b4b-dev: Change some _U and _R history fields to be on by default #3667
Conversation
Derived from half-degree run started 1959-01-01, forced with CRU-JRA, at tag alpha-ctsm5.4.CMIP7.15.ctsm5.3.079. Growing seasons defined by files [sh]dates_ggcmi_crop_calendar_phase3_v1.01_nninterp-hcru_hcru_mt13.2000-2000.20230728_165845.nc. Maximum growing season length in generate_gdds.py derived from mxmat in clm60_params.crop_omni02.c251120.nc with no 'cushion'. generate_gdds.py first season 1995 and last season 2005.
Derived from half-degree run started 1959-01-01, forced with CRU-JRA, at tag alpha-ctsm5.4.CMIP7.15.ctsm5.3.079. generate_gdd20_baseline -v GDDB20 -y1 1995 -yN 2005.
…leName_gdd20_baseline.
post-5.4.n00: New crop calendar inputs and updated params (omni02)
…4b_changes_to_ctsm5.4 Conflicts: bld/unit_testers/build-namelist_test.pl
…if the file wasn't found use the cmip6 version approach
Post 5.4 namelist updates
olyson
added
the
next
|
clm_short tests PASS on Derecho except for FIELDLIST failures as expected. |
|
@ekluzek and I agreed I should rebase to b4b-dev when the current b4b-dev gets merged to master. |
slevis-lmwg
left a comment
slevis-lmwg
left a comment
There was a problem hiding this comment.
Very straightforward. Approving. From discussing with Erik, seems ok to merge to b4b-dev.
ekluzek
left a comment
ekluzek
left a comment
There was a problem hiding this comment.
This is of course simple and fine to bring in. @slevis-lmwg will merge this to b4b-dev after the b4b-dev goes to master.
I worry about adding more output to the default history. But, there is an established reason based on removing them that we need to add them back in. And these are 2D fields so not that big of a deal.
It does remind me that we need better ways of handling history fields, so we can turn on different levels of output. Some thinking around a way to implement that is here: #528 (comment)
It also seems like there could be a way to flag history fields in "scientific interest" categories, something like:
urban, crop, hydrology, biogeochemistry
where turning on a category would flag turning on extra fields for those categories. This way standard simulations wouldn't have them on, but production simulations and simulations for a scientific interest would turn on that area to get more output.
We have a few ad-hoc ways of doing this sort of thing with specific namelist settings:
- hist_wrtch4diag
- calc_human_stress_indices (turns on calculation and history output of diagnostic fields)
- hist_wrt_matrixcn_diag
Then of course turning on certain options turns on default fields that go with those options.
ekluzek
added
enhancement
slevis-lmwg
changed the title
|
I rebased to b4b-dev expecting that "files changed" may look temporarily messed up (which they do). This will get fixed (we will go back to 6 files changed) when the current b4b-dev gets merged to master, and then I merge the updated b4b-dev to this branch. |
5 participants
Description of changes
Change some _U and _R history fields to be on by default
Specific notes
For CMIP6, we had eliminated a bunch of _U and _R history fields from the default monthly history files because we determined that they could be reconstructed from auxiliary sub-grid history tapes. However, feedback from CLMU community was that it was an onerous process to do so and made it difficult to work with the urban (and rural) variables. This PR reinstates some of these _U and _R fields, mainly so that urban/rural contrasts in temperature and humidity as well as the main urban energy balance components can be more easily analyzed.
Contributors other than yourself, if any: NA
CTSM Issues Fixed (include github issue #): NA
Are answers expected to change (and if so in what way)? No. New history fields will appear by default.
Any User Interface Changes (namelist or namelist defaults changes)? No
Does this create a need to change or add documentation? Did you do so? No. No.
Testing performed, if any:
I will run clm_short on Derecho once it comes back up to make sure the new code compiles and answers don't change.