A deadly flood that helped improve weather forecasting

A look at progress in 40 years since Colorado's Big Thompson flood of July 31, 1976

As Coloradans celebrated the centennial of the state’s founding late on a hot summer afternoon 40 years ago, a brewing thunderstorm parked itself at the head of the Big Thompson Canyon.

Up to a foot of rain fell over the next three hours, unleashing a torrent of floodwater that swept 10-foot boulders down the canyon in the dark. Many of the estimated 2,500 to 3,000 people in the canyon were caught between the canyon’s steep walls. One hundred forty-four lives were lost, and 570 homes and businesses were destroyed in the worst flood in Colorado history. 

The Big Thompson flood of July, 31, 1976, was one of three major flash floods during the span of five years in the 1970s that killed more than 450 people across the country -- tragic events that helped spur the modernization of NOAA’s National Weather Service flood forecasting system.

“These three floods really provided the impetus for research into how to develop tools to better forecast and issue warnings for flash flooding events,” said Leroy Spayd, former chief of NWS Meteorological Services Division. 

A natural hazard

Flash floods are a naturally recurring aspect of weather along Colorado’s Front Range – and the West.  In the 21st century, Americans have come to expect timely, accurate forecasts and instantaneous warnings of severe weather.

But in 1976, teletypes, fax machines and telephones were state-of-the-art communication tools. Computers were in their infancy and numerical models produced columns of numbers spat out by dot-matrix printers. There was one radar station serving all of eastern Colorado. NOAA’s lone satellite sent a picture of the western United States every half-hour. Forecasters had to manually assemble information to compile a forecast. 

“In 1976, data was old by the time we got it,” said Nezette Rydell, lead forecaster with National Weather Service’s Denver-Boulder Forecast Office. “It took a long time to get observations transmitted to the forecast offices. The biggest change  in my mind is we now have the ability  to overlay radar on top of satellite imagery on top of surface readings and we can integrate it in front of us almost in real time.”

For Coloradoans, a big test of the new technologies came in 2013, when a massive storm dumped a year’s worth of rain on the northern Front Range in a week. Despite much more widespread damage from flooding, many fewer lives were lost.

Forecast improvements

Preventing tragedies such as these spurred Congress to fund research to develop better tools to provide more timely and accurate severe weather warnings. These three were a direct result of that initial research push.

• Better Radar: In the 1970s, the development of NEXRAD Doppler Radar technology, primarily at NOAA’s National Severe Storms Laboratory in Norman, Oklahoma, gave forecasters the ability to monitor precipitation rates and develop more precise estimates of precipitation amounts to improve flood and flash flood warnings. Deployed in the 1990s, more than 160 NEXRAD stations now in operation across the country provide blanket coverage of the continental United States and give forecasters a  powerful tool to monitor the development of  severe weather.  

• Faster Processing: In 1980, NOAA launched research program that led deployment of the Advanced Weather Interactive Processing System, or AWIPS. The system debuted in Denver in 1986, giving forecasters for the first time the ability to see integrated and animated meteorological and hydrological data along with satellite and radar imagery. It also generated simple, speedy severe weather warnings. AWIPS has undergone many upgrades and is still considered the cornerstone of NWS operations. The research program evolved into the Global Systems Division, a part of NOAA Research’s Earth Systems Research Lab in Boulder, which continues to develop new tools for the NWS.

• Automatic Observations: In the early 1990s, the Automated Surface Observing System began providing real-time weather observations for forecasters. Today, the system gathers weather data from 900 sites across the country, and serves as the nation’s a primary climatological monitoring network.

Global Systems Division researchers went on to develop revolutionary computing techniques such as Massively Parallel Processing, which became the foundation for the Rapid Update Cycle weather model, predecessor to NOAA’s current High-Resolution Rapid Refresh weather model.

Recently, National Severe Storms Laboratory’s Multi-Radar, Multi-Sensor system, went operational, providing better visual depictions of high-impact weather events such as heavy rain, snow, hail, tornadoes, and other threats. MRMS delivers more than 100 high-resolution products every two to five minutes, improving forecasters’ ability to issue public warnings and advisories for severe weather such as tornadoes, hail and flash floods. 

In November, forecasters will gain another powerful new tool with the launch of NOAA’s new GOES-R satellite, which will deliver three times the data with four times the resolution and five times the transmission speed. “We’re going to have so much more data - it will be another leap in our forecasting tool kit,” Rydell said.

For more information, please contact Theo Stein, NOAA Communications, at 303-497-6288 or by email at theo.stein@noaa.gov