Jump to content

Model Output Discussion 01/09/17


Recommended Posts

More fascinating charts and scenarios. Two very differing scenarios with GFS/ECM in the one corner and UKMO in the other. Fight night part 2. Who's everyone' money on? Has to be a blended solution I reckon but how strong the blend. Anyone recall when UKMO backed down towards GFS.

What a great period of model watching we are living in

  • Like 4
Link to post
Share on other sites
  • Replies 6.4k
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

Good morning Gang ,Its great to be back reading the Posts on Net weather .My Wife passed away two weeks ago ,she had Pancreatic Cancer which spread ,but enjoyed life since she was diagnosed back in Ap

Good afternoon gang ,its nice to be back reading all the posts again ,although i have been lurking .(Thanks to all friends on here for my much appreciated messages over the loss of my wife recently ,T

So sorry to hear that mate Dont want to clutter up this thread but im sure i speak for everyone in wishing you the very best and hope you will be ok.

Posted Images

Actually the ECM 240  mean is a very slight upgrade in terms of 850's, the -4c line is now Wirral / North Wales (ish) to London (ish), the 0z it was North of Liverpool to Suffolk, synoptically although some might say the ridge isn't as higher latitude so its a downgrade, I think shortly afterwards there would be a decent cluster which produces a snow event.

Edited by feb1991blizzard
Link to post
Share on other sites

MODEL VERIFICATION

There is always a lot of disagreement on here as to which model has performed better or worse compared to another. This has been strongly in evidence again today.  I know that verification charts are sometimes posted but I imagine I am not the only one who finds these difficult to read and compare. I remember when @Gibby used to produce his superb daily analysis (up to about 18 months ago, I believe) and he used to provide simple model verification statistics. I’m not sure if these are easily available.

Anyway, it occurred to me that we can do our own verification exercise whenever we want to and in a way we can all easily relate to. The Meteociel site provides archive records for the main models. So, I’m going to do a cross model comparison to today’s actual 12z chart. Then I’ll show each models’ chart for this time at Day 5 and at Day 10 (where available, as UKMO only goes to day 6 and NAVGEM to day 7/8).

Current Chart showing 1300 on November 26th ECM 12z T+0:

ECH1-0.GIF?12 

 

Day 5 T+120 Forecast Charts from 1300 on November 21st – all are 12z:

                  ECM                                                             GFS                                                           UKMO                                         

ECH1-120.GIF?12                         gfsnh-0-120.png?12                          UN120-21.GIF?21-12   

                  GEM                                                         NAVGEM                                                        JMA

 gemnh-0-120.png                        navgemnh-0-120.png?21-18                          JN120-21.GIF?26-12 

 

Day 10 T+240 Charts from 1300 on November 16th – all are 12z:

                   ECM                                                            GFS                                                           GEM                                                            JMA

ECH1-240.GIF?12                   gfsnh-0-240.png?12                    gemnh-0-240.png                  JN240-21.GIF?26-12

Now we can all disagree upon which are the best and worst matches!  Remember most of the models flipped flopped all over the place from run to run, not just the FI charts. It may well have been the case that a particular model went off track on the 12z run only to nail it on their next run! I cannot go into ensemble charts and all the perturbations, that would take me all night and in any case, I could not find an archive link for the GEFS ensemble mean or any of their earlier run charts.

None of the Day 5 charts are too far from the current chart and in my view, perhaps JMA takes the prize with ECM a close second. Possibly the two USA models GFS and NAVGEM get the wooden spoon. .Not surprisingly, the Day 10 charts are much further away from the current position, in fact they fail quite abysmally and I do not feel any of them deserve a prize. This type of exercise could be done with just one model taking 10 of their runs for each day or even every single run for a certain period.

In future if an argument crops up on who performed well or badly, remember that the evidence is available!!! 

Edited by Guest
Correct typos
Link to post
Share on other sites

Hoping for cross model agreement on the Gfs 12z solution..I'm greedy!:D:cold:

Anyway, cold shot number 1 is locked and loaded so enjoy the northern / arctic chill after Monday.:drinks:

Edited by Frosty.
  • Like 1
Link to post
Share on other sites

I was planning to slip in quietly to open my winter account. But not much chance of that! Just to say that those ECM means are the best I as a coldie have seen for several winters past and we are technically only just entering that season. In some ways they remind me of December 1981, one model and too early to get carried away I know. ? Just enjoy the ride.

  • Like 3
Link to post
Share on other sites
1 hour ago, Daniel* said:

Folk are missing out what is in front of them. Is this eye candy? At day 2 a sight to behold after all this woeful muck of the last few years. The first wave of cold has been picked up very well. 

14B72679-D6A2-4E13-B667-1B04D7BC5FA0.thumb.gif.da1daa411f4274f9ecc03d3e7121b6df.gifEBFE8D2D-D5FE-4215-9A67-E24663348CF4.thumb.png.43673d76de7ac6f3493ca429838dce4c.png4898E654-8AAF-4DA1-88E5-F9D86FE8EA16.thumb.png.1ddeab74e4b3aa0d57c3d101467ada38.png

EDIT

I just got home and posted thebelow as I came across the post and responded to it before reading through the rest of the thread but see this has already been covered by BB1962. (That's twice in 2 days:sorry:)

LOL

 

Agree with the eye candy and what is in front but not so sure about the models performing well.

None of them spotted the Atlantic ridge on the 27th until it was fairly close in (well within 10 days if memory serves. They have been all over the map since then with the placement of the ridge, amplification and sharpness of the trough.

TBF it has been quite a complex pattern but the look ahead to a possible sustained cold spell has less moving parts so they should have it resolved much quicker.

Edited by Mucka
  • Like 2
Link to post
Share on other sites

December 1981 began with high pressure over the UK which then pulled back into the Atlantic and joined up with a large Greenland High.

A low moving south east across England on the 8th ushered in bitterly cold northerly winds and the rest as they say is history.

Tonight's GFS shows a remarkably similar scenario as does ECM, lets hope history is repeating itself 36 years later 

Andy

  • Like 5
Link to post
Share on other sites
4 minutes ago, bobbydog said:

we also have backup from higher up.

the strat i mean, not some higher being...

gfsnh-10-384-7.thumb.png.b69fa01bf26655ecdd55a47215db64e4.png

npst30.thumb.png.9c9d61a846ade1beafc6a7c220a70b08.png

u_65N_10hpa_gefs.thumb.png.5b16e8857c21345a01ef8968bbd31217.png

u_65N_10hpa-1.thumb.png.eb5e77ffd340d200cc359e044de317ab.png

that should have a big impact further down the line and will probably have knocker hiding in his woodshed....

jesse-fast-show.thumb.jpg.8a1141153689815f7a789c915b20d1c8.jpg

The stratospheric aspect was the missing part of the puzzle RE this winter as far as I was concerned...hence the reason for my negativity early doors. If we can see a warming through to the finish line then all options are on the table...

  • Like 3
Link to post
Share on other sites
1 minute ago, CreweCold said:

The stratospheric aspect was the missing part of the puzzle RE this winter as far as I was concerned...hence the reason for my negativity early doors. If we can see a warming through to the finish line then all options are on the table...

in the winter thread, we were drawing comparisons with 1962. that year had similar patterns and background signals to this year. it also had a minor warming in early december then a major one in january. as it stands, a minor warming will do and that seems to be on the cards. :wink:

  • Like 6
Link to post
Share on other sites

The ECM ens shows it turning colder from Tuesday through to the end of the week over the weekend it briefly turns less cold before another plunge of colder air into the 1st full week of December

London

graphe_ens3_php.png1.thumb.png.c226cc5d739f0ec666dd31cd01761694.png

Northern Scotland

graphe_ens3_php.thumb.png.8446f2c0ef7ba6eb5901ccfecea1ce2c.png

  • Like 2
Link to post
Share on other sites

Very clear trend on De bilt ensembles. Steady cool down this week then a notable warm up for a few days. this is then followed by a steady decrease again - remarkable support for this at this range with only a few members going differently (including control).

 

ensemble-verwachtingen-detail10.thumb.png.1f8cb51313556adc545c70977825a801.png

Link to post
Share on other sites
3 hours ago, carinthian said:

Looking at the latest chart from UKMO, not good . Seems if it wants to take the Atlantic trough northwards away from the Azores. This can only lead to sinking of the high close to the West of British Isles. There will hopefully be a correction in later outputs to dig in the trough and undercut the high. This will be the best solution to retrogress the high polewards again and open up Northerlies again as mentioned in there medium range forecasts. Its a shame we have no access to these charts past 144t but I am sure this latest run at 144t does not fit in with the general consensus shown by the other models. The form horse this past few weeks has been the persistent lower heights near to the Azores and this has been a prominent factor of the blocking scenario.This latest run from UKMO wants to scupper that.

C

UKMOPEU12_144_1.png

Following on from the above post, the UKMO extended chart at 168t seems to confirm a big difference compared the other 2 big models, GFS and ECM. The UKMO replaces the lower heights and builds the Azores high towards the British Isles with the Atlantic trough heading north. In stark contrast at the same time,GFS and ECM maintain the lower Azores heights with retrogressive high moving poleward. So will UKMO be the winner or loser against all odds .?

ukm2.2017120312.168.lant.troplant.prp.fcst.gentracker.png

Edited by carinthian
  • Sad 2
Link to post
Share on other sites

No idea of late but a general pattern for the models success is indentifable. ECM top dog of course, the canadian model GEM (the black line) ought to be wheeled out more often, greater success at day 5 than GFS in many occasions. On rare occasions the canadian model has outperformed the ECM at day 5, however when it goes wrong.. it can be the worst of the bunch. Sadly nothing of UKMO but I sense it will just be knocking second.. 

 

Link to post
Share on other sites
36 minutes ago, bobbydog said:

we also have backup from higher up.

the strat i mean, not some higher being...

gfsnh-10-384-7.thumb.png.b69fa01bf26655ecdd55a47215db64e4.png

npst30.thumb.png.9c9d61a846ade1beafc6a7c220a70b08.png

u_65N_10hpa_gefs.thumb.png.5b16e8857c21345a01ef8968bbd31217.png

u_65N_10hpa-1.thumb.png.eb5e77ffd340d200cc359e044de317ab.png

that should have a big impact further down the line and will probably have knocker hiding in his woodshed....

jesse-fast-show.thumb.jpg.8a1141153689815f7a789c915b20d1c8.jpg

Great to have some humour/character alongside decent knowledge...

Im dying here???onto the 18z...

  • Haha 1
Link to post
Share on other sites
3 minutes ago, Daniel* said:

No idea of late but a general pattern for the models success is indentifable. ECM top dog of course, the canadian model GEM (the black line) ought to be wheeled out more often, greater success at day 5 than GFS in many occasions. On rare occasions the canadian model has outperformed the ECM at day 5, however when it goes wrong.. it can be the worst of the bunch. Sadly nothing of UKMO but I sense it will just be knocking second.. 

 

It says in the text UKMO was tied 4th (i.e behind theses three).

Link to post
Share on other sites
Guest
This topic is now closed to further replies.
×
×
  • Create New...