Integration of Criteria and Toxic Pollutants in SMOKE

1
Integration of Criteria and Toxic Pollutants in
SMOKE
For the Models 3 Users Workshop October 22, 2002

• Madeleine Strum, OAQPS
• Collaborators Marc Houyoux, MCNC/EMC
• Ron Ryan Greg Stella, OAQPS
• Bill Benjey Gerald Gipson, ORD
• Rich Cook, OTAQ

2
EPAs Toxics and Criteria Modeling Systems
Separate
Issues Sound Science Consistency Efficiency

• Photochemical vs Gaussian
• SMOKE vs EMS-HAP
• Criteria vs NTI

Grid
EPA is moving towards one-atmosphere modeling to
address these issues
3
One-Atmosphere Approach
COMMENTS
National Emissions Inventory
Still two separate inventories for 1999 toxics
(HAPs) criteria
Modify to process multipollutant inventories
SMOKE
Add capabilities for toxics
CMAQ
4
Main Issue for SMOKE is to Address Overlap in
Pollutants
VOC from criteria inventory
Volatile organic HAPs from toxics inventory

• No double-counting
• Use of toxics inventory pollutants, where
  possible
• Conservation of VOC mass, where possible

Note even though were not looking at
particulates, addressing the VOC HAP overlap
will impact PM modeling
5
Steps in Determining Changes Needed to SMOKE

• Examine criteria and toxics inventories select
  when to use toxics in cases of overlap
• Mobile sources EPA-generated nonroad and
  MOBILE6 onroad inventories for toxics and critera
  are consistent
• Stationary source Generally not consistent
• Examine potential modeling applications
• Modeling ozone/PM using emission estimates for
  toxics from the toxics inventory in place of the
  criteria inventory
• Modeling toxics using emission estimates for
  toxics from the toxics inventory

6
Steps in Determining Changes Needed to SMOKE
(continued)

• Determine how CMAQ could be modified for the
  applications
• No modifications needed for modeling ozone/PM
• Develop new mechanisms and solvers to handle
  toxics
• Modify CB-IV to treat key toxics explicitly
  (e.g., formaldehyde and acetaldyde)
• Add capability to approximate chemical loss of
  other toxics outside chemical mechanism via
  pseudo-first order reaction rates
• Determine model species needed and how SMOKE will
  produce them see next slide for details

7
How SMOKE Will Produce Model Species
For Ozone/PM Modeling
Mobile sources integration - toxic
pollutants (gaseous organic HAP) from toxics
inventory - non toxics (nonHAP VOC) from
criteria inventory Stationary sources speciate
criteria
For Toxics Modeling
Mobile sources integration (same as
above) Stationary sources - speciate
criteria for mechanism species that are not
toxics - use toxics inventory for mechanism
species that are explicit toxics
(e.g., formaldehyde) - use toxics inventory
for toxics modeled outside of mechanism
8
Integration Approach for Mobile Sources
VOC or TOG (criteria) emissions
Gaseous HAP emissions
Create NONHAPTOG or NONHAPVOC VOC
sum(gaseous toxics species)
New SMOKE function
Existing SMOKE function but new speciation
profiles are needed
Speciation
Model species
9
Practical Considerations for Integration

• The emission sources must be exactly the same for
  the toxics and the VOC/TOG
• The methodology to estimate emissions should be
  consistent between the two inventories

Both of these hold true for EPA-generated nonroad
emissions and MOBILE6 generated
emissions Generally, these dont hold true for
stationary sources
10
Integration for Onroad Mobile Sources

• SMOKE runs MOBILE6.2 to generate TOG and 26
  gaseous HAPs such as benzene, acetaldehdye, PAHs
• For each emission source, SMOKE creates a
  nonHAPTOG pollutant by subtracting the toxics
  from the TOG
• SMOKE applies speciation profiles for nonHAPTOG
  and individual HAPs to generate emissions of
  model species.
• For a toxic modeled outside the mechanism, SMOKE
  maps the toxic to itself and its mechanism species

EXAMPLE SPECIATION PROFILES FOR ONROAD
SOURCES 1313A NONHAPTOG OLE
0.00127373705 1 0.0357011 1313A
NONHAPTOG PAR 0.03439269933
1 0.491116 0000 BENZENE
BENZENE 1.00 78.11
1.0 0000 BENZENE PAR
1.00 78.11
0.1666667 0000 BENZENE NR
5.00 78.11
0.8333333
11
Integration for Nonroad Mobile SourcesSimilar to
Onroad

• SMOKE imports toxic and criteria nonroad emission
  sources to be integrated- user assures
  consistency in sources
• SMOKE creates a nonHAPVOC pollutant by
  subtraction
• SMOKE applies speciation profiles

12
Approach For Stationary Sources is Replacement,
not Integration

• SMOKE imports both criteria and toxic inventories
  
• SMOKE speciates criteria inventory using
  speciation profiles that zero out the explicit
  toxics model species (e.g., formaldehyde)
• SMOKE maps the toxics inventory pollutants to the
  CB-IV explicit toxics model species
• For a toxic modeled outside the mechanism, SMOKE
  maps the toxic to itself, but not its mechanism
  species

EXAMPLE SPECIATION PROFILES FOR STATIONARY
SOURCES 0307 TOG OLE
0.00126649899 1
0.03553 0307 TOG PAR
0.03791738749 1
0.5397267 0000 FORM FORM
0.00126649899 1
0.03553 0000 BENZENE_ST BENZENE
1.00 30.03 1
13
Summary

• We have explored one-atmosphere modeling using
  toxics and criteria inventories in the SMOKE/CMAQ
  modeling system, focusing on gaseous HAPS
• Benefits to ozone/PM modeling better estimates
  of a subset of the pollutants (gaseous TOG) than
  current VOC speciation approach
• Benefits to Toxics modeling allows for use of
  photochemical grid models with toxics inventory
• Wider integration will be possible when
  inventories become more consistent