The Internet hype machine is starting to hum with regards to this, so we might as well talk the potential for a “snowstorm” to impact parts of the South — including East Tennessee — on Thursday.
Yes, there is potential for snow to fall as we get into the latter part of this week. Is it going to amount to anything? Probably not something you’d write home about.
Various models are indicating the potential for snow on Thursday, after a weak system forms along the Gulf Coast late Wednesday. Some models are much more bullish than others, but the general trend of the computer-generated guidance has been to bring precipitation from this system a little further north as time passes. So whereas it once looked as though we wouldn’t see precipitation here on the northern Cumberland Plateau, it now appears that we might. The National Weather Service is warily watching the modeling trend, and has introduced a slight chance (20%) for a rain/snow mix to the forecast on Thursday.
This “event,” such as it is, is still 48 hours away, so there is plenty of time for change. But, for now, it doesn’t look like anything of significance … at least not in our neck of the woods.
That hasn’t stopped the hype machine — it never does — and in the hours and days ahead you’re likely to see maps on social media indicating major accumulations. Keep in mind that these are likely to be computer-generated maps that are based on the interpretation of a single model, without human input. Some weather enthusiasts have been known to refer to these maps as “clown maps” because they’re often comically inaccurate, and that seems like as good a description as any. Earlier today, I watched a YouTube video from an amateur meteorologist who was touting the potential for a major Southern snowstorm across Tennessee and into northern Georgia and the Carolinas. His entire video was based on a single run of the NAM (the North American Mesoscale Forecast System, a numerical model run by NOAA). The problem is that the NAM is just one of several models, it is the only one that is really showing significant accumulations for most areas, and it is historically not especially accurate. But it fit his narrative and drives viewers to his video.
It’s not that I’m trying to rain on anyone’s parade (the pun wasn’t intended, really); as I said, there is 48 hours separating us from the onset of precipitation, and that’s time for change to potentially occur. But, for now, I’m here to tell you that a serious snowstorm — a disrupt-your-day kind of snowstorm — is highly unlikely.
Okay, so let’s go ahead and show the NAM that has everyone buzzing, just for illustration purposes. This is the 12z run today (the model generates every six hours — 0z, 6z, 12z and 18z, with the “z” standing for Zulu time). You can see that it shows healthy accumulations for the entire eastern two-thirds of Tennessee, including a solid four inches of snow along much of the northern Cumberland Plateau and more than that along the western edge of the central plateau and in the mountains. That’s the kind of accumulation that, if it verified, would meet the criteria for this region’s first Winter Storm Warning in a very long time.
Now let’s talk about the caveats. First, that’s a single run of this model. The previous run, from just six hours earlier, isn’t nearly as bullish, especially this far north:
Second, that accumulation map is showing the standard 10:1 snow ratio, without too much regard for things like surface temperatures. Here’s a look at the same run (12z NAM) with the Kuchera method applied, which does take some of those external factors into account:
So, to be fair, it’s still showing fairly healthy accumulations across large swaths of East Tennessee. But the third caveat is that the NAM is a single model — and it just happens to be the most bullish in terms of snow potential on Thursday. It’s interesting that most folks in the weather community are quick to toss out the NAM … except when it fits their desire for snow. Here’s a look at what the GFS model, a global model that is also run by NOAA, shows:
And here’s what the ECMWF, a very accurate global model run by NOAA’s counterpart across the pond, shows:
Even a higher-resolution version of the same NAM model isn’t throwing nearly as much snow as far north as its lower-resolution counterpart:
So, with all that said … does that mean the NAM can’t score a coup and be right? No, of course not. The NAM could very well be right. The precipitation could very well be this far north. Even the GFS is showing almost two-tenths of an inch of precipitation falling along the northern plateau on Thursday.
The problem is temperatures, temperatures, temperatures. Even if it snows, and snows a bunch, snow is going to have a very difficult time accumulating due to the thermal profiles. The potential for Thursday’s system has been showing up for some time, but temperatures have always looked marginal at best, and that continues to be the case.
Temperatures may not get much below freezing Wednesday night, depending on how quickly cloud cover builds. The current forecast from the NWS is for a low of 31 degrees in Oneida. Even the NAM shows temperatures only dropping to just below freezing on Thursday morning. And once the sun rises, temperatures become more of an issue. Cold air is not going to be advecting into the region during the day on Thursday. You don’t get that when you’ve got a southern system that’s slinging precipitation northward, unless you’re dealing with an upper level low that generates its own cold air — and that’s not the case here.
The lower levels of the atmosphere will be relatively dry as the onset of potential precipitation nears on Thursday morning, so evaporational cooling will likely occur. This is known as the wet-bulb affect. Essentially, as precipitation falls into dryer air and evaporates, it has a cooling affect that continues to drop surface temperatures until the atmosphere is fully saturated. However, evaporational cooling can only carry you so far, and eventually the dynamics of warm air advection and the late February sun angle will overcome any cooling that occurs as the atmosphere is being moistened up.
By 1 p.m. Thursday afternoon, when the NAM projects precipitation to be just past its peak here on the northern plateau, the ECMWF model has us at 40 degrees. The GFS has us at 38 degrees. There are those who will quickly say, “But wait! Some of our biggest snowstorms have occurred with temperatures above freezing!” That’s true. If precipitation rates are high enough, snow can certainly accumulate when temperatures are above freezing — and what results is a wet, heavy snow that sticks to everything in sight. And if the NAM were to verify, we would see some fairly robust snowfall rates. However, once you start talking about temperatures in the 37-40 degree range, you might as well stop talking about accumulation potential.
To be fair, the NAM does keep parts of the northern Cumberland Plateau just below freezing for the duration of precipitation on Thursday — and if we’re going to assume that the NAM might be the outlier in terms of how much precipitation we see, we can also assume that it’s going to be right with regard to our temperatures.
However, there’s another fly in the ointment, and that’s the ground temperatures. We’re going to be in the 60s today across much of the region. Ground temperatures are quite warm, even for this time of year. With marginal surface temperatures, and the late February sun angle that will work against any snow that falls during the daytime even with thick cloud cover in place, snowfall rates would have to be quite high to accumulate on anything more than elevated surfaces.
That’s why meteorologists aren’t biting. In a forecast discussion this morning, forecasters at the NWS’s Morristown office noted that the precipitation appears to be trending further north, but also pointed out the temperature issues. As far as accumulation potential? “Some minor snow accumulations are possible mainly above 3,500 feet in the Tennessee mountains,” the NWS said in the forecast discussion. My old buddy Eric Miller may have said it best this morning: If the 12z run of that NAM model were to verify, “maybe some slushy accums on mailboxes and car hoods across the Plateau.”
The bottom line? Don’t take these snow maps (like the first one I posted above) as the gospel when you see them circulating on social media today, as you inevitably will. Don’t believe the hype videos on YouTube. For example, the maker of the hype video I referred to above, that was based exclusively on a single run of the NAM, showed a snowfall accumulation map like the one I posted, based on the model’s depiction of precipitation that will fall during the morning hours on Thursday. Then he showed a surface temperature map from late Thursday night, depicting temperatures around the 28-degree range across much of Tennessee, and talked about that being the “sweet spot” for heavy snow accumulation. The obvious problem is that, by that time, the precipitation would have long since ended for everyone west of the mountains. That’s worse than building a “forecast” around a single run of a single model. It’s cherry-picking data from that single model run to build a narrative. That’s asinine and irresponsible. But it’s also par for the course when the word “snow” gets mentioned anywhere in the South.
Update: The 12z run of the GFS model has generated since I wrote this post Tuesday morning, and it continues the trend of slightly increased snow potential for the northern Cumberland Plateau. However, all of the above caveats still apply. It’s worth keeping an eye on … but, for now, significant snow accumulation still seems extremely unlikely on Thursday.