Model output discussion - heading into 2018

[Ice Man 85]

might be jumping the gun with my limited knowledge, but the 18z seems to have reverted to mild again and earlier too. Oh well, was nice while it lasted...

That Mormon succumbed a lot faster than even I expected.

Edited December 22, 2017 by Ice Man 85
more to add

[Daniel*]

This seems like a new development on GFS 18z a picture postcard white Christmas for much of Scotland? IMO a very good chance of a white xmas for places like Inverness, Glasgow & Edinburgh and more.

Not restricted to northern / northwest Scotland..

[Man With Beard]

EC clusters tonight at D10, D12 and D15. Average westerly conditions the best bet really, but with the odd mini-ridge possible at times. Much the same as recent days for the period. I'd guess a little above average generally but the possibility of the odd cooler day as troughs pull past the UK.

Edited December 22, 2017 by Man With Beard

[feb1991blizzard]

5 hours ago, feb1991blizzard said:

Met offices in house model has a bit of sleetyness but for the most part  heavy rain with next weeks slider.

Much more snow on offer on the next run - certainly for the pennines.

[northwestsnow]

2 minutes ago, feb1991blizzard said:

Much more snow on offer on the next run - certainly for the pennines.

Sounds good feb

Bit of an upgrade by the sounds of it..

[bluearmy]

2 minutes ago, northwestsnow said:

Sounds good feb

Bit of an upgrade by the sounds of it..

Isn't it a bit far away to be getting excited on a 'will it/won't it' scenario ?? 

mwb - the extended clusters really aren't convincing me - I have no idea which is more likely (or a blend of two of them.)

[feb1991blizzard]

1 minute ago, northwestsnow said:

Sounds good feb

Bit of an upgrade by the sounds of it..

Yes - haven't looked at the map yet but this BBC location forecast (which I believe is derived from same model) didn't have one snowflake symbol in 10 days worth of forecast range earlier, now not only does it have a spell of snow with the slider overnight tue into wed, the next system / or showers on Friday are now wintry too.

http://www.bbc.co.uk/weather/ol3

Also icon 18z had a fair amount of snow for NW eng  Cheshire and parts of wales but also had snow as  the system clears as far south as parts of the SE.

[northwestsnow]

9 minutes ago, feb1991blizzard said:

Yes - haven't looked at the map yet but this BBC location forecast (which I believe is derived from same model) didn't have one snowflake symbol in 10 days worth of forecast range earlier, now not only does it have a spell of snow with the slider overnight tue into wed, the next system / or showers on Friday are now wintry too.

http://www.bbc.co.uk/weather/ol3

Also icon 18z had a fair amount of snow for NW eng  Cheshire and parts of wales but also had snow as  the system clears as far south as parts of the SE.

Oh no!!!

Not the BBC app!!!

Its dreadful Feb- it changes like the clappers..

[blizzard81]

7 minutes ago, bluearmy said:

Isn't it a bit far away to be getting excited on a 'will it/won't it' scenario ?? 

mwb - the extended clusters really aren't convincing me - I have no idea which is more likely (or a blend of two of them.)

Mmm! Four days out. Not exactly in the far distant future is it? Some might argue the same point about analysing clusters at days 10 to 15. Live and let live is what I say. 

[ronan]

13 minutes ago, feb1991blizzard said:

Yes - haven't looked at the map yet but this BBC location forecast (which I believe is derived from same model) didn't have one snowflake symbol in 10 days worth of forecast range earlier, now not only does it have a spell of snow with the slider overnight tue into wed, the next system / or showers on Friday are now wintry too.

http://www.bbc.co.uk/weather/ol3

Also icon 18z had a fair amount of snow for NW eng  Cheshire and parts of wales but also had snow as  the system clears as far south as parts of the SE.

Didn't the last slider change its position 4 times a day for a solid week? Locations forecasts are a load of crap to be honest.

Edited December 22, 2017 by ronan

[northwestsnow]

1 minute ago, blizzard81 said:

Mmm! Four days out. Not exactly in the far distant future is it? Some might argue the same point about analysing clusters at days 10 to 15. Live and let live is what I say. 

I suppose 4 days is a long time for pinning down specifics- so i hear Blue..

But no harm in looking at the potential

[feb1991blizzard]

Just now, northwestsnow said:

Oh no!!!

Not the BBC app!!!

Its dreadful Feb- it changes like the clappers..

Yes I know, because its updated really often the nearer timeframes (all it is is like a simplistic view of Met Office model data as its updated by them.   -    To say they are not worth bothering with is like saying that that particular model is not worth bothering with!

[Paul]

2 minutes ago, feb1991blizzard said:

Yes I know, because its updated really often the nearer timeframes (all it is is like a simplistic view of Met Office model data as its updated by them.   -    To say they are not worth bothering with is like saying that that particular model is not worth bothering with!

Except of course this is the model thread, so can we actually keep it to that topic please. And yes, I know the app is derived from a model, but there again so is every forecast, hence the reason we say model discussion = model discussion, not loosely derived forecast icon discussion. Although does have a ring to it, I admit. 

[blizzard81]

2 minutes ago, northwestsnow said:

I suppose 4 days is a long time for pinning down specifics- so i hear Blue..

But no harm in looking at the potential

Looking at the potential is why we are here after all isn't it? Whether it be T0, T48, T120, T300

[Guest]

CONTINUING THE SEARCH FOR BROADER PATTERN CHANGES WITH CROSS-MODEL ANALYSES

In this post, I want to take a closer look at any general signs of pattern changes showing up in the models and attempt to put these into the context of some of the shorter, medium and slightly longer term periods that have been suggested as "possible" outcomes by some of our more experienced members. I shall focus on output from the 12z runs (I know that the 18z output will have rolled out while I'm writing this long post and I have't even glanced at it yet). I will be looking at day 6, 11 and 16 charts. I shall start off with the "big 3" models for surface pressure but UKMO only goes up to D6 and ECM up to D10. For the "Jet Stream" and Stratosphere charts I'll only refer to the GFS and GEFS output.

DAY 5: 1300 December 27th - Surface Pressure  

                           UKMO                                                      ECM                                                GFS Operational                             GEFS Ensemble Panel

                                  

I picked this period as it's just after the time when rather colder Polar Maritime/Arctic Maritime air has cleared southwards of the UK and the start of a cool unsettled spell. I'll leave it to other posters to comment on the UK and any possible wintry weather. I want to focus upstream and look at the Arctic pressure patterns as these are likely to have some significant impacts for us as we move through the first half of January. The big 3 all show the WAA that pushed across into the Arctic from the Canadian/Pacific/Alaska side. This is already disrupting the tropospheric PV.  As I discussed in several recent posts (including yesterday's on page 5 of this thread) my research of previous winters and split PVs suggests that it's not the angle of attack that is particularly important or whether we see a full split PV or how long it remains split (rarely more than several days) it's the impact of successive or multiple attacks that can result in the PV being very disrupted, displaced and thoroughly weakened. These three charts as well as all of the GEFS ensembles show the PV coming under huge strains with attacks from various angles even at this very early stage.

"If" this continued uninterrupted the warm air mixing into the Arctic would probably lead to some unusual weather patterns for much of the remainder of this winter. Cold air is released (or forced out of the Arctic and towards the middle latitudes (with North America normally the most favoured but sometimes EurAsia and sometimes both). There is generally a time lag of up to two weeks for the colder patterns to develop. Unfortunately, it's rarely as simple as that and much can go wrong. At this time of year the PV is usually approaching its strongest and most organised and the jet stream is also close to its seasonal peak strength and frequently taking a much more direct route close to or over the UK. Obviously this is not always the case but given the chance the PV will almost always try to revert to its climatological norm. Some of the frequently blocked and colder patterns that we have seen so far this winter, have been typical of an early winter La Nina state (I will not elaborate on that again now). Our posters who report on the background signals have highlighted for some time that this period (from late December through to early January) that we'll have this battle between a strengthening PV and some renewed amplification signals (again I will not go deeper into this, this time). Later in this post, I'll look at the jet stream and also the stratospheric PV. With these pressure patterns, I'm looking for signs of the continuing disruption to off-set or prevent the PV  from becoming a much stronger "beast". More on this below. 

DAY 10: 1300 January 1st - Surface Pressure  

                             ECM                                                GFS Operational                             GEFS Ensemble Panel

                         

Moving on to day 10, I selected this period as it might be getting very close to the critical period in terms of which direction we go in. The "winter on a knife edge" description really does seem to apply here. Will the PV strengthen or will it continue to come under attack? it could even be a bit of both. In fact my research shows that in a number of past cold winters, the PV does recover to some extent prior to rather more decisive disruption. Several winters have seen two or more cold spells but with a mild(er) interlude for exactly these reasons. The ECM and GFS operational charts do look rather different but, under closer inspection, not quite as much as one might think. The ECM shows the PV coming under renewed attack, mostly from the Russian/Asian side and it remains pretty weak. The GFS shows a slight strengthening of the PV and under less attack but still rather disorganised. It also has a strong build of pressure on the Asian side but with slightly less HP in the Arctic. Turning to the GEFS ensemble members, the vast majority show the PV remaining under various forms of attack with only a small number showing some strengthening. In fact some have it weakening further with one fully split PV and quite a few getting close to that position. Whether GFS operational is an outlier at this point or whether it's leading the way or not is far from certain (and subject to change of course).

DAY 16: 1300 January 7th - Surface Pressure  

                      GFS Operational                                GEFS Ensemble Panel

                   

Then we move into the unreliable FI period but we do need to see whether the GFS is moving closer towards the "climatological norm" with a strongly recovering PV or if it's still toying with the idea of a continuing disrupted PV pattern with colder evolutions still on the menu. While by the end of the first week in January we may well have a much better idea of where we stand, we cannot rely at all on the extended model output as so many critical factors are involved (I cover some of these later) and we do not know to what extent which of these is being factoring into the input. Let's see how the GFS is handling it on this particular run. The operational chart at D16 is not that different to D10. A slightly stronger and a little more organised PV but certainly not a raging beast at this stage. Again the ensemble members tell a slightly different story. About six show a stronger and more organised PV, ten show it disrupted or very disrupted, three have it half split and one fully split. So, at this point, we could still go either side of that knife edge. So, we need to look at other factors.                                                                   

DAY 5: 1300 December 27th - Jet Stream   

                     GFS Operational                                        GEFS Control                                     GEFS Ensemble Panel

                                   

The jet stream on the operational and control charts at day 5 is just starting to dip south of the UK which is ushering in the colder interlude. The ensemble charts are not available on the full northern hemisphere view but the North Atlantic view is good enough to go on. I am a little surprised that there is quite a variation in what should be the reliable period. Broadly, they all show quite a strong jet (certainly stronger than we've seen for a while) pushing from west to east towards the UK. Three members have the jet blowing right through the UK into the continent. The others showing it beginning to buckle and meander to various degrees. Most of them have the jet as it reaches the UK taking a more north-west to south-east tilt. Some of these are more north-north-west to south-south-east and one is north to south. Most of these have the UK just on the cold side of the jet (hence the rather cold to colder conditions for next week). Two show this more decisively with one of them showing a strong loop or "U" bend with the UK clearly on the cold side. Given this spread at D5, I would say that there is already some uncertainty creeping into the run.    

DAY 10: 1300 January 1st - Jet Stream   

                     GFS Operational                                       GEFS Control                                       GEFS Ensemble Panel

                               

By day 10, the op and control both show the jet weakening and buckling further, taking a more southerly route with the UK on the cold side. There is a much greater spread amongst the ensemble members. About six have maintained the stronger jet with three of these showing it blasting into the UK. The other three have it looping with quite a strong north to south shape across the UK (the stronger Arctic incursions shown for this period). Six more have the UK on the cold side of the loop. A few show some looping around Scandinavia (rising pressure there). Two have highly unusual strong loops taking routes up across Greenland, Svalbard and south-east into Scandinavia with one pushing out south-eastwards through eastern Europe - suggesting HP over central Europe (a "Euro-Slug") but the other one with it moving back on a north-east to south-west path through north France - indicating a Scandi HP (or close to it). A very mixed picture showing even greater uncertainty and an awful lot to be resolved.

DAY 16: 1300 January 7th - Jet Stream   

                     GFS Operational                                           GEFS Control                                    GEFS Ensemble Panel

                                

By day 16, the op and control have the jet further south, with some meandering and moderate strength with the UK on the cold side.Needless to say the GEFS ensemble members show a wide variety of evolutions. In fact the GFS seems to be exploring almost every possible outcome! There are northerly routes, southerly routes and a couple right at us with varying strengths. Almost all are looping, meandering or buckling to some extent. All this tells me that there is extreme uncertainty in terms of the GFS settling upon any the longer term outcome - yes, partly as expected in deep FI but with such a variety of jet stream outcomes shown in early January it does not really help us to see which direction we are going in.

I normally find that the jet stream is one of the better indicators during periods of high uncertainty but not on this occasion. So, we now need to consider other factors. Returning to the disrupted tropical PV and possible amplification signals versus the strong tendency for the PV ( and jet stream) to strengthen, what forces could have a decisive impact on the outcome? I covered most of these in my post (on page 5) yesterday. I will not go into all the teleconnection signals again other than to consider the impacts from the stratosphere (in a very simplified way). Up until a few days ago, we had been told that there was a disconnect between the stratosphere and the troposphere. As long as this continued (with changes in the various layers up there, any warmings and coolings) the attacks on the tropospheric PV might be able to carry on to allow the chance of the broader change to a more amplified pattern and likely colder regime to take place. Later on a mid to late winter warming or even a sudden stratospheric warming event might change things but we are dealing with possible changes running through early January. Then just two days ago, our strat specialists started to see changes that would very likely impact on this earlier period. There would be downwelling from the stratosphere through to the lower layers of the troposphere and towards the surface. The strat PV would start to imprint on the trop PV. This would likely return it closer to its more normal position nearer the Pole, cause it to strengthen considerably and  become much more organised. This would completely transform the outlook for much of January. I, however detect a little renewed uncertainty and there have been various conflicting signals. I haven't seen any more updates on the strat thread (I last looked early this afternoon) so not much to go on. I have my limitations in dealing with this but let's have a look at the GFS strat charts from the same time slots.

DAY 5: 1300 December 27th - 10hPa  Stratosphere Temperature (ensemble panels not shown on Meteoceil, so control and mean charts shown instead)  

                     GFS Operational                                          GEFS Control                                             GEFS Mean                

                                 

The GFS was showing hints of warming on successive runs for some time but mostly in the unreliable period. Then they latched on to a more consistent signal for warming in the more reliable period. Then a few days ago they temporarily dropped it almost altogether only for it to reappear again in the last few runs. Again ,this uncertainty gives me a little more encouragement that there is still much to be resolved. The other thing has been the position of the strat PV moving away from to right over the Pole and out again. At day 5, the operational, control and mean charts all show the strat PV centred south of the Pole over Svalbard. There are signs of modest warming on the north Asian side.  

DAY 10: 1300 January 1st - 10hPa  Stratosphere Temperature      

                     GFS Operational                                          GEFS Control                                             GEFS Mean 

                                

By day 10 all three charts show the strat PV edging very slightly north and closer to the Pole but the warming over north Asia has weakened considerably.

DAY 16: 1300 January 7th - 10hPa  Stratosphere Temperature   

                     GFS Operational                                        GEFS Control                                               GEFS Mean

                                

By day 16 we have some discrepancies showing up. All three charts show the strat PV moving much further away from the Pole. This has changed from the last few runs on day 16 charts which had it much closer to the Pole and "may" lead to a change of opinion from our strat experts? The operational chart shows strat PV centred just north of Scandinavia and stronger warming has reappeared, this time over north-east Russia/Siberia. The control chart has moved it even further and centred it over north-west Russia but with only slight signs of any warming. The mean (which averages out the op, the control and all 20 ensemble members)  is very similar to the control in terms of position and little warming showing (with obviously a greater spread over the ensemble members).

Now my point that I've suggested in several recent posts (but with no one reacting to or commenting on it) is that could this be a sign of the "downwelling" from the strat not getting down properly through the lower layers of the trop (or delayed, interrupted or even not at all and reversed)? I also "suggest" that with the trop PV in a pretty disrupted, displaced and weak state that this might make any re-connection or coupling far from a straightforward process? Furthermore, there have already been some indications that some of the much earlier amplification imprints (residues from the frequent periods with at least some HLB over the past 8 weeks and with the +PNA and mid-Atlantic ridge) might still be an embedded signature in the lower layers of the trop? Is this simply me as a "coldie" just clutching at straws or could there be something in this? I'm sure that many readers on the model thread would appreciate a fuller account and explanation from someone from the strat thread. This is so relevant for us in looking out for when the models start to sniff out these changes (if they haven't already).

Having delayed it for a few days, I will do one of my thorough temperature analyses some time over the weekend before I take a few days break over Christmas. I am really fascinated by the current unusual patterns and set up and I feel that whatever happens it's likely to be a learning curve for all of us and probably a very bumpy ride following the fluctuating model output. . 

EDIT: Please take note of @bluearmy's useful comment below this post which does show up my limited understanding of the stratosphere processes - although I would like the analysis that I outlined above from our strat experts + some more answers to some of the key points that I raised.

Edited December 22, 2017 by Guest
Correct typos and check charts

[bluearmy]

BB1962  - you can't rely on seeing the location of the vortex from the temp profile. Heights are visible on instantweathermaps although take care the further the run goes as output becomes less reliable (better than the trop the higher up you go but post day 12/13 I would be v cautious re over reliance on detail and rely on continuity)

Edited December 22, 2017 by bluearmy

[feb1991blizzard]

4 minutes ago, bluearmy said:

BB1962  - you can't see the location of the vortex from the temp profile. Heights are visible on instantweathermaps although take care the further the run goes as output becomes less reliable (better than the trop the higher up you go but post day 12/13 I would be v cautious re over reliance on detail and rely on continuity)

Pretty decent continuity 300-384 range wrt warming though, the question is though can we see the warming continue to intensify and get into the heart of the vortex the further we go, once it starts to wane before disrupting the vortex, that's usually the signal that its not strong enough a warming from past experience, also temps need to be up near freezing point right into the polar regions.

[Guest]

Having produced another long post, I'm suffering from post and model fatigue! Just had a quick spin through GFS 18z and saw this:

What on earth is Marge Simpson doing in the Arctic!

Goodnight!

[ArHu3]

I'm guessing the mjo won't help us for at least a month ?

http://www.cpc.ncep.noaa.gov/products/precip/CWlink/MJO/CLIVAR/ALL_emean_phase_full.gif

Edited December 23, 2017 by ArHu3

[feb1991blizzard]

GEFS are zonal rubbish.

[knocker]

Most areas over the weekend will be cloudy with patchy drizzle in the west in a quite breezy moist south westerly airstream with temps a little above average. The exception to this will be north west Scotland where a trailing front is loitering (and later a low pressure area) thus rain which becomes heavier over the weekend and moves slowly south to affect N. Ireland and also quite windy.

By 12 Monday the wave is over the north Irish Sea with the front down across Ireland and west of Cornwall with England and Wales in an increasingly brisk south westerly Over the next 12 hours the low deepens to 980mb and tracks east into the North Sea whilst the front, and rain, traverse England and Wales clearing the east in the early hours of Tuesday. It continues to deepen and track east and by 12 Tuesday has initiated some very inclement weather into southern Norway with possible blizzard and gales in Denmark

Whilst all this is going on another wave forms in mid Atlantic, and driven by a very strong low level jet,swings around a transient ridge, to be 987mb over southern England by 00 Wednesday. The precise track is quite important as it could well produce snow along the northern edge as it tracks into the southern North Sea leaving the UK in a much colder airmass.

After this it all becomes very complicated with another wave and front arriving in the west by 12 Thursday which could also bring some snow along the northern edge But the whole shebang is dominated by the displaced vortex lobe in North America and major upper troughs across the Atlantic and the second vortex lobe Franz Joseph Land and the trough running south through Scandinavia to southern Europe. Ergo a very complex surface analysis that I'm not even going to attempt to analyze as a chart will suffice. And certainly best left here.

Edited December 23, 2017 by knocker

[geordiekev]

Just a couple of charts from the 0z that caught my eye 

[knocker]

Further to my earlier post the lobe over N. America is dragging some cold polar air a long way south as New year approaches.

Edited December 23, 2017 by knocker

[knocker]

Well you really couldn't make this up. The ecm has a wave forming on the front north west of Coruna at 12 Monday and by 00 Tuesday it is over St Malo 986mb with gales north west France. The front is Scotland/Wales/Cornwall with heavy rain over the latter. It continues tracking north east into the North Sea, via Kent, before merging with the other low by 12 Tuesday. It then tracks the Atlantic wave further south and whips it up the Channel Tuesday evening through Wednesday night bringing rain along the south coast.

Edited December 23, 2017 by knocker

[northwestsnow]

And todays big winners on this mornings snow-code lottery (ec snow depth charts ) is.........the midlands!!!

There are some big differences on the 850s between 96-120 this morning-

UKMO96

120

EC96

120

UKMO quite a bit colder-could be the difference between cold rain and cold snow for some !!