GEM model output

All model "data" output is written in RPN standard files.

There are different types of data the model can output:

Click on the following link for a description of the most common variables:  variable dictionaries

All  files above are stored in 'machine:directory' as specified by 'CLIMAT_arch_mach' and 'CLIMAT_archdir' in the file 'configexp.cfg'.
The files are organized into the following subdirectories:

And then you should also have this file:
     maingemdm_Linux_x86-64.Abs 
or
     maingemdm
which is a copy of the executable.

2-D and 3-D output - "normal" output used for evaluation/analysis

There are four types of output files:

Samples (direct model output)

All fields requested in 'outcfg.out' are written into the four types of output files for the requested frequencies.

Diagnostics

If 'CLIMAT_diagnos=1' in 'configexp.cfg' the Diagnosticsmodel output will get further processed.
The diagnostics contain:

• monthly means and variances calculated for all model output fields
• 2-D time series for selected fields (if requested and available)

Fixed fields

Fixed fields can generally be found under the 'Samples' directory inside the directory ending on '*_step0'.
ME will be in the dm-file, all other fixed fields are usually in the pm-file.

Time series (station data)

In addition to the general 2-D and 3-D output, GEM/GEMCLIM can output time series at selected grid points (stations).

Requesting time series

A few parameters need to be added to the config-file 'gemclim_settings.nml' to output time series. These parameters are are part of the namelist 'gem_cfgs':

Note: Whenever asking for accumulators in time series output, one needs to also ask for them as 2-D fields in the regular output in the file 'outcfg.out' with the same output frequency. Otherwise they will never get reset to zero but accumulate forever!

Looking at time series

Although the time series are written in RPN standard format you will not be able to look at them with 'xrec' or to use 'r.diag' or 'editfst'. But you can have a look at the content using 'voir'.

There is also a program called 'ts2tab' which converts the RPN standard file time series into ASCII files.

Zonal means

Similar to requesting time series one can also request zonal means.

Analysis files

Analysis files are used to initialize the model, which is why we also call them "initial conditions".

Environment and Climate Change Canada (and other institutes) uses comprehensive assimilation processes, combining observation data and previous model output to create the analyses files used to initial their weather forecast simulations.

At UQAM we sometimes use our own model output from previous simulations to initialize other simulations. During a model simulation one can automatically output files which can later get used to initialize another simulation by setting the parameter 'CLIMAT_out_anal' in 'configexp.cfg' to the frequency at which one wants to output these initial condition/analysis files. For example for creating monthly initial condition/analysis files set:

     CLIMAT_out_anal="1m"

These files will get archive in the directory named "Analysis". They contain all fields needed to initialize a simulation, using the same surface and microphysics schemes.

Pilot files

Pilot files contain all data needed to to drive a LAM (Limited Area Model) simulation at the lateral boundaries.

Usually we use re-analysis data from other institutes (like ERA5) or analysis data from ECCC or data created by our own simulations. During a model simulation one can automatically output files which can later get used to drive another simulation at the lateral boundaries by setting the parameter 'CLIMAT_out_pilot' in 'configexp.cfg' to the frequency at which one wants to output these driving data. For example for creating 1-hourly pilot files set:

     CLIMAT_out_pilot="1h"

These files will get archive in the directory named "Pilots". They contain all fields needed to drive a simulation at the lateral boundaries.

Listings

While the model is running the listings (log files) will appear under the directory:
            ~/listings/${TRUE_HOST}
There names will all start with ${GEM_exp}_* (as defined in your 'configexp.cfg').

Once a job/month finished properly they will all get archived under:
            ${CLIMAT_archdir}/Listings/listings_${GEM_exp}.zip
'CLIMAT_archdir' is also defined in your 'configexp.cfg'.

Actually, there are up to four zip files under ${CLIMAT_archdir}/Listings/listings_${GEM_exp}.zip:

Restart files

At the end of each job, a restart file gets written. This contains the physics and dynamics files needed to restart the model. There will be files for each tile. These files contain a copy of the atmosphere and soil fields at the last time step of the job. When the next month gets started it reads the restart files of the previous month and continues from there. After the new month has started, the restart files from the previous month get archived on CLIMAT_arch_mach in CLIMAT_archdir as specified in your 'configexp.cfg'. It is always possible to restart the model at any given month from a restart file.

Since the restart files contain a full copy of the model atmosphere and soil they are quite large. So usually we do not keep all of them but erase some by hand afterwards. The name of a restart file looks like this:

    ${GEM_exp}_step###.tarz

With '###' being the number of the last timestep that was run.