Dispersion Modeling 101: ISCST3 vs. AERMOD

1
Dispersion Modeling 101ISCST3 vs. AERMOD

• Iowa Chapter AWMA
• February 14, 2006
• Mick Durham
• Stanley Consultants, Inc.

2
What we are going to talk about

• Brief History of Dispersion Modeling
• Industrial Source Complex Model
• AMS/EPA Regulatory Model (AERMOD)
• Comparisons
• The IDNR Connection
• Questions Answers

3
Brief History of Modeling

• Earliest Studies Simulated the Movement of Air
• G.I. Taylor, 1915, Eddy Motion in the Atmosphere
• O.G. Sutton, 1932, A Theory of Eddy Diffusion
• Dispersion of Pollutants (Mainly Particulate)
  Followed WW II
• E.W. Hewson, 1945, Meteorological Control of
  Atmospheric Pollutants by Heavy Industry
• E.W. Hewson, 1955, Stack Heights Required to
  Minimize Ground Level Concentrations
• Gale, Stewart Crooks, 1958, The Atmospheric
  Diffusion of Gases Discharged from a Chimney

4
Brief History of Modeling

• Birth of Dispersion Parameters
• F.A. Gifford, 1960, Atmospheric Dispersion
  Calculations Using the Gaussian Plume Model
• F. Pasquill, 1961, The Estimation of the
  Dispersion of Windborne Material
• D. Bruce Turner, 1967, Workbook on Atmospheric
  Dispersion Estimates
• Briggs, Gary, 1969, Plume Rise

5
Brief History of Modeling

• Modeling and the Computer Age
• PTMAX, PTMIN, PTMTP, 1972
• Air Quality Display Model (AQDM), 1974
• Single Source (CRSTER) Model, 1977
• Complex Terrain (VALLEY) Model, 1977
• Multiple Source (MPTER) Model, 1980
• Pollutant and Environment Specific Models
• APRAC, CALINE, HIWAY Carbon Monoxide Models
• BLP (Bouyant Line and Point Sources) PAL (Point
  Area and Line Source), 1979 TEM (Texas Episode
  for Urban Areas)
• RPM (Reactive Plume Model) for Ozone, 1980

6
Brief History of Modeling

• Guideline on Air Quality Models
• The Guidelines on Air Quality Models, 1978
• 40 CFR Part 58, Appendix W
• Refined and More Complex Models
• Industrial Source Complex (ISC), 1979
• Industrial Short-Term ST
• Industrial Long-Term LT
• Complex Terrain (COMPLEX)
• Dense Gas (DEGADIS)
• Urban Airshed Model (UAM)

7
Brief History of Modeling

• Refined and More Complex Models (cont.)
• Screening Model (SCREEN)
• California Line Source (CALINE) and Mobile Source
  Emission Factors (MOBILE)
• Puff Models (INPUFF)
• Visibility (VISCREEN)

8
Brief History of Modeling

• Advanced Models
• Industrial Source Complex Version 2 (ISC2), 1990
• Industrial Source Complex Version 3 (ISC3), 1995
• California Line Source (CAL3QH3)
• Urban Airshed Model (UAM-V)
• Complex Terrain Dispersion Model (CTDMPLUS)
• Offshore and Coastal Dispersion Model (OCD)
• Bouyant Line and Point Source (BLP)
• Area Locations of Hazardous Atmospheres (ALOHA)
• Dense Gas Dispersion Model (DEGADIS 2.1)

9
Brief History of Modeling

• Todays Models
• AERMOD
• Point, Area, Line Sources
• Simple or Complex Terrain
• Transport distance up to 50 km
• CALPUFF
• Transport from 50 to hundreds of kilometers
• Visibility, Regional Haze
• Dispersion in Complex Terrain
• Complex Dispersion Model Plus Algorithms for
  Unstable Conditions (CTDMPLUSS)
• Dispersion in Complex Terrain

10
Brief History of Modeling

• Todays Models (Continued)
• Caline3 or CAL3QHC, MOBILE6
• Highway Line sources
• Simple Terrain
• Carbon Monoxide
• Buoyant Line and Point Source (BLP)
• Aluminum Reduction plants with buoyant line and
  point sources
• Rural location
• Simple Terrain
• Community Multi-scale Air Quality Model (CMAQ)
• Ozone

11
Industrial Source Complex Model

• Introduced in 1979
• First adopted as Preferred Model in 1983
• Major Revisions 4 times in 27 year history
• Can remain acceptable as a preferred model until
  November 9, 2006

12
Industrial Source Complex Model

• Gaussian Plume Model
• Building Downwash
• Particulate Deposition
• Point, Area, and Line Sources
• Complex Terrain
• Simple Meteorological Data Input

13
Industrial Source Complex Model

• Has been primary model in Iowa for 27 years
• Over 100 facilities have modeled compliance with
  ISC
• Generally the short-term standards have caused
  greatest predicted non-compliance

14
Industrial Source Complex Model

• Problems with ISCST3
• Modeling of Plume Dispersion is Crude
• Only 6 possible states (Stability Classes)
• No variation in most meteorological variables
  with height
• No use of observed turbulence data
• No information about surface characteristics
• Erroneous depiction of dispersion in convective
  conditions
• Substantial overprediction in complex terrain
• Crude building downwash algorithm

15
AERMOD

• AERMOD stands for American Meteorological
  Society/ Environmental Protection Agency
  Regulatory Model
• Formally Proposed as replacement for ISC in 2000
• Adopted as Preferred Model November 9, 2005

16
AERMOD

• 3 COMPONENTS
• AERMET THE METEOROLOGICAL PREPROCESOR
• AERMAP THE TERRAIN DATA PREPROCESSOR
• AERMOD THE DISPERSION MODEL
• 2 SUPPORT TOOLS
• AERSURFACE PROCESSES SURFACE CHARACTERISTICS
  DATA
• AERSCREEN PROVIDES A SCREENING TOOL

17
AERMOD

• AERMOD IS SIMILAR TO ISC IN SETUP
• THE CONTROL FILE STRUCTURE IS THE SAME
• VIRTUALLY ALL THE CONTROL KEYWORDS AND OPTIONS
  ARE THE SAME

18
AERMOD

• AERMOD IS DIFFERENT FROM ISC
• REQUIRES SURFACE CHARACTERISTICS (ALBEDO, BOWEN
  RATIO, SURFACE ROUGHNESS) IN AERMET
• HAS PRIME FOR BUILDING DOWNWASH AND THE BUILDING
  PARAMETERS ARE MORE EXTENSIVE
• REQUIRES LONGER COMPUTER RUN TIMES (up to 5 times
  longer!)

19
Comparison of Dispersion Model FeaturesMeteorolo
gical Data Input

• ISCST3
• One level of data accepted
• AERMOD
• An arbitrarily large number of data levels can
  be accommodated

20
Comparison of Dispersion Model FeaturesPlume
Dispersion and Plume Growth Rates

• ISCST3
• Based upon six discrete stability classes only
• Dispersion curves are Pasquill-Gifford
• Choice of rural or urban surfaces only
• AERMOD
• Uses profiles of vertical and horizontal
  turbulence variable with height
• Uses continuous growth function
• Uses many variations of surface characteristics

21
Comparison of Dispersion Model FeaturesComplex
Terrain Modeling

• ISCST3
• Elevation of each receptor point input
• Predictions are very conservative in complex
  terrain
• AERMOD
• Controlling hill elevation and point elevation
  at each receptor are input
• Predictions are nearly unbiased in complex
  terrain

22
Comparisons ISC Vs AERMOD

• CONSEQUENCE ANALYSIS - ratios of AERMOD
  predicted high concentrations to ISCST3 predicted
  high concentrations
• flat and simple terrain
• point, volume and area sources.
• 1hour 3hour 24hour annual
• average 1.04 1.09 1.14 1.33
• high 4.25 2.82 3.15 3.89
• low 0.32 0.26 0.24 0.30
• Total 48 48 48 48
• AN OVERVIEW FOR THE 8TH MODELING CONFERENCE
  SEPTEMBER 22, 2005

23
Comparisons ISC Vs AERMOD

• CONSEQUENCE ANALYSIS - ratios of AERMOD
  predicted high concentrations to ISCST3 (and
  PRIME) predicted high concentrations
• flat terrain
• point sources with significant bldg downwash
• ANNUAL 24 H2H
  3 H2H
• AER/ISC3 AER/ISCP AER/ISC3
  AER/ISCP AER/ISC3 AER/ISCP
• ave 1.08 1.05 1.25
  1.01 0.71 1.05
• max 1.35 1.29 1.87
  1.14 1.20 1.17
• min 0.69 0.79 0.69
  0.84 0.38 0.93
• No cases 6 6
  6
• AN OVERVIEW FOR THE 8TH MODELING CONFERENCE
  SEPTEMBER 22, 2005

24
Comparisons ISC Vs AERMOD

• Duane Arnold Energy Center Data (Palo, IA)
• Ratio of Modeled Conc to Observed
• AERMOD 0.69 (1-hr avg 46m release)
• ISC-Prime 0.76 (1-hr avg 46m release)
• AERMOD 0.25 (1-hr avg 24m release)
• ISC-Prime 0.29 (1-hr avg 24m release)
• AERMOD 0.51 (1-hr avg 1m release)
• ISC-Prime 0.38 (1-hr avg 1m release)

25
Comparisons ISC Vs AERMOD

• Presentation at EUEC conference by Bob Paine,
  TRC
• AERMOD consistently showed better or comparable
  performance with ISCST3
• In flat terrain, AERMOD and ISCST3 predictions
  are comparable, but AERMOD has higher annual
  averages
• In complex terrain, AERMOD predictions are
  markedly lower
• Building downwash predictions will often be
  lower, especially for stacks located some
  distance from controlling buildings
• Overall, more confidence in accuracy of AERMOD
  results

26
Comparisons ISC Vs AERMOD

• Our Recent Experience
• Annual concentrations higher with AERMOD by
  10-15
• Short term concentrations similar without
  downwash
• Short-term concentrations generally lower with
  building downwash by 20

27
The IDNR Connection

• IDNR will allow use of either ISCST3 or AERMOD
  until November 9, 2006
• Meteorological Data will be provided by IDNR for
  eight stations
• Compliance with ISCST3 and non-compliance by
  AERMOD must be addressed

28
Questions Answers
29
(No Transcript)
30
AERMOD
31
AERMOD
32
AERMOD