JAMSTEC's Frigid Global Fall

[donsutherland1]

I continue to believe blocking will be key. While the latest JAMSTEC winter forecast shows cold over most of the U.S. and Canada, such an outcome is probably highly unlikely.

El Niño events that followed either neutral or La Niña ENSO winters and had warm summers in North America were 1953, 1957, 1986, 1991, and 2002. ENSO-Teleconnection analogs have persistently been highlighting 1953, 1991, and 2002 for the summer.

The following are charts showing temperature anomalies in the U.S. for autumn and fall based on those years:

[CoastalWx]

Coastal,

We're probably on similar wavelengths. I don't know whether or not you saw my edit to my prior post. I added this:

"Based on history, if we can get a WEAK El Nino (which is more likely than a moderate Nino or neutral positive imho) with good blocking, we can have quite a cold eastern US winter. So, that part is quite plausible imo. Weak El Nino's as a whole have seen the coldest E US winters over the last 130 years of any one ENSO phase. However, I wouldn't think it would be cold in the western US and that much of Canada based on the same history."

Cool,GA. It wasn't directed at you at all, just noting the Jamstec seemed too cold out west and the Plains.

[Go Kart Mozart]

FWIW, the euro for the next 10 days looks anomalously cold for eatern North america, especially the Ungava region. A harbinger of good things to come?

[buckeye]

JB seems to like the JAMSTEC's forecast for this summer in the great lakes, which is calling for a cooler than normal summer in the areas that are experiencing the heatwave right now. He's also intrigued by its forecast for the Fall.

....how's that working out? Chicago posters?

[Brewers]

....how's that working out? Chicago posters?

Terribly, Milwaukee is on pace for its warmest summer ever.

[csnavywx]

....how's that working out? Chicago posters?

A failure even bigger than the '11-'12 winter Accuweather forecast you have in your avatar, which is quite a feat.

[carley5]

joe bastardi is very excited about the super cold fall coming.

[mattie g]

A failure even bigger than the '11-'12 winter Accuweather forecast you have in your avatar, which is quite a feat.

To be fair, it was a model "forecast" months out, so it's probably not quite as bad.

[jdrenken]

A failure even bigger than the '11-'12 winter Accuweather forecast you have in your avatar, which is quite a feat.

Actually...JB's prior winter, spring, and summer forecast is 0/3 while Paul has done much better in spring and summer.

[Rainshadow]

I continue to believe blocking will be key. While the latest JAMSTEC winter forecast shows cold over most of the U.S. and Canada, such an outcome is probably highly unlikely.

El Niño events that followed either neutral or La Niña ENSO winters and had warm summers in North America were 1953, 1957, 1986, 1991, and 2002. ENSO-Teleconnection analogs have persistently been highlighting 1953, 1991, and 2002 for the summer.

The following are charts showing temperature anomalies in the U.S. for autumn and fall based on those years:

Bingo Don. Around our area the last three winters have shown if the AO/NAO deviate much from around neutrality they can trump the Tropical Pacific. Regardless there are not too many El Nino Octobers that are warm around our area (I think most of those ensuing winters were duds).

[Ottawa Blizzard]

JB is now tweeting that the CFS has joined the Japanese model in calling for a cool Fall. He seems to like the 1976-77 analog, but this summer has been nothing like 1976. Still, he nailed the 2002-2003 winter when everyone else seemed to be calling for a warm winter.

[WxMidwest]

LOL @ Late 70's winter analog.....

[icebreaker5221]

JB is now tweeting that the CFS has joined the Japanese model in calling for a cool Fall. He seems to like the 1976-77 analog, but this summer has been nothing like 1976. Still, he nailed the 2002-2003 winter when everyone else seemed to be calling for a warm winter.

If you call for a cold winter every year, every once in a while you get it right, even in a warm decade.

[Ottawa Blizzard]

If you call for a cold winter every year, every once in a while you get it right, even in a warm decade.

That's true, but in JBs defence I can recall him laying out his reasons very early on. I think it had something to do with warm water in the Gulf of Alaska and a mxture of warm and cold water off the east coast.

[uncle W]

JB is now tweeting that the CFS has joined the Japanese model in calling for a cool Fall. He seems to like the 1976-77 analog, but this summer has been nothing like 1976. Still, he nailed the 2002-2003 winter when everyone else seemed to be calling for a warm winter.

what...he couldn't find a colder fall?...a more realistic analog would be 1957 or 1986...not everyone was calling for a mild winter that year...At the time I thought it was going to be a clasic around here...

[Ottawa Blizzard]

what...he couldn't find a colder fall?...a more realistic analog would be 1957 or 1986...not everyone was calling for a mild winter that year...At the time I thought it was going to be a clasic around here...

I hope it's not like 1957. That December was really mild up in Ontario.

JB always goes for the most extreme, as I'm sure you know. There's no way we get a winter even close to 1976-77. That was a once-in-a-lifetime event, as was 1993-94.

[MJO812]

FWIW

[IntenseBlizzard2014]

FWIW

The CFS v2 looks warmer than this.

[IntenseBlizzard2014]

Interesting. CFS v1 and v2 split after September.

v1 becomes Cold and Stormy for December and January.

v2 becomes Cold and Dry for December, then Average Temps and Very Dry for January and February

This is for the Eastern Half of the Mississippi River.

It's probably because the Warmest Positive Anomalies redevelop further west for v1, which enables -NAO/-AO episodes. Then for v2, the Warmest Positive Anomalies stay near Nino 1+2, but then eventually drifts into Nino 3.4 and Nino 4 until December, then drifts back into Nino 1+2 after December.

http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/lanina/images3/glbSSTMonPDFC.gif

http://www.cpc.ncep.noaa.gov/products/people/wwang/cfsv2fcst/htmls/usT2me3Mon.html

http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/lanina/images3/usPrecMon.gif

http://www.cpc.ncep.noaa.gov/products/people/wwang/cfsv2fcst/htmls/usPrece3Mon.html

[Jaguars]

Well looking at those cfs maps I'd say that would manifest itself in a warmer to colder fall by november and probably averaging out around average climo. which, after this summer I would sign on the dotted line for right now. As we head into late summer it will be interesting to me to see if the heat waves

will have basically now peaked or if we eclipse the episodes we've already had. CFS clearly keeps it hot in the Central US but looks like it wants to start and reduce that heat as evidenced by the slightly cooler anomalies in northern NE in August and then the big reduction of intense heat Central US come Sept.

[Ender]

I continue to believe blocking will be key. While the latest JAMSTEC winter forecast shows cold over most of the U.S. and Canada, such an outcome is probably highly unlikely.

El Niño events that followed either neutral or La Niña ENSO winters and had warm summers in North America were 1953, 1957, 1986, 1991, and 2002. ENSO-Teleconnection analogs have persistently been highlighting 1953, 1991, and 2002 for the summer.

Using CPC's tri-monthy anomaly numbers for region 3.4 I came up with the following graph for the last two winters, I then extended the graph (dashed) via inclusion of the Dynamical model mean; though I have assumed that the upcoming NDJ will actually end-up at +1.1C and DJF at +1C rather than +1C and +.9C currently shown by the mean. Anyhow, here's the resulting 27 month profile:

That's a fairly pronounced "Nina, Nina, Nino" tri-winter sequence. Note the true double dip as the Nina re-intensified last winter.

A more typical multi-winter Nina is one in which the first winter sees an impressive minima followed by protracted return to normal rather than a second pronounced dip. Sequences that mirrored the 2011-2013 pattern were only seen in 2009-2010, less so, in 1950-1952. The more typical, "gradual recovery extended Nina" was seen in 1956-1957, 1971-1973, 1985-1987, 1996-1998 and 2001-2003. Here are the 10 "Nina, Nina, Nino" like winters along with the two non-fit examples ('52-'54 and '90-'92) referenced in your post above.

I remember reading a paper once wherein the author had constructed a meta index based on a composite of high latitude index values into a sort of "blockiness index" (AO, NAO, PNA, etc). He then correlated the winter blockiness index against the just concluded Atlantic + Pacific tropical season. What I don't recall was whether the meta index he used was some sort of standardized index/methodology...or just something he invented for his paper. Does anyone recall what I'm referring to?

I believe others have already shown that first winter Ninas and Ninos tend to feature quite a bit of high latitude blocking compared to a more neutral year. I also recall, perhaps incorrectly, that second winter Ninas tend to behave more like a neutral year than did the first Nina winter. That is, the high latitude flow behaves "more normally" in the second of a multi Nina winter even while region 3.4 may remain quite a bit below normal. It'd be interesting to see whether something stands out when comparing the 2008 and 2009 DJF/JFM along with the 2011 and 2012 DJF/JFM Ninas against what was seen in the "more typical protracted Nina recovery" examples of say 1985-1987 or 1971-1973.

[donsutherland1]

Very nice work, Ender.

The peak NDJ figure appears reasonable to me. Hopefully, there will be enough blocking for some decent snowstorms and also sufficient cold in Canada to benefit winter industries that have suffered in recent years.

[WxMidwest]

Nice work ender, haven't really taken a look at the ENSO analogs quite in that manner using 2 year analogs. You know what would be neat, a suggestion scatterplott these points for the periods against the current one, and seeing which one has the highest explained Coefficient of Variance. The R^2's with the highest value should trend the best. About the "Blocking Index" deal, I thought "NineInchNails" did work on that quite a while ago on Eastern/American, I'm not for sure though...

[Ender]

Very nice work, Ender.

The peak NDJ figure appears reasonable to me. Hopefully, there will be enough blocking for some decent snowstorms and also sufficient cold in Canada to benefit winter industries that have suffered in recent years.

Nice work ender, haven't really taken a look at the ENSO analogs quite in that manner using 2 year analogs.

Thank you both, though it's not really "work" at this point since I can't support the conclusions that I'd like to draw. Though, I suppose I can accept some praise for making pretty pictures.

You know what would be neat, a suggestion scatterplott these points for the periods against the current one, and seeing which one has the highest explained Coefficient of Variance. The R^2's with the highest value should trend the best. About the "Blocking Index" deal, I thought "NineInchNails" did work on that quite a while ago on Eastern/American, I'm not for sure though...

I'm not sure about the use of CV for this data set since it's basically temp anomaly either side of a 0C mean. I guess I could convert everything to K, that'd eliminate problems posed by negative values, and the asymptotic/infinite CVs. Maybe I'm missing your point, though.

[QVectorman]

I think you can use the Spearmans correlation or CORREL functions in Excel to computer it. But I may be wrong since it is non-linear. It would also eliminate having to worry about neg. values if you used the excel functions.

[WxMidwest]

Spearman's Rank would work too, So will pearson with standardized data, in reality there should not be much of a difference between them CC of either method.. Because the trends run so close, it's fairly linear in nature...

[Ottawa Blizzard]

Very nice work, Ender.

The peak NDJ figure appears reasonable to me. Hopefully, there will be enough blocking for some decent snowstorms and also sufficient cold in Canada to benefit winter industries that have suffered in recent years.

I just hope it isn't like 2009-2010. That winter wasa bust in Canada and much of the northern tier of the US

[mnweather]

I just hope it isn't like 2009-2010. That winter wasa bust in Canada and much of the northern tier of the US

09-10 did stink bad for us northerners. I really doubt next Winter will be as bad as that one. It might not be above average in terms of snowfall but I think it will be an improvement at least.

[Rainshadow]

Using CPC's tri-monthy anomaly numbers for region 3.4 I came up with the following graph for the last two winters, I then extended the graph (dashed) via inclusion of the Dynamical model mean; though I have assumed that the upcoming NDJ will actually end-up at +1.1C and DJF at +1C rather than +1C and +.9C currently shown by the mean. Anyhow, here's the resulting 27 month profile:

That's a fairly pronounced "Nina, Nina, Nino" tri-winter sequence. Note the true double dip as the Nina re-intensified last winter.

A more typical multi-winter Nina is one in which the first winter sees an impressive minima followed by protracted return to normal rather than a second pronounced dip. Sequences that mirrored the 2011-2013 pattern were only seen in 2009-2010, less so, in 1950-1952. The more typical, "gradual recovery extended Nina" was seen in 1956-1957, 1971-1973, 1985-1987, 1996-1998 and 2001-2003. Here are the 10 "Nina, Nina, Nino" like winters along with the two non-fit examples ('52-'54 and '90-'92) referenced in your post above.

I remember reading a paper once wherein the author had constructed a meta index based on a composite of high latitude index values into a sort of "blockiness index" (AO, NAO, PNA, etc). He then correlated the winter blockiness index against the just concluded Atlantic + Pacific tropical season. What I don't recall was whether the meta index he used was some sort of standardized index/methodology...or just something he invented for his paper. Does anyone recall what I'm referring to?

I believe others have already shown that first winter Ninas and Ninos tend to feature quite a bit of high latitude blocking compared to a more neutral year. I also recall, perhaps incorrectly, that second winter Ninas tend to behave more like a neutral year than did the first Nina winter. That is, the high latitude flow behaves "more normally" in the second of a multi Nina winter even while region 3.4 may remain quite a bit below normal. It'd be interesting to see whether something stands out when comparing the 2008 and 2009 DJF/JFM along with the 2011 and 2012 DJF/JFM Ninas against what was seen in the "more typical protracted Nina recovery" examples of say 1985-1987 or 1971-1973.

Dr. Hart (FSU) did a research paper on recurving tropical systems as to whether or not a) they recurved and if they did, maintained tropical characteristics above 40N. While the paper I don't believe talked about blockiness per se, its conclusions were colder winters with a low number of recurving tropical systems.

[am19psu]

Dr. Hart (FSU) did a research paper on recurving tropical systems as to whether or not a) they recurved and if they did, maintained tropical characteristics above 40N. While the paper I don't believe talked about blockiness per se, its conclusions were colder winters with a low number of recurving tropical systems.

Yeah, Dr. Hart's paper hypothesized that more recurving tropical systems in the warm season decreased the overall hemispheric baroclinicity in the cold season.