Model output discussion - 7th January onwards

[chionomaniac]

9 minutes ago, ALL ABOARD said:

Anyone worrying about what the GFS is showing for a period 13 days away .

GFS 11th Of February 2018 for 24th Of February.




Actual chart

Haha, I remember Catacol and I fighting against the despair - mind you that was one of the easier forecasts to make against the GFS at that time. This time I think it will take a little longer to get a block of that magnitude!

[Singularity]

GFS & GEFS seem to be struggling to imprint the split on the troposphere in a sustained manner. Hard to be entirely sure that it's down to model bias (poor resolution of low height divisions and split jet flows along with general zonal overcook), but given the much more positive trends from EPS, it seems pretty likely.

Yeah, I know, famous last words... 

 

[Lloyds32]

32 minutes ago, bluearmy said:

The fi gefs look more mobile to me than the extended eps 

not ‘feeling the love’ for a sustained height rise to our nw yet ...

I think we will be waiting another week at least.

[Kirkcaldy Weather]

My interest is beginning to increase as are the number of members going down the scandi high route

ensembles FNMOC/NOGAPS 0z

  

ensembles GEM 0z

      

ensemble NAEFS 0z now gets the linking of above average heights  Atlantic / Greenland / arctic by 20th Jan (although I think this is just a bit too quick) good positives going blocked quicker

  and by 384h  

GEFS

 

I am not concerned with the GFS / GEFS / GFS P , think it is a case of uncertainty so going with its default of developing low pressures as much as it can.

My first point of focus - get the lows that start to develop on that NW - SE track bringing some cooler / colder conditions (north westerlies, northerlies, north easterlies) then attention turns to blocking going through last week of Jan (keep monitoring blocking trends til then of course)

33 minutes ago, Quicksilver1989 said:

PS where are people finding the GFS parallel run? Its still stuck on yesterday on wetterzentrale.

https://www.meteociel.fr/modeles/gfse_cartes.php?ech=6&code=code&mode=0&mode3h=0&runpara=1&carte=1&runpara=1 

[feb1991blizzard]

Just now, Singularity said:

GFS & GEFS seem to be struggling to imprint the split on the troposphere in a sustained manner. Hard to be entirely sure that it's down to model bias (poor resolution of low height divisions and split jet flows along with general zonal overcook), but given the much more positive trends from EPS, it seems pretty likely.

Yeah, I know, famous last words... 

 

There was a brilliant strat GFS run yesterday (6z) - since then its struggling to imprint the full pattern on the trop again, although even where the strat split in the mid strat isn't favourable, there is a 3 vortice split maintained right out to 384, so one things for sure, for a while yet, at least any failure is unlikely to be our last attempt at cold, cannot see with current synoptics, being able to write off winter until Early - mid March.

[karlos1983]

12 minutes ago, Singularity said:

GFS & GEFS seem to be struggling to imprint the split on the troposphere in a sustained manner. Hard to be entirely sure that it's down to model bias (poor resolution of low height divisions and split jet flows along with general zonal overcook), but given the much more positive trends from EPS, it seems pretty likely.

Yeah, I know, famous last words... 

 

I am reliably told that the "EC can be overly sensitive to Stratospheric weakening", so probably wise for cautious approach at this stage. 

Hopefully over the next few days we will start to see the signalled EC46 pattern start to play out on NWP, just to ease some tension! In the mean time if the EPS remain steadfast with the pattern for the last 3rd of Jan onward, that's definitely a good thing!

Edited January 9, 2019 by karlos1983

[Timmytour]

3 hours ago, johnholmes said:

Well the first tentative signal from the 8-14 NOAA 500 mb anomaly that a change in pattern may be about to show?

My comments on my 500 mb anomaly file this morning.

Wed 9 january

Ec-gfs show no signal for marked blocking, just minor +ve height signal mid atlantic on ec bit more ridging even on gfs

However noaa does show more suggestion of the flow buckling more on the 8-14 than the 6-10, also the contour heights are slowly falling, just s of the uk is now 546 DM, some 10-12 below its suggested value several days ago.

All 3 have some form of troughing over europe, although the main centre of this is still way east

http://www.cpc.ncep.noaa.gov/products/predictions/814day/500mb.php

http://mp1.met.psu.edu/~fxg1/ECMWF_0z/hgtcomp.html

Too early for this to be looked on as a strong indicator but if the same charts continue over the next couple of days then the % probability would change. Currently about 20%, on the later period, 10-14 days say.

Delighted to see this post.  I have found over many winters that promises of nirvana in the models and background signals have never ever delivered without some support from JH's anomaly files analysis and the lack of such support to date has tempered my optimism for the promises of what 10 days ahead will deliver us now.  I am still cautious - which is only natural having seen so many huge let downs in the past - but much happier to see that now at least JH can see some signs of a change in pattern 

Edited January 9, 2019 by Timmytour
two words too many used!

[northwestsnow]

GEFS still look decent to me ..

Systematic of PM influence..

 

Click image to see the updated 6z GEFS.

 

Edited January 9, 2019 by northwestsnow

[MattStoke]

20 minutes ago, karlos1983 said:

I am reliably told that the "EC can be overly sensitive to Stratospheric weakening", so probably wise for cautious approach at this stage. 

Hopefully over the next few days we will start to see the signalled EC46 pattern start to play out on NWP, just to ease some tension! In the mean time if the EPS remain steadfast with the pattern for the last 3rd of Jan onward, that's definitely a good thing!

On the other hand, isn’t the GFS normally slow with the trop response due to there being fewer layers in its modelling?

[karlos1983]

The control for De Bilt is insane brhhh 

Very cold set in general

[karlos1983]

5 minutes ago, MattStoke said:

On the other hand, isn’t the GFS normally slow with the trop response due to there being fewer layers in its modelling?

Not as good in the strat as the EC, but to be honest I'd rather the GFS underplays it and we get a nice surprise than the alternative  

I'm referring to the trop response at High Latitudes in terms of blocking, rather than Strat. (EC)

Edited January 9, 2019 by karlos1983

[Lloyds32]

1 hour ago, karlos1983 said:

The control for De Bilt is insane brhhh 

Very cold set in general

That's because the models actually show cold uppers dropping down over that part of Europe. Here we miss out.

Edited January 9, 2019 by Blessed Weather
Final comment removed.

[frosty ground]

39 minutes ago, feb1991blizzard said:

There was a brilliant strat GFS run yesterday (6z) - since then its struggling to imprint the full pattern on the trop again, although even where the strat split in the mid strat isn't favourable, there is a 3 vortice split maintained right out to 384, so one things for sure, for a while yet, at least any failure is unlikely to be our last attempt at cold, cannot see with current synoptics, being able to write off winter until Early - mid March.

The same pretty much every year then, you know with winter ending at the same point every year and the thermal lag always allowing for winter weather into spring.

[snowfish1]

Boys can we play nicely. We're a great Gang and us newbies appreciate your input but not the punch ups. Cheers. Gfs still looking very reserved by comparison lol. Anyone wanna take a punt as to when gfs will sniff out and start showing something more exciting? 

Edited January 9, 2019 by snowfish1

[bluearmy]

1 hour ago, ALL ABOARD said:

Anyone worrying about what the GFS is showing for a period 13 days away .

GFS 11th Of February 2018 for 24th Of February.




Actual chart

can i accuse you of 'selective posting' ?

four gfs ops that day

2 brought in a cold easterly from a scandi ridge

1 brought in a scandi ridge without a frigid easterly

you posted the other one

in general, the gfs ops were good on the feb reversal from about five days out of th ssw onset (6th feb) - yes some runs dropped the baton but generally good. this downwelling is clearly a very different animal

[ALL ABOARD]

1 minute ago, bluearmy said:

can i accuse you of 'selective posting' ?

four gfs ops that day

2 brought in a cold easterly from a scandi ridge

1 brought in a scandi ridge without a frigid easterly

you posted the other one

in general, the gfs ops were good on the feb reversal from about five days out of th ssw onset (6th feb) - yes some runs dropped the baton but generally good. this downwelling is clearly a very different animal

Indeed.

But my point still stands, I was referring to the 6z this morning.

Who's to say the 12z wont show a better outcome?

I was merely highlighting that each individual run should not be taken at face value.

 

[Skyraker]

14 minutes ago, Lloyds32 said:

That's because the models actually show cold uppers dropping down over that part of Europe. Here we miss out. I wouldn't call it insane though, bit over the top. 

True, but it's on it's own there, which is good for us.

[bluearmy]

Just now, ALL ABOARD said:

Indeed.

But my point still stands, I was referring to the 6z this morning.

Who's to say the 12z wont show a better outcome?

I was merely highlighting that each individual run should not be taken at face value.
 

a few days ago i mused that it would be good for one of the the gfs/gfsp to pick a run that fitted the expected direction of travel for 20/23 jan on each suite … that worked for the next three runs but has been less reliable since.  i still feel the ncep model looks too mobile (that shouldn't be a surprise) but i remain uncomfortable whilst the para fails to read the script for more than the odd run.

[Weathizard]

2 minutes ago, bluearmy said:

can i accuse you of 'selective posting' ?

four gfs ops that day

2 brought in a cold easterly from a scandi ridge

1 brought in a scandi ridge without a frigid easterly

you posted the other one

in general, the gfs ops were good on the feb reversal from about five days out of th ssw onset (6th feb) - yes some runs dropped the baton but generally good. this downwelling is clearly a very different animal

Yeah, there’s definitely an issue with selective memory in here for when people have an agenda for certain things(not saying that is what this is) it’s very easy to cherry pick and say a model was wrong or right... in the long run I would say all the main models are equally as fallible as each other, there’s a reason why the verifation stats are always so close to each other.

I think it’s fair to say a cold spell counting down to t0 with no hiccups from day 10+ is always very unlikely, it’s exceptionally rare. So not surprising to see the models moving back and fourth slightly.

[bluearmy]

Just now, Weathizard said:

Yeah, there’s definitely an issue with selective memory in here for when people have an agenda for certain things(not saying that is what this is) it’s very easy to cherry pick and say a model was wrong or right... in the long run I would say all the main models are equally as fallible as each other, there’s a reason why the verifation stats are always so close to each other.

I think it’s fair to say a cold spell counting down to t0 with no hiccups from day 10+ is always very unlikely, it’s exceptionally rare. So not surprising to see the models moving back and fourth slightly.

for whatever reason, the feb reversal was pretty textbook stuff and the modelling reflected that …... 

[northumberland snowman]

27 minutes ago, karlos1983 said:

The control for De Bilt is insane brhhh 

Very cold set in general

Brilliant to see.

[No Balls Like Snow Balls]

This might get a few backs up, but it's not meant to and it is just my opinion, but I find the use of the De Bilt ensembles useless for predicting cold for the UK.

I have seen time and time again over the years the really cold air stopping at the start of the English Channel, and it can be below freezing in De Bilt, while it is can be +5c in the SE/East of the UK.  All the De Bilt ensembles tell me is that more than likely Northern Mainland Europe will be very cold, whilst in reality, the UK 8/9 times out of 10 is much warmer. 

I welcome comments on what value they can add when looking for UK specific cold.

 

Edited January 9, 2019 by No Balls Like Snow Balls

[Timmytour]

20 minutes ago, Weathizard said:

Yeah, there’s definitely an issue with selective memory in here for when people have an agenda for certain things(not saying that is what this is) it’s very easy to cherry pick and say a model was wrong or right... in the long run I would say all the main models are equally as fallible as each other, there’s a reason why the verifation stats are always so close to each other.

I think it’s fair to say a cold spell counting down to t0 with no hiccups from day 10+ is always very unlikely, it’s exceptionally rare. So not surprising to see the models moving back and fourth slightly.

Neither is it unknown to see a mild spell, after a cold spell has been counted down from day 10+ to 2....

[karlos1983]

8 minutes ago, No Balls Like Snow Balls said:

This might get a few backs up, but it's not meant to and it is just my opinion, but I find the use of the De Bilt ensembles useless for predicting cold for the UK.

I have seen time and time again over the years the really cold air stopping at the start of the English Channel, and it can be below freezing in De Bilt, while it is can be +5c in the SE/East of the UK.  All the De Bilt ensembles tell me is that more than likely Northern Mainland Europe will be very cold, whilst in reality, the UK 8/9 times out of 10 is much warmer. 

I welcome comments on what value they can add when looking for UK specific cold.

 

Because they are pretty, that's about it for me... Jokes!

Actually if an easterly is on the cards they are useful, also if we can cool the continent down and we can tap into that source, even if it's briefly from a sinking low etc, then it is useful. imo 

Edited January 9, 2019 by karlos1983

[winterof79]

18 minutes ago, No Balls Like Snow Balls said:

This might get a few backs up, but it's not meant to and it is just my opinion, but I find the use of the De Bilt ensembles useless for predicting cold for the UK.

I have seen time and time again over the years the really cold air stopping at the start of the English Channel, and it can be below freezing in De Bilt, while it is can be +5c in the SE/East of the UK.  All the De Bilt ensembles tell me is that more than likely Northern Mainland Europe will be very cold, whilst in reality, the UK 8/9 times out of 10 is much warmer. 

I welcome comments on what value they can add when looking for UK specific cold.

 

For a start the Control is an out and outlier but may be a trend setter as it shows a stonking northerly come Easterly 

The set is ok other than that but trending colder. They are useful when a Scandi high forms and cold is definitely heading our way but as for a developing Atlantic ridge who knows?

Edited January 9, 2019 by winterof79