Add CICE-QC testing script #5969
Conversation
|
This test uses two netcdf files, which I've put in the public inputdata directory here: https://web.lcrc.anl.gov/public/e3sm/inputdata/ice/mpas-seaice/general/ The path to these files is currently hard-coded in the python script to the LCRC path, so this will only work on those machines as is. |
|
@eclare108213 @proteanplanet do either of you have a pair of runs sitting on LCRC I can use to test this? The runs I had in the sample job scripts here have been purged. |
|
Yes, the latest qc tests that I ran are in /lcrc/group/e3sm/ac.eclare/E3SM-Polar: D12.qcPR25.emc.qcPR25.thermo_itd2.eclare108213.anvil Those passed -- I don't have any that failed :) |
|
Back-to-back B-case simulations of Icepack and ColPkg: /lcrc/group/e3sm/ac.afroberts/E3SM-Polar/B12.merge.icepack.icepack-integration.eclare108213.anvil k000 --> config_column_physics_type = 'column_package' I'd love to know what the QC test shows. |
|
The QC scripts here need to be modified for the k000 / k001 directory structure produced using Andrew's scripts. E.g. my modified QC scripts are in /home/ac.eclare/MPAS-Analysis. |
|
and I should clarify that I'm not suggesting the scripts as submitted in this PR need to be modified, but that a user would need to modify them for, for example, Andrew's data above. |
|
@eclare108213 @darincomeau N.B. I'm changing that directory structure, because it doesn't play well across a range of E3SM applications (PACE, zppy), so no need to rewire this work. I'll rewire my script. |
@proteanplanet this test requires non-standard output, which is daily iceThickness snapshots. Unfortunately I can't run it on the pair of runs you've pointed to. @eclare108213 you've been turning this output on for your tests - do you have an automated way of modifying the |
|
and thanks @eclare108213 for pointing to your changes. I should loosen the requirement here that the run directory doesn't need to be named exactly |
There was a problem hiding this comment.
In looking through the code, I think this testing would be most versatile if test runs could be automatically generated with necessary output automatically turned on. Adding that functionality here is quite a bit of work and it's also unneeded because of the E3SM-Polar-Developer.sh script. As a result, I think the best approach would be to leave this script as is, and call it from that test script with a command line option that sets up the simulations, then performs the T-test when they are finished.
Approved.
There was a problem hiding this comment.
I have been using these scripts on anvil as part of the icepack merge testing process, slightly modified to work with the directory structure produced by the current E3SM_Polar_Developer.sh script. This quality control testing functionality is extremely useful. Future improvements could include more automation, but the current implementation works.
Add CICE-QC testing script Adds a script to perform "CICE-QC" test, which determines if non-BFB changes in the sea ice model are climate-changing. This has been adapted from the CICE implementation for MPAS-Seaice. Documentation on usage is provided in a README.md alongside test script. [BFB]
|
passes sanity testing, merged to next |
|
merged to master |
Adds a script to perform "CICE-QC" test, which determines if non-BFB changes in the sea ice model are climate-changing. This has been adapted from the CICE implementation for MPAS-Seaice.
Documentation on usage is provided in a
README.mdalongside test script.[BFB]