Transcript RegAPstudy copy

Regional Air Pollution Study
Alissa Dickerson, M.S.
Environmental Specialist
Enviroscientists, Inc.
Goal of Study
Western Regional Air Partnership (WRAP)
http://wrapair.org
Causes of Haze Assessment (COHA)
Goal: provide assessment of Class I areas
through integrated approach
www.coha.dri.edu
2
Overview
Introduction
Methodology
Analysis
Results & Discussion: Case Studies
Summary
3
What is Spatial Representativeness?
Area within which pollutant
concentrations are approximately
constant
Quantitative and qualitative approach to
investigate equivalency of
measurements
4
Why is it important?
Data assessments can determine
dependence and elicit solutions
Comprehensive picture of a complex
system
Tool to assess degree to which
measured concentrations can be
derived from reference points
Optimal network design
5
Why is it Important? (cont.)
Evaluation tool to help more efficiently
in mediation of environmental problems
Understanding regional visibility &
reduction
6
Introduction
Visibility reduction 1977 CAA
USEPA Regional Haze Rule, Final
(40 CFR 51, 1999)
Interagency Monitoring of Protected Visual
Environments = IMPROVE (1985)
5 regional organizations
7
The IMPROVE Network: Objectives
Federally mandated Class I areas
National parks, monuments, wilderness
areas
Identify current conditions of visibility
Determine aerosol species and sources
Document trends
Cultivate representative monitoring
network
8
The IMPROVE Network
 163 sites
 1-in-3 day
sampling
 4 cyclone-based
modules
 Coarse mass &
speciated fine
aerosols
9
The Improve Network
bext
visibility
Light Extinction Formula
bext= 3*f(RH)[Sulfate] + 3*f(RH)[Nitrate] +
4*[Organic Carbon] + 10*[Elemental Carbon] +
1*[ Fine Soil] + 0.6*[Coarse Mass]+ 10
Concentrations [ ] Units=μg/m3
Units= Mm-1, proportional to amount of light lost over
distance of 1 million meters
Rayleigh Scattering= 10 Mm-1, proportional 0.0
deciviews or 400 km
10
Research Objectives
Cascade Range
Southern Pacific Rainforests
Superior Upland
North Central Lowland Plains
Northern Great Plains
Columbia Plateau
Central Rocky Mountains
Great Plains
Klamath Mountains
Great Basin
Southern Rocky Mountains
Sierra Nevada Range
California Central Valleys
California Coast Ranges Colorado Plateau
Ouachita and Ozark Plateau
Central Lowland Plains
Southern California Ranges
Southwest DesertsMexican Highlands
Southern Great Plains
West Gulf Coastal and Mississippi Alluvial Plains
Determine spatial
representativeness of
IMPROVE monitorsWRAP
WA, OR, CA, NV, ID,
ND, SD, CO, AZ, NM,
TX
14 Physiographic
Regions
11
Considerations
What are most dominant chemical
species during 20% worst visibility days
within a region?
What are practical statistical and spatial
analysis methods?
How do concentrations vary by season?
12
Considerations
How can expected average concentrations
be determined for a region?
What is a method to test validity?
13
Methodology
Data
1997-2002, 54 monitors w/most complete data
Six aerosol species
Sulfates, nitrates, organic carbon (OC),
elemental carbon (EC), fine soil, coarse mass
(CM)
Focus: Upper 20% of calculated visibility
impairment values or 20% worst visibility days
14
Assumptions
All elemental sulfur is from sulfate ->
ammonium sulfate
All nitrate -> ammonium nitrate
Total organic carbon= C released in four
steps (OC1-OC4) + pyrolized organics (OP)
Thermal Optical Reflectance (TOR) analysis of quartz
filter
15
Assumptions
Elemental carbon (light absorbing carbon) =
EC fractions (EC1-EC3) – pyrolized organics
(OP)
TOR analysis of quartz filter
Fine soil = sum of Al, Si, K, Ca, Ti
particle-induced X-ray emission (PIXE)
& Fe
X-ray fluorescence (XRF)
Coarse mass = total mass - fine mass
16
Analysis Procedures
1) Characterize dynamics of regions
Climate & meteorology: wind patterns &
back-trajectory analysis (transport)
Graphically displays % of time an air mass
spent in an area
Color coded (shading increases w/ residence)
Topography: elevation & intervening
terrain
Emission sources and population centers
17
Analysis Procedures (cont.)
2) Regional spatial correlation analysis:
correlation expected to decrease w/distance
Correlation matrix of aerosol measurements
Distance matrix (km)
 Consideration
Correlation of site vs. itself = unity
[Artificial]= uncertainty * random #+measurement
 [Artificial] plotted at distance of 0
18
Analysis (cont.)
3) Criteria correlation cut-off = 0.7
Rationalize association between monitoring
sites
Validation of spatial representativeness
4) Seasons
Warm months: April to September
Cold months: October to March
19
Analysis (cont.)
5) Expected average concentrations
density (like temp.) of atmosphere varies w/
altitude
[Estimated] = [aerosol]* site density
density @ sea level
Put conc. into elevation ranges based on natural
breaks, then averaged= regional estimated conc.
Uncertainty= standard deviation of average
concentrations within elevation range (applicable
only with 2 or more sites)
20
Analysis (cont.)
6) Test of representativeness
 Analyzed sites within each region
Calculated seasonal average
concentrations
Uncertainty= average measurement
uncertainty
Compared to estimated concentrations
21
3.The Northern Great Plains Region
Characteristics
(E) Montana, (NE) Wyoming, & (W) portions of
North and South Dakota
Terrain: mostly prairie & rolling hills, mix of forest
and grassland
Badlands composed of steep buttes and pinnacles
Sparse population centers
Several coal-fired power plants, west-central ND
22
The N. Great Plains
6-IMPROVE sites
Site Name
Abbreviation
Elevation (m)
Badlands National Park
BADL1
736
Lostwood Wilderness Area
LOST1
692
Medicine Lake Wilderness Area
MELA1
605
Theodore Roosevelt Nat'l Park
THRO1
853
UL Bend Wilderness Area
ULBE1
893
Wind Cave National Park
WICA1
1300
23
Residence Time Analysis WICA1
 Warm months
 Prevailing winds
SE
 Bring in dry air
from SW U.S.
 Moist warm air
masses from Gulf
of Mexico
 Few inversions
24
Residence Time Analysis MELA1
Cold months
Cold continental
air flowing from
N/NW from
Canada
L system
typical, flushes
atmosphere
25
Aerosol Summary
Average Concentration (ug/m3)
Average Aerosol Concentration During the
20% Worst Visibility Days
18
16
14
CM
12
Soil
10
8
LAC
OMC
6
Nitrate
4
Sulfate
2
0
BADL1
LOST1
MELA1
THRO1
ULBE1
WICA1
26
Aerosol Summary (cont.)
Contribution to Bext (1/Mm)
Average Contribution to Light Extinction During the 20% Worst
Visibility Days
70.0
60.0
CM
50.0
Soil
40.0
LAC
30.0
OMC
Nitrate
20.0
Sulfate
10.0
0.0
BADL1
LOST1
MELA1
THRO1
ULBE1
WICA1
27
Estimated Concentration (µg/m3)
Elevation
500-1000m
UNC
1000-1500m
UNC
WARM Months
0.31
0.02
0.26
0.03
COLD Months
0.30
0.03
0.22
0.00
WARM Months
0.21
0.06
0.17
0.05
COLD Months
0.71
0.27
0.34
0.10
WARM Months
1.10
0.15
1.13
0.02
COLD Months
0.52
0.07
0.44
0.01
WARM Months
0.16
0.01
0.16
0.01
COLD Months
0.13
0.01
0.11
0.01
WARM Months
0.78
0.10
0.67
0.03
COLD Months
0.37
0.03
0.29
0.06
WARM Months
7.27
0.61
4.67
0.12
COLD Months
3.27
0.19
2.08
0.12
SO4
NO3
OC
EC
Soil
CM
28
Test Sites
FOPE1 (2yr) NOCH1 (2 yr)
Expected
Elevation
UNC
500-1000m
FOPE1
UNC
638m
Expected
UNC
1000-1500m
NOCH1
UNC
1332m
SO4
WARM Months
0.31
0.02
0.32
0.02
0.26
0.03
0.28
0.01
COLD Months
0.30
0.03
0.29
0.01
0.22
0.00
0.17
0.01
WARM Months
0.21
0.06
0.21
0.03
0.17
0.05
0.20
0.02
COLD Months
0.71
0.27
0.90
0.04
0.34
0.10
0.21
0.01
WARM Months
1.10
0.15
1.08
0.29
1.13
0.02
1.51
0.34
COLD Months
0.52
0.07
0.54
0.17
0.44
0.01
0.33
0.14
WARM Months
0.16
0.01
0.14
0.01
0.16
0.01
0.19
0.02
COLD Months
0.13
0.01
0.10
0.01
0.11
0.01
0.07
0.01
WARM Months
0.78
0.10
0.26
0.02
0.67
0.03
0.28
0.02
COLD Months
0.37
0.03
0.11
0.01
0.29
0.06
0.14
0.01
WARM Months
7.27
0.61
6.62
0.20
4.67
0.12
4.79
0.15
COLD Months
3.27
0.19
2.31
0.09
2.08
0.12
1.72
0.08
NO3
OC
FOPE1
30m elev.
difference
MELA1
[NO3]=0.9
µg/m3
EC
Soil
CM
29
Northern Great Plains
Regional Conclusions
Relatively flat terrain with good
dispersion of air
Atypical stagnation alleviates regional
haze problems during most days
SO4
representative ~ 180km
Colder months show good agreement out
to 700 km
30
Northern Great Plains
Regional Conclusions (cont.)
NO3
Rep. Distance ~ 450 km, 200km warm months
Factor – chemical nature to volatilize quickly in
warmer temperatures or not form at all
OC
Southerly located IMPROVE samplers recorded
higher OC concentrations on worst visibility days
Forest fire episodes
Rep. distance (Southern region) ~250 km
31
Thank you
Questions?
32