Welcome, Guest
Username: Password: Remember me
General topics

TOPIC: User Manual for beginners

User Manual for beginners 3 years 7 months ago #1761


I read the documentation available online and I found quite a big help on how to run an Harmonie experiment at ECMWF. However, I would like to know if there is a doc (for absolute beginners like me) explaining how are the output files organized and a short description of the parameters included in each file?


User Manual for beginners 3 years 7 months ago #1764

Dear Isabel,

Sadly it is not available yet.
During last System Working Week in Copenhagen this was one of the topics. We plan a pdf, online tutorial or video tutorial for beginners.
Meanwhile, feel free to ask any of us about your doubts.



User Manual for beginners 3 years 7 months ago #1767

Dear Daniel,
I wanted to conduct a number of observing system experiments of a storm event between 06-02-2014 and 11-02-2014 and assess the 24 h forecasts with 3h assimilation cycles. I manage to run the experiment considering conventional observations. Output files were stored at cca in $SCRATCH. Now the doubts:
1)It seems that only the complete files for 10-02-2014 to 11-02-2014 were kept.Files are stored in /scratch/ms/pt/ptq/hm_home/IberiaScatt_EXP0/archive and divided into 2014 and extract dirs.
The 2014 dir, seems to have all the files for 10-02 and 11-02 but not for the previous days. extract dir seems to have only information about the observations used throughout the run. Is that so?
2)in 2014 dir when I do grib_ls to /02/10/09/fc2014021009+000grib_sfx (surface fields?) for example I find 10u and 10v but I don't find the mslp field.
3) What is the logic behind the file names (only the grib files):
anyyyymmdd09+000grib = analysis?
bayyyymmdd09+000grib ?
fcyyyymmdd09+000grib = forecast?
fcyyyymmdd09+000grib_fp = forecast?
fcyyyymmdd09+000grib_sfx =forecast surface parameters?
fcyyyymmdd09+00Hgrib_fp = forecast?
sayyyymmdd09+000grib = ?

Sorry for such a long post :(


User Manual for beginners 2 years 3 weeks ago #2110

Again a basic question, I'm affraid:
Where can I find land sea mask ? in fcyyymmdd* files I cannot file "lsm" shortname...

User Manual for beginners 2 years 3 weeks ago #2111

Hi Isabel,
In the climate version (cy38), it is in Const.Clim.sfx in the climate folder (SFX.COVER001).

User Manual for beginners 2 years 3 weeks ago #2112

Hi Danijel,
It should be similar in cy40, thanks.
PS - I just found out that there is "lsm" field in anyyyymmdd*grib (cy40)
Last Edit: 2 years 3 weeks ago by Isabel Monteiro.

User Manual for beginners - Spectral Analysis 2 years 1 week ago #2115

Hi all,
As anyone ever use ALTO (compiled by default in harmonie) to perform kinetic energy spectral analysis?
There is some documentation online about ecto:
But if anyone has some more information on "how to use it" I would appreciate.

User Manual for beginners 2 years 1 week ago #2116


I will ask Alvaro Subia about this tools, I don't know when he starts his holidays today or tomorrow.

Other candidate to know about that could be Colm Clancy.

I will send an email to both of them.


User Manual for beginners 2 years 1 week ago #2117

  • Eoin Whelan
  • Eoin Whelan's Avatar
  • Gold Boarder
  • Posts: 209
  • Thank you received: 41
Daniel, Isabel,

a previous thread may provide some useful information: http://www.hirlam.org/index.php/forum/19-general-topics/1206-spectral-grid#1486

Alternatively, if you have access to EPyGrAM you should be able to plot directly from a model state file.


User Manual for beginners 2 years 1 week ago #2118

Thanks Daniel

User Manual for beginners 2 years 1 week ago #2119

Thank you Eoin, I think it is opensource.umr-cnrm.fr/projects/epygram
that you are referring to. I'm trying to get access to this, it seems quite an interesting tool (for other stuff as well)

Regarding the older posts that you are referring to: I understand where is ALTO and I can obtain the binary from my experiments, but I don't understand what is the meaning of
"These are the steps:
1) Copy the binary "ALTO" to "ECTO" " in Lisa Bengtsson post

Why can't we just run ./ALTO namlist infile${EXP}.fa, after constructing the namelist and doing the link ?

Last Edit: 2 years 1 week ago by Isabel Monteiro.

User Manual for beginners 2 years 1 week ago #2120

Hi all
I just found out that there is an EPyGrAM at ECMWF:
/perm/ms/spsehlam/hlam/EPyGrAM/ with tar files that allow one to install epygram (EPyGrAM-1.2.1-Source.tar and vortex-1.0.0.tar). I downloaded the files and I've tried to install epygram at my $HOME.
I'm stuck in the arpifs4py library/package.(installation instructions in opensource.umr-cnrm.fr/projects/epygram/wiki/Wiki)
arpifs4py package contains a set of Fortran interface routines, wrapping the calls to existing ARPIFS routines (from ifsaux, trans and etrans projects); these routines are gathered in a .so compiled shared object library
As far as I understood I will need to
1) To create a pack with gmkpack (version 6.6.2 or newer), based (at least) on the latest export version (CY40T1_bf.05), and with target program libs4py
Not clear to me how to do it.
Did anyone tried doing this?

User Manual for beginners -from first guess 2 years 6 days ago #2121

Hi all,
Another basic question:
I would like to continue an experiment, at ECMWF, using the first guess from the previous cycle. It is an old experiment that I would like to run with longer (than the original one) forecast length for some days.
Is it enough to change the progress.log to the day from which I want to run and do:
../Harmonie prod LL=new_forecast_length ?

-->>>Of course I'm assuming that first guess files are on the right places

Last Edit: 1 year 11 months ago by Isabel Monteiro.

User Manual for beginners -from first guess 1 year 11 months ago #2126

Hi All,

I think forums are quite useful particularly to someone who is starting and with no colleague next door working with Harmonie. So I'll answer to myself hoping that this answer can be useful to someone somewhere :).
1) Because it was an old experiment it had all the files stored at ECFS. I've copied the first guess files from
ec:/$user_id/harmonie/$exper/$year/$month/$day/$an/"ICMSHHARM+0003" and "ICMSHHARM+0003.sfx"
to cca

2) I modified progress.log and progressPP.log to the date and time from which I wanted to continue my experiment

3) I continued the experiment
../Harmonie prod DTGEND=new_yearmonthdayanalysis LL=24

Time to create page: 0.084 seconds