2013 Fall Medium Range Discussion

[tacoman25]

You left off quite a few years with DJF ONIs of .3-.7. I did exclude 06-07 and 87-88 even though the ONI was .7 in DJF, because both peaked at 1.0 or 1.1 in NDJ which I agree is too Ninoish. The resulting map has a significant Nino signal (I should have posted it the first time). Based on this I would conclude a +.5 is more similar to +1-1.5 than to a -.5 to 0. A -.5 to 0 has a significant Nina-ish signal, while a +.5 has a significant Nino-ish signal.

You left off 51-52, 69-70, 76-77, 77-78

ALL years with a DJF ONI of +.3-.7 (except 06-07 and 87-88):

Well, I was going by peak ONI. Which I think is more important overall. 1951-52 peaked at 1.2 in the fall, WAY higher than this year, and 1976-77 and 1977-78 peaked at .8 in the fall, also easily higher than this year. And 1969-70 peaked at .9 in the fall. The ONI peak in 2012 has been only .4, and the next trimonthly is unlikely to be much higher than that, if at all.

The fact that the two Ninos from the 1970s change the map so much should also tell you that sample size and extreme years is an issue in this evaluation.

I would also urge you to look at the individual years in the whole -.9 to .9 (weak/neutral ENSO years) range. There is a wide variance between different years and a few extreme years that tend to skew the smaller groupings (like .3 to .8), and I really don't see an overall tendency towards the type of Nino/Nina climatology that we see with stronger events. There are other, more important factors than ENSO to consider with most of these years (PDO, AO/NAO, QBO, pattern tendencies in fall, etc), and I believe the same is true with 2012-13.

[uncle W]

Agree taco. Mid/northern latitude teleconnections exert a stronger influence nationwide in the neutral ENSO years, hence quite a bit more variability. There's a reason why Ninas and Ninos, particularly mod/strong, are generally more predictable for LR forecasters. This is especially the case for the southern tier.

I did a little research with the oni index for DJF and how cold it was in NYC...I'll just list the ten coldest/warmest winters and the DJF ONI...

warmest.......................

season.....temp...ONI...

2001-02.....41.5......-0.2

2011-12.....40.5......-0.9

1997-98.....39.6.....+2.2

1990-91.....39.2.....+0.3

1998-99.....38.7......-1.5

1952-53.....38.1.....+0.5

1996-97.....37.8......-0.5

1949-50.....37.5......-1.4

1974-75.....37.5......-0.5

1953-54.....37.4.....+0.7

Coldest...........................

1976.77.....28.5.....+0.6

1962-63.....30.0......-0.4

1977-78.....30.3.....+0.7

1969-70.....30.5.....+0.6

1958-59.....30.9.....+0.6

1993-94.....31.2.....+0.1

2002-03.....31.2.....+1.1

1967-68.....31.3......-0.6

1960-61.....31.7.......0.0

1970-71.....32.1......-1.2

..............................................................................

It's hard to base a temperature forecast on enso alone...Let alone a neutral year...

four of the top warmest had an oni between 0.0 and -0.9...

two had an oni -1.0 and lower...

three had an oni of +0.1 to +0.9

One had an oni of +2.2...

seven of the coldest winters had an oni of 0.0 to +0.9...

one had an oni +1.1...

two had an oni -0.1 to -0.9...

..................................................................................

Weak enso of -0.4 to +0.4 had three cold and one warm...4

weak enso of --0.5 to -0.9 had two cold and three warm...5

stronger enso -1.0 and lower had one cold and two warms...3

weak enso +0.5 to +0.9 had four cold and two warm...6

Stronger enso +1.0 or higher...one cold and one warm...2

........................................................................................

[tacoman25]

I agree the signal is weaker for a weaker event, however ~ +.5s often behave fairly Ninoish, unlike -neutrals which are fairly Ninaish. As such, I believe Don's use of 0-1.5 is preferable to OSU's -.5 to +.5. The latter includes many years which behaved fairly Nina-ish in contrast to the Nino-ish signal of ~ +.5s.

Many of the -neutrals include substantial SE ridging through the winter, which explains the much lower snow totals at DC and BWI despite the -AO. Years with an ONI near +.5 look much more Ninoish and have better snowfall in the SE.

Let's break this down year by year, looking at all years that had a DJF peak in the -.2 to -.8 range or +.2 to .8 range.

1950-51 had a DJF ONI of -.8. The winter had a mixed signal, but actually closer to Nino climo.

1951-52 had a DJF ONI of +.6. The winter had a very Ninaish pattern.

1952-53 had a DJF ONI of +,5. The winter had a national blowtorch, not really Nina or Ninoish.

1953-54 had a DJF ONI of +.7. Another national blowtorch.

1954-55 had a DJF ONI of -.7. It had more of a Ninaish pattern.

1956-57 had a DJF ONI of -.3. Very Ninaish pattern.

1958-59 had a DJF ONI of +.6. It had a very mixed pattern.

1961-62 had a DJF ONI of -.2. Very Ninaish pattern.

1962-63 had a DJF ONI of -.4. More of a Nino-ish pattern.

1964-65 had a DJF ONI of -.6. More of a Ninaish pattern, but not strongly so.

1966-67 had a DJF ONI of -.3. More of a Ninoish pattern.

1967-68 had a DJF ON of -.6. Mixed, but more of a Ninoish pattern.

1969-70 had a DJF ONI of +.6. Very Nino-ish pattern.

1971-72 had a DJF ONI of -.6. Very Ninaish pattern.

1974-75 had a DJF ONI of -.5. Mixed, but more of a Ninaish pattern.

1976-77 had a DJF ONI of +6. Very Ninoish.

1977-78 had a DJF ONI of +7. Somewhat mixed, but more Ninoish.

1979-80 had a DJF ONI of +.5. Definite Nino signal.

1980-81 had a DJF ONI of -.4. Very Ninoish signal.

1983-84 had a DJF ONI of -.5. No real signal either way.

1985-86 had a DJF ONI of -.5. Mixed, but more of a Nino pattern.

1987-88 had a DJF ONI of +.7. More of a Nino pattern, but this was a faded strong Nino.

1990-91 had a DJF ONI of +.3. More of a Nina pattern.

1992-93 had a DJF ONI of +2. More of a Nina pattern.

1996-97 had DJF ONI of -.6. More of a Nina pattern.

2000-01 had a DJF ONI of -.7. Cold nationally, no real signal.

2001-02 had a DJF ONI of -2. National blowtorch, not Nino or Ninaish.

2003-04 had a DJF ONI of +.3. More of a Nino pattern.

2004-05 had a DJF ONI of +.5. Kind of Ninoish, but mostly national blowtorch.

2006-07 had a DJF ONI of +.7. Mixed pattern.

So, of the 14 winters more on the -ENSO side of neutral, 7 had more of a Nina-ish pattern, 5 had more of a Nino-ish pattern, and the rest were mixed. So half the winters did not have much of a Nina pattern overall.

Of the 14 winters more on the +ENSO side of neutral, 6 had more of a Nino-ish pattern, 3 had a Nina-ish pattern, and the rest had a very mixed signal. So more than half of these winters did not have much of a Nino signal overall.

[skierinvermont]

Nice work, taco. If I have time I might go through and see if I get the same results as you. Going on what you've done though I still see a decent signal. A Nina-ish pattern is over 2X more likely at ~ -.5 than at ~ +.5. I agree there is more variance than with stronger events.

It still seem to me that a +.5 meteorologically is about halfway between a -.5 and a +1.5. You go from weak/variable Nina signal to a weak/variable Nino signal to a strong/less variable Nino signal. Even with +1.5s there is variance with some years being national blowtorches etc.

I agree Don's inclusion of +1-1.5s may introduce too much of a Nino signal into the dataset. The best range is likely centered right around .5.

Ideally I'd use to represent this year:

1) 0-1.0

2) -.2 to +1.2 to broaden the SS

3) -.2 to 1.0 if you want to broaden the SS without including the stronger Nino signals of the >1.0s.

4) -.2 to +.8 (similar reasoning to #3 but more so)

Which would you select? From the sound of it it sounds like you would favor #3 or #4. Can we agree that the use of -.5 to +.5 is not good for this year?

[tacoman25]

Nice work, taco. If I have time I might go through and see if I get the same results as you. Going on what you've done though I still see a decent signal. A Nina-ish pattern is over 2X more likely at ~ -.5 than at ~ +.5. I agree there is more variance than with stronger events.

It still seem to me that a +.5 meteorologically is about halfway between a -.5 and a +1.5. You go from weak/variable Nina signal to a weak/variable Nino signal to a strong/less variable Nino signal. Even with +1.5s there is variance with some years being national blowtorches etc.

I agree Don's inclusion of +1-1.5s may introduce too much of a Nino signal into the dataset. The best range is likely centered right around .5.

Ideally I'd use to represent this year:

1) 0-1.0

2) -.2 to +1.2 to broaden the SS

3) -.2 to 1.0 if you want to broaden the SS without including the stronger Nino signals of the >1.0s.

4) -.2 to +.8 (similar reasoning to #3 but more so)

Which would you select? From the sound of it it sounds like you would favor #3 or #4. Can we agree that the use of -.5 to +.5 is not good for this year?

I think any of those are ok. Personally, I would probably start with #3, and then if I wanted a broader representation, move it to -.2 to 1.2 (by peak ONI). But yeah, once you get above 1.3 or so, it's definitely a much stronger Nino signal. And even .6 to 1.0 has a stronger Nino signal than events in the 0 to .5 range. So if I used years in the 0 to -.5 range, I'd probably weight them about the same as .8 to 1.2, with the most weight on years in the 0 to .7 range.

But that's just me. Main thing is I'd want to avoid -.9 or greater years and 1.4+ greater years, as that probably represents the point where things really start deviating from the +ENSO neutral non-tendencies.

[wxmx]

Tell me how do you want the temps anomalies to behave and I can produce the map, with convincent arguments. Also, the tendency of SSTAs will probably be near flat for DJF, but in the medium range, after the current Niñoish GLAAM orbit, it should go back to low amplitude/Niñaish phases, with an overall cooling for ~30-45 days

[Wow]

Tell me how do you want the temps anomalies to behave and I can produce the map, with convincent arguments. Also, the tendency of SSTAs will probably be near flat for DJF, but in the medium range, after the current Niñoish GLAAM orbit, it should go back to low amplitude/Niñaish phases, with an overall cooling for ~30-45 days

My thoughts exactly, Jorge.

[JamieOber]

What effect, if any, does the very dry and quite cold November in the East have on analogs?

[donsutherland1]

What effect, if any, does the very dry and quite cold November in the East have on analogs?

I took autumn precipitation into consideration in coming up with my seasonal idea. My thoughts on seasonal snowfall and precipitation were as follows:

Snowfall should return to near normal figures for the Great Lakes, Ohio Valley, northern Mid-Atlantic, and southern New England regions. Unlike last winter, the potential exists for a Kocin-Uccellini-type snowstorm at some point during the winter on account of the blockiness.

In terms of precipitation, the western half of the U.S. and southern Canada will likely see near to above normal precipitation. The Gulf Coast and Southeastern States could see below normal precipitation.

The CAS model, which is based on soil moisture, forecasts dry conditions up the Eastern Seaboard, with driest ones in the Gulf Coast region and Southeast. There's another large area of dry conditions for the Southwest. My final thoughts were not quite as dry, but there was agreement with regard to the Southeast and Gulf Coast, even when precpitation was excluded. So, at least as far as precipitation ideas go, it appears that my early thoughts might be reasonable.

Of course, we won't know until after the winter. December, though, could offer a clue. I took a more detailed look at NYC, with its lengthy monthly precipitation data. I examined all cases (29) where October had < 3.00" precipitation and November < 2.50" precipitation. Following the most recent storm, NYC still meets those criteria: October precipitation: 2.92" and November precipitation: 1.81". I compared average monthly precipitation for the December-January-February period vs. the October-November period. If December turned notably wetter than the October-November period (200% or above), the winter typically proved wetter.

All 29 cases:

Monthly Average DJF < 150% Montly Average (ON): 21%

Monthly Average DJF 150% or more of the Montly Average (ON): 79%

Monthly Average DJF 200% or more of the Montly Average (ON): 45%

A December-February monthly average of 200% or more of the October-November monthly average was 2.2 times as likely as one that was < 150% of the October-November average.

December < 200% of the October-November Average:

Monthly Average DJF < 150% Montly Average (ON): 33%

Monthly Average DJF 150% or more of the Montly Average (ON): 67%

Monthly Average DJF 200% or more of the Montly Average (ON): 13%

A December-February monthly average of < 150% of the October-November monthly average was 2.5 times as likely as one that was 200% or more above the October-November average.

December < 200% of the October-November Average:

Monthly Average DJF < 150% Montly Average (ON): 7%

Monthly Average DJF 150% or more of the Montly Average (ON): 93%

Monthly Average DJF 200% or more of the Montly Average (ON): 64%

A December-February monthly average of 200% or more of the October-November monthly average was 9.0 times as likely as one that was < 150% of the October-November average.

In short, even as the winter monthly average was almost always higher than the October-November average (only 1 case had a lower figure), a notably wetter December provided a good indication that winter monthly precipitation could average 200% or more of the October-November precipitation.

I also refined the above 29 cases strictly for November precipitation. There was a tendency that monthly winter precipitation had an inverse correlation with November precipitation.

November Precipitation: <2.00":

Monthly Average DJF < 150% Montly Average (ON): 6%

Monthly Average DJF 150% or more of the Montly Average (ON): 94%

Monthly Average DJF 200% or more of the Montly Average (ON): 44%

A December-February monthly average of 200% or more of the October-November monthly average was 7.0 times as likely as one that was < 150% of the October-November average.

November Precipitation: 2.00" or more:

Monthly Average DJF < 150% Montly Average (ON): 38%

Monthly Average DJF 150% or more of the Montly Average (ON): 62%

Monthly Average DJF 200% or more of the Montly Average (ON): 31%

A December-February monthly average of < 150% of the October-November monthly average was 1.3 times as likely as one that was 200% or more above the October-November average.

As I didn't have December data, I considered the risk of lower precipitation than might be the case. Hence, I suggested that snowfall totals would return to near normal. I believe there is more upside potential than downside potential with regard to seasonal snowfall in the Great Lakes, Ohio Valley, northern Mid-Atlantic, and southern New England regions. The elevated November snowfall in parts of the region has typically indicated above normal snowfall. The idea of a pickup of precipitation (though how much remains to be seen) to coincide with a near normal to perhaps somewhat colder than normal winter argues for more snowfall. A number of my analogs had a major snowstorm for the aforementioned region at some point during the winter (and the same idea shows up based on some of the blocky winters I cited in Message #25).

December will provide some important insight.

[Eduardo]

Following the most recent storm, NYC still meets those criteria: October precipitation: 2.92" and November precipitation: 1.81".

Interesting Don. Great stuff, as always.

Question: Are we then to assume that if, say, Sandy had come in further north and dropped additional preciptation on NYC that this somehow would have beared on the outcome of this upcoming winter in terms of real weather?

[JamieOber]

Interesting Don. Great stuff, as always.

Question: Are we then to assume that if, say, Sandy had come in further north and dropped additional preciptation on NYC that this somehow would have beared on the outcome of this upcoming winter in terms of real weather?

Thanks Don, and that's a great question, Eduardo. I've wondered about such things, how large single events effect thinking on long-range forecasts since they can skew precip.

[wxmx]

Thanks Don, and that's a great question, Eduardo. I've wondered about such things, how large single events effect thinking on long-range forecasts since they can skew precip.

That's why a good sample universe is essential for statistics. In this case, if the sample is too small, one single event can skew things, but won't in large enough samples. How large does the sample needs to be? There's where it comes statistics as a branch of math, confidence intervals, sampling, etc.

[donsutherland1]

Interesting Don. Great stuff, as always.

Question: Are we then to assume that if, say, Sandy had come in further north and dropped additional preciptation on NYC that this somehow would have beared on the outcome of this upcoming winter in terms of real weather?

I don't believe it would have made much difference, at least as far as my thinking is concerned. Again, using the NYC data (1869-present):

Wet October (5.00" or more) + Dry November (< 2.50"): DJF: 60% cases averaged 3.00" or more, 50% cases averaged 4.00" or more; JF: 60% cases averaged 3.00" or more, 40% cases averaged 4.00" or more

Somewhat Dry to Dry October (<3.00") + Dry November (<2.50"): DJF: 76% cases averaged 3.00" or more, 24% cases averaged 4.00" or more; JF: 69% cases averaged 3.00" or more, 28% cases averaged 4.00" or more

All Years: DJF: 66% cases averaged 3.00" or more; 28% cases averaged 4.00" or more; JF: 60% cases averaged 3.00" or more; 27% cases averaged 4.00" or more

My overall point is that it is likely that Winter 2012-13 will be wetter than October-November 2012 in terms of average monthly precipitation. It's too soon to tell whether it will turn out to be a wet winter (my guess was near normal precipitation based on a combination of analogs and modeling). Even if Sandy had tracked farther north and brought a significant rainfall to the NYC Metro Area, my winter thoughts would have remained essentially the same. The sample size for the wet October-wet November cases was only 10, so there's a large amount of uncertainty involved.

Finally, it should be noted that the CFSv2 has been trending wetter for Winter 2012-13 in recent runs for eastern North America.

[uncle W]

The last winter that had well above average snowfall From Washington DC to Boston was 2002-03...Some forecasts are calling for above average snowfall in these areas for 2012-13...Since 1947-48 there have been six other winters with at least 60" in Boston...40" in NYC and 20" in Washington DC...1995-96 was the greatest I95 winter for recent times...

season......DCA.....NYC.....Boston...

1947-48.....23.4".....63.2".....89.2"

1960-61.....40.3".....54.7".....61.5"

1963-64.....33.6".....44.7".....63.0"

1966-67.....37.1".....51.4".....60.1"

1977-78.....22.7".....50.7".....85.1"

1995-96.....46.0".....75.6"...107.6"

2002-03.....40.4".....49.3".....67.7"

[Isotherm]

Wanted to make a post addressing some recent comments/concerns about the surge positive in NAO/AO that's possible over the next 7-10 days, namely it's potential ramifications for down the road.

I did some research, looking back at daily NAO/AO fluctuations, and interestingly enough, my top analog for this winter (1968-69) had similar behaviour NAO/AO wise throughout the autumn.

NAO/AO values plummeted in October/early November w/ numbers as low as -3 SD. However, by late November, daily NAO readngs went positive, and the AO reached a peak of +1.7 SD on December 3rd 1968.

As a result, the following H5 pattern developed in the last week of Nov through early December 1968.

Looks like a terrible pattern across the northern latitudes, right? That's because it is a terrible pattern. Low heights dominating from AK eastward to Greenland and in the Arctic as well. For that period, a significant pattern relaxation occurred, and in turn, sfc temps responded mild in much of sern Canada and the US for the first week of December 1968:

Not an all out blowtorch pattern, but certainly biased warmer than normal for the early portion of December. We might see a similar outcome here, with a bit more warmth.

Moving onward, the global regime changed dramatically post December 10th, and the H5 pattern for Dec 10-30th 1968 looked like this:

Is this the same winter?

AO values went from near +2 in early December, to near neutral by December 15th, then in the tank around Christmas, near -4.5 SD. Thus the monthly AO value of slightly negative for DEC 1968 does not tell the whole story of the progression of the month. NAO readings went slightly negative, but nothing crazy, for the remainder of December.

The temperature change at the surface was also impressive for the Dec 10-30th period 1968:

Note the major cold over NW Canada that we've already seen show up in NOV 2012. The source region was not too dissimilar back in 1968. The Canadian cold began making attacks into the nern tier by the middle of December, and the East Coast shared in some of the cold fun by the second half of December.

As far as snowfall, that's not what this post is about, but let's see if the 1968-69 pattern analogged here will hold merit. The 2012 pattern will obviously not evolve in an identical way - the timing could be slightly different, but the idea of a pattern break w/ +NAO/AO period for a couple weeks is not uncommon. The turn positive in the indices does not necessarily mean it's game over - quite the contrary. This is in fact a transitional period of the year, and part of the reason we're seeing such extreme run to run model variance is due to all the re-shuffling that's occuring at H5 across the northern hemisphere. Time will tell if 2012 follows a similar route to 1968, or maybe not. But the point is, the pattern looking unfavorable for the next couple weeks is not reason to worry about the winter as a whole going down the tubes or even writing off the rest of December.

[GaWx]

Isotherm,

Excellent and very interesting/encouraging post. My main concern is that 68-9 was a borderline weak/ moderate Nino vs. the current significantly cooler SST's neutral positive. I wonder if that may make it more difficult for a similar evolution this winter. Thoughts?

[uncle W]

Wanted to make a post addressing some recent comments/concerns about the surge positive in NAO/AO that's possible over the next 7-10 days, namely it's potential ramifications for down the road.

I did some research, looking back at daily NAO/AO fluctuations, and interestingly enough, my top analog for this winter (1968-69) had similar behaviour NAO/AO wise throughout the autumn.

NAO/AO values plummeted in October/early November w/ numbers as low as -3 SD. However, by late November, daily NAO readngs went positive, and the AO reached a peak of +1.7 SD on December 3rd 1968.

As a result, the following H5 pattern developed in the last week of Nov through early December 1968.

Looks like a terrible pattern across the northern latitudes, right? That's because it is a terrible pattern. Low heights dominating from AK eastward to Greenland and in the Arctic as well. For that period, a significant pattern relaxation occurred, and in turn, sfc temps responded mild in much of sern Canada and the US for the first week of December 1968:

Not an all out blowtorch pattern, but certainly biased warmer than normal for the early portion of December. We might see a similar outcome here, with a bit more warmth.

Moving onward, the global regime changed dramatically post December 10th, and the H5 pattern for Dec 10-30th 1968 looked like this:

Is this the same winter?

AO values went from near +2 in early December, to near neutral by December 15th, then in the tank around Christmas, near -4.5 SD. Thus the monthly AO value of slightly negative for DEC 1968 does not tell the whole story of the progression of the month. NAO readings went slightly negative, but nothing crazy, for the remainder of December.

The temperature change at the surface was also impressive for the Dec 10-30th period 1968:

Note the major cold over NW Canada that we've already seen show up in NOV 2012. The source region was not too dissimilar back in 1968. The Canadian cold began making attacks into the nern tier by the middle of December, and the East Coast shared in some of the cold fun by the second half of December.

As far as snowfall, that's not what this post is about, but let's see if the 1968-69 pattern analogged here will hold merit. The 2012 pattern will obviously not evolve in an identical way - the timing could be slightly different, but the idea of a pattern break w/ +NAO/AO period for a couple weeks is not uncommon. The turn positive in the indices does not necessarily mean it's game over - quite the contrary. This is in fact a transitional period of the year, and part of the reason we're seeing such extreme run to run model variance is due to all the re-shuffling that's occuring at H5 across the northern hemisphere. Time will tell if 2012 follows a similar route to 1968, or maybe not. But the point is, the pattern looking unfavorable for the next couple weeks is not reason to worry about the winter as a whole going down the tubes or even writing off the rest of December.

there wasn't many positive nao days during the 1968-69 winter...It went slightly positive around the Lindsay storm for two days...You can count the positive nao days on one hand...

1968 12 1 -0.212

1968 12 2 -0.194

1968 12 3 -0.417

1968 12 4 -0.787

1968 12 5 -1.215

1968 12 6 -1.697

1968 12 7 -1.436

1968 12 8 -1.032

1968 12 9 -0.808

1968 12 10 -0.701

1968 12 11 -0.381

1968 12 12 -0.079

1968 12 13 0.018

1968 12 14 -0.118

1968 12 15 -0.236

1968 12 16 -0.477

1968 12 17 -0.732

1968 12 18 -0.796

1968 12 19 -0.484

1968 12 20 -0.103

1968 12 21 -0.196

1968 12 22 -0.080

1968 12 23 -0.074

1968 12 24 -0.169

1968 12 25 -0.347

1968 12 26 -0.869

1968 12 27 -1.114

1968 12 28 -1.095

1968 12 29 -1.194

1968 12 30 -0.885

1968 12 31 -0.526

1969 1 1 -0.422

1969 1 2 -0.462

1969 1 3 -0.596

1969 1 4 -0.923

1969 1 5 -1.096

1969 1 6 -0.908

1969 1 7 -0.961

1969 1 8 -1.302

1969 1 9 -1.410

1969 1 10 -1.293

1969 1 11 -1.052

1969 1 12 -0.748

1969 1 13 -0.526

1969 1 14 -0.432

1969 1 15 -0.392

1969 1 16 -0.408

1969 1 17 -0.280

1969 1 18 -0.288

1969 1 19 -0.189

1969 1 20 -0.258

1969 1 21 -0.466

1969 1 22 -0.390

1969 1 23 -0.297

1969 1 24 -0.171

1969 1 25 -0.017

1969 1 26 -0.021

1969 1 27 -0.205

1969 1 28 -0.255

1969 1 29 -0.087

1969 1 30 -0.024

1969 1 31 -0.027

1969 2 1 -0.243

1969 2 2 -0.290

1969 2 3 -0.160

1969 2 4 -0.241

1969 2 5 -0.599

1969 2 6 -0.678

1969 2 7 -0.524

1969 2 8 -0.180

1969 2 9 0.044

1969 2 10 0.172

1969 2 11 -0.022

1969 2 12 -0.506

1969 2 13 -0.612

1969 2 14 -0.551

1969 2 15 -0.652

1969 2 16 -0.973

1969 2 17 -1.075

1969 2 18 -1.015

1969 2 19 -1.086

1969 2 20 -1.129

1969 2 21 -0.944

1969 2 22 -0.790

1969 2 23 -0.685

1969 2 24 -0.681

1969 2 25 -0.783

1969 2 26 -0.808

1969 2 27 -0.828

1969 2 28 -0.694

[Isotherm]

Isotherm,

Excellent and very interesting/encouraging post. My main concern is that 68-9 was a borderline weak/ moderate Nino vs. the current significantly cooler SST's neutral positive. I wonder if that may make it more difficult for a similar evolution this winter. Thoughts?

Thanks Larry. You bring up a good point. I'm sure if we had stratosphere data for back then, we'd probably see a more favorable picture up there given the Nino. However, the tropospheric NAO/AO pattern of 1968 shares a strong similarity to 2012. 1968 had several months of -NAO from the summer straight through November; 2012 has also had a very impressive, long duration stretch of -NAO beginning in June this year, with no break since then (in the monthly values, there have been daily fluctuations positive). Point being, the tropospheric pattern has been conducive for the recurrence of high latitude blocking, and given the severe AO and NAO drops we've seen in the past month, history demonstrates that the tendency would continue through the DJF period.

In fact, since 1950, I can't find many other years in which the NAO remained negative from mid summer through November with no interruption. The other closest years appear to be fairly recent, 2008 (very similar November in the US) and 2010, which had most of the year -NAO. I definitely believe the very low solar is a strong forcing for this since 2008.

There were many Ninos in which the high latitude pattern was terrible, with little blocking. So while the Nino definitely helps warm the stratosphere, increasing the chance for a downwelling/vortex disrupting event, the tropospheric patterns of these Ninos vary greatly.

[Isotherm]

there wasn't many positive nao days during the 1968-69 winter...It went slightly positive around the Lindsay storm for two days...You can count the positive nao days on one hand...

1968 12 1 -0.212

1968 12 2 -0.194

1968 12 3 -0.417

1968 12 4 -0.787

1968 12 5 -1.215

1968 12 6 -1.697

1968 12 7 -1.436

1968 12 8 -1.032

1968 12 9 -0.808

1968 12 10 -0.701

1968 12 11 -0.381

1968 12 12 -0.079

1968 12 13 0.018

1968 12 14 -0.118

1968 12 15 -0.236

1968 12 16 -0.477

1968 12 17 -0.732

1968 12 18 -0.796

1968 12 19 -0.484

1968 12 20 -0.103

1968 12 21 -0.196

1968 12 22 -0.080

1968 12 23 -0.074

1968 12 24 -0.169

1968 12 25 -0.347

1968 12 26 -0.869

1968 12 27 -1.114

1968 12 28 -1.095

1968 12 29 -1.194

1968 12 30 -0.885

1968 12 31 -0.526

1969 1 1 -0.422

1969 1 2 -0.462

1969 1 3 -0.596

1969 1 4 -0.923

1969 1 5 -1.096

1969 1 6 -0.908

1969 1 7 -0.961

1969 1 8 -1.302

1969 1 9 -1.410

1969 1 10 -1.293

1969 1 11 -1.052

1969 1 12 -0.748

1969 1 13 -0.526

1969 1 14 -0.432

1969 1 15 -0.392

1969 1 16 -0.408

1969 1 17 -0.280

1969 1 18 -0.288

1969 1 19 -0.189

1969 1 20 -0.258

1969 1 21 -0.466

1969 1 22 -0.390

1969 1 23 -0.297

1969 1 24 -0.171

1969 1 25 -0.017

1969 1 26 -0.021

1969 1 27 -0.205

1969 1 28 -0.255

1969 1 29 -0.087

1969 1 30 -0.024

1969 1 31 -0.027

1969 2 1 -0.243

1969 2 2 -0.290

1969 2 3 -0.160

1969 2 4 -0.241

1969 2 5 -0.599

1969 2 6 -0.678

1969 2 7 -0.524

1969 2 8 -0.180

1969 2 9 0.044

1969 2 10 0.172

1969 2 11 -0.022

1969 2 12 -0.506

1969 2 13 -0.612

1969 2 14 -0.551

1969 2 15 -0.652

1969 2 16 -0.973

1969 2 17 -1.075

1969 2 18 -1.015

1969 2 19 -1.086

1969 2 20 -1.129

1969 2 21 -0.944

1969 2 22 -0.790

1969 2 23 -0.685

1969 2 24 -0.681

1969 2 25 -0.783

1969 2 26 -0.808

1969 2 27 -0.828

1969 2 28 -0.694

Yep, very negative NAO year. Solar parameters in the late 60s were quite weak as well which likely helped with the blocking maintenance.

[donsutherland1]

Some morning thoughts...

1. The Arctic Oscillation (AO) has now dropped below -3.000. This morning's reading was -3.236.

2. In my winter thoughts (Message #6), one of the blocking analogs was 2002. Today, in commenting on the excessive precipitation event now impacting the West Coast, Wes pointed that the leading CIPPS analog is 2002.

3. One of my top analogs for the first week of December was the 12/7-9/1980 period. Highest readings from that period for select cities were:

Boston: 61°

New York City: 64°

Philadelphia: 67°

Washington, DC: 71°

The 11/30 12z MEX MOS is showing the following highest readings for the first week of December:

Boston: 60°

New York City: 62°

Philadelphia: 63°

Washington, DC: 67°

December 1980 would later feature a flip to a colder pattern.

At this point in time, with the blocking likely to persist, I have little reason to change my thinking that a transition to a colder patter will likely get underway at some point during the second week in December. The second half of December could wind up colder than normal.

[IntenseBlizzard2014]

I have some old stratosphere data. I'll post it at the Stratosphere Thread in a little while.

[Ottawa Blizzard]

Some morning thoughts...

1. The Arctic Oscillation (AO) has now dropped below -3.000. This morning's reading was -3.236.

2. In my winter thoughts (Message #6), one of the blocking analogs was 2002. Today, in commenting on the excessive precipitation event now impacting the West Coast, Wes pointed that the leading CIPPS analog is 2002.

3. One of my top analogs for the first week of December was the 12/7-9/1980 period. Highest readings from that period for select cities were:

Boston: 61°

New York City: 64°

Philadelphia: 67°

Washington, DC: 71°

The 11/30 12z MEX MOS is showing the following highest readings for the first week of December:

Boston: 60°

New York City: 62°

Philadelphia: 63°

Washington, DC: 67°

December 1980 would later feature a flip to a colder pattern.

At this point in time, with the blocking likely to persist, I have little reason to change my thinking that a transition to a colder patter will likely get underway at some point during the second week in December. The second half of December could wind up colder than normal.

Don, 1980 was the year I was born, so I have no memory of that December, but from stats I can see that it was a pretty impressive month. In fact, it was the coldest Christmas morning over recorded at both Toronto Pearson and Ottawa airport.

Brett Anderson of accuweather posted hisi nterpretation of the latest european weeklies and (according to him) they seem to show very little cold making their way into this region in December and in fact show a very mild Christmas week. Do you have access to the weeklies? What is your interpretation.

Judging from Joe Bastardi's tweets, he would seem to be in your camp, calling for sustained cold to move in weeks 3 and 4.

[tacoman25]

Some morning thoughts...

1. The Arctic Oscillation (AO) has now dropped below -3.000. This morning's reading was -3.236.

2. In my winter thoughts (Message #6), one of the blocking analogs was 2002. Today, in commenting on the excessive precipitation event now impacting the West Coast, Wes pointed that the leading CIPPS analog is 2002.

3. One of my top analogs for the first week of December was the 12/7-9/1980 period. Highest readings from that period for select cities were:

Boston: 61°

New York City: 64°

Philadelphia: 67°

Washington, DC: 71°

The 11/30 12z MEX MOS is showing the following highest readings for the first week of December:

Boston: 60°

New York City: 62°

Philadelphia: 63°

Washington, DC: 67°

December 1980 would later feature a flip to a colder pattern.

At this point in time, with the blocking likely to persist, I have little reason to change my thinking that a transition to a colder patter will likely get underway at some point during the second week in December. The second half of December could wind up colder than normal.

Given the current state of the PDO phase and fall patterns we have seen, I have to lean more towards Isotherm's 1968-69 analog than 1980-81 or 2002-03.

[easternsnowman]

Adam in the philly forum and others have said that the euro weeklies are not reliable beyong week 2, they had a cold pattern recently...it changes on a regular basis back and forth.

[usedtobe]

Given the current state of the PDO phase and fall patterns we have seen, I have to lean more towards Isotherm's 1968-69 analog than 1980-81 or 2002-03.

The Dec 2002 analog is one only used based on the model forecasts of a number of fields over a fairly small window over the west to give an idea of the potential of the current storm hitting the west coast. Too many people try to extend the use of analogs beyond what they are intended for. It was not meant to be used to try to forecast this winter.

[usedtobe]

Don, I think we get through the 1st two weeks of Dec and that they will be in the warm camp. Week 3 we may start seeing a change but I'm not even sold on that yet.

[easternsnowman]

Decembers over the last 15 years or so have been nearly 1 degree above the 1981-2010 norms. There have been many more decembers like last year than the one in 2010.

[tacoman25]

The Dec 2002 analog is one only used based on the model forecasts of a number of fields over a fairly small window over the west to give an idea of the potential of the current storm hitting the west coast. Too many people try to extend the use of analogs beyond what they are intended for. It was not meant to be used to try to forecast this winter.

Agreed. Don seemed to mention it as a winter analog in this case, tying it into your post about 2002 in the West Coast event thread and his previous comments about it as a blocking analog.

[tacoman25]

Decembers over the last 15 years or so have been nearly 1 degree above the 1981-2010 norms. There have been many more decembers like last year than the one in 2010.

However, over the last 5 years, December has been quite the cold month nationally, and near normal overall for much of the east coast.

[icebreaker5221]

Raging +EPO, -PNA on today's long range 12Z ECMWF. You can pop an east coast trough with that pattern, but it will be transient. The zonal jet will just be too strong in the western half of the country, pushing things along.