a "situational awareness" that severe weather is possible on any given day.

The fact that modern meteorology can provide that awareness may be the biggest meteorological advance of our lifetime.