technical highlights of epa’s 7 conference on air quality ... · iii fax question sheet apti...
TRANSCRIPT
EnvironmentalPrograms
Technical Highlights ofEPA’s 7th Conference onAir Quality Modeling
Workshop GuideAPTI Workshop T-029DAY 2
Developed by Environmental Programs - North Carolina State UniversityEPA Cooperative Assistance Agreement CT-825724________________________________________________________________________________
Industrial Extension Service College of Engineering North Carolina State University
ii
This project has been funded wholly or in part by the United States Environmental Protection Agency(U.S. EPA) under Cooperative Assistance Agreement CT-825724 to North Carolina State University.The contents of this document do not necessarily reflect the views and policies of U.S. EPA, nor doesmention of trade names or commercial products constitute endorsement or recommendation for use.
2000 North Carolina State University
All rights reserved. State and local air pollution control agencies, U.S. EPA offices, and federal officesdesignated by EPA are authorized to make royalty-free copies of this document in connection withtelecourses. Otherwise, no part of this publication may be reproduced or transmitted in any form or byany means electronic or mechanical, including photocopying, recording, or information storage andretrieval systems, for any purpose without the express written permission of North Carolina StateUniversity.
Printed on recycled paper in the United States of America.
(Revision: 8/00)
iii
Fax Question Sheet
APTI Workshop T-029Technical Highlights of EPA’s 7th Conference onAir Quality Modeling DAY 2August 2, 2000
Voice: (800) 742-9813 Fax: (800) 553-7656
____________________________________________________________________________________Please write your question and direct it to the appropriate presenter if possible.
Question for:__________________________________________________________________
Question from: ________________________________________________________________
City/State:____________________________ Office Phone Number: ( ) ______________
KeyPoint Services International, Inc. Trouble Line: (800) 231-8945
iv
(This page intentionally left blank.)
v
APTI Workshop T-029
Technical Highlights of EPA’s 7th Conference on Air Quality Modeling
Presented by OAQPS
Broadcast Agenda
August 2, 2000 1:00pm ET DAY 2
SECTION TOPIC
1 Introduction Jim Dicke
2 Alternative ModelsADMS David Carruthers, Ph.D.CAMx Ralph MorrisSCIPUFF Ian Sykes
10 MIN. BREAK
3 Alternative ModelsHYROAD Introduction Edward CarrHYROAD Intersection Model Robert IresonUAM-V Edward Carr
4 SummaryJoe Tikvart
10 MIN. BREAK
Questions and Answers and Wrap up
Technical Highlights of EPA’s 7th Conference on Air Quality Modeling
Presenters
Edward CarrICF ConsultingSan Francisco, [email protected]
Dr. David J. CarruthersCambridge Environmental ResearchConsultants, Cambridge, [email protected]
Alan J. CimorelliRegional Modeler, EPA Region IIIPhiladelphia, [email protected]
Julie Ann DraperFederal Aviation AdministrationWashington, [email protected]
Chuck HakkarinenEPRIPalo Alto, [email protected]
Robert IresonICF ConsultingSan Francisco, [email protected]
John S. IrwinMeteorologist, NOAAAir Quality Modeling Group, [email protected]
Ralph E. MorrisENVIRON International Corp.Arlington, [email protected]
Robert J. Paine
ENSR CorporationActon, [email protected]
Warren D. PetersEnvironmental EngineerAir Quality Modeling Group, [email protected]
Joseph S. ScireVice President, Earth Tech, Inc.Concord, [email protected]
R. Ian SykesARAP GroupTitan [email protected]
Theodore G. ThrasherCSSI, Inc.Washington, [email protected]
Joe TikvartLeader, Air Quality Modeling Group, EPAResearch Triangle Park, [email protected]
John VimontNational Park ServiceDenver, [email protected]
Roger L. Wayson, Ph. D.University of Central Florida and Visiting Scientist, [email protected]
Jeffrey C. WeilCIRES, University of ColoradoBoulder, [email protected]
Rob WilsonRegional Modeler, EPA Region 10
Seattle, [email protected]
Presentation 1 1
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingADMS Atmospheric Dispersion Modeling System
Dr. David J Carruthers
Dr. David J CarruthersCambridge Environmental
Research Consultants
ADMSAtmospheric Dispersion
Modelling System
ADMS! Development commissioned in
1988 following a CERC report toregulatory authorities in the UK
! The CERC report highlighted theadvantages of the use ofsurface/boundary layer scalingover Pasquill Gifford stabilitycategories
ADMS! Sponsors include UK’s
Environment Agency, UK Healthand Safety Executive, major powerand chemical companies
Presentation 1 2
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingADMS Atmospheric Dispersion Modeling System
Dr. David J Carruthers
! Development by:
" CERC(including Prof. Julian Hunt,Dr. David Carruthers, Dr. ChristineMcHugh, Dr. Rex Britter)
" University of Surrey(Prof. Alan Robins)
" UK Meteorological Office(Dr. David Thomson)
ADMS
ADMS! ADMS is the leading European
Short Range Air Dispersion Modeland is used extensively in the UKand across Europe
! ADMS has featured in all 6European Workshops onHarmonisation of DispersionModels (1991-present)
Key Features of ADMS! Continuous or discrete releases
! Point, line, area, volumeand jet sources
" treatment depends onreceptor location
Presentation 1 3
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingADMS Atmospheric Dispersion Modeling System
Dr. David J Carruthers
! Skewed-Gaussian model usinglocal boundary layer variables
! Meteorological preprocessor
! Integral plume rise model
Key Features of ADMS
Key Features of ADMS! Building effects
! Complex terrain
! Coastline
! Wet and dry deposition
! Chemical transformation
! Radioactive decay & gamma dose
! Jets and directional releases
! Concentration fluctuations module
! Condensed plume visibility module
Key Features of ADMS
Presentation 1 4
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingADMS Atmospheric Dispersion Modeling System
Dr. David J Carruthers
Regulatory Applications! Multiple buoyant or passive
industrial emissions
! Surface, near surfaceor elevated releases
! Urban or rural areas
! Short (seconds) to long (annual)term averaging times
Flat Terrain Validation
Table 1
Table 2
Summary Scores for ISC3, ADMS and AERMOD(Different model input parameters)
Table 1 from Hanna et al, 6th Workshop on Harmonisation, France Oct 1999Table 2 from Hanna et al, AWMA Meeting, US, June 2000
ISC3 ADMS AERMODBest 4 8 10Middle 10 15 11Worst 10 1 3
ISC3 ADMS AERMODBest 5 19 6Middle 2 5 11Worst 17 0 7
5000SO2 emission rate (g/s)
130Temperature (°°°°C)
22Exit velocity (m/s)
13Stack diameter (m)
200Stack height (m)
Typical input data
Meteorological data:
1 year of hourly sequential datafrom Manchester, UK, 1995
7.52.6Annual mean
3.20.290th
543.995th
1275198th
1809799th
52118899.9th
66931099.99th
681398100th
AERMODADMS 3
Maximum values of Percentilesand Annual Mean Concentration
Power StationComparison
Presentation 1 5
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingADMS Atmospheric Dispersion Modeling System
Dr. David J Carruthers
ModellingBuildingEffects
Two plumemodel
Based onflow field
large building
Building Validation -Wind Tunnel Experiment
small building
Building Validation - Results
0.00001
0.0001
0.001
0.01
0.1
1
0.0
000
1
0.0
001
0.0
01
0.0
1
0.1 1
BRE Dimensionless Concentration K
AD
MS
Dim
ensi
on
les
s C
on
cen
trat
ion
K S 1-opening
W 1-opening
X 1-opening
S 4-openings
W 4-openings
X 4-openings
S 9-openings
W 9-openings
X 9-openings
S 15-openings
W 15-openings
X 15-openings
ADMS=BRE
ADMS=FACTOR 2 > BRE
ADMS=FACTOR 2 < BRE
Comparison of ADMS with wind tunnel results for the large buildingwith 1, 4, 9 and 15 openings, increasing buoyancy flux (S, W and X)
Presentation 1 6
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingADMS Atmospheric Dispersion Modeling System
Dr. David J Carruthers
Above: View of idealized hill
-3000 0 3000-3000
0
3000
Complex Terrain EffectsBased onmodelledflow field
Right: Flow field at 80mabove terrain and groundlevel concentration froman 80m stack
ADMS and AERMOD Comparison:Terrain amplification factors
-2000 -1500 -1000 -500 0 500 1000 1500 2000-2000
-1500
-1000
-500
0
500
1000
1500
2000
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2.2
-2000 -1500 -1000 -500 0 500 1000 1500 2000-2000
-1500
-1000
-500
0
500
1000
1500
2000
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0
3.2
3.4
3.6
3.8
4.0
4.2
4.4
-2000 -1500 -1000 -500 0 500 1000 1500 2000-2000
-1500
-1000
-500
0
500
1000
1500
2000
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
-2000 -1500 -1000 -500 0 500 1000 1500 2000-2000
-1500
-1000
-500
0
500
1000
1500
2000
200
250
300
350
400
450
500
550
Left: ADMS CmaxRight: AERMOD Cmax
Ratio of complexterrain to flatterrain as functionof stack location
• Neutral conditions• 50m stack• Idealised hill
Left: ADMS XmaxRight: AERMOD Xmax
-1500 -1000 -500 0 500 1000 1500 20000
250
500
750
-1500 -1000 -500 0 500 1000 1500 20000
250
500
750
0.0
0.5
1.0
1.5
2.0
2.5
5.0
10.0
15.0
20.0
25.0
Ratio of complex terrainto flat terrain maximumconcentrations asfunction of stack heightand location
Wind tunnel, fromLawson et al
ADMS
AERMOD
ADMS and AERMOD Comparison:Complex terrain, Neutral flow
Presentation 1 7
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingADMS Atmospheric Dispersion Modeling System
Dr. David J Carruthers
-1500 -1000 -500 0 500 1000 1500 20000
100
200
300
400
-1500 -1000 -500 0 500 1000 1500 20000
100
200
300
400
0
1
2
3
4
5
6
7
8
9
10
20
30
40
45
50
ADMS
AERMOD
ADMS and AERMOD Comparison:Complex terrain, Stable flow
Ratio of complex terrainto flat terrain maximumconcentrations asfunction of stack heightand location
• Wind Speed (10m)= 0.76 m/s
• Monin-Obukhov length= 23 m
ADMS and AERMOD Comparison:Complex terrain results
369000 375000 381000437000
443000
449000
25
50
75
100
125
150
175
200
225
250
369000 375000 381000437000
443000
449000
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
MaximumConcentration (ug/m3)
Long Term AverageConcentration (ug/m3)
ADMS (Max=178) ADMS (Max=4.0)
AERMOD (Max=1162) AERMOD (Max=10.3)369000 375000 381000
437000
443000
449000
369000 375000 381000437000
443000
449000
Stack andsurrounding
terrain
ADMS-Urban• Advanced airquality modelfor major citiesand airports
• Includesalgorithms fordispersion fromroads andstreet canyons
• Used in majorcities in Europeand beyond
LondonAnnual average NO2 concentration, 2005
Presentation 1 8
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingADMS Atmospheric Dispersion Modeling System
Dr. David J Carruthers
Summary! ADMS includes in one model all
the features of AERMOD (exceptinput of observed boundary layerprofiles), ISC-PRIME and CTDMPLUS – potential difficulties arisingas to whether to use AERMOD orISC-PRIME avoided (e,.g, at sitewith buildings and tall stacks)
! Additionally ADMS includesconcentration fluctuation, plumechemistry and condensed plumevisibility algorithms
Summary
Summary! ADMS-Urban includes capabilities
of CALINE, most features of EDMSand other features
! The costs of ADMS are similarto commercially availableversions of AERMOD and ISC
Presentation 1 9
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingADMS Atmospheric Dispersion Modeling System
Dr. David J Carruthers
! ADMS was first released in1993 and has been used inmany critical applications
! There are over 500licenses worldwide
Summary
http://www.cerc.co.uk
Presentation 2 1
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
ComprehensiveAir-quality
Model with extensions(CAMx)
Ralph E. MorrisENVIRON International Corp.
! 3-D Eulerian troposphericphotochemical transport model
" treats emissions, chemistry,dispersion, removal of gaseous andaerosol air pollution
" scales range from individual pointsources (< 1 km) to regional (>1000 km)
CAMx Version 2.00
! Combines features required of“state-of-the-science” models
" new coding of several industry-accepted algorithms
" computationally and memoryefficient
" easy to use
CAMx Version 2.00
Presentation 2 2
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
" modular framework permits easysubstitution of revised and/oralternate algorithms
" publicly available (www.camx.com)
CAMx Version 2.00
! Technical Features:
! Grid nesting
" two-way horizontal and verticalnesting
" supports multiple levels
" variable meshing factors
CAMx Version 2.00
! Plume-in-Grid (PiG) sub-model
! Multiple, fast and accuratechemical mechanisms
! Mass conservative and massconsistent transport scheme
CAMx Version 2.00
Presentation 2 3
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
0 1 2 3 4 5 6 7 8 9 100
1
2
3
4
5
6
7
8
9
10
Coarse Grid Boundary Cells
Coarse Grid Boundary Cells
Co
ars
e G
rid
Bo
un
da
ry C
ells
Co
ars
e G
rid
Bo
un
da
ry C
ells
Fine Buffer Cells
Fine Buffer Cells
Fin
e B
uff
er
Ce
lls
Fin
e B
uffe
r C
ells
Boundary of Fine Grid 1
Boundary of Fine Grid 2
! Multiple map projections
" curvi-linear latitude/longitude
" Universal Transverse Mercator
" Lambert Conformal (MM5)
" Rotated Polar Stereographic (RAMS)
CAMx Version 2.00
! Ozone Source Apportionment(OSAT)
" tracks source region/categorycontributions to receptor ozoneconcentrations
" indicates if ozone formed in NOxor VOC-limited conditions
CAMx Version 2.00
Presentation 2 4
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
! Ability to use historical airquality model databasesdeveloped for other models
" OTAG, LMOS, COAST/Houston,Atlanta, Northeast Corridor
CAMx Version 2.00
Key Technical Components! Overview
" solves continuity equationfor each species
" time splitting operation
# each process solved individuallyfor each grid, each time step
" time step size maintains stablesolution of transport on each grid
# multiple transport steps permaster grid step required for
# nested grids
# multiple chemistry stepsper transport step required
" model developed to run onmeteorological modeling grid
# reduces error due tointerpolation and averaging
# multiple map projections available
Key Technical Components
Presentation 2 5
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
! Transport
" advection and diffusion solvers aremass conservative
" horizontal and vertical advectionlinked through the divergentcompressible atmospheric continuityequation
# mass consistency
Key Technical Components
" order of east-west andnorth-south advectionalternates each master grid step
" three options available forhorizontal advection solvers:
# Smolarkiewicz (1983): diffusion-corrective forward-upstream scheme
# Bott (1989): area-preservingflux-form solver
# Piecewise Parabolic Method (PPM)
Key Technical Components
! Transport (concluded)
" vertical transport and diffusionsolved with an implicit scheme
" dry deposition rates are used asthe surface boundary condition
" horizontal diffusion solvedwith an explicit scheme intwo directions simultaneously
Key Technical Components
Presentation 2 6
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
! Pollutant Removal
" dry deposition velocities for eachspecies determined using resistanceapproach (Wesely, 1989)
# dependent upon: season, land cover,solar flux, near-surface stability,surface wetness, species solubilityand diffusivity
# for aerosols: size spectrumdictates sedimentation velocity
Key Technical Components
" wet scavenging based onMaul (1980) as implementedin CALPUFF (EPA, 1995)
# exponential decay
# decay rate dependent upon:rainfall rate, species solubility
# species removed from entiregrid column (all layers)
Key Technical Components
! Photochemistry
" CBM-IV (Gery et al., 1989)
# 3 variations available
" SAPRC97 (Carter, 1990)
# chemically up-to-date
# tested extensively againstenvironmental chamber data
# uses a different approachfor VOC lumping
Key Technical Components
Presentation 2 7
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
" all mechanisms are balancedfor nitrogen conservation
" photolysis rates derivedfrom TUV preprocessor
# generates lookup table over:zenith angle, altitude,ozone column, albedo, turbidity
# first two determined foreach grid cell internally
# last three provided by input files
Key Technical Components
" photolysis rates affected by clouds
# UAM-V approach: rates scaled byfractional cloud coverage only
# RADM approach: rates scaled byoptical depth and cloud coverage
Key Technical Components
! Chemistry Solver
" most “expensive” component ofphotochemical grid simulations
" CAMx solver increases efficiencyand flexibility
" adaptive hybrid approach:
# radicals (fastest reacting species)solved using implicit steady stateapproximation
Key Technical Components
Presentation 2 8
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
# fast state species solved usingsecond-order Runge-Kutta method
# slow state species solved explicitly
# “Adaptive” = number of fast statespecies changes according to thechemical regime
Key Technical Components
! Plume-in-Grid (PiG)
" fine resolution needed for near-source chemistry/dispersion of largeNOx plumes
" tracks stream of plume segments(puffs) in a Lagrangian frame
# each puff moved by winds in host cell
# puff growth (dispersion) determinedby diffusion coefficients in host cell
Key Technical Components
" GREASD PiG: faster, conceptuallysimpler
# reduced NOx chemistry set(NO-NO, NOx/ozone equilibrium, HNO3production)
# cross-sectional Gaussianpollutant distribution
Key Technical Components
Presentation 2 9
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
# puffs leak mass according togrowth rates and grid cell size
# puffs terminated due to ageor sufficiently dilute NOx
Key Technical Components
! Ozone Source Apportionment(OSAT)
" determines source area/categorycontributions to ozone anywherein the domain
" uses tracers to track precursoremissions and ozoneproduction/destruction
Key Technical Components
Presentation 2 10
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
" also tracks contribution of initialand boundary conditions
" estimates whether ozone is producedunder NOx- or VOC-limited conditions
Key Technical Components
" removes need to run modelrepeatedly to understand:
# chemical regime
# influences of various sources
" HOWEVER: cannot quantify ozoneresponse to NOx or VOC controls
Key Technical Components
! CAMx Version 2.00 PM Treatment
" Primary Particulate Matter (PM)
" Secondary Organic Aerosols(SOA) treated using aerosolyield approach in photochemistry
" Sulfate/Nitrate/Ammonia equilibriumaerosol thermodynamics usingempirical UAM/LC approach
" Empirical aqueos-phase (Sulfate)
Key Technical Components
Presentation 2 11
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
CAMx Version 3 AttributesComing Summer 2000
(www.camx.com)! Flexi-nesting
" Ability to add/delete nested-gridsduring a simulation
" Real time interpolation of fine-gridinputs from next coarser grid
CAMx Version 3 AttributesComing Summer 2000
(www.camx.com)! Decoupled Direct Method (DDM)
" Sensitivity coefficients providesinformation on the relationship ofCAMx-estimated ozone (or otherspecies) and sources of precursors(emissions, boundary conditions,and initial concentrations)
" Information is useful for
# Control Strategy Development
# Model performance evaluation
# Diagnostic analysis
CAMx Version 3 AttributesComing Summer 2000
(www.camx.com)
Presentation 2 12
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
Moving Sectional
Hybrid
CAMx Version 3 Attributes
AerosolSize Composition Distribution (fixed size grid)
Bulkaqueous-phase
chemistry
Size-resolvedaqueous-phase
chemistry
Equilibrium Dynamics
SCAPE2AERO-LC SEQUILIB2.1
Fixed Sectional
Aerosol Module
Cloud Microphysics/Chemistry Module
Advanced PM Treatment
ISORROPIA
Mass Transfer:
Advanced PM Treatment
Gas-PhaseChemistry:
Size Section:
1) SAPRC972) Enhanced CBM-IV
(monoterpines)
1) Fixed Section2) Moving Section
1) Equilibrium2) Hybrid3) Dynamic
CAMx Version 3 Attributes
SecondaryOrganicAerosol:
1) SOAM22) Aerosol Yields (existing)
Advanced PM TreatmentCAMx Version 3 Attributes
AerosolThermo-dynamic:
Aqueous-PhaseChemistry:
1) LCAERO (parameterized RFM)2) SCAPE2 (full science)3) ISORRPIA (significantly faster)
1) Bulk2) Size Resolved3) Empirical (existing)
Presentation 2 13
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
WetDeposition:
1) Rainout and washout as part of aqueous-phase module2) Existing gaseous wet deposition interface with aqueous-phase module
Advanced PM TreatmentCAMx Version 3 Attributes
Coagulation:
Nucleation:
DryDeposition:
1) CMU algorithm
1) CMU algorithm
1) Wesley gaseous (existing)2) AERO particle dry deposition
CAMx Postprocessingand Analysis Tools
! CAMxtrct
" Extracts and reformats CAMx outputfor multiple grids (Fortran)
! SURFER (by Golden Software)
" Visualization (PC based)
! MAPS (by Alpine Geophysics)
" Model evaluation and visualization(Fortran/NCAR graphics)
! PAVE (by MCNC)
" Visualization (Unix/LINUX)
CAMx Postprocessingand Analysis Tools
Presentation 2 14
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
CAMx Postprocessingand Analysis Tools
! VIS5D
" Visualization (various)
! CAMxDESK (by EnviroModeling)
" Visualization and analysis software(PC based)
Contribution to 1-Hour Ozone > 124ppb in the Lake Michigan Region for theJuly, 1995 Episode Composite using Base Case EPA1A (bar 1) and SIP Call
SIP2A (bar 2). EPA1A Conc = 141ppb; SIP2A Conc = 133ppb; n =1515
0
10
20
30
40
50
60
70
LakeMI IL BC W-MO OK E-MO AR KS TX KY TN IN
Source Region
Ozo
ne
Co
ntr
ibu
tio
n (
pp
b)
IC/BC
Area
Vehicles
Point
Biogenic
Contribution to 8-Hour Ozone > 84ppb in the Lake MichiganRegion for the July, 1995 Episode Composite using Base Case
EPA1A (bar 1) and SIP Call SIP2A (bar 2). EPA1A Conc =105ppb; SIP2A Conc = 99ppb; n =5383
0
5
10
15
20
25
30
35
40
LakeMI IL BC W-MO E-MO OK AR KS IN KY TX IA
Source Region
Ozo
ne
Co
ntr
ibu
tio
n (
pp
b)
IC/BC
Area
Vehicles
Point
Biogenic
Presentation 2 15
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingComprehensive Air-quality Model with extensions (CAMx)
Ralph E. Morris
Contributions to 8-Hour Ozone > 84 ppb by SourceType for the July, 1995 Episode for 2007 Episode
Composite Base Case
0
20
40
60
80
100
120
Atlanta
Birmin
gham
Boston
Buffalo
Charle
ston
Charlo
tte
Chatta
nooga
Chicag
o+Milw
auke
e
Cinci
nnati
Cleve
land
Colum
bus
Dayto
n
Detro
it
Evansv
ille
Ft Way
ne
Huntingto
n
India
napolis
Knoxvill
e
Louisvi
lle
Mem
phis
Nashvi
lle
New Y
ork
Parke
rsbur
g
Philadel
phia+B
alt
Pittsb
urgh
Portlan
d
Richm
ond
SW-M
ichig
an
St Louis
Terre
Hau
te
Toronto
Was
hingto
n DC
Receptors
Ozo
ne
Co
ntr
ibu
tio
n (
pp
b)
IC/BC
Biogenics
Area
Mobile
Utilities
Industrial
84.5ppb
-97.5 -90 -82.5 -75
Longitude
28.5
31
33.5
36
38.5
41
43.5
46
La
titu
de
1 m/s
Wind
0.02
0.04
0.06
0.08
0.1
ppm
O3
CAMxDESK Overlay of Ozone and Winds
290
294298
298
302
302
302
306
306
306
Longitude
28.5
31
33.5
36
38.5
41
43.5
46
La
titu
de
10
30
50
70
90
%
Urban
286
290
294
298
302
306
310
ºK
Tsfc
CAMxDESK Overlay of SurfaceTemperature and Urban Land use
10298
294
-75-82.5-90-97.5
Presentation 3 1
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingSecond-order Closure Integrated Puff (SCIPUFF)
R. Ian Sykes
R. Ian SykesARAP Group
Titan Corporation
Second-order ClosureIntegrated Puff (SCIPUFF)
Overview! Modeling Approach
! Graphical User Interface
! Input/Output
! Model Evaluation Studies
Lagrangian Puff Model! Concentration field - collection of
overlapping puffs with Gaussiandistributions
! Concentration givenby sum over all puffs
! Solve ODE’s for puff moments
Presentation 3 2
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingSecond-order Closure Integrated Puff (SCIPUFF)
R. Ian Sykes
Lagrangian Puff Attributes! Arbitrary range of scales without
numerical grid and associateddiffusion errors
! Arbitrary time-dependent, spatiallyinhomogeneous conditions
! Multiple sources with arbitrarytime-dependence
( ) ( ) ( )
( )( )
( ) ( ) ( )
( ) ( ) ( )
( )
( )
( )
1 1 ( )exp2( )
-1
i i j j
i
ij
c c Q G
G x x x xijV
Q
x
V
α α α
α α
α α α
α
α
α
ασ
σ
= =
= − − −
=
=
=
= =
∑ ∑x x x
where
total mass of puff (zeroth moment)
puff centroid (first moment)
spatial spread (second moment)
puff volume ( )13 2( )2
2α
σπ
( ) ( ) ( )
( )( )
( ) ( ) ( )
( ) ( ) ( )
( )
( )
( )
1 1 ( )exp2( )
-1
i i j j
i
ij
c c Q G
G x x x xijV
Q
x
V
α α α
α α
α α α
α
α
α
ασ
σ
= =
= − − −
=
=
=
= =
∑ ∑x x x
where
total mass of puff (zeroth moment)
puff centroid (first moment)
spatial spread (second moment)
puff volume ( )13 2( )2
2α
σπ
Concentration Field
Puff Moment Equations
( )
( ) ( )
( ) ( )
( )
( )
( ) ( )
0
( , )i i
i j j i
ij
j iik jk
k k
dQ
dtd
x u tdt
x u c x u cd
dt Q
u u
α
α α
α α
αα
α α
σ
σ σ
=
=
′ ′ ′ ′ ′ ′+=
∂ ∂+ +
∂ ∂
x
( )
( ) ( )
( ) ( )
( )
( )
( ) ( )
0
( , )i i
i j j i
ij
j iik jk
k k
dQ
dtd
x u tdt
x u c x u cd
dt Q
u u
α
α α
α α
αα
α α
σ
σ σ
=
=
′ ′ ′ ′ ′ ′+=
∂ ∂+ +
∂ ∂
x
Presentation 3 3
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingSecond-order Closure Integrated Puff (SCIPUFF)
R. Ian Sykes
Turbulence Closure
( ) ( )( )
i j i j i j
d Aqx u c Q u u x u c
dt
α αα′ ′ ′ ′ ′ ′ ′ ′= −Λ
( ) ( )( )
i j i j i j
d Aqx u c Q u u x u c
dt
α αα′ ′ ′ ′ ′ ′ ′ ′= −Λ
where and Λ is theturbulence length scale.
2i iq u u′ ′=2i iq u u′ ′=
Turbulent Dispersion! Closure model gives direct
relationship between turbulencequantities and diffusion rates
! Provides single diffusion modelframework for a wide range ofatmospheric scales
Model Efficiency! Puff splitting allows accurate
treatment of wind shear
! Puff merging minimizesnumber of puffs
! Efficient adaptivetime-step algorithm
! “Static” puffs forsteady-state section of plume
Presentation 3 4
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingSecond-order Closure Integrated Puff (SCIPUFF)
R. Ian Sykes
c
! Turbulent dispersion implies arandom concentration field
! 2nd-order closure model gives bothfluctuation variance, , and
! The probability distribution ofc is then modeled by the clippednormal distribution
Concentration Fluctuations
2c′2c′
Fackrell & Robins (1982)
0.1 1 10
10
100
1000
1
SCIPUFFMaximum mean dataGround-level data
x / H
CUH2
Q
0 100
2 4 6 8
1
2
3
4
5
6
x / H
ˆ c
cm
SCIPUFF
Concentration fluctuation intensityfor different source sizes
Mean concentration vs downwinddistance
0.1 1 10
10
100
1000
1
SCIPUFFMaximum mean dataGround-level data
x / H
CUH2
Q
0 100
2 4 6 8
1
2
3
4
5
6
x / H
ˆ c
cm
SCIPUFF
Concentration fluctuation intensityfor different source sizes
Mean concentration vs downwinddistance
Plume Rise! Associate dynamic vertical
momentum and temperatureperturbation integrals with each puff
! Add evolution equations for puffdynamics based on conservationof momentum and temperature
Presentation 3 5
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingSecond-order Closure Integrated Puff (SCIPUFF)
R. Ian Sykes
Momentum Rise
2.54.06.08.0
13.215.4
R R=15.4
13.2
8.0
6.0
4.0
2.5
0 10 20 30 40 50 60 700
10
20
30
40
50
60
70
x r0
"z r0
Buoyant Rise
z
U
Wind Shear! Use complete (six-moment)
Gaussian specification so thatshear distortions can beaccurately calculated
Presentation 3 6
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingSecond-order Closure Integrated Puff (SCIPUFF)
R. Ian Sykes
Model Input! Source Data
# Pollutant physical andchemical properties
# Release type
! Meteorological Data
# Fixed winds
# Observational Input(surface and/or profile)
# Time-dependent3-dimensional gridded
Model Input
Model Input! Terrain for mass
consistent wind field
! Turbulence Data
# Planetary boundary layer
# Large scale variability
Presentation 3 7
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingSecond-order Closure Integrated Puff (SCIPUFF)
R. Ian Sykes
! Boundary layer turbulence
# Profiles based on wind speed,roughness, and surface heat fluxcalculation
Model Input
GUI Input
Model Output! Sampler file
# Time history at receptor locations
! GUI plots
# Horizontal slices
# Vertical slices
# Integrated surfacedeposition and dose
# Probability
Presentation 3 8
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingSecond-order Closure Integrated Puff (SCIPUFF)
R. Ian Sykes
GUI Output
GUI Output
Model Evaluation Studies! PGT curves
! Instantaneousdispersion data
! Lab dispersion andfluctuation data
! Continental-scaleANATEX fieldexperiment
! EPRI PMV&D tall-stack emissions
! CONFLUX (shortrange, fluctuations)
! Dugway field tests
! Model Data Archive
! ETEX
Presentation 3 9
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingSecond-order Closure Integrated Puff (SCIPUFF)
R. Ian Sykes
MDA Passive
Pre
dic
ted C
on
centr
atio
n (
ppm
)
Observed Concentration (ppm)
103
10–1
101
10–1 101 103
a - Prairie Grassb - Hanford (Long time average)c - Hanford (Short time average)
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
aa
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
b
b
b
b
b
b
b
c
c
c
c
c
c
c
c
c
c
P
red
icte
d C
on
centr
atio
n (
ppm
)
Observed Concentration (ppm)
103
10–1
101
10–1 101 103
a - Prairie Grassb - Hanford (Long time average)c - Hanford (Short time average)
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
aa
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
b
b
b
b
b
b
b
c
c
c
c
c
c
c
c
c
c
Conflux
0 4 8 12 16 0 4 8 12 16C (ppm) C (ppm)
0
4
8
12
16
0
4
8
12
16
Z(m)
Z(m)
0 4 8 12 16 0 4 8 12 16C (ppm) C (ppm)
0
4
8
12
16
0
4
8
12
16
Z(m)
Z(m)
MeanConcentration
Conflux
ConcentrationFluctuation
Intensity
0 4 8 12 16 20 0 4 8 12 16 20$ c/C $ c/C
0
4
8
12
16
0
4
8
12
16
Z(m)
Z(m)
0 4 8 12 16 20 0 4 8 12 16 20$ c/C $ c/C
0
4
8
12
16
0
4
8
12
16
Z(m)
Z(m)
Presentation 3 10
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingSecond-order Closure Integrated Puff (SCIPUFF)
R. Ian Sykes
ETEX24 Hours
SCIPUFF
Observed
44
49
54
59
64
Latit
ude
(deg
rees
)
48 Hours
44
49
54
59
64
Latit
ude
(de
gree
s)
-5 0 5 10 15 20 25
Longitude (degrees)
72 Hours
-5 0 5 10 15 20 25
Longitude (degrees)
-5 0 5 10 15 20 25
Longitude (degrees)
Model Availability
Available for downloading from theTitan website:www.titan.com/systems/prod.htm
Presentation 4 1
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
The Hybrid RoadwayIntersection Model:
HYROAD
Edward Carr andRobert Ireson
ICF Consulting
Contact InformationCrawford Jencks, P.E.Program Officer, NCHRP202-334-2379 / [email protected]
Edward L. Carr, CCMProject Manager, ICF Consulting415-507-7186 / [email protected]
Robert G. Ireson, Ph.D.Principal Investigator415-925-1440 / [email protected]
Overview! Background
! Scientific basis and modelformulation
! Model application -resource needs
Presentation 4 2
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
! Model sensitivity and performance
! Proposed applications
! Project status and next steps
Overview
Background! SIP, conformity, and EIS
“Hot-Spot” analysis forcarbon monoxide (CO)
! Sponsorship: NCHRP, FHWA
! Four phase study:
" Problem assessment -Site Monitoring Plan (1993-1995)
" Site monitoring and evaluation -Data Collection & Analysis(1993-1997)
" Model development -Evaluation & Testing (1997-2000)
" Development GUI & user guide(2000-2001)
Background
Presentation 4 3
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
PrimaryObjectives of Research
! Assemble a comprehensivenational database
# Model testing and evaluation
! Developed an improved fullyintegrated roadway intersectionmodel
# Dispersion
# Emissions
# Traffic
PrimaryObjectives of Research
Approach toModel Development
! Analysis of field program data tocharacterize dynamic processes
# Assess those elements mostimportant and incorporate intomodel framework
! Develop model based onunderstanding of key processes
Presentation 4 4
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
Scientific Basis andModel Formulation
! Intersection model components
! Limitations of existing models
! Field study design and findings
! HYROAD formulation
Intersection Dynamics! Traffic
# Queuing
# Acceleration, deceleration,and cruise
# Non-steady state
! Emissions and dispersion
# Modal effects(e.g., power enrichment)
# Buoyancy and vehiclewake turbulence
# Short transport distances
Intersection Dynamics
Presentation 4 5
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
Intersection ModelComponents
Outputs
The need for integration
DispersionModule
EmissionModule
TrafficModule
EmissionFactors
MeteorologicalInputs
Roadwayand Traffic
Inputs
Limitations ofExisting Models
! CAL3QHC simplifications
# Two emission states -- idle or cruise
# Steady state meteorology(PG sigma-y and sigma-z)
# Queuing based onquality of progression
! CALINE4 approaches
# Empirical emissionadjustment for acceleration
# Roadway turbulence and buoyancy
# Use of sigma-theta in place of sigma-y
Limitations ofExisting Models
Presentation 4 6
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
Field StudyElements - Continuous
! Traffic approach volumeand signal timing
! Meteorology
# 3 m and 10 m wind speedand direction
# Temperature, RH, sigma-theta, stability
(15 minute average)
! CO and CO2 at 16+ locations
! Sonic anemometers (2- and 3-axis)
Field StudyElements - Continuous
(15 minute average)
Field Study Layout
Presentation 4 7
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
Field Study Elements --Short-term Studies
! Floating car runs
# Time-distance data(1 second resolution)
! Coldstart survey(distance from trip origin)
! Tracer study (SF6)
# Light wind periods
# 15-minute averagesat multiple receptors
Field Study Elements --Short-term Studies
Field Study Findings --Flows and Turbulence
! Induced flows of 3+ m/s at roadside
! Induced flows and enhancedturbulence observable at >25 mfrom roadside
Presentation 4 8
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
! Observed vertical dispersionrate exceeds both PG andCALINE3/CAL3QHC
! SF6 tracer observed >100 m‘upwind’ of release point due toinduced flows
Field Study Findings --Flows and Turbulence
Field Study Findings --Traffic and Emissions
! Speed/acceleration distributionsdo not resemble any emissiontest cycles
! Power enrichmentoccurs on depart legs
! Constant g/gal emission ratesobserved at all locations forhigh concentration periods
! Enrichment does NOT appreciablycontribute to high concentrations
Field Study Findings --Traffic and Emissions
Presentation 4 9
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
Emissions (g/gal) ShowEnrichment on Acceleration
0
1000
2000
3000
0 1000 2000 3000
Approach Leg
De
pa
rt L
eg Emissions > 3x Mean
High CO Events ShowLittle Enrichment
-2
0
2
4
6
8
10
12
14
16
18
20
-50 0 50 100 150 200 250 300
[COx] (ppm)
[CO
] (p
pm
)
Emissions > 3x Mean
Average Emissions
Concentrations > 6 ppm
HYROAD Formulation
Microsimulation
Speed-distribution-based, Modal-ready
Hybrid steady-state and puff
OutputsDispersion
ModuleEmissionModule
TrafficModule
EmissionFactors
MeteorologicalInputs
Roadwayand Traffic
Inputs
Presentation 4 10
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
Traffic Module -Micro-simulation
! Based on TRAF-NETSIM -simulation of vehicle movementsat 1 second resolution
! Explicit treatment of traffic patterns
# Turn lanes, signal phases, queuing
# Coordination of upstream signals
! Output: Speed/accelerationdistribution by 10 m roadwaysegment and signal phase
Traffic Module -Micro-simulation
Denver Speed-Acceleration Profile
-16
-12
-8 -4 0 4 8
12
16
0
20
40
60
0.0%
0.2%
0.4%
0.6%
0.8%
1.0%
Speed (fps)
Accelera t ion
(fps/s )
Id le = 38.6%
Idle = 38.6 %
Accel (fps/s)
Speed (fps)
Presentation 4 11
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
HYROAD Speed-AccelerationProfile (LOS C Test Case)
-12-9
-6-3
03
69
12
0
7
15
22
29
37
44
5159
66
0
100
200
300
400
500
600
700
800
900
1000
Vehic le-seconds
Acceleration (fps/s)
Speed (fps)
e
Idle = 32.3 %
Emission Module! Objective: make the best
possible use of cycle-basedemission factors
! Core rates from MOBILE5for the speeds of the speedcorrection cycles
! Multivariate regression weightsspeed correction cycles to matchnon-idle speed distribution fromNETSIM
Emission Module
Presentation 4 12
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
Emission Module! Total emissions calculated from
weighted average g/mi rate plusexcess idle (idle time not explainedby weighted cycles)
! Speed/acceleration distributionsused to calculate fuel consumptionby 10 m roadway segment andsignal phase
! Fuel consumption used as asurrogate to spatially andtemporally allocate emissions
Emission Module
HYROAD Speed Distribution
0
20000
40000
60000
80000
100000
120000
140000
160000
180000
200000
Idle 5 10 15 20 25 30 35 40 45 50 55 60Speed
Ve
hic
le s
eco
nd
s
(15
min
ute
sim
ula
tio
n)
Traffic Module
Emissions Module
Excess Idle Time Added
Presentation 4 13
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
CAL3QHC Speed Distribution
0
20000
40000
60000
80000
100000
120000
140000
160000
180000
200000
1 2 3 4 5 6 7 8 9 10 11 12 13Speed
Ve
hic
le S
eco
nd
s
Traffic ModuleCAL3QHC
Queued vehicle idle time
Induced Turbulenceand Flow Fields
! Based on ROADWAY(Eskridge, 1987)
! Turbulence and induced flowcalculated for each 10 mroadway segment using trafficvolume and speed
! Output for each signal phase:Gridded wind speed and eddydiffusivity for a 2.5 x 2.5 kmgrid of 10 m cells
Induced Turbulenceand Flow Fields
Presentation 4 14
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
Wind Field with Induced Flows
Dispersion Module! Gaussian model (moderate wind)
# Uses CALINE4 mixing zone, sigma-z, sigma-theta
! Puff model
# 1 puff per second per10 m roadway segment
# Puff transport with griddedwinds by phase
# Puff growth based on local stabilityfrom gridded eddy diffusivity
# Domain wind and stabilityused after 1 cycle
Dispersion Module
Presentation 4 15
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
Model Formulation -Conclusions
! HYROAD integrates acceptedmodeling approaches to treatimportant processes affectingintersection CO concentrations
# Induced flows and turbulence
# High spatial and temporalvariability of emissions
! Modular design allows updating(e.g., for modal emissions)
Model Formulation -Conclusions
Model Application -Resource Needs
! Standard inputs
# Intersection geometry(lanes, medians, etc.)
# Traffic (volume, turns,signal cycles, speeds, coordination)
Presentation 4 16
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
# Meteorology (wind speed, direction,stability, temperature, optional
sigma-theta
# Emission factors by temperature(constant cold-start), or hour-specific
Model Application -Resource Needs
Model Application -Resource Needs
! Input preparation time
# Geometry and NETSIM: 8 hours
# Emissions and Dispersion: 12 hours
! Run time(500 MHz Pentium, 128 Mb)
# Netsim: 30 sec / simulation hour
# Dispersion: 4 min / simulation hour
# Optimization possibilities
Model Application -Resource Needs
Presentation 4 17
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
HYROAD Sensitivityand Performance
! Sensitivity analyses
# Nominal intersection with gridof receptors in NW quadrant
# Modeling with bothGaussian and puff models
# Results for wind speed,wind direction, and stability
1 2 3 4 5 6 7 8
S1
S2
S3
S4
S5
S6
S7
S8
16-18
14-16
12-14
10-12
8-10
6-8
4-6
2-4
0-2
25 m
Concentration Patterns --Puff v. CALINE4
Puff Module CALINE4 Module
CO(ppm)
1 2 3 4 5 6 7 8
Wind
HYROAD Sensitivityand Performance
! Performance evaluation data sets
# Intensive data sets forthree intersections
" 528 hours with 15 minute data(6000 available)
" 10 or more receptor locations
Presentation 4 18
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
# 8 SLAMS/NAMS sites
" 1728 hours (75,000 available)
" Uncertain background concentrations
HYROAD Sensitivityand Performance
HYROAD Sensitivityand Performance
! Performance evaluation approach
# Concurrent evaluation of HYROAD(Gaussian and puff) with CAL3QHCwithout regulatory constraints(D, 1 m/s)
# Scatterplots, stratified by receptor,wind speed, and wind direction
# Standard statistics
" All data
" Max 25 paired
" Max 25 unpaired
HYROAD Sensitivityand Performance
Presentation 4 19
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
Predicted vs. Observed [CO]
-1
0
1
2
3
4
5
6
7
8
9
10
11
-2 -1 0 1 2 3 4 5 6 7 8 9 10 11
Observed (ppm)
Pre
dic
ted
(p
pm
)
HYROAD-Puff
HYROAD-CL4
CAL3QHC
Quantile-Quantile Plot
-1
0
1
2
3
4
5
6
7
8
9
10
11
-2 -1 0 1 2 3 4 5 6 7 8 9 10 11
Observed (ppm)
Pre
dic
ted
(p
pm
)
HYROAD-Puff
HYROAD-CL4
CAL3QHC
PreliminaryPerformance Results
! HYROAD produces more accuraterobust high concentration thanCAL3QHC
! Performance differences observedbetween receptor locations
! HYROAD appears to provideimproved treatment of problematicworst-case conditions
Presentation 4 20
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
PreliminaryPerformance Results
! Performance evaluation forother intersections is under way
! Screening methodology usingHYROAD will be developedand evaluated
Proposed Applications! Refined CO Applications
# SIPs
# Conformity
# EIS/EIR
! Particulate matter“hot-spots” assessment
! Air toxic risk assessment
Project Status! Current Status
# Select transition frompuff to line-segment
# Complete modelperformance evaluation
" Completed Tucson
" Virginia & Denver plus 8SLAMS/NAMS sites
Presentation 4 21
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Hybrid Roadway Intersection Model: HYROAD
Robert Ireson,Edward Carr
# Present Evaluation andRecommendations to NCHRP panel
Project Status
Next Steps! Develop Graphical User Interface
# Facilitate communicationbetween modules
# Ease burden for userin developing inputs
! Update Draft User Guide
# Reflect changes
! Schedule for Completion
# Complete Evaluation (August 2000)
# Develop GUI and Beta Testing(Fall 2000)
# Update User Guide(Winter 2000-2001)
Next Steps
Presentation 5 1
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
Edward Carr
The Variable Grid UrbanAirshed Model
(UAM-V, UAM-VPM)
Overview of the Variable-Grid Urban Airshed Model
(UAM-V)! Simulates the physical and
chemical processes governing theformation and transport of ozone inthe troposphere
" three-dimensional, Eulerian(grid-based) model
" requires specification ofmeteorological, emissions, land-use,and other geographic inputs
" output includes hourly concentrationsof ozone and precursor pollutants foreach grid cell within a(three-dimensional) modeling domain
Overview of the Variable-Grid Urban Airshed Model
(UAM-V)
Presentation 5 2
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
Overview of UAM-V! Core model, supporting software,
user’s manuals, and examplemodeling database available fromSAI at no charge
! www.uamv.saintl.com
! Version 1.24 – OTAG version
" Updated isoprene chemistry (1996)
! Version 1.30 – Latest version
" Toxics chemistry, process analysis
Overview of UAM-V
UAM-V ModelingSystem Features
! Carbon-Bond-IV chemicalmechanism with enhancedisoprene and toxics chemistry
! Two-way interactivenested-grid capabilities
! Plume-in-grid (P-i-G) treatment
Presentation 5 3
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
! Accepts output from a variety ofdynamic meteorological models(e.g. MM5)
! Contains “process analysis”capabilities
UAM-V ModelingSystem Features
Treatment ofProcesses in UAM-V
! Advective pollutant transport
" Smolarkiewicz scheme
! Turbulent diffusion
" Dispersion proportional toconcentration gradient – K Theory
! Surface removal
" Uptake of pollutants by varioussurface features - land use
! Chemistry
" Carbon Bond IV chemical mechanismwith updated isoprene chemistry andtoxics mechanism
Treatment ofProcesses in UAM-V
Presentation 5 4
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
UAM-V Modeling System! Core model and input processing
software
! Emissions Preprocessing System(EPS2.5) to prepare ozone andparticulate emission inventories
! UAM-V Postprocessing System(UPS)
! Process analysis modelingsystem software
! Model Output Visualization andInput Evaluation Software(MOVIES) – Color animations
UAM-V Modeling System
UAM-V InputFile Requirements
! Meteorological input files
" wind
" temperature
" water-vapor concentration
" pressure
Presentation 5 5
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
" vertical diffusivity(effective mixing height)
" cloud cover
" rainfall rate
UAM-V InputFile Requirements
UAM-V InputFile Requirements
! Emissions input files
" low-level anthropogenic emissions
# point sources
# area sources
# motor-vehicles
" elevated point source emissions
" biogenic emission estimates
UAM-V InputFile Requirements
! Air quality related input files
" initial conditions
" boundary conditions
! Chemistry input files
" chemical reaction rates
" photolysis rates
Presentation 5 6
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
! Geographic/other input files
" land-use
" albedo, turbidity, and ozone column
UAM-V InputFile Requirements
UAM-V ProcessAnalysis Capabilities
! UAM-V process analysis providesdetailed information on thephysical and chemical simulationprocesses
! Process-level information includes
" photochemicalproduction/consumption
" horizontal advection/diffusion
" vertical advection
" vertical diffusion
" deposition
" emissions (for precursor pollutants)
UAM-V ProcessAnalysis Capabilities
Presentation 5 7
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
Simulated Process Contributions to Ozone
-25-20-15-10
-505
101520
0 2 4 6 8 10 12 14 16 18 20 22
Chem Horiz. V. Adv. V Diff. Dep.
-25-20-15-10
-505
101520
0 2 4 6 8 10 12 14 16 18 20 22
Chem Horiz. V. Adv. V Diff. Dep.
Ozone
ppb
Hour
0
100
Simulated —and observedozone concentration
Simulated Process Contributions to Ozone
UAM-V Users/Applications! Over 60 registered users worldwide
(Version 1.30), unknown number ofunregistered users
" U.S. EPA, state/local agencies,and industry
" Environment Canada
Presentation 5 8
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
" European research andregulatory agencies
" European and Japanese auto-makers; European oil companies
UAM-V Users/Applications
UAM-V Users/Applications! Registered users
" Research and regulatory groups inCentral and South America, Australia,New Zealand, and several Asiancountries including
# China
# Taiwan
# South Korea
# Philippines
# India
# Thailand (AIT)
UAM-V Users/Applications! Some completed applications
" Numerous U.S. regions/cities
# OTAG, Atlanta, Houston, BatonRouge, Chicago
" Vancouver,B.C.
" U.K.
" Paris
" Milan
" Mexico City
" Athens
Presentation 5 9
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
Worldwide Users of theUAM-V Modeling System
(2)
(2)
(4)
(2)
(2)(3)
(3)(2)(2) (2)
(2) (2)
(3)
(2)
(5)
Current Applications! Gulf Coast Ozone Study (GCOS) –
Assessment of ozone formationand transport processes affecting1- and 8-hour ozone along the U.S.Gulf Coast
! Arkansas-Tennessee-MississippiOzone Study (ATMOS) –Assessment of potential 8-hourozone issues for Memphis,Nashville, Knoxville, Tupelo,Chattanooga, and Little Rock
! Mexico City - Demonstration for anew area/preliminary emissionssensitivity analysis
Current Applications
Presentation 5 10
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
Example of a Multiple Nested Domain: Gulf Coast Ozone Study
4-kmgrids(GridsA,B & C)
2-kmgrid(Grid D)
36-kmgrid(Grid 1)
12-kmgrid(Grid 2)
Overview of theUAM-VPM
! UAM-V photochemical model(CB-2000)
! Particulate matter (PM) stand-alonebox model employing hybridmodal-sectional approach to PMrepresentation
! Gas-phase chemical mechanismgenerator for the UAM-V
Structure of the UAM-VPM! Features of PM dynamics that can
be well characterized by knownalgorithms are hard coded
! Features which are not well knownare user inputs or dynamicallyselected
! Allows for the best research gradealgorithms to be used in aregulatory and planning platform
Presentation 5 11
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
PM Processes in UAM-VPM1) Modal discretization (new)
2) Nucleation(Fitzgerald, Hoppel, Gelbard, 1998)
3) Coagulation(Jacobson, 1994, 1999)
4) Condensation(Jacobson, 1997, 1999)
5) Dissolution (Jacobson, 1997, 1999)
6) Reversible chemistry (various)
7) Sectional remodalization (new)
PM Processes in UAM-VPM
Gas-to-particleTransformations
EmissionsInjected
(Gas ∧∧∧∧ PM)
HorizontalAdvection/Diffusion
(Gas ∧∧∧∧ PM)
VerticalAdvection/Diffusion
(Gas ∧∧∧∧ PM)
Deposition(Gas ∧∧∧∧ PM)
Chemistry(Gas Phase)
Gas Concentrations and PM Modes (low size resolution)UAM-VPM Flow Chart
DiscretizeModes
Nucleationadd mass to
smallest size bin
Solve homo-coagulationequations
Calculate diffusioncoefficients knudsennumbers, coagulation
kernels etc., for allcombinations of particles
Calculate diffusioncoefficients, knudsennumbers, coagulationkernels etc., for self-
similar particles
Solve hetero-coagulationequations
Determinedeliquescence points forsolids, and if below r.h.,eliminate reactions fromequilibrium mechanism
Equilibrate ions, solids, andgases to get ionic activitycoefficients (equilibrium
chemistry solver)
Determine watercontent w/Zdanovskii-
Stokes-Robinson (ZSR)relationship
DetermineeffectiveHenry’s
constants
Dissolution/Condensation(add or subtract mass)
Inert OrganicHenry’s constants, or
“effective” Henry’s constantssupplied by user
Solid ∧∧∧∧ AqueousInorganic
Calculate effectiveHenry’s constants
Solve Analytic Prediction ofDissolution (APD) for Each
Condensable Species
Solve Analytic Prediction ofCondensation (APC) for
Each Condensable Species
Renormalize SizeDistributionCoagulation
Homo-Coagulation
Hetero-Coagulation
Presentation 5 12
Technical Highlights of EPA’s 7th Conference on Air Quality ModelingThe Variable Grid Urban Airshed Model
Edward Carr
ExampleUAM-VPM Species
H2O(aq) water Na+ sodium ion
H2SO4(aq) sulfuric acid Cl- chloride ion
HNO3(aq) nitric acid Na2SO4(s) sodium sulfate
NH3(aq) ammonia NaHSO4(s) sodium bisulfate
HCl(aq) hydrochloric acid NaCl(s) sodium chloride
H+ hydrogen ion NaNO3(s) sodium nitrate
OH- hydroxy ion (NH4)2SO4(s) ammonium sulfate
NH4+ ammonium ion NH4HSO4(s) ammonium bisulfate
NO3- nitrate ion NH4Cl(s) ammonium chloride
HSO4- bisulfate NH4NO3(s) ammonium nitrate
SO42- sulfate
ChemicalFormula
Chemical Name
ChemicalFormula
Chemical Name
Status of UAM-VPM: Plansfor 2000 and beyond
! Rigorous testing of box model
! Testing of full modeling system
! Complete initial application toVancouver for a 10-day 1993episode
! Initiate application to Alberta
Presentation 6 1
Technical Highlights of EPA’s 7th Conference on Air Quality Modeling7th Conference on Air Quality Modeling: Prepared Comments on the Second Day - June 29
Joseph A. Tikvart
7th Conference on AirQuality Modeling:
Prepared Comments on theSecond Day - June 29
Joseph A. Tikvart
AMS Committee PerspectiveJ. Weil
AWMA Committee PerspectiveR. Paine
Prognostic Meteorological Model PanelR. Schulze
STAPPA/ALAPCO AgenciesP. Hanrahan
Department of EnergyP. Lunn
Gas Research InstituteD.Blewitt
American Petroleum InstituteH. FeldmanK. Steinberg
Utility Air Regulatory GroupA. FieldR. Paine
Southern CompanyS. Vasa
Trinity ConsultantsR. Schulze
Personal StatementM. Sharan
Presentation 6 2
Technical Highlights of EPA’s 7th Conference on Air Quality Modeling7th Conference on Air Quality Modeling: Prepared Comments on the Second Day - June 29
Joseph A. Tikvart
7th Conference on AirQuality Modeling
Questions EPA Asked on New Modeling Systems
Q1. Scientific Merit
Q2. Model Accuracy
Q3. Appropriate Regulatory Applications
7th Conference on AirQuality Modeling
Questions EPA Asked on New Modeling Systems
Q4. Implementation Issues
Q5. Resource Constraints
Q6. Additional Analyses
Notes on Summaryof Comments
! Public comment periodis open until 8/21/00
! Summary is for oral comments only
Presentation 6 3
Technical Highlights of EPA’s 7th Conference on Air Quality Modeling7th Conference on Air Quality Modeling: Prepared Comments on the Second Day - June 29
Joseph A. Tikvart
! No responses have beenformulated at this time
! Comments on the scientificmerit of AERMOD, CALPUFF,and ISC-PRIME were generallyfavorable
Notes on Summaryof Comments
Notes on Summaryof Comments
! Summary of comments is basedon the previous 6 questions
" AERMOD
" CALPUFF
" Numerical Grid Models
" Data from Meteorological Models
" General Comments
! There were no specific oralcomments on other models
Notes on Summaryof Comments
Presentation 6 4
Technical Highlights of EPA’s 7th Conference on Air Quality Modeling7th Conference on Air Quality Modeling: Prepared Comments on the Second Day - June 29
Joseph A. Tikvart
AERMODQ1, Q2. Scientific Merit
& Model Accuracy
" Improvements over ISC, especiallyregarding PBL dispersion andcomplex terrain, are desirable
Q3. Appropriate Regulatory Applications
" Add PRIME and depositionalgorithms to AERMOD
" Develop AERSCREEN
AERMOD
AERMODQ4, Q6. Implementation Issues
& Additional Analyses
" Define separate uses ofAERMOD and ISC-PRIME
" Expand period for transitionto AERMOD (>12 months)
Presentation 6 5
Technical Highlights of EPA’s 7th Conference on Air Quality Modeling7th Conference on Air Quality Modeling: Prepared Comments on the Second Day - June 29
Joseph A. Tikvart
Q5. Resource Constraints
" Application of PRIME -- separate fromAERMOD or included with AERMOD
" Use of electronic terrain data --need more training
" Additional EPA support
AERMOD
CALPUFFQ1. Scientific Merit
" Significant advancement as State-of-Practice for long range transport
" Model flexibility providesroom for growth
Q2. Model Accuracy
" Testing is adequate forinclusion in the Guideline
" More testing for short-rangeapplications is desirable
CALPUFF
Presentation 6 6
Technical Highlights of EPA’s 7th Conference on Air Quality Modeling7th Conference on Air Quality Modeling: Prepared Comments on the Second Day - June 29
Joseph A. Tikvart
CALPUFFQ3. Appropriate
Regulatory Applications
" Appropriate for longrange transport (50 - 200km)
" Need more specific guidelinelanguage for other applications
Q4. Implementation Issues
" Clarify user's guideand default options
CALPUFFQ5. Resource Constraints
" Computer requirements
" User skills
" Additional EPA support
CALPUFFQ6. Additional Analyses
" Requirement for 5 years ofmeteorological data
" Dispersion coefficient treatment
" Example protocol
" Better chemistry for SOx/NOx
" Overall experience of dispersionmodel community
Presentation 6 7
Technical Highlights of EPA’s 7th Conference on Air Quality Modeling7th Conference on Air Quality Modeling: Prepared Comments on the Second Day - June 29
Joseph A. Tikvart
Numerical Grid Models forUrban/Regional Scales
Q1. Scientific Merit
" Support removal of old grid modelfrom guideline
Q6. Additional Analyses
" Need more testing of Models-3/CMAQ
" Insure that models are in the publicdomain
Data fromMeteorological Models
Q1, Q2. Scientific Merit & Model Accuracy
" Use of data from meteorologicalmodels (e.g., RUC, MM5) is desirableand meets needs
Q3. Appropriate Regulatory Applications
" Applications need aresolution finer than 80 km
Data fromMeteorological Models
Presentation 6 8
Technical Highlights of EPA’s 7th Conference on Air Quality Modeling7th Conference on Air Quality Modeling: Prepared Comments on the Second Day - June 29
Joseph A. Tikvart
Data fromMeteorological Models
Q4. Implementation Issues
" Need detailed resolution ofterrain and meteorological data
" CALMET can be used inconjunction with MM5 data
" What are the consequencesof using these data bases inregulatory programs
Data fromMeteorological Models
Q5. Resource Issues
" Need more modelers using RUC &MM5 meteorological model productsand greater exposure to data bases
" Need a repository of routine andeasily accessible prognosticmeteorological data (NCEP,RUC)
" Need training for MM5 and similarmeteorological models
Data fromMeteorological Models
Presentation 6 9
Technical Highlights of EPA’s 7th Conference on Air Quality Modeling7th Conference on Air Quality Modeling: Prepared Comments on the Second Day - June 29
Joseph A. Tikvart
General Comments
Summary! Public comment period is open
until 8/21/00
! No responses have beenformulated at this time
! Comments on the scientific merit ofAERMOD, CALPUFF, and ISC-PRIME were generally favorable --implementation / resource issues
! There were no specificcomments on other models
! Next steps
Summary