Jan 23, 2014: this directory contains files for the tidal mixing scheme described in Schmittner and Egbert (2014, Geoscientific Model Development)

The energy flux out of the barotropic tide as estimated by Egbert using satellite altimetry is contained in the ascii file global_dissip_mWperm2_lat_lon_m2_s2_k1_o1.ascii

You need ferret to use these scripts. It is available for free. Just google ferret noaa and follow the download and installation instructions.

To create a file of subgridscale dissipation follow these steps:

start ferret

run make_diss (type "go make_diss") 
this converts the ascii file into the netcdf file diss.nc

diss.nc contains dissipation on the same resolution as the original (1/6 deg)

quit ferret

start ferret

run make_diss20 averages onto a 1/3 deg grid

quit ferret

start ferret

run make_sg_m2 (type "go make_sg_m2")
run make_sg_s2
run make_sg_k1
run make_sg_o1

quit ferret
start ferret

run make_tidal_egbert_3D

-------------
Previous scripts are in the subdirectory "old". These have been used to actually calculate the files used in the paper. However, those scripts are more convoluted and the new scripts produce almost identical output. If you really want to use them here are instructions:

make_tidal_hr is used by make_*_hr, which produces the files tidal_*_hr.nc.
These contain the subgrid-scale dissipation on the climate model grid.
make_all_3D collects all of the tidal_*_hr.nc files in one file: tidal_egbert_3D.nc, which I use as input for the UVic model.

make_tidal_hr uses information about the climate model grid, which is contained in grid.nc and kmt.nc. You may want to exchange these files and variable names in make_tidal_hr with the appropriate files and names from your model.

Note that make_tidal_hr also uses a 1/3 deg bathymetry that comes with ferret (etopo20).
------------

vmixc.F and vmixc_2D.F contain the source code of the climate model using the subgrid-scale scheme and Harper's original (2D) scheme. You'll see that the main difference is an additional loop over the vertical. There are also a few other minor differences between the two files, like limiting N2 to > 10^-8, how diffusivities in the Southern Ocean are limited to > 1, and the calculation of hab. IMO the original code is incorrect in using zw(kmt(i,jrow) - 1) since this can lead to zero or negative values for hab.

Andreas Schmittner (aschmitt@coas.oregonstate.edu)


