ANALYSIS: Why All The Data In The World Isn't Enough For Storm Predictions

It’s been the kind of winter that has left residents and officials weary, not just because of the snow and cold, but also because of weather forecasts gone awry at the Jersey Shore.

There was the blizzard-that-wasn’t in late January, where weather forecasters were calling for as much as three feet of snow as the storm approached, resulting in widespread closures of businesses, schools and a state of emergency being declared by Gov. Chris Christie.

When the worst-case scenario fizzled into a storm that dropped less than half of the predicted totals, forecasters were blistered by residents and government officials. Several even issued apologies.

In the weeks following, forecasts were scaled back -- and were for the most part on the mark.

Until Friday. Forecasts leading into the day called for Wall and the Jersey Shore to get perhaps 2 inches, if that much.

As the snow continued to fall well into Friday evening, however, residents were howling, over plowing that started later, over snow total predictions that rose throughout the day as the National Weather Service continued to extend its winter weather advisory -- initially scheduled to be lifted at 8 p.m., the advisory was finally lifted at the Shore at 4 a.m. Saturday. The National Weather Service’s snowfall total report for Wall said 5.5 inches fell.

Bob “Weatherman” Burger, a popular forecaster in the Shore area who posts his forecasts on Facebook, said in status update Friday evening that heavy snow bands simply surprised everyone: “I don’t know of a weather service or forecaster that saw this coming ... in a word, oooopps ...”

How does that happen, you wonder? An article that appeared in the New York Times Magazine in September 2012 sheds some light on the subject.

The article by Nate Silver, titled, “The Weatherman Is Not A Moron,” details how weather forecasting has improved significantly over the last several decades, including when it comes to things such as predicting where a hurricane will make landfall. (Interestingly, the article came out more than a month before Hurricane Sandy made a direct hit on the Jersey Shore, devastating thousands of homes, particularly in Monmouth and Ocean counties -- a direct hit that was predicted for almost five days before Sandy arrived.)

The Times’ piece gives some history of weather prediction in the United States, and efforts over the years to improve it.

But the basic problem with weather forecasting, it says, is “our knowledge of its initial conditions is highly imperfect, both in theory and in practice ... we can observe our surroundings with only a certain degree of precision. No thermometer is perfect, and it isn’t physically possible to stick one into every molecule in the atmosphere.”

Silver’s analysis also noted another issue:

“Weather also has two additional properties that make forecasting even more difficult. First, weather is nonlinear, meaning it abides by exponential rather than arithmetic relationships. Second, it’s dynamic -- its behavior at one point in time influences its behavior in the future.”

In some ways, predicting the weather might be likened to herding cats: a perfect outcome is almost impossible, because there’s always at least one that will do something you won’t expect.

If you really want to understand the complexity of predicting the weather, read more of Silver’s piece here.