Model Output Discussion 29th December - Into mid-Winter.

[bluearmy]

19 minutes ago, nick sussex said:

I'm dubious of any output that removes the now cemented higher pressure over southern Europe until its shown within T168hrs.

We've been chasing lower heights for weeks and as yet nothing has materialized. There seems to have been an inbuilt bias this winter to lower heights in that region in the longer range modelling which subsequently implodes.

I'd advise people to keep their sanity intact and wait till we see a strong constant signal which gets through to the day 7 timeframe and not be seduced by the GFS or any other model which suggests the demise of that higher pressure.

 

 

 

Afraid this remains the salient point and I was expecting to see the eps change to maintain these longer, given the latter stages of the op. However, they didn't but if the 12z op ECM is consistent then suspect the eps will slowly follow. 

Clealry the failure of the northerly this coming weekend has moved the pattern back again but interesting that it sets up the same and this time with cross model ens support of the mean/anomoly pattern. so if we can avoid the shortwave dramas then we could be in business. The models didn't see the Atlantic low conundrum which scuppered the retrogression this time and sustained the euro heights as the ridge sank se. clealry, these micro scale evolutions are invisible to the mean/anomolies and probably all but the smallest clusters. 

so much more interesting than last winter although we were chasing some fi cold charts early Jan last year as I recall. 

[booferking]

6 minutes ago, Ali1977 said:

Definitely of interest, this isn't deep FI .

Deep interest again watch it get closer but to get watered down  then to be replaced with epic easterly for a couple of days the thread goes overload me included to then flip to flat setup high sitting to south east the end.

Edited January 1, 2017 by booferking

[bluearmy]

The difference this time could be the predicted -AO. remember we didn't have any polar forcing to back the amplification in the modelling. 

[Day 10]

Happy New Year to all.

So our search for true cold goes on.

It would be nice to get one of these this Winter, here's hoping and praying.

Time for a model break I think, good luck!

 

 

[Banbury]

1 hour ago, Weathizard said:

Really? Just going through the ENS I don't see anything great within 10 days or so...

 

edit it just seen above: day 10 mean is abojt as useful as a chocolate fireguard to be honest! 

Some very good ENS at 240 without a doubt

[bobbydog]

43 minutes ago, Gibby said:

Despite the wild switches in the output of the models this chart has remained largely steadfastly constant over the last week to 10 days with just minor changes day to day. It's been proven to be largely right over the last 10 days and I see no driven reason why we won't remain in a largely benign weather pattern with winds between SW and North over the next 10-14 days due to High pressure just to the SW as this mornings offering shows. It never has showed an Easterly or a strong enough Northerly over the last 10 day period to support the musings of the operational and other permutations almost each run has shown since Christmas especially from GFS. If I was to use the data from each run to my local community my credibility on my local website and Facebook page would reduce to zero in no time. I'm confident that the first half of January will see no marked cold with frosts and a gentle wind from a West or NW point with a lot of cloud at times with little rain or snow- all based on the mean chart illustrated below-in other words the main message I've maintained since Christmas. Sorry 'coldies' but that's how I see it and that from a no bigger snow fan than me.

http://old.wetterzentrale.de/pics/Reem2401.gif

Hi Gibby, I can't see any reason to disagree with that and I don't think at this point, anyone realistically expects any different within the timescales you mention. After that point however, there does seem to be leanings towards colder weather in the output. So whilst us 'coldies' will be disappointed in the near to mid term, I see no reason to give up hope yet.

[Purga]

We are just entering ostensibly -ve NAO & AO phases which are built into the model outputs.

A casual glance at the ensembles recently shows that beyond about 7 days there is so much divergence that the situation is effectively unforecastable to any
useful level of reliability beyond 5 days and the only guide for forecasting further ahead is to analyse clusters and try to spot consistency over several runs.

I believe that looking at 'mean' charts can be pretty useless - The mean is only useful if there's only one cluster!

EPS

& GEFS 06z

After D10 there remain significant cold clusters and this is being reflected in the GFS operation runs recently.

Of course it's frustrating for all us cold & snow lovers that things always seem to be always 'out of reach' but

there is a continuing signal for blocking and the potential for cold coming from the NWP.

It may not give us exactly what we 'want' but the potential is there and much better than it has ben for some years.

 

 

Edited January 1, 2017 by Purga

[Ali1977]

Last 2 ENS very consistent 

[knocker]

There was noticeable more interest today in the Atlantic ridge late on vis the 100mb analysis.

[bluearmy]

1 minute ago, knocker said:

There was noticeable more interest today in the Atlantic ridge late on vis the 100mb analysis.

Even the bias corrected GEFS which are always keen on a big Canadian vortex segment and high heights to our south are showing the opposite on the 06z GEFS run knocks - perhaps something is stirring .....

[knocker]

Just now, bluearmy said:

Even the bias corrected GEFS which are always keen on a big Canadian vortex segment and high heights to our south are showing the opposite on the 06z GEFS run knocks - perhaps something is stirring .....

Yes I'm wondering the same myself blue and I can't help wondering where we will end up. Surely not low pressure centred to the SE as mooted some weeks ago?

[feb1991blizzard]

4 minutes ago, bluearmy said:

Even the bias corrected GEFS which are always keen on a big Canadian vortex segment and high heights to our south are showing the opposite on the 06z GEFS run knocks - perhaps something is stirring .....

Not as pronounced Atlantic high though on the BC GEFS as there is on the regular GEFS.

[Scandinavian High.]

well maybe something is lurking cold wise.

[phil nw.]

Yes was just glancing through the later gefs abbie.Later in week 2 we could well have to look north or north west instead of east for some cold with the trend to swing systems se around a ridging Atlantic again.

Quite a downward trend in the 850's again later on in the 06z set.

         

If that pattern sets up then there's always a chance of an Arctic northerly,at the best time of year.

BA or anyone do we know what the eps are looking like days 10-15?

[Penrith Snow]

1 hour ago, Quicksilver1989 said:

Unfortunate to see another cold spell consigned to the bin. I think we can safely say that now.

I think there was a number of warning signs:

The sinking of high pressure to the south of the UK which really was a big blow, a northerly at the end of this week would have been a big help

Good agreement on the operational runs - Yes we should also look at the ensembles but we saw good agreement between the operational runs yesterday afternoon. The operational runs use a higher resolution then the accompanied ensemble members. The fact that three different models (UKMO,GFS and ECM) were showing stubborn heights to our SSE was a big warning sign, despite cold ensemble members still being shown. Most of these ensemble members saw the -10C isotherm barely clip the East Coast anyway.

Cold in Greenland and NE Canada - Pretty simple but with cold in these regions the Atlantic is always at risk of firing up. The low to the West of Iceland now goes against us given we don't have a northerly. The result is a flatter ridge to Scandi that doesn't get anywhere.

As for why this pattern keeps repeating... I don't think it's the westerly QBO as December 2010 was westerly. I did post this on NW a while back which is worth repeating for those who missed it and I stick by this interpretation:

Over the past few years we have had references to the North Atlantic cold blob, that has persisted. Some people say its gone but I think its still there. There is something in oceanography called the re-emergence mechanism. Basically in the winter the mixing layer (area that interacts with the atmosphere) is at its deepest due to the stormy weather. In the Summer months the mixing layer shoals. However the cold SSTs from the previous winter remain at depth, only to re-emerge again when it becomes stormier. Here is an example. First is March 2015, then December 2015.

 - March 2015

 

 - December 2015

See? the same SST pattern re-emerges again the following winter. It is interesting to note that this SST pattern is associated with increased cold air outbreaks filtering through the North Atlantic from NE Canada.

Unfortunately, here is March 2016....

 - March 2016

I get the feeling that unless some great atmospheric forcing overwhelms this then we will continue to see more westerly atmospheric circulation patterns then normal. I hope I'm wrong but in 2010 we had the opposite:

 - March 2010

 - December 2010

In 2013 we got bailed out by a good SSW event and I think we will need the same again. An SSW isn't a guarantee of cold but it increases the chances.

Now don't get me wrong I think this could easily turn around. Numerous ensemble members show northerlies for example. This however is well into FI.. I hope that the effects of low solar activity will kick in soon, like they did in 2008.

However for the reasons listed above, I am keeping my expectations very low. I am usually a cold ramper but this winter the background signals are favouring mild and this is what we have seen.

Thats the most informative post I have read in a long time and unlike so many so called teleconnections these sea temperature anomalies actually make sense!

The difference between recent years and 2010 is remarkable, no wonder we have difficulty getting a decent northerly or easterly with endless depressions being spawn off Newfoundland.

The question is however when is this SST anomaly going to change?

Andy

 

[Jason M]

There are a number of Greenland high scenarios in amongst the GEFS which look highly unlikely to my eye and will presumably scew the means / anomalies. From memory one had a 1080 high pressure cell over Greenland which is clearly utter nonsense. The GEFS in FI tend to overdo Greenland highs and they hardly ever make it nearer than day 10 in practice. Ignoring the Greenland Cluster I thought the 06z suite was pretty much random scatter. Some cold looking charts that in isolation that looked blocked but we're actually just a snapshot in a quick moving pattern.

certainly my conclusion is that there is no coherent signal whatsoever for anything other than brief, watered down cold shots.

[knocker]

5 minutes ago, phil nw. said:

Yes was just glancing through the later gefs abbie.Later in week 2 we could well have to look north or north west instead of east for some cold with the trend to swing systems se around a ridging Atlantic again.

Quite a downward trend in the 850's again later on in the 06z set.

         

If that pattern sets up then there's always a chance of an Arctic northerly,at the best time of year.

BA or anyone do we know what the eps are looking like days 10-15?

The mean shows ridging to our west, a positively tilted trough to our E/SE thus a a cool NW upper flow.with some retrogression during the period.

Edited January 1, 2017 by knocker

[bluearmy]

5 minutes ago, knocker said:

The mean shows ridging to our west, a positively tilted trough to our E/SE thus a a cool NW upper flow.with some retrogression during the period.

and the 10/15 days anomoly about the same as the GEFS/geps Phil.  Posted this about 9am ...... where were you! 

[Weather Boy]

1 hour ago, nick sussex said:

I'm dubious of any output that removes the now cemented higher pressure over southern Europe until its shown within T168hrs.

We've been chasing lower heights for weeks and as yet nothing has materialized. There seems to have been an inbuilt bias this winter to lower heights in that region in the longer range modelling which subsequently implodes.

I'd advise people to keep their sanity intact and wait till we see a strong constant signal which gets through to the day 7 timeframe and not be seduced by the GFS or any other model which suggests the demise of that higher pressure.

 

 

 

Hi Nick

Interesting post, particularly the part I have emphasised.  Although we don't know for sure yet that you're right, it certainly does look that way.  If it is, the big question is why?

It seems to me that we have the following possibilities.

1. The models have inbuilt bias towards lowering pressure on the continent and therefore are not doing their job effectively;

2. There is a signal that the models are latching on to which is then removed or watered down meaning that other factors overide it; or

3. A bit of both.

Now, if it is pure no.1, then you would expect this to happen a lot.  Certainly it has happened before, but I would suggest that it is not happened enough for us to be able to say that it is this factor alone.  No. 1 alone therefore seems unlikely to me.

If I'm right, we're therefore left with nos 2 or 3.  In that case, what signals are we talking about and why are they being watered down.  Perhaps most importantly, are there signals there which might indicate that the heights WILL lower at some point, perhaps later in time than the models are currently predicting?

From my limited knowledge, I'm at a bit of a loss as to what these signals might be.  There's no SSW or other significant strat warming to latch onto.

Any thoughts?

[phil nw.]

2 minutes ago, knocker said:

The mean shows ridging to our west, a positively tilted trough to our E/SE thus a a cool NW upper flow.with some retrogression during the period.

Thanks Knocker sounds a similar pattern.All change again,something more typical of mid-Winter by the looks a cool Atlantic setup,maybe an arctic northerly if we see enough ridging out west.

[phil nw.]

1 minute ago, bluearmy said:

and the 10/15 days anomoly about the same as the GEFS/geps Phil.  Posted this about 9am ...... where were you! 

lol,Just about surfacing!

I missed that one but thanks to you and Knocker.I am just pleased to see some movement in the pattern as i am sure many of us were dreading another 2 weeks of a UK high. 

[chris55]

delete

 

 

Edited January 1, 2017 by chris55

[chris55]

2 hours ago, Summer Sun said:

 

That 20% chance is still a chance eh

 

You can see on ECM 144 how its not an impossibility to get some sort of a colder north/north east/east flow from this point.

We just need that trough off Norway to drop south east into Europe and that could allow a much colder feed to build and establish.

ECM 144

 

P18 on the 6z ens does this and it leads to very nice outcome for colder conditions.

 

 

 

 

 

A real long shot I know, but hey, we can dream .....

[bluearmy]

7 minutes ago, Weather Boy said:

Hi Nick

Interesting post, particularly the part I have emphasised.  Although we don't know for sure yet that you're right, it certainly does look that way.  If it is, the big question is why?

It seems to me that we have the following possibilities.

1. The models have inbuilt bias towards lowering pressure on the continent and therefore are not doing their job effectively;

2. There is a signal that the models are latching on to which is then removed or watered down meaning that other factors overide it; or

3. A bit of both.

Now, if it is pure no.1, then you would expect this to happen a lot.  Certainly it has happened before, but I would suggest that it is not happened enough for us to be able to say that it is this factor alone.  No. 1 alone therefore seems unlikely to me.

If I'm right, we're therefore left with nos 2 or 3.  In that case, what signals are we talking about and why are they being watered down.  Perhaps most importantly, are there signals there which might indicate that the heights WILL lower at some point, perhaps later in time than the models are currently predicting?

From my limited knowledge, I'm at a bit of a loss as to what these signals might be.  There's no SSW or other significant strat warming to latch onto.

Any thoughts?

Well we would be looking at them disappearing by next weekend if it weren't for the Azores low conundrum which sank the Atlantic ridge se. Now we have to wait for the six/seven day cycle to repeat. (Without more dramas). clearly we live in a world which turns west to east so much easier for the semi permanent Azores to throw upper ridges which end up to our south rather than retrogress, especially under a wqbo 

Then there is always climatology which has high heights to our south rather than low ones!  Getting them to drop is away from the norm.   

anyway, hopefully, this question will be redundant within ten days! 

[chris55]

1 hour ago, bluearmy said:

Afraid this remains the salient point and I was expecting to see the eps change to maintain these longer, given the latter stages of the op. However, they didn't but if the 12z op ECM is consistent then suspect the eps will slowly follow. 

Clealry the failure of the northerly this coming weekend has moved the pattern back again but interesting that it sets up the same and this time with cross model ens support of the mean/anomoly pattern. so if we can avoid the shortwave dramas then we could be in business. The models didn't see the Atlantic low conundrum which scuppered the retrogression this time and sustained the euro heights as the ridge sank se. clealry, these micro scale evolutions are invisible to the mean/anomolies and probably all but the smallest clusters. 

so much more interesting than last winter although we were chasing some fi cold charts early Jan last year as I recall. 

This is interesting Nick.

If the means/anomalies struggle with such synoptics/details, that ultimately have a profound effect on the outcome then how do we have confidence in the means/anomalies? And how do we read them with confidence?

I guess its a case of "best available tool" but not flawless by any means.