Impact of Light Winds on AERMOD A Case Study

1
Impact of Light Winds on AERMODA Case Study

• Joe Sims
• AL Department of Environmental Management
• Region 4 Modelers WorkshopMarch 18, 2009

2
The Issue

• Introduction of 1-minute averaged ASOS winds will
  generally result in lighter wind speeds than
  currently available.
• Recent (since mid-90s) archived met data at NCDC
  comes from METAR observations.
• Observed winds less than 3 knots reported as calm
  in the METAR code.
• Winds 2-6 knots with variable direction reported
  as variable (and ignored by AERMOD).
• Even though AERMOD can process winds as low as
  0.28m/s, it ignores calms.

How will use of light winds impact AERMOD
predicted concentrations?
3
The Study

• Find a met data set with known higher incidence
  of light winds
• For the same location, obtain a standard ASOS
  data set.
• Obtain/create a source set with a wide variety of
  stack heights, buoyancy characteristics, emission
  types, etc.
• Run AERMOD and compare results.

4
Met Data

• Met Data (Test) (Hybrid)
• Birmingham Airport (BHM) 1-minute ASOS data used
  to create hourly average winds for 2002.
• This data set created by OAQPS for use in the
  Birmingham Annual PM2.5 SIP.
• Met data treated in AERMET as On-site data and
  augmented as necessary by regular ASOS reports
  from BHM.
• Upper air data from the Shelby County Airport
  (EET) used.
• Surface characteristics developed from
  AERSURFACE.
• Met Data (Control) (Std ASOS)
• Standard BHM ASOS data.

5
Wind Speed Frequency Distribution
6
Receptor Grid, Etc.

• Receptors
• Polar grid, centered at BHM.
• 36 radials.
• Rings 25m apart out 200m, 50m apart from 200m to
  1000m and 100m apart from 1000m to 5000m.
• RURAL/URBAN Ran both conditions.
• Terrain - Flat.

7
Sources

• Set of hypothetical sources used by the AIWG
  Surface Characteristic Subgroup.
• 12 non-buoyant point sources
• 10 buoyant point sources
• 5 volume sources
• 4 area sources

8
Point Sources
9
Volume Sources
10
Area Sources
11
Results - Point Sources Ratio of H1H (RURAL)
12
Results Volume/Area Sources - Ratio of H1H
(RURAL)
13
Results - Point Sources - Ratio of H1H (URBAN)
14
Results Volume/Area Sources - Ratio of H1H
(URBAN)
15
Comments

• 25 of the winds were in the 0-3kt category,
  which were not available for use in the Std
  ASOS AERMOD runs. In other words, this
  experiment gives a reasonable test of AERMOD
  performance using light winds.
• Significant differences in the results between
  rural and urban. This is seen especially for
  taller stacks, where concentrations are generally
  greater than 4 times higher (and as much as 13
  times) for the hybrid wind set. The opposite is
  seen with the lowest release height area sources.
• Point source B2-35 was not included in the charts
  for the short term averages because it was an
  outlier. B2-35 is a 35m tall stack with some
  buoyancy (293K). Other stacks near the same
  height with more buoyancy behaved better.
  Apparently B2-35 did not have enough buoyancy to
  penetrate the inversion.

16
Comments (Contd)

• In almost all cases, the predicted concentrations
  using lighter winds were higher than when using
  standard ASOS as expected.
• I was especially interested in point sources in a
  rural environment. This combination constitutes
  the vast majority of PSD applications we see in
  Alabama. Except for the 3hr averages for
  low-level releases (and the troublesome B2-35),
  the ratios are pretty much in the 1 to 2 range.
  Applicants wont like this but it could be a lot
  worse.
• Using building downwash could significantly alter
  these results. James Thurmans results
  (presented at the Denver workshop) demonstrated
  this.
• Using terrain could also alter the results.
• This was only a one year sample. James runs
  showed a significant year-to-year variability.

17
Conclusions

• Using the hourly averaged 1-minute ASOS data to
  better represent dispersion potential from a
  source makes a lot of sense.
• It generally seems to give more conservative
  results than standard ASOS, therefore is more
  protective of human health.
• We as regulators must be prepared for challenges
  (and complaints) from the regulated community.