Nouska

The 18Z GFS has developed a bit of a reputation on weather communities - not wholly deserved IMO - it has sometimes been the first to pick out runs that verify in the long range. Mr Holmes is a proponent of comparing like for like runs: so let's do that. Dates and times slightly different but the theme is what matters. Yesterday PGFS 12Z left and todays on the right. Same for the PGFS 18Z runs. Now, I've been following charts a long number of years and to my mind that is quite a signal of continuity - it has been seen before at such a long lead - Dec 2010 and Christmas day/Boxing day of last year are just a couple of examples. Edited to add the trend from the NAEFs - chosen Paris as it indicates if low heights are moving south at that time scale. 00Z 12Z

The Reading EPSgram has finally got down to -3C on the scale and there seems to be a drying out trend as well. http://i.imgur.com/QHBlnUs.gif

Couple of points on this: - just as the ensembles blend out nuances, so do the composites that we look at when considering Cohen's hypothesis. There will be a variety of options possible - probably none exactly like the composite! The Met Office 15 - 30 day outlook is heavily weighted on the ECM monthly - it was last updated on Thursday of last week - a lot of water has flowed under the bridge (NWP output) in that time-span. If it is still the same on Wednesday, time to scratch the head and wonder where we are going.

Right at the end of the PGFS 18Z, a fair representation of the Z500 anomalies presented with the OPI winter forecast.

The GFS run with precipitation charts - the white hatched lines indicate snow. This is too far ahead to be of any use as a forecast but you can follow to see how it changes. http://www.meteociel.fr/modeles/gfse_cartes.php?ech=204&code=0&mode=2 The low res part of the run is a good example of how quickly Europe can cool down with the right synoptic set up.

A little diigging around in the archives of the MJO to see if any years matched current forecast - it produced three matches - '75 - '84 and '93. I suspect the winter that would most appeal among those three is 84/85 so a quick recap of the state of play on the first of December of that year. http://cawcr.gov.au/staff/mwheeler/maproom/RMM/phasediag/pd.1984.12.1.gif Looking very familiar? .....thought so..... The archive - to have a look at the other options:- http://www.meteociel.fr/modeles/archives/archives.php?mode=2&month=12&day=1&year=1984&map=0&hour=0 The MJO archive. http://cawcr.gov.au/staff/mwheeler/maproom/RMM/phasediag.list.htm

I see what you are saying but the McRitchie one has today's date as low amplitude, cusp of phase eight/one. whereas multi-model, for 20th November is well into phase two. The Hendon and Wheeler one is lower amplitude but still phase two. I'm not meaning to be pedantic, just needing clarification.

IDO, something amiss with your MJO graphic - the forecast start point is not the same as on the current multi-model one. No clear picture ahead from phase three.

The CFS four weekly Z500 outlook - they are also seeing the trend to a Scadinavian high pressure as the follow-on from the digging trough.

John, you were mentioning the MJO phase in an earlier post - I did some checking last night after reading Mike Ventrice's blog. The phase two and three composites are as follows for December. I've used SLP as it replicates the 500 heights well: easy to see why some are looking on with interest if it should follow composite.

Ongoing research on this subject. http://gtr.rcuk.ac.uk/project/CBAD3031-B011-4FFF-B839-3095EC5366FB This is also a good read, from a multi-discipline level, of how solar effects can impact different aspects of the climate. http://science.nasa.gov/science-news/science-at-nasa/2013/08jan_sunclimate/

Since this seems to be focussed on an American theme, have a read of Dr Ventrice's latest blog on the December outlook for the US. The headline is about the massive Indian Ocean MJO event that is forecast. http://www.wsi.com/blog/energy/sub-seasonal-u-s-temperature-outlook-through-mid-december/ Being that an IO, MJO gave us our only, northerly dominated, below average month this year (August) and that the portrayed US temperature departure for December looks nothing like the pattern of last December - which some fear will take hold - what might be on our menu for December? December high amplitude phase 2. December high amplitude phase 3. Composites from http://www.americanwx.com/raleighwx/MJO/MJO.html Another source of composites that are less familiar on here from JMA. Z500 and T2m anomaly. If that should transpire as illustrated, nothing at all like last December - a gradually cooling picture under a nice seasonal, frosty high.

Hi Knocker - is this chart from the FIM the same parameter as the one Dr Maue posted on the tweet.That one is from GFS but dont see the potential temp pv=2 option on the GFS we get to view.

For all the dataholics, ECMWF have a page devoted to data collection sources for its daily model runs. http://old.ecmwf.int/products/forecasts/d/overview/monitoring/coverage/dcover!AMVs-Infrared!00!pop!od!mixed!w_coverage!latest!obs/

Here is the Reading epsgram of the ECM 12Z output - the high uppers do not reflect in surface temperatures but both deterministic and control are at the bottom of the range. I've used an imgur link in case of copyright issues. http://i.imgur.com/UId8M3x.gif

http://www.metoffice.gov.uk/barometer/science/2014-08/ENDGame

At first glance, the model output may not look as good as last night but a look at the De Bilt temps shows a strengthening signal for colder options. I've used max temps rather than minimums as we know nights will be cold under high pressure - the days can be variable. 12Z 00Z

Yes, indeed - a very clever man whose writings have been a useful tutorial in my baby steps to further my knowledge of tropical meteorology. From reading the tweets, he's not saying moderate Nino in fifteen days - rather, the USA will possibly see a Nino type pattern in the fifteen day timescale (indicative of ocean and atmosphere finally cooperating) and CFS is forecasting moderate/strong Nino to develop as winter progresses.

What are the odds of synoptics, rare as rocking horse pooh, turning up twice in four years on the exact same day? Probably no chance of winning that bet but there is something in the current atmosphere state that must be conducive to such an extreme possibility - charts like these require a very special set of circumstances.

I'm a great devotee of the excellent tools that JMA provide for creating composites. This one allows you to play around with the different ENSO states and apply different parameters for each month. For example, December 850 temps with a west based Nino. http://ds.data.jma.go.jp/gmd/tcc/tcc/products/clisys/newoceanindex/index.html

The SLP was a good match in October but has gone out of step in the first trimester of November. You can do a quick compare of other levels on the JMA link where I've used the ten day increments. http://ds.data.jma.go.jp/gmd/tcc/tcc/products/clisys/figures/db_hist_jun_tcc.html

This animation of the 500mb Arctic circulation, from Global weather logistics, gives a fantastic insight into the forecast disturbance. Model is the GFS. http://globalweatherlogistics.com/seaiceforecasting/gfs.500mb.temps.arctic.html

You got there before I realised the cursor had moved on the timescale - no difference actually.

The situation in the next couple of days is a good example where there is very little differnce between 850 temps and ground temps - the atmosphere 'cuts' available on meteociel are a useful tool for a quick check. http://www.meteociel.fr/modeles/coupes_gfs.php?mode=7&ech=3

I think we have been seeing variations of this pattern in recent winters where the jet tracks southerly but it depends on the progression into Europe and the tilt/shape of the low as to what weather we get. My hunch is that the OPI forecast, using the analogue years, is a little too extreme. By that, I mean, the inclusion of 62/63 skews too much to cold overall. When I looked at their Z500 composite, I immediately thought of Feb/March 2013..... .....a little tweaking of orientation would make a good fit as long as we assume the latitude variabilty of the lows approach.