Glacier Point

Ironically, the year with the warmer Arctic was the colder for western Europe:

1951 and 1952 are however getting quite similar in magnitude and probably our best comparables in SSTs. 1951 in particular has many similarities and would serve quite nicely as an analogue: A cold night today, followed by a wet and chilly tomorrow - come on CET, drop down to 12.9

the 1940 - 1960 period did see anomalously warm waters in the North Atlantic and that didn't seem to stop those winters from being quite nifty. We are however in a totally different league in respect of the magnitude of this warmth.

Thanks Paul. Yes a restablishment of a pattern following an anomaly would be suggestive of the trend for increased warming throughout the year. Two anomalies on the bounce, some inconvenient questions might be asked. The problem comes if we get a highly average winter, but with the first widespread snowfall event in the last xx years. That wouldn't help us too much either way.

Not entirely balanced here Stratos. Rewind the tapes. 2005, one of the warmest hemishpheric and global years. Temperatures across northern and western Europe widely +1.5 to 2 C for the summer months, a highly diminshed cold polar pool in historical terms. And yet..... A significantly cold winter for central and SW Europe, in some cases 2.5 to 3 C below the 1968 - 96 average. True, the cold never really got to the UK's shores, but in a climatological sense, that is but a hair's breadth due to the prevelant synoptics as the origin our sustained winter cold episodes are essentially the same as the rest of western Europe. We must ask why did such a turnaround take place given such positive forcing of hemishperic surface temperatures ? Was there a masking of a really harsh winter ? The answer to the second challenge is one that neither you nor I can confidently respond to and it likely take many more years to be able to conclude either way. This year in many respects poses a re-run of 2005. The Atlantic SSTAs are very similar. The summer warmth is greater, and yet I would conclude at this stage the prospects for another winter with an increased probability of cold synoptics is entirely feasible - this time perhaps more so during January. Should be a fascinating few months though either way. GP

Those two comparables are illuminating. 1968 was the only year recorded with an August to October -ve NAO whilst the SSTAs for both years was very similar to todays. This October will come in as the only other recorded incidence of the three consecutive months NAO for the same time. I singled out 1969 as a useful comparable for the Atlantic because it most matched the Summer variables - and it continues to throw up the comparisons. GP

The key to understanding why january 1969 was so mild despite a -ve NAO (and AO for that matter): 1) SSTA: Big -ve anomaly in the central Atlantic helping to influence a large Atlantic trough, perfectly placed to maintain posiitve height anomalies over Greenland; 2) ENSO neutral conditions (there was a very weak powder-puff El Nino that Spring) allowing the SSTA in the Atlantic to prevail; 3) A stonking -ve PDO helping to force the PNA -ve throughout January: http://www.cdc.noaa.gov/Correlation/pdo.data http://www.cdc.noaa.gov/Correlation/pna.data bringing about about the ingredients for a highly amplified jet pattern and troughing in the Atlantic maintaining a run of warm southerly or sw'lys interspersed by some easterly phases as the jet fractured. History is always relevant to today and the conclusions to be drawn are that a -ve PDO / -ve NAO is never a good combination for winter cold for the UK, especially so when the SSTA points towards mean troughing in the Atlantic. This year, the PDO has been negative for the last two months: http://jisao.washington.edu/pdo/PDO.latest although there are indications for some weakening of this to more neutral values. The east Pacific may however offset and impact of this factor with a tendency towards with a strong jet flow across the southern US brought about the developing El Nino and warmer sea temperatures to the NE Pacific: http://weather.unisys.com/surface/sst_anom.html GP

It's very hard to make a direct link between one month's CET and SSTAs, especially so during January which is highly influenced by stratospheric and broader hemispheric factors. From 2001 to present, those plots illustrate a range of ENSO conditions and we should also point towards the AO being in very different leading modes for these months. However, a more trend-based analysis might provide more useful. Here are the reanalysis plots for the period August - November preceding those years and the plot for August - September this year.... and 2006 so far... Arguably, this year has seen the greatest differential in anomalies between latitudes 55 - 70 N and 35 - 45 N which should translate to an increased probability of the NAO in a -ve state this winter. GP

September's MEI value was 0.892, and a value of somewhere between 1 and 1.5 looks a very strong candidate for our DJF calculations - weak to moderate category. GP

not so much of the tripole but dipoles or couplets these days Brick, but yes, very much a return to 1940-1960 dodecadal patterns of pretty warm north Atlantics: +ve anomalies in the field 55 - 70 N, less +ve or -ve in the field 35 - 45 N will still do the job.

Does a positive orientated SSTA couplet in the Atlantic have the same impact as a more clear cut positive - negative one ? This is essentially the debate at present regarding those warm anomalies to the north Atlantic. Most contemporary research has concluded that extra-topical SSTAs vary in their influences between latitude, orientation and season with summer-time infleunces less so (but still significant) in the northern hemishpere. What has not been successfully demonstrated yet is the response (particulalry in terms of the NAO signal) of warm anomalies over less warm anomalies. This is due to this being quiet a recent phenomenon, witnessed over the last 15 years: compare 1980s to 1990s+ SSTAs Now this may or may not be related to the lack of wintry weather experienced by western Europe (the timing certainly fits), but this could be attributed to other factors, not least decadal trends in the oceans. Last year provided a really good test of the thoery that warm waters around 55 degrees north over less warmer waters in the Atlantic below 45 degrees north would bring about a -ve NAO and colder winter conditions. It didn't bring about a negative NAO, but it did bring to a juddering halt the long run of above average months and very mild winters. I would speculate that the NAO was heavily influenced and often over-riden last year by the strong easterly phase Pacific Jet and a developing La Nina. By taking the NAO as a measurement of Atlantic SSTAs, for winters with all three months in a -ve NAO leading mode, the SSTA composite looks like this: with warm waters to the north and cold anomalies to the SW Atlantic, and some hint of positive anomalies over the Pole. The CDC (following Barnston and Livezey 1987) calculation of the NAO and therefore the impact of SSTAs on lower tropospheric height variability would appear to suggest that a warm and less warm scenario in the Atlantic will still result in a -ve NAO: http://www.cdc.noaa.gov/map/wx/indices.shtml My assessment of this is that warm waters in the north Atlantic and neutral or even marginally warm waters southwards will still inherently infleunce a -ve NAO pattern (allowing for the AO in its leading mode). This year the SSTAs in the Atlantic are pretty similar to last: and we will once again test this theory, particularly so as the weak El Nino will, if persistent, promote -ve mean height anomalies in the SW Atlantic where cooler anomalies (cold anomalies = winter time low pressure) are required. In so far as 1985/6 is concerned SF, that is a poor example of how cold water anomalies are required for cold - a winter with the NAO values +.0.22, +1.11 and -1 despite some strong loading from a -ve AO throughout and a La Nina infleunced pattern to boot. ..no surprises that the SSTA for the preceeding 4 months in 1985 period looked like this... GP

1995: 2005: 2006: I'm struck by the similarity with last year really. Imagine how that would have turned out without the strongly -ve QBO and la Nina ?

indeed, another -ve EPO phase coming up with that cold anomaly south-eastern Pacific combined with the stronger sub-tropical jet which will be in evidence. I think this area will be crucial in determining amplification of the polar jet across North America - the more amplified the better within a +ve PNA regime which could be a real hum dinger for alternating cold and warm spells. It would be worth just revisiting the compsote anomalies for a -ve EPO during DJF to see how compatible this would be with a weak El Nino signature. RE Bartlett, absolutely no way with a stronger sub-tropical flow. The danger would be of a quasi Scandinavian high not quite orientated correctly for cold advection. GP

I disagree. I posted mid October last year how, under ENSO neutral conditions as we were then, that there was a reasonably strong relationship between warm Septembers and cold winters.... taken from my winter forecast 2005/6 issued late November 2005: "The lesson from historical parallels is therefore that under similar conditions, excessive Autumn warmth is met by a sudden reversal in temperatures during the winter months. Crucially, none of the warm winters experienced in the last 10 years outside of 1999/2000 have come close to the SSTA and ENSO conditions now prevalent.." and in the winter review.... "Back in November I identified a combination of an active hurricane season (increased tropical Atlantic convection), neutral El Nino Southern Oscillation (ENSO) conditions allied to a warm September as possible signals of an impending reversal in the monthly warming trend. I think the dramatic shift in patterns during November which took an above average first half of the month to a below average second half of the month in some way may be attributed to this, as can in part the following December pattern. So if we arrive in October with ENSO neutral (as they may well be) conditions, and a warm September behind us, take the tip, the cold will not be far away. The shift to weak La Nina conditions may however have promoted alterations in the Pacific and jet patterns over North America." It may not have been the degree of cold forecast - I think the switch to La Nina did have an impact as did the strongly -ve QBO - but I would maintain that Autumn warmth, under the right teleconnective regime is a positive signal for winter cold. The case is however hampered by a low sample. GP

Welcome back sceptical West. Your claim that the Atlantic is full blown is slightly erroneous. The story so far...... Low pressure manifested to our SW all the way to the Azores, and much above heights recorded over Greenland. Not a classic autumnal zonal pattern with low pressures crashing in on a strong jet. The Autumn so far has been one of strong meridional flows and a very weak southward displaced west-east pattern. This could all reverse spectacularly over the coming weeks however, but right now it is no way full blown. GP

September's MEI confirms we are on course for the desired weak to moderate strength event with most pleasingly no evidence of strong event in the offing. http://www.cdc.noaa.gov/people/klaus.wolter/MEI/table.html pieces of the winter jigsaw now beginning to fall into place. GP

September NAO data out - second big negative in a row following August. http://www.cdc.noaa.gov/Correlation/nao.data fairly unsual in its occurence - the last time was during the mid eighties - but how significant it is impossible to say. I suspect that phasing between La Nina and El Nino episodes may have some contribution but this would only be a hunch. Of more significance perhaps is the high degree of disconnect between AO and NAO. During September, the AO - NAO correlation is 0.7. So a big -ve NAO (-1.73) should have seen a big -ve AO ? No way Jose - http://www.cpc.ncep.noaa.gov/products/prec...ex/ao.sprd2.gif more a positive AO from the look of it although we still await the final data from CPC. What does this mean ? Basically the two oscillations are behaving independently of each other which is very, very unsual. What does this mean for the Winter ? Very hard to draw any firm conclusions yet and all we can safely say that this a statistical quirk. It's worth watching the NAO over the next month or two to see if a trend develops, particulalry given the SSTAs in the northern Atlantic which IMO are highly conducive to a -ve state. Of note so far has been the modality of the NAO trending very +ve or very -ve with no neutral values this year. Also, 5/9 months were with the NAO in -ve state this year. GP

October if anything could be more than the 1.5 anomaly. This could actually be good news for coldies as it would be a portent for increased strength of anomalies switching between strongly positive and negative. This has some logic in terms of sustained ridge - trough development through the Autumn - more cold air gets pumped southwards sustaining blocking west-east setting up bigger cold pools into the winter and so on..... I think we commented early on this summer of a possible winter of extremes. Perhaps ? time for a lie down, that's far too much ramping for one day.

I would concur. This old chestnut seems to get as many outings as the 1947, 1963, 1987 (and now seemingly 1969) charts!

Certainly Tamara. The warm SSTAs that may be a significant factor in delivering cold could equally be our downfall for the Feb period if (and that at this juncture is a large if) with all that mild air swamping in from low pressures to the west and SW. Very much speculative at this stage though, but constructive debate because this pattern will start to verify at the end of this month. I would suggest that we should look for cold surface temp anomalies across western / central Russia and a coninuation of the present pattern across western Europe with low pressure anomalies out to the SW. currently over the last 10 days of September ..... not bad, and GFS / ECM out to t144 maintain the ridging over Scandinavia which makes dowsntream trough digging cold air south into Russia a good bet to continue. Promising to watch and monitor over the next few weeks.

The German Climate Model (Roeder) predictions are based on mean sea level pressure anomalies. Blue is not necessarily high pressure and red not necessarily low pressure but the pattern between the two is important. It must also be remember that these are mean monthly projections where a +6 anomaly might vary considerably between say 1000 and 1030 mb in terms of daily values.

Most promising set of runs I've seen for cold lovers, and ones which very much chime with current thinking on weak to moderate +ENSO, a QBO becoming 0 by early January and AO / NAO becoming quite modal and overall neutral -neg: October very much as anticipated from the ENSO composites http://www.netweather.tv/forum/index.php?s...st&p=791954 http://www.netweather.tv/forum/index.php?s...st&p=783215 with low pressure anchored to our SW throwing up warm and another above average month - but also check out events over Siberia with that negative anomaly propogating some serious early cold southwards. November continues the ENSO trend, mean low pressure to our west and over the UK, mild and wet - but cold air still being drawn southwards to our east. December - classic Atlantic pattern for El Nino with lowering of pressures from Florida to the Azores and interestingly, the jetstream going very quiet and southwards with a probable strong sub-tropical jet in evidence. First glimpse and instinct might be for those easterlies not to pack the punch, but bear in mind what may have gone on since October in that locale. January - a doozy - strong sub-tropical flow and weak, displaced polar jet flirting with the SW - potentially a very snowy projection. Big -ve NAO I would suggest and neutralish AO with a possible very weakly +ve or even switched QBO phase. Feb - the breakdown, with the NAO still negative so some potential there before the milder stuff heads in. At present (we still have over 6 weeks to go) my thoughts would be very similar for October and November. Whether December follows suite is the major question at the moment. By far the biggest encouragement is the overlay of ENSO composites with current trend in Atlantic SSTAs and a strong hint for the sub-topical jet to be the major player on the hemispheric stage this winter. GP

Verification of September's CET prediction - coments in green bold: Obviously 1.9 C below the observed record temp outturn but encouragingly the overall synoptic prediction was quite accurate - perhaps too much emphasis on a -ve AO when +ve was more the call. Score 1 pt or deduct 1.9 under the new league table scoring system

easy SB, the league hasn't started yet - then you can try and psych us out.. GP

September's data now in so we have reached what I would regard as halfway point in terms of the 3-4 month lead in period for Atlantic SSSTAs influencing the broad weather patterns for the winter of 06/07. .. strong positive anomalies above 50 N and weakly positive or neutral values southwards of this latitude translates to a weakly negative or neutral NAO signal - based on the oceanic signal alone. Predictions of a +ve NAO based on SSTAs really have to start see this pattern reversed and at present, this looks unlikely given the current situation: http://weather.unisys.com/surface/sst_anom.html Also of note are the continuing development of El Nino in the Pacific and this will have a bearing on the Atlantic SSTAs over the next 2-3 months at least. Take a look at the composite anomaly pattern for Sep-Oct-Nov weak/moderate developing El Ninos: and the observed pattern for September... Height anomalies at mid and lower levels in the mid-Atlantic more or less coincide with the weakly +ve or neutral SSTAs. This will allow tropical storms to continue to stir up waters and increased convection in these areas further prohibiting warm water development at the surface - in effect maintaining the current trend towards a neutral or weakly -ve NAO for the winter in terms of the oceanic signal. What I would stop short at is forecasting a neutral or weakly -ve for January / February period as this will likely be more influenced by the AO when the linear correlation between the two oscillations is more signficant. December might be different but still a long way off just yet, especially with the QBO trending +ve at this stage (-ve AO and therefore -ve NAO less likely). GP