GRAND JUNCTION – There are lots of jokes about weathermen getting it wrong. George Carlin mocked the profession, and its seemingly obvious conclusions: “Weather forecast for tonight: dark.”
Larry David has openly accused weathermen of predicting rain just so they can hog golf courses.
Jim Pringle, warning coordination meteorologist at the National Weather Service in Grand Junction, has heard them all. When he started with the Weather Service 37 years ago, laymen would constantly rib Pringle for failed predictions.
This week, when a reporter asked Pringle about the incorrect weather predictions he most regrets from his early career, Pringle demurred, saying he couldn’t remember one specifically. While answering, Pringle seemed suddenly uncomfortable, abruptly shifting his weight in his seat several times.
But, he said, in the last few years, he’s been getting a lot less guff from laymen.
“I don’t want to brag,” he said, “so I better not say anything. But there have been times when weather has happened, and we’ve issued a forecast for a storm that hit pretty good,” he said, beaming.
Pringle’s career with the National Weather Service has coincided with a boom in forecasting accuracy.
In 1972, the service’s temperature forecasts for weather three days out missed by six degrees on average. These days, it’s down to three degrees.
In this line of work, small improvements in accuracy can have huge social repercussions. In 1940, Americans stood a 1 in 400,000 chance of dying by lightning strike. Today, because of better storm predicting, it’s 1 in 11 million.
Breakthroughs in weather prediction are founded on Edward Lorenz’s 1972 insight that a butterfly flapping its wings in Brazil might set off a tornado in Texas.
“It’s based on chaos theory,” Pringle said.
He said increased computing power combined with meteorologists’ understanding that weather is based on exponents, not arithmetic, has yielded much more sophisticated weather modelling.
Some of the greatest strides have been in hurricane forecasting. Even 25 years ago, when trying to predict hurricane landings three days in advance, the National Hurricane Center missed by 350 miles – on average. Yet in 2012, the National Hurricane Center correctly predicted that Hurricane Sandy would descend on southern New Jersey nearly five days ahead of the storm’s landfall.
The center’s accuracy saved lives, allowing people within its path of destruction to evacuate. Meanwhile, there was no false alarm for the majority of the East Coast, where people could continue life as usual – attending weddings, closing business deals, making flights and letting kids bike to school.
As forecasting science advances, the nuances of microclimates such as Durango’s can still fell the most mathematically sophisticated weather models and data-gathering technology.
Joe Ramey, climate forecaster with the National Weather Service in Grand Junction, said, “Durango is very difficult for us. We don’t have very good radar coverage down there.”
Yet, in rural Colorado, where weather is a near-obessive subject of lay speculation, forecasting science is eclipsing its oldest detractor: local wisdom.
In late April, Julie Malingowski with the National Weather Service in Grand Junction predicted a storm would shower at least an inch of snow on Durango over the weekend – a laughingstock on the Friday she forecast it when Durango was awash in sun and temperatures flirted with the 70s.
Sure enough, by Sunday morning, Durango had received an inch of snow. As she predicted, temperatures remained wintry for the next five days.
cmcallister@durangoherald.com
