Thursday, November 20, 2014

Snow Day Forecasts

It was cold this week. The Raleigh-Durham airport (RDU) recorded a record loW of 19°F Wednesday morning. That's plenty cold for mid-November, but cold enough to cause the local schools to delay opening? ... Nope, not even close. If last winter is any indication, it takes, at minimum, a "Wind Chill Advisory" - issued for wind chills below 0°F - for schools to delay simply because it's "too cold", and even then, not every Wind Chill Advisory results in a delay.

Speaking of which...last winter was cold, too, and there were plenty of school delays and closings. Wake County schools had 9 snow days last year, delayed opening twice, and closed early twice more. (None of those were in November or December, by the way.) Schools farther north in Virginia had even more snow days than we did. And in the lead-up to each winter weather event, of all the meteorologists I follow on Twitter and read on the internet, this question generally went unanswered: "Given the forecast, how likely is it that schools will be closed or delayed?" The excellent Capital Weather Gang communicates the likelihood of school closings on a scale of 0 apples to 4 apples, but that's the closest thing I've seen, and I'd prefer something more percentage-based anyway. (And, of course, something local to the Triangle.)

I actually thought about doing this myself. I could start up a Twitter feed and/or blog and/or something that specializes in "snow day forecasts". I wouldn't be forecasting the weather itself, because I know I couldn't do any better than NWS Raleigh or the local TV meteorologists. Instead, I'd read all of the forecasts and technical discussions, compare that to past events - I chronicled each potential event last winter in order to determine the thresholds for school delays and closures - and combined with the uncertainty in the weather forecast itself, come up with something like, "Durham County Schools are 40% likely to be closed for the day on Thursday, 60% likely to have at least a delay". I think I could make snow day forecasts as well as anyone. Maybe even better than the school systems themselves!

Forecasting snow days might even be harder than forecasting the weather itself because of the element of human behavior, on top of the already uncertain weather, so I expect this to be a bit challenging. Maybe that's why nobody does it!

So, what's stopping me from starting now? Well, first off, given that we just had our second child a few months ago, now is probably not the best time to start a new project. But also, I think I need one winter to "practice". So, maybe I'll tweet out some "beta" snow day predictions this winter on my personal Twitter. We'll see how it goes, and if it goes well, maybe I'll start something a little more formal the following winter. Hopefully I'll get plenty of opportunities this season! But not too many.

1 comment:

Spartangoogle said...

North East had no delays or closings due to the recent storm.