Shawn Roselle, Deborah Luecken, William Hutzell, Russell Bullock, Golam Sarwar, and Kenneth Schere

1
Development of a Multipollutant Version of the
Community Multiscale Air Quality (CMAQ) Modeling
System

• Shawn Roselle, Deborah Luecken, William Hutzell,
  Russell Bullock, Golam Sarwar, and Kenneth Schere
  
• U.S. EPA/ORD/NERL NOAA/OAR/ARL
• Atmospheric Modeling Division
• Research Triangle Park, NC

2
Background

• Humans and ecosystems can be exposed to multiple
  pollutants at the same time
• Chemistry, transport and fate of pollutants are
  interrelated
• CMAQ was developed to provide aone-atmosphere
  approach for air quality modeling
• State-of-science capabilities for modeling
  multiple air quality issues, including
  tropospheric ozone, fine particles, hazardous air
  pollutants (HAPs), acid deposition, and
  visibility degradation

3
Background (continued)

• CMAQ utilized by EPA/OAR/OAQPS to evaluate
  benefits and effectiveness of various control
  programs
• Clean Air Interstate Rule (CAIR)
• Clean Air Mercury Rule (CAMR)
• Clean Air Visibility Rule (CAVR)
• Increasing interest for modeling multipollutants,
  including criteria and hazardous air pollutants,
  in a single modeling framework for air quality
  management
• Predict ozone, PM, mercury, and other HAPs
  concentrations and interactions all within the
  same model simulation
• Ability to examine co-benefits of emission
  reductions

4

Current Capabilities (CMAQv4.6)

• Several configurations available for CMAQ
• Base model1
• Carbon Bond 05 (CB05) and SAPRC99
• Aerosol module version 4 (aero4 or AE4)
• Cloud chemistry
• HAPs1
• Includes gas-phase HAPs, toxics metals, and
  diesel PM
• CB05-HAPs chemical mechanism includes chlorine
  chemistry
• SAPRC99 version is also available
• Mercury1
• CB05-Mercury mechanism includes chlorine
  chemistry
• Additional gas-phase species
• Elemental mercury and reactive gaseous mercury
• Aerosol species
• Particulate Hg (Aitken and Accumulation mode)
• Additional gas-phase reactions
• Major modifications to aqueous phase chemistry
• 1Although versions exist for CB4 and aero3 (AE3),
  these will no longer be supported beginning with
  the 2008 CMAQ release

5
Approach

• Merge capabilities from HAPs and mercury model
  versions
• Differences in model results were related to
  chlorine emissions and chemistry
• Normalize results across the different model
  configurations

6
Number of Species for Different Model
Configurations
Configuration Gas-phase Aerosol Non-reactive Total Species (total transported)
Base CB05 cb05_ae4_aq 56 34 12 102 (79)
CB05 w/ Chlorine cb05cl_ae4_aq 62 34 11 107 (82)
CB05 Mercury cb05hg_ae4_aq 65 36 11 112 (88)
CB05 HAPs cb05cltx_ae4_aq 73 58 33 164 (141)
CB05 Multipollutant cb05txhg_ae4_aq 76 60 33 169 (147)
7
Multipollutant Model Tests

• Model platform
• Simulation period July 22-31, 2001
• Domain continental U.S. 36 km grid resolution
  14 layers
• Meteorological Data
• MM5 Simulations (36 km grid resolution 34
  layers)
• Emissions
• Merged existing emissions files for criteria air
  pollutants (CAPs), HAPs and mercury (1999 NEI
  added mercury emissions)
• Simulations
• Multipollutant, HAPs, mercury, CB05 w/ chlorine
  chemistry, and base-CB05
• Other model sensitivity tests

8
Ozone (10-day Maximum)
Multipollutant Model
Mercury Model
HAPs Model
Max diff over all hours Multipollutant vs. HAPs

• Differences caused by Cl2 emissions being
  zeroed-out in HAPs Model to keep O3 same as base
  CB05

9
Chlorine (10-day Average)
Multipollutant Model
Mercury Model
HAPs Model
CB05 w/ Chlorine Chem. Model

• Cl2 emissions were not included in HAPs Model

10
Hydrochloric Acid (10-day Average)
Multipollutant Model
Mercury Model
HAPs Model
CB05 w/ Chlorine Chem. Model

• Cl2 emissions not included in HAPs model
• HAPs model includes reaction
• ltCL21gt HCL OH CL 6.58E-131.16 _at_ -58
• Minimum value set for Cl- in mercury version of
  AQCHEM
• Cl- (aq) ? HCl and ACLJ

11
Hydrochloric Acid (10-day Average)
Mercury Model
Multipollutant w/ AQCHEM mod
HAPs Model w/ Cl2 emis
Multipollutant w/ AQCHEM mod with HCl emissions
CB05 w/ Chlorine Chem. Model

• Cl2 emissions now included in HAPs model
• Minimum value for Cl- reduced by several orders
  of magnitude in multi-pollutant version of AQCHEM
• None of the models include HCl emissions

12
Summary of the Normalization Process

• Turned on Cl2 emissions in HAPs model
• Turned on HCl emissions in all models with
  chlorine chemistry HAPs, Mercury, CB05 with
  chlorine chemistry, and Multipollutant
• Reduced background aqueous Cl- concentration in
  aqueous chemistry to improve mass balance (in
  multipollutant and mercury models)
• Added HClOH gas phase reaction to mercury model
  and CB05 w/ chlorine chemistry model

13
Ozone 10-day Maximum
Multipollutant Model
Mercury Model
HAPs Model
All Hours (ppb) MAX DIFF MIN DIFF 0.063
-0.037
All Hours (ppb) MAX DIFF MIN DIFF 0.456
-0.081
14
Ozone 10-day Maximum
Multipollutant Model
Base CB05 Model
CB05 w/ Chlorine Chem. Model
All Hours (ppb) MAX DIFF MIN DIFF 36.55
-0.995
All Hours (ppb) MAX DIFF MIN DIFF 0.471
-0.073
15
Sulfate (10-day Average)
Multipollutant Model
Mercury Model
HAPs Model
(ug/m3) MAX DIFF MIN DIFF 0.032 -0.026
(ug/m3) MAX DIFF MIN DIFF 0.001 -0.001
16
Sulfate (10-day Average)
Multipollutant Model
Base CB05 Model
CB05 w/ Chlorine Chem. Model
(ug/m3) MAX DIFF MIN DIFF 0.023
-0.051
(ug/m3) MAX DIFF MIN DIFF 0.030
-0.028
17
Average Elapsed CPU Time (minutes) per Simulation
Day
Model Chemical Solver Chemical Solver
Model ros3 ebi
Base CB05 19.9 16.6
CB05 w/ chlorine chemistry 22.2 18.9
CB05 mercury 23.8 20.7
CB05 HAPs 33.7 30.7
CB05 Multipollutant 35.9 32.0
SGI-Altix 4700, 8 PEs
18
Summary

• A multipollutant version of CMAQ has been
  developed and tested
• An internal version of the model is being applied
  and evaluated by OAQPS
• Sharon Phillips will present results from
  multipollutant model applications in Session 8
• Multipollutant model will be included in next
  release

19
2008 CMAQ Release

• Based on current model test results, plan to use
  a version of CB05 w/chlorine chemistry as the
  base configuration for the next model release
• Consistent results for the different model
  configurations
• Requires emissions for Cl2 and HCl
• Next release Fall 2008

20
Disclaimer The research presented here was
performed under the Memorandum of Understanding
between the U.S. Environmental Protection Agency
(EPA) and the U.S. Department of Commerces
National Oceanic and Atmospheric Administration
(NOAA) and under agreement number DW13921548.
This work constitutes a contribution to the NOAA
Air Quality Program. It has not been reviewed by
EPA or NOAA and has not been approved for
publication, and does not necessarily reflect
their views or policies.