Only convert units for species that require it #2072

yantosca · 2023-12-12T20:37:04Z

Name and Institution (Required)

Name: Bob Yantosca
Institution: Harvard + GCST

Confirm you have reviewed the following documentation

Contributing guidelines

Describe the update

This PR updates the unit conversion routines as follows:

Introduces State_Chm%Species(N)%Units and State_Chm%Species(N)%Previous_Units fields
Updates in GeosUtil/unitconv_mod.F90:
- In Convert_Spc_Units:
  - Now accepts a mapping argument, specifying the modelId species indices
  - Argument outUnit is renamed to new_units
  - Argument origUnit is renamed to previous_units
- Added function Check_Units
- Added function Check_Previous_Units

Calls to Convert_Spc_Units throughout the code have been updated correspondingly.

Expected changes

We will see small differences at the level of numerical noise.

Related Github Issue(s)

yantosca · 2023-12-12T20:46:38Z

Still shaking down some bugs in the implementation.

Headers/species_mod.F90 - Add SpcConc%Units and SpcConc%Previous_Units to store the integer flags for the current and previous species units. - Add comments Headers/state_chm_mod.F90 - Add State_Chm%Map_All(:) mapping vector, which returns all species indices. This is needed for species-specific unit conversions. - Set State_Chm%Species(:)%Units = 0 at init Set State_Chm%Species(:)%Previous_Units = 0 at init CHANGELOG.md - Updated accordingly Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

GeosUtil/unitconv_mod.F90 - Added routine Check_Input_Units - In each unit conversion routine: - Accepts a mapping array as an argument. This will allow the calling routine to pass one of the State_Chm%Map_* vectors, which specifies the ModelId's of the species for which unit conversions will be done. - Removed references to State_Chm%Spc_Units. - Added new loop index S. - N (the modelID) is now obtained with N = mapping(S) - Update units metadata in State_Chm%Species(N)%Units and State_Chm%Species(N)%Previous_Units - Pull computations outside of (I,J,L) loops where expedient and store the results in !$OMP PRIVATE loop variables for use within (I,J,L) loops. - Store conversion factors in an !$OMP PRIVATE loop variable, both for clarity and to avoid repeated computations - Now place adjoint code in IF/THEN blocks (instead of using a single IF statement with continuation characters. - Change e.g. 1e-3fp to 1.0e-3_fp, etc. - Updated comments CHANGELOG.md - Updated accordingly

GeosUtil/unitconv_mod.F90 - Rename function Check_Input_Units to Check_Units. - Add function Check_Previous_Units - In routine Convert_Spc_Units: - Add a single call to Check_Units before the CASE statement. - Rename "outUnit" argument to "new_units" - Rename "origUnit" argument to "previous_units" - Add errUnits error message variable - Add a single call to Check_Previous_Units before exiting, to ensure that all species have the same Previous_Units value - In routine "Print_Species_Kg" - Update calls to "Convert_Spc_Units" - In other routines - Remove calls to Check_Input_Units Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

This commit updates the various calls to Convert_Spc_Units in modules throughout the GeosCore/*.F90, Interfaces/GCHP/*.F90, and Interfaces/GEOS/*.F90. Updates include: - Adding the "mapping" argument to pass one of the State_Chm%Map_* arrays. This is the list of modelId's for each species. - Renaming "outUnit" -> "new_units" - Renaming "origUnit" -> "previous_units NOTE: Some additional work will be needed, as some code (mostly in TOMAS) still relies on the old State_Chm%Spc_Units variable. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

This commit fixes several minor issues that were left over from the prior commit. GeosCore/aerosol_mod.F90 - Add missing comma after "previous_units" GeosCore/fullchem_mod.F90 - Rename "origUnit" -> "previous_units" GeosCore/hco_interface_gc_mod.F90 - Add missing continuation character & - Rename "outUnit" -> "new_units" GeosCore/hco_utilities_gc_mod.F90 GeosCore/vdiff_mod.F90 - Rename "origUnit" -> "previous_units" - Rename "outUnit" -> "new_units" - Add mapping argument in calls to Convert_Spc_Units GeosUtil/unitconv_mod.F90 - Rename "inUnit" -> "in_units" - Rename "outUnit" -> "new_units" ObsPack/obspack_mod.F90 - Fixed typo: "State_Chem" -> "State_Chm" GeosCore/mercury_mod.F90 - Bug fix: previous_units should be INTEGER, not LOGICAL GeosCore/sulfate_mod.F90 - Rename "origUnit" -> "previous_units" - Rename "outUnit" -> "new_units" Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

GeosCore/calc_met_mod.F90 - Add "units" variable to store units in UpdtMR block - Only convert units for species that have dry units (moved IF block to within the N loop over species0 GeosCore/chemistry_mod.F90 - Remove reference to State_Chm%Spc_Units GeosCore/hco_utilities_gc_mod.F90 - Remove reference to State_Chm%Spc_Units - Set initial units and previous_units values for each species N GeosCore/ucx_mod.F90 - Remove reference to State_Chm%Spc_Units - Update error check for H2O units using Spc(id_H2O)%Units Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

GeosUtil/unitconv_mod.F90 - Undo numerical optimizations where we pull terms outside of the IJL loops. This seems to cause some numerical differences. We can return to this later. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

GeosUtil/unitconv_mod.F90 - Now declare Check_Units and Check_Previous_Units as PUBLIC, so that they can be seen outside the module - Rename the "unit" argument to "units" in both Check_Units and Check_Previous_Units routines, for consistency Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

GeosCore/aerosol_mod.F90 - State_Chm is now INTENT(INOUT) in routine Set_Aermass_Diagnostic - Now call Check_Units; Remove reference to State_Chm%Spc_Units GeosCore/chemistry_mod.F90 GeosCore/diagnostics_mod.F90 GeosCore/set_boundary_conditions.F90 - Now call Check_Units; Remove reference to State_Chm%Spc_Units Headers/state_chm_mod.F90 - Remove Spc_Units from TYPE(ChmState) History/history_mod.F90 - Replace State_Chm%Spc_Units with State_Chm%Species(1)%Units Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

GeosUtil/unitconv_mod.F90 - Remove loops over (I,J,L) and replace with array operations, which are more efficient - Compute constant terms only once where expedient - Reduce the number of local variables where expedient Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

yantosca · 2023-12-15T17:17:44Z

The unit conversions update result in a noticeable speedup in GEOS-Chem fullchem simullations.

Both runs on huce_cascade, --exclusive, 48 cores

7day fullchem benchmark run, ref (dev/no-diff-to-benchmark)

G E O S - C H E M   T I M E R S
 
  Timer name                       DD-hh:mm:ss.SSS     Total Seconds
-------------------------------------------------------------------------------
  GEOS-Chem                     :  00-00:52:27.125          3147.125
  HEMCO                         :  00-00:09:27.250           567.250
  All chemistry                 :  00-00:18:57.500          1137.500
  => Gas-phase chem             :  00-00:11:14.875           674.875
  => Photolysis                 :  00-00:01:17.250            77.250
  => Aerosol chem               :  00-00:06:06.875           366.875
  => Linearized chem            :  00-00:00:05.000             5.000
  Transport                     :  00-00:03:52.125           232.125
  Convection                    :  00-00:04:02.875           242.875
  Boundary layer mixing         :  00-00:07:14.250           434.250
  Dry deposition                :  00-00:00:10.625            10.625
  Wet deposition                :  00-00:01:32.250            92.250
  Diagnostics                   :  00-00:06:09.250           369.250
  Unit conversions              :  00-00:02:00.500           120.500

7day fullchem benchmark run, ref (feature/units-array-ops)

G E O S - C H E M   T I M E R S
 
  Timer name                       DD-hh:mm:ss.SSS     Total Seconds
-------------------------------------------------------------------------------
  GEOS-Chem                     :  00-00:48:45.750          2925.750
  HEMCO                         :  00-00:07:58.250           478.250
  All chemistry                 :  00-00:17:58.250          1078.250
  => Gas-phase chem             :  00-00:11:10.375           670.375
  => Photolysis                 :  00-00:01:18.875            78.875
  => Aerosol chem               :  00-00:05:10.375           310.375
  => Linearized chem            :  00-00:00:04.125             4.125
  Transport                     :  00-00:03:37.500           217.500
  Convection                    :  00-00:03:59.625           239.625
  Boundary layer mixing         :  00-00:06:43.750           403.750
  Dry deposition                :  00-00:00:08.250             8.250
  Wet deposition                :  00-00:01:24.250            84.250
  Diagnostics                   :  00-00:06:08.000           368.000
  Unit conversions              :  00-00:01:26.250            86.250

yantosca · 2023-12-15T17:19:47Z

From the 7-day fullchem benchmark runs shown above, most species have numerical noise differences w/r/t the prior code. This may be because numerical computations (1) may be done in a slightly different order, which may incur roundoff, and (2) Some species will have less unit conversions than others.

yantosca · 2023-12-20T16:28:34Z

As part of this PR, I am making sure that unit conversion is timed separately (i.e. halting other timers before calls to Convert_Spc_Units and starting them again afterwards).

This commit GeosCore/aerosol_mod.F90 GeosCore/airs_ch4_mod.F90 GeosCore/carbon_mod.F90 GeosCore/chemistry_mod.F90 GeosCore/fullchem_mod.F90 GeosCore/gosat_ch4_mod.F90 GeosCore/hco_interface_mod.F90 GeosCore/hco_utilities_mod.F90 GeosCore/linear_chem_mod.F90 GeosCore/mercury_mod.F90 GeosCore/mixing_mod.F90 GeosCore/pbl_mix_mod.F90 GeosCore/rrtmg_rad_transfer_mod.F90 GeosCore/sulfate_mod.F90 GeosCore/tracer_mod.F90 GeosCore/vdiff_mod.F90 GeosCore/wetscav_mod.F90 ObsPack/obspack_mod.F90 - USE Timer_Start and Timer_End from GeosUtil/timers_mod.F90 - Halt the active timer (by calling Timer_End) before each call to unit conversion routine Convert_Spc_Units - Start the halted timer (by calling Timer_Start) after each call to Convert_Spc_Units - This will make sure that the time spent in unit conversions will be logged separately from other timers Interfaces/GCClassic/main.F90 - Halt the Diagnostics timer (with Timer_Start) before each of the AIRS, GOSAT, TCCON observation operators calls - Restart the diagnostics timer (with Timer_End) after each of the AIRS, GOSAT, TCCON observation operators calls CHANGELOG.md - Updated accordingly Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

yantosca · 2023-12-20T21:53:22Z

In 8d05296, we have reorganized calls to timers so that unit conversion is timed separately from other categories. This now results in the following timing output:

All simualtions were run on the Cannon cluster on 48 cores (Intel Cascade Lake CPUs).

1-month carbon simulations

Ref version: geoschem/geos-chem commit `69ae58a`

G E O S - C H E M   T I M E R S
 
  Timer name                       DD-hh:mm:ss.SSS     Total Seconds
---------------------------------------------------------------------------
  GEOS-Chem                     :  00-00:16:56.312          1016.312
  HEMCO                         :  00-00:10:03.796           603.797
  All chemistry                 :  00-00:01:45.750           105.750
  => Gas-phase chem             :  00-00:01:43.250           103.250
  => Photolysis                 :  >>>>> THE TIMER DID NOT RUN <<<<<
  => Aerosol chem               :  >>>>> THE TIMER DID NOT RUN <<<<<
  => Linearized chem            :  >>>>> THE TIMER DID NOT RUN <<<<<
  Transport                     :  00-00:02:46.484           166.484
  Convection                    :  00-00:00:03.156             3.156
  Boundary layer mixing         :  00-00:00:43.593            43.594
  Dry deposition                :  >>>>> THE TIMER DID NOT RUN <<<<<
  Wet deposition                :  >>>>> THE TIMER DID NOT RUN <<<<<
  Diagnostics                   :  00-00:01:20.171            80.172
  Unit conversions              :  00-00:00:09.937             9.938

Dev version: geoschem/geos-chem commit `8d05296`

G E O S - C H E M   T I M E R S
 
  Timer name                       DD-hh:mm:ss.SSS     Total Seconds
---------------------------------------------------------------------------
  GEOS-Chem                     :  00-00:16:11.093           971.094
  HEMCO                         :  00-00:09:43.796           583.797
  All chemistry                 :  00-00:01:19.515            79.516
  => Gas-phase chem             :  00-00:01:19.500            79.500
  => Photolysis                 :  >>>>> THE TIMER DID NOT RUN <<<<<
  => Aerosol chem               :  >>>>> THE TIMER DID NOT RUN <<<<<
  => Linearized chem            :  >>>>> THE TIMER DID NOT RUN <<<<<
  Transport                     :  00-00:02:45.984           165.984
  Convection                    :  00-00:00:02.875             2.875
  Boundary layer mixing         :  00-00:00:35.343            35.344
  Dry deposition                :  >>>>> THE TIMER DID NOT RUN <<<<<
  Wet deposition                :  >>>>> THE TIMER DID NOT RUN <<<<<
  Diagnostics                   :  00-00:01:21.515            81.516
  Unit conversions              :  00-00:00:08.312             8.312

7-day fullchem benchmark simulations:

Ref version: geoschem/geos-chem commit `69ae58a`

G E O S - C H E M   T I M E R S
 
  Timer name                       DD-hh:mm:ss.SSS     Total Seconds
---------------------------------------------------------------------------
  GEOS-Chem                     :  00-00:55:33.500          3333.500
  HEMCO                         :  00-00:09:27.500           567.500
  All chemistry                 :  00-00:23:48.531          1428.531
  => Gas-phase chem             :  00-00:12:48.968           768.969
  => Photolysis                 :  00-00:01:32.218            92.219
  => Aerosol chem               :  00-00:09:02.234           542.234
  => Linearized chem            :  00-00:00:08.781             8.781
  Transport                     :  00-00:03:04.812           184.812
  Convection                    :  00-00:03:58.906           238.906
  Boundary layer mixing         :  00-00:05:53.156           353.156
  Dry deposition                :  00-00:00:12.468            12.469
  Wet deposition                :  00-00:01:32.015            92.016
  Diagnostics                   :  00-00:06:16.000           376.000
  Unit conversions              :  00-00:01:56.640           116.641

Dev version: geoschem/geos-chem commit `8d05296`

G E O S - C H E M   T I M E R S
 
  Timer name                       DD-hh:mm:ss.SSS     Total Seconds
---------------------------------------------------------------------------
  GEOS-Chem                     :  00-00:50:26.390          3026.391
  HEMCO                         :  00-00:08:35.781           515.781
  All chemistry                 :  00-00:20:16.921          1216.922
  => Gas-phase chem             :  00-00:11:03.578           663.578
  => Photolysis                 :  00-00:01:18.250            78.250
  => Aerosol chem               :  00-00:07:21.562           441.562
  => Linearized chem            :  00-00:00:03.343             3.344
  Transport                     :  00-00:03:00.921           180.922
  Convection                    :  00-00:03:58.671           238.672
  Boundary layer mixing         :  00-00:04:44.875           284.875
  Dry deposition                :  00-00:00:07.734             7.734
  Wet deposition                :  00-00:01:14.750            74.750
  Diagnostics                   :  00-00:06:30.062           390.062
  Unit conversions              :  00-00:01:13.703            73.703

yantosca · 2024-04-09T18:45:05Z

@lizziel: Also note, I made sure that the passive species in the restart file was set to a constant value. I used this script to edit the PassiveTracer starting concentration:

#!/usr/bin/env python3

import numpy as np
import xarray as xr

ifile="./GEOSChem.Restart.20190101_0000z.nc4"
ofile="./new.GEOSChem.Restart.20190101_0000z.nc4"

with xr.set_options(keep_attrs=True):

    data = xr.open_dataset(ifile)

    arr = data["SpeciesRst_PassiveTracer"].values
    arr[:,:,:,:] = 1.0e-7

    data["SpeciesRst_PassiveTracer"].values = arr

    data.to_netcdf(ofile)

yantosca · 2024-04-09T18:54:01Z

@lizziel: If I shut off wetdep, I get this output:

===============================================================================
Global mass of PassiveTracer

Ref = alpha.9
Dev = PR_2072
===============================================================================

 Date & Time                  Ref mass [Tg]             Dev mass [Tg]      Abs Diff    % Diff
 -----------------     --------------------      -------------------- -------------  --------
 2019-01-01 00:00          17.6563096646018          17.6563096646018    0.0000e+00     0.000
 2019-01-02 00:00          17.6563097334271          17.6563097334271    0.0000e+00     0.000
 2019-01-03 00:00          17.6563096472464          17.6563096472464    0.0000e+00     0.000
 2019-01-04 00:00          17.6563097381031          17.6563097381031    0.0000e+00     0.000
 2019-01-05 00:00          17.6563096420086          17.6563096420086    0.0000e+00     0.000
 2019-01-06 00:00          17.6563096769969          17.6563096769969    0.0000e+00     0.000
 2019-01-07 00:00          17.6563097517568          17.6563097517568    0.0000e+00     0.000
 2019-01-08 00:00          17.6563096589453          17.6563096589453    0.0000e+00     0.000
 
 Summary                                Ref                       Dev      Abs Diff    % Diff
 -----------------     --------------------      -------------------- -------------  --------
 Maximum mass [Tg]         17.6563097517568          17.6563097517568    0.0000e+00     0.000
 Minimum mass [Tg]         17.6563096420086          17.6563096420086    0.0000e+00     0.000
 Abs diff [g]          109748.2034140284668      109748.2034140284668    0.0000e+00     0.000
 % difference               0.0000006215806           0.0000006215806    0.0000e+00     0.000

 Start mass [Tg]           17.6563096646018          17.6563096646018    0.0000e+00     0.000
 End mass [Tg]             17.6563096589453          17.6563096589453    0.0000e+00     0.000
 Abs diff [g]           -5656.5063744073996       -5656.5063744073996    0.0000e+00    -0.000
 % difference              -0.0000000320367          -0.0000000320367    0.0000e+00    -0.000

 Mean mass [Tg]            17.6563096891357          17.6563096891357    0.0000e+00     0.000
 Variance [Tg]          1.7396662828534e-15       1.7396662828534e-15    0.0000e+00     0.000

yantosca · 2024-04-09T19:22:16Z

@lizziel: I was able to get this output from the compare_diags.py script:

... Creating single-level plots
... Creating zonal mean plots
... Printing totals and differences
... Only showing variables with |absolute difference| > 0.0
Variable               Ref=alpha.9              Dev=PR #2072             Dev - Ref
SpeciesRst_PassiveTracer : 0.02384641             | 0.02384641             | -7.105427357601002e-15

But this may just be numerical noise at the limit of representation.

In this commit, we now make sure to convert all species units (not just advected or wetdep) for TOMAS. This will make sure not to break the internal unit conversions. GeosCore/carbon_mod.F90 - Remove mapping=State_Chm%Map_Advect from calls to Convert_Spc_Units, which will convert all species units GeosCore/chemistry_mod.F90 - Deleted whitespace GeosCore/convection_mod.F90 - Deleted whitespace - Rearranged call to WASHOUT by placing input arguments first, then input/output, then output - For TOMAS only, added fromWetDep=.FALSE. keyword so that the proper unit conversion inside WASHOUT will be applied GeosCore/fullchem_mod.F90 - Update comments - Now call TOMAS routine AQOXID with fromWetDep=.FALSE. so that the proper internal unit conversion will be applied GeosCore/sulfate_mod.F90 - Remove mapping=State_Chm%Map_Advect from calls to Convert_Spc_Units, which will convert all species units - Add an error trap after routine CHEM_SO4_AQ - Now call TOMAS routine AQOXID with fromWetDep=.FALSE. so that the proper internal unit conversion will be applied GeosCore/tomas_mod.F90 - Remove commented out, obsolete unit conversion code - Remove mapping=State_Chm%Map_Advect from calls to Check_Units, which will check if all species have units of in kg. - Added fromWetDep argument to routine AQOXID. If fromWetDep is TRUE then convert units from kg/m2 -> kg. Otherwise, do no unit conversion (as units are already in kg in the calling routine). - Added UnitFactor argument to routine GETFRACTION. The unit conversion factor (either State_Met%AD(I,J,L) or State_Grid%AREA_M2(I,J)) will be passed from the calling program, depending on whether units need to be converted from kg/kg dry or from kg/m2. GeosCore/wetscav_mod.F90 - In routine COMPUTE_F: - Pass State_Met%AD(I,J,L) to GETFRACTION via the UnitFactor argument. This converts kg/kg dry -> kg for TOMAS. - In routine WETDEP: - Convert all species units for TOMAS. - Convert only wetdep species units if we are not using TOMAS. - In routine DO_RAINOUT_ONLY: - Pass State_Grid%Area_M2(I,J) to GETFRACTION via UnitFactor argument. This will convert kg/m2 -> kg for TOMAS. - In routine DO_WASHOUT_ONLY and DO_WASHOUT_AT_SFC, - Arrange arguments to WASHOUT by inputs, input/outputs, and outputs. - Pass fromWetDep=.TRUE. for TOMAS only. - Call AQOXID with fromWetDep=.TRUE. - In routine DO_COMPLETE_REEVAP, - Call AQOXID with fromWetDep-.TRUE. GeosCore/unitconv_mod.F90 - In the ConvertBox_* routines, restore loop over all species - Trimmed trailing whitespace Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

lizziel · 2024-04-11T20:44:56Z

With respect to no diff, looks good to me!

yantosca · 2024-04-11T21:32:55Z

@lizziel @msulprizio: Thanks! I am doing integration test off of alpha.9 (where the branch is now). Then will merge up to dev/14.4.0 (locally) and test after the volcano update.

GeosCore/wetscav_mod.F90 - Adding missing "=" sign in call to WASHOUT at line 4475 - Changed capitalization: "phRain" -> "pHRain" GeosCore/convection_mod.F90 - Changed capitalization: "phRain" -> "pHRain" Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

Interfaces/GCHP/gchp_chunk_mod.F90 - In routine GCHP_CHUNK_RUN: - Add loop variable N - Removed reference to State_Chm%allSpeciesInDryMixingRatio - Now define species units in a loop over 1..nSpecies. This would avoid a compilation error when using a mask to a pointer field. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

@msulprizio

This merge brings the dev/14.4.0 branch as of PR #2246 (Update volcano emissions through 2023, by @msulprizio). This is necesary in order to include various updates and bug fixes from the 14.3.1 release. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

This fixes the issue described in geoschem/geos-chem #2250. run/shared/setupConfigFiles.sh - We now reset EFYO -> CYS for the SPC_ container in HEMCO_Config.rc (which is the container for restart file variables). CHANGELOG.md - Updated accordingly Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

GeosCore/wetscav_mod.F90 - Now define REEVAPSO2 in routines DO_WASHOUT_ONLY and DO_COMPLETE_REEVAP when the units are in kg species. CHANGELOG.md - Updated accordingly Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

@msulprizio

…ory( ) This merge brings the hotfix for the volcano climatology file (by @msulprizio) into the feature/unitconv-per-species-category branch. This should have been merged previously but was omitted due to an oversight. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

GeosCore/wetscav_mod.F90 - Initialize REEVAPSO2 and KMIN in the loop over wetdep species - Define REEVAPSO2 and convert from kg/m2 to kg if necessary - Remove error traps, they seem to be obsolete as as all species are either in kg or kg/m2 for TOMAS Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

yantosca · 2024-04-16T20:52:44Z

I've merged 14.4.0-alpha.12 into the feature/unitconv-per-species-category branch and re-ran integration tests. The GEOS-Chem Classic tests all passed:

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

GCClassic #965e66d GEOS-Chem submodule update: Add hotfix to correct typo in volcano emission file name
GEOS-Chem #49782e849 Make sure REEVAPSO2 is properly defined in kg
HEMCO     #fddcae5 Update version to 3.8.1 for release

Using 24 OpenMP threads
Number of execution tests: 29

Submitted as SLURM job: 28408502
==============================================================================

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

yantosca · 2024-04-22T19:43:17Z

@msulprizio, @lizziel: I ran a set of integration tests w/o the carbon simulation tests. All integration tests passed. We will look into the carbon simulation issues in a separate PR. I think it is OK to merge so as not to hold up the 14.4.0 release.

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

GCHP      #85f9efc GEOS-Chem submodule update: Add hotfix to correct typo in volcano emission file name
GEOS-Chem #49782e849 Make sure REEVAPSO2 is properly defined in kg
HEMCO     #fddcae5 Update version to 3.8.1 for release

Number of execution tests: 6

Submitted as SLURM job: 29233300
==============================================================================
 
Execution tests:
------------------------------------------------------------------------------
gchp_merra2_fullchem................................Execute Simulation....PASS
gchp_merra2_fullchem_benchmark......................Execute Simulation....PASS
gchp_merra2_fullchem_RRTMG..........................Execute Simulation....PASS
gchp_merra2_fullchem_TOMAS15........................Execute Simulation....PASS
gchp_merra2_tagO3...................................Execute Simulation....PASS
gchp_merra2_TransportTracers........................Execute Simulation....PASS
 
Summary of test results:
------------------------------------------------------------------------------
Execution tests passed: 6
Execution tests failed: 0
Execution tests not yet completed: 0

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

yantosca · 2024-04-22T21:39:29Z

Am merging locally up to 14.4.0 and re-running integration tests.

yantosca · 2024-04-23T14:03:02Z

After merging on top of 14.4.0-alpha.12, all GEOS-Chem Classic integration tests passed:

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

GCClassic #a4231d5 GEOS-Chem submod update: Merge PR #2260 (Use "stale" GitHub action)
GEOS-Chem #2f7c05bba Merge PR #2072 (Only convert units for species that require it)
HEMCO     #fddcae5 Update version to 3.8.1 for release

Using 24 OpenMP threads
Number of execution tests: 29

Submitted as SLURM job: 29253161
==============================================================================

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

yantosca · 2024-04-23T14:07:06Z

All GCHP integration tests also passed:

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

GCHP      #ad4667a GEOS-Chem submod update: Merge PR #2260 (Use "stale" GitHub action)
GEOS-Chem #2f7c05bba Merge PR #2072 (Only convert units for species that require it)
HEMCO     #fddcae5 Update version to 3.8.1 for release

Number of execution tests: 11

Submitted as SLURM job: 29252666
==============================================================================
 
Execution tests:
------------------------------------------------------------------------------
gchp_merra2_carbon..................................Execute Simulation....PASS
gchp_merra2_carbon_CH4..............................Execute Simulation....PASS
gchp_merra2_carbon_CO...............................Execute Simulation....PASS
gchp_merra2_carbon_CO2..............................Execute Simulation....PASS
gchp_merra2_carbon_OCS..............................Execute Simulation....PASS
gchp_merra2_fullchem................................Execute Simulation....PASS
gchp_merra2_fullchem_benchmark......................Execute Simulation....PASS
gchp_merra2_fullchem_RRTMG..........................Execute Simulation....PASS
gchp_merra2_fullchem_TOMAS15........................Execute Simulation....PASS
gchp_merra2_tagO3...................................Execute Simulation....PASS
gchp_merra2_TransportTracers........................Execute Simulation....PASS
 
Summary of test results:
------------------------------------------------------------------------------
Execution tests passed: 11
Execution tests failed: 0
Execution tests not yet completed: 0

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

NOTE: The carbon simulation integration tests passed, but runs having CH4 data ran much more slowly than usual. We are currently looking into this.

gopikrishnangs44 · 2024-11-17T04:55:18Z

Hi @yantosca

I am trying to simulate GCAP.

Encountered the same error, as below

HEMCO: Opening ./Restarts/GEOSChem.Restart.20450101_0000z.nc4

HEMCO ERROR: Cannot find field SPC_BUTDI. Please check variable name in the config. file

HEMCO ERROR: Error encountered in routine HCOIO_Read!

HEMCO ERROR: Error in HCOIO_DATAREAD called from HEMCO ReadList_Fill: SPC_BUTDI
 --> LOCATION: ReadList_Fill (HCO_ReadList_Mod.F90)

HEMCO ERROR: Error in called ReadList_Fill (7) from HEMCO ReadList_Read
 --> LOCATION: ReadList_Read (HCO_ReadList_Mod.F90)
 Error in ReadList_Read called from hco_run
===============================================================================
GEOS-Chem ERROR: Error encountered in "HCO_Run"!
 -> at HCOI_GC_Run (in module GeosCore/hco_interface_gc_mod.F90)

THIS ERROR ORIGINATED IN HEMCO!  Please check the HEMCO log file for
additional error messages!
===============================================================================


Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Backtrace for this error:
#0  0x15555363eb4f in ???
#1  0x155554a63169 in sflush
        at /local/gg2995/spack-stage/spack-stage-gcc-10.2.0-ca5qgfppgl4tppua3svba6vehsuvjayp/spack-src/libgfortran/io/unix.h:89
#2  0x456c62 in __emissions_mod_MOD_emissions_run
        at /burg/fiore_new/users/gg2995/test2/gc_2x25_ModelE2.1_fullchem/CodeDir/src/GEOS-Chem/GeosCore/emissions_mod.F90:184
#3  0x4068c4 in geos_chem
        at /burg/fiore_new/users/gg2995/test2/gc_2x25_ModelE2.1_fullchem/CodeDir/src/GEOS-Chem/Interfaces/GCClassic/main.F90:667
#4  0x405556 in main
        at /burg/fiore_new/users/gg2995/test2/gc_2x25_ModelE2.1_fullchem/CodeDir/src/GEOS-Chem/Interfaces/GCClassic/main.F90:32

lizziel · 2024-11-18T18:18:50Z

@gopikrishnangs44, please create a new issue to request help.

gopikrishnangs44 · 2024-11-18T18:31:59Z

HI @lizziel

I figured it out by changing the SPC restart from EY to CYS.

Thank for the response.

yantosca self-assigned this Dec 12, 2023

yantosca added 4 commits December 13, 2023 12:01

yantosca force-pushed the feature/unitconv-per-species-category branch from 31cc208 to d838f7e Compare December 13, 2023 17:02

yantosca added 5 commits December 13, 2023 12:32

Undo numerical optimizations in unitconv_mod.F90

80c3650

GeosUtil/unitconv_mod.F90 - Undo numerical optimizations where we pull terms outside of the IJL loops. This seems to cause some numerical differences. We can return to this later. Signed-off-by: Bob Yantosca <yantosca@seas.harvard.edu>

yantosca changed the base branch from dev/no-diff-to-benchmark to main December 14, 2023 16:28

yantosca modified the milestone: 14.3.1 Dec 20, 2023

yantosca requested review from lizziel and msulprizio December 20, 2023 22:02

yantosca marked this pull request as ready for review December 20, 2023 22:02

yantosca changed the title ~~[WIP] Only convert units for species that require it~~ Only convert units for species that require it Dec 20, 2023

yantosca mentioned this pull request Dec 21, 2023

Implement updates to allow unit conversions for all species or per species #1701

Closed

1 task

yantosca added 3 commits April 12, 2024 11:03

yantosca mentioned this pull request Apr 16, 2024

Out-of-the-Box GEOS-Chem Classic TOMAS simulations fail due to species not found in restart file #2250

Closed

4 tasks

yantosca added 3 commits April 16, 2024 11:45

yantosca linked an issue Apr 16, 2024 that may be closed by this pull request

Out-of-the-Box GEOS-Chem Classic TOMAS simulations fail due to species not found in restart file #2250

Closed

4 tasks

yantosca added the topic: Math and Units Issues pertaining to math expressions and/or unit conversions label Apr 16, 2024

yantosca requested a review from lizziel April 22, 2024 19:01

yantosca mentioned this pull request Apr 22, 2024

Add missing gas postmeter CH4 emissions for the GHGI express extension inventory #2261

Merged

yantosca requested a review from msulprizio April 22, 2024 19:44

yantosca added this to the 14.4.0 milestone Apr 23, 2024

yantosca merged commit 2f7c05b into dev/14.4.0 Apr 23, 2024

yantosca deleted the feature/unitconv-per-species-category branch April 23, 2024 15:59

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Only convert units for species that require it #2072

Only convert units for species that require it #2072

yantosca commented Dec 12, 2023 •

edited

Loading

yantosca commented Dec 12, 2023

yantosca commented Dec 15, 2023 •

edited

Loading

yantosca commented Dec 15, 2023 •

edited

Loading

yantosca commented Dec 20, 2023

yantosca commented Dec 20, 2023

yantosca commented Apr 9, 2024

yantosca commented Apr 9, 2024

yantosca commented Apr 9, 2024

lizziel commented Apr 11, 2024

yantosca commented Apr 11, 2024

yantosca commented Apr 16, 2024

yantosca commented Apr 22, 2024

yantosca commented Apr 22, 2024

yantosca commented Apr 23, 2024

yantosca commented Apr 23, 2024

gopikrishnangs44 commented Nov 17, 2024

lizziel commented Nov 18, 2024

gopikrishnangs44 commented Nov 18, 2024

Only convert units for species that require it #2072

Only convert units for species that require it #2072

Conversation

yantosca commented Dec 12, 2023 • edited Loading

Name and Institution (Required)

Confirm you have reviewed the following documentation

Describe the update

Expected changes

Related Github Issue(s)

yantosca commented Dec 12, 2023

yantosca commented Dec 15, 2023 • edited Loading

yantosca commented Dec 15, 2023 • edited Loading

yantosca commented Dec 20, 2023

yantosca commented Dec 20, 2023

1-month carbon simulations

Ref version: geoschem/geos-chem commit 69ae58a

Dev version: geoschem/geos-chem commit 8d05296

7-day fullchem benchmark simulations:

Ref version: geoschem/geos-chem commit 69ae58a

Dev version: geoschem/geos-chem commit 8d05296

yantosca commented Apr 9, 2024

yantosca commented Apr 9, 2024

yantosca commented Apr 9, 2024

lizziel commented Apr 11, 2024

yantosca commented Apr 11, 2024

yantosca commented Apr 16, 2024

yantosca commented Apr 22, 2024

yantosca commented Apr 22, 2024

yantosca commented Apr 23, 2024

yantosca commented Apr 23, 2024

gopikrishnangs44 commented Nov 17, 2024

lizziel commented Nov 18, 2024

gopikrishnangs44 commented Nov 18, 2024

yantosca commented Dec 12, 2023 •

edited

Loading

yantosca commented Dec 15, 2023 •

edited

Loading

yantosca commented Dec 15, 2023 •

edited

Loading

Ref version: geoschem/geos-chem commit `69ae58a`

Dev version: geoschem/geos-chem commit `8d05296`

Ref version: geoschem/geos-chem commit `69ae58a`

Dev version: geoschem/geos-chem commit `8d05296`