Jump to content
Sign in to follow this  
Riccardo

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

Recommended Posts

Basically every single Winter thats had an OPI -2 or less since OPI records began has been cold, you can't get a better correlation than that!

Edited by Barry95

Share this post


Link to post
Share on other sites

how many have been -2 or below? Can anyone post the years please?

 

 2009 John.

 

An admittedly small sample field but  a 100% correlation none the less. LOL

Edited by mcweather

Share this post


Link to post
Share on other sites

how many have been -2 or below? Can anyone post the years please?

 Just 1, -3.15 in 2009. BFTV posted a list on this thread early on.

 

Here we are:

 

https://forum.netweather.tv/topic/81494-october-pattern-index-opi-monitoring-winter-season-2014-2015/page-4#entry3051011

Edited by Yarmy

Share this post


Link to post
Share on other sites

I read that chart wrong, theirs only been 2 OPI's below -2, all OPI's below -1.5 have had cold winters though I believe.

Edited by Barry95

Share this post


Link to post
Share on other sites

I hate to pour water on the enthusiasm for the OPI, at this stage before any peer related test but to expect the small sample available to be sure of a 'cold' winter is asking for jam on it really.

The reality is that, IF other factors are favourable, then OPI values of -1.5 or lower do 'seem' to give a better chance of cold winters.

Factual is, out of 14 winters that rate cold for at least 2 out of the 3 winter months (December to February) 6 were cold. OPI did not indicate 5 out of 14 including the very cold one of 1981/82 (mean=2.6C).

I do admit to the theory being very interesting but further work does need to be done as those involved with it readily admit. Of course if this October, probably around -2 or a little below turns up a cold winter when, at the moment, none of the professional models, suggest anything like this, that really will make folk sit up!

Share this post


Link to post
Share on other sites

I hate to pour water on the enthusiasm for the OPI, at this stage before any peer related test but to expect the small sample available to be sure of a 'cold' winter is asking for jam on it really.

The reality is that, IF other factors are favourable, then OPI values of -1.5 or lower do 'seem' to give a better chance of cold winters.

Factual is, out of 14 winters that rate cold for at least 2 out of the 3 winter months (December to February) 6 were cold. OPI did not indicate 5 out of 14 including the very cold one of 1981/82 (mean=2.6C).

I do admit to the theory being very interesting but further work does need to be done as those involved with it readily admit. Of course if this October, probably around -2 or a little below turns up a cold winter when, at the moment, none of the professional models, suggest anything like this, that really will make folk sit up!

 

I prefer to think of it as an analysis tool in that it is measuring the state of the atmosphere ie. what the past/current indicators have actually created.

 

The OPI posits these indicators are predisposed to an overall negative Arctic Oscillation (based on small sample size) but that does not mean that it will produce cold in the UK. As I've mentioned earlier, SSWs will make a difference to how winter pans out and whether we are the beneficiaries of the cold.

 

I doubt the OPI was intended to be sufficiently specific as to be useful for such a tiny area as CET covers.

Share this post


Link to post
Share on other sites

Of course if this October, probably around -2 or a little below turns up a cold winter when, at the moment, none of the professional models, suggest anything like this, that really will make folk sit up!

 

I think this is what makes the subject so intriguing.  As you say, the traditional models are not pointing at anything notable, so if we do get a severe winter the OPI's credibility will certainly be massively enhanced.  

Share this post


Link to post
Share on other sites

I hate to pour water on the enthusiasm for the OPI, at this stage before any peer related test but to expect the small sample available to be sure of a 'cold' winter is asking for jam on it really.

The reality is that, IF other factors are favourable, then OPI values of -1.5 or lower do 'seem' to give a better chance of cold winters.

Factual is, out of 14 winters that rate cold for at least 2 out of the 3 winter months (December to February) 6 were cold. OPI did not indicate 5 out of 14 including the very cold one of 1981/82 (mean=2.6C).

I do admit to the theory being very interesting but further work does need to be done as those involved with it readily admit. Of course if this October, probably around -2 or a little below turns up a cold winter when, at the moment, none of the professional models, suggest anything like this, that really will make folk sit up!

And we all know how accurate the professional models are! Lol

At the moment the limited sample provided by the OPI is the best we have and I can not remember any autumn in the last 6 years where the professional models forecasted a cold winter despite 4 out of the 6 being cold or very cold!

These same models did a crap job predicting the good UK summer this year and last winters storm loaded washout.

I am not a betting man but if I was my money would go with the OPI than the dubious model runs compiled by the professionals.

Andy

Share this post


Link to post
Share on other sites

I can not remember any autumn in the last 6 years where the professional models forecasted a cold winter despite 4 out of the 6 being cold or very cold!

Where were you last Autumn? All I kept reading about in Sep/Oct, was the longer range models going for a cold Winter.

Share this post


Link to post
Share on other sites

I hate to pour water on the enthusiasm for the OPI, at this stage before any peer related test but to expect the small sample available to be sure of a 'cold' winter is asking for jam on it really.

The reality is that, IF other factors are favourable, then OPI values of -1.5 or lower do 'seem' to give a better chance of cold winters.

Factual is, out of 14 winters that rate cold for at least 2 out of the 3 winter months (December to February) 6 were cold. OPI did not indicate 5 out of 14 including the very cold one of 1981/82 (mean=2.6C).

I do admit to the theory being very interesting but further work does need to be done as those involved with it readily admit. Of course if this October, probably around -2 or a little below turns up a cold winter when, at the moment, none of the professional models, suggest anything like this, that really will make folk sit up!

I think though john, that the OPI may correlate better with colder winters for Western Europe as a whole, rather than just the UK.

Its worth noting that in many peoples' minds, a week long spell of heavy snow makes a 'cold winter'. The rest of the 3 months could be mild mush and be easily forgotten!

Share this post


Link to post
Share on other sites

An Analysis of the October Pattern Index vs Central England Temperature

 

another fascinating piece of research-thank you

I think though john, that the OPI may correlate better with colder winters for Western Europe as a whole, rather than just the UK.

Its worth noting that in many peoples' minds, a week long spell of heavy snow makes a 'cold winter'. The rest of the 3 months could be mild mush and be easily forgotten!

that is probably very true, 5-10 days of severe cold and dislocating snowfall for large parts of lowland Britain would certainly tend to be remembered by most of us, even if our bit of the world fared less well in terms of snowfall.

Share this post


Link to post
Share on other sites

another fascinating piece of research-thank you

 

Thanks, John.

 

Of course, what I didn't mention, but maybe I should do: is that the OPI January prediction rules out approximately the 50% warmest January CET values during the time frame measured. Which is something, I suppose, for snow lovers. Furthermore, there is a 50/50 chance of the January CET recording a score, according to this, of the lowest 4 scores of the period analysed.

 

November and December are simply not predicted by the OPI whatsoever, the weighting from January is so strong that results for the winter are skewed so far right to make it useless ie it is January's correlation that skews the winter's average, which is why a lot of people have commented about the winter correlation. Unfortunately, not true.

Edited by Sparkicle

Share this post


Link to post
Share on other sites

Sparkicle

 

Try Easte QBO years & Years below 1.5 - OPI  V CET months below average

Edited by Steve Murr

Share this post


Link to post
Share on other sites

Thanks, John.

 

Of course, what I didn't mention, but maybe I should do: is that the OPI January prediction rules out approximately the 50% warmest January CET values during the time frame measured. Which is something, I suppose, for snow lovers. Furthermore, there is a 50/50 chance of the January CET recording a score, according to this, of the lowest 4 scores of the period analysed.

 

November and December are simply not predicted by the OPI whatsoever, the weighting from January is so strong that results for the winter are skewed so far right to make it useless ie it is January's correlation that skews the winter's average, which is why a lot of people have commented about the winter correlation. Unfortunately, not true.

 

I think, as has been previously discussed, it may be beneficial to eliminate OPI neutral winters and play around with neutral at 0-.5 to +0.5, -0.75 to +0.75, -1.0 to + 1.0 and so on and see if correlation improves as we stretch the scope of  neutral and increasingly filter the data. If the OPI has any effect on CET then would expect correlation to improve as we proceed. 

Edited by Mucka

Share this post


Link to post
Share on other sites

Thanks, John.

 

Of course, what I didn't mention, but maybe I should do: is that the OPI January prediction rules out approximately the 50% warmest January CET values during the time frame measured. Which is something, I suppose, for snow lovers. Furthermore, there is a 50/50 chance of the January CET recording a score, according to this, of the lowest 4 scores of the period analysed.

 

November and December are simply not predicted by the OPI whatsoever, the weighting from January is so strong that results for the winter are skewed so far right to make it useless ie it is January's correlation that skews the winter's average, which is why a lot of people have commented about the winter correlation. Unfortunately, not true.

 

Interesting analysis. Is the winter mean NDJ or DJF? And what about February? When I looked at the Scottish data the correlation seemed to be strongest for January and reasonably strong for February too. Looking at the pattern for Novembers following -ve OPI the Atlantic-south Greenland mean trough pretty much precludes a decisively below average 'mean' November, although it is rather a good pattern for wave breaking.

Share this post


Link to post
Share on other sites

Sparkicle

 

Try Easte QBO years & Years below 1.5 - OPI  V CET months below average

 

If you got the data, I can do the analysis.

I think, as has been previously discussed, it may be beneficial to eliminate OPI neutral winters and play around with neutral at 0-.5 to +0.5, -0.75 to +0.75, -1.0 to + 1.0 and so on and see if correlation improves as we stretch the scope of  neutral and increasingly filter the data to shoe higher anomalies. If the OPI has any effect on CET then would expect correlation to improve as we proceed. 

 

It's unlikely we will see step change behaviour, although it is possible.

Interesting analysis. Is the winter mean NDJ or DJF? And what about February? When I looked at the Scottish data the correlation seemed to be strongest for January and reasonably strong for February too. Looking at the pattern for Novembers following -ve OPI the Atlantic-south Greenland mean trough pretty much precludes a decisively below average 'mean' November, although it is rather a good pattern for wave breaking.

 

NDJ. February isn't winter. It could well be that the restrictions of the UK definition of winter holds up analysis here.

Share this post


Link to post
Share on other sites

 Looking at the pattern for Novembers following -ve OPI the Atlantic-south Greenland mean trough pretty much precludes a decisively below average 'mean' November

 

My analysis shows that this isn't true. There is no correlation, either way, from the OPI to Nov CET.

Share this post


Link to post
Share on other sites

I think, as has been previously discussed, it may be beneficial to eliminate OPI neutral winters and play around with neutral at 0-.5 to +0.5, -0.75 to +0.75, -1.0 to + 1.0 and so on and see if correlation improves as we stretch the scope of  neutral and increasingly filter the data. If the OPI has any effect on CET then would expect correlation to improve as we proceed. 

 

Isn't there the danger of further back engineering of a product that is already based on hindcasting. We could probably filter to give something that looked plausible but would it be of value as a forecast tool?

Share this post


Link to post
Share on other sites

Isn't there the danger of further back engineering of a product that is already based on hindcasting. We could probably filter to give something that looked plausible but would it be of value as a forecast tool?

 

I am not data analyst but I would of thought filtering out a possible neutral value OPI would be valid method. After all if there is anything to the OPI it stands to reason that stronger positive or negative results would give stronger correlation and therefore we are simply filtering out possible noise. 

Edited by Mucka

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×
×
  • Create New...