Carvers Gap

Model whiplash....12z CFSv2 and 18z CFSv2.

Bonus post....sorry. This is why the LR is a mess in terms of, "What will it do?" On this map, we have many conflicting signals. We have a second tropical-ish system south of the Yucatan, a low east of Hawaii(good), a PNA ridge(good), a cold front crashing eastward (good), a -NAO (not good), a block and a half over Alaska and the Bering Sea( I can't even remember the Bering Sea rules), a -AO(not good), a GOA low(normally not great), a western Atlantic Ridge (WAR), and cold over the interior of Canada(good). I could make a pretty good case that if the "movie is played forward" that this is has the potential to empty a lot of cold into NA and eventually the EC by the second half of December. Again, if any model is going to sniff out a cold shot at this time of year(December), my money is on the GFS. By my count the 18z GFS has (4) cool or cold fronts which make it to the west side of the Apps. And that is it for a bit...I have been watching things today, and wanted to sit on a few model cycles before posting. Plus, it has been a busy weekend.

The last in a flurry of posts...I do like the GFS at this time of year. By mid-late December, I like the Euro better for medium and long range accuracy. The 18z GFS isn't a bad pattern. Maybe keep an eye on the Dec 6 timeframe. For the past several runs, the GFS really wants to lift something out of the Gulf. I have definitely noted over the years that the GFS will often spot a cold outbreak more quickly on the deterministic d10-15.

For clarification of my above model verification comments. @Terpeastposted this. For future reference, based on MJO verification over the last 30 days, GEFS is really the one to watch when MJO is medium to high amplitude. And it’s not even competitive.

Really there are two BIG things models are trying to navigate. It is going to take some time for them to "get right." 1. SSW - to what extant (if any) does this impact NA or the surface at all??? 2. The cutoffs in the SW. It sure looks like feedback after system one. I have been fooled by modeled cool forecasts for December before. Conversely, I have seen LR extended modeling show cold, reverse course to warm, and then revert back to the original cold look. So, I am gonna just "chill," enjoy the Florida beat down aftermath, and eat some great Thanksgiving food this week. We certainly could see a warm temps in phases 8-1-2. It has happened before, and could happened again. But I have also seen deterministic modeling flip very cold once they recognize the MJO phases - almost all at once.

FWIW, the 12z GEFS looked good in terms of cold. I noticed that one of the mets in the MA noted that the GEFS is running circles around other modeling right now in the extended range. I don't have that data, and am to lazy to look...so just look in their December thread for reference.

One last note, the Euro Weeklies are very dependent upon the first two weeks of the 0z deterministic. I knew today's Weeklies would be warm. It was simply an extrapolation of the deterministic which is struggling greatly with what is likely Southwest feedback.. It would not surprise me to see them snap back cold if the MJO continues to rotate. I have definitely seen models not recognize the MJO impact until day 10-12, and we are not to that point yet. I would suggest modeling will trend much colder as it becomes clear to modeling that we are entering a phase 7-8 rotation, and I wouldn't be surprised to see it rotate right back to 6-7-8.

The good thing about the Weeklies.....it is still shoulder season and they apparently have missed days 1-10 of December badly - just speaking of the hemispheric long wave pattern. In other words, do we trust a model which likely is going to miss badly? The Weeklies will almost undoubtedly miss embedded cold. Models are swinging back and forth wildly right now. That tells me cold air is incoming into NA. The Mountain West looks likely simply because that has been the pattern since 2015 or so. I really doubt models are done with their wild swings. The GFS trying to "cook up" a Gulf hurricane tells us this could be wild. But really, I think the first week of December has been pretty well documented as being warm. This was a textbook kick the can or Lucy football situation. The second week of December should be a transition w/ the time around Christmas "maybe" being very cold. I also think SSWs upends thing. Within days of it being mentioned, the LR pattern went warm-ish. We have seen that occur several times. That said, I do think we have some chances upcoming. Either way, early December is rarely good snow climatology. I do think there is a bit more pressure to have cold weather in December during Nina years as they are front loaded. If Cosgrove is right, the last half of winter should be good. He certainly has stuck to his ideas and appears to be ready to score a coup of sorts for December. Again, after models missed the Baja low (multiple lows in succession), I have been waiting for the dust to settle so to speak. If the end result of this is getting a cold or seasonal air mass around the holidays...I will gladly take it.

That is insanely good information, John.

Haha. The 18z GFS has freaking hurricane in the Gulf on December 6th. Not sure I have seen that before!

Here is the difference in the control for all of December...Notice what is gone. This also correlates with the MJO looks we have seen. The SER wasn't pumped by multiple Baja lows.

I think that trough is baked in the cake. It should be COLD. Nice animation. How did you get the GIF for WxBell to work?

These are the changes for week 3 of December. Ridge is out of the Aleutians. There is no ridge bellying into the SE. This would allow for northern stream systems and sliders. The control is MUCH colder than this.

Comparing week 2 of December. This is the change which is seen on the Euro Weeklies. This was derived from the 0z run. The 12z derivation would have certainly been even better. Notice the SER is gone. Notice the EPO ridge? It isn't in the Aleutians which teleconnects to AN heights over the SE. That is a 1,000mi change in the Eastern Pac ridge. There are AN heights over Greenland. Those changes IMHO are a direct result of things resolving w/ the erroneous(?) Baja low. The SER doesn't get pumped long term. addendum: Yesterday, the blues and yellows were exactly where u didn't want them. Today, they are generally where you want them

Noticeable shift colder today on most models with the exception of the CMC deterministic. The Euro Weeklies have returned to their cold look and serviceable 500 pattern for December. By the second week of December, they begin to push the cold for the rest of that month - equilibrium restored. The Weeklies, contrary to some comments on other forums, had a major change about two days ago. That first week of December has pretty much been warm on those Weeklies for some time. Nobody is upset about that. The thing that was frustrating to see was the Weeklies completely lost the rest of the month of December, especially after nearly 4-5 weeks of cold being shown. The Euro Weeklies have been anything but steady this past week - welcome to shoulder season. I really think the problem was feed back over the Southwest. Now that is being resolved with each passing run, modeling(deterministic and ensembles) is beginning to correlate again with the MJO and LR ext modeling.

The 12z Euro deterministic also kicks out the energy out of the Southwest. The EPS is a little slower to do so, but I am going to guess it does the same. Both the 12z GEFS and EPS have a formidable cold look now. The SER flexes are short lived. Just file this one away for later used. I would say 75% chance that what we saw yesterday was feedback across all modeling. It caused havoc downstream from that. I would suggest modeling in the d8-14 is still in the process of correcting.

The 12z Euro has correct as well. Storms are kicking out of the Southwest as one would expect. Now that modeling is getting that fixed, we should see some downstream corrections. I would expect for something to track pretty soon.

I think the multiple cutoffs to Baja and/or Southern Cal....was an error. The GFS has a bias to be progressive, but because of that it will work this out more quickly. The 12z gfs and ensemble matched the 6z run. The CMC is still coming around. In fact, the GFS is flirting with snow around Dec7-8. My guess is that cuts along the Ohio River Valley. But things don't hang up...that cold front will come roaring through at the end of the first week of December.

Overnight, it looks like the GFS started to get rid of the feedback over the SW. The 6z run pretty much eliminates it after the 6th. There will be a cutoff out there, but the GFS(and its ensemble) kick out everything after it. Is this because the model has a progressive bias? Maybe. But it would also be the first to fix whatever is going on(if indeed there is a feedback error which seems to be growing more likely).

Hour 276....we have some things to get worked out. The difference in the Euro AIFS and GFS ta 18z over Nashville is 58 degrees. The Real Feel on the AIFS is 3. The GFS is 61. The AIFS did not dig into the SW w/ the SLP FWIW, but it did the second. That still allowed a massive cold front to crash eastward on Dec 3 on the AIFS. The AIFS even dropped 1-3" of snow north of I-40 from Nashville eastward.

Option 1(and there will sure be options 3,4, and 5!) is that the trailer vortex spins out to sea. That is the 18z version. Option 2 is what happened at 12z. The trailing vortex followed and dug out a huge trough which allows a third vortex to drop into. We need that first vortex (even if it goes to the FC...to dig less and kick earlier). I think the first vortex digging into the SW is legit. I am not sure I buy the second and third vortices doing the same - that smacks of feedback. If the second and third vortex simply "don't go to Baja" then the trough likely digs into the EC. Take a minute and looks at the little vortices which modeling is juggling. There are four rotating around the EPO ridge - not counting the Four Corners low. This set-up is an absolute thorn to work out. The downstream effects are huge. It will be interesting to see how this works itself out. OK, enough stream of though posting. It is Friday night, and I am chilling. Will check back later.

Night and day differences when the SW doesn't have multiple lows cutting off...I don't know which is right. But the 18z GFS gives me some hope that the multitude of cutoffs in the SW may well be a feedback error.

The second vortex (which follows the Four Corners low) is the problem. During the 18z run it literally cuts directly through the ridge in the GOA. At 12z, it cuts over the top and dives into Baja which deepens the western trough. You can see the variations on the panels of the ensembles. About half have a trough in the East and half have a ridge - thanks to that little vortex. The EPO ridge is so tall that it is rotating energy back under it which is causing westward flow of Cali. If that vortex doesn't sit over Cali, I do wonder if Santa Anna winds might develop as a result. edit: When the first Four Corners low doesn't dig as much, it doesn't really leave much room for the second to follow. All of this is something to watch on future runs. Until this gets worked out, everything from Weeklies to deterministic runs are going to be impact IMHO. Why? That series of cutoffs takes almost a week to resolve. In other words, there are 5-7 days of cutoffs if not more (on previous runs). If we are looking at just 24-48 hours of a single, cutoff low...that allows for the trough to form in the East. It is a great example of the butterfly effect. The difference between record highs at 12z and the PV heading south at 18z....look nor further than the SW sequence.

Look at the different in the eastern half of the US. The first low(now in the Plains) is shallow and digs less. The second vortex spins off the Cali coast(westward). Not sure how much of that is believable, but this illustrates what happens when perpetual lows don't form in the SW.

Just watching the 18z GFS run, it is already very different over the Southwest. The second shortwave does not dig over the Southwest. That yields an entirely different outcome. That subtle change illustrates how important that sequence is....