Chaocean

The goal of this project is to diagnose the origin of the low-frequency variability in the North Atlantic.

Initial condition, open boundaries and atmospheric forcing

All the scripts needed to build the script directory. The main script is mygendata.m. To generate new initial condition you need to set

• the output directory

dir_o = '/tank/groups/climode/chaocean/init_cond97_12/';

• Then tere are a couple of flags; if you start from scratch, it should be

build_init = 1; % 1: generate initial conditions, 0: no
build_obcs = 1; % 1: generate open boundaries files, 0: no

build_atmosphere = 1; % 1: generate atmospheric forcing files, 0: no

flag_cut = 1;  % 1: put eastern atlantic in south america, 0: no
flag_interp = 0; % 1: matlab interpolation (slower), 0: mygriddata
flag_plot = 0; % 1: plot a couple of diags, 0: no

• In the NEW GRID section, adjust the grid size. For example, to set the resolution to 1/12, do

dla = 1/12.;
dlo = 1/12;

• Choose the vertical grid: you can comment/uncomment the dz_fi line you like or create a new one.

Configuration files for MITgcm

All these files are in the MITgcm directory. The 2 sub-directories are input and code (usual MITgcm configurations files). Each time you change the resolution, you need to make sure the parameters in data are still right. When the grid changes, you also need to update data.obcs.

The tiles are defined in SIZE.h. Here is an exemple of a tiling configuration (this plot was produced with mygendata.m with nxp = 10; and nyp = 9;)

1/4 degree configuration

to run a 1/4 degree run, you need to

• change the resolution in mygendata.m
• in input dir, copy data.obcs_025 to data.obcs and data_025 to data
• in code dir, copy SIZE.h_025 to SIZE.h

This configuration is supposed to run on 25 procs

Runs

Once you started a run, fill out the table in runs/README.md