GEOS-Chem v9-02 benchmark history

From Geos-chem
Revision as of 19:36, 17 January 2014 by Bmy (Talk | contribs) (v9-02o)

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 v9-02.

1-year benchmarks

v9-02r-geosfp-Run1

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

Color Quantity Plotted Met Type Year Emissions Tropospheric chemistry mechanism Photolysis Linoz Annual Mean OH
[105 molec/cm3]
Red v9-02r-geos5-Run0 GEOS-5, 4x5 2005 Same as v9-02l-Run0
+ AEIC aircraft emissions
+ Fix seasonality in CAC NH3 emissions
+ Bug fix for NEI2005 SO4 emissions in sulfate_mod.F
Same as v9-02j-Run0
+ Cloud water pH for sulfate formation
+ Update molecular weight of sea salt tracers
Same as v8-03-01
+ Better representation of OC growth with RH and correction to sulfate optics
ON 12.926
Green v9-02r-geosfp-Run0 GEOS-FP, 4x5 2013 Same as v9-02r-geos5-Run0 (scaled all emissions to 2005) " " " " ON 12.389
Blue v9-02r-geosfp-Run1 GEOS-FP, 4x5 2013 Anthropogenic emissions scaled to 2010 (latest available)
+ 2011 GFED3 emissions (latest available)
+ Olson 2001 land map
" " " " ON 12.689
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/v9-02/v9-02r/geosfp/Run1/output/ps
mget *
cd gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02r/geosfp/Run1/output/pdf
mget *

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

http://ftp.as.harvard.edu/gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02r/geosfp/Run1/output/pdf/

--Melissa Sulprizio 15:36, 10 January 2014 (EST)

v9-02r-geosfp-Run0

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

Color Quantity Plotted Met Type Year Emissions Tropospheric chemistry mechanism Photolysis Linoz Annual Mean OH
[105 molec/cm3]
Red v9-02l-Run0 GEOS-5, 4x5 2005 Same as v9-01-03r-Run0
+Updated NH3 seasonal scaling factors over the US
+ Updated Canadian NH3 inventory
+ Fix diurnal NOx scale factors and fix double counting of biofuels over Mexico, Canada, and Europe
+ Bug fix for 3-hourly GFED3 emissions
+ Bug fix for anthropogenic scaling factors for years 2006 and later
+ Correct additional bugs in ship emissions w/ non-local PBL mixing
+ Bug fix for ship CO emissions
Same as v9-02j-Run0 Same as v8-03-01 ON 12.570
Green v9-02r-geos5-Run0 GEOS-5, 4x5 2005 + AEIC aircraft emissions
+ Fix seasonality in CAC NH3 emissions
+ Bug fix for NEI2005 SO4 emissions in sulfate_mod.F
+ Cloud water pH for sulfate formation
+ Update molecular weight of sea salt tracers
+ Better representation of OC growth with RH and correction to sulfate optics ON 12.926
Blue v9-02r-geosfp-Run0 GEOS-FP, 4x5 2013 Same as v9-02r-geos5-Run0 (scaled all emissions to 2005) " " " " ON 12.389
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/v9-02/v9-02r/geosfp/Run0/output/ps
mget *
cd gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02r/geosfp/Run0/output/pdf
mget *

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

http://ftp.as.harvard.edu/gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02r/geosfp/Run0/output/pdf/

--Melissa Sulprizio 15:36, 10 January 2014 (EST)

v9-02r-geos5-Run0

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

Color Quantity Plotted Met Type Anthro emissions Tropospheric chemistry mechanism Photolysis Linoz Annual Mean OH
[105 molec/cm3]
Red v9-02j-Run0 GEOS-5, 4x5 Same as v9-01-03r-Run0 Same as v9-02g-Run0
+ Removal of NOx-Ox partitioning
+Inhibition of N2O5 uptake by nitrate aerosol
+Improved HO2 uptake by aerosol
Same as v8-03-01 ON 12.238
Green v9-02l-Run0 GEOS-5, 4x5 +Updated NH3 seasonal scaling factors over the US
+ Updated Canadian NH3 inventory
+ Fix diurnal NOx scale factors and fix double counting of biofuels over Mexico, Canada, and Europe
+ Bug fix for 3-hourly GFED3 emissions
+ Bug fix for anthropogenic scaling factors for years 2006 and later
+ Correct additional bugs in ship emissions w/ non-local PBL mixing
+ Bug fix for ship CO emissions
" " " " ON 12.570
Blue v9-02r-geos5-Run0 GEOS-5, 4x5 + AEIC aircraft emissions
+ Fix seasonality in CAC NH3 emission files
+ Bug fix for NEI2005 SO4 emissions in sulfate_mod.F
+ Cloud water pH for sulfate formation
+ Update molecular weight of sea salt tracers
+ Better representation of OC growth with RH and correction to sulfate optics ON 12.926
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/v9-02/v9-02r/geos5/Run0/output/ps
mget *
cd gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02r/geos5/Run0/output/pdf
mget *

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

http://ftp.as.harvard.edu/gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02r/geos5/Run0/output/pdf/

--Melissa Sulprizio 15:15, 23 December 2013 (EST)

v9-02l-Run0

This 1-year benchmark simulation was approved by the GEOS-Chem Steering Committee on 02 Jul 2013.

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

Color Quantity Plotted Met Type Anthro emissions Tropospheric chemistry mechanism Linoz Annual Mean OH
[105 molec/cm3]
Red v9-02g-Run0 GEOS-5, 4x5 Same as v9-01-03r-Run0 Same as v9-02f-Run0
+ Adopt Paulot isoprene scheme as standard
+ Fix RO2+HO2 rate constant
ON 12.472
Green v9-02j-Run0 GEOS-5, 4x5 " " + Removal of NOx-Ox partitioning
+Inhibition of N2O5 uptake by nitrate aerosol
+Improved HO2 uptake by aerosol
ON 12.238
Blue v9-02l-Run0 GEOS-5, 4x5 +Updated NH3 seasonal scaling factors over the US
+ Updated Canadian NH3 inventory
+ Fix diurnal NOx scale factors and fix double counting of biofuels over Mexico, Canada, and Europe
+ Bug fix for 3-hourly GFED3 emissions
+ Bug fix for anthropogenic scaling factors for years 2006 and later
+ Correct additional bugs in ship emissions w/ non-local PBL mixing
+ Bug fix for ship CO emissions
" " ON 12.570
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/v9-02/v9-02l/Run0/output/ps
mget *
cd gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02l/Run0/output/pdf
mget *

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

http://ftp.as.harvard.edu/gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02l/Run0/output/pdf/

--Melissa Payer 17:14, 26 June 2013 (EDT)

v9-02j-Run0

This 1-year benchmark simulation was approved by the GEOS-Chem Steering Committee on 03 Jun 2013.

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

Color Quantity Plotted Met Type Anthro emissions Tropospheric chemistry mechanism Linoz Annual Mean OH
[105 molec/cm3]
Red v9-02f-Run0 GEOS-5, 4x5 Same as v9-01-03r-Run0 Same as v9-01-03r-Run1+ Inorganic chemistry updates
+ Methyl peroxy nitrate (MPN) chemistry
ON 11.503
Green v9-02g-Run0 GEOS-5, 4x5 " " + Adopt Paulot isoprene scheme as standard
+ Fix RO2+HO2 rate constant
ON 12.472
Blue v9-02j-Run0 GEOS-5, 4x5 " " + Removal of NOx-Ox partitioning
+Inhibition of N2O5 uptake by nitrate aerosol
+Improved HO2 uptake by aerosol
ON 12.238
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/v9-02/v9-02j/Run0/output/ps
mget *
cd gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02j/Run0/output/pdf
mget *

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

http://ftp.as.harvard.edu/gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02j/Run0/output/pdf/

--Melissa Payer 13:02, 31 May 2013 (EDT)

v9-02g-Run0

This 1-year benchmark simulation was approved by the GEOS-Chem Steering Committee on 01 Apr 2013.

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

Color Quantity Plotted Met Type Anthro emissions Soil NOx emissions Tropospheric chemistry mechanism Linoz Annual Mean OH
[105 molec/cm3]
Red v9-02b-Run0 GEOS-5, 4x5 Same as v9-01-03r-Run0 New soil NOx emission module Same as v9-01-03r-Run1 ON 11.424
Green v9-02f-Run0 GEOS-5, 4x5 " " " " + Inorganic chemistry updates
+ Methyl peroxy nitrate (MPN) chemistry
ON 11.503
Blue v9-02g-Run0 GEOS-5, 4x5 " " " " + Adopt Paulot isoprene scheme as standard
+ Fix RO2+HO2 rate constant
ON 12.472
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/v9-02/v9-02g/Run0/output/ps
mget *
cd gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02g/Run0/output/pdf
mget *

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

http://ftp.as.harvard.edu/gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02g/Run0/output/pdf/

--Melissa Payer 12:15, 29 March 2013 (EDT)

v9-02f-Run0

This 1-year benchmark simulation was approved by the GEOS-Chem Steering Committee on 20 Mar 2013.

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

Color Quantity Plotted Met Type Anthro emissions Soil NOx emissions Tropospheric chemistry mechanism Linoz Annual Mean OH
[105 molec/cm3]
Red v9-01-03r-Run1 GEOS-5, 4x5 Same as v9-01-03r-Run0 Original soil NOx algorithm based on Yienger & Levy [1995] Same as v8-02-01-Run0
+ Bromine chemistry mechanism
ON 11.356
Green v9-02b-Run0 GEOS-5, 4x5 " " New soil NOx emission module " " ON 11.424
Blue v9-02f-Run0 GEOS-5, 4x5 " " " " + Inorganic chemistry updates
+ Methyl peroxy nitrate (MPN) chemistry
ON 11.503
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/v9-02/v9-02f/Run0/output/ps
mget *
cd gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02f/Run0/output/pdf
mget *

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

http://ftp.as.harvard.edu/gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02f/Run0/output/pdf/

--Melissa Payer 18:06, 18 March 2013 (EDT)

v9-02b-Run0

NOTE: A 1-year benchmark simulation was performed for v9-02b to serve as a reference when benchmarking the chemistry updates implemented in v9-02f.

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

Color Quantity Plotted Met Type Anthro emissions Soil NOx emissions Tropospheric chemistry mechanism Linoz Annual Mean OH
[105 molec/cm3]
Red v9-01-03r-Run0 GEOS-5, 4x5 Same as v9-01-02-Run1
+ C2H6 emissions from offline simulation
+ PARANOX ship plume emissions
Original soil NOx algorithm based on Yienger & Levy [1995] Same as v8-02-01-Run0 ON 11.858
Green v9-01-03r-Run1 GEOS-5, 4x5 " " " " + Bromine chemistry mechanism ON 11.356
Blue v9-02b-Run0 GEOS-5, 4x5 " " New soil NOx emission module " " ON 11.424
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/v9-02/v9-02b/Run0/output/ps
mget *
cd gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02b/Run0/output/pdf
mget *

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

http://ftp.as.harvard.edu/gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02b/Run0/output/pdf/

--Melissa Payer 18:06, 18 March 2013 (EDT)

1-month benchmarks

v9-02r

NOTE: For v9-02r, we performed 1-month benchmarks with both GEOS-FP and GEOS-5 meteorology. We shall now use GEOS-FP meteorology for all 1-month (and 1-year) benchmarks for all future GEOS-Chem versions.

1-month benchmark with GEOS-FP meteorology

Here is the assessment form for 1-month benchmark simulation v9-02r with GEOS-FP meteorology.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • GEOS-FP updates: GEOS-Chem Support Team
  • Additional minor bug fixes: GEOS-Chem Support Team, Colette Heald (MIT)
  • Reprocessed CAC NH3 files: Wai Ho Lo (Dalhousie)
  • Sea salt update: Colette Heald (MIT)
Version, resolution, met fields used: v9-02, GEOS-FP (47L), 4x5, July 2013
1-month benchmark finished on: Thu Dec 12 23:08:16 EST 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-08.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:47
  • Scalability: 6.9529
Compared to previous benchmark: v9-02r with GEOS-5 meteorology
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):
  1. This benchmark used July 2013 GEOS-FP met, but the v9-02r GEOS-5 benchmark used July 2005 met;
  2. This benchmark used the identical initial conditions as the the v9-02r GEOS-5 benchmark simulation.
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-02/v9-02r.geosfp/
Metrics
Global mean OH (from log file): 13.7371733863556 x 105 molec/cm3
Methyl chloroform lifetime: 4.6673 years
Did either of these change by more than 5%? No. The mean OH differs by -0.79%, and the MCF lifetimes differ by 3.951%.
At the SURFACE, list all species that changed by 10% or more: All species except CHBr3 change by 10% or more.
Comments on SURFACE differences: Observed differences at the surface are most likely attributed to:
  1. Differences between the GEOS-FP and GEOS-5 met fields, and
  2. Interannual variability between July 2005 and July 2013.
At 500 hPa, list all species that changed by 10% or more: All species except CHBr3 change by 10% or more.
Comments on 500 hPa differences: Observed differences at the surface are most likely attributed to:
  1. Differences between the GEOS-FP and GEOS-5 met fields, and
  2. Interannual variability between July 2005 and July 2013.
In the ZONAL MEAN differences, list all species that changed by 10% or more: All species change by 10% or more.
Comments on ZONAL MEAN differences: Observed differences in zonal mean quantities are most likely attributed to:
  1. Differences between the GEOS-FP and GEOS-5 met fields and
  2. Interannual variability between July 2005 and July 2013.
In the EMISSION RATIO maps, list all species that changed by 10% or more:
  1. Anthropogenic emissions: CO, NH3, NO, SO2, SO4
  2. Aerosol emissions: DST1, DST2, DST3, DST4, SALA, SALC
  3. Biofuel emissions: CO, NO
  4. Biomass burning emissions: All species
  5. Biogenic emissions: All species except CHBr3
Comments on EMISSION RATIO differences:
  1. Changes in aerosol emissions can be attributed to interannual variability.
    • The dust and sea-salt tracers are very sensitive to the 10-meter wind speed, which differs between July 2005 and July 2013.
  2. Changes in anthropogenic and biofuel emissions can be attributed to interannual variability.
    • The EPA/NEI05 inventory relies on weekday/weekend scaling, which differs for July 2005 and July 2013.
    • Likewise, EMEP, BRAVO, EMEP, and CAC inventories are computed with different inventory years in the 2 benchmark simulations.
  3. Changes in biomass burning emissions can be attributed to interannual variability.
    • The v9-02r GEOS-5 benchmark (July 2005 met) uses GFED year 2005.
    • The v9-02r GEOS-FP benchmark (July 2013 met) uses GFED year 2011, which is the last available GFED3 year.
  4. Changes in biogenic emissions can be attributed to:
    • Differences in the GEOS-FP and GEOS-5 met fields;
    • Interannual variability between July 2005 and July 2013
    • Difference in land maps used for biogenic emissions and dry deposition:
Additional or summary comments:

Despite using a different set of met fields for a different year, the v9-02r GEOS-FP mean OH only differs by less than -1% from the v9-02r GEOS-5 benchmark. We may take this as evidence that the GEOS-FP simulation in GEOS-Chem is performing adequately, and that there are no major surprises. Further testing will be needed with the 1-year benchmark simulation.

We also note the following similarities and differences between v9-02 GEOS-FP and v9-02 GEOS-5 benchmark simulations:

  1. Both v9-02r GEOS-5 and v9-02r GEOS-FP benchmark simulations used identical initial conditions. Any differences between these two simulations should therefore faithfully reflect:
    • Differences in GEOS-FP vs. GEOS-5 met, and
    • Interannual variability between July 2005 and July 2013.
  2. Both benchmark simulations used different assimilated ozone columns.
    • The v9-02r GEOS-5 benchmark (as with all past benchmarks with GEOS-5 met) used the assimilated ozone columns from the TOMS/SBUV merged total ozone data. Ozone columns from the GEOS-5 met fields are only used when TOMS/SBUV data is not available. The last available TOMS year is 2010.
    • The v9-02r GEOS-FP benchmark used the assimilated ozone columns taken directly from the GEOS-FP met field archive.
  3. The GEOS-FP met fields used for the v9-02r GEOS-FP benchmark take the 3-D cloud fraction field ("CLOUD") directly from the GMAO GEOS-FP "raw" data files.
    • The previous v9-02r GEOS-FP benchmark that we submitted used met field files that computed
      CLOUD = min( CFAN + CFLS, 1.0 ),
      which proved to be an inaccurate approximation.
  4. The scalability of the v9-02r GEOS-FP benchmark on 8 CPUs (6.9529) is lower than in the corresponding v9-02r GEOS-5 benchmark (7.2794). We attribute this to the following reasons:
    • The GEOS-FP meteorology must be read from disk more frequently (at hourly and 3-hourly intervals) than GEOS-5 meteorology (3-hourly and 6-hourly intervals).
    • For compatibility with our Grid-Independent GEOS-Chem project, we have stored the GEOS-FP met field data in netCDF data files. GEOS-Chem must utilize I/O routines from the netCDF library—which incur a fixed amount of overhead—in order to read the GEOS-FP data from disk. On the other hand, GEOS-5 met fields are stored in flat binary data files, which can be read from disk without such additional overhead.
Approval
Requires further investigation: No
Approved by: Bob Yantosca, Melissa Sulprizio
Date of approval: 14 Dec 2013

--Bob Y. 14:05, 14 January 2014 (EST)

1-month benchmark with GEOS-5 meteorology

Here is the assessment form for 1-month benchmark simulation v9-02r with GEOS-5 meteorology.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • GEOS-FP updates: GEOS-Chem Support Team
  • Additional minor bug fixes: GEOS-Chem Support Team, Colette Heald (MIT)
  • Reprocessed CAC NH3 files: Wai Ho Lo (Dalhousie)
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5, July 2005
1-month benchmark finished on: Fri Nov 8 15:49:34 EST 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-10.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:38
  • Scalability: 7.2794
Compared to previous benchmark: v9-02q
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): Minor numerical differences may occur as a result of the bug fixes
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-02/v9-02r.geos5/
Metrics
Global mean OH (from log file): 13.8467219229137 x 105 molec/cm3
Methyl chloroform lifetime: 4.4899 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: NH3, NIT
Comments on SURFACE differences:

NH3 differences over Canada are to be expected because we now use corrected CAC emissions files.
NIT differences over Canada (mirroring those of NH3) appear to be caused by ISORROPIA.

At 500 hPa, list all species that changed by 10% or more: NH3, NIT
Comments on 500 hPa differences:

NH3 differences over Canada are to be expected because we now use corrected CAC emissions files.
NIT differences over Canada (mirroring those of NH3) appear to be caused by ISORROPIA.

In the ZONAL MEAN differences, list all species that changed by 10% or more: PMN, PRPE, CH38, SO4s, NH3, NIT, NITs, BCPO, OCPO, DST2, SALC
Comments on ZONAL MEAN differences:

NH3: Small number differences caused by corrected CAC emissions files
NIT: Small number differences likely caused by ISORROPIA (this is a known issue)
PMN: A single box is nonzero at approx. 600 hPa (appears to be small-number difference)
PRPE, C3H8, SO4s, NITs, BCPO, OCPO, DST2, SALC: Observed small-number differences are visible near top of atmosphere.

In the EMISSION RATIO maps, list all species that changed by 10% or more: NH3
Comments on EMISSION RATIO differences: NH3: Observed differences over Canada are to be expected because we are now using corrected CAC emissions files.
Additional or summary comments: Most of the changes in this version were to include the GEOS-FP met fields. As a result, we observe mostly minor differences in the output. NH3 emissions over Canada have changed due to the fact that we are now using corrected CAC emissions files with the proper seasonality. We have also made several minor bug fixes to increase numerical stability, which may be responsible for some of the minor differences in the other tracers that we observe.

We will perform 1-month and 1-year benchmarks with the new GEOS-FP met fields. Those results will be posted here shortly.

Approval
Requires further investigation: No
Approved by: Bob Yantosca
Date of approval: 14 Nov 2013

--Bob Y. 14:25, 14 November 2013 (EST)

v9-02q

Here is the assessment form for 1-month benchmark simulation v9-02q.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • jv_spec.dat updates: David Ridley (MIT), Randall Martin (Dalhousie)
  • jv_spec_aod.dat fix: Gabriele Curci (U. L'Aquila)
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri Sep 13 21:06:31 EDT 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-08.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:37
  • Scalability: 7.2660
Compared to previous benchmark: v9-02p
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/v9-02/v9-02q/
Metrics
Global mean OH (from log file): 13.8470394111778 x 105 molec/cm3
Methyl chloroform lifetime: 4.4898 years
Did either of these change by more than 5%? No. The mean OH differs by 0.82%, and the MCF lifetimes differ by -0.89%.
At the SURFACE, list all species that changed by 10% or more: NO, ISOP, HNO3, H2O2, MVK, MACR, PMN, PPN, PRPE, N2O5, HNO4, MP, DMS, NH3, NIT, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, MPN, ISOPN, MOBA, HAC, GLYC, MMN, RIP, IEPOX, MAP, NO2, NO3, HNO2, OH, HO2
Comments on SURFACE differences:
  • Differences in NO, ISOP, MVK, MACR, PMN, ISOPN, MOBA, RIP, and IEPOX are mainly over the oceans or Antarctica. These large ratios can be attributed to small number differences where concentrations are very low.
  • Differences in NH3 and NIT can be attributed to numerical noise from ISORROPIA.
At 500 hPa, list all species that changed by 10% or more: NO, ISOP, MVK, MACR, PMN, PRPE, N2O5, HNO4, DMS, NH3, NIT, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, MPN, ISOPN, MOBA, GLYC, MMN, RIP, IEPOX, NO3, HNO2
Comments on 500 hPa differences:
  • Differences in NO, ISOP, MVK, MACR, PMN, ISOPN, MOBA, RIP, and IEPOX are mainly over the oceans or Antarctica. These large ratios can be attributed to small number differences where concentrations are very low.
  • Differences in NH3 and NIT can be attributed to numerical noise from ISORROPIA.
In the ZONAL MEAN differences, list all species that changed by 10% or more: NO, ALK4, ISOP, H2O2, MVK, MACR, PMN, PRPE, C3H8, N2O5, HNO4, DMS, SO4s, NH3, NIT, NITs, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, MPN, ISOPN, MOBA, HAC, GLYC, MMN, RIP, IEPOX, NO2, NO3, HNO2, OH, HO2
Comments on ZONAL MEAN differences:
  • Differences in NO, ALK4, ISOP, MACR, PMN, C3H8, SO4s, ISOPN, and RIP can be attributed to small number differences where concentrations are very low.
  • Differences in NH3, NIT, and NITs can be attributed to numerical noise from 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: The changes in jv_spec.dat generally reduce AOD and aerosol surface area, in turn increasing boundary layer OH which causes changes in other species.
Approval
Requires further investigation: No
Approved by: Randall Martin, David Ridley, Gabriele Curci
Date of approval: 17 Sep 2013

--Melissa Sulprizio 17:28, 13 September 2013 (EDT)

v9-02p

Here is the assessment form for 1-month benchmark simulation v9-02p.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Tue Sep 10 23:50:47 EDT 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-09.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:36
  • Scalability: 7.2645
Compared to previous benchmark: v9-02o
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/v9-02/v9-02p/
Metrics
Global mean OH (from log file): 13.7340615285201 x 105 molec/cm3
Methyl chloroform lifetime: 4.5300 years
Did either of these change by more than 5%? No. The mean OH differs by -0.12%, and the MCF lifetimes differ by 0.17%.
At the SURFACE, list all species that changed by 10% or more: ISOP, HNO3, H2O2, MVK, MACR, PMN, N2O5, HNO4, DMS, SO2, SO4, SO4s, NH3, NH4, NIT, NITs, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, ISOPN, MOBA, HAC, GLYC, RIP, IEPOX, NO3, HNO2, HO2
Comments on SURFACE differences:
  • Differences in SO2 and SO4 over the oceans are related to enhanced reaction rates of SO2 by O3 related to increased cloud pH due to alkaline sea salt aerosol.
  • Increased SO4 over the oceans impacts the gas-particle partitioning of NH3/NH4 and HNO3/NIT. Increased aerosol acidity (due to increased SO4) increases NH4 (while decreasing NH3) and increases HNO3 (while decreasing NIT).
  • Increased SO4 aerosol also increases the rate of N2O5 hydrolysis leading to lower N2O5. This should also contribute to higher total nitrate (HNO3 + NIT).
  • Increased in-cloud oxidation of SO2 by O3 in the MBL due to higher cloud pH will decrease the other SO2 oxidation pathways. This is shown in lower SO4S due to decreases in heterogeneous oxidation of SO2 on the surface of sea salt aerosol. Since oxidation of SO2 and condensation of HNO3 compete for available sea salt aerosol alkalinity, lower SO4S results in higher NITS.
  • Small increases in H2O2 result from less SO2 oxidation by H2O2 in favor of oxidation by O3 at the higher cloud pH in the MBL.
  • Differences in ISOP, HNO3, MVK, MACR, PMN, N2O5, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, ISOPN, MOBA, GLYC, RIP, IEPOX are mainly over the oceans or Antarctica. These large ratios can be attributed to small number differences where concentrations are very low.
  • Differences in NH3, NH4, NIT, NITs can be attributed to numerical noise from ISORROPIA.
At 500 hPa, list all species that changed by 10% or more: NO, ISOP, MVK, MACR, PMN, DMS, SO2, SO4, SO4s, NH3, NIT, NITs, ISOPN, MOBA, HAC, GLYC, RIP, IEPOX
Comments on 500 hPa differences:
  • Differences in SO2, SO4, NH3, SO4s, NIT and NITs are due due to the same processes described for the surface.
  • Differences in NO, ISOP, MVK, MACR, PMN, ISOPN, MOBA, HAC, GLYC, RIP, IEPOX are mainly over the oceans or Antarctica. These large ratios can be attributed to small number differences where concentrations are very low.
  • Differences in NH3, NIT, NITs can be attributed to numerical noise from ISORROPIA.
In the ZONAL MEAN differences, list all species that changed by 10% or more: NO, ISOP, HNO3, H2O2, MVK, MACR, PMN, PRPE, C3H8, N2O5, SO2, SO4, SO4s, NH3, NH4, NIT, NITs, HBr, ISOPN, MOBA, HAC, GLYC, RIP, IEPOX, HNO2
Comments on ZONAL MEAN differences:
  • Differences in SO2, SO4, NH3, NH4, SO4s, H2O2, N2O5, NIT and NITs are due due to the same processes described for the surface. Differences tend to be higher in the southern hemisphere because there is more ocean surface than in the northern hemisphere.
  • Differences in NO, ISOP, MVK, MACR, PMN, PRPE, C3H8, ISOPN, MOBA, GLYC, RIP, IEPOX can be attributed to small number differences where concentrations are very low.
  • Differences in NH3, NH4, NIT, NITs can be attributed to numerical noise from 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:
Approval
Requires further investigation: No
Approved by: Becky Alexander
Date of approval: 13 Sep 2013

--Melissa Sulprizio 16:47, 6 September 2013 (EDT)

v9-02o

Here is the assessment form for 1-month benchmark simulation v9-02o.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • SOA with semi-volatile POA: Havala Pye (EPA)
  • Non-local PBL mixing in Rn-Pb-Be simulation: Jintai Lin (PKU); GEOS-Chem Support Team
  • TOMAS bug fix: Sal Farina (CSU)
  • ND36 bug fix: Yanko Davila (U. Colorado)
  • Modifications for GC Unit Tester: GEOS-Chem Support Team
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Wed Aug 28 22:40:20 EDT 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-09.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:35
  • Scalability: 7.2649
Compared to previous benchmark: v9-02n
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 v9-02n
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-02/v9-02o.tar.gz
Metrics
Global mean OH (from log file): 13.7511714559642 x 105 molec/cm3
Methyl chloroform lifetime: 4.5225 years
Did either of these change by more than 5%? No. They are identical to the v9-02n values.
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: None
Comments on ZONAL MEAN differences:
In the EMISSION RATIO maps, list all species that changed by 10% or more: None
Comments on EMISSION RATIO differences:
Additional or summary comments:
  1. The v9-02o output is 100% identical to the v9-02n output. This indicates that the full-chemistry simulation has not been impacted by the updates, bug fixes, and structural changes that we have made in this version.
  2. Special 1-month benchmark simulations were performed for v9-02n using the traditional SOA simulation and for v9-02o using the updated SOA simulation. For more information on these simulations, please see this wiki post.
Approval
Requires further investigation: No
Approved by: Jeff Pierce, Havala Pye
Date of approval: 03 Sep 2013

--Melissa Sulprizio 16:36, 28 August 2013 (EDT)

v9-02n

Here is the assessment form for 1-month benchmark simulation v9-02n.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Sebastian Eastham (MIT)
  • Steven Barrett (MIT)
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Mon Aug 5 14:53:17 EDT 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-08.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:37
  • Scalability: 7.2908
Compared to previous benchmark: v9-02m
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/v9-02/v9-02n.tar.gz
Metrics
Global mean OH (from log file): 13.7511714559642 x 105 molec/cm3
Methyl chloroform lifetime: 4.5225 years
Did either of these change by more than 5%? No. The mean OH differs by 1.1%, and the MCF lifetimes differ by -0.73%.
At the SURFACE, list all species that changed by 10% or more: NO, ISOP, MVK, MACR, PMN, NIT, Br, HBr, BrNO3, MPN, ISPON, MOBA, IEPOX, NO2, NO3, HNO2
Comments on SURFACE differences:
  • Changes in Br and NOx species were isolated to a small band in the Antarctic
  • Perturbations to hydrocarbon species were mainly isolated to regions where the initial value was very low (e.g. remote ocean)
  • NIT delta highly isolated and likely a result of existing issues with aerosol formation
At 500 hPa, list all species that changed by 10% or more: NO, ALK4, ISOP, HNO3, MVK, MACR, PMN, PRPE, N2O5, HNO4, DMS, NH3, NIT, ISOPN, MOBA, GLYC, MMN, RIP, IEPOX, HNO2
Comments on 500 hPa differences:
  • Changes in NO isolated to Antarctic
  • Large-scale hydrocarbon perturbations, such as for PRPE, due to increase in overall HC loading as aircraft HC emissions are new to the model
  • Small-scale, "noisy" HC perturbations result from small absolute changes in regions with low initial values
In the ZONAL MEAN differences, list all species that changed by 10% or more: NO, ISOP, HNO3, H2O2, MVK, MACR, PMN, PRPE, N2O5, HNO4, MP, DMS, SO2, SO4s, NH3, NIT, NITs, BrNO2, BrNO3, MPN, ISOPN, MOBA, HAC, GLYC, MMN, RIP, IEPOX, MAP, NO2, HNO2, OH, HO2
Comments on ZONAL MEAN differences:
  • NO, NO2, BrNO2, BrNO3, HNO2, HNO3, HNO4 and N2O5 changes result from new aircraft NOx emissions and are small in absolute terms
  • Hydrocarbon changes are either "noisy", occurring in regions with low initial values (i.e. high altitude), or closely follow aircraft flight patterns due to addition of aircraft HC emissions
  • Small drop in H2O2, again at high altitude, largely due to increased aerosol activity
  • SO2 and DMS perturbations likely result from enhanced aircraft sulfur emission modeling
  • NIT and NH3 deltas are also high-altitude and "noisy", likely due to noisy aerosol formation
  • NITs and SO4s changes restricted to very high altitude, where small number changes lead to large ratios
  • OH and HO2 deltas small and restricted to very high altitude
In the EMISSION RATIO maps, list all species that changed by 10% or more: SO2, NO
Comments on EMISSION RATIO differences:
  • New aircraft SO2, NOx emissions at higher resolution than before
  • Other new aircraft emissions are not explicitly shown on the emission ratio maps
Additional or summary comments: We are currently emitting NOx as NO, but gridded data for the NO/NO2/HNO2 split emissions are available for aircraft NOx. We should consider implementing the split emissions in a future version.
Approval
Requires further investigation:
  • Emission of high-altitude hydrocarbons may result in formation of a stratospheric HC reservoir (see PRPE)
  • Stratospheric NITs and SO4s difficult to explain, but are likely due to numerical noise
Approved by: Sebastian Eastham
Date of approval: 12 Aug 2013

--Melissa Sulprizio 09:36, 31 July 2013 (EDT)

v9-02m

Here is the assessment form for 1-month benchmark simulation v9-02m.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • RCP scenarios: Chris Holmes (UC Irvine)
  • EDGARv4.2 for CH4 simulation: Kevin Wecht (Harvard)
  • Annual scale factors update: Aaron van Donkelaar (Dalhousie)
  • GFED3 update: Prasad Kasibhatla (Duke)
  • TOMS O3 columns update: Jenny Fisher (U. Wollongong)
  • GCAP fixes: GEOS-Chem Support Team
  • Bug fix for Canadian emissions: Fabien Paulot (Harvard); Christoph Keller (Harvard)
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Thu Jul 25 21:14:47 EDT 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-08.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:34
  • Scalability: 7.3134
Compared to previous benchmark: v9-02l
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/v9-02/v9-02m.tar.gz
Metrics
Global mean OH (from log file): 13.6057246332902 x 105 molec/cm3
Methyl chloroform lifetime: 4.5559 years
Did either of these change by more than 5%? No. The mean OH differs by -0.047%, and the MCF lifetimes differ by 0.053%.
At the SURFACE, list all species that changed by 10% or more: ISOP, MVK, MACR, PMN, NIT, ISOPN, MOBA, RIP, IEPOX
Comments on SURFACE differences:
  • The species listed above show differences mainly over the oceans or Antarctica, where the large ratios can be attributed to small number differences where concentrations are very low.
  • Differences in NIT can be attributed to numerical noise from ISORROPIA.
At 500 hPa, list all species that changed by 10% or more: ISOP, MVK, NH3, NIT, ISOPN, MOBA, RIP
Comments on 500 hPa differences:
  • The species listed above show differences mainly over the oceans or Antarctica, where the large ratios can be attributed to small number differences where concentrations are very low.
  • Differences in NH3 and NIT can be attributed to numerical noise from ISORROPIA.
In the ZONAL MEAN differences, list all species that changed by 10% or more: ALK4, ISOP, MVK, PMN, PRPE, C3H8, NH3, NIT, NITs, BCPO, OCPO, ISOPN, RIP
Comments on ZONAL MEAN differences:
  • Differences in most of the species listed above can be attributed to small number differences where concentrations are very low.
  • Differences in NH3 and NIT can be attributed to numerical noise from ISORROPIA.
In the EMISSION RATIO maps, list all species that changed by 10% or more: SO2, SO4
Comments on EMISSION RATIO differences:
  • The update for annual anthropogenic scale factors is responsible for:
    • Differences in SO2 anthropogenic emissions over Alaska and SO2 ship emissions along the coasts of US and Canada.
    • Differences in SO4 anthropogenic emissions over the continental US and Alaska
Additional or summary comments:
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 30 Jul 2013

--Melissa Sulprizio 10:25, 24 July 2013 (EDT)

v9-02l

Here is the assessment form for 1-month benchmark simulation v9-02l.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • NH3 seasonal scaling: Lin Zhang (Peking University)
  • Updated CAC NH3: Wai-Ho Lo (Dalhousie)
  • Wetdep fix: Fabien Paulot (Harvard)
  • DRYACET fix: Emily Fischer (Harvard)
  • Diurnal NOx scale factor fix: Christoph Keller (Harvard)
  • Double counting of biofuels fix: Christoph Keller (Harvard)
  • 3-hr GFED3 fix: Christoph Keller (Harvard)
  • Fix in grid_mod.F90: Christoph Keller (Harvard)
  • Scale factor fix: Jenny Fisher (U. Wollongong)
  • Ship emission fixes: Chris Holmes (UC Irvine)
  • Ship CO fix: Chris Holmes (UC Irvine)
  • VOLCPRESS fix: Thibaud Thonat (U. Edinburgh)
  • Removal of array temporaries: GEOS-Chem Support Team
  • Day-of-week fix: GEOS-Chem Support Team
  • Fixes in diag48_mod.F: Jenny Fisher (U. Wollongong)
  • Offline aerosol simulation fix: GEOS-Chem Support Team
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Tue Jun 18 20:40:16 EDT 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-10.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:35
  • Scalability: 7.3577
Compared to previous benchmark: v9-02k
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/v9-02/v9-02l.tar.gz
Metrics
Global mean OH (from log file): 13.6120550574916 x 105 molec/cm3
Methyl chloroform lifetime: 4.5535 years
Did either of these change by more than 5%? No. The mean OH differs by 1.74%, and the MCF lifetimes differ by -2.29%.
At the SURFACE, list all species that changed by 10% or more: NO, O3, PAN, ALK4, ISOP, HNO3, H2O2, MEK, ALD2, RCHO, MVK, MACR, PMN, PPN, R4N2, PRPE, C3H8, CH2O, C2H6, N2O5, HNO4, MP, DMS, SO2, SO4, SO4s, MSA, NH3, NH4, NIT, NITs, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, ISOPN, MOBA, PROPNN, HAC, GLYC, MMN, RIP, IEPOX, MAP, NO2, NO3, HNO2, OH, HO2
Comments on SURFACE differences:
At 500 hPa, list all species that changed by 10% or more: NO, ISOP, HNO3, MVK, MACR, PMN, PRPE, N2O5, HNO4, MP, DMS, SO2, MSA, NH3, NH4, NIT, NITs, Br, HOBr, BrNO2, BrNO3, CH2Br2, CH3Br, MPN, ISOPN, MOBA, PROPNN, HAC, GLYC, MMN, RIP, IEPOX, NO2, NO3, HNO2, OH
Comments on 500 hPa differences:
  • Some species only have a few grid boxes having significant differences at 500 hPa.
In the ZONAL MEAN differences, list all species that changed by 10% or more: NO, O3, PAN, CO, ALK4, ISOP, HNO3, H2O2, ALD2, RCHO, PPN, R4N2, PRPE, C3H8, CH2O, C2H6, N2O5, HNO4, MP, DMS, SO2, SO4, SO4s, MSA, NH3, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, CHBr3, HAC, GLYC, MMN, RIP, IEPOX, MAP, NO2, NO3, HNO2, OH, HO2
Comments on ZONAL MEAN differences:
  • The following species have percent differences larger than 10%, but very small absolute differences: NO, O3, PAN, ALK4, ISOP, ALD2, RCHO, MVK, MACR, PMN, R4N2, C3H8, CH2O, C2H6, N2O5, SO2, SO4, SO4s, MSA, NH3, NH4, NIT, NITs, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, CHBr3, MPN, MPN, ISOPN, PROPPNN, GLYC, IEPOX, NO2, NO3, HNO2, and OH.
In the EMISSION RATIO maps, list all species that changed by 10% or more:
  1. anthropogenic CO (along coastlines and over China)
  2. anthropogenic NH3 (over Canada)
  3. anthropogenic NO (mostly over oceans)
  4. anthropogenic SO2 and SO4 (over China)
  5. lightning NO (slight decrease over Russia)
Comments on EMISSION RATIO differences:
  1. The differences in anthropogenic CO over oceans are caused by the bug fix for ship CO emissions.
  2. The differences in anthropogenic NH3 are caused by the updated Canadian NH3 emissions inventory.
    • Wai-Ho Lo wrote: In my old report, it shows lower emission for the new CAC NH3 emission for Canada in Alberta. The rest of the country barely has any NH3 emission to begin with so the ratio could be unreliable. The benchmark test ratios seems correct.
    • The updates of the US NH3 emission seasonality mostly increase NH3 emissions in spring; the change in July (the benchmark month) is minor (<1%). In the previous code, the seasonal scaling factors are further scaled to the August value: SCALE(ThisMonth)/SCALE(August). Now we do not scale it for NH3.
  3. The differences in anthropogenic NO over oceans are caused by the fix for bug fix for PARANOX ship plume model when using non-local PBL mixing.
    • The NO emissions ratio over oceans should be 3.3, but the colorbar saturates at 2.0.
    • There are small, but non-zero ship emissions almost everywhere over the ice-free oceans.
    • The ICOADS-AMVER inventory uses real ship positions reported via GPS transponders.
  4. The decrease in CO, SO2, and SO4 emissions over China can be attributed to the bug fix for anthropogenic scaling factors for years 2006 and later. The base year for Streets emissions in the simulation is 2006, so these emissions are affected by the fix for annual scaling factors.
  5. About the decrease in lightning NO over Russia, Chris Holmes writes:
    • A small portion (1-3%) of lightning and aircraft NOx emissions would occur in the stratosphere, but this is not allowed in routine setemis.F. Now, these stratospheric emissions are also excluded from the ND32 diagnostics (NOx sources).
Additional or summary comments:
  1. The computation of the day of week w/r/t local time and the computation of diurnal NOx scale factors w/r/t local time may change the emissions totals slightly, but not to the 10% level. The base emissions remain the same, but the time of day used to scale emissions will now differ from Greenwich Mean Time (GMT).
    • For example, the EPA/NEI US emissions specify "weekday" or "weekend" emissions. In versions prior to v9-02l, the day of week was computed w/r/t GMT instead of local times. Since the local time in the USA lags 5-8 hours behind GMT, this means that we were transitioning from "weekday" to "weekend" 5-8 hours sooner than we should have been. This is now corrected.
  2. The fixes to remove array temporaries do not change any scientific results.
  3. Rigorous tests were performed to ensure that we obtained identical results when OpenMP parallelization was turned on and off. This indicates that we did not introduce any parallelization errors while adding these updates to GEOS-Chem.
Approval
Requires further investigation: No
Approved by: Jintai Lin, Chris Holmes, Lin Zhang, Jenny Fischer, Christoph Keller, Wai-Ho Lo
Date of approval: 26 Jun 2013

--Melissa Payer 16:48, 26 June 2013 (EDT)

v9-02k

Here is the assessment form for 1-month benchmark simulation v9-02k.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Thu Jun 6 14:38:12 EDT 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-11.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:30
    • NOTE: 1/2 hour faster than v9-02j
  • Scalability: 7.4285
    • NOTE: Scales much better than v9-02j (6.2)
Compared to previous benchmark: v9-02j
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 v9-02j
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-02/v9-02k.tar.gz
Metrics
Global mean OH (from log file): 13.3796050179548 x 105 molec/cm3
Methyl chloroform lifetime: 4.6600 years
Did either of these change by more than 5%? No. They are identical to the v9-02j values.
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: None
Comments on ZONAL MEAN differences:
In the EMISSION RATIO maps, list all species that changed by 10% or more: None
Comments on EMISSION RATIO differences:
Additional or summary comments:
  1. The v9-02k output file (aka ctm.bpch.v9-02k) is 100% binary identical to the v9-02j output file (ctm.bpch.v9-02j). This indicates that the full-chemistry simulation in v9-02k has not been impacted by the structural changes that we have made.
  2. In v9-02k, we have segregated all of the places where GEOS-Chem interacts with TOMAS. If TOMAS is used, then these sections of code are activated; otherwise they are ignored. TOMAS was not turned on during the v9-02k 1-month benchmark simulation. Sal Farina will continue to develop and validate the TOMAS simulation in GEOS-Chem.
  3. A separate Hg benchmark simulation (1-year) is being performed by Team Hg. Results are forthcoming. Because we have obtained identical results in the 1-month full-chemistry benchmarks for v9-02k and v9-02j, we recommend that v9-02k be approved immediately.. This will allow the SEAC4RS Team to adopt v9-02k as their base code. We feel that further fixes to the Hg simulation can be added into the next benchmark (v9-02l) if necessary.
  4. IMPORTANT NOTE: v9-02k scales much better than previous versions. We have shaved 1/2 hour per model month off of the wall clock time at 4° x 5° resolution. This translates to a speedup of 6 hours per model year at 4° x 5°. We attribute this speedup to our rewriting of inefficient subroutine calls. We also anticipate that GEOS-Chem will scale better at 2° x 2.5° global resolution, as well as for all nested grids (0.5° x 0.666° and 0.25° x 0.3125°).
Approval
Requires further investigation:
  • Full-chemistry simulation: NO
  • Hg simulation: We shall ask "Team Hg" to review the 1-year Hg benchmark when it is completed.
Approved by: Bob Y.
Date of approval: 07 Jun 2013

--Bob Y. 10:58, 7 June 2013 (EDT)

v9-02j

Here is the assessment form for 1-month benchmark simulation v9-02j.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Jingqiu Mao (NOAA/GFDL)
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri May 24 10:41:55 EDT 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-11.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 3:05
  • Scalability: 6.2866
Compared to previous benchmark: v9-02i
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/v9-02/v9-02j.tar.gz
Metrics
Global mean OH (from log file): 13.3796050179548 x 105 molec/cm3
Methyl chloroform lifetime: 4.6600 years
Did either of these change by more than 5%? No. The mean OH differs by -1.22%, and the MCF lifetimes differ by 2.19%.
At the SURFACE, list all species that changed by 10% or more: NO, O3, PAN, ALK4, ISOP, HNO3, H2O2, MEK, ALD2, RCHO, MVK, MACR, PMN, PPN, R4N2, PRPE, C3H8, CH2O, C2H6, N2O5, HNO4, MP, DMS, SO2, SO4, SO4s, NH3, NH4, NIT, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, MPN, ISOPN, MOBA, PROPNN, HAC, GLYC, MMN, RIP, IEPOX, MAP, NO2, NO3, HNO2, OH, HO2
Comments on SURFACE differences: mostly driven by the change in OH.
At 500 hPa, list all species that changed by 10% or more: NO, ALK4, ISOP, H2O2, ALD2, RCHO, MVK, MACR, PMN, PPN, R4N2, PRPE, C3H8, N2O5, HNO4, DMS, SO2, SO4s, NH3, NIT, Br2, Br, BrO, BrNO3, ISOPN, MOBA, PROPNN, HAC, GLYC, MMN, RIP, IEPOX, MAP, NO2, NO3, HNO2, OH, HO2
Comments on 500 hPa differences: mostly driven by the change in OH.
In the ZONAL MEAN differences, list all species that changed by 10% or more: NO, PAN, ALK4, ISOP, H2O2, MEK, ALD2, RCHO, MVK, MACR, PMN, PPN, R4N2, PRPE, C3H8, CH2O, N2O5, HNO4, MP, DMS, SO2, SO4, SO4s, NH3, NIT, NITs, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, MPN, ISOPN, MOBA, PROPNN, HAC, GLYC, MMN, RIP, IEPOX, MAP, NO2, NO3, HNO2, OH, HO2
Comments on ZONAL MEAN differences: mostly driven by the change in OH.
In the EMISSION RATIO maps, list all species that changed by 10% or more: N/A
Comments on EMISSION RATIO differences: N/A
Additional or summary comments: We recommend gamma_HO2 = 0.2 (producing H2O) for the standard model, although gamma_HO2 could be potentially high than 0.2.
Approval
Requires further investigation: No
Approved by: Jingqiu Mao
Date of approval: 05/28/2013

--Melissa Payer 17:09, 20 May 2013 (EDT)

v9-02i

Here is the assessment form for 1-month benchmark simulation v9-02i.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Lin Zhang (Peking University)
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri May 17 10:16:11 EDT 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-11.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 3:06
  • Scalability: 6.2832
Compared to previous benchmark: v9-02h
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/v9-02/v9-02i.tar.gz
Metrics
Global mean OH (from log file): 13.5441691420451 x 105 molec/cm3
Methyl chloroform lifetime: 4.5603 years
Did either of these change by more than 5%? No. The mean OH differs by 0.23%, and the MCF lifetimes differ by -0.23%.
At the SURFACE, list all species that changed by 10% or more:
Comments on SURFACE differences:
At 500 hPa, list all species that changed by 10% or more:
Comments on 500 hPa differences:
In the ZONAL MEAN differences, list all species that changed by 10% or more:
Comments on ZONAL MEAN differences:
In the EMISSION RATIO maps, list all species that changed by 10% or more:
Comments on EMISSION RATIO differences:
Additional or summary comments:
Approval
Requires further investigation: No
Approved by: Lin Zhang, Mat Evans
Date of approval: 17 May 2013

--Melissa Payer 10:52, 16 May 2013 (EDT)

v9-02h

Here is the assessment form for 1-month benchmark simulation v9-02h.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri May 10 12:35:12 EDT 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-09.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 3:04
  • Scalability: 6.2645
Compared to previous benchmark: v9-02g
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/v9-02/v9-02h.tar.gz
Metrics
Global mean OH (from log file): 13.5130950197591 x 105 molec/cm3
Methyl chloroform lifetime: 4.5706 years
Did either of these change by more than 5%? No. The mean OH differs by 0.52%, and the MCF lifetimes differ by -0.93%.
At the SURFACE, list all species that changed by 10% or more: NO/NOx, O3/Ox, ISOP, HNO3, MVK, MACR, PMN, PPN, PRPE, N2O5, HNO4, DMS, NH4, NIT, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, MPN, ISOPN, MOBA, HAC, GLYC, MMN, RIP, IEPOX, OH, HO2
Comments on SURFACE differences: The removal of NOx-Ox partitioning has resulted in less NO with respect to NOx. O3 remains unchanged with respect to Ox, except for in the Antarctic, where O3 is larger than Ox by 3 ppb or less. This change is primarily due to the restart file. A restart file was spun up for the 5 new tracers and combined with the existing benchmark restart file, resulting in slightly different initial conditions for O3 with respect to Ox. The introduction of the new tracers has also caused slight differences in N2O5 over source regions.

Differences in NH4 and NIT can be attributed to numerical noise from ISORROPIA.

Differences in the remaining species primarily occur over the oceans or over Antarctica and can be attributed to small number differences where concentrations are very low.

At 500 hPa, list all species that changed by 10% or more: NO/NOx, ISOP, MVK, MACR, PMN, PRPE, N2O5, DMS, NH3, NIT, Br, BrO, BrNO3, MPN, ISOPN, MOBA, HAC, GLYC, MMN, RIP, IEPOX, OH, HO2
Comments on 500 hPa differences: The removal of NOx-Ox partitioning has resulted in less NO with respect to NOx.

Differences in NH3 and NIT can be attributed to numerical noise from ISORROPIA.

Differences in the remaining species primarily occur over the oceans or over Antarctica and can be attributed to small number differences where concentrations are very low.

In the ZONAL MEAN differences, list all species that changed by 10% or more: NO/NOx, O3/Ox, ALK4, ISOP, HNO3, H2O2, MVK, MACR, PMN, PRPE, C3H8, N2O5, HNO4, MP, DMS, SO4s, NH3, NIT, NITs, Br, BrO, HBr, BrNO2, BrNO3, MPN, ISOPN, MOBA, HAC, GLYC, MMN, RIP, IEPOX, MAP, OH, HO2
Comments on ZONAL MEAN differences: Differences in NO/NOx and O3/Ox can be attributed to the removal of NOx-Ox partitioning.

Differences in NH3 and NIT can be attributed to numerical noise from ISORROPIA.

Differences in the remaining species primarily occur above 200 hPa or near the poles, where concentrations are low and small changes produce large % difference.

In the EMISSION RATIO maps, list all species that changed by 10% or more: None
Comments on EMISSION RATIO differences: N/A
Additional or summary comments: Please note that family tracers NOx and Ox have been removed in this version and replaced with tracers NO, O3, NO2, NO3, and HNO2. Therefore, the first two panels on several of the benchmark plots (i.e. differences, ratios, zonal differences) compare NO and O3 from the current version with NOx and Ox from the previous version (v9-02g). For concentration maps of the five new tracers, see http://ftp.as.harvard.edu/pub/exchange/mpayer/1mo_quick_look/v9-02h/v9-02h.concentrations.pdf (NO and O3 are on page 1; NO2, NO3, and HNO2 are on page 22).
Approval
Requires further investigation: No
Approved by: Melissa Payer, Jingqiu Mao
Date of approval: 15 May 2013

--Melissa Payer 16:26, 16 April 2013 (EDT)

v9-02g

Here is the assessment form for 1-month benchmark simulation v9-02g.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Isoprene scheme: Fabien Paulot (Harvard), Eloise Marais (Harvard), Jingqiu Mao (NOAA/GFDL), Mat Evans (York)
  • RO2+HO2 rate: Fabien Paulot (Harvard)
  • GEOS-5.7.2 fix:: GEOS-Chem Support Team
  • Streets fix: Jenny Fisher (U. Wollongong)
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Wed Mar 20 20:11:47 EDT 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-08.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:58
  • Scalability: 6.3565
Compared to previous benchmark: v9-02f
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/v9-02/v9-02g.tar.gz
Metrics
Global mean OH (from log file): 13.4422368196455 x 105 molec/cm3
Methyl chloroform lifetime: 4.6134 years
Did either of these change by more than 5%? No. The mean OH differs by 3.73%, and the MCF lifetimes differ by -4.67%.
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, N2O5, HNO4, MP, DMS, MSA, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, MPN, OH
Comments on SURFACE differences: Differences in NOx and Ox are mainly due to the treatment of isoprene nitrates. Higher NOx leads to higher OH in northern mid-latitude. Higher OH in tropics is due to isomerization of RIO2. The changes in other species are mainly due to the changes in OH and NOx.
At 500 hPa, list all species that changed by 10% or more: NOx, PAN, ALK4, ISOP, HNO3, H2O2, MEK, ALD2, RCHO, MVK, MACR, PMN, PPN, R4N2, PRPE, C3H8, N2O5, HNO4, MP, DMS, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, MPN, OH
Comments on 500 hPa differences: Same as surface plots.
In the ZONAL MEAN differences, 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, N2O5, HNO4, MP, DMS, MSA, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, MPN, OH
Comments on ZONAL MEAN differences: Same as surface plots.
In the EMISSION RATIO maps, list all species that changed by 10% or more: No
Comments on EMISSION RATIO differences: N/A
Additional or summary comments: I also increase gamma(NO3) to 0.1, according to recent literature.
Approval
Requires further investigation: TBD
Approved: Jingqiu Mao
Date of approval: 03/22/2013

--Melissa Payer 12:17, 21 March 2013 (EDT)

v9-02f

Here is the assessment form for 1-month benchmark simulation v9-02f.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Inorganic chemistry updates: Mat Evans (York)
  • MPN chemistry: Ellie Browne (UC Berkeley)
  • Nested Hg bug fixes: Yanxu Zhang (U. Washington)
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Wed Jan 30 16:36:48 EST 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-10.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:32
  • Scalability: 6.3256
Compared to previous benchmark: v9-02e
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/v9-02/v9-02f.tar.gz
Metrics
Global mean OH (from log file): 12.9589397348900 x 105 molec/cm3
Methyl chloroform lifetime: 4.8396 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, PMN

N2O5

HNO4

DMS, MSA

Br, BrO, Other bromine compounds

Comments on SURFACE differences: Isoprene related tracers: Concentrations are changing over remote oceanic regions reflecting changes in the initial oxidation steps of these compounds.

N2O5 changes are again is restricted to remote ocean regions. Changes to the NO3+DMS rate constant have made this faster which will tend to reduce NO3 and N2O5 concentrations in regions where DMS is the dominant VOC

DMS, MSA Concentrations are changing due to changes in the DMS+OH and DMS+NO3 rate constants. At the surface changes are small and restricted to continental regions.

Br compounds. These are restricted to the polar regions where concentrations are low and reflects the combination of other changes of longer lived species. The same is the case for HNO3.

At 500 hPa, list all species that changed by 10% or more: NOx

ISOP, MVK, MACR, PMN

N2O5

HNO4

PRPE

DMS, MSA

NIT, NH4

Br species

Comments on 500 hPa differences: The largest changes here are to the DMS and MSA. This reflects the changes in the DMS+OH and DMS+NO3 oxidation routes which essentially both changes the location of DMS oxidation and the branching ratio of the DMS Oxidation

isoprene chemistry changes again reflect changes in the primary oxidation of isoprene and the subsquent branching ration

NIT, NH4 Isoropia seems unstable at cold temperatures and small changes leads to large relative changes in these compounds even though they are at low concentrations.

N2O5, HNO4, Br species see above.

In the ZONAL MEAN differences, list all species that changed by 10% or more: NOx

ISOP, MVK, MACR, PMN

PRPE

CH2O

N2O5

HNO4

MP

DMS, MSA, SO2

Br compounds

Comments on ZONAL MEAN differences: NOx, HNO3, H2O2, CH2O, MP Reflect that changes seen in Browne et al., 2011

DMS, MSA, SO2 reflect changes in the initial DMS oxidation of DMS+OH and DMS+NO3

ISOP, MACR, MVK, PMN changes reflect changes in the ISOP oxidation ISOP+OH, O3, NO3

The N2O5 decrease at ~200 hPa in the zonal mean is also consistent with Browne et al., 2011

In the EMISSION RATIO maps, list all species that changed by 10% or more: No
Comments on EMISSION RATIO differences: N/A
Additional or summary comments: This all seems fine. In general changes are small in absolute concentrations but can be relatively large fractionally.
Approval
Requires further investigation: No
Approved: MJE as long as others are happy.
Date of approval: 07 Feb 2013

--Melissa Payer 09:31, 1 February 2013 (EST)

v9-02e

Here is the assessment form for 1-month benchmark simulation v9-02e.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri Jan 4 17:01:07 EST 2013
Performance statistics:
  • Ran on 8 CPUs of bench@titan-10.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:32
  • Scalability: 6.3584
Compared to previous benchmark: v9-02d
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/v9-02/v9-02e.tar.gz
Metrics
Global mean OH (from log file): 12.9399839759816 x 105 molec/cm3
Methyl chloroform lifetime: 4.8585 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: ISOP, NIT
Comments on SURFACE differences: Differences are attributed to the numerical noise issue in ISORROPIA II.
At 500 hPa, list all species that changed by 10% or more: ISOP, NH3, NIT
Comments on 500 hPa differences: Differences are attributed to the numerical noise issue in ISORROPIA II.
In the ZONAL MEAN differences, list all species that changed by 10% or more: ALK4, PMN, PRPE, C3H8, SO4s, NH3, NIT, NITs
Comments on ZONAL MEAN differences: Differences are seen in ALK4, PMN, PRPE, C3H8, SO4s, NH3, and NITs are in locations where concentrations are already very low. These differences can be attributed to numerical noise.

Differences in NH3 and NITs are attributed to the numerical noise issue in ISORROPIA II.

In the EMISSION RATIO maps, list all species that changed by 10% or more: None
Comments on EMISSION RATIO differences: The bug fix for regridding of Streets anthropogenic emissions caused CO anthropogenic emissions to change by 0.004 Tg (~0.01%).
Additional or summary comments:
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 07 Jan 2013

--Melissa Payer 16:01, 7 January 2013 (EST)

v9-02d

Here is the assessment form for 1-month benchmark simulation v9-02d.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Wed Dec 19 09:33:39 EST 2012
Performance statistics:
  • Ran on 8 CPUs of bench@titan-08.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:32
  • Scalability: 6.3638
Compared to previous benchmark: v9-02c
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 expected to see no changes from v9-02d. We have only made structural modifications but not scientific modifications.
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-02/v9-02d.tar.gz
Metrics
Global mean OH (from log file): 12.9397354478399 x 105 molec/cm3
Methyl chloroform lifetime: 4.8586 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: Differences in NIT are attributed to the numerical noise issue in ISORROPIA II.
At 500 hPa, list all species that changed by 10% or more: ISOP, NH3, NIT
Comments on 500 hPa differences: Differences are seen in ISOP in locations where concentrations are already very low (i.e. over oceans). These differences are attributed to the bromine bug fix.

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: ALK4, ISOP, PRPE, C3H8, SO4s, NH3, NIT, NITs, Br2,
Comments on ZONAL MEAN differences: Differences are seen in ALK4, ISOP, PRPE, C3H8, SO4s, and Br2 are in locations where concentrations are already very low. These differences are attributed to the bromine bug fix.

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: None
Comments on EMISSION RATIO differences:
Additional or summary comments: When ISORROPIA is turned off, and when the Bug fix for Br2 emissions is also added to v9-02c, then we obtain identical results between v9-02c and v9-02d. This demonstrates that the structural modifications made for grid independence do not change GEOS-Chem's scientific results. Most of these modifications were made to carry data fields in derived-type objects rather than in module arrays, but the data fields themselves have not changed.
Approval
Requires further investigation: No
Approved: Yes
Date of approval: 19 Dec 2012

--Bob Y. 10:39, 19 December 2012 (EST)

v9-02c

Here is the assessment form for 1-month benchmark simulation v9-02c.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • NEI05 Hg emissions: Yanxu Zhang (U. Washington); Lyatt Jaeglé (U. Washington)
  • Nested-grid Hg: Yanxu Zhang (U. Washington); Lyatt Jaeglé (U. Washington)
  • Streets future Hg emissions: Bess Corbitt (Harvard)
  • Updated Hg(0) oxidation kinetics: Bess Corbitt(Harvard); Justin Parrella (Harvard)
  • GEOS-Chem Br/Br fields in Hg simulation: Bess Corbitt(Harvard); Justin Parrella (Harvard)
  • POPs simulation: Carey Friedman (MIT); Noelle Selin (MIT)
  • CAC bug fix: Aaron van Donkelaar (Dalhousie)
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Wed Nov 28 22:37:00 EST 2012
Performance statistics:
  • Ran on 8 CPUs of 8 CPUs of bench@titan-08.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:20
  • Scalability: 6.5218
Compared to previous benchmark: v9-02b
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 v9-02b
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-02/v9-02c.tar.gz
Metrics
Global mean OH (from log file): 12.9398281506753 x 105 molec/cm3
Methyl chloroform lifetime: 4.8585 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: None
Comments on ZONAL MEAN differences:
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: 29 Nov 2012

--Melissa Payer 16:58, 27 November 2012 (EST)

v9-02b

Here is the assessment form for 1-month benchmark simulation v9-02b.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Soil NOx: Rynda Hudman Kay (EPA); Neil Moore (formerly of Dalhousie); Bram Maasakkers (Eindhoven)
  • Bug fix in hdf_mod.F: Sajeev Philip (Dalhousie)
  • Tagged CO simulation fixes: Jenny Fisher (Wollongong); GEOS-Chem Support Team
  • Bug fix in nei_anthro_mod.F: Sajeev Philip (Dalhousie)
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Wed Oct 24 16:10:25 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
  • Scalability: 6.5256
Compared to previous benchmark: v9-02a
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): Soil NOx emissions
Plots may be viewed at: ftp://ftp.as.harvard.edu/gcgrid/geos-chem/1mo_benchmarks/v9-02/v9-02b.tar.gz
Metrics
Global mean OH (from log file): 12.9398281506753 x 105 molec/cm3
Methyl chloroform lifetime: 4.8585 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, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, and OH
Comments on SURFACE differences: The changes are as expected. The new soil NOx parameterization significantly increases the NOx concentrations over North America, the southern part of South America, Central Asia and parts of Africa. Decreases occur over the northern part of South America, central Africa and Indonesia.

Other species are influenced by resulting changes in O3 (and consequently OH). However, some of the reported changes are mainly caused by noise since the absolute differences are very close to zero.

At 500 hPa, list all species that changed by 10% or more: NOx, PAN, ALK4, ISOP, HNO3, MEK, ALD2, RCHO, MVK, MACR, PMN, PPN, PRPE, N2O5, HNO4, MP, DMS, SO2, MSA, NH3, NIT, NITs, Br2, Br, BrO, HOBr, HBr, BrNO2, BrNO3, and OH
Comments on 500 hPa differences: See 'Comments on SURFACE differences'
In the ZONAL MEAN differences, list all species that changed by 10% or more: NOx, PAN, ISOP, HNO3, MEK, RCHO, MVK, MACR, PMN, PPN, PRPE, N2O5, HNO4, DMS, SO4s, NH3, NIT, NITs, Br, BrO, BrNO2, BrNO3 and OH.
Comments on ZONAL MEAN differences: See 'Comments on SURFACE differences'
In the EMISSION RATIO maps, list all species that changed by 10% or more: NOx fertilizer emissions and NOx soil emissions
Comments on EMISSION RATIO differences: The NOx soil emissions diagnostic now includes the NOx fertilizer emissions since the emissions related to fertilizer are now calculated using the same procedure as the “regular” soil emissions based on the biome emission coefficients. The soil NOx emission total increases with 0.310 Tg N to 1.028 Tg N for July.

The fertilizer emissions diagnostics gives the soil NOx emissions related to fertilizer use and deposited nitrogen.

Additional or summary comments: The occurring changes are as expected. The benchmark is performed using a new soil NOx restart file which saves per grid cell:
  • The Dry Period (hours)
  • Pulsing factor (unitless)
  • GWET from the previous timestep (unitless)
  • N reservoir from dry/wet deposition (ngN/m2)

The overall changes are in NOx emissions are consistent with those in Hudman et al. (2012) with the additional canopy reduction that Bram added. The changes in O3 and OH follow the changes in NOx emissions.

Approval
Requires further investigation: No
Approved: Randall Martin, Bram Maasakkers
Date of approval: 29 Oct 2012

--Melissa Payer 13:42, 24 October 2012 (EDT)

v9-02a

Here is the assessment form for 1-month benchmark simulation v9-02a.

Description
New features added into GEOS-Chem:
Developer name(s) and institution(s):
  • Sajeev Philip (Dalhousie)
Version, resolution, met fields used: v9-02, GEOS-5 (47L), 4x5
1-month benchmark finished on: Fri Oct 12 17:58:19 EDT 2012
Performance statistics:
  • Ran on 8 CPUs of bench@titan-10.as.harvard.edu (2.659 GHz x 8 CPU)
  • Wall time: 2:19
  • Scalability: 6.5621
Compared to previous benchmark: v9-01-03-Fix_for_IFORT12
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/v9-02/v9-02a.tar.gz
Metrics
Global mean OH (from log file): 12.8727104044567 x 105 molec/cm3
Methyl chloroform lifetime: 4.8983 years
Did either of these change by more than 5%? No
At the SURFACE, list all species that changed by 10% or more: All species
Comments on SURFACE differences: Expected. Change in boundary layer mixing
At 500 hPa, list all species that changed by 10% or more: DST4, NITS, NH3, DMS, PRPE, MACR, ISOP (over Ocean) and NIT (over Land)
Comments on 500 hPa differences: Change in boundary layer mixing
In the ZONAL MEAN differences, list all species that changed by 10% or more: All species
Comments on ZONAL MEAN differences: Change in boundary layer mixing
In the EMISSION RATIO maps, list all species that changed by 10% or more: None
Comments on EMISSION RATIO differences: None
Additional or summary comments: None
Approval
Requires further investigation: No
Approved: Yes
Date of approval: Approved 16 Oct 2012

--Melissa Payer 12:00, 15 October 2012 (EDT)

Special benchmarks

v9-02r Rn-Pb-Be

Karen Yu performed 1-year Rn-Pb-Be simulations using GEOS-Chem v9-02r to evaluate the GEOS-FP met fields. The simulations were done for June 2012–May 2013, with a 2-month spinup. For comparison of the Pb-210 and Be-7 budgets to previous versions, please see the following posts on the Rn-Pb-Be simulation wiki page:

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

You may view the benchmark plots for the simulation by pointing your browser to:

http://ftp.as.harvard.edu/gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02r/geos5/RnPbBe/output/pdf/

and

http://ftp.as.harvard.edu/gcgrid/geos-chem/1yr_benchmarks/v9-02/v9-02r/geosfp/RnPbBe/output/pdf/

--Melissa Sulprizio 13:41, 20 November 2013 (EST)

v9-02o SOA

One-month benchmark simulations were performed for v9-02n using the traditional SOA simulation and for v9-02o using the updated SOA simulation. The optional semivolatile POA and IVOC features were turned off in the updated SOA simulation. The standard 1-month benchmark plots comparing these simulations were created to evaluate tracers 1-66. The benchmark plots may be viewed at:

   http://ftp.as.harvard.edu/pub/exchange/mpayer/1mo_quick_look/v9-02o_SOA/output/

Jeff Pierce wrote:

I've gone through and compared the SOA and SOG tracer concentrations for the simulations using Havala's SOA scheme vs. the old-school SOA scheme. Everything seems to make sense including (1) the partitioning between SOA and SOG for the various volatility bins, (2) the spatial dependence of the terpene-based, isoprene-based and anthro-based SOA, and (3) the magnitude of the concentrations compared to the old-school scheme. There may be some subtleties that Havala may be able to spot, but there was nothing I was able to see that looked odd. There seems to be no impact on the other species in the ratio and difference maps.

Havala Pye wrote:

I concur with Jeff. The v9-02o benchmark looks good in terms of magnitude and spatial variability.

--Melissa Sulprizio 14:42, 13 September 2013 (EDT)