Ending out December with a Potential Pattern Change

[Brian5671]

less than an inch here YTD.   From here south is where the real snowfall bust may lie.  Can't imagine NYC, PHL, DCA getting much in this pattern unless the NAO mysteriously comes to life (and no sign of that)   further northeast into NE maybe ok, as latitude certainly helps, but very difficult to imagine many getting an above normal snowfall year unless there's some sort of 2/8/13 type deal.  If we lose 12/30 and 1/4 to rain or misses, we're done yet another week to 10 days with zip.

[TauntonBlizzard2013]

But should someone really be expecting a 40" month? It's not like they grow on trees.

I think thats the problem. I'm not banking on a 35-40" month to save above average snowfall guesses in this area especially.

It can happen, but it isn't common.

The other problem is, as has been pointed out, this period for at least the first third or half on jan does not even resemble a nino look.

I think its fair to say that by 1/10 if we are still in this position, and the outlook isn't good, its time to reassess. We need something over the next 10-14 I think

[TauntonBlizzard2013]

No way. One month can feature snowfall of greater than 40"....especially in an el Nino season.

How likely is that though? It can't be very common. The pattern also really doesn't look to resemble nino in the near future.

[ice1972]

I apologize to everyone for getting my yard ready for winter last week....

[ORH_wxman]

But should someone really be expecting a 40" month? It's not like they grow on trees.

No. But I'm referring to people basically punting a good winter.

Looking at the numbers, ORH gets 70"+ from January onward 17 times since 1950. That is between 1 in 3 and 1 in 4 years.

So while expecting a 40 inch month isn't very reasonable, a very strong finish to the winter with probably a 30"+ month in there isnt too far fetched. Especially if the pattern look favorable in an already climatologically-favored ENSO state for backloaded winters.

[ice1972]

Once we get to the middle of feb the Ct RV is done.....that's 6-7 weeks......no bueno

[CapturedNature]

Once we get to the middle of feb the Ct RV is done.....that's 6-7 weeks......no bueno

 

I believe that BDL get's about 20% of annual snowfall after that.  I think you're thinking of mid-March because more things need to align after that but as history shows, it still can happen.

[weathafella]

Wow many melted including Mahk. Seriously folks. All some were saying is if we're in this exact state 1/10 we may want to reconsider but that is 2 weeks away. I still think BOS ends the winter more than modesty above normal. As many have pointed out, 2 years ago flipped niño style. The guidance is there. I want to see NAO and maybe 3-4 weeks from now I will.

[512high]

can someone tell me what "torch" means,i have been reading that alot on this thread, again im a newbee just trying to get with the lingo and terms, ty

[weathafella]

can someone tell me what "torch" means,i have been reading that alot on this thread, again im a newbee just trying to get with the lingo and terms, ty

Much above normal temperatures.

[SR Airglow]

can someone tell me what "torch" means,i have been reading that alot on this thread, again im a newbee just trying to get with the lingo and terms, ty

No official definition of it, but it's basically a sustained period of well above normal temps. My "definition" for a torch is 3 days of >=+10 but everyone's got their own.

[ice1972]

I believe that BDL get's about 20% of annual snowfall after that. I think you're thinking of mid-March because more things need to align after that but as history shows, it still can happen.

Yeah I don't know the numbers but seems the middle of feb here in Hartford is this barrier where everything after is elevation dependent and in west Hartford I don't have that.....certainly big events happen in feb like 2006 and 2012 but they were before or right at the middle.....sucks so bad to lose December.....and the first week of jan.....damn that means only 4-5 weeks to recover....****

[CoastalWx]

Yeah I don't know the numbers but seems the middle of feb here in Hartford is this barrier where everything after is elevation dependent and in west Hartford I don't have that.....certainly big events happen in feb like 2006 and 2012 but they were before or right at the middle.....sucks so bad to lose December.....and the first week of jan.....damn that means only 4-5 weeks to recover....****

 

LOL, you'll be fine right through mid March in W HFD. The events that start to suck after early March are the 2-4SM -SN events that amount to fiddle sticks anyways.

[alex]

I feel like I'm starting to crave a couple sunny days more than a snowstorm, haha. Greybird after greybird after greybird.

It's a gorgeous day here in the MWV. Family wanted to do some shopping in North Conway and it feels like April. Never thought I'd say that but it's actually enjoyable. I'll hit the slopes this afternoon

[ice1972]

LOL, you'll be fine right through mid March in W HFD. The events that start to suck after early March are the 2-4SM -SN events that amount to fiddle sticks anyways.

I will take that as encouragement and keep the faith.....I hate Nickle and dime events....I need a couple bad boys and the clock is ticking.....

[Damage In Tolland]

can someone tell me what "torch" means,i have been reading that alot on this thread, again im a newbee just trying to get with the lingo and terms, ty

Yesterday thru tomorrow

[CoastalWx]

I will take that as encouragement and keep the faith.....I hate Nickle and dime events....I need a couple bad boys and the clock is ticking.....

 

The clock starts ticking after Presidents Day relatively speaking and even then there is still plenty of time.

[rogue]

There are many problems with this. 

 

The first is that there are so many variables.  The atmosphere is continuous and dynamic across the globe, so certainly the state of the atmosphere in one location will affect the flow someplace else.  There are very general recurrent "patterns."  But the specific state of these patterns and their evolution can take on an infinite degree of variation.

 

We try to categorize indices with algorithms and weighted averages to define them and then ascribe characteristics to these numbers.  But the thing we are categorizing is much more complicated than our categorization system.  The height field in Alaska or the western Pacific might yield favorable conditions for snow in one instance, but that doesn't mean it always will.  We try to invent "characteristic patterns" that we think we've seen before and then expect a repeatable outcome.  Even with a favorable NAO, PNA, AO, and recurving Typhoon, we just might not get favorable weather.  It is even more challenging in mid-latitudes with such transitional and dynamic atmospheric flow.  Those indices are averaged and generalized and useful only on a very large scale.  There is not a strong enough correlation between the regional or local scale that we are interested in for weather and the large or global scale that we are looking at for "patterns."  There is definitely a link, but it's not as sure of a thing as a lot of people think.

 

Another problem is the relatively small sample size in historical correlations.  All the time I see people filter data by ENSO or MJO state and also by month or some other variable and then come up with a really pretty picture that looks convincing.   But the mathematician in me knows that if you only have 2 or 3 or even 5 instances of those variables, your result is not statistically significant.

 

The last problem concerns causality.  I often see people talk about the forecast state for various climate indices and teleconnectors.  They then infer based on the modeled future state of the global atmosphere what the weather "pattern" over the US is likely to be, and then further speculate what that might mean for the weather in the northeast US.  This exercise employs a very tenuous logical framework.  The first problem is that climate does not cause weather.  You can associate certain states of one to specific conditions, but climate describes rather than causes weather.  When the weather (i.e. the atmosphere) changes, the climate indices will also change.  It is the weather that drives our measurement of climate.  So it seems a bit backwards to be looking at forecasts for the climate to discern the future weather.  And the forecast indices are as likely to be wrong as a 10 day GFS prog.  So to make a reliable prediction about regional weather and the likelihood for snow you have to make one assumption on top of another, none of which is well supported.

 

There are clearly characteristic atmospheric patterns and they recur and correlate in somewhat repeatable forms on a global scale.  But we are still not good enough at predicting their formation and evolution, and especially the regional and local impact. 

 

A lot of the "pattern" talk on here I feel is just leading a herd of "weenies" off a cliff.

This is an outstanding post and should be annual must reading before the winter outlooks are released. Not that it will make a difference because the same folks who put out seasonal outlooks, both pro and hobbyist, will continue to do so and many people will read and hang on every word, regardless of how inaccurate and woefully poor the history of results. Amazing how many outlooks are so beautifully written, complete with lots of pretty graphs and maps, but at the end of the day they amount to nothing valuable.

[CoastalWx]

This is an outstanding post and should be annual must reading before the winter outlooks are released. Not that it will make a difference because the same folks who put out seasonal outlooks, both pro and hobbyist, will continue to do so and many people will read and hang on every word, regardless of how inaccurate and woefully poor the history of results. Amazing how many outlooks are so beautifully written, complete with lots of pretty graphs and maps, but at the end of the day they amount to nothing valuable.

 

It depends on the index. Many of our indices have had a huge and statistically significant sample size. Some have not. I think the pros know what they are doing for the most part, but as always..no two similar index values are alike.

[Baroclinic Zone]

Bring on the sun and 50s.

delightful day today

[ORH_wxman]

Yeah the monstrous EPO ridge shoving big cold into the CONUS is not a small sample size or anything close to heuristic analysis. We know it's favorable.

OTOH something like an MJO composite phase 6 in the month of January in El ninos is likely to have sample size problems.

[Ginx snewx]

Eduggs you are applying meso standards to a hemispheric outlook. Correlations have limits and outliers.However understanding the probabilities increases your confidence. You being a mathematician should further your ability to understand probabilistic forecasts. The weenie comments are unnecessary. If you indeed read the pretty maps and correlations you would understand the limitations. However as many have pointed out some correlations have greater weight, EPO example. Does NOT mean the highest variable, snow in EDuggs back yard, will occur, although last look you are above normal. Point being it's not always chaos but yes it is an extremely difficult science.

[Ginx snewx]

This is an outstanding post and should be annual must reading before the winter outlooks are released. Not that it will make a difference because the same folks who put out seasonal outlooks, both pro and hobbyist, will continue to do so and many people will read and hang on every word, regardless of how inaccurate and woefully poor the history of results. Amazing how many outlooks are so beautifully written, complete with lots of pretty graphs and maps, but at the end of the day they amount to nothing valuable.

totally disagree,sorry your snowfall is below normal, so far

[eyewall]

We are back to bare ground in the CPV and it looks dry and cold to start. Hopefully we can get something going eventually.

[Typhoon Tip]

 

 

"...The last problem concerns causality.  I often see people talk about the forecast state for various climate indices and teleconnectors.  They then infer based on the modeled future state of the global atmosphere what the weather "pattern" over the US is likely to be, and then further speculate what that might mean for the weather in the northeast US.  This exercise employs a very tenuous logical framework.  The first problem is that climate does not cause weather.  You can associate certain states of one to specific conditions, but climate describes rather than causes weather.  When the weather (i.e. the atmosphere) changes, the climate indices will also change.  It is the weather that drives our measurement of climate.  So it seems a bit backwards to be looking at forecasts for the climate to discern the future weather.  And the forecast indices are as likely to be wrong as a 10 day GFS prog.  So to make a reliable prediction about regional weather and the likelihood for snow you have to make one assumption on top of another, none of which is well supported.

 

There are clearly characteristic atmospheric patterns and they recur and correlate in somewhat repeatable forms on a global scale.  But we are still not good enough at predicting their formation and evolution, and especially the regional and local impact. 

 

A lot of the "pattern" talk on here I feel is just leading a herd of "weenies" off a cliff.

 

I have a couple of problems with the "logic framework" from which you derive your conclusions/intimations (bold above).. .

 

Firstly, the use of Climate and Teleconnectors are mutually exclusive applications in operational Meteorology.  What is meant by that is that Climate and Teleconnectors cannot be bundled into the same meaning, because they are independent in their definition.

 

Climate ... as you insinuated correctly, is the sum of weather events divided by the number of events in the sum to produce a mean; and of course, which you also insinuated correctly, the less number of events in a solution set implies a lower chi test, and so a given mean could be a result of mere noise.  However, given a sufficiently large number of events ... while extended periods of fractal born "false patterns" cannot be entirely precluded, there is lesser possibility that the given pattern is the result of repeating accidents.  In short, 1,000 years of climate is a great deal more indicative than a 30 year climate (which could be argued against if there are loud geological events that should impose ... interpretive spacial-temporal boundaries to in the data used..)

 

However, Teleconnectors have no such mathematical/statistical evolution.  They are essentially a great representation of the atmosphere's conservation of mass (or space, depending upon which agency one uses; more on that later).  If the heights in the NAO domain are in positive anomaly, there absolutely has to be counter-balancing region or regions that when in summation the totals equal/off-set the anomaly of the NAO domain space.  That's not arguable. The overall mass of the atmosphere does not change. There are regions of higher and lower pressure/density, but these regions have flux at their boundaries, which we call storms. 

 

The CDC agency employs lower troposphere flux anomalies (mass) to determine the characteristics of the domain spaces.  The CPC employs mid level geopotential height anomalies (space) in theirs. Both are calculated by empirical orthogonal functions, for those that care to research further.  The short version is diagonal observations across a field of data are taken, and then a polynomial expression is derived, which is then solved quadratic to determine the roots of the expression.  ...hence negative and positive values.  Then later, ... there is a comparison to climate, perhaps, but that process of derivation is different and has a different meaning therein.  

 

So in conclusion ... if a Met or knowledgeable user is inferring pattern via teleconnectors, and said user knows their stuff, their discussion points are going to be cogent, or at least .. more cogent than the user that attempts to infer pattern via Climate alone.  

 

****

 

I have not been on blog in a couple of days ... so I cannot speak to everyone's contribution, but I wasn't aware there was any leading... The expectations (particularly by my self) and from what I read from other Mets ...we've all been on the same page for days really; this pattern may not produce the kind of weather the average user of SNE's forum really want.  The pattern changes alright, but to one that may not deliver snow much to the chagrin of most.  

 

In fact, going over things with a bit more intimacy this morning, I am forced to conclude that for the next 7 to 10 days, nothing has really changed comparing to the ideas discussed last week, and/or since.  I took two days away to be with friends and fam for the Holiday, I come back ... and more than figuratively I almost don't/didn't have to even look -- it's that bad.

 

The ridge in the west is till too biased west.  As I said to others three or so days ago, it is nearly impossible given R-wave mathematics to have a ridge at 130W longitude, and have a counterbalancing trough be smooth and unperturbed ... extending across the girth of North America at middle latitudes.  The reason there is this "tucked" trough repeatedly showing up over the southern Rockies really is not so much guided by endemic model biases, but is a result of physics this time.  The other smoking gun is that all models have it... so -

 

But, for larger organized precipitation events (regardless of ptype), they are less likely.  The flow is in a "deconstructive interference" pattern down wind of the Rockies.  The flow becomes confluent over the OV, and depending on the axis of that mean confluence ... dictates where comparatively weaker systems will translate.  

 

The current Euro chooses a west track for an "iffy" D9 system, whereas the GFS is right over head, and the GGEM is S of SNE and the upper MA.  Meanwhile ... it is entirely possible that there may not even be an event there, because the models could correct toward damping (which is what is happening to the 30th deal) as said day(s) get closer in time.  The 30th could still happen...but again... these are all "limiting factors,"  not preventers. 

[ORH_wxman]

The problem is a lot of weenies don't understand probabilistic forecasts. They read them like they would a deterministic forecast.

I've always sort of cringed at long range pattern talk for this very reason. The uncertainty is very high. Even if you know a pattern change is coming, all people here really care about is if it snows in their backyard. That is just one more layer of variance on top of an already uncertain science.

That said, I do think there is value in it. We clearly saw the change to a colder regime coming a while ago. It's going to happen. We knew that because of the ridging shooting north into Alaska and even well up into the Arctic Ocean. It wasn't because of statistically insignificant data either.

We also said a bunch of times that it doesn't mean you are going to get 18" of snow in the first 10 days of the new regime. You might get skunked. It sucks when that happens but it is perfectly within the envelope of reasonable solutions for those ten days. If it wasn't, then they would average 100" of snow per year on the coastal plain.

The good news is:

1. We haven't gotten skunked yet. We could still easily get a couple events.

2. Even if we do get skunked through 1/10, there doesn't look to be a shift away from the -EPO. So we will likely have some more shots.

[Ginx snewx]

The problem is a lot of weenies don't understand probabilistic forecasts. They read them like they would a deterministic forecast.

I've always sort of cringed at long range pattern talk for this very reason. The uncertainty is very high. Even if you know a pattern change is coming, all people here really care about is if it snows in their backyard. That is just one more layer of variance on top of an already uncertain science.

That said, I do think there is value in it. We clearly saw the change to a colder regime coming a while ago. It's going to happen. We knew that because of the ridging shooting north into Alaska and even well up into the Arctic Ocean. It wasn't because of statistically insignificant data either.

We also said a bunch of times that it doesn't mean you are going to get 18" of snow in the first 10 days of the new regime. You might get skunked. It sucks when that happens but it is perfectly within the envelope of reasonable solutions for those ten days. If it wasn't, then they would average 100" of snow per year on the coastal plain.

The good news is:

1. We haven't gotten skunked yet. We could still easily get a couple events.

2. Even if we do get skunked through 1/10, there doesn't look to be a shift away from the -EPO. So we will likely have some more shots.

perfectly stated.

[CoastalWx]

It's the same weenie posts over and over know matter how many times we say it, regarding snow. 

[40/70 Benchmark]

Except it's not a Nino pattern in the least thru mid Jan

Which is good.Save it for Feb/March.....Nino climo favors that, anyway.

[40/70 Benchmark]

But should someone really be expecting a 40" month? It's not like they grow on trees.

Which is why I suggested reevaluating top end winter notions should we make it past 1/10