September 28, 2006
Page 1
ACCOUNTABILITY CASE STUDY: DETROIT
INTRODUCTION
The goal of this project was to identify and quantify changes in ambient air quality that
are a result of the implementation of known regulations. The Detroit area was chosen for a case
study that focuses on acid rain program regulations (1995 and 2000) and the 2003/2004 nitrogen
oxides (NOx) State Implementation Plan (SIP) Call. Two approaches were used:
(1) identification of when control measures were implemented and which pollutants were
targeted and examination of ambient data during the two periods of regulation; and
(2) examination of ambient data for trends to determine whether any changes in concentrations
occurred at the same time that known and documented regulations were in place. Both these
approaches require long data records and significant changes in ambient concentration to identify
trends above the "noise" of the data, that is, year-to-year variability due to meteorology,
fluctuations in emissions, etc. Long data records are particularly important for regional
pollutants, such as NOx, sulfate, and particulate matter (PM) mass, which typically are spatially
homogeneous. Urban-rural site pairs are also useful to segregate local and regional impacts. A
rural site is expected to have few local sources nearby and be representative of regional impact.
An urban site is expected to have many local sources and be representative of local impact.
In this case study, acid rain program regulations were expected to impact sulfur dioxide
(S02) concentrations, with the potential for impact on particulate sulfate, acid deposition, and
visibility degradation due to sulfate aerosol. A long SO2 data record was available for Detroit
and the nearby area, and emissions were well-quantified. NOx SIP call regulations were
expected to impact NOx and ozone concentrations in the summer months. However, NOx data
are only available beginning in 2002; this data set is likely not sufficiently long to see any impact
in ambient NOx concentrations resulting from the NOx SIP Call in 2003-2004.
SULFUR - ACID RAIN PROGRAM REGULATIONS
SO2 is both a local and regional pollutant, so intra-urban differences are expected. If
local sources are close to monitors, they may obscure long-term regional trends. Continuous SO2
data for 1993-2005 are available from the U.S. Environmental Protection Agency's (EPA) Air
Quality System (AQS) for five sites in the Detroit area. National SO2 emissions trends estimates
are available for 1993-2002,1 and electric generating facility emissions are available from 1995-
20052. In addition to SO2, sulfate aerosol, visibility extinction from sulfate aerosol, and acid
deposition should be impacted by the acid rain program. To understand the multipollutant effect
of SO2 regulations, ambient sulfate aerosol concentration, sulfur deposition, and light extinction
due to sulfate aerosol data for 1993-2005 were obtained from the Ann Arbor, Michigan, Clean
Air Status and Trends Network (CASTNET) site.
1 National Emission Inventory; .
2 Clean Air Markets; .
-------
September 28, 2006
Page 2
Between 1993 and 2002, national S02 emissions reductions were gradual, with a large
decrease in 1995 according to the NEI.1 Emissions from electrical generating facilities in
Michigan and regionally showed a large decrease in concentrations from 1998-2001.2 Specific
dates and locations of local SO2 controls in the Detroit area are not known; regional controls may
also impact concentrations in the Detroit area.
Overall, all sites showed a decrease in ambient SO2 concentrations from 1993 to 2005;
Figures 1 and 2 illustrate the results. Three-year averages were used for most of this analysis to
limit year-to-year variability. A large decrease (about 30%) in year-to-year concentrations of
S02 is evident between 1994 and 1995, corresponding to the largest decrease in year-to-year
emissions nationally (28%). Changes noted include a
14% decrease in Michigan S02 emissions from electric power generation (1995-1997 to
2003-2005);
26% region-wide decrease in S02 emissions from electric power generation (1995-1997 to
2003-2005);
26% decrease in Detroit average S02 (1993-1995 to 2003-2005);
24% decrease in sulfate concentrations in Ann Arbor (1991-1993 to 2003-2005);
7% decrease in sulfate concentrations in Allen Park (2001-2003 to 2003-2005);
26% decrease in total sulfur deposition in Ann Arbor (1991-1993 to 2003-2005); and
17% decrease in light extinction due to sulfate (1991-1993 to 2003-2005).
Figure 1. Annual average S02 concentrations at Detroit area sites, 1993-2005,
and national S02 emissions trends, 1993-2002. Average S02 trend across sites is
shown as a black line.
-------
September 28, 2006
Page 3
Legend notes: (1) Excluding 1996-1997 (incomplete data); (2) Ann Arbor CASTNET data
(representative of other Michigan sites); (3) Speciation Trends Network (STN) network data;
(4) utility emissions from http://www.epa.gov/airmarkets/emissions/prelimarp/index.html.
Axis note: Light extinction calculated from b=(3)ft(RH)[S042"], where RH is relative humidity
Figure 2. Annual total SO2 emissions in Michigan and regional area; three-year
averaged concentrations of sulfur species in Michigan (end year shown on graph).
PM2.5 mass concentrations may have been impacted by a decrease in the sulfate aerosol
component; however, PM2.5 data were available only from 1999-2005, and sulfate data from
2002-2005 in the Detroit area. Because decreases in sulfate aerosol after 2001 were small,
meteorology and transport confound any trends that may be due to changes in SO2 emissions.
Figure 3 shows that no PM2.5 mass trends were evident from 1999 to 2005 (years for which data
were available).
-------
September 28, 2006
Page 4
Figure 3. Annual average PM2.5 concentrations for the Detroit area, 1999-2005.
NOx SIP CALL
Like SO2, NOx is both a local and regional pollutant. Intra-urban differences are likely,
and mobile sources are expected to be the largest source of NOx in an urban area. NOx from
power generation (the target of the NOx SIP call) is about 40% of total NOx emissions in the
Detroit area, so changes in NOx emissions from other sources (such as mobile sources) could
confound the results. In addition, NOx data are only available from two sites in the Detroit area
for 2002-2005, and regulations in Michigan to reduce NOx were not implemented until 2004.
Decreases in NOx concentrations because of these regulations are probably not large enough to
be noticeable with such a short data record. Ozone concentrations are expected to decrease
corresponding to a decrease in NOx concentrations, but nitrate and PM2.5 mass are not expected
to change as a result of regulation because the regulation is only in effect during summer months.
In summer months, nitrate formation is minimal; thus, nitrate contribution to PM2.5 mass is very
small.
When summer-only yearly box whisker plots were examined for the two Detroit NOx
measurement sites—East 7 Mile and Linwood, no change in concentrations after 2004 was seen.
East 7 Mile measurements showed a decrease in NOx concentrations in 2004, followed by an
increase in 2005, but it is not clear that controls were in place prior to the summer of 2004. This
change in ambient NOx concentrations was not observed at the Linwood site, even though it is
closer to NOx point sources. Differences between the sites could be due to stack height and
mixing as distance from point sources increases. Concentrations were segregated by hour to
examine rush-hour (i.e., mobile source-dominated) versus non-rush hour trends, nighttime hours
(lowest mobile source contribution), and daytime hours, but no consistent trend was evident
-------
September 28, 2006
Page 5
across sites. Since mobile source activity is lower on weekends but electricity generation
activity generally is not, ambient NOx concentrations were also segregated by day of week and
hour to determine if examining periods when mobile source emissions are low could reveal
trends from electrical generation sources. No consistent trend was observed at sites from which
data were available (see example, Figure 4).
2002
_L_
_L_
~ Weekday
~ Weekend
2003 2004
Summers Only
2005
Figure 4. NOx concentrations at the Linwood site, segregated by day of week.
Because the large mobile source contribution to NOx may confound any changes in
concentration due to the NOx SIP Call, wind direction analysis was also performed to isolate the
point sources of NOx. Point sources were expected to dominate concentrations when winds were
180-225 degrees from the monitors. The remaining data were divided into two sectors: (1) winds
from the Detroit area—mobile-dominated and (2) winds from Canada—no emissions
information available. Concentrations were significantly higher at East 7 Mile and Linwood
when the wind was from 180-225 degrees, supporting the hypothesis that large point sources in
this direction impact ambient concentrations. However, no significant year-to-year change in
concentrations was evident at either site (see example, Figure 5). Data were divided by hours to
further isolate the point source-versus-mobile source contribution, but no consistent change
across years was seen with morning hour data only or nighttime hour data only.
3 The box shows the 25th, 50th (median), and 75th percentiles. The whisker shows the highest or lowest data point
with a maximum length of 1.5 times the interquartile range, IQR. Data outside this range are shown as "outliers"
identified with asterisks representing the points that fall within three times the IQR and circles representing points
beyond this. The boxes are notched (narrowed) at the median and return to full width at the 95% lower and upper
confidence interval values.
-------
September 28, 2006
Page 6
75
_ 50
.Q
Q.
Q.
X
O
Z 25
0
2002 2003 2004 2005
Figure 5. NOx concentrations at East 7 Mile when wind is from the southwest
(direction of major point sources).
Ratio analysis was also conducted using ratios of NOx with mobile source-dominated
pollutants. If the mobile source species (benzene and TNMOC) do not change with time, a
change in their ratios to NOx could indicate a change in the point source contribution. However,
no consistent year-to-year change was seen in these ratios at the sites (see example, Figure 1-6).
0.5
0.4
£ 0.3
CL
DO
£L 0.2
0.1
0.0
2002 2003 2004 2005
Figure 6. Benzene:NOx ratio, East 7 Mile site, 2002-2005.
-------
September 28, 2006
Page 7
Continuous Emissions Monitoring Systems (CEMS) data for the Detroit area are only
available from 2003-2005 (Figure 7). The available data show similar NOx emissions from
2003-2004 followed by a large increase in emissions from 2004-2005. However, not enough
information is currently available to determine a trend in emissions.
Figure 7. NOx CEMS data for the Detroit area, 2003-2005. Only June-September
data were included.
The NOx SIP call is intended to decrease ozone concentrations. Because ozone is a
photochemical pollutant, and because we want to understand the change in ozone resulting from
changes in ozone precursor emissions rather than inter-annual variability in meteorology, ozone
concentrations need to be adjusted for meteorology before trends can be examined.
Meteorologically adjusted ozone concentrations are available for the Detroit area from 1997
through 2005 (a much longer time frame than that for available NOx data). Figure 8 shows that
ozone concentrations decreased slightly over the 1997-2005 period.
-------
September 28, 2006
Page 8
ao
J3
CL
B60
o
c
o
8
40 -
20
Adjusted
Observed
I
2002
1993
2000
2004
Year
Figure 8. Meteorologically-adjusted ozone, Detroit area, 1997-2005 (source: Bill Cox, EPA).
CONCLUSIONS
In response to acid rain program regulations, a large decrease in SO2 emissions coupled
with a long ambient SO2 data record was critical in observing a decrease in ambient
concentrations of SO2. A 28% decrease in emissions in the Michigan area was coincident with a
30% decrease in ambient SO2 in the Detroit area. Secondary pollutants and other environmental
measures were also affected by the decrease in SO2, including sulfate aerosol, visibility
extinction due to sulfate aerosol, and sulfur deposition. The data record was not long enough to
determine the impact on PM2.5 mass concentrations.
The NOx SIP Call affected Michigan only in 2004. Because the data record prior to 2004
was short and the influence of mobile sources was confounding, no trend was established in
ambient NOx concentrations in response to the NOx SIP Call at the Detroit urban sites. Several
analyses (wind direction, time of day, weekday/weekend, and ratio) were used to isolate the
point source contributions; however, in all cases, no consistent trend was evident at any site.
More data are needed to establish a measurable temporal trend.
-------
September 28, 2006
Page 9
APPENDIX A
ADDITIONAL ANALYSES FOR DETROIT CASE STUDY
Additional analyses of the NOx data were conducted for the Detroit case study:
• Seasonal trends - determine whether changes in concentrations would be confounded by
seasonal variation.
• Weekday/weekend trends - look for a difference in mobile-dominated (i.e., weekday)
versus nonmobile-dominated (weekend) days.
• Morning only analysis - look for a difference in mobile-dominated (i.e., rush hour)
versus nonmobile dominated (non-rush hour) times.
• Ratio analyses - examine ratios of mobile source pollutants to NOx; a change in ratios
could indicate a change in nonmobile NOx emissions, assuming the mobile source
pollutant used remains constant.
• Ozone/total oxidant (Ox) analysis - as a product of NOx, ozone concentrations should be
affected by any change in NOx concentrations; Ox is the sum of NOx + ozone and would
also reflect changes in NOx concentrations.
No consistent trend was evident in any of these analyses to support a decrease in NOx
concentrations resulting from the NOx SIP Call. Graphs generated in the additional analyses
follow.
-------
September 28, 2006
Page 10
SEASONAL TRENDS
Lin wood
i i i i i i i i i i i i i i i
E 7 Mile
NW Haven
CD
Cl
Q_
n$y \Vr0p<$y \V- \t*rfiP'>}? ^ ^ \^>j<> ofP ^
Figure A-l. Seasonal distributions of NOx at three Detroit area sites. Seasonal
trends do not appear to change over time.
WEEKDAY/WEEKEND ANALYSIS
E 7 Mile, 24-hr avg
Linwood, 24-hr avg
NWHaven, 24-hr
2002
2003 2004
Summers Only
2005 2002
2003 2004
Summers Only
~ Weekday
~ Weekend
2005 1993 1994 1995 1996 1997
Summers Only
Figure A-2. Weekday versus weekend NOx concentrations at Detroit area sites.
Only summer data (June-August) were used.
-------
September 28, 2006
Page 11
MORNING ONLY ANALYSIS
E 7 Mile, hourly data
Linwood, hourly data
NW Haven, hourly data
2002
2003 2004
Summers Only
2005
2002
2003 2004
Summers Only
~ Morning
~ Non Morning
2005 1993 1994 1995 1996
Summers Only
Figure A-3. NOx concentrations at Detroit area sites comparing morning hours
(5 a.m.-9 a.m.) with the rest of the day. Only summer data (June-August) were
used. Concentrations above ppb are not shown.
1997
RATIO ANALYSES
Linwood
E 7 Mile
2002 2003 2004
Summer Only
2005
2002 2003 2004
Summer Only
2005
Figure A-4. S02:NOx ratio for Detroit area sites. Only summer data (June-
August) were used.
-------
September 28, 2006
Page 12
2002
2003 2004
Summer Only
2005 2002
2003 2004
Summer Only
~ Morning
~ Not Morning
2005 2002
2003 2004
Summer Only
~ Weekday
~ Weekend
Figure A-5. TNMOC:NOx ratio at East 7 Mile site. All hours/days, morning
versus non-morning, and weekday/weekend groupings presented. Only summer
data (June-Aug) were used.
13 5 7
Day of Week (Sun-Sat)
Figure A-6. TNMOC:NOx by day of week, East 7 Mile site, 2002-2005.
-------
September 28, 2006
Page 13
m
Q.
Q.
6
m
Q.
X
o
O
O
TNMOC:NOx
NOx 5
TNMOC
~l 1 1
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Figure A-7. NOx, TNMOC, and TNMOC :NOx diurnal profiles at East 7 Mile
site; 2002-2005 average.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Hour
Figure A-8. TNMOC:NOx diurnal profile by year for East 7 Mile site,
2002-2005. Median concentrations for each hour are shown.
-------
September 28, 2006
Page 14
0.4
£ 0.3
CL
00
t 0.2
0.1
0.0
2002 2003 2004 2005
Figure A-9. Benzene:NOx, East 7 Mile site, 2002-2005.
Hour
Figure A-10. Diurnal profile of benzene:NOx by year, East 7 Mile, 2002-2005.
Median concentrations by hour are shown; only summer data (June-August) were
used.
-------
September 28, 2006
Page 15
Linwood
Hour
E 7 Mile
Hour
Figure A-l 1. Ozone:NOx diurnal profile by year for Detroit area sites. Only
summer data (June-August) are shown. Ratios above 1000 were excluded from
analysis.
-------
September 28, 2006
Page 16
OZONE/Ox ANALYSIS
Warren, 8-hr max
125-
1993
1997 2001
Summers Only
2005
Pthuron, 8-hr max
1993
1997 2001
Summers Only
2005
Figure A-12. Seasonal ozone concentrations at Detroit area sites, 1993-2005. Only
summer data (June-Aug) are shown.
Linwood
30
20
10
0
0 1 2 3 4 5 6 7
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Hour
70
60
50
40
30
20
10
0
E7M ile
0 1 2 3 4 5 6 7
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Hour
Figure A-13. Ox (ozone + NOx) diurnal profile by year at Detroit area sites.
------- |