Large case benchmarks: North-Atlantic 1/4° NEMO/LOBSTER configuration

The large case benchmark is based on a realistic configuration of the NEMO ocean model, for the North Atlantic Ocean, at a 1/4° resolution (Barnier et al, 2006), and including an ecosystem model component (Ourmières et al., 2009), figuring the operational MyOcean systems. This model was selected because it has been used in numerous assimilation studies before (Ourmières et al., 2009, Béal et al, 2010) and because it is based on the NEMO ocean model which is used by most MyOcean Monitoring and Forecasting Centres. This implementation will be used for selected simulations to assess new techniques with real data.

Download the NEMO model

Register to the NEMO website http://www.nemo-ocean.eu/: get login and password. After login, you can access to the full NEMO documentation: reference manuals, user's guides, description of the reference configurations,...
However, NATL025 is a NEMO configuration developed by the DRAKKAR project. It is thus easier to download the NEMO code with all required specific development together with the DRAKKAR configuration manager (DCM).
More information about DCM (login and password required).

Download NEMO and associated tools from the DCM distribution (with the same login and password):

svn https://servforge.legi.grenoble-inp.fr/svn/DCM/DCM/trunk NEMODRAK_3.4

svn https://servforge.legi.grenoble-inp.fr/svn/DCM/BUILDNC/trunk/TIME TIME

svn https://servforge.legi.grenoble-inp.fr/svn/DCM/BUILDNC/trunk/DIMGPROC DIMGPROC

svn https://servforge.legi.grenoble-inp.fr/svn/DCM/BUILDNC/trunk/Macrolib Macrolib

Please note that DCM requires to set up several environment variables (check the documentation).

Prepare the NATL025 configuration

Create the directory structure for the new configuration/case:
```
mkconfdir NATL025 "your_case_name"
```
A new case should be created each time you want to perform a new simulation with different settings (modified code, parameters, input data). A standard naming convention for the case is proposed in the DCM documentation.

Edit the configuration "makefile":

$UDIR/CONFIG_NATL025/NATL025-"your_case_name"/makefile

Modify especially the following variables:

    CASEREF = 'none'
    PREV_CONFIG = '${HOMEDCM}/CONFIGS/NATL025-GREF3.4'
    MACHINE = PW6_VARGAS
    NCOMPIL_PROC = 64

Copy all relevant files from the reference configuration:
```
make copyconfig
```

Mofify the precompilation options separately (in the file CPP.keys), for instance:

P_P = key_natl025 key_dynspg_flt key_zdftke key_traldf_c2d key_dynldf_c2d \
      key_ldfslp key_dimgout key_lim2 key_lim2_vp key_mpp_mpi

Compile the NATL025 configuration

Prepare the code before compilation:
```
make install
```
Compile NEMO (current configuration and case):
```
make
```
Compile the peripheric tools in the directories TIME and DIMGPROC (the second one is to combine the binary output files produced by each process into a single NetCDF file), e.g.:
```
    cd DIMGPROC
    ln -s ../Macrolib/macro.vargas make.macro
    make ; make install
    
```

Run the NATL025 configuration

Download the DRAKKAR scripts to run the model:
```
svn co https://servforge.legi.grenoble-inp.fr/svn/DCM/RUNTOOLS/trunk RUN_TOOLS
```
These must be considered as example scripts, which need to be adjusted to your computer.
Prepare the control script for the current configuration/case
```
    cd RUN_TOOLS/TEMPLATE
    mkctl NATL025-"your_case_name"
    
```
The following assumes that you install the scripts for the machine "vargas" (IBM SP Power6).
Go to the directory with the control script for the current configuration/case
```
    cd $HOME/RUN_NATL025/NATL025-"your_case_name"/CTL
    
```
Prepare the model parameter files:
- namelist: template in NEMODRAK_3.4/CONFIGS/NATL025-GREF3.4
- namelist_ice: template in NEMODRAK_3.4/CONFIGS/NATL025-GREF3.4
- namelistio: template in DIMGPROC
Edit the file "includefile.ksh", and adjust the parameters and directories.
In particular, set the number of jobs to be resubmitted (MAXSUB).
Check class limits in the script:
```
NATL025-"your_case_name"_vargas.ksh
```
Set the first and last time step of the first job in:
```
NATL025-"your_case_name".db
```
All subsequent jobs are automatically resubmitted with the same number of timesteps.
Prepare the input data files in the data directory ($SDIR/NATL025/NATL025-I).
Run the model:
```
./run_nemo_vargas.ksh
```
The results of the simulations are produced in the directory:
```
$SDIR/NATL025/NATL025-"your_case_name"-S
```

References

Barnier B., Madec G., Penduff T., Molines J.M., Tréguier A.M., Beckmann A., Biastoch A., Boning C., Dengg J., Gulev S., Le Sommer J., Rémy E., Talandier C., Theetten S., Maltrud M., Mc Lean J. Impact of partial steps and momentum advection schemes in a global ocean circulation model at eddy permitting resolution. Ocean Dynamics, 56(5-6), 543-567, 2006.
Béal D., Brasseur P., Brankart J.-M., Ourmières Y. and Verron J. Characterization of mixing errors in a coupled physical biogeochemical model of the North Atlantic: implications for nonlinear estimation using Gaussian anamorphosis. Ocean Science, 6, 247-262, 2010.
Ourmières Y., Brasseur P., Levy M., Brankart J.M. and Verron J. On the key role of nutrient data to constrain a coupled physical-biogeochemical assimilative model of the North Atlantic Ocean. Journal of Marine Systems, 75(1-2), 100-115, 2009.