Add fixes to allow carbon simulations to run with a single advected species #1963

yantosca · 2023-09-15T14:07:55Z

Name and Institution (Required)

Name: Bob Yantosca
Institution: Harvard + GCST

Confirm you have reviewed the following documentation

Contributing guidelines

Describe the update

This PR adds several updates & fixes to allow the carbon simulations to use only a single advected species. A new script run/shared/singleCarbonSpecies is called to edit the "out-of-the-box" run directory configuration files (which by default advect CH4, CO, CO2, and OCS) so that only the species of interest is advected (and that emissions/diagnostics for the other species are disabled).

The singleCarbonSpecies.sh script is called from the createRunDir.sh script for both GCClassic and GCHP, which now asks if you wish to use all species or a single species:

===========================================================
GEOS-CHEM RUN DIRECTORY CREATION
===========================================================

-----------------------------------------------------------
Choose simulation type:
-----------------------------------------------------------
   1. Full chemistry
   2. Aerosols only
   3. CH4
   4. CO2
   5. Hg
   6. POPs
   7. Tagged CH4
   8. Tagged CO
   9. Tagged O3
  10. TransportTracers
  11. Trace metals
  12. Carbon
>>> 12

-----------------------------------------------------------
Do you wish to use a single advected species?
-----------------------------------------------------------
  1. Use all species
  2. Use CH4 only
  3. Use CO2 only
  4. Use CO only
  5. Use OCS only
>>>

Also, an integration test has been added (for GEOS-Chem Classic only at this time, GCHP to follow) for the carbon simulation with CH4 as the only advected species.

Expected changes

This is a zero-diff update w/r/t the fullchem benchmark simulations. It will only affect the carbon simulation by giving the user the option to advect a single species.

Reference(s)

N/A

Related Github Issue(s)

N/A

run/GCClassic/HEMCO_Config.rc.templates/HEMCO_Config.rc.carbon run/GCHP/HEMCO_Config.rc.templates/HEMCO_Config.rc.carbon - Now read all CHEMISTRY_INPUT data regardless of the settings of USE_CH4_DATA, USE_CO_DATA, USE_CO2_DATA. - The only exception is that if USE_CO_DATA is false, we have to read alternate entries of GMI_PROD_CO and GMI_LOSS_CO with "*" as the species name instead of CO. Otherwise, this will cause a HEMCO error, as HEMCO will try to place these data into the nonexistent CO species. CHANGELOG.md - Updated accordingly Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

In this commit, we have merged GEOS-Chem 14.1.1 into the the development branch "bugfix/carbon-single-tracer", which contains updates that allow the carbon simulation to be run with a single species. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

GeosCore/chemistry_mod.F - Wrap the call to "Chem_Carbon_Gases with the FlexChem timer GeosCore/carbon_gases_mod.F(09 - Add KPP and RCONST timers Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

This merge brings the bugfix/carbon-single-tracer branch up-to-date with the GEOS-Chem 14.2.0-alpha tag. The 14.2.0-alpha.2 tag was shown to have zero differences w/r/t GEOS-Chem 14.1.1 so this is a good place to update to. This will also allow further updates to the carbon simulation to take advantage of the updated HEMCO 3.7.0 in development. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

This merge brings the "bugfix/carbon-single-tracer" up to 14.2.0-rc.1. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

run/GCClassic/HEMCO_Config.rc.templates/HEMCO_Config.rc.carbon run/GCHP/HEMCO_Config.rc.templates/HEMCO_Config.rc.carbon - Change "Verbose: 0" to "Verbose: false" for HEMCO 3.7.0 - Remove "Warnings: 1" for HEMCO 3.7.0 - Move "2009 CHINA_MASK" out of the USE_CO_DATA.or.USE_CO2_DATA block, this was causing HEMCO to crash with a I/O read error. run/GCClassic/HEMCO_Diagn.rc.templates/HEMCO_Diagn.rc.carbon run/GCHP/HEMCO_Diagn.rc.templates/HEMCO_Diagn.rc.carbon - Removed Tagged CO entries - Fixed comment headers run/shared/single_carbon_species.sh - Script that updates carbon-simulation configuration files so that it is possible to run with a single species. More testing and development is needed. This will eventually be called from the run/GCClassic/createRunDir.sh and run/GCHP/createRunDir.sh. - Currently seems to work for GCClassic; Needs more modifications to edit the ExtData.rc file properly. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

This merge brings the updates for the GEOS-Chem carbon simulation (with single tracer rundirectory option) up to date with the latest state of the 14.2.1 development stream. This also brings updates from 14.2.0-alpha.19 into the bugfix/carbon-single-tracer branch. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

run/shared/singleCarbonSpecies.sh - Renamed from run/single_carbon_species.sh run/shared/single_carbon_species.sh - Moved to singleCarbonSpecies.sh run/GCClassic/createRunDir.sh run/GCHP/createRunDir.sh - Use functions from run/shared/singleCarbonSpecies.sh to edit rundir configuration files for a single species simulation CHANGELOG.md - Updated accordingly Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

run/shared/singleCarbonSpecies.sh - Add function updateExtData, which removes entries in ExtData.rc for species that are not being retained in a single-species carbon simulation. - Updated comments - Trimmed trailing whitespace - Replaced tabs with spaces Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

run/GCClassic/createRunDir.sh run/GCHP/createRunDir.sh - Set sim_extra_option to "none" if option 1 is selected - Only call singleCarbonSpecies.sh for the carbon simulation if sim_extra_option is not none These fixes prevented an infinite loop because the sim_extra_option variable was not being set to the default value of "none" if the user selected to "use all species". Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

test/integration/GCClassic - Add integration test for carbon simulation with CH4 as the only advected species (gc_4x5_merra2_carbon_CH4only) - Update run directory creation commands in integrationTestCreate.sh for GEOS-Chem Classic NOTE: We will add this functionality to GCHP integration tests soon. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

yantosca · 2023-09-15T14:11:46Z

GEOS-Chem Classic integration tests on the bugfix/single-carbon-tracer branch all passed, except for TOMAS, which is a known issue that will be fixed in 14.3.0.

==============================================================================
GEOS-Chem Classic: Execution Test Results

GCClassic #f73eb7e GEOS-Chem submod update: Merge PR #1957 and PR #1958
GEOS-Chem #24a6b77ef Update int test scripts for single-species carbon simulation
HEMCO     #daf54d0 Merge PR #235 (Bug fix for vertical regridding)

Using 24 OpenMP threads
Number of execution tests: 27

Submitted as SLURM job: 2217582
==============================================================================
 
Execution tests:
------------------------------------------------------------------------------
...
gc_4x5_47L_merra2_fullchem_TOMAS15..................Execute Simulation....FAIL
gc_4x5_47L_merra2_fullchem_TOMAS40..................Execute Simulation....FAIL
...
 
Summary of test results:
------------------------------------------------------------------------------
Execution tests passed: 25
Execution tests failed: 2
Execution tests not yet completed: 0

All GCHP integration tests also passed:

==============================================================================
GCHP: Execution Test Results

GCClassic #d5180cc GEOS-Chem submod update: Merge PR #1957 and PR #1958
GEOS-Chem #24a6b77ef Update int test scripts for single-species carbon simulation
HEMCO     #daf54d0 Merge PR #235 (Bug fix for vertical regridding)

Number of execution tests: 5

Submitted as SLURM job: 2217719
==============================================================================

...
 
Summary of test results:
------------------------------------------------------------------------------
Execution tests passed: 5
Execution tests failed: 0
Execution tests not yet completed: 0

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%  All execution tests passed!  %%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

yantosca · 2023-09-15T15:05:42Z

Also tagging @hannahnesser @kbowman77 @dbajones for your reference.

yantosca · 2023-09-15T15:16:54Z

Integration tests on bugfix/carbon-single-tracer locally merged into dev/14.2.1 are now running.

yantosca · 2023-09-15T16:10:20Z

After merging locally atop PR #1957 and PR #1958, all GEOS-Chem Classic integration tests passed, except TOMAS (which will be fixed later).

==============================================================================
GEOS-Chem Classic: Execution Test Results

GCClassic #f73eb7e GEOS-Chem submod update: Merge PR #1957 and PR #1958
GEOS-Chem #155c6b66d Merge PR #1963 (Fixes to allow carbon sims w/ a single advected species)
HEMCO     #daf54d0 Merge PR #235 (Bug fix for vertical regridding)

Using 24 OpenMP threads
Number of execution tests: 27

Submitted as SLURM job: 2284433
==============================================================================
 
Summary of test results:
------------------------------------------------------------------------------
Execution tests passed: 25
Execution tests failed: 2
Execution tests not yet completed: 0

Also all integration tests were zero-diff w/r/t the prior integration test, except for these:

Checking gc_4x5_merra2_carbon
   -> 1 difference found in OutputDir
      * GCC_r24/rundirs/gc_4x5_merra2_carbon/OutputDir/HEMCO_diagnostics.201901010000.nc 
        GCC_r25/rundirs/gc_4x5_merra2_carbon/OutputDir/HEMCO_diagnostics.201901010000.nc 

Checking gc_4x5_merra2_fullchem_APM
   -> 2 differences found in OutputDir
      * GCC_r24/rundirs/gc_4x5_merra2_fullchem_APM/OutputDir/GEOSChem.Metrics.20190701_0000z.nc4 
        GCC_r25/rundirs/gc_4x5_merra2_fullchem_APM/OutputDir/GEOSChem.Metrics.20190701_0000z.nc4 
      * GCC_r24/rundirs/gc_4x5_merra2_fullchem_APM/OutputDir/GEOSChem.SpeciesConc.20190701_0000z.nc4 
        GCC_r25/rundirs/gc_4x5_merra2_fullchem_APM/OutputDir/GEOSChem.SpeciesConc.20190701_0000z.nc4 
   -> 1 difference found in Restarts
      * GCC_r24/rundirs/gc_4x5_merra2_fullchem_APM/Restarts/GEOSChem.Restart.20190701_0100z.nc4 
        GCC_r25/rundirs/gc_4x5_merra2_fullchem_APM/Restarts/GEOSChem.Restart.20190701_0100z.nc4 

Checking gc_4x5_merra2_fullchem_RRTMG
   -> 1 difference found in OutputDir
      * GCC_r24/rundirs/gc_4x5_merra2_fullchem_RRTMG/OutputDir/GEOSChem.RRTMG.20190701_0000z.nc4 
        GCC_r25/rundirs/gc_4x5_merra2_fullchem_RRTMG/OutputDir/GEOSChem.RRTMG.20190701_0000z.nc4

The difference in the carbon simulation is because in this PR, we have now activated the CO2 biomass burning for GFED2, where it was deactivated before:

< 
< #NOTE: Biomass burning doesn't seem to work.  Keep here for now.
< # Uncomment this if GFED is used
< #EmisCO2_Biomass_GFED  CO2    111    1    1   2  kg/m2/s  CO2_biomass_burning_emissions_from_GFED4
< 
< # Uncomment this if FINN is used
---
> EmisCO2_Biomass_GFED  CO2    111    1    1   2  kg/m2/s  CO2_biomass_burning_emissions_from_GFED4
75,148d68

yantosca

I will also push the following changes as a post-release commit:

commit c8e0ebefaa143c6d36289858a3fd531c78d9b8ad (HEAD -> dev/14.2.1)
Author: Bob Yantosca <yantosca@seas.harvard.edu>
Date:   Fri Sep 15 12:39:00 2023 -0400

    PR #1963 post-merge fix: Change Emis*_Bioburn to Emis*_BiomassBurn
    
    run/GCClassic/HEMCO_Diagn.rc.templates/HEMCO_Diagn.rc.carbon
    run/GCHP/HEMCO_Diagn.rc.templates/HEMCO_Diagn.rc.carbon
    - Renamed EmisCO_Bioburn to EmisCO_BiomassBurn
    - Renamed EmisCO2_Bioburn to EmisCO2_BiomassBurn
    - Renamed EmisOCS_Bioburn to EmisOCS_BiomassBurn
    - Now use Cat = -1 and Hier = -1 for Emis*_BiomassBurn containers
      as well as extension 111 (GFED)
    
    run/GCHP/HISTORY.rc.templates/HISTORY.rc.carbon
    - Renamed EmisCO_Bioburn to EmisCO_BiomassBurn
    - Renamed EmisCO2_Bioburn to EmisCO2_BiomassBurn
    - Added OCS emission diagnostics
    
    Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

msulprizio

I just had some minor questions and comments. See below.

CHANGELOG.md

run/GCClassic/createRunDir.sh

run/shared/singleCarbonSpecies.sh

run/GCClassic/HEMCO_Diagn.rc.templates/HEMCO_Diagn.rc.carbon run/GCHP/HEMCO_Diagn.rc.templates/HEMCO_Diagn.rc.carbon - Renamed EmisCO_Bioburn to EmisCO_BiomassBurn - Renamed EmisCO2_Bioburn to EmisCO2_BiomassBurn - Renamed EmisOCS_Bioburn to EmisOCS_BiomassBurn - Now use Cat = -1 and Hier = -1 for Emis*_BiomassBurn containers as well as extension 111 (GFED) run/GCHP/HISTORY.rc.templates/HISTORY.rc.carbon - Renamed EmisCO_Bioburn to EmisCO_BiomassBurn - Renamed EmisCO2_Bioburn to EmisCO2_BiomassBurn - Added OCS emission diagnostics Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

run/GCClassic/createRunDir.sh - Removed leftover ls command from debugging run/GCClassic/singleCarbonSpecies.sh - Removed commented-out code Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

CHANGELOG.md - Removed the >>>>>>> line that was left over at line 137 Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

yantosca · 2023-09-19T14:33:16Z

@msulprizio: I have merged these fixes into dev/14.2.1 locally and am running integration tests. Results to follow.

yantosca · 2023-09-19T16:01:43Z

After making the fixes requested by @msulprizio, all GEOS-Chem Classic integration tests passed (except TOMAS, which is a known issue):

==============================================================================
GEOS-Chem Classic: Execution Test Results

GCClassic #f73eb7e GEOS-Chem submod update: Merge PR #1957 and PR #1958
GEOS-Chem #5d6b1f53e Merge PR #1963 (Add fixes for single-species carbon simulation )
HEMCO     #daf54d0 Merge PR #235 (Bug fix for vertical regridding)

Using 24 OpenMP threads
Number of execution tests: 27

Submitted as SLURM job: 2568433
==============================================================================
 
Summary of test results:
------------------------------------------------------------------------------
Execution tests passed: 25
Execution tests failed: 2
Execution tests not yet completed: 0

All of the GEOS-Chem classic integration tests were identical to the prior integration test except:

Checking gc_4x5_merra2_carbon
   -> 1 difference found in OutputDir
      * GCC_r24/rundirs/gc_4x5_merra2_carbon/OutputDir/HEMCO_diagnostics.201901010000.nc 
        GCC_r25/rundirs/gc_4x5_merra2_carbon/OutputDir/HEMCO_diagnostics.201901010000.nc 
   -> No differences in Restarts

Checking gc_4x5_merra2_fullchem_APM
   -> 2 differences found in OutputDir
      * GCC_r24/rundirs/gc_4x5_merra2_fullchem_APM/OutputDir/GEOSChem.Metrics.20190701_0000z.nc4 
        GCC_r25/rundirs/gc_4x5_merra2_fullchem_APM/OutputDir/GEOSChem.Metrics.20190701_0000z.nc4 
      * GCC_r24/rundirs/gc_4x5_merra2_fullchem_APM/OutputDir/GEOSChem.SpeciesConc.20190701_0000z.nc4 
        GCC_r25/rundirs/gc_4x5_merra2_fullchem_APM/OutputDir/GEOSChem.SpeciesConc.20190701_0000z.nc4 
   -> 1 difference found in Restarts
      * GCC_r24/rundirs/gc_4x5_merra2_fullchem_APM/Restarts/GEOSChem.Restart.20190701_0100z.nc4 
        GCC_r25/rundirs/gc_4x5_merra2_fullchem_APM/Restarts/GEOSChem.Restart.20190701_0100z.nc4 

Checking gc_4x5_merra2_fullchem_RRTMG
   -> 1 difference found in OutputDir
      * GCC_r24/rundirs/gc_4x5_merra2_fullchem_RRTMG/OutputDir/GEOSChem.RRTMG.20190701_0000z.nc4 
        GCC_r25/rundirs/gc_4x5_merra2_fullchem_RRTMG/OutputDir/GEOSChem.RRTMG.20190701_0000z.nc4 
   -> No differences in Restarts

We expect the carbon simulation not to be zero-diff, as this simulation contains diagnostics that have been renamed w/r/t the prior integration test. Also the APM and RRTMG are known issues that will be fixed later.

yantosca · 2023-09-19T16:03:38Z

All GCHP integration tests passed

==============================================================================
GCHP: Execution Test Results

GCClassic #d5180cc GEOS-Chem submod update: Merge PR #1957 and PR #1958
GEOS-Chem #5d6b1f53e Merge PR #1963 (Add fixes for single-species carbon simulation )
HEMCO     #daf54d0 Merge PR #235 (Bug fix for vertical regridding)

Number of execution tests: 5

Submitted as SLURM job: 2568462
==============================================================================
 

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%  All execution tests passed!  %%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

And all were zero-diff w/r/t the prior integration test.

msulprizio

Looks good to merge now!

yantosca added 11 commits February 23, 2023 11:43

Add timers for carbon simulation

362a55a

GeosCore/chemistry_mod.F - Wrap the call to "Chem_Carbon_Gases with the FlexChem timer GeosCore/carbon_gases_mod.F(09 - Add KPP and RCONST timers Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

Merge branch 'dev/14.2.0' into bugfix/carbon-single-tracer

ea3c4a5

This merge brings the "bugfix/carbon-single-tracer" up to 14.2.0-rc.1. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

yantosca added this to the 14.2.1 milestone Sep 15, 2023

yantosca requested a review from msulprizio September 15, 2023 14:07

yantosca self-assigned this Sep 15, 2023

yantosca commented Sep 15, 2023

View reviewed changes

msulprizio reviewed Sep 15, 2023

View reviewed changes

CHANGELOG.md Outdated Show resolved Hide resolved

run/GCClassic/createRunDir.sh Outdated Show resolved Hide resolved

run/shared/singleCarbonSpecies.sh Outdated Show resolved Hide resolved

yantosca added 3 commits September 19, 2023 10:00

Clean up run directory scripts for carbon simulation

610d00b

run/GCClassic/createRunDir.sh - Removed leftover ls command from debugging run/GCClassic/singleCarbonSpecies.sh - Removed commented-out code Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

Removed leftover text from a Git merge from CHANGELOG.md

c273374

CHANGELOG.md - Removed the >>>>>>> line that was left over at line 137 Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

yantosca requested a review from msulprizio September 19, 2023 14:29

msulprizio approved these changes Sep 20, 2023

View reviewed changes

yantosca merged commit 5d6b1f5 into dev/14.2.1 Sep 20, 2023

yantosca deleted the bugfix/carbon-single-tracer branch September 20, 2023 15:33

yantosca mentioned this pull request Sep 20, 2023

[feature] add config variable to perturb CH4 boundary conditions #1934

Merged

1 task

yantosca mentioned this pull request Nov 6, 2023

Adding new dummy species with loss rate dependent on meteorology #2008

Closed

4 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add fixes to allow carbon simulations to run with a single advected species #1963

Add fixes to allow carbon simulations to run with a single advected species #1963

yantosca commented Sep 15, 2023 •

edited

Loading

yantosca commented Sep 15, 2023

yantosca commented Sep 15, 2023

yantosca commented Sep 15, 2023

yantosca commented Sep 15, 2023

yantosca left a comment •

edited

Loading

msulprizio left a comment

yantosca commented Sep 19, 2023

yantosca commented Sep 19, 2023 •

edited

Loading

yantosca commented Sep 19, 2023

msulprizio left a comment

Add fixes to allow carbon simulations to run with a single advected species #1963

Add fixes to allow carbon simulations to run with a single advected species #1963

Conversation

yantosca commented Sep 15, 2023 • edited Loading

Name and Institution (Required)

Confirm you have reviewed the following documentation

Describe the update

Expected changes

Reference(s)

Related Github Issue(s)

yantosca commented Sep 15, 2023

yantosca commented Sep 15, 2023

yantosca commented Sep 15, 2023

yantosca commented Sep 15, 2023

yantosca left a comment • edited Loading

Choose a reason for hiding this comment

msulprizio left a comment

Choose a reason for hiding this comment

yantosca commented Sep 19, 2023

yantosca commented Sep 19, 2023 • edited Loading

yantosca commented Sep 19, 2023

msulprizio left a comment

Choose a reason for hiding this comment

yantosca commented Sep 15, 2023 •

edited

Loading

yantosca left a comment •

edited

Loading

yantosca commented Sep 19, 2023 •

edited

Loading