Jump to content
Cold?
Local
Radar
Snow?

Model Output Discussion - 10th Jan Onwards


Recommended Posts

Posted
  • Location: just south of Doncaster, Sth Yorks
  • Location: just south of Doncaster, Sth Yorks

    Well there you have it. MOGREPS is clearly completely dropped the idea of decent cold and snow for to next couple of weeks. Time for a break, This one is gone folks.

     

    of course you MIGHT be misunderstanding what Ian means?

    and 2+2=4 not any other number.

    how do you know it is MOGREPS, of course it could be, it may also be from another source Exeter have?

    • Like 1
    Link to post
    Share on other sites
    Spotted a post you think may be an issue? Please help the team by reporting it.
    • Replies 2.6k
    • Created
    • Last Reply

    Top Posters In This Topic

    Posted
  • Location: Newcastle upon Tyne
  • Weather Preferences: Cold winters and cool, wet summers.
  • Location: Newcastle upon Tyne

    Mods, if you move this post to another thread can you PM and tell me where to so that I can check back for replies please.

     

    Ok, genuine questions ...

     

    Why do different models offer different solutions? Why do models rarely agree until almost T24? Why do they get so many details wrong the further out they go?

     

    The reason I ask is that with modern technology these days it appears (to me anyway) that forecasting (certainly long range in Winter) is no further forward than it was say a decade or so ago.

     

    Why do I not see any tangible improvement in forecasting the weather, particularly but not exclusively, during Winter (I must admit I only really have an interest in Winter or extreme weather). How can various models predict varying solutions until almost T24?

     

    Is this an investment issue?

     

    Furthermore, if FI is almost futile, why do the models go this far out? What is the point? Would it not be better to produce more accurate modelling say to 5/7 days rather than say 10 or more?

     

    These questions aren't borne out of frustration at lack of cold/snow, more why do we seem to be no further forward on long range forecasting than I think we were roughly 10 years ago. Or are we further forward and if we are, then why do we see models flip about like a fish in hot water?

     

    Thanks.

    • Like 3
    Link to post
    Share on other sites
    Posted
  • Location: Reigate Hill
  • Weather Preferences: Anything
  • Location: Reigate Hill

    ^^^ Latest Met update:

     

    "

    UK Outlook for Sunday 26 Jan 2014 to Sunday 9 Feb 2014:

    Whilst current indications point towards changeable conditions persisting into late January and early February, they are not expected to be as stormy as the start of this month. The most likely scenario during this period would see northwestern parts of the UK affected by the most frequent bouts of unsettled weather leaving southern and eastern regions tending to see the most prolonged spells of drier and brighter conditions. Some periods of colder weather are likely, particularly between weather systems, but overall temperatures may well remain near normal.

    Updated: 1159 on Sat 11 Jan 2014"

    • Like 1
    Link to post
    Share on other sites
    Posted
  • Location: Solihull, WestMidlands, 121m asl -20 would be nice :-)
  • Weather Preferences: Cold and Snow -20 would be nice :)
  • Location: Solihull, WestMidlands, 121m asl -20 would be nice :-)

    At least FEB & MARCH look interestingPosted Image

     

    Posted Image

    Posted Image

    • Like 1
    Link to post
    Share on other sites
    Posted
  • Location: West Northants
  • Weather Preferences: Cold Winters, Warm Summers.
  • Location: West Northants

    fergieweather

    Hey @Claire7Carter, your 'Say No to Snow' t-shirt clearly having impact if v latest Met Office assessment into 10-15d period proves correct11/01/2014 11:49

    Edited by JOPRO
    Link to post
    Share on other sites
    Posted
  • Location: just south of Doncaster, Sth Yorks
  • Location: just south of Doncaster, Sth Yorks

    Well that update from the mogreps model is the final nail in our winter coffin I feel. That outlook takes us to end of jan almost. Roll on spring.

     

    again I ask has Ian actually quoted MOGREPS as his source of information, whatever that is, as I have not yet seen a quote from him?

    • Like 6
    Link to post
    Share on other sites
    Posted
  • Location: Pontypridd, Wales 240m asl
  • Location: Pontypridd, Wales 240m asl

    fergieweatherHey @Claire7Carter, your 'Say No to Snow' t-shirt clearly having impact if v latest Met Office assessment into 10-15d period proves correct11/01/2014 11:49

    Clair Carters t shirt, this is the reason for our crap winter, it's the one factor that has been controlling the models, someone better let James madden and the express know. That way they can factor it in to their forecasts! I bet MOGREPS was updated to include this factor and is why it's been performing so well! LOL
    Edited by andymusic
    • Like 1
    Link to post
    Share on other sites
    Posted
  • Location: st albans
  • Location: st albans

    incidentally, despite the gem op being the most cold biased model throughout this period of watching, the gem ens have been the least cold on uppers despite what looked like a decent mean and anomolys. i assumed it was a model bias but perhaps not.  perhaps if we get a proper cold spell this winter, we will get a chance to investigate this further. 

     

    tamara- this flushing down of the zonal winds into the trop was shown on the berlin charts some time ago and i mentioned it as the models began to amplify. what confuses me is that the ecm op, which generates the berlin strat data, was amongst the keenest output to amplify and slow the jet too much. so on the one hand, you had the same model run, slowing the lower trop momentum in the high lats and yet also showing the stronger zonal flow flushing down at the same time ?  maybe you or ed can help on this ?

    Edited by bluearmy
    • Like 2
    Link to post
    Share on other sites
    Posted
  • Location: West Northants
  • Weather Preferences: Cold Winters, Warm Summers.
  • Location: West Northants

    Ignor button on overtime this morning,normal doom merchants at it again! Can't we just see what rolls out of the models rather than having hysterical responses to each run/MO/CFS/EC 32dayer update. Stop flapping people and just see what transpires,if every model was right we would not be on here talking about the weather.

    Edited by JOPRO
    • Like 5
    Link to post
    Share on other sites
    Posted
  • Location: West Sussex
  • Weather Preferences: Extreme cold & snow
  • Location: West Sussex

    Well that update from the mogreps model is the final nail in our winter coffin I feel. That outlook takes us to end of jan almost. Roll on spring.

    Well the medium term 6-15 day update from METO today mentions it perhaps becoming a little colder from the east of north east which is an improvement on yesterday!

     

    I won't post it because it will just be deleted so her is the link http://www.metoffice.gov.uk/weather/uk/se/se_forecast_temp.html

    Link to post
    Share on other sites
    Posted
  • Location: Alresford, Near Colchester, Essex
  • Weather Preferences: As long as it's not North Sea muck, I'll cope.
  • Location: Alresford, Near Colchester, Essex

    Thanks Tamara for your well considered and detailed, analytical postings here. They are always a welcome sight! :)

    • Like 9
    Link to post
    Share on other sites
    Posted
  • Location: Broxbourne, Herts
  • Weather Preferences: Snow snow and snow
  • Location: Broxbourne, Herts

    again I ask has Ian actually quoted MOGREPS as his source of information, whatever that is, as I have not yet seen a quote from him?

    Likewise, even if he did,  was it the same infallible MOGREPS now saying there;s no chance of snow that was telling us a couple of days ago there was a chance?All models apply data to algorithms to forecast the weather ahead..... as recent days have proved, there's no exact science involved....as days ahead will prove again 

    • Like 2
    Link to post
    Share on other sites
    Posted
  • Location: Reigate Hill
  • Weather Preferences: Anything
  • Location: Reigate Hill

    again I ask has Ian actually quoted MOGREPS as his source of information, whatever that is, as I have not yet seen a quote from him?

    Hi JohnMogreps has a bit of street cred, as it was the only model that dismissed the phantom easterly that all other models picked up on a few days ago.The latest ec32 is probably what the Met update is based on. That wasn't far from the above.
    Link to post
    Share on other sites
    Posted
  • Location: Wombwell, South Yorkshire (75m / 246ft ASL)
  • Weather Preferences: Summer Sunshine / Winter Snow
  • Location: Wombwell, South Yorkshire (75m / 246ft ASL)

    Even if Ian is quoting MOGREPS (we don't know) and even if MOGREPS is correct (it isn't always) then why is winter being written off when it only forecasts the next two - four weeks?

     

    We are in a fascinating setup at the moment and as BA says nothing has been decided for this time next week let alone the rest of winter!

     

     

     

    I think a little perspective is required here...for example this time last week an Easterly seemed a million miles away and look what a ride the models have taken us on since!

    • Like 2
    Link to post
    Share on other sites
    Posted
  • Location: just south of Doncaster, Sth Yorks
  • Location: just south of Doncaster, Sth Yorks

    Mods, if you move this post to another thread can you PM and tell me where to so that I can check back for replies please.

     

    Ok, genuine questions ...

     

    Why do different models offer different solutions? Why do models rarely agree until almost T24? Why do they get so many details wrong the further out they go?

     

    The reason I ask is that with modern technology these days it appears (to me anyway) that forecasting (certainly long range in Winter) is no further forward than it was say a decade or so ago.

     

    Why do I not see any tangible improvement in forecasting the weather, particularly but not exclusively, during Winter (I must admit I only really have an interest in Winter or extreme weather). How can various models predict varying solutions until almost T24?

     

    Is this an investment issue?

     

    Furthermore, if FI is almost futile, why do the models go this far out? What is the point? Would it not be better to produce more accurate modelling say to 5/7 days rather than say 10 or more?

     

    These questions aren't borne out of frustration at lack of cold/snow, more why do we seem to be no further forward on long range forecasting than I think we were roughly 10 years ago. Or are we further forward and if we are, then why do we see models flip about like a fish in hot water?

     

    Thanks.

     

    hi

    Best to get the comments from Jo or Ian F but I will try and give you my answer.

    Overall, as was iluustrtaed in the video clip about the 60 years Met have done BBC weather forecasts, the 4 day forecast is more accurate than the 1 day was 30 years ago. For verification and who says this I would e mail Exeter I am sure they will reply.

    As to why different models give different solutions on the same run then has to be down to how they deal with the data. GFS, belonging to NOAA, they readily admit they have problems and are seeking help from ECMWF to try to overcome these issues. Several years ago GFS was quite clearly the leader, I did some work on the accuracy at 168h and it was certainly better, for the UK, than it is now.

    When you question winter being worse than summer-is this because you feel the models predict cold and snow and it does not arrive. If it is that then predicting cold blocks and once formed predicting the arrival of milder weather has always been a major issue. This in the days prior to models being available and human forecasting going no further 1-2 days ahead or now when models run out to 144h minimum and 384h, that we see, as a maximum. Basic physics but I have no idea why both parts of that last sentence have proved so difficult to resolve. If we get the cold air in then the problem is will it be rain or snow. As you may have seen regular posts from me, predicting rain is harder than predicting the over pressure pattern at the surface. Predicting snow rather than rain requires a further 7 variables on top of has the model got the precipitation in the right place at the right time?

     

    hope this helps in some way to answer your questions but please pm me if you want to chat further.

    Hi JohnMogreps has a bit of street cred, as it was the only model that dismissed the phantom easterly that all other models picked up on a few days ago.The latest ec32 is probably what the Met update is based on. That wasn't far from the above.

     

    but you are guessing-I have pm's Ian to ask if he will give the definitive answer from himself, so best we wait and see what he says?

    • Like 5
    Link to post
    Share on other sites
    Posted
  • Location: NR LOURDES SW FRANCE
  • Location: NR LOURDES SW FRANCE

    Mods, if you move this post to another thread can you PM and tell me where to so that I can check back for replies please.

     

    Ok, genuine questions ...

     

    Why do different models offer different solutions? Why do models rarely agree until almost T24? Why do they get so many details wrong the further out they go?

     

    The reason I ask is that with modern technology these days it appears (to me anyway) that forecasting (certainly long range in Winter) is no further forward than it was say a decade or so ago.

     

    Why do I not see any tangible improvement in forecasting the weather, particularly but not exclusively, during Winter (I must admit I only really have an interest in Winter or extreme weather). How can various models predict varying solutions until almost T24?

     

    Is this an investment issue?

     

    Furthermore, if FI is almost futile, why do the models go this far out? What is the point? Would it not be better to produce more accurate modelling say to 5/7 days rather than say 10 or more?

     

    These questions aren't borne out of frustration at lack of cold/snow, more why do we seem to be no further forward on long range forecasting than I think we were roughly 10 years ago. Or are we further forward and if we are, then why do we see models flip about like a fish in hot water?

     

    Thanks.

    You pose some interesting questions JP, generally though its the type of synoptic set up that means reliability drops, so in a typical zonal flow its easier to get the detail correct in later timeframes, anytime theres blocking to the ne reliability drops considerably because of the more complex nature of the set up.

     

    NWP has though made a lot of improvements over the years however I remain firmly of the belief that LRF's still remain very error prone, theres just too much chaos in the atmosphere to put more detail on weather months ahead.

     

    I think the way forward in that respect was more what Glacier Point was doing rather than  a simple start point and running a model out to months ahead,of course the use of ensemble forecasting has made a difference and that's probably been the best step taken by the various global model providers, but you still in a sense have a built in error because you simply can't map every part of the atmosphere.

     

    The GFS lower resolution is really an exercise in how to waste a lot of money for little reward, the output past T240hrs is prone to so many changes, NCEP of course must have verification stats for past day ten but are obviously too embarrassed to publish them.

     

    Of course you need scientists pushing the boundaries to move forward but sometimes you have to hold your hands up and say that in terms of weather detail is simply impossible past a certain point, its better they use their resources to perhaps move towards the more teleconnective route and forget about dishing up worthless charts at T360hrs which have no value other than to simply reinforce the belief that you'd be better off pulling a synoptic outcome out of  a hat at that range.

    Edited by nick sussex
    • Like 3
    Link to post
    Share on other sites
    Posted
  • Location: Worcester
  • Location: Worcester

    Hi JohnMogreps has a bit of street cred, as it was the only model that dismissed the phantom easterly that all other models picked up on a few days ago.The latest ec32 is probably what the Met update is based on. That wasn't far from the above.

    But it didn't dismiss the Easterly completely was it not 50/50 as opposed to 60/40 for the EC according to Fergie on Wednesday?
    • Like 1
    Link to post
    Share on other sites
    Posted
  • Location: Solihull, WestMidlands, 121m asl -20 would be nice :-)
  • Weather Preferences: Cold and Snow -20 would be nice :)
  • Location: Solihull, WestMidlands, 121m asl -20 would be nice :-)

    I hope some on here take the time to read your post Tamara, instead of franticly chasing every detail and run for snow. 

     

    It's the 11th Jan, not even half way through winter and if some understood just how strong the Jet Stream has been over past month or 2 then nothing really comes as a surprise, but with further SSW model'd and an ever weakening Vortex leading to a meandering Jet and the continent cooling, we are now in a far better position for something more seasonal than we have been all winter.

    Regarding an ever weakening Vortex looking at this it seems to be going that way Posted Image

     

    Posted Image

    Edited by Dancerwithwings
    • Like 7
    Link to post
    Share on other sites
    Posted
  • Location: Lytchett Matravers - 301 ft ASL
  • Weather Preferences: Snowy Winters, Torrential Storm Summers
  • Location: Lytchett Matravers - 301 ft ASL

    Just to clear something up, I asked Ian F on Twitter and he said MOGREPS never really went for the easterly at all.

     

    Thanks for the reply Ian, saves speculating

    Edited by karlos1983
    • Like 4
    Link to post
    Share on other sites
    Posted
  • Location: Darlington
  • Weather Preferences: Warm dry summers
  • Location: Darlington

    At least FEB & MARCH look interestingPosted Image

     

    Posted Image

    Posted Image

     

    Sorry but if that CFS model was correct from what it was showing during most of Autumn we'd have had non stop northern blocking since December

     

    I'm afraid CFS is no better or worse than any other model

    • Like 6
    Link to post
    Share on other sites
    Guest
    This topic is now closed to further replies.
    ×
    ×
    • Create New...