Conversation
…n cime, and add a usermod for CUPiD
ekluzek
added
enhancement
…ppens if env_postprocessing.xml hasn't been setup yet
| @@ -0,0 +1,44 @@ | |||
| # QuickStart to Running Diagnostics of your Case with CUPiD (CESM Unified Postprocessing and Diagnostics) | |||
There was a problem hiding this comment.
Can we point to the documentation for running CUPiD via the CESM workflow rather than duplicating and ending up with outdated information?
I'm thinking it would also be good for me to put the land-specific 'cheat sheet' as a docs page for component-specific example of running a STANDALONE version
.gitmodules
Outdated
| path = tools/CUPiD | ||
| #url = https://github.com/NCAR/CUPiD.git | ||
| url = https://github.com/TeaganKing/CUPiD.git | ||
| #fxtag = v0.4.1 |
There was a problem hiding this comment.
The tag will need to be updated.
There was a problem hiding this comment.
Let's go through this and figure out which of these should be default settings for ALL CTSM standalone cases—i.e., not just in this usermod. (And also—which of these are already set correctly by default without us in CTSM needing to do anything.)
There was a problem hiding this comment.
Maybe in cime_config/config_component.xml. But need to check that these would only apply in a standalone CTSM checkout.
There was a problem hiding this comment.
I've been looking at other repositories for other things, but I see this sort of thing in config_compsets.xml for several components.
<entries>
<entry id="RUN_STARTDATE">
<values>
<value compset="1850_">0001-01-01</value>
</values>
</entry>So it looks like that's the sort of thing that can be done. The entry is defined elsewhere, but you can set the value according to the compset as above. I think in our case we want it to be in config_component.xml, so I'll check if it can be added there.
If not we could add %CUPID as part of the CLM compset options. I don't think I like that, but it's probably something to discuss. We do have compsets that set a particular user-mod directory as part of their definition. And this would be doing that same thing here.
README.DIAGNOSTICS
Outdated
|
|
||
| This document has more details on more options and such | ||
|
|
||
| https://docs.google.com/document/d/1ziZWGgaj9FxgR6WRHyCAZnzIc6-4ezqfUDDpQv4MzUI/edit?tab=t.0 |
There was a problem hiding this comment.
Point to this:
README.DIAGNOSTICS
Outdated
| And you only need to do it again if the CUPiD environment changes. | ||
|
|
||
| ``` shell | ||
| # Setup the conda environments |
There was a problem hiding this comment.
Here, point to the CUPiD documentation
Merge b4b-dev to master - Change some _U and _R history fields to be on by default PR #3667 by Keith Oleson - Update to cime version that fixes the check_input_data --download issue PR #3647 by Erik Kluzek; #3647 updated to cime6.1.145, while updating b4b-dev to master in this PR gets us to cime6.1.146. - Initial fixes to generate_gdd20_baseline PR #3543 - Decomp mod unittest PR #3699
…e new land_only example, add a hist_2000-2005 user-mod _include for CUPiD
…e interprets it as a _C NINST test
wwieder
removed
the
next
…s: LDF, ROF, ILAMB
| # | ||
|
|
||
| # Set the standard CUPiD baseline case to compare to | ||
| ./xmlchange --force CUPID_BASELINE_CASE='$CASE' |
There was a problem hiding this comment.
For the base case I need to point to @wwieder base case.
/glade/campaign/cesm/development/cross-wg/diagnostic_framework/CESM_output_for_testing/ctsm5.4.002_clm6_BGCcrop_crujra_4x5_HIST/lnd/hist
make the clm6_0 physics in the ctsm5.,4.002 tag 4x5 case the base case
There was a problem hiding this comment.
Use the start and extent for the base case.
There was a problem hiding this comment.
Do the case as a I2000 starting 2000-2004
| ./xmlchange CUPID_RUN_ALL=FALSE | ||
| # Use the land_only example | ||
| ./xmlchange CUPID_EXAMPLE=land_only | ||
| # TODO: Get CUPiD working with LDF under CESM |
There was a problem hiding this comment.
LND without either LDF or ILAMB won't work except with a ne30 case
…RUN_TYPE elsewhere
… this didn't work, add setting of the base case to usefor comparision, adjust some of the settings and get it setup for the general case
…a different clone of CTSM
…n, LDF with SP, LDF with I2000, and ROF, as well as web
…test takes a half hour
4 participants
Description of changes
The changes that are needed to turn CUPiD diagnostics from a CTSM case.
Specific notes
Contributors other than yourself, if any:
@TeaganKing @samrabin
CTSM Issues Fixed (include github issue #):
Are answers expected to change (and if so in what way)? No
Any User Interface Changes (namelist or namelist defaults changes)? Allows
RUN_POSTPROCESSING==TRUE to be used in CTSM cases
Does this create a need to change or add documentation? Did you do so? Yes
I worked on the README files, but we'll need to add to the User's Guide
Testing performed, if any:
I'm running a single test case to show the workflow. But, there are issues...