Jump to content

  • Member Statistics

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

February 12-13 Storm, Part III: Trilogy ends and then Obs thread soon!

stormtracker

By stormtracker,
in Mid Atlantic

Recommended Posts

  • Replies 1.5k
  • Created
  • Last Reply
wxmeddler

wxmeddler

Posted
Posted

Hires:  4kNAM, RGEM, RAP, HRRR, NAM?, Euro?

Globals:  GGEM, GFS, UKIE

 

I thought I previously read that the Euro is a hires model but I can't remember.  Just trying to get a handle on which models are the higher res vs. the others.  

 

Euro is a global but has the highest resolution of the globals.

Link to comment
Share on other sites
vinylfreak89

vinylfreak89

Posted
Posted

Convective precipitation isn't necessarily driven by the main (upper level) large scale dynamic forcing.  I would have to look at profiles, cap, and then find the trigger(s).  The convection looks to be driven from the bottom, since at hour 30 there is strong low level warm advection, some frontogenetical forcing, and perhaps some low level convergence.  With the correct profiles, the model has no choice but to trigger convection in the warm sector to adjust.

 

I have not looked at lower level stuff, but I thought that warm sector convection in winter on the NE side of a ULL is a pretty big statistical anomaly?

 

Don't see much convergence on this map as modeled? What are some good panels (other than 850 winds) for looking at LL convergence?

 

gfs_namer_030_850vor_500ht_200wd_s.gif

Link to comment
Share on other sites
Ian

Ian

Posted
Posted

Neither?  To be honest, I'm not sure.  One certainly wouldn't normally want to rely on a coarser resolution model for a short range, dynamic forecast like this.  However, the GFS is generally decent guidance (with weaknesses and biases in certain regimes/regions/situations not being forgotten about) and I don't think it can simply be ignored.

The GFS is super good these days overall. It has some known weaknesses that seemingly show up on the regular.  I think the fact that people didn't immediately all hop on the Euro with this one shows that the gap has closed.. as it probably will continue to do.

Link to comment
Share on other sites
dtk

dtk

Posted
Posted

Hires:  4kNAM, RGEM, RAP, HRRR, NAM?, Euro?

Globals:  GGEM, GFS, UKIE

 

I thought I previously read that the Euro is a hires model but I can't remember.  Just trying to get a handle on which models are the higher res vs. the others.  

 

The ECMWF is a global (spectral) model as well but run at higher spatial resolution than the other globals.  I think they are at T1279 right now, or ~15km or so (GFS is currently ~27km, off the top of my head the Canadian is now under ~25km, and UKMet probably also in the low 20km range). 

 

We are in the late stages of finalizing our 13km GFS package to be implemented later this year.

Link to comment
Share on other sites
Deck Pic

Deck Pic

Posted
Posted

Hires:  4kNAM, RGEM, RAP, HRRR, NAM?, Euro?

Globals:  GGEM, GFS, UKIE

 

I thought I previously read that the Euro is a hires model but I can't remember.  Just trying to get a handle on which models are the higher res vs. the others.  

 

In my experience once you get inside 24 hours, the euro's superiority is mitigated when it comes to details....sometimes it takes a step back...it can be useful inside of 24 hours when blended and also as a check on other models when its QPF is lower....it is absolutely the best model at QPF...but when you get close, it isn't necessarily going to account for banding and other mesoscale features...Tonight's run is important...I dont think 12z tomorrow matters much unless it is a big shift from 0z...

Link to comment
Share on other sites
tcutter

tcutter

Posted
Posted

The ECMWF is a global (spectral) model as well but run at higher spatial resolution than the other globals. I think they are at T1279 right now, or ~15km or so (GFS is currently ~27km, off the top of my head the Canadian is now under ~25km, and UKMet probably also in the low 20km range).

We are in the late stages of finalizing our 13km GFS package to be implemented later this year.

Wow, so the GFS is soon to have the best resolution?

Link to comment
Share on other sites
dtk

dtk

Posted
Posted

I have not looked at lower level stuff, but I thought that warm sector convection in winter on the NE side of a ULL is a pretty big statistical anomaly?

 

Don't see much convergence on this map as modeled? What are some good panels (other than 850 winds) for looking at LL convergence?

 

Look at the lower level winds (not 200/jet level).

Link to comment
Share on other sites
vinylfreak89

vinylfreak89

Posted
Posted

The ECMWF is a global (spectral) model as well but run at higher spatial resolution than the other globals.  I think they are at T1279 right now, or ~15km or so (GFS is currently ~27km, off the top of my head the Canadian is now under ~25km, and UKMet probably also in the low 20km range). 

 

We are in the late stages of finalizing our 13km GFS package to be implemented later this year.

 

will the 13 km implementation include 4dVAR assimilation?

Link to comment
Share on other sites
vinylfreak89

vinylfreak89

Posted
Posted

Look at the lower level winds (not 200/jet level).

 

derp i misread that... thought it was 850 winds as well 850 vort... confusing graphics are confusing

 

ETA: ok with the right map....convergence signature looks well inland (S C VA to me) though I do admit I see some convergence along the coast where it is indicating convection

 

gfs_namer_030_850_vort_ht_s.gif

Link to comment
Share on other sites
dtk

dtk

Posted
Posted

will the 13 km implementation include 4dVAR assimilation?

No, the implementation this year will be mostly "model only".  All of the DA changes that go with it will be observation related (qc, bias correction, etc.).  They plan to implement a hybrid 4D EnVar (not the same as traditional 4DVAR) sometime in FY15.

Link to comment
Share on other sites
dtk

dtk

Posted
Posted

The GFS is super good these days overall. It has some known weaknesses that seemingly show up on the regular.  I think the fact that people didn't immediately all hop on the Euro with this one shows that the gap has closed.. as it probably will continue to do.

NCEP's global model has lagged behind ECMWF for decades now.  There is no simple fix to get us caught up to them.  They have a much narrower mission, dedicated (huge) computer, pretty good (financial) resources, and a lot of brilliant people.  That being said, we (and this is my opinion) can and should do better.

Link to comment
Share on other sites
nj2va

nj2va

Posted
Posted

In my experience once you get inside 24 hours, the euro's superiority is mitigated when it comes to details....sometimes it takes a step back...it can be useful inside of 24 hours when blended and also as a check on other models when its QPF is lower....it is absolutely the best model at QPF...but when you get close, it isn't necessarily going to account for banding and other mesoscale features...Tonight's run is important...I dont think 12z tomorrow matters much unless it is a big shift from 0z...

 

Thanks Matt.  I would expect the Euro to come a bit east tonight but thats 50% weenie-ism/50% reaction based on the 0z runs tonight (sans the GFS).  Or at least I don't expect it to skip further west.

Link to comment
Share on other sites
wxmeddler

wxmeddler

Posted
Posted

Thanks.  So with a system as complex as this storm, we should be paying closer attention to the higher resolution while using the others as a "secondary" tool now that we're within 24ish hours?

 

ETA:  Thanks dtk..looking forward to the upgrades.

 

Each has their strengths and weaknesses. Higher resolution models (under 10km or so) tend to develop "runaway" biases the further in time you get, that's why things like the RAP and HRRR are really only good out to the first 10-12 hours or so. Globals are much less prone to that as they "average" things out over their larger resolution.

 

Re: dtk, I personally love the GFS in every day situations if I were to pick one model for any given day of the year it would be the GFS, can't wait to see the 13km res parallel.

Link to comment
Share on other sites
Ian

Ian

Posted
Posted

NCEP's global model has lagged behind ECMWF for decades now.  There is no simple fix to get us caught up to them.  They have a much narrower mission, dedicated (huge) computer, pretty good (financial) resources, and a lot of brilliant people.  That being said, we (and this is my opinion) can and should do better.

Yeah with ECMWF continuing to advance quickly it probably will be difficult to get fully equal soon.  One thing (among many) the US weather community lacks is a strong voice in the real policy world. With all the hullabaloo surrounding Sandy including tons of mischaracterizations of where we stand compared to the Europeans, at least it got some money flowing. 

Link to comment
Share on other sites
vinylfreak89

vinylfreak89

Posted
Posted

Yeah with ECMWF continuing to advance quickly it probably will be difficult to get fully equal soon.  One thing (among many) the US weather community lacks is a strong voice in the real policy world. With all the hullabaloo surrounding Sandy including tons of mischaracterizations of where we stand compared to the Europeans, at least it got some money flowing. 

 

I know this is sort of OT, but has the option of distributed computing ever been breached to eek out more power on a lower budget?

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.
Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...