All the computer models are pointing in different directions. The implication is there will be a big snowstorm. I just can’t tell you where.
Up in Buffalo, safely removed from this particular tumult, Don Paul threw out a forecast suggestion for us in the Northeast: “Accumulations may range from 3″ to 3′.”
He suggested we mumble that line and move on.
The models are seeded with real world data to get them initialized, but my usual favorite (GFS) has initialized with a poor understanding of a strong low in the Gulf of Mexico. Other models are forecasting snowfall in the multi-foot range. Are they any better initialized?
Our data is limited and often flawed. There are many places for which no data is received. I don’t quite understand how these models attempt to compensate for all the things we don’t measure. It has to weaken the model output.
As a meteorologist the uncertainty bothers me no end.
We’ve had a little discussion going on over at Facebook. Some of the talk is technical. This is generally a time when meteorologists talk it out. The scenarios are so complex it’s possible to miss something so talking is valuable.
In the end all our individual forecasts will vary a little. Most of the time most of us will be mostly right… and we’ll relax knowing we’ve dodged the bullet for now. There is no upside to being wrong.
After twenty five plus years these winter forecasts have gotten no easier to make. I’m sweating it out with you.