Skip to content

Answer Changing Tag NEW PARAMS FILE (ctsm51d154_2deg_GSWP3V1_hist) #38

Closed

Description

@slevis-lmwg

Description:

For updating the Answer Changing Tags wiki page. Previous one was #28.

Used script /glade/work/slevis/Answer_Changing_Tags/setup_run_answer_changing_tags.csh_dev145-154 to start the case.
(Remove _dev145-154 from file name to run it.)


Case directory:
/glade/work/slevis/git_ctsm_tags/ctsm5.1.dev154/cime/scripts/ctsm51d154_2deg_GSWP3V1_hist


Sandbox:
/glade/work/slevis/git_ctsm_tags/ctsm5.1.dev154


user_nl_ changes:
Other than the usual user_nl mods for Answer Changing Tags:

 finidat = '/glade/campaign/cgd/tss/people/slevis/CLM5_restarts/ctsm51d145_2deg_GSWP3V1_hist.clm2.r.1989-01-01-00000.nc'
 use_init_interp = .true.

Note that the finidat used here originates in #28.


SourceMods:
The simulation failed at the year transition from 2001 to 2002. I have tried the following:

  1. WORKED. In issue Meier2022 option fails in BiogeophysPreFluxCalcsMod.F90 line 208 in IHist case ESCOMP/CTSM#2264 I explain this first fix as:
    "remove the new if istcrop, as well as the endrun and just set z0m and displa to 0 for tiny htop."
  2. FAILED. In the same issue, I explain the second fix in this post and why it failed.
  3. DONE same as (1) but with these suggestions to set z0m = z0mg and run with DEBUG.
  4. DONE same as (1) but reverting to the pre-dev154 default z0param_method = 'ZengWang2007'.

Diagnostics:
1 vs. previous answer-changing tag:
https://webext.cgd.ucar.edu/I20TR/ctsm51d154_2deg_GSWP3V1_hist/lnd/ctsm51d154_2deg_GSWP3V1_hist.1995_2014-ctsm51d145_2deg_GSWP3V1_hist.1995_2014/setsIndex.html

3 vs. 4: https://webext.cgd.ucar.edu//I20TR/ctsm51d154_2deg_GSWP3V1_hist/lnd/ctsm51d154_2deg_GSWP3V1_hist.1995_2014-ctsm51d154like156_2deg_GSWP3V1_hist.1995_2014/setsIndex.html


Output:

  1. /glade/campaign/cgd/tss/people/slevis/ANSWER_CHANGING_TAGS/ctsm51d154_2deg_GSWP3V1_hist/iss38_opt1
  2. /glade/campaign/cgd/tss/people/slevis/ANSWER_CHANGING_TAGS/ctsm51d154_2deg_GSWP3V1_hist
  3. /glade/campaign/cgd/tss/people/slevis/ANSWER_CHANGING_TAGS/ctsm51d154like156_2deg_GSWP3V1_hist

Contacts:
@slevis-lmwg


Extra details:
1850-1919 cycle GSWP3 1901-1920 data
1920-2014 use GSWP3 1901-2014 data

Activity

  1. self-assigned this
    on Nov 22, 2023
  2. slevis-lmwg commented on Nov 30, 2023

    @slevis-lmwg
    CollaboratorAuthor

    I have posted diagnostics from this run.
    I have not looked though.

  3. wwieder commented on Nov 30, 2023

    @wwieder
    Collaborator

    It's kind of hard for me to understand this one, since it has the FUN-Bug and Meier roughness changes. I'm assuming fun is causing reductions in LAI, and then is Meier driving increases in ET over some of these same regions (tropical and boreal forests)?

    The new model has higher vegetation temps in some regions

  4. olyson commented on Nov 30, 2023

    @olyson
    Collaborator

    I think the decrease in sensible heat flux and increase in surface temperature (TG) in warm desert regions (e.g., Sahara) is consistent with the Meier paper results. Caused by lower z0mg.
    The z0mv of vegetation is generally increased in Meier, so given the availability of moisture in the tropics that is probably driving an increase in the latent heat flux in those regions, against generally consistent with Meier.
    We ourselves actually never did a clean comparison of Meier against a control because of bugs we ran into and the discussion we had regarding zetamaxstable and the changes made there. Our efforts were focused on sorting out those issues. Now that Meier in the model has settled down, it might be worth doing a simulation as in #38 here but with Meier off, all else being the same. A comparison with the Meier paper results might be complicated by the fact that they put each pft on its own soil column, I'm not sure what effect that might have with respect to roughness results.

  5. wwieder commented on Nov 30, 2023

    @wwieder
    Collaborator
  6. slevis-lmwg commented on Nov 30, 2023

    @slevis-lmwg
    CollaboratorAuthor

    Can do.

    First, I have started #38 again with Keith's and Erik's suggestions.

  7. slevis-lmwg commented on Nov 30, 2023

    @slevis-lmwg
    CollaboratorAuthor

    Easiest way to do the new run is to clone #38 and add to user_nl_clm:
    z0param_method = 'ZengWang2007'
    ...rather than starting from scratch by checking out dev156. I hope you're both ok with this choice.

  8. olyson commented on Dec 1, 2023

    @olyson
    Collaborator

    Seems ok to me.

  9. slevis-lmwg commented on Dec 5, 2023

    @slevis-lmwg
    CollaboratorAuthor

    Just posted new diags up top:
    Comparing dev154 (Meier2022 on) to dev156 (Meier2022 off)

  10. wwieder commented on Dec 6, 2023

    @wwieder
    Collaborator

    This is great, Sam. Thanks for posting the clear Meier on-off posts. It seems like this is working as intended (e.g. change in evaporative fraction in the tropics). Once all the bugs are fixed it seems like we're ready to turn this on by default?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions