Comparisons of CALPUFF and AERMOD for Vermont Applications

1
Comparisons of CALPUFF and AERMOD for Vermont
Applications

• Examining differing model performance for a 76
  meter and 12 meter (stub) stack with emission
  characteristics for wood combustion.
• With Building Downwash and Examining Near Field
  Impacts.

2
The Modeled Scenario

• For both models using a 2 km square domain
  centered over Burlington High School.
• Building Downwash occurs from the High School
  Building Structure.
• The Stub modeled at 1 meter above the building
  structure with emission characteristics of an
  outdoor wood boiler.
• The Tall Stack (76 Meters), with an 18
  meter/second exit Velocity, characteristics of
  wood combustion for electrical generation.
• Same Emission Rate for Both Stacks.

3
A Necessary Initial Step Comparison of AERMOD
Impacts using Current and Outdated AERSURFACE
Geophysical Processing for AERMET.

• Where version 1 was directionally independent,
  out to 3 km. radial extent, ASOS locations /-
  0.5 km. accuracy (Prior EPA Guidance).
• Version 2 (Current EPA Guidance) smaller radial
  extent (1 km.), requires better ASOS siting -
  /- 100 meters after contacting ASOS sites for
  assistance. Surface roughness calculations were
  performed on a directionally specific basis which
  will improve estimates of the turbulence
  parameters.  The turbulence estimated for a given
  wind direction is very sensitive to land use and
  terrain variations.
• Monthly Variation Specified to represent longer
  winters.

4
Normalized Comparison of Impacts for Old and New
AERSURFACE Processing
5

• Significantly Lower Impacts Predicted by the Tall
  Stack (76 Meters), for 1 km. Radial Extent of
  Landuse. (The airport area constitutes a larger
  percent of the land use, so the Friction Velocity
  averages lower, so Plume Does not impact the
  surface as much).

6
for the 76 meter and 12 meter (stub) stack with
emission characteristics for wood combustion

• Associating the maximum hourly impacts with
  meteorological measures paired in time and
  space.
• Examining CALPUFF model performance for the
  subset of CALM and variable wind direction
  (WDIR), hours that AERMOD discards.

7
Meteorological Field Production

• CALPUFF - Albany, NY Upper Air Combined with
  Burlington, VT Sfc. Meteorology.
• AERMOD Burlington, VT Sfc. Meteorology.

8
Dispersion Model Options

• In this Comparison it was assumed that AERMOD was
  the better tool, and efforts were made to achieve
  similarity in predicted impacts between the 2
  models by altering CALPUFF options.
• Comparisons were drawn for different downwash
  algorithms, dispersion estimate methods, many
  other aspects of CALPUFF Simulation.
• A Failing The CTDM-Like dispersion calculation
  option for stable and neutral conditions was not
  possible to evaluate because on-site Met data was
  not available.

9

• Method used to compute dispersion
• coefficients (MDISP)
  Default 3 ! MDISP 3 !
• 1 dispersion coefficients computed from
  measured values
• of turbulence, sigma v, sigma w
• 2 dispersion coefficients from
  internally calculated
• sigma v, sigma w using
  micrometeorological variables
• (u, w, L, etc.)
• 3 PG dispersion coefficients for RURAL
  areas (computed using
• the ISCST multi-segment
  approximation) and MP coefficients in
• urban areas
• 4 same as 3 except PG coefficients
  computed using
• the MESOPUFF II eqns.
• 5 CTDM sigmas used for stable and
  neutral conditions.
• For unstable conditions, sigmas are
  computed as in
• MDISP 3, described above. MDISP
  5 assumes that
• measured values are read

10
CALPUFF Options

• Terrain Adjustment Method.
• Transitional plume rise.
• Downwash - ISC method, PRIME method.
• Partial plume penetration of elevated inversion.
• Dispersion Coefficients.
• For CALPUFF Maximum hourly impacts, these
  variables did not alter predictions more than
  20.

11
(No Transcript)
12
Averaged Wind and Mixing Height values for
CALPUFF Maximum Impact Hours
13
Averaged Wind and Mixing Height values for AERMOD
Maximum Impact Hours
14

• AERMOD Maximum Impacts for Stub Stack Mixing
  Heights Appear too high.
• CALPUFF is simulating maximum impacts from the
  Tall (76 meter), stack under calm conditions
  throughout lowest 200 meters of the atmosphere
  with significant stagnation / puff buildup.

15
Associating Maximum Hourly Predictions from
CALPUFF and AERMOD with Meteorological Measures
(August 4, 1998).
16
Examining Meteorology for August 4, 1998, 7 a.m.
(Hour of Maximum impacts for Tall Stack with
CALPUFF).
17
(No Transcript)
18
UTC versus Local (EST) Time?

• From Raw ASOS DATA - EST - Matches with CALPUFF-
• 08,04,1998,06, 61, 59, 0, 0.0, ,-999,
  29.85,0.5,1000, 0.00,
• 08,04,1998,07,-999,-999,-999,-999,
  ,-999,-999,-999,-999,-999,
• 08,04,1998,08, 68, 61, 0, 0.0, ,-999,
  29.86,0.0,1000, 0.00,
• From AERMET .pfl File
• 98 8 4 6 6.1 1 62. 1.50 16.7 99.0
  99.00
• 98 8 4 7 6.1 1 57. 1.50 16.1 99.0
  99.00
• 98 8 4 8 6.1 1 83. 1.50 15.6 99.0
  99.00
• 98 8 4 11 6.1 1 0. 0.00 16.1 99.0
  99.00
• 98 8 4 12 6.1 1 68. 1.50 17.2 99.0
  99.00
• 98 8 4 13 6.1 1 0. 0.00 20.0 99.0
  99.00
• AERMET is Specified in UTC?

19
Examining CALM hours

• For AERMOD, with the stub stack 3 hours were
  discarded out of the top 5 hours of maximum
  predicted impacts, 18 hours for top 50 (the CALM
  Hours).
• For the tall stack no hours were discarded.

20
Examination of Calpuff Predictions for set of
CALM and variable wind direction Hours in ASOS
Data excluded by AERMOD
21
Examination of Calpuff Predictions for set of
CALM and variable wind direction Hours in ASOS
Data excluded by AERMOD for Maximum 50 Hourly
Impacts.
22
Future Work

• Examine CALPUFF for CALM,Variable Wind Direction
  Hours for VT ASOS Sites in sheltered locations
  (Springfield, Rutland). For these locations calm
  hours are greater than 4000 hours per year.
• Current Intercomparison may not be paired in
  time. This will be reexamined.
• Continue to compare CALPUFF and AERMOD handling
  of specific hours of meteorology after rerunning
  CALPUFF with CTDM-like Dispersion.