Incorporation of the Model of Aerosol Dynamics, Reaction, Ionization and Dissolution (MADRID) into CMAQ Yang Zhang, Betty K. Pun, Krish Vijayaraghavan, Shiang-Yuh Wu and Christian Seigneur AER, San Ramon, CA CMAQ Workshop, October 2002
Jan 02, 2016
Incorporation of the Model of Aerosol Dynamics, Reaction, Ionization and Dissolution (MADRID)
into CMAQ
Yang Zhang, Betty K. Pun, Krish Vijayaraghavan, Shiang-Yuh Wu and Christian Seigneur
AER, San Ramon, CA
CMAQ Workshop, October 2002
MADRIDModel of Aerosol Dynamics, Reaction, Ionization,
and Dissolution
Gas/particle mass transfer
• Hybrid algorithm
• Full equilibrium algorithm
Coagulation not important under polluted conditions
Condensable gases
ExistingParticles
Nucleation
Condensation
Coagulation
Gas-to-Particle Conversion Processes in MADRID
• Nucleation (McMurry and Friedlander,1979)• Thermodynamic equilibrium for inorganic species
– ISORROPIA (SO4=, NO3
-, NH4+, Na+, Cl-, water)
• Equilibrium for organic species– Absorption based on empirical data– Dissolution and absorption from first principles
• Diffusion-limited condensation/volatilization– Hybrid mass transfer from Calpado & Pandis or from
Meng et al.– Moving-center algorithm of Jacobson
Major Differences between MADRID and Original CMAQ Module
CMAQ• Modal size distribution
• NH4+, SO4
=, NO3-, Na+, Cl-
• Coagulation• Nucleation• Full equilibrium approach to
simulate mass transfer• Standard dry deposition• Absorption (irreversible) of 6
SOA using chamber data
MADRID• Sectional representation• Same species • Not treated• New particle formation• Hybrid or full equilibrium
approach• Revised flux approach• Two SOA modules
available
SOA Modules in MADRID
MADRID 1• Modified CBM-IV &
RADM2• 4 anthropogenic SOA
(aromatics)• 34 biogenic SOA
(monoterpenes)• Absorption based on
smog chamber data (Odum et al., 1997; Griffin et al., 1999)
MADRID 2• CACM(1)
– 42 condensable products– hydrophobic surrogate SOA
• 4 anthropogenic, 1 biogenic– hydrophilic surrogate SOA
• 3 anthropogenic, 2 biogenic
• Absorption based on estimated properties
• Dissolution into existing aqueous particles
(1) Caltech atmospheric chemistry mechanism
Meteorology
Sectional Modal
CBM-IV / RADM2 + 19 biogenic
reactions or CACM
Sectional PM module
PM concentrations
Sectional
PM chemicalconcentrationsby size section
PM depositionflux by chemical
Pre-Processors
ChemicalTransportModel
Output
Conversion from modal to sectional
PM concentrations
Gas-phase: CBM-IV + 3 biogenic
reactions
Modal PM module
PM chemicalconcentrations
by mode
Dry Deposition(sectional Vdep)
Dry Deposition(modal Vdep)
Emissions, initial conditions, boundary conditions (modal)
ModalSectional
Modal
Incorporation of MADRID into Models-3
Los Angeles Application
• SCAQS episode of 27-28 August 1987
• Simulation using MM5 and CMAQ-MADRID 1
SCAQS 1987 Episode
• 25-29 August 1987• Domain: 63 x 28 grid cells, consistent with previous
modeling exercises• Grid Resolution: 5 km• MM5 used to generate input meteorology• Emission inventory developed from previous
simulations
H AW T
C ELAR IVR
Ventura Los Angeles S an B ernard ino
R ivers ide
S an D iego
Orange
SCAQS Modeling Domain
HAWT
CELARIVR
Model Performance Ozone and PM2.5
Species Error Bias
O3 34% 9%
PM2.5 44% 14%
Model Performance PM2.5 Components
Species Error Bias
Sulfate 38% 11%
Nitrate 45% -38%
EC 54% -20%
OC 49% -22%
Observed and Simulated PM2.5 Composition
27 August 28 August
Sulfate
Nitrate
Ammonium
EC
OC
Others
Observations
MADRID 1
11%
42%
17%
2%
8%
20% 13%
37%
16%
2%
9%
23%
10%
42%
13%
4%
16%
15%9%
42%
13%
4%
15%
17%
Nashville, Tennessee Application
• SOS episode of 15-18 July 1995
• Simulation using MM5 and CMAQ-MADRID 2
Model Performance Ozone, PM2.5 and Sulfate
Species Error Bias
O3 17% 4%
PM2.5 17% -15%
Sulfate 13% -11%
Formation of Condensable Organics
0
2
4
6
8
10
12
14
16
18
20
0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68
Time (hr)
Gases
Particles
Condensable products in Nashville
Time (hour)
Con
cent
rati
ons
(g/
m3 )
Formation of Particulate Organics
AEC
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68
SO
A (
g/m
3 )
Biogenic
Anthropogenic
Nashville
Time (hour)
SO
A (g
/m3 )
Hydrophobic vs. Hydrophilic Organics
NAS
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68
SO
A (
g/m
3 )
Hydrophilic SOA
Hydrophobic SOA
Nashville
Time (hour)
SO
A (g
/m3 )
0
2
4
6
8
10
12
14
0 24 48 72
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Sensitivity of Hydrophilic Organics to Henry’s Law Constant
Nashville
Time (hour)
Hyd
roph
ilic
SO
A (g
/m3 )
RH
Base case; H = 1.6 x 106 M/atm Sensitivity case; H = 109 M/atm
RH
Other Applications of MADRID
• Nashville– comparison of three SOA modules
• BRAVO– regional simulation with RADM2 and MADRID 1
• Southeast– applications of MADRID 1 and MADRID 2
• Eastern United States– application of MADRID 1 for one year for nitrogen
deposition
Lessons from PM Simulations
• Accurate PM emission inventories are critical • Secondary organic aerosols remain a major source of
uncertainty• Boundary conditions can have significant effects on
O3 and PM predictions
• Effects of clouds on sulfate need to be simulated for regional haze
• Models yet to be tested for wintertime conditions
Acknowledgments
• Funding for this work was provided by EPRI and CARB
• We would like to thank – J.H. Seinfeld, S. Pandis, M. Jacobson, R. Griffin,
and A. Nenes for providing source codes used in MADRID
– S. Leduc and F. Binkowski for discussions regarding CMAQ