Model Output Discussion - 1st February Onwards 12z--->

[Jason M]

In fairness Dorset, John said 'deep cold'.

The charts he uses won't preclude temp pm shots etc, but it's reasonable to say no deep cold for two weeks and if there is any evidence this morning that suggests otherwise I sure can't see it anywhere. The balance of probabilities this morning suggests notable warmth is more likely that notable cold. By notable cold I want to stress I'm not ruling out a 24 / 48 hour northerly.

Also worth noting that although John has been saying no deep cold for two years, we haven't had any deep cold for two years. I guess the test will come if we get a notable cold shot next winter to see how johns method picks that up.

I think a similar thing applies to the METO seasonal view of a warmer than average winter. Looking against the CET ranges for the last 100 years they show a steady climb so to predict above average against these, is probably not a hard call. The skill is picking out the rare cold winter. That's the true test. Unfortunately, where there is no cold winter or even deep cold spell in a winter we can't criticise people for consistently calling mild.

I said last night that I thought a 10 to 20 per cent chance in late feb, but seeing the changes in the ouput this morning I'm revising that to 5 per cent. Not that I pretend to be a forecasting guru :-) just going off experience.

[damianslaw]

Must admit things are looking less and less promising now for a cold spell second half of the month but GEM is just an example of what could happen 10/11 days from now. I appreciate this is unlikely but at t+240 the models have got things wrong before and will do in the future. Mother Nature has a way of bringing surprises and history tells us that March can bring us cold spells. March 2013 ring any bells??

 

As I have said before some members have just been disappointed and frustrated as it is back to back winters that have failed to deliver 1 nationwide snowy and cold spell and as a result have thrown the towel in from a cold perspective. Premature I think and history tells us that.

 

We just need Vortex to move towards Siberia and even in this happens in 14 days time that is still only 22nd Feb. Lets just enjoy the next 10 days with a bit of frost hopefully and pleasant sunny days. After that, lets just see.

 

 

General theme from the models is for high pressure to move eastwards allowing a temporary incursion of atlantic air into the north by the end of the week, with weak trough feature, then its back to high pressure again for the following week, with centre of core heights to our SW allowing some milder conditions to filter through especially for the north, so we are most likely to lose the foggy frosty conditions.

 

Where we go thereafter, remains to be seen, as ever much will depend on the position of the Jetstream and Polar Vortex, we could see a more amplified flow develop once again before the end of the month ushering in a similar set up as we saw at the end of Jan into Feb with heights ridging back westwards and a trigger low from the NW digging SE with the trough going into scandi allowing a colder influence from the NW once again. At this juncture its far too early to call no chance of anything cold and wintry surfacing again this month. If we enter the third week with the models continuing the fluid mobile pattern we have now, then yes we can start to begin to dismiss such chances, but until then everything will remain conjecture. Lets see where we are 8/9 days time. I said if we are to see a colder wintry evolution it is likely to occur in a short time scale and abruptly much like occurred in January within the 144 hr timeframe.

 

For those seeking a colder outlook, we are in a far better starting position given we have a displaced azores high and a major high pressure on the scene, rather than a raging jet on a SW-NE trajectory with the azores sat in place in its comfort zone or even worse over Europe. There is no signal in the charts such a pattern to occur, the current high can could easily align in a favourable position for a colder outcome, but it also may not..

Edited February 8, 2015 by damianslaw

[johnholmes]

hi Dorset

But I do treat them as means, perhaps our understanding of what I am doing are not the same? Happy to discuss this via pm if you wish as I always am on anything to do with the weather.

In fairness they picked up the change to colder weather over Christmas and again in January. yes the synoptic models did, I stress at times, but at other times they did not. That is the benefit of smoother outputs as means must always be and how one interprets what they show. I honestly believe they offer to we amateurs the best possible advice on what the upper flow as an average over a 6-15 day time scale is most likely to be. With practice it is also possible to suggest, as I try to do, when they offer the possibility of some variation from what they indicate as the mean or average flow. Like I say DW I am very happy to chat further via pm?

The ECM MEAN anomolys picked every colder period thus far this winter at two weeks. The signal weakish to begin with but firming up quickly to show the euro troughing. The gefs less successful.

 

yes I would agree with that, hence why I always use them, plus looking at every other bit of data in the 6-20 day time scale is readily available to me.

[johnholmes]

In fairness Dorset, John said 'deep cold'.

The charts he uses won't preclude temp pm shots etc, but it's reasonable to say no deep cold for two weeks and if there is any evidence this morning that suggests otherwise I sure can't see it anywhere. The balance of probabilities this morning suggests notable warmth is more likely that notable cold. By notable cold I want to stress I'm not ruling out a 24 / 48 hour northerly.

Also worth noting that although John has been saying no deep cold for two years, we haven't had any deep cold for two years. I guess the test will come if we get a notable cold shot next winter to see how johns method picks that up.

I think a similar thing applies to the METO seasonal view of a warmer than average winter. Looking against the CET ranges for the last 100 years they show a steady climb so to predict above average against these, is probably not a hard call. The skill is picking out the rare cold winter. That's the true test. Unfortunately, where there is no cold winter or even deep cold spell in a winter we can't criticise people for consistently calling mild.

I said last night that I thought a 10 to 20 per cent chance in late feb, but seeing the changes in the ouput this morning I'm revising that to 5 per cent. Not that I pretend to be a forecasting guru :-) just going off experience.

 

 

UK Met and the 500mb anomalies both picked out the unusually cold spell for late November-December 2010 and the return to less cold weather. 

[Frosty.]

It will be interesting to see the 6z perturbations, I noticed the 0z control run showed several incursions of polar maritime air during the run, even more than the ecm 0z op run..so I'm hoping the high which incidentally looks a rather cloudy high, the worst kind. I'm hoping the high will be pushed and pulled this way and that with at least occasional bouts of unsettled weather and a risk of at least short lived pm shots with wintry showers at times. In my opinion, two weeks plus of generally benign, dull and calm weather at the tail end of a frankly poor winter would be insufferable,the final insult..so...I hope the outlook is spiced up by some colder outbreaks.

Edited February 8, 2015 by Frosty.

[bluearmy]

John, the issue here being that we do not see the ECM mean /anomolys 10/15 day unless we are prepared to pay. If these were freely available, I think the view on here would be quite different, especially this season.

[Cloud 10]

The GFS has finally climbed aboard the trough train for the end of the week,but certainly was last to the "party"

 

yesterday..  today..

[bluearmy]

The GFS has finally climbed aboard the trough train for the end of the week,but certainly was last to the "party"

 

yesterday..gfs-2015020706-0-144.png  today..gfs-0-120.png

Indeed, the increase in resolution has certainly helped it in general but it still fails to see stuff.

Read an interesting article on weatherbell where they looked at hurricane sandy - the catalyst for the upgrade to the model. Evidently, the new model would also have failed because the physics that failed on that occasion hasn't been changed in the upgrade.

[nick sussex]

I think the patient is in critical condition, that would be the best analogy for todays outputs in terms of wintry potential. The priest hasn't yet been called to deliver the last rites but the nurse has his number on speed dial!

 

Although we don't see the rounded PV blob to the north theres just a lack of amplification in the Atlantic so even with a more elongated stretched PV any cold seems to be diverted south well to the east.

 

The more favourable MJO phase 8 signal has been trampled to death by the limpet Canadian PV, at this point in the winter chasing the odd PM interlude is to be honest underwhelming and the model fatigue seems evident in here.

 

Going forward then that's the issue, either the models find some amplification to retrogress the high or the odds will be on the winter going out with a sad whimper.

 

Earlier we still see some divergence re the weekend, a few GEFS bring some colder air se briefly behind that low, the models have really dragged a lot of drama out over the split jet pattern.

 

The ECM operational run looks on the mild side of its ensembles which are more interested in keeping high pressure further north.

 

Apart from the weekend it looks pretty quiet weather wise, personally I'd rather that then wet windy average conditions.

 

So overall then unless theres some changes re amplification then a period of boredom coming up, even though the current MJO signal is being muted out whilst that's still favourable a small chance something might pop up especially if the limpet Canadian PV can edge further west.

Edited February 8, 2015 by nick sussex

[IDO]

The GFS has finally climbed aboard the trough train for the end of the week,but certainly was last to the "party"

 

yesterday..gfs-2015020706-0-144.png  today..gfs-0-120.png

 

Yes a small shortwave digging slightly south is not a major fail from the GFS op. If you looked at its GEFS this was always a possibility. It is transient and of minor detail. Yet if you look at the ECM op from 2 days ago when it sent a trough south and then cut it off, that was a fail. It hinted at a pattern change and of course a complete wrong solution:

 

 

In any case the 5 day ECM member's clusters are split 50:50 with regard to how far south that transient low dives. The other 50% (cluster 1) keep heights over the UK:  http://old.ecmwf.int/samples/d/inspect/teasers/samplers/banner/mean_sea_level_pressure_and_temperature_at_850hpa!120!pop!od!enfo!enplot!2015020800!!/

 

So that may still be uncertain. But in the great scheme of things, whether we get a diving low next weekend, it seems to make little difference to the further outlook.

 

Looking at the D10 verification stats and the GFS op leads the field with the ECM one from last, behind the much maligned GEM and CFS!!!

 

 

The GEM op was an outlier and at D10 the mean shows a UK high:

 

 

The GEM mean keeps HP in charge till D16 though a trend to a N/S divide. The GFS 06z suite is HP dominated till D16 apart from the D5 period where we have some phasing between the two HP systems. D10 and D14 means:

 

 

 

The MJO looks like it is sleeping, just meandering at low amplitude in phases 7 & 8, so I am not sure that will be a driver in the next two weeks:

 

 

Warming up in week 2 in a very quite and dry period.

Edited February 8, 2015 by IDO

[johnholmes]

Indeed, the increase in resolution has certainly helped it in general but it still fails to see stuff.

Read an interesting article on weatherbell where they looked at hurricane sandy - the catalyst for the upgrade to the model. Evidently, the new model would also have failed because the physics that failed on that occasion hasn't been changed in the upgrade.

image.jpg

 

That omission seems a bit odd when the problem was known so long ago!

[johnholmes]

John, the issue here being that we do not see the ECM mean /anomolys 10/15 day unless we are prepared to pay. If these were freely available, I think the view on here would be quite different, especially this season.

 

have to agree with you ba.

EC and Met appear to have a rather similar idea on release of information that may help the independent forecast centres, sensibly, although perhaps it is not seen that way by weather enthusiasts, they choose to sell it at quite high costs and not release to the general public.

[Cloud 10]

Indeed, the increase in resolution has certainly helped it in general but it still fails to see stuff.

Read an interesting article on weatherbell where they looked at hurricane sandy - the catalyst for the upgrade to the model. Evidently, the new model would also have failed because the physics that failed on that occasion hasn't been changed in the upgrade.

image.jpg

 

Perhaps NOAA's new awesome super-duper computer can give the GFS a helping hand.

 

https://forum.netweather.tv/topic/76236-in-depth-model-discussion-and-summaries/page-11#entry3166652

[bluearmy]

Perhaps NOAA's new awesome super-duper computer can give the GFS a helping hand.

 

https://forum.netweather.tv/topic/76236-in-depth-model-discussion-and-summaries/page-11#entry3166652

IMO, only if they increase the number of vertical levels and, as the piece I referred to illustrates, change the algorithms where they know them to be suspect.

[Paul]

There's another upgrade to the GFS planned for the end of this year with physics tuning etc.

[nick sussex]

Not a model to put too much faith in but the GME  is close to bringing in some colder air at T132hrs. Its likely to be a brief affair even if the pattern does undergo some last minute amplification but with nothing much else interesting so far today this is the best so far in terms of potential:

 

SLP

 

 

850's

 

 

 

 

 

[bluearmy]

I read IDO' s earlier post where he poked fun at the ecm op solution the other day where it dropped the trough through and built the ridge back in.

I was going to post that this is not so ridiculous and then along comes the ukmo 12z op !!

[polar vortex]

Positive signs on the gfs 12z with stronger ridging over on the Pacific side of north America. This could be very 

beneficial down the line.

[492]

http://www.meteociel.com/modeles/gfse_cartes.php

 

Big changes from around T+120 on 12Z GFS backed up by latest UKMO as trough approaching UK slips south giving lower heights to S of the UK as weak ridge tries to build to the N/NE. Seem to remember that the recent cold spell appeared out of the blue at around T+96.      

[igloo]

first signs possibly  of a scandi HP trying to form did the ECM32 control not build one last week for the 3rd week of february not likely to come to anything as the PV is sitting in the usual place 

 

[polar vortex]

With the much stronger ridging up into Alaska on 12z run we hopefully will see a much better outcome 

in FI. This could turn into a very good run for us.

[Frank Trough]

so the GFS goes from this yesterday:

 

 

to this today:

 

 

 

 

possibly just a blip in the overall pattern but a fairly notable oversight. In fact, the GFS is now looking a lot like the ECM from yesterday or whenever it was.

[johnholmes]

http://www.meteociel.com/modeles/gfse_cartes.php

 

Big changes from around T+120 on 12Z GFS backed up by latest UKMO as trough approaching UK slips south giving lower heights to S of the UK as weak ridge tries to build to the N/NE. Seem to remember that the recent cold spell appeared out of the blue at around T+96.      

 

Not true it was predicted about 10-12 days before it arrived or at least that is how I saw it?

 

Just to show I have not made the above up, sorry for the long post of 10 Jan into Net Wx and remember the links will not show the data on that day or I don't think they will.

 

Written 7 Jan

Just looked at the GFS MJO and it has a reasonable predicted orbit to take it into 6 (no help from deep cold) and then perhaps into 7 at a reasonable orbit, which could help towards late January perhaps?

tentative signs only as yet.

 

And on the 8^th re NOAA 6-10

Is this the first indication on noaa of some kind of move to a more n of west flow, upper trough perhaps e of uk?? Next 2 days should show this and especially if by 2 maybe 3 days from now similar shows on 6-10 version, even more so if this idea is starting to show also on ec-gfs?

 

And on the 9^th

 

Noaa

6-10 and it is showing the veering 500 flow with +ve ehights along with troughing and –ve heights w-e of uk-early to go for it  and 8-14 is not really that far into it but along with trends etc along with 30mb and mjo prob it would look more like a colder slot, quite unclear how many days as yet but could be 4-7 and that MIGHT extend if the 8-14 takes on a similar pattern with the 6-10 not moving away from the idea?

6-10 height=540 and 8-14 shows 546dm tonight far s of uk=colder by perhaps 6dm than 2-3 days ago.

 

And this posted I think the same day into Net Wx

 

Posted into net wx am sat 10/01/14

A more detailed post than the one I gave last evening.

 

I think it is quite clear that  a colder overall pattern is developing, the anomaly charts, links below, have been reasonably consistent on this over the last 2-3 days. Again similar to their over/post Xmas predicted change, it was the ECMWF-GFS version that first gave this idea. In the 5 years I have made daily notes on the 3 it is usually NOAA at 6-10 sometimes even the 8-14 that picks up pattern changes first. Although it may be a moot point if it is actually a pattern change. To me it is a sufficient shift in the long wave pattern to say it is. Time will tell if it is a major one or a minor one.

 

Above doubt is that the 500mb flow WILL change from its westerly, twitching a touch south and a touch north of to a touch south of west over the past weeks or two, to a quite definite north of west direction. This will mean that ‘most’ of the time fairly cold and relatively deep Polar Maritime air will affect most of the country, at least in the 6-10 day time scale. So far the 8-14 NOAA does not indicate this moving into a more northerly Arctic air mass. There is still no signal for a general height rise Iceland to Greenland for this to look feasible. It may develop but it certainly is not showing yet.

NOAA 6-10 and 8-14, also have a read of the NOAA forecasters assessment, they can give some guide to what the main wave pattern is likely to be in their view. This can sometimes help in giving a bit more guidance for the eastern Atlantic/western Europe area.

http://www.cpc.noaa.gov/products/predictions/610day/500mb.php

 

ECMWF-GFS output

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

 

Looking at the synoptic models and those that show this idea out to at least 240h are pretty close to the overall upper pattern. The surface detail, troughs, even the much loved (!) short wave features are not discernible this far out on the anomaly charts but remember it is the upper air that is the major cause of these not the surface, although like much in meteorology there are times when it is not easy to separate one from the other. How much snow, where, when again is for the synoptic models at no more than 72h and more likely 12-24 h to give this detail.

Below is the link on Wetter which gives a quick glimpse of how GFS and ECMWF suggest things may look at 240h

http://www.wetterzentrale.de/topkarten/fsecmeur.html

As you might expect with small features running in this flow then timing is going to be problematical at this time range, hence the differences between the 2 GFS versions and ECMWF, but all 3 really following the idea suggested above.

 

 

Turning to further out than 10 days and we can see how the AO and NAO are looking although they are not drivers but indicators of what the atmosphere is being suggested by the other outputs. The link below show AO, NAO and MJO data (actual and forecast)

http://www.cpc.ncep.noaa.gov/index.php

I tend to mistrust AO, NAO and PNA much after about 7 days, as mentioned they are really only showing what the synoptic models are giving and are as likely to be incorrect at the range of much beyond about 7 days.

Turning to the GFS MJO it currently is in phase 5/6 in reasonable amplitude. The link below shows how this tends to fit in at 500mb in January

http://www.americanwx.com/raleighwx/MJO/MJO.html

Both of them show much as one should expect, that is a 500mb pattern not that far removed from what we have had over the past week or more.

What of its forecast?

This suggests with a lower amplitude but possibly still sufficient to have some effect on northern hemisphere weather edging into phase 7, follow the link above to see what it suggests for the 500mb pattern?

It does suggest a colder upper air pattern with ridging occurring N-E-SE of the UK with –ve heights out west.

That leaves the problem what sort of surface flow might get IF this is correct, and is the upper pattern likely to be that? This is where I leave the stage as I have not enough knowledge to be able to make sensible and constructive suggestions on this. I suspect on balance it may help for the colder spell to be longer than 5-7 days, possibly we MIGHT be looking into 10+ days and of course in that time the feed backs from such a situation will have impacts on how any of the teleconnections will then show.

So enjoy the colder spell and keep your fingers crossed that in that time some snow may fall on you, enjoy any frosty mornings and see what old man weather actually does. He may well make a complete fool of me but never mind.

Edited February 8, 2015 by johnholmes

[Summer Sun]

Agreement now for the end of the coming week with some rain likely for a time it may fall as snow over the highest ground in Scotland

 

 

Into the Weekend and pressure slowly starts to rise again with any rain starting to become more patchy again though a lot of cloud

 

 

Beyond next weekend it does look like pressure will rebuild leaving a lot of dry weather in any sunshine it should feel pleasant enough with temperatures around or slightly above average

 

[Eugene]

Well after the latest GFS debacle my confidence in anything that model shows past 5 days has hit rock bottom.