Model output discussion - 20th Feb onwards

[Polar Maritime]

Lets please keep discussion polite and on topic, Thanks.

Edited March 8, 2016 by Polar Maritime

[MP-R]

33 minutes ago, Timmytour said:

Think Tim knows that. Look at the The "shifty" bit of his reply as ironic comment aimed at both the models and the radar projections rather than anything aimed at you. That's how I read it anyway

Fair enough. I never set out to annoy anyone and don't consider people lying if they say it's snowing or anything else for that matter. No offence intended. Obviously got a bit lost in translation. Moving on...

[jethro]

4 hours ago, feb1991blizzard said:

Cold Knocking on the door of our South West members (pardon the pun)

I've got the door opened wide, it's welcome to step inside.

[bobbydog]

32 minutes ago, jethro said:

I've got the door opened wide, it's welcome to step inside.

you might want to shut it again for a while....

[feb1991blizzard]

1 hour ago, jethro said:

I've got the door opened wide, it's welcome to step inside.

I thought for one minute you had invited me inside your house then!!!!!

 

3 hours ago, Singularity said:

I think a straight-forward replication of the strat. pattern in the trop. would produce the convective easterly - but as usual there looks to be complications such as how a few peripheral lows interact with the main Siberian trough as it drops south and then possible edges west. An undercut of blocking N/NE of the UK by either that or a secondary low into Scandinavia is needed to import air that is cold enough, and without too much modification of the lower levels by the North Sea. Perhaps a 1 in 4 chance of all the pieces coming together prior to week 3 of March. I have to add that limiter because the longer range is just impossible to fathom at the moment beyond the idea that a standard zonal regime is the least likely outcome.

Thanks, it looks like the GFS 12z is taken notice, much better run, better shaped Norhern blocking means aa nice cold plunge into Europe ready for us to tap into late month.

[jethro]

1 hour ago, bobbydog said:

you might want to shut it again for a while....

I think I might have the odd tree barricading it shut for me!

[jethro]

23 minutes ago, feb1991blizzard said:

I thought for one minute you had invited me inside your house then!!!!!

 

If you come weighted down with a sack or three of snow for me to play with, you're more than welcome.

[Singularity]

Well, the GFS 12z is the best example I have ever seen of the model failing hopelessly to handle a split jet scenario.

At this point, with height rising rapidly in the arctic, the UK high should link up and the Atlantic trough be forced to disrupt as one arm of the jet undercuts into Europe.

FAIL.

I'm rarely this blunt but come on GFS, get your act together. If you want to see how it should work out, check out the ECM 00z:

No sudden return of Atlantic westerlies there!

[Singularity]

 

Well this is most amusing. Quite the event if it comes off, but that would be in the face of Met Office expectations given that it's shown to be ice/snow right down to the coast - I believe these models do try to represent what we see at the surface based on terrain height? If not then this may be a red herring for areas below 2-300 m based on Met Office warning details.

[Summer Sun]

High pressure starting to dominate our weather from the end of this week makes a nice change saying that rather low pressure dominating!

Edited March 8, 2016 by Summer Sun

[fergieweather]

12 minutes ago, Singularity said:

 

Well this is most amusing. Quite the event if it comes off, but that would be in the face of Met Office expectations given that it's shown to be ice/snow right down to the coast - I believe these models do try to represent what we see at the surface based on terrain height? If not then this may be a red herring for areas below 2-300 m based on Met Office warning details.

No - it's actually exactly in line with UKMO expectations, which are, quoting the Chief Forecaster earlier this afternoon:

"Allow for temporary snowfall – maybe even heavy wet snow/sleet at times – within the warning area, mainly in its southern half, with some possible (though low point prob) accumulations of 1-2cm at low-level. This should melt again quickly, especially as ppn readily turns back to rain as intensities ease slightly. Greater risk (and possible slightly higher accumulations, say 2-4cm) on higher ground along this track – Salisbury Plain, Mendips – incl Bristol Airport, Cotswolds and perhaps even the Chilterns. All of this low confidence and again the track of the Low is critical."

Tricky, in short.

[Singularity]

A few more interesting charts

  

Those areas that see the snow on the ARPEGE run wake up to temps of 8-9*C so it would be truly bizzarre to see snow just a few hours later! I have added in HIRLAM which also has that snow - and this model is associated with the Met Office I think?

  

The idea of mid-teens for Friday seems to have been lost and annoyingly this carries on through the weekend for all but a few lucky spots in the northwest corner of England and southwest corner of Scotland. 

This being a consequence of the high being adjusted further north with more of a flow off the North Sea. GFS still breaks up the cloud well so that's something - though east coasts may see a lot of murk unfortunately. Slight risk of frost overnight if skies are clear enough.

Edited March 8, 2016 by Singularity
more is more

[Singularity]

9 minutes ago, fergieweather said:

No - it's actually exactly in line with UKMO expectations, which are, quoting the Chief Forecaster earlier this afternoon:

"Allow for temporary snowfall – maybe even heavy wet snow/sleet at times – within the warning area, mainly in its southern half, with some possible (though low point prob) accumulations of 1-2cm at low-level. This should melt again quickly, especially as ppn readily turns back to rain as intensities ease slightly. Greater risk (and possible slightly higher accumulations, say 2-4cm) on higher ground along this track – Salisbury Plain, Mendips – incl Bristol Airport, Cotswolds and perhaps even the Chilterns. All of this low confidence and again the track of the Low is critical."

Tricky, in short.

So a revision from the previous guidance of snow down to 300 m ASL?  Thanks for the update 

The threat of coinciding with the rush hour must be making them hot under the collar this evening!

I was about to add that rain is expected to follow the snow and persist for a great many hours, so any accumulations would soon be gone again with no time to really appreciate the results. 

HIRLAM tries to turn the precip. back to snow in the evening, but it seems to be overly fond of doing that via diurnal cooling.

WRF-NMM has only a very small area of snow in line with Weymouth for a short time. So enough uncertainty to cause headaches 

[fergieweather]

6 minutes ago, Singularity said:

So a revision from the previous guidance of snow down to 300 m ASL?  Thanks for the update 

The threat of coinciding with the rush hour must be making them hot under the collar this evening!

I was about to add that rain is expected to follow the snow and persist for a great many hours, so any accumulations would soon be gone again with no time to really appreciate the results. 

 

Yes, tricky. The issue here is that various models generate very little snow, but UKMO were suspicious of that - considering (even their own models) to be "over optimistic", as they termed it, in not generating the risk sufficiently. So, they've run the very clever tools they have at Ops Centre that employ the Wilderspin and Lumb methodology to gauge potential of model rain signal changing to snow, given varied background parameters.  This experimentation and tweaking showed just how sensitive the relationship is between intensity and PPN phase within the yellow (rain) warning area, and that's even allowing for the strength of the wind, which normally would create sufficient mixing in the boundary layer to minimise risk of snow getting down much below the WBFL.

[Guest]

POSSIBLE COLDER IMPACTS OF THE SSW EVENT, WAYS TO MONITOR IT AND WHAT THE MODELS MIGHT TELL US

The Models are Struggling:

Some of the models are perhaps just starting to toy with the forthcoming impacts associated with the major stratospheric warming event which is now well underway. There have been several charts posted here today offering some of the possible scenarios, which are all over the place in terms of HLB or flatter more zonal patterns and colder or warmer possibilities. I feel that we shall see further wild swings in the different model outputs during the next few days and particularly in what is currently the 6 to 10 day period (T+144 to T+240).

I closely follow Gibby’s excellent daily reports and always take note of the 31 day mean model verification statistics. I have noticed a big change in the 10 day reliability. This was in the low 50s until just a few days ago and has dropped rapidly into the mid to high 40s since last weekend with ECM at 46.5% and GFS at 46.4% this morning. As this is a 31 day mean, this could be due to some much better 10 day verifications a month ago falling out the equation or, as I strongly suspect, is due to a very poor recent period. If this is correct (perhaps someone can examine the daily figures more closely, to ensure that I am not making an unrealistic assumption) then the figures for the last few days are more likely in the 30s! This would reflect the enormous uncertainties the models are currently facing and have been for the last 10 days or so. Even Fergie admits that GloSea5 and EC-32 have been struggling and these are the most respected, reliable and sophisticated computer models out there.

Following on from Dr Judah Cohen’s AO Report:

Although the current SSW event is far from straight forward in terms of intensity, timing and impact, I shall continue to go with Dr Judah Cohen’s latest predictions which I commented on yesterday (see my post on page 65). Here is the link again for those of you who missed it: http://www.aer.com/science-research/climate-weather/arctic-oscillation

If Judah Cohen is correct we should expect to see some strong HLB in the Arctic developing during the next week and becoming dominant by around mid-March. He expects a lobe of the polar vortex to move down into Europe with lower heights from northern Scandinavia extending right down into central and southern Europe and another lobe of the vortex pushing into Siberia. He thinks that there will be a very long fetch easterly established almost all the way from the north-east Asia coast right through to Western Europe and the UK.

How Cold Might it Get in the UK?:

Judah Cohen is careful not to overstate the intensity of the cold. For the UK he talks of “relatively cold” initially which I would take as “for this time of year”. A little later, once the easterlies are established, he refers to “cold polar continental air”. Again, he does not say how cold. He seems much more certain about the longevity of the cold spell “lasting well into April”. He does mention that this set up would bring snow in January but more like “cold rain” in April. With this type of broad easterly pattern, the UK might be favoured for a lot of dry weather (as Singularity said yesterday was “one” of the possibilities). Initially, we might be under northerlies or north-easterlies being on the western edge of the low heights spilling down into Europe. This would depend upon how cold the source of the “Arctic” airstream is. There could well be uppers in the -10c to 15c range but as we will be in the second half of March, this could be cancelled out by the ever increasing solar energy. A cloudy airstream would supress daytime temperatures. Given the right more local set-up there could be some convectivity producing on shore showers (possibly wintry) and with any troughs or disturbances perhaps leading to some more widespread wintriness moving inland.

The same might apply later on with the more direct easterly blast which might have even colder uppers. March 2013 did give many days of stratus cloud with often very dry conditions but there were some significant snow events in parts of the UK. Even with similar synoptics to that epic cold spell, we might find that the temperatures are just that bit higher.

Last week I examined some of the possible colder scenarios in 2 posts (part 1 on page 48 of this thread on Tuesday, March 1st and part 2 on page 51, the following day). I looked at some of the key factors that would need to be right to deliver sufficient cold. This included, the right synoptics, the source of any cold air, whether it is cold enough and assuming that we can tap into it,  establishing a deep European cold pool and much of Europe developing a decent snow cover to help maintain any cold pool. All those factors were prevalent when the March 2013 cold spell started but we are nowhere near these right now.

Some Cold Indicators:

I will not repeat what I said in last week’s post on this subject but I will now update several of the indicators and links that I used then. It seems to me that the initial cold spell that Judah predicts may come to us on a northerly or north-easterly from the high Arctic. I looked at Svalbard as a good indicator of surface temperatures in our part of the Arctic. I showed that with much of the Barents Sea and Kara Sea being almost ice free and following a record warm Arctic winter temperatures in Svalbard had been much above average almost continuously not just for winter 2015-16 but for almost all of the last 3 years. Almost all the Arctic deep cold has been on the opposite side of the North Pole.

Firstly, here is the current Arctic temperature chart

GFS – Northern Hemisphere Current Temperatures (as at 1900 today, March 8th) :

http://www.meteociel.fr/modeles/gfse_cartes.php?ech=6&code=0&mode=9&carte=1

*** Please note:   At the time of posting this link it was showing 1900 on Tuesday, March 8th. The charts are automatically updated 4 times a day, so the temperatures shown will be different to those I just mentioned above. As this is GFS and just one of the models, I am not really focussing on forecast temperatures this far out. It is vital to note the time of day to take account of daytime/nighttime variations. So for like for like comparisons, for example the 1900 charts for each day should be available to view from the 12z (T+6) updates which are published around about 4 hours later. This also applies to some of the other charts below that I link to in this post.

The deep cold is still on the other (wrong) side of the Arctic but we will need to see this shift towards Svalbard. Now let’s take a look at the 3 Svalbard stations and their 9 day temperature trends as “predicted” by the Norwegian met office (I have stated the short, medium and longer-term forecast temperature for each station below each link) .

Central Svalbard – Longyearbyen 28 m asl:   

http://www.yr.no/place/Norway/Svalbard/Longyearbyen/forecast.pdf

March 9th - -4c;   March 13th  0c;    March 17th  -12c.

West Svalbard – Ny-Alesund:                                                            

http://www.yr.no/place/Norway/Svalbard/Ny-%C3%85lesund/forecast.pdf

March 9th  -7c;   March 13th  0c;    March 17th  -13c.

South Svalbard – Sveagruva:                                          

http://www.yr.no/place/Norway/Svalbard/Sveagruva/forecast.pdf

March 9th - -4c;   March 13th  -2c;    March 17th  -14c.

These links also update automatically several times a day.

This looks very good to me – very much colder in 9 days’ time but I think that we need to see -15c to -20c in that region, if not lower. The Norway met office’s predictions for March 17th are more or less in line with Judah Cohen’s timing of the initial SSW surface impacts. In fact they also show the warm up this weekend before the cold deepens significantly. Of course, we need to be aware that these are only a forecast that is subject to change and I am told that the Arctic surface temperature forecasts are not completely reliable even at short range. We are, however, looking for the general trend. “Singularity” highlighted yesterday that we need to see the Arctic high building in the Svalbard area. So, we need this as part of the HLB sequence of events and we need much deeper cold up there. I shall keep a close eye on the figures and provide several updates between now and March 17th, assuming that the rest of Judah’s predictions are looking to be falling into place.

I shall also start looking again at the Eurasia snow cover. I may, if I have time, report on this tomorrow. No more time now, as want to give a brief update on the AO and MJO ensemble charts within an hour or so.

Please remember that this is my interpretation of Dr Judah Cohen’s predictions. He and I could both be wrong but I feel that these charts will assist the monitoring of the initial SSW impacts together with all the usual model analysis that appears on this thread.

 

Edited March 8, 2016 by Guest
Spelling, missing words grammar.

[The PIT]

ECM showing plenty of dry settled spring weather cool by night but pleasant and warm during the day as does the GFS although the ECM shows it to be more in control than what the GFS suggests. It does look like a prolonged settled spell is on the cards.

[shotski]

27 minutes ago, The PIT said:

ECM showing plenty of dry settled spring weather cool by night but pleasant and warm during the day as does the GFS although the ECM shows it to be more in control than what the GFS suggests. It does look like a prolonged settled spell is on the cards.

It would be nice but that high will probably be ladened with cloud and we'll struggle to get double figures. 

[Frosty.]

It will soon be good riddance to low pressure and a warm welcome to our first proper anticyclonic spell for a long time according to the Ecm 12z this evening. High pressure builds in and becomes centred over the UK and intensifies, there are signs of retrogression later but the high stays in control out to T+240. After all the unsettled wet and windy filth we have had through the winter that never was, it  looks like it will become very pleasant with spells of sunshine and lighter winds..Bring it on!

[Guest]

AO Ensemble Charts: 

http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/ao_index_ensm.shtml

A slight change tonight with several more ensemble members remaining positive in 10 days time. On the other hand several of the members that were back into negative territory by March 15th, go more strongly negative. This wider spread of outcomes must be reflective of the even greater uncertainty of the timing and impacts of the SSW. Overall, NOAA are still favouring some HLB. I imagine that the “mean” 8-14 day anomaly charts will not show either scenario strongly. To block or not to block, that is the question? We will go one way or the other and perhaps there is no halfway house! I would love to see individual ensemble members from NOAA as I believe that the “mean” anomaly charts right now are not very useful!.

If they are available, please can someone post them.

MJO Ensemble charts:

Here are todays MJO ensemble charts for the big 3 and Kyle MacRitchie’s for March 8th. 

UKMO: http://www.cpc.ncep.noaa.gov/products/precip/CWlink/MJO/CLIVAR/ukme.shtml

ECM: http://www.cpc.ncep.noaa.gov/products/precip/CWlink/MJO/CLIVAR/ecmm.shtml

NCEP/GEFS:  http://www.cpc.ncep.noaa.gov/products/precip/CWlink/MJO/CLIVAR/ncpe.shtml

Kyle MacRitchie: http://www.kylemacritchie.com/real-time-maps/realtime-mjo/

UKMO has changed very little but does “just” avoid the circle of death now albeit at very low amplitude. They enter phase 2 tomorrow and move steadily on to phase 3 and phase 4 before stalling.

 ECM is yet again very much in line with UKMO until it reaches phase 4 at marginally higher (still low) amplitude. Then they go for phases 5 and 6 but with a much wider ensemble spread. 

NCEP/GEFS take quite a different view which has changed somewhat from yesterday. One member stays in phase 1 and seems to sticks there at good amplitude (this might be a charting error?). All the other ensemble members, after leaving phase 1 tomorrow, enter phase 2 for 3 days and then dive into the circle of death but the stalling there is briefer than shown yesterday. There is then a wide spread of outcomes with one ensemble members entering phase 5, another to phase 6, 2 to phase 7 and the majority straight back towards phase 8 but at uncertain timing and amplitude. Overall, this model’s outcome is still pretty confused and must reflect considerably uncertainty.

Kyle MacRitchie continues to take a much more positive view. He shows the MJO only leaving phase 7 and entering phase 8 today and then pushing through phase 8 and on into phase 1 on March12th (slightly slower progression than yesterday) and then on to phase 2 around March 14th/15th (slightly earlier than yesterday). All 4 of his ensemble members progress all the way though phases 3, 4, 5 and 6 between about March 23rd to April 2nd. Then they re-enter the important phase 7 around April 3rd/4th. They retain good amplitude throughout the whole period and one member goes into very high amplitude.

General Comment:

All 4 charts show some changes from yesterday. I wonder if GloSea5 has changed from its movement and stalling in the circle of death. So, we really have 5 rather different model predictions! I wonder if Kyle MacRitchie is on to something or is he completely wrong, continuing to defy the big 3 and GloSea5? There is huge uncertainty with the SSW. Ideally we would like to see the MJO in phases 7, 8 and 1 to favour the HLB, particularly in the North Atlantic and the Arctic.

 

Please can Fergie state if the GloSea5  MJO charts that he posts are based on one model like the UKMO or take account of all the main models and then, perhaps, modify these predictions according to their parameters and assessment. Do they take account of the ENSO impacts like Kyle MacRitchie?

 

Edited March 8, 2016 by Guest
The ECM link should work okay now

[Singularity]

The cross-polar heights are missing on this run. There certainly is a strange lack of tropospheric response to the SSW on this run; you'd expect to see a strong height rise at the high latitudes, not a weak ridge with low pressure continuing over the pole.

Could it be that the rather fast movement of the stratospheric pattern is preventing the models from establishing a defined change in the troposphere; they get pulled in one direction, then another, as the location of displaced vortex remnants and the upper level ridge changes at a considerable pace?

The way this 12z ECM run evolves looks more like an MJO-driven amplification via an upstream trough digging down in the Atlantic. It has a weak MJO in phase 2 and that does produce such a response despite the low level of activity:

 

Note the proximity of the trough in the Atlantic - perhaps a reason for the trouble some ensemble members have had shutting the Atlantic out (that signal for lowering heights that Fergie has advised caution over)?

Interesting to then consider what a greater level of activity in phases 3 and 4 might lend support to:

 

...this being something UKME is looking keen on. GEFS is a bit 'out there' with it's idea of the MJO fading out and then looking to re-emerge in the Central Pacific, but after it's success with the recent 7/8 activity, I hesitate to rule that idea out. It does explain some of the added momentum that the det. run finds in lower-res, though I'm pretty sure it went too far with it.

Still a high risk of unseasonably cold conditions in the second half of this month, even with that chance of snow. I saw a whole afternoon of light snow on 4th April 2013, albeit not settling. It carried on as pure, almost powdery snow even when the sun broke through the clouds a bit around noon, which left me staggered. That such can happen at just 50 m elevation in CS England, a fortnight after the spring equinox, reveals the power of a deep pool of cold continental air - and if you're imagining that the uppers were something like -10 or lower on that day, think again;

-7 to -8*C at the 850 hPa level. This is why it raises my eyebrows when I see the output for mid-March suggesting uppers getting to -10*C or below over Scandinavia and with an easterly flow to the UK. That's what ECM would look like for day 11 I expect.

Yes, most likely nuisance cold, but fascinating from a meteorological standpoint. I do feel sad for the wildlife though - mixed emotions on such events.

[Captain Shortwave]

Another day and yet again we sit with cross model support for heights remaining near or over the UK at day 10.

ECM(update ), GFS and GEM

      

GFS and GEM pretty similar, the GEM probably more progressive in terms of where we go from there. The ECM shows a ridge in the Atlantic moving north east and linking with the high over the UK. Could we see retrogression or does the entirely lot get flattened by the low moving towards Greenland with a strengthening of the jet.

Overall I remain unconvinced of any movement away from the settled theme. The details at the surface look uncertain as slight changes in orientation and position could mean the difference from a sunny high with a continental feed and a cloudy one with low cloud pushing in off the north sea. There remains the risk of cold air being pulled towards the UK even in the semi-reliable timeframe as illustrated again by the JMA. Though again the precision position of the high will determine whether we can fully establish the southern arm of the jet.

Edited March 8, 2016 by Captain Shortwave

[metrosnow]

You have to admire the work put in on this thread from the honourable members from Exmoor & The New Forest above. This forum is all the better for it. Many thanks to you both.

[phil nw.]

The projected 500hPa pattern over the next couple of weeks continues to defy the warmings and wind reversal going on higher up.

If we look at the last GFS/ECM Strat forecasts at day 10 we can see how the upper vortex has been demolished over the pole-this at 30hPa-

 

but at the lower levels of the Strat we can see that familiar axis of twin vortex segments over Canada and Siberia-albeit in a less robust state.I should add these are the last runs available so are a day or so behind now but they are suitable for this purpose i think.

Looking at the mean day 10 GFS and the NOAA 8-14 forecast heights 

we can see the NH pattern close to the lower Strat pattern so i think currently this is why we are not seeing the polar heights as the down welling of zonal wind reversal is not as yet complete.

This may still filter down in time but the last ECM forecast we have  does not yet show this

 

the blue areas are the -ve zonal winds.

So what we see is a blocking high both at the lower Strat. and troposphere level forecasted around the UK locale from the week end and continuing through the following week. There does look a trend to gradually send more of the vortex split down through the Siberian side but unfortunately if looking for a late dose of deep cold the main thrust of Arctic air looks like spilling much further east into E.Europe and Russia.

 

[knocker]

The search for the elusive cold easterly continues apace but alas still no sign.

NOAA 8-14 this evening has the weakening HP to the west and still a slack flow from a westerly quadrant. The GEFS and EPS 10-15 are in fact quite similar with the trough in NE Europe with the colder air. Both spread the LP west by the end of the run although not the trough significantly with the EPS going zonal and the GEFS building heights south of Greenland thus veering the wind. Thus the weather turning more unsettled and the temps returning to around average.

[Singularity]

Thanks for your take on things Phil 

I'm aware of the lack of propagation down but that's just the thing - for a SSW that splits the vortex clean in two, there really should be propagation right down to the troposphere and at a fair old pace too. Unless, of course, the theory I have read through is not entirely correct or has overlooked possible exceptions to the rule. On the other hand, Fergie has mentioned the issues with models struggling to handle the SSW (though the latest such advice was a few days back I think) and past experience does suggest that for whatever reason, the propagation mechanism is something the models have a hard time capturing effectively.

The actual outcome this month will have major ramifications in terms of how far the models have progressed since the last major SSW in 2009.

 

Edit: @knocker just wanted to check, is this an accidental contradiction in your post? "GEFS building heights south of Greenland thus veering the wind. Thus the weather turning more unsettled and the temps returning to around average"

tia.

Edited March 8, 2016 by Singularity