#1264 Postby wxman57 » Tue Dec 12, 2017 1:09 pm
I was just talking with my cold-mongering coworker in the cubicle next to me. We've been working together since the late 1980s, by the way. I was telling him that the 500 mb pattern in recent GFS runs reminds me of the Thanksgiving week Texas ice storm of 1993. I still have those model forecasts printed out in my file cabinet. Back in 1993, the GFS was predicting that Arctic air would plunge down the plains. However, it developed an upper-level low over Nevada/Utah as the front reached central Texas. The main model (may have been the AVN or MRF back then) didn't handle shallow Arctic air very well at all (sound familiar?).
What the model did was to move the Arctic air northward as a warm front as the upper-low dipped southeastward toward New Mexico. The model had the front moving northward across northern Oklahoma a couple days out. Wrong! Arctic air doesn't do that. The front was out in the NW Gulf of Mexico, not northern Oklahoma. As a result, freezing rain and sleet covered much of Texas, and Leon Lett went down in Dallas Cowboys history for his bad play in the Thanksgiving game.
The moral of this story is that whatever the American model has been named over the past 25 years, it still doesn't handle Arctic air very well. To me, this is looking like more of a freezing rain and sleet event rather than a snow event for Texas. However, it's still 12-14 days out, so I don't feel too confident about anything until the Arctic air is actually present in southern Canada and on the move south. that won' happen for another 10 days.
5 likes