Jump to content
  • Member Statistics

    17,510
    Total Members
    7,904
    Most Online
    Toothache
    Newest Member
    Toothache
    Joined

February 12-13 Storm: Title TBD


stormtracker

Recommended Posts

Isn't there a point where the ensembles become less useful, this close to zero hour? This one starts on wednesday.

In terms of deterministic versus ensemble mean for the global models, the deterministic's higher resolution is of increasing importance inside of 120hr and dramatically so inside 96hr. But...

There are cases where the global ensembles still play an important role all the way down to 60-72 hours in advance. Particularly during very fast flow regimes, timing of full phasing, etc.

My opinion, would love to see verification numbers, is that the meso ensemble mean is more useful in the 30/36hr - 60hr range than the NAM and GFS. Mixed feelings on whether or not to lump the EC in to that last group.

The SREF will often have a sort of binomial distribution of solutions, say 40% - 50% with one solution, 30% - 40% tightly clustered around a second solution and 10% - 20% with scattered solutions. I've certainly seen a number of occasions where the primary or secondary solution cluster ends up being "more right" than the EC's verbatim 36hr - 60hr depiction.

Link to comment
Share on other sites

  • Replies 1.2k
  • Created
  • Last Reply

Even if it is right just because it caught on first does not mean the GFS is crap. This has been gone over ad nasuem and DTK said that the EURO is the best but the GFS is not trash which i and any other sensible person would agree with.

 

I don't disagree with what DTK said, am just loling at people who are saying the GFS sucks because its showing something completely different.

 

For all we know, the GFS is correct and we get nothing.

Link to comment
Share on other sites

Ignore the fact that the GFS has done well this winter and  the Euro that's had to play catch up a few times as well.

 

exactly, humans are pretty terrible at providing an unbiased sample of reality -- we are however very adept at selectively sampling points of a distribution to attempt to explain the perceived outcome. 

Link to comment
Share on other sites

PURP.

 

Hoping this doesn't fall prey to the curse of the bulseye/north trend. It probably will lol

 

 

Well, that's the furthest ampe now-- so, we have wiggle room on the other models. :)

Yeah, I'm talking to you GFS. 

 

I'd be more worried about sleet and freezing rain mixing than wide east. 

Link to comment
Share on other sites

Can't believe this is acceptable to people on charge of this stuff that the model is so poor

 

Dude...there've been a number of occasions where the GFS "get's it right" and the Euro stubbornly insists on a solution that ultimately fails. All the way down to the 72 hour timeframe. Especially when dealing with full phasing along the SE to Mid-Atlantic coast and especially so in the spring. Usually the EC will be all on it's lonesome (no NAM support) in those cases, though. 

Link to comment
Share on other sites

Dude...there've been a number of occasions where the GFS "get's it right" and the Euro stubbornly insists on a solution that ultimately fails. All the way down to the 72 hour timeframe. Especially when dealing with full phasing along the SE to Mid-Atlantic coast and especially so in the spring. Usually the EC will be all on it's lonesome (no NAM support) in those cases, though. 

I get that but it seems like in these "big ticket" events....its always falling short. I am sure it out performs the euro in many areas but when its playoff time...it seems to do a peyton manning:)

Link to comment
Share on other sites

1.  We do not even have verification, so thanks for chiming in already on model performance for a future event.

2.  I don't think either of you are experts on data assimilation or numerical weather prediction.

3.  The ECMWF is still the best in the world (by quite a bit I might add), but the UKMO/GGEM/GFS are all good models and not *that* far behind.

4.  Issues regarding resources at NCEP have been discussed time and time again.  Can they do things better?  Of course.  Is it all trash?  Don't be ridiculous.

5.  The GFS is being upgraded again this year to include a resolution increase to ~13 km, modified physical parameterizations, increased resolution in the ensemble part of the DA, and use of better resolution (spacial and temporal) SST initialization.

 

/rant

 

Thanks for the updates. I like hearing what's happening w/ these upgrades and changes. 

Link to comment
Share on other sites

fwiw, 6z NAM was 6-8 hours of snow for DC and then sleet and then back....

 

I think that is a pretty good call with almost any track right now given the thump.....4-6" of snow then probably flip, then back, but how soon depends on track....

 

I do like that the NAM is picking up on the fact that it will be 31-32 in deform, and not like 34.....

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Guest
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.

×
×
  • Create New...