Forecasting the Big Game — 1967 vs 2014

Read NCAR's AtmosNews post:

A high of 38 and a low of 23. Mild winds and a small chance of precipitation.

After all the hand-wringing over the first Super Bowl to be played outdoors in a cold-weather climate, the Super Bowl XLVIII forecast as of late Tuesday called for what most would consider fully acceptable conditions for a football game.

It might not seem like such a great leap forward to be able to make a forecast with some certainty five days before the Denver Broncos and Seattle Seahawks finally face off.

But in January 1967, the year that the Green Bay Packers topped the Kansas City Chiefs 35-10 in Los Angeles in the first-ever Super Bowl — the league wouldn't actually start calling the event the Super Bowl until the following year — weather forecasts did not look beyond 48 hours.

David Hosansky, media relations manager and science writer for the National Center for Atmospheric Research, authored a blog post for NCAR's AtmosNews in which he highlighted just one example of the potential costs of such limited past forecasting skills.


On Wednesday afternoon, Jan. 25, 1967, just 10 days after the first Super Bowl, the forecast for Chicago by the U.S. Weather Bureau — the precursor to the National Weather Service — showed "cloudy with rain or snow likely" for the next day, a Thursday, with a 50 percent chance of precipitation. For that Friday, the bureau foresaw "rain or snow ending."

What actually happened over those next two days was the city's worst blizzard on record, with 23 inches of snow, 4-to-6-foot drifts, 60 deaths, the closing of O'Hare Airport for three days and $1 billion in damages in 2014 dollars.

But huge weather events sneaking up on a region are not, of course, entirely a thing of the past.

Many in the Boulder area believe the historic rains of September managed to descend and unleash their fury well under the NWS radar. But Hosansky says an expanding array of tools gives forecasters the ability to clearly see what's coming in East Rutherford, N.J., five days away — and also to not miss a cataclysmic storm such as what hit Chicago when it is looming just 24 hours away.

Seeing how the global can influence the local

"I think it's important for society to realize how much weather forecasting has improved and perhaps even revolutionized our lives," Hosansky said. "Fifty years ago, it was unfathomable that we would know what the weather might be like a week from now. To be able to predict the future is an incredible achievement, I think."

Hosansky's post highlights several forecasting developments by NCAR and its collaborators that he sees as pushing the window of reliable forecasting forward. One is the Model for Prediction Across Scales, with which researchers can run computer simulations of the whole planet, while also targeting specific geographic regions. MPAS, as it's called, recently was successful in showing a cluster of severe thunderstorms in the Midwest four days in advance and tropical cyclones in the Pacific six days out.

"The way weather works is that the weather here might be affected by what's happening in the atmosphere thousands of miles away," Hosansky said. "MPAS right now is being used for research, but that type of approach might work its way into forecasts in the near future."

One potential use for the MPAS technology being examined, Hosansky said, might soon lead to reliable seven-day forecasts for the paths of hurricanes, whereas the outside range on such forecasts now is considered to be only three to five days.

'The progress is slow'

Other projects through which NCAR researchers are hoping to advance forecasting include a three-year national project funded by the Department of Energy to develop 36-hour, fine-scale forecasts of incoming energy from the sun, which could have implications for utilities anticipating solar energy output. Another is examining the so-called Madden-Julian Oscillation, an atmospheric event in the Indian Ocean capable of sending out atmospheric waves that can trigger storms in the United States more than a week later.

"We're making progress, but the progress is slow," said Bill Skamarock, a senior scientist at NCAR involved in the MPAS research. "It's a hard problem, and there are a lot of facets to it that we don't have a lot of understanding in, and that lack of understanding is translated in our models.

"We know there are deficiencies in our models, but we need to better understand the key physics of the atmosphere to know where we need to improve our models."

Faster computing speeds — the goal is to be increasing them by factors of 100 — and ever-higher-resolution satellite scans of features such as mountains, coastal regions, clouds and weather systems will be pivotal in developing reliable longer-range forecasts.

But whatever happens Sunday, it can be boldly forecast that there will be much less talk about the weather in advance of Super Bowl XLIX, to be played in January 2015 — unless it's about the heat. The game will be contested in Glendale, Ariz.

Contact Camera Staff Writer Charlie Brennan at 303-473-1327 or