Weather Accuracy

[midd]

I defer to the weather experts on the board, but what is the outside range of accuracy in a forecast? 10 days? a week? 5 days?

Just trying to get a rough idea what conditions we'll be looking at next weekend, and I'm not sure when to start believing the forecasts.

 

[Greg]

I defer to the weather experts on the board, but what is the outside range of accuracy in a forecast? 10 days? a week? 5 days?

Just trying to get a rough idea what conditions we'll be looking at next weekend, and I'm not sure when to start believing the forecasts.

3 days out of a pretty safe bet, unless we're talking about a storm where the eventual track can change the outcome. I look out up to 5-7 days. Beyond that, I don't base any plans on an extended forecast and will only look at ones that are positive. The bottom line is this is the weekend to ski. There is no telling what it will be like on T-Day weekend.

 

[Justin10]

1-3 day forecasts are generally pretty reliable, I wouldnt expect any major changes within that period. It gets more tricky the farther out you go. Forecasts 4-7 days out can point you in the right direction. A lot can change in the better part of a week, sometimes they're spot on, sometimes they're not. Forecasts for over a week are actually pretty bad, you really can't believe them at all. Its nice however when the models show a huge coastal low, but more than likley these long range forecasts will change....a lot.

 

[Marc]

I'm not a weather expert, although I do specialize in fire protection, and the modeling of compartment fires (and free burning fires in general) is remarkably similar in complexity to weather modeling and forecasting.

That being said, the answer is, as Greg said, depends. It all depends on the area, the current conditions and the complexity of approaching systems and conditions. In New England, what you may consider an accurate forecast may only be realistic for a few hours up to I'd bet at least five days.

Similar to modeling fires. Considering a typical residential compartment with ordinary combustibles and a given ignition source and location, conditions can be fairly accurately predicted almost to the point of flashover. Contrast that to predicting the behavior of a rack storage warehouse fire, which, especially when sprinkler activations are involved, is nearly impossible to predict the behavior of after the first sprinkler activation, so much so that we still run full scale tests at around $65,000 a pop to see what happens.