Active Pattern Continues: Weekend Forecast and a Next-Week-Peek

  Well we had a double-team of cold fronts pass through yesterday and Winter is back. It was never really gone I don't suppose but still, after getting back out to chase a few severe storms in January and experiencing temps that climbed all the way up to about 60F I may not have been completely honest with myself about what time of year it is. Alas, it is still Winter and we have been rewarded with this brief reprieve from mild, comfortable, 50+degree days with temps in the teens. Awesome...

Yesterday's "Snowstorm" (Or Not)

Surface analysis I performed Thursday morning (1 Feb 16z) preceding the onset of precipitation. One of the main things I was interested in was the narrow corridor of moisture (assessed with 700mb Relative Humidity) with a "vort max" (associated with lift) trailing behind it. This moisture was trying to exit the region just ahead of the colder polar-air-mass and strongest ascent. Some overlap occurred resulting in the snow we did see but more widespread moisture that lingered may have aided more snowfall. 

Same analysis chart as the one previously pictured. This analysis was performed Thursday morning (1Feb 16z) to assess the environment before precipitation began. 

What about that snow we saw? If you remember just over a week ago there were whispers about the possibility of major accumulating snowfall yesterday (Thursday 1 Feb). What happened? As I look outside I see about a half-inch of snow on the grass and in eastern KY there are locations measuring just about 1". But the models said 8"! They did in fact, and even more, they were somewhat consistent for a few days. But then it happened. First it was just subtle shifts in location. Then they started trending lower, and lower, and lower. Before you know it we had a whopping 0.5" of snow before the actual system moved out. I'm not going to waste much time on this because there are more interesting things I want to get into but this was a great example of why it's important (to me anyway) to utilize ALL of the tools at your disposal when forecasting. Personally, I was never sold on this being a major snowstorm and I said for days that I was expecting less than an inch of snow with a little more in Eastern Kentucky. Before you think I'm bragging allow me to bring myself down for you real quick: The last system that came through brought 5"-6" of snow to many locations and do you know what I predicted for that event? 1"-3". That wasn't a terrible prediction but I was unable to foresee the storm getting "hung up" and stalling out. The snowfall rates I was forecasting were correct but that detail meant that those rates fell over locations for a longer period of time resulting in more snow than expected. I just want you guys to know that I am not interested in "glossing" things over or anything like that and when I'm wrong I always admit it and figure out why so I can do better. Nature is notorious for bluffing so the more you know about her hand the better chance you have of making the right choice. It's also worth noting that many of the local Meteorologists were on top of yesterday's system and I don't want to sound like I'm trying to bring their skills into question.

"Spaghetti plot" from 25 Jan.
Contours=various model "predictions" for where snow will fall on
a given date (1 Feb).
That's a large spread. That basically said that
snow may fall anywhere in the US from the Rockies to
the Appalachians. Hard to work anything specific into a forecast
based on that. That's why posting one select model
graphic online without context is misleading.

  There were a bunch of folks posting things online as far back as last Wednesday (8 days out!) about major snowfall however. So why? What were they seeing? Well for starters they were seeing aggressive model runs and we have been in a pattern that supports possible systems impacting the region. Actually, the pattern is even supportive of sending one wave after another, and another, and another, much like an assembly line, across the U.S and through the Ohio Valley, each one with potential. Ok, so based on that why did many of us think the models were overdoing it? The short answer is, it was 8 DAYS OUT! I really start to take model guidance regarding large-scale features seriously when it gets to about 96hrs, give or take depending on the scenario. The models were also struggling with a system that was moving through at the time so that really discredits its output. Every time-step of a model run plants the seed for the next time-step and these details amplify exponentially, sometimes into crazy outcomes (check out "chaos theory" as it applies to weather models). The more assumptions you have to make for something to be true, the more likely it is that it's not true. That was the case. IF it's handling the current system correctly, and IF it's handling the initial evolution of the upstream conditions, and IF it's resolving the phasing of upper-level energy correctly, and IF it's handling the moisture trajectories and forcing for ascent correctly, and IF it's handling the precipitation formation processes/physics correctly then we may be able to trust the output. That's a lot of IF's! That doesn't mean you don't look at the models or trust the data just ask yourself the relevant questions and think about the details, particularly how many assumptions have to be made to verify. The trend is your friend when it comes to models. You still have to investigate it further but trends can point you in the right direction. More telling were the actual environmental conditions that were setting up early this week. The biggest thing limiting greater snowfall was the fact that in order to maximize accumulation we would have to see a series of upper-level waves phase flawlessly with an upper-level jet-streak timed perfectly as it entered the base of the parent trough and crossed into the region. Another issue was that we were looking at a deep surface-low with very strong low-level flow, specifically in the warm sector ahead of the cold front/s. Surface gusts were reaching 30kts and this was in effect aiding strong warm air advection in the boundary-layer (Snow can accumulate even when surface temps are above freezing as long as the snowfall rate exceeds the rate of melting). This would have to be overcome by colder air to maintain frozen precip at the surface and the result was a transiton event with rain eventually changing over to snow. This meant less snowfall. The cold air did eventually win out and we even saw some heavy snow for awhile but there can be a big difference in heavy snow falling for a few hours in the afternoon and snow falling all night long.

Unsettled Pattern Brings Rain/Snow/Storm Chances Through Next Week

  A pretty stubborn but active pattern aloft sets the stage going forward. Currently we're looking at a single-stream polar-jet with a longwave trough over the eastern CONUS and a Pacific-ridge off of the coast of California. These features have shown no signs of breaking down anytime soon and that means that as mid/upper-level waves develop upstream they have a fairly favorable track across the central and eastern U.S. 

500mb analysis for 12z (6amCST) Fri 2 Feb. Depicting the Longwave Trough Over the Eastern Part of the Country and the Ridge Out West. These Features are Locked in Place for now and Provide the General Track for Systems to Bring Unsettled Weather Throughout Next Week. 

  You can see a shortwave trough along the West coast of British Columbia and that is one of the pieces of energy of our Sunday system. In my opinion the track of this wave is all but decided. There are some questions related to the temperature profiles and moisture quality come Sunday but I'm pretty confident in how the shortwave itself will evolve over the next 48-60hrs. Interestingly enough the upper-level pattern is progressive right now with a high number of longwaves in the Northern Hemisphere but a pair of broad ridges are helping keep our parent-low relatively stationary. We could very likely see our shortwave flatten out and become somewhat disorganized as it crosses the mountains of British Columbia and then the Rockies but the energy should maintain and re-emerge and reorganize over the central CONUS sometime Saturday PM. It's at this point I believe it will deepen as it enters the base of the parent-trough while a strong jet-streak enters its base as well. That's a lot I know, but if you can anticipate these things and time them correctly then you may be able to correctly predict the evolution of these features and in turn you'll have a much better shot at being able to forecast regardless of whether or not the models are performing well (Notice I have purposely avoided mentioning any model data in this part of the thought process). An associated surface-low will be in the Midwest and should deepen a little as it tracks toward the Great Lakes, dragging a cold front through the Ohio Valley. The precip types late Saturday night is where confidence drops off. Right now it's looking like temperatures will be in the low/mid-30's, except for far western Kentucky southward where warm air advection will be a little stronger ahead of the front. Precipitation will likely move in from the SW Saturday evening and temperatures should be near or just above freezing. Initially it should be rain until later Saturday night when the atmosphere has cooled enough and a transition to snow or a rain/snow mix is possible. "Wet-bulbing" (evaporative cooling) could play a big role in precip type as it could serve to cool the low-levels enough to transition to snow fairly quickly. As warmer air continues to work in throughout the afternoon on Sunday a freezing rain/rain/snow mix becomes possible, particularly East of I-65 in Kentucky and Tennessee. Along and North of the Ohio River I expect primarily snow since colder air should be established. A transition to mainly rain could exist along the backside of the precip-shield as it moves eastward. Temps will climb into the 40's Sunday afternoon at which point a transition back to rain will occur. After the front passes Sunday afternoon I think we cold see another quick shot of snow flurries with the NW flow wrapping around the low-pressure system and very cold air interacting with lingering moisture. 

How Much Snow Are We Talking Here?

  I haven't seen anything that suggests a major event but there could certainly be some impacts, especially considering many folks may be having Super Bowl parties or something. If the freezing rain is able to over-achieve then that would be a bigger issue than currently expected so that's worth following. I don't think 1"-3" of snowfall by Sunday night is unrealistic for areas East of I-65, especially in North-Central Kentucky into Indiana, with up to 1" possible West. Temperatures will drop to the teens Sunday night so slick spots will be likely and should be accounted for. 

  So there you have it, rain-to-rain/snow mix-to-snow-to-rain-to-snow. Fun stuff for sure. I didn't post as many graphics as I usually do so I apologize for that. We've put everything together about as well as we can with what we know at this time though so all that's left to do now is monitor the data and observations and we'll be following in real-time tomorrow night.

And Next Week...

  It definitely looks like active weather continues through next week with more chances for rain, storms, and snow across the Midwest. I even think severe weather is at least on the table for the Southern Plains into the Arkansas/Missouri/Mississippi River Valley but there are still some things to work out with all of that. Luckily there's plenty of time and we'll dive into that tomorrow and see if we can put some of the pieces into place but with one disturbance after another set to eject across the U.S within strong flow aloft and a Polar air mass that hasn't entirely modified or cut off the Gulf source I think we have something to work with. Hey, it's a start!

See ya tomorrow, stay warm.

DK

The Big Thaw and...Thunderstorms?

  Good afternoon! Snow is melting away a little at a time, although the big melt will be this weekend. With the snow keeping me stuck inside for at least another day or so I thought it would be a perfect opportunity for pining over days with surging dew points and triple points, warm sectors that are actually warm, and when the rumbling you hear outside is from thunder and not just the snow plow going back down the road. Days when you spend hours and hours on the road jacked up on caffeine eating lunch from a Love's Truck Stop in search of a moment in the perfect position on the perfect storm.
  Maybe those days aren't too far away anymore? There are some signals that we could see thunderstorms around East Texas/Arkansas as we round out the weekend. You guys know I like to be a patient forecaster so we won't jump to conclusions, and honestly things aren't necessarily jumping out screaming "severe weather outbreak" or anything but we will look at what we know and examine some of the possibilities going forward and, as always, we will set a "checkpoint" to come back and reassess things. It's something at least, right.


  It's only been two months since I was preparing a forecast for chasing severe storms but it feels like many more. Let's see if I can remember how this goes...I believe best practices dictate that you first analyze and diagnose the current state of the atmosphere so here we go.

GOES16 water vapor imagery shows a series of shortwaves/upper-lows across North America. I've numbered them 1-4 from East to West. The disturbance off of the East Coast (#1) over the Atlantic Ocean was responsible for bringing quite a bit of snow to the Eastern CONUS but it is caught up in the flow of the northern and southern stream and will continue to depart. #2 is a shortwave over the Great Lakes and will track within the northern stream and stay North of our area of concern. #3 has been giving the models fits actually but even with all of the model divergence I expect it to stay South of our area. I'm not saying these shortwaves are insignificant because every detail should always be analyzed, and I have analyzed these as well, but the feature we are focused on for the possible thunderstorms Sunday is #4. Currently located off of the Northwest Coast, it is a very well-defined system already. 12-hour height-falls are significant in the Pacific Northwest in response to this low (200m+). Strong flow is associated with this low as well with 100kts+ sampled in the 00z soundings throughout the region.  

Stepping down to the surface real quick we can see the high-pressure that's dominating our local weather for the next few days, as discussed in yesterday's post. Upstream in the NW CONUS you can see a deep surface-low and attendant frontal boundaries, a clear surface reflection of the impressive mid-upper wave we looked at on water vapor. This surface chart actually says a lot about what may happen, even looking forward 3-4 days. The main thing I'm getting at here is the impact of the surface-high over the Southeast right now. This high is weakening but it's stubborn and has already taken steps to limit severe potential in the near-future. Yes, the southerly flow along the western periphery of it is beginning the return-flow cycle for moisture and warm air advection but I'm concerned if that will be enough to support severe weather. With the divergent flow within the center of the sfc-high straddling the Gulf Coast and an Arctic air mass that has been established for what seems like forever, extensive modification of the air residing over the GofM has taken place. Cooler air has made its way well over the usually warm waters which serve as a common source region for Southeast severe weather. I do think we will see sufficient moisture make its way across the ArkLaTex region as the broad high-pressure slowly makes its way Eastward but it is definitely something to consider going forward, especially if the shortwave over the Southern US is able to develop a related surface system (not highly likely but not impossible either).

Surface dew point observations across the US. The influence of Arctic air is evident and the intrusion of this drier air has resulted in modification extending well into the Gulf of Mexico. This will have to be overcome by sustained return of deep moisture before severe storms would be likely.

We'll take a look at some of the model data next because there is, believe it or not, pretty good consistency regarding the large-scale features that set the stage. Now I always say that you don't just jump on board with the models because they're consistent because they can be consistently wrong, but in this case I agree with a lot of the output regarding the evolution of our shortwave & upper-flow going into Sunday and I don't see anything they are tripping up on either so we'll briefly touch on it. Where I get concerned is with the usual differences in timing. The GFS has initiation of storms a little earlier in the day when diurnal heating could help destabilize things a little more, although there could be showers early on anyway. The Euro's latest "master" run is much later with showers throughout the afternoon and thunderstorms firing up after 9pmCST. These are the details we'll have to work out just like we always do in these situations.

GFS Ensemble Mean 500mb Heights 

Above are the recent GEFS ensemble runs for 72hr/day 3. The valid time of these images is about 18-24hrs before expected thunderstorms across the Southern US. I wanted to show the relative consensus with the upper-level wave as it approaches the Rocky Mountains. You can see that the GEFS has pretty similar solutions in its most recent output. You can also see the struggle with the shortwave over the SE CONUS that I mentioned earlier. From this point I suspect we will see our system round the Rockies and emerge over the Central Plains while also assuming a negative-tilt as it amplifies. I expect amplification in response to a strong upper-level jet passing over the low which will lead to upper-level divergence/mass removal. We should also see a surface-low develop in response to this as well as it interacts with the thermal boundary in place from the retreating cold air and returning warm air. The low-level low-pressure will help advect higher theta-e into the region. With a strong southwesterly jet up top and southerly low-level winds, coupled with a strong kicker at 850mb, wind profiles look favorable for both damaging winds and even a tornado or two. That isn't really in question. Storms should be well-ventilated and deep-layer shear would be supportive of organized updraft/downdraft relationships. Forecast soundings are showing sickle-shaped hodographs supportive of rotation in both the low and mid-levels. Although there are the timing differences I think there isn't much doubt in sufficient forcing for ascent arriving at some point, when will be important though.

 I'm just not sold on instability. There will be a moisture/theta-e axis across East Texas up into Arkansas most likely. It currently looks like we could see surface dew points approaching 60F and dew points at 850mb at 8-10C (I like to see 10-12C). I do prefer to see the deepest moisture established over the target area for about 18-24hrs before storms but it is what it is. The destabilization just doesn't look overly impressive right now. The GFS is suggesting 500j/kg of surface-based CAPE with marginal lapse rates (6.5C/km) during the afternoon centered over ArkLaTex. The ECMWF has slightly more instability and pushes 1,000j/kg of surface-based buoyancy across the area but it shows it after dark with later initiation and also has rain showers throughout much of the day. The ECMWF seems like it's overdoing it personally so I'm leaning more toward the GFS for now. This is one of those "subject to change" things though as I would prefer to lock in the timing better before I make a call on that.

Mutli-Model "Spaghetti" Plot of 500mb Heights at 72hrs (Sunday 6amCST)

Multi-Model "Spaghetti" Plot of 500mb Heights at 96hrs (Monday 6amCST)

GFS MOS (Model Output Statistics) of Max Temp and Probability of Precipitation at 6pmCST Sunday. The Surge of Warm Air Northward Can Be Seen Across East Texas into Missouri. 

Based on everything I've seen I don't see any signals for a "widespread" severe weather outbreak Sunday-Monday. I do see a chance for some storms however with some strong-severe embedded. I won't make a call on specific threats yet either but I can't rule out a tornado or two right now so that's worth watching too. I'd like to see how things go over the next 24hrs regarding moisture return, shortwave evolution, and model output and then make an update Saturday morning so that's the plan. At that point I'd expect to have a better grasp on how smaller-scale details look so we could take a shot at a more specific forecast too.

Regardless of how we see this unfold one thing seems likely; everyone will crawl out of their winter hibernation, throw the Snuggie away, and start making their calls. It's not quite Spring yet but it might be that time for a small taste of chasing model trends, SPC outlooks (and the criticism from the know-it-alls of course), and copious amounts of coffee while impatiently living life on a 12z to 00z schedule!