eduggs

Do you really have confidence that we can predict February weather 5+ weeks in advance. Other than February being further away from a snowless period, I see no compelling reason to think January couldn't be better than February.

IMO you are a little loose with your terminology. People get so used to certain phrases that they start taking them for granted. Terms like "nail" and "lock into place" are subjective. The parameters and spatial scale in question as well as your criteria for assessing model accuracy are not clear. Even the term "pattern" is only vaguely defined. It's easy to rationalize having a good handle on something if details and definitions are kept fuzzy. The magnitude and orientation of 500mb height values at the continental scale are modestly predictable out to about 10 days. But the point I've been trying to make is that regional weather forecasting at and beyond this time frame requires model accuracy that exceeds the current average error. Even if longwave trofs and ridges are roughly predictable, local sensible weather is highly dependent on fine-scale features and evolution that is outside the scope of model skill and only modestly correlated to large-scale features. It's hard enough to see a regional cold snap coming 10 days out. To detect a snowstorm at that range is really hard. And while everybody is looking far into the future for the perfect pattern, a decaying lake effect streamer could drop an inch or two almost without warning.

Let's get it solidly inside 7 days before we celebrate. LR ensembles hedge towards climo at the extended ranges. And we've seen hints of this kind of change already this year that did not materialize.

That's fair. My preference is to just not look out past 10 days at all. But since I sometimes can't resist, I just assume that any model ensemble run out in that range is very very low accuracy. You seem to profess more certainty with LR forecasting than I think is warranted. That's really my only subtle disagreement. Maybe it's more enthusiasm than anything.

I can't believe any coach would actually make that claim. The average NBA basketball player is in the 99th+ percentile for human height. But the average very tall person is not great at basketball, and certainly nowhere near good enough to be an NBA player. If we plotted anomaly charts of nba basketball player's heights we would see that height is extremely well correlated to playing in the NBA. This is analogous to bluewave's favorite historical anomaly charts. Unfortunately in both cases, the underlying metrics are not very predictive of the thing we are trying to forecast because of rarity (northeast snowstorms and NBA skill) and poor correlation.

We clearly disagree then. LR ensembles are already averaged values, which significantly distorts magnitude and location of features, though still usefully out to day 10 or so. Additionally averaging across time scales modulates the resulting values to a degree that renders them almost useless IMO. This completely masks synoptic evolution of features, which is critical to regional forecasting. I believe that's a significant cause for so many head fakes and false alarms. Just look at the 5-day averaged anomaly charts for next weekend and compare that to the operational models. A post mortem of the past few forecasting seasons should shine some light on this issue. But based on what you've already written, I don't think we're going to agree on this point. I wonder who or what it would take for you to ease off the LR multi days averaged ensemble charts.

I don't think there is a more misleading model chart than a 5 day averaged 500mb anomaly chart. If I had a dime for every time Brooklynwx posted one of those and said how good it looked, I'd buy twitter.

My first post today was about next weekend i.e., about 180 hrs out. I pointed out that during this time period we briefly lose the western trof which you suggested was something we needed to get snow. My sense is that you are good at diagnosing the big picture. I prefer to look at the details, which I believe are critical for regional and particularly local snowfall. In truth I believe both scales are important. If the big picture is unfavorable, the details don't get you squat.

By the way, the reason why next weekend fails to deliver is not because of lack of cold. It's because the modeled PVA is either too far offshore of fails to fully round the base of the northern stream trof to initiate surface low formation close enough to our region to produce precipitation. The difference between snow and no snow is in those fine-scale synoptic details and their evolution in time. Snapshot anomaly charts and "pattern" recognition just can't capture those details.

I have been following intently, for years. I appreciate you enthusiasm a lot. I think your concept of "pattern" is on shaky ground. I also don't think the current state of LR forecasting allows you or anyone else to identify productive snow periods more than about 10 days in advance. I would respectfully encourage you and everyone else to follow Walt's lead and focus more of specific synoptic feature combinations in the mid-range and less of fleeting fantasy "patterns" out in fantasy land.

I'll say it another way. The physical and psychological attributes that make a good basketball player are complex. If we relied only on simplistic metrics like height to predict basketball prowess, we would not be very successful basketball scouts. The forecasting of complex patterns requires very precise identification of causal factors and large practice-set sample sizes, both of which are currently lacking in LR weather forecasting.

What you posted is not a "pattern." It's a graphical representation of a set of numerical values at the continental-scale. It's a purely static depiction. Any meaningful definition of weather "pattern" should incorporate the wave dynamics associated with evolution and propagation of airmasses. In that way, "pattern" and "cold" should always be interrelated. In fairness, everyone on this forum would be better off if we all stopped using the term "pattern" because it usually just leads to misunderstandings and unfulfilled expectations.

IMO the coupled global atmosphere and ocean systems are way too complex for easily identifiable and repeatable patterns. Personally I think there are general recurrent features, but not as predicable as you suggest. "El Nino" is a numerical range used to represent a particular geophysical variable in a general region. No two El Nino seasons are even close to being the same.

Are you suggesting that "cold" is independent from "pattern?" That doesn't make a lot of sense to me. You might as well just wait until "cold" and "moisture" are collocated at the same time and place before forecasting snow. A so called favorable "pattern" is never sufficient for snow around here.

As you mention, certain key features are correlated to NYC snowfall, particularly if you isolate major snowstorms. So this is a situation where the features are necessary but not sufficient. That's part of the frustration with long range forecasting. Even if we see these features modeled, we usually don't get a snowstorm. The other major issue for the LR is that ensemble modeling is not stable or reliable enough beyond 10 days to make accurate regional forecasts.

Ensembles have a transient relaxation of the western trof next weekend. 500mb anomaly charts look decent around that time too. Unfortunately the actual surface weather is likely to be rain to dry NW flow. That's the weakness of 500mb height anomaly charts. Favorable anomalies are only loosely correlated to local snowfall. Snow threat identification requires a closer examination of the full synoptic evolution, not just 500mb and not just a snapshot in time. 500mb anomaly charts are also better suited for identifying setups that won't deliver as opposed to ones that will.

They also teach that there is near zero skill at the regional-scale beyond 10 days. Yet all I read on these forums is how good the 15 day ensembles look. I really thought after last year there would be some serious rethinking about super long range forecasting and lessons learned. But instead people keep doubling and tripling down on failed approaches. Does anybody bother to go back a few weeks or months to see what they wrote about the 15 day ensemble means back then? How did that turn out? There's so much wishing and hoping that it's hard to believe. Go back and look at what you posted almost every day all of last year. Seriously. Go check. Always a great pattern out just beyond the rainbow.

I know it doesn't matter since it's 6 days out and it will change every run, but the UKIE was still snowing at the end of its 12z run.

Unfortunately the modeled negative 500mb anomalies centered around December 11 are very transient. Seasonable cold is gone in a day or two and there are currently no obvious shortwaves to key on in that timeframe. Hopefully something trackable materializes, but right now there's nothing much encouraging on the horizon out at least 10-15 days.

Yeah not a ton of cold air around this fall. But I want to reiterate my point that 2m temp. anomalies are misleading in terms of describing airmasses. The 2m temp maps that you posted yesterday gave a false impression of warm air when in fact the forecasted airmass was seasonably cold and theoretically supportive of wintry precip. for part of our region during that time period. Cloudy nights usually show up as strongly positive 2m temp. anomalies even with a fairly cold airmass because of the lack of nighttime thermal radiation relative to average.

The 2m temp. anomalies are pretty misleading. They depict cloudiness and time of day more than anything. It's not frigid but there's plenty of cold air during that time period. And if you loop the NA 850mb temperatures, you can see the source region is eastern Canada. The anomalies look like they advect from the Pacific, but anomalies don't advect, heat does... through the movement of air.

15 day 500mb anomaly charts are pure fantasy. Even if they weren't wildly inaccurate and we could take them for future reality, favorable anomalies don't necessarily mean snow. We need successive run, multi-model ensemble snow threats in the 7-10 range, at minimum, to justify excitement.

H5 has really improved on the 0z ICON and CMC. Multi-run trend towards a strong, more consolidated mid-level low instead of a dampening open wave. PVA and cold pool continue to tick further south. Temps below 925mb level will always be a challenge this time of year. But H5 is finally starting to look like a legit interior snowstorm.

The trend towards a higher impact coastal storm on Wed. continues with the 0z CMC. The cold pool aloft on that run now almost passes overhead. A quick glance at those charts in any winter month and you would think this was a snowstorm for NEPA, NNJ, SENY, CT. It's really only the surface to near 925mb that's a problem. Even the City would probably be upper 30s in steady precipitation. Highly trackable event after a long break from winter. Good 0z so far.

The 0z ICON now has a legitimate far interior snowstorm for Wed - Wed night. Mixing to NW suburbs. Also a significant shift towards a storm on the 0z GFS. Most guidance has been ticking deeper, sharper, and further southwest with the approaching shortwave for the past several runs. Even if frozen precipitation stays far to our north, this is a very trackable winter storm. If the 0z ICON or even 12z/18z ECM is close to being correct, this could be a decent snowstorm for elevated spots in the Catskills and Taconics. And there's still time for this to continue evolving. The GFS still has further to come... the UKMET I think as well.