William T. Hutzell, Shawn J. Roselle, Annmarie G. Carlton, and O. Russell Bullock

1
Changes to the Multi-Pollutant version in the
CMAQ 4.7

• William T. Hutzell, Shawn J. Roselle, Annmarie G.
  Carlton, and O. Russell Bullock
• Atmospheric Modeling Division
• US EPA

2
Background Multi-Pollutant Model

• Developed to simulate criteria and hazardous air
  pollutants in a single modeling study
• examine co-benefits of emission reductions over
  multiple interest
• O3, Particulate Matter, and Hazardous Air
  Pollutants (HAPs)
• Roselle et al. (2007) presented a prototype of
  the Multi-Pollutant Model
• mechanism called CB05TXHG_AE4_AQ
• Based on CB05 mechanisms for mercury and other
  HAPs in CMAQ version 4.6

3
Multi-Pollutant Model in version 4.7

• Updates prototype in Roselle et al (2007)
• photochemical production of Secondary Organic
  Aerosols (SOA)
• aerosol physics for SOA and interactive coarse
  mode
• cloud chemistry of organic compounds and SOA
• In-calculations for dry deposition and emission
  processes
• New mechanism called CB05TXHG_AE5_AQ
• Specific settings needed to build and run the
  Multi-Pollutant model.
• Consult release notes on building and running the
  model.

4
Goals of this Presentation

• Describe major changes to new science options for
  Multi-Pollutant Model
• Present differences in predictions from CMAQ
  using CB05CL_AE5_AQ, the standard version
• Highlight differences in predictions from the
  prototype of the Multi-Pollutant Model

5
Gas Chemistry

• Adds to CB05CL_AE5_AQ the species and reactions
  for Hg compounds and other HAPs from
  CB05HG_AE4_AQ and CB05CLTX_AE4_AQ
• Two methods compute the chemical transformation
  of gas phase Hg and other HAPs
• One participates in ozone and radical
  photochemistry
• Other does not affect ozone and radical
  concentrations and serves as reactive tracers.
• Changes from CB05TXHG_AE4_AQ
• adds a new reaction, HG CL ? HGIIGAS
• This reaction should produce Hg(I)
• HGIIGAS redefined as reactive gaseous mercury
• i.e., Hg(I) Hg(II).
• Hg(I) assumed to quickly convert into Hg(II)
• the oxidation process not explicitly represented

6
Aerosols

• Adds aerosol species representing mercury and
  other toxic metals
• They do not affect aerosol microphysics and
  deposition rates
• These aerosols species do coagulate and mode
  merge
• Mercury species differ from the other metallic
  aerosols
• photochemical source of particulate mercury
• goes directly into the accumulation mode
• assumes accumulation mode dominates condensation
  onto the three aerosol modes
• unlike version 4.6 where fine modes divided
  condensation based on their surface area

7
In-Cloud Chemistry

• Adds in-scavenging for mercury and other metallic
  aerosols.
• Adds cloud chemistry for atmospheric mercury
• Based on Bullock and Brehme (2002)
• Indirectly affects other aqueous species
• modifies the minimum time step used in the
  numerical solution when the mercury species have
  the fastest rate of change.
• mercury chemistry requires using the gas phase
  HO2, HOCl and Cl2 based on Lin et al. (1998).
• affects pH and ion balance in cloud droplets

8
In-Cloud Chemistry (cont.)

• Possible effects from changes
• change particulate sulfate
• wet deposition of HO2, HOCl and Cl2
• can produce HOCl from clouds with low or no
  participation.

9
Inline Vertical Diffusion

• Aerosol emissions includes routines for
  particulate mercury and other metals
• Emissions includes source of Cl2 over open
  oceans.
• Set off by default
• Set on with the environment variable,
  CTM_CL2_SEAEMIS
• Source mimics implied heterogeneous production
  for sea salt aerosols (Spicer et al. 1998)
• Knipping and Dadbub (2002 and 2003) proposed a
  mechanism but not used
• reaction efficiencies are not well defined
• CB05CL_AE5_AQ does not include ClNO2 and ClONO2.

10
Comparison to Standard CMAQ, i.e.,CB05CL_AE5_AQ

• Using two weekly periods in January and July 2002
• Domain covered Continental US
• grid cell had 36X36 km2 horizontal dimensions
• 14 vertical layers in s pressure coordinates
• Comparison used the weekly averages
• tile plots of differences from CB05CL_AE5_AQ,
  i.e., standard model
• scatter plots showing difference versus
  standards prediction

11
HOCl
Higher HOCl production appears higher in January
12
Ozone
Ozone higher in winter and correlate with HOCL
but generally within the accuracy of the
chemistry solver.
13
Impact of Ozone DifferenceBenzene
Januarys ozone differences alter OH thereby
Benzene
14
Particulate Sulfate
Januarys ozone differences also have small
effect on sulfate production
15
Hg Changes between CB05TXHG version 4.6 and 4.7
16
Areas for further improvement

• Gas Phase Chemistry for Mercury
• explicitly representing products
• Oceanic Cl2 source
• representing heterogeneous production
• possible with the proposed SAPRC07 mechanism
• Bi-Directional Surface Flux for Mercury
• addressed science model release in version 4.7
• see release notes by Jesse Bash
• Cloud chemistry does not include Cr(III) and
  Cr(VI) redox reactions

Disclaimer Although this work has been reviewed
by EPA and approved for publication, it does not
necessarily reflect their policies or views.