Well, just when everything seemed well...

Everything seemed to be coming into agreement last night, with most models showing SE Michigan getting a nice snowstorm on Sunday. I was one more set of similar forecast model runs from fully jumping on the snowstorm bandwagon, but then the latest run of the GFS model went way south of its previous run. In addition, the GGEM model went a little further south as well. Meanwhile, the ECMWF model is back to it's far northern solution. This storm is turning out to be one of the more interesting/challenging storms I've seen in the few years that I've been seriously following the weather. We're essentially 4 days out from this storm impacting SE Michigan, and there is still very little agreement among the models that I look at. Both solutions are perfectly plausible, yet one or both of them are going to be incorrect, which from a weather lover's standpoint, is fascinating. So, let's take a look at what the models show and what this could mean for SE Michigan...

Note that if you don't want to read the whole thing (because it's a book), the last two paragraphs adequately summarize everything.

The 12z GFS has the low pressure system tracking from Nashville, TN to Raleigh, NC to New York, NY. Although SE Michigan would still get a few inches of snow with this track, the best snowfall accumulations would be to your south. The image below shows the GFS's storm total precipitation, and as you can see, the highest amounts are to the south.



The 12z ECMWF is WAY further north, tracking the low pressure system from Des Moines, IA to Chicago, IL to Detroit, MI. SE Michigan would get a lot of precipitation from this track, but there would be the issue of rain mixing in with the snow, especially south of I-696. Areas that did mix with rain would obviously have their accumulations cut, but areas that avoided that would pick up a lot of snow. If this track happened, figuring out which areas would mix with rain would likely be a nowcasting situation, requiring the use of short-range models and observations once the storm started. The image below shows where the ECMWF has the low pressure system on Sunday Morning.



Finally, the 12z GGEM has the low tracking from around Paducah, KY to around Cincinnati, OH to around Pittsburgh, PA. This track would allow much of SE Michigan to get moderate-heavy snow, but Lake Orion would be cutting it very very close to being too far north to get significant accumulations. The image below shows where the GGEM has the low pressure system on Sunday Morning.



The fundamental difference that I see in all three models was the amount of time that it took each of them to close off the upper-level low. While the ECMWF closes off the upper-level low quickly (by the time it gets into Iowa), the GFS and GGEM keep the energy weak, allowing it to move further south, and don't close it off until it's well east of the Great Lakes states. The end result of this (the reasoning behind it is complicated to explain) is that the reason the ECMWF is much further north than the GGEM and the GFS is because it closes the low off so early relative to the other two models.

The images below show, in order, where the GFS, ECMWF, and GGEM close off the upper level low. Note the large difference in the areas where the low closes off.

The GFS closes it off in Canada.



The ECMWF closes it off all the way back in the Midwest.



The GGEM closes it off in the Mid Atlantic.



While all three solutions are fully believable at this point in time, I've been thinking about this lately, and I'm really starting to favor the stronger/further north solution of the ECMWF. As you know if you've been reading my posts on this storm, this is a fairly strong/dynamic system. With this in mind, I tend to believe that something closer to the ECMWF is likely to happen. In my experience following these storms, I just don't remember seeing these dynamic systems take as long to get their act together as this one has on the GFS and GGEM. It's entirely possible that something like what is shown on the GFS and GGEM could happen, but I just feel that the ECMWF/northern solution is more likely to happen.

To summarize, there is a large spread of solutions among the models I look at. The GFS has SE Michigan receiving little snow. The GGEM has SE Michigan receiving a decent amount of snow, but it would be a very close call (if it went even a little more south, Lake Orion would be largely left out of the best snow accumulations). The ECMWF has SE Michigan receiving a lot of precipitation, but mixing issues would be present, especially south of I-696. The reason why these models are showing such different solutions is that the GFS and GGEM take way longer to close off the upper level low than the ECMWF, resulting in a further south track. Due to reasons mentioned in the previous paragraph, I feel the ECMWF solution is the most likely, and that is the solution I am currently leaning towards. However, there is obviously still a lot of uncertainty.

If it all works out, SE Michigan could definitely see a great snowstorm with a high likelihood of a snow day on Monday. However, there is a higher than average bust potential with this system than there normally is at 3.5-4 days out. Issues such as the storm tracking too far south resulting in SE Michigan being north of the best snow accumulations, and on the other hand, the storm tracking too far north resulting in SE Michigan dealing with mixed precipitation could both greatly diminish the snow day possibilities. We can only hope that the models come to a better agreement soon so I can start discussing the likelihood of a snow day, but at this time I am still unsure as to what exactly will happen, and am in fact less confident than I was in my last post. This is an extremely interesting storm to follow, and I'll likely have my next post up by the time you leave for school tomorrow morning.

Until then, feel free to leave questions/comments in the comments section below.