This morning, while making the forecast for the next 5 days, I saw an interesting scenario on the GFS long-range model. My first thought was “That looks impressive!” Then I wondered how many young, reckless, let-me-just-make-a-name-for-myself meteorologists were going to flood social networks like Facebook and Twitter with screaming headlines about a terrible hurricane hitting the southeastern coast exactly two weeks from today.
Yep. There’s a tropical system striking Georgia in that model.
I’m not mentioning it here for click bait or trying to be the first to talk about it. I mention it because it’s there and it will be interesting to watch how the model develops and changes the forecast from day to day. This morning, I took a screenshot of what I saw two weeks out for that exact time and date. I will continue to take screenshots between now and 06Z Thursday, August 17, 2017, of what the model is predicting for that exact time. In two weeks, I will post them, and we will see how things have truly developed from today through this morning ’s projected day of landfall.
I will give you a hint. I’m not expecting it to happen the way the GFS is predicting today.
I’ve written before about how the accuracy of the numerical weather prediction models decreases the farther out you go. Factors that the models aren’t even aware of can affect the forecast just a week out, even more so two weeks out. It would be irresponsible to say that we’ll be dealing with a hurricane two weeks from today knowing what I know.
Or would it?
Would anyone on the coast be more likely to prepare if they had two-weeks warning? Or are they savvy enough about the models to think they should ignore it for at least another twelve days?
My guess is that they would wait to see how it progresses because they live in an hurricane prone area and are used to forecast tracks changing. On the other hand, the not-as-savvy people inland who are considering a vacation at the beach that week may panic and start worrying about whether they should have bought trip insurance.
Until the computer models are more consistent with long-range forecast accuracy, anything more than a passing mention of a potential tropical system somewhere on the southeastern coast of the United States is imprudent. Still, I’m going to do my little screenshot experiment to see how this plays out. I’ll let you know.