2013-14 MA Winter Disco - 'cause it's never too early.

[mitchnick]

I saw on the news tonight that Australia had a whole lot of fires and Sydney was all smoked-in for the past day or so, all because of drought conditions

don't know if anyone knows or recalls, NINOs bring drought to much of Australia, which is why their weather service is pretty ENSO centric

the point being, the NINO-like conditions we're starting to experience with the ridge on the west coast and the trough coming our way this week must be Pacific wide with the Weenie Mitch (OK, sorry to repeat myself!) hoping this has got some staying power through the winter, at least at times, in light of its current extent

[H2O]

DT has been humping the early snow cover hope a ton on FB of late. 

[WxUSAF]

DT has been humping the early snow cover hope a ton on FB of late. 

I (politely) commented on one of his snow cover posts that the SAI will not necessarily be that high given that October started with much above normal snowcover.  We're now much closer to normal for this date.  It's not about the raw AMOUNT of snowcover at the end of October that is important, but the CHANGE in snowcover over the month.  We very well could have above normal snowcover on Halloween with a rather pedestrian SAI for the month.  He of course responded with something like "I know better than you", but with more words, misspellings and capital letters.  

[ORH_wxman]

I (politely) commented on one of his snow cover posts that the SAI will not necessarily be that high given that October started with much above normal snowcover.  We're now much closer to normal for this date.  It's not about the raw AMOUNT of snowcover at the end of October that is important, but the CHANGE in snowcover over the month.  We very well could have above normal snowcover on Halloween with a rather pedestrian SAI for the month.  He of course responded with something like "I know better than you", but with more words, misspellings and capital letters.  

 

 

Interestingly the last year with very high "raw" snow cover in October but a muted SAI was 2002...and that winter produced a nice looking arcitc. I wonder if the Cohen correlation is weighted more heavily toward years that start off below normal (like 2009 and last year) and then increase rapidly to well above normal....but those years that stay above normal all month in a sort of steady-state are still good. The daily snow cover data only goes back to 1997...so before that you have to use weekly data.

[usedtobe]

Interestingly the last year with very high "raw" snow cover in October but a muted SAI was 2002...and that winter produced a nice looking arcitc. I wonder if the Cohen correlation is weighted more heavily toward years that start off below normal (like 2009 and last year) and then increase rapidly to well above normal....but those years that stay above normal all month in a sort of steady-state are still good. The daily snow cover data only goes back to 1997...so before that you have to use weekly data.

 

I think that is a good question.  It will be interesting to watch what the AO does this winter. 

[WxUSAF]

Interestingly the last year with very high "raw" snow cover in October but a muted SAI was 2002...and that winter produced a nice looking arcitc. I wonder if the Cohen correlation is weighted more heavily toward years that start off below normal (like 2009 and last year) and then increase rapidly to well above normal....but those years that stay above normal all month in a sort of steady-state are still good. The daily snow cover data only goes back to 1997...so before that you have to use weekly data.

So is the correlation between the SAI and DJF AO not as high for the 02-03 winter? Well, there's 2 factors at play, right?  The one Cohen is talking about is the disruptive forcing caused by the rapid change in snow cover, right?  But having large amounts of snow cover alone should allow airmasses to become cold, remain cold and get colder with radiative cooling.  

 

All things being equal, I'd always take larger amounts of snowcover than not, but I think that is something different than what Cohen suggested.   

[isohume]

DT has been humping the early snow cover hope a ton on FB of late. 

 

I've completely hidden DT and his friends from any FB postings. Sure I may miss something important, but it's turned out to work out pretty well so far. 

[ORH_wxman]

So is the correlation between the SAI and DJF AO not as high for the 02-03 winter? Well, there's 2 factors at play, right?  The one Cohen is talking about is the disruptive forcing caused by the rapid change in snow cover, right?  But having large amounts of snow cover alone should allow airmasses to become cold, remain cold and get colder with radiative cooling.  

 

All things being equal, I'd always take larger amounts of snowcover than not, but I think that is something different than what Cohen suggested.   

 

 

Well it wasn't terrible, it predicted slightly positive and the AO was modestly negative that winter. The biggest whiff on the SAI was actually 2005-2006 when it predicted a positive AO and instead we had a solid -AO (not that it helped our part of the globe that winter).

 

The SAI only has a sample size of 15 years, so we have to be careful. The correlation is very high, so there's definitely statistical significance even with 15 years. Cohen's original research was based just on total October snow cover, but then found that the rate of increase south of 60N had a better relationship...even when he used the more crude weekly data back to 1970 since daily data only went to 1997.

[Isotherm]

Well it wasn't terrible, it predicted slightly positive and the AO was modestly negative that winter. The biggest whiff on the SAI was actually 2005-2006 when it predicted a positive AO and instead we had a solid -AO (not that it helped our part of the globe that winter).

 

The SAI only has a sample size of 15 years, so we have to be careful. The correlation is very high, so there's definitely statistical significance even with 15 years. Cohen's original research was based just on total October snow cover, but then found that the rate of increase south of 60N had a better relationship...even when he used the more crude weekly data back to 1970 since daily data only went to 1997.

 

 

What is the physical reason why a year with a rapid increase in snow cover would produce better Arctic blocking than a year with steady above normal snow cover? It seems to me that both paths would generally lead to more efficient building of pressures in the high latitudes than a year with low snow cover.

[Bob Chill]

Rutgers weekly values are still @ week 39 for 2013. This is the last week of September. 2013 Eurasian snow cover came in @ 5.88m sq km which is the the 4th highest total since 1967 (rutgers dataset). The only years that came in higher are 77, 72, & 73. 

 

I suppose the statistical significance of this is that starting Oct with a high value goes against the potential for the rate of increase during October. 

 

Just for fun I pulled all the years where week 39 came in above 4m sq km. I compared against week 43 (end of October) and caculated % of increase for each year. I also pulled DJF AO for the years to see if there is anything to glean here. 

 

 

 

One interesting thing of note is the 2 years with the slowest rate of increase had a solid late season -AO. Also, the years with the highest rate of increase seem to favor a -AO December. We'll have to wait and see what Oct #'s look like before trying to fit 2013 into a box. I was just interested in the possibility of what a top 4 September value may imply. 

[usedtobe]

Rutgers weekly values are still @ week 39 for 2013. This is the last week of September. 2013 Eurasian snow cover came in @ 5.88m sq km which is the the 4th highest total since 1967 (rutgers dataset). The only years that came in higher are 77, 72, & 73. 

 

I suppose the statistical significance of this is that starting Oct with a high value goes against the potential for the rate of increase during October. 

 

Just for fun I pulled all the years where week 39 came in above 4m sq km. I compared against week 43 (end of October) and caculated % of increase for each year. I also pulled DJF AO for the years to see if there is anything to glean here. 

 

Snowcover.JPG

 

 

One interesting thing of note is the year with the slowest rate of increase had a solid late season -AO. Also, the years with the highest rate of increase seem to favor a -AO December. We'll have to wait and see what Oct #'s look like before trying to fit 2013 into a box. I was just interested in the possibility of what a top 4 September value may imply. 

Too bad there is no way to look at the snow cover south of 60N.  Cohen speculated that the snow cover at the lower latitudes might be more important than those at high latitudes and that might be one of the reasons the SAI worked better than the October snow cover alone. 

[ORH_wxman]

Too bad there is no way to look at the snow cover south of 60N.  Cohen speculated that the snow cover at the lower latitudes might be more important than those at high latitudes and that might be one of the reasons the SAI worked better than the October snow cover alone. 

 

 

Yes, I think this answers Isotherm's question too. Above average snow cover in late September and early October is often mostly in areas north of 60N. Later in the month, we need it south of 60N to remain above average. Perhaps years where the snow cover is almost down to 60N very early (giving huge positive anomalies) in the season but then has trouble advancing past it all that much to where it is closer to normal by late October are when the overall snow cover metric could be misleading...but that is just speculation.

[Bob Chill]

Too bad there is no way to look at the snow cover south of 60N.  Cohen speculated that the snow cover at the lower latitudes might be more important than those at high latitudes and that might be one of the reasons the SAI worked better than the October snow cover alone. 

 

True. Best I can do is draw an imaginary 60* line on this map and say at the very least we are doing better than last year. But not by much considering the difference in disbursement over the 2 years. 

 

 

I really haven't changed any of my ideas about winter for months. That's why I haven't posted much. Nothing has has changed in my eyes. We're going to have to fight an unfavorable storm track overall but the pac won't be a disaster and completely rule out a 4-6 week period of opportunity like it did last winter. 

 

We may end up relying on difficult to predict tc's like the nao more often than not. Hard to speculate there. If the storm track is going to favor nw of us then the only thing that can really save us is a -nao or some miracle 50-50 that can't be predicted with reliability 4 days in advance. 

 

I'll just stick with my hunch that Dec will be aob in the temp department and each month of the DJF period will have a favorable period for potential. 75% of climo snow seems like a fair wag too. That's all I got. 

[WinterWxLuvr]

Rutgers weekly values are still @ week 39 for 2013. This is the last week of September. 2013 Eurasian snow cover came in @ 5.88m sq km which is the the 4th highest total since 1967 (rutgers dataset). The only years that came in higher are 77, 72, & 73. 

 

I suppose the statistical significance of this is that starting Oct with a high value goes against the potential for the rate of increase during October. 

 

Just for fun I pulled all the years where week 39 came in above 4m sq km. I compared against week 43 (end of October) and caculated % of increase for each year. I also pulled DJF AO for the years to see if there is anything to glean here. 

 

Snowcover.JPG

 

 

One interesting thing of note is the 2 years with the slowest rate of increase had a solid late season -AO. Also, the years with the highest rate of increase seem to favor a -AO December. We'll have to wait and see what Oct #'s look like before trying to fit 2013 into a box. I was just interested in the possibility of what a top 4 September value may imply. 

 

I think you could say that there's a definite tendency toward a neg AO in what you posted Bob.

[Bob Chill]

I missed this update from the 20th. Looks solid N of 60. South of 60 not as much. 

 

[WinterWxLuvr]

True. Best I can do is draw an imaginary 60* line on this map and say at the very least we are doing better than last year. But not by much considering the difference in disbursement over the 2 years. 

 

ea snow.JPG

 

I really haven't changed any of my ideas about winter for months. That's why I haven't posted much. Nothing has has changed in my eyes. We're going to have to fight an unfavorable storm track overall but the pac won't be a disaster and completely rule out a 4-6 week period of opportunity like it did last winter. 

 

We may end up relying on difficult to predict tc's like the nao more often than not. Hard to speculate there. If the storm track is going to favor nw of us then the only thing that can really save us is a -nao or some miracle 50-50 that can't be predicted with reliability 4 days in advance. 

 

I'll just stick with my hunch that Dec will be aob in the temp department and each month of the DJF period will have a favorable period for potential. 75% of climo snow seems like a fair wag too. That's all I got. 

 

I think the temp predictions have at least some skill (by you guys that look at that stuff) but I'd still argue that snowfall is a completely different animal.  One or two well timed disturbances/lows/storms/whatever can make for a completely different winter wrt snow.  

 

2010-2011 missed by an eyelash down here of being a memorable winter.  I can still remember the storms and the dates.  It could have ended much differently.

 

So, while it may look less than perfect from several angles, it's not inconceivable that it could end up as a good winter from an overall perspective.

[WxUSAF]

Cohen's 2011 paper doesn't go into the "why" part of it at all.  I think HM has opined on why it might be important in one of the winter threads somewhere.  But, I think the general connection with snowcover (of any sort) is that snowcover enhances troughs through radiative cooling.  These troughs are vertically propagating baroclinic waves, which can disrupt the high altitude/stratospheric polar vortex and lead to cold air being displaced farther south.  Combined with cold airmasses can stay colder if moving over snow covered terrain.  I think that's the general gist...someone correct me if I'm wrong.  

 

 

Here's the link to the paper, but it's paywalled for most average folks.  http://onlinelibrary.wiley.com/doi/10.1029/2011GL049626/full

[usedtobe]

Cohen's 2011 paper doesn't go into the "why" part of it at all.  I think HM has opined on why it might be important in one of the winter threads somewhere.  But, I think the general connection with snowcover (of any sort) is that snowcover enhances troughs through radiative cooling.  These troughs are vertically propagating baroclinic waves, which can disrupt the high altitude/stratospheric polar vortex and lead to cold air being displaced farther south.  Combined with cold airmasses can stay colder if moving over snow covered terrain.  I think that's the general gist...someone correct me if I'm wrong.  

 

 

Here's the link to the paper, but it's paywalled for most average folks.  http://onlinelibrary.wiley.com/doi/10.1029/2011GL049626/full

He did speculate a little about albedo as being a possible player but didn't really go into the mechanism of how in one of his articles.

 

Have you looked at the thread on the OCTOBER PATTERN INDEX(OPI)?  Because of the language problems I couldn't really get a handle on exactly how they formulated the index.  Hopefully,  they'll publish it in a peer reviewed journal in English.  Until then,I won't give it too much credence.   Ricardo suggested that his index would be forecasting a positive AO this winter. 

[Bob Chill]

I think the temp predictions have at least some skill (by you guys that look at that stuff) but I'd still argue that snowfall is a completely different animal.  One or two well timed disturbances/lows/storms/whatever can make for a completely different winter wrt snow.  

 

2010-2011 missed by an eyelash down here of being a memorable winter.  I can still remember the storms and the dates.  It could have ended much differently.

 

So, while it may look less than perfect from several angles, it's not inconceivable that it could end up as a good winter from an overall perspective.

 

True, but looking at the goodies definitely points towards those type of events being lower probability. Sure, we "could" do well but given the state of the pdo/enso, the types of events you are describing are flukes in the grand scheme. We are due to time something though so I'll give it that. 

 

As far as the ao is concerned, a -ao definitely helps but not as much as we would like when other things don't cooperate along with it. A solid -nao has much more influence as a standalone feature than the ao during djf. The dec ao last year came in @ -1.749 but we were warm and when it did precipitate it was mostly wet. Especially near the cities. 

[ORH_wxman]

Cohen's 2011 paper doesn't go into the "why" part of it at all.  I think HM has opined on why it might be important in one of the winter threads somewhere.  But, I think the general connection with snowcover (of any sort) is that snowcover enhances troughs through radiative cooling.  These troughs are vertically propagating baroclinic waves, which can disrupt the high altitude/stratospheric polar vortex and lead to cold air being displaced farther south.  Combined with cold airmasses can stay colder if moving over snow covered terrain.  I think that's the general gist...someone correct me if I'm wrong.  

 

 

Here's the link to the paper, but it's paywalled for most average folks.  http://onlinelibrary.wiley.com/doi/10.1029/2011GL049626/full

 

 

The paper can be read for free here:

 

http://web.mit.edu/jlcohen/www/papers/CohenandJones_GRL11.pdf

[WxUSAF]

He did speculate a little about albedo as being a possible player but didn't really go into the mechanism of how in one of his articles.

 

Have you looked at the thread on the OCTOBER PATTERN INDEX(OPI)?  Because of the language problems I couldn't really get a handle on exactly how they formulated the index.  Hopefully,  they'll publish it in a peer reviewed journal in English.  Until then,I won't give it too much credence.   Ricardo suggested that his index would be forecasting a positive AO this winter. 

I have looked at it, Wes.  I feel the same as you regarding it's publication status.  Ricardo admits they are amateurs.  i'm not going to knock his work out of hand for that reason alone, but I am going to be cautious about jumping to conclusions/predictions based off it.  

[Bob Chill]

Cohen's 2011 paper doesn't go into the "why" part of it at all.  I think HM has opined on why it might be important in one of the winter threads somewhere.  But, I think the general connection with snowcover (of any sort) is that snowcover enhances troughs through radiative cooling.  These troughs are vertically propagating baroclinic waves, which can disrupt the high altitude/stratospheric polar vortex and lead to cold air being displaced farther south.  Combined with cold airmasses can stay colder if moving over snow covered terrain.  I think that's the general gist...someone correct me if I'm wrong.  

 

 

Here's the link to the paper, but it's paywalled for most average folks.  http://onlinelibrary.wiley.com/doi/10.1029/2011GL049626/full

 

I agree with the logic. Source region is important. Having a bounce year with arctic ice combined with a pretty substantial area of higher latitude snow cover early in the season can only help and not hurt irt temps at lower latitudes. 

 

Delivery mechanism has been very problematic for our latitude. It can snow with above normal temps to our north but not down here for the most part. Russia, China, and Europe have definitely has more than there share of really frigid temps for the last 5+ years. Hard to say why. I've speculated that it has at least something to do with the open water on the atlantic side of the pole. There has been a stubborn blocking high centered over Russia to the south of the area I'm talking about. This shows it pretty well:

 

 

The pv and bitter arctic airmass just seems bottled up on the other side of the globe for years now. I certainly don't have the knowledge to understand why but it sure it persistent. 

 

Oct this year is a little promising in that department. It's the exact opposite of 2012 so far. 

 

Oct 1-18 2012

 

 

 

Oct 1-18 2013

 

 

 

Maybe this is the year we finally break the trend and have the bitter cold available for north america. We might not be in the crosshairs in our area but having bitter outbreaks in the northern plains sure helps when it modifies and heads our way compared to standard continental air that's seasonal at best by the time it gets here. 

[WxUSAF]

 

 

Maybe this is the year we finally break the trend and have the bitter cold available for north america. We might not be in the crosshairs in our area but having bitter outbreaks in the northern plains sure helps when it modifies and heads our way compared to standard continental air that's seasonal at best by the time it gets here. 

I certainly hope so.  We're certainly in a bit of uncharted waters regarding the persistent cold ENSO.  But I do think it's hard (or it SHOULD be hard) to go so long without snowless winters.  I tend to agree with WinterWxLuvr's point regarding snow forecasts, particularly for our region.  One big storm can be 50-100% of our seasonal totals.  Sometimes that needle gets threaded.  Given the ENSO and PDO regimes (and we'll see what happens with SAI), I'd hedge toward the lower side of climo for our region, but I'm an eternal optimist regarding one lucky break in an otherwise bad winter.   

[usedtobe]

I certainly hope so.  We're certainly in a bit of uncharted waters regarding the persistent cold ENSO.  But I do think it's hard (or it SHOULD be hard) to go so long without snowless winters.  I tend to agree with WinterWxLuvr's point regarding snow forecasts, particularly for our region.  One big storm can be 50-100% of our seasonal totals.  Sometimes that needle gets threaded.  Given the ENSO and PDO regimes (and we'll see what happens with SAI), I'd hedge toward the lower side of climo for our region, but I'm an eternal optimist regarding one lucky break in an otherwise bad winter.   

 

Right now in terms of snow I'd lean towards 8-12" for DCA, towards the median but with more chance of staying below it than above.  I have no real reason except that most years since 1987 we've stayed below the median.  I still don't feel I have much of a feel for the pattern but I rarely do this early. 

[griteater]

The preceding Cohen paper from June 2011 does a better job of explaining the atmospheric response to the snow cover advance. - http://www.nws.noaa.gov/ost/climate/STIP/FY11CTBSeminars/jcohen_062211.pdf

 

Here's a snippet from the conceptual model...

 

"October is the month snow cover makes its greatest advance, mostly across Siberia.

October is also the month that the Siberian high, one of the three dominant centers of action across the

Northern Hemisphere (NH), forms. In years when snow cover is above normal this leads to a strengthened

Siberian high and colder surface temperatures across Northern Eurasia in the fall. We suggest that the

intensification of the Siberian high, along with the thermal impacts of enhanced snow cover and topographic

forcing, corresponds to a positive wave activity flux anomaly in the late fall and early winter, leading to

stratospheric warming and to the January tropospheric negative winter AO response we have mentioned

above."

 

Here's an example of an anomalously strong Siberian High in the 2nd half of October preceding a strongly negative AO/NAO winter (1959-1960)...

 

 

[WxUSAF]

CFSv2 says +PNA/+NAO for December and January and then +PNA/-NAO for February.  

http://www.cpc.ncep.noaa.gov/products/people/wwang/cfsv2fcst/htmls/glbz700e3Mon.html

 

Someone give us some good analogs for that type of winter in the Mid-Atlantic.  I think 93-94 had a predominant +NAO and maybe one of the cold winters in the 70s?  

[Bob Chill]

Last year was a raging +NAO in Dec coming in @ 2.52 on the month and Jan was 1.17. It's one of the reasons that the massive -ao was useless (along with a ghastly pac), There's no reason not to use last year as an analog either. 

 

I'll pull #'s and try to line up a few with a -pdo / enso neutral later today. 

 

ETA: I got my years screwed up. 11-12 was the big +nao and it isn't a good analog. Last year is still a decent one irt to storm track. PAC shouldn't be so awful this year. At least I sure hope it freekin isn't. 

[WxUSAF]

Last year was a raging +NAO in Dec coming in @ 2.52 on the month and Jan was 1.17. It's one of the reasons that the massive -ao was useless (along with a ghastly pac), There's no reason not to use last year as an analog either. 

 

I'll pull #'s and try to line up a few with a -pdo / enso neutral later today. 

Good point, last year is certainly an analog.  With the ENSO state as well.  ENSO forecasts seem fairly tightly clustered to bring us to warm neutral in another couple/few months.  Probably won't do much for us in that regard.  

[Huffwx]

First mention of SNOW in my point and click-

 

http://forecast.weather.gov/MapClick.php?site=rnk&FcstType=text&zmx=1&zmy=1&map.x=260&map.y=111&site=RNK#.UmfaL_nrzKg

 

20% shot of rain, mixing with snow after 5am. 

 

Set your alarms! 

[Bob Chill]

We can't totally overlook 61-62 and 62-63. They were neg neutral and 2 years removed from the last enso event (weak Nino). They were also at the beginning of a long term -pdo period. Great Decembers thanks to a robust -nao.

 

Yes please