Using GFS 0.25° grid dataset

Compiling NMM/ARW code for EMS system, upgrading WRF cores on your own, changing EMS scripts to suit your particular needs, and other modifications to original EMS distribution goes in this forum. These are officially unsupported actions.
meteoadriatic
Posts: 1512
Joined: Wed Aug 19, 2009 10:05 am

Re: Using GFS 0.25° grid dataset

Post by meteoadriatic » Mon Nov 03, 2014 7:28 pm

surgeon wrote:NOMADS grib filter offers GFS 0.25 since few days.
How stable this NOMADS works? I know only bad things about it's reliability in past. What is your experience?

And... how to use that service (grib filter)?

Gippox
Posts: 59
Joined: Sat Mar 02, 2013 1:16 am

Re: Using GFS 0.25° grid dataset

Post by Gippox » Mon Nov 03, 2014 9:24 pm

surgeon wrote:
Gippox wrote:...unfortunately i also have to wait NOMADS grib filter service since my DSL is not sufficiently fast ... :roll: .
NOMADS grib filter offers GFS 0.25 since few days.
Thank you, I have seen now ...it was on the bottom of the page... ;)

Hello Meteo you must create yourself a little script ftp, where for example you can use curl cycle in this way:

Code: Select all

curl "http://nomads.ncep.noaa.gov/cgi-bin/filter_gfs_0p25.pl?file=gfs.t"$CYCLE"z.pgrb2.0p25.f0"$TIME"&all_lev=on&all_var=on&subregion=&leftlon=-10.00&rightlon=30.50&toplat=57.50&bottomlat=33.00&dir=%2Fgfs."$(date +"%Y")""$DATE_MONTH""$START_DAY""$CYCLE"" -o /home/andrea/wrfems/runs/arw/grib/"$(date +"%#y")""$DATE_MONTH""$START_DAY""$CYCLE".gfs.t"$CYCLE"z.pgrb2f"$TIME"
So you extract all levels and variables available, for coordinates instead put the corners (&leftlon=-10.00&rightlon=30.50&toplat=57.50&bottomlat=33.00) of your external domain ...

meteoadriatic
Posts: 1512
Joined: Wed Aug 19, 2009 10:05 am

Re: Using GFS 0.25° grid dataset

Post by meteoadriatic » Mon Nov 03, 2014 10:01 pm

Thx! Did you found that server and service stable enough for serious production use?

surgeon
Posts: 44
Joined: Wed Dec 08, 2010 1:34 pm
Location: Poland
Contact:

Re: Using GFS 0.25° grid dataset

Post by surgeon » Tue Nov 04, 2014 10:20 am

meteoadriatic wrote:How stable this NOMADS works? I know only bad things about it's reliability in past. What is your experience?
meteoadriatic wrote:Thx! Did you found that server and service stable enough for serious production use?
They sometimes catch some delays, but usually it works properly.
In example right now GFS 0.5 download is very slow, and GFS 0.25 is quite fast (download speed is dependent on the size of the domain). But GFS 0.25 is available little delayed to GFS 0.5.

Gippox
Posts: 59
Joined: Sat Mar 02, 2013 1:16 am

Re: Using GFS 0.25° grid dataset

Post by Gippox » Tue Nov 04, 2014 9:37 pm

surgeon wrote:
meteoadriatic wrote:How stable this NOMADS works? I know only bad things about it's reliability in past. What is your experience?
meteoadriatic wrote:Thx! Did you found that server and service stable enough for serious production use?
They sometimes catch some delays, but usually it works properly.
In example right now GFS 0.5 download is very slow, and GFS 0.25 is quite fast (download speed is dependent on the size of the domain). But GFS 0.25 is available little delayed to GFS 0.5.
Sometimes this error occurs

Code: Select all

0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0
  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0
100   782    0   782    0     0    644      0 --:--:--  0:00:01 --:--:--   644
100 45576    0 45576    0     0  20672      0 --:--:--  0:00:02 --:--:-- 20669
100 89049    0 89049    0     0  26627      0 --:--:--  0:00:03 --:--:-- 26621
100  126k    0  126k    0     0  30023      0 --:--:--  0:00:04 --:--:-- 30019
100  169k    0  169k    0     0  32550      0 --:--:--  0:00:05 --:--:-- 34175
100  224k    0  224k    0     0  36322      0 --:--:--  0:00:06 --:--:-- 44763
100  296k    0  296k    0     0  42143      0 --:--:--  0:00:07 --:--:-- 51611
100  453k    0  453k    0     0  56360      0 --:--:--  0:00:08 --:--:-- 76695
100  641k    0  641k    0     0  71338      0 --:--:--  0:00:09 --:--:--  105k
100  832k    0  832k    0     0  82482      0 --:--:--  0:00:10 --:--:--  132k
100 1023k    0 1023k    0     0  93519      0 --:--:--  0:00:11 --:--:--  164k
100 1252k    0 1252k    0     0   102k      0 --:--:--  0:00:12 --:--:--  191k
100 1469k    0 1469k    0     0   110k      0 --:--:--  0:00:13 --:--:--  200k
100 1660k    0 1660k    0     0   116k      0 --:--:--  0:00:14 --:--:--  202k
100 1853k    0 1853k    0     0   121k      0 --:--:--  0:00:15 --:--:--  209k
100 2055k    0 2055k    0     0   126k      0 --:--:--  0:00:16 --:--:--  205k
100 2255k    0 2255k    0     0   131k      0 --:--:--  0:00:17 --:--:--  200k
100 2466k    0 2466k    0     0   135k      0 --:--:--  0:00:18 --:--:--  203k
100 2660k    0 2660k    0     0   138k      0 --:--:--  0:00:19 --:--:--  201k
100 2854k    0 2854k    0     0   141k      0 --:--:--  0:00:20 --:--:--  199k
100 3045k    0 3045k    0     0   143k      0 --:--:--  0:00:21 --:--:--  199k
100 3238k    0 3238k    0     0   145k      0 --:--:--  0:00:22 --:--:--  196k
100 3412k    0 3412k    0     0   147k      0 --:--:--  0:00:23 --:--:--  189k
100 3594k    0 3594k    0     0   148k      0 --:--:--  0:00:24 --:--:--  186k
100 3792k    0 3792k    0     0   150k      0 --:--:--  0:00:25 --:--:--  188k
100 3994k    0 3994k    0     0   152k      0 --:--:--  0:00:26 --:--:--  188k
100 4185k    0 4185k    0     0   153k      0 --:--:--  0:00:27 --:--:--  188k
100 4380k    0 4380k    0     0   155k      0 --:--:--  0:00:28 --:--:--  193k
100 4583k    0 4583k    0     0   156k      0 --:--:--  0:00:29 --:--:--  197k
100 4792k    0 4792k    0     0   158k      0 --:--:--  0:00:30 --:--:--  200k
100 4987k    0 4987k    0     0   159k      0 --:--:--  0:00:31 --:--:--  200k
100 5033k    0 5033k    0     0   160k      0 --:--:--  0:00:31 --:--:--  201k
curl: (18) transfer closed with outstanding read data remaining
He's annoying, because Metgrid fails, i do not know if it depends on my DSL or Nomad server.

emsiwx
Posts: 75
Joined: Sun Aug 12, 2012 11:07 am

Re: Using GFS 0.25° grid dataset

Post by emsiwx » Wed Nov 05, 2014 7:03 am

Just FYI:

besides the resolution change, these are the major improvements:

*T1534 Semi-Lagrangian.
*More accurate spectral-grid transforms.
*Reduced grid consistency when creating orographic fields.
*Use of high resolution daily SST analysis.
*Use of high resolution sea ice.
*Physics:
**Cloud estimate modifications.
**EDMF parameterization.
**Dissipation heating paramterization.
**Convective gravity wave drag.
**McICA.
**Modify how GFS diagnoses snow accumulation to be consistent with post.
**Land Surface - Removal of soil moisture nudging.
*Unification of analysis ensemble with GEFS ensemble.
*Ensemble resolution increase to T574L64.
*Analysis upgrades - See global GSI implementation roadmap.
*Post Processor:
**Faster/less memory version.
**Updated version to use new crtm 2.1.3.
**Add user requested fields:
***H/T/U/V at FL410.
***Ozone at 150, 200, 250, 300, 350, and 400 mb.
***2m dew point.
***Wind chill and heat index.
***Instantaneous precipitation type.
***Membrane SLP in GDAS pgb files.
**Improved in flight icing algorithm in post.
**Consolidation of GFS master file creation with GFS DNG, similation of GOES radiances and WAFS.

oib
Posts: 117
Joined: Mon Jan 31, 2011 3:29 pm

Re: Using GFS 0.25° grid dataset

Post by oib » Wed Nov 05, 2014 10:14 am

surgeon wrote:
Gippox wrote:...unfortunately i also have to wait NOMADS grib filter service since my DSL is not sufficiently fast ... :roll: .
NOMADS grib filter offers GFS 0.25 since few days.
thank you so much , I had not noticed before! ;)

meteo60
Posts: 96
Joined: Tue Apr 17, 2012 4:50 pm

Re: Using GFS 0.25° grid dataset

Post by meteo60 » Wed Nov 05, 2014 11:51 am


oib
Posts: 117
Joined: Mon Jan 31, 2011 3:29 pm

Re: Using GFS 0.25° grid dataset

Post by oib » Thu Nov 06, 2014 10:39 am

GFS Parallel gribs 0.25° at the moment have only 27 vertical levels. This is for both the "whole earth" gribs and for the tiles from "grib filter" service from NCEP.

Instead for GFS 0.50° and also for the new GFS Parallel 0.50° NCEP provied gribs with the usual 47 levels full set, the so called "master gribs".

That's the grib inventory reported by ungrib.exe for GFS Parallel 0.25° and you can see it has only 27 levels.

Code: Select all

2014-11-06 08:29:37.327 --- INFORM: Reading Grib Edition 2
2014-11-06 08:29:37.349 --- Inventory for date = 2014-11-06 00:00:00
2014-11-06 08:29:37.350 ---      PRES        TT        UU        VV        RH       HGT ... omissis ...
2014-11-06 08:29:37.350 --- -------------------------------------------------
2014-11-06 08:29:37.351 ---    201300         O         O         O         O         O ... omissis ...
2014-11-06 08:29:37.351 ---    200100         X         X         X         X         O ... omissis ...
2014-11-06 08:29:37.351 ---    100000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.352 ---     97500         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.352 ---     95000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.352 ---     92500         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.352 ---     90000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.353 ---     85000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.353 ---     80000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.353 ---     75000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.353 ---     70000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.354 ---     65000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.354 ---     60000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.354 ---     55000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.355 ---     50000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.355 ---     45000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.355 ---     40000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.356 ---     35000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.356 ---     30000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.356 ---     25000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.357 ---     20000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.357 ---     15000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.357 ---     10000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.358 ---      7000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.358 ---      5000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.358 ---      3000         X         X         X         X         X ... omissis ...
2014-11-06 08:29:37.359 ---      2000         X         X         X         -         X ... omissis ...
2014-11-06 08:29:37.359 ---      1000         X         X         X         X         X ... omissis ...
And that's the same but from GFS and GFS Parallel 0.50° "master gribs" and you can see it has 47 levels.

Code: Select all

2014-11-06 08:38:03.951 --- INFORM: Reading Grib Edition 2
2014-11-06 08:38:03.971 --- Inventory for date = 2014-11-06 00:00:00
2014-11-06 08:38:03.971 ---      PRES        TT        UU        VV        RH       HGT ... omissis ... 
2014-11-06 08:38:03.971 --- -------------------------------------------------
2014-11-06 08:38:03.971 ---    201300         O         O         O         O         O ... omissis ...  
2014-11-06 08:38:03.971 ---    200100         X         X         X         X         O ... omissis ...  
2014-11-06 08:38:03.971 ---    100000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.972 ---     97500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.972 ---     95000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.972 ---     92500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.972 ---     90000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.972 ---     87500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.972 ---     85000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.972 ---     82500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.973 ---     80000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.973 ---     77500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.973 ---     75000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.973 ---     72500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.973 ---     70000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.973 ---     67500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.973 ---     65000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.974 ---     62500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.974 ---     60000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.974 ---     57500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.974 ---     55000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.974 ---     52500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.974 ---     50000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.974 ---     47500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.975 ---     45000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.975 ---     42500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.975 ---     40000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.975 ---     37500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.975 ---     35000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.975 ---     32500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.975 ---     30000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.976 ---     27500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.976 ---     25000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.976 ---     22500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.976 ---     20000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.976 ---     17500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.976 ---     15000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.976 ---     12500         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.977 ---     10000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.977 ---      7000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.977 ---      5000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.977 ---      3000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.977 ---      2000         X         X         X         -         X ... omissis ... 
2014-11-06 08:38:03.977 ---      1000         X         X         X         X         X ... omissis ... 
2014-11-06 08:38:03.977 ---       700         X         X         X         -         X ... omissis ... 
2014-11-06 08:38:03.977 ---       500         X         X         X         -         X ... omissis ... 
2014-11-06 08:38:03.978 ---       300         X         X         X         -         X ... omissis ... 
2014-11-06 08:38:03.978 ---       200         X         X         X         -         X ... omissis ... 
2014-11-06 08:38:03.978 ---       100         X         X         X         -         X ... omissis ...
In your opinion NCEP will provide the missing vertical layers in the incoming GFS upgrade also for the higher resolution 0.25° gribs ? :?:

If I'm not mistaken even the STRC EMS servers has always returned GFS 0.50° grib tiles with only 27 levels: the so called "grib a" GFS gribs.

I never understood the reason why STRC EMS tiling service provides not the full gribs ... frankly I do not know how this could impact on the WRF forecast.

And it would interesting to know if someone has done some test on this :)

meteoadriatic
Posts: 1512
Joined: Wed Aug 19, 2009 10:05 am

Re: Using GFS 0.25° grid dataset

Post by meteoadriatic » Thu Nov 06, 2014 3:10 pm

Yes, there are no "master-ptiles" from STRC. Maybe something to add on whishlist for Robert :) I hope that after new GFS enters into main line, there will be master's but those will be almost 500MB big /file. So ptiles of those would be really must have. But it would finally be pretty complete solution.

Somebody wrote here that they get little better results with full levels than with "a" gribs only. I also use master gribs where I have enough bandwidth and haven't switched yet to 0.25°.

Post Reply

Who is online

Users browsing this forum: No registered users and 5 guests