And we begin... Part Deux

[skierinvermont]

Thats the problem I've had with the snowcover theory, why would this supposed snowcover forcing mechanism just shut off in early november? Do physics change? Both 2009 and 2010 featured well above average snowcover in october yet in november one had the weak -NAO avg (2010) and one had the +NAO avg (2009). Thats why I theorize there is something in the pattern that leads to above average snowcover in the autumn, and that pattern may evolve and/or lead to a -NAO in winter as wavelengths increase given natural flow tendancies. If it were the snowcover, why would october correlate well, yet november not correlate when physics don't change? And why would one month of snowcover deviation affect the blocking for the rest of the winter? That in itself may argue for natural tendancies in the pattern that led to the increased snowcover serving as the culprit for increased blocking rather than a theorized october snowcover correlation with a mechanism in place for 6 weeks only then shutting off somehow. Perhaps it is some connection in the stratosphere.

If it is somehow the snowcover, the only think I can think of would be that the amount of sunlight diminishing in november could mess with something, but what would it be? It just seems iffy to me. And in case I know what you're thinking, the changes in the geomag sun correlating to changes in blocking cannot be placed in the same camp as the snowcover theory because the correlation exists year round with the geomag theory rather than in 6/52 weeks with the snowcover theory.

It'll be interesting to watch thigns unfold.

People have posted papers explaining the mechanism many times in this thread (in response to your requests) and wxmx just provided a nice summary as well. I suggest you read them. I just explained the mechanism and why it is not relevant in November in the quote you responded to. There is a good correlation between October Eurasian snowcover and winter blocking and a clear causative mechanism.

http://www.nws.noaa....ohen_062211.pdf

Briefly:

The snow cover leads to a Siberian high pressure which then promotes a transfer of energy from the troposphere to the stratosphere. This warms the polar vortex, weakening the N-S temperature gradient in the stratosphere, and thus weakening the polar vortex. A weaker polar vortex allows for higher pressures in the arctic and sometimes the polar vortex splits. This promotes a -AO.

Fall 2009 and Fall 2010 both had above normal snow cover in October. This lead to a warming of the stratosphere in late October and November and eventually an extremely negative AO in December, January and February '09-10 and December and January '10-11.

The reason why November snow cover would not matter as much is that Eurasia is already covered by November. Since the mechanism is snow in Eurasia --> Siberian high --> stratospheric warming, the mechanism becomes irrelevant by November when Eurasia is already covered every year.

[WxUSAF]

This is just MO, but I would surmise that post-October, snowcover on our side of the pole becomes increasingly important. Increased snowcover in Canada allows the source airmasses to be colder and suffer less modification as they move south.

[SVT450R]

Latest comparison last year to this year lookin good.

[Ellinwood]

http://www.nws.noaa....ohen_062211.pdf

Briefly:

The snow cover leads to a Siberian high pressure which then promotes a transfer of energy from the troposphere to the stratosphere. This warms the polar vortex, weakening the N-S temperature gradient in the stratosphere, and thus weakening the polar vortex. A weaker polar vortex allows for higher pressures in the arctic and sometimes the polar vortex splits. This promotes a -AO.

Fall 2009 and Fall 2010 both had above normal snow cover in October. This lead to a warming of the stratosphere in late October and November and eventually an extremely negative AO in December, January and February '09-10 and December and January '10-11.

The reason why November snow cover would not matter as much is that Eurasia is already covered by November. Since the mechanism is snow in Eurasia --> Siberian high --> stratospheric warming, the mechanism becomes irrelevant by November when Eurasia is already covered every year.

Excellent post! As someone who isn't fully versed in the Siberian snow cover correlation, I found this to be rather helpful in understanding the mechanisms involved.

[skierinvermont]

Excellent post! As someone who isn't fully versed in the Siberian snow cover correlation, I found this to be rather helpful in understanding the mechanisms involved.

And if you have a second check out the 70mb temp graphs I posted 15 or 20 posts back. There was a very clear rise in stratospheric temps in late October/November in 2009 and 2010 (both had good snow cover) which preluded the extreme blocking both winters. Other factors probably involved, but it's pretty cool to see the stratospheric warming in late fall actually occurred just as one would expect! The polar stratosphere has been much colder this fall thus far.

[Kevin Druff]

The snow cover leads to a Siberian high pressure which then promotes a transfer of energy from the troposphere to the stratosphere. This warms the polar vortex, weakening the N-S temperature gradient in the stratosphere, and thus weakening the polar vortex. A weaker polar vortex allows for higher pressures in the arctic and sometimes the polar vortex splits. This promotes a -AO.

...

The reason why November snow cover would not matter as much is that Eurasia is already covered by November. Since the mechanism is snow in Eurasia --> Siberian high --> stratospheric warming, the mechanism becomes irrelevant by November when Eurasia is already covered every year.

Excellent post! As someone who isn't fully versed in the Siberian snow cover correlation, I found this to be rather helpful in understanding the mechanisms involved.

While quite interested in this subject, I'm not as educated on it as others. However, I'm curious if the quickly dwindling sunlight sunlight also plays a role. For example, in north-central Siberia total daylight is 11.2 hours on October 1, 9.3 hours on October 15, and 7.1 hours on October 31. They lose 4 hours in a month. By comparison we lose 5 hours total between solstice and solstice.

It's possible that, in addition to Skier's point above about anomalies being almost non-existent by November (as everything is, essentially, always covered), there's simply not enough energy reaching the ground for the albedo to matter.

Edit: I calculated times using the fantastic http://www.suncalc.net ... if you're interested in amount of daylight, sun path, etc., it's a great site. Great help when you're siting a garden!.

[BethesdaWX]

People have posted papers explaining the mechanism many times in this thread (in response to your requests) and wxmx just provided a nice summary as well. I suggest you read them. I just explained the mechanism and why it is not relevant in November in the quote you responded to. There is a good correlation between October Eurasian snowcover and winter blocking and a clear causative mechanism.

http://www.nws.noaa....ohen_062211.pdf

Briefly:

The snow cover leads to a Siberian high pressure which then promotes a transfer of energy from the troposphere to the stratosphere. This warms the polar vortex, weakening the N-S temperature gradient in the stratosphere, and thus weakening the polar vortex. A weaker polar vortex allows for higher pressures in the arctic and sometimes the polar vortex splits. This promotes a -AO.

Fall 2009 and Fall 2010 both had above normal snow cover in October. This lead to a warming of the stratosphere in late October and November and eventually an extremely negative AO in December, January and February '09-10 and December and January '10-11.

The reason why November snow cover would not matter as much is that Eurasia is already covered by November. Since the mechanism is snow in Eurasia --> Siberian high --> stratospheric warming, the mechanism becomes irrelevant by November when Eurasia is already covered every year.

Even though what you mention is a proposed mechanism enough to fit the definition of theory, that actually isn't what I asked however I don't want to get into this again. I said there was no specific mechanism as to how one month of snowcover deviation over the pole enhances blocking 10-15 weeks afterwards, and while november apparently doesn't mean anything much. I guess we'll see how it plays out.

And no the point wasn't wasn't intra basic either.

[skierinvermont]

Even though what you mention is a proposed mechanism enough to fit the definition of theory, that actually isn't what I asked however I don't want to get into this again. I said there was no specific mechanism as to how one month of snowcover deviation over the pole enhances blocking 10-15 weeks afterwards, and while november apparently doesn't mean anything much. I guess we'll see how it plays out.

And no the point wasn't wasn't intra basic either.

That's what was just explained, and what the many papers have explained. Oh well.

Also I am curious what intra basic means.

[BethesdaWX]

That's what was just explained, and what the many papers have explained. Oh well.

Also I am curious what intra basic means.

No it wasn't, theorized mechanisms and potential reasonings were not sold in one month of snowcover deviation = influence on blocking for the rest of winter, reflected SW is reflected SW and does not pick up in LW. Point is it's a theory, and there are reasons to doubt it, as there are all in all theories. Big bang theory is one example.

By intra basic I meant my question was not intricate detail layed out simply because nothing is settled in theory. As I hear you say all the time correlation does not = causation. Theorized mechanisms are not necessarily factual mechanisms because there are complexities not fully taken into account as there always are in atmospheric science.

[Neblizzard]

Latest comparison last year to this year lookin good.

Huge differences since last week, good news to say the least.

[skierinvermont]

No it wasn't, theorized mechanisms and potential reasonings were not sold in one month of snowcover deviation = influence on blocking for the rest of winter, reflected SW is reflected SW and does not pick up in LW. Point is it's a theory, and there are reasons to doubt it, as there are all in all theories. Big bang theory is one example.

By intra basic I meant my question was not intricate detail layed out simply because nothing is settled in theory. As I hear you say all the time correlation does not = causation. Theorized mechanisms are not necessarily factual mechanisms because there are complexities not fully taken into account as there always are in atmospheric science.

The mechanism relies specifically upon high latitude snow cover. Specifically in Siberia. Thus strengthening the Siberian high. Thus warming the stratosphere. Given Sibriea is covered by November every year, the mechanism becomes irrelevant by November. The mechanism relies specifically upon Siberian snow cover, not snow cover generally.

Snow in Iowa does not affect the stratosphere. Snow in Siberia does. Get it?

This has been explained many times already.

[Alpha5]

The mechanism relies specifically upon high latitude snow cover. Specifically in Siberia. Thus strengthening the Siberian high. Thus warming the stratosphere. Given Sibriea is covered by November every year, the mechanism becomes irrelevant by November. The mechanism relies specifically upon Siberian snow cover, not snow cover generally.

Snow in Iowa does not affect the stratosphere. Snow in Siberia does. Get it?

This has been explained many times already.

How does the presence of a siberian high warm the stratosphere?

[BethesdaWX]

The mechanism relies specifically upon high latitude snow cover. Specifically in Siberia. Thus strengthening the Siberian high. Thus warming the stratosphere. Given Sibriea is covered by November every year, the mechanism becomes irrelevant by November. The mechanism relies specifically upon Siberian snow cover, not snow cover generally.

Snow in Iowa does not affect the stratosphere. Snow in Siberia does. Get it?

This has been explained many times already.

Not the point at all, you cannot quantitatively demonstrate that October snowcover in the high lattitudes was the culprit for the Siberian high in the first place, you can correlate it but the correlation can be due to a number of factors. There are years where the correlation hasn't worked out in October due to an unfavorable Solar and/or QBO phase, and it so happens that the correlation only seems to work when one of those indices is favoring blocking. Snowcover could be irrelavent, it could have a minor impact, it could be more. But it is a theory and it's my opinion that there are holes within its logic.

[skierinvermont]

Not the point at all, you cannot quantitatively demonstrate that October snowcover in the high lattitudes was the culprit for the Siberian high in the first place, you can correlate it but the correlation can be due to a number of factors. There are years where the correlation hasn't worked out in October due to an unfavorable Solar and/or QBO phase, and it so happens that the correlation only seems to work when one of those indices is favoring blocking. Snowcover could be irrelavent, it could have a minor impact, it could be more. But it is a theory and it's my opinion that there are holes within its logic.

Actually if you read the papers you would know why snowcover promotes the formation of anomalously high pressure. This is meteorology 101.

The correlation and causation between October Eurasian snowfall and winter blocking is very clear and it is widely used forecasting tool in the field of meteorology.

[BethesdaWX]

Actually if you read the papers you would know why snowcover promotes the formation of anomalously high pressure. This is meteorology 101.

The correlation and causation between October Eurasian snowfall and winter blocking is very clear and it is widely used forecasting tool in the field of meteorology.

I did read the papers as I've already said. Increased snowcover supporting increased blocking doesn't necessarily tie the knot to it's causation. I didn't say I believe there is no effect, but blocky winters always seem to feature at least a favorable QBO and/or Sun, and in the 1980s and 1990s the correlation between the October snow anoms and the winter AO does break down.

Many years in the 1980s and 1990s were -QBO, the Sun was unfavorable pretty much all of the time, but if October snowfall isn't driven by the same pattern forcing that drives the blocking, and is rather variable in October, then how can you say that snowcover in October correlates when a -QBO and high October Snowfall in October are followed by a raging +AO?

[skierinvermont]

I did read the papers as I've already said. Increased snowcover supporting increased blocking doesn't necessarily tie the knot to it's causation. I didn't say I believe there is no effect, but blocky winters always seem to feature at least a favorable QBO and/or Sun, and in the 1980s and 1990s the correlation between the October snow anoms and the winter AO does break down.

Many years in the 1980s and 1990s were -QBO, the Sun was unfavorable pretty much all of the time, but if October snowfall isn't driven by the same pattern forcing that drives the blocking, and is rather variable in October, then how can you say that snowcover in October correlates when a -QBO and high October Snowfall in October are followed by a raging +AO?

The papers thoroughly explain both causation and correlation which is in fact more than can be said for some hypothesized solar relationships which as of yet are unexplained correlations with no clear causative mechanism. Unlike the sun, the relation between October Eurasian snow cover has both a good correlation and very clear well explained causation.

Is snow the only factor? No of course not. But the correlation and causation are both there.

[BethesdaWX]

The papers thoroughly explain both causation and correlation which is in fact more than can be said for some hypothesized solar relationships which as of yet are unexplained correlations with no clear causative mechanism.

No, the papers propose a mechanism that likely is legitimate, however that doesn't mean the mechanism has adequate forcing potential to affect the PV relative to the other processes that occur on a regular basis, such as changes in the QBO, changes in SSTA's of the PDO/AMO, and changes in the Sun; Which BTW is not an "unknown" mechanism by any means because there are several possible explanations, it's just our lack of measurement in that department that has resulted in some gray areas. The forcing from the Sun is potentially immense.

Unlike the sun, the relation between October Eurasian snow cover has both a good correlation and very clear well explained causation.

Tell that to the 1980s and 1990s. How many Octobers featured above average northern pole snowcover and how many of them resulted in a -AO? There is not a strong correlation.

The PDO correlation.

What drives the PDO? Likely it is a random 30 year oscillation, but it also does show correlation with geomag.

Is snow the only factor? No of course not. But the correlation and causation are both there.

There is a proposed mechanism and a correlation that works well with many other factors as well, including the Sun and AMO/PDO, with QBO interspersed. However it cannot be proven that the mechanism has the adequate potency to force a signal in the AO contrary to stronger drivers. That is why it is a theory.

[skierinvermont]

Who said it would overwhelm other factors?

There is a well understood causative mechanism which makes physical sense and which weather models based in physics confirm. And there is a good correlation. And good predictive value.

That is as good as any theory gets. I don't know what you are objecting to or trying to prove.

[ORH_wxman]

Above avg Oct snow cover in Siberia works to the winter time AO/NAO about 65% of the time or something like that, so plenty of times it doesn't work. Its physical relation to the Siberia high is noted, but what is less understood is exactly how the stratospheric connection directly ties into the state of the AO/NAO...the theories are there, but they are still just that...theories.

As for the PDO, its really a reflection of ENSO...the PDO does not drive ENSO, it seems to be the other way around. How it relates directly to Oct snow cover in Siberia, Iam not sure, but there is definitely an oscillation in both over the last 50-60 years that seems to be loosely in tandem.

[Mallow]

Above avg Oct snow cover in Siberia works to the winter time AO/NAO about 65% of the time or something like that, so plenty of times it doesn't work. Its physical relation to the Siberia high is noted, but what is less understood is exactly how the stratospheric connection directly ties into the state of the AO/NAO...the theories are there, but they are still just that...theories.

As for the PDO, its really a reflection of ENSO...the PDO does not drive ENSO, it seems to be the other way around. How it relates directly to Oct snow cover in Siberia, Iam not sure, but there is definitely an oscillation in both over the last 50-60 years that seems to be loosely in tandem.

ENSO certainly projects onto the PDO pattern, but the PDO isn't merely a "reflection of ENSO". I would strongly agree that the PDO doesn't drive ENSO, though.

[ORH_wxman]

ENSO certainly projects onto the PDO pattern, but the PDO isn't merely a "reflection of ENSO". I would strongly agree that the PDO doesn't drive ENSO, though.

I should probably say that the PDO reflects a longer term period of ENSO...definitely not short term such as a seasonal or monthly type of relationship. That was poorly worded by me not using the longer term (as in decade) adjective. I guess this stuff is still being studied, but from what I have seen in some recent analysis is that a longer term period of ENSO will help (in a significant manner) support a longer term period of N PAC SSTs that produce the PDO index.

[BethesdaWX]

Above avg Oct snow cover in Siberia works to the winter time AO/NAO about 65% of the time or something like that, so plenty of times it doesn't work. Its physical relation to the Siberia high is noted, but what is less understood is exactly how the stratospheric connection directly ties into the state of the AO/NAO...the theories are there, but they are still just that...theories.

As for the PDO, its really a reflection of ENSO...the PDO does not drive ENSO, it seems to be the other way around. How it relates directly to Oct snow cover in Siberia, Iam not sure, but there is definitely an oscillation in both over the last 50-60 years that seems to be loosely in tandem.

This is where I think we can tie the knots together. It seems that the PDO and the NAO do correlate fairly well, and the geomag Sun seems to connect to both the NAO and the PDO phase except in the 1950s assuming that data is correct, so perhaps they all have a common driver. All the mets in my family theorize that geomag is the principal driver of ENSO, which as you said, seems to be the frontseat driver to the PDO anomaly. It might make sense that the terrestrial effect of geomag would reflect in the PDO anomaly, which correlates well to the NAO state despite the fact that the two may be completely seperate and irrelavent entities.

You have to say this at least somewhat interesting. The 1950s break the correlation briefly and I'm not sure whether or not I believe the nuclear war theory on ENSO.

The PDO and snowcover anomaly also correlate. They all seem to fit in. The only wierd deviation is in the 1950s with the PDO and the AP index. I normally wouldn't think much of it since the PDO can vary without solar influence but the NAO dropped as well which is puzzling to me.

Who said it would overwhelm other factors?

There is a well understood causative mechanism which makes physical sense and which weather models based in physics confirm. And there is a good correlation. And good predictive value.

That is as good as any theory gets. I don't know what you are objecting to or trying to prove

You are confusing me. You said the correlation is strong. It isn't strong. Yes the mechanism makes sense but there is no evidence that it's forcing is adequately potent enough to modulate blocking, and it's correlation could very well be the result of the one driver that everything correlates to, the Sun. Yes that is also a theory, but it happens to differ from the rest knowing that blocking doesn't modulate the Sun, and that it has to be the other way around unless you believe in "relative planetary mass" theories, which states that the sun and the paterns here are driven that way. And I'm sure you don't believe that.

[skierinvermont]

there is no evidence that it's forcing is adequately potent enough to modulate blocking, and it's correlation could very well be the result of the one driver that everything correlates to, the Sun.

If you read the papers posted in this thread, you would know that the effect IS strong enough to effect the siberian high and then the stratosphere and the correlation corroborates this.

and lol @ geomag being the primary driver of ENSO.. I think you should sit down and have a nice long talk with Jim Hughes. I like how you absolutely refuse to believe that Oct Snowcover effects the stratosphere and thus blocking, but accept on far less evidence that geomagnetic activity is the primary driver of ENSO.

[CoastalWx]

Bethesda, I have seen some work done showing a decent correlation between the PDO value in September and the DJF NAO. The signal was positively correlated in December and January and weakly negative in February. I guess the assumption there, is that the La Nina signal takes over in the back half of the winter.

[usedtobe]

Bethesda, I have seen some work done showing a decent correlation between the PDO value in September and the DJF NAO. The signal was positively correlated in December and January and weakly negative in February. I guess the assumption there, is that the La Nina signal takes over in the back half of the winter.

I ran correlation between the sept PDO index versus the 500 heights for Dec and Jan and found little correlation. Maybe a very,very weak one for Dec but none for Jan.

[ORH_wxman]

I ran correlation between the sept PDO index versus the 500 heights for Dec and Jan and found little correlation. Maybe a very,very weak one for Dec but none for Jan.

What about the actual PDO number? A lot of times the heights won't agree with the SSTs...esp if the PDO isn't that strong.

[usedtobe]

The mechanism relies specifically upon high latitude snow cover. Specifically in Siberia. Thus strengthening the Siberian high. Thus warming the stratosphere. Given Sibriea is covered by November every year, the mechanism becomes irrelevant by November. The mechanism relies specifically upon Siberian snow cover, not snow cover generally.

Snow in Iowa does not affect the stratosphere. Snow in Siberia does. Get it?

This has been explained many times already.

Also, the impact is not immediate, I think that is why the snow cover later in the fall (late Nov into December is not as important is earlier in the year. Once the polar night jet weakens, then it's easier to get blocking. Certainly, snow cover is not the only player in the game, the Atlantic SSTA pattern also has an impact. There is also work that suggests the MJO may play a role as well as the solar factor. The NAO is a pretty complicated animal which is why many climo people think it cannot be forecast much more than a week or two in advance. I think some years it can, this year, to me the signals are more mixed. Judah Cohen will come out with his forecast based on Oct snow cover this week (I think tomorrow).

[usedtobe]

What about the actual PDO number? A lot of times the heights won't agree with the SSTs...esp if the PDO isn't that strong.

I didn't look to correlate the two indexes. It would be pretty easy for someone who has them downloaded into a spreadsheet. That's too much work.

[BethesdaWX]

I've been under the impression that the PDO lags and is influenced by ENSO in the means? ENSO itself doesn't correlate with the NAO state but it does correlate to the AP index very well, and the PDO anomaly also correlates well with the AP index value, which makes me wonder if the PDO anomaly is forced by same driver modulating the NAO? Is the consensus here that the strong correlation between the PDO, AP index, and NAO is just a freak Wtf coincidence? I find that hard to believe, the statistical odds of that being a coincidence seem quite low.

The PDO does seem to lag the AP index but I find thateven more interesting. The AP index versus the NAO would make sense if indeed it is stratospheric connection in the PV structure, but how and why that signal is showing up in the PDO anomaly seems like anyones guess since ENSO and PDO are connected, so the Geomag forcing on ENSO to get than PDO anomaly doesn't seem farfetched.

This is something I'm very interested in, my older brother who got me into the wx tells me it's "clearly" the geomag sun, so I'm biased, but it seems like there really is something there.

If you read the papers posted in this thread, you would know that the effect IS strong enough to effect the siberian high and then the stratosphere and the correlation corroborates this.

How many times have I told you that I have read the papers.....? I mean wow really. Point me to where it is proven that snowcover directly forces blocking in the AO/NAO area. Not that it correlates, not that there is a plausible mechanism...but that the mechanism that is proposed directly forced the AO/NAO blocking.

[Mallow]

I should probably say that the PDO reflects a longer term period of ENSO...definitely not short term such as a seasonal or monthly type of relationship. That was poorly worded by me not using the longer term (as in decade) adjective. I guess this stuff is still being studied, but from what I have seen in some recent analysis is that a longer term period of ENSO will help (in a significant manner) support a longer term period of N PAC SSTs that produce the PDO index.

I believe I read somewhere once that if you do an EOF analysis on Pacific SSTs, ENSO is the first EOF (not surprisingly), and the PDO is the second. There is definitely some orthogonality between the two indices, as the PDO is heavily focused on the North Pac with some signal over the tropics, while ENSO is heavily focused on the tropics with some signal over the North/South Pac.

EDIT: Here it is, figure on second page...

http://www.cgd.ucar.edu/cas/cdeser/Docs/deser.pacsstvariations.jclim95.pdf

Fig 1.a is ENSO, 1.b is PDO.