Transcript U.S. EPA’s Models-3 : An Integrated “One

Applications and Evaluation of USEPA’s
Models-3/CMAQ System: From Regional and
Urban Air Pollution to Global Climate Change
Carey Jang, Pat Dolwick, Norm Possiel,
Brian Timin, Joe Tikvart
U.S. EPA
Office of Air Quality Planning and Standards
(OAQPS)
Research Triangle Park, North Carolina
OUTLINE

Overview of USEPA’s “One-Atmosphere”
Models-3/CMAQ Modeling System

Applications and Evaluation of Models-3
/CMAQ System

OAQPS Modeling Initiative on Intercontinental
Transport and Climatic Effects of Pollutants
Third-Generation Air Quality Models:
U.S.EPA’s Models-3/CMAQ System

“Open-Access” Community-Based Models :
• User-friendly, Modular, Common modeling framework for
scientists and policy-makers.

Advanced Computer Technologies :
• High performance hardware and software technologies
(Cross-platform, GUI, distributed computing, visualization
tools, etc.).
 “One-Atmosphere”
Modeling :
• Multi-pollutant (Ozone, PM, visibility, acid
deposition, air toxics, etc.), Multi-scale.
“One-Atmosphere” Management and Modeling
Mobile
Sources
Ozone
NOx, VOC,
PM, Toxics
PM
(Cars, trucks, planes,
boats, etc.)
Industrial
Sources Chemistry
NOx, VOC,
SOx, PM,
Toxics
(Power plants, refineries/
chemical plants, etc.)
Area
Sources
NOx, VOC,
PM, Toxics
(Residential, farming
commercial, biogenic, etc.)
Acid Rain
Meteorology
Visibility
Air Toxics
Atmospheric
Deposition
Climate
Change
NOx-Related Air Quality Issues
(NO3-, NH4+)
(NOx + VOC + hv) -->
PM
Ozone
NOx
Acid Rain
Visibility
(NO3- deposition)
(Fine PM)
Water Quality
(Nitrogen deposition,
Lake Acidification)
SOx-Related Air Quality Issues
(Fine PM)
(SO42-, NH4+)
Visibility
PM
SOx
Acid Rain
Water Quality
(SO42- deposition)
(Lake acidification,
Toxics deposition)
.
OH role in pollutants formation : One-Atmosphere
PM2.5
VOC + OH --->
Orgainic PM
Fine PM Visibility
(Nitrate, Sulfate,
Organic PM)
Ozone
.
OH
NOx + VOC + OH
+ hv ---> O3
Acid Rain
Water
Quality
NOx + SOx + OH
(Lake Acidification,
Eutrophication)
SOx [or NOx] + NH3 + OH
---> (NH4)2SO4 [or NH4NO3]
SO2 + OH ---> H2SO4
NO2 + OH ---> HNO3
Air Toxics
OH <---> Air Toxics
(POM, PAH, Hg(II), etc.)
Example of “One-Atmosphere” Modeling
Impact of
50 % NOx
Emission Reduction
on PM 2.5
Impact of 50% NOx emission reduction
Nitrate PM decrease
Sulfate PM decrease
Impact of 50% NOx emission reduction
O3 decrease
HOx decrease
Formation of Secondary PM :
Sulfate PM formation:
H2SO4 + 2 NH3 ---> (NH4)2SO4 (s)
Gas Phase:
O2,H2O
SO2 + OH
---> H2SO4
Aqueous Phase:
H2O
SO2 + H2O2 ---> H2SO4 (Dominate over low pH)
SO2 + O3 ---> H2SO4
Oraginc PM formation:
Nitrate PM formation:
HNO3 + NH3 <---> NH4NO3 (aq,s) Gas Phase :
Gas Phase : (daytime)
VOC + OH --->
NO2 + OH ---> HNO3
Organic PM(semiGas &Aq Phase : (nighttime)
volatile)
N2O5 + H2O ---> HNO3
(Long-chain VOCs,
Aromatics, Biogenic VOCs)
Impact of 50% NOx emission reduction
O3 decrease
HOx decrease
Impact of 50% NOx emission reduction
Nitrate PM decrease
Sulfate PM decrease
Third-Generation Air Quality Models:
U.S.EPA’s Models-3/CMAQ

“Open-Access” Community-Based Models :
• User-friendly, Modular, Common modeling framework for
scientists and policy-makers.

Advanced Computer Technologies :
• High performance hardware and software technologies
(Cross-platform, GUI, distributed computing, visualization
tools, etc.).
 “One-Atmosphere”
Modeling :
• Multi-pollutant (Ozone, PM, visibility, acid
deposition, air toxics, etc.), Multi-scale.
Models-3/CMAQ System Framework
Meteorology
Processor
or
RAMS
Emission
Processor
or
SMOKE
Air Quality
Model
PAVE
Models-3/CMAQ Demo & Evaluation:
USEPA/ORD
Domain :
Eastern U.S.A.
Grid Resolution :
36-km/12-km/4-km
(Nested Modeling)
Episode :
July 6 - 16, 1995
O3 Episode in the Northeast U.S. (7/12-15, 1995)
Nested 4 km grid domain (144 x 147 cells)
Measurement Sites and Terrain Features
(Courtesy of USEPA/ORD, Daewon Byun)
Model Predictions vs. Observations
160
120
( E)
O3
(p p b ) 2 5 0 1 3 0 0 0 8
80
40
0
120
obs.
04 km
( D )
12 km
36 km
80
40
25 01300 03
0
( C)
120
09 00310 03
80
40
0
120
( B )
80
40
09 00700 07
0
( A )
120
09 00911 23
80
40
0
19
EST
07
July 13
19
07
July 14
19
EST
07
July 15
19
Models-3/CMAQ Applications
at EPA/OAQPS

Western U.S. Application
• Episodic O3, July 96, 36/12 km, Evaluation completed

Annual Nationwide U.S. Application
• 1-atmosphere, annual 1996, 36-km, evaluation &
diagnostics, on-going annual 2000

Eastern U.S. Application
• 1-atmosphere, July 95, urban applications, 36/12/4-km,
emissions control & growth

Intercontinental Transport/Air Quality & Climate Change
• Intercontinental transport and climatic effects of air
pollutants
Models-3/CMAQ Modeling: Domain Maps
36 km western US ozone domain
36 km Annual National US domain
Western U.S. Application

Objectives :
• New M3/CMAQ Domain
• New Episode (July 1996)

177
Model Setup :
•
•
•
•
•
Episodic O3 modeling
Meteorology : MM5
Emissions : Tier-2 regridded
36km/12km, 12 layers
Compared against UAM-V
153
Modeled
Observed
O3 Conc. and Trend
( ppm & ppm / hr)
Process Contribution to O3
(ppm / hr)
Process Analysis : (Los Angeles grid)
Diffusion
Chem
O3 Conc.
dO3/dt
Time Step (7/19 - 7/31/96)
Western U.S. CMAQ Ozone Modeling
BEIS3 Sensitivity Testing -- Western U.S.
Annual Nationwide U.S. Application

Features :
• Annual CMAQ Run
• Nationwide CMAQ Domain

Model Setup :
•
•
•
•
•
Annual PM and O3 (1996)
36-km, 8 vertical layers
Meteorology : MM5
Emissions Processing: SMOKE
Model Evaluation: Compared against observed
data (IMPROVE & CASTNET) & REMSAD
NOx Emissions
SO2 Emissions
July 1, 1996
Models-3/CMAQ Simulation: Annual Average
PM 2.5
Sulfate PM
Nitrate PM
Organic PM
Models-3/CMAQ : Monthly Average (July)
PM 2.5
Sulfate PM
Nitrate PM
Organic PM
Models-3/CMAQ : Monthly Average (January)
PM 2.5
Sulfate PM
Nitrate PM
Organic PM
National 1996 CMAQ Modeling:
CB4
vs.
RADM2
Nitrate PM (Jan. 1996)
Formation of Secondary PM :
Sulfate PM formation:
H2SO4 + 2 NH3 ---> (NH4)2SO4 (s)
Gas Phase:
O2,H2O
SO2 + OH
---> H2SO4
Aqueous Phase:
H2O
SO2 + H2O2 ---> H2SO4 (Dominate over low pH)
SO2 + O3 ---> H2SO4
Nitrate PM formation:
HNO3 + NH3 <---> NH4NO3
(aq,s)
Gas Phase : (daytime)
NO2 + OH ---> HNO3
Gas &Aq Phase : (nighttime)
N2O5 + H2O ---> HNO3
Oraginc PM formation:
Gas Phase :
VOC + OH --->
Organic PM(semivolatile)
(Long-chain VOCs,
Aromatics, Biogenic VOCs)
NH3 Sensitivity Modeling
Nitrate PM : (January Avg.)
Base
50% NH3 reduction
0.62
0.70
0.09
1.69
0.08
0.53
1.16
1.24
Seasonal Average Total Nitrate* Concentrations
Castnet Dry Deposition Network
Annual CMAQ 1996 Modeling
18
CMAQ Predicted (ug/m3)
16
14
12
Summer
10
Fall
Spring
8
Winter
6
4
2
0
0
2
4
*Total nitrate= Nitric acid + particulate nitrate
6
8
10
12
Castnet Observed (ug/m3)
14
16
18
National 1996 CMAQ Modeling: O3
(July Max in ppb)
149
56
99
62
134
76
137
107
215
132
139
126
131
127
144
113
194
119
National 1996 CMAQ Modeling: Visibility
(January Average in Deciview)
18
11
16
19
14
9
16
26
10
9
24
26
15
22
12
22
26
24
24
14
23
National 1996 CMAQ Modeling (January average)
Sulfur Wet Deposition
Nitrogen Wet Deposition
OAQPS Modeling Initiative on Intercontinental
Transport and Climatic Effects of Air Pollution
Air Pollution/Climate Change Modeling Initiative
Background :

O3 and PM are not only key air pollutants, but also major climateforcing substances;

Reduction of non-CO2 substances (e.g., O3 and PM, especially
black carbon) could be a viable alternative to CO2 reduction to
curb global warming. A key strategy suggested was to focus on
air pollution to benefit regional and local air quality and global
climate simultaneously (Hansen et al., PNAS, 2000);

Black carbon could be the second largest heating component after
CO2 contributing to global warming; Control of fossil-fuel black
carbon could be the most effective method of slowing glabal
warming (Jacobson, Nature, 2001);
Climatic Effects of Air Pollution
O3 (0.3+0.1)
Black (0.8)
Carbon
(Hansen et al., PNAS, 2001)
“Kyoto Protocol also failed to address
two major pollutants that have an impact
on warming, black soot and
tropospheric ozone. Both are proven
health hazards. Reducing both would
not only address climate change, but
also dramatically improve people’s
health.”
(President Bush, June 12, 2001, New
York Times)
Air Pollution/Climate Change Modeling Initiative
Background (continued):

There is also mounting evidence that criteria pollutants originating from
some developing countries, especially those in Asia such as China and
India, could impact U.S. domestic air quality as well as contribute to the
global background of climate-forcing substances. This intercontinental
transport issue is expected to worsen with the rapid growth in emissions
in these regions.

For example, recent modeling studies showed that by 2020 Asian
emissions could contribute as much as 2 ~ 6 ppb of O3 in the western
U.S., offsetting the Clean Air Act efforts up to 25% in that region (Jacob et
al., Geophys. Res. Letts., 1999) and increase global mean O3 level up to
10% (Collins et al., Atmos. Env., 2000); Asian and Sahara dust could
contribute a significant amount of PM in the western and southeastern
U.S. (Husar, http://capita.wustl.edu/CAPITA/).
Asian Dust Storm Event: April 2001 (NASA/TOMS)
(4/7)
(4/9)
(4/11)
(4/12)
(4/13)
(4/14)
Transport of CO : March 2000 (NASA/MOPITT)
(3/10)
(3/12)
(3/13)
(3/15)
(Byun and Uno, 2000)
Air Pollution/Climate Change Modeling Initiative
Objectives :



To assess available approaches for evaluating the
linkage of air pollution to climate change and
enhancing modeling capacity within EPA to address
these linkage issues.
To explore the impacts of intercontinental transport of
O3 and PM as well as their implications for US domestic
and regional air quality and global climate change
To design integrated emissions control strategies to
benefit global climate and regional and local air quality
simultaneously
Air Pollution/Climate Change Modeling Initiative
Work Plan :
Phase I : Short-Term (~6 months)
 Establish a better scientific foundation to address the
issues related to intercontinental transport and climatic
effects of air pollutants by leveraging current studies
1. Global Modeling of O3 and PM
2. Global Radiative Forcing of Aerosols
3. Emission Inventories for Climate-Forcing Pollutants
 Hold a “Workshop on Air Quality and Climate Change”
and establish an expert advisory panel to provide
guidance in developing a conceptual model and
modeling protocol for Phase II work.
Air Pollution/Climate Change Modeling Initiative
Work Plan : Phase II - Long-Term (~2 years)
Based on Phase I effort, a series of activities will be conducted.
These efforts may include, but not limited to:
1. Support continued development of global and regional modeling
capabilities for studying “policy-relevant” climatic effects of air
pollution and the impacts of intercontinental transport
2. Improve global and regional emission inventories for global and
regional modeling of O3 and PM
3. Develop nesting capability between global chemistry/climate models
and regional air quality models
4. Simulate hemispheric or regional air quality under a variety of
current and future global and regional emission scenarios
5. Evaluate global and regional air quality models using a diverse set
of observational data sets, including data from satellites, surface
networks, intensive field studies, etc
6. Assessment of the potential radiative forcing and climate benefits
resulting from planned and alternative non-CO2 control strategies
Air Pollution/Climate Change Modeling Initiative
Models-3/CMAQ Run Example
Ozone (ppm) 1998 April 11: 1200 UTC
Trans-Pacific O3 simulation
(Byun and Uno, 2000)
3D Tracer from Gobi Desert
(MCNC, 2000)
Charge Questions
1. Support continued development of global and regional modeling
capabilities for studying “policy-relevant” climatic effects of air pollution
and the impacts of intercontinental transport
• Are there existing global/regional models that can be practically used
for the assessment of these two issues simultaneously?
• If not, which models are better for addressing the climatic effects of air
pollution (global climate/chemistry model)? Which are better for
addressing the impacts of intercontinental transport (global
chemistry/regional air quality models)?
2. Improve global and regional emission inventories for global and regional
modeling of O3 and PM
• Are current global and regional EI sufficient for O3 and PM modeling?
• If not, what are the weaknesses of exising regional/global EI?
– O3 and PM precursors (NOx, SOx, VOC, NH3, etc.), black carbon?
– Geographic distribution (Asia, America, Europe?) and
resolution (global, regional, and urban)?
– Source categories (biomass burning, biogenic emissions, domestic,
mobile/point/area sources, etc.)?
– EI modeling tools to convert EI to data needed for modeling?
Charge Questions
3. Develop nesting capability between global chemistry/climate models and
regional air quality models
• Are the chemical boundary conditions sufficiently represented in
regional air quality models?
• Do global models have sufficient resolution to address regional
impacts?
• Is grid-nesting between global and regional models a good approach
to bridge these gaps?
4. Simulate hemispheric or regional air quality under a variety of current and
future global and regional emission scenarios
• What current and future scenarios are to be simulated?
– Emissions sensitivity scenarios (NOx, SOx, VOC, BC, CH4, etc.)?
– Source categories sensitivity scenarios (fossil fuel, transportation,
biomass burning, etc.)?
– IPCC & LRTAP emission scenarios?
– Climate change scenarios? Energy use scenarios?
Charge Questions
5. Evaluate global and regional air quality models using a diverse set of
observational data sets
• What data sets are available for evaluating the model results?
– Satellite and spacecraft? Surface network? Remote/Sentinel
monitoring stations? Special field studies?
– How to effectively use the observed data to evaluate against the
model results?
6. Assessment of the potential radiative forcing and climate benefits resulting
from planned and alternative non-CO2 control strategies
• Can global climate models be used to estimate the climate-forcing
effects of air pollutants?
– How to account for the photochemistry of O3 and PM?
•
Can regional air quality models be used to estimate the climateforcing effects
– How to translate changes in pollutant concentrations to climaticforcing? Can we assume changes in pollutant concentration is
linear to changes in radiative properties?
– How to extrapolate from regional-/hemispheric-scale modeling
results to global-scale climate change?