Comprehensive Particulate Matter Modeling: A One Atmosphere Approach

1
GEOS-CHEM Modeling for Boundary Conditions and
Natural Background
James W. Boylan Georgia Department of Natural
Resources - VISTAS National RPO Modeling
Meeting Denver, CO May 25, 2004
2
Outline

• Objectives and Background
• Summary of Modeling Results
• Update on RPO GEOS-CHEM Project
• Natural Background Options
• Group Discussion

3
Objectives

• Run the GEOS-CHEM global chemical transport model
  for CY2002 to develop temporally and spatially
  varying boundary conditions (BCs) for the 36-km
  national modeling grid to be used for ozone,
  PM2.5, and regional haze modeling.
• Run GEOS-CHEM to evaluate natural background
  visibility and transboundary pollutant transport.

4
GEOS-CHEM Global Chemical Transport Model

• Driven by Goddard Earth Observing System (GEOS)
  assimilated meteorological data from the NASA
  Data Assimilation Office (DAO)
• Horizontal resolution 1o x 1o to 4o x 5o
  (user-selected), 48 levels in vertical
  (80 km)
• Ozone-NOx-VOC (oxidant) chemistry 80 species,
  400 reactions
• Aerosols H2SO4-HNO3-NH3, organic carbon (OC),
  elemental carbon (EC), soil dust (four size
  classes), sea salt (two size classes)
• Oxidant and aerosol simulations coupled by
  photolysis frequencies, heterogeneous chemistry,
  sulfate/nitrate formation, HNO3(g)/NO3-
  partitioning
• Multi-process wet deposition scheme
• http//www-as.harvard.edu/chemistry/trop/geos

5
(No Transcript)
6
(No Transcript)
7
3-Hourly GEOS-CHEM Sulfate BCs
EPA Default
8
SO4 Seasonal Boundary Condition Time Series
Summer
9
PM diff. (GCM BCs/ICs Default BCs/ICs)
D PM 2.5 D PM Sulfate
(July monthly avg., 2001)
Slide provided by Carey Jang, U.S. EPA
10
PM diff. (GCM BCs/ICs Default BCs/ICs)
D PM 2.5 D PM Sulfate
(July 22, 2001, daily average)
Slide provided by Carey Jang, U.S. EPA
11
PM diff. (GCM BCs/ICs Default BCs/ICs)
D PM 2.5 D PM Nitrate
(January monthly avg., 2001)
Slide provided by Carey Jang, U.S. EPA
12
PM diff. (GCM BCs/ICs Default BCs/ICs)
D PM 2.5 D PM Nitrate
(Jan. 13, 2001, daily average)
Slide provided by Carey Jang, U.S. EPA
13
January 2002 EpisodeGEOS-CHEM (S) vs. EPA-Default
CASTNET TNO3 (VISTAS) CASTNET TNO3 (US)
14
Summary of Results

• Summer Episodes
• Significant change in ozone in WRAP, MRPO, and
  MANE-VU
• Significant change in PM2.5 (sulfate) in WRAP
• Winter Episodes
• Significant change in PM2.5 (nitrate) in MRPO,
  MANE-VU, and VISTAS related to ozone or
  reactive nitrogen BCs (?)
• Ozone performance worse (low bias) BC low in
  north
• NO2, NO3, and TNO3 performance better
• Mexican Fires
• Significant change in PM2.5 (OC EC) in CENRAP,
  WRAP, VISTAS, and MANE-VU

15
Comparison of Wind Fields
GEOSCHEM DAO
GEOSCHEN Westerly (inflow) MM5 Northerly
(outflow)
16
GEOSCHEM Easterly and northerly MM5 Clock
wise rotation motion
17
Linking GEOS-CHEM and CMAQ

• Possible inconsistencies between the global and
  regional scale dynamics
• MM5 Outflow and GEOS-CHEM Inflow ? OK
• GEOS-CHEM Outflow and MM5 Inflow ? Problems
• Remedy (being tested by UH)
• Run regional scale model with global scale output
  as input for initialization and analysis nudging
• Too late to use this remedy for VISTAS modeling
• Other Options?
• In the case where GEOS-CHEM outflow and MM5
  inflow, could replace hourly BC with seasonal
  average BC or seasonal average inflow BC

18
GEOS-CHEM Modeling for RPOs

• Daniel Jacob (Harvard) will be performing annual
  (CY2002) GEOS-CHEM modeling for the RPOs
• Project is being managed by a steering committee
  comprised of representatives from all 5 RPOs
• VISTAS volunteered to take the lead
• Workplan identifies three major tasks to be
  completed by August 31, 2004
• VISTAS is funding Daewon Byun (UH) to convert
  GEOS-CHEM outputs to boundary conditions for CMAQ
  (CAMx?)

19
Harvard Deliverables

• Global Emissions Inventory for CY2002
• Biomass burning inventory with 1x1 resolution
• GEOS-CHEM global 3-D concentration fields (4x 5
  grid) with 3-hour resolution for three full-year
  simulations
• Baseline 2002 simulation with all anthropogenic
  emissions
• Sensitivity simulation with U.S. anthropogenic
  emissions turned off
• Sensitivity simulation with worldwide
  anthropogenic emissions turned off
• Model Performance Evaluation (MPE)
• IMPROVE and CASTNET
• Monthly mean SO4, NO3, TNO3, NH4, OC, EC, and
  dust
• Extensive MPE conducted under EPRI funding

20
Natural Background Runs

• The GEOS-CHEM model results can be used to
  evaluate natural conditions at Class I areas
• Directly Method Turn off U.S. anthropogenic
  emissions in GEOS-CHEM and examine pollutant
  concentrations at Class I areas
• 4x5 resolution too coarse to be useful
  (especially for mountainous and costal sites)
• 1x1 resolution will be more useful than 4x 5
  resolution, but still might not meet our needs
• Results available through EPRI funding in Fall
  2004

21
4x 5 GEOS-CHEM Grid
22
Natural Background Runs

• Indirect Method Turn off U.S. anthropogenic
  emissions in GEOS-CHEM to generate a new set of
  BCs that will be used for a CMAQ run at 36/12 km
  resolution with U.S. anthropogenic emissions
  turned-off.
• New boundary conditions must be generated with
  GEOS-CHEM (no U.S. emissions) to eliminate
  recirculated U.S. anthropogenic emissions from
  the boundary conditions
• Advantages finer grid resolution, better
  meteorological fields, more advanced air quality
  model
• Disadvantages time and
• Same concepts for evaluating true natural
  conditions by removing worldwide anthropogenic
  emissions in GEOS-CHEM.

23
4x 5 GEOS-CHEM Grid
24
4x 5 GEOS-CHEM Grid
25
Discussion Issues

• Options for addressing inconsistencies between
  the global and regional meteorological fields?
• Natural Background Simulations
• Direct or Indirect Method?
• Turn-off U.S. anthropogenic emissions or turn-off
  all North American anthropogenic emissions?
• Boundary Conditions for 2002 typical and 2018
  typical CMAQ simulation
• Use 3-hour GEOS-CHEM BCs or use seasonal average
  GEOS-CHEM BCs or use seasonal average
  inflow/outflow BCs?
• Other Issues?