Jump to content
Sign in to follow this  
Riccardo

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

Recommended Posts

Apparently, in the graph showing the correlation, the years with the weakest correlation are those where there is a significant change in the ENSO state.  So this would appear to be something worth keeping an eye on over the next month or two.  As far as I'm aware it's not expected to change much in this timeframe from the very weak state?

Share this post


Link to post
Share on other sites

This is a fascinating subject matter but surely is it not just one piece of the jigsaw that is needed for HLB, I tend to look at the SST around Greenland and the Pacific coast of Alsaka for an idea of what we may expect over the coming winter months?

Edited by Hocus Pocus

Share this post


Link to post
Share on other sites

Perhaps a read of Riccardo's forecast for the US, last November, will help clarify some points being raised. I hope it is OK to reproduce the 500mb chart from that forecast as it was so stunningly accurate.

 

http://www.americanwx.com/bb/index.php/topic/41639-seasonal-winter-20132014-forecast-based-on-opi-index/

 

 

 

A strong -ve EPO pattern was a persistent feature for much of last winter so the forecast did appear to be accurate in this respect. However, the AO forecast which is touted as the main ability of the OPI was well out, unfortunate for its first public showing. Without seeing a little more how it has been calculated, it is hard to judge the OPI-AO correlation seen in the graphs. But taken at face value, two winters with similar average AO indices and thus supposedly OPI, could be very different in character as to when and how strong the AO index is which begs the question how good is OPI for making more than a vague forecast and how does it relate to the physical processes involved?

Share this post


Link to post
Share on other sites

Riccardo commented on this in terms of how the index is poor once in 15 years relating to the alignment of the vortex -

 

*IF* that graph is correct & there was no bias to producing a OPI result close to the AO index ( bearing in mind all the historic information was available) then even some error once in 15 isn't a problem.

 

Also the correlation to OPI less than -1.5 to -AO winters covering 3 months was 1 !!!!

 

I think we may be about to test that theory this winter!!

 

As I mentioned though there is a measure to check at the closing stages of November to validate the following Winters AO index, which is a real time measure.

S

Edited by Steve Murr

Share this post


Link to post
Share on other sites

A strong -ve EPO pattern was a persistent feature for much of last winter so the forecast did appear to be accurate in this respect. However, the AO forecast which is touted as the main ability of the OPI was well out, unfortunate for its first public showing. Without seeing a little more how it has been calculated, it is hard to judge the OPI-AO correlation seen in the graphs. But taken at face value, two winters with similar average AO indices and thus supposedly OPI, could be very different in character as to when and how strong the AO index is which begs the question how good is OPI for making more than a vague forecast and how does it relate to the physical processes involved?

The comment was related solely to the forecast for N. America. A negative AO can happen due to a variety of vortex placements/alignments and it is rare for the pattern to be fixed for all of the winter months - it is this that I have difficulty with when applying a 'one size fits all' forecast in the late autumn.

 

I see similar very positive SSTs in the NE Pacific and a continuation of positive PNA pattern across the US - if it is such a good indicator, does this suggest more of the same for this winter? Will the QBO negative be sufficient to change the outlook and allow more chance of wave penetration to disrupt any vortex via SSW? Already we see a tendency to an arctic dipole anomaly - temporary while the ice losses re-establish or indicative?

 

So many questions and variables - it cannot just hinge on one aspect alone.

Share this post


Link to post
Share on other sites

 

 

So many questions and variables - it cannot just hinge on one aspect alone.

 

very true G_F

I truly doubt that it will be the answer to the coldies hopes or indeed the end of their hopes. There are just too many questions about this. No one has yet shown a 'null' test hypothesis on it. Its' fine showing how accurate it seems for predicting the -AO for the NORTHERN hemisphere but as last winter showed parts can be in the freezer and other parts the opposite. Even over that until I see solid statistical evidence to back it up I remain rather dubious about its use, certainly on its own. Used CAREFULLY in conjuction with other teleconnections then I would think it another tool to give some reasonable guidance.

Share this post


Link to post
Share on other sites

Totally agree John, the O.P.I may well be a good indicator of the coming Winters A.O index but we must be careful not to assume that a negative A.O automatically means a colder Winter for the UK, many other factors involved but to be fair to Riccardo, I don't think he's suggesting that.

Thanks to Chiono for explaining the O.P.I in a little more detail, in terms of European,height anomalies with axis of N.H Rossby Waves factored into the equation as well.

Regards,

Tom.

Share this post


Link to post
Share on other sites

very true G_F

I truly doubt that it will be the answer to the coldies hopes or indeed the end of their hopes. There are just too many questions about this. No one has yet shown a 'null' test hypothesis on it. Its' fine showing how accurate it seems for predicting the -AO for the NORTHERN hemisphere but as last winter showed parts can be in the freezer and other parts the opposite. Even over that until I see solid statistical evidence to back it up I remain rather dubious about its use, certainly on its own. Used CAREFULLY in conjuction with other teleconnections then I would think it another tool to give some reasonable guidance.

 

To be fair, I think that's pretty much all that Steve, and certainly all that Riccardo, was saying - it's a very useful tool for calculating the AO, particularly in conjunction with the QBO, and a very negative OPI and negative/East based QBO is/would be a very good indicator of a negative AO and, hemispherically at least, a more blocked pattern. The analogue years Steve put up for a very negative OPI with east QBO are also particularly encouraging for those of us who are looking for a colder than average winter in our locality. Nothing is guaranteed of course, and there are numerous ways to skin a negative AO, but the early indications are far more promising for a blocked NH pattern than at this stage last year.

 

Share this post


Link to post
Share on other sites

 

To be fair, I think that's pretty much all that Steve, and certainly all that Riccardo, was saying - it's a very useful tool for calculating the AO, particularly in conjunction with the QBO, and a very negative OPI and negative/East based QBO is/would be a very good indicator of a negative AO and, hemispherically at least, a more blocked pattern. The analogue years Steve put up for a very negative OPI with east QBO are also particularly encouraging for those of us who are looking for a colder than average winter in our locality. Nothing is guaranteed of course, and there are numerous ways to skin a negative AO, but the early indications are far more promising for a blocked NH pattern than at this stage last year.

 

 

 

Not just the AO, Steve illustrated that with the east based QBO, then the NAO is more likely to follow the AO - I think I saw an 80% chance of a negative NAO - if the OPI is low enough - if you have an east QBO?

 

So unless the QBO suddenly reverses, a strongly negative OPI (if we get that) implies a very good chance of a negative NAO.  Of course it is still possible to be mild even with a negative NAO - but that is very unlikely when combined with a negative AO at the same time.

Share this post


Link to post
Share on other sites

Ive presented the statistics about 3 times now....

 

the only concern with them is the short sample size!!!

 

I will write it again though...

 

For OPI index years sub -1.5 & Strong easterly QBOS theres almost a full on correlation of 1 for the AO to be ALL negative DJF for the follwing winter.

 

I will also break that down in my preliminary round up into 2 groups

Junior AO's- which had indexes up to -1.5

Super AO's- which had indexes below -1.5 -

 

This was correlated last night in my post to the NAO - which again in Easterly QBO years is ALMOST 1 as well for negative Months.

 

The angle of this years vortex is driving the results on the link as much as the HLBs we currently have.

In terms of this pattern we have another 4/5 days of this so the first third of the month looks like being in around -2 index..

 

S

Share this post


Link to post
Share on other sites

Ive presented the statistics about 3 times now....

 

the only concern with them is the short sample size!!!

 

 

 

S

I think that the hypothesis does need to be tested independently as well, Steve. It looks a great piece of work to me as well, but....

 

 I could present a hypothesis and show good correlation that every time I put my foot on the accelerator of my car that it moves faster - and present a graph to this effect. Then someone else may come along and press down the clutch pedal (or brake pedal) whilst also pressing the accelerator and all of a sudden my hypothesis has a major flaw. I would say that my hypothesis stood up every time I tested it and that it has 100% correlation - when in real terms I just had excluded the clutch pedal from my hypothesis because it wasn't what I was testing. In the same way the OPI would need the same scrutinisation, and when it passes that test with flying colours then hey presto. To me, last year was the first test and there was a little bit of 'clutch pressing' present that skewed the correlation.

Share this post


Link to post
Share on other sites

I think that the hypothesis does need to be tested independently as well, Steve. It looks a great piece of work to me as well, but....

 

 I could present a hypothesis and show good correlation that every time I put my foot on the accelerator of my car that it moves faster - and present a graph to this effect. Then someone else may come along and press down the clutch pedal (or brake pedal) whilst also pressing the accelerator and all of a sudden my hypothesis has a major flaw. I would say that my hypothesis stood up every time I tested it and that it has 100% correlation - when in real terms I just had excluded the clutch pedal from my hypothesis because it wasn't what I was testing. In the same way the OPI would need the same scrutinisation, and when it passes that test with flying colours then hey presto. To me, last year was the first test and there was a little bit of 'clutch pressing' present that skewed the correlation.

 

 

I especially think the OPI index appears more in the first part of winter, from mid-November to mid-January. Beyond this area, the trend OPI October away from the second half of January, the brightness begins to point the nose of the northern hemisphere so the waves waves tend to increase with time, be accentuated in February / March. So I think the IPO is very important to predict the capacity of VP on the first part of winter when there is no sunshine on the Arctic and not the entire winter.

 

Now, if we take the average of the AO on the first part of winter and the strength and the inclination of the VPS, we note that the prediction made ​​by the OPI index was very good

Share this post


Link to post
Share on other sites

I especially think the OPI index appears more in the first part of winter, from mid-November to mid-January. Beyond this area, the trend OPI October away from the second half of January, the brightness begins to point the nose of the northern hemisphere so the waves waves tend to increase with time, be accentuated in February / March. So I think the IPO is very important to predict the capacity of VP on the first part of winter when there is no sunshine on the Arctic and not the entire winter.

 

Now, if we take the average of the AO on the first part of winter and the strength and the inclination of the VPS, we note that the prediction made ​​by the OPI index was very good

That may be the case but it is not the hypothesis though - more a case of fitting the data to suit after the event.

Share this post


Link to post
Share on other sites

Moreover, it seemed that Riccardo had said that the second part of winter would be different for the reasons I mentioned. After the VP was so compact in the first part of winter that waves 1 and 2 propagating at regular intervals could not cause strong changes in atmospheric circulation during the second part of the winter 2013/2014

Share this post


Link to post
Share on other sites

That may be the case but it is not the hypothesis though - more a case of fitting the data to suit after the event.

 

Indeed I did make that comment earlier as well-

 

However the hypothesis did look well presented last year. & certainly makes sense.

remember though the November information though that will support this winters forecast will land irrelevent to the OPI -

 

As it stands though from the standpoint of today you would expect a slam dunk - AO/ -NAO winter with at least 2 Cold months.

Share this post


Link to post
Share on other sites

The correlation with eQBO in Ens Neutral is remarkable, the fly in the ointment for me is that having followed last year forecast closely it is difficult not to feel sceptical. Perhaps this is unfounded, and it was just an unlucky year for OPI to come to the fore.

 

Maybe some of the doubts are grounded in not seeing full re-analysis, nor seeing specifically how the telemappa software creates it's axis, or calculates likely wave intrusions on the vortex. 

 

Knowing the above would perhaps help explain why, as Riccardo put it, the OPI predicted value busts once every 14/15 years, I do appreciate though that this will be subject of the official release paper. It's intriguing to guess away at the technical elements involved.

 

If the figures ring true and the eQBO correlation, all other variable excluded, then this Winter presents the OPI a great opportunity to get back on track.

 

Good to see this talked about more this year and generating interest. Perhaps we can solve the clues before the paper is released !

 

 

 

 

 

 

 

Share this post


Link to post
Share on other sites

It's a case of Princess Leia(cold lover), help me OPI Kenobe you're my only hope! versus the hideous Darth Vader( Bartlett High)!

 

I fear for the emotional state of this thread if the OPI implodes after a great start. However I think we should take some comfort from the fact that we're very unlikely to see last years horror show repeated.

 

Last winter was really facing an uphill struggle from the outset because so many factors were against cold, so its safe to assume it will be a more interesting winter.

Share this post


Link to post
Share on other sites

If the OPI carries on this way - south of -2 then we will be left with very little to forecast at the end of October-

 

There will only be 3 Analogues

 

62 / 63

76 / 77

09 / 10

 

... What a choice.

 

Interestingly the lowest OPI on record for October was 2009 - Also 2009 has the lowest AO on record @ -1.5

If we look at the AO ensembles the first half of the months landing zone looks to be around -2....

Edited by Steve Murr

Share this post


Link to post
Share on other sites

I'm watching this thread from afar but it is already permanently a tab open on my phone to check each day and will continue to be with great interest. The reason I am generally holding fire for now is that I am all too aware of how much of a tease following a month-long index can be after a few years of SAI watching.

 

May I say as ever some really brilliant analysis from the usual suspects - Steve in particular has really produced some top notch insight and some incredibly useful data.

 

I would still urge caution for now, however for anyone wondering why the current reading shows such a negative index - taking in to account the fact that the OPI website takes forecast conditions for out to 10 days ahead too - it is worth comparing the current 500mb Anomaly forecast from the GFS side-by-side with the anomaly chart Steve provided with his analysis of October 2009:

 

post-1235-0-36591100-1412629695.png

 

If we then take a look at the 8 and 10 day forecasts from the 12z:

 

gfsnh-12-192.png?12gfsnh-12-240.png?12

 

Note in particular the very strong anomalies present around The Aleutians and stretching in to the Pole in all three charts and you find your reasoning for the "current" state of the OPI.

 

However caution should be urged as these are still forecast charts. That becomes even more apparent when we take a look at the ECMWF forecast for day 10:

 

ECH101-240.GIF?07-0

 

Whilst we may not have the strong anomaly suggestion in the same zone, we do still generally have positive height anomalies around that 65N mark, but the main point here is to highlight that much of this is still based around forecasts stretching out in to an unreliable timeframe. Something which may throw some forecasts out a little more in terms of the specific detail beyond a few days is a renewed MJO wave:

 

ALL_emean_phase_full.gif

 

Something worth bearing in mind when following the twists and turns of our new found favourite winter tool.

 

All in all though a very intriguing watch this year and I must extend my thanks to Ricardo and the team for the monitoring tools. This thread will make fascinating reading in the weeks to come I am sure.

 

SK

 

I could be missing the point....so your saying the current OPI readings are not the actual readings, just a forecast that can change? So that chart with the - figures could just change to what actually happens and we could be just led up the garden path?

Share this post


Link to post
Share on other sites

The rolling average can be worked out on the link at the bottom by taking the mean average of all the days ( 1-7 )

the forecast is days 1-7 + 10 Day GFS geopotential heights in the areas. ( which also includes a calculation in relation to the angle of the vortex) so the top figure is the current rolling average + the forecast period 'averaged'

 

S

Share this post


Link to post
Share on other sites

Is there an actual index list out there, with the numerical OPI values for each year? I've made my own just using the graph of the OPI (and the correlation with the winter AO is 0.91, as expected) but the added precision of the actual index would be handy!

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...