GEOS-Chem v9-01-03 benchmark history

From Geos-chem
Jump to: navigation, search

On this page we have posted complete information about all benchmark simulations (both 1-month and 1-year) for GEOS-Chem v9-01-03.

1-year benchmarks

v9-01-03-Release

Run0

Three GEOS-Chem model versions were compared to each other:

Color Quantity Plotted Met Type Anthro
emissions
Biomass
emissions
Wet deposition Stratospheric chemistry mechanism Linoz Annual Mean OH
[105 molec/cm3]
Red v9-01-02-Run1 GEOS-5, 4x5 Same as v9-01-02-Run0
+RETRO anthropogenic VOC emissions
Same as v9-01-02-Run0
+ GFED3 monthly inventory
Same as v9-01-01-Run0
(Original algorithm described in Jacob et al [2000]
+ Allow both washout and rainout when precip. forms
+ Updates for MERRA met fields)
Simple linearized stratospheric chemistry scheme ON 12.332
Green v9-01-03e-Run0 GEOS-5, 4x5 " " " " + Improved snow scavenging and washout parameterization " " ON 11.935
Blue v9-01-03r-Run0 GEOS-5, 4x5 + C2H6 emissions from offline simulation
+ PARANOX ship plume emissions
" " " " Updated linearized stratospheric chemistry scheme ON 11.858
Black Observations              


We have archived the files for v9-01-03-Release_Run0 into TARBALL (i.e. tar.gz) format to save disk space. To download the TARBALL containing all relevant files, type:

ftp ftp.as.harvard.edu
get gcgrid/geos-chem/1yr_benchmarks/v9-01-03/v9-01-03-Release_Run0.tar.gz

And then to extract the data, type:

tar xvzf v9-01-03-Release_Run0.tar.gz

The comparison plots will be in located in the output/pdf folder.

--Melissa Payer 15:56, 7 September 2012 (EDT)
--Bob Yantosca (talk) 20:55, 20 June 2017 (UTC)

Run1

Three GEOS-Chem model versions were compared to each other:

Color Quantity Plotted Met Type Anthro
emissions
Wet deposition Tropospheric chemistry mechanism Stratospheric chemistry mechanism Linoz Annual Mean OH
[105 molec/cm3]
Red v9-01-03e-Run0 GEOS-5, 4x5 Same as v9-01-02-Run1 Same as v9-01-01-Run0
+ Improved snow scavenging and washout parameterization
Same as v8-02-01-Run0 " " ON 11.935
Green v9-01-03r-Run0 GEOS-5, 4x5 + C2H6 emissions from offline simulation
+ PARANOX ship plume emissions
" " " " Updated linearized stratospheric chemistry scheme ON 11.858
Blue v9-01-03r-Run1 GEOS-5, 4x5 " " " " + Bromine chemistry mechanism " " ON 11.356
Black Observations              


We have archived the files for v9-01-03-Release-Run1 into TARBALL (i.e. tar.gz) format to save disk space. To download the TARBALL containing all relevant files, type:

ftp ftp.as.harvard.edu
get gcgrid/geos-chem/1yr_benchmarks/v9-01-03/v9-01-03-Release-Run1.tar.gz

And then to extract the data, type:

tar xvzf v9-01-03-Release-Run1.tar.gz

The comparison plots will be in located in the output/pdf folder.

--Melissa Payer 15:56, 7 September 2012 (EDT)
--Bob Yantosca (talk) 20:57, 20 June 2017 (UTC)

v9-01-03o

Run0

NOTE: This version was not approved by the GCSC, but subjected to further investigation due to a larger effect of bromine chemistry on ozone and OH than reported in Parrella et al. (2012). New 1-year benchmark simulations were run for v9-01-03-Release as reported above.

Three GEOS-Chem model versions were compared to each other:

Color Quantity Plotted Met Type Anthro
emissions
Biomass
emissions
Wet deposition Tropospheric chemistry mechanism Stratospheric chemistry mechanism Linoz Annual Mean OH
[105 molec/cm3]
Red v9-01-02-Run1 GEOS-5, 4x5 Same as v9-01-02-Run0
+RETRO anthropogenic VOC emissions
Same as v9-01-02-Run0
+ GFED3 monthly inventory
Same as v9-01-01-Run0
(Original algorithm described in Jacob et al [2000]
+ Allow both washout and rainout when precip. forms
+ Updates for MERRA met fields)
Same as v8-02-01-Run0 Simple linearized stratospheric chemistry scheme ON 12.332
Green v9-01-03e-Run0 GEOS-5, 4x5 " " " " + Improved snow scavenging and washout parameterization " " " " ON 11.935
Blue v9-01-03o-Run0 GEOS-5, 4x5 + C2H6 emissions from offline simulation
+ PARANOX ship plume emissions
" " " " + Bromine chemistry mechanism Updated linearized stratospheric chemistry scheme ON 11.143
Black Observations                


The output plots for Run0 (both PostScript and PDF format) may be downloaded from:

ftp ftp.as.harvard.edu
cd gcgrid/geos-chem/1yr_benchmarks/v9-01-03/v9-01-03o/Run0/output/ps
mget *
cd gcgrid/geos-chem/1yr_benchmarks/v9-01-03/v9-01-03o/Run0/output/pdf
mget *

You may also view the PDF files online by pointing your browser to

ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1yr_benchmarks/v9-01-03/v9-01-03o/Run0/output/pdf/

--Melissa Payer 11:43, 23 July 2012 (EDT)

v9-01-03e

Run0

Three GEOS-Chem model versions were compared to each other:

Color Quantity Plotted Met Type Anthro
emissions
Biofuel
emissions
Biomass
emissions
Wet deposition Linoz Annual Mean OH
[105 molec/cm3]
Red v9-01-02-Run0 GEOS-5, 4x5 Same as v9-01-01-Run0
+ Fix for GEIA emissions scaling factor
+ Imposed seasonality on Streets 2000 NH3 emissions
+ Bug fix in ship NOx emissions
Same as v9-01-01-Run0 + Bug fix for Streets biofuel emissions
Same as v9-01-01-Run0
(GFED2 monthly inventory)
Same as v9-01-01-Run0
(Original algorithm described in Jacob et al [2000]
+ Allow both washout and rainout when precip. forms
+ Updates for MERRA met fields)
ON 12.233
Green v9-01-02-Run1 GEOS-5, 4x5 + RETRO anthropogenic VOC emissions " " + GFED3 monthly inventory " " ON 12.332
Blue v9-01-03e-Run0 GEOS-5, 4x5 " " " " " " + Improved snow scavenging and washout parameterization ON 11.935
Black Observations              


We have archived the files for v9-01-03e_Run0 into TARBALL (i.e. tar.gz) format to save disk space. To download the TARBALL containing all relevant files, type:

ftp ftp.as.harvard.edu
get gcgrid/geos-chem/1yr_benchmarks/v9-01-03/v9-01-03e_Run0.tar.gz

And then to extract the data, type:

tar xvzf v9-01-03e_Run0.tar.gz

The comparison plots will be in located in the output/pdf folder. /v9-01-03-Release-Run1.tar.gz

--Melissa Payer 13:19, 1 February 2012 (EST)
--Bob Yantosca (talk) 20:58, 20 June 2017 (UTC)

Rn-Pb-Be simulation

We also conducted a 1-year Rn-Pb-Be simulation with GEOS-Chem v9-01-03e. Please follow the links below to see the budget of Pb210 and Be7 from 1-year Rn-Pb-Be simulations for v9-01-03e and previous versions:

  1. Budget of Pb210 from 1-year benchmark simulations
  2. Budget of Be7 from 1-year benchmark simulations

Plots comparing modeled Rn222, Pb210, and Be7 to GEOS-Chem v9-01-02 are now available. The plotting programs were provided by Hongyu Liu. For more information on the plots, see Liu et al. [2001].


We have archived the files for v9-01-03e_RnPbBe into TARBALL (i.e. tar.gz) format to save disk space. To download the TARBALL containing all relevant files, type:

ftp ftp.as.harvard.edu
get gcgrid/geos-chem/1yr_benchmarks/v9-01-03/v9-01-03e_RnPbBe.tar.gz

And then to extract the data, type:

tar xvzf v9-01-03e_RnPbBe.tar.gz

The comparison plots will be in located in the output/pdf folder.

--Melissa Payer 16:48, 24 January 2012 (EST)
--Bob Yantosca (talk) 20:59, 20 June 2017 (UTC)

1-month benchmarks

v9-01-03-Fix_for_IFORT12

Here is the assessment form for 1-month benchmark simulation v9-01-03-Fix_for_IFORT12. This unofficial benchmark included last-minute updates and various post-release patches to GEOS-Chem v9-01-03.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri Oct 12 13:14:55 EDT 2012
Performance statistics:
  • Ran on 8 CPUs of bench@titan-08.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:19
  • CPU time / wall time: 6.5073
  • % of ideal performance: 81.34%
Compared to previous benchmark: v9-01-03r
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify): SSA debromination update will affect Br2 emissions. We expect to see no change from the other post-release patches.
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03-Fix_for_IFORT12.tar.gz
Metrics
Global mean OH (from log file): 12.8643920212567 x 105 molec/cm3
Methyl chloroform lifetime: 4.9020 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: NOx, Ox, PAN, ISOP, HNO3, H2O2, ALD2, RCHO, MVK, MACR, PMN, PPN, PRPE, CH2O, N2O5, HNO4, MP, DMS, SO2, SO4, SO4s, MSA, NIT, NITs, Br2, Br, BrO, HOB4, HBr, BrNO2, BrNO3, OH
Comments on SURFACE differences: Concentrations of bromine species are reduced over the oceans due to the use of size-dependent depletion factors in sea salt aerosol debromination. Changes in other species occur in regions where tracer concentrations are low, resulting in large % differences.
At 500 hPa, list all species that changed by 10% or more: NOx, ISOP, ALD2, PMN, PRPE, N2O5, HNO4, DMS, SO2, Br2, Br, BrO, HOB4, HBr, BrNO2, BrNO3, OH
Comments on 500 hPa differences: Changes at the surface propagate into the free troposphere.
In the ZONAL MEAN differences, list all species that changed by 10% or more: NOx, Ox, PAN, CO, ALK4, HNO3, H2O2, ACET, MEK, ALD2, RCHO, MVK, MACR, PMN, PPN, R4N2, PRPE, CH2O, C2H6, HNO4, MP, DMS, SO2, SO4, SO4s, MSA, NH4, NITs, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, CHBr3, CH2Br2, CH3Br, OH
Comments on ZONAL MEAN differences: See above
In the EMISSION RATIO maps, list all species that changed by 10% or more: Br2
Comments on EMISSION RATIO differences: Emissions of Br2 are reduced due to changes to sea salt aerosol debromination
Additional or summary comments:

--Melissa Payer 11:57, 15 October 2012 (EDT)

v9-01-03r

Here is the assessment form for 1-month benchmark simulation v9-01-03r.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Mon Aug 27 17:48:22 EDT 2012
Performance statistics:
  • Ran on 8 CPUs of titan-04 (2.792 GHz x 16 CPU)
  • Wall time: 01:35
  • CPU time / wall time: 6.522
  • % of ideal performance: 81.53%
Compared to previous benchmark: v9-01-03q
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify): Differences should only be at the level of numerical noise.
Plots may be downloaded from: http://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03.tar.gz
Metrics
Global mean OH (from log file): 12.8111553738048 x 105 molec/cm3
Methyl chloroform lifetime: 4.9276 years
Did either of these change by more than 5%? No. The mean OH differs only by 0.0004%, and the MCF lifetimes are identical.
At the SURFACE, list all species that changed by 10% or more: None
Comments on SURFACE differences:
At 500 hPa, list all species that changed by 10% or more: None
Comments on 500 hPa differences:
In the ZONAL MEAN differences, list all species that changed by 10% or more: ALK4, PRPE, C3H8, SO4s, NH3, NIT, NITs, BCPO, OCPO,
Comments on ZONAL MEAN differences: Differences occur near the top of the atmosphere. These appear to be small-number differences. We attribute this to ISORROPIA II.
In the EMISSION RATIO maps, list all species that changed by 10% or more: None
Comments on EMISSION RATIO differences:
Additional or summary comments: Differences in tracers appear to be due to ISORROPIA II and due to the elimination of numerical noise in the regridding.
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 28 Aug 2012

--Bob Y. 10:32, 28 August 2012 (EDT)

v9-01-03q

Here is the assessment form for 1-month benchmark simulation v9-01-03q.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Tue Aug 21 22:09:21 EDT 2012
Performance statistics:
  • Ran on 8 CPUs of titan-01 (2.925 GHz x 16 CPU)
  • Wall time: 3:49
  • CPU time / wall time: 4.8670
  • % of ideal performance: 60.84%
Compared to previous benchmark: v9-01-03p
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify): The bug fix impacts the regridding of regional anthro emissions (EMEP, CAC, BRAVO, NEI05, Streets).
Plots may be downloaded from: http://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03q.tar.gz
Metrics
Global mean OH (from log file): 12.8111028654046 x 105 molec/cm3
Methyl chloroform lifetime: 4.9276 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: None
Comments on SURFACE differences:
At 500 hPa, list all species that changed by 10% or more: None
Comments on 500 hPa differences:
In the ZONAL MEAN differences, list all species that changed by 10% or more: ALK4, ISOP, MVK, MACR, PRPE, C3H8, N2O5, SO4s, NH3, NIT, NITs, DST1, DST2, DST3, SALC, Br2
Comments on ZONAL MEAN differences:
  • SO4s, NH3, NIT, and NITs differences are caused by ISORROPIA II.
  • Differences in other species appear to be small number differences near the top of the atmosphere. These appear also to be caused by the ISORROPIA II package.
In the EMISSION RATIO maps, list all species that changed by 10% or more: None
Comments on EMISSION RATIO differences:
Additional or summary comments:
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 22 Aug 2012

--Melissa Payer 14:58, 22 August 2012 (EDT)

v9-01-03p

Here is the assessment form for 1-month benchmark simulation v9-01-03p.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri Aug 17 16:19:07 2012
Performance statistics:
  • Ran on 8 CPUs of titan-08.as.harvard.edu
  • Wall time: 04:39
  • CPU time / wall time: 6.390
  • % of ideal performance: 79.88%
Compared to previous benchmark: v9-01-03o
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify): We should see no significant scientific changes. Only structural modifications were made.
Plots may be downloaded from: http://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03p.tar.gz
Metrics
Global mean OH (from log file): 12.8110847670390 x 105 molec/cm3
Methyl chloroform lifetime: 4.9276 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: None
Comments on SURFACE differences:
At 500 hPa, list all species that changed by 10% or more: None
Comments on 500 hPa differences:
In the ZONAL MEAN differences, list all species that changed by 10% or more: ALK4, C3H8, SO4s, NH3, NIT, NITs, SALC
Comments on ZONAL MEAN differences:
  • SO4s, NH3, NITs, differences are caused due to the ISORROPIA II ATE package.
  • Differences in ALK4, C3H8, SALC appear to be small number differences near the top of the atmosphere. These appear also to be caused by the ISORROPIA II package.
In the EMISSION RATIO maps, list all species that changed by 10% or more: None
Comments on EMISSION RATIO differences:
Additional or summary comments:
  • Differences appear to be minor.
  • We also performed benchmarks for v9-01-03o and v9-01-03p with ISORROPIA II turned off, named v9-01-03o.NoIso and v9-01-03p.NoIso.
  • The binary punch files for v9-01-03o.NoIso and v9-01-03p.NoIso are 100% identical (determined with the Unix diff command).
  • The frequency distribution plot for v9-01-03p.NoIso shows 100% of all tracers in the central histogram bar.
  • Therefore we can attribute the differences between the benchmark simulations v9-01-03o and v9-01-03p to ISORROPIA II.
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 21 Aug 2012

--Bob Y. 10:36, 21 August 2012 (EDT)

v9-01-03o

Here is the assessment form for 1-month benchmark simulation v9-01-03o.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Lee Murray (Harvard)
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Wed Jul 11 13:23:00 EDT 2012
Performance statistics:
  • Ran on 8 CPUs of titan-02.as.harvard.edu (2.925 GHz x 16 CPU)
  • Wall time: 2:14
  • CPU time / wall time: 6.7995
  • % of ideal performance: 84.99%
Compared to previous benchmark: v9-01-03n
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify):
Plots may be downloaded from: http://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03o.tar.gz
Metrics
Global mean OH (from log file): 12.8111028654046 x 105 molec/cm3
Methyl chloroform lifetime: 4.9276 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: All tracers somewhere except for CHBr3, CH2Br2, CH3Br, and CO.
Comments on SURFACE differences: See summary comments.
At 500 hPa, list all species that changed by 10% or more: All tracers somewhere except for CHBr3, CH2Br2, CH3Br, CO, and Ox.
Comments on 500 hPa differences: See summary comments.
In the ZONAL MEAN differences, list all species that changed by 10% or more: All tracers somewhere.
Comments on ZONAL MEAN differences: See summary comments.
In the EMISSION RATIO maps, list all species that changed by 10% or more: Acetone ocean sink.
Comments on EMISSION RATIO differences: Assumably dependent on tropospheric acetone burden, which has decreased.
Additional or summary comments: NOTE: This benchmark simulation used a new restart file, spun up for the stratospheric P and k update for over 10 years (provided by Lee Murray). Plots for the 1-month benchmark simulation using the previous restart file are located at:

ftp://ftp.as.harvard.edu/pub/exchange/mpayer/1mo_quick_look/v9-01-03o.old_restart/

Summary of changes. Stratospheric production rates and loss frequencies are updated for the first time since 2002.

NOx, PAN, CO, ALK4, ISOP, HNO3, H2O2, ACET, MEK, ALD2, RCHO, MVK, MACR, PMN, PPN, R4N2, PRPE, C3H8, CH2O, C2H6, N2O5, HNO4, MP

Stratospheric concentrations are substantially different, and now match those from the GMI Combo model's fully resolved stratosphere, as shown in these plots. The subsequent changes in strat-trop exchange fluxes alter tropospheric composition.

Ox

A bug fix for Linoz has slightly changed stratospheric ozone. (Linoz was previously solved over the chemistry time step each dynamic time step).

DMS, SO2, SO4, SO4s, MSA, NH3, NH4, NIT, NITs, BCPI, OCPI, BCPO, OCPO, DST1, DST2, DST3, DST4, SALA, SALC, BrNO2

These species do not have any prod/loss rates applied themselves, but are affected by

  • (1) Secondary effects of the above new strat-trop fluxes. Most changes are in remote marine and polar regions, and absolute changes are small.
  • (2) Some species were not at equilibrium with respect to STE in the original restart file. Dust, sea salt, and BC/OC all had erroneously high stratospheric burdens that fell out even without the new stratospheric prod/loss rates.

Br2, Br, BrO, HOBr, HBr, BrNO3

These species are prescribed in the stratosphere and should be identical to v9-01-03n. However, there are differences, most notable at the poles. I suspect a bug in the original implementation that was corrected when the code was changed from SCHEM.f to strat_chem_mod.F90. As shown on pages 44-47, the v9-01-03o stratosphere in the Antarctic now matches the nighttime value and in the Arctic matches the daytime value, as it should. Tropospheric values reflect these changes in stratosphere, and changes in their reactants.

CHBr3, CH2Br2, CH3Br

These gases are emitted from the surface and are lost by reaction with OH in the stratosphere. The GMI OH fields are lower than the old 2D zonal OH fields.

OH

Troposphere reflects changes from strat-trop fluxes. New stratospheric OH fields from GMI are lower.

Approval
Requires further investigation: No
Approved: Yes
Date of approval: 11 Jul 2012

--Melissa Payer 09:43, 11 July 2012 (EDT)

v9-01-03n

Here is the assessment form for 1-month benchmark simulation v9-01-03n.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • OMI NOx: Lok Lamsal (NASA GSFC)
  • MAP_A2A fixes: Lee Murray (Harvard)
  • Dicarbonyl bug fixes: GEOS-Chem Support Team
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Thu Jun 7 21:05:35 EDT 2012
Performance statistics:
  • Ran on 8 CPUs of titan-02.as.harvard.edu (2.925 GHz x 16 CPU)
  • Wall time: 3:09
  • CPU time / wall time: 4.7143
  • % of ideal performance: 58.92%
Compared to previous benchmark: v9-01-03m
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify): We expect to see no changes from v9-01-03m
Plots may be downloaded from: http://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03n.tar.gz
Metrics
Global mean OH (from log file): 12.6219171815788 x 105 molec/cm3
Methyl chloroform lifetime: 4.9629 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: NOx, PAN, ALK4, ISOP, HNO3, MEK, ALD2, RCHO, MVK, MACR, PMN, PPN, R4N2, PRPE, CH2O, N2O5, HNO4, MP, DMS, SO2, SO4, SO4s, NH3, NH4, NIT, NITs, OCPI, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, OH
Comments on SURFACE differences: Small numerical noise where concentrations are low cause large percent differences. Absolute differences are generally small. Larger differences are caused by slight shifts in the location of emissions resulting from bug fixes in the MAP_A2A regridding algorithm.
At 500 hPa, list all species that changed by 10% or more: NOx, ALK4, ISOP, MEK, ALD2, RCHO, MVK, MACR, PMN, PRPE, N2O5, HNO4, DMS, NH3, NIT, Br2, Br, BrO, HOBr, BrNO2, BrNO3, OH
Comments on 500 hPa differences: Small numerical noise where concentrations are low cause large percent differences. Absolute differences are generally small.
In the ZONAL MEAN differences, list all species that changed by 10% or more: ALK4, ISOP, ALD2, RCHO, MVK, MACR, PMN, PRPE, N2O5, DMS, SO4s, NH3, NIT, Br, BrO
Comments on ZONAL MEAN differences: Small numerical noise where concentrations are low cause large percent differences. Absolute differences are generally small.
In the EMISSION RATIO maps, list all species that changed by 10% or more: ACET, ALD2, ALK4, CO, ISOP, MEK, NH3, NOx, OC, PRPE, SO2, SO4
Comments on EMISSION RATIO differences: Differences are due to slight shifts in location of some emissions resulting from bug fixes in the MAP_A2A regridding algorithm.
Additional or summary comments:
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 07 Jun 2012

--Melissa Payer 10:43, 8 June 2012 (EDT)

v9-01-03m

Here is the assessment form for 1-month benchmark simulation v9-01-03m.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Thu May 31 19:29:08 EDT 2012
Performance statistics:
  • Ran on 8 CPUs of terra-06-c.as.harvard.edu (2.792 GHz x 16 CPU)
  • Wall time: 1:46
  • CPU time / wall time: 7.7233
  • % of ideal performance: 96.54%
Compared to previous benchmark: v9-01-03l
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify):
Plots may be downloaded from: http://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03m.tar.gz
Metrics
Global mean OH (from log file): 12.6371702448654 x 105 molec/cm3
Methyl chloroform lifetime: 4.9566 years
Did either of these change by more than 5%? No.

Global mean OH change: -3.66%
MCL change: +4.11%

At the SURFACE, list all species that changed by 10% or more: NOx, PAN, N2O5, HNO4, Ox, Alk4, isoprene, H2O2, ALD2, CH2O, DMS, NIT, MP, MVK, MACR, PMN, PPN, SO2, and SO4.
Comments on SURFACE differences: NOx (but only over the ocean, where concentrations were small), surface PAN over oceans, N2O5 over oceans, HNO4 over oceans, Ox over oceans at latitudes above 45N (but global means are <10% consistent with v8-02-02), Alk4 increases over oceans at lats above 45N (consistent with decreases in OH and Ox), isoprene increases exceed 10% in remote regions (in response to OH decreases), H2O2 decreases exceed 10% in remote regions, ALD2 and CH2O decreases exceed 10% in remote regions (dominant reaction partners with Br producing HBr), DMS increases exceed 10% (mostly in N extratropics, tracking decreased OH), and NIT and MP in the remote northern extratropics. Also, SO2 increases and SO4 decreases by >10% in the Arctic due to lower OH. The MVK and MACR increases are in response to lower OH. PMN and PPN decreases are >10% in remote regions due to NOx decreases.
At 500 hPa, list all species that changed by 10% or more: NOx (extratropics), N2O5, NH3, HNO4, isoprene, ALD2 in Southern extratropics only (low concentrations), PMN, PRPE, DMS, NIT, SO2.
Comments on 500 hPa differences: NIT changes are just noise where concentrations are near 0, NH3 increased over the antarctic where concentrations are near 0. N2O5 decreases in remote regions globally but also increases North of 80N, which is not seen in the surface simulation; this is happening because concentrations of N2O5 are near 0 there. Changes in the other listed species are consistent with explanations given for the surface difference plots.
In the ZONAL MEAN differences, list all species that changed by 10% or more: NOx in the extratropics (low concentrations), Ox in the Arctic PBL, PAN in the PBL and remote FT (southern extratropics), Alk4 in the northern hemisphere (due to OH decrease), global isoprene (due to OH decrease), H2O2 mostly in the northern hemisphere PBL, PPN in the PBL and remote FT (southern extratropics), MP increases in the Arctic PBL due to lower OH. Other species that change >10% in the zonal mean due so where there are near-0 starting concentrations.
Comments on ZONAL MEAN differences:
In the EMISSION RATIO maps, list all species that changed by 10% or more: No emissions changed other than the new bromine emissions that were introduced: sea salt Br2, CHBr3, CH2Br2, and fixed concentrations of CH3Br in the PBL based on CMDL surface observations.
Comments on EMISSION RATIO differences:
Additional or summary comments: The 1-month benchmark output looks good and in line with what's expected from the v8-02-02 bromine simulation. I think that it should be approved - Justin Parrella
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 06 Jun 2012

--Melissa Payer 13:21, 21 May 2012 (EDT)

v9-01-03l

Here is the assessment form for 1-month benchmark simulation v9-01-03l.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Prasad Kasibhatla (Duke)
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri May 4 19:40:32 EDT 2012
Performance statistics:
  • Ran on 8 CPUs of terra-11-c.as.harvard.edu (2.660 GHz x 8 CPU)
  • Wall time: 1:43
  • CPU time / wall time: 7.6201
  • % of ideal performance: 95.25%
Compared to previous benchmark: v9-01-03k
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify):
Plots may be downloaded from http://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03l.tar.gz
Metrics
Global mean OH (from log file): 13.1174291218146 x 105 molec/cm3
Methyl chloroform lifetime: 4.7610 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: Various species at discrete locations
Comments on SURFACE differences: Related to change in code to regrid on the fly from native GFED3 resolution to model grid
At 500 hPa, list all species that changed by 10% or more: A few species at a few locations
Comments on 500 hPa differences: Some related to changes in surface; a bit puzzled by changes to a few species like NH3, NIT, and ISOP (one grid box over the Amazon), but likely related to changes to chemistry at surface caused by regridding change (ISOP) or in regions with very low concentratiomns (eg NIT)
In the ZONAL MEAN differences, list all species that changed by 10% or more: ISOP, MVK, MACR, N2O5 at isolated latitudes, but more widespread changes in NIT and NH3
Comments on ZONAL MEAN differences: I think the changes in ISOP, MVK, etc are related to changes in chemistry caused by changes in regridding, but am a bit puzzled by NIT and NH3 changes.
In the EMISSION RATIO maps, list all species that changed by 10% or more: Some species at discrete locations
Comments on EMISSION RATIO differences: Due to changes associated with regridding on the fly from native GFED3 resolution
Additional or summary comments: Note: Special benchmarks were also requested for (1) 3-hourly and (2) daily GFED3 biomass burning emissions. The benchmark plots for these simulations may be viewed at:
  1. 3-hourly GFED3: ftp://ftp.as.harvard.edu/pub/exchange/mpayer/1mo_quick_look/v9-01-03l.3h_GFED3/
  2. Daily GFED3: ftp://ftp.as.harvard.edu/pub/exchange/mpayer/1mo_quick_look/v9-01-03l.daily_GFED3/

Differences are consistent with Daily and 3-hourly variation in emisisons

Approval
Requires further investigation: No
Approved: Yes
Date of approval: 18 May, 2012

--Melissa Payer 17:33, 9 May 2012 (EDT)

v9-01-03k

Here is the assessment form for 1-month benchmark simulation v9-01-03k.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Matt Cooper (Dalhousie)
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Wed Apr 25 14:48:08 EDT 2012
Performance statistics:
  • Ran on 8 CPUs of terra-05-c.as.harvard.edu (2.793 GHz x 16 CPU)
  • Wall time: 1:27
  • CPU time / wall time: 7.6049
  • % of ideal performance: 95.06%
Compared to previous benchmark: v9-01-03j
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify): We expect to see no changes from v9-01-03j
Plots may be downloaded from: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03k.tar.gz
Metrics
Global mean OH (from log file): 13.1180503320529 x 105 molec/cm3
Methyl chloroform lifetime: 4.7607 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: NOx, PAN, ALK4, ISOP, HNO3, MEK, ALD2, RCHO, MVK, MACR, PMN, PPN, R4N2, PRPE, C3H8, CH2O, N2O5, HNO4, MP, DMS, SO2, SO4, SO4s, NH3, NH4, NIT, NITs, BCPI, OCPI, BCPO, OCPO, OH
Comments on SURFACE differences: For most species, small numerical noise where concentrations are low cause large percent differences. Absolute differences are generally small. Other species (ie HNO3, NOx) with large differences generally have a grid box with positive change next to box with similar negative change. This is due to slight shifts in location of some emissions under new regrid algorithm.
At 500 hPa, list all species that changed by 10% or more: ISOP, MEK, RCHO, MVK, MACR, PMN, PRPE, N2O5, DMS, SO2, SO4, NH3, NIT, BCPO, OCPO
Comments on 500 hPa differences: Small numerical noise where concentrations are low cause large percent differences. Absolute differences are generally small.
In the ZONAL MEAN differences, list all species that changed by 10% or more: ALK4, ISOP, MVK, MACR, PRPE, N2O5, NH3,NIT, OCPO
Comments on ZONAL MEAN differences: Small numerical noise where concentrations are low cause large percent differences. Absolute differences are generally small.
In the EMISSION RATIO maps, list all species that changed by 10% or more: ACET, ALD2, ALK4, BC, C3H8, CH2O, ISOP, MEK, NH3, NOx, OC, PRPE, SO2, SO4
Comments on EMISSION RATIO differences: Small numerical noise where concentrations are low cause large percent differences. Absolute differences are generally small.
Additional or summary comments:
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 27 April 2012

--Melissa Payer 14:45, 26 April 2012 (EDT)

v9-01-03j

Here is the assessment form for 1-month benchmark simulation v9-01-03j.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri Apr 13 15:02:26 EDT 2012
Performance statistics:
  • Ran on 8 CPUs of terra-10-c.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 1:42
  • CPU time / wall time: 7.6407
  • % of ideal performance: 95.51%
Compared to previous benchmark: v9-01-03i
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify):
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03j.tar.gz
Metrics
Global mean OH (from log file): 13.1149651381533 x 105 molec/cm3
Methyl chloroform lifetime: 4.7612 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: ISOP, MVK, MACR
Comments on SURFACE differences: Differences are seen in ISOP in locations where concentrations are already very low (i.e. over oceans). MVK and MACR differences are presumably tied ot to the differences in ISOP. As described in this wiki post, we attribute these changes to the numerical differences caused by regridding leaf area index data from 0.5° x 0.5° native resolution (instead of from 1° x 1°, as was done in the past).
At 500 hPa, list all species that changed by 10% or more: ISOP, MVK, MACR, NH3, NIT
Comments on 500 hPa differences: Differences are seen in ISOP in locations where concentrations are already very low (i.e. over oceans). MVK and MACR differences are presumably tied to to the differences in ISOP. As described in this wiki post, we attribute these changes to the numerical differences caused by regridding leaf area index data from 0.5° x 0.5° native resolution (instead of from 1° x 1°, as was done in the past).

Differences in NH3 and NIT are attributed to the Numerical noise issue in ISORROPIA II.

In the ZONAL MEAN differences, list all species that changed by 10% or more: ISOP, NH3, NIT
Comments on ZONAL MEAN differences: ISOP shows % differences of about ~5% in places where concentrations are already very low. As described in this wiki post, we attribute these changes to the numerical differences caused by regridding leaf area index data from 0.5° x 0.5° native resolution (instead of from 1° x 1°, as was done in the past).

Differences in NH3 and NIT are attributed to the Numerical noise issue in ISORROPIA II.

In the EMISSION RATIO maps, list all species that changed by 10% or more: Biogenic ACET, Biogenic ISOP, Biogenic PRPE
Comments on EMISSION RATIO differences: All of the above-mentioned species show differences, especially near coastlines. All of these species are computed by the MEGAN biogenic emissions module, which now takes in the leaf area indices regridded from 0.5° x 0.5°, As described in this wiki post. We attribute the differences in these species to the numerical differences in the leaf area index data.
Additional or summary comments: The updates added to v9-01-03j were made to primarily assist development of the Grid-independent GEOS-Chem. We needed to rewrite the leaf area index module in order to (1) remove obsolete ASCII data files that were prohibiting the grid-independent model from functioning efficiently, and (2) compute leaf area indices consistently by regridding them from the native resolution (in this case, 0.5° x 0.5°) instead of using leaf area index data at several different resolutions. Also, we have merged into the standard repository several updates from Mike Long that are necessary in order to have GEOS-Chem driven by the NASA/GEOS-5 GCM.

We performed benchmark v9-01-03j with the Olson 1992 land map (which is the default in GEOS-Chem), and the MODIS leaf area indices, both at 0.5° x 0.5° resolution. The updated LAI code introduced into v9-01-03j will eventually allow us to use higher resolution versions of both the Olson land map and the MODIS leaf area indices. This will be an important step that will facilitate GEOS-Chem nested-grid simulations 0.25° x 0.25°.

Approval
Requires further investigation: No
Approved: Yes
Date of approval: 17 Apr 2012

--Bob Y. 11:06, 17 April 2012 (EDT)

v9-01-03i

Here is the assessment form for 1-month benchmark simulation v9-01-03i.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Thu Mar 22 18:19:43 EDT 2012
Performance statistics:
  • Ran on 8 CPUs of terra-11-c.as.harvard.edu (2.660 GHz x 8 CPU)
  • Wall time: 1:42
  • CPU time / wall time: 7.6384
  • % of ideal performance: 95.48%
Compared to previous benchmark: v9-01-03h
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify):
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03i.tar.gz
Metrics
Global mean OH (from log file): 13.1142905403284 x 105 molec/cm3
Methyl chloroform lifetime: 4.7616 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: PAN, ALD2, C2H6
Comments on SURFACE differences:

Concentrations of C2H6 increase 10-100% globally, except in eastern North America and coastal China. Typically, regional emission inventories overwrite the global emissions inventories; however, in this update, the C2H6 anthropogenic emissions overwrite has been removed for EPA/NEI2005, Streets, and EMEP because anthropogenic emissions of C2H6 were found to be too low in these regional inventories. Aside from decreases over eastern North America and coastal China, the offline C2H6 emissions improve concentrations over the rest of the North American and Asian regions.

Smaller increases in PAN (~10%) and ALD2 (10-60%) are expected with an increase in C2H6, as C2H6 is a source of these species.

Note: Differences in NIT are numerical noise caused by ISORROPIA.

At 500 hPa, list all species that changed by 10% or more: ALD2, C2H6
Comments on 500 hPa differences:

The changes at the surface propagate into the free troposphere for ALD2 and C2H6.

Note: Differences in NH3 and NIT are numerical noise caused by ISORROPIA.

In the ZONAL MEAN differences, list all species that changed by 10% or more: PAN, ALD2, C2H6
Comments on ZONAL MEAN differences:

Patterns are consistent with the surface changes described above.

Note: Differences in NH3 and NIT are numerical noise caused by ISORROPIA.

In the EMISSION RATIO maps, list all species that changed by 10% or more: C2H6
Comments on EMISSION RATIO differences: There are widespread increases in anthropogenic emissions of C2H6 by as much as 100%. Decreases of up to 50% occur in some regions, primarily in regions where C2H6 anthropogenic emissions are low in the first place. Overall, this is an improvement from using RETRO C2H6 anthropogenic emissions, which were found to be too low.
Additional or summary comments:
Approval
Requires further investigation: No
Approved: Yes (--Melissa Payer 12:22, 23 March 2012 (EDT))
Date of approval: 26 March 2012

--Melissa Payer 17:49, 20 March 2012 (EDT)

v9-01-03h

Here is the assessment form for 1-month benchmark simulation v9-01-03h.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Hg(II) partitioning: Helen Amos (Harvard)
  • Nested-grid CO2: Yuxuan Wang (Tsinghua)
  • Historical emission inventories: Eric Leibensperger (MIT)
  • CH4 simulation updates: Kevin Wecht (Harvard)
  • Remove support for GEOS-3: GEOS-Chem Support Team
  • Fix in streets_anthro_mod.F: Fabien Paulot
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Thu Mar 8 19:02:33 EST 2012
Performance statistics:
  • Ran on 8 CPUs of terra-02-c.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 1:40
  • CPU time / wall time: 7.6326
  • % of ideal performance: 95.41%
Compared to previous benchmark: v9-01-03g
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify): We expect to see no changes from v9-01-03g
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03h.tar.gz
Metrics
Global mean OH (from log file): 13.1255094316462 x 105 molec/cm3
Methyl chloroform lifetime: 4.7576 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: NIT
Comments on SURFACE differences: Small numerical noise where concentrations are low cause large % differences. This is due to ISORROPIA.
At 500 hPa, list all species that changed by 10% or more: NH3, NIT
Comments on 500 hPa differences: Small numerical noise where concentrations are low cause large % differences. This is due to ISORROPIA.
In the ZONAL MEAN differences, list all species that changed by 10% or more: NH3, NIT
Comments on ZONAL MEAN differences: Small numerical noise where concentrations are low cause large % differences. This is due to ISORROPIA.
In the EMISSION RATIO maps, list all species that changed by 10% or more: None.
Comments on EMISSION RATIO differences:
Additional or summary comments: To eliminate the effect from ISORROPIA, we re-ran 1-month benchmarks for v9-01-02g and v9-01-02h with ISORROPIA turned off. We have created this frequency distribution histogram plot. As you can see, both benchmarks have identical results (i.e. 100% of all grid boxes in the central bar of the histogram).
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 09 March 2012

--Melissa Payer 10:36, 9 March 2012 (EST)

v9-01-03g

Here is the assessment form for 1-month benchmark simulation v9-01-03g.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • emep_mod.F fix: GEOS-Chem Support Team
  • nei2005_anthro_mod.F fix: GEOS-Chem Support Team
  • PRECTOT fix: GEOS-Chem Support Team
  • Rn-Pb-Be drydep diagnostic fix: GEOS-Chem Support Team
  • Offline aerosol fix: GEOS-Chem Support Team
  • diag3.F fixes: Mat Evans (Leeds) and GEOS-Chem Support Team
  • Nested-grid bug fixes: Dylan Millet (U. Minnesota)
  • Hg emissions fix: Jenny Fisher (Harvard)
  • MERRA 2x2.5 lightning NOx: Lee Murray (Harvard) and Matt Cooper (Dalhousie)
  • GEOS-5.7.2: GEOS-Chem Support Team
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Mon Feb 27 15:00:20 EST 2012
Performance statistics:
  • Ran on 8 CPUs of terra-03-c.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 1:39
  • CPU time / wall time: 7.6499
  • % of ideal performance: 95.62%
Compared to previous benchmark: v9-01-03f
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify): We expect to see no changes from v9-01-03f
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03g.tar.gz
Metrics
Global mean OH (from log file): 13.1253265985999 x 105 molec/cm3
Methyl chloroform lifetime: 4.7576 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: NIT
Comments on SURFACE differences: Small numerical noise where concentrations are low cause large % differences. This is due to ISORROPIA.
At 500 hPa, list all species that changed by 10% or more: NH3, NIT
Comments on 500 hPa differences: Small numerical noise where concentrations are low cause large % differences. This is due to ISORROPIA.
In the ZONAL MEAN differences, list all species that changed by 10% or more: NH3, NIT
Comments on ZONAL MEAN differences: Small numerical noise where concentrations are low cause large % differences. This is due to ISORROPIA.
In the EMISSION RATIO maps, list all species that changed by 10% or more: None.
Comments on EMISSION RATIO differences:
Additional or summary comments: To eliminate the effect from ISORROPIA, we re-ran 1-month benchmarks for v9-01-02f and v9-01-02g with ISORROPIA turned off. We have created this frequency distribution histogram plot. As you can see, both benchmarks have identical results (i.e. 100% of all grid boxes in the central bar of the histogram).


NOTE: The kludge implemented to prevent crashes in wetdep when encountering anomalously low PRECTOT values causes very small (mostly on the order of 0.0001 ppbv) changes in tracer concentration, as seen in this dynamic range difference plot with ISORROPIA turned off. The kludge is still an improvement to the code because the fraction of the precipitating box should always be between 0 and 1.

Approval
Requires further investigation: No
Approved: Yes
Date of approval: 27 Feb 2012

--Melissa Payer 16:36, 27 February 2012 (EST)

v9-01-03f

Here is the assessment form for 1-month benchmark simulation v9-01-03f.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • PARANOX: Geert Vinken (Eindhoven)
  • Streets bug fixes: Jenny Fisher (Harvard)
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Wed Feb 15 15:30:13 EST 2012
Performance statistics:
  • Ran on 8 CPUs of terra-10-c.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 1:40
  • CPU time / wall time: 7.6161
  • % of ideal performance: 95.20%
Compared to previous benchmark: v9-01-03e
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify):
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03f.tar.gz
Metrics
Global mean OH (from log file): 13.1254879976765 x 105 molec/cm3
Methyl chloroform lifetime: 4.7576 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: NOx, Ox, PAN, ALK4, ISOP, HNO3, H2O2, MEK, ALD2, RCHO, MVK, MACR, PMN, PPN, R4N2,PRPE, C3H8, CH2O, N2O5, HNO4, MP, DMS, SO2, SO4, SO4s, MSA, NH3, NH4, NIT, NITs, OH
Comments on SURFACE differences: In original model, ship NOx was released as 10*O3 and 1*HNO3 molecules (continuously, day and night). Now we release a fraction of the NOx, and O3 and HNO3 produced. This explains large increase in NOx, as no ship NOx was present before this update. Other species are influenced by changes in O3 (and consequently OH).
At 500 hPa, list all species that changed by 10% or more: NOx, ISOP, MVK, MACR, PMN, PRPE, N2O5, HNO4, DMS, SO2, MSA, NH3 (noisy), NIT (noisy), NITs
Comments on 500 hPa differences:
In the ZONAL MEAN differences, list all species that changed by 10% or more: NOx, PAN, ISOP, HNO3, H2O2, RCHO, MVK, MACR, PMN, PPN, PRPE, CH2O, N2O5, HNO4, MP, DMS, MSA, NH3, NIT (noisy), NITs, OH
Comments on ZONAL MEAN differences: Most of the effects are concentrated near the surface and in the lat range 0..60, where most of the ship emissions are located.
In the EMISSION RATIO maps, list all species that changed by 10% or more: NOx
Comments on EMISSION RATIO differences: Instead of emitting no ship NOx, we now release a fraction of the ship NOx (+O3 and HNO3 produced).
Additional or summary comments: Streets bugfix only influences model runs before 2000 or nested Asia simulations (see Bug fixes for David Streets emissions for years 2005 and earlier)
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 17 Feb 2012

--Geert Vinken 13:46, 17 February 2012 (CET)

--Melissa Payer 16:35, 15 February 2012 (EST)

v9-01-03e

Here is the assessment form for 1-month benchmark simulation v9-01-03e.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Scavenging improvements: Qiaoqiao Wang (Harvard)
  • Washout structural update: Helen Amos (Harvard)
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri Jan 20 13:26:06 EST 2012
Performance statistics:
  • Ran on 8 CPUs of terra-08-c.as.harvard.edu (2.926 GHz x 8 CPU)
  • Wall time: 1:13
  • CPU time / wall time: 7.7027
  • % of ideal performance: 96.28%
Compared to previous benchmark: v9-01-03d
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify):
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03e.tar.gz
Metrics
Global mean OH (from log file): 13.0850715760517 x 105 molec/cm3
Methyl chloroform lifetime: 4.7757 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: NOx, PAN, ISOP, HNO3, H2O2, MVK, PPN, PRPE, N2O5, HNO4,SO2, SO4, SO4s, MSA, NH3, NH4, NIT, NITs, BCPI, BCPO, OCPI, OCPO, DST1, DST2, DTS3, DST4, SALA, SALC
Comments on SURFACE differences: The differences are mainly due to the changes in below-cloud scavenging for aerosols. The below-cloud scavenging is very efficient for coarse sea salt and coarse dust, but not for other aerosols.
At 500 hPa, list all species that changed by 10% or more: NOx, ALK4, ISOP, HNO3, MVK, MACR, PMN, PRPE, N2O5, HNO4, MP, DMS, SO2, SO4, SO4s, MSA, NH3, NH4, NIT, NITs, BCPI, BCPO, OCPI, OCPO, DST1, DST2, DTS3, DST4, SALA, SALC
Comments on 500 hPa differences: The significant decreases in DUST, BCPO and HNO3 are because of the in-cloud scavenging by snow.
In the ZONAL MEAN differences, list all species that changed by 10% or more: NOx, ALK4, ISOP, HNO3, H2O2, MVK, MACR, PMN, PRPE, N2O5, HNO4, MP, DMS, SO2, SO4, SO4s, MSA, NH3, NH4, NIT, NITs, BCPI, BCPO, OCPI, OCPO, DST1, DST2, DTS3, DST4, SALA, SALC, OH
Comments on ZONAL MEAN differences: Changes in below-cloud scavenging reduces concentrations of coarse dust and coarse sea salt but increases other aerosol concentrations. Changes in in-cloud scavenging decrease concentrations of BCPO, dust and HNO3 especially at high altitudes. As a result, dust concentrations are decreased at all altitudes. BCPO concentrations tend to decrease at high altitude but increase at low altitude. Concentrations of other aerosols tend to increase at high altitude especially over the tropics and subtropics. This is because we suppress the in-cloud scavenging at low temperature during convective updrafts so as to be consistent with the parameterizations in large-scale precipitation.
In the EMISSION RATIO maps, list all species that changed by 10% or more: No
Comments on EMISSION RATIO differences:
Additional or summary comments:
Approval
Requires further investigation: NO
Approved: Yes
Date of approval: 02 Feb 2012

--Melissa Payer 16:15, 23 January 2012 (EST)

v9-01-03d

Here is the assessment form for 1-month benchmark simulation v9-01-03d.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Sea salt improvements: Lyatt Jaeglé (U. Washington)
  • Dry deposition: Lyatt Jaeglé (U. Washington) and GEOS-Chem Support Team
  • Dust submicron size distribution: David Ridley (Colo. State)
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Tue Jan 10 19:18:47 EST 2012
Performance statistics:
  • Ran on 8 CPUs of terra-12-c.as.harvard.edu (2.670 GHz x 8 CPU)
  • Wall time: 1:32
  • CPU time / wall time: 7.6956
  • % of ideal performance: 96.20%
Compared to previous benchmark: v9-01-03c
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify): Sea salt emissions & deposition, Dust AOD
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03d.tar.gz
Metrics
Global mean OH (from log file): 13.3024776896910 x 105 molec/cm3
Methyl chloroform lifetime: 4.7075 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: SALA, SALC, N2O5, HNO3, NIT, NITs, NH4, NH3, SO4s, HNO4, NOx, PAN
Comments on SURFACE differences: Concentrations of sea salt aerosols (SALA and SALC) increase in the tropics by ~50-100% and decrease at mid-high latitudes by 50-100% as expected from the implementation of the SST dependent sea salt source function.

The impact on other species is largest for nitrogen and sulfur species that directly interact with sea salt. While the levels for some of these species change by factors of 1.5-2 over the tropical and high latitude oceans (HNO3, NIT, NITs, SO4s, NH3, NH4), their concentrations are very small in these remote regions, so the absolute changes are small. For example, HNO3 increases by more than a factor of 2 over the remote southern ocean, and decreases by a factor of 2 over the tropics. But the absolute changes are a few pptv at the surface.

At 500 hPa, list all species that changed by 10% or more: SALA, SALC, SO4s, NH3, NIT, NITs
Comments on 500 hPa differences:
In the ZONAL MEAN differences, list all species that changed by 10% or more: SALA, SALC, HNO3, N2O5, HNO4, SO4s, NH4, NIT, NITs
Comments on ZONAL MEAN differences: Most of the effects are concentrated near the surface, where the sea salt burden is highest.
In the EMISSION RATIO maps, list all species that changed by 10% or more: SALA, SALC
Comments on EMISSION RATIO differences: As expected from the inclusion of a SST dependence in the sea salt source function, sea salt emissions for accumulation mode sea salt aerosols (SALA) increase by 30-80% in warm tropical regions, and decrease by ~50% over high-latitudes. Overall SALA emissions decrease by 23%. Global emission of coarse mode sea salt aerosols (SALC) increase by 70%, with the largest increases in the tropics. This is driven by both the SST dependence but also the expansion of the size distribution of sea salt from 0.5-4 to 0.5-8 microns.
Additional or summary comments:
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 12 Jan 2012

--Melissa Payer 11:44, 11 January 2012 (EST)

v9-01-03c

Here is the assessment form for 1-month benchmark simulation v9-01-03c.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Jingqiu Mao (NOAA/GFDL)
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri Dec 16 00:16:12 EST 2011
Performance statistics:
  • Ran on 8 CPUs of terra-11-c.as.harvard.edu (2.660 GHz x 8 CPU)
  • Wall time: 1:58
  • CPU time / wall time: 6.2283
  • % of ideal performance: 77.85%
Compared to previous benchmark: v9-01-03b
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify):
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03c.tar.gz
Metrics
Global mean OH (from log file): 13.3104192214225 x 105 molec/cm3
Methyl chloroform lifetime: 4.7056 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: PAN, ISOP, ALD2, RCHO, MVK, MACR, PMN, PPN, PRPE, CH2O, N2O5, MP, DMS
Comments on SURFACE differences: The biggest difference were found for MVK and MACR, as they were not deposited in previous versions. The difference in HCHO is about 10%. The differences in organic nitrates are expected due to the supply of OVOCs.
At 500 hPa, list all species that changed by 10% or more: ISOP, ALD2, RCHO, MVK, MACR, PMN, PRPE, N2O5, DMS, NH3, NIT
Comments on 500 hPa differences: Expected.
In the ZONAL MEAN differences, list all species that changed by 10% or more: MVK,MACR,PMN,NH3,NIT
Comments on ZONAL MEAN differences: Expected.
In the EMISSION RATIO maps, list all species that changed by 10% or more: None
Comments on EMISSION RATIO differences:
Additional or summary comments: The small increase in global mean OH is likely due to the small increase of NOx (similar pattern in the difference plots).
Approval
Requires further investigation: No
Approved: Yes (Jingqiu Mao)
Date of approval: 12-19-2011

--Melissa Payer 10:46, 19 December 2011 (EST)

v9-01-03b

Here is the assessment form for 1-month benchmark simulation v9-01-03b.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Acetone: Emily Fischer (Harvard)
  • Retirement of GEIA: Dylan Millet (U. Minnesota)
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Tue Dec 13 12:20:47 EST 2011
Performance statistics:
  • Ran on 8 CPUs of terra-10-c.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 1:39
  • CPU time / wall time: 7.2830
  • % of ideal performance: 91.04%
Compared to previous benchmark: v9-01-03a
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify):
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03b.tar.gz
Metrics
Global mean OH (from log file): 13.2811407896285 x 105 molec/cm3
Methyl chloroform lifetime: 4.7161 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: PAN, ACET, R4N2
Comments on SURFACE differences: ACET generally decreases 10-50%, strongest decreases over the oceans due to elimination of the (incorrect) ocean source. Over continental regions, the largest changes are away from terrestrial acetone sources. There are smaller relative changes over the main continental emission regions. PAN decreases by 10-25% over most of the NH oceans, but the absolute differences are very small due to low surface PAN mixing ratios in these regions (< 30 pptv). Weak decreases for R4N2 in a couple spots.
At 500 hPa, list all species that changed by 10% or more: ACET, R4N2
Comments on 500 hPa differences: ACET decreases of 10-25% over many regions at 500hPa. Weak decreases for R4N2 in a couple spots.
In the ZONAL MEAN differences, list all species that changed by 10% or more: ACET, PAN
Comments on ZONAL MEAN differences: Changes exceeding 10% for ACET and PAN, consistent with the above.
In the EMISSION RATIO maps, list all species that changed by 10% or more: ACET
Comments on EMISSION RATIO differences: Changes as expected based on emission updates.
Additional or summary comments:
Approval
Requires further investigation: No
Approved: Yes (--Dbm 11:17, 14 December 2011 (EST))
Yes (--[[User:Emily Fischer] 3:32, 14 December 2011 (EST))
Date of approval: 12-14-2011

--Melissa Payer 13:42, 13 December 2011 (EST)
--Bob Y. 17:06, 14 December 2011 (EST)

v9-01-03a

Here is the assessment form for 1-month benchmark simulation v9-01-03a.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Fabien Paulot (Harvard)
Version, resolution, met fields used: v9-01-03, GEOS-5 (47L), 4x5
1-month benchmark finished on: Wed Dec 7 16:06:38 EST 2011
Performance statistics:
  • Ran on 8 CPUs of terra-04-c.as.harvard.edu (2.925 GHz x 8 CPU)
  • Wall time: 0:57
  • CPU time / wall time: 7.4849
  • % of ideal performance: 93.56%
Compared to previous benchmark: v9-01-02s
This update will impact:
(select all that apply with boldface)
Advection, BL Mixing, Convection, Met Fields, Dry Dep, Wet Dep, Stratosphere, Anthro Emiss, Biogenic Emiss, Biomass Emiss, Photolysis, Chemistry, Other (please specify):
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-01-03/v9-01-03a.tar.gz
Metrics
Global mean OH (from log file): 13.2473124589166 x 105 molec/cm3
Methyl chloroform lifetime: 4.7300 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: PAN (downwind of South America), ALK4 over oceanic basins, ISOP downwind of South America, MEK tropics, MVK, MACR (downwind of South America), PMN, PPN, PRPE, N2O5, DMS
Comments on SURFACE differences: Sort of makes sense. Changes are in regions that are NOx starved (South America) or cold (the loss from GLPAN will be greatest). Changes are small overall and almost always negative. The changes for DMS are relatively large (10-20% decrease)
At 500 hPa, list all species that changed by 10% or more: ALK4, ISOP, MEK, MVK, MACR, PRPE (large changes in polar regions > 20%), DMS
Comments on 500 hPa differences: Changes are generally bigger than at the surface (temperature dependence of GLPAN)
In the ZONAL MEAN differences, list all species that changed by 10% or more: ISOP (but at high altitude and in polar regions), ALK4 in polar regions, MEK, MVK, MACR, PMN, PRPE
Comments on ZONAL MEAN differences: Small changes overall (in an absolute sense). DMS has the largest change.
In the EMISSION RATIO maps, list all species that changed by 10% or more: none
Comments on EMISSION RATIO differences:
Additional or summary comments:
Approval
Requires further investigation: No
Approved: Yes
Date of approval: Approved 08 Dec 2011

--Bob Y. 11:21, 8 December 2011 (EST)
--Melissa Payer 13:48, 13 December 2011 (EST)