Jump to content
Sign in to follow this  
Riccardo

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

Recommended Posts

Is that a daily value or is it a cummulative value (i.e. first 6 days+10 days)?

 

2.10 is the daily value, scroll down and underneath is a bar graph for the whole month.

 

Great news but I am gutted you got there before me!  :wink:

 

Karyo

 

Yes, I need to read up on it but does seem a good correlation between it and the winter AO, as per Steve's excellent posts and others up top of the thread.

Share this post


Link to post
Share on other sites

Exactly mulzy.

I for one am now a little confused (easily achieved nowadays!) as to what the “Daily Value†of the OPI shown on Riccardo’s web link actually represents. In my post (above) I assumed the value shown each day was the running monthly average up to that day, based on the model taking all previous daily “actuals†for the month and adding 10 forecast days. My assumption being based on what Riccardo posted at the start of the thread:

 

“The software analyzes the previous consolidated daily charts plus 10 forecast charts as given by the GFS model. For example on the 15th of October, the software calculates the OPI index based on the 15 consolidated daily charts (1-15 October) adding 10 forecast charts as given by the GFS model (16-25 October).â€

 

Additionally, if you look below the headline Index value at the box called “Calculation Parameters†it says “OPI Calculate at 16/10/14†which looks like it supports the above methodology (i.e. 6 actual days plus 10 forecast days = 16h).

 

Steve (above) does not appear to have interpreted the “Daily Value†in the same way and has taken each days Index and worked out the average himself.

 

It would be good if someone (Riccardo himself?) could clarify this please. Many thanks.

Edited by Blessed Weather

Share this post


Link to post
Share on other sites

Exactly mulzy.

I for one am now a little confused (easily achieved nowadays!) as to what the “Daily Value†of the OPI shown on Riccardo’s web link actually represents. In my post (above) I assumed the value shown each day was the running monthly average up to that day, based on the model taking all previous daily “actuals†for the month and adding 10 forecast days. My assumption being based on what Riccardo posted at the start of the thread:

 

“The software analyzes the previous consolidated daily charts plus 10 forecast charts as given by the GFS model. For example on the 15th of October, the software calculates the OPI index based on the 15 consolidated daily charts (1-15 October) adding 10 forecast charts as given by the GFS model (16-25 October).â€

 

Additionally, if you look below the headline Index value at the box called “Calculation Parameters†it says “OPI Calculate at 16/10/14†which looks like it supports the above methodology (i.e. 6 actual days plus 10 forecast days = 16h).

 

Steve (above) does not appear to have interpreted the “Daily Value†in the same way and has taken each days Index and worked out the average himself.

 

It would be good if someone (Riccardo himself?) could clarify this please. Many thanks.

Yes the way you understand it is how i interpreted it BW.

 

Btw folks to avoid any confusion my post above does not detract from the impressive link between the Oct OPI and the Winter AO following.

Knowing that a -AO can sometimes not work in our favour,ie we could have one that's too west or east based and end up on the wrong side of any blocking i thought a study related to our own Winter CET and the OPI was more useful for us.

 

As it turns out nothing proved either way except that recent cold UK Winters have been recorded from both - and +ve OPI indexes. 

In fact based on my findings we could still have a slightly positive OPI and get a colder Winter,so no reason to pin everything on this. 

That really was my point ,maybe i should have made this clearer in my earlier post.

Edited by phil nw.

Share this post


Link to post
Share on other sites

It calculates a value for the whole of October from daily scores, so to start is mostly from forecast data. The eventual value becomes closer from the middle of October onwards and the actual value is only known at the end of the month.

Edited by Interitus

Share this post


Link to post
Share on other sites

Just like to say thanks to Riccardo for sharing his work with us, e benvenuto a NW, it will be very interesting to see how the rest of October pans out.

 

Very interesting post from Phil a little earlier and the linking of the October AO index and Winter CETs. Phil, I see you've just hinted at the point I'm about to make.

We have to be cautious about putting all our eggs in one basket re. a Winter -AO index. It may surprise some members that the Winter of 2013/2014 came out just on the positive side of neutral, + 0.18 by my calculations, indeed Jans figure was -0.969. As we know, from around mid December a very active mobile Atlantic pattern set in and delivered an endless conveyor belt of deep lows to our north-western doorstep, which continued virtually unabated into Jan and Feb. 

This very dominant mobile pattern came on the back of a very powerful jet-stream fuelled by bitter cold blasting down over central parts of Canada on bitter N/NW winds exiting out of the E.Seaboard over relatively mild Atlantic SSTs, resulting in explosive cyclogenesis and those deep bombing lows, rushing towards the UK.

As we now know, a very strong, compact Polar Vortex settled over E.Canada/N.E. US, as Winter progressed and the pattern became self perpetuating and continued into late Winter. But as important as all these elements, if not more so, was the part that a very strong high pressure cell in the North Pacific played in the synoptic pattern for large parts of the NH, last winter. It persisted for long periods of the season and from time to time threw strong ridges high into the Arctic, via the Aleutians and Alaska, which in turn helped lock in the PV, over E.Canada/N.E US. To our NE, helping that pattern persist, positive heights were to be found over NW Russia but unfortunately with such a powerful jet-stream, aided by the position of the PV, any blocking over Scandinavia was never able to take hold. At some point in January cold and snow got as far west as the Dutch/German border, briefly.

 

The extremely laboured point I'm making is that purely taking the AO index in isolation, as an indicator of a certain synoptic type in Winter, conducive to delivering cold and snow to NW Europe, needs to be treated with caution, as Phil quite rightly mentions above. As we know the AO index measures height anomalies over the Arctic and last Winters values were neutral because those higher heights recorded over the Arctic, adjacent to the N.Pacific, negated those lower height anomalies recorded over the Arctic, adjacent to the N.Atlantic. It's worth mentioning though that when the AO goes deeply negative, we stand much more chance of developing a synoptic pattern that will deliver wintry weather to our shores.

 

Lets hope then this Winter that the area affectionately referred to as the GIN (Greenland, Iceland, Norway) corridor on NW, sees plenty of yellow and orange on the 500 hpa charts and not those disgusting purples and blues, its certainly the type of -AO I like to see.

 

Just a quick point to Riccardo, as has been stated above, could you please explain the OPI in a little more detail, I assume it's a height anomaly index for Europe but how is that calculated exactly.

Also would you have the OPI figure for Oct.1978, from your historical reanalysis? As we know, the Uk and other parts of Europe, experienced a cold and snowy Winter, in part.

 

Much thanks,

Tom.

Edited by Kentish Blues Man

Share this post


Link to post
Share on other sites

Exactly mulzy.

I for one am now a little confused (easily achieved nowadays!) as to what the “Daily Value†of the OPI shown on Riccardo’s web link actually represents. In my post (above) I assumed the value shown each day was the running monthly average up to that day, based on the model taking all previous daily “actuals†for the month and adding 10 forecast days. My assumption being based on what Riccardo posted at the start of the thread:

 

“The software analyzes the previous consolidated daily charts plus 10 forecast charts as given by the GFS model. For example on the 15th of October, the software calculates the OPI index based on the 15 consolidated daily charts (1-15 October) adding 10 forecast charts as given by the GFS model (16-25 October).â€

 

Additionally, if you look below the headline Index value at the box called “Calculation Parameters†it says “OPI Calculate at 16/10/14†which looks like it supports the above methodology (i.e. 6 actual days plus 10 forecast days = 16h).

 

Steve (above) does not appear to have interpreted the “Daily Value†in the same way and has taken each days Index and worked out the average himself.

 

It would be good if someone (Riccardo himself?) could clarify this please. Many thanks.

From Riccardo ( OLD POST )

 

Finally, I want to point out one thing that I give will create confusion in many people.

Until October 31, the date for which will be released on the final value of the index, the program takes into meal cards issued by gfs forecast (for example, October 25, will be taken 25 cards + 6 to the consolidated forecast). So it is clear that the more we advance, the more the value will become reliable as it will tend to get progressively closer to the final value. To date, Stano still 8th of the month, we only have 8 cards consolidated more than 10 forecast (we stop at Gfs 240 hours), so we are still in a very early stage and therefore very little concrete. For this reason, in the next few days you can also see the very radical changes of the index. For this reason we decided to color the gray bars: only when the index will begin to stabilize better (more or less after passing the first half), bars assume color: those upwards of red and those down blue (according to the sign of the AO duqnue OPI and winter).

 

 

So if you want accuracy then just go with the rolling mean as the overall index has GFS data FORECAST included.

Edited by Steve Murr

Share this post


Link to post
Share on other sites

I don't seem to have got my head around this yet, so am I right in thinking that it wont really become reliable until late in month, due to it incorporating more GFS FI data than verified data?

Share this post


Link to post
Share on other sites

PS heres why the October AO comparison & the NAO negative comparison ( that was discussed last night ) isn't the nest 'fit' to the OPI

 

the OPI index works on 2 parameters of which I cant fully explain yet

 

The Geopotential Anomaly across 3 Areas

* Polar Field

* Eurasia

* American Sector

( Sadly I cannot see what specific co-ordinates are covered)

 

However the measure is how opposed these areas are ( in terms of geopotential ) they are to the Norm- so for example the pole & the other 2 would be have to have very strong Positive heights to create a negative OPI anomaly..

THe index will become more negative as the 3 areas become strongly opposed to the Norm.

You would image the Lattitude is close to around 65-70N

 

Theres is also then a trigonometric function in relation to the alignment of the vortex relating to its Orthogonally to the Imaginary line that you would draw along an Axis of the line joining the 2 major oceans,  which in lamens terms means the alignment of the Vortex is usually West > East, the index will be more Negative when the vortex alignment trends towards North South - instead of West East.

 

Please see the Long term Geopotential Average for October from 1948-2014

post-1235-0-80866600-1412629430_thumb.pn

 

 

If we look at the October geopotential anomaly @ 500 MB so far this October we see-

post-1235-0-29811200-1412629570_thumb.gi

 

This shows the index being skewed towards Negative by all the PLUS pressure anomalies over Eurasia, US & the pole.

 

Heres the NH plot for 2009 where the index was -3.4

post-1235-0-36591100-1412629695_thumb.pn

 

You can see the very strong index over the pole ( inverse ) & the ring of positive anomalies across 65N in the chosen other 2 sectors...

 

 

If you look at the 10 day comparison ( forecast) from the Ewall PSU site - the negative trend ( strong blocking ) looks to contine

 

http://mp1.met.psu.edu/~fxg1/ECMWF_0z/hgtcomp.html

 

best regards

steve

 

 


I don't seem to have got my head around this yet, so am I right in thinking that it wont really become reliable until late in month, due to it incorporating more GFS FI data than verified data?

 

the forecast wont become reliable ( as in the final landing figure ) however the rolling figure will be 100% accurate, & as we progress through the month you will get a more robust rolling average as it covers more days.

 

The 5 day rolling average is at a stab now around -1.15

 

S

Edited by Steve Murr

Share this post


Link to post
Share on other sites

Steve as i understand it the OPI is simply another way of showing the NH 500hPa pattern AO for October  and Recarrdo's findings on the graph show  a .91 correlation to the subsequent Winter AO ?

In any event it doesn't really alter my presentation re measuring the coldest Winters CET against the preceding October.

It's an interesting discussion and i appreciate the work you have also done which i see you widened to iclude the NAO.

 

Kentish Blues Man in a very good post above explains the possible pitfalls of concentrating on just the AO for the Winter quarter and  underlines more fully why i took the CET series rather than the Winter indexes to compare with the preceding Octobers.

 

In any event it does no harm to come at this from different perspectives,it's the way we learn more and hey this isn't an excercise to show who is right or wrong.

Edited by phil nw.

Share this post


Link to post
Share on other sites

Great thread. Been looking forward the impending annual opening of the Strat thread and this is a very welcome bonus!

I have to say whilst I knew about the correlation between the OPI and the following DJF AO, I am truly staggered by the graph above which sheds no doubt on the connection.

Share this post


Link to post
Share on other sites

Does anybody know if the UKMO actually looks at this OPI? Have they used this before when they issued their previous seasonal forecasts?  I think a more interesting correlation would be between this and the CET series.

 

On the face of it a 0.9 correlation is very good in terms of AO so lets at least hope this October delivers some hope for the winter especially after last winters feeble attempt at cold.

Share this post


Link to post
Share on other sites

Hi Phil-

 

but see my post, it is however its focussed on 3 specific areas & in addition the alignment of the Vortex through a function - which isn't quite the same as the AO metric - although is probably supported by the AO in terms of having similarities.

 

S


Interestingly Steve, that October plot has a significant positive anomaly across the East Siberian/Alaskan arctic. It is here that Cohen hypothesises that we are more likely to see raised pressure due to the reduced amount of sea ice. And in turn, that this induces feedbacks later seen in winter that can lead to negative AO conditions (through changes in the storm tracks, jet stream changes and planetary waves inducing stratospheric change. (see bottom of page 7 here: http://web.mit.edu/jlcohen/www/papers/Cohenetal_NGeo14.pdf)

 

The OPI system not only measures the pressure differentials, but also takes into consideration the axis of the NH Rossby cross pole waves and it is these combined together that I believe Riccardo et al have encorporated into their index.

 

Interestingly, totally seperate from the OPI winter composites that you have been posting in this thread, there are other analogues that also suggest later winter Arctic blocking - more of that to come at a later date......

 

Yes - which I eluded to in my post to Phil & in my post around( my 'other' index for November ) :)  Which I would hazard a guess you may already know -

its not an official index but heres a clue-

the index for 1976 / 77 for November is +10. & its different to the QBO.

 

S

 

PS it also answers the questions of why we are seeing more extremes in the last 10 or so years....... I wonder if we can get a -5 AO month ...!!!

Edited by Steve Murr

Share this post


Link to post
Share on other sites

Does anybody know if the UKMO actually looks at this OPI? Have they used this before when they issued their previous seasonal forecasts?  I think a more interesting correlation would be between this and the CET series.

 

On the face of it a 0.9 correlation is very good in terms of AO so lets at least hope this October delivers some hope for the winter especially after last winters feeble attempt at cold.

The OPI has yet to be scutinised and vetted independently and submitted to journals - and until it is, it will not be taken on board by the met (rightly too). But luckily we can watch the hypothesis being played out in the meantime. It is one thing building an index to fit retrospective dates, but quite another to get this to work in the future - as last year showed.

Share this post


Link to post
Share on other sites

Does anybody know if the UKMO actually looks at this OPI? Have they used this before when they issued their previous seasonal forecasts?  I think a more interesting correlation would be between this and the CET series.

 

On the face of it a 0.9 correlation is very good in terms of AO so lets at least hope this October delivers some hope for the winter especially after last winters feeble attempt at cold.

Hi Nick,

See my post no 25 above where i have done that. :)

Share this post


Link to post
Share on other sites

Hi Phil-

 

but see my post, it is however its focussed on 3 specific areas & in addition the alignment of the Vortex through a function - which isn't quite the same as the AO metric - although is probably supported by the AO in terms of having similarities.

 

S

 

Yes - which I eluded to in my post to Phil & in my post around( my 'other' index for November ) :)  Which I would hazard a guess you may already know -

its not an official index but heres a clue-

the index for 1976 / 77 for November is +10. & its different to the QBO.

 

S

 

PS it also answers the questions of why we are seeing more extremes in the last 10 or so years....... I wonder if we can get a -5 AO month ...!!!

Interesting Steve so i wonder how they get the rolling forecast of their version of the index from the GFS as the month goes on?

Would that not simply be the AO index as i understand it?

Is there a link to a full explanation of the OPI as opposed to the basic October AO index ,if there is it would be beneficial to all. 

edit 

just had another re read of the first post and i think i have answered my own question.

They feed into their dedicated software those daily GFS  charts and the forecast ones for up to 10 days ahead and it works out from the 500hPa patterns their OPI.Indeed a basic -AO pattern would underpin a -ve OPI but i think i get the difference now.

Someone correct me if i have it wrong.

Edited by phil nw.

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/

 

7kss.png

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/

 

7kss.png

That is very good.  That Alaskan/W Canadian ridge anomaly I think was very much anticipated by the unusual warmth in the north pacific last year. 

 

BFTP

Edited by BLAST FROM THE PAST

Share this post


Link to post
Share on other sites

"Kentish Blues Man in a very good post above explains the possible pitfalls of concentrating on just the AO for the Winter quarter and  underlines more fully why i took the CET series rather than the Winter indexes to compare with the preceding Octobers."

 

Thanks Phil, exactly, fully understandable why you've cited Winter CET series, rather than Winter AO index.

 

Just as a west based -NAO is not conducive to delivering wintry synoptics to the UK, a -AO with positive height anomalies over the other side of the Arctic, i.e adjacent to the Pacific is not necessarily good news either for those of us who want to see cold and snow delivered to our shores.

 

No finer examples of two totally different -NAO months came back to back in Jan and Feb 1969, obviously resulting in totally different weather types for the UK.

 

500 hpa charts from those 2 months.

 

West based -NAO.

 

Rrea00119690109.gif

 

 

East based -NAO.

 

Rrea00119690208.gif

Share this post


Link to post
Share on other sites

I've been musing, why October? Will the 'NPI' (as in November) bring an even higher correlation to the DJF AO? By rights surely it should, shouldn't it? Being a month closer... I probably need to read Cohen again but can anybody give a simple answer? This is going to seem very unscientific but it does seem winter is 'moulded' in this month.

But then maybe that isn't unscientific as although, yes, over-simplistic, is it as relatively straightforward as the QBO? To my mind it has to be a.main player in all of this (with ENSO serving to muddy the waters). In its Easterly phase, aka winter 14/15, it will inhibit ( to one degree or another) the natural formation of the early winter vortex, this in itself leading to a more negative AO. Compare to a westerly QBO (as in the debacle that was winter 13/14) and you have in scenario in which the early winter vortex will be able to form more freely. This in itself leads to a SELF-PERPETUATING +AO pattern that will be near damn impossible to shift through the majority of the winter months. The slight fly in the ointment was the Jan 14 -AO output but I think it was anomalous as what holds true is that the vortex remained relatively compact. From memory, wave1 events served to displace the raging PV and allow the ridging over the pole seen that month.

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...