Skip to content

JMMP-Group/CO_AMM7

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

77 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

CO9 AMM7 (Coastal Ocean Atlantic Margin Model 7km)

The Joint Marine Modelling Programme (JMMP) provides world-class and easily accessible national capability, ocean modelling infrastructure and configurations to support the UK’s scientific research and operational prediction systems for ocean, weather and climate. It is partnership between the Met Office and British Antarctic Survey, National Oceanography Centre and Centre for Polar Observation and Modelling.

Model configurations are underpinned by the Nucleus for European Modelling of the Ocean (NEMO) framework. JMMP works closely with the NEMO consortium to develop the underpinning model capability. 

The key update in this version of the AMM7 configuration is the change to a multi-envelope vertical coordinate system (from a quasi-sigma terrain following system). Details of this vertical coordinate system change can be found in the Ocean Modelling article The effect of vertical coordinates on the accuracy of a shelf sea model and accompanying repository.


Configuration

**Configuration ** Specification
Nemo-ocean repository http://forge.ipsl.jussieu.fr/nemo/svn/NEMO
Branch releases/r4.0/r4.0.2
Components OCE
CPP keys key_mpp_mpi key_vectopt_loop key_nosignedzero key_iomput
Grid ORCA
Resolution 7 km
Horizontal Gridpoints 111,375 (y=375, x=297)
Vertical Levels 51
Vertical Coordinates Multi-Envelope - 2 envelopes of quasi-sigma terrain following
Time Step [s] 300 / 30
Boundaries 2 sets of unstructured 2d and 3d open boundaries (1 for open ocean, 1 for Baltic)

Install guidance:

To clone the git repository locally

git clone [email protected]:JMMP-Group/CO9_AMM7.git

To then download NEMO and copy files from the git repo into the appropriate directories

./CO9_AMM7/scripts/setup/amm7_setup_light -w $PWD/CO9_AMM7_instance -s $PWD/CO9_AMM7

For ARCHER2 follow the instructions here to compile XIOS and then setup the compiler configuration file for NEMO. NEMO can then be compiled with

cd CO9_AMM7_instance/nemo
./makenemo -m X86_ARCHER2-Cray -r AMM7 -j 16

Create a link to xios in the experiment (i.e. EXP00) directory, e.g.

ln -s ${PRFX}/xios/2.5/cmpich8-ucx/cce12/bin/xios_server.exe xios_server.exe

Input files can then be downloaded into the experiment directory and the configuration should then be setup. A sample runscript is included to run on ARCHER2 in the EXP00 directory.


Configuration Input Files

Input Download Location
Domain_cfg.nc https://gws-access.jasmin.ac.uk/public/jmmp/AMM7/CO9_repo/domain_cfg_co9amm7_MEsL51r10-07.nc
Open ocean boundary coordinates.bdy.nc http://gws-access.jasmin.ac.uk/public/jmmp/AMM7/grid/coordinates.bdy.nc
Baltic coordimates.bdy.nc http://gws-access.jasmin.ac.uk/public/jmmp/AMM7/grid/coordinates.skagbdy.nc

Sample Forcing Files

Forcing Download Location
Surface boundary http://gws-access.jasmin.ac.uk/public/jmmp/AMM7/inputs/SBC/
Open ocean boundary http://gws-access.jasmin.ac.uk/public/jmmp/AMM7/inputs/BDY/
Baltic boundary http://gws-access.jasmin.ac.uk/public/jmmp/AMM7/inputs/BDY_SKAG/
River runoff http://gws-access.jasmin.ac.uk/public/jmmp/AMM7/inputs/RIV/
Tide https://gws-access.jasmin.ac.uk/public/jmmp/AMM7/inputs/TIDE/
Initial condition https://gws-access.jasmin.ac.uk/public/jmmp/AMM7/inputs/IC/

NOTES

Land sea mask

For the atmospheric boundary data a land sea mask can be applied to mask the data over land on the native atmospheric grid. This masked data can be filled by the user during preproccessing (effectively not applying a land sea mask) or the lasn sea mask can be supplied and NEMO can fill the relevent grid points at runtime. The namelist option nn_lsm specifies how the atmospheric data on the model grid is extrapolated from the unmasked data around the coastlines when a land sea mask is supplied. Although nn_lsm=1 should in prinicple be ok, it has been found that inconsistencies between model runs using different core setups can then accumulate. It is therefore recommended to fill the land points during preprocessing or set nn_lsm=10, which simply mean more model grid points are filled at runtime.

About

AMM7 configuration for testing various vertical coordinate options

Resources

Stars

Watchers

Forks

Packages

No packages published

Contributors 3

  •  
  •  
  •