[MITgcm-support] grid-independent "broken lines" code

[Rose, Brian]

Tangentially related to this:
If anyone is looking for a Python equivalent of the Matlab code calcZonalAvgCube.m, I put this up on github a couple of years ago:
https://github.com/brian-rose/MITgcmutils/blob/master/MITgcmutils/cs/zonal.py
extending the functionality of the existing MITgcmutils python package for CS grids.

If folks wants to check this into the main repository they are welcome. I probably won't be developing it further at this point.

- Brian

__________________________________

  Brian E. J. Rose
     Assistant Professor
     Atmospheric & Environmental Sciences
     University at Albany

     Earth Sciences 315
     (518) 442-4477
     http://www.atmos.albany.edu/facstaff/brose/

On Mar 17, 2017, at 2:30 PM, gael forget <gforget at mit.edu<mailto:gforget at mit.edu>> wrote:

Hi Alexandre,

I am answering this question via mitgcm-support since the answer may be of interest to others.

Within the gcmfaces framework (https://github.com/gaelforget/gcmfaces) you will find two routines called gcmfaces_lines_zonal.m and gcmfaces_lines_transp.m which are grid-independent alternatives to the "broken lines" code you have been using with CS32. These routines should work for all grids in ftp://mit.ecco-group.org/ecco_for_las/version_4/grids/grids_input/ (cube spheres, LLCs, etc.) and ftp://mit.ecco-group.org/ecco_for_las/version_4/grids/grids_output/ (ECCOv2, ECCOv3, ECCOv4, OCCA, ECCO2, and SOSE) although I reckon that I have not tested every combination.

The example_transport.m routine listed in the gcmfaces documentation (https://github.com/gaelforget/gcmfaces/blob/master/gcmfaces.pdf) illustrates how these functions can be used for zonal average, meridional transport, etc. computations by taking ECCO v4 output on the LLC90 grid as an example. For the CS48 grid you may not have any model output to try transport computations yet. However, after having updated your gcmfaces clone / copy to present time, you can readily try a zonal average test as follows:
grid_load_native('cs_nostretch/cube_48/',6,0);%load grid
gcmfaces_lines_zonal;%compute broken lines
gcmfaces_global;%add mygrid to current workspace
zm_YC=calc_zonmean_T(mygrid.YC);%use broken lines
figure; plot(zm_YC);%plot zonal mean latitude

Cheers,
Gael


I was wondering: is there any repository where I could find the broken lines (i.e., the 'isoLat' files) associated with the ECCO grids ? I tried to generate them using 'gener_bk_line.m’ but the output files do not match the structure required by the scripts used for the diagnostics (e.g., calcEulerPsiCube.m). For example, the fields 'blk_IJuv' and 'bkl_Zon' are missing.

I found a thread on the MITgcm mailing list referring to the following directory: 'http://wwwcvs.mitgcm.org/viewvc//MITgcm/MITgcm_contrib/dfer/matlab_stuff/'.<http://wwwcvs.mitgcm.org/viewvc//MITgcm/MITgcm_contrib/dfer/matlab_stuff/'./> Unfortunately there is only the cs24 isoLat there.

Best,
Alexandre

__

Alexandre POHL
Postdoctoral Researcher
CEREGE, TECHNOPOLE ENVIRONNEMENT ARBOIS-MEDITERRANEE
BP80, 13545 AIX en PROVENCE, CEDEX 04, FRANCE
https://alexandrepohl.org<https://alexandrepohl.org/>

_______________________________________________
MITgcm-support mailing list
MITgcm-support at mitgcm.org<mailto:MITgcm-support at mitgcm.org>
http://mitgcm.org/mailman/listinfo/mitgcm-support

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20170317/5a082236/attachment-0001.htm>