Extract Meteo-France data

Meso-NH can be initialized and forced at its lateral boundary conditions by Météo-France operationnal data (ARPEGE, ALADIN and/or AROME). Hereafter you will find the procedure to extract these data.

Warning

AROME forecasts are only available for a moving 3-year period.

Note

For Meteo-France data, the namelist NAM_PREP_SURF_ATM must be added to PRE_REAL1.nam (namelist of PREP_REAL_CASE program). This namelist will contain the files to be used by SURFEX: the surface analysis and the associated PGD file.

For non-Belenos user

For security reasons, it is not possible to extract Météo-France operational data from outside Météo-France (belenos). If you don’t have access to belenos, data extraction from Météo-France files can be performed upon request, within reasonable limits. For that please send an e-mail to Meso-NH’s support containing :

  • the desired model and type of data :

    • AROME (operational since December 2008) : analysis, forecast, first guess

    • ARPEGE : analysis, forecast, first guess, initialized analysis, short-cycle forecast

    • ALADIN (before 26/03/2012) : analysis, analysis with digital filters, forecast, first guess, initialized analysis

    • AROME or ALADIN-REUNION : analysis, analysis with digital filters, forecast, first guess, initialized analysis

    • MOCAGE : please indicate the directories where SM and HM data are stored

  • dates (start and end dates, step, …)

For Belenos user

If you have access to belenos, you can extract data by yourself. For that you need to use extractMF procedure.

Warning

  • Its use is limited to ARPEGE and AROME-France.

  • For Aladin-Reunion and Forecast-Olive, continue with extarome at CNRM.

  • For AROME-OM, the procedure is under development, extractions are available on request.

  • For Mocage and Arpege-Climat, there is currently no way of extracting these data automatically.

To extract ARPEGE or AROME-France data by yourself you need to copy the $MESONH/procedures/extractMFrc in the directory from which you will launch extractMF. You need to insert in this file the desired extraction parameters, they are documented in the extractMFrc file and the main parameters are presented hereafter :

  • The MODEL variable can take the values arome or arpifs,

    MODEL=arome

    or

    MODEL=arpifs

    In the case of AROME extraction, surface files are automatically renamed with the date (see extractMF return for exact names).

  • The TYPE variable can take the values AN for analyses (every 1h for arome) or FC for forecasts.

    TYPE = AN

    or

    TYPE = FC

  • AROME can zoom in to extract a sub-domain of the AROME-FRANCE domain with the LZOOM=T flag. You then need to enter the edges of the domain, for example :

    LZOOM=T
LATMIN=45.5
LATMAX=48.5
LONMIN=3.0
LONMAX=7.9

    The use of zoom is highly recommended in order to reduce memory requirements at the stage of interpolating AROME fields onto the Meso_NH grid at the PREP_REAL_CASE step.

  • OUTDIR corresponds to the directory where the data will be extracted (in your $WORKDIR if you keep HOST=tosupc).

Then you can launch the extraction with :

. ~rodierq/SAVE/mesonh/profile_mnhp
extractMF

A slurm job is launch and you can follow the extraction with :

squeue --me

Tip

  • In the extractMF return file, you can find at the end a message indicating the namelist NAM_PREP_SURF_ATM to be transferred to your PRE REAL1.nam file.

  • Your data will be stored in the OUTDIR directory.

Note

Historically, at the start of AROME, we only had surface data from the AROME coupler model (ALADIN then ARPEGE), which are still available in the AROME analysis file. However, with the introduction of SURFEX in AROME, and in particular of a surface analysis performed by AROME, it became essential to be able to use data from AROME’s surface analysis rather than that of its coupler model.

  • Arome operational, ExtractMF get following files :

    • analysis : ˜mxpt001/arome/oper/production/YYYY/MM/DD/RR/analyse

    • guess : ˜mxpt001/arome/oper/production/YYYY/MM/DD/RR/guess

    • forecast : ˜mxpt001/arome/oper/production/YYYY/MM/DD/RR/ICMSHAROM+000x

    • surface analysis :

      • before cycle 37 : ˜mxpt001/arome/oper/production/YYYY/MM/DD/RR/INIT_SURF.lfi, this file is automatically rename in INIT_SURF.YYYYMMDD.RR.lfi

      • after cycle 38 : ˜mxpt001/arome/oper/production/YYYY/MM/DD/RR/analyse.sfx, this file is automatically transform in lfi and called INIT_SURF.YYYYMMDD.RR.lfi

  • Files from OLIVE experiment:

    • analysis : xp/XXXX/YYYYMMDDHHHP/pseudotraj/analyse

    • forecast : xp/XXXX/YYYYMMDDHHHP/forecast/ICMSHAROM+000x

    • surface analysis : xp/XXXX/YYYYMMDDHHHP/surfan/ (name depend of the experience and AROME’s cycle)

    • PGD_AROME : if AROME’s cycle higher than 37 you have to get PGD file in LFI format from OLIVE experiment

Use of extracted GRIB files

Depending on the simulated date, the NAM_PREP_SURF_ATM changes which are described in the following table :

NAM_PREP_SURF_ATM options

Start date

Arome cycle

SURFEX version

Namelist NAM_PREP_SURF_ATM

before juillet 2013

cy37 and after

6

CFILE=’INIT_SURF.20120924.12’

CFILETYPE = ‘MESONH’

CFILEPGD = ‘INIT_SURF.20120924.12’

CFILEPGDTYPE = ‘MESONH’

02/07/2013 00h

cy38 / cy39

7.2

CFILE=’INIT_SURF.20130702.12’

CFILETYPE=’MESONH’

CFILEPGD=’pgd_frangp.02km50.02’

CFILEPGDTYPE=’MESONH’

13/04/2015 06h

cy40

7.2

CFILE=’INIT_SURF.20150624.12’

CFILETYPE=’MESONH’

CFILEPGD=’pgd_franmg.01km30.01’

CFILEPGDTYPE=’MESONH’

08/12/2015 03h

cy41

7.3

CFILE=’INIT_SURF.20160124.12’

CFILETYPE=’MESONH’

CFILEPGD=’PGD_oper_41t1.01km30’

CFILEPGDTYPE=’MESONH’

02/07/2019 03h

cy43

8.0

CFILE=’INIT_SURF.20190702.12’

CFILETYPE=’MESONH’

CFILEPGD=’PGD_oper_43t2.01km30.05’

CFILEPGDTYPE=’MESONH’

22/06/2022 06h

cy46

8.1

CFILE=’INIT_SURF.20220622.06’

CFILETYPE=’MESONH’

CFILEPGD=’PGD_oper_46t1.01km30.05’

CFILEPGDTYPE=’MESONH’

15/10/2024 00h

cy48

8.1

CFILE=’INIT_SURF.20241015.00’

CFILETYPE=’MESONH’

CFILEPGD=’PGD_oper_48t1.01km30’

CFILEPGDTYPE=’MESONH’

Note

  • Since version 7.2 of surfex, surface fields are separated into two files: the PGD part and the PREP part. Up to AROME cycle 37, everything is stored in a single file (CFILE=CFILEPGD).

  • PGD files are available under $MESONH/PGD (don’t forget to create the empty .des file). By default, the operational PGD associated with the date of your extraction is not available. If you wish to retrieve it, you must add the line LGET_PGD_AROME=T in your extaromerc.

  • Starting with AROME cycle 41 (December 2015), AROME analyses are available every hour. However, surface analyses are only available on the main cycles (03 06 09 12 15 18 21). Therefore, if a loop is requested with an hourly frequency for analyses, no surface analysis will be extracted by extarome. In this case, you need to request the first file (from a main cycle) separately to obtain the AROME surface analysis.

  • From MESO-NH version 5-4-0 onwards, add &NAM_CONFIO NIO_ABORT_LEVEL=1 NGEN_ABORT_LEVEL=1. Since this version we can have Meso-NH PGD files in NetCDF and AROME surface files (+AROME PGD) in lfi.

Example of PRE REAL1.nam namelist with AROME surface analysis :

&NAM_CONFIO
  LCDF4            = T,
  NIO_ABORT_LEVEL  = 1,
  NGEN_ABORT_LEVEL = 1
/
&NAM_FILE_NAMES
  HATMFILE     = 'arome.FC.2026013000.12',
  HATMFILETYPE = 'GRIBEX',
  HPGDFILE     = 'PGD_AZF',
  CINIFILE     = 'AROME_COUPLING'
/
&NAM_VER_GRID
  NKMAX        = 50,
  YZGRID_TYPE  = 'FUNCTN',
  ZDZGRD       = 60.,
  ZDZTOP       = 700.,
  ZZMAX_STRGRD = 2500.,
  ZSTRGRD      = 9.,
  ZSTRTOP      = 7.
/
&NAM_PREP_SURF_ATM
  CFILE        = 'INIT_SURF.20260130.12',
  CFILETYPE    = 'MESONH',
  CFILEPGD     = 'PGD_oper_48t1.01km30',
  CFILEPGDTYPE = 'MESONH'
/