GEOS–Chem: Benchmark Simulations

GEOS–Chem
Benchmark Simulations

Last Updated April 4, 2006

Each release of the GEOS–Chem model is benchmarked with a standard 1 month oxidant–aerosol chemistry simulation before it is distributed to users worldwide. The results from each benchmark simulation are compared to results from the previous GEOS–Chem release (e.g. Version 5.02 is tested against Version 5.01). In this way, potential errors in coding or in algorithm design can be spotted easily.

An optional 1 month ²²²Rn–²¹⁰Pb–⁷Be simulation may also be performed to test transport, convective, and wet deposition process.

As of summer 2003, certain versions of GEOS–Chem are benchmarked with a 1 year oxidant–aerosol chemistry simulation. The results of this simulation are compared to observations as well as to 1 year benchmarks of previous versions. This allows both model-model and model-observation comparisons to be made.

Prof. Daniel Jacob is currently responsible for examining the results of the GEOS–Chem benchmark simulations.

1 month Oxidant–Aerosol Chemistry Benchmark
with coupled sulfate aerosol chemistry

Dates of Run:	July 1, 2001 to August 1, 2001
Vertical Resolution:	GEOS-3 with 30 sigma levels. (GEOS-3 normally has 48 sigma levels; we vertically bunch levels 22 and higher into groups of 2 or 4 levels in order to gain computational advantage.)
Horizontal Resolution:	4^° latitude x 5° longitude
Chemical tracers:	Gas-phase tracers: NO_x, O_x, PAN, CO, Lumped >=C4 Alkanes, ISOP, HNO₃, H₂O₂, MEK, Acetaldehyde, Lumped Aldehyde >=C3, MVK, Methacrolein, PMN, PPN, Lumped Alkyl Nitrate, >=C3 Alkenes, C₃H₈, HCHO, C₂H₆, N₂O5, HNO₄, Methyl Hydro Peroxide, DMS, SO2, SO4, MSA, NH3, NH4, NIT, Hydrophilic BC, Hydrophilic OC, Hydrophobic BC, Hydrophobic OC, Dust Aerosol (4 size bins), Sea Salt Aerosol (2 size bins) Aerosol tracers: DMS, SO2, SO4, MSA, NH3, NH4, NIT, Hydrophilic BC, Hydrophilic OC, Hydrophobic BC, Hydrophobic OC, Dust Aerosol (4 size bins), Sea Salt Aerosol (2 size bins)
Chemical Mechanism:	SMVGEAR II (Jacobson et al, 1997), using 90 chemical species, approximately 300 kinetic reactions (including rapid recycling of peroxides as well as H2SO4—HNO3—NH3 chemistry), and 51 photolysis reactions. View a detailed description of the SMVGEAR II chemistry mechanism in PDF format.
Photolysis Mechanism:	FAST-J algorithm (Wild et al, 2000)
Operations:	Transport (Lin & Rood, 1996, 1997 Cloud Convection (Lin, 1996) Boundary Layer Mixing (Allen, 1996) Emissions Dry Deposition (Jacob et al, 1995) Chemistry Wet Deposition (Jacob et al, 2000)
Diagnostic Output:	Optical Depths Noontime J(O₃) and J(NO₂) Transport fluxes (N/S, E/W, up/down) Biomass burning emission fluxes Biogenic emission fluxes Anthropogenic emission fluxes Dry deposition fluxes and velocities Surface Pressure Sources of CO Sources of NO_x Sources and sinks of biogenic acetone Concentrations of chemically produced OH and NO Tracer concentrations Chemical production & loss of O_x and CO Lifetime of Methylchloroform (CH₃CCl₃) DAO 3-D winds & temperatures Air masses, grid box heights All diagnostics above are saved as monthly means.

The following types of plots and summaries are created from the oxidant-aerosol simulation:

Budget of O_x and CO
Summary of total emissions (this version and previous version)
Differences of total emissions (this version - previous version)
Frequency distribution (this version - previous version) for tracers
Ratios (this version / previous version) of JO3, JNO2 at the surface
Ratios (this version / previous version) of JO3, JNO2 at 500 hPa
Ratios (this version / previous version) of tracers at the surface
Ratios (this version / previous version) of tracers at 500 hPa
Tracer concentration at the surface
Tracer concentration at 500 hPa
Zonal mean tracer concentration

1 month ²²²Rn–²¹⁰Pb–⁷Be Benchmark

Dates of Run:	January 1, 2001 to January 1, 2002
Vertical Resolution:	GEOS-3 with 48 sigma levels
Horizontal Resolution:	4° latitude x 5° longitude
Chemical tracers:	²²²Rn, ²¹⁰Pb, ⁷Be
Chemical Mechanism:	Radioactive decay (first-order loss)
Photolysis Mechanism:	None
Operations:	Transport (Lin & Rood, 1996, 1997) Cloud Convection (Lin, 1996) Boundary Layer Mixing (Allen, 1996) Emissions Dry Deposition (Jacob et al, 1995) Chemistry Wet Deposition (Jacob et al, 2000)
Diagnostic Output:	Source and decay terms Dry deposition fluxes and velocities Surface Pressure Tracer concentrations DAO 3-D winds & temperatures Air masses, grid box heights All diagnostics above are saved as monthly means.

The following types of plots and summaries are created from the ²²²Rn-²¹⁰Pb-⁷Be simulation:

Budgets of ²²²Rn, ²¹⁰Pb, ⁷Be
Frequency distribution (this version - previous version) for tracers
Ratios (this version / previous version) of tracers at the surface
Ratios (this version / previous version) of tracers at 500 hPa
Tracer concentration at the surface
Tracer concentration at 500 hPa
Zonal mean tracer concentration

1 year Oxidant–Aerosol Chemistry Benchmark
with coupled sulfate aerosol chemistry

Spin-up:	January 1, 2000 to January 1, 2001
Dates of Run:	January 1, 2001 to January 1, 2002
Vertical Resolution:	GEOS–4 with 30 hybrid levels OR GEOS–3 with 30 sigma levels. GEOS–3 normally has 48 sigma levels and GEOS–4 normally has 55 hybrid levels. We lump stratosphereic levels together in groups of 2 or 4 levels in order to gain computational advantage.
Horizontal Resolution:	4° latitude x 5° longitude
Chemical tracers:	Gas-phase tracers: NO_x, O_x, PAN, CO, Lumped >=C4 Alkanes, ISOP, HNO₃, H₂O₂, MEK, Acetaldehyde, Lumped Aldehyde >=C3, MVK, Methacrolein, PMN, PPN, Lumped Alkyl Nitrate, >=C3 Alkenes, C₃H₈, HCHO, C₂H₆, N₂O5, HNO₄, Methyl Hydro Peroxide, DMS, SO2, SO4, MSA, NH3, NH4, NIT, Hydrophilic BC, Hydrophilic OC, Hydrophobic BC, Hydrophobic OC, Dust Aerosol (4 size bins), Sea Salt Aerosol (2 size bins) Aerosol tracers: DMS, SO2, SO4, MSA, NH3, NH4, NIT, Hydrophilic BC, Hydrophilic OC, Hydrophobic BC, Hydrophobic OC, Dust Aerosol (4 size bins), Sea Salt Aerosol (2 size bins) Secondary Organic Aerosols (optional): ALPH, LIMO, ALCO, SOA1, SOA2, SOA3, SOG1, SOG2, SOG3.
Chemical Mechanism:	SMVGEAR II (Jacobson et al, 1997), using 90 chemical species, approximately 300 kinetic reactions (including rapid recycling of peroxides as well as H2SO4—HNO3—NH3 chemistry), and 51 photolysis reactions. View a detailed description of the SMVGEAR II chemistry mechanism in PDF format.
Photolysis Mechanism:	FAST-J algorithm (Wild et al, 2000)
Operations:	Transport (Lin & Rood, 1996, 1997) Cloud Convection GEOS–3: (Lin, 1996) GEOS–4: (Rasch, 2001) Boundary Layer Mixing (Allen, 1996) Emissions Dry Deposition (Jacob et al, 1995) Chemistry Wet Deposition (Jacob et al, 2000)
Diagnostic Output:	Optical Depths Noontime J(O₃) and J(NO₂) Biomass burning emission fluxes Biogenic emission fluxes Anthropogenic emission fluxes Dry deposition fluxes and velocities Surface Pressure Sources of CO Sources of NO_x Sources and sinks of biogenic acetone Concentrations of chemically produced OH & NO Tracer concentrations Chemical production & loss of O_x and CO Lifetime of Methylchloroform (CH₃CCl₃) DAO 3-D winds & temperatures Air masses, grid box heights All diagnostics above are saved as monthly means.

The following types of plots and summaries are created from the 1-year NO_x–O_x–hydrocarbon simulation:

C2H6 from 3 benchmark simulations vs. aircraft data
C3H8 from 3 benchmark simulations vs. aircraft data
CO from 3 benchmark simulations vs. aircraft data
H2O2 from 3 benchmark simulations vs. aircraft data
HNO3 from 3 benchmark simulations vs. aircraft data
NO from 3 benchmark simulations vs. aircraft data
O3 from 3 benchmark simulations vs. aircraft data
PAN from 3 benchmark simulations vs. aircraft data
Model CO vs ship track data
Model column CO from 3 benchmark simulations vs. observations
Model O3 from 3 benchmark simulations vs. ozonesonde data

http://www.as.harvard.edu/ctm/geos/geos_benchmark.html