Test climo and ts for land variables
#533
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The monthly land As for the globally averaged |
|
As for |
forsyth2
changed the title
climo and ts for land variables
|
Using a smaller subset of vars, from https://acme-climate.atlassian.net/wiki/spaces/EIDMG/pages/3476979729/zppy+New+Feature+Requests: It looks like it's looking for the wrong years for some reason? |
|
@chengzhuzhang Above are my initial findings on computing |
|
The cfg is |
|
Hi Ryan, thank you for starting on this, let's try a set of more standard variables: I currently don't have a good understanding of the Land output especially those with BGC options. We can get input from land groups later. At this point lets use above basic variables for testing purpose. Based on Charlie's note (copied below), there is a way for ncclimo can handle fault-tolerant variable lists via regular expressions: The option -v FSNT,AODVIS requires that both variables be present. You can indicate variables that are not required to be present by using a regular expression that matches the variable name exactly. For example, -v ^FSNT$,AODVIS allows FSNT to be absent and requires AODVIS to be present. In general, prefix each optional variable with ^ and postfix it with $. Nothing else changes. The climo will complete successfully with only the variables that are present that match the regular expressions. You can experiment with ncks to print the variable names that match the specified var_lst: |
|
@chengzhuzhang I tried to see what variables are supported -- and it looks like it is very few of them.
I'm wondering if some of the unsupported variables are because of the specific simulation I was using as input. Is it possible land variables weren't saved for that simulation (which was part of a Water Cycle campaign)? |
|
Hi Ryan, thanks for sharing the testing results. Definitely more variables are listed than what is provided. Not sure where the original list came from, but I think the list includes some BGC specific varaibles where the standard v2.historical run doesn't have. Though one thing I notice is that, the list of variables are not consistence cross all land related operations: climo, ts_monthly_glb, ts_monthly. I assume 3 of them should support the same set of available variables from the native output. What's more puzzling is that none of the variables are supported by |
Before your suggested shorter variable list, I was using variables listed at https://acme-climate.atlassian.net/wiki/spaces/EIDMG/pages/3476979729/zppy+New+Feature+Requests (linked at #385) and https://bitbucket.org/minxu/alm2ilamb_wkflow/src/8b8fb98911911d930cc7f598ea64b83fdb199d4c/elm_singlevar_ts.bash#lines-304 (linked at #507 (comment)).
Yes, that was quite puzzling that they should differ so much.
Agreed. |
|
Okay, I think I found why none of the variables are supported by climo--land_monthly_180x360_aave. In /lcrc/group/e3sm/ac.forsyth2/zppy_test_complete_run_output/issue_531/v2.LR.historical_0201/post/scripts/climo_land_monthly_180x360_aave_1850-1851.o434621 the ncclimo command used is: Obviously, it is looking for land variabls in atmosphere output. |
|
Re: ts -- land_monthly_glb, I don't see why variables needs to be deleted. Based on the log, It looks like all variables are generated with the short variable lists being tested? /lcrc/group/e3sm/ac.forsyth2/zppy_test_complete_run_output/issue_531/v2.LR.historical_0201/post/scripts/ts_land_monthly_glb_1855-1859-0005.o434631 |
Oh wow, good catch. While the parameter inheritance can be useful, I'll admit it can make errors like this less obvious.
Hmm I was getting errors on the variables I deleted -- maybe they still ended up generating output though. |
|
Ok setting up the land I get: It looks like the |
|
Note for self: do |
|
@chengzhuzhang do you have any input on the following errors? This is my cfg: cfg(1) From (2) As mentioned in #533 (comment), it looks like (3) It looks like |
|
The |
|
Hmm, that's going to require a code change. That's only implemented for I'll see if that fixes it though. |
|
@forsyth2 For testing purpose, if -v is not list, climatology will be generated for all variables including "landfrac".This can give an idea if missing |
|
for |
Actually this problem is related to |
Yes, omitting the variable list does in fact make |
I'm wondering if there is a zppy configuration that can omit the variable list? |
I think this is possible by setting In If |
This is good to know, either using |
|
Latest run: That matches the issue in #534 (comment). I'll rerun with Unified 1.9.1. |
There appears to be a different
@chengzhuzhang So, do we have any idea what's causing the Regarding the other errors mentioned in the comment linked above:
|
|
For trouble shooting, could you exam the input files for e3sm_to_cmip/cmor, to see if the dimension of the variables are remapped to lat lon coordinate, as expected? |
In Then: @chengzhuzhang How can I tell if it's remapped? |
Based on the dimensions, it looks like global time series files are used as input for e3sm_to_cmip. The expected files should come from regrided files with lat lon dimensions, or native land output is also supported for v3 tri-grid. |
|
Looking at Then: Indeed, the dimensions are very different. So, we want the
The only way I can think to do that would be to feed in the output of the I also argue that such a design would elevate #467/#496 in priority. The dependency chain should be " @chengzhuzhang I feel like I'm not following your intended design here. Can you clarify further? Thanks! |
I created #542 to track this. We can choose to do that as part of this PR or separately, if we're ok with always setting |
Just to clarify, setting |
No, I don't think so. I think it would need the |
|
@chengzhuzhang I set Although, I am now puzzled why we needed
So it looks like the issues were wildcards? |
@chengzhuzhang Following up from this week's meeting, in which you said that I did a diff of the |
|
@chengzhuzhang Given the issues in the above 2 comments have been resolved, I think we have proof that |
The |
I think so. It would be useful to starting a configuration file to summarize findings from this issue discussion. |
|
Therefore, I am closing this PR. As mentioned above, a sample land configuration can be found in #545. |
14 tasks
3 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Test
climoandtstasks for land variables. Resolves #531.