We are just a couple of weeks short of May when the MetO will be starting to work on their LRF for the winter 2008/9 NAO index. One of their key indicators will be the spatial pattern of N. Atlantic SST anomalies and as we approach May I thought it would be worth looking at the seasonal pattern emerging in the N. Atlantic to see if we can pre-empt the MetO's NAO call this year.

Here is the latest seasonal pattern:-

and this can be compared with the MetO's own predictor pattern for a +ive winter NAO based upon May SST anomalies.

Ummm.....some evidence of a negative image there

Game on...it's looking like a cold one for sure...remember you heard it here first

I know you are serious and mean well but, come on...

We are well overdue a cold and snowy winter 

Anyway on the converse side - we said that summers were always hot and sunny after 2006 - then 2007 comes along 

so anything is possible for winter

I think we will get a cold and snowy winter - but when ? 

Matty H:

I know you are serious and mean well but, come on...

http://www.metoffice.gov.uk/research/seasonal/regional/nao/index.html

your point being......

brogdale:

Matty H:

I know you are serious and mean well but, come on...

http://www.metoffice.gov.uk/research/seasonal/regional/nao/index.html

your point being......

Oh I give up

I recall Jonny got a bit of flack last year for doing something similar in May.

Out of interest Jonny, were there any conclusions reached from last years thread and, more importantly, what was the comparison with reality?

Sevendust:

I recall Jonny got a bit of flack last year for doing something similar in May.

Out of interest Jonny, were there any conclusions reached from last years thread and, more importantly, what was the comparison with reality?

Spotted 'Hawkeye'

You're quite right...I did post a similar thread last April trying to promote some discussion in advance of the  May SST pattern analysis and I think a good deal of the 'flak' you refer to came from the SW direction

The thread was titled something similar to this ("will MetO call cold winter?") and went on for some time...but when, in the Autumn, it became clear that the MetO were not calling a cold winter Martyn kindly put the thread out of its misery.

On a (semi) serious note..it is the case that the MetO do factor the preceeding May's N. Atlantic SST anomaly pattern into their NAO prediction, and that, in turn, is one element of their Winter seasonal forecast. I know there are a number of members on here who pay attention to SST patterns, and I thought it would be useful to raise the issue again. (No offence was intended Matty)

Edit: to answer your substantive point...I recall that last year's May SST pattern was a little inconclusive but appeared to favour the -ive NAO ie. cold side of things. As it happened the MetO's seasonal forecast was itself somewhat inconclusive but the crucial issue of Autumnal/Winter re-emergence of SST anomalies is a complex one.

Nonetheless one to watch...especially with the increasingly data-rich environment provided by the Argo array.

brogdale:

to answer your substantive point...I recall that last year's May SST pattern was a little inconclusive but appeared to favour the -ive NAO ie. cold side of things. As it happened the MetO's seasonal forecast was itself somewhat inconclusive but the crucial issue of Autumnal/Winter re-emergence of SST anomalies is a complex one.

Yes during the autumn we saw a la nina influenced pattern drive a big ridge up the East Coast of America (creating warmth there) - which put paid to any re-emergence of the spring sea temps.  It was in the Autumn that the MO actually adjusted their spring NAO forecast to more positive - and in hindsight they were very right to do so.

I think apart from the NAO signal it would be good to start the autumn without huge positive anomalies across much of the Atlantic (a feature of recent years - perhaps relating to the positive AMO pattern we're in) - at least that way it will be harder for any Eastern blocking to be blown away by the atlantic.

brogdale:

On a (semi) serious note..it is the case that the MetO do factor the preceeding May's N. Atlantic SST anomaly pattern into their NAO prediction, and that, in turn, is one element of their Winter seasonal forecast. I know there are a number of members on here who pay attention to SST patterns, and I thought it would be useful to raise the issue again. (No offence was intended Matty)

LOL None was taken

Just can't help it that it makes me laugh when one winter is barely out the way and people are trying to figure out what's going to happen next winter already. I know that is someone like Codge posted something regarding next summer in November he'd be accused of stirring up trouble and mild ramping

Hopefully next winter will be as mild as the one just passed. Probably a decent chance of that too

I'll be holding fire on this coming winter until I see what La Nina does during the autumn. My suspicion is that La Nina will re-stregnthen and have another go at us. In which case, even if we once again have reasonably favourable Atlantic signal, zonal induced La Nina will be the dominant player.

sriram:

We are well overdue a cold and snowy winter 

But years (or days for that matter) are not independent events. Therefore we are more likely to see a re-run of the previous day/week/year than to see a break in the pattern, in general terms; and only if there is specific evidence (NAOs or change in course of jet or whatever) for a break in the pattern is that prediction likely to change

DEW:

sriram:

We are well overdue a cold and snowy winter 

But years (or days for that matter) are not independent events. Therefore we are more likely to see a re-run of the previous day/week/year than to see a break in the pattern, in general terms; and only if there is specific evidence (NAOs or change in course of jet or whatever) for a break in the pattern is that prediction likely to change

but surely Sriram was just making the fairly straightforward point that if sustained severe (winter) weather (1962/3) has a statistical recurrence time period, say 200 years (), the each year we go without such events must surely 'shorten the odds' on a recurrence?

Anyway here are some SST maps:-

This first map is from mid May of 2005 and, in part, led the MetO to issue their first 'colder than average' winter seasonal forecast:-

Whilst this map from mid May of 2006 helped to produce a less cold (+ive NAO) forecast:-

and here is the latest offering :-

well, as they say....pick the bones out of that

Looks pretty neutral to me. Certainly not the warm/cold/warm band that we saw in 2005. Neutral to positive looks the call to me, but we still have another couple of weeks to go.

Gavin P:

Looks pretty neutral to me. Certainly not the warm/cold/warm band that we saw in 2005. Neutral to positive looks the call to me, but we still have another couple of weeks to go.

That would mean an average winter wouldn't it 

brogdale:

DEW:

sriram:

We are well overdue a cold and snowy winter 

But years (or days for that matter) are not independent events. Therefore we are more likely to see a re-run of the previous day/week/year than to see a break in the pattern, in general terms; and only if there is specific evidence (NAOs or change in course of jet or whatever) for a break in the pattern is that prediction likely to change

but surely Sriram was just making the fairly straightforward point that if sustained severe (winter) weather (1962/3) has a statistical recurrence time period, say 200 years (), the each year we go without such events must surely 'shorten the odds' on a recurrence?

If indeed he and you are making that point, then you are (I regret to say) statistically speaking quite wrong.

Consider tossing a coin. By your argument, a sequence of, say, 10 heads makes  the next toss virtually certain to be a tail. Alas, the penny doesn't remember what it's been doing and the next toss is still equally likely to be heads or tails. In fact, you might well argue that the next toss is actually more likely to be heads, as someone might be trying to wind you up with a double-headed penny

My argument is that the situation as regards weather is worse than that of the penny. The weather may well 'remember'  what it has been doing, but only in the sense that a pattern once established tends to persist. So repetition is more likely than variation. In some respects, that's what global warming predicts.