NEW YORK (MainStreet) – Monday is Labor Day, and a lot of people will use the day to hold the last cookout of the season. There’s just one problem: In many parts of the country, it’s forecasted to rain that day.
According to Weather.com, the New York City area is due for scattered thunderstorms on Monday, with a 40% chance of precipitation. Accuweather is even more pessimistic, predicting rain for most of the day and a thunderstorm at some point. Weather Underground, yet another service, gives a somewhat sunnier outlook, but still says that there’s a 30% chance of rain that day.
Regardless of the predictions, don’t go canceling your barbecue just yet. While the science of meteorology has come a long way, weather is still very hard to predict if you try to peer too far into the future. And when you get into the business of predicting weather conditions a week or more in advance, the crystal ball gets very hazy indeed.
“Once you get out to the seven-plus-day range, the specifics are often difficult to pin down,” says Jon Nese, a professor of meteorology at Pennsylvania State University. “In general, the farther out you go, the less certainty you have.”
To test this out, we printed out the 10-day forecast for New York from Weather.com on the Friday before last, then observed how well it matched up to actual weather conditions. While some forecasted “isolated thunderstorms” did not materialize on Saturday, the forecast was otherwise spot-on for the next four days. It wasn’t until day six that things broke down: Thursday was forecasted to be partly cloudy but was rainy, and Friday was supposed to have isolated thunderstorms but was instead a clear day. And this past weekend, forecasted a week in advance to be cloudy with a slim chance of precipitation, instead saw New York hit by Hurricane Irene.
Nese says that it’s fair to give the Weather Channel a pass on the weekend, as hurricanes are singularly unpredictable events; while hurricane tracking is also much better than it used to be, he says that even 72 hours in advance there’s still a margin of error of about 200 miles when projecting a hurricane’s path. So it’s hard to criticize the Weather Channel for not seeing it coming more than a week in advance.
Even if we take the hurricane out of the equation, though, the fact remains that services like the Weather Channel are on fairly shaky ground when they try to predict the weather 10 days in advance. Nese says that at that timeframe you’ll likely have just as much success looking at the “normal” weather conditions for that area during that time of year rather than trying to predict things based on computer models. Yet that hasn’t stopped weather services from telling you exactly what to expect 10 days in advance – or in the case of Accuweather, a full 15 days early.
“Forecasting services put a specific number on there just like they do on day one, and that’s always bothered me,” says Nese, who used to work at the Weather Channel. “It gives the false impression that we have the same certainty seven days down the road.”
Indeed, none of the weather services communicate their level of uncertainty on these long-term forecasts, except by hedging through less forceful language (like “isolated thunderstorms” instead of just “thunderstorms”). As such, it’s fair to say that many people may be reading too much into these long-term forecasts when planning events down the road, whether that’s a vacation or a Labor Day cookout.
“I would say the trend is that clients do look at that,” says Julie Sturgeon, who owns the travel agency Curing Cold Feet. Still, she thinks that on some level most travelers realize there’s a degree of uncertainty with such long-term forecasts, and that they mainly look at them for peace of mind.
“They have this desire for their vacation to be good, so if they see crappy weather 10 days out, they say, ‘How do they know that?’,” she says. “But if it’s giving them the answer they want, it’s true.”
—For the best rates on loans, bank accounts and credit cards, enter your ZIP code at BankingMyWay.com.