Jump to content

Model output discussion - here comes the beast!

Message added by Paul

Please only post model discussion in this thread 

For more general chat and banter, or moans and ramps loosely around the models, please head to the banter thread:

For general weather chat including about the snow/cold chances around the country, please go to the regional threads:

We also have a special post SSW cold spell discussion open here:

Thank you!

Recommended Posts

  • Replies 4k
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

Early guidance forewarning of next weeks weather was issued earlier on, and this has been further enhanced with extended outlooks in bulletins. AFAIK (comms not being my area) public agencies and majo

Come on those looking for the breadown before the major cold and snow has begun You will be very lucky if that is the correct word to see this in ANY winter in the rest of your lives I would sugg

Microscale detail absorbing an awful lot of attention on here, when in reality it is a waste of emotion. The macroscale pattern is now fixed - it will not change significantly now for the start of nex

Posted Images

The ens-and variations-are in tune with the 18z raw.

Now the adjustment sits..

With lack of mobile energy fixed into the gfs.

We gain on nao-then switch...

Its indeed correct in its evolution....


18Z has been on the button..and you will see this evoulution via its sister outs..

However'again' the dynamical exacts are fraught..and miss-matched for secondary warm(ssw)..

196hrs is the cut off point ..via gfs..

But the 18z by time definintion is the lead player!!


Vigorous eastern lobe vortex!!..

The gfs is on miss programe at this point.

And you can see the clear diabolics for want of revert.



Edited by tight isobar
  • Like 3
  • Thanks 1
Link to post
Share on other sites
4 minutes ago, Mike Poole said:

GFS 18z is a totally nonsense run.  The T132 to T150 period looks very dodgy to me and so I think the rest of the run can be discounted as too many pints in the pub, a dodgy rancid kebab and a massive chunder!

Yes agree, I think it’s trying to hard with the general evolution, if you get my drift. Just looks odd. However the odds seem to rising on a channel type low, southern counties should probably keep an eye on this, how far north is the question! Elongated is mine and probably most in here’s preference. Interesting for sure. 

  • Like 2
Link to post
Share on other sites
25 minutes ago, snowbob said:

I think everyone is just now staring at the screen in disbelief.

we are so used to watching model runs from 300 hours get closer and closer then a shortwave will kill the cold.

or watching a 24hour cold easterly or northerly or even a north westerly get watered down to marginal sleet fest.  watching our 2cm of slush drip drip dripping away.

this last week has been amazing 

0z upgrading on its 6z then 12z totally amazing model watching.

i have been a member on this site since 2009 and haven’t seen anything like this.

this really I believe will not be seen to this level for a good 30 odd years.

some posters on this thread (you know who you are) in my opinion should actually be getting paid for their skills

i.e @chionomaniac , @Steve Murr, @nick sussex, @Nick F 

Amazing analysis and tell it how it is and what to look for

keep up the good work fellas

and let’s all enjoy this

Yes, I want to echo those sentiments.  Lurking here as I often do, it is amazing how much one can learn from some of the people here.  I really appreciate the time and careful thought that so many people put in - I'm sure I'm not the only one who reads and learns so much but rarely comments.  The combined pool of knowledge and experience here really is a good thing. 

Right, onto the models.  The latest GFS model seems to show quite a lot of instability ploughing into southeastern UK from early Friday onwards, that low pressure blob and so on. My unashamedly newbie question is - to what extent is this a low-resolution prediction of an actual weather event, and to what extent is it merely the result of the amplification of noise further down the line of a very complex forecast with lots of variables?     


Edited by CloudyBay
  • Like 8
Link to post
Share on other sites
6 minutes ago, Steel City Skies said:

Do you think the 18z is realistic in terms of its northwards progression? GFS often does this and it's annoying! 

I'd rather it stay a little further south and keep us in the cold for longer.

It will continue north until it meets the next push of cold uppers coming in from the north / north-east. That looks to be around days 10-12, so yes scope for northern parts of the U.K but questionable how much it can dent the embedded cold airmass. 

  • Like 6
  • Thanks 1
Link to post
Share on other sites

There's brutal cold, severe wind-chill, heavy drifting snow and penetrating frosts on the way from early next week..nothing the Gfs 18z shows from the end of next week can stop that!...we are in for a very severe siberian blast which has cross model support, the wheels are already in motion..it's coming!!:shok::cold:

  • Like 8
Link to post
Share on other sites

FI is colder on 18z GFS than 12z. Even classic cold spells have milder blips. We may have to accept a break in the snow before a renewing of northern heights and another cold plunge? 





  • Like 1
Link to post
Share on other sites

Oddly enough it is the extreme blocking which seems to cause this run to go a bit iffy.


That deep cold is completely cut off from the arctic for a significant amount of thime causing it to barrel around with no other steering like the others which tend to redevelop the NE/SW trough extension from the developing Asian vortex. This allows the low to deepen and approach the UK. The long term trend is similar to the GEFs/EPs but this action allows the jet to lift far enough north to turn the south milder. This run looks out of kilter early on with the extent of the blocking and hence a solution squeezing the cold pool more tightly early on is probably favoured.

This does not however resolve the track of the low at the end of the UK because the UKMO pushes this up into the south of England regardless. Still some deep cold on offer before this and plenty of snow for many. Also it would not take much to keep the whole of the UK on the cold side with plenty of precipitation on offer that could deliver snow.

Edited by Captain Shortwave
  • Like 1
Link to post
Share on other sites

So it’s a race between the block to the north and the low to the sw.

Interesting to see how this plays out although I’m still dubious of outputs that push the low right into the UK,

The low is a little unusual , it tracks more east towards Iberia before moving ne , it’s not what one would call either a typical Channel low scenario or a normal breakdown from the sw.

It’s possible the low might get further east before it’s turn towards the ne.



Edited by nick sussex
  • Like 3
Link to post
Share on other sites
3 minutes ago, bluearmy said:

There are six eps clusters until T192. That says we are still some way from knowing how this plays out, even days 4/6. All clusters are bloody freezing. 

The spreads say that the first low doesn’t get north of the channel coast with less cold uppers and neither does the day 10. that a tad surprising. 

Is it really? It’s going to slow down due to how it encroaches, in very cold dense air. This slow down itself what could dump a lot of snow will be interesting to see IMO, its better staying south in terms of keeping cold on for longer I’d say people also should note even though uppers in SE are much less cold. Because it has been so cold it could well still fal as snow I believe the worst snow event in SW England had insanely marginal uppers.

  • Like 2
Link to post
Share on other sites

I quite surprised how quiet it is in here tbh, think everyone just wants next week to start as we have been counting this down for quite a while now.

got to say though how nice is it to not be seeing any downgrades as so often is the case.

upgrades just keep coming.

bring on the snow...it’s going to be bitter

  • Like 2
Link to post
Share on other sites

Unbelievable model viewing. I'm lost for words, and teeming with excitement. The block in question is monumental, and any supposed breakdown is very unlikely to be modelled with any real accuracy. Agree with other posters, GFS loosing the plot and "defaulting"to norm, there is no way imo such deep cold would be removed so quickly from Europe. The trend imo had been to increase the strength of the block and the severity of the cold that hits the uk. Are we in for a direct hit ? I unfortunately took a break from model watching, and almost jumped out of my chair when I saw the charts. I think we are in for a memorable event folks 



Edited by Zak94
  • Like 5
Link to post
Share on other sites
Just now, Daniel* said:

Is it really? It’s going to slow down due to how it encroaches, in very cold dense air. This slow down itself what could dump a lot of snow will be interesting to see IMO, its better staying south in terms of keeping cold on for longer I’d say people also should note even though uppers in SE are much less cold. Because it has been so cold it could well still fal as snow I believe the worst snow event in SW England had insanely marginal uppers.

Not just because we'll have entrenched surface cold but also we'll have very low dew points due to the airmass originating over the continent. Uppers could get up to 0 and it would all be snow.

  • Like 3
Link to post
Share on other sites


The last time I produced one of these Eurasia reports was on February 2nd. This post will be a full report with all my usual features. With the upcoming severe cold spell shaping up to break a few records, I plan to post quite regular short updates throughout this truly fascinating period. I will post one of my Arctic temperature profiles, sea ice analysis and sea surface temperature reports later tomorrow.

European 2m Surface Temperatures:

       Feb 23rd  Current =  2230                            Feb 23rd 0650                                      Feb 22nd 1450                        GIF: 24 hours to 2230 Feb 23rd

  temp_eur2_wtd0.png     temp_eur2-06.png     temp_eur2-14.png     tempresult_wew7.gif

Note that it is important to allow for min/max temps when comparing the charts. The 0650s are usually close to the minimums and the 1450s are usually close to maximums when other conditions remain similar. This is not the case right now with progressively colder air surging in from Siberia and Russia. The "GIF" chart shows the changes over the last 24 hours (receding and expanding cold). Notice that the values on it at 2230 are already matching those at 0650 this morning. The first chart "should" auto update roughly every 10 minutes but it doesn't seem to want to play ball. I'm always having copy problems with these - when I don't want charts to auto update they often do and now that I want one to update automatically, it won't! As many of you will want to watch the progress of the deep cold moving in, here's the link to it: 


The dark purple colours (sub -16c surface temperatures) are pushing in from Russia across north-east Europe and into eastern Scandinavia. The bright purple (or deep pink) colours (sub -20c to -30c) are following on behind. There should be some white colours (sub -30c) and even lower values appearing within a couple of days.

  European Dew Point Temperatures:            

        Feb 23rd  Current =  2230                           Feb 23rd 0650                                       Feb 22nd 1450                         GIF: 24 hours to 2150 Feb 23rd

   pointrosee_eur2_ttr4.png      pointrosee_eur2-06.png     pointrosee_eur2-14.png      tempresult_mtb1.gif


As is typical with a long fetch easterly sourced from northern Russia and Siberia, there is some very dry air with extremely low dew points moving in. In fact the whole of Europe (except the far south) including the UK has dew point values below 0c. Many central parts have values between -10c and -20c and eastern parts have values below -20c. Some sub -30c dew points are just starting to move in from north-west Russia.  Again the GIF chart shows that the dew point this evening are already matching the 0650 values from this morning.

Overall, these exceptionally low dew point temperatures and the increasingly low surface temperatures, combined with the near record low 850s temperatures (see my post on page 104 which compared the predicted 850s against the lowest recorded in past severe winter spells) demonstrate the sheer depth of the severe cold approaching us. I for one, will not be at all surprised to see the UK's late February and March maximum and minimum temperature records broken during the next two weeks. It is quite possible than many of the UK forecast temperatures for this period have been thoroughly understated and will need to be substantially lowered. Time will tell of course. UPDATE: While I'm writing this bit, the 12z output is just rolling out and I see that some of the models are starting to get more of a handle on these excessively low temperatures - still some way to go. There is nothing at all "marginal " about this spell.     

Temperature Forecasts for Key Cities:          

I normally include northern hemisphere temperature and 850 charts in this section but so many posters are reporting on these these right now, so instead I'll take a closer look at the specific forecast temperatures for several cities from London eastwards. These show the GEFS 6z ensemble output (source: Netweather). By the time I finish this report the 12z charts will be out and I'll show them directly below each 6z chart.for a good comparison. I believe that most of the surface and 850 temperatures are still very much overstated and we'll see progressively lower values for at least week 1. Then as the deep cold pool becomes entrenched, the week 2 values may well go down too in perhaps 3 to 4 days time (but that's a little too far off at this stage). 

GEFS 6z:       London 2m temps                                     London 850 temps                                         Berlin 2m temps                                     Berlin 850 temps    

     t2mLondon.png    t850London.png     t2mBerlin.png    t850Berlin.png

The average of London's surface temps remains close to freezing throughout, starting off a little above, then dipping a little below around the turn of the month and rising slightly during week 2. The GFS operational run was one of the warmest and the GEFS control run one of the coldest solutions. There is a substantial spread within the ensemble members in week 2. This is reflective of the fairly consistent attempts by GFS to bring in less cold air during week 2. I do expect them to fall into line with the other models very soon as evidenced in several of my recent posts. The lowest 850s of around -16c on March 1st/2nd is more in line with the other models but again in week 2 GFS is mixing out those uppers and pushing them back north-eastwards. I feel that they are completely underestimating the exceptional strength of this cold block. There is a similar trend for Berlin, albeit with temps around 5c lower throughout. I'm really struggling to interpret their 850s in week 2. The control stays pretty low but the op gets to +5c which would suggest that the whole block is being pushed away by a powerful jet stream, which couldn't be further from the truth. This is NOT ramping it is plain common sense and GFS simply cannot handle full pattern reversals, a blocked Atlantic and a weaker jet stream diverted a long way to our south.

GEFS 12z:    London 2m temps                                     London 850 temps                                         Berlin 2m temps                                          Berlin 850 temps   

    t2mLondon.png     t850London.png      t2mBerlin.png     t850Berlin.png 

Well, as expected there is a slight trend downward from the 6z to the 12z output for London. The op run remains one of the warmer solutions. The 850s generally recover more slowly but the control run joins the op amongst the warmest solutions and there is a slightly narrower spread amongst the ens members. Berlin is little changed but the 850s are shown to recover more quickly from their low values. 

GEFS 6z :     Moscow 2m temps                                    Moscow 850 temps                                       Helsinki 2m temps                                     Helsinki 850 temps 

     t2mMoscow.png    t850Moscow.png      t2mHelsinki.png      t850Helsinki.png

Going further east, the Moscow temps during week 1 drop off the chart, the average is probably below -20c. Then the is a recovery to nearer -5c - this usually occurs during a snowfall. I must check this elsewhere and this is as representative as any right now.   https://www.bbc.co.uk/weather/524901   Moscow's 850s drop well below -20c but are shown to recover during week 2.to nearer -5c on average but with a large ens spread. Helsinki's surface temps also drop off the scale starting off around -10c then falling (probably) below -20c and gradually recovering to around -5c to -10c with a moderate spread of ens members. The 850s are similar to Moscow's but with a slightly tigher spread. 

GEFS 12z:    Moscow 2m temps                                     Moscow 850 temps                                       Helsinki 2m temps                                     Helsinki 850 temps     

      t2mMoscow.png    t850Moscow.png     t2mHelsinki.png     t850Helsinki.png

Moscow is also little changed early on with extremely low values for late February but the recovery is marginally stronger on the 12z and more so for the 850s.  Helsinki is also little changed. On both the surface and 850 charts the op run has moved from a colder to a warmer solution. The control on the 850 has moved the other way. There is a large spread amongst the ens members from around day 5 onwards suggesting that GFS still needs to get a better handle on this (see my post on page 64 for my many reasons why I feel that the GFS output for week 2 is just plain wrong)..

Asian and European Snow Cover:

I show animations for snow cover and sea ice changes. These are produced by the National Ocean and Atmospheric Administration (NOAA).  When you go to their site you can change the date range and go back over 10 years. You can change the speed and pause on any particular day. You can also switch to a northern hemisphere view. These are brilliant, very informative charts and great to play around with. I’ve re-set the link below to show the period from Feb 1st to Feb 22nd but you can change the dates on the site and choose your own options.

Animated Europe and Asia Day Snow Cover  (updated by NOAA February 23rd):


Here are the start and end charts:                               February 1st                                               February 22nd


Source: NOAA US National Ice Data        ims2018032_asiaeurope.gif       ims2018053_asiaeurope.gif            

North Asian snow cover has been well above average since mid-October. During February the extreme southern extent in central Asia has receded slightly. Scandinavia has a complete snow cover. The snow cover in Europe has been much more inconsistent and come and gone several times throughout the winter. The snow cover is once again extending south-westwards from Russia into eastern and parts of central Europe. As the deep cold pool develops over Europe, any snowfall will accumulate and I expect much more of Europe to turn white during the next two weeks, along with much of the UK, which "could" get considerably more in this set-up compared to the near continent. I will keep this under review and provide regular updates. 

British Sea Surface Temperatures:

                 Feb 22nd 2018                                     Feb 1st 2018                                   Feb 22nd 2017                                 Feb 22nd 2016                   Global SST Anomalies Feb 22nd

sstuk.gif   2018-02-01uk.gif   2017-02-22uk.gif  2016-02-22uk.gif  sst_anom.gif                  

Source: Meteoceil                                                                                                                                                                                                                                                               Source: NetWeather

Sea surface temperatures (SSTs) around our shores are approaching their seasonal lows. This is typically around mid-March but can be any time from late February to early April depending upon the prevailing weather patterns, conditions and surface air temperatures. You can see that values have dropped about about another degree or so since the beginning of February. The lowest values are in the North Sea around East Anglia and Lincolnshire. I also show the charts for the same dates for the last two winters. At this time last winter and in the previous winter SSTs were around 1c higher than they are now in the North Sea and between 0.5c and 1c higher elsewhere. With the incoming easterly, we can see that North Sea temperatures will modify the surface layers once they leave the European land mass. I commented on "convective easterlies" in my post yesterday (now buries on page 82 of this thread)

The final chart above shows the global (excluding the Arctic which I'll report on tomorrow) SST anomalies. This compares current values to the 30 year mean (1970 to 2000). Apart from a colder patch south of Iceland and Greenland most of the Atlantic has above average anomalies. Global SSTs have generally been above their average values for some time. 

Right, that's enough for today. I'll be back with my Arctic temperature report much later tomorrow.





Edited by Guest
Correct typos and check charts & links
Link to post
Share on other sites
6 minutes ago, bluearmy said:

There are six eps clusters until T192. That says we are still some way from knowing how this plays out, even days 4/6. All clusters are bloody freezing. 

The spreads say that the first low doesn’t get north of the channel coast with less cold uppers and neither does the day 10. that a tad surprising. 

EPS have a habit of leaning againt ECM right? Even so, a glancing blow at most still makes more sense to me than the low getting right across the UK.

Maybe the low getting across the far south as per UKMO 12z in the most extreme case - but the major letdown of March 2013 has made me completely unable to trust such outcomes at more than a few days’ range.

Having said that, while a halfway house between GFS and ECM remains a very snowy outcome, I will be feeling at least a little positive with respect to that potential event.

Just a whole bunch of potential snow streamers to resolve first...!

  • Like 3
Link to post
Share on other sites
This topic is now closed to further replies.
  • Create New...