OCTOBER PATTERN INDEX (OPI) MONITORING WINTER SEASON 2014-2015

[Weather-history]

If we have a pattern change to negative AO say last week of Jan and lasting through Feb will people be quite so ready to write off the OPI theory.

I would be more inclined to let the rest of the winter pan first before making any decisions as to it's viability or lack thereof.

I can just imagine if the OPI had been presented back in autumn 1946 and a first half of january 47 being mild after a brief bit of cold around Christmas 1946 and people queuing up to call it a busted flush. they wouldn't have looked so clever come March 47.

Just musing to myself...................................................

As I mentioned in another thread today, December 1946 was overall a cold month, it had a CET of 3.1C

I very much doubt people would have said it was a busted flush at the time if it existed.

I would say the OPI theory is looking a bit ropey at the moment.

Edited January 2, 2015 by Weather-history

[Tom Quintavalle]

Hi Kevin,

Yes, one would have to say its under a bit of pressure. I think part of the problem is the miniscule sample of years that have been researched, hardly amounting to a grain of sand, against a desert of climatological records.

Its quite conceivable that we could be staring down the barrel of a "fail" year, of course. The "flies in the ointment", that have been mentioned earlier and have been flagged up by the likes of Tamara, Nick F, etc, in such well reasoned posts on other threads, are no doubt contributing to what is seemingly a disconnect with the SAI theory, at the moment, and could prove a severe test to the OPI.

But I still feel its too early to start writing obituaries, about its demise.

Tom.

Edited January 2, 2015 by TomBR7

[Interitus]

Well as described previously in the thread, for comparison below is a graph of winter AO from 1950-2013 alongside a Taymyr index from the previous Octobers. This is derived from the October 500mb geopotential anomaly calculated over the same time period, from an arbitrary location near the northern tip of the Taymyr peninsula (77.5°N 104°E), simply divided by -100.

The data is a timeseries obtained from NOAA/ESRL NCEP reanalysis - http://www.esrl.noaa.gov/psd/cgi-bin/data/timeseries/timeseries1.pl - and is not optimised for location, nor adjusted to improve fit.

 

 

The mean geopotential from last October of 5245.177m represents a 1950-2013 anomaly of 53.96 metres giving a forecast winter AO of about -0.54 which stands a chance of being reasonably close given any favourable conditions in the next couple of months. Predictions near to -3 always had to be considered as unlikely at best.

 

 

 

[SE Blizzards]

OPI - It's all gone

 

 

shaped!

 

 

[picog]

OPI - It's all gone

 

download.jpg

 

shaped!

There are still 5.5 weeks of winter left, things could change. Mild and wet is not a done deal by any means.

[mountain shadow]

There are still 5.5 weeks of winter left, things could change. Mild and wet is not a done deal by any means.

Still time for what? The extremely negative OPI was forecasted to have at least two Winter months with a net -AO, it just hasn't happened. With the first quarter of February likely to be lacking in a HLB it would take an 1947 style turnaround to save the OPI's credibility.

Edited January 20, 2015 by mountain shadow

[Hotspur62]

Still time for what? The extremely negative OPI was forecasted to have at least two Winter months with a net -AO, it just hasn't happened. With the first quarter of February likely to be lacking in a HLB it would take an 1947 style turnaround to save the OPI's credibility.

I have to agree with the above and think we are pretty certain of no HLB in next 2 weeks and that is 2 months of winter gone. It just goes to prove that OPI is only an indicator and is not a "straw to clutch" for definite cold winters for the future.I think the problem was that it caused so much hype on here when the final calculations were made that some on here saw it as Red that we would have a colder than average winter and a possible December 2010 scenario.However,it is another tool and it can be used going forward but obviousily not a guarantee for cold.

[Continental Climate]

Its definitely a tool that can be looked at when making a winter forecast but IMO not one of the more important ones. There are many other factors and it seems that each season has its own new fad. I for one will only be looking at three main factors in the future these being the NAO the AO and most importantly of all the activity on the surface of the sun. Sunspot activity was record low in 2009/10. The last two years it has recovered. Now the NAO and AO and sunspots are probably all related to some degree as the jet stream does go south with a quiet sun, so what Im saying is the prime driver of a cold winter in the UK is how active the sun is. (IMO of course)

Edited January 21, 2015 by Continental Climate

[Gael_Force]

Its definitely a tool that can be looked at when making a winter forecast but IMO not one of the more important ones. There are many other factors and it seems that each season has its own new fad. I for one will only be looking at three main factors in the future these being the NAO the AO and most importantly of all the activity on the surface of the sun. Sunspot activity was record low in 2009/10. The last two years it has recovered. Now the NAO and AO and sunspots are probably all related to some degree as the jet stream does go south with a quiet sun, so what Im saying is the prime driver of a cold winter in the UK is how active the sun is. (IMO of course)

 

I would go further than that and say some of the indices have been thwarted/influenced by being looked at as a reflection of ice loss or snowcover. The Taymyr connection was discussed earlier in this thread - it is in an area considered to be strongly influenced by high pressure in low solar conditions.

 

Some recent research, behind a paywall, unfortunately.

 

http://www.sciencedirect.com/science/article/pii/S0273117714007340

[Weather-history]

so what Im saying is the prime driver of a cold winter in the UK is how active the sun is. (IMO of course)

What I find interesting is that of the 6 great winters of the 20th century, the ones that had a sub 2C CET, not one actually occurred at or just after a solar minimum. 4 of them in fact occurred close to a solar maximum.

[Interitus]

I would go further than that and say some of the indices have been thwarted/influenced by being looked at as a reflection of ice loss or snowcover. The Taymyr connection was discussed earlier in this thread - it is in an area considered to be strongly influenced by high pressure in low solar conditions.

 

Some recent research, behind a paywall, unfortunately.

 

http://www.sciencedirect.com/science/article/pii/S0273117714007340

 

From the abstract that paper refers to reconstruction of summer temperatures but more importantly looks at solar cycles which are in the region 60-140 years long, this year is not going to be dramatically different to any recent year in that time scale.

The Taymyr October index is an annual circulation signal which was referenced in this paper - http://onlinelibrary.wiley.com/doi/10.1002/joc.3968/abstract

Unfortunately it is also paywalled but similar results are very simple to reproduce and verify.

[damianslaw]

What I find interesting is that of the 6 great winters of the 20th century, the ones that had a sub 2C CET, not one actually occurred at or just after a solar minimum. 4 of them in fact occurred close to a solar maximum.

 

 

I may be wrong, but I think I read a paper which said periods of low solar activity combined with a cold PDO tend to produce cold winters, but the odd year/period will buck the trend. I think it is this combination which has greatly attributed to the run of cold winters in recent years notably 08-09 to 10/11 and to an extent 12/13, incidentally winter 11/12 coincided with a spike in solar activity and obviously 13/14 coincided with peak of solar cycle and this winter has been on the back of quite strong activity, the current mild period since late 13 could well be one of those periods bucking the trend but we will go back to conditions pre summer 13 soon.. as solar activity relents and we remain in the cold phase of the PDO.

 

Correct me if I'm wrong.

 

Back to the topic at hand, there was certainly a lot of frenzied feeling in October based on the OPI, I suspect this coming October will be very different with people dismissing any signs and signals to take from it. Once again it proves our weather perplexes with its complexity.

Edited January 21, 2015 by damianslaw

[Weather-history]

I may be wrong, but I think I read a paper which said periods of low solar activity combined with a cold PDO tend to produce cold winters, but the odd year/period will buck the trend. I think it is this combination which has greatly attributed to the run of cold winters in recent years notably 08-09 to 10/11 and to an extent 12/13, incidentally winter 11/12 coincided with a spike in solar activity and obviously 13/14 coincided with peak of solar cycle and this winter has been on the back of quite strong activity, the current mild period since late 13 could well be one of those periods bucking the trend but we will go back to conditions pre summer 13 soon.. as solar activity relents and we remain in the cold phase of the PDO.

 

Correct me if I'm wrong.

 

Back to the topic at hand, there was certainly a lot of frenzied feeling in October based on the OPI, I suspect this coming October will be very different with people dismissing any signs and signals to take from it. Once again it proves our weather perplexes with its complexity.

The issue I have with solar activity being used as a reason for why this winter hasn't gone to what some may have thought is that activity compared to previous recent maxima is lower than these peaks, around which 4 of the severest winters of the 20th century occurred. Going further back 1894-95 occurred just after a maximum. 1968-69 occurred around a maximum.

Now solar activity didn't prevent these winters being cold to very notable. Other factors must have played part.

[Woollymummy]

Quote from the TIL -OPI site:

"Specifically over Europe, our OPI evolution and October pattern analysis, suggest that the first cold anti-zonal period should arrive with good probability in early winter, followed by the onset of a stronger Polar Jet over the North Atlantic and a milder but more unsettled period. This period with higher zonality should be interrupted by a new increasing of the wave activity over the North Atlantic, with a renewed colder pattern; differently from the first period, this second colder period could be characterized by a lower anti-zonality, with higher probabilities for arctic air masses to reach lower latitudes."

........are we nearly on that bit yet?

 

see the page:https://app.til.it/opi/

[snowking]

For interest purposes

 

CET is currently running at +0.88c, which means we are unlikely to finish below average now.

 

That being the case, this will be the first January within the OPI recorded years (back to 1978) with figure < -1 which has not produced a below average January. Every such January before it, be it under varying ENSO conditions and indeed QBO direction, finished below the 71-00 average.

 

Does this mean OPI is a bust? Absolutely not. It was never designed to be correlated to the CET, it just so happened that up until now, it did. But this simply means we look more closely at the data now and perform some sort of post-mortem to see if there's anything we can pick out which may have caused this (and it may be that nothing comes to light directly). It does go to show the risks of working with such a limited dataset though (partially the reason I remain sceptical about various other datasets within climatology).

 

But this is a classic example of scientific advancement. To leave you with a rather overly-philisophical Friday night piece of advice (especially being prior to any sort of alcohol consumption too) you should never be afraid to fail.

 

SK

[Interitus]

 

Does this mean OPI is a bust? Absolutely not. It was never designed to be correlated to the CET

 

Yes, it's the failure to predict the AO which means it's a bust.

[Midlands Ice Age]

Yes, it's the failure to predict the AO which means it's a bust.

 

Interitus and SK

 

I agree about the bust AO for Dec and Jan.....

.

I thought that Cohen had admitted that things were not going quite right at the end of December (or was it November) when he wrote about the abnormal retention of low heights over Northern Siberia  and it could change things somewhat this year...

 

However looking at the patterns he predicted for January for low heights widely across Europe (and moving towards Scandinavia by the end of January into February) as WMummy suggested above are we not moving currently into that position now? All be it 3 - 4 weeks later than he predicted by the looks of it. His winter forecast also included February so lets what happens in February. I do not consider it a total bust (yet) but let us see.

 

Currently the US appear to have the winter synoptics he predicted with the east of the US having a lot of snowy colder  cyclonic conditions pretty much straight through.

 

The only real bust seems to be the greenland high. It appears to have been located at the other side of the pole.

It may well be that the Siberian low had snookered the early  winter pattern, but it looks as if is trying to get back onto his predicted path by February (at least to me) and even without the existance of the 'Greenie'  high.

 

I must admit I have been amazed this year by the continuation of the cold theme for the last 2 weeks, withoiut the HLBlocking which have accompanied most other cold winters.. 

 

Could we end up with a back loaded winter this year? It has happened  occasionally in the past and we could end up with a cold february/march instead of January/February. 

 

As to the solar variations theory. I am still a believer, but this solar cycle sure is strange with an extended (but relatively low strength) solar maximum happening at the current time..I have seen papers that it is the length of the solar cycle which seems to affect the northern blocking pattern. Many of the cycles 19 to 23 were very high maxes, but all of the standard 11 year pattern in length. Hence it is suggested  - little HLB.

 

Several of the cold winter periods  in the 18th and 19th centuries occured during low maximums its true  but also they were during periods of extended length of the solar cycle.

Just maybe they are connected.

 

Solar scientists have predicted this current cycle (24) to be extended to the range of 13 to 14 years so once the solar max's have past we may enter a period when more blocking (as per 2009-2011) is prevalent.

 

It certainly is an interesting time to be watching how these things evolve.

 

MIA

Edited January 31, 2015 by Midlands Ice Age

[Interitus]

Interitus and SK

 

I agree about the bust AO for Dec and Jan.....

.

I thought that Cohen had admitted that things were not going quite right at the end of December (or was it November) when he wrote about the abnormal retention of low heights over Northern Siberia  and it could change things somewhat this year...

 

However looking at the patterns he predicted for January for low heights widely across Europe (and moving towards Scandinavia by the end of January into February) as WMummy suggested above are we not moving currently into that position now? All be it 3 - 4 weeks later than he predicted by the looks of it. His winter forecast also included February so lets what happens in February. I do not consider it a total bust (yet) but let us see.

 

 

The correlation and forecast is for the mean AO for the whole of the winter, the NCEP monthly AO for December was 0.413 and January will be even higher, so the February value is going to have be lower than the 2010 record of -4.266 to get anywhere near the final OPI of -2.18.

It is also nothing to do with Cohen, but even assuming that his analysis provides an explanation, when has a forecast being 3-4 weeks late ever been considered good? Having a 3-month window in which the AO could be low at the beginning, middle or end or any combination of these and would count as verification is plenty of leeway already.

[damianslaw]

Given how the winter is panning out, I'll be very surprised to see this thread resurrected with any real gusto come October..

[VillagePlank]

Yes, it's the failure to predict the AO which means it's a bust.

 

Even cursory research suggested that this might be the case, unfortunately.

 

https://forum.netweather.tv/topic/81494-october-pattern-index-opi-monitoring-winter-season-2014-2015/?p=3063183

 

The search for the definitive UK winter prediction index marches on ....

[mountain shadow]

Even cursory research suggested that this might be the case, unfortunately.

 

https://forum.netweather.tv/topic/81494-october-pattern-index-opi-monitoring-winter-season-2014-2015/?p=3063183

 

The search for the definitive UK winter prediction index marches on ....

Indeed,

 

I would love for the more learned, qualified members on here to attempt a Southern Greenland Predictive Height Anonomaly Index. If we stuck a pin in Southern Greenland and drew a box 1000 square miles from that point and attempted a seasonal height index from say Mid November, trying to forecast the Winter, i think it would work.

 

Calculating it would be far far beyond my intellect.

[sundog]

Given how the winter is panning out, I'll be very surprised to see this thread resurrected with any real gusto come October..

Indeed cant see next octobers OPI thread being nearly as busy as this one was last October.

[jonboy]

As was said at the outset this is a new tool with not just snow cover as its main predictor. Come october I am sure we will see a whole new set solar influences as we come of a prolonged maximum ( not expected)  which is likely to fall rapidly once it goes into decline.

I for one will be following the OPI with great interest

[johnholmes]

It is pretty obvious that some of the claims made in here, NOT by the authors has been wide of the mark.

It is possible, once they have had time to review their work, get it peer reviewed, and for it to be accepted as yet another tool in the attempt to predict a winter before winter has started then that will be fine.

[Gael_Force]

It is pretty obvious that some of the claims made in here, NOT by the authors has been wide of the mark.

It is possible, once they have had time to review their work, get it peer reviewed, and for it to be accepted as yet another tool in the attempt to predict a winter before winter has started then that will be fine.

 

I think it's worth pointing out that the authors were themselves doubting the figure that they arrived at. Some other research, based on a Zonality index, contradicted the signal. I managed to read the cached version of their paper when it was let slip on another site - I'm still puzzled by the time span chosen when they were using the Meteociel reanalysis archive. There's more than a hundred years on there, could a longer term analysis have yielded less robust results.