By noon three inches had fallen and by nightfall we had nearly half a foot of the white stuff on the ground. The extreme cold made for a fluffy, powdery snow rather than the wet stuff that fell a few days earlier, so I guess you could argue that it was a “light” snow. All six inches of it.
Weather forecasters are easy targets. Forecasting is an inexact science, subject to countless variables that can’t always be measured or properly interpreted. Still, it’s hard not to be envious of a job where you can be wrong half the time and still get a sizeable raise at the end of the year.
But I’m not blaming the weathermen (or women). They do the best they can with the available information, and the differences in their forecasts only underscore the difficulty of making the right predictions.
AccuWeather itself is another matter.
Entire regions of the world depend on AccuWeather’s forecasts. But how much faith can you put in them? Jason Samenow of The Washington Post did a study of AccuWeather’s 25-day forecasts and found that predicted temperatures often were wrong by as much as 10 degrees -- basically the same rate as if one merely guessed -- and that half the rain that fell was never forecast.
Many people believe that Accu-Weather provides its forecasts as a public service. Not so. The for-profit American media company uses much of the data provided by the National Weather Service, repackaging it and selling it to newspapers and radio and TV stations around the world.
The National Weather Service, on the other hand, provides its information for free in the public domain.
(Interestingly, back in 2005, U.S. Sen. Rick Santorum, R-PA, introduced a bill that would have prohibited the National Weather Service from providing ANY information directly to the public. The bill, which thankfully never came up for a vote, would have ensured that someone [Accu-Weather, perhaps?] would have profited from the NWS’ taxpayer-funded research by banning the free delivery of that information. [I’m sure it’s just a coincidence that Santorum received campaign contributions from Joel Myers, president of AccuWeather.] But I digress.)
OK, so AccuWeather wants to make a buck. I’d have no problem with that, if we were getting any bang for our buck. But what we get are unreliable forecasts, predictions so far into the future that most weather experts agree they have no bearing on reality, and a constant stream of weather warnings designed to scare the pants off of us (and keep us listening/reading for updates).
What irks me is that AccuWeather apparently can’t or won’t staff its offices round-the-clock to ensure that the information it gives IS the latest, most up-to-date data available.
Instead what we get are same-day forecasts that often are proven wrong at the very moment you read or hear them.
Again, I understand it’s not easy to predict the weather. What I don’t understand is how AccuWeather can allow itself to appear idiotic by failing to update those forecasts. For example, after I finished shoveling last Sunday, with the snow still steadily falling, the forecast still called for “light snow” We had more snow Sunday than fell during several “winter storm warnings” There was no warning for Sunday, yet the roads were slick and dangerous, more so because salt crews rely on forecasts to plan their staffing.
Why isn’t someone on staff available to correct such obvious contradictions?
It’s far from rare that forecasts clash with reality.
For three straight days in the first week of February, I logged onto the Web about 9 a.m. to check the local forecast. The reports called for highs in the mid-20s with lows in the upper teens. At the very moment I was reading the forecasts, the temperature outside was in the single digits. One forecast called for a low of 22. At 9 a.m. that morning, it was 3 degrees. Now that’s accuracy.
It doesn’t inspire much confidence in AccuWeather’s predictions when they’re telling you the day’s low will be 19 degrees warmer than the current temperature.
Just to avoid embarrassment, you’d think AccuWeather would have someone available, or at least some sort of computer program, capable of revising the forecast to reflect current conditions. Instead, we get predictions of rain when there’s not a cloud to be seen, rain when it’s supposed to be sunny and temperatures that have no relationship to what’s predicted.
Evidently the company doesn’t really care if its forecast is correct — as long as it’s getting paid.