Transboundary ManMade Impacts at Class I Areas

1
Transboundary Man-Made Impacts at Class I Areas

• Jim Boylan
• Georgia Department of Natural Resources
• VISTAS STAD Meeting
• August 17, 2005

2
Background

• Presentation by John Jansen will demonstrate the
  impacts of various assumptions on meeting the
  reasonable progress glide slope
• EPA defaults
• Alternative IMPROVE equation
• Sea salt, OC multiplier, dust
• Transboundary man-made impacts
• All these assumptions can potentially impact
  conclusions about reasonable progress
• Largest adjustments to the glide slope are due to
  transboundary man-made impacts
• Ammonium sulfate from GEOS-CHEM simulations

3
GEOS-CHEM Simulations

• VISTAS runs on a 4 x 5 grid for 2002
• EPRI runs on a 1 x 1 grid for 2001
• Harvard performed three GEOS-CHEM runs
• Run 1 Basecase with all emissions (MPE BCs)
• Run 2 Sensitivity with no U.S. anthropogenic
  emissions
• Run 3 Sensitivity with no global anthropogenic
  emissions (natural conditions)
• Transboundary Man-Made Impacts at Class I Areas
  calculated by (Conc. from Run 2) (Conc.
  from Run 3)

4
EPRI (1 x 1)
5
VISTAS (4 x 5)
Sulfate is 2-3 times smaller than EPRI (1 x
1)
6
EPRI (1 x 1)
7
VISTAS (4 x 5)
OC is 4 times smaller than EPRI (1 x 1) OC
at EVER is 10 times smaller than EPRI (1 x 1)
8
Major Differences

• Grid resolution and grid structure
• VISTAS 4 x 5 vs. EPRI 1 x 1
• Meteorology
• VISTAS 2002 vs. EPRI 2001
• Biomass burning emissions inventory
• VISTAS 2002 vs. EPRI 2001

9
GEOS-CHEM Grids
10
4 x 5 Mexican Emissions
11
1 x 1 Mexican Emissions
12
4 x 5 Canadian Emissions
13
1 x 1 Canadian Emissions
14
Summary

• Grid resolution and definition likely responsible
  for much of the discrepancies
• Also, model year and biomass burning emissions
  inventory
• VISTAS 4 x 5 sensitivities likely too low
• VISTAS 4 x 5 grid definition includes Canadian
  emissions as U.S. emissions
• EPRI 1 x 1 sensitivities likely too high
• EPRI 1 x 1 grid definition includes U.S.
  emissions as Canadian emissions
• Canadian emissions 30 too high in both model runs

15
Additional Modeling

• Harvard is re-running the GEOS-CHEM simulations
  for EPRI with a new 1 x 1 grid and revised
  Canadian emissions
• Canadian emissions reduced by 30
• New 1 x 1 grid definition better distinguishes
  U.S. and Canadian emissions
• VISTAS will be performing an annual CMAQ
  simulation on their 36 km grid with clean
  GEOS-CHEM BCs and Mexican and Canadian
  anthropogenic emissions removed.

16
Updated 1 x 1 GEOS-CHEM Grid
17
VISTAS 36 km Grid
18
State Options

• If a state is going to consider Transboundary
  Man-Made Impacts on the Reasonable Progress
  Glide Slope, there are 2 options
• Use VISTAS 36 km CMAQ simulation with clean
  GEOS-CHEM BCs and Mexican and Canadian
  anthropogenic emissions removed.
• Use GEOS-CHEM re-runs currently being performed
  by Harvard for EPRI with new 1x1 grid and
  revised Canadian emissions.
• Pick one? Average? Maximum impact? Minimum
  impact?