Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration

Office of Planning, Environment, & Realty (HEP)
PlanningEnvironmentReal Estate

HEP Events Guidance Publications Awards Contacts

FHWA and EPA National Near-Road Study - Detroit: Executive Summary

Part V: Results and Discussion

During this study gigabytes of data were collected- including data from continuous monitors such as the CO, NOX, BC and TEOM (PM10, PM2.5, PM Coarse) analyzers, etc.; integrated sample data; traffic data and video data. Results (i.e., data and graphs) presented herein were selected based on the original objectives of the study: "…to determine MSAT concentrations and variations in concentrations as a function of distance from the highway and to establish relationships between MSAT concentrations as related to highway traffic flows including traffic count, vehicle types, and speed; and meteorological conditions such as wind speed and wind direction; and other pollutants primarily emitted from motor vehicles such as CO, NO, NO2, NOX and BC."2

Traffic Activity.

Traffic activity for this location exhibited the typical bi-modal distribution as shown in Figure 3. It should be noted that construction activity on the M-39, an adjacent connecting freeway, may have influenced traffic patterns during a portion of this study. The field team noted that construction did take place on the M-39 during the late-winter/early-spring of 2011.

Line graph. Click image for source data.

Figure 3. Average Hourly Traffic Volume at I-96 Site from Sept. 29, 2010 through June 20, 2011.

Meteorology.

Figure 4 shows the wind direction and wind speeds observed during the course of the study. As can be seen from the wind roses, there were strong northwesterly winds during the morning commute hours. One implication of these wind conditions is that the study sites may have been influenced by nearby sources such as roadway traffic from US-24, Telegraph Road. Telegraph Road is a 6-lane divided highway approximately 440 meters west of Station 2 and 380 meters west of Station 3. An additional nearby source of air pollutants emissions was the apartment complex immediately adjacent (west of Station 2 and 3). The source of emissions at the apartment complex was most probably cold-start emissions from passenger vehicles.

Figure 4a: Wind Roses from the 100 meter station north of I-96 by time of day depicting wind direction and wind speeds observed during the course of the study indicating strong northwesterly winds during the morning commute hours.

Figure 4b: Vertical bar graph showing percent of variable, upwind, parallel and downwind wind direction from September 2010 through June 2011 showing the percentages of wind direction varying month to month

Figure 4 a. Wind Roses from the 100 m Station North of I-96 by time of day and Figure 4 b. Wind Direction in Percent, September 2010 through June 2011

Air Quality - Continuous Analyzers.

Figure 5 shows the spatial gradient for NO, NO2, and NOX for the study period. The spatial gradient is similar to the spatial gradient for the Las Vegas study. As shown by the figure the mean values for Station 3 are slightly higher than Station 2. This may be the result of air pollutant influences from nearby sources.

(a) Detroit

Line graphs. Click image for source data.

(b) Las Vegas.

2nd image for Figure 5. Source data linked on 1st image.

Figure 5 Average concentrations of NO, NO2 and NOx measured at all four monitoring stations indicating long-term trends in concentration gradients for each pollutant (a) Detroit. (b) Las Vegas - (source FHWA and EPA National Near-Road Study Las Vegas Final Report).

Figure 5 note: The lines connecting the points are provided as a visual aid to the reader and do not imply statistically significant differences in concentrations.

Figure 6 shows a polar plot for NO2 for all stations and all wind directions. The radial dimension is an indicator of wind speed (m/sec). Further away from the center of the plot, the higher the wind speed.

Figure 6: Polar plot of NO2 for all stations and all wind directions showing higher average NO2 concentrations at the 100 meter downwind and 10 meter roadside sites.

Figure 6 Polar plot for NO2 for all stations, all wind directions.

Note that higher average NO, NO2 and NOX concentrations and generally steeper gradients were observed during conditions when the winds are from the roadway as opposed to all wind directions (Figure 7). This may also be observed in CO concentration plots, Figure 8 and Figure 9. As shown by these figures (Figures 8, and 9) the mean values for Station 3 are slightly higher than Station 2. This may be the result of air pollutant influences from nearby sources.

Line graph. Click image for source data.

Figure 7 Average concentrations of NO, NO2, and NOx measured at all four monitoring stations indicating long-term trends in concentration gradients for each pollutant- winds from south.

Figure 7 note: The lines connecting the points are provided as a visual aid to the reader and do not imply statistically significant differences in concentrations.

Line graph. Click image for source data.

Figure 8 Average CO concentrations - winds from all wind directions.

Figure 8 note: The lines connecting the points are provided as a visual aid to the reader and do not imply statistically significant differences in concentrations.

Line graph. Click image for source data.

Figure 9. Average CO Concentrations -- winds from south.

Figure 9 note: The lines connecting the points are provided as a visual aid to the reader and do not imply statisticially significant differences in concentrations.

Air Quality - Integrated Samples -- VOC

Table 2 shows the number of observations, mean and 95% confidence intervals for the VOC data (TO-15 method).

Table 2. VOC -- averages for all wind directions (09/29/2010-06/15/2011)

Site name

Distance from Road

N (Obs.)

Mean (ppb)

95% CI (ppb)

Acrolein

4

100 m upwind

32

0.72

0.55 - 0.89

1

10 m roadside

37

0.69

0.56 - 0.82

2

100 m downwind

21

0.68

0.54 - 0.81

3

300 m downwind

34

0.63

0.51 - 0.74

1,3-Butadiene

4

100 m upwind

31

0.13

0.08 - 0.18

1

10 m roadside

37

0.19

0.14 - 0.24

2

100 m downwind

21

0.13

0.07 - 0.18

3

300 m downwind

34

0.12

0.07 - 0.16

Benzene

4

100 m upwind

32

0.36

0.27 - 0.44

1

10 m roadside

37

0.46

0.36 - 0.55

2

100 m downwind

21

0.29

0.19 - 0.40

3

300 m downwind

34

0.32

0.24 - 0.40

NOTE: Data are for valid samples only.

Data Caveats- Integrated Samples -- VOC

All sample results are presented with no blank or recovery correction. This was deemed unnecessary as the field blank values were either zero, below the method detection limit, or not statistically significant. While acrolein data is reported for the TO-15 method, caution should be used when assessing the data.

Air Quality - Integrated Samples -- Carbonyl

Table 3 shows the number of observations, mean and 95% confidence intervals for the carbonyl data (TO-11a method).

Table 3. Carbonyl -- averages for all wind directions (09/29/2010-06/15/2011)

Site name

Distance from Road

N (Obs.)

Mean (ppb)

95% CI (ppb)

Acrolein

4

100 m upwind

32

3.83

1.02 - 6.64

1

10 m roadside

34

4.08

1.43 - 6.74

2

100 m downwind

30

1.21

0.17 - 2.25

3

300 m downwind

36

1.12

0.34 - 1.90

Acetaldehyde

4

100 m upwind

32

2.16

1.38 - 2.94

1

10 m roadside

34

2.67

1.10 - 4.23

2

100 m downwind

30

2.05

1.24 - 2.87

3

300 m downwind

36

1.68

0.90 - 2.45

Formaldehyde

4

100 m upwind

32

3.14

1.93 - 4.35

1

10 m roadside

34

3.27

2.02 - 4.53

2

100 m downwind

30

2.60

1.30 - 3.89

3

300 m downwind

36

3.13

1.94 - 4.32

NOTE: Data are for valid samples only.

Data Caveats- Integrated Samples - Carbonyl

All sample results are presented with no blank or recovery correction. This was deemed unnecessary as the field blank values were either zero, below the method detection limit, or not statistically significant. While acrolein data is reported for the TO-11a method, caution should be used when assessing the data.

Data Caveats- Integrated Samples - Acrolein

Acrolein results are presented for both TO-15 (VOCs) and TO-11a (carbonyls) methods and the results should be used with caution. Method TO-15 utilizes passivated stainless steel canisters under vacuum to be filled at a constant rate to near ambient pressure for a specified time period. The air collected in the canisters undergoes laboratory analysis using a GC/MS. Method TO-11a utilizes cartridges containing DNPH coated media. These cartridges are connected to a sampler that draws ambient air at a constant rate for a specified time period. These cartridges undergo laboratory analysis using High-Performance Liquid Chromatography (HPLC)

Both methods are considered problematic as both methods have issues with the measurement of acrolein. The TO-15 method is considered problematic due to the "growth" of acrolein inside cleaned canisters. Acrolein concentrations inside cleaned canisters containing zero humidified air have been shown to increase over time due to unknown reasons. The TO-11a method is considered inaccurate due to the retention instability on the DNPH coated absorbent and low acrolein capture efficiency.

Moreover, caution should be used when comparing Las Vegas acrolein measurements to Detroit acrolein measurements. Acrolein values for Las Vegas were reported using TO-11a, while acrolein values for Detroit are reported using TO-11a and TO-15. We observed "growth" of acrolein in canisters during the Las Vegas study and for this reason we had very low confidence in the data and did not report acrolein results using TO-15. We did not observe "growth" of acrolein in canisters during the Detroit study.

EPA is continuing to research acrolein measurement methods, specifically focusing on the TO-15 method. This research is currently identified as a priority in EPA ORD's research action plan.

Air Quality - Integrated Samples - PM2.5

A summary of PM2.5 averages and confidence intervals are shown in Table 4. Figure 10 shows box-whisker plots PM2.5 integrated filter samples.

Table 4. PM2.5 Filters -- averages for all wind directions (09/29/2010-06/15/2011)

Site name

Distance from Road

N (Obs.)

Mean (µg/m3)

95% CI (µg/m3)

Station 4

-100

17

11.46

8.14 - 14.78

Station 1

10

19

12.87

8.84 - 16.90

Station 2

100

16

12.12

8.53 - 15.71

Station 3

300

18

10.40

7.10 - 13.69

Figure 10: Box-Whisker Plot of data in Table 4 showing PM2.5 concentrations for all stations, all sample times and all wind directions

Figure 10 Box-Whisker Plot for PM2.5 for all stations, all sample times, all wind directions.

Updated: 05/29/2014
HEP Home Planning Environment Real Estate
Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000