NittanyWx

Believe it or not, I'm capable of being able to see the globe. Because I forecast for...the globe. I forecast for 4 continents. Germany is not in a monthlong cold spell shattering records. Berlin is +10 anomaly today lol. Did you know Brazil is locked in it's hottest and also many areas driest Novy-Dec on record? I can cherrypick data too. The good news is we have a tool to determine how EVERYTHING is performing at a given time. Joe's a book salesman now. All his energy clients left him because he kept forecasting too cold. He went from widely followed in 2010 to little more than noise.

Someone's been following Joe Bastardi on twitter. You gonna blame underwater volcanoes next?

Honest question...did you read what I wrote? I laid out a multifactor reason for why this is warm. I've repeated it about 5 times over the course of the month and never once mentioned 'oversimplified El Nino', so not sure why I'm being quoted with that lexicon. Yes, background warming plays an issue too, but how much of that +10 you want to ascribe to that is anyone's guess. +1F? +2.5F? I honestly don't know. What I have said, and will continue to say, is that the EC weeklies gave you a very, very warm signal for end Dec, have been consistent with it since the first week of the month, and a lot of folks dismissed it for reasons relating to 'well MJO 8 is a colder phase and that'll solve it'. The Hovvies show you that VP200 progression alone is not enough to solve a problem. Even if you believe that the weeklies were underdamped (welcome to ensemble forecasting), you still had a massive, massive warm signal staring you in the face with a high degree of conviction. So even if it really did miss on amplifying p7, it STILL verified very close to final outcome. What does that tell you? It tells you that there were a confluence of factors...about 5 of them...as I'm now saying the 4th time...that led you this way. Tldr; the data was there to go very warm. People thought they could outperform the model at T2m and got smoked on it. We can call it what it is.

For the record, these were the temp anomalies shown for this Christmas week during the first week of December: and current (yes, I know shorter time horizon): If you had copy/pasted EC Weeklies as a forecaster, you feel pretty damn good about how this turned out. Rare that you get modeling in that strong of a conviction aligned with the seasonal work.

This is why these tools aren't meant to be viewed in isolation, there are more tools at your disposal than that. I did use a stratospheric signal, however, in conjunction with a view that PV displacement into Siberia would cause a significant source region issue for the month across the continent. When you overlaid that with December Nino analogs and an ongoing tropical forcing signal which argued dominant forcing in a bad place you had about 5 different factors pointing very warm into the continental US. I see what's put out on weather twitter, I see what gets put out by vendors...I see a lot of folks blaming the MJO and I don't think that's right in this case. I think it was forecasters underweighting other signals in lieu of overweighting MJO phase progression. For the life of me I don't understand sometimes the obsession with the tropical space as the answer to all. In this case, it wasn't. You had VP progression, you had some h5 response and....you still didn't have a source region. The reason you didn't have a source region is a combination of about 3 or 4 different things happening at the same time.

I boil the late Dec cold misses from a lot of folks down to an error in analyzing source region, not an error in model diagnosis of MJO. Yes the MJO 'signal' as it were lost some steam as it rolled forward, which may have tripped people up. But those weeklies, even with that h5 look were still very warm and will verify pretty well considering lead time.

I'm very aware of how amplified the signal was...which is why I don't use the RMM's operationally. I think they're perhaps the worst way to illustrate tropical forcing to a forecaster. The reason VP was amplified is due to the strongly anomlous ENSO 4' SSTs yes, but there was also a traversing Kelvin Wave in the region which is and was clearly identifiable. Also had a couple of SHEM cyclones spawn out of that pattern. My point is that during the month forcing did shift east, h5 pattern had the appearance of more favorable for cold, but source region was still not conducive for colder conditions. I understand a lot of folks want to boil weather down to one single element causing a forecast miss. The MJO component wasn't as badly forecasted as you think. And the PV displacement had an absolutely crucial role to play in this moment, particularly as VP200 signature improved.

I don't think the MJO 'slowed' at all. I think you had a Kelvin wave continue to traverse as expected, but the signal was overamplified by modeling. It was less coherent as you rolled forward. I also think the biggest issue was PV displacement on the other side of the pole. Can't blame that entirely on tropical forcing. It was a source region issue. I asked this same question when the weeklies were showing this +PNA for the last week of Dec...what air are you advecting?

Ongoing source region issue, and will continue to be an issue in early Jan.

I think the best way for this to change that is PV consolidation and then some sort of Aleutian ridge to displace it towards NA. Need to have the cold air shunted towards NA after this coming PNA spike to fix the snowcover issue.

The Indian Ocean Dipole. Coupled atmospheric/ocean temperature pattern, measured by the difference between ocean temps off the coast of NE Africa and Sumatra. Indian Ocean climate influences (bom.gov.au)

He's making a fair point about the seasonal IOD collapse and subsequent weakening of the standing wave signal in the IO. Not sure that's necessarily the way out on it's own like he's indicating here, but the path analysis to source region shift makes sense conceptually.

Sure, in certain instances it doesn't work out with a colder Canadian airmass nearby to tap into for a plethora of reasons. We have certainly seen cases where promising Canadian airmasses and cold air supply are accompanied by jet suppression. However, historically, patterns with a normal to cooler than normal Canadian airmass providing available air to tap into are significantly better performing for the local area than alternative when it comes to snowfall. At least I've certainly found that to be the case in my career. You're talking about statistical odds. As I discussed yesterday, no one is saying it cannot snow ever in this pattern, I'm certainly not and I do like the 500mb pattern from a 1000 foot view. But it's not unreasonable to say it is statistically less likely than the alternative.

Missing Canadian HP from your equation.

Persistence forecasting for persistence sake is normally not a smart way to go about things. However, in certain patterns in certain years you can meteorologically have a good reason for going with it. That said, it depends on the meteorological variables on the board at a given time. I think there's been several years where if you took the persistence view on an extended/monthly basis it worked significantly better than trying to force a pattern change for X reason. I personally need strong evidence to copy in a persistence view, but I've certainly had bimonthly periods where I felt the pattern was just stuck, I had no real reason to believe it would change in any meaningful way for any appreciable length of time and went with it on a longer range basis. Yeah, there's some inherent variability in there for small windows but the big picture monthly degree day view made it somewhat negligible. For the record, I'm not taking the persistence view today.

I also agree.

I don't disagree here, but persistence is a view and sometimes it rewards a forecaster who has a healthy dose of skepticism. It works until it doesnt.

The timing is literally half of the forecast verification, but to some it is a fungible concept

There's a large temperature spread between this December anomaly and the number's you're talking about. However, historically we do better when there's a source airmass in W Canada. I'm pretty clear in that I think the PNA is 'favorable' for a stormier solution, source is not favorable for a decent cold air supply and that leaves us with chances for snow, but less room for error.

I agree with this, but I also think people need to stop being so sensitive when it comes to pushback on that snow idea. If someone who's done this for a long time comes out and says 'I think that's wrong and a bad idea', it's not shutting down debate. It's literally the debate process. If you've got a view, take it. If someone pushes back on your view that's not shutting down debate. I'm sorry, but you're going to get challenged if you're going to make a forecast. Need to accept that. Especially if you continually have the same view and it blows up in your face. What I don't accept however, is selective verification. Far too many people try to convince people they were right after the fact or come with 'well I was wrong on timing and right on idea'...guys timing is as important to the forecast as the directional view.

2015 and 1982...wonder if those years have something in common with this one

I'd rather have a quote tweet reply from pretty much anyone else other than BAMWX

I think on the surface a reversion to +PNA with some sort of trof in the east is justified view for early Jan. There is a lack of significant source air to advect however which has been a common theme. Now if this PNA spike ends up being more GOA focused, perhaps there's an ability to dislodge some colder air. But in stormy situations with marginal airmasses, relying on a lot to go 'right' to thread the needle. That's where my head is at right now. I still see source region issues.

So just to update my thinking here from a few weeks back...we're in our much above normal temp phase post mid-month transient cool shot. Now we're left with the following: still no significant source region in Canada, dire snowpack in the mid-con and while we're seeing a PNA spike on most models...it doesn't have much of a source to tap into. I think we are likely going to see an erosion of the much aboves heading into early January, but I'd be lying if I didn't say I'm not loving what source air there is to work with at this stage and was a little more hopeful to see the EPO region shaken up instead of the bulk of this PV piece retro'ing back to Siberia. The EPS has been struggling, significantly at that, with 2mT cold bias so far this winter. I suspect it's doing so again late 11-15 day. That said, I am expecting to see the much aboves ease off a bit. H5 improvements are likely in early January. Tropical forcing should be more C Pac/E Pac based as well, but lacking available cold air to tap into I don't see much more than a marginally cooler period ushering in the first week of Jan.

There is an inherent disbelief in his post that a clear climatic trend is occurring because we somehow don't know what happened more than 150 years ago. It was a silly statement and wrong.