BUFFALO, N.Y. — One minute you hear there is going to be a major lake effect snowfall is expected over Thanksgiving, and the next minute the forecast changes 180°.
Why did that happen and why was so much concern relayed by the National Weather Service and other government agencies only to have those warnings pulled back so quickly?
First and foremost the reaction of the National Weather Service was spot on in its decision to be out front of this potential weather event.
The various forecast computer models that were generating for Thanksgiving week on Wednesday and Thursday of last week, were all in agreement.
The three long range computer models, the GFS, the Canadian and the European, were all predicting the same relative forecast. A very strong area of low pressure was to move from the Plains through the Great Lakes and draw down large amounts of cold air to produce a long duration lake effect snow event during a busy travel week.
For all of the models to be showing the same forecast led to enough confidence to get the message out for a potential significant and even hazardous weather situation for the days leading up to and including Thanksgiving.
Computer model agreement does not always happen five or six days before an event, so the fact that all models were on the same page was critical in this decision.
So, what happened?
From a meteorological standpoint, the forecast changed dramatically as two pieces of energy needed to join or phase together to create such a large area of low pressure did not do so in subsequent forecasts.
On Wednesday and Thursday of last week, computer model solutions showed this “phasing” occurring. Then on Friday, those same models predicted that that phasing of energy was no longer going to happen.
So, instead of having a deep area of low pressure to draw down cold air, there was literally no area of low pressure, and the large amount of cold air no longer existed in the forecast.
That forecast for a potential significant lake effect snow event was replaced by partly sunny skies and temperatures in the 40s.
Here is an example of what one computer model showed in its forecast for the day before Thanksgiving last Thursday and the same forecast time period 24 hours later.
Last week when the forecast was starting to look potentially ominous for this Thanksgiving week, some computer model generated predictions for snowfall were posted on social media sites.
Those posts did not necessarily reflect the caution and uncertainty that goes into those predictions.
So, without any explanation of that uncertainty, it appeared that a forecast was changed radically without any perspective. The truth is the forecast that came from reputable sources, including the National Weather Service did suggest caution and uncertainty, a perspective not delivered in many social media posts.
In hindsight, given what meteorological factors were present at the time, the caution that was advised well ahead of the storm was the correct action.
RELATED VIDEO:
