Model output discussion pm 13/01/2016

[nick sussex]

A quick look at all the main outputs in terms of pressure, 850's and precip , you can clearly see how much of an outlier the UKMO is in terms of pressure:

The good thing about these meteograms is you can see the differences between the 00hrs and 12hrs outputs projections.

You can find these here for other UK locations aswell:

http://meteocentre.com/models/get_mgram.php?stn=London&mod=compar&run=12&var=std&lang=en&map=UK

 

Edited January 15, 2016 by nick sussex

[gottolovethisweather]

4 minutes ago, johnholmes said:

apologies for quoting this again, it should really be in the other thread but for checking will it snow look at the Guides, it really does work, IF the model has it correctly but best left until no more than 24 hours away and maybe better still at no more than 12 hours away. Remember how hard it is to forecast rain in the right place at the right time and intensity-add 7 more variables and you can see how complex it is!

 

Indeed within 24 hours of the event, I and a great many others will and should be revisiting that thread. Thanks for the reminder John.

[Catacol]

3 hours ago, Nouska said:

The models at 120 - serious resolution issues there.

GFS   UKMO     ARP 

Haha - brilliant, isnt it? You couldnt make it up! They have no idea whatsoever what is coming. And neither do we. I'm loving it. It's like forecasts from 100 years ago... not that any of us (I think) were alive then, but people must have routinely taken forecasts with a massive pinch of salt back in those days. 

It'll be a bit gutting if we get nothing out of this chaos pattern over the next couple of weeks. And February still looking ok for cold..

[Spurry]

The GEM 12Z is by far the best for coldies,the Atlantic never really breaks through and then the HP to the east re-asserts itself ,granted it's in the minority but given the last few days of model switching who's to say it won't be the right one!!

[nick sussex]

The ECM postage stamps are in:

http://old.ecmwf.int/samples/d/inspect/teasers/samplers/banner/mean_sea_level_pressure_and_temperature_at_850hpa!120!pop!od!enfo!enplot!2016011512!!/

A quick count around 15 members  at T120hrs have high pressure further north, some of those also bring a chance of more snow in from the west/sw.

Theres also another 5 that might bring some snow in but are more borderline as to what would happen post T120hrs.

Edited January 15, 2016 by nick sussex

[I remember Atlantic 252]

6 minutes ago, Spurry said:

The GEM 12Z is by far the best for coldies,the Atlantic never really breaks through and then the HP to the east re-asserts itself ,granted it's in the minority but given the last few days of model switching who's to say it won't be the right one!!

how's GEM rated guys? I class it as the ultimate cannon fodder model

[phil nw.]

You know peeps in spite of all the ups and downs recently the models still show the milder trend later next week.

Going back a while the cold spell, such as it is, was modeled to commence around 12/13th January nationwide and has duly arrived.

Pretty much all ensemble modeling was looking at a duration of around a week maybe a day or 2 longer so say to around 20/21st January.

I would say GFS/ECM Op runs have gone for that over the last few days but i think maybe a lot of us-incl.me to a certain extent have pushed this to the backburner-  and got caught up in the possible battleground that is building btween the cold and the Atlantic.That's understandable as many of us are looking for some snow.

A look at the t72hrs fax and already we can see the approach of the milder Atlantic air to the south west behind those fronts

a look at the 2 main models at t96hrs

show further inroads being made into the UK.

Yes there are subtle differences in speed and it could well be the ECM will delay things another day or so but we can see already the cold uppers retreating into the N.Sea by then.I would think ECM would have the better chance of any snowfall as the surface flow is better angled off the continent for a day or so longer than GFS on this run.

A look at the gefs 2mtr temperatures for Warks.

also support temps rising by the 20th.

I think the current uncertainty is more around how the breakdown happens in the next 3/4 days- whether there will be much snow and so on and will it be delayed another day or 2- but i believe now it is coming later next week.The milder interlude may not be for long though.

 

 

[bobbydog]

1 hour ago, Ali1977 said:

Is it time to start looking for the next cold spell, been a bit of a damp squib so far for many, especially down South....5-6c daytime temps isn't really that cold nor is -1 to -2c nightime temps.

to be fair mate, its only been like that for a couple of days and it might not be siberia but considering we had a constant 12c for about 6 weeks, its bloody freezing now!!

[PolarWarsaw]

1 hour ago, The PIT said:

Well in FI land all models going to for a return to milder weather.  UKMO probably now being the most progressive of the the big three.

Just like they did at that range a week ago..and the week before that...and so on. The default weather in this country is mild. The chance of having a cold snap (as we are) and cold showing in ALL model runs from T0 to T360 is erm...somewhat unlikely. 

[chicken soup]

9 minutes ago, Greenland1080 said:

Fantastic model rated by the experts

Supposed to be very good around the northwest atlantic area ..its a canadian model.

Edited January 15, 2016 by chicken soup

[bobbydog]

1 hour ago, The PIT said:

Well in FI land all models going to for a return to milder weather.  UKMO probably now being the most progressive of the the big three.

well... not strictly true...

[Spurry]

14 minutes ago, I remember Atlantic 252 said:

how's GEM rated guys? I class it as the ultimate cannon fodder model

You're probably correct , but i'll clutch at the straw

 

[Singularity]

Not much I can add this evening really. Another decent ICON run is worth viewing on Meteociel but really we can't put much faith in the details despite how many model runs have most of the snow falling across central parts on Monday - I have seen such consistency come to nothing with this sort of thing before. Notable changes nearly always occur within 48 and then 24 hours.

 

Thinking about how unreliable the model output for next week is based on numerous expert opinions, another clue comes from the way the MJO has behaved compared to projections. Remember yesterday I noted ECMF keeping the MJO close to the COD and GEFS only slightly more amplified? Well, the actual behaviour has turned out to be close to the notably more amplified UKM:

Which suggests a poor handling of the tropical forcing by the models. Though that's not really news to some on here I suppose! Continue the trends and it could be back in phase 6/7 before too long.

Curious that UKM has once again shown a +24 hour solution that's basically an extrapolation from the previous 24 hours. It was exactly like that yesterday as well. I'm not sure how genuine this MJO signal is though, in light of GP writing about residual signals in the Pacific the other day...?

I know I asked a similar question yesterday, but if I got a reply it was lost among the many as nobody quoted me (i.e. there was no notification). Sorry if it is a repeated theme.

 

Edit: incidentally, the MJO phase 3 composite for an El Nino January is a nice modification from the pattern evolving this weekend.

 

If only the signal was strong enough!

Edited January 15, 2016 by Singularity

[nick sussex]

Further to Phils post the ECM ensembles for De Bilt do show a  signal towards next Friday to start lifting the temps in that region.

There is however a few more members hanging onto the cold. In terms of the UKMO output theres not much support in the ECM ensembles for its solution at T120hrs.

I don't know what MOGREPS has for it but currently it does look like an outlier solution.

[phil nw.]

1 hour ago, The PIT said:

Well in FI land all models going to for a return to milder weather.  UKMO probably now being the most progressive of the the big three.

In fairness the output we see from the UK model has now come into range of the change Pit so will now show the block giving way in it's later frames now.

[ANYWEATHER]

Evening All!  The Semi Reliable time frame comes in at T+72. Unfortunately the Atlantic comes in but judging by the data we have most will have rain! Unfortunately the Atlantic wins for a while, There is a a very strong  block to the East ......... and if you look at the gfs right at the end of the run , it will make Coldies spark with delight....

Edited January 15, 2016 by ANYWEATHER

[DAVID SNOW]

9 minutes ago, bobbydog said:

well... not strictly true...

Well.. yes it is true, the Pit commented on the UKMO model which goes out to 144 hours(does go further out though, but we cant see that) and the other 'big' models  which all now show a return to milder conditions in the 5 to 7 day range, that's far enough into the future for me  GFS @384 hours was not his point I think.

[Bristawl Si]

Anyone thinking the Atlantic will just bulldoze its way through during Monday should pop into the SW thread for a read. Still uncertainty.

[A Frayed Knot]

Fergies' post just now from the SW Thread .

Key uncertainties begin from T+84. It's impossible to gauge outcome on regional level. MOGREPS has snow risk Taunton eastwards for SW/W Country by Monday; WBFL modelling by then considered highly unreliable as for now are UKMO Best Data Tmin/Max as shown via BBC graphics. A concerted W'ly push across all the south Sun night-Mon night only rated 30% prob at present. Rebuild of cold block W'wards into Tues complicates matters enormously. Some folk on forums are doubtless slavishly following/making rash pronouncements based on modelled 850hPa temps as the singular 'snow line' delineation but it's not: the higher Theta-W's will *override* cold boundary layer, offering a very difficult forecast mix ranging from rain-sleet-snow and, moreover, profiles suggestive of freezing rain in places, too. PPN phase algorithms (eg snow charts from EC, GFS, ARPEGE etc) won't synthesise these nuances at all well - yet - due to their granularity and the ongoing output uncertainty at that range. However, currently a fair signal for fronts generally weakening as they run east, and only a minority of ENS presently offer *significant* snow amounts (these in central-E UK where cold air may linger much of the week, whilst W/SW back earlier to average/above average temps). So, fun and games looking likely for my forecast colleagues Sun eve on into Mon, but thankfully I'm not having to deal with this headache until back on shift later Monday!!

[bobbydog]

6 minutes ago, DAVID SNOW said:

Well.. yes it is true, the Pit commented on the UKMO model which goes out to 144 hours(does go further out though, but we cant see that) and the other 'big' models  which all now show a return to milder conditions in the 5 to 7 day range, that's far enough into the future for me  GFS @384 hours was not his point I think.

he said "all the models in FI land" (which includes the whole range of their output) so as i said- not strictly true.

i should add, i'm fully aware, as are the majority of members on here that it won't look anything like that at +384hrs so it was really just a bit of fun. though if by chance it does look like that, i'm sure we wouldn't be complaining!

Edited January 15, 2016 by bobbydog

[lorenzo]

Fantastic debate today, Stewart, Steve, Tamara - great to see such a meeting of minds here. So different in approach, so absolutely impassioned to see the weather people want and examine the evolution - hat's off to you all.

Love the fact that there are so many factors rolling around that can impact short term, medium term, mid range and long range.

Wave One standing impact on vortex from El Nino
Brutal chain reaction of the Kara High and previous wave break on boxing day - this is a beauty complete atmospheric cascade of chaos.
Composite lift out of strength in Canadian climate home for vortex.
Projected AAM dip - not being an AAM dip at all in big scheme of things with baseline AAM simply outrageous. Relative AAM with a drop that would make La Nina state sing vs. our current state that threatens re-cycle and game on for blocking.
Our Polar Vortex - completely off the grid cold vs. Anomalous off the grid blocking ( Kara / NAO via Davis Strait - yes Davis strait blocking is at 90th percentile of climatological mean).
Siberian High evolution on modelling - huge hit to vortex vs. zero hit within 6 hrs on GFS. Chaotic atmospheric dynamics, via,
Ex-Alex- if ever something needed to write a signature for this winter then Alex is it. Ignore the record breaking vortex dynamics for cold temps. Ignore the record breaking Nino temps for regions, the weather is apparently a spin doctor ( no pun intended)..

Cue Alex, model curveball and gives us the best short range modelling cannonball I have ever seen.

So, with that what can anyone predict, would hazard a guess at ' not a helluva much'  that's speaking politely!

Couple of other thoughts - the GWO Jan Feb - hope to analog these over the weekend with new composites.= - daily modes considering MJO Waves and GWO cycle.. will post as soon as built.

The strat trop disconnect - fascinating this year, W1 Nino vs. the oddness of Kara and now potentially Siberia leaves a layer of the atmosphere modified not by upwelling or down welling but by other factors. Forget OPI or SAI or QBO, this year I feel brings EPF to the fore, there is a possible mode here similar to the QBO. It originates in the mesosphere so beyond our science right now, however it is there.

Here are a few chart to display the awaiting roller coaster

Angular momentum nuclear - 

Oh it's on it's way down...

No worries - it is still rampant....

That is one hastened atmosphere and we have an up and coming monstrous torque event. This winter is not done with it's chaos yet. 

As far as looking for cold, we have it, as far as timing, it's here not withstanding 5-7 days and technical descriptions re SSW, and quantification. Amy Butler for her Strat Trop pathways should be relatively happy with skill. Loosely fits forecast so will take that - especially in the face of all the elements outlined above for near term modelling.

The reduction in move to SSW outlined earlier this week - am wondering if this has been reduced due to classic tightening of the vortex and AO profile, with a pulse of relaxation to follow, hence the move away from u reversal. Maybe will relax in 24-48 hrs as the wave recession and drop occurs, if not I feel it is academic as of more interest is the perturbation created by the AAM state.

Over to the close up models !

Did anyone see Ed's tweet today of backwards air trajectory from last Friday - cold air we have now originated in Svalbard, incidentally round about the same time he predicted it. good work !

 

 

 

[Jason M]

38 minutes ago, gottolovethisweather said:

Indeed within 24 hours of the event, I and a great many others will and should be revisiting that thread. Thanks for the reminder John.

As a rough guide re 850s to guarantee snow I tend to use the following as a guide (generally it stands up well unless it's for a location right on the coast).

For a NW -8c 

For a straight northerly -7c

for any NE -6c

for an E or SE -5c

examples of where the above needs modification is where the upper values have not been in place for long as it takes time for a cold air mass to displace a warmer one at all levels through the lower amosphere. So in this example a NW sourced airmass might deliver rain / sleet at low levels for a while until it's fully embedded. Also, the opposite is true where a frontal system brings warmer air against an embedded cold pool. For a period of time it might snow with 850s of -1 to -3

As others have said its a complex issue and there are other parameters but usually you won't be far off with those values. That's why I generally don't get excited about NWers as 850s of -5 to -6 just won't cut it for most on here due to modification of the air mass by the long sea track.

Where 850s are -10c or below you are pretty much guaranteed snow. if it rains at those values you have run over a black cat or live on the Isle of Sheppy 

[I remember Atlantic 252]

4 minutes ago, Jason M said:

As a rough guide re 850s to guarantee snow I tend to use the following as a guide (generally it stands up well unless it's for a location right on the coast).

For a NW -8c 

For a straight northerly -7c

for any NE -6c

for an E or SE -5c

examples of where the above needs modification is where the upper values have not been in place for long as it takes time for a cold air mass to displace a warmer one at all levels through the lower amosphere. So in this example a NW sourced airmass might deliver rain / sleet at low levels for a while until it's fully embedded. Also, the opposite is true where a frontal system brings warmer air against an embedded cold pool. For a period of time it might snow with 850s of -1 to -3

As others have said its a complex issue and there are other parameters but usually you won't be far off with those values. That's why I generally don't get excited about NWers as 850s of -5 to -6 just won't cut it for most on here due to modification of the air mass by the long sea track.

Where 850s are -10c or below you are pretty much guaranteed snow. if it rains at those values you have run over a black cat or live on the Isle of Sheppy 

I need -10 off a NW'ly

and had just rain off these uppers, but ad snow off -1 uppers off battleground style setup

Edited January 15, 2016 by I remember Atlantic 252

[bradymk]

Colder ECM tonight. So many changes the past week. Some very cold nights monday and tuesday next week. Possible snow Sunday for some north western areas and then central areas. Tonight ECM -5c monday night in Birmingham. A bit further south, an ice day in Oxford Tuesday with cold hanging on in most areas until early Thursday. Plenty of changes to come.

[Singularity]

I was just looking at the Sat/Sun precip showing up on a number models and there is the same story of high wet-bulb potential temps riding over the top of colder air near the surface that Fergie mentions for Monday. Given a hard frost the previous night and 2m temps probably dipping below freezing ahead of any precip, early Sunday could turn out pretty slippery as well if I'm looking at things correctly?
 

That's where the upper layers are too warm for snow, of course. Some parts may escape that and get a dusting instead. Or the precip. might not develop at all... hard to call. Recent slight shower potential in these parts has come to nothing, after all.