Benchmark/GEOS-Chem 12.3.2: Difference between revisions

From Geos-chem
Jump to navigation Jump to search
No edit summary
Line 1: Line 1:
== 1-month benchmark ==
=== Benchmark assessment form ===
{| border=1 cellspacing=0 cellpadding=5
|-bgcolor="#CCCCCC"
!width="325px"|Description
!
|-valign="top"
|New features added into GEOS-Chem
|
{| border=1 cellspacing=0 cellpadding=5
|-bgcolor="#CCCCCC"
!width="500px"|Feature
!width="200px"|Submitted by
|-
!colspan="4" bgcolor="#CCFFFF"|Features affecting the full-chemistry simulation:
|-valign="top"
|[[#Inconsistency in H2402 concentrations in GCHP vs GEOS-Chem Classic|Fix inconsistency for species H2402 between GCHP and GEOS-Chem "Classic"]]
|Sebastian Eastham (MIT)<br>[[GCST]]
|-valign="top"
|Bug fixes for the FAST-JX photolysis mechanism:
*[[FAST-JX_v7.0_photolysis_mechanism#Duplicate_entries_for_species_ClNO2_in_FJX_j2j.dat_file|Remove duplicate entry for species ClNO2 in the FJX_j2j.dat file]]
*[[FAST-JX_v7.0_photolysis_mechanism#Species_DHDC_was_not_listed_as_photolyzing_in_the_species_database|Species DHDC was not listed as photolyzing in the GEOS-Chem species database]]
|Bob Yantosca ([[GCST]])
|-valign="top"
|Updates and fixes for HEMCO diagnostics, including:
*Enable saving out fertilizer NOx emissions
*Lightning NOx emissions are 3D not, 2D
*HNO3 ship emissions are 2D, not 3D
*Update and fill in missing inventory diagnostics (currently used in only in benchmark simulations)
|Melissa Sulprizio ([[GCST]])
|-
!colspan="4" bgcolor="#CCFFFF"|Features NOT affecting the full-chemistry simulation:
|-valign="top"
|[[CEDS_anthropogenic_emissions#Updates_introduces_in_GEOS-Chem_12.3.2|Extending the CEDS anthropogenic emissions inventory to include 1750-1949 and additional VOC species]]
|Melissa Sulprizio (GCST)
|-
!colspan="4" bgcolor="#CCFFFF"|Features only affecting GCHP:
|-valign="top"
|[[#Disable Br2 offline seasalt emissions to match GEOS-Chem Classic default|Disable Br2 offline seasalt emissions to match GEOS-Chem Classic default]]
|Sebastian Eastham (MIT)
|-valign="top"
|Assorted GCHP run directory feature updates
*Rename Makefile clean and build options
*Remove deprecated Makefile/build.sh clean and build options
*Remove aliases from sample run scripts
*Remove output file runConfig.log; send to gchp.log instead
*Update benchmark run directory to match GEOS-Chem Classic settings
|Lizzie Lundgren (GCST)
|-valign="top"
|Embed GEOS code within select GCHP files
|Lizzie Lundgren (GCST)
|Structural
|}
|-valign="top"
|Version, resolution, met fields used:
|GEOS-FP (72L), 4x5, July 2016
|-valign="top"
|1-month benchmark finished on:
|
|-valign="top"
|Performance statistics:
|
*Compiled with ifort 17.0.4
*Ran on _ CPUs of _ (Intel(R) Xeon(R) CPU E5-2680 v3 "Haswell" @ 2.50 GHz)
*Wall time:
*[[Scalability|CPU time / wall time]]:
*% of ideal performance: %
*Memory: TBD GB
|-valign="top"
|Compared to previous benchmark:
|[[GEOS-Chem_12#12.3.0|12.3.0]]
|-valign="top"
|This update will impact:<br>(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): Volcano emissions
|-valign="top"
|[[GEOS-Chem Unit Tester|Unit test]] results may be viewed at:
|
|-valign="top" bgcolor="#cccccc"
!Metrics
!
|-valign="top"
|Global mean OH from log file (% change):
|12.5308895343096 x 10<sup>5</sup> molec/cm<sup>3</sup> (0.04%)
|-valign="top"
|Methyl chloroform lifetime (% change):
|4.9406 years (-0.05%)
|-valign="top"
|Methane lifetime (% change):
|8.3253 years (-0.05%)
|-valign="top"
|At the SURFACE, list all species that changed by 10% or more:
|
|-valign="top"
|Comments on SURFACE differences:
|
|-valign="top"
|
|-valign="top"
|Comments on 500 hPa differences:
|
*See comments for surface differences
|-valign="top"
|In the ZONAL MEAN differences, list all species that changed by 10% or more:
|-valign="top"
|Comments on ZONAL MEAN differences:
|
*See comments for surface differences
|-valign="top"
|In the EMISSION RATIO maps, list all species that changed by 10% or more:
|
|-valign="top"
|Comments on EMISSION RATIO differences:
|
|-valign="top"
|Additional or summary comments:
|
|-valign="top"
|Comments on additional differences in GCHP:
|''To be filled out by GCHP WG''
|-bgcolor="#cccccc"
!Approval
!
|-valign="top"
|Requires further investigation:
|
|-valign="top"
|Approved by:
|
|-valign="top"
|Date of approval:
|
|}
--[[User:Bmy|Bob Yantosca]] ([[User talk:Bmy|talk]]) 20:19, 25 April 2019 (UTC)
== GCHP 1-month benchmark ==
== GCHP 1-month benchmark ==



Revision as of 20:19, 25 April 2019

1-month benchmark

Benchmark assessment form

Description
New features added into GEOS-Chem
Feature Submitted by
Features affecting the full-chemistry simulation:
Fix inconsistency for species H2402 between GCHP and GEOS-Chem "Classic" Sebastian Eastham (MIT)
GCST
Bug fixes for the FAST-JX photolysis mechanism: Bob Yantosca (GCST)
Updates and fixes for HEMCO diagnostics, including:
  • Enable saving out fertilizer NOx emissions
  • Lightning NOx emissions are 3D not, 2D
  • HNO3 ship emissions are 2D, not 3D
  • Update and fill in missing inventory diagnostics (currently used in only in benchmark simulations)
Melissa Sulprizio (GCST)
Features NOT affecting the full-chemistry simulation:
Extending the CEDS anthropogenic emissions inventory to include 1750-1949 and additional VOC species Melissa Sulprizio (GCST)
Features only affecting GCHP:
Disable Br2 offline seasalt emissions to match GEOS-Chem Classic default Sebastian Eastham (MIT)
Assorted GCHP run directory feature updates
  • Rename Makefile clean and build options
  • Remove deprecated Makefile/build.sh clean and build options
  • Remove aliases from sample run scripts
  • Remove output file runConfig.log; send to gchp.log instead
  • Update benchmark run directory to match GEOS-Chem Classic settings
Lizzie Lundgren (GCST)
Embed GEOS code within select GCHP files Lizzie Lundgren (GCST) Structural


Version, resolution, met fields used: GEOS-FP (72L), 4x5, July 2016
1-month benchmark finished on:
Performance statistics:
  • Compiled with ifort 17.0.4
  • Ran on _ CPUs of _ (Intel(R) Xeon(R) CPU E5-2680 v3 "Haswell" @ 2.50 GHz)
  • Wall time:
  • CPU time / wall time:
  • % of ideal performance: %
  • Memory: TBD GB
Compared to previous benchmark: 12.3.0
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): Volcano emissions
Unit test results may be viewed at:
Metrics
Global mean OH from log file (% change): 12.5308895343096 x 105 molec/cm3 (0.04%)
Methyl chloroform lifetime (% change): 4.9406 years (-0.05%)
Methane lifetime (% change): 8.3253 years (-0.05%)
At the SURFACE, list all species that changed by 10% or more:
Comments on SURFACE differences:
Comments on 500 hPa differences:
  • See comments for surface differences
In the ZONAL MEAN differences, list all species that changed by 10% or more:
Comments on ZONAL MEAN differences:
  • See comments for surface differences
In the EMISSION RATIO maps, list all species that changed by 10% or more:
Comments on EMISSION RATIO differences:
Additional or summary comments:
Comments on additional differences in GCHP: To be filled out by GCHP WG
Approval
Requires further investigation:
Approved by:
Date of approval:

--Bob Yantosca (talk) 20:19, 25 April 2019 (UTC)

GCHP 1-month benchmark

Comparison of GCHP and GEOS-Chem Classic

NOTE: This section is still under development

This table contains plots comparing GEOS-Chem Classic 12.3.2 (Ref) against GCHP 12.3.2 (Dev) for a 1-month simulation.

Category1 Plots2 Additional Information
Level Map Zonal Mean Emissions
Primary sfc 500hPa full column stratosphere sfc
Aerosols sfc 500hPa full column stratosphere sfc
Bromine sfc 500hPa full column stratosphere sfc
Chlorine sfc 500hPa full column stratosphere --
Iodine sfc 500hPa full column stratosphere sfc
Nitrogen sfc 500hPa full column stratosphere sfc
Primary Organics sfc 500hPa full column stratosphere sfc
Secondary Organics sfc 500hPa full column stratosphere sfc
Sulfur sfc 500hPa full column stratosphere sfc
Emissions by HEMCO category

Aircraft, Anthro, Bioburn, Biofuel, Biogenic,
Lightning, Monoterp, Natural, Ocean, Seabirds, SenescingPlants,
Ship, Soil, Total, VolcDegas, VolcErupt,

1Benchmark category definitions: benchmark_categories.json
2Lumped species definitions: lumped_species.json

Comparison of GCHP versions

This table contains plots comparing GCHP 12.2.1 (Ref) and GCHP 12.3.0 (Dev) for a 1-day simulation. This is a prototype only and the real GCHP benchmarks will compare 1-month simulation data.

Category1 Plots2 Additional Information
Level Map Zonal Mean Emissions
Primary sfc 500hPa full column stratosphere sfc
Aerosols sfc 500hPa full column stratosphere sfc
Bromine sfc 500hPa full column stratosphere sfc
Chlorine sfc 500hPa full column stratosphere --
Iodine sfc 500hPa full column stratosphere sfc
Nitrogen sfc 500hPa full column stratosphere sfc
Primary Organics sfc 500hPa full column stratosphere sfc
Secondary Organics sfc 500hPa full column stratosphere sfc
Sulfur sfc 500hPa full column stratosphere sfc
Emissions by HEMCO category

Aircraft, Anthro, Bioburn, Biofuel, Biogenic,
Lightning, Monoterp, Natural, Ocean, Seabirds, SenescingPlants,
Ship, Soil, Total, VolcDegas, VolcErupt,

1Benchmark category definitions: benchmark_categories.json
2Lumped species definitions: lumped_species.json

NOTES:

  1. Emissions tables generated by gcpy are not currently compatible with cubed-sphere data and so are excluded from the GCHP benchmark.

--Lizzie Lundgren (talk) 17:08, 12 April 2019 (UTC)