&EPA
United! S*K
Enviraimerilfll PiutmBmi
Agamy
2010 National Monitoring Programs Annual
Report (UATMP, NATTS, CSATAM)
Volume 1: Main
November 2012
Final Report
-------
EPA-454/R-12-006a
November 2012
2010 National Monitoring Programs Annual Report (UATMP, NATTS, CSATAM)
Volume 1: Main
By:
Eastern Research Group, Inc.
Morrisville, NC 27560
Prepared for:
Margaret Dougherty and David Shelow
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Contract No. EP-D-09-048
Delivery Orders 2, 6, 7, 8, 9, 10, & 11
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Emissions, Monitoring and Analysis Division
Research Triangle Park, NC 27711
-------
2010 National Monitoring Programs
Annual Report
(UATMP, NATTS, and CSATAM)
Final Report
EPA Contract No. EP-D-09-048
Delivery Orders 2, 6, 7, 8, 9, 10, & 11
Prepared for:
Margaret Dougherty and David Shelow
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Prepared by:
Eastern Research Group, Inc.
601 Keystone Park Drive, Suite 700
Morrisville, NC 27560
November 2012
-------
DISCLAIMER
Through its Office of Air Quality Planning and Standards, the U.S. Environmental Protection
Agency funded and managed the research described in this report under EPA Contract
No. EP-D-09-048 to Eastern Research Group, Inc. This report has been subjected to the
Agency's peer and administrative review and has been approved for publication as an EPA
document. Mention of trade names or commercial products in this report does not constitute
endorsement or recommendation for their use.
11
-------
TABLE OF CONTENTS (Continued)
TABLE OF CONTENTS
Page
List of Appendices xix
List of Figures xx
List of Tables xxxiv
List of Acronyms xliv
Abstract xlvi
1.0 Introduction 1-1
1.1 Background 1-1
1.2 The Report 1-2
2.0 The 2010 National Monitoring Programs Network 2-1
2.1 Monitoring Locations 2-1
2.2 Analytical Methods and Pollutants Targeted for Monitoring 2-12
2.2.1 VOC and SNMOC Concurrent Sampling and Analytical Methods ... 2-14
2.2.2 Carbonyl Compound Sampling and Analytical Method 2-18
2.2.3 PAH Sampling and Analytical Method 2-19
2.2.4 Metals Sampling and Analytical Method 2-20
2.2.5 Hexavalent Chromium Sampling and Analytical Method 2-21
2.3 Sample Collection Schedules 2-22
2.4 Completeness 2-29
3.0 Summary of the 2010 National Monitoring Programs Data Treatment and
Methods 3-1
3.1 Approach to Data Treatment 3-1
3.2 Human Health Risk and the Pollutants of Interest 3-3
3.3 Noncancer Risk Screening Evaluation Using Minimum Risk Levels 3-7
3.4 Additional Program-Level Analyses of the 2010 National Monitoring Programs
Dataset 3-8
3.4.1 The Effect of Mobile Source Emissions on Spatial Variations 3-8
3.4.2 Variability Analyses 3-9
3.4.3 Greenhouse Gas Assessment 3-10
in
-------
TABLE OF CONTENTS (Continued)
Page
3.5 Additional Site-Specific Analyses 3-11
3.5.1 Site Characterization 3-11
3.5.2 Meteorological Analysis 3-12
3.5.2.1 Back Trajectory Analysis 3-13
3.5.2.2 Wind Rose Analysis 3-14
3.5.3 Site-Specific Comparison to Program-level Average Concentrations 3-15
3.5.4 Site Trends Analysis 3-16
3.5.5 Risk Screening and Pollutants of Interest 3-17
3.5.5.1 Emission Tracer Analysis 3-17
3.5.5.2 Cancer and Noncancer Surrogate Risk Approximations 3-18
3.5.5.3 Risk-Based Emissions Assessment 3-19
4.0 Summary of the 2010 National Monitoring Programs Data 4-1
4.1 Statistical Results 4-1
4.1.1 Target Pollutant Detection Rates 4-1
4.1.2 Concentration Range and Data Distribution 4-13
4.1.3 Central Tendency 4-14
4.2 Preliminary Risk Screening and Pollutants of Interest 4-15
4.2.1 Concentrations of the Pollutants of Interest 4-21
4.2.2 Risk Screening Assessment Using MRLs 4-30
4.3 The Impact of Mobile Sources 4-32
4.3.1 Mobile Source Emissions 4-32
4.3.2 Hydrocarbon Concentrations 4-35
4.3.3 Motor Vehicle Ownership 4-35
4.3.4 Estimated Traffic Volume 4-37
4.3.5 Vehicle Miles Traveled 4-38
4.4 Variability Analysis 4-39
4.4.1 Coefficient of Variation and Inter-site Variability 4-39
4.4.2 Quarterly Variability Analysis 4-66
4.5 Greenhouse Gases 4-93
5.0 Sites in Arizona 5-1
5.1 Site Characterization 5-1
5.2 Meteorological Characterization 5-7
5.2.1 Climate Summary 5-7
5.2.2 Meteorological Conditions in 2010 5-8
5.2.3 Back Trajectory Analysis 5-10
5.2.4 Wind Rose Comparison 5-13
iv
-------
TABLE OF CONTENTS (Continued)
Page
5.3 Pollutants of Interest 5-16
5.4 Concentrations 5-18
5.4.1 2010 Concentration Averages 5-18
5.4.2 Concentration Comparison 5-22
5.4.3 Concentration Trends 5-26
5.5 Additional Risk Screening Evaluations 5-29
5.5.1 Risk Screening Assessment Using MRLs 5-29
5.5.2 Cancer and Noncancer Surrogate Risk Approximations 5-29
5.5.3 Risk-Based Emissions Assessment 5-32
5.6 Summary of the 2010 Monitoring Data for PXSS and SPAZ 5-36
6.0 Sites in California 6-1
6.1 Site Characterization 6-1
6.2 Meteorological Characterization 6-11
6.2.1 Climate Summary 6-11
6.2.2 Meteorological Conditions in 2010 6-11
6.2.3 Back Trajectory Analysis 6-14
6.2.4 Wind Rose Comparison 6-19
6.3 Pollutants of Interest 6-23
6.4 Concentrations 6-25
6.4.1 2010 Concentration Averages 6-25
6.4.2 Concentration Comparison 6-28
6.4.3 Concentration Trends 6-30
6.5 Additional Risk Screening Evaluations 6-31
6.5.1 Risk Screening Assessment Using MRLs 6-31
6.5.2 Cancer and Noncancer Surrogate Risk Approximations 6-31
6.5.3 Risk-Based Emissions Assessment 6-33
6.6 Summary of the 2010 Monitoring Data for CELA, RUCA, and SJJCA 6-39
7.0 Sites in Colorado 7-1
7.1 Site Characterization 7-1
7.2 Meteorological Characterization 7-14
7.2.1 Climate Summary 7-14
7.2.2 Meteorological Conditions in 2010 7-14
-------
TABLE OF CONTENTS (Continued)
Page
7.2.3 Back Trajectory Analysis 7-17
7.2.4 Wind Rose Comparison 7-24
7.3 Pollutants of Interest 7-32
7.4 Concentrations 7-34
7.4.1 2010 Concentration Averages 7-34
7.4.2 Concentration Comparison 7-39
7.4.3 Concentration Trends 7-45
7.5 Additional Risk Screening Evaluations 7-50
7.5.1 Risk Screening Assessment Using MRLs 7-50
7.5.2 Cancer and Noncancer Surrogate Risk Approximations 7-55
7.5.3 Risk-Based Emissions Assessment 7-58
7.6 Summary of the 2010 Monitoring Data for the Sites in Colorado 7-66
8.0 Site in District of Columbia 8-1
8.1 Site Characterization 8-1
8.2 Meteorological Characterization 8-6
8.2.1 Climate Summary 8-6
8.2.2 Meteorological Conditions in 2010 8-6
8.2.3 Back Trajectory Analysis 8-8
8.2.4 Wind Rose Comparison 8-10
8.3 Pollutants of Interest 8-12
8.4 Concentrations 8-13
8.4.1 2010 Concentration Averages 8-13
8.4.2 Concentration Comparison 8-15
8.4.3 Concentration Trends 8-17
8.5 Additional Risk Screening Evaluations 8-18
8.5.1 Risk Screening Assessment Using MRLs 8-18
8.5.2 Cancer and Noncancer Surrogate Risk Approximations 8-18
8.5.3 Risk-Based Emissions Assessment 8-19
8.6 Summary of the 2010 Monitoring Data for WADC 8-23
9.0 Sites in the Florida 9-1
9.1 Site Characterization 9-1
9.2 Meteorological Characterization 9-13
vi
-------
TABLE OF CONTENTS (Continued)
Page
9.2.1 Climate Summary 9-13
9.2.2 Meteorological Conditions in 2010 9-13
9.2.3 Back Trajectory Analysis 9-16
9.2.4 Wind Rose Comparison 9-23
9.3 Pollutants of Interest 9-30
9.4 Concentrations 9-32
9.4.1 2010 Concentration Averages 9-32
9.4.2 Concentration Comparison 9-36
9.4.3 Concentration Trends 9-41
9.5 Additional Risk Screening Evaluations 9-49
9.5.1 Risk Screening Assessment Using MRLs 9-49
9.5.2 Cancer and Noncancer Surrogate Risk Approximations 9-50
9.5.3 Risk-Based Emissions Assessment 9-52
9.6 Summary of the 2010 Monitoring Data for the Florida Sites 9-60
10.0 Site in Georgia 10-1
10.1 Site Characterization 10-1
10.2 Meteorological Characterization 10-6
10.2.1 Climate Summary 10-6
10.2.2 Meteorological Conditions in 2010 10-6
10.2.3 Back Trajectory Analysis 10-7
10.2.4 Wind Rose Comparison 10-10
10.3 Pollutants of Interest 10-12
10.4 Concentrations 10-13
10.4.1 2010 Concentration Averages 10-13
10.4.2 Concentration Comparison 10-14
10.4.3 Concentration Trends 10-16
10.5 Additional Risk Screening Evaluations 10-18
10.5.1 Risk Screening Assessment Using MRLs 10-18
10.5.2 Cancer and Noncancer Surrogate Risk Approximations 10-18
10.5.3 Risk-Based Emissions Assessment 10-19
10.6 Summary of the 2010 Monitoring Data for SDGA 10-23
vn
-------
TABLE OF CONTENTS (Continued)
Page
11.0 Sites in Illinois 11-1
11.1 Site Characterization 11-1
11.2 Meteorological Characterization 11-7
11.2.1 Climate Summary 11-8
11.2.2 Meteorological Conditions in 2010 11-8
11.2.3 Back Trajectory Analysis 11-10
11.2.4 Wind Rose Comparison 11-13
11.3 Pollutants of Interest 11-16
11.4 Concentrations 11-18
11.4.1 2010 Concentration Averages 11-19
11.4.2 Concentration Comparison 11-24
11.4.3 Concentration Trends 11-29
11.5 Additional Risk Screening Evaluations 11-39
11.5.1 Risk Screening Assessment Using MRLs 11-39
11.5.2 Cancer andNoncancer Surrogate Risk Approximations 11-43
11.5.3 Risk-Based Emissions Assessment 11-46
11.6 Summary of the 2010 Monitoring Data for NBIL and SPIL 11-50
12.0 Sites in Indiana 12-1
12.1 Site Characterization 12-1
12.2 Meteorological Characterization 12-8
12.2.1 Climate Summary 12-9
12.2.2 Meteorological Conditions in 2010 12-9
12.2.3 Back Trajectory Analysis 12-11
12.2.4 Wind Rose Comparison 12-14
12.3 Pollutants of Interest 12-18
12.4 Concentrations 12-19
12.4.1 2010 Concentration Averages 12-19
12.4.2 Concentration Comparison 12-21
12.4.3 Concentration Trends 12-22
12.5 Additional Risk Screening Evaluations 12-25
12.5.1 Risk Screening Assessment Using MRLs 12-25
12.5.2 Cancer and Noncancer Surrogate Risk Approximations 12-25
12.5.3 Risk-Based Emissions Assessment 12-26
viii
-------
TABLE OF CONTENTS (Continued)
Page
12.6 Summary of the 2010 Monitoring Data for INDEM and WPIN 12-30
13.0 Site in Kentucky 13-1
13.1 Site Characterization 13-1
13.2 Meteorological Characterization 13-6
13.2.1 Climate Summary 13-6
13.2.2 Meteorological Conditions in 2010 13-6
13.2.3 Back Trajectory Analysis 13-8
13.2.4 Wind Rose Comparison 13-10
13.3 Pollutants of Interest 13-12
13.4 Concentrations 13-13
13.4.1 2010 Concentration Averages 13-13
13.4.2 Concentration Comparison 13-15
13.4.3 Concentration Trends 13-17
13.5 Additional Risk Screening Evaluations 13-17
13.5.1 Risk Screening Assessment Using MRLs 13-17
13.5.2 Cancer andNoncancer Surrogate Risk Approximations 13-18
13.5.3 Risk-Based Emissions Assessment 13-19
13.6 Summary of the 2010 Monitoring Data for GLKY 13-23
14.0 Site in Massachusetts 14-1
14.1 Site Characterization 14-1
14.2 Meteorological Characterization 14-6
14.2.1 Climate Summary 14-6
14.2.2 Meteorological Conditions in 2010 14-6
14.2.3 Back Trajectory Analysis 14-8
14.2.4 Wind Rose Comparison 14-10
14.3 Pollutants of Interest 14-12
14.4 Concentrations 14-13
14.4.1 2010 Concentration Averages 14-13
14.4.2 Concentration Comparison 14-15
14.4.3 Concentration Trends 14-18
14.5 Additional Risk Screening Evaluations 14-21
14.5.1 Risk Screening Assessment Using MRLs 14-21
ix
-------
TABLE OF CONTENTS (Continued)
Page
14.5.2 Cancer and Noncancer Surrogate Risk Approximations 14-21
14.5.3 Risk-Based Emissions Assessment 14-22
14.6 Summary of the 2010 Monitoring Data for BOMA 14-26
15.0 Site in Michigan 15-1
15.1 Site Characterization 15-1
15.2 Meteorological Characterization 15-6
15.2.1 Climate Summary 15-6
15.2.2 Meteorological Conditions in 2010 15-6
15.2.3 Back Trajectory Analysis 15-8
15.2.4 Wind Rose Comparison 15-10
15.3 Pollutants of Interest 15-12
15.4 Concentrations 15-13
15.4.1 2010 Concentration Averages 15-13
15.4.2 Concentration Comparison 15-16
15.4.3 Concentration Trends 15-20
15.5 Additional Risk Screening Evaluations 15-23
15.5.1 Risk Screening Assessment Using MRLs 15-23
15.5.2 Cancer and Noncancer Surrogate Risk Approximations 15-23
15.5.3 Risk-Based Emissions Assessment 15-25
15.6 Summary of the 2010 Monitoring Data for DEMI 15-29
16.0 Site in Missouri 16-1
16.1 Site Characterization 16-1
16.2 Meteorological Characterization 16-6
16.2.1 Climate Summary 16-6
16.2.2 Meteorological Conditions in 2010 16-6
16.2.3 Back Trajectory Analysis 16-8
16.2.4 Wind Rose Comparison 16-10
16.3 Pollutants of Interest 16-12
16.4 Concentrations 16-13
16.4.1 2010 Concentration Averages 16-14
16.4.2 Concentration Comparison 16-18
16.4.3 Concentration Trends 16-23
x
-------
TABLE OF CONTENTS (Continued)
Page
16.5 Additional Risk Screening Evaluations 16-29
16.5.1 Risk Screening Assessment Using MRLs 16-30
16.5.2 Cancer and Noncancer Surrogate Risk Approximations 16-30
16.5.3 Risk-Based Emissions Assessment 16-32
16.6 Summary of the 2010 Monitoring Data for S4MO 16-36
17.0 Sites in New Jersey 17-1
17.1 Site Characterization 17-1
17.2 Meteorological Characterization 17-12
17.2.1 Climate Summary 17-13
17.2.2 Meteorological Conditions in 2010 17-13
17.2.3 Back Trajectory Analysis 17-15
17.2.4 Wind Rose Comparison 17-20
17.3 Pollutants of Interest 17-26
17.4 Concentrations 17-29
17.4.1 2010 Concentration Averages 17-29
17.4.2 Concentration Comparison 17-34
17.4.3 Concentration Trends 17-38
17.5 Additional Risk Screening Evaluations 17-49
17.5.1 Risk Screening Assessment Using MRLs 17-49
17.5.2 Cancer and Noncancer Surrogate Risk Approximations 17-49
17.5.3 Risk-Based Emissions Assessment 17-52
17.6 Summary of the 2010 Monitoring Data for the New Jersey Monitoring
Sites 17-58
18.0 Sites in New York 18-1
18.1 Site Characterization 18-1
18.2 Meteorological Characterization 18-13
18.2.1 Climate Summary 18-13
18.2.2 Meteorological Conditions in 2010 18-14
18.2.3 Back Trajectory Analysis 18-16
18.2.4 Wind Rose Comparison 18-21
18.3 Pollutants of Interest 18-28
18.4 Concentrations 18-30
xi
-------
TABLE OF CONTENTS (Continued)
Page
18.4.1 2010 Concentration Averages 18-30
18.4.2 Concentration Comparison 18-35
18.4.3 Concentration Trends 18-36
18.5 Additional Risk Screening Evaluations 18-36
18.5.1 Risk Screening Assessment Using MRLs 18-36
18.5.2 Cancer and Noncancer Surrogate Risk Approximations 18-37
18.5.3 Risk-Based Emissions Assessment 18-39
18.6 Summary of the 2010 Monitoring Data for the New York Monitoring Sites.. 18-45
19.0 Sites in Oklahoma 19-1
19.1 Site Characterization 19-1
19.2 Meteorological Characterization 19-13
19.2.1 Climate Summary 19-13
19.2.2 Meteorological Conditions in 2010 19-14
19.2.3 Back Trajectory Analysis 19-16
19.2.4 Wind Rose Comparison 19-22
19.3 Pollutants of Interest 19-28
19.4 Concentrations 19-33
19.4.1 2010 Concentration Averages 19-33
19.4.2 Concentration Comparison 19-41
19.4.3 Concentration Trends 19-48
19.5 Additional Risk Screening Evaluations 19-52
19.5.1 Risk Screening Assessment Using MRLs 19-52
19.5.2 Cancer and Noncancer Surrogate Risk Approximations 19-52
19.5.3 Risk-Based Emissions Assessment 19-58
19.6 Summary of the 2010 Monitoring Data for the Oklahoma Monitoring Sites.. 19-66
20.0 Site in Rhode Island 20-1
20.1 Site Characterization 20-1
20.2 Meteorological Characterization 20-6
20.2.1 Climate Summary 20-6
20.2.2 Meteorological Conditions in 2010 20-6
20.2.3 Back Trajectory Analysis 20-8
20.2.4 Wind Rose Comparison 20-10
xn
-------
TABLE OF CONTENTS (Continued)
Page
20.3 Pollutants of Interest 20-12
20.4 Concentrations 20-13
20.4.1 2010 Concentration Averages 20-13
20.4.2 Concentration Comparison 20-15
20.4.3 Concentration Trends 20-17
20.5 Additional Risk Screening Evaluations 20-18
20.5.1 Risk Screening Assessment Using MRLs 20-18
20.5.2 Cancer and Noncancer Surrogate Risk Approximations 20-18
20.5.3 Risk-Based Emissions Assessment 20-19
20.6 Summary of the 2010 Monitoring Data for PRRI 20-23
21.0 Site in South Carolina 21-1
21.1 Site Characterization 21-1
21.2 Meteorological Characterization 21-6
21.2.1 Climate Summary 21-6
21.2.2 Meteorological Conditions in 2010 21-6
21.2.3 Back Trajectory Analysis 21-8
21.2.4 Wind Rose Comparison 21-10
21.3 Pollutants of Interest 21-12
21.4 Concentrations 21-13
21.4.1 2010 Concentration Averages 21-13
21.4.2 Concentration Comparison 21-14
21.4.3 Concentration Trends 21-16
21.5 Additional Risk Screening Evaluations 21-17
21.5.1 Risk Screening Assessment Using MRLs 21-18
21.5.2 Cancer and Noncancer Surrogate Risk Approximations 21-18
21.5.3 Risk-Based Emissions Assessment 21-19
21.6 Summary of the 2010 Monitoring Data for CHSC 21-22
22.0 Sites in South Dakota 22-1
22.1 Site Characterization 22-1
22.2 Meteorological Characterization 22-9
22.2.1 Climate Summary 22-9
22.2.2 Meteorological Conditions in 2010 22-9
xiii
-------
TABLE OF CONTENTS (Continued)
Page
22.2.3 Back Trajectory Analysis 22-11
22.2.4 Wind Rose Comparison 22-14
22.3 Pollutants of Interest 22-17
22.4 Concentrations 22-19
22.4.1 2010 Concentration Averages 22-20
22.4.2 Concentration Comparison 22-23
22.4.3 Concentration Trends 22-26
22.5 Additional Risk Screening Evaluations 22-26
22.5.1 Risk Screening Assessment Using MRLs 22-26
22.5.2 Cancer and Noncancer Surrogate Risk Approximations 22-27
22.5.3 Risk-Based Emissions Assessment 22-29
22.6 Summary of the 2010 Monitoring Data for SSSD and UCSD 22-33
23.0 Sites in Texas 23-1
23.1 Site Characterization 23-1
23.2 Meteorological Characterization 23-8
23.2.1 Climate Summary 23-9
23.2.2 Meteorological Conditions in 2010 23-9
23.2.3 Back Trajectory Analysis 23-11
23.2.4 Wind Rose Comparison 23-15
23.3 Pollutants of Interest 23-18
23.4 Concentrations 23-19
23.4.1 2010 Concentration Averages 23-20
23.4.2 Concentration Comparison 23-21
23.4.3 Concentration Trends 23-24
23.5 Additional Risk Screening Evaluations 23-24
23.5.1 Risk Screening Assessment Using MRLs 23-24
23.5.2 Cancer and Noncancer Surrogate Risk Approximations 23-24
23.5.3 Risk-Based Emissions Assessment 23-25
23.6 Summary of the 2010 Monitoring Data for CAMS 35 and CAMS 85 23-29
24.0 Site in Utah 24-1
24.1 Site Characterization 24-1
xiv
-------
TABLE OF CONTENTS (Continued)
Page
24.2 Meteorological Characterization 24-6
24.2.1 Climate Summary 24-6
24.2.2 Meteorological Conditions in 2010 24-6
24.2.3 Back Trajectory Analysis 24-7
24.2.4 Wind Rose Comparison 24-10
24.3 Pollutants of Interest 24-12
24.4 Concentrations 24-13
24.4.1 2010 Concentration Averages 24-14
24.4.2 Concentration Comparison 24-18
24.4.3 Concentration Trends 24-22
24.5 Additional Risk Screening Evaluations 24-28
24.5.1 Risk Screening Assessment Using MRLs 24-28
24.5.2 Cancer and Noncancer Surrogate Risk Approximations 24-32
24.5.3 Risk-Based Emissions Assessment 24-34
24.6 Summary of the 2010 Monitoring Data for BTUT 24-38
25.0 Sites in Vermont 25-1
25.1 Site Characterization 25-1
25.2 Meteorological Characterization 25-10
25.2.1 Climate Summary 25-10
25.2.2 Meteorological Conditions in 2010 25-10
25.2.3 Back Trajectory Analysis 25-12
25.2.4 Wind Rose Comparison 25-16
25.3 Pollutants of Interest 25-21
25.4 Concentrations 25-24
25.4.1 2010 Concentration Averages 25-24
25.4.2 Concentration Comparison 25-29
25.4.3 Concentration Trends 25-34
25.5 Additional Risk Screening Evaluations 25-35
25.5.1 Risk Screening Assessment Using MRLs 25-35
25.5.2 Cancer and Noncancer Surrogate Risk Approximations 25-36
25.5.3 Risk-Based Emissions Assessment 25-39
25.6 Summary of the 2010 Monitoring Data for the Vermont Monitoring Sites .... 25-45
xv
-------
TABLE OF CONTENTS (Continued)
Page
26.0 Site in Virginia 26-1
26.1 Site Characterization 26-1
26.2 Meteorological Characterization 26-6
26.2.1 Climate Summary 26-6
26.2.2 Meteorological Conditions in 2010 26-6
26.2.3 Back Trajectory Analysis 26-8
26.2.4 Wind Rose Comparison 26-10
26.3 Pollutants of Interest 26-12
26.4 Concentrations 26-13
26.4.1 2010 Concentration Averages 26-13
26.4.2 Concentration Comparison 26-14
26.43 Concentration Trends 26-16
26.5 Additional Risk Screening Evaluations 26-16
26.5.1 Risk Screening Assessment Using MRLs 26-17
26.5.2 Cancer and Noncancer Surrogate Risk Approximations 26-17
26.5.3 Risk-Based Emissions Assessment 26-18
26.6 Summary of the 2010 Monitoring Data for RIVA 26-21
27.0 Site in Washington 27-1
27.1 Site Characterization 27-1
27.2 Meteorological Characterization 27-6
27.2.1 Climate Summary 27-6
27.2.2 Meteorological Conditions in 2010 27-6
27.2.3 Back Trajectory Analysis 27-8
27.2.4 Wind Rose Comparison 27-10
27.3 Pollutants of Interest 27-12
27.4 Concentrations 27-13
27.4.1 2010 Concentration Averages 27-13
27.4.2 Concentration Comparison 27-16
27.4.3 Concentration Trends 27-21
27.5 Additional Risk Screening Evaluations 27-21
27.5.1 Risk Screening Assessment Using MRLs 27-22
27.5.2 Cancer and Noncancer Surrogate Risk Approximations 27-22
27.5.3 Risk-Based Emissions Assessment 27-24
xvi
-------
TABLE OF CONTENTS (Continued)
Page
27.6 Summary of the 2010 Monitoring Data for SEW A 27-28
28.0 Site in Wisconsin 28-1
28.1 Site Characterization 28-1
28.2 Meteorological Characterization 28-6
28.2.1 Climate Summary 28-6
28.2.2 Meteorological Conditions in 2010 28-7
28.2.3 Back Trajectory Analysis 28-7
28.2.4 Wind Rose Comparison 28-10
28.3 Pollutants of Interest 28-12
28.4 Concentrations 28-13
28.4.1 2010 Concentration Averages 28-13
28.4.2 Concentration Comparison 28-15
28.4.3 Concentration Trends 28-15
28.5 Additional Risk Screening Evaluations 28-16
28.5.1 Risk Screening Assessment Using MRLs 28-16
28.5.2 Cancer and Noncancer Surrogate Risk Approximations 28-16
28.5.3 Risk-Based Emissions Assessment 28-17
28.6 Summary of the 2010 Monitoring Data for HOWI 28-21
29.0 Data Quality 29-1
29.1 Completeness 29-1
29.2 Method Precision 29-1
29.2.1 VOC Method Precision 29-4
29.2.2 SNMOC Method Precision 29-10
29.2.3 Carbonyl Compound Method Precision 29-13
29.2.4 Metals Method Precision 29-15
29.2.5 Hexavalent Chromium Method Precision 29-16
29.2.6 PAH Method Precision 29-17
29.3 Analytical Precision 29-18
29.3.1 VOC Analytical Precision 29-20
29.3.2 SNMOC Analytical Precision 29-27
29.3.3 Carbonyl Compound Analytical Precision 29-30
29.3.4 Metals Analytical Precision 29-33
29.3.5 Hexavalent Chromium Analytical Precision 29-33
29.3.6 PAH Analytical Precision 29-34
xvii
-------
TABLE OF CONTENTS (Continued)
Page
29.4 Accuracy 29-35
30.0 Results, Conclusions, and Recommendations 30-1
30.1 Summary of Results 30-1
30.1.1 National-level Summary 30-1
30.1.2 State-level Summary 30-2
30.1.3 Composite Site-level Summary 30-25
30.1.4 Data Quality Summary 30-27
30.2 Conclusions 30-28
30.3 Recommendations 30-29
31.0 References 31-1
xvin
-------
TABLE OF CONTENTS (Continued)
List of Appendices
Appendix A AQS Site Descriptions for the 2010 NATTS, UATMP, and CSATAM
Monitoring Sites
Appendix B Program Method Detection Limits (MDLs)
Appendix C 2010 VOC Raw Data
Appendix D 2010 SNMOC Raw Data
Appendix E 2010 Carbonyl Compounds Raw Data
Appendix F 2010 PAH Raw Data
Appendix G 2010 Metals Raw Data
Appendix H 2010 Hexavalent Chromium Raw Data
Appendix I Summary of Invalidated 2010 Samples
Appendix J 2010 Summary Statistics for VOC Monitoring
Appendix K 2010 Summary Statistics for SNMOC Monitoring
Appendix L 2010 Summary Statistics for Carbonyl Compounds Monitoring
Appendix M 2010 Summary Statistics for PAH Monitoring
Appendix N 2010 Summary Statistics for Metals Monitoring
Appendix O 2010 Summary Statistics for Hexavalent Chromium Monitoring
Appendix P Glossary of Terms
Appendix Q Risk Factors Used Throughout the 2010 NMP Report
xix
-------
LIST OF FIGURES
Page
2-1 Locations of the 2010 National Monitoring Programs Monitoring Sites 2-3
4-1 a Coefficient of Variation Analysis of Acenaphthene Across 26 Sites 4-42
4-lb Inter-Site Variability for Acenaphthene 4-42
4-2a Coefficient of Variation Analysis of Acetaldehyde Across 30 Sites 4-43
4-2b Inter-Site Variability for Acetaldehyde 4-43
4-3a Coefficient of Variation Analysis of Acrylonitrile Across 24 Sites 4-44
4-3b Inter-Site Variability for Acrylonitrile 4-44
4-4a Coefficient of Variation Analysis of Arsenic Across 14 Sites 4-45
4-4b Inter-Site Variability for Arsenic 4-45
4-5a Coefficient of Variation Analysis of Benzene Across 24 Sites 4-46
4-5b Inter-Site Variability for Benzene 4-46
4-6a Coefficient of Variation Analysis of Benzo(a)pyrene Across 26 Sites 4-47
4-6b Inter-Site Variability for Benzo(a)pyrene 4-47
4-7a Coefficient of Variation Analysis of Beryllium Across 14 Sites 4-48
4-7b Inter-Site Variability for Beryllium 4-48
4-8a Coefficient of Variation Analysis of 1,3-Butadiene Across 24 Sites 4-49
4-8b Inter-Site Variability for 1,3-Butadiene 4-49
4-9a Coefficient of Variation Analysis of Cadmium Across 14 Sites 4-50
4-9b Inter-Site Variability for Cadmium 4-50
4-10a Coefficient of Variation Analysis of Carbon Tetrachloride Across 24 Sites 4-51
4-10b Inter-Site Variability for Carbon Tetrachloride 4-51
4-1 la Coefficient of Variation Analysis of Chloroform Across 24 Sites 4-52
4-1 Ib Inter-Site Variability for Chloroform 4-52
4-12a Coefficient of Variation Analysis ofp-Dichlorobenzene Across 24 Sites 4-53
4-12b Inter-Site Variability for/>-Dichlorobenzene 4-53
4-13a Coefficient of Variation Analysis of 1,2-Dichloroethane Across 24 Sites 4-54
4-13b Inter-Site Variability for 1,2-Dichloroethane 4-54
4-14a Coefficient of Variation Analysis of Ethylbenzene Across 24 Sites 4-55
4-14b Inter-Site Variability for Ethylbenzene 4-55
4-15a Coefficient of Variation Analysis of Fluorene Across 26 Sites 4-56
4-15b Inter-Site Variability for Fluorene 4-56
4-16a Coefficient of Variation Analysis of Formaldehyde Across 30 Sites 4-57
4-16b Inter-Site Variability for Formaldehyde 4-57
4-17a Coefficient of Variation Analysis of Hexavalent Chromium Across 23 Sites 4-58
4-17b Inter-Site Variability for Hexavalent Chromium 4-58
4-18a Coefficient of Variation Analysis of Lead Across 14 Sites 4-59
4-18b Inter-Site Variability for Lead 4-59
4-19a Coefficient of Variation Analysis of Manganese Across 14 Sites 4-60
4-19b Inter-Site Variability for Manganese 4-60
4-20a Coefficient of Variation Analysis of Naphthalene Across 26 Sites 4-61
4-20b Inter-Site Variability for Naphthalene 4-61
4-21a Coefficient of Variation Analysis of Nickel Across 14 Sites 4-62
4-21b Inter-Site Variability for Nickel 4-62
xx
-------
LIST OF FIGURES (Continued)
Page
4-22a Coefficient of Variation Analysis of Tetrachloroethylene Across 24 Sites 4-63
4-22b Inter-Site Variability for Tetrachloroethylene 4-63
4-23a Coefficient of Variation Analysis of Trichloroethylene Across 24 Sites 4-64
4-23b Inter-Site Variability for Trichloroethylene 4-64
4-24a Coefficient of Variation Analysis of Vinyl Chloride Across 24 Sites 4-65
4-24b Inter-Site Variability for Vinyl Chloride 4-65
4-25 Comparison of Average Quarterly Acenaphthene Concentrations 4-69
4-26 Comparison of Average Quarterly Acetaldehyde Concentrations 4-70
4-27 Comparison of Average Quarterly Acrylonitrile Concentrations 4-71
4-28a Comparison of Average Quarterly Arsenic (PMio) Concentrations 4-72
4-28b Comparison of Average Quarterly Arsenic (TSP) Concentrations 4-72
4-29 Comparison of Average Quarterly Benzene Concentrations 4-73
4-30 Comparison of Average Quarterly Benzo(a)pyrene Concentrations 4-74
4-3 la Comparison of Average Quarterly Beryllium (PMio) Concentrations 4-75
4-3 Ib Comparison of Average Quarterly Beryllium (TSP) Concentrations 4-75
4-32 Comparison of Average Quarterly 1,3-Butadiene Concentrations 4-76
4-33a Comparison of Average Quarterly Cadmium (PMio) Concentrations 4-77
4-33b Comparison of Average Quarterly Cadmium (TSP) Concentrations 4-77
4-34 Comparison of Average Quarterly Carbon Tetrachloride Concentrations 4-78
4-3 5 Comparison of Average Quarterly Chloroform Concentrations 4-79
4-36 Comparison of Average Quarterly /?-Dichlorobenzene Concentrations 4-80
4-37 Comparison of Average Quarterly 1,2-Dichloroethane Concentrations 4-81
4-3 8 Comparison of Average Quarterly Ethylbenzene Concentrations 4-82
4-39 Comparison of Average Quarterly Fluorene Concentrations 4-83
4-40 Comparison of Average Quarterly Formaldehyde Concentrations 4-84
4-41 Comparison of Average Quarterly Hexavalent Chromium Concentrations 4-85
4-42a Comparison of Average Quarterly Lead (PMio) Concentrations 4-86
4-42b Comparison of Average Quarterly Lead (TSP) Concentrations 4-86
4-43a Comparison of Average Quarterly Manganese (PMio) Concentrations 4-87
4-43b Comparison of Average Quarterly Manganese (TSP) Concentrations 4-87
4-44 Comparison of Average Quarterly Naphthalene Concentrations 4-88
4-45a Comparison of Average Quarterly Nickel (PMio) Concentrations 4-89
4-45b Comparison of Average Quarterly Nickel (TSP) Concentrations 4-89
4-46 Comparison of Average Quarterly Tetrachloroethylene Concentrations 4-90
4-47 Comparison of Average Quarterly Trichloroethylene Concentrations 4-91
4-48 Comparison of Average Quarterly Vinyl Chloride Concentrations 4-92
5-1 Phoenix, Arizona (PXSS) Monitoring Site 5-2
5-2 South Phoenix, Arizona (SPAZ) Monitoring Site 5-3
5-3 NEI Point Sources Located Within 10 Miles of PXSS and SPAZ 5-4
5-4 2010 Composite Back Trajectory Map for PXSS 5-11
5-5 Back Trajectory Cluster Map for PXSS 5-11
5-6 2010 Composite Back Trajectory Map for SPAZ 5-12
5-7 Back Trajectory Cluster Map for SPAZ 5-12
xxi
-------
LIST OF FIGURES (Continued)
Page
5-8 Wind Roses for the Phoenix Sky Harbor International Airport Weather Station
nearPXSS 5-14
5-9 Wind Roses for the Phoenix Sky Harbor International Airport Weather Station
nearSPAZ 5-15
5-10 Program vs. Site-Specific Average Arsenic (PMio) Concentration 5-23
5-11 Program vs. Site-Specific Average Benzene Concentration 5-23
5-12 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 5-23
5-13 Program vs. Site-Specific Average 1,3-Butadiene Concentration 5-24
5-14 Program vs. Site-Specific Average Hexavalent Chromium Concentration 5-24
5-15 Program vs. Site-Specific Average Manganese (PMio) Concentration 5-24
5-16 Program vs. Site-Specific Average Naphthalene Concentration 5-25
5-17 Three-Year Rolling Statistical Metrics for Arsenic (PMio) Concentrations Measured
atPXSS 5-27
5-18 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at PXSS 5-27
5-19 Three-Year Rolling Statistical Metrics for Manganese (PMio) Concentrations
Measured at PXSS 5-28
6-1 Los Angeles, California (CELA) Monitoring Site 6-2
6-2 Rubidoux, California (RUCA) Monitoring Site 6-3
6-3 San Jose, California (SJJCA) Monitoring Site 6-4
6-4 NEI Point Sources Located Within 10 Miles of CELA 6-5
6-5 NEI Point Sources Located Within 10 Miles of RUCA 6-6
6-6 NEI Point Sources Located Within 10 Miles of SJJCA 6-7
6-7 2010 Composite Back Trajectory Map for CELA 6-15
6-8 Back Trajectory Cluster Map for CELA 6-15
6-9 2010 Composite Back Trajectory Map for RUCA 6-16
6-10 Back Trajectory Cluster Map for RUCA 6-16
6-11 2010 Composite Back Trajectory Map for SJJCA 6-17
6-12 Back Trajectory Cluster Map for SJJCA 6-17
6-13 Wind Roses for the Downtown Los Angeles/USC Campus Weather Station
near CELA 6-20
6-14 Wind Roses for the Riverside Municipal Airport Weather Station near RUCA 6-21
6-15 Wind Roses for the San Jose International Airport Weather Station near SJJCA 6-22
6-16 Program vs. Site-Specific Average Arsenic (PMio) Concentration 6-29
6-17 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 6-29
6-18 Program vs. Site-Specific Average Manganese (PMio) Concentration 6-30
6-19 Program vs. Site-Specific Average Naphthalene Concentration 6-30
7-1 Grand Junction, Colorado (GPCO) Monitoring Site 7-2
7-2 Battlement Mesa, Colorado (BMCO) Monitoring Site 7-3
7-3 Silt, Colorado (BRCO) Monitoring Site 7-4
7-4 Parachute, Colorado (PACO) Monitoring Site 7-5
7-5 Rifle, Colorado (RICO) Monitoring Site 7-6
7-6 Rulison, Colorado (RUCO) Monitoring Site 7-7
xxii
-------
LIST OF FIGURES (Continued)
Page
7-7 NEI Point Sources Located Within 10 Miles of GPCO 7-8
7-8 NEI Point Sources Located Within 10 Miles of BMCO, BRCO, PACO, RICO, and
RUCO 7-9
7-9 2010 Composite Back Trajectory Map for GPCO 7-18
7-10 Back Trajectory Cluster Map for GPCO 7-18
7-11 2010 Composite Back Trajectory Map for BMCO 7-19
7-12 2010 Composite Back Trajectory Map for BRCO 7-19
7-13 Back Trajectory Cluster Map for BRCO 7-20
7-14 2010 Composite Back Trajectory Map for PACO 7-20
7-15 Back Trajectory Cluster Map for PACO 7-21
7-16 2010 Composite Back Trajectory Map for RICO 7-21
7-17 Back Trajectory Cluster Map for RICO 7-22
7-18 2010 Composite Back Trajectory Map for RUCO 7-22
7-19 Back Trajectory Cluster Map for RUCO 7-23
7-20 Wind Roses for the Walker Field Airport Weather Station near GPCO 7-25
7-21 Wind Roses for the Garfield County Regional Airport near BMCO 7-26
7-22 Wind Roses for the Garfield County Regional Airport near BRCO 7-27
7-23 Wind Roses for the Garfield County Regional Airport near PACO 7-28
7-24 Wind Roses for the Garfield County Regional Airport near RICO 7-29
7-25 Wind Roses for the Garfield County Regional Airport near RUCO 7-30
7-26 Program vs. Site-Specific Average Acetaldehyde Concentration 7-40
7-27a Program vs. Site-Specific Average Benzene (Method TO-15) Concentration 7-40
7-27b Program vs. Site-Specific Average Benzene (SNMOC) Concentration 7-41
7-28 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 7-41
7-29a Program vs. Site-Specific Average 1,3-Butadiene (Method TO-15) Concentration 7-42
7-29b Program vs. Site-Specific Average 1,3-Butadiene (SNMOC) Concentration 7-42
7-30 Program vs. Site-Specific Average Formaldehyde Concentration 7-43
7-31 Program vs. Site-Specific Average Hexavalent Chromium Concentration 7-43
7-32 Program vs. Site-Specific Average Naphthalene Concentration 7-43
7-33 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations Measured
at GPCO 7-46
7-34 Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured at
GPCO 7-47
7-35 Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations Measured
at GPCO 7-47
7-36 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations Measured
at GPCO 7-48
7-37 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at GPCO 7-48
7-38 Dichloromethane Pollution Rose for GPCO 7-54
8-1 Washington, D.C. (WADC) Monitoring Site 8-2
8-2 NEI Point Sources Located Within 10 Miles of WADC 8-3
8-3 2010 Composite Back Trajectory Map for WADC 8-9
8-4 Back Trajectory Cluster Map for WADC 8-9
xxiii
-------
LIST OF FIGURES (Continued)
Page
8-5 Wind Roses for the Ronald Reagan Washington National Airport Weather Station
nearWADC 8-11
8-6 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 8-16
8-7 Program vs. Site-Specific Average Hexavalent Chromium Concentration 8-16
8-8 Program vs. Site-Specific Average Naphthalene Concentration 8-16
8-9 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at WADC 8-17
9-1 St. Petersburg, Florida (AZFL) Monitoring Site 9-2
9-2 Pinellas Park, Florida (SKFL) Monitoring Site 9-3
9-3 Plant City, Florida (SYFL) Monitoring Site 9-4
9-4 Winter Park, Florida (ORFL) Monitoring Site 9-5
9-5 Orlando, Florida (PAFL) Monitoring Site 9-6
9-6 NEI Point Sources Located Within 10 Miles of the Tampa/St. Petersburg, Florida
Monitoring Sites 9-7
9-7 NEI Point Sources Located Within 10 Miles of ORFL and PAFL 9-8
9-8 2010 Composite Back Trajectory Map for AZFL 9-16
9-9 Back Trajectory Cluster Map for AZFL 9-17
9-10 2010 Composite Back Trajectory Map for SKFL 9-17
9-11 Back Trajectory Cluster Map for SKFL 9-18
9-12 2010 Composite Back Trajectory Map for SYFL 9-18
9-13 Back Trajectory Cluster Map for SYFL 9-19
9-14 2010 Composite Back Trajectory Map for ORFL 9-19
9-15 Back Trajectory Cluster Map for ORFL 9-20
9-16 2010 Composite Back Trajectory Map for PAFL 9-20
9-17 Back Trajectory Cluster Map for PAFL 9-21
9-18 Wind Roses for the St. Petersburg/Whitted Airport Weather Station near AZFL 9-24
9-19 Wind Roses for the St. Petersburg/Clearwater International Airport Weather
Station near SKFL 9-25
9-20 Wind Roses for the Plant City Municipal Airport Weather Station near SYFL 9-26
9-21 Wind Roses for the Orlando Executive Airport Weather Station near ORFL 9-27
9-22 Wind Roses for the Orlando Executive Airport Weather Station near PAFL 9-28
9-23 Program vs. Site-Specific Average Acetaldehyde Concentration 9-36
9-24 Program vs. Site-Specific Average Arsenic (PMio) Concentration 9-37
9-25 Program vs. Site-Specific Average Benzo(a)Pyrene Concentration 9-37
9-26 Program vs. Site-Specific Average Formaldehyde Concentration 9-38
9-27 Program vs. Site-Specific Average Hexavalent Chromium Concentration 9-38
9-28 Program vs. Site-Specific Average Manganese (PMio) Concentration 9-39
9-29 Program vs. Site-Specific Average Naphthalene Concentration 9-39
9-30 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentration
Measured at AZFL 9-42
9-31 Three-Year Rolling Statistical Metrics for Formaldehyde Concentration
Measured at AZFL 9-42
9-32 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at ORFL 9-43
xxiv
-------
LIST OF FIGURES (Continued)
Page
9-33 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at ORFL 9-43
9-34 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at SKFL 9-44
9-35 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at SKFL 9-44
9-36 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at SYFL 9-45
9-37 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at SYFL 9-45
9-38 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at SYFL 9-46
10-1 Decatur, Georgia (SDGA) Monitoring Site 10-2
10-2 NEI Point Sources Located Within 10 Miles of SDGA 10-3
10-3 2010 Composite Back Trajectory Map for SDGA 10-9
10-4 Back Trajectory Cluster Map for SDGA 10-9
10-5 Wind Roses for the Hartsfield International Airport Weather Station near SDGA 10-11
10-6 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 10-15
10-7 Program vs. Site-Specific Average Hexavalent Chromium Concentration 10-15
10-8 Program vs. Site-Specific Average Naphthalene Concentration 10-15
10-9 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at SDGA 10-17
11-1 Northbrook, Illinois (NBIL) Monitoring Site 11-2
11-2 Schiller Park, Illinois (SPIL) Monitoring Site 11-3
11-3 NEI Point Sources Located Within 10 Miles of NBIL and SPIL 11-4
11-4 2010 Composite Back Trajectory Map for NBIL 11-10
11-5 Back Trajectory Cluster Map for NBIL 11-11
11-6 2010 Composite Back Trajectory Map for SPIL 11-11
11-7 Back Trajectory Cluster Map for SPIL 11-12
11-8 Wind Roses for the Palwaukee Municipal Airport Weather Station near NBIL 11-14
11-9 Wind Roses for the O'Hare International Airport Weather Station near SPIL 11-15
11-10 Program vs. Site-Specific Average Acetaldehyde Concentration 11 -24
11-11 Program vs. Site-Specific Average Arsenic (PMio) Concentration 11-25
11-12 Program vs. Site-Specific Average Benzene Concentration 11-25
11-13 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 11-25
11-14 Program vs. Site-Specific Average 1,3-Butadiene Concentration 11-26
11-15 Program vs. Site-Specific Average Formaldehyde Concentration 11-26
11-16 Program vs. Site-Specific Average Hexavalent Chromium Concentration 11 -27
11-17 Program vs. Site-Specific Average Manganese (PMio) Concentration 11-27
11-18 Program vs. Site-Specific Average Naphthalene Concentration 11 -27
11-19 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations Measured
at NBIL 11-30
XXV
-------
LIST OF FIGURES (Continued)
Page
11-20 Three-Year Rolling Statistical Metrics for Arsenic (PMio) Concentrations Measured
atNBIL 11-30
11-21 Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured at
NBIL 11-31
11-22 Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations Measured
atNBIL 11-31
11-23 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations Measured
atNBIL 11-32
11-24 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at NBIL 11-32
11-25 Three-Year Rolling Statistical Metrics for Manganese (PMio) Concentrations
Measured at NBIL 11-33
11-26 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations Measured
atSPIL 11-33
11-27 Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured at
SPIL 11-34
11-28 Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations Measured
at SPIL 11-34
11-29 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations Measured
at SPIL 11-35
11-30 Formaldehyde Pollution Rose for NBIL 11-42
12-1 Gary, Indiana (INDEM) Monitoring Site 12-2
12-2 Indianapolis, Indiana (WPIN) Monitoring Site 12-3
12-3 NEI Point Sources Located Within 10 Miles of INDEM 12-4
12-4 NEI Point Sources Located Within 10 Miles of WPIN 12-5
12-5 2010 Composite Back Trajectory Map for INDEM 12-11
12-6 Back Trajectory Cluster Map for INDEM 12-12
12-7 2010 Composite Back Trajectory Map for WPIN 12-12
12-8 Back Trajectory Cluster Map for WPIN 12-13
12-9 Wind Roses for the Lansing Municipal Airport Weather Station near INDEM 12-15
12-10 Wind Roses for the Indianapolis International Airport Weather Station near WPIN... 12-16
12-11 Program vs. Site-Specific Average Acetaldehyde Concentration 12-21
12-12 Program vs. Site-Specific Average Formaldehyde Concentration 12-22
12-13 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at INDEM 12-23
12-14 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at INDEM 12-23
13-1 Grayson, Kentucky (GLKY) Monitoring Site 13-2
13-2 NEI Point Sources Located Within 10 Miles of GLKY 13-3
13-3 2010 Composite Back Trajectory Map for GLKY 13-9
13-4 Back Trajectory Cluster Map for GLKY 13-9
13-5 Wind Roses for the Tri-State/MJ. Ferguson Field Airport Weather Station
near GLKY 13-11
xxvi
-------
LIST OF FIGURES (Continued)
Page
13-6 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 13-15
13-7 Program vs. Site-Specific Average Hexavalent Chromium Concentration 13-16
13-8 Program vs. Site-Specific Average Naphthalene Concentration 13-16
14-1 Boston, Massachusetts (BOMA) Monitoring Site 14-2
14-2 NEI Point Sources Located Within 10 Miles of BOMA 14-3
14-3 2010 Composite Back Trajectory Map for BOMA 14-8
14-4 Back Trajectory Cluster Map for BOMA 14-9
14-5 Wind Roses for the Logan International Airport Weather Station near BOMA 14-11
14-6 Program vs. Site-Specific Average Arsenic (PMio) Concentration 14-15
14-7 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 14-16
14-8 Program vs. Site-Specific Average Hexavalent Chromium Concentration 14-16
14-9 Program vs. Site-Specific Average Manganese (PMio) Concentration 14-16
14-10 Program vs. Site-Specific Average Naphthalene Concentration 14-17
14-11 Three-Year Rolling Statistical Metrics for Arsenic (PMio) Concentrations
Measured at BOMA 14-18
14-12 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at BOMA 14-19
14-13 Three-Year Rolling Statistical Metrics for Manganese (PMio) Concentrations
Measured at BOMA 14-19
15-1 Dearborn, Michigan (DEMI) Monitoring Site 15-2
15-2 NEI Point Sources Located Within 10 Miles of DEMI 15-3
15-3 2010 Composite Back Trajectory Map for DEMI 15-9
15-4 Back Trajectory Cluster Map for DEMI 15-9
15-5 Wind Roses for the Detroit City Airport Weather Station near DEMI 15-11
15-6 Program vs. Site-Specific Average Acetaldehyde Concentration 15-16
15-7 Program vs. Site-Specific Average Benzene Concentration 15-17
15-8 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 15-17
15-9 Program vs. Site-Specific Average 1,3-Butadiene Concentration 15-17
15-10 Program vs. Site-Specific Average Formaldehyde Concentration 15-18
15-11 Program vs. Site-Specific Average Hexavalent Chromium Concentration 15-18
15-12 Program vs. Site-Specific Average Naphthalene Concentration 15-18
15-13 Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured at
DEMI 15-21
15-14 Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations Measured
at DEMI 15-21
15-15 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at DEMI 15-22
16-1 St. Louis, Missouri (S4MO) Monitoring Site 16-2
16-2 NEI Point Sources Located Within 10 Miles of S4MO 16-3
16-3 2010 Composite Back Trajectory Map for S4MO 16-8
16-4 Back Trajectory Cluster Map for S4MO 16-9
16-5 Wind Roses for the St. Louis Downtown Airport Weather Station near S4MO 16-11
xxvii
-------
LIST OF FIGURES (Continued)
Page
16-6 Program vs. Site-Specific Average Acetaldehyde Concentration 16-19
16-7 Program vs. Site-Specific Average Arsenic (PMio) Concentration 16-19
16-8 Program vs. Site-Specific Average Benzene Concentration 16-19
16-9 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 16-20
16-10 Program vs. Site-Specific Average 1,3-Butadiene Concentration 16-20
16-11 Program vs. Site-Specific Average Formaldehyde Concentration 16-20
16-12 Program vs. Site-Specific Average Hexavalent Chromium Concentration 16-21
16-13 Program vs. Site-Specific Average Manganese (PMio) Concentration 16-21
16-14 Program vs. Site-Specific Average Naphthalene Concentration 16-21
16-15 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations Measured
atS4MO 16-24
16-16 Three-Year Rolling Statistical Metrics for Arsenic (PMio) Concentrations Measured
atS4MO 16-24
16-17 Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured at
S4MO 16-25
16-18 Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations Measured
atS4MO 16-25
16-19 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations Measured
atS4MO 16-26
16-20 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at S4MO 16-26
16-21 Three-Year Rolling Statistical Metrics for Manganese (PMio) Concentrations
Measured at S4MO 16-27
17-1 Chester, New Jersey (CHNJ) Monitoring Site 17-2
17-2 Elizabeth, New Jersey (ELNJ) Monitoring Site 17-3
17-3 New Brunswick, New Jersey (NBNJ) Monitoring Site 17-4
17-4 Paterson, New Jersey (PANJ) Monitoring Site 17-5
17-5 NEI Point Sources Located Within 10 Miles of CHNJ 17-6
17-6 NEI Point Sources Located Within 10 Miles of ELNJ and NBNJ 17-7
17-7 NEI Point Sources Located Within 10 Miles of PANJ 17-8
17-8 2010 Composite Back Trajectory Map for CHNJ 17-16
17-9 Back Trajectory Cluster Map for CHNJ 17-16
17-10 2010 Composite Back Trajectory Map for ELNJ 17-17
17-11 Back Trajectory Cluster Map for ELNJ 17-17
17-12 2010 Composite Back Trajectory Map for NBNJ 17-18
17-13 Back Trajectory Cluster Map for NBNJ 17-18
17-14 2010 Composite Back Trajectory Map for PANJ 17-19
17-15 Back Trajectory Cluster Map for PANJ 17-19
17-16 Wind Roses for the Summerville-Somerset Airport Weather Station near CHNJ 17-21
17-17 Wind Roses for the Newark International Airport Weather Station near ELNJ 17-22
17-18 Wind Roses for the Summerville-Somerset Airport Weather Station near NBNJ 17-23
17-19 Wind Roses for the Essex County Airport Weather Station near PANJ 17-24
17-20 Program vs. Site-Specific Average Acetaldehyde Concentration 17-35
17-21 Program vs. Site-Specific Average Benzene Concentration 17-35
xxviii
-------
LIST OF FIGURES (Continued)
Page
17-22 Program vs. Site-Specific Average 1,3-Butadiene Concentration 17-36
17-23 Program vs. Site-Specific Average Formaldehyde Concentration 17-36
17-24 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations Measured
atCHNJ 17-38
17-25 Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured at
CHNJ 17-39
17-26 Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations Measured
at CHNJ 17-39
17-27 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations Measured
at CHNJ 17-40
17-28 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations Measured
atELNJ 17-40
17-29 Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured at
ELNJ 17-41
17-30 Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations Measured
atELNJ 17-41
17-31 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations Measured
atELNJ 17-42
17-32 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations Measured
atNBNJ 17-42
17-33 Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured at
NBNJ 17-43
17-34 Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations Measured
atNBNJ 17-43
17-35 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations Measured
atNBNJ 17-44
18-1 Public School 52, New York City, New York (BXNY) Monitoring Site 18-2
18-2 Morrisania, New York City, New York (MONY) Monitoring Site 18-3
18-3 Rochester, New York (ROCH) Monitoring Site 18-4
18-4 Tonawanda, New York (TONY) Monitoring Site 18-5
18-5 NEI Point Sources Located Within 10 Miles of BXNY and MONY 18-6
18-6 NEI Point Sources Located Within 10 Miles of ROCH 18-7
18-7 NEI Point Sources Located Within 10 Miles of TONY 18-8
18-8 2010 Composite Back Trajectory Map for BXNY 18-16
18-9 2010 Composite Back Trajectory Map for MONY 18-17
18-10 Back Trajectory Cluster Map for MONY 18-17
18-11 2010 Composite Back Trajectory Map for ROCH 18-18
18-12 Back Trajectory Cluster Map for ROCH 18-18
18-13 2010 Composite Back Trajectory Map for TONY 18-19
18-14 Back Trajectory Cluster Map for TONY 18-19
18-15 Wind Roses for the LaGuardia International Airport Weather Station near BXNY.... 18-24
18-16 Wind Roses for the Central Park Weather Station near MONY 18-25
18-17 Wind Roses for the Greater Rochester International Airport Weather Station
near ROCH 18-26
xxix
-------
LIST OF FIGURES (Continued)
Page
18-18 Wind Roses for the Niagara Falls International Airport Weather Station near
TONY 18-27
18-19 Program vs. Site-Specific Average Hexavalent Chromium Concentration 18-35
19-1 Tulsa, Oklahoma (TOOK) Monitoring Site 19-2
19-2 Tulsa, Oklahoma (TMOK) Monitoring Site 19-3
19-3 Pryor Creek, Oklahoma (PROK) Monitoring Site 19-4
19-4 Midwest City, Oklahoma (MWOK) Monitoring Site 19-5
19-5 Oklahoma City, Oklahoma (OCOK) Monitoring Site 19-6
19-6 NEI Point Sources Located Within 10 Miles of TMOK and TOOK 19-7
19-7 NEI Point Sources Located Within 10 Miles of PROK 19-8
19-8 NEI Point Sources Located Within 10 Miles of MWOK and OCOK 19-9
19-9 2010 Composite Back Trajectory Map for TOOK 19-16
19-10 Back Trajectory Cluster Map for TOOK 19-17
19-11 2010 Composite Back Trajectory Map for TMOK 19-17
19-12 Back Trajectory Cluster Map for TMOK 19-18
19-13 2010 Composite Back Trajectory Map for PROK 19-18
19-14 Back Trajectory Cluster Map for PROK 19-19
19-15 2010 Composite Back Trajectory Map for MWOK 19-19
19-16 Back Trajectory Cluster Map for MWOK 19-20
19-17 2010 Composite Back Trajectory Map for OCOK 19-20
19-18 Back Trajectory Cluster Map for OCOK 19-21
19-19 Wind Roses for the Richard Lloyd Jones Jr. Airport Weather Station near TOOK 19-23
19-20 Wind Roses for the Tulsa International Airport Weather Station near TMOK 19-24
19-21 Wind Roses for the Claremore Regional Airport Weather Station near PROK 19-25
19-22 Wind Roses for the Tinker Air Force Base Airport Weather Station near MWOK 19-26
19-23 Wind Roses for the Wiley Post Airport Weather Station near OCOK 19-27
19-24 Program vs. Site-Specific Average Acetaldehyde Concentration 19-42
19-25 Program vs. Site-Specific Average Arsenic (TSP) Concentration 19-43
19-26 Program vs. Site-Specific Average Benzene Concentration 19-44
19-27 Program vs. Site-Specific Average 1,3-Butadiene Concentration 19-45
19-28 Program vs. Site-Specific Average Formaldehyde Concentration 19-46
19-29 Program vs. Site-Specific Average Manganese (TSP) Concentration 19-47
19-30 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations Measured
at TOOK 19-49
19-31 Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured at
TOOK 19-49
19-32 Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations Measured
at TOOK 19-50
19-33 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations Measured
at TOOK 19-50
20-1 Providence, Rhode Island (PRRI) Monitoring Site 20-2
20-2 NEI Point Sources Located Within 10 Miles of PRRI 20-3
20-3 2010 Composite Back Trajectory Map for PRRI 20-9
XXX
-------
LIST OF FIGURES (Continued)
Page
20-4 Back Trajectory Cluster Map for PRRI 20-9
20-5 Wind Roses for the T.F. Green State Airport Weather Station near PRRI 20-11
20-6 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 20-16
20-7 Program vs. Site-Specific Average Hexavalent Chromium Concentration 20-16
20-8 Program vs. Site-Specific Average Naphthalene Concentration 20-16
20-9 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at PRRI 20-17
21-1 Chesterfield, South Carolina (CHSC) Monitoring Site 21-2
21-2 NEI Point Sources Located Within 10 Miles of CHSC 21-3
21-3 2010 Composite Back Trajectory Map for CHSC 21-9
21-4 Back Trajectory Cluster Map for CHSC 21-9
21-5 Wind Roses for the Monroe Airport Weather Station near CHSC 21-11
21-6 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 21-15
21-7 Program vs. Site-Specific Average Hexavalent Chromium Concentration 21-15
21-8 Program vs. Site-Specific Average Naphthalene Concentration 21-15
21-9 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at CHSC 21-17
22-1 Sioux Falls, South Dakota (SSSD) Monitoring Site 22-2
22-2 Union County, South Dakota (UCSD) Monitoring Site 22-3
22-3 NEI Point Sources Located Within 10 Miles of SSSD 22-4
22-4 NEI Point Sources Located Within 10 Miles of UCSD 22-5
22-5 2010 Composite Back Trajectory Map for SSSD 22-12
22-6 Back Trajectory Cluster Map for SSSD 22-12
22-7 2010 Composite Back Trajectory Map for UCSD 22-13
22-8 Back Trajectory Cluster Map for UCSD 22-13
22-9 Wind Roses for the Joe Foss Field Airport Weather Station near SSSD 22-15
22-10 Wind Roses for the Sioux Gateway Airport Weather Station near UCSD 22-16
22-11 Program vs. Site-Specific Average Acetaldehyde Concentration 22-24
22-12 Program vs. Site-Specific Average Benzene Concentration 22-24
22-13 Program vs. Site-Specific Average 1,3-Butadiene Concentration 22-25
22-14 Program vs. Site-Specific Average Formaldehyde Concentration 22-25
23-1 Deer Park, Texas (CAMS 35) Monitoring Site 23-2
23-2 Karnack, Texas (CAMS 85) Monitoring Site 23-3
23-3 NEI Point Sources Located Within 10 Miles of CAMS 35 23-4
23-4 NEI Point Sources Located Within 10 Miles of CAMS 85 23-5
23-5 2010 Composite Back Trajectory Map for CAMS 35 23-13
23-6 Back Trajectory Cluster Map for CAMS 35 23-13
23-7 2010 Composite Back Trajectory Map for CAMS 85 23-14
23-8 Back Trajectory Cluster Map for CAMS 85 23-14
23-9 Wind Roses for the William P. Hobby Airport Weather Station near CAMS 35 23-16
23-10 Wind Roses for the Shreveport Regional Airport Weather Station near CAMS 85 23-17
23-11 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 23-22
xxxi
-------
LIST OF FIGURES (Continued)
Page
23-12 Program vs. Site-Specific Average Hexavalent Chromium Concentration 23-22
23-13 Program vs. Site-Specific Average Naphthalene Concentration 23-23
24-1 Bountiful, Utah (BTUT) Monitoring Site 24-2
24-2 NEI Point Sources Located Within 10 Miles of BTUT 24-3
24-3 2010 Composite Back Trajectory Map for BTUT 24-9
24-4 Back Trajectory Cluster Map for BTUT 24-9
24-5 Wind Roses for the Salt Lake City International Airport Weather Station near
BTUT 24-11
24-6 Program vs. Site-Specific Average Acetaldehyde Concentration 24-18
24-7 Program vs. Site-Specific Average Arsenic (PMio) Concentration 24-18
24-8 Program vs. Site-Specific Average Benzene Concentration 24-19
24-9 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 24-19
24-10 Program vs. Site-Specific Average 1,3-Butadiene Concentration 24-19
24-11 Program vs. Site-Specific Average Formaldehyde Concentration 24-20
24-12 Program vs. Site-Specific Average Hexavalent Chromium Concentration 24-20
24-13 Program vs. Site-Specific Average Manganese (PMio) Concentration 24-20
24-14 Program vs. Site-Specific Average Naphthalene Concentration 24-21
24-15 Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations Measured
at BTUT 24-23
24-16 Three-Year Rolling Statistical Metrics for Arsenic (PMio) Concentrations Measured
at BTUT 24-23
24-17 Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured at
BTUT 24-24
24-18 Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations Measured
at BTUT 24-24
24-19 Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations Measured
at BTUT 24-25
24-20 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at BTUT 24-25
24-21 Three-Year Rolling Statistical Metrics for Manganese (PMio) Concentrations
Measured at BTUT 24-26
24-22 Dichloromethane Pollution Rose for BTUT 24-31
25-1 Burlington, Vermont (BURVT) Monitoring Site 25-2
25-2 Underbill, Vermont (UNVT) Monitoring Site 25-3
25-3 Rutland, Vermont (RUVT) Monitoring Site 25-4
25-4 NEI Point Sources Located Within 10 Miles of BURVT and UNVT 25-5
25-5 NEI Point Sources Located Within 10 Miles of RUVT 25-6
25-6 2010 Composite Back Trajectory Map for BURVT 25-13
25-7 Back Trajectory Cluster Map for BURVT 25-13
25-8 2010 Composite Back Trajectory Map for RUVT 25-14
25-9 Back Trajectory Cluster Map for RUVT 25-14
25-10 2010 Composite Back Trajectory Map for UNVT 25-15
25-11 Back Trajectory Cluster Map for UNVT 25-15
xxxii
-------
LIST OF FIGURES (Continued)
Page
25-12 Wind Roses for the Burlington International Airport Weather Station near BURVT.. 25-18
25-13 Wind Roses for the Rutland State Airport Weather Station near RUVT 25-19
25-14 Wind Roses for the Morrisville-Stowe State Airport Weather Station near UNVT 25-20
25-15 Program vs. Site-Specific Average Arsenic (PMio) Concentration 25-29
25-16 Program vs. Site-Specific Average Benzene Concentration 25-30
25-17 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 25-30
25-18 Program vs. Site-Specific Average 1,3-Butadiene Concentration 25-31
25-19 Program vs. Site-Specific Average Hexavalent Chromium Concentration 25-31
25-20 Program vs. Site-Specific Average Manganese (PMio) Concentration 25-32
25-21 Program vs. Site-Specific Average Naphthalene Concentration 25-32
25-22 Three-Year Rolling Statistical Metrics for Hexavalent Chromium Concentrations
Measured at UNVT 25-35
26-1 Richmond, Virginia (RIVA) Monitoring Site 26-2
26-2 NEI Point Sources Located Within 10 Miles of RIVA 26-3
26-3 2010 Composite Back Trajectory Map for RIVA 26-8
26-4 Back Trajectory Cluster Map for RIVA 26-9
26-5 Wind Roses for the Richmond International Airport Weather Station near RIVA 26-11
26-6 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 26-15
26-7 Program vs. Site-Specific Average Hexavalent Chromium Concentration 26-15
26-8 Program vs. Site-Specific Average Naphthalene Concentration 26-15
27-1 Seattle, Washington (SEWA) Monitoring Site 27-2
27-2 NEI Point Sources Located Within 10 Miles of SEW A 27-3
27-3 2010 Composite Back Trajectory Map for SEW A 27-9
27-4 Back Trajectory Cluster Map for SEW A 27-9
27-5 Wind Roses for the Boeing Field/King County International Airport Weather
Station near SEW A 27-11
27-6 Program vs. Site-Specific Average Acetaldehyde Concentration 27-17
27-7 Program vs. Site-Specific Average Arsenic (PMio) Concentration 27-17
27-8 Program vs. Site-Specific Average Benzene Concentration 27-17
27-9 Program vs. Site-Specific Average Benzo(a)pyrene Concentration 27-18
27-10 Program vs. Site-Specific Average 1,3-Butadiene Concentration 27-18
27-11 Program vs. Site-Specific Average Formaldehyde Concentration 27-18
27-12 Program vs. Site-Specific Average Hexavalent Chromium Concentration 27-19
27-13 Program vs. Site-Specific Average Manganese (PMio) Concentration 27-19
27-14 Program vs. Site-Specific Average Naphthalene Concentration 27-19
28-1 Horicon, Wisconsin (HOWI) Monitoring Site 28-2
28-2 NEI Point Sources Located Within 10 Miles of HOWI 28-3
28-3 Composite Back Trajectory Map for HOWI 28-9
28-4 Back Trajectory Cluster Map for HOWI 28-9
28-5 Wind Roses for the Dodge County Airport Weather Station near HOWI 28-11
28-6 Program vs. Site-Specific Average Hexavalent Chromium Concentration 28-15
xxxin
-------
LIST OF TABLES
Page
1-1 Organization of the 2010 National Monitoring Programs Report 1-4
2-1 2010 National Monitoring Programs Sites and Past Program Participation 2-4
2-2 Site Characterizing Information for 2010 National Monitoring Programs Sites 2-7
2-3 2010 VOC Method Detect!on Limits 2-16
2-4 2010SNMOC Method Detect!on Limits 2-17
2-5 2010 Carbonyl Compound Method Detection Limits 2-19
2-6 2010 PAH Method Detect on Limits 2-20
2-7 2010 Metals Method Detect!on Limits 2-21
2-8 20 lOHexavalent Chromium Method Detect!on Limits 2-22
2-9 2010 Sampling Schedules and Completeness Rates 2-24
2-10 Method Completeness Rates for 2010 2-29
3-1 Overview and Lay out of Data Presented 3-1
3-2 NATTS MQO Core Analytes 3-6
3-3 NATTS MQO Core Analytes Selected for Comparative Analysis 3-15
3-4 NATTS MQO Core Analytes Selected for Trends Analysis 3-16
3-5 POM Groups for PAHs 3-20
4-1 Statistical Summaries of the VOC Concentrations 4-3
4-2 Statistical Summaries of the SNMOC Concentrations 4-6
4-3 Statistical Summaries of the Carbonyl Compound Concentrations 4-9
4-4 Statistical Summaries of the PAH Concentrations 4-10
4-5 Statistical Summaries of the Metals Concentrations 4-11
4-6 Statistical Summaries of the Hexavalent Chromium Concentrations 4-12
4-7 Program-Level Risk Screening Summary 4-16
4-8 Site-Specific Risk Screening Comparison 4-19
4-9 Annual Average Concentration Comparison of the VOC Pollutants of Interest 4-22
4-10 Annual Average Concentration Comparison of the Carbonyl Compound Pollutants
of Interest 4-24
4-11 Annual Average Concentration Comparison of the PAH Pollutants of Interest 4-25
4-12 Annual Average Concentration Comparison of the Metals Pollutants of Interest 4-26
4-13 Program-Level MRL Risk Screening Assessment 4-31
4-14 Summary of Mobile Source Information by Monitoring Site 4-33
4-15 Greenhouse Gases Measured by Method TO-15 4-93
5-1 Geographical Information for the Arizona Monitoring Sites 5-5
5-2 Population, Motor Vehicle, and Traffic Information for the Arizona Monitoring
Sites 5-7
5-3 Average Meteorological Conditions near the Arizona Monitoring Sites 5-9
5-4 Risk Screening Results for the Arizona Monitoring Sites 5-17
5-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Arizona Monitoring Sites 5-20
xxxiv
-------
LIST OF TABLES (Continued)
Page
5-6 Cancer and Noncancer Surrogate Risk Approximations for the Arizona
Monitoring Sites 5-30
5-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Arizona Monitoring Sites 5-33
5-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Arizona Monitoring
Sites 5-34
6-1 Geographical Information for the California Monitoring Sites 6-8
6-2 Population, Motor Vehicle, and Traffic Information for the California Monitoring
Sites 6-10
6-3 Average Meteorological Conditions near the California Monitoring Sites 6-13
6-4 Risk Screening Results for the California Monitoring Sites 6-24
6-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
California Monitoring Sites 6-26
6-6 Cancer and Noncancer Surrogate Risk Approximations for the California Monitoring
Sites 6-32
6-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the California Monitoring Sites 6-34
6-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the California Monitoring
Sites 6-36
7-1 Geographical Information for the Colorado Monitoring Sites 7-10
7-2 Population, Motor Vehicle, and Traffic Information for the Colorado Monitoring
Sites 7-13
7-3 Average Meteorological Conditions near the Colorado Monitoring Sites 7-15
7-4 Risk Screening Results for the Colorado Monitoring Sites 7-32
7-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Colorado Monitoring Sites 7-35
7-6 Noncancer Risk Screening Summary for the Colorado Monitoring Sites 7-52
7-7 Cancer and Noncancer Surrogate Risk Approximations for the Colorado
Monitoring Sites 7-56
7-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Colorado Monitoring Sites 7-59
7-9 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Colorado Monitoring
Sites 7-62
8-1 Geographical Information for the Washington, D.C. Monitoring Site 8-4
8-2 Population, Motor Vehicle, and Traffic Information for the Washington, D.C.
Monitoring Site 8-5
8-3 Average Meteorological Conditions near the Washington, D.C. Monitoring Site 8-7
8-4 Risk Screening Results for the Washington, D.C. Monitoring Site 8-12
xxxv
-------
LIST OF TABLES (Continued)
Page
8-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Washington, D.C. Monitoring Site 8-14
8-6 Cancer and Noncancer Surrogate Risk Approximations for the Washington, D.C.
Monitoring Site 8-19
8-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Washington, D.C. Monitoring Site 8-21
8-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Washington, D.C.
Monitoring Site 8-22
9-1 Geographical Information for the Florida Monitoring Sites 9-9
9-2 Population, Motor Vehicle, and Traffic Information for the Florida Monitoring
Sites 9-12
9-3 Average Meteorological Conditions near the Florida Monitoring Sites 9-14
9-4 Risk Screening Results for the Florida Monitoring Sites 9-31
9-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Florida Monitoring Sites 9-33
9-6 Cancer and Noncancer Surrogate Risk Approximations for the Florida
Monitoring Sites 9-51
9-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Florida Monitoring Sites 9-53
9-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Florida Monitoring
Sites 9-56
10-1 Geographical Information for the Georgia Monitoring Site 10-4
10-2 Population, Motor Vehicle, and Traffic Information for the Georgia Monitoring
Site 10-5
10-3 Average Meteorological Conditions near the Georgia Monitoring Site 10-8
10-4 Risk Screening Results for the Georgia Monitoring Site 10-12
10-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Georgia Monitoring Site 10-14
10-6 Cancer and Noncancer Surrogate Risk Approximations for the Georgia
Monitoring Site 10-19
10-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Georgia Monitoring Site 10-20
10-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Georgia Monitoring
Site 10-21
11-1 Geographical Information for the Illinois Monitoring Sites 11-5
11-2 Population, Motor Vehicle, and Traffic Information for the Illinois Monitoring
Sites 11-7
11-3 Average Meteorological Conditions near the Illinois Monitoring Sites 11-9
11-4 Risk Screening Results for the Illinois Monitoring Sites 11-17
xxxvi
-------
LIST OF TABLES (Continued)
Page
11-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Illinois Monitoring Sites 11-19
11-6 Noncancer Risk Screening Summary for the Illinois Monitoring Sites 11-41
11-7 Cancer and Noncancer Surrogate Risk Approximations for the Illinois Monitoring
Sites 11-44
11-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Illinois Monitoring Sites 11 -47
11-9 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Illinois Monitoring
Sites 11-48
12-1 Geographical Information for the Indiana Monitoring Sites 12-6
12-2 Population, Motor Vehicle, and Traffic Information for the Indiana Monitoring
Sites 12-8
12-3 Average Meteorological Conditions near the Indiana Monitoring Sites 12-10
12-4 Risk Screening Results for the Indiana Monitoring Sites 12-18
12-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Indiana Monitoring Sites 12-20
12-6 Cancer and Noncancer Surrogate Risk Approximations for the Indiana
Monitoring Sites 12-26
12-7 Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Indiana Monitoring Sites 12-27
12-8 Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Indiana Monitoring
Sites 12-28
13-1 Geographical Information for the Kentucky Monitoring Site 13-4
13-2 Population, Motor Vehicle, and Traffic Information for the Kentucky Monitoring
Site 13-5
13-3 Average Meteorological Conditions near the Kentucky Monitoring Site 13-7
13-4 Risk Screening Results for the Kentucky Monitoring Site 13-12
13-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Kentucky Monitoring Site 13-14
13-6 Cancer and Noncancer Surrogate Risk Approximations for the Kentucky
Monitoring Site 13-18
13-7 Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Kentucky Monitoring Site 13-20
13-8 Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Kentucky Monitoring
Site 13-21
14-1 Geographical Information for the Massachusetts Monitoring Site 14-4
14-2 Population, Motor Vehicle, and Traffic Information for the Massachusetts
Monitoring Site 14-5
14-3 Average Meteorological Conditions near the Massachusetts Monitoring Site 14-7
xxxvii
-------
LIST OF TABLES (Continued)
Page
14-4 Risk Screening Results for the Massachusetts Monitoring Site 14-12
14-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Massachusetts Monitoring Site 14-14
14-6 Cancer and Noncancer Surrogate Risk Approximations for the Massachusetts
Monitoring Site 14-22
14-7 Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Massachusetts Monitoring Site 14-23
14-8 Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Massachusetts
Monitoring Site 14-24
15-1 Geographical Information for the Michigan Monitoring Site 15-4
15-2 Population, Motor Vehicle, and Traffic Information for the Michigan Monitoring
Site 15-5
15-3 Average Meteorological Conditions near the Michigan Monitoring Site 15-7
15-4 Risk Screening Results for the Michigan Monitoring Site 15-12
15-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Michigan Monitoring Site 15-14
15-6 Cancer and Noncancer Surrogate Risk Approximations for the Michigan
Monitoring Site 15-24
15-7 Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Michigan Monitoring Site 15-26
15-8 Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Michigan Monitoring
Site 15-27
16-1 Geographical Information for the Missouri Monitoring Site 16-4
16-2 Population, Motor Vehicle, and Traffic Information for the Missouri Monitoring
Site 16-5
16-3 Average Meteorological Conditions near the Missouri Monitoring Site 16-7
16-4 Risk Screening Results for the Missouri Monitoring Site 16-13
16-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Missouri Monitoring Site 16-14
16-6 Cancer and Noncancer Surrogate Risk Approximations for the Missouri Monitoring
Site 16-30
16-7 Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Missouri Monitoring Site 16-33
16-8 Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Missouri Monitoring
Site 16-34
17-1 Geographical Information for the New Jersey Monitoring Sites 17-9
17-2 Population, Motor Vehicle, and Traffic Information for the New Jersey Monitoring
Sites 17-12
17-3 Average Meteorological Conditions near the New Jersey Monitoring Sites 17-14
xxxviii
-------
LIST OF TABLES (Continued)
Page
17-4 Risk Screening Results for the New Jersey Monitoring Sites 17-27
17-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
New Jersey Monitoring Sites 17-30
17-6 Cancer and Noncancer Surrogate Risk Approximations for the New Jersey
Monitoring Sites 17-50
17-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the New Jersey Monitoring Sites 17-53
17-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the New Jersey Monitoring
Sites 17-55
18-1 Geographical Information for the New York Monitoring Sites 18-9
18-2 Population, Motor Vehicle, and Traffic Information for the New York Monitoring
Sites 18-12
18-3 Average Meteorological Conditions near the New York Monitoring Sites 18-15
18-4 Risk Screening Results for the New York Monitoring Sites 18-29
18-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
New York Monitoring Sites 18-31
18-6 Cancer and Noncancer Surrogate Risk Approximations for the New York
Monitoring Sites 18-38
18-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the New York Monitoring Sites 18-40
18-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the New York Monitoring
Sites 18-42
19-1 Geographical Information for the Oklahoma Monitoring Sites 19-10
19-2 Population, Motor Vehicle, and Traffic Information for the Oklahoma Monitoring
Sites 19-13
19-3 Average Meteorological Conditions near the Oklahoma Monitoring Sites 19-15
19-4 Risk Screening Results for the Oklahoma Monitoring Sites 19-29
19-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Oklahoma Monitoring Sites 19-34
19-6 Cancer and Noncancer Surrogate Risk Approximations for the Oklahoma
Monitoring Sites 19-53
19-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Oklahoma Monitoring Sites 19-59
19-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Oklahoma Monitoring
Sites 19-62
20-1 Geographical Information for the Rhode Island Monitoring Site 20-4
20-2 Population, Motor Vehicle, and Traffic Information for the Rhode Island Monitoring
Site 20-5
20-3 Average Meteorological Conditions near the Rhode Island Monitoring Site 20-7
xxxix
-------
LIST OF TABLES (Continued)
Page
20-4 Risk Screening Results for the Rhode Island Monitoring Site 20-12
20-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Rhode Island Monitoring Site 20-14
20-6 Cancer and Noncancer Surrogate Risk Approximations for the Rhode Island
Monitoring Site 20-19
20-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Rhode Island Monitoring Site 20-21
20-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Rhode Island
Monitoring Site 20-22
21-1 Geographical Information for the South Carolina Monitoring Site 21-4
21-2 Population, Motor Vehicle, and Traffic Information for the South Carolina
Monitoring Site 21-5
21-3 Average Meteorological Conditions near the South Carolina Monitoring Site 21-7
21-4 Risk Screening Results for the South Carolina Monitoring Site 21-12
21-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
South Carolina Monitoring Site 21-14
21-6 Cancer and Noncancer Surrogate Risk Approximations for the South Carolina
Monitoring Site 21-18
21-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the South Carolina Monitoring Site 21 -20
21-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the South Carolina
Monitoring Site 21-21
22-1 Geographical Information for the South Dakota Monitoring Sites 22-6
22-2 Population, Motor Vehicle, and Traffic Information for the South Dakota
Monitoring Sites 22-8
22-3 Average Meteorological Conditions near the South Dakota Monitoring Sites 22-10
22-4 Risk Screening Results for the South Dakota Monitoring Sites 22-19
22-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
South Dakota Monitoring Sites 22-20
22-6 Cancer and Noncancer Surrogate Risk Approximations for the South Dakota
Monitoring Sites 22-28
22-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the South Dakota Monitoring Sites 22-30
22-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the South Dakota
Monitoring Sites 22-31
23-1 Geographical Information for the Texas Monitoring Sites 23-6
23-2 Population, Motor Vehicle, and Traffic Information for the Texas Monitoring Sites... 23-8
23-3 Average Meteorological Conditions near the Texas Monitoring Sites 23-10
23-4 Risk Screening Results for the Texas Monitoring Sites 23-19
xl
-------
LIST OF TABLES (Continued)
Page
23-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Texas Monitoring Sites 23-20
23-6 Cancer and Noncancer Surrogate Risk Approximations for the Texas Monitoring
Sites 23-25
23-7 Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Texas Monitoring Sites 23-26
23-8 Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Texas Monitoring
Sites 23-27
24-1 Geographical Information for the Utah Monitoring Site 24-4
24-2 Population, Motor Vehicle, and Traffic Information for the Utah Monitoring Site 24-5
24-3 Average Meteorological Conditions near the Utah Monitoring Site 24-8
24-4 Risk Screening Results for the Utah Monitoring Site 24-13
24-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Utah Monitoring Site 24-14
24-6 Noncancer Risk Screening Summary for the Utah Monitoring Site 24-30
24-7 Cancer and Noncancer Surrogate Risk Approximations for the Utah Monitoring
Site 24-33
24-8 Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Utah Monitoring Site 24-35
24-9 Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Utah Monitoring
Site 24-36
25-1 Geographical Information for the Vermont Monitoring Sites 25-7
25-2 Population, Motor Vehicle, and Traffic Information for the Vermont
Monitoring Sites 25-9
25-3 Average Meteorological Conditions near the Vermont Monitoring Sites 25-11
25-4 Risk Screening Results for the Vermont Monitoring Sites 25-23
25-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Vermont Monitoring Sites 25-25
25-6 Cancer and Noncancer Surrogate Risk Approximations for the Vermont
Monitoring Sites 25-36
25-7 Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Vermont Monitoring Sites 25-40
25-8 Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Vermont
Monitoring Sites 25-42
26-1 Geographical Information for the Virginia Monitoring Site 26-4
26-2 Population, Motor Vehicle, and Traffic Information for the Virginia Monitoring
Site 26-5
26-3 Average Meteorological Conditions near the Virginia Monitoring Site 26-7
26-4 Risk Screening Results for the Virginia Monitoring Site 26-12
xli
-------
LIST OF TABLES (Continued)
Page
26-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Virginia Monitoring Site 26-14
26-6 Cancer and Noncancer Surrogate Risk Approximations for the Virginia Monitoring
Site 26-17
26-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Virginia Monitoring Site 26-19
26-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Virginia Monitoring
Site 26-20
27-1 Geographical Information for the Washington Monitoring Site 27-4
27-2 Population, Motor Vehicle, and Traffic Information for the Washington
Monitoring Site 27-5
27-3 Average Meteorological Conditions near the Washington Monitoring Site 27-7
27-4 Risk Screening Results for the Washington Monitoring Site 27-12
27-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Washington Monitoring Site 27-14
27-6 Cancer and Noncancer Surrogate Risk Approximations for the Washington
Monitoring Site 27-23
27-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Washington Monitoring Site 27-25
27-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Washington
Monitoring Site 27-26
28-1 Geographical Information for the Wisconsin Monitoring Site 28-4
28-2 Population, Motor Vehicle, and Traffic Information for the Wisconsin
Monitoring Site 28-5
28-3 Average Meteorological Conditions near the Wisconsin Monitoring Site 28-8
28-4 Risk Screening Results for the Wisconsin Monitoring Site 28-13
28-5 Quarterly and Annual Average Concentrations of the Pollutants of Interest for the
Wisconsin Monitoring Site 28-14
28-6 Cancer and Noncancer Surrogate Risk Approximations for the Wisconsin
Monitoring Site 28-17
28-7 Top 10 Emissions, Toxi city-Weighted Emissions, and Cancer Risk Approximations
for Pollutants with Cancer UREs for the Wisconsin Monitoring Site 28-18
28-8 Top 10 Emissions, Toxi city-Weighted Emissions, and Noncancer Risk
Approximations for Pollutants with Noncancer RfCs for the Wisconsin
Monitoring Site 28-19
29-1 Method Precision by Analytical Method 29-3
29-2 VOC Method Precision: Average Coefficient of Variation Based on Duplicate and
Collocated Samples by Site 29-4
29-3 SNMOC Method Precision: Average Coefficient of Variation Based on Duplicate
and Collocated Samples by Site 29-11
xlii
-------
LIST OF TABLES (Continued)
Page
29-4 Carbonyl Compound Method Precision: Average Coefficient of Variation Based on
Duplicate and Collocated Samples by Site 29-13
29-5 Metals Method Precision: Average Coefficient of Variation Based on Collocated
Samples by Site 29-16
29-6 Hexavalent Chromium Method Precision: Average Coefficient of Variation Based
on Collocated Samples by Site 29-17
29-7 PAH Method Precision: Average Coefficient of Variation Based on Collocated
Samples by Site 29-18
29-8 Analytical Precision by Analytical Method 29-19
29-9 VOC Analytical Precision: Average Coefficient of Variation based on Replicate
Analyses by Site 29-20
29-10 SNMOC Analytical Precision: Average Coefficient of Variation Based on Replicate
Analyses by Site 29-28
29-11 Carbonyl Compound Analytical Precision: Average Coefficient of Variation Based
on Replicate Analyses by Site 29-31
29-12 Metals Analytical Precision: Average Coefficient of Variation Based on Replicate
Analyses by Site 29-33
29-13 Hexavalent Chromium Analytical Precision: Average Coefficient of Variation Based
on Replicate Analyses by Site 29-34
29-14 PAH Analytical Precision: Average Coefficient of Variation Based on Replicate
Analyses by Site 29-35
29-15 VOC NATTS PT Audit Samples-Percent Difference from True Value 29-36
29-16 Carbonyl Compound NATTS PT Audit Samples-Percent Difference from
True Value 29-36
29-17 Metals NATTS PT Audit Samples-Percent Difference from True Value 29-36
29-18 Hexavalent Chromium PT Audit Samples-Percent Difference from True Value 29-37
29-19 PAH NATTS PT Audit Samples-Percent Difference from True Value 29-37
xliii
-------
LIST OF ACRONYMS
AADT Average annual daily traffic
AGL Above ground level
AIRS Aerometric Information and Retrieval System
AQS Air Quality System (of the Aerometric Information and Retrieval System)
ASE Accelerated Solvent Extractor
ATSDR Agency for Toxic Substances and Disease Registry
CBS A Core-based statistical area(s)
CFR Code of Federal Regulations
CNG Compressed Natural Gas
CSATAM Community-Scale Air Toxics Ambient Monitoring
CV Coefficient of variation
DNPH 2,4-dinitrophenylhydrazine
DQO Data Quality Objective(s)
EPA U.S. Environmental Protection Agency
ERG Eastern Research Group, Inc.
F Fahrenheit
FHWA Federal Highway Administration
GC/MS-FID Gas chromatography/mass spectrometry and flame ionization detection
GHG Greenhouse gas(es)
GIS Geographical Information System
GWP Global Warming Potential
HAP Hazardous Air Pollutant(s)
HPLC High-performance liquid chromatography
HQ Hazard Quotient
HYSPLIT Hybrid Single-Particle Lagrangian Integrated Trajectory
1C Ion Chromatography
ICP-MS Inductively coupled plasma/mass spectrometry
IPCC Intergovernmental Panel on Climate Change
kt Knots
mb Millibar
MDL Method Detection Limit
mg/m3 Milligrams per cubic meter
mL Milliliter
MQO Method Quality Objective(s)
MRL Minimal risk level
MSA Metropolitan or Micropolitan Statistical Area(s)
MTBE Methyl tert-buty\ ether
NATA National Air Toxics Assessment
xliv
-------
LIST OF ACRONYMS (Continued)
NATTS National Air Toxics Trends Site
ND Non-detect
NEI National Emissions Inventory
ng/m3 Nanograms per cubic meter
NMOC Non-Methane Organic Compound(s)
NMP National Monitoring Programs
NOAA National Oceanic and Atmospheric Administration
NOx Oxides of Nitrogen
NWS National Weather Service
PAMS Photochemical Assessment Monitoring Stations
PAH Polycyclic Aromatic Hydrocarbons
PM Particulate Matter
PMio Particulate matter less than 10 microns
POM Polycyclic Organic Matter
ppbC Parts per billion carbon
ppbv Parts per billion by volume
ppm Parts per million
PT Proficiency Test
PUF Polyurethane foam
QAPP Quality Assurance Project Plan
RfC Reference Concentration(s)
RFG Reformulated gasoline
SATMP School Air Toxics Monitoring Program
SIM Selected ion monitoring
SIP State Implementation Plan(s)
SNMOC Speciated Nonmethane Organic Compound(s)
UATMP Urban Air Toxics Monitoring Program
VOC Volatile Organic Compound(s)
TAD Technical Assistance Document
TNMOC Total Nonmethane Organic Compound(s)
tpy Tons per year
TRI Toxics Release Inventory
TSP Total Suspended Particulate
TSV Total spatial variance
|ig/m3 Micrograms per cubic meter
jiL Microliter
URE Unit Risk Estimate(s)
VMT Vehicle miles traveled
WBAN Weather Bureau/Army/Navy ID
xlv
-------
Abstract
This report presents the results and conclusions from the ambient air monitoring conducted
as part of the 2010 National Monitoring Programs (NATTS, UATMP, and CSATAM) - three
individual programs with different goals, but together result in a better understanding and
appreciation of the nature and extent of toxic air pollution. The 2010 NMP includes data from
samples collected at 52 monitoring sites that collected 24-hour air samples, typically on a l-in-6
or l-in-12 day schedule. Twenty-four sites sampled for 61 volatile organic compounds (VOC); 30
sites sampled for 14 carbonyl compounds; nine sites sampled for 80 speciated nonmethane
organic compounds (SNMOC); 26 sites sampled for 22 polycyclic aromatic hydrocarbons (PAH);
14 sites sampled for 11 metals; and 23 sites sampled for hexavalent chromium. Over 214,900
ambient air concentrations were measured during the 2010 NMP. This report uses various
graphical, numerical, and statistical analyses to put the vast amount of ambient air monitoring
data collected into perspective. Not surprisingly, the ambient air concentrations measured during
the program varied significantly from city-to-city and from season-to-season.
The ambient air monitoring data collected during the 2010 NMP serve a wide range of
purposes. Not only do these data characterize the nature and extent of air pollution close to the
52 individual monitoring sites participating in these programs, but they also identify trends and
patterns that may be common to both urban and rural environments, and across the country.
Therefore, this report presents results that are specific to particular monitoring locations and
presents other results that are common to all environments. The results presented provide
additional insight into the complex nature of air pollution. The raw data are included in the
appendices of this report.
xlvi
-------
1.0 Introduction
Air pollution contains many components that originate from a wide range of stationary,
mobile, and natural emissions sources. Because some of these components include air toxics that
are known or suspected to have the potential for negative human health impacts, the
U.S. Environmental Protection Agency (EPA) encourages state, local, and tribal agencies to
understand and appreciate the nature and extent of toxic air pollution in their respective
locations. To achieve this goal, EPA sponsors the National Monitoring Programs (NMP), which
include the Photochemical Assessment Monitoring Stations (PAMS) network, Urban Air Toxics
Monitoring Program (UATMP), National Air Toxics Trends Stations (NATTS) network,
Community-Scale Air Toxics Ambient Monitoring (CSATAM) Program, and monitoring for
other pollutants such as Non-Methane Organic Compounds (NMOC). This report focuses on
monitoring sites participating in the UATMP, NATTS, and CSATAM programs. These
programs have the following program-specific objectives:
• The primary purpose of the UATMP is to characterize the composition and
magnitude of air toxics pollution through ambient air monitoring.
• The primary purpose of the CSATAM program is to conduct local-scale investigative
air toxics monitoring projects.
• The primary goal of the NATTS network is to obtain a statistically significant
quantity of high-quality representative air toxics measurements such that long-term
trends can be identified.
1.1 Background
EPA began the NMOC program in 1984. Monitoring for selected NMOC was performed
during the morning hours of the summer ozone season. NMOC data were to be used to better
understand ozone formation and to develop ozone control strategies. The UATMP was initiated
by EPA in 1988 as an extension of the existing NMOC program to meet the increasing need for
information on air toxics. Over the years, the program has grown in both participation and
targeted pollutants (EPA, 2009a). The program has allowed for the identification of compounds
that are prevalent in ambient air and for participating agencies to screen air samples for
concentrations of air toxics that could potentially result in adverse human health effects.
The NATTS network was created to generate long-term ambient air toxics concentration
data at specific fixed sites across the country. The NATTS Pilot program was developed and
implemented during 2001 and 2002, leading to the development and initial implementation of
1-1
-------
the NATTS network during 2003 and 2004. The goal of the program is to estimate the
concentrations of air toxics on a national level at fixed sites that remain active over an extended
period of time (EPA, 2009a). The generation of large quantities of high-quality data over an
extended period may allow concentration trends (i.e., any substantial increase or decrease over a
period of time) to be identified. The data generated are also used for validating modeling results
and emissions inventories, assessing current regulatory benchmarks, and assessing the potential
for developing cancerous and noncancerous health effects (EPA, 2012a). The initial site
locations were based on results from preliminary air toxics pilot programs such as the 1996
National Air Toxics Assessment (NATA), which used air toxics emissions data to model
ambient monitoring concentrations across the nation. Monitoring sites were placed in both urban
and rural locations. Urban areas were chosen to measure population exposure, while rural areas
were chosen to determine background levels of air pollution (EPA, 2009b). Currently, 27
NATTS sites are strategically placed across the country (EPA, 2010a).
The CSATAM Program was initiated in 2004 and is intended to support state, local, and
tribal agencies in conducting discreet, investigative projects of approximately 2-year durations
via periodic grant competitions (EPA, 2009a). The objectives of the CSATAM Program include
identifying and profiling air toxics sources; developing and assessing emerging measurement
methods; characterizing the degree and extent of local air toxics problems; and tracking progress
of air toxics reduction activities (EPA, 2009a).
Many environmental and health agencies have participated in these programs to assess
the sources, effects, and changes in air pollution within their jurisdictions. In past reports,
measurements from NATTS, UATMP, and CSATAM monitoring sites have been presented
together and referred to as "UATMP sites." In more recent reports, a distinction is made among
the three programs due to the increasing number of sites covered under each program. As such, it
is appropriate to describe each program; to distinguish among their purposes and scopes; and to
integrate the data, which allows each program's objectives and goals to complement each other.
1.2 The Report
This report summarizes and interprets the 2010 NATTS, UATMP, and CSATAM
monitoring efforts of the NMP. Data collected at 52 sites around the country are included in this
1-2
-------
report. The operating agencies of these 52 sites have opted to have their samples analyzed by
EPA's contract laboratory, Eastern Research Group, Inc. Agencies operating sites under the
NMP are not required to have their samples analyzed by ERG or may not have samples for all
methods analyzed by ERG, as they may have their own laboratories or use other contract
laboratories. In these cases, data are generated by sources other than ERG and are not included in
this report. The 52 sites included in this report are located in or near 34 urban or rural locations
in 23 states and the District of Columbia, including 33 metropolitan or micropolitan statistical
areas (MSA). Much of the data analysis and interpretation contained in this report focuses on
pollutant-specific risk potential.
This report provides both a qualitative overview of air toxics pollution at selected urban
and rural locations and a quantitative data analysis of the factors that appear to affect the
behavior of air toxics in urban and rural areas most significantly. This report also focuses on data
characterizations for each of the 52 different air sampling locations, a site-specific approach that
allows for a much more detailed evaluation of the factors (e.g., emissions sources, natural
sources, meteorological influences) that affect air quality differently from one location to the
next.
This report offers participating agencies useful insight into important air quality issues.
For example, participating agencies can use trends and patterns in the monitoring data to
determine whether levels of air pollution present public health concerns, to identify which
emissions sources contribute most to air pollution, or to forecast whether proposed pollution
control initiatives might significantly improve air quality. Monitoring data may also be
compared to modeling results, such as from EPA's NATA.
Policy-relevant questions that the monitoring data may help answer include the
following:
• Which anthropogenic sources substantially affect air quality?
• Have pollutant concentrations decreased as a result of regulations?
• Which pollutants contribute the greatest health risk on a short-term, intermediate-
term, and long-term basis?
1-3
-------
The data analyses contained in this report are applied to every participating NATTS,
UATMP, or C SAT AM monitoring site, depending upon pollutants sampled and duration of
sampling. Although many types of analyses are presented, state and local environmental
agencies are encouraged to perform additional evaluations of the monitoring data so that the
many factors that affect their specific ambient air quality can be understood fully.
To facilitate examination of the 2010 NATTS, UATMP, and CSATAM monitoring data,
henceforth referred to as NMP data, the complete set of measured concentrations is presented in
the appendices of this report. In addition, these data are publicly available in electronic format
from the Air Quality System (AQS) of EPA's Aerometric Information Retrieval System (AIRS)
at http ://www. epa. gov/ttn/airs/airsaq s/.
This report is organized into 31 sections and 17 appendices. While each state section is
designed to be a stand-alone section to allow those interested in a particular site or state to
understand the data analyses without having to read the entire report, it is recommended that
Sections 1 through 4 (Introduction, Monitoring Programs Network Overview, Data Treatments
and Methods, and Summary of Results) and Sections 29 and 30 (Data Quality and Summary of
Results and Recommendations) be read as complements to the individual state sections.
Table 1-1 highlights the contents of each section.
Table 1-1. Organization of the 2010 National Monitoring Programs Report
Report
Section
Section Title
Overview of Contents
Introduction
This section serves as an introduction to the
background and scope of the National Monitoring
Programs (specifically, the NATTS, UATMP, and
CSATAM).
The 2010 National Monitoring
Programs Network
This section provides information on the 2010 National
Monitoring Programs and network:
• Monitoring locations
• Pollutants selected for monitoring
• Sampling and analytical methods
• Sampling schedules
• Completeness of the air monitoring programs.
1-4
-------
Table 1-1. Organization of the 2010 National Monitoring Programs Report (Continued)
Report
Section
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Section Title
Summary of the 2010 National
Monitoring Programs Data
Treatments and Methods
Summary of the 2010 National
Monitoring Programs Results
Sites in Arizona
Sites in California
Sites in Colorado
Site in the District of Columbia
Sites in Florida
Site in Georgia
Sites in Illinois
Sites in Indiana
Site in Kentucky
Site in Massachusetts
Site in Michigan
Site in Missouri
Sites in New Jersey
Overview of Contents
This section presents and discusses the data treatments
used on the 2010 National Monitoring Programs data
to determine significant trends and relationships in the
data; characterize data based on how ambient air
concentrations varied with monitoring location and
with time; interpret the significance of the observed
spatial and temporal variations; and evaluate risk.
This section presents and discusses the results of the
data treatments from the 2010 National Monitoring
Programs data.
Monitoring results for the sites in the Phoenix-Mesa-
Glendale, AZ MSA (PXSS and SPAZ)
Monitoring results for the sites in the Los Angeles-
Long Beach-Santa Ana, CA MSA (CELA), Riverside-
San Bernardino-Ontario, CA MSA (RUCA), and San
Jose-Sunnyvale-Santa Clara, CA MSA (SJJCA)
Monitoring results for the sites in the Grand Junction,
CO MSA (GPCO) and Garfield County (BMCO,
BRCO, PACO, RICO, and RUCO)
Monitoring results for the site in the Washington-
Arlington-Alexandria, DC-VA-MD-WV MSA
(WADC)
Monitoring results for the sites in the Orlando-
Kissimmee-Sanford, FL MSA (ORFL and PAFL) and
Tampa-St. Petersburg-Clearwater, FL MSA (AZFL,
SKFL, and SYFL)
Monitoring results for the site in the Atlanta-Sandy
Springs-Marietta, GA MSA (SDGA)
Monitoring results for the sites in the Chicago- Joliet-
Naperville, IL-IN-WI MSA (NBIL and SPIL)
Monitoring results for the sites in the Chicago-Joliet-
Naperville, IL-IN-WI MSA (INDEM) and
Indianapolis-Carmel, IN MSA (WPIN)
Monitoring results for the site in Gray son, KY
(GLKY)
Monitoring results for the site in the Boston-
Cambridge-Quincy, MA-NH MSA (BOMA)
Monitoring results for the site in the Detroit-Warren-
Livonia, MI MSA (DEMI)
Monitoring results for the site in the St. Louis, MO-IL
MSA (S4MO)
Monitoring results for the sites in the New York-
Northern New Jersey-Long Island, NY-NJ-PA MSA
(CHNJ, ELNJ, NBNJ, and PANJ)
1-5
-------
Table 1-1. Organization of the 2010 National Monitoring Programs Report (Continued)
Report
Section
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Section Title
Sites in New York
Sites in Oklahoma
Site in Rhode Island
Site in South Carolina
Sites in South Dakota
Sites in Texas
Site in Utah
Sites in Vermont
Site in Virginia
Site in Washington
Site in Wisconsin
Data Quality
Summary of Results and
Recommendations
References
Overview of Contents
Monitoring results for the sites in the New York-
Northern New Jersey-Long Island, NY-NJ-PA MSA
(BXNY and MONY), Rochester, NY MSA (ROCH),
and Buffalo-Niagara Falls, NY MSA (TONY)
Monitoring results for the sites in the Tulsa, OK MSA
(TOOK and TMOK), Oklahoma City, OK MSA
(MWOK and OCOK), and Pryor Creek, OK (PROK)
Monitoring results for the site in the Providence-New
Bedford-Fall River, RI-MA MSA (PRRI)
Monitoring results for the site in Chesterfield, SC
(CHSC)
Monitoring results for the sites in the Sioux City, IA-
NE-SD MSA (UCSD) and the Sioux Falls, SD MSA
(SSSD)
Monitoring results for the sites in the Houston-Sugar
Land-Baytown, TX MSA (CAMS 35) and the
Marshall, TX MSA (CAMS 85)
Monitoring results for the site in the Ogden-Clearfield,
UT MSA (BTUT)
Monitoring results for the sites in the Burlington-South
Burlington, VT MSA (BURVT and UNVT) and the
Rutland, VT MSA (RUVT)
Monitoring results for the site in the Richmond, VA
MSA (RIVA)
Monitoring results for the site in the Seattle-Tacoma-
Bellevue, WA MSA (SEWA)
Monitoring results for the site in the Beaver Dam, WI
MSA (HOWI)
This section defines and discusses the concepts of
precision and accuracy. Based on quantitative and
qualitative analyses, this section comments on the
precision and accuracy of the 2010 National
Monitoring Programs ambient air monitoring data.
This section summarizes the most significant findings
of the report and makes several recommendations for
future projects that involve ambient air monitoring.
This section lists the references cited throughout the
report.
1-6
-------
2.0 The 2010 National Monitoring Programs Network
Agencies operating NATTS, UATMP, or CSATAM sites may choose to have their
samples analyzed by EPA's contract laboratory, Eastern Research Group, Inc. (ERG) in
Morrisville, NC. Data from 52 monitoring sites that collected 24-hour integrated ambient air
samples for up to 12 months, at l-in-6 or l-in-12 day sampling intervals, and sent them to ERG
for analysis are included in this report. Samples were analyzed for concentrations of selected
hydrocarbons, halogenated hydrocarbons, and polar compounds from canister samples
(Speciated Nonmethane Organic Compounds
(SNMOC) and/or Method TO-15), carbonyl
, r. u.^-j i/A/r.ij not required to have their samples
compounds trom sorbent cartridge samples (Method , ,. __ _, „. .
r ° r \ analyzed by ERG. They may have
TO-11 A), poly cyclic aromatic hydrocarbons (PAH)
from polyurethane foam (PUF) and XAD-2® resin , . . , , ^ . _
have their own laboratories. In
samples (Method TO-13 A), hexavalent chromium from
Agencies operating these sites are
samples for only select methods
analyzed by ERG, as they may
these cases, data are generated by
.. ,. , , r-i /T-™ A 1 sources other than ERG and are
sodium bicarbonate-coated filters (EPA-approved . . , , ,. .,. .
not included in this report.
method), and trace metals from filters (Method IO-3.5).
Section 2.2 provides further details on each of the
sampling methodologies used to collect and analyze samples.
The following sections review the monitoring locations, pollutants selected for
monitoring, collection schedules, sampling and analytical methods, and completeness of the
2010NMPdataset.
2.1 Monitoring Locations
For the NATTS Program, monitor siting is based on the need to assess population
exposure and background-level concentrations. For the UATMP and CSATAM programs,
representatives from the state, local, and tribal agencies that voluntarily participate in the
programs select the monitoring locations based on specific siting criteria and study needs.
Among these programs, monitors were placed in urban areas near the centers of heavily
populated cities (e.g., Chicago, IL and Phoenix, AZ), while others were placed in moderately
populated rural areas (e.g., Horicon, WI and Chesterfield, SC). Figure 2-1 shows the locations of
the 52 monitoring sites participating in the 2010 programs, which encompass 34 different urban
and rural areas. Outlined in Figure 2-1 are the associated core-based statistical areas (CBSA), as
2-1
-------
designated by the U.S. Census Bureau, where each site is located (Census Bureau, 2009). A
CBSA refers to either a metropolitan or micropolitan statistical area (Census Bureau, 2012).
Table 2-1 lists the respective monitoring program and the years of program participation
for the 52 monitoring sites. Forty-eight monitoring sites have been included in previous annual
reports, while four new sites began sampling in 2010.
As Figure 2-1 and Table 2-1 show, the 2010 NMP sites are widely distributed across the
country. Detailed information about the monitoring sites is provided in Table 2-2 and
Appendix A. Monitoring sites that are designated as part of the NATTS network are indicated by
bold italic type in Table 2-1 and subsequent tables throughout this report in order to distinguish
this program from the other two programs. Table 2-2 shows that the location types of the
monitoring sites vary significantly, based on elevation, population, land use, climatology, and
topography. A more detailed look at each monitoring site's surroundings is provided in the
individual state sections.
For record-keeping and reporting purposes, each site was assigned the following:
• A unique four- or five-letter site code used to track samples from the monitoring site
to the ERG laboratory.
• A unique nine-digit AQS site code used to index monitoring results in the AQS
database.
This report cites the four- or five-letter site code when presenting selected monitoring
results. For reference, each site's AQS site code is provided in Table 2-2.
2-2
-------
Figure 2-1. Locations of the 2010 National Monitoring Programs Monitoring Sites
Underbill, VT
Burlington, VT?f I \ r'
~lj W Rutland, VT
Rochester, NY TLJ^Boston, MA
Tonawanda, NY
Sioux Falls, SD
Union CountyTSD
Bountiful, UT
Co?
_ ParachuteTCO
San Jose, CA / Rulison, CO / Rifle, CO
Battlement Mesa, CO h^t SJH co
Northbrook,
Schiller Park,
Elizabeth, NJ
New Brunswick, NJ
Indianapolis, IN
Grand Junction, CO
Grayson, KY^ Richmond,',VA
Tulsa.pK
Oklahoma City, OKrO »^ Rryor Creek, OK
L-J-MidwestCity, OK
I
Karnack, TX t<
Los Angeles, CA
- A
Chesterfield, SC
Winter Park, FL
Orlando, FL
Pinellas Park, FL
St. Petersburg, FL
Legend
• CSATAM
o NATTS
• UATMP
Metropolitan/Micropolitan Statistical Area
-------
Table 2-1. 2010 National Monitoring Programs Sites and Past Program Participation
Monitoring Location
and Site
Battlement Mesa, CO (BMCO)
Boston, MA (BOMA)
Bountiful, UT (BTUT)
Burlington, VT (BURVT)
Chester, NJ (CHNJ)
Chesterfield, SC (CHSC)
Dearborn, MI (DEMI)
Decatur, GA (SDGA)
Deer Park, TX (CAMS 35)
Elizabeth, NJ (ELNJ)
Gary, IN (INDEM)
Grand Junction, CO (GPCO)
Grayson, KY (GLKY)
Horicon, WI (HOWI)
Indianapolis, IN (WPIN)
Karnack, TX (CAMS 85)
Los Angeles, CA (CELA)
Midwest City, OK (MWOK)
Program
UATMP
NATTS
NATTS
UATMP
UATMP
NATTS
NATTS
NATTS
NATTS
UATMP
UATMP
NATTS
NATTS
NATTS
UATMP
NATTS
NATTS
UATMP
2000 and Earlier
1999-2000
2001
S
S
^
2002
S
S
s
2003
S
s
s
s
s
2004
S
S
S
S
S
S
S
2005
S
S
s
s
s
s
s
s
s
2006
S
S
S
S
S
S
S
S
S
S
2007
S
S
S
S
S
S
S
S
S
S
S
S
S
2008
S
S
S
S
S
S
S
S
S
S
S
S
S
2009
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
2010
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
to
Yellow shading indicates a new site for year 2010;
BOLD ITALICS = EPA-designaled NATTS site.
green shading indicates past participation with a gap in sampling under the NMP.
-------
Table 2-1. 2010 National Monitoring Programs Sites and Past Program Participation (Continued)
Monitoring Location
and Site
New Brunswick, NJ (NBNJ)
New York, NY (BXNY)
New York, NY (MONY)
Northbrook, IL (NBIL)
Oklahoma City, OK (OCOK)
Orlando, FL (PAFL)
Parachute, CO (PACO)
Paterson, NJ (PANJ)
Phoenix, AZ (PXSS)
Phoenix, AZ (SPAZ)
Pinellas Park, FL (SKFL)
Plant City, FL (SYFL)
Providence, RI (PRRI)
Pryor Creek, OK (PROK)
Richmond, VA (SIVA)
Rifle, CO (RICO)
Rochester, NY (ROCH)
Rubidoux, CA (RUCA)
Program
UATMP
NATTS
NATTS
NATTS
UATMP
UATMP
UATMP
CSATAM
NATTS
UATMP
NATTS
NATTS
NATTS
UATMP
NATTS
UATMP
NATTS
NATTS
2000 and Earlier
2001
S
S
^
2002
S
S
2003
S
S
S
2004
S
S
S
S
S
2005
S
S
S
S
S
2006
S
S
S
S
S
S
S
S
2007
S
S
S
S
S
S
S
S
S
S
2008
S
S
S
S
S
S
S
S
S
S
S
s
s
s
s
2009
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
2010
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
to
Yellow shading indicates a new site
BOLD ITALICS = EPA-designaled
for year 2010; green shading indicates past participation with a gap in sampling under the NMP.
NATTS site.
-------
Table 2-1. 2010 National Monitoring Programs Sites and Past Program Participation (Continued)
Monitoring Location
and Site
Rulison, CO (RUCO)
Rutland, VT (RUVT)
San Jose, CA (SJJCA)
Schiller Park, IL (SPIL)
Seattle, WA (SEWA)
Silt, CO (BRCO)
Sioux Falls, SD (SSSD)
St. Louis, MO (S4MO)
St. Petersburg, FL (AZFL)
Tonawanda, NY (TONY)
Tulsa, OK (TMOK)
Tulsa, OK (TOOK)
Underbill, VT (UNVT)
Union County, SD (UCSD)
Washington, D.C. (WADC)
Winter Park, FL (ORFL)
Program
UATMP
UATMP
NATTS
UATMP
NATTS
UATMP
UATMP
NATTS
UATMP
CSATAM
UATMP
UATMP
NATTS
UATMP
NATTS
UATMP
2000 and Earlier
1995-1999
1991-1992
1990-1991
2001
S
2002
S
s
s
s
2003
s
s
s
s
2004
s
s
s
s
2005
s
s
s
s
s
s
s
2006
S
S
S
s
s
s
s
s
2007
S
S
S
S
S
S
S
S
2008
S
S
S
S
S
S
S
S
S
S
S
S
2009
S
S
s
s
s
s
s
s
s
s
s
s
s
s
s
s
2010
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
to
Yellow shading indicates a new site for year 2010;
BOLD ITALICS = EPA-designaled NATTS site.
green shading indicates past participation with a gap in sampling under the NMP.
-------
Table 2-2. Site Characterizing Information for 2010 National Monitoring Programs Sites
Site
Code
AZFL
BMCO
BOMA
BRCO
BTUT
BURVT
BXNY
CAMS
35
CAMS
85
CELA
CHNJ
AQS
Code
12-103-0018
NA
25-025-0042
08-045-0009
49-011-0004
50-007-0014
36-005-0110
48-201-1039
48-203-0002
06-037-1103
34-027-3001
Location
St. Petersburg, FL
Battlement Mesa, CO
Boston, MA
Silt, CO
Bountiful, UT
Burlington, VT
New York, NY
Deer Park, TX
Karnack TX
Los Angeles, CA
Chester NT
Land Use
Residential
Residential
Commercial
Agricultural
Residential
Commercial
Residential
Residential
Agricultural
Residential
Agricultural
Location
Setting
Suburban
Rural
Urban/City
Center
Rural
Suburban
Urban/City
Center
Urban/City
Center
Suburban
Rural
Urban/City
Center
Rural
Estimated
Daily Traffic,
AADTa
(Year)
41,500
(2010)
2,527
(2002)
31,400
(2007)
150
(2002)
113,955
(2010)
4,000
(2010)
100,230
(2008)
31,043
(2004)
1,400
(2010)
235,000
(2010)
12,917
(2010)
Population
Residing Within
10 Miles of the
Monitoring Siteb
554,850
5,941
1,670,959
24,174
259,066
116,261
6,590,357
715,640
3,034
3,679,965
244,577
County-level
Vehicle
Registration,
# of Vehicles
(Year)
879,317
(2010)
74,847
(2009)
501,587
(2010)
74,847
(2009)
239,754
(2010)
223,316
(2010)
248,600
(2010)
3,115,974
(2010)
69,883
o :oi m
7,410,625
(2010)
389,359
(2010)d
County-level
Stationary
Source HAP
Emissions
from the 2008
NEIC
(tpy)
1,381.26
1,364.26
572.38
1,364.26
391.90
347.53
2,171.17
9,322.29
593.11
14,794.19
198.46
County-level
Mobile
Source HAP
Emissions
from the 2008
NEIC
(tpy)
3,808.72
353.08
1,156.01
353.08
1,198.09
623.35
1,217.06
11,313.66
413.72
14,628.66
1,907.47
to
BOLD ITALICS = EPA-designated NATTS site.
aAADT is average annual daily traffic.
Reference: Xionetic, 2011.
c Reference: EPA, 2012b.
dThe proportion of county-level population to the state-level population was applied to state-level vehicle registration figure and used as a surrogate when county-
level vehicle registration counts were not available.
eGPCO's hexavalent chromium monitor is at a separate, but adjacent, location; as such, this site has two AQS codes.
f The 10-mile population estimate for BXNY was used as a surrogate for MONY.
NA = Data not loaded into AQS per agency request.
-------
Table 2-2. Site Characterizing Information for 2010 National Monitoring Programs Sites (Continued)
Site
Code
CHSC
DEMI
ELNJ
GLKY
GPCOe
HOW
INDEM
MONY
MWOK
NBIL
NBNJ
AQS
Code
45-025-0001
26-163-0033
34-039-0004
21-043-0500
08-077-0017
08-077-0018
55-027-0001
18-089-0022
36-005-0080
40-109-0041
17-031-4201
34-023-0006
Location
Chesterfield SC
Dearborn, MI
Elizabeth, NJ
Grayson, KY
Grand Junction, CO
Horicon, WI
Gary, IN
New York, NY
Midwest City, OK
Northbrook, IL
New Brunswick NT
Land Use
Forest
Industrial
Industrial
Residential
Commercial
Agricultural
Industrial
Residential
Commercial
Residential
Agricultural
Location
Setting
Rural
Suburban
Suburban
Rural
Urban/City
Center
Rural
Urban/City
Center
Urban/City
Center
Urban/City
Center
Suburban
Rural
Estimated
Daily Traffic,
AADTa
(Year)
550
(2010)
106,900
(2010)
250,000
(2006)
428
(2009)
12,000
(2010)
5,000
(2008)
52,440
(2009)
134,421
(2008)
41,200
(2010)
34,100
(2009)
114,322
(2010)
Population
Residing Within
10 Miles of the
Monitoring Siteb
5,605
1,082,362
2,180,662
16,880
117,098
21,539
406,979
6,590,357f
361,698
859,738
783,724
County-level
Vehicle
Registration,
# of Vehicles
(Year)
40,431
(2009)
1,336,940
(2010)
424,894
(2010)d
36,031
(2010)
180,119
(2009)
98,211
(2010)
182,989
(2010)
248,600
(2010)
809,783
(2010)
2,083,141
(2010)
640,893
(2010)d
County-level
Stationary
Source HAP
Emissions
from the 2008
NEIC
(tpy)
97.19
7,384.27
360.61
55.08
532.80
531.88
1,486.55
2,171.17
1,242.77
15,376.26
475.76
County-level
Mobile
Source HAP
Emissions
from the 2008
NEIC
(tpy)
209.23
7,014.06
1,342.05
179.45
573.11
467.91
1,857.03
1,217.06
3,717.21
11,796.13
2,290.35
to
oo
BOLD ITALICS = EPA-designated NATTS site.
aAADT is average annual daily traffic.
Reference: Xionetic, 2011.
c Reference: EPA, 2012b.
dThe proportion of county-level population to the state-level population was applied to state-level vehicle registration figure and used as a surrogate when county-
level vehicle registration counts were not available.
eGPCO's hexavalent chromium monitor is at a separate, but adjacent, location; as such, this site has two AQS codes.
f The 10-mile population estimate for BXNY was used as a surrogate for MONY.
NA = Data not loaded into AQS per agency request.
-------
Table 2-2. Site Characterizing Information for 2010 National Monitoring Programs Sites (Continued)
Site
Code
OCOK
ORFL
PACO
PAFL
PANJ
PROK
PRRI
PXSS
RICO
RIVA
ROCH
AQS
Code
40-109-1037
12-095-2002
08-045-0005
12-095-1004
34-031-0005
40-097-0187
44-007-0022
04-013-9997
08-045-0007
51-087-0014
36-055-1007
Location
Oklahoma City, OK
Winter Park, FL
Parachute, CO
Orlando, FL
Paterson, NJ
Pryor Creek, OK
Providence, RI
Phoenix, AZ
Rifle, CO
Richmond, VA
Rochester, NY
Land Use
Residential
Commercial
Residential
Commercial
Commercial
Industrial
Residential
Residential
Commercial
Residential
Residential
Location
Setting
Suburban
Urban/City
Center
Urban/City
Center
Suburban
Urban/City
Center
Suburban
Urban/City
Center
Urban/City
Center
Urban/City
Center
Suburban
Urban/City
Center
Estimated
Daily Traffic,
AADTa
(Year)
41,600
(2010)
31,500
(2010)
2,600
(2010)
43,500
(2010)
22,272
(2010)
15,900
(2010)
136,800
(2009)
193,000
(2009)
17,000
(2010)
74,000
(2009)
116,725
(2008)
Population
Residing Within
10 Miles of the
Monitoring Siteb
380,090
1,003,746
7,898
872,658
1,332,800
26,739
660,225
1,473,228
17,641
460,195
639,090
County-level
Vehicle
Registration,
# of Vehicles
(Year)
809,783
(2010)
1,037,369
(2010)
74,847
(2009)
1,037,369
(2010)
396,602
(20 Wy
40,832
(2010)
485,837
(2010)d
3,739,918
(2010)
74,847
(2009)
347,790
(2010)
552,184
(2010)
County-level
Stationary
Source HAP
Emissions
from the 2008
NEIC
(tpy)
1,242.77
1,791.25
1,364.26
1,791.25
162.17
329.16
906.46
1,618.22
1,364.26
740.28
1,809.55
County-level
Mobile
Source HAP
Emissions
from the 2008
NEIC
(tpy)
3,717.21
4,785.53
353.08
4,785.53
1,064.24
256.05
1,485.96
11,681.75
353.08
1,020.76
2,250.12
to
BOLD ITALICS = EPA-designated NATTS site.
aAADT is average annual daily traffic.
Reference: Xionetic, 2011.
c Reference: EPA, 2012b.
dThe proportion of county-level population to the state-level population was applied to state-level vehicle registration figure and used as a surrogate when county-
level vehicle registration counts were not available.
eGPCO's hexavalent chromium monitor is at a separate, but adjacent, location; as such, this site has two AQS codes.
f The 10-mile population estimate for BXNY was used as a surrogate for MONY.
NA = Data not loaded into AQS per agency request.
-------
Table 2-2. Site Characterizing Information for 2010 National Monitoring Programs Sites (Continued)
Site
Code
RUCA
RUCO
RUVT
S4MO
SDGA
SEWA
SJJCA
SKFL
SPAZ
SPIL
SSSD
AQS
Code
06-065-8001
NA
50-021-0002
29-510-0085
13-089-0002
53-033-0080
06-085-0005
12-103-0026
04-013-4003
17-031-3103
46-099-0008
Location
Rubidoux, CA
Rulison, CO
Rutland, VT
St. Louis, MO
Decatur, GA
Seattle, WA
San Jose, CA
Pinellas Park, FL
Phoenix, AZ
Schiller Park, IL
Sioux Falls, SD
Land Use
Residential
Agricultural
Commercial
Residential
Residential
Industrial
Commercial
Residential
Residential
Mobile
Commercial
Location
Setting
Suburban
Rural
Urban/City
Center
Urban/City
Center
Suburban
Suburban
Urban/City
Center
Suburban
Urban/City
Center
Suburban
Urban/City
Center
Estimated
Daily Traffic,
AADTa
(Year)
145,000
(2010)
699
(2002)
7,200
(2010)
81,174
(2009)
145,890
(2010)
234,000
(2010)
103,000
(2010)
49,500
(2010)
130,000
(2009)
170,700
(2009)
21,340
(2010)
Population
Residing Within
10 Miles of the
Monitoring Siteb
990,029
17,641
34,336
811,927
793,817
952,319
1,486,476
672,114
898,861
2,046,549
190,685
County-level
Vehicle
Registration,
# of Vehicles
(Year)
1,707,950
(2010)
74,847
(2009)
118,002
(2010)
1,121,528
(2010)
472,535
(2011)
1,763,504
(2010)
1,517,995
(2010)
879,317
(2010)
3,739,918
(2010)
2,083,141
(2010)
208,911
(2010)
County-level
Stationary
Source HAP
Emissions
from the 2008
NEIC
(tpy)
2,552.70
1,364.26
135.82
1,054.65
779.22
3,191.49
3,325.51
1,381.26
1,618.22
15,376.26
382.22
County-level
Mobile
Source HAP
Emissions
from the 2008
NEIC
(tpy)
3,490.17
353.08
308.74
1,157.32
3,044.68
9,694.40
2,772.68
3,808.72
11,681.75
11,796.13
600.33
BOLD ITALICS = EPA-designated NATTS site.
aAADT is average annual daily traffic.
Reference: Xionetic, 2011.
c Reference: EPA, 2012b.
dThe proportion of county-level population to the state-level population was applied to state-level vehicle registration figure and used as a surrogate when county-
level vehicle registration counts were not available.
eGPCO's hexavalent chromium monitor is at a separate, but adjacent, location; as such, this site has two AQS codes.
f The 10-mile population estimate for BXNY was used as a surrogate for MONY.
NA = Data not loaded into AQS per agency request.
-------
Table 2-2. Site Characterizing Information for 2010 National Monitoring Programs Sites (Continued)
Site
Code
SYFL
TMOK
TONY
TOOK
UCSD
UNVT
WADC
WPIN
AQS
Code
12-057-3002
40-143-1127
36-029-1013
40-143-0235
46-127-0001
50-007-0007
11-001-0043
18-097-0078
Location
Plant City, FL
Tulsa, OK
Tonawanda, NY
Tulsa, OK
Union County SD
Underbill, VT
Washington, D.C.
Indianapolis, IN
Land Use
Residential
Residential
Industrial
Industrial
Agricultural
Forest
Commercial
Residential
Location
Setting
Rural
Urban/City
Center
Urban/City
Center
Urban/City
Center
Rural
Rural
Urban/City
Center
Suburban
Estimated
Daily Traffic,
AADTa
(Year)
10,700
(2010)
12,700
(2010)
74,406
(2008)
62,566
(2010)
156
(2007)
1,200
(2005)
7,700
(2009)
143,410
(2009)
Population
Residing Within
10 Miles of the
Monitoring Siteb
323,844
320,319
598,180
456,229
6,153
35,228
1,911,152
787,003
County-level
Vehicle
Registration,
# of Vehicles
(Year)
1,125,844
(2010)
604,284
(2010)
669,746
(2010)
604,284
(2010)
25,051
(2010)
223,316
(2010)
219,173
(2009)
204,908
(2010)
County-level
Stationary
Source HAP
Emissions
from the 2008
NEIC
(tpy)
2,633.02
1,219.02
2,387.82
1,219.02
62.28
347.53
632.23
2,965.43
County-level
Mobile
Source HAP
Emissions
from the 2008
NEIC
(tpy)
4,579.82
3,065.07
2,632.06
3,065.07
122.79
623.35
1,257.69
3,380.45
BOLD ITALICS = EPA-designated NATTS site.
aAADT is average annual daily traffic.
bReference: Xionetic, 2011.
c Reference: EPA, 2012b.
dThe proportion of county-level population to the state-level population was applied to state-level vehicle registration figure and used as a surrogate when county-
level vehicle registration counts were not available.
eGPCO's hexavalent chromium monitor is at a separate, but adjacent, location; as such, this site has two AQS codes.
f The 10-mile population estimate for BXNY was used as a surrogate for MONY.
NA = Data not loaded into AQS per agency request.
-------
The proximity of the monitoring locations to different emissions sources, especially
industrial facilities and heavily traveled roadways, often explains the observed spatial variations
in ambient air quality. To provide a first approximation of the potential contributions of
stationary and mobile source emissions on ambient air quality at each site, Table 2-2 also lists
the following:
• Stationary and mobile source HAP emissions in the monitoring site's residing county,
according to the 2008 National Emissions Inventory (NEI).
• The number of people living within 10 miles of each monitoring site.
• The county-level number of motor vehicles registered in each site's respective
county, based on total vehicle registrations.
• The number of vehicles passing the nearest available roadway to the monitoring site,
generally expressed as average annual daily traffic (AADT).
This information is discussed in further detail in the individual state sections.
2.2 Analytical Methods and Pollutants Targeted for Monitoring
Air pollution typically contains hundreds of components, including, but not limited to,
volatile organic compounds (VOC), metals, and particulate matter. Because the sampling and
analysis required to monitor for every component of air pollution has been prohibitively
expensive, the NMP focuses on specific pollutants that are analyzed using specific methods, as
listed below. The target pollutants varied significantly from monitoring site to monitoring site.
• Compendium Method TO-15 was used to measure ambient air concentrations of
61 VOC.
• EPA-approved SNMOC Method was used to measure 80 ozone precursors. This
method was often used concurrently with Method TO-15.
• Compendium Method TO-11A was used to measure ambient air concentrations of
14 carbonyl compounds.
• Compendium Method TO-13A was used to measure ambient air concentrations of
22 PAH.
• Compendium MethodIO-3.5 was used to measure ambient air concentrations of
11 metals.
2-12
-------
• EPA-approved hexavalent chromium method was used to measure ambient air
concentrations of hexavalent chromium.
At each monitoring site, the sample collection equipment was installed either as a stand-
alone sampler or in a temperature-controlled enclosure (usually a trailer or a shed) with the
sampling probe inlet exposed to the ambient air. With these common setups, most monitoring
sites sampled ambient air at heights approximately five to 20 feet above local ground level.
The detection limits of the analytical methods must be considered carefully when
interpreting the corresponding ambient air monitoring data. By definition, method detection
limits (MDLs) represent the lowest concentrations at which laboratory equipment have been
experimentally determined to reliably quantify concentrations of selected pollutants to a specific
confidence level. If a chemical concentration in ambient air is below the method sensitivity (as
gauged by the method detection limit), the analytical method might not differentiate the pollutant
from other pollutants in the sample or from the random "noise" inherent in laboratory analyses.
While quantification below the MDL is possible, the measurement reliability is lower. Therefore,
when samples contain concentrations at levels below their respective detection limits, multiple
analyses of the same sample may lead to a wide range of measurement results, including highly
variable concentrations or "non-detect" observations (i.e., the pollutant was not detected by the
instrument). Data analysts should exercise caution when interpreting monitoring data with a high
percentage of reported concentrations at levels near or below the corresponding detection limits.
MDLs are determined annually at the ERG laboratory using 40 CFR, Part 136
Appendix B procedures (EPA, 2012c) in accordance with the specifications presented in the
NATTS Technical Assistance Document (TAD) (EPA, 2009b). This procedure involves
analyzing at least seven replicate standards prepared on/in the appropriate sampling media (per
analytical method). Instrument-specific detection limits (replicate analysis of standards only) are
not determined because sample contamination and preparation variability would not be
considered.
For the metals, however, the MDL procedure described by "Appendix D: DQ FAC
Single Laboratory Procedure v2.4" (FAC, 2007) was used to determine MDLs for chromium for
both quartz and Teflon filter types, as well as manganese, cobalt, nickel, cadmium, and lead for
2-13
-------
the quartz filters. The method involves analyzing at least seven replicate samples extracted from
blank sampling and calculating the MDLs from the results. For all other metals analytes, the
MDL procedure described in 40 CFR was used.
Tables 2-3 through 2-8 identify the specific target pollutants for each method and their
corresponding MDLs. For the VOC and SNMOC analyses, the experimentally-determined
MDLs do not change within a given year unless the sample was diluted. The 2010 VOC and
SNMOC MDLs are presented in Tables 2-3 and 2-4, respectively. For the rest of the analyses,
the MDLs vary due to the actual volume pulled through the sample or if the sample was diluted.
For these analyses, the range and average of each MDL is presented for each pollutant in
Tables 2-5 through 2-8. Pollutant-specific MDLs are also presented in Appendix B.
The following discussion presents an overview of the sampling and analytical methods.
For detailed descriptions of the methods, refer to EPA's original documentation of the
Compendium Methods (EPA, 1998; EPA, 1999a; EPA, 1999b; EPA, 1999c; EPA, 1999d; EPA,
2006a).
2.2.1 VOC and SNMOC Concurrent Sampling and Analytical Methods
VOC and SNMOC sampling and analysis can be performed concurrently in accordance
with a combination of EPA Compendium Method TO-15 (EPA, 1999a) and the procedure
presented in EPA's "Technical Assistance Document for Sampling and Analysis of Ozone
Precursors" (EPA, 1998). When referring to SNMOC, this report may refer to this method as the
"concurrent SNMOC method" or "concurrent SNMOC analysis" because both methods were
often employed at the same time to analyze the same sample. Ambient air samples for VOC
and/or concurrent SNMOC analysis were collected in passivated stainless steel canisters. The
ERG laboratory distributed the prepared canisters (i.e., cleaned and evacuated) to the monitoring
sites before each scheduled sample collection event, and site operators connected the canisters to
air sampling equipment prior to each sample day. Prior to field sampling, the passivated canisters
had internal pressures much lower than atmospheric pressure. Using this pressure differential,
ambient air naturally flowed into the canisters automatically once an associated system solenoid
valve was opened. A mass flow controller on the sampling device inlet ensured that ambient air
entered the canister at an integrated constant rate across the collection period. At the end of the
2-14
-------
24-hour sampling period, the solenoid valve automatically closed and stopped ambient air from
flowing into the canister. Site operators recovered and returned the canisters, along with the
Chain of Custody forms and all associated documentation, to the ERG laboratory for analysis.
By analyzing each sample with gas chromatography incorporating mass spectrometry
(operating in the Selected Ion Monitoring (SIM) mode) and flame ionization detection
(GC/MS-FID), laboratory staff determined ambient air concentrations of 61 VOC and/or 80
SNMOC, and calculated the total nonmethane organic compounds (TNMOC) concentration.
TNMOC is the sum of all hydrocarbon concentrations within the sample. Because isobutene and
1-butene elute from the GC column at the same time, the SNMOC analytical method reports
only the sum of the concentrations for these two compounds, and not the separate concentration
for each compound. The same approach applies to w-xylene and/>-xylene for both the VOC and
concurrent SNMOC methods. These raw data are presented in Appendices C and D.
Table 2-3 presents the MDLs for the laboratory analysis of VOC samples with Method
TO-15 and Table 2-4 presents the MDLs for the analysis of SNMOC samples. The MDL for
every VOC is lower than 0.04 parts per billion by volume (ppbv). SNMOC detection limits are
expressed in parts per billion Carbon (ppbC). All of the SNMOC MDLs are less than 0.40 ppbC.
2-15
-------
Table 2-3. 2010 VOC Method Detection Limits
Pollutant
Acetonitrile
Acetylene
Acrolein
Acrylonitrile
tert-Amyl Methyl Ether
Benzene
Bromochloromethane
Bromodichloromethane
Bromoform
Bromomethane
1,3 -Butadiene
Carbon Bisulfide
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
Chloromethane
Chloromethylbenzene
Chloroprene
Dibromochloromethane
2010
MDL
(ppbv)
0.016
0.025
0.036
0.027
0.013
0.019
0.018
0.021
0.011
0.013
0.010
0.011
0.024
0.014
0.012
0.017
0.016
0.017
0.014
0.011
Pollutant
1 ,2-Dibromoethane
/w-Dichlorobenzene
o-Dichlorobenzene
£>-Dichlorobenzene
Dichlorodifluoromethane
1 , 1 -Dichloroethane
1 ,2-Dichloroethane
1 , 1 -Dichloroethene
cis- 1 ,2-Dichloroethylene
/ra«5-l,2-Dichloroethylene
Dichloromethane
1 ,2-Dichloropropane
cis- 1 ,3 -Dichloropropene
trans- 1 , 3 -Dichloropropene
Dichlorotetrafluoroethane
Ethyl Acrylate
Ethyl ter/-Butyl Ether
Ethylbenzene
Hexachloro- 1 ,3 -butadiene
Methyl Ethyl Ketone
2010
MDL
(ppbv)
0.012
0.010
0.012
0.010
0.012
0.017
0.015
0.013
0.036
0.014
0.023
0.025
0.015
0.016
0.012
0.011
0.009
0.012
0.012
0.026
Pollutant
Methyl Isobutyl Ketone
Methyl Methacrylate
Methyl tert-Butyl Ether
w-Octane
Propylene
Styrene
1 , 1 ,2,2-Tetrachloroethane
Tetrachloroethylene
Toluene
1 ,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1,1,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Trichlorotrifluoroethane
1,2,4-Trimethylbenzene
1 ,3 ,5-Trimethylbenzene
Vinyl Chloride
/w.^-Xylene1
o-Xylene
2010
MDL
(ppbv)
0.010
0.021
0.009
0.011
0.028
0.010
0.011
0.011
0.013
0.018
0.020
0.018
0.017
0.012
0.014
0.011
0.010
0.013
0.014
0.010
sum of/w-xylene and^-xylene concentrations and not concentrations of the individual isomers.
2-16
-------
Table 2-4. 2010 SNMOC Method Detection Limits1
Pollutant
Acetylene
Benzene
1,3 -Butadiene
w-Butane
c/s-2-Butene
/raws-2-Butene
Cyclohexane
Cyclopentane
Cyclopentene
w-Decane
1-Decene
/w-Diethylbenzene
£>-Diethylbenzene
2,2-Dimethylbutane
2,3 -Dimethy Ibutane
2, 3 -Dimethy Ipentane
2,4-Dimethylpentane
w-Dodecane
1-Dodecene
Ethane
2-Ethyl-l-butene
Ethylbenzene
Ethylene
/w-Ethyltoluene
o-Ethyltoluene
/>-Ethyltoluene
2010
MDL
(ppbC)
0.195
0.178
0.180
0.169
0.177
0.141
0.189
0.124
0.240
0.230
0.240
0.241
0.141
0.195
0.201
0.374
0.234
0.286
0.290
0.121
0.360
0.181
0.377
0.146
0.178
0.236
Pollutant
w-Heptane
1-Heptene
w-Hexane
1-Hexene
c/s-2-Hexene
/ra«s-2-Hexene
Isobutane
Isobutene/ 1 -Butene2
Isopentane
Isoprene
Isopropylbenzene
2-Methy 1-1 -Butene
3 -Methy 1-1 -Butene
2-Methy 1- 1 -Pentene
4-Methyl- 1 -Pentene
2-Methyl-2-Butene
Methylcyclohexane
Methylcyclopentane
2-Methy Iheptane
3-Methylheptane
2-Methy Ihexane
3 -Methy Ihexane
2-Methy Ipentane
3 -Methy Ipentane
w-Nonane
1-Nonene
2010
MDL
(ppbC)
0.178
0.370
0.236
0.357
0.360
0.360
0.128
0.153
0.185
0.237
0.205
0.240
0.240
0.360
0.360
0.240
0.193
0.140
0.169
0.115
0.112
0.151
0.135
0.196
0.185
0.240
Pollutant
w-Octane
1-Octene
w-Pentane
1 -Pentene
c/s-2-Pentene
/ra«s-2-Pentene
a-Pinene
6-Pinene
Propane
w-Propylbenzene
Propylene
Propyne
Styrene
Toluene
w-Tridecane
1-Tridecene
1,2,3-Trimethylbenzene
1 ,2,4-Trimethylbenzene
1 ,3 ,5-Trimethylbenzene
2,2,3 -Trimethy Ipentane
2,2,4-Trimethy Ipentane
2,3,4-Trimethy Ipentane
w-Undecane
1-Undecene
7w-Xylene/£>-Xylene2
o-Xylene
2010
MDL
(ppbC)
0.166
0.280
0.094
0.121
0.189
0.140
0.240
0.240
0.202
0.205
0.160
0.200
0.275
0.237
0.290
0.290
0.169
0.242
0.170
0.280
0.171
0.140
0.216
0.220
0.280
0.188
1 Concentration in ppbC = concentration in ppbv * number of carbon atoms in compound.
2 Because isobutene and 1-butene elute from the GC column at the same time, the SNMOC analytical method reports
the sum of concentrations for these two compounds and not concentrations of the individual compounds. For the
same reason, the /w-xylene and^-xylene concentrations are reported as a sum.
2-17
-------
2.2.2 Carbonyl Compound Sampling and Analytical Method
Following the specifications of EPA Compendium Method TO-11A (EPA,1999b),
ambient air samples for carbonyl compound analysis were collected by passing ambient air
through an ozone scrubber and then through cartridges containing silica gel coated with
2,4-dinitrophenylhydrazine (DNPH), a compound known to react selectively and reversibly with
many aldehydes and ketones. Carbonyl compounds in ambient air are retained in the sampling
cartridge, while other compounds pass through the cartridge without reacting with the DNPH-
coated matrix. The ERG laboratory distributed the DNPH cartridges to the monitoring sites prior
to each scheduled sample collection event and site operators connected the cartridges to the air
sampling equipment. After each 24-hour sampling period, site operators recovered and returned
the cartridges, along with the Chain of Custody forms and all associated documentation, to the
ERG laboratory for analysis.
To quantify concentrations of carbonyl compounds in the sampled ambient air, laboratory
analysts extracted the exposed DNPH cartridges with acetonitrile. High-performance liquid
chromatography (HPLC) analysis and ultraviolet detection of these solutions determined the
relative amounts of individual carbonyl compounds present in the original air sample. Because
the three tolualdehyde isomers elute from the HPLC column at the same time, the carbonyl
compound analytical method reports only the sum of the concentrations for these isomers, and
not the separate concentrations for each isomer. These raw data are presented in Appendix E.
Table 2-5 lists the MDLs reported by the ERG laboratory for measuring concentrations
of 14 carbonyl compounds. Although the sensitivity varies from pollutant-to-pollutant and from
site-to-site due to the different volumes pulled through the samples, the average detection limit
reported by the ERG laboratory for every pollutant is less than 0.008 ppbv.
2-18
-------
Table 2-5. 2010 Carbonyl Compound Method Detection Limits
Pollutant
Acetaldehyde
Acetone
Benzaldehyde
Butyraldehyde
Crotonaldehyde
2,5-Dimethylbenzaldehyde
Formaldehyde
Hexaldehyde
Isovaleraldehyde
Propionaldehyde
Tolualdehydes1
Valeraldehyde
Minimum
MDL
(ppbv)
0.0020
0.0030
0.0007
0.0009
0.0010
0.0004
0.0020
0.0006
0.0007
0.0010
0.0010
0.0009
Maximum
MDL
(ppbv)
0.0230
0.0300
0.0070
0.0090
0.0100
0.0050
0.0400
0.0060
0.0090
0.0130
0.0170
0.0110
Average
MDL
(ppbv)
0.0060
0.0078
0.0019
0.0023
0.0026
0.0012
0.0044
0.0013
0.0025
0.0026
0.0031
0.0027
1 The three tolualdehyde isomers elute from the HPLC column at the same time; thus,
the analytical method reports only the sum concentration for these three isomers and
not the individual concentrations.
2.2.3 PAH Sampling and Analytical Method
PAH sampling and analysis was performed in accordance with EPA Compendium
Method TO-13A (EPA, 1999c) and ASTM D6209-98 (ASTM, 2004). The ERG laboratory
prepared sampling media and supplied them to the sites before each scheduled sample collection
event. The clean sampling PUF/ XAD-2® cartridge and quartz filter are installed in a high
volume sampler by the site operators and allowed to sample for 24 hours. Sample collection
modules and Chain of Custody forms and all associated documentation were returned to the
ERG laboratory after sample collection. Within 14 days of sampling, the filter and cartridge are
extracted together using a toluene in hexane solution using the Dionex Accelerated Solvent
Extractor (ASE) 350 or ASE 300. The sample extract is concentrated to a final volume of
1.0 milliliter (mL). A volume of 1 microliter (uL) is injected into the GC/MS operating in the
SIM mode to analyze 22 PAH. PAH raw data are presented in Appendix F.
Table 2-6 lists the MDLs for the 22 PAH target pollutants. Although the sensitivity varies
from pollutant-to-pollutant and from site-to-site due to the different volumes pulled through the
samples, the average MDLs for PAH ranged from 0.029 (coronene) to 1.44 (naphthalene)
nanograms per cubic meter (ng/m3).
2-19
-------
Table 2-6. 2010 PAH Method Detection Limits
Pollutant
Acenaphthene
Acenaphthylene
Anthracene
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Chrysene
Coronene
Cyclopenta[cd]pyrene
Dibenz(a,h)anthracene
Fluoranthene
Fluorene
9-Fluorenone
Indeno(l,2,3-cd)pyrene
Naphthalene
Perylene
Phenanthrene
Pyrene
Retene
Minimum
MDL
(ng/m3)
0.036
0.024
0.027
0.027
0.025
0.036
0.028
0.018
0.024
0.021
0.015
0.029
0.021
0.022
0.026
0.030
0.017
0.215
0.021
0.052
0.021
0.050
Maximum
MDL
(ng/m3)
0.367
0.243
0.269
0.270
0.250
0.362
0.279
0.180
0.245
0.213
0.147
0.294
0.210
0.222
0.268
0.306
0.170
7.400
0.209
0.694
0.217
0.661
Average
MDL
(ng/m3)
0.072
0.048
0.053
0.054
0.050
0.071
0.055
0.035
0.049
0.042
0.029
0.058
0.042
0.044
0.053
0.060
0.034
1.437
0.041
0.135
0.043
0.129
2.2.4 Metals Sampling and Analytical Method
Sampling for the determination of metals in or on particulate matter was performed by
the sites in accordance with EPA Compendium Method IO-3.5 (EPA, 1999d). Ambient air
samples for metals analysis were collected by passing ambient air through either 47mm Teflon®
filters or 8 x 10" quartz filters, depending on the separate and distinct sampling apparatus used to
collect the sample; the 47mm Teflon® filter is used for low-volume samplers, whereas the
8" x 10" quartz filter is used for high-volume samplers. EPA provides the filters to the
monitoring sites. Sites sampled for either parti culate matter less than 10 microns (PMio) or total
suspended particulate (TSP). Particulates in ambient air were collected on the filters and after a
24-hour sampling period, site operators recovered and returned the filters, along with the Chain
of Custody forms and all associated documentation, to the ERG laboratory for analysis.
Upon receipt at the laboratory, the whole filters (47mm Teflon®) or filter strips
(8" x 10" quartz) were digested using a dilute nitric acid solution. The digestate was then
2-20
-------
quantified using inductively coupled plasma mass spectrometry (ICP-MS) to determine the
concentration of individual metals present in the original air sample. These raw data are
presented in Appendix G.
Table 2-7 lists the MDLs for the analysis of the metals samples. Due to the difference in
sample volume/filter collection media, there are two sets of MDLs listed in Table 2-7. Although
the sensitivity varies from pollutant-to-pollutant and from site-to-site due to the different
volumes pulled through the samples, the average MDLs ranged from 0.001 ng/m3 (beryllium) to
2.92 ng/m3 (chromium) for the quartz filters and from 0.007 ng/m3 (nickel) to 4.64 ng/m3
(chromium) for the Teflon® filters.
Table 2-7. 2010 Metals Method Detection Limits
Pollutant
Minimum
MDL
(ng/m3)
Maximum
MDL
(ng/m3)
Average
MDL
(ng/m3)
8 X 10" Quartz Filters
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Cobalt
Lead
Manganese
Mercury
Nickel
Selenium
0.003
0.007
0.001
0.047
1.710
0.011
0.497
0.205
0.002
1.020
0.010
0.041
1.330
0.008
0.085
28.700
0.185
82.800
3.450
1.100
42.000
2.180
0.031
0.052
0.001
0.068
2.916
0.019
1.019
0.351
0.013
1.736
0.085
Pollutant
Minimum
MDL
(ng/m3)
Maximum
MDL
(ng/m3)
Average
MDL
(ng/m3)
47mm Teflon® Filters
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Cobalt
Lead
Manganese
Mercury
Nickel
Selenium
0.006
0.003
0.006
0.009
0.320
0.003
0.009
0.030
0.040
0.004
0.080
0.010
0.070
0.030
0.050
5.510
0.010
0.020
0.050
0.230
0.200
0.140
0.009
0.062
0.028
0.011
4.644
0.010
0.011
0.044
0.053
0.007
0.120
2.2.5 Hexavalent Chromium Sampling and Analytical Method
Hexavalent chromium was measured using an EPA-approved approach. For a detailed
description of the method, refer to the "Standard Operating Procedure for the Determination of
Hexavalent Chromium in Ambient Air Analyzed by Ion Chromatography (1C)" (EPA, 2006a).
Ambient air samples for hexavalent chromium analysis were collected by passing ambient air
through sodium bicarbonate impregnated acid-washed cellulose filters. ERG prepared and
distributed filters secured in Teflon cartridges to the monitoring sites prior to each scheduled
sample collection event and site operators connected the cartridges to the air sampling
equipment. After a 24-hour sampling period, site operators recovered the cartridges and Chain of
2-21
-------
Custody forms and returned them to the ERG laboratory for analysis. Upon receipt at the
laboratory, the filters were extracted using a sodium bicarbonate solution. Ion chromatography
(1C) analysis and ultraviolet-visible detection of these extracts determined the amount of
hexavalent chromium present in each sample.
Although the sensitivity varies from site-to-site due to the different volumes pulled
through the samples, the average MDL for the program, which is presented in Table 2-8, was
0.0028 ng/m3. Raw data are presented in Appendix H.
Table 2-8. 2010 Hexavalent Chromium Method Detection Limits
Pollutant
Hexavalent Chromium
Minimum
MDL
(ng/m3)
0.0008
Maximum
MDL
(ng/m3)
0.0665
Average
MDL
(ng/m3)
0.0028
2.3 Sample Collection Schedules
Table 2-9 presents the first and last date on which sample collection occurred for each
monitoring site sampling in 2010. The first sample date for each site is generally in January 2010
and continued through December 2010, although there were a few exceptions. The following
sites began sampling after January 2010 or ended sampling before December 2010:
• The Karnack, TX site (CAMS 85) started sampling hexavalent chromium in
February.
• The Paterson, NJ site (PANJ) began sampling VOC in April.
• The Tonawanda, NY site (TONY) stopped sampling in July.
in 2010:
In some instances, an existing site began sampling additional methods under the NMP
In February, the Deer Park, TX site (CAMS 35) began sampling hexavalent
chromium under the NMP, in addition to PAH.
In June, the Grayson, KY site (GLKY) began sampling VOC under the NMP, in
addition to hexavalent chromium and PAH.
2-22
-------
In other instances, an existing site stopped sampling certain methods under the NMP in
2010:
• The Horicon, WI site (HOWI) stopped sampling PAH under the NMP because the
Wisconsin Department of National Resources laboratory began performing the
analysis.
• The Underbill, VT site (UNVT) stopped sampling carbonyl compounds under the
NMP because the Vermont Department of Environmental Conservation laboratory
began performing this analysis.
Additionally, the instruments at several monitoring sites moved to alternative locations
mid-year:
• The Mayville, WI NATTS site (MVWI) stopped sampling in December 2009 and the
instrumentation was moved to the Horicon, WI site (HOWI). Sampling began at
HOWI in late December 2009. Because only two samples were collected at HOWI in
2009, that site was not included in the 2008-2009 NMP annual report; thus, the two
2009 samples have been included with HOWFs 2010 samples.
• The New York, NY site (BXNY) stopped sampling in June 2010 and the
instrumentation was moved another New York site (MONY). Sampling at MONY
began in July 2010.
• The Rulison, CO site (RUCO) stopped sampling in mid-September 2010 and the
instrumentation was moved to the Battlement Mesa, CO site (BMCO). Sampling at
BMCO began later in September 2010.
2-23
-------
Table 2-9. 2010 Sampling Schedules and Completeness Rates
Site
AZFL
BMCO
BOMA
BRCO
BTUT
BURVT2
BXNY
CAMS 35
CAMS 85
CELA
CHNJ
CHSC
DEMI
ELNJ
Monitoring Period1
First
Sample
1/2/10
9/17/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
2/7/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
Last
Sample
12/28/10
12/29/10
12/28/10
12/29/10
12/28/10
12/28/10
6/13/10
12/28/10
12/28/10
12/28/10
12/28/10
12/28/10
12/28/10
12/28/10
Carbonyl
Compounds
A
61
7
-
17
53
-
-
-
-
-
59
-
60
59
B
61
92
-
242
61
-
-
-
-
-
61
-
61
61
C
100
78
-
71
87
-
-
-
-
-
97
-
98
97
voc
A
-
-
-
-
57
31
-
-
-
-
57
-
61
59
B
-
-
-
-
61
31
-
-
-
-
61
-
61
61
C
-
-
-
-
93
100
-
-
-
-
93
-
100
97
Hexavalent
Chromium
A
-
-
61
-
59
-
28
52
51
-
-
62
59
-
B
-
-
61
-
61
-
28
55
55
-
-
61
61
-
C
-
-
100
-
97
-
100
95
93
-
-
>100
97
-
Metals
A
-
-
61
-
59
-
-
-
-
-
-
-
-
-
B
-
-
61
-
61
-
-
-
-
-
-
-
-
-
C
-
-
100
-
97
-
-
-
-
-
-
-
-
-
SNMOC
A
-
18
-
61
57
-
-
-
-
-
-
-
-
-
B
-
18
-
61
61
-
-
-
-
-
-
-
-
-
C
-
100
-
100
93
-
-
-
-
-
-
-
-
-
PAH
A
-
-
60
-
57
-
28
57
-
59
-
58
59
-
B
-
-
61
-
61
-
28
61
-
61
-
61
61
-
C
-
-
98
-
93
-
100
93
-
97
-
95
97
-
to
to
A = Number of valid samples collected.
B = Number of valid samples that should be collected based on sample schedule and start/end date of sampling.
C = Completeness (%).
1 Begins with 1st sample collected and ends with last sample collected; date range presented may not be representative of each method-specific date range.
2 Sampling schedule was a l-in-12 day schedule not a l-in-6 schedule.
3 Includes two samples from December 2009.
BOLD ITALICS = EPA-designated NATTS site.
Shading indicates that completeness is below the DQO of 85%.
-------
Table 2-9. 2010 Sampling Schedules and Completeness Rates (Continued)
Site
GLKY
GPCO
HOWf
INDEM
MONY
MWOK
NBIL
NBNJ
OCOK
ORFL
PACO
PAFL2
PANJ2
PROK
PRRI
Monitoring Period1
First
Sample
1/2/10
1/2/10
12/21/09
1/2/10
7/13/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/8/10
4/26/10
1/2/10
1/2/10
Last
Sample
12/28/10
12/28/10
12/28/10
12/28/10
12/28/10
12/28/10
12/28/10
12/28/10
12/28/10
12/28/10
12/29/10
12/22/10
12/22/10
12/28/10
12/28/10
Carbonyl
Compounds
A
-
61
-
61
-
60
55
58
60
60
28
-
-
60
-
B
-
61
-
61
-
61
61
61
61
61
302
-
-
61
-
C
-
100
-
100
-
98
90
95
98
98
93
-
-
98
-
voc
A
35
59
-
-
-
61
55
55
61
-
-
-
21
61
-
B
35
61
-
-
-
61
61
61
61
-
-
-
21
61
-
C
100
97
-
-
-
100
90
90
100
-
-
-
100
100
-
Hexavalent
Chromium
A
61
58
63
-
27
-
61
-
-
-
-
-
-
-
60
B
61
61
63
-
29
-
61
-
-
-
-
-
-
-
61
C
100
95
100
-
93
-
100
-
-
-
-
-
-
-
98
Metals
A
-
-
-
-
-
61
61
-
61
-
-
30
-
61
-
B
-
-
-
-
-
61
61
-
61
-
-
30
-
61
-
C
-
-
-
-
-
100
100
-
100
-
-
100
-
100
-
SNMOC
A
-
-
-
-
-
-
55
-
-
-
58
-
-
-
-
B
-
-
-
-
-
-
61
-
-
-
61
-
-
-
-
C
-
-
-
-
-
-
90
-
-
-
95
-
-
-
-
PAH
A
60
57
31
-
29
-
59
-
-
-
-
-
-
-
58
B
61
61
32
-
29
-
61
-
-
-
-
-
-
-
61
C
98
93
97
-
100
-
97
-
-
-
-
-
-
-
95
to
to
A = Number of valid samples collected.
B = Number of valid samples that should be collected based on sample schedule and start/end date of sampling.
C = Completeness (%).
1 Begins with 1st sample collected and ends with last sample collected; date range presented may not be representative of each method-specific date range.
2 Sampling schedule was a l-in-12 day schedule and not a l-in-6 schedule.
3 Includes two samples from December 2009.
BOLD ITALICS = EPA-designated NATTS site.
Shading indicates that completeness is below the DQO of 85%.
-------
Table 2-9. 2010 Sampling Schedules and Completeness Rates (Continued)
Site
PXSS
RICO
RIVA
ROCH
RUCA
RUCO
RUVT2
S4MO
SDGA
SEWA
SJJCA
SKFL
SPAZ2
SPIL
SSSD
Monitoring Period1
First
Sample
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/11/10
1/2/10
1/2/10
Last
Sample
12/28/10
12/29/10
12/28/10
12/28/10
12/28/10
9/11/10
12/28/10
12/28/10
12/28/10
12/28/10
12/28/10
12/28/10
12/22/10
12/28/10
12/28/10
Carbonyl
Compounds
A
20
24
-
-
-
18
-
54
-
59
-
61
-
58
61
B
61
302
-
-
-
212
-
61
-
61
-
61
-
61
61
C
33
80
-
-
-
86
-
89
-
97
-
100
-
95
100
voc
A
61
-
-
-
-
-
28
53
-
59
-
-
29
60
60
B
61
-
-
-
-
-
31
61
-
61
-
-
30
61
61
C
100
-
-
-
-
-
90
87
-
97
-
-
97
98
98
Hexavalent
Chromium
A
57
-
61
59
-
-
-
57
60
59
-
61
-
-
-
B
61
-
61
61
-
-
-
61
61
61
-
61
-
-
-
C
93
-
100
97
-
-
-
93
98
97
-
100
-
-
-
Metals
A
59
-
-
-
-
-
-
60
-
58
58
-
-
-
-
B
61
-
-
-
-
-
-
61
-
61
58
-
-
-
-
C
97
-
-
-
-
-
-
98
-
95
100
-
-
-
-
SNMOC
A
-
60
-
-
-
40
-
-
-
-
-
-
-
-
60
B
-
61
-
-
-
43
-
-
-
-
-
-
-
-
61
C
-
98
-
-
-
93
-
-
-
-
-
-
-
-
98
PAH
A
59
-
60
3
60
-
-
58
59
58
59
59
-
-
-
B
61
-
61
61
61
-
-
61
61
61
61
61
-
-
-
C
97
-
98
5
98
-
-
95
97
95
97
97
-
-
-
to
to
A = Number of valid samples collected.
B = Number of valid samples that should be collected based on sample schedule and start/end date of sampling.
C = Completeness (%).
1 Begins with 1st sample collected and ends with last sample collected; date range presented may not be representative of each method-specific date range.
Sampling schedule was a l-in-12 day schedule and not a l-in-6 schedule.
Includes two samples from December 2009.
BOLD ITALICS = EPA-designated NATTS site.
Shading indicates that completeness is below the DQO of 85%.
-------
Table 2-9. 2010 Sampling Schedules and Completeness Rates (Continued)
Site
SYFL
TMOK
TONY
TOOK
UCSD
UNVT
WADC
WPIN
Monitoring Period1
First
Sample
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
1/2/10
Last
Sample
12/28/10
12/28/10
7/1/10
12/28/10
12/29/10
12/28/10
12/28/10
12/28/10
Carbonyl
Compounds
A
61
60
-
60
58
30
-
56
B
61
61
-
61
61
30
-
61
C
100
98
-
98
95
100
-
92
voc
A
-
61
-
61
59
60
-
-
B
-
61
-
61
61
61
-
-
C
-
100
-
100
97
98
-
-
Hexavalent
Chromium
A
57
-
-
-
58
60
-
B
61
-
-
-
61
61
-
C
93
-
-
-
95
98
-
Metals
A
-
60
-
61
-
61
-
-
B
-
61
-
61
-
61
-
-
C
-
98
-
100
-
100
-
-
SNMOC
A
-
-
-
59
-
-
-
B
-
-
-
61
-
-
-
C
-
-
-
97
-
-
-
PAH
A
60
29
-
-
60
58
-
B
61
30
-
-
61
61
-
C
98
94
-
-
98
95
-
to
to A = Number of valid samples collected.
^ B = Number of valid samples that should be collected based on sample schedule and start/end date of sampling.
C = Completeness (%).
1 Begins with 1st sample collected and ends with last sample collected; date range presented may not be representative of each method-specific date range.
2 Sampling schedule was a l-in-12 day schedule and not a l-in-6 schedule.
3 Includes two samples from December 2009.
BOLD ITALICS = EPA-designated NATTS site.
Shading indicates that completeness is below the DQO of 85%.
-------
According to the NMP schedule, 24-hour integrated samples were to be collected at each
monitoring site every l-in-6 or l-in-12 days (dependent upon location and monitoring
objectives) and each sample collection began and ended at midnight, local standard time.
However, there were some exceptions:
• The Garfield County, CO sites (BMCO, BRCO, PACO, RICO, and RUCO) collected
samples by initiating the samplers manually. For these sites, samples were generally
collected from mid-morning of one day to mid-morning of the next. In addition,
SNMOC samples were collected on a l-in-6 day schedule while carbonyl compounds
were on a l-in-12 day schedule.
• The South Phoenix, AZ site (SPAZ) collected VOC samples on a l-in-12 day
schedule.
• The Paterson, NJ site (PANJ) collected VOC samples on a l-in-12 day schedule.
• The Orlando, FL site (PAFL) collected metals samples on a l-in-12 day schedule.
• The Burlington, VT and Rutland, VT sites (BURVT and RUVT) collected VOC
samples on a l-in-12 day schedule.
Table 2-9 shows the following:
• 24 sites collected VOC samples and 30 sites collected carbonyl compound samples;
VOC and carbonyl compound samples were collected concurrently at 19 sites.
• 9 sites collected SNMOC samples.
• 26 sites collected PAH samples.
• 14 sites collected metals samples.
• 23 sites collected hexavalent chromium samples.
As part of the sampling schedule, site operators were instructed to collect duplicate (or
collocated) samples on roughly 10 percent of the sample days for select methods when duplicate
(or collocated) samplers were available. Field blanks were collected once a month for carbonyl
compounds, hexavalent chromium, metals, and PAH. Sampling calendars were distributed to
help site operators schedule the collection of samples, duplicates, and field blanks. In cases
where a valid sample was not collected for a given scheduled sample day, site operators were
instructed to reschedule or "make up" samples on other days. This practice explains why some
monitoring locations periodically strayed from the l-in-6 or l-in-12 day sampling schedule.
2-28
-------
The l-in-6 or l-in-12 day sampling schedule provides cost-effective approaches to data
collection for trends characterization of toxic pollutants in ambient air and ensures that sample
days are evenly distributed among the seven days of the week to allow weekday/weekend
comparison of air quality. Because the l-in-6 day schedule yields twice the number of
measurements than the l-in-12 day schedule, data characterization based on this schedule tends
to be more representative.
2.4 Completeness
Completeness refers to the number of valid samples collected and analyzed compared to
the number of total samples expected based on a l-in-6 or l-in-12 day sample schedule.
Monitoring programs that consistently generate valid samples have higher completeness than
programs that consistently have invalid samples. The completeness of an air monitoring
program, therefore, can be a qualitative measure of the reliability of air sampling and laboratory
analytical equipment and a measure of the efficiency with which the program is managed. The
completeness for each monitoring site is presented in Table 2-9. Table 2-10 presents method-
specific completeness. Appendix I identifies samples that were invalidated and lists the reason
for invalidation, based on the applied AQS null code.
Table 2-10. Method Completeness Rates for 2010
Method
voc
SNMOC
Carbonyl Compounds
PAH
Metals Analysis
Hexavalent Chromium
# of Valid
Samples
1,264
468
1,499
1,354
811
1,291
#of
Samples
Scheduled
1,307
488
1,608
1,462
820
1,328
Method
Completeness
(%)
96.71
95.90
93.22
92.61
98.90
97.21
Minimum
Site-Specific
Completeness
(%)
86.89
(S4MO)
90.16
(NBIL)
32.79
(PXSS)
4.92
(ROCH)
95.08
(SEWA)
92.73
(CAMS 85)
Maximum
Site-Specific
Completeness
(%)
100.00
(10 Sites)
100.00
(2 Sites)
100.00
(7 Sites)
100.00
(2 Sites)
100.00
(9 Sites)
>100
(CHSC)
2-29
-------
The following observations summarize the completeness of the monitoring datasets for
samples collected during the 2010 NMP sampling year, as shown in Tables 2-9 and 2-10.
• For VOC sampling, the site-specific completeness for 2010 ranged from 87 to 100
percent, with an overall completeness of 97 percent.
• For SNMOC sampling, the site-specific completeness for 2010 ranged from 90 to 100
percent, with an overall completeness of 96 percent.
• For carbonyl compound sampling, the site-specific completeness for 2010 ranged
from 33 to 100 percent, with an overall completeness of 93 percent.
• For PAH sampling, the site-specific completeness for 2010 ranged from 5 to
100 percent, with an overall completeness of 93 percent.
• For metals sampling, the site-specific completeness for 2010 ranged from 95 to 100
percent, with an overall completeness of 99 percent.
• For hexavalent chromium sampling, the site-specific completeness for 2010 ranged
from 93 to greater than 100 percent, with an overall completeness of 97 percent.
The data quality objective (DQO) for completeness based on the EPA-approved Quality
Assurance Project Plan (QAPP) specifies that at least 85 percent of samples from a given
monitoring site must be collected and analyzed successfully to be considered sufficient for data
trends analysis (ERG, 2009). The data in Table 2-9 shows that five datasets from a total of 126
datasets from the 2010 NMP monitoring sites did not meet this data quality objective (cells
shaded in Table 2-9).
• Three of the five site-method combinations for which completeness was less than
85 percent were for Garfield County carbonyl compound sites (BRCO, BMCO,
and RICO). The instrumentation at RUCO was moved to the new BMCO location
in September; as such, BMCO did not have time to make up invalid carbonyl
compound samples. In addition, these sites tended to experience issues with their
carbonyl compound samplers.
• Maintenance of the primary carbonyl compound sampler at PXSS in mid-
February led to a problem with the ozone denuder. As a result, the sampling
results from mid-February 2010 through the end of the year were invalidated.
• Problems with the PAH sampler at ROCH led to the invalidation of nearly all of
ROCH's PAH data for 2010. The sampler was re-certified at the end of 2010 and
the final three samples from 2010 were kept.
2-30
-------
3.0
Summary of the 2010 National Monitoring Programs Data Treatment and Methods
This section summarizes the data treatment
and approaches used to evaluate the measurements
generated from samples collected during the 2010
NMP sampling year. These data were analyzed on
a program-wide basis as well as a site-specific
basis.
Results from the program-wide data
analyses are presented in Section 4
and results from the site-specific data
analyses are presented in the
individual state sections, Sections 5
through 28.
A total of 214,954 valid air toxics concentrations (including non-detects, duplicate
analyses, replicate analyses, and analyses for collocated samples) were produced from 8,529
valid samples collected at 52 sites during the 2010 reporting year. A tabular presentation of the
raw data and statistical summaries are found in Appendices C through O, as presented
in Table 3-1. Appendix P serves as the glossary for the NMP report and many of the terms
discussed and defined throughout the report are provided here.
Table 3-1. Overview and Layout of Data Presented
Pollutant Group
voc
SNMOC
Carbonyl Compounds
PAH
Metals
Hexavalent Chromium
Number of Sites
24
9
30
26
14
23
Appendix
Raw Data
C
D
E
F
G
H
Statistical Summary
J
K
L
M
N
O
3.1 Approach to Data Treatment
This section examines the various statistical tools employed to characterize the data
collected during the 2010 sampling year. Certain data analyses were performed at the program-
level, other data analyses were performed at both the program-level and on a site-specific basis,
and still other approaches were reserved for site-specific data analyses only. Regardless of the
data analysis employed, it is important to understand how the concentration data were treated.
The following paragraphs describe techniques used to prepare this large quantity of data for data
analysis.
5-1
-------
Pairs of duplicate (or collocated) and replicate measurements were averaged together in
order to calculate a single concentration for each pollutant for each method for each sample day
at each monitoring site. This is referred to as the preprocessed daily measurement.
Concentrations of m,/>-xylene and o-xylene were summed together and are henceforth
referred to as "total xylenes," "xylenes (total)," or simply "xylenes" throughout the remainder of
this report, with a few exceptions. One exception is Section 4.1, which examines the results of
basic statistical calculations performed on the dataset. Table 4-1 and Table 4-2, which are the
method-specific statistics for VOC and SNMOC, respectively, present the xylenes results
retained as m,p-xy\ene and o-xylene species. This is also true of the Data Quality section
(Section 29).
The treatment of non-detects in the 2010 NMP report differs from previous reports. For
the 2010 NMP, where statistical parameters are calculated based on the preprocessed daily
measurements, zeros have been substituted for non-detect results. In past reports, the substitution
of zeros was applied to only risk-related analyses; however, in the 2010 NMP report, the
substitution of zeros was applied to all analyses. This approach is consistent with how data are
loaded into AQS per the NATTS TAD (EPA, 2009b) as well as other EPA air toxics monitoring
programs such as the School Air Toxics Monitoring Program (SATMP) (EPA, 201 la). The
substitution of zeros for non-detects would result in lower average concentrations of pollutants
that are rarely measured at or above the associated method detection limit and/or have a
relatively high MDL.
In order to compare concentrations across multiple sampling methods, all concentrations
have been converted to a common unit of measure: microgram per cubic meter (ug/m3).
However, whenever a particular sampling method is isolated from others, such as in Tables 4-1
through 4-6, the statistical parameters are presented in the units of measure associated with the
particular sampling method. As such, it is important to pay very close attention to the unit of
measure associated with each data analysis discussed in this and subsequent sections of the
report.
5-2
-------
In addition, this report presents various duration-based averages to summarize the
measurements for a specific site; where applicable, quarterly and annual averages were
calculated for each site. The quarterly average of a particular pollutant is simply the average
concentration of the preprocessed daily measurements over a given calendar quarter. Quarterly
averages include the substitution of zeros for all non-detects. The first quarter in a calendar year
includes concentrations from January, February, and March; the second quarter includes April,
May, and June; the third quarter includes July, August, and September; and the fourth quarter
includes October, November, and December. A site must have a minimum of 75 percent valid
samples of the total number of samples possible within a given quarter to have a quarterly
average. For sites sampling on a l-in-6 day sampling schedule, 12 samples represents 75 percent;
for sites sampling on a l-in-12 day schedule, 6 samples represents 75 percent. Sites that do not
meet these minimum requirements do not have a quarterly average concentration presented. Sites
may not meet this minimum requirement due to invalidated or missed samples or because of a
shortened sampling duration.
An annual average includes all measured detections and substituted zeros for non-detects
for a given calendar year (2010). Annual average concentrations were calculated for monitoring
sites where three quarterly averages could be calculated and where method completeness was
greater than or equal to 85 percent. Sites that do not meet these requirements do not have an
annual average concentration presented.
The concentration averages presented in this report are often provided with their
associated 95 percent confidence intervals. Confidence intervals represent the interval within
which the true average concentration falls 95 percent of the time. The confidence interval
includes an equal amount of quantities above and below the concentration average. For example,
an average concentration may be written as 1.25 ± 0.25 |ig/m3, thus the interval over which the
true average would be expected to fall would be between 1.00 to 1.50 |ig/m3 (EPA, 201 la).
3.2 Human Health Risk and the Pollutants of Interest
A practical approach to making an assessment on a large number of measurements is to
focus on a subset of pollutants based on the end-use of the dataset. Thus, a subset of pollutants is
selected for further data analyses for each annual NMP report. In NMP annual reports prior to
3-3
-------
2003, this subset was based on the frequency and magnitude of concentrations (previously called
"prevalent compounds"). Since the 2003 NMP annual report, health risk-based calculations have
been used to identify "pollutants of interest." For the 2010 NMP report, the pollutants of interest
are also based on risk potential. The following paragraphs provide an overview of health risk
terms and concepts and outline how the pollutants of interest are determined and then used
throughout the remainder of the report.
EPA defines risk as "the probability that damage to life, health, or the environment will
occur as a result of a given hazard (such as exposure to a toxic chemical)" (EPA, 201 Ib). Human
health risk can be defined in terms of time. Chronic effects develop from repeated exposure over
long periods of time; acute effects develop from a single exposure or from exposures over short
periods of time (EPA, 201 Ob). Health risk is also route-specific; that is, risk varies depending
upon route of exposure (i.e., oral vs. inhalation). Because this report covers air toxics in ambient
air, only the inhalation route is considered. Hazardous air pollutants (HAPs) are those pollutants
known or suspected to "cause cancer or other serious health effects, such as reproductive effects
or birth defects, or adverse environmental and ecological effects" (EPA, 201 Ic).
Health risks are typically divided into cancer risk and noncancer health risks when
referring to human health risk. Cancer risk is defined as the likelihood of developing cancer as a
result of exposure over a 70-year period, and is presented as the number of people at risk for
cancer per million people. Noncancer health effects include conditions such as asthma;
noncancer health risks are presented as a value below which no adverse health effects are
expected (EPA, 20lib).
In order to assess health risk, EPA and other agencies develop screening values, such as
cancer unit risk estimates (UREs) and noncancer reference concentrations (RfCs), to estimate
cancer and noncancer risks and to identify (or screen) where air toxics concentrations may
present a human health risk.
EPA has published a guidance document outlining a risk screening approach that utilizes
a risk-based methodology for performing an initial screen of ambient air toxics monitoring
datasets (EPA, 201 Ob). This preliminary risk screening process provides a risk-based
3-4
-------
methodology for analysts and interested parties to identify which pollutants may pose a risk in
their area. Not all pollutants analyzed under the NMP have screening values; of the 172
pollutants sampled under the NMP, 72 pollutants have screening values in the guidance
document. The screening values used in this analysis are presented in Appendix Q1.
The preprocessed daily measurements of the target pollutants were compared to these
chronic risk screening values in order to identify pollutants of interest across the program. The
following risk screening process was used to identify pollutants of interest:
1. The TO-15 and SNMOC methods have 12 pollutants in common. If a pollutant was
measured by both the TO-15 and SNMOC methods at the same site, the TO-15
results were used. The purpose of this data treatment is to have one concentration per
pollutant per day per site.
2. Each preprocessed daily measurement was compared against the screening value.
Concentrations that were greater than the screening value are described as "failing the
screen."
3. The number of failed screens was summed for each applicable pollutant.
4. The percent contribution of the number of failed screens to the total number of failed
screens program-wide was calculated for each applicable pollutant.
5. The pollutants contributing to the top 95 percent of the total failed screens were
identified as pollutants of interest.
In regards to Step 5 above, the actual cumulative contribution may exceed 95 percent in
order to include all pollutants contributing to the minimum 95 percent criteria (refer to
acenaphthene in Table 4-7 for an example). In addition, if the 95 percent cumulative criterion is
reached, but the next pollutant contributed equally to the number of failed screens, that pollutant
was also designated as a pollutant of interest. Results for the program-wide risk screening
process are provided in Section 4.2.
Laboratory analysts have indicated that acetonitrile values may be artificially high (or
non-existent) due to site conditions and potential cross-contamination with concurrent sampling
of carbonyl compounds using Method TO-11 A. The inclusion of acetonitrile in data analysis
1 The risk screening process used in this report comes from guidance from EPA Region 4's report "A Preliminary
Risk-Based Screening Approach for Air Toxics Monitoring Datasets" but the screening values referenced in that
report have been updated (EPA, 2012d).
3-5
-------
calculations must be determined on a site-specific basis by the agency responsible for the site.
Thus, acetonitrile results are excluded from certain program-wide and site-specific data analyses,
particularly those related to risk.
For the 2010 NMP report, another step for identifying the pollutants of interest was
added. In addition to the preliminary risk-screening approach described above, the pollutants of
interest designation was further refined based on the NATTS TAD (EPA, 2009b). This document
identifies 19 pollutants ("Method Quality Objective (MQO) Core Analytes") that participating
sites are required to sample and analyze for under the NATTS program. Table 3-2 presents these
19 NATTS MQO Core Analytes. Monitoring for these pollutants is required because they are
major health risk drivers according to EPA (EPA, 2009b).
Table 3-2. NATTS MQO Core Analytes
Pollutant
Acrolein
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Acetaldehyde
Formaldehyde
Naphthalene
Benzo(a)pyrene
Arsenic
Beryllium
Cadmium
Manganese
Lead
Nickel
Hexavalent chromium
Class/Method
VOC/TO-15
Carbonyl Compounds/
TO-11A
PAH/TO-13A
Metals/IO-3.5
Metals/EPA
With the exception of acrolein, all of the pollutants listed in Table 3-2 are inherently
considered pollutants of interest due to their designation as NATTS MQO Core Analytes. If a
pollutant listed in Table 3-2 did not meet the pollutant of interest criteria based on the
preliminary risk screening approach outlined above, that pollutant was added to the list of
program-wide pollutants of interest.
-------
Although it is a NATTS MQO Core Analyte, acrolein was excluded from the preliminary
risk screening process due to questions about the consistency and reliability of the measurements
(EPA, 2010c). Thus, the results from sampling and analysis of this pollutant have been excluded
from any risk-related analyses presented in this report, similar to acetonitrile (as discussed
above).
The "pollutants of interest" designation is reserved for pollutants targeted for sampling
through the NMP that meet the identified criteria. As discussed in Section 2.0, agencies
operating monitoring sites that participate under the NMP are not required to have their samples
analyzed by EPA's contract laboratory or may measure analytes other than those targeted under
the NMP. In these cases, data are generated by sources other than ERG and are not included in
the preliminary risk screening process or any other data analysis contained in this report.
3.3 Noncancer Risk Screening Evaluation Using Minimum Risk Levels
In addition to the preliminary risk screening described above, a second risk screening was
conducted using the Agency for Toxic Substances and Disease Registry (ATSDR) Minimal Risk
Level (MRL) health benchmarks (ATSDR, 2010). An MRL is a concentration of a hazardous
substance that is "likely to be without appreciable risk of adverse noncancer health effects over a
specified duration of exposure" (ATSDR, 2012). MRLs are intended to be used as screening
tools, similar to the risk screening approach discussed above, and "exposure to a level above the
MRL does not mean that adverse health effects will occur" (ATSDR, 2012). ATSDR defines
MRLs for three durations of exposure: acute, intermediate, and chronic exposure. Acute risk
results from exposures of 1 to 14 days; intermediate risk results from exposures of 15 to 364
days; and chronic risk results from exposures of 1 year or greater (ATSDR, 2012). MRLs, as
published by ATSDR, are presented in parts per million (ppm) for gases and milligrams per
cubic meter (mg/m3) for particulates. The MRLs used in this report have been converted to
|ig/m3, have one significant figure, and are presented in Appendix Q.
For this risk screening evaluation, the preprocessed daily measurements were compared
to acute MRLs; quarterly averages were compared to intermediate MRLs; and annual averages
were compared to chronic MRLs. Section 4.2.2 presents the number of preprocessed daily
measurements, quarterly averages, and/or annual averages that were greater than their respective
3-7
-------
MRL for each pollutant, summed to the program level. The number of site-specific
concentrations and/or time period averages that were greater than their respective MRLs is
expanded upon in the individual state sections.
3.4 Additional Program-Level Analyses of the 2010 National Monitoring Programs
Dataset
This section summarizes additional analyses performed on the 2010 NMP dataset at the
program level. Additional program-level analyses include an examination of the potential effect
of motor vehicles and a review of how concentrations vary among the sites themselves and from
quarter-to-quarter. The results of these analyses are presented in Sections 4.3 and 4.4.
3.4.1 The Effect of Mobile Source Emissions on Spatial Variations
Mobile source emissions from motor vehicles contribute significantly to air pollution.
"Mobile sources" refer to emitters of air pollutants that move, or can be moved, from place to
place and include both on-road and non-road emissions (EPA, 2012e). Pollutants found in motor
vehicle exhaust generally result from incomplete combustion of vehicle fuels. Although modern
vehicles and, more recently, vehicle fuels have been engineered to minimize air emissions, all
motor vehicles with internal combustion engines emit a wide range of pollutants. The magnitude
of these emissions primarily depends on the volume of traffic, while the chemical profile of these
emissions depends more on vehicle design and fuel formulation. This report uses a variety of
parameters to quantify and evaluate the effect of motor vehicle emissions on ambient air quality,
which are discussed further in Section 4.3:
• Emissions data from the NEI
• Total hydrocarbon concentrations
• Motor vehicle ownership data
• Estimated daily traffic volume
• Vehicle miles traveled (VMT)
This report uses Pearson correlation coefficients to measure the degree of correlation
between two variables, such as the ones listed above. By definition, Pearson correlation
coefficients always lie between -1 and +1. Three qualification statements apply:
-------
• A correlation coefficient of-1 indicates a perfectly "negative" relationship, indicating
that increases in the magnitude of one variable are associated with proportionate
decreases in the magnitude of the other variable, and vice versa.
• A correlation coefficient of+1 indicates a perfectly "positive" relationship, indicating
that the magnitudes of two variables both increase and both decrease proportionately.
• Data that are completely uncorrelated have Pearson correlation coefficients of 0.
Therefore, the sign (positive or negative) and magnitude of the Pearson correlation coefficient
indicate the direction and strength, respectively, of data correlations. In this report, correlation
coefficients greater than or equal to 0.50 and less than or equal to -0.50 are classified as strong,
while correlation coefficients less than 0.50 and greater than -0.50 are classified as weak.
The number of observations used in a calculation is an important factor to consider when
analyzing the correlations. A correlation using relatively few observations may skew the
correlation, making the degree of correlation appear higher (or lower) than it may actually be.
Thus, in this report, five data points must be available to present a correlation.
3.4.2 Variability Analyses
Variability refers to the degree of difference among values in a dataset. Three types of
variability are analyzed for this report. The first type examines the coefficient of variation for
each of the program-level pollutants of interest across the program sites. The coefficient of
variation provides a relative measure of variability by expressing standard deviation to the
magnitude of the arithmetic mean for each of the program-level pollutants of interest, as
identified in Section 4.2. It is particularly useful when comparing different sets of data because it
is unitless (Pagano, P. and Gauvreau, K., 2000). In this report, variability across data
distributions for different sites and different pollutants are compared. The coefficients of
variation are shown in the form of scatter plots, where data points represent the coefficients of
variation and a trend line is plotted to show linearity. In addition, the "R2" value is also shown
on each scatter plot. R2 is the coefficient of determination and is an indicator of how dependant
one variable is on the other. If R2 is equal to 1.0, the data exhibit perfect linearity; the lower R2,
the less dependent the variables are each other (Pagano, P. and Gauvreau, K., 2000). Pollutants
of interest whose data points are clustered together indicate uniformity in how the concentrations
are dispersed among the sites. This suggests that concentrations are affected by typical and
3-9
-------
consistent sources (e.g., mobile sources). Data points that are not clustered suggest the likelihood
of a stationary source not typically found in most urban areas (e.g., coke manufacturing facility).
The second type of variability assessed in this report is inter-site variability. The annual
average concentration for each program-wide pollutant of interest for each site is plotted in the
form of a bar graph. The criteria for calculating an annual average is discussed in Section 3.1 and
sites that do not meet these requirements do not have an annual average concentration presented.
This assessment allows the reader to visualize how concentrations varied across the sites for a
particular pollutant of interest. In order to further this analysis, the program-level average
concentrations, as presented in Tables 4-1 through 4-6 in Section 4.1, are plotted against the site-
specific annual averages. This allows the reader to see how the site-specific annual averages
compared to the program-level average for each pollutant. Note that the average concentrations
shown for VOC, SNMOC, and carbonyl compounds in Tables 4-1 through 4-3 are presented in
method-specific units, but have been converted to a common unit of measurement (|ig/m3) for
the purposes of this analysis.
Quarterly variability is the third type of variability assessed in this report. The
concentration data for each site were divided into four quarters for each year, as described in
Section 3.1. The completeness criteria, also described in Section 3.1, are maintained here as well.
The site-specific quarterly averages are illustrated by bar graphs for each program-level pollutant
of interest. This analysis allows for a determination of a quarterly (or seasonal) correlation with
the magnitude of concentrations for a specific pollutant.
3.4.3 Greenhouse Gas Assessment
Currently, there is considerable discussion about climate change among atmospheric and
environmental scientists. Climate change refers to an extended period of change in
meteorological variables used to determine climate, such as temperature and precipitation.
Researchers are typically concerned with greenhouse gases (GHGs), which are those that cause
heat to be retained in the atmosphere (EPA, 2012f).
Agencies researching the effects of greenhouse gases tend to concentrate primarily on
tropospheric levels of these gases. The troposphere is the lowest level of the atmosphere, whose
3-10
-------
height varies depending on season and latitude. This is also the layer in which weather
phenomenon occur (NOAA, 2012a). A few of VOC measured with Method TO-15 are
greenhouse gases, although these measurements reflect the concentration at the surface, or in the
breathing zone, and do not represent the entire troposphere. Section 4.5 presents the 10 GHGs
currently measured with Method TO-15, their Global Warming Potential (GWP), and the
average concentration across the NMP program. GWP is a way to determine a pollutant's ability
to retain heat relative to carbon dioxide, which is one of the predominant anthropogenic GHGs in
the atmosphere (EPA, 2012g and NOAA, 2012b). In the future, additional GHG pollutants may
be added to the NMP Method TO-15 target pollutant list in order to assess their surface-level
ambient concentrations.
3.5 Additional Site-Specific Analyses
In addition to many of the analyses described in the preceding sections, the state-specific
sections contain additional analyses that are applicable only at the local level. This section
provides an overview of these analyses but does not discuss their results. Results of these site-
specific analyses are presented in the individual state-specific sections (Sections 5 through 28).
3.5.1 Site Characterization
For each site participating in the NMP for 2010, a site characterization was performed.
This analysis includes a review of the nearby area surrounding the monitoring site; plotting of
emissions sources surrounding the monitoring site; and obtaining population, vehicle
registration, traffic data, and other characterizing information.
Regarding the plotting of emissions sources: for the 2010 NMP report, point sources
plotted near the monitoring sites were obtained from Version 2 of the 2008 NEI (EPA, 2012b).
The 2008 NEI was compiled using a more streamlined approach by EPA from state, local, and
tribal agencies, as well as limited emission inventory data from other federal programs, such as
EPA's Toxics Release Inventory (TRI). By comparison, the 2008-2009 NMP report used Version
3 of the 2005 NEI, which included additional datasets not available for the 2008 NEI. As such,
the total number of emission sources surrounding the monitoring sites is generally lower in the
2008 NEI vs. the 2005 NEI. Thus, when comparing facility maps and emission estimates
presented in the 2010 NMP report to those presented in the 2008-2009 NMP report, it should be
3-11
-------
noted that the emissions inventory used in each report was for different base years and was
compiled differently.
3.5.2 Meteorological Analysis
Several site-specific meteorological analyses were performed in order to help readers to
determine which meteorological factors may play a role in a given site's air quality. First, an
overview of general climatology is provided, based on the area in which each site in located, to
give readers a general idea of what types of meteorological conditions likely affect the site. Next,
the average (or mean) for several meteorological parameters (such as temperature and relative
humidity) are provided. Two averages are presented, one average for all days in 2010 and one
average for sample days only. These two averages allow for the determination of how
meteorological conditions on sample days varied from typical conditions throughout the year.
These averages are based on hourly meteorological observations collected from the National
Weather Service (NWS) weather station nearest each site and obtained from the
National Climatic Data Center (NCDC, 2009 and 2010). Although some monitoring sites have
meteorological instruments on-site and report this data to AQS, NWS data were chosen for this
analysis for several reasons:
• Some sites do not have meteorological instruments on-site.
• Some sites collect meteorological data but do not report it to AQS; thus it is not readily
available.
• There are differences among the sites in the meteorological parameters reported to AQS.
Although there are limitations to using NWS data, the data used is standardized.
In addition to the climate summary and the statistical calculations performed on
meteorological observations collected near each monitoring site, the following sections describe
the additional meteorological analyses that were performed for each monitoring site. These
analyses were performed to further characterize the meteorology at or near each monitoring site
and to determine if the meteorological conditions on days samples were collected were
representative of conditions typically experienced near each site.
3-12
-------
3.5.2.1 Back Trajectory Analysis
A back trajectory traces the origin of an air parcel in relation to the location where it is
currently being measured. The method of constructing a back trajectory uses the Lagrangian
frame of reference. In simplest terms, an air parcel can be traced back 1 hour to a new point of
reference based on the current measured wind speed and direction. At this new point of reference
(that is now 1 hour prior to the current observation), the wind speed and direction are used again
to determine where the air was 1 hour before. Back trajectory calculations are also governed by
other meteorological parameters, such as pressure and temperature. Each time segment is
referred to as a "time step."
Gridded meteorological data and the model used for back trajectory analyses were
prepared and developed by the National Oceanic and Atmospheric Administration (NOAA)
using data from the NWS and other cooperative agencies. The model used is the Hybrid Single-
Particle Lagrangian Integrated Trajectory (HYSPLIT) model (Draxler, R.R. and Rolph, G.D.,
1997 and 1998; Draxler, R.R., 1999). Back trajectories were computed using the HYPLIT model
to represent four times for each sample day, one at OOZ, 06Z, 12Z, and 18Z. Although back
trajectories can be modeled for extended periods of time, trajectories were constructed for
durations of 24 hours to match the 24-hour sampling duration. Trajectories are modeled with an
initial height of 50 meters above ground level (AGL), and each sample day's trajectories are
plotted to create a composite back trajectory map. A composite back trajectory map was
constructed for each monitoring site using Geographical Information System (GIS) software.
The composite back trajectory map can be used in the estimation of a 24-hour air shed domain
for each site. An air shed domain is the geographical area surrounding a site from which an air
parcel may typically travel within the 24-hour time frame. Agencies can use the air shed domain
to evaluate regions where long-range transport may affect their monitoring site.
In addition to the composite back trajectory map, the HYSPLIT model was used to
perform trajectory cluster analysis. This analysis is a grouping technique that allows the model to
create a subset of trajectories or "clusters" that represent trajectories originating from similar
locations. For each monitoring site, data from each sample day's trajectories were used as input
for the cluster analysis program. The model compares the end points between each trajectory and
calculates a spatial variance. Trajectories that are similar to each other have lower spatial
3-13
-------
variances while trajectories that are dissimilar have larger spatial variances. The model then
provides the user with information about total spatial variance (TSV) among the trajectories,
which allows the user to determine how many clusters best represent a given group of
trajectories (Draxler, R.R., et. al., 2009). Similar to the composite map, once the cluster
trajectories for each site were computed, a cluster map was constructed for each monitoring site
using GIS software. Both the direction and the distance from monitoring site are considered in
the clustering process. A minimum of 30 trajectories must be available for the model to run the
cluster analysis. Since four trajectories were computed for each sample day, a minimum of 30
sample days was needed to run the cluster analysis. The cluster analysis is useful for
scientifically and quantitatively determining where the air most often originates for a given
location.
3.5.2.2 Wind Rose Analysis
Wind roses were constructed for each site to help identify the predominant direction from
which the wind blows. A wind rose shows the frequency of wind directions as petals positioned
around a 16-point compass, and uses color or shading to represent wind speeds. Wind roses are
constructed by uploading hourly NWS surface wind data from the nearest weather station (with
sufficient data) into a wind rose software program, WRPLOT (Lakes, 2011). For each site, three
wind roses were constructed: first, historical data were used to construct a historical wind rose
for years prior to sampling; second, a wind rose presenting the wind data for the entire calendar
year; and lastly, a wind rose presenting the wind data for sample days only. In addition to the
wind roses, a map showing the distance between the NWS station used and the monitoring site is
presented.
A wind rose is often used in determining where to install an ambient monitoring site
when trying to capture emissions from an upwind source. A wind rose may also be useful in
determining whether high concentrations correlate with a specific wind direction. While the
composite back trajectory maps show where a parcel of air originated on a number of days, the
wind rose shows the frequency at which wind speed and direction are measured near the
monitoring site. Thus, the back trajectory analysis focuses on long range transport, while the
wind rose captures day-to-day fluctuations at the surface. Both are used to identify potential
meteorological influences on a monitoring site.
3-14
-------
3.5.3 Site-Specific Comparison to Program-level Average Concentrations
To better understand how a site's concentrations compare to the program-level
concentrations, as presented in Tables 4-1 through 4-6 of Section 4.1, the site-specific and
program-level concentrations are presented together graphically for the selected NATTS MQO
Core Analytes listed in Table 3.3. This analysis is an extension of the analysis discussed in
Section 3.4.2 and utilizes box and whisker plots, or simply boxplots, to visually show this
comparison. These boxplots were created in Microsoft Excel, using the Peltier Box and Whisker
Plot Utility (Peltier, 2012). This analysis was conducted for the selected NATTS MQO Core
Analytes shown in Table 3-3. Note that for sites that sampled SNMOC, benzene and
1,3-butadiene are showed only in comparison to those sites sampling SNMOC as opposed to
sites sampling these pollutants with Method TO-15, to match Tables 4-1 through 4-6 in
Section 4.1.
Table 3-3. NATTS MQO Core Analytes Selected for Comparative Analysis
Pollutant
Benzene
1,3 -Butadiene
Acetaldehyde
Formaldehyde
Benzo(a)pyrene
Naphthalene
Arsenic
Manganese
Hexavalent Chromium
Class/Method
VOC/TO-15 and SNMOC
Carbonyl
Compounds/TO-1 1 A
PAH/TO-13A
MetaMO-3.5
Metals/EPA
The boxplots used in this analysis overlay the site-specific minimum, annual average, and
maximum concentrations over several program-level statistical metrics. For the program-level,
the first, second (median), third, and fourth (maximum) quartiles are shown as colored segments
on a "bar" where the color changes indicates the exact numerical value of the quartile. The thin
vertical line represents the program-level average concentration. The site-specific annual average
is shown as a white circle plotted on top of the bar and the horizontal lines represent the
minimum and maximum concentration measured at the site. An example of this figure can be
seen in Figure 5-10. Note that the program-level average concentrations shown for VOC,
3-15
-------
SNMOC, and carbonyl compounds in Tables 4-1 through 4-3 are presented in method-specific
units, but have been converted to a common unit of measurement (|ig/m3) for the purposes of this
analysis. These graphs are presented in Section 5 through 28, and are grouped by pollutant
within each state section. This allows for both a "site vs. program" comparison, and an inter-site
comparison within a given state.
3.5.4 Site Trends Analysis
Table 2-1 presents current monitoring sites that have participated in the NMP in previous
years. Site-specific trends analyses were conducted for sites with at least 5 consecutive years of
method-specific data analyzed under the NMP. The approach to this trends analysis is described
below and the results are presented in the individual state sections (Sections 5 through 28).
In 2009, EPA expanded the list of Core Analytes for the NATTS program to 19
pollutants, as discussed in Section 3.2. For this report, a trends analysis was conducted for the
selected NATTS MQO Core Analytes shown in Table 3-4. This table is very similar to
Table 3-3; however, PAHs were not included in this analysis because they have not been
analyzed for under the NMP for 5 consecutive years. Twenty-five of the 52 sites met the criteria
for 3-year rolling statistical metrics to be calculated.
Table 3-4. NATTS MQO Core Analytes Selected for Trends Analysis
Pollutant
Benzene
1,3 -Butadiene
Acetaldehyde
Formaldehyde
Arsenic
Manganese
Hexavalent Chromium
Class/Method
VOC/TO-15
Carbonyl Compounds/
TO-11A
Metals/IO-3.5
Metals/EPA
The trends figures and analyses are presented as 3-year rolling statistical metrics. The
following criteria were used to calculate valid rolling statistical metrics:
• Analysis performed under the NMP.
• A minimum of at least 5 years of consecutive data.
3-16
-------
The five individual 3-year rolling statistical metrics are presented as box and whisker
plots, or simply boxplots, an example of which can be seen in Figure 5-17. The statistical metrics
shown include the minimum and maximum concentration measured during each 3-year period
(as shown by the upper and lower value of the lines extending from the box); the 5th percentile,
50thpercentile (or median), and 95th percentile (as shown by the y-values corresponding with the
bottom, gray line, or top of the box, respectively); and the 3-year rolling average concentration
(as denoted by the white diamond). Each of the five rolling metrics represents all measurements
from that 3-year period. The use of rolling averages allows for a smoothing of raw data in order
to identify long-term trends (NIST, 2012).
Data used in this analysis were downloaded from EPA's AQS database (EPA, 2012h).
Non-detects are uploaded into AQS as zeros (EPA, 2009b). Similar to other analyses presented
in this report, zeros representing these non-detects were incorporated into the statistical
calculations. Samples with precision data (duplicates, collocates, and/or replicates) were
averaged together to allow for the determination of a single concentration per pollutant per site
per date, reflecting the data treatment described in Section 3.1.
3.5.5 Risk Screening and Pollutants of Interest
The risk screening process described in Section 3.2 and applied at the program-level was
also completed for each individual monitoring site to determine site-specific pollutants of
interest. Once these were determined, the time-period averages (quarterly and annual) described
in Section 3.1 were calculated for each site and these were used for various risk-related analyses
at the site-specific level, as described below.
• Comparison to ATSDR MRLs, as described in Section 3.3, including the
emission tracer analysis described below.
• The calculation of cancer and noncancer surrogate risk approximations.
• Risk-based emissions assessment.
3.5.5.1 Emission Tracer Analysis
The preprocessed daily measurements and time-period average concentrations for each
site-specific pollutant of interest were compared to the ATSDR MRL health benchmarks in the
same fashion described in Section 3.3. To further this analysis, pollution roses were created for
3-17
-------
each of the site-specific pollutants of interest that had preprocessed daily measurements greater
than their respective ATSDR acute MRL health benchmark. This analysis is performed to help
identify the geographical area where the emissions sources of these pollutants may have
originated. A pollution rose is a plot of the ambient concentration versus the wind speed and
direction; high concentrations may be shown in relation to the direction of potential emissions
sources.
3.5.5.2 Cancer and Noncancer Surrogate Risk Approximations
Risk was further examined by calculating cancer and noncancer surrogate risk
approximations for each of the site-specific pollutants of interest. The cancer risk
approximations presented in this report estimate the cancer risk due to exposure at the annual
average concentration over a 70-year period, not the risk resulting from exposure over the time
period covered in this report. A cancer risk approximation less than 1 in-a-million is considered
negligible; a cancer risk greater than 1 in-a-million but less than 100 in-a-million is generally
considered acceptable; and a cancer risk greater than 100 in-a-million is considered significant
(EPA, 2009c). Noncancer risk is presented as the Noncancer Hazard Quotient (HQ). According
to EPA, "If the HQ is calculated to be equal to or less than 1.0, then no adverse health effects are
expected as a result of exposure. If the HQ is greater than 1.0, then adverse health effects are
possible" (EPA, 20lib).
The risk factors applied to calculate cancer and noncancer surrogate risk approximations
are typically UREs or RfCs (respectively), which are developed by EPA. However, UREs and
RfCs are not available for all pollutants. In the absence of EPA values, risk factors developed by
agencies with credible methods and that are similar in scope and definition were used (EPA,
2012d). Cancer URE and noncancer RfC risk factors can be applied to the annual averages to
approximate surrogate chronic risk estimates based on ambient monitoring data. While these risk
approximations do not incorporate human activity patterns and therefore do not reflect true
human inhalation exposure, they may allow analysts to further refine their focus by identifying
concentrations of specific pollutants that may present health risks. Cancer UREs and/or
noncancer RfCs, site-specific annual averages, and corresponding annual average-based
surrogate chronic risk approximations are presented in each state section (Sections 5
through 28).
3-18
-------
3.5.5.3 Risk-Based Emissions Assessment
A pollutant emitted in high quantities does not necessarily present a higher risk to human
health than a pollutant emitted in very low quantities. The more toxic the pollutant, the more risk
associated with its emissions in ambient air. The development of various health-based risk
factors has allowed analysts to apply weight to the emissions of pollutants based on toxicity
rather than mass emissions. This approach considers both a pollutant's toxicity potential and the
quantity emitted.
This assessment compares county-level emissions to toxicity-weighted emissions based
on the EPA-approved approach described below (EPA, 2007). The 10 pollutants with the highest
total mass emissions and the 10 pollutants with the highest associated toxicity-weighted
emissions for pollutants with cancer and noncancer toxicity factors are presented in each state
section. While the absolute magnitude of the pollutant-specific toxicity-weighted emissions is
not meaningful, the relevant magnitude of toxicity-weighted emissions is useful in identifying
the order of potential priority for air quality managers. Higher values suggest greater priority;
however, even the highest values may not reflect potential cancer effects greater than the level of
concern (100 in-a-million) or potential noncancer effects above the level of concern
(e.g., HQ = 1.0). The pollutants exhibiting the 10 highest annual average-based surrogate chronic
cancer and noncancer risk approximations are also presented in each state section. The results of
this data analysis may help state, local, and tribal agencies better understand which pollutants
emitted, from a toxicity basis, are of the greatest concern.
The toxicity-weighted emissions approach consists of the following steps:
1. Obtain HAP emissions data for all anthropogenic sectors from the NEI. For point
sources, sum the process-level emissions to the county-level.
2. Apply the mass extraction speciation profiles to extract metal and cyanide mass. The
only exception is for two chromium species: chromium and chromium compounds.
3. For chromium and chromium compounds, trivalent chromium (non-toxic) must be
separated from hexavalent chromium (toxic). To do this, apply the chromium
speciation profile to extract the hexavalent chromium mass by industry group.
3-19
-------
4. Apply weight to the emissions derived from the steps above based on their toxicity.
a. To apply weight based on cancer toxicity, multiply the emissions of each
pollutant by its cancer URE.
b. To apply weight based on noncancer toxicity, divide the emissions of each
pollutant by its noncancer RfC.
The PAH measured using Method TO-13A are a sub-group of Polycyclic Organic Matter
(POM). Because these compounds are often not speciated into individual compounds in the NEI,
the PAH are grouped into POM Groups in order to assess risk attributable to these pollutants
(EPA, 201 Id). Thus, emissions data and toxicity-weighted emissions for PAH are presented by
POM Groups for this analysis. Table 3-5 presents the 22 PAH measured by Method TO-13A and
their associated POM Groups. The POM groups are sub-grouped in Table 3-5 because toxicity
research has lead to the refining of UREs for certain PAH (EPA, 2012d). Note that naphthalene
emissions are reported to the NEI individually; therefore, naphthalene is not included in one of
the POM Groups. Also note that four additional pollutants analyzed by Method TO-13A and
listed in Table 3-5 do not have assigned POM Groups.
Table 3-5. POM Groups for PAHs
Pollutant
Acenaphthene
Acenaphthylene
Anthracene
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Chrysene
Coronene
Cyclopenta[cd]pyrene
Dibenz(a,h)anthracene
Fluoranthene
Fluorene
9-Fluorenone
Indeno( 1 ,2,3 -cd)pyrene
Naphthalene*
POM Group
Group 2
Group 2
Group 2
POM
Subgroup
Group 2b
Group 2b
Group 2d
Group 6
Group 5
GroupSa
Group 6
Group 2
Group 2
Group 2b
Group 2b
Group 6
Group 7
NA
NA
Group 5
Group 2
Group 2
GroupSb
Group 2b
Group 2b
NA
Group 6
NA
* Naphthalene emissions are reported to the NEI individually;
therefore, naphthalene is not included in one of the POM Groups.
NA = no POM Group assigned.
3-20
-------
Table 3-5. POM Groups for PAHs (Continued)
Pollutant
Perylene
Phenanthrene
Pyrene
Retene
POM Group
Group 2
Group 2
Group 2
NA
POM
Subgroup
Group 2b
Group 2d
Group 2d
NA
* Naphthalene emissions are reported to the NEI individually;
therefore, naphthalene is not included in one of the POM Groups.
NA = no POM Group assigned.
3-21
-------
4.0 Summary of the 2010 National Monitoring Programs Data
This section summarizes the results of the data analyses performed on the NMP dataset
as described in Section 3.
4.1 Statistical Results
This section examines the following statistical parameters for each analytical method:
1) detection rates of the target pollutants, 2) concentration ranges and data distribution, and
3) central tendency statistics. Tables 4-1 through 4-6 present statistical summaries for the target
pollutants and Sections 4.1.1 through 4.1.3 review the basic findings of these statistical
calculations.
4.1.1 Target Pollutant Detection Rates
Every pollutant has an MDL, as described in Section 2.2. Quantification below the MDL
is possible, although the measurement's reliability is lower. If a concentration does not exceed
the MDL, it does not mean that the pollutant is not present in the air. If the instrument does not
generate a numerical concentration, the measurement is marked as "ND," or "non-detect." As
explained in Section 2.2, data analysts should exercise caution when interpreting monitoring data
with a high percentage of reported concentrations at levels near or below the corresponding
MDLs. A thorough review of the number of measured detections, the number of non-detects, and
the total number of samples is beneficial to understanding the representativeness of the
interpretations made.
Tables 4-1 through 4-6 summarize the number of times the target pollutants were
detected out of the number of valid samples collected and analyzed. Approximately 55 percent of
the reported measurements (based on the preprocessed daily measurements) were above the
MDLs. The following list shows the percent of measurements that were above the MDLs:
• 42.8 percent of VOC
• 53.8 percent of SNMOC
• 80.9 percent of carbonyl compounds
• 64.7 percent of PAH
4-1
-------
• 75.0 percent of metals
• 72.2 percent of hexavalent chromium samples.
Some pollutants were always detected while others were infrequently detected or not
detected at all. Similar to previous years' reports, acetaldehyde, formaldehyde, and acetone had
the greatest number of measured detections (1,499), using the preprocessed daily measurements.
These pollutants were reported in every valid carbonyl compound sample collected (1,499).
Benzene, chloromethane, dichlorodifluoromethane, propylene, toluene, trichlorofluoromethane,
and trichlorotrifluoroethane were detected in every VOC sample collected (1,264). Twelve
pollutants, including acetylene, benzene, ethylene, and toluene, were detected in every SNMOC
sample collected (468). Naphthalene, phenanthrene, pyrene, and fluoranthene were detected in
every valid PAH sample collected (1,354). Antimony, chromium, lead, manganese, and nickel
were detected in every metal sample collected (811). Hexavalent chromium was detected in 933
(out of 1,291) samples.
Similar to previous years' reports, BTUT and NBIL had the greatest number of measured
detections (7,174 for BTUT and 6,768 for NBIL). They were also the only two sites that
collected samples for all six analytical methods/pollutant groups. Yet, the detection rates for
these sites (69 and 66 percent, respectively) were not as high as other sites. Detection rates for
sites that sampled suites of pollutants that are frequently detected tended to be higher (refer to
the list of method-specific percentages of measurements above the MDL listed above). For
example, metals were rarely reported as non-detects. As a result, sites that sampled only metals
(such as PAFL) would be expected to have higher detection rates. PAFL's detection rate is 100
percent. Conversely, VOCs had the lowest percentage of concentrations greater than the MDLs
(42.5 percent). A site measuring only VOC would be expected to have lower detection rates,
such as SPAZ (50.5 percent).
4-2
-------
Table 4-1. Statistical Summaries of the VOC Concentrations
Pollutant
Acetonitrile
Acetylene
Acrolein
Acrylonitrile
tert-Amyl Methyl Ether
Benzene
Bromochloromethane
Bromodichloromethane
Bromoform
Bromomethane
1,3 -Butadiene
Carbon Tetrachloride
Carbon Bisulfide
Chlorobenzene
Chloroethane
Chloroform
Chloromethane
Chloromethylbenzene
Chloroprene
Dibromochloromethane
1,2-Dibromoethane
/w-Dichlorobenzene
o-Dichlorobenzene
£>-Dichlorobenzene
Dichlorodifluoromethane
1 , 1 -Dichloroethane
1,2-Dichloroethane
1, 1-Dichloroethene
#of
Measured
Detections1
1,260
1,263
1,241
151
4
1,264
6
71
13
1,040
1,033
1,219
1,258
7
347
1,021
1,264
7
11
110
21
72
31
725
1,264
13
214
18
#
of Non-
Detects1
4
1
23
1,113
1,260
0
1,258
1,193
1,251
224
231
45
6
1,257
917
243
0
1,257
1,253
1,154
1,243
1,192
1,233
539
0
1,251
1,050
1,246
Minimum2
(ppbv)
0.036
0.079
0.053
0.010
0.003
0.041
0.010
0.003
0.001
0.008
0.004
0.004
0.005
0.005
0.006
0.010
0.031
0.006
0.013
0.001
0.002
0.002
0.002
0.002
0.022
0.002
0.010
0.005
Maximum
(ppbv)
5,530
9.08
46.2
1.36
0.010
2.17
0.013
1.31
0.014
0.884
0.409
10.9
0.193
0.029
0.217
2.47
3.18
0.011
0.090
0.200
0.011
0.109
0.018
3.52
1.30
0.013
0.061
0.030
Arithmetic
Mean
(ppbv)
26.3
0.862
0.590
0.017
0.00002
0.311
0.00005
0.004
0.00006
0.013
0.038
0.567
0.099
0.00008
0.007
0.038
0.631
0.00005
0.0004
0.001
0.0001
0.0008
0.0002
0.019
0.578
0.00009
0.003
0.0002
Median
(ppbv)
0.361
0.659
0.306
0
0
0.245
0
0
0
0.011
0.026
0.037
0.102
0
0
0.020
0.613
0
0
0
0
0
0
0.007
0.579
0
0
0
Mode
(ppbv)
0.190
1.10
0
0
0
0.260
0
0
0
0
0
0.010
0.110
0
0
0
0.560
0
0
0
0
0
0
0
0.530
0
0
0
First
Quartile
(ppbv)
0.197
0.424
0.196
0
0
0.173
0
0
0
0.009
0.012
0.013
0.086
0
0
0.014
0.567
0
0
0
0
0
0
0
0.534
0
0
0
Third
Quartile
(ppbv)
3.09
1.01
0.483
0
0
0.373
0
0
0
0.014
0.048
0.272
0.116
0
0.010
0.031
0.680
0
0
0
0
0
0
0.019
0.619
0
0
0
Standard
Deviation
(ppbv)
184
0.753
2.09
0.084
0.0005
0.223
0.001
0.047
0.001
0.028
0.044
1.38
0.025
0.001
0.017
0.119
0.122
0.001
0.005
0.009
0.001
0.005
0.001
0.105
0.070
0.001
0.008
0.001
Coefficient
of
Variation
6.97
0.874
3.54
5.11
18.9
0.716
14.6
10.7
12.0
2.23
1.17
2.43
0.254
15.2
2.33
3.10
0.193
13.6
12.7
7.17
8.08
5.68
7.07
5.55
0.122
10.4
2.32
9.69
1 Out of 1,264 valid samples.
2 Excludes zeros for non-detects.
-------
Table 4-1. Statistical Summaries of the VOC Concentrations (Continued)
Pollutant
c/'s-l,2-Dichloroethylene
/ra«5-l,2-Dichloroethylene
Dichloromethane
1 ,2-Dichloropropane
cis- 1 , 3 -Dichloropropene
trans- 1 , 3 -Dichloropropene
Dichlorotetrafluoroethane
Ethyl Aery late
Ethyl tort-Butyl Ether
Ethylbenzene
Hexachloro- 1 ,3 -butadiene
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Methyl Methacrylate
Methyl fer/-Butyl Ether
w-Octane
Propylene
Styrene
1, 1,2,2-Tetrachloroethane
Tetrachloroethylene
Toluene
1 ,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1, 1,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Trichlorotrifluoroethane
1,2,4-Trimethylbenzene
#of
Measured
Detections1
14
21
1,263
2
3
0
1,263
7
19
1,258
22
1,263
1,102
54
27
1,098
1,264
1,106
12
1,054
1,264
44
1,067
11
297
1,264
1,264
1,243
#
of Non-
Detects1
1,250
1,243
1
1,262
1,261
1,264
1
1,257
1,245
6
1,242
1
162
1,210
1,237
166
0
158
1,252
210
0
1,220
197
1,253
967
0
0
21
Minimum2
(ppbv)
0.035
0.007
0.023
0.011
0.004
Maximum
(ppbv)
0.150
0.207
1,510
0.012
0.009
Arithmetic
Mean
(ppbv)
0.001
0.0004
3.05
0.00002
0.00002
Median
(ppbv)
0
0
0.124
0
0
Mode
(ppbv)
0
0
0.080
0
0
First
Quartile
(ppbv)
0
0
0.086
0
0
Third
Quartile
(ppbv)
0
0
0.221
0
0
Standard
Deviation
(ppbv)
0.013
0.007
49.9
0.0005
0.0004
Coefficient
of
Variation
9.97
15.5
16.3
25.1
21.5
Not Detected
0.006
0.020
0.002
0.005
0.002
0.030
0.005
0.009
0.002
0.006
0.090
0.005
0.005
0.004
0.033
0.001
0.006
0.004
0.001
0.009
0.007
0.003
0.088
0.050
0.029
7.25
0.033
9.03
0.389
0.390
0.054
1.80
66.0
15.9
0.012
0.199
22.7
0.067
0.185
0.014
1.79
2.07
0.301
0.562
0.020
0.0002
0.0002
0.082
0.0001
0.502
0.038
0.003
0.0003
0.046
0.560
0.072
0.00009
0.025
0.585
0.0004
0.010
0.00009
0.011
0.295
0.094
0.072
0.018
0
0
0.053
0
0.397
0.030
0
0
0.031
0.367
0.022
0
0.016
0.350
0
0.010
0
0
0.281
0.093
0.049
0.018
0
0
0.020
0
0.190
0
0
0
0
0.290
0
0
0
0.240
0
0.010
0
0
0.260
0.090
0.030
0.017
0
0
0.030
0
0.256
0.018
0
0
0.018
0.260
0.014
0
0.008
0.183
0
0.009
0
0
0.262
0.087
0.026
0.020
0
0
0.100
0
0.616
0.050
0
0
0.053
0.572
0.039
0
0.030
0.707
0
0.012
0
0
0.306
0.101
0.089
0.010
0.003
0.002
0.216
0.001
0.443
0.036
0.021
0.003
0.084
1.92
0.573
0.0009
0.029
1.06
0.004
0.008
0.001
0.073
0.089
0.013
0.073
0.519
14.3
9.59
2.63
10.5
0.883
0.935
7.77
8.82
1.84
3.42
7.99
10.6
1.18
1.81
8.32
0.816
11.1
6.40
0.302
0.135
1.02
Out of 1,264 valid samples.
2 Excludes zeros for non-detects.
-------
Table 4-1. Statistical Summaries of the VOC Concentrations (Continued)
Pollutant
1,3,5 -Trimethylbenzene
Vinyl chloride
m,p-Xylene
o-Xylene
#of
Measured
Detections1
1,195
53
1,256
1,247
#
of Non-
Detects1
69
1,211
8
17
Minimum2
(ppbv)
0.002
0.002
0.009
0.004
Maximum
(ppbv)
0.187
0.046
2.70
0.888
Arithmetic
Mean
(ppbv)
0.025
0.0004
0.198
0.075
Median
(ppbv)
0.019
0
0.126
0.051
Mode
(ppbv)
0.010
0
0.020
0.040
First
Quartile
(ppbv)
0.011
0
0.061
0.027
Third
Quartile
(ppbv)
0.031
0
0.260
0.099
Standard
Deviation
(ppbv)
0.023
0.002
0.220
0.075
Coefficient
of
Variation
0.937
6.21
1.11
0.997
1 Out of 1,264 valid samples.
2 Excludes zeros for non-detects.
-------
Table 4-2. Statistical Summaries of the SNMOC Concentrations
Pollutant
Acetylene
Benzene
1,3 -Butadiene
w-Butane
c/s-2-Butene
/ra«s-2-Butene
Cyclohexane
Cyclopentane
Cyclopentene
w-Decane
1-Decene
/w-Diethylbenzene
£>-Diethylbenzene
2,2-Dimethylbutane
2,3 -Dimethylbutane
2,3 -Dimethylpentane
2,4-Dimethylpentane
w-Dodecane
1-Dodecene
Ethane
2-Ethyl-l-butene
Ethylbenzene
Ethylene
/w-Ethyltoluene
o-Ethyltoluene
£>-Ethyltoluene
w-Heptane
#of
Measured
Detections1
468
468
164
425
415
396
463
463
135
446
4
282
184
466
442
460
453
457
377
468
7
462
468
432
404
442
466
#
of Non-
Detects1
0
0
304
43
53
72
5
5
333
22
464
186
284
2
26
8
15
11
91
0
461
6
0
36
64
26
2
Minimum2
(ppbC)
0.155
0.180
0.061
0.599
0.067
0.056
0.069
0.094
0.057
0.072
0.234
0.069
0.061
0.099
0.088
0.116
0.069
0.057
0.065
1.76
0.182
0.096
0.579
0.057
0.036
0.051
0.101
Maximum
(ppbC)
10.9
7.89
4.91
162
6.06
6.69
19.2
26.4
2.43
5.84
1.49
1.82
3.18
3.98
7.67
4.97
2.12
12.2
2.10
549
8.70
65.1
33.3
2.89
1.59
1.31
12.5
Arithmetic
Mean
(ppbC)
1.74
1.97
0.114
17.2
0.263
0.280
2.50
0.734
0.111
0.641
0.008
0.162
0.186
0.735
1.07
1.03
0.496
0.413
0.296
55.3
0.035
1.04
2.87
0.355
0.266
0.267
2.18
Median
(ppbC)
1.24
1.54
0
10.3
0.184
0.143
1.35
0.499
0
0.442
0
0.117
0
0.579
0.808
0.964
0.428
0.249
0.214
23.0
0
0.416
2.30
0.298
0.225
0.235
1.30
Mode
(ppbC)
1.04
1.54
0
0
0
0
0
0
0
0
0
0
0
1.25
0
0
0
0
0
114
0
0
1.44
0
0
0
1.52
First
Quartile
(ppbC)
0.820
0.926
0
3.38
0.131
0.095
0.257
0.270
0
0.212
0
0
0
0.324
0.277
0.610
0.229
0.149
0.106
5.67
0
0.258
1.72
0.163
0.129
0.157
0.310
Third
Quartile
(ppbC)
2.07
2.63
0.121
24.6
0.285
0.278
3.94
0.885
0.112
0.853
0.000
0.208
0.137
0.965
1.52
1.33
0.637
0.458
0.390
77.6
0.000
0.648
3.16
0.469
0.369
0.333
3.45
Standard
Deviation
(ppbC)
1.51
1.39
0.329
19.9
0.394
0.526
2.96
1.32
0.302
0.652
0.097
0.235
0.486
0.561
1.03
0.595
0.359
0.719
0.320
73.7
0.466
4.57
2.67
0.295
0.213
0.181
2.29
Coefficient
of
Variation
0.866
0.705
2.89
1.15
1.50
1.87
1.18
1.80
2.72
1.02
12.6
1.45
2.61
0.763
0.961
0.576
0.724
1.74
1.08
1.33
13.3
4.41
0.929
0.832
0.802
0.678
1.05
1 Out of 468 valid samples.
2 Excludes zeros for non-detects.
-------
Table 4-2. Statistical Summaries of the SNMOC Concentrations (Continued)
Pollutant
1-Heptene
c/s-2-Hexene
/raws-2-Hexene
w-Hexane
1-Hexene
Isobutane
Isobutene/ 1 -Butene
Isopentane
Isoprene
Isopropylbenzene
2-Methyl-l-butene
3 -Methyl- 1 -butene
2-Methyl- 1 -pentene
4-Methy 1- 1 -pentene
2-Methyl-2-butene
Methylcyclohexane
Methylcyclopentane
2-Methylheptane
3-Methylheptane
2-Methylhexane
3 -Methy Ihexane
2-Methylpentane
3-Methylpentane
w-Nonane
1-Nonene
w-Octane
1-Octene
w-Pentane
#of
Measured
Detections1
315
45
40
468
439
468
441
465
434
311
259
6
19
141
265
466
468
437
415
426
421
452
468
456
334
461
250
468
#
of Non-
Detects1
153
423
428
0
29
0
27
o
J
34
157
209
462
449
327
203
2
0
31
53
42
47
16
0
12
134
7
218
0
Minimum2
(ppbC)
0.082
0.093
0.076
0.159
0.075
0.245
0.184
0.396
0.047
0.068
0.049
0.095
0.055
0.088
0.050
0.097
0.120
0.077
0.055
0.190
0.039
0.140
0.135
0.064
0.073
0.090
0.029
0.173
Maximum
(ppbC)
4.18
0.527
0.533
40.4
1.05
262
32.1
868
10.4
3.44
20.2
0.514
0.402
1.03
54.5
34.5
19.3
3.54
2.70
10.2
10.4
35.4
17.6
4.02
1.77
13.4
1.17
6,315
Arithmetic
Mean
(ppbC)
0.628
0.019
0.015
4.17
0.230
15.1
2.09
15.8
0.665
0.106
0.292
0.003
0.008
0.094
0.406
4.84
2.37
0.698
0.527
1.58
1.70
4.73
2.39
0.764
0.155
1.50
0.144
23.0
Median
(ppbC)
0.342
0
0
2.56
0.223
7.25
0.663
8.78
0.287
0.113
0.104
0
0
0
0.105
2.26
1.52
0.499
0.35
1.22
1.30
3.51
1.75
0.465
0.133
0.844
0.099
6.11
Mode
(ppbC)
0
0
0
1.16
0
26.6
0
14.2
0
0
0
0
0
0
0
10.1
0.423
0
0
0
0
0
1.24
0
0
0
0
10.1
First
Quartile
(ppbC)
0
0
0
0.695
0.175
1.42
0.464
3.98
0.159
0
0
0
0
0
0
0.362
0.518
0.171
0.136
0.574
0.506
1.41
0.609
0.184
0
0.224
0
1.99
Third
Quartile
(ppbC)
1.04
0
0
6.23
0.287
21.2
1.41
17.9
0.593
0.147
0.287
0
0
0.171
0.246
7.75
3.51
1.04
0.774
2.27
2.52
6.73
3.41
1.04
0.209
2.19
0.241
13.2
Standard
Deviation
(ppbC)
0.743
0.065
0.057
4.83
0.114
22.3
4.48
43.3
1.11
0.183
1.13
0.027
0.040
0.172
3.43
5.87
2.47
0.658
0.512
1.37
1.54
4.42
2.38
0.780
0.184
1.71
0.186
291
Coefficient
of
Variation
1.18
3.49
3.83
1.16
0.493
1.48
2.14
2.74
1.67
1.72
3.88
10.9
5.34
1.83
8.45
1.21
1.05
0.943
0.972
0.864
0.907
0.936
0.995
1.02
1.19
1.14
1.29
12.6
1 Out of 468 valid samples.
2 Excludes zeros for non-detects.
-------
Table 4-2. Statistical Summaries of the SNMOC Concentrations (Continued)
Pollutant
1-Pentene
c/s-2-Pentene
/ra«s-2-Pentene
a-Pinene
6-Pinene
Propane
w-Propylbenzene
Propylene
Propyne
Styrene
Toluene
w-Tridecane
1-Tridecene
1 ,2,3 -Trimethylbenzene
1 ,2,4-Trimethylbenzene
1 ,3 ,5-Trimethylbenzene
2,2,3 -Trimethylpentane
2,2,4-Trimethylpentane
2,3 ,4-Trimethylpentane
w-Undecane
1-Undecene
7w-Xylene//?-Xylene
o-Xylene
SNMOC (Sum of Knowns)
Sum of Unknowns
TNMOC
#of
Measured
Detections1
464
354
412
347
59
468
405
468
0
209
468
50
23
292
465
416
291
356
408
458
179
467
455
468
468
468
#
of Non-
Detects1
4
114
56
121
409
0
63
0
468
259
0
418
445
176
3
52
177
112
60
10
289
1
13
0
0
0
Minimum2
(ppbC)
0.090
0.043
0.051
0.052
0.075
0.168
0.073
0.376
Maximum
(ppbC)
26.8
19.2
4.67
18.9
2.70
389
2.89
7.13
Arithmetic
Mean
(ppbC)
0.619
0.213
0.249
0.310
0.043
38.1
0.192
1.22
Median
(ppbC)
0.238
0.130
0.172
0.198
0
19.8
0.175
0.958
Mode
(ppbC)
0.192
0
0
0
0
60.1
0
1.03
First
Quartile
(ppbC)
0.176
0.076
0.122
0
0
6.10
0.119
0.761
Third
Quartile
(ppbC)
0.350
0.183
0.314
0.334
0
52.1
0.242
1.36
Standard
Deviation
(ppbC)
2.17
1.12
0.318
0.945
0.186
46.7
0.188
0.858
Coefficient
of
Variation
3.50
5.25
1.28
3.05
4.34
1.23
0.980
0.701
Not Detected
0.062
0.196
0.051
0.065
0.043
0.067
0.054
0.099
0.083
0.059
0.035
0.043
0.100
0.086
17.4
11.9
34.1
105
63.1
3.58
3.58
20.6
2.85
1.71
1.46
7.41
2.65
15.6
4.99
14.6
4.32
7,053
849
7,535
0.390
4.20
0.033
0.016
0.196
0.565
0.353
0.205
0.548
0.249
0.598
0.239
2.04
0.560
228
82.7
311
0
3.31
0
0
0.115
0.464
0.274
0.170
0.312
0.165
0.388
0
1.62
0.448
136
51.9
231
0
1.33
0
0
0
0.563
0
0
0
0
0
0
1.54
0
263
69.1
103
0
1.47
0
0
0
0.279
0.149
0
0.101
0.109
0.203
0
0.601
0.240
51.6
34.2
102
0.157
5.59
0
0
0.191
0.740
0.484
0.303
0.734
0.269
0.697
0.192
2.89
0.741
293
91.7
397
4.87
4.40
0.192
0.170
0.978
0.424
0.298
0.232
0.756
0.302
0.926
0.573
1.91
0.464
386
102
423
12.5
1.05
5.88
10.8
5.00
0.751
0.845
1.13
1.38
1.21
1.55
2.40
0.939
0.829
1.69
1.23
1.36
oo
1 Out of 468 valid samples.
2 Excludes zeros for non-detects.
-------
Table 4-3. Statistical Summaries of the Carbonyl Compound Concentrations
Pollutant
Acetaldehyde
Acetone
Benzaldehyde
Butyraldehyde
Crotonaldehyde
2,5-Dimethylbenzaldehyde
Formaldehyde
Hexaldehyde
Isovaleraldehyde
Propionaldehyde
Tolualdehydes
Valeraldehyde
#of
Measured
Detections1
1,499
1,499
1,481
1,496
1,486
0
1,499
1,473
32
1,486
1,164
1,438
#
of Non-
Detects1
0
0
18
3
13
1,499
0
26
1,467
13
335
61
Minimum2
(ppbv)
0.089
0.055
0.006
0.006
0.006
Maximum
(ppbv)
5.99
7.40
0.244
0.859
1.12
Arithmetic
Mean
(ppbv)
1.06
1.20
0.029
0.091
0.104
Median
(ppbv)
0.893
1.03
0.023
0.074
0.044
Mode
(ppbv)
1.03
1.01
0.017
0.077
0.024
First
Quartile
(ppbv)
0.597
0.659
0.016
0.049
0.026
Third
Quartile
(ppbv)
1.29
1.56
0.034
0.111
0.112
Standard
Deviation
(ppbv)
0.718
0.801
0.024
0.069
0.153
Coefficient
of
Variation
0.677
0.668
0.806
0.756
1.47
Not Detected
0.136
0.005
0.008
0.006
0.006
0.004
43.5
0.885
0.108
1.01
0.224
0.567
2.01
0.032
0.001
0.118
0.023
0.029
1.66
0.023
0
0.097
0.020
0.021
1.67
0.014
0
0.076
0
0
1.09
0.014
0
0.063
0.009
0.013
2.47
0.036
0
0.150
0.034
0.034
1.80
0.046
0.006
0.089
0.021
0.036
0.894
1.46
8.77
0.753
0.911
1.22
1 Out of 1,499 valid samples.
2 Excludes zeros for non-detects.
-------
Table 4-4. Statistical Summaries of the PAH Concentrations
Pollutant
Acenaphthene
Acenaphthylene
Anthracene
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Chrysene
Coronene
Cyclopenta[cd]pyrene
Dibenz(a,h)anthracene
Fluoranthene
Fluorene
9-Fluorenone
Indeno( 1 ,2,3 -cd)pyrene
Naphthalene
Perylene
Phenanthrene
Pyrene
Retene
#of
Measured
Detections1
1,332
624
793
627
683
1,184
1,044
1,198
721
1,288
874
228
196
1,354
1,353
1,347
933
1,354
517
1,354
1,354
1,304
#
of Non-
Detects1
22
730
561
727
671
170
310
156
633
66
480
1,126
1,158
0
1
7
421
0
837
0
0
50
Minimum2
(ng/m3)
0.090
0.029
0.031
0.010
0.007
0.013
0.011
0.009
0.007
0.013
0.009
0.007
0.012
0.030
0.078
0.048
0.014
0.309
0.015
0.057
0.022
0.023
Maximum
(ng/m3)
175
175
69.1
35.8
42.7
38.7
22.2
29.2
13.5
28.8
13.6
55.1
3.52
111
152
25
30.7
1,490
6.88
239
154
61.1
Arithmetic
Mean
(ng/m3)
3.98
0.872
0.479
0.123
0.131
0.274
0.142
0.169
0.072
0.266
0.074
0.066
0.012
2.40
4.82
1.63
0.145
95.3
0.055
9.63
1.51
0.360
Median
(ng/m3)
2.04
0
0.138
0
0.020
0.110
0.059
0.071
0.022
0.128
0.037
0
0
1.29
2.91
1.07
0.055
66.4
0
5.51
0.813
0.169
Mode
(ng/m3)
0
0
0
0
0
0
0
0
0
0
0
0
0
1.05
2.26
1.47
0
104
0
5.40
1.33
0
First
Quartile
(ng/m3)
0.983
0
0
0
0
0.047
0.022
0.036
0
0.074
0
0
0
0.694
1.75
0.634
0
36.6
0
2.98
0.440
0.098
Third
Quartile
(ng/m3)
4.30
0.540
0.428
0.086
0.100
0.279
0.146
0.161
0.071
0.271
0.074
0
0
2.57
5.39
2
0.140
117
0.047
11.2
1.56
0.307
Standard
Deviation
(ng/m3)
7.69
6.38
2.76
1.14
1.32
1.27
0.696
0.947
0.440
0.955
0.429
1.51
0.106
4.78
8.09
1.77
0.966
117
0.227
14.4
5.44
1.72
Coefficient
of
Variation
1.93
7.32
5.77
9.25
10.0
4.62
4.89
5.61
6.11
3.59
5.77
23.0
8.89
1.99
1.68
1.09
6.68
1.23
4.10
1.49
3.61
4.77
1 Out of 1,354 valid samples.
2 Excludes zeros for non-detects.
-------
Table 4-5. Statistical Summaries of the Metals Concentrations
Pollutant
Antimony (PM10)
Arsenic (PM10)
Beryllium (PM10)
Cadmium (PM10)
Chromium (PM10)
Cobalt (PM10)
Lead (PM10)
Manganese (PM10)
Mercury (PM10)
Nickel (PM10)
Selenium (PM10)
Antimony (TSP)
Arsenic (TSP)
Beryllium (TSP)
Cadmium (TSP)
Chromium (TSP)
Cobalt (TSP)
Lead (TSP)
Manganese (TSP)
Mercury (TSP)
Nickel (TSP)
Selenium (TSP)
#of
Measured
Detections1'2
507
504
393
506
507
504
507
507
421
507
504
304
304
303
304
304
304
304
304
299
304
304
#
of Non-
Detects1'2
0
3
114
1
0
o
J
0
0
86
0
3
0
0
1
0
0
0
0
0
5
0
0
Minimum
(ng/m3)3
0.030
0.010
0.00005
0.010
0.260
0.000002
0.080
0.280
0.00003
0.070
0.001
0.079
0.113
0.000055
0.033
0.832
0.055
0.593
1.79
0.000875
0.211
0.076
Maximum
(ng/m3)
9.60
4.77
0.040
2.05
10.9
1.68
47.8
200
0.290
10.6
3.95
16.7
2.24
0.081
1.64
10.3
8.02
26.0
60.4
0.064
7.82
2.21
Arithmetic
Mean
(ng/m3)
1.07
0.558
0.003
0.164
2.26
0.127
3.67
6.82
0.012
1.06
0.453
0.653
0.544
0.016
0.175
1.75
0.532
3.20
14.8
0.010
0.877
0.588
Median
(ng/m3)
0.757
0.415
0.002
0.084
2.00
0.077
2.32
4.03
0.007
0.845
0.330
0.450
0.457
0.012
0.130
1.58
0.323
2.57
13.6
0.009
0.732
0.536
Mode
(ng/m3)
0.200
0.240
0
0.040
1.79
0.060
1.56
11.0
0
1.01
0.200
0.267
0.339
0.011
0.122
1.49
0.254
2.36
13.7
0.007
0.299
1.03
First
Quartile
(ng/m3)
0.460
0.240
0.0003
0.050
1.66
0.048
1.45
2.15
0.003
0.594
0.184
0.292
0.341
0.008
0.088
1.35
0.212
1.83
7.55
0.006
0.514
0.347
Third
Quartile
(ng/m3)
1.28
0.700
0.004
0.176
2.92
0.140
3.74
7.97
0.013
1.22
0.554
0.712
0.667
0.020
0.208
1.91
0.552
3.87
19.0
0.013
1.03
0.789
Standard
Deviation
(ng/m3)
1.08
0.535
0.005
0.238
0.907
0.177
4.95
11.6
0.021
0.915
0.435
1.14
0.311
0.013
0.172
0.816
0.722
2.26
9.61
0.007
0.647
0.331
Coefficient
of
Variation
1.01
0.959
1.42
1.45
0.401
1.39
1.35
1.71
1.71
0.861
0.959
1.75
0.572
0.828
0.982
0.466
1.36
0.707
0.651
0.678
0.738
0.563
1 For PM10, out of 507 valid samples.
2 For TSP, out of 304 valid samples.
3 Excludes zeros for non-detects.
-------
Table 4-6. Statistical Summaries of the Hexavalent Chromium Concentrations
Pollutant
Hexavalent Chromium
#of
Measured
Detections1
933
#
of Non-
Detects1
358
Minimum
(ng/m3)2
0.0024
Maximum
(ng/m3)
3.51
Arithmetic
Mean
(ng/m3)
0.037
Median
(ng/m3)
0.018
Mode
(ng/m3)
0
First
Quartile
(ng/m3)
0
Third
Quartile
(ng/m3)
0.032
Standard
Deviation
(ng/m3)
0.129
Coefficient
of
Variation
3.49 |
1 Out of 1,291 valid samples.
2 Excludes zeros for non-detects.
to
-------
4.1.2 Concentration Range and Data Distribution
The concentrations measured during the 2010 NMP show a wide range of variability. The
minimum and maximum concentration measured (excluding zeros substituted for non-detects)
for each target pollutant are presented in Tables 4-1 through 4-6 (in respective pollutant group
units). Some pollutants, such as acetonitrile, had a wide range of concentrations measured, while
other pollutants, such as carbon tetrachloride, did not, even though they were both detected
frequently. The pollutant for each method-specific pollutant group with the largest range in
measured concentrations is as follows:
• For VOC, acetonitrile (0.036 to 5,530 ppbv)
• For SNMOC, w-pentane (0.173 to 6,315 ppbC)
• For carbonyl compounds, formaldehyde (0.136 to 43.5 ppbv)
• For PAH, naphthalene (0.309 to 1,490 ng/m3)
• For metals, both size fractions, manganese (0.28 to 200 ng/m3 for PMio and 1.79 to
60.35 ng/m3 for TSP)
• For hexavalent chromium, 0.0024 to 3.51 ng/m3.
A large number of monitoring sites that sampled for hexavalent chromium measured
elevated concentrations on July 4, 2006. Hexavalent chromium is a component in fireworks
(NLM, 2012) and it is possible that Independence Day fireworks celebrations may have caused
this increased concentration level. Based on the l-in-6 sampling schedule for 2010, samples
were collected on July 1 and July 7, thereby missing any potential holiday celebrations. For
2010, the maximum hexavalent chromium concentrations were measured on various days across
the calendar year. Three sites measured their maximum hexavalent chromium in January, one in
February, two in April, two in June, six in July, two in August, one in September, two in
October, one in November, and three in December. One site (S4MO) took a measurement on
July 4 because they were making up a missed sample; the maximum hexavalent chromium
concentration for this site was from the July 4 sample. S4MO is one of the six sites that
measured its highest hexavalent chromium concentration in July. Additional examples of this
phenomenon can be seen in the site-specific trends analysis section of the individual state
sections. Additional studies of this phenomenon were recommended in the 2006 UATMP
4-13
-------
Report. The next year hexavalent chromium is scheduled to be sampled on Independence Day is
July 4, 2014.
4.1.3 Central Tendency
In addition to the number of measured detections and the concentration ranges,
Tables 4-1 through 4-6 also present a number of central tendency and data distribution statistics
(arithmetic mean, mode, median, first and third quartiles, standard deviation, and coefficient of
variation) for each of the pollutants sampled during the 2010 NMP by respective pollutant group
units. A multitude of observations can be made from these tables. The pollutants with the three
highest average concentrations, by mass, for each pollutant group are provided below, with
respective confidence intervals, although the 95 percent confidence interval is not provided in
the table.
The top three VOC by average mass concentration, as presented in Table 4-1, are:
• acetonitrile (26.3 ±10.1 ppbv)
• dichloromethane (3.05 ± 2.75 ppbv)
• acetylene (0.862 ± 0.042 ppbv).
The top three SNMOC by average mass concentration, as presented in Table 4-2, are:
• ethane (55.3 ± 6.69 ppbC)
• propane (38.1 ± 4.24 ppbC)
• w-pentane (23.0 ± 26.46 ppbC).
The top three carbonyl compounds by average mass concentration, as presented in
Table 4-3, are:
• formaldehyde (2.01 ± 0.09 ppbv)
• acetone (1.20 ± 0.04 ppbv)
• acetaldehyde (1.06 ± 0.04 ppbv).
The top three PAH by average mass concentration, as presented in Tables 4-4, are:
• naphthalene (95.3 ±6.25 ng/m3)
4-14
-------
• phenanthrene (9.63 ± 0.77 ng/m3)
• fluorene (4.82 ± 0.43 ng/m3).
The top three metals by average mass concentration for both PMio and TSP fractions, as
presented in Table 4-5, are;
• manganese (PMW = 6.82 ± 1.01 ng/m3, TSP = 14.8 ± 1.08 ng/m3)
• lead (PMio = 3.67 ± 0.43 ng/m3, TSP = 3.20 ± 0.26 ng/m3)
• total chromium (PMio = 2.26 ± 0.08 ng/m3, TSP = 1.75 ± 0.09 ng/m3).
The average mass concentration of hexavalent chromium, as presented in Table 4-6, is
0.037 ± 0.007 ng/m3.
Appendices J through O present similar statistical calculations on a site-specific basis.
4.2 Preliminary Risk Screening and Pollutants of Interest
Based on the preliminary risk screening approach described in Section 3.2, Table 4-7
identifies the pollutants that failed at least one screen; summarizes each pollutant's total number
of measured detections, percentage of screens failed, and cumulative percentage of failed
screens; and highlights those pollutants contributing to the top 95 percent of failed screens
(shaded in gray) and thereby designated as program-wide pollutants of interest.
The results in Table 4-7 are listed in descending order by number of screens failed.
Table 4-7 shows that benzene failed the highest number of screens (1,500), although
formaldehyde and acetaldehyde were not far behind (1,499 and 1,467, respectively). These three
pollutants were also among those with the highest number of measured detections. Conversely,
vinyl chloride and xylenes failed only one screen each. The number of measured detections for
these three pollutants varied significantly. Xylenes were detected in 1,498 samples while vinyl
chloride was detected in only 53 samples (both out of 1,501).
4-15
-------
Table 4-7. Program-Level Risk Screening Summary
Pollutant
Benzene
Formaldehyde
Acetaldehyde
Carbon Tetrachloride
Naphthalene
1,3-Butadiene
Arsenic
Manganese
Ethylbenzene
p-Dichlorobenzene
1 ,2-Dichloroethane
Acrylonitrile
Fluorene
Acenaphthene
Hexavalent Chromium
Trichloroethylene
Nickel
Propionaldehyde
Cadmium
Fluoranthene
Benzo(a)pyrene
Dichloromethane
1,2-Dibromoethane
Lead
Hexachloro- 1 ,3 -butadiene
1, 1,2,2-Tetrachloroethane
Chloroprene
Acenaphthylene
Chloromethylbenzene
Bromomethane
1, 1,2-Trichloroethane
Benzo(a)anthracene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Chloroform
Dibenz(a,h)anthracene
Indeno( 1,2,3 -cd)pyrene
Vinyl chloride
Xylenes
Total
Screening
Value
(Hg/m3)
0.13
0.077
0.45
0.17
0.029
0.03
0.00023
0.005
0.4
0.091
0.038
0.015
0.011
0.011
0.000083
0.2
0.0021
0.8
0.00056
0.011
0.00057
7.7
0.0017
0.015
0.045
0.017
0.0021
0.011
0.02
0.5
0.0625
0.0057
0.0057
0.011
0.011
0.0057
9.8
0.00052
0.0057
0.11
10
#of
Failed
Screens
,500
,499
,467
,244
,082
,013
672
478
391
367
214
151
100
92
87
55
52
39
37
35
33
33
21
15
13
12
11
8
7
5
3
2
2
2
2
2
2
2
2
1
1
10,754
#of
Measured
Detections
1,501
1,499
1,499
1,258
1,354
1,115
808
811
1,494
725
214
151
1,353
1,332
933
297
811
1,486
810
1,354
683
1,263
21
811
22
12
11
624
7
1,040
11
627
1,184
1,044
1,198
721
1,021
196
933
53
1,498
33,785
%of
Failed
Screens
99.93
100.00
97.87
98.89
79.91
90.85
83.17
58.94
26.17
50.62
100.00
100.00
7.39
6.91
9.32
18.52
6.41
2.62
4.57
2.58
4.83
2.61
100.00
1.85
59.09
100.00
100.00
1.28
100.00
0.48
27.27
0.32
0.17
0.19
0.17
0.28
0.20
1.02
0.21
1.89
0.07
31.83
%of
Total
Failures
13.95
13.94
13.64
11.57
10.06
9.42
6.25
4.44
3.64
3.41
1.99
1.40
0.93
0.86
0.81
0.51
0.48
0.36
0.34
0.33
0.31
0.31
0.20
0.14
0.12
0.11
0.10
0.07
0.07
0.05
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.01
0.01
Cumulative
%
Contribution
13.95
27.89
41.53
53.10
63.16
72.58
78.83
83.27
86.91
90.32
92.31
93.71
94.64
95.50
96.31
96.82
97.30
97.67
98.01
98.34
98.64
98.95
99.14
99.28
99.40
99.52
99.62
99.69
99.76
99.80
99.83
99.85
99.87
99.89
99.91
99.93
99.94
99.96
99.98
99.99
100.00
BOLD = EPA MQO NATTS Core Analyte.
4-16
-------
While seven pollutants exhibited a failure rate of 100 percent, most of them were
infrequently detected. Of these seven, formaldehyde was detected in all 1,499 samples, while
other pollutants (chloroprene and chloromethylbenzene, for example) were detected relatively
few times. Thus, the number of failed screens, the number of measured detections, and the
failure rate must all be considered when reviewing the results of the risk screening process.
Other pollutants with relatively high failure rates include benzene, acrylonitrile,
1,2-dichloroethane, 1,2-dibromoethane, 1,1,2,2-tetrachloroethane, and carbon tetrachloride.
While each of these pollutants failed more than 98 percent of screens, four of them (acrylonitrile,
1,2-dichloroethane, 1,2-dibromoethane, 1,1,2,2-tetrachloroethane) were detected in fewer than
15 percent of samples collected.
EPA periodically revises the risk screening values based on new studies or updated
information. The screening values are often adjusted upward or downward, added, or removed
altogether. Screening values that have increased since the last report include dichloromethane
(from 2.1 |ig/m3to 7.7 |ig/m3) and tetrachloroethylene (from 0.17 |ig/m3to 3.8 |ig/m3). Screening
values that have decreased since the last report include nickel (from 0.009 |ig/m3 to
0.0021 |ig/m3), chloroprene (from 0.7 |ig/m3to 0.0021 |ig/m3), trichloroethylene (from 0.5 |ig/m3
to 0.2 |ig/m3), and many of the PAH (such as benzo(a)pyrene from 0.00091 to 0.00057 |ig/m3).
Perylene now has a risk screening value (0.011 |ig/m3) while anthracene, phenanthrene, and
pyrene no longer have screening values.
The 18 NATTS MQO Core Analytes (excluding acrolein) are bolded in Table 4-7.
Several NATTS MQO Core Analytes failed screens, but did not contribute to the top 95 percent
of failed screens (such as hexavalent chromium). However, as described in Section 3.2, all
NATTS MQO Core Analytes are inherently designated as program-wide pollutants of interest.
Two pollutants, beryllium and tetrachoroethylene, were added as pollutants of interest because
they are NATTS MQO Core Analytes, even though they did not fail any screens. These two
pollutants are not shown in Table 4-7. Note that six of the pollutants contributing to the top
95 percent of failed screens (ethylbenzene,/>-dichlorobenzene, 1,2-dichloroethane, acrylonitrile,
fluorene, and acenaphthene) are not NATTS MQO Core Analytes.
4-17
-------
The program-level pollutants of interest, as indicated by the shading and/or holding in
Table 4-7, were identified as follows:
• Acenaphthene • 1,2-Dichloroethane
• Acetaldehyde • Ethylbenzene
• Acrylonitrile • Fluorene
• Arsenic • Formaldehyde
• Benzene • Hexavalent Chromium
• Benzo(a)pyrene • Lead
• Beryllium • Manganese
• 1,3-Butadiene • Naphthalene
• Cadmium • Nickel
• Carbon Tetrachloride • Tetrachloroethylene
• Chloroform • Trichloroethylene
• />-Dichlorobenzene • Vinyl Chloride
The 2010 list of pollutants of interest identified via the preliminary risk screening
approach is similar to the 2008-2009 list of pollutants of interest. However, there are a few
exceptions. For the 2008-2009 NMP report, 1,2-dichloroethane was just outside the 95 percent
criteria cut-off for the pollutants of interest designation. Fluorene and acenaphthene are also new
to the list but they have revised screening values, as discussed above. Tetrachloroethylene was
not identified as a pollutant of interest for 2010 via the preliminary risk screening approach as its
screening value increased significantly.
Of the 80 pollutants sampled for under the NMP that have corresponding screening
values, concentrations of 41 pollutants failed at least one screen (or roughly 51 percent). Of
these, a total of 10,754 of 33,785 concentrations (31.8 percent) failed screens. If the measured
detections for tetrachloroethylene and beryllium (the two NATTS MQO Core Analytes that did
not fail any screens) are included in the total number of concentrations (35,535), the percentage
4-18
-------
of failed screens is approximately 30 percent. If all of the pollutants with screening values are
considered (including those that did not fail any screens), the percentage of concentrations
failing screens is much less (10,754 of 51,634, or 20.8 percent).
Table 4-8 presents the total number of failed screens per site, in descending order, as a
means of comparing the results of the risk screening process across the sites. As shown, S4MO
had the largest number of failed screens (574), followed by PXSS (526) and TMOK (503). In
addition to the number of failed screens, Table 4-8 also provides the total number of screens
conducted (one screen per valid preprocessed daily measurement for each site for all pollutants
with screening values). The failure rate, as a percentage, was determined from the number of
failed screens and the total number of screens conducted (based on applicable measured
detections) and is also provided in Table 4-8.
Table 4-8. Site-Specific Risk Screening Comparison
Site
S4MO
PXSS
TMOK
TOOK
BTUT
NBIL
OCOK
DEMI
MWOK
SEWA
PROK
GPCO
ELNJ
SPIL
NBNJ
SSSD
UCSD
CHNJ
UNVT
SKFL
PJCO
#of
Failed
Screens
574
526
503
499
483
468
467
466
466
444
434
424
387
373
320
303
283
276
223
177
176
Total # of
Measured
Detections1
2,538
2,389
1,861
1,855
2,383
2,572
1,895
2,085
1,859
2,348
,845
,933
,284
,271
,202
,289
,256
,070
,974
888
487
%of
Failed
Screens
22.62
22.02
27.03
26.90
20.27
18.20
24.64
22.35
25.07
18.91
23.52
21.93
30.14
29.35
26.62
23.51
22.53
25.79
11.30
19.93
36.14
#of
Pollutant
Groups
Analyzed
5
5
3
3
6
6
3
4
3
5
3
4
2
2
2
3
3
2
5
o
6
2
1 Total number of measured detections for all pollutants with screening
values, not just those failing screens.
BOLD ITALICS = EPA-designaled NATTS Site
4-19
-------
Table 4-8. Site-Specific Risk Screening Comparison (Continued)
Site
SYFL
SPAZ
PACO
INDEM
AZFL
ORFL
BOMA
WPIN
PANJ
BURVT
GLKY
BRCO
SJJCA
RUVT
RUCO
MONY
CELA
PRRI
RIVA
TONY
WADC
CAMS 35
SDGA
RUCA
BXNY
BMCO
PAFL
CAMS 85
HOW
CHSC
ROCH
#of
Failed
Screens
158
152
146
122
122
120
114
113
105
105
103
100
96
95
87
73
70
65
61
61
60
59
58
57
54
39
36
33
10
6
2
Total # of
Measured
Detections1
775
551
454
183
183
179
1,392
168
394
531
1,163
424
1,134
461
299
395
589
757
637
417
638
580
642
600
401
138
300
51
396
449
69
%of
Failed
Screens
20.39
27.59
32.16
66.67
66.67
67.04
8.19
67.26
26.65
19.77
8.86
23.58
8.47
20.61
29.10
18.48
11.88
8.59
9.58
14.63
9.40
10.17
9.03
9.50
13.47
28.26
12.00
64.71
2.53
1.34
2.90
#of
Pollutant
Groups
Analyzed
3
1
2
1
1
1
3
1
1
1
3
2
2
1
2
2
1
2
2
1
2
2
2
1
2
2
1
1
2
2
2
1 Total number of measured detections for all pollutants with screening
values, not just those failing screens.
BOLD ITALICS = EPA-designaled NATTS Site
The total number of screens and the number of pollutant groups measured by each site
must also be considered when interpreting the results in Table 4-8. For example, sites sampling
three, four, or five pollutant groups tended to have a higher number of failed screens. Although
WPIN, ORFL, INDEM, and AZFL had the highest failure rates (approximately 67 percent each),
each of these sites sampled only one pollutant group (carbonyl compounds). Three pollutants
measured with Method TO-11A (carbonyl compounds) have screening values (acetaldehyde,
4-20
-------
formaldehyde, and propionaldehyde) and two of these pollutants typically fail all or most of the
screens conducted, as shown in Table 4-7. Thus, sites sampling only carbonyl compounds have
relatively high failure rates. Conversely, sites that sampled several pollutant groups tended to
have lower failure rates due to the larger number of HAPs screened, as is the case with S4MO,
PXSS, NBIL, BTUT, and SEW A, to name a few. For this reason, the number of pollutant groups
for which sampling was conducted is also presented in Table 4-8. Every site had at least one
pollutant fail a screen.
The following sections from this point forward focus only on those pollutants designated
as program-level pollutants of interest.
4.2.1 Concentrations of the Pollutants of Interest
Concentrations of the program-level pollutants of interest vary significantly, among the
pollutants and among the sites. Tables 4-9 through 4-12 present the top 10 annual average
concentrations and 95 percent confidence intervals by site for each of the program-level
pollutants of interest (for VOC, carbonyl compounds, PAH, and metals respectively). As
described in Section 3.1.1, an annual average is the average concentration of all measured
detections and zeros substituted for non-detects for a given year. Further, an annual average is
only considered valid where there are at least three quarterly averages and where the site-specific
method completeness is at least 85 percent. The annual average concentrations for PAH in
Table 4-11 and metals in Table 4-12 are reported in ng/m3 for ease of viewing, while annual
average concentrations in Tables 4-9 and 4-10, for VOC and carbonyl compounds, respectively,
are reported in ug/m3. Note that not all sites sampled each pollutant; thus, the list of possible
sites presented in Tables 4-9 through 4-12 is limited to those sites sampling each pollutant. For
example, only five sites sampled TSP metals; thus, all five sites appear in Table 4-12 for each
metal pollutant of interest shown.
4-21
-------
Table 4-9. Annual Average Concentration Comparison of the VOC Pollutants of Interest
to
to
Rank
1
2
3
4
5
6
7
8
9
10
Acrylonitrile
(Hg/m3)
SPAZ
0.39 ±0.23
S4MO
0.17 ±0.14
NBNJ
0.14±0.11
UCSD
0.05 ±0.03
PXSS
0.05 ±0.03
CHNJ
0.04 ± 0.02
OCOK
0.03 ±0.03
SPIL
0.02 ± 0.02
TMOK
0.02 ± 0.02
GPCO
0.02 ±0.01
Benzene
(jig/m )
TOOK
2.34 ±0.36
PACO
1.72 ±0.21
SPAZ
1.69 ±0.37
RUCO
1.62 ±0.18
TMOK
1.57 ±0.20
RICO
1.46 ±0.17
GPCO
1.40 ±0.16
PXSS
1.38 ±0.20
BTUT
1.22 ±0.16
BRCO
1.10±0.18
1,3-Butadiene
(Hg/m3)
SPAZ
0.26 ±0.08
PXSS
0.21 ±0.05
RICO
0.16 ±0.06
SPIL
0.14 ±0.02
GPCO
0.14 ±0.02
S4MO
0.12 ±0.03
ELNJ
0.12 ±0.01
RUVT
0.12 ±0.05
BTUT
0.10 ±0.02
TMOK
0.10 ±0.02
Carbon
Tetrachloride
(Hg/m3)
SEWA
0.72 ± 0.04
NBIL
0.72 ±0.03
SPIL
0.71 ±0.03
DEMI
0.69 ±0.03
SPAZ
0.66 ± 0.04
PXSS
0.66 ±0.03
PROK
0.65 ±0.03
MWOK
0.64 ± 0.04
CHNJ
0.64 ±0.03
OCOK
0.64 ± 0.04
Chloroform
(Hg/m3)
NBIL
1.06 ±0.53
DEMI
0.63 ±0.08
PXSS
0.37 ±0.05
BTUT
0.28 ±0.40
SPAZ
0.24 ± 0.04
S4MO
0.19 ±0.04
PROK
0.14±0.11
SEWA
0.14 ±0.01
ELNJ
0.13 ±0.02
NBNJ
0.12 ±0.02
/7-Dichlorobenzene
(Hg/m3)
BTUT
0.54 ±0.74
S4MO
0.35 ±0.18
SPAZ
0.28 ±0.06
OCOK
0.24 ± 0.07
MWOK
0.18 ±0.04
PXSS
0.16 ±0.03
TOOK
0.15 ±0.03
TMOK
0.14 ±0.03
PROK
0.10 ±0.03
ELNJ
0.09 ±0.02
BOLD ITALICS = EPA-designaled NATTS Site.
-------
Table 4-9. Annual Average Concentration Comparison of the VOC Pollutants of Interest (Continued)
J^.
K>
Rank
1
2
3
4
5
6
7
8
9
10
1,2-Dichloroethane
(Hg/m3)
PROK
0.02 ±0.01
MWOK
0.02 ±0.01
CHNJ
0.02 ±0.01
BTUT
0.02 ±0.01
OCOK
0.02 ±0.01
SPIL
0.02 ±0.01
SEWA
0.02 ±0.01
S4MO
0.02 ±0.01
NBNJ
0.02 ±0.01
ELNJ
0.02 ±0.01
Ethylbenzene
(Hg/m3)
PACO
0.98 ±0.98
RICO
0.95 ±0.86
SPAZ
0.76 ±0.18
UCSD
0.75 ±1.05
PXSS
0.60 ±0.10
TMOK
0.55 ±0.13
DEMI
0.50 ±0.11
GPCO
0.50 ±0.07
BTUT
0.48 ±0.15
TOOK
0.47 ±0.08
Tetrachloroethylene
(Hg/m3)
PXSS
0.40 ± 0.08
GPCO
0.39 ±0.07
SPAZ
0.33 ±0.09
SPIL
0.32 ±0.06
S4MO
0.23 ± 0.06
MWOK
0.22 ± 0.06
DEMI
0.21 ±0.05
NBIL
0.21 ±0.05
ELNJ
0.20 ± 0.03
RUVT
0.18 ±0.04
Trichloroethylene
(Hg/m3)
SPIL
0.79 ±0.40
BTUT
0.09 ±0.11
UCSD
0.08 ±0.10
NBIL
0.07 ± 0.02
S4MO
0.05 ±0.02
SPAZ
0.05 ±0.02
PXSS
0.03 ±0.03
ELNJ
0.03 ±0.01
GPCO
0.03 ±0.01
DEMI
0.02 ±0.01
Vinyl Chloride
(Hg/m3)
DEMI
0.003 ± 0.002
MWOK
0.002 ± 0.004
BTUT
0.002 ± 0.002
RUVT
0.002 ± 0.004
NBNJ
0.002 ±0.001
S4MO
0.001 ±0.001
OCOK
0.001 ±0.001
UNVT
0.001 ±0.001
TOOK
0.001 ±0.001
UCSD
0.001 ±0.002
BOLD ITALICS = EPA-designated NATTS Site.
-------
Table 4-10. Annual Average Concentration Comparison of the Carbonyl Compound Pollutants of Interest
J^.
K>
Rank
1
2
3
4
5
6
7
8
9
10
Acet aldehyde
(Ug/m3)
S4MO
4.10 ±0.59
SKFL
3. 36 ±0.48
AZFL
2.94 ±0.40
NBNJ
2.92 ±0.37
ELNJ
2.73 ±0.39
WPIN
2.56 ±0.33
BTUT
2.25 ±0.27
TOOK
2.20 ±0.29
GPCO
2.00 ± 0.20
TMOK
2.00 ± 0.20
Formaldehyde
(Ug/m3)
ELNJ
4.46 ± 0.64
BTUT
3.66 ±0.63
NBIL
3.59 ±2.18
WPIN
3.58 ±0.45
TMOK
3.35 ±0.40
TOOK
3. 14 ±0.45
DEMI
2.80 ±0.31
GPCO
2.78 ±0.23
SYFL
2.76 ±0.49
S4MO
2.74 ±0.33
BOLD ITALICS = EPA-designated NATTS Site.
-------
Table 4-11. Annual Average Concentration Comparison of the PAH Pollutants of Interest
to
Rank
1
2
3
4
5
6
7
8
9
10
Acenaphthene
(ng/m3)
DEMI
13.74 ±7.06
NBIL
10.46 ±3. 16
GPCO
7.30 ±1.53
CELA
6.15±1.18
S4MO
5.76 ± 1.35
CAMS 35
4.07 ±0.71
RIVA
3.63 ±0.65
PRRI
3.34 ±0.57
JK4DC
2.98 ±0.51
SDG4
2.86 ±0.57
Benzo(a)pyrene
(ng/m3)
PRRI
0.20 ± 0.07
GPCO
0.18 ±0.08
DEMI
0.17 ±0.04
S4MO
0.16 ±0.05
NBIL
0.11 ±0.03
BOMA
0.10 ±0.02
fl/^/4
0.08 ±0.04
PAS'S
0.07 ±0.03
CELA
0.07 ±0.05
SDG4
0.06 ± 0.04
Fluorene (ng/m3)
DEMI
12.62 ±6.23
NBIL
10.69 ±2.98
C£Z/4
6.90±1.11
SVM0
6.57 ±1.39
GPCO
6.44 ±0.78
C4MS35
4.57 ±0.78
PRRI
4.56 ±0.70
fl/^/4
4.50 ±0.60
WADC
4.41 ±0.67
RUCA
3.75 ±0.51
Naphthalene
(ng/m3)
GPCO
147.04 ±22.61
CEZ/4
143. 33 ±24.02
DEMI
137.84 ±20.80
S4MO
135.13 ±35.06
SDGA
127.84 ±23. 35
WADC
110.77 ±18.56
RIVA
106. 17 ±16.89
NBIL
105.54 ±31.79
CAMS 35
92.93 ± 14.64
SKFL
90.08 ± 17.04
BOLD ITALICS = EPA-designated NATTS Site.
-------
Table 4-12. Annual Average Concentration Comparison of the Metals Pollutants of Interest
to
Rank
1
2
o
J
4
5
6
7
8
9
10
Arsenic
(PM10)
(ng/m3)
S4MO
1.02 ±0.20
NBIL
0.75 ±0.17
BTUT
0.61 ±0.15
S£^
0.58 ±0.09
PAFL
0.57 ±0.10
PXSS
0.56 ±0.14
SJJCA
0.37 ±0.07
BOMA
0.36 ±0.05
UNVT
0.21 ±0.04
Arsenic (TSP)
(ng/m3)
TOOK
0.66 ±0.08
TMOK
0.62 ±0.09
PROK
0.55 ±0.07
OCOK
0.44 ±0.08
MWOK
0.44 ± 0.06
Beryllium
(PM10)
(ng/m3)
S4MO
0.007 ±0.001
PXSS
0.006 ± 0.002
PAFL
0.004 ± 0.002
NBIL
0.004 ± 0.001
BTUT
0.003 ± 0.001
BOMA
0.003 ± 0.001
SJJCA
0.002 ±<0.001
SEWA
0.002 ± 0.001
UNVT
0.001 ±<0.001
Beryllium
(TSP)
(ng/m3)
PROK
0.023 ± 0.005
TOOK
0.019 ±0.003
OCOK
0.014 ±0.003
TMOK
0.012 ±0.002
MWOK
0.010 ±0.002
Cadmium
(PM10)
(ng/m3)
S4MO
0.62 ±0.11
BOMA
0.19 ±0.02
NBIL
0.13 ±0.02
PXSS
0.12 ±0.03
BTUT
0 1 0 ± 0 03
SEWA
0.09 ±0.02
PAFL
0.06 ±0.01
SJJCA
0.06 ±0.01
UNVT
0.06 ±0.01
Cadmium
(TSP)
(ng/m3)
TOOK
0.28 ±0.06
TMOK
0.23 ± 0.04
PROK
0.17 ±0.04
OCOK
0.10 ±0.01
MWOK
0.10 ±0.01
Hexavalent
Chromium
(ng/m3)
CAMS 85
0.31 ±0.16
PXSS
0.13 ±0.03
CAMS 35
0.05 ±0.01
DEMI
0.04 ±0.01
S4MO
0.03 ±0.01
SEWA
0.03 ±0.01
BTUT
0.03 ±0.01
SKFL
0.02±<0.01
NBIL
0.02±<0.01
BOMA
0.02 ±0.01
BOLD ITALICS = EPA-designaled NATTS Site.
-------
Table 4-12. Annual Average Concentration Comparison of the Metals Pollutants of Interest (Continued)
to
Rank
1
2
o
J
4
5
6
7
8
9
10
Lead
(PM10)
(ng/m3)
S4MO
11.66 ±2.60
PXSS
3.42 ±0.71
NBIL
3.11±0.53
PAFL
3.08 ±1.34
BTUT
2.68 ±0.57
SEWA
2.63 ±0.32
BOMA
2.53 ±0.35
SJJCA
2.13 ±0.44
UNVT
1.45 ±0.32
Lead
(TSP1
(ng/m5)
TOOK
4.46 ±0.51
TMOK
3.72 ±0.40
PROK
3.06 ±0.58
MWOK
2.69 ±0.80
OCOK
2.08 ±0.22
Manganese
(PM10)
(ng/m3)
S4MO
17. 15 ±7.07
PXSS
12.38 ±2.53
NBIL
6.74 ±1.57
SEWA
5.75 ± 1.79
BTUT
5.61 ±0.79
SJJCA
3.76 ±0.68
BOMA
3. 18 ±0.48
PAFL
3.04 ±0.97
UNVT
1.94 ±0.38
Manganese
(TSP1
(ng/nr5)
TOOK
23.61 ±3. 15
TMOK
15.88 ±1.95
PROK
12.84 ±2.00
OCOK
11. 98 ±1.68
MWOK
9.55 ± 1.41
Nickel
(PM10)
(ng/m3)
SEWA
1.91 ±0.48
BOMA
1.25 ±0.14
PXSS
1.23 ±0.21
NBIL
1.06 ±0.19
PAFL
1.05 ±0.24
S4MO
1.04 ±0.14
BTUT
0.94 ±0.12
SJJCA
0.84 ±0.10
UNVT
0.26 ± 0.04
Nickel
(TSP1
(ng/m5)
TOOK
1.10±0.12
MWOK
0.98 ±0.17
TMOK
0.89 ±0.11
PROK
0.82 ±0.26
OCOK
0.59 ±0.06
BOLD ITALICS = EPA-designated NATTS Site.
-------
Some observations from Tables 4-9 through 4-12 include the following:
• The highest annual average concentration among the program-wide pollutants of
interest was calculated for formaldehyde for ELNJ (4.46 ± 0.64 |ig/m3). All 10 of the
annual averages of formaldehyde and six of the 10 annual averages of acetaldehyde
shown in Table 4-10 are higher than the next highest annual average concentration
for another analytical method/pollutant.
• Behind ELNJ, BTUT and NBIL have the next highest annual average concentrations
of formaldehyde (3.66 ± 0.63 |ig/m3 and 3.59 ± 2.18 |ig/m3, respectively). However,
the confidence interval for NBIL's annual average is more than three times that of
BTUT or ELNJ. This likely indicates that NBIL's annual average is influenced by
outliers while ELNJ and BTUT's concentrations tend to be consistently higher. While
less than 1 |ig/m3 separates these three sites' annual averages, their median
concentrations range from 1.32 |ig/m3 for NBIL, to 2.94 |ig/m3 for BTUT, to
3.79|ig/m3forELNJ.
• Among the VOC, the annual average concentrations of benzene are the only annual
averages consistently greater than 1 ug/m3. TOOK's annual average benzene
concentration (2.34 ± 0.36 |ig/m3) is significantly higher than the next highest annual
average (1.72 ± 0.21 |ig/m3 for PACO). TOOK has the four highest benzene
measurements and 12 of the 20 highest benzene concentrations among all sites
sampling this pollutant. Five of the six Colorado sites have one of the top 10 benzene
concentrations; only BMCO does not appear in Table 4-9 for benzene. This site did
not begin sampling until September 2010; thus, annual averages could not be
calculated.
• The difference between the highest and tenth highest annual average concentration of
carbon tetrachloride is only 0.08 |ig/m3. The difference between the highest and
lowest annual average concentration of this pollutant among all NMP sites is
0.20 |ig/m3, indicating the relative uniformity in concentrations of this pollutant in
ambient air.
• NBIL's annual average chloroform concentration (1.06 ± 0.53 |ig/m3) is nearly two
times higher than the next highest annual average (DEMI, 0.63 ± 0.08 |ig/m3). Of the
24 concentrations of chloroform greater than 1 |ig/m3, 13 were measured at NBIL and
eight were measured at DEMI.
• BTUT's annual average concentration ofp-dichlorobenzene has a large confidence
interval associated with it, indicating the likely influence of outliers. The highest
concentration ofp-dichlorobenzene measured at BTUT was 21.2 |ig/m3, nearly seven
times higher than the next highest measurement of 3.11 |ig/m3 (also measured at
BTUT). These were the two highest measurements of this pollutant among sites
sampling VOC.
4-28
-------
• The top 10 annual average concentrations of 1,2-dichloroethane range from 0.021
|ig/m3 (PROK) to 0.015 |ig/m3 (ELNJ). Although not shown in Table 4-9, the lowest
annual average concentration of this pollutant is 0.007 |ig/m3 (TOOK), indicating the
relatively small range of concentrations measured for this pollutant.
• Note the relatively large confidence intervals associated with three of the top four
annual average concentrations of ethylbenzene. This likely indicates the presence of
outliers. Of the seven measurements of ethylbenzene greater than 5 |ig/m3, one was
measured at UCSD (31.5 |ig/m3,) two were measured at PACO (26.7 |ig/m3 and
11.9 |ig/m3), and two at RICO (25.7 |ig/m3and 6.73 |ig/m3). Note that the second
highest concentration of ethylbenzene measured at UCSD is 2.08 |ig/m3. Also, the
median concentrations of ethylbenzene for each of these three sites is less than half
the annual averages, while SPAZ's median concentration (0.62 |ig/m3) is much closer
to the respective annual average (0.76 jig/m3), indicating that SPAZ's concentrations
tended to be consistently higher compared to the other three sites.
• SPIL's annual average concentration of trichloroethylene (0.79 ± 0.40 |ig/m3) is
more than seven times higher than the next highest annual average
(BTUT, 0.09 ±0.11 |ig/m3). Of the 18 concentrations of trichloroethylene greater
than 1 |ig/m3, 15 were measured at SPIL.
• Among the sites sampling PAHs, the annual averages for DEMI, GPCO, and S4MO
are among the top five for each of the four PAH pollutants of interest.
• Although GPCO has the highest annual average concentration of naphthalene, the
highest measurements of this pollutant were not measured at this site. Three of the
New York sites measured some of the highest concentrations of naphthalene. In
particular, the TONY site has the two highest measurements among all sites sampling
naphthalene. TONY has 9 of 15 concentrations greater than 500 ng/m3 and BXNY
and MONY have one each. However, annual averages could not be calculated for
these three sites due to abbreviated sampling durations.
• CAMS 85 has the highest annual average concentration of hexavalent chromium.
CAMS 85 is the only site with hexavlanet chromium concentrations greater than
1 ng/m3 (ranging from 1.17 ng/m3 to 3.51 ng/m3). The annual average concentration
of hexavalent chromium for CAMS 85 is more than twice the next highest annual
average (PXSS). Both of these annual averages are an order of magnitude higher than
the other listed sites. Of the 25 measurements of hexavalent chromium greater than
0.25 ng/m3, 17 were measured at CAMS 85 and eight at PXSS.
• S4MO has the highest annual average concentration of five of the six PMio metals:
arsenic, beryllium, cadmium, lead, and manganese. In addition, S4MO's annual
average nickel concentration ranks sixth. Moreover, S4MO's annual averages of
cadmium, lead, and manganese are significantly higher than the other annual averages
listed. S4MO is the only site sampling PMio metals with cadmium concentrations
greater than 1 ng/m3 (11 ranging from 1.01 to 2.05 ng/m3). The 12 highest
concentrations of lead were measured at S4MO (ranging from 18.0 ng/m3 to
47.8 ng/m3). The highest manganese concentration among all sites sampling
4-29
-------
metals was measured at S4MO (200 ng/m3) and was an order of magnitude higher
than the next highest manganese concentration (84.5 ng/m3), also measured at S4MO.
• TOOK has the highest annual average concentration of five of the six TSP metals:
arsenic, cadmium, lead, manganese, nickel. TOOK's annual average beryllium
concentration ranked second.
• S4MO was on the top 10 list for 21 of the 24 program-level pollutants of interest;
PXSS and BTUT were both on the top 10 list for 17 of the 24 program-level
pollutants of interest. NBIL appears in Tables 4-9 through 4-12 a total of 16 times.
Conversely, 13 sites do not appear in Table 4-9 through 4-12 at all. However, some
sites did not meet the criteria for annual averages to be calculated.
4.2.2 Risk Screening Assessment Using MRLs
A summary of the program-level MRL risk assessment is presented in Table 4-13.
Dichloromethane and formaldehyde are the only pollutants with at least one concentration or
time-period average concentration greater than their respective ATSDR heath risk benchmarks.
Out of 1,495 measured detections of formaldehyde, one concentration is higher than the ATSDR
acute MRL (50 |ig/m3), which was measured at NBIL. Two measured detections of
dichloromethane are greater than the ATSDR acute MRL for dichloromethane (2,000 |ig/m3),
which were measured at BTUT and GPCO. Concentrations that are greater than their respective
acute MRL are discussed on a site-specific basis in further detail in Sections 5 through 28.
Out of 103 quarterly averages of formaldehyde, none of the quarterly averages are greater
than the ATSDR intermediate MRL (40 |ig/m3). None of the quarterly averages of
dichloromethane are greater than the ATSDR intermediate MRL (1,000 |ig/m3). In addition,
none of the annual averages of either formaldehyde or dichloromethane are greater than their
respective ATSDR chronic MRLs (10 |ig/m3 and 1,000 |ig/m3, respectively). Graphical displays
of the site-specific quarterly averages for the program-level pollutants of interest are presented
and discussed in Section 4.4.2.
4-30
-------
Table 4-13. Program-Level MRL Risk Screening Assessment
Sampling
Method
TO-11A
TO- 15
Pollutant
Formaldehyde
Dichloromethane
Acute Risk
ATSDR
Acute
MRL1
(Hg/m3)
50
2,000
#of
Concentrations
>MRL/
# of Measured
Detections
1/1,495
2/1,263
Intermediate Risk
ATSDR
Intermediate
MRL1
(Hg/m3)
40
1,000
# of 1st
Quarter Avg
Cone > MRL/
# of Quarterly
Averages
0/23
0/22
# of 2nd
Quarter Avg
Cone > MRL/
# of Quarterly
Averages
0/28
0/22
# of 3rd
Quarter Avg
Cone > MRL/
# of Quarterly
Averages
0/27
0/23
# of 4th
Quarter Avg
Cone > MRL/
# of Quarterly
Averages
0/25
0/23
Chronic Risk
ATSDR
Chronic
MRL1
(Hg/m3)
10
1,000
# of Annual
Avg Cone
>MRL/
# of Annual
Averages
0/24
0/22
1 Reflects the use of one significant digit for MRLs.
-------
4.3 The Impact of Mobile Sources
Ambient air is significantly impacted by mobile sources, as discussed in Section 3.4.1.
Table 4-14 contains several parameters that are used to assess mobile source impacts on air
quality near the monitoring sites, including emissions data from the NEI, concentration data, and
site-characterizing data, such as vehicle ownership.
4.3.1 Mobile Source Emissions
On-road emissions come from mobile sources such as automobiles, buses, and
construction vehicles that use roadways; non-road emissions come from the remaining mobile
sources such as locomotives, lawn mowers, and boats (EPA, 2012e). Table 4-14 contains county-
level on-road and non-road HAP emissions from the 2008 NEI. Mobile source emissions tend to
be highest in large urban areas and lowest in rural areas. Estimated on-road county emissions
were highest in Los Angeles County, CA (where CELA is located), followed by Harris County,
TX (where CAMS 35 is located), and Maricopa County, AZ (where PXSS and SPAZ are
located) while estimated on-road emissions were lowest in Union County, SD, and Chesterfield
County, SC (where UCSD and CHSC are located, respectively). Estimated non-road county
emissions were also highest in Los Angeles County, CA, followed by Cook County, IL and
Maricopa County, AZ. Estimated non-road county emissions were lowest in Carter County, KY
(where GLKY is located), and Union County, SD.
4-32
-------
Table 4-14. Summary of Mobile Source Information by Monitoring Site
Site
AZFL
BMCO
BOMA
BRCO
BTUT
BURVT
BXNY
CAMS 35
CAMS 85
CELA
CHNJ
CHSC
DEMI
ELNJ
GLKY
GPCO
HOW
INDEM
MONY
MWOK
NBIL
NBNJ
OCOK
ORFL
PACO
PAFL
County-level
Motor Vehicle
Registration
(# of Vehicles)
879,317
74,847
501,587
74,847
239,754
223,316
248,600
3,115,974
69,883
7,410,625
389,359
40,431
1,336,940
424,894
36,031
180,119
98,211
182,989
248,600
809,783
2,083,141
640,893
809,783
1,037,369
74,847
1,037,369
Estimated
10-Mile Vehicle
Ownership
(# of Vehicles)
532,212
7,921
1,158,723
32,230
201,757
165,680
1,181,520
542,457
3,224
2,774,128
193,281
4,856
796,952
1,723,298
21,977
144,154
23,836
150,152
1,181,520
406,137
344,352
619,349
426,788
905,833
10,530
787,532
Annual
Average Daily
Traffic
(# of Vehicles)
41,500
2,527
31,400
150
113,955
4,000
100,230
31,043
1,400
235,000
12,917
550
106,900
250,000
428
12,000
5,000
52,440
134,421
41,200
34,100
114,322
41,600
31,500
2,600
43,500
County-level
Daily VMT
23,138,726
1,942,038
NA
1,942,038
7,360,752
4,027,945
NA
NA
NA
211,876,660
14,256,044
1,302,685
47,115,093
12,485,902
1,164,000
2,047,739
2,659,643
11,801,000
NA
NA
89,621,776
20,415,685
NA
35,657,527
1,942,038
35,657,527
County-Level
On-road
Emissions1
(tpy)
2,650.97
260.03
715.05
260.03
861.85
371.91
825.34
8,521.88
284.17
9,556.40
1,202.20
128.86
5,900.70
951.86
163.87
392.28
247.41
1,222.76
825.34
2,900.47
7,721.47
1,617.22
2,900.47
3,198.03
260.03
3,198.03
County-Level
Non-road
Emissions1
(tpy)
1,157.75
93.05
440.97
93.05
336.23
251.44
391.72
2,791.78
129.54
5,072.27
705.27
80.37
1,113.36
390.19
15.58
180.83
220.50
634.27
391.72
816.74
4,074.66
673.13
816.74
1,587.49
93.05
1,587.49
Hydrocarbon
Average2
(ppbv)
NA
NA
NA
NA
4.37
2.76
NA
NA
NA
NA
1.13
NA
3.01
5.55
1.24
4.66
NA
NA
NA
2.10
2.27
1.98
2.13
NA
NA
NA
Reference: EPA, 2012b
2This parameter is only available for monitoring sites sampling VOC.
BOLD ITALICS = EPA-designated NATTS Site.
NA = Data not available.
-------
Table 4-14. Summary of Mobile Source Information by Monitoring Site (Continued)
Site
PANJ
PROK
PRRI
PXSS
RICO
RIVA
ROCH
RUCA
RUCO
RUVT
S4MO
SDGA
SEWA
SJJCA
SKFL
SPAZ
SPIL
SSSD
SYFL
TMOK
TONY
TOOK
UCSD
UNVT
WADC
WPIN
County-level
Motor Vehicle
Registration
(# of Vehicles)
396,602
40,832
485,837
3,739,918
74,847
347,790
552,184
1,707,950
74,847
118,002
1,121,528
472,535
1,763,504
1,517,995
879,317
3,739,918
2,083,141
208,911
1,125,844
604,284
669,746
604,284
25,051
223,316
211,653
204,908
Estimated
10-Mile Vehicle
Ownership
(# of Vehicles)
1,053,264
26,447
511,525
1,439,566
23,520
520,602
474,074
767,438
23,520
65,811
690,875
541,355
866,590
1,262,220
644,692
878,323
819,706
234,348
295,497
319,719
436,105
455,374
10,630
50,202
669,201
178,215
Annual
Average Daily
Traffic
(# of Vehicles)
22,272
15,900
136,800
193,000
17,000
74,000
116,725
145,000
699
7,200
81,174
145,890
234,000
103,000
49,500
130,000
170,700
21,340
10,700
12,700
74,406
62,566
156
1,200
7,700
143,410
County-level
Daily VMT
8,178,167
NA
NA
89,448,000
1,942,038
8,260,273
NA
55,167,650
1,942,038
1,766,027
23,385,327
21,057,000
23,454,115
39,402,370
23,138,726
89,448,000
89,621,776
3,716,475
34,745,256
NA
NA
NA
790,541
4,027,945
NA
35,081,000
County-Level
On-road
Emissions1
(tpy)
616.98
172.07
1,104.51
7,862.48
260.03
831.85
1,566.25
2,486.42
260.03
158.14
974.72
2,272.55
6,932.11
1,960.08
2,650.97
7,862.48
7,721.47
467.40
3,252.93
2,197.21
1,954.05
2,197.21
91.81
371.91
929.71
2,664.97
County-Level
Non-road
Emissions1
(tpy)
447.26
83.98
381.46
3,819.27
93.05
188.91
683.88
1,003.76
93.05
150.60
182.60
772.13
2,762.29
812.60
1,157.75
3,819.27
4,074.66
132.93
1,326.89
867.85
678.01
867.85
30.98
251.44
327.98
715.48
Hydrocarbon
Average2
(ppbv)
7.05
1.96
NA
4.59
NA
NA
NA
NA
NA
2.77
3.28
NA
1.99
NA
NA
5.85
2.91
1.74
NA
3.57
NA
4.15
1.51
0.85
NA
NA
Reference: EPA, 2012b
2This parameter is only available for monitoring sites sampling VOC.
BOLD ITALICS = EPA-designated NATTS Site.
NA = Data not available.
-------
4.3.2 Hydrocarbon Concentrations
Hydrocarbons are organic compounds that contain only carbon and hydrogen.
Hydrocarbons are derived mostly from crude petroleum sources and are classified according to
their arrangement of atoms as alicyclic, aliphatic, and aromatic. Hydrocarbons are of prime
economic importance because they encompass the constituents of the major fossil fuels,
petroleum and natural gas, as well as plastics, waxes, and oils. Hydrocarbons in the atmosphere
originate from natural sources and from various anthropogenic sources, such as the combustion
of fuel and biomass, petroleum refining, petrochemical manufacturing, solvent use, and gas and
oil production and use. In urban air pollution, these components, along with oxides of nitrogen
(NOX) and sunlight, contribute to the formation of tropospheric ozone. According to the EPA,
approximately 47 percent of hydrocarbon emissions are from mobile sources (both on-road and
non-road) (EPA, 20121). Thus, the concentration of hydrocarbons in ambient air may act as an
indicator of mobile source activity levels. Several hydrocarbons are sampled with Method
TO-15, including benzene, ethylbenzene, and toluene.
Table 4-14 presents the average of the sum of hydrocarbon concentrations for each site
sampling VOC. Note that only sites sampling VOC have data in this column. Table 4-14 shows
that PANJ, SPAZ, and ELNJ had the highest hydrocarbon averages among the sites monitoring
VOC. Each of these sites is located in a highly populated urban area and in close proximity to
heavily traveled roadways. For example, ELNJ is located near Exit 13 on 1-95 near New York
City.
The sites with the lowest hydrocarbon averages are UNVT, CHNJ, and GLKY. All three
sites are located in rural areas. However, CHNJ is still located within the New York City MSA,
although on the periphery. The daily average hydrocarbon concentration can be compared to
other indicators of mobile source activity, such as the ones discussed below, to determine if
correlations exist.
4.3.3 Motor Vehicle Ownership
Another indicator of motor vehicle activity near the monitoring sites is the total number
of vehicles owned by residents in the county where each monitoring site is located, which
includes passenger vehicles, trucks, and commercial vehicles, as well as vehicles that can be
4-35
-------
regional in use such as boats or snowmobiles. Actual county-level vehicle registration data were
obtained from the applicable state or local agency, where possible. If data were not available,
vehicle registration data are available at the state-level (FHWA, 2011). The county proportion of
the state population was then applied to the state registration count.
The county-level motor vehicle ownership data and the average hydrocarbon
concentration are presented in Table 4-14. As previously discussed, PANJ, SPAZ, and ELNJ had
the highest average hydrocarbon concentrations, respectively, while UNVT, CHNJ, and GLKY
had the lowest. Table 4-14 also shows that SPAZ, PXSS, NBIL, and SPIL had the highest
county-level vehicle ownership of the sites sampling VOC, while UCSD, GLKY, and PROK
have the lowest. CELA, which had the highest county-level vehicle ownership of all the sites,
did not sample VOC. The Pearson correlation coefficient calculated between these two datasets
is 0.32. While this correlation falls below the "strong" classification, it does indicate a positive
correlation between hydrocarbon concentrations and vehicle registration.
The vehicle ownership at the county-level may not be completely indicative of the
ownership in a particular area. As an illustration, for a county with a large city in the middle of
its boundaries and less populated areas surrounding it, the total county-level ownership may be
more representative of areas inside the city limits than in the rural outskirts. Therefore, a vehicle
registration-to-population ratio was developed for each county with a monitoring site. Each ratio
was then applied to the 10-mile population surrounding the sites (from Table 2-2) to estimate a
10-mile population, which is presented in Table 4-14. Table 4-14 shows that ELNJ, PXSS, and
PANJ have the highest 10-mile estimated vehicle ownership of the sites sampling VOC, while
UCSD, GLKY, and PROK have the lowest. Again, CELA, which had the highest 10-mile
estimated vehicle ownership of all the sites, did not sample VOC under the NMP. The Pearson
correlation coefficient calculated between the average hydrocarbon calculations and the 10-mile
vehicle registration estimates is 0.63. This represents a strong positive correlation, indicating that
as vehicle registration inside the 10-mile radius increases, concentration of hydrocarbons tend to
proportionally increase.
Other factors may affect the reliability of motor vehicle ownership data as an indicator of
ambient air monitoring data results:
4-36
-------
• Estimates of higher vehicle ownership surrounding a monitoring site do not
necessarily imply increased motor vehicle use in the immediate vicinity of a
monitoring site. Conversely, sparsely populated regions often contain heavily
traveled roadways.
• Emissions sources in the area other than motor vehicles may significantly affect
levels of hydrocarbons in the ambient air.
4.3.4 Estimated Traffic Volume
In NMP reports prior to 2007, traffic data, which represents the average number of
vehicles passing a monitoring site on a daily basis, were obtained from AQS. However, much of
the populated traffic data reflected traffic conditions during site initiation, and were often 5 or
more years old. Thus, beginning with the 2007 NMP report, updated traffic data were obtained
from state and local agencies, primarily departments of transportation. Most of the traffic counts
in this report reflect AADT, which is "the total volume of traffic on a highway segment for 1
year, divided by the number of days in the year," and incorporates both directions of traffic
(FL DOT, 2007). AADT counts obtained were based on data from 2002 to 2010, primarily 2008-
2010. The updated traffic values are presented in Table 4-14. The traffic data presented in
Table 4-14 represent the most recently available data applicable to the monitoring site.
There are several limitations to obtaining the AADT near each monitoring site. AADT
statistics are developed for roadways, such as interstates, state highways, or local roadways,
which are managed by different municipalities or government agencies. AADT is not always
available in rural areas or for secondary roadways. For monitoring sites located near interstates,
the AADT for the interstate segment closest to the site was obtained. For other monitoring sites,
the highway or secondary road closest to the monitoring site was used. Only one AADT value
was obtained for each monitoring site. The intersection or roadway chosen for each monitoring
site is discussed in each individual state section (Sections 5 through 28).
Table 4-14 shows that ELNJ, SEW A, and PXSS have the highest daily traffic volume of
the sites sampling VOC, while UCSD, GLKY, and UNVT, have the lowest. For all monitoring
sites (not just those sampling VOC), the highest daily traffic volume occurs near ELNJ, CELA,
and SEWA. ELNJ is located near Exit 13 on 1-95; CELA is located in downtown Los Angeles;
and SEWA is located in Seattle near the intersection of 1-5 and 1-9. ELNJ has the highest traffic
4-37
-------
volume and the third highest hydrocarbon average (behind PANJ and SPAZ), but SEW A, PXSS,
and SPIL, which have the second, third, and fourth highest traffic volumes, have the 17th, 5th, and
11th highest hydrocarbon averages, respectively. CELA did not measure VOC. A Pearson
correlation coefficient calculated between the average hydrocarbon calculations and the traffic
counts is 0.36. While this correlation is not a "strong" correlation, it does indicate a positive
correlation between hydrocarbon concentrations and traffic volumes.
4.3.5 Vehicle Miles Traveled
Another approach to determine how mobile sources affect urban air quality is to review
VMT. VMT is "the sum of distances traveled by all motor vehicles in a specified system of
highways for a given period of time" (OR DOT, 2012). Thus, VMT values tend to be large (in
the millions). In past NMP reports, daily VMT data from the Federal Highway Administration
(FHWA) were obtained. However, VMT was only available by urban area; thus no VMT was
available for sites located in rural areas. Beginning with the 2010 NMP, county-level VMT was
obtained from state organizations, primarily departments of transportation. However, these data
are not readily available from all states. In addition, not all states provide this information on the
same level. For example, many states provide VMT for all public roads, while the state of
Colorado provided this information for state highways only. VMT are presented in Table 4-14,
where available.
The sites with the highest county-level VMT, where available, are CELA (Los Angeles
County, CA), SPIL and NBIL (Cook County, IL), and PXSS and SPAZ (Maricopa County, AZ).
The sites with the lowest county-level VMT, where available, are UCSD (Union County, SD),
GLKY (Carter County, KY), and CHSC (Chesterfield County, SC). A Pearson correlation
coefficient calculated between the average hydrocarbon concentrations and VMT, where
available is relatively weak (0.23), although indicating a slight positive relationship between the
two. However, this correlation is higher than correlations calculated for the 2007 and 2008-2009
NMP reports (-0.02 and 0.06, respectively). It is important to note that many of the sites with
larger VMT did not measure VOC (such as CELA, RUCA, and SJJCA). In addition, county-
level VMT were not readily available for Rhode Island, New York, Oklahoma, Massachusetts,
Texas, and the District of Columbia.
4-38
-------
4.4 Variability Analysis
This section presents the results of the three variability analyses described in
Section 3.4.2.
4.4.1 Coefficient of Variation and Inter-site Variability
The site-specific coefficient of variations and the inter-site comparison analyses are
discussed together in this section. Figures 4-la through 4-24a are graphical displays of site-
specific coefficient of variations (standard deviation vs. annual average concentration) for the
program-level pollutants of interest. Figures 4-lb through 4-24b are graphs depicting the site-
specific annual averages overlain on the program-level average, as discussed in Section 4.1. For
each program-wide pollutants of interest, the coefficient of variation graph is shown first,
followed by the inter-site variability graph. The figures are aligned this way because they tend to
complement each other; the data point with the highest average concentration and/or standard
deviation in the coefficient of variation graph is easily identified in the inter-site variability
graph. Further, the inter-site variability graphs allow the reader to see how the individual site-
specific annual averages feed into the program-level averages (i.e., if a specific site(s) is driving
the program average).
A couple of items to note about these figures: Some sites do not have annual averages
presented on the inter-site variability graphs because they did not meet the criteria specified in
Section 3.1. These same sites without annual averages on the inter-site variability graphs are not
represented by a data point on the coefficient of variation graphs either. For the sites sampling
metals, sites collecting PMio samples are presented in green while sites collecting TSP samples
are presented in pink.
The coefficient of variation figures show that few of the pollutants appear to exhibit the
"clustering" discussed in Section 3.4.2. Figure 4-10a for carbon tetrachloride exhibits clustering,
or uniformity in concentrations. Carbon tetrachloride is a pollutant that was used worldwide as a
refrigerant. However, it was identified as an ozone-depleting substance in the stratosphere and its
use was banned at the Kyoto Protocol. This pollutant has a long lifetime in the atmosphere, but
slowly degrades over time. Today, its concentration in ambient air is fairly ubiquitous regardless
of where it is measured. The coefficients of variation shown in Figure 4-10 not only support the
4-39
-------
expected uniformity (i.e., lack of variability) in "background" concentrations of carbon
tetrachloride, but are also a testament to the representativeness of the data generated under the
NMP. Figure 4-10b supports what is shown in Figure 4-10a. The inter-site variability is
relatively low, with the annual average concentrations of carbon tetrachloride ranging from
0.53 |ig/m3 (GPCO) to 0.72 |ig/m3 (SEWA and NBIL). Further, the confidence intervals for all
sites are less than 0.1 |ig/m3.
Figure 4-13a shows that 1,2-dichloroethane also appears to exhibit clustering. However,
it should be noted that the y-axis scale for the standard deviation is larger than the x-axis scale
for the annual average concentration. This indicates that there is more variability in the annual
average concentrations themselves, as indicated in Figure 4-13b by the relatively large
confidence intervals. This pollutant was not detected frequently (17 percent of samples) and
therefore has many zero substitutions included in each annual average, which contributes to both
the low range of annual average concentrations and the large confidence intervals.
Several of the coefficients of variations for the program-wide pollutants of interest follow
a linear trend line. Many of the annual averages of acetaldehyde, benzene, and
tetrachloroethylene exhibit this trend, as shown in Figures 4-2a, 4-5a, and 4-22a, respectively.
This means that as the annual averages increase, so do the standard deviations, indicating
increasing variability. This increased variability is a result of an increased range of individual
measurements that are used to calculate the annual average. This is supported by the inter-site
variability shown in Figures 4-2b, 4-5b, and 4-22b. The annual averages that extend well above
the program average for each pollutant tend to have a wider confidence interval associated with
them, indicating the likely influence of outliers. The annual averages well below the program
average tend to have much smaller confidence intervals.
Trichloroethylene appears to exhibit clustering in Figure 4-23a; however,
trichloroethylene is detected infrequently and yields relatively low annual averages and standard
deviations, due in part to the substitution of zeros for many non-detects. If the data point that
represents SPIL's annual average and standard deviation was removed and the scales adjusted,
the trichloroethylene concentrations would show more variability. The annual average
trichloroethylene concentration for SPIL is nearly eight times the annual average
4-40
-------
trichloroethylene concentration for any other site, as shown in both Figure 4-23a and 4-23b. If
the four highest concentrations (those greater than 2 |ig/m3) were removed from the calculation,
SPIL's annual average would still be 0.44 ± 0.14 |ig/m3, and more than four times the next
highest annual average (BTUT). SPIL's annual average trichloroethylene concentration is
discussed in Section 4.2.1.
Although many of the other pollutants do not exhibit easily classifiable clustering, or
even appear to follow a linear pattern, some of them are influenced by one or more data points
that do not fall in line with the others. For example, the larger standard deviation exhibited for
/>-dichlorobenzene in Figure 4-12a indicates that this particular average is likely influenced by
outliers. Figure 4-12b shows that this is BTUT's annual average, which is discussed in more
detail in Section 4.2.1. Excluding this data point would allow the rest to follow a more linear
trend line. The same is true for formaldehyde. The larger standard deviation exhibited for NBIL
in Figure 4-16a indicates that this particular annual average was likely influenced by outliers.
Although NBIL's annual average concentration of formaldehyde is not the highest among sites
sampling this pollutant, its confidence interval is the largest, as shown in Figure 4-16b. This
annual average is discussed also discussed in Section 4.2.1 as well as Section 4.2.2. Excluding
this data point would allow the rest to follow a more linear trend line.
Another example is shown in Figures 4-3a and 4-3b for acrylonitrile. The annual
averages for NBNJ, S4MO, and SPAZ are shown by the data points with the high standard
deviations in Figure 4-3a and by the large confidence intervals in Figure 4-3b. Without these
three data points on the coefficient of variation graph, Figure 4-3 a would exhibit clustering
around the annual average concentration. However, the standard deviations are relatively large
due predominantly to the large number of zeros substituted for non-detects (a 12 percent
detection rate). Vinyl chloride is another infrequently detected pollutant (less than 5 percent
detection rate) for which the annual averages have very large standard deviations in Figure 4-24a
and large confidence intervals in Figure 4-24b. While Figure 4-24a appears to show that
coefficient of variations for this pollutant follow a linear pattern, this figure also indicates a high
variability in the concentrations, as shown by the standard deviations (note how many of the
standard deviations are more than double the corresponding annual averages). Further, the
confidence intervals for nearly all the sites shown in Figure 4-24b are large.
4-41
-------
Figure 4-la. Coefficient of Variation Analysis of Acenaphthene Across 26 Sites
V = 1.6478x-2.3826
R; = 0.8526
I
10
O
o o
6 8 10
Annual Average Concentration (ng/m3)
Figure 4-lb. Inter-Site Variability for Acenaphthene
25
-n
ntfL
Monitoring Site
a Program Average D Site-Specific Annual Average
4-42
-------
Figure 4-2a. Coefficient of Variation Analysis of Acetaldehyde Across 30 Sites
1.5 1 2.5 3
Annual Average Concentration (ng/m3)
Figure 4-2b. Inter-Site Variability for Acetaldehyde
Monitoring Site
D Program Average
3 Site-Specific Annual Average
4-43
-------
Figure 4-3a. Coefficient of Variation Analysis of Acrylonitrile Across 24 Sites
0.15 0.2 0,25 0}
Annual Average Concentration (ug/m3)
035 0.4
Figure 4-3b. Inter-Site Variability for Acrylonitrile
• ^
Monitoring Site
D Program Average
D Site-Specific Annual Average
4-44
-------
Figure 4-4a. Coefficient of Variation Analysis of Arsenic Across 14 Sites
,=08255x-0.0374
R2 = 0.7895
0.6
fn
I
v = 0.3576x-0.1002
--"'6 R2 = 0.5663
0.2
0,4 0.6 0.8
Annual Average Concentration (ng/m3)
O PM10 O TSP
Linear(PMlO) ------ Linear(TSP)
1.2
Figure 4-4b. Inter-Site Variability for Arsenic
14
|
BOMA BTUT HBIL PAFL PXSS S4MO SEWA SJJCA UNVT MWOK OCOK PROK TMOK TOOK
Monitoring Site
D Program AveragePM10
D Program Average TSP
3 Site-Specific Annual Average
4-45
-------
Figure 4-5a. Coefficient of Variation Analysis of Benzene Across 24 Sites
I
o /c
-ti O
1 1.5
Annual Average Concentration (ng/m3)
Figure 4-5b. Inter-Site Variability for Benzene
I
rih
rii
rii
d^X ^ / ^
-------
Figure 4-6a. Coefficient of Variation Analysis of Benzo(a)pyrene Across 26 Sites
0.3
V = 0.9802x^0.0476
R2 = 0.647
T?
005
0.1 0.15
Annual Average Concentration jng/m3)
0.2
Figure 4-6b. Inter-Site Variability for Benzo(a)pyrene
0.3 -
0.25
E
c
Average Concentration
o
p ^
0.05
0
„
[
4
*
~
"
LA f
Monitoring Site
D Program Average
D Site-Specific Annual Average
4-47
-------
Figure 4-7a. Coefficient of Variation Analysis of Beryllium Across 14 Sites
0.025
0.02
-| 0.015
c
o
1
1
-D
I 0.01
•
0.005
0
V = 0.8352x- 0.0017
R! = 0.7196 .-••"
o.. --•'' o
o
o ..•-'"
V = 0.791x» 0.0013 ..••' O
R^O.621^/
QS^
j^~ O
o o
0 0.005 0.01 0.015 0.02 0.025
Annual Average Concentration (ng/rn3)
Figure 4-7b. Inter-Site Variability for Beryllium
tration (ng/m3)
3
3 f
•» i
Average Concer
c
* c
3 1-
* (.
-r
•
J-
T
I
1 [
I r4n /
I 1 /
T /
I . . /
T T 1 /
II m
T T
+ I-A
BOMA BTUT HBIL PAFL PXSS S4MO SEWA SJJCA UNVT MWOK OCOK PROK TMOK TOOK
Monitoring Site
D Program Aver Age PM10 D Program Average TSP D Site-Specific Annual Average
4-48
-------
Figure 4-8a. Coefficient of Variation Analysis of 1,3-Butadiene Across 24 Sites
025
0.15
5 o.i
y,
y = 0.7849x-0.002
0,1 0.15 0.2
Annual Average Concentration (ug/m3)
Figure 4-8b. Inter-Site Variability for 1,3-Butadiene
Monitoring Site
D Program Average D Site-Specific Annual Average
4-49
-------
Figure 4-9a. Coefficient of Variation Analysis of Cadmium Across 14 Sites
0.3
"m
-o 0.2
= l,0503x-0.0499
R* = 0.9373
V = 0.666X1-0.0053 O
R' = 0.9484 /
005
0.2 0.3 0.4 0.5
Annual Average Concentration (ng/m3)
O PM10 O TSP Lineai (PM10J Linear (TSP)
Figure 4-9b. Inter-Site Variability for Cadmium
0.7
f05
X
3
I
BOMA BTUT NBIL PAFL PXSS S4MO SEWA SJJCA LINVT MWOK OCOK PROK TMOK TOOK
Monitoring Site
DProgram Average PM10
D Program Average TSP
D Site-Specific Annual Average
4-50
-------
Figure 4-10a. Coefficient of Variation Analysis of Carbon Tetrachloride Across 24 Sites
0.2
E 0.15
1
o o
°o0o\°o
= -0.4139x-0.4046
R2 = 0.5823
0.5 0.4 0.5
Annual Average Concentration (ug/m3)
Figure 4-10b. Inter-Site Variability for Carbon Tetrachloride
Monitoring Site
D Program Average
D Site-Specific Annual Average
4-51
-------
Figure 4-lla. Coefficient of Variation Analysis of Chloroform Across 24 Sites
o
y = 1.6356x-0.0617
R' = 0.6136
O
e 1.2
.2 1
O
0.2
04 0.6 0.8
Annual Average Concentration (ug/m3)
1.2
Figure 4-1 Ib. Inter-Site Variability for Chloroform
Monitoring Site
D Program Average
D Site-Spe(ific Annual Average
4-52
-------
Figure 4-12a. Coefficient of Variation Analysis of/J-Dichlorobenzene Across 24 Sites
V = 3.5798x-0.1793
R2 = 0.6799
0.2 03 0.4
Annual Average Concentration (|.ig/m 3}
05
0.6
Figure 4-12b. Inter-Site Variability for/7-Dichlorobenzene
T
-h
T
1
r
j.
T T
r"n
f*T
rfi
T=^=>- L
-------
Figure 4-13a. Coefficient of Variation Analysis of 1,2-Dichloroethane Across 24 Sites
0.06
•| 0.03
o
I
-n
c
0.01
= 1.3094x- 0.0134
R' = 0.5282
0°
0.005
001 0.015
Annual Average Concentration (ug/m3)
D.02
0.025
Figure 4-13b. Inter-Site Variability for 1,2-Dichloroethane
Monitoring Site
D Program Average
D Site-Specific Annual Average
4-54
-------
Figure 4-14a. Coefficient of Variation Analysis of Ethylbenzene Across 24 Sites
r
a
~S 2
0.1 0.2
0.3 0.4 0.5 0,6
Annual Average Concentration (ug/m3)
0,7 0.8 0.9
Figure 4-14b. Inter-Site Variability for Ethylbenzene
Monitoring Site
3 Program Average
n Site-Specific Annual Average
4-55
-------
Figure 4-15a. Coefficient of Variation Analysis of Fluorene Across 26 Sites
E 15
I
V = 1.57G3x-3.2564
R! = 0.7966
2 4 6 8 10
Annual Average Concentration (ng/m3)
12
14
Figure 4-15b. Inter-Site Variability for Fluorene
Monitoring Site
D Program Average D Site-Specific Annual Average
4-56
-------
Figure 4-16a. Coefficient of Variation Analysis of Formaldehyde Across 30 Sites
1
>
01
O
I4
V = 0.9529x-0.839
RJ = 0.3096
O
CD
2 2.5 3
Annual Average Concentration (ug/m3)
Figure 4-16b. Inter-Site Variability for Formaldehyde
Monitoring Site
D Program Average D Site-specific Annual Average
4-57
-------
Figure 4-17a. Coefficient of Variation Analysis of Hexavalent Chromium Across 23 Sites
0.5
04
T
I
0.1
= l,6984x-0.019
R2 = 0.9405
0.15 0.2
Annual Average Concentration (ng/m3)
Figure 4-17b. Inter-Site Variability for Hexavalent Chromium
Monitoring Site
D Program Average
D Site-Specific Annual Average
4-58
-------
Figure 4-18a. Coefficient of Variation Analysis of Lead Across 14 Sites
ation (ng/m3)
r. c
1
Standard Dev
a, e
1
V = 0.8901x -0.3272
R' = 0.9588
~^_
/
/
O ,/
0 jS
£f V = 0.1219x- 1.576
O
024
Annu
O PM10 O T
6 8 10 12 14
al Average Concentration (ng/m3)
Figure 4-18b. Inter-Site Variability for Lead
16
14
12
i10
i
1 8
S
I
j.
4
2
0
1
T rh
T i ^
T
1 T FT
J 'r:, 1 1
BOMA BTUT NBIL PAFL PXSS S4MO SEWA SJJCA UNVT MWOK OCOK PROK TMOK TOOK
Monitoring Site
D Program Average PM10 Q Program Average TSP H Site-Specific Annual Average
4-59
-------
Figure 4-19a. Coefficient of Variation Analysis of Manganese Across 14 Sites
ation(ng/m3)
-> r-
71 C
Standard Dev
3 i
To"
|
o
iu
41
£,
*«
0
V = 1.5258x- 3.2535
R! = 0.8736 .X
/
/y = 0.4662x- 1.0625
^ = 0.9441 ..---U
7 ° -•"'
o..--- o
> .- --*""
v^
) S 10 15 20
Annual Average Concentration (ng/mS)
Figure 4-19b. Inter-Site Variability for Manganese
I
1 1
I
/ 1 nn
1 1 /
1
/T 1
i
i i F
ju i ^^ \;
j- \/
T T '
,— -, i /
+ ,/
BOMA BTUT HBIL PAFL PXSS S4MO SEWA SJJCA UNVT MWOK OCOK F'ROK TMOK TOOK
Monitoring Site
D Program Average PM 10 n Program Average TSP D Site-S|iecific Annual Average
15
4-60
-------
Figure 4-20a. Coefficient of Variation Analysis of Naphthalene Across 26 Sites
60 80 100
Annual Average Concentration (ng/m3|
120 140
Figure 4-20b. Inter-Site Variability for Naphthalene
"m
£
-Jj
J
£
u
£> 80
2
i
*
60
40
20
i
r
1
T
it
1 ,
r^
*
—
-
— •
///<3^^»
^ ^S-°
^
^
//
n
•
/
T
i
,&. p>
•^ v*"
^
*
-------
Figure 4-21a. Coefficient of Variation Analysis of Nickel Across 14 Sites
Q
14
= 1.001x-O.J876
R' = 0.8358
0.8
04
0.2
= 0,3238x»0.2805
R2 = 0.0431
o
0,8 1 1.2 1.4
Annual Average Concentration (ng/m3)
O TSP linear (PM10) Linear (TSP)
Figure 4-12b. Inter-Site Variability for Nickel
2.5
i
BOMA BTUT NBIL PAFL PXSS S4MO SEWA SJJCA UNVT MWOK OCOK PROK TMOK TOOK
Monitoring Site
D Program Average PM10
D Program Average TSP
D Site-Specific Annual Average
4-62
-------
Figure 4-22a. Coefficient of Variation Analysis of Tetrachloroethylene Across 24 Sites
0.3
V
Q
S 0.15
•Q
C
in
01
= 0.764x^0.0064 O
R2 = 0.8981
o
0.05 0.1
0.15 0.2 0.25 03
Annual Average Concentration (ng/m3)
0.35 0.4 0.45
Figure 4-22b. Inter-Site Variability for Tetrachloroethylene
Monitoring Site
D Program Average
Site-Specific Annual Average
4-63
-------
Figure 4-23a. Coefficient of Variation Analysis of Trichloroethylene Across 24 Sites
1.6
14
•o 08
•
-D
I
0.6
= 1.9387x + 0.0285
R2 = 0.9485
0.1 0.2
0.5 0.4 0.5 0.6
Annual Average Concentration (ug/m3)
0.7 0.8
0.9
Figure 4-23b. Inter-Site Variability for Trichloroethylene
«" o.^ O
Monitoring Site
n Program Average
n Site-Specific Annual Average
4-64
-------
Figure 4-24a. Coefficient of Variation Analysis of Vinyl Chloride Across 24 Sites
0.016
•2 0.008
o
1
00
0
0.0005
V = 4.2654x^0.0003
R1 = 0.846
0
0.001 00015 0.002
Annual Average Concentration (ng/m3)
0.0025
0.003
Figure 4-24b. Inter-Site Variability for Vinyl Chloride
0.001
0 4-
Monitoring Site
D Program Average
Site-Specific Annual Average
4-65
-------
4.4.2 Quarterly Variability Analysis
Figures 4-25 through 4-48 provide a graphical display of the quarterly average
concentrations for each of the program-level pollutants of interest. Quarterly averages are
calculated based on the criteria specified in Section 3.1.1. If the pollutant of interest has a
corresponding ATSDR Intermediate MRL, as defined in Section 3.3, then this value is indicated
on the graph and is plotted where applicable. The National Ambient Air Quality Standard
(NAAQS) for lead (TSP) is a 3-month standard. Because this time period aligns well with the
quarterly averages presented in this section, the NAAQS for lead (TSP) is also provided in
Figure 4-42b. Note that the scales on the PMio and TSP graphs are the same for a given
pollutant. The scales are also the same when a graph is split into two graphs for a given
pollutant.
Gaps, or missing quarterly averages, in the figures for the pollutants of interest can be
attributed to two reasons. First, some of the program-wide pollutants of interest were frequently
not detected in some quarters but were in others and have a quarterly average concentration of
zero as a result of the substitution of zeros for non-detects. One of the most apparent examples of
this is Figure 4-27 for acrylonitrile. This pollutant was frequently not detected (151 measured
detections out of 1,264 valid samples); of the 94 possible quarterly averages of this pollutant, 42
of them are zero. Thus, few quarterly averages appear in Figure 4-27. Further, most of the
remaining quarterly averages have relatively few measurements and many zero substitutions for
non-detects, resulting in relatively low quarterly averages. (Although this pollutant was detected
in only 12 percent of VOC samples collected, its risk screening value is relatively low; thus, all
151 measured detections of this pollutant failed screens.)
Another reason for gaps in the figures is due to the sampling duration of each site. Some
sites started late or ended early, which may result in a lack of quarterly averages. For example,
benzene is almost always detected in VOC samples, thus the gaps in Figure 4-29 are primarily
due to sampling duration. BMCO did not begin sampling VOC until the third quarter of 2012, in
September. Thus, the first and second quarterly averages are blank. Because the criteria in
Section 3.1.1 require a site to have 75 percent of the possible samples within a quarter (12 for a
site sampling on a l-in-6 day schedule), BMCO could not get a quarterly average for the third
4-66
-------
quarter because it did not sample long enough within that quarter. Therefore, the first quarterly
average that could be calculated for BMCO was the fourth quarter.
Both examples can be shown in Figure 4-36 for/?-dichlorobenzene. For example, PANJ
started sampling at the end of April 2010; thus, quarterly averages could not be calculated for the
first quarter. There were not enough valid samples in the second quarter to meet the 75 percent
criteria for PANJ for a quarterly average to be calculated, thus no second quarter average is
shown either. A similar situation is shown for GLKY, where sampling began in June 2010.
Conversely, CFINJ, RUVT, and UCSD sampled continuously in 2010 but did not detect this
pollutant in a single sample in the first quarter. As such, both the start and stop dates of each site
and the quarterly average criteria must be considered when interpreting the quarterly average
concentration graphs.
Some pollutants of interest, such as acetaldehyde, benzene, carbon tetrachloride,
ethylbenzene, formaldehyde, and naphthalene, were detected year-round. Comparing the
quarterly averages for the sites with four valid quarterly averages in a year may reveal a trend for
these pollutants. For example, formaldehyde averages tended to be highest in the third quarter, as
shown in Figure 4-40, with 18 of the 29 sites sampling formaldehyde exhibiting the highest
quarterly average during July through September, followed by second quarter (with seven), both
of which include warmer months of the year. Conversely, benzene averages tended to be higher
during the fourth quarter followed by the first quarter, or the colder months, as shown in
Figure 4-29. The seasonal behavior of benzene and formaldehyde suggests the influence of
reformulated gasoline (RFG) because the benzene content is typically lowered during warmer
periods (i.e., summer and spring). Refineries typically begin production of RFG during the
spring and end in the autumn. Additionally, methyl fert-butyl ether (MTBE) is often used as an
RFG additive in fuels to replace the lowered benzene content. Research has shown that the
combustion of fuels containing MTBE leads to the secondary production of formaldehyde. Thus,
while benzene concentrations decrease during the summer months, formaldehyde concentrations
may increase if MTBE is used in the gasoline blend.
4-67
-------
Other notable trends include benzo(a)pyrene with higher concentrations in the first
quarter; acenaphthene and ethylbenzene with higher concentrations in the third quarter; and
1,3-butadiene with higher concentrations in the fourth quarter.
Other notable trends may also be revealed in these graphs. Figure 4-37 for
1,2-dichlorethane shows that nearly all (over 95 percent) of the measured detections of this
pollutant were measured during the first and second quarter of 2010. Figure 4-48 for vinyl
chloride shows that this pollutant was infrequently detected, but like 1,2-dichloroethane, was
most frequently detected in the first two quarters of the year.
The quarterly average comparison also allows for the identification of sites with
unusually high concentrations of the pollutants of interest compared to other sites and when
those high concentrations were measured. This is evident in Figures 4-33a, 4-35, 4-41, 4-42a,
4-44, and 4-47 for cadmium, chloroform, hexavalent chromium, lead, naphthalene, and
trichloroethylene, respectively. For example, Figure 4-47 shows that the quarterly averages of
trichloroethylene for SPIL are significantly higher than for other sites sampling VOC.
Figures 4-33a and 4-42a show that S4MO's quarterly averages of cadmium and lead are
significantly higher than the quarterly averages for the other sites sampling metals. Conversely,
these graphs may also reveal when there is very little variability in the quarterly averages across
other sites. Figure 4-34 for carbon tetrachloride shows that the quarterly averages of this
pollutant did not vary significantly across the sites. Other pollutants may not exhibit such trends.
These graphs also show that only 10 of the 24 program-level pollutants of interest have
ATSDR Intermediate MRLs. For the 10 that do, the quarterly average concentrations were
significantly below their respective ATSDR Intermediate MRLs, generally by an order of
magnitude or more, which is also discussed in Section 4.2.2. In all 10 cases, the scale on the
graph is well below the ATSDR Intermediate MRL.
4-68
-------
Figure 4-25. Comparison of Average Quarterly Acenaphthene Concentrations
BOMA BTUT BXNY CAMS 35 CELA CHSC DEMI GLKY GPCO HOWI MONY NBIL PRRI
Monitoring Site
• 2nd Quarter
D 3rd Quarter
14th Quarter
Figure 4-25. Comparison of Average Quarterly Acenaphthene Concentrations (Continued)
PXSS R1VA ROCH RUCA S4MO SDGA 5EWA SJJCA 5KFL SYFL TONY UNVT WADC
Monitoring Site
12nd Quarter
D 3rd Quarter
14th Quarter
4-69
-------
Figure 4-26. Comparison of Average Quarterly Acetaldehyde Concentrations
AZFL BMCO BRCO BTUT CHHJ DEMI ELWJ GPCO INDEM MWOK MBIL NBNJ OCOK OREL PACO
Monitoring Site
• lit Quarter Bind Quarter n 3rd Quarter • 4th Quarter
Figure 4-26. Comparison of Average Quarterly Acetaldehyde Concentrations (Continued)
PROK PXSS RICO RLICO S4MO SEWA SKFL SPIL SSSD SYFL TMOK TOOK UCSD UNVT WPIN
Monitoring Site
• IstQuarter •2ndQuarter a 3idQnaiter •4thQuarter
4-70
-------
Figure 4-27. Comparison of Average Quarterly Acrylonitrile Concentrations
CO
I
S 0.8
:;:'
S 0.6
BTLIT BURVT CHHJ DEMI ELNJ GLKY GPCO MWOK NBIL NBHJ OCOK PANJ
Monitoring Site
• 1st Quarter • 2nd Quarter D 3rd Quarter • 4th Quarter
Figure 4-27. Comparison of Average Quarterly Acrylonitrile Concentrations (Continued)
1.4
1.2
_ 1
: 0.8
i 0.6
PROK PXSS RUVT S4MO SEWA SPAZ 5PIL SSSD TMOK TOOK UCSD UNVT
Monitoring Site
• IstQuarter Bind Quarter D 3rd Quarter • 4th Quarter
4-71
-------
Figure 4-28a. Comparison of Average Quarterly Arsenic (PMio) Concentrations
BOMA BTUT HBIL
• 1st Quarter
PAFL PXSS S4MO SEWA
Monitoring Site
• 2nd Quarter D 3rd Quarter
SJJCA UNVT
14th Quarter
Figure 4-28b. Comparison of Average Quarterly Arsenic (TSP) Concentrations
11st Quarter
PROK
Monitoring Site
• 2nd Quarter D 3rd Quarter
14th Quarter
4-72
-------
Figure 4-29. Comparison of Average Quarterly Benzene Concentrations
BMCO BRCO BTUT BURVT CHNJ DEMI ELNJ GLKY GPCO MWOK NBIL MBMJ OCOK PACO PAN)
Monitoring Site
• 1st Quarter Bind Quarter D 3rd Quarter • 4th Quarter
Figure 4-29. Comparison of Average Quarterly Benzene Concentrations (Continued)
ATSDFi Intermediate MRL = 10 ug/m3
PROK PXSS RICO RUCO RUVT S4MO SEWA SPAZ SPIL S5SO TMOK TOOK UOD UNVT
Monitoring Site
• 1st Quarter • 2nd Quarter D 3rd Quarter • 4th Quarter
4-73
-------
Figure 4-30. Comparison of Average Quarterly Benzo(a)pyrene Concentrations
3.5
3 -
_ 2.5
i
icentration
o
u
1
3
0.5
0 -
LI •_
• •. 1 1
i . rh h n r •_
L.L
BOMA BTUT BXNY CAMS 35 CELA CHSC DEMI GLKY GPCO HOWI MONY NBIL PRRI
Monitoring Site
• 1st Quarter • 2nd Quarter D 3rd Quarter • 4th Quarter
Figure 4-30. Comparison of Average Quarterly Benzo(a)pyrene Concentrations Continued)
_2.5
m
E
1.5
• J Ll
PXSS RIVA ROCH RUCA S4MO SDGA SEWA SJJCA SKFL SYFL TONY UHVT WADC
Monitoring Site
• IstOuarter •2ndQuarter D3rdQuarter •4thQuarter
4-74
-------
Figure 4-31a. Comparison of Average Quarterly Beryllium (PMio) Concentrations
003
BOMA BTUT NBIL PAFL PXSS S4MO SEWA SJJCA UNVT
Monitoring Site
• 1st Quarter • 2nd Quarter D 3rd Quarter • 4th Quarter
Figure 4-31b. Comparison of Average Quarterly Beryllium (TSP) Concentrations
0.035
• 1st Quarter
PROK
Monitoring Site
12nd Quarter
TOOK
• 4th Quarter
4-75
-------
Figure 4-32. Comparison of Average Quarterly 1,3-Butadiene Concentrations
BMCO BRCO BTUT BURVT CHNJ DEMI ELNJ GLKY GPCO MWOK NBIL NBNJ OCOK PACO PAN)
• 1st Quarter
Monitoring Site
• 2nd Quarter D 3rd Quarter • 4th Quarter
Figure 4-32. Comparison of Average Quarterly 1,3-Butadiene Concentrations (Continued)
045
PROK PXSS RICO RUCO RUVT S4MO SEWA SPAZ SPIL «SD TMOK TOOK UCSD IJNVT
• 1st Quarter
Monitoring Site
12nd Quarter D 3rd Quarter • 4th Quarter
4-76
-------
Figure 4-33a. Comparison of Average Quarterly Cadmium (PMio) Concentrations
BOMA BTUT NBIL PAFL PXSS S4MO SEWA SJJCA UNVT
• 1st Quarter
Monitoring Site
• 2nd Quarter D 3rd Quarter • 4th Quarter
Figure 4-33b. Comparison of Average Quarterly Cadmium (TSP) Concentrations
05
PROK
Monitoring Site
• 1st Quarter
• 4th Quarter
4-77
-------
Figure 4-34. Comparison of Average Quarterly Carbon Tetrachloride Concentrations
ATSt'F; Intermediate MRl = 200 ug/m?
BTUT BURVT CHNJ DEMI ELNJ GLKY GPCO MWOK NBIL NBNJ OCOK PAH)
11st Quarter
Monitoring Site
• 2nd Quarter D 3rd Quarter
Figure 4-34. Comparison of Average Quarterly Carbon Tetrachloride Concentrations
(Continued)
ATSC'R Intermediate MRL = 200 Hg/m3
PROK PXSS RUVT S4MO SEWA SPAZ SPIL SSSD TMOK TOOK UCSD UNVT
Monitoring Site
• 1st O.uarter • 2nd Quarter P ird Quarter • 4th Quarter
4-78
-------
Figure 4-35. Comparison of Average Quarterly Chloroform Concentrations
ATSDR Intermediate MRL = 200 pg/m3
BTUT BIJRVT CHNJ DEMI ELMJ GLKY GPCO MWOK NBIL NBNJ OCOK PANJ
Monitoring Site
• 1st Quarter B2nd Quarter D 3rd Quarter • 4th Quarter
Figure 4-35. Comparison of Average Quarterly Chloroform Concentrations (Continued)
16
m" 1.4
I
0 0.6
0.2
ATSDR Intermediate MRL = 200 ug/m3
libanrt*
PROK PXSS RUVT S4MO SEWA SPAZ SPIL SSSD TMOK TOOK UCSD UNVT
Monitoring Site
• 1st Quarter •2nd Quarter D 3rd Quarter B4th Quarter
4-79
-------
Figure 4-36. Comparison of Average Quarterly /7-Dichlorobenzene Concentrations
I
o
1
<_l U.B
-Dichlorobenzene Concentrations
(Continued)
a
o 0.8
ATSC'R Intermediate MRL = 1000 |ig/m3
PROK PXSS RUVT S4MO 5EWA SPAZ SPIL SSSD TMOK TOOK UCSD LINVT
Monitoring Site
12nd Quarter n 3rd Quarter
14th Quarter
4-80
-------
Figure 4-37. Comparison of Average Quarterly 1,2-Dichloroethane Concentrations
BTUT BURVT CHNJ DEMI ELHJ GLKY SPCO MWOK NBIL HBNJ OCOK PANJ
Monitoring Site
• 1st Quarter
• 2nd Quarter
n ird Quarter
14th Quarter
Figure 4-37. Comparison of Average Quarterly 1,2-Dichloroethane Concentrations
(Continued)
PROK PXSS RUVT S4MQ SEWA SPAZ SPIL SSSD TMOK TOOK UCSD UNVT
Monitoring Site
12nd Quarter
a 3rd Quarter
14th Quarter
4-81
-------
Figure 4-38. Comparison of Average Quarterly Ethylbenzene Concentrations
3 -
2.5
m"
1 *
c
o
age Concentr
Vi
HI
<
V
§ '
0,5
0 -
ATSC'R Intermediate
MRL = 9000 ug/m3
I
BMCO
g
n
BRCO
—
^h
,
i
1 MM I
BTUT BURVT CHHJ
[i
DEMI
ELNJ
^
GLKY
~
|
J
iL
-
k" ^IM
GPCO MWOK NBIL MBNJ OCOK
I
1
PACO
Monitoring Site
• 1st Quarter
• 2nd Quarter
D 3rd Quarter
• 4th Quarter
Figure 4-38. Comparison of Average Quarterly Ethylbenzene Concentrations (Continued)
s
i 1.5
1 i
3
05
ATSDR Intermediate MRL = 9000 Mg/m3
JIMlil
PAWJ PROK PXSS RICO RLICO RUVT S4MO SEWA 5PAZ SPIL SSSD TMOK TOOK LIC5D UNVT
Monitoring Site
• IstQuarter •2ndQuarter D3rdOuarter •4thOuarter
4-82
-------
Figure 4-39. Comparison of Average Quarterly Fluorene Concentrations
BOMA BTLIT BXNY CAMS ?,5 CELA CHSC DEMI GLKY GPCO HOWI MONY NBIL PRRI
• 1st Quarter
Monitoring Site
• 2nd Quarter D 3rd Quarter • 4th Quarter
Figure 4-39. Comparison of Average Quarterly Fluorene Concentrations (Continued)
RIVA (•''
-------
Figure 4-40. Comparison of Average Quarterly Formaldehyde Concentrations
rtT'.C'Fi Intermediate MFlL=40 iig.'ni
AZFL BMCO BRCO BTLIT CHNJ DEMI ELHJ GPCO INDEM M'A'OK NBIL UBHJ OCOK ORFL PACO
• 1st Quarter
Monitoring Site
• 2nd Quarter D 3rd Quarter • 4th Quarter
Figure 4-40. Comparison of Average Quarterly Formaldehyde Concentrations (Continued)
PROK PXVS RICO RLICO S4MO SEWA SKFL SPIL SSSD 5YFL TMOK TOOK UCSD LIN^T WPIN
Monitoring Site
• IstQuarter •2ndQuarter D3rdQuarter •4thQuarter
4-84
-------
Figure 4-41. Comparison of Average Quarterly Hexavalent Chromium Concentrations
£. 0.5
ATSC'R Intermediate MRL = 300 ng/m3
BOMA BTUT BXNY CAMS 35 CAMS 85 CHSC DEMI GLKY GPCO HOWI MONV NBIL
11st Quarter
Monitoring Site
12nd Quarter n 3rd Quarter • 4th Quarter
Figure 4-41. Comparison of Average Quarterly Hexavalent Chromium Concentrations
(Continued)
!,
s
o 0.4
ATSDR Intermediate MRL = 300 ng/m3
PRRI PXSS RIVA ROCH S4MO SDGA SEWA SKFL SYFL UNVT WADC
Monitoring Site
• lit Quarter • 2nd Quarter G 3rd Quarter • 4th Quarter
4-85
-------
Figure 4-42a. Comparison of Average Quarterly Lead (PMio) Concentrations
£ 10
BOMA BTUT NBIL PAFL PXSS S4MO SEWA SJJCA UHVT
• 1st Quarter
Monitoring Site
• 2nd Quarter D 3rd Quarter
14th Quarter
Figure 4-42b. Comparison of Average Quarterly Lead (TSP) Concentrations
16
NAAQS for Lead (TSP) = 150 ng/m3
10
I
a
MWOK
• 1st Quarter
OCOK PROK
Monitoring Site
• 2nd Quarter n 3rd Quarter
TMOK
TOOK
• 4th Quarter
4-86
-------
Figure 4-43a. Comparison of Average Quarterly Manganese (PMio) Concentrations
BOMA BTUT MBIL F'AFL PXSS S4MO SEWA SJJCA UNVT
• 1st Quarter
Monitoring Site
• 2nd Quarter D 3rd Quarter • 4th Quarter
Figure 4-43b. Comparison of Average Quarterly Manganese (TSP) Concentrations
• 1st Quarter
Monitoring Site
• 2nd Quartet D 3rd Quarter
• 4th Quarter
4-87
-------
Figure 4-44. Comparison of Average Quarterly Naphthalene Concentrations
550
E 400
*
3 300
200
BOMA BTUT BXNY CAMS 35 CELA CHSC DEMI GLKY GPCO HOWI MONY NBIL
Monitoring Site
• 1st Quarter Bind Quarter D 3rd Quarter • 4th Quarter
Figure 4-44. Comparison of Average Quarterly Naphthalene Concentrations (Continued)
550
500
PXSS RIVA ROCH RUCA S4MO SDGA SEWA SJJCA SKFL SYFL TONY IJHVT WADC
• 1st Quarter
Monitoring Site
• 2nd Quarter D ?.rd Quarter
• 4th Quarter
4-S
-------
Figure 4-45a. Comparison of Average Quarterly Nickel (PMio) Concentrations
ATSC'R Intermediate MRL = 200 ng/m3
BOMA
BTUT HBIL
• 1st Quarter
PAFL PXSS S4MO SEWA
Monitoring Site
• 2nd Quarter D 3rd Quarter
SJJCA UNVT
14th Quarter
Figure 4-45b. Comparison of Average Quarterly Nickel (TSP) Concentrations
ATSt R Intermediate MRL = 200 ng/m3
1
tt
I
I
3
I
I
• 1st Quarter
PROK
Monitoring Site
• 2nd Quarter D 3rd Quarter
• 4th Quarter
4-89
-------
Figure 4-46. Comparison of Average Quarterly Tetrachloroethylene Concentrations
BTUT BURVT CHHJ DEMI
11st Quarter
ELNJ GLKY GPCO MWOK NEIL
Monitoring Site
• 2nd Quarter D 3rd Quarter
NBNJ OCOK PANJ
• 4th Quarter
Figure 4-46. Comparison of Average Quarterly Tetrachloroethylene Concentrations
(Continued)
PROK PXSS RUVT S4MO SEWA
SPAZ SPIL
Monitoring Site
2nd Quarter D 3rd Quarter
SSSD TMOK TOOK LICSD UNVT
14th Quarter
4-90
-------
Figure 4-47. Comparison of Average Quarterly Trichloroethylene Concentrations
ATSDR Intermediate MRL = 500 Ug/m3
I °8
BTUT BURVT CHNJ DEMI ELNJ GLKY GPCO MWOK HBIL MBMJ OCOK PAN)
Monitoring Site
12nd Quarter D 3rd Quarter • 4th Quarter
Figure 4-47. Comparison of Average Quarterly Trichloroethylene Concentrations
(Continued)
ATSt'R Intermediate MRL = 500 irg/nr?
PROK PXSS RUVT S4MO SEWA SPAZ SPIL SSSD TMOK TOOK UCSD LINVT
Monitoring Site
• lit Quarter • 2nd Quarter n 3rd Quarter • 4tli Quarter
4-91
-------
Figure 4-48. Comparison of Average Quarterly Vinyl Chloride Concentrations
BTUT BURVT CHNI DEMI ELNJ GLKY GPCO MWOK NBIL NBNJ OCOK PANJ
Monitoring Site
• 1st Quarter Bind Quarter D 3rd Quarter • 4th Quarter
Figure 4-48. Comparison of Average Quarterly Vinyl Chloride Concentrations (Continued)
ATSDR Intermediate MRL = SO ug/m3
PROK PXSS RUVT S4MO SEWA SPAZ SPIL SSSD TMOK TOOK LIC5D IJNVT
Monitoring Site
• IstQuarter •2ndQuarter D3rdQuaiter •4thQuartei
4-92
-------
4.5 Greenhouse Gases
Table 4-15 presents the program-level average concentrations for the 10 GHGs measured
using Method TO-15, in descending order by GWP. As shown, most of the GHGs were detected
in nearly every sample collected (a total 1,264 valid VOC samples). Chloroform, bromomethane,
and 1,1,1-trichloroethane were the only pollutants detected in less than 85 percent of VOC
samples collected, although all 10 were still detected in greater than 80 percent of samples.
Dichlorodifluoromethane and dichlorotetrafluoroethane have the highest GWPs of the GHGs
measured by Method TO-15 (10,900 and 10,000 respectively), while bromomethane and
dichloromethane have the lowest GWP (5 and 8.7, respectively). Dichloromethane has the
highest program average concentration among the GHGs measured, although the associated
confidence interval indicates that this concentration is likely influenced by outliers. A review of
the data shows that two sites contributed to this high average concentration. Four concentrations
of this pollutant greater than 1000 jig/m3 were measured at BTUT and GPCO (one at GPCO and
three at BTUT). An additional six concentrations greater than 100 |ig/m3 were measured at
BTUT. Besides dichloromethane, only three additional GHGs shown in Table 4-15 have
program averages greater than 1 |ig/m3: dichlorodifluoromethane, trichlorofluoromethane, and
chloromethane.
Table 4-15. Greenhouse Gases Measured by Method TO-15
Pollutant
Dichlorodifluoromethane
Dichlorotetrafluoroethane
Trichlorotrifluoroethane
Trichlorofluoromethane
Carbon Tetrachloride
1,1,1 -Trichloroethane
Chloroform
Global
Warming
Potential1
(100 yrs)
10,900
10,000
6,130
4,750
1,400
146
31
Total # of
Measured
Detections
1,264
1,263
1,264
1,264
1,258
1,067
1,021
2010
Program
Average
(Ug/m3)
2.86
±0.02
0.14
±0.01
0.73
±0.01
1.66
±0.03
0.63
±0.01
0.06
±0.01
0.19
±0.03
:GWP presented here are taken from the Intergovernmental Panel on Climate Change
(IPCC) Fourth Assessment Report (AR4) (IPCC, 2012).
4-93
-------
Table 4-15. Greenhouse Gases Measured by Method TO-15 (Continued)
Pollutant
Chloromethane
Dichloromethane
Bromomethane
Global
Warming
Potential1
(100 yrs)
13
8.7
5
Total # of
Measured
Detections
1,264
1,263
1,040
2010
Program
Average
(Hg/m3)
1.31
±0.01
10.63
±9.59
0.05
±0.01
:GWP presented here are taken from the Intergovernmental Panel on Climate Change
(IPCC) Fourth Assessment Report (AR4) (IPCC, 2012).
4-94
-------
5.0 Sites in Arizona
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS and UATMP sites in Arizona, and integrates these
concentrations with emissions, meteorological, and risk information. Data generated by sources
other than ERG are not included in the data analyses contained in this report. Readers are
encouraged to refer to Sections 1 through 4 and the glossary (Appendix P) for detailed
discussions and definitions regarding the various data analyses presented below.
5.1 Site Characterization
This section characterizes the Arizona monitoring sites by providing geographical and
physical information about the location of the sites and the surrounding areas. This information
is provided to give the reader insight regarding factors that may influence the air quality near the
sites and assist in the interpretation of the ambient monitoring measurements.
The Arizona monitoring sites are located in Phoenix, Arizona. Figures 5-1 and 5-2 are
composite satellite images retrieved from ArcGIS Explorer showing the monitoring sites in their
urban locations. Figure 5-3 identifies point source emissions locations by source category, as
reported in the 2008 NEI for point sources. Note that only sources within 10 miles of the sites are
included in the facility counts provided in Figure 5-3. Thus, sources outside the 10-mile radius
have been grayed out, but are visible on the map to show emissions sources outside the 10-mile
boundary. A 10-mile boundary was chosen to give the reader an indication of which emissions
sources and emissions source categories could potentially have a direct effect on the air quality at
the monitoring sites. Further, this boundary provides both the proximity of emissions sources to
the monitoring sites as well as the quantity of such sources within a given distance of the sites.
Table 5-1 describes the area surrounding each monitoring site by providing supplemental
geographical information such as land use, location setting, and locational coordinates.
5-1
-------
Figure 5-1. Phoenix, Arizona (PXSS) Monitoring Site
to
-------
Figure 5-2. South Phoenix, Arizona (SPAZ) Monitoring Site
-------
Figure 5-3. NEI Point Sources Located Within 10 Miles of PXSS and SPAZ
112'25'CTW 112'2Q'0*W
112-10'0-W 112'5'Q-W
Legend
PXSS NATTS site
Note: Due to facili'.v density and collocation, the total facilities
displayed may not represent a\\ facilities within the area of interest.
10 mile radius
•jjf SPAZ UATMP site County boundary
Source Category Group (No. of Facilities)
•fi Aerospace/Aircraft Manufacturing (1)
4> Aircraft Operations (39)
c Chemical Manufacturing (1)
* Electricity Generation via Combustion (4)
-& Flexible Polyurethane Foam Production (1 )
R Furniture Plant (3)
? Miscellaneous Commercial/Industrial (4)
M Miscellaneous Manufacturing (1)
R Rubber and Miscellaneous Plastics Products (1 )
5-4
-------
Table 5-1. Geographical Information for the Arizona Monitoring Sites
Site
Code
PXSS
SPAZ
AQS Code
04-013-9997
04-013-4003
Location
Phoenix
Phoenix
County
Maricopa
Maricopa
Micro- or
Metropolitan
Statistical Area
Phoenix-Mesa-
Glendale, AZ
MSA
Phoenix-Mesa-
Glendale, AZ
MSA
Latitude
and
Longitude
33.503731,
-112.095809
33.40316,
-112.07533
Land Use
Residential
Residential
Location
Setting
Urban/City
Center
Urban/City
Center
Additional Ambient Monitoring Information1
Haze, CO, SO2, NO, NO2, NOX, PAMS, O3,
Meteorological parameters, PM10, PM25, PM Coarse,
PM2 5 Speciation.
CO, PAMS, O3, Meteorological parameters, PM2 5,
PM10,PM Coarse.
BOLD ITALICS = EPA-designated NATTS Site.
-------
PXSS is located in central Phoenix while SPAZ is located farther south. Figure 5-1 shows
that PXSS is located in a highly residential area on North 17th Avenue in central Phoenix. The
Grand Canal is shown at the bottom of Figure 5-1. The monitoring site is approximately
three-quarters of a mile east of 1-17 and 2 miles north of 1-10. Figure 5-2 shows that SPAZ is
located in South Phoenix, near the intersection of W. Tamarisk Avenue and S. Central Avenue.
SPAZ is bounded on the west side by residential properties and commercial properties on the
east side. SPAZ is located approximately 1 mile south of 1-17.
As Figure 5-3 shows, SPAZ and PXSS are located within 7 miles of each other. The
majority of emissions sources are located between the sites, to the south of PXSS and north of
SPAZ. The source category with the highest number of sources near these monitoring sites is the
aircraft operations source category, which includes airports as well as small runways, heliports,
or landing pads. The emissions source nearest PXSS is a landing strip at a hospital while the
source nearest SPAZ is a landing strip at a police station.
Table 5-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the areas surrounding the Arizona
monitoring sites. Table 5-2 also includes a vehicle registration-to-county population ratio
(vehicles-per-person) for each site. In addition, the population within 10 miles of each site is
presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-level
vehicle registration-to-population ratio to the 10-mile population surrounding each monitoring
site. Table 5-2 also contains annual average daily traffic information. For both sites, traffic data
for locations along 1-17 were selected. Finally, Table 5-2 presents the daily VMT for Maricopa
County.
5-6
-------
Table 5-2. Population, Motor Vehicle, and Traffic Information for the Arizona Monitoring
Sites
Site
PXSS
SPAZ
Estimated
County
Population1
3,827,371
County-level
Vehicle
Registration2
3,739,918
Vehicles per
Person
(Registration:
Population)
0.98
Population
within 10
miles3
1,473,228
898,861
Estimated
10-mile
Vehicle
Ownership
1,439,566
878,323
Annual
Average
Daily
Traffic4
193,000
130,000
County-
level Daily
VMT5
89,448,000
1 County-level population estimates reflect 2010 data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the Arizona DOT (AZ DOT, 2010)
310-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2009 data from the Arizona DOT (AZ DOT, 2009)
5 County-level VMT reflects 2010 data for all public roads from the Arizona DOT (AZ DOT, 2011)
BOLD ITALICS = EPA-designated NATTS Site
Observations from Table 5-2 include the following:
• Maricopa County has the fourth highest county population and second highest
county-level vehicle registration compared to other counties with NMP sites.
• The vehicle-per-person ratio is just less than one vehicle per person. This ratio falls in
the middle of the range compared to other NMP sites.
• The 10-mile radius population and estimated vehicle ownership are higher near PXSS
than SPAZ.
• PXSS experiences a higher annual average traffic volume compared to SPAZ, based
on locations along 1-17. The traffic volume near PXSS is the fourth highest compared
to traffic volumes near other NMP sites.
• The Maricopa County daily VMT is the third highest compared to other counties with
NMP sites (where VMT data were available).
5.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in Arizona on sample days, as well as over the course of the year.
5.2.1 Climate Summary
The Phoenix area is located in the Salt River Valley, which is part of the Sonora Desert.
The area experiences mild winters and extremely hot and dry summers. Differences between the
daytime maximum temperature and overnight minimum temperature can be as high as 50°F. A
summer "monsoon" period brings precipitation to the area for part of the summer, while storms
5-7
-------
originating off the Pacific Ocean bring rain in the winter and early spring. Winds are generally
light (Bair, 1992, and WRCC, 2012).
5.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest these sites were
retrieved for 2010 (NCDC, 2010). The closest weather station to PXSS and SPAZ is located at
Phoenix Sky Harbor International Airport (WBAN 23183). Additional information about the Sky
Harbor weather station, such as the distance between the sites and the weather station, is
provided in Table 5-3. These data were used to determine how meteorological conditions on
sample days vary from normal conditions throughout the year.
Table 5-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 5-3 is the 95
percent confidence interval for each parameter. As shown in Table 5-3, average meteorological
conditions on sample days were representative of average weather conditions throughout the
year. Table 5-3 also shows that these sites experienced the lowest relative humidity levels among
NMP sites.
-------
Table 5-3. Average Meteorological Conditions near the Arizona Monitoring Sites
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
Average
Temperature
Average
Dew Point
Temperature
Average
Wet Bulb
Temperature
Average
Relative
Humidity
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Phoenix, Arizona - PXSS
Phoenix Sky
Harbor Intl
Airport
23183
(33.44, -111.99)
7.19
miles
136°
(SE)
Sample
Day
2010
86.3
±3.7
85.6
±1.7
75.5
±3.6
75.2
± 1.6
38.3
±2.8
38.8
±1.3
56.0
±2.1
56.0
±1.0
32.0
±3.9
32.8
±1.7
1010.8
±1.1
1010.7
±0.5
5.1
±0.5
5.2
±0.2
South Phoenix, Arizona - SPAZ
Phoenix Sky
Harbor Intl
Airport
23183
(33.44, -111.99)
5.46
miles
70°
(ENE)
Sample
Day
2010
84.8
±5.9
85.6
±1.7
74.1
±5.8
75.2
±1.6
38.6
±4.4
38.8
±1.3
55.5
±3.5
56.0
±1.0
34.0
±6.1
32.8
± 1.7
1010.6
±1.6
1010.7
±0.5
5.2
±0.8
5.2
±0.2
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
5.2.3 Back Trajectory Analysis
Figure 5-4 is the composite back trajectory map for days on which samples were
collected at the PXSS monitoring site in 2010. Included in Figure 5-4 are four back trajectories
per sample day. Figure 5-5 is the corresponding cluster analysis for 2010. Similarly, Figure 5-6 is
the composite back trajectory map for days on which samples were collected at SPAZ and
Figure 5-7 is the corresponding cluster analysis. An in-depth description of these maps and how
they were generated is presented in Section 3.5.2.1. For the composite maps, each line represents
the 24-hour trajectory along which a parcel of air traveled toward the monitoring site on a given
sample day and time. For the cluster analyses, each line corresponds to a back trajectory
representative of a given cluster of trajectories. For all maps, each concentric circle around the
sites in Figures 5-4 through 5-7 represents 100 miles.
Observations from Figures 5-4 and 5-5 for PXSS include the following:
• The 24-hour air shed domain was smaller for PXSS than for many other NMP
monitoring sites. The farthest away a trajectory originated from PXSS was off Baja
California, or less than 450 miles away. However, most trajectories (86 percent)
originated less than 250 miles from PXSS.
• Back trajectories originated from a variety of directions at PXSS, although many
trajectories originated from the southwest and west. A secondary group of trajectories
originated from the southeast. Trajectories also originated from the northwest, north,
and northeast.
• The cluster analysis map supports the observations above regarding the direction of
trajectory origin as well as the observations about trajectory distances. Nearly all of
the cluster trajectories originated within 300 miles of PXSS, and four of the six are
less than 200 miles long.
Observations from Figures 5-6 and 5-7 for SPAZ include the following:
• Samples were collected every 12 days at SPAZ, which is half the frequency of sample
collection at PXSS. As a result, fewer trajectories are shown in Figure 5-6.
• The composite trajectory map for SPAZ has a trajectory distribution pattern similar to
PXSS. The cluster analysis maps are also similar to each other. This is expected given
their close proximity to each other.
• Similar to PXSS, most trajectories (90 percent) originated within 250 miles of SPAZ.
5-10
-------
Figure 5-4. 2010 Composite Back Trajectory Map for PXSS
Figure 5-5. Back Trajectory Cluster Map for PXSS
5-11
-------
Figure 5-6. 2010 Composite Back Trajectory Map for SPAZ
Figure 5-7. Back Trajectory Cluster Map for SPAZ
5-12
-------
5.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at Phoenix Sky Harbor International
Airport were uploaded into a wind rose software program to produce customized wind roses, as
described in Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals"
positioned around a 16-point compass, and uses different colors to represent wind speeds.
Figure 5-8 presents three different wind roses for the PXSS monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figure 5-9 presents the three
wind roses and distance map for SPAZ.
Observations from Figures 5-8 and 5-9 for the Arizona monitoring sites include the
following:
• The NWS weather station at Phoenix Sky Harbor International Airport is the closest
weather station to both PXSS and SPAZ. The Phoenix Sky Harbor weather station is
located approximately 7.2 miles southeast of PXSS and 5.5 miles east-northeast of
SPAZ.
• Because the Phoenix Sky Harbor weather station is the closest weather station to both
sites, the historical and 2010 wind roses for PXSS are the same as those for SPAZ.
• The historical wind rose shows that easterly, westerly, and east-southeasterly winds
were the most commonly observed wind directions near PXSS and SPAZ. Winds
from the northwest, north, and northeast were infrequently observed, as were winds
from the south. Calm winds (< 2 knots) account for more than 15 percent of the
hourly wind measurements from 1999 to 2009.
• The 2010 wind patterns are similar to the historical wind patterns. Further, the sample
day wind patterns also resemble the historical and 2010 wind patterns, indicating that
conditions on sample days were representative of those experienced over the entire
year and historically.
5-13
-------
Figure 5-8. Wind Roses for the Phoenix Sky Harbor International Airport Weather Station
near PXSS
1999-2009 Historical Wind Rose
2010 Wind Rose
Calms: 15.63%
WIND SPEED
(Knots)
o *=
^| 17 • 21
^| 11 - 17
^| 7- 11
CH 4-7
!• 2- 4
Calms: 18.64%
2010 Sample Day Wind Rose
Distance between PXSS and NWS Station
5-14
-------
Figure 5-9. Wind Roses for the Phoenix Sky Harbor International Airport Weather Station
near SPAZ
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between SPAZ and NWS Station
i r u,.,.»,.,.««;
.1- L, j . E ,.,„ L,
5-15
-------
5.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Arizona monitoring sites in
order to allow analysts and readers to focus on a subset of pollutants through the context of risk.
For each site, each pollutant's preprocessed daily measurement was compared to its associated
risk screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by each monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 5-4 presents the pollutants of interest for PXSS and SPAZ. The pollutants that
failed at least one screen and contributed to 95 percent of the total failed screens for each
monitoring site are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of interest
are shaded and/or bolded. PXSS sampled for VOC, carbonyl compounds, PAH, metals (PMio),
and hexavalent chromium; SPAZ sampled for VOC only.
Observations from Table 5-4 include the following:
• The number of pollutants failing screens varied significantly between the two
monitoring sites; this is expected given the different pollutants measured at each site.
• Twenty-two pollutants failed at least one screen for PXSS, of which 14 are NATTS
MQO Core Analytes.
• Twelve pollutants, of which nine are NATTS MQO Core Analytes, were initially
identified as pollutants of interest for PXSS. Benzo(a)pyrene, cadmium, lead, nickel,
and trichloroethylene were added to PXSS's pollutants of interest because they are
NATTS MQO Core Analytes, even though they did not contribute to 95 percent of
PXSS's total failed screens. Four additional NATTS MQO Core Analytes were added
to PXSS's pollutants of interest, even though their concentrations did not fail any
screens: beryllium, chloroform, tetrachloroethylene, and vinyl chloride. These four
pollutants are not shown in Table 5-4.
• For PXSS, approximately 60 percent of the measured detections failed screens (of the
pollutants failing at least one screen).
5-16
-------
Table 5-4. Risk Screening Results for the Arizona Monitoring Sites
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
%of
Total
Failures
Cumulative
%
Contribution
Phoenix, Arizona - PXSS
Benzene
Carbon Tetrachloride
1,3-Butadiene
Naphthalene
Manganese (PM10)
Arsenic (PM10)
£>-Dichlorobenzene
Ethylbenzene
Hexavalent Chromium
Acetaldehyde
Formaldehyde
Acrylonitrile
1 ,2-Dichloroethane
Nickel (PM10)
Chloroprene
Benzo(a)pyrene
Cadmium (PM10)
1 ,2-Dibromoethane
Dichloromethane
Lead (PM10)
1 , 1 ,2,2-Tetrachloroethane
Trichloroethylene
0.13
0.17
0.03
0.029
0.005
0.00023
0.091
0.4
0.000083
0.45
0.077
0.015
0.038
0.0021
0.0021
0.00057
0.00056
0.0017
7.7
0.015
0.017
0.2
Total
61
61
57
57
54
44
42
40
31
20
20
13
9
7
o
3
526
61
61
59
59
59
59
52
61
57
20
20
13
9
59
3
21
59
1
61
59
1
21
875
100.00
100.00
96.61
96.61
91.53
74.58
80.77
65.57
54.39
100.00
100.00
100.00
100.00
11.86
100.00
4.76
1.69
100.00
1.64
1.69
100.00
4.76
60.11
11.60
11.60
10.84
10.84
10.27
8.37
7.98
7.60
5.89
3.80
3.80
2.47
1.71
1.33
0.57
0.19
0.19
0.19
0.19
0.19
0.19
0.19
11.60
23.19
34.03
44.87
55.13
63.50
71.48
79.09
84.98
88.78
92.59
95.06
96.77
98.10
98.67
98.86
99.05
99.24
99.43
99.62
99.81
100.00
South Phoenix, Arizona - SPAZ
Benzene
1,3-Butadiene
Carbon Tetrachloride
£>-Dichlorobenzene
Ethylbenzene
Acrylonitrile
1 ,2-Dichloroethane
Chloroprene
Trichloroethylene
0.13
0.03
0.17
0.091
0.4
0.015
0.038
0.0021
0.2
Total
29
29
29
26
23
9
4
2
1
152
29
29
29
29
29
9
4
2
13
173
100.00
100.00
100.00
89.66
79.31
100.00
100.00
100.00
7.69
87.86
19.08
19.08
19.08
17.11
15.13
5.92
2.63
1.32
0.66
19.08
38.16
57.24
74.34
89.47
95.39
98.03
99.34
100.00
PXSS failed the second highest number of screens (526) among all NMP sites, behind
only S4MO with 574 failed screens (refer to Table 4-8 of Section 4.2). However, the
failure rate for PXSS, when incorporating all pollutants with screening values, is
much lower, at 22 percent. This is due primarily to the relatively high number of
pollutants sampled at this site, as discussed in Section 4.2.
Nine pollutants failed screens for SPAZ, of which four are NATTS MQO Core
Analytes. Six pollutants were initially identified as pollutants of interest for SPAZ.
Trichoroethylene was added to SPAZ's pollutants of interest because it is a NATTS
5-17
-------
MQO Core Analyte, even though this pollutant did not contribute to 95 percent of
SPAZ's total failed screens. Two additional NATTS MQO Core Analytes were added
to SPAZ's pollutants of interest, even though their concentrations did not fail any
screens: chloroform and tetrachloroethylene. These two pollutants are not shown in
Table 5-4. While vinyl chloride is also a NATTS MQO Core Analyte, it was not
detected at SPAZ, and therefore not added to the list of pollutants of interest.
• For SPAZ, nearly 88 percent of the measured detections failed screens (of the
pollutants failing at least one screen).
• The following pollutants of interest failed 100 percent of screens for both sites:
acrylonitrile, benzene, carbon tetrachloride, chloroprene, and 1,2-dichloroethane.
1,3-Butadiene also failed 100 percent of screens for SPAZ. Acetaldehyde,
1,2-dibromoethane, 1,1,2,2-tetrachloroethane, and formaldehyde failed 100 percent of
screens atPXSS. However, 1,2-dibromoethane and 1,1,2,2-tetrachloroethane were
each detected only once at PXSS.
5.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Arizona monitoring sites. Concentration averages are provided for the pollutants of interest
for each Arizona site, where applicable. Concentration averages for select pollutants are also
presented graphically for each site, where applicable, to illustrate how each site's concentrations
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at
each site, where applicable. Additional site-specific statistical summaries are provided in
Appendices J, L, M, N, and O.
5.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each Arizona site, as described in Section 3.1. The quarterly average of a particular pollutant
is simply the average concentration of the preprocessed daily measurements over a given
calendar quarter. Quarterly average concentrations include the substitution of zeros for all non-
detects. A site must have a minimum of 75 percent valid samples of the total number of samples
possible within a given quarter for a quarterly average to be calculated. An annual average
includes all measured detections and substituted zeros for non-detects for the entire year of
sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the Arizona
5-18
-------
monitoring sites are presented in Table 5-5, where applicable. Note that concentrations of the
PAH, metals, and hexavalent chromium for PXSS are presented in ng/m3 for ease of viewing.
Also note that if a pollutant was not detected in a given calendar quarter, the quarterly average
simply reflects "0" because only zeros substituted for non-detects were factored into the
quarterly average concentration.
Observations for PXSS from Table 5-5 include the following:
• The pollutant with the highest annual average concentration by mass is benzene. This
is the only pollutant with an annual average concentration greater than 1 |ig/m3
(1.38 ± 0.20 |ig/m3). Benzene appears to exhibit a seasonal trend, with higher
quarterly averages for the colder months of the year. However, the confidence
intervals indicate that this difference is not statistically significant.
• Similar to benzene, concentrations of benzo(a)pyrene, 1,3-butadiene, ethylbenzene,
/>-dichlorobenzene, naphthalene, and tetrachloroethylene also appear to be higher
during the colder months of the year. Again, the difference in quarterly average
concentrations is not statistically significant for these pollutants.
• Acrylonitrile concentrations appear much higher during the first quarter of 2010 and
then decrease across the remaining quarters. However, the confidence intervals for
each are relatively high compared to the averages themselves. This pollutant was
detected in only 13 of 61 samples, resulting in many zero substitutions into the
averages. This pollutant was detected more frequently in the first half of the year than
the second (seven times in the first quarter, three times in the second, twice in the
third, and once in the fourth).
• The fourth quarter average concentration of trichloroethylene has a confidence
interval higher than the concentration itself. The highest concentration of
trichloroethylene (0.829 |ig/m3) was measured on November 22, 2010 and is nearly
six times higher than the next highest concentration (0.144 |ig/m3), measured on
January 26, 2010.
• Note that neither acetaldehyde nor formaldehyde have quarterly or annual average
concentrations presented in Table 5-5. This is because maintenance of the primary
carbonyl compound sampler at PXSS led to a problem with the ozone denuder,
resulting in the invalidation of the sampling results from mid-February 2010 through
the end of the year.
5-19
-------
Table 5-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Arizona Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Ug/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(Ug/m3)
4th
Quarter
Average
(Ug/m3)
Annual
Average
(Ug/m3)
Phoenix, Arizona - PXSS
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
Ethylbenzene
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Arsenic (PM10)a
Benzo(a)pyrene a
Beryllium (PM10)a
Cadmium (PM10)a
Hexavalent Chromium3
Lead(PM10)a
Manganese (PM10) a
Naphthalene a
Nickel (PM10)a
16/20
13/61
61/61
59/61
61/61
61/61
52/61
61/61
16/20
57/61
21/61
1/61
59/59
21/59
42/59
59/59
57/57
59/59
59/59
59/59
59/59
NA
0.11
± 0.07
1.60
± 0.47
0.28
± 0.12
0.73
± 0.03
0.29
± 0.07
0.16
± 0.06
0.66
± 0.22
NA
0.53
± 0.21
0.02
± 0.02
<0.01
± <0.01
0.51
± 0.21
0.13
± 0.10
O.01
± <0.01
0.11
± 0.04
0.10
± 0.05
3.33
± 1.03
8.35
± 2.33
116.50
± 27.38
1.08
± 0.42
NA
0.05
± 0.06
1.03
± 0.22
0.13
± 0.04
0.66
± 0.07
0.40
± 0.11
0.13
± 0.04
0.47
± 0.11
NA
0.25
± 0.06
0.02
± 0.01
0
0.53
± 0.40
0.01
± 0.01
<0.01
± <0.01
0.11
± 0.09
0.13
± 0.06
3.46
± 1.76
17.02
± 8.10
84.39
± 17.84
1.45
± 0.57
NA
0.03
± 0.04
0.95
± 0.30
0.10
± 0.05
0.61
± 0.08
0.37
± 0.13
0.11
± 0.06
0.42
± 0.15
NA
0.24
± 0.11
0.03
± 0.02
0
0.39
± 0.16
0
0.01
± <0.01
0.08
± 0.03
0.12
± 0.05
2.06
± 0.49
9.58
± 2.20
51.97
± 11.84
1.10
± 0.37
NA
0.01
± 0.02
1.97
± 0.42
0.33
± 0.10
0.66
± 0.04
0.42
± 0.11
0.23
± 0.06
0.85
± 0.22
NA
0.60
± 0.18
0.08
± 0.11
0
0.81
± 0.30
0.14
± 0.09
0.01
± 0.01
0.17
± 0.06
0.16
± 0.07
4.85
± 1.82
14.20
± 3.73
109.87
± 29.97
1.27
± 0.31
NA
0.05
± 0.03
1.38
± 0.20
0.21
± 0.05
0.66
± 0.03
0.37
± 0.05
0.16
± 0.03
0.60
± 0.10
NA
0.40
± 0.08
0.03
± 0.03
O.01
± <0.01
0.56
± 0.14
0.07
± 0.03
0.01
± <0.01
0.12
± 0.03
0.13
± 0.03
3.42
± 0.71
12.38
± 2.53
89.15
± 12.41
1.23
± 0.21
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of
viewing.
5-20
-------
Table 5-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Arizona Monitoring Sites (Continued)
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Ug/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(Ug/m3)
4th
Quarter
Average
(Ug/m3)
Annual
Average
(Ug/m3)
South Phoenix, Arizona - SPAZ
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
Ethylbenzene
Tetrachloroethylene
Trichloroethylene
9/29
29/29
29/29
29/29
28/29
29/29
29/29
29/29
13/29
1.34
± 0.28
1.94
± 0.66
0.32
± 0.15
0.75
± 0.03
0.19
± 0.08
0.23
± 0.08
0.79
± 0.34
0.40
± 0.22
0.07
± 0.07
0.19
± 0.42
1.22
± 0.49
0.16
± 0.09
0.64
± 0.09
0.23
± 0.09
0.22
± 0.08
0.62
± 0.34
0.19
± 0.07
0.02
± 0.03
0.07
± 0.14
1.26
± 0.68
0.12
± 0.08
0.65
± 0.06
0.29
± 0.12
0.27
± 0.15
0.63
± 0.38
0.22
± 0.09
0.02
± 0.04
0
2.39
± 1.05
0.43
± 0.24
0.62
± 0.10
0.26
± 0.07
0.40
± 0.12
1.02
± 0.45
0.53
± 0.27
0.06
± 0.06
0.39
± 0.23
1.69
± 0.37
0.26
± 0.08
0.66
± 0.04
0.24
± 0.04
0.28
± 0.06
0.76
± 0.18
0.33
± 0.09
0.05
± 0.02
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of
viewing.
Observations for SPAZ from Table 5-5 include the following:
• Similar to PXSS, the pollutant with the highest annual average concentration by mass
is benzene. This is also the only pollutant with an annual average concentration
greater than 1 |ig/m3 (1.69 ± 0.37 |ig/m3). The fourth quarter average concentration of
benzene has a relatively high confidence interval compared to the other quarterly
averages. The two highest concentrations of benzene were measured on
December 10, 2010 (4.64 |ig/m3) and December 1, 2010 (3.65 |ig/m3). The
December 10, 2010 concentration was the fifth highest benzene measurement among
all NMP sites sampling benzene.
• Concentrations of benzene, 1,3-butadiene, ethylbenzene, and tetrachloroethylene
appear to be higher during the colder months of the year. Similar to concentrations of
these pollutants for PXSS, the difference in quarterly average concentrations is not
statistically significant.
• The first quarter average acrylonitrile concentration is significantly higher than the
other quarterly averages. Of the nine measured detections of this pollutant (out of 29
valid samples), seven were measured during the first quarter of 2010 (with one in the
second quarter and one in the third quarter). This explains the large confidence
intervals associated with these quarterly averages as well as the zero for the fourth
quarter average.
5-21
-------
• Each of the quarterly average concentrations of trichloroethylene has a confidence
interval equal to or greater than the concentration itself. This pollutant was detected in
fewer than half of the valid samples collected and analyzed. The average
concentrations of both acrylonitrile and trichloroethylene demonstrate the variability
introduced by substituting zeros for non-detects where a pollutant is detected
infrequently.
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for PXSS and
SPAZ from those tables include the following:
• PXSS and SPAZ appear in Tables 4-9 through 4-12 a total of 26 times.
• SPAZ has the highest annual average concentration of acrylonitrile and 1,3-butadiene
among all NMP sites sampling these pollutants. PXSS ranks fifth and second for
these pollutants, respectively.
• PXSS has the highest annual average concentration of tetrachloroethylene among all
NMP sites sampling this pollutant. SPAZ ranks third for this pollutant.
• SPAZ also has the third highest annual average concentrations of ethylbenzene and
/>-dichlorobenzene, while PXSS has the third highest annual average concentration of
chloroform.
• PXSS has the second highest annual average concentration of hexavalent chromium,
behind only CAMS 85. The annual averages of hexavalent chromium for these two
sites are an order of magnitude higher than the annual averages for the remaining
eight sites shown in Table 4-12.
• For the PMio metals, PXSS has the second highest annual average concentration of
beryllium, lead, and manganese and the third highest annual average concentration of
nickel.
5.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for benzene and 1,3-butadiene
were created for both PXSS and SPAZ. Box plots were also created for arsenic, benzo(a)pyrene,
hexavalent chromium, manganese, and naphthalene for PXSS. Figures 5-10 through 5-16 overlay
the sites' minimum, annual average, and maximum concentrations onto the program-level
minimum, first quartile, average, median, third quartile, and maximum concentrations, as
described in Section 3.5.3.
5-22
-------
Figure 5-10. Program vs. Site-Specific Average Arsenic (PMi0) Concentration
•
2 2.5 3
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
o
Figure 5-11. Program vs. Site-Specific Average Benzene Concentration
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
O
Figure 5-12. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
| Program Max Concentration = 42.7 ng/m3
1
0.8 1 1.2
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
5-23
-------
Figure 5-13. Program vs. Site-Specific Average 1,3-Butadiene Concentration
1
1 1 1 1 1
•!'
0 0.1 0.2 0.3
Program: IstQuartile
Site: Site Average
0
1 1 1 1 1
0.4 0.5 0.6 0.7 0.8 0.9 1
Concentration (ng/m3)
2ndQuartile SrdQuartile 4thQuartile Average
• D D
Site Minimum/Maximum
Figure 5-14. Program vs. Site-Specific Average Hexavalent Chromium Concentration
S-
~
D 0.15
Program: IstQuartile
Site: Site Average
O
! Program Max Concentration = 3.51 ng/m3 i
L
0.3 0.45 0.6
Concentration (ng/m3)
2ndQuartile SrdQuartile 4thQuartile Average
Site Minimum/Maximum
0.
Figure 5-15. Program vs. Site-Specific Average Manganese (PMio) Concentration
E
20 40 60 80 100 120 140 160
Concentration (ng/m3)
180 200
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
5-24
-------
Figure 5-16. Program vs. Site-Specific Average Naphthalene Concentration
1
1 1
600 800
Concentration (ng/m3)
Program
Site:
: IstQuartile
Site Average
0
2ndQuartile SrdQuartile 4thQuartile Avt
• n
Site Minimum/Maximum
;rage
Observations from Figures 5-10 through 5-16 include the following:
• Figure 5-10 shows that PXSS's annual average arsenic (PMio) concentration is
nearly identical to the program-level average for arsenic (PMio). There were no
non-detects of arsenic measured at PXSS.
• Figure 5-11 for benzene shows both sites, as both SPAZ and PXSS sampled
VOC. While neither Arizona site measured the maximum benzene concentration
across the program, both annual averages are greater than the program-level
average concentration. In addition, SPAZ's benzene concentrations are slightly
higher than PXSS's concentrations, as illustrated by the annual average and the
maximum concentrations measured. There were no non-detects of benzene
measured at either site (or among sites sampling VOC).
• Figure 5-12 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the program-level first quartile for this pollutant is zero
and is not visible on this box plot. This box plot shows that the annual average
concentration for PXSS is below the program-level average concentration.
Figure 5-12 also shows that the maximum concentration measured at PXSS is
well below the maximum concentration measured across the program. Several
non-detects of benzo(a)pyrene were measured at PXSS.
• Figure 5-13 for 1,3-butadiene also shows both sites. The annual average
concentrations for both sites are more than twice the program-level average
concentration. Further, these two sites have the highest annual average
concentrations of this pollutant across the program, as mentioned in Section 5.4.1
and shown in Table 4-9. SPAZ's 1,3-butadiene annual average concentration is
slightly higher than PXSS's annual average concentration. Note that the
maximum concentration measured across the program was measured at SPAZ
(0.907 |ig/m3), although PXSS's maximum concentration was not much lower
(0.896 |ig/m3) and was the second highest 1,3-butadiene concentration (for
5-25
-------
Method TO-15) across the program. There were no non-detects of 1,3-butadiene
measured at SPAZ, but there were two measured at PXSS.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 5-14 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 5-14 shows the annual average concentration of hexavalent chromium for
PXSS is greater than the program-level average (more than three times higher).
While the maximum concentration measured at PXSS is well below the program
maximum concentration, PXSS has the second highest annual average
concentration among NMP sites sampling hexavalent chromium (behind only
CAMS 85), as discussed in Section 5.4.1 and shown in Table 4-12. There were no
non-detects of hexavalent chromium measured at PXSS.
• Figure 5-15 shows the annual average concentration of manganese (PMio) for
PXSS is greater than the program-level average (nearly twice as high). While the
maximum concentration measured at PXSS is well below the program maximum
concentration, PXSS has the second highest annual average concentration among
the NMP sites sampling manganese (PMio), behind only S4MO, as shown in
Table 4-12. There were no non-detects of manganese measured at PXSS.
• Figure 5-16 shows that the annual naphthalene average for PXSS is very similar
to the program-level average concentration. The maximum naphthalene
concentration measured at PXSS is well below the program-level maximum
concentration. There were no non-detects of naphthalene measured at PXSS.
• Recall that annual averages could not be calculated for formaldehyde and
acetaldehyde, as discussed in Section 5.4.1.
5.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. PXSS has sampled PMio metals and hexavalent chromium for 5 years as part of
the NMP; thus, Figures 5-17 through 5-19 present the 3-year rolling statistical metrics for
arsenic, hexavalent chromium, and manganese, respectively. SPAZ has not sampled
continuously for 5 years as part of the NMP; therefore, the trends analysis was not conducted for
this site. The statistical metrics presented for assessing trends include the substitution of zeros for
non-detects.
5-26
-------
Figure 5-17. Three-Year Rolling Statistical Metrics for Arsenic (PMi0) Concentrations
Measured at PXSS
ioo i 2009
Three-Year Period
- Median - Maximum • 'i',s|i ['n, ,.•„tj|,.- * -. ..M^ •
Figure 5-18. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at PXSS
200/ 200H
Three-Year Period
• 5thF-er«nti1e - Minimum - Median - Maximum • ^5HiPcnronttk -• -+•- Average
5-27
-------
Figure 5-19. Three-Year Rolling Statistical Metrics for Manganese (PMi0) Concentrations
Measured at PXSS
11)1}
so
Ion (H|/rn3]
s s
!
•3 10
1U
• '
^^
^
-
i
20 Z008 200 / 200"J ^OOS ^01
Three-Year Period
• ltd Perreiitile — Miniiiiuin — Median — Mitxiniiiin * q^rli Perfeiilil?
••+•• Average
Observations from Figure 5-17 for arsenic measurements at PXSS include the following:
• PXSS began sampling arsenic under the NMP in January 2006.
• The maximum arsenic concentration was measured on December 26, 2007 and is
more than twice the next highest concentration, measured in December 2006. The
maximum concentration for each year was measured in December or January, with
the exception of 2008, which was measured in September.
• The average rolling concentrations show little change over the years of sampling,
which is also true of most for the other statistical parameters, with the exception of
the maximum concentration. The maximum concentration for all years of sampling
ranged from 2 to 3 ng/m3, with the exception of 2007.
Observations from Figure 5-18 for hexavalent chromium measurements at PXSS include
the following:
• PXSS began sampling hexavalent chromium in January 2006.
• The maximum hexavalent chromium concentration shown was measured on
July 10, 2006. The maximum concentrations for subsequent years were nearly half
that measurement or less.
• The average rolling concentrations exhibit a slight decrease from 2006-2008 to
2007-2009 and then return to initial levels for 2008-2010.
5-28
-------
• The 95th percentile increased for the third 3-year period, indicating an increase in the
range of concentrations measured.
Observations from Figure 5-19 for manganese measurements at PXSS include the
following:
• The two highest manganese concentrations were measured in July and August of
2009.
• The rolling average, median, and 95th percentile decreased slightly for the second and
third 3-year periods shown, even with the highest manganese concentrations
measured in 2009.
5.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at each
Arizona monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
5.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Arizona monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where available.
As described in Section 3.3, acute risk results from exposures of 1 to 14 days; intermediate risk
results from exposures of 15 to 364 days; and chronic risk results from exposures of 1 year or
greater. The preprocessed daily measurements of the pollutants of interest for each site were
compared to the acute MRL; the quarterly averages were compared to the intermediate MRL;
and the annual averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the Arizona monitoring sites were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
the Arizona monitoring sites.
5.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Arizona monitoring sites and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
5-29
-------
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 5-6, where applicable.
Table 5-6. Cancer and Noncancer Surrogate Risk Approximations for the Arizona
Monitoring Sites
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
# of Measured
Detections vs.
# of Samples
Annual
Average
(jig/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Phoenix, Arizona - PXSS
Acetaldehyde
Acrylonitrile
Arsenic (PM10)a
Benzene
Benzo(a)pyrene a
Beiy Ilium (PM10)a
1,3 -Butadiene
Cadmium (PM10) a
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
Ethylbenzene
Formaldehyde
Hexavalent Chromium3
Lead(PM10)a
Manganese (PM10) a
Naphthalene a
Nickel (PM10) a
0.0000022
0.000068
0.0043
0.0000078
0.00176
0.0024
0.00003
0.0018
0.000006
0.000011
0.0000025
0.000013
0.012
0.000034
0.00048
0.009
0.002
0.000015
0.03
0.00002
0.002
0.00001
0.1
0.098
0.8
1
0.0098
0.0001
0.00015
0.00005
0.003
0.00009
16/20
13/61
59/59
61/61
21/59
42/59
59/61
59/59
61/61
61/61
52/61
61/61
16/20
57/57
59/59
59/59
59/59
59/59
NA
0.05
±0.03
O.01
±<0.01
1.38
±0.20
O.01
±<0.01
0.01
±0.01
0.21
±0.05
0.01
±0.01
0.66
±0.03
0.37
±0.05
0.16
±0.03
0.60
±0.10
NA
0.01
±0.01
O.01
±0.01
0.01
±0.01
0.09
±0.01
0.01
±0.01
NA
3.22
2.40
10.76
0.12
0.01
6.22
0.21
3.97
1.72
1.50
NA
1.57
3.03
0.59
NA
0.02
0.04
0.05
0.01
0.10
0.01
O.01
0.01
O.01
0.01
NA
0.01
0.02
0.25
0.03
0.01
— = a Cancer URE or Noncancer RfC is not available.
NA = Not available due to the criteria for calculating an annual average.
a For the annual average concentration of this pollutant in ng/m3, refer to Table
5-5.
5-30
-------
Table 5-6. Cancer and Noncancer Surrogate Risk Approximations for the Arizona
Monitoring Sites (Continued)
Pollutant
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Cancer
URE
(Hg/m3)1
0.00000026
0.0000048
0.0000088
Noncancer
RfC
(mg/m3)
0.04
0.002
0.1
# of Measured
Detections vs.
# of Samples
57/61
21/61
1/61
Annual
Average
(jig/m3)
0.40
±0.08
0.03
±0.03
0.01
±0.01
Cancer Risk
Approximation
(in-a-million)
0.10
0.16
0.01
Noncancer Risk
Approximation
(HQ)
0.01
0.02
0.01
South Phoenix, Arizona - SPAZ
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
Ethylbenzene
Tetrachloroethylene
Trichloroethylene
0.000068
0.0000078
0.00003
0.000006
0.000011
0.0000025
0.00000026
0.0000048
0.002
0.03
0.002
0.1
0.098
0.8
1
0.04
0.002
9/29
29/29
29/29
29/29
28/29
29/29
29/29
29/29
13/29
0.39
±0.23
1.69
±0.37
0.26
±0.08
0.66
±0.04
0.24
±0.04
0.28
±0.06
0.76
±0.18
0.33
±0.09
0.05
±0.02
26.40
13.15
7.67
3.98
3.07
1.90
0.09
0.22
0.19
0.06
0.13
0.01
O.01
0.01
0.01
0.01
0.02
— = a Cancer URE or Noncancer RfC is not available.
NA = Not available due to the criteria for calculating an annual average.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 5-5.
Observations for PXSS from Table 5-6 include the following:
• The pollutants with the highest annual average concentrations by mass are benzene,
carbon tetrachloride, and ethylbenzene.
• Based on the annual averages and cancer UREs, benzene, 1,3-butadiene, and carbon
tetrachloride have the three highest cancer risk approximations. An additional six
pollutants have cancer risk approximations greater than 1.0 in-a-million.
• None of PXSS's pollutants of interest have noncancer risk approximations greater
than 1.0. The pollutant with the highest noncancer risk approximation is manganese
(0.25).
• Annual averages (and therefore cancer and noncancer surrogate risk approximations)
could not be calculated for acetaldehyde and formaldehyde, as discussed in Section
5.4.1.
5-31
-------
Observations for SPAZ from Table 5-6 include the following:
• The pollutants with the highest annual average concentrations by mass are benzene,
ethylbenzene, and carbon tetrachloride.
• Based on the annual averages and cancer UREs, acrylonitrile, benzene, and
1,3-butadiene have the three highest cancer risk approximations. An additional three
pollutants have cancer risk approximations greater than 1.0 in-a-million.
• SPAZ's annual acrylonitrile average concentration is two orders of magnitude higher
than PXSS's annual average for this pollutant. Thus, SPAZ's cancer risk
approximation for acrylonitrile is more than eight times higher than PXSS's cancer
risk approximation.
• None of SPAZ's pollutants of interest have noncancer risk approximations greater
than 1.0. The pollutant with the highest noncancer risk approximation is acrylonitrile
(0.19).
5.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 5-7 and 5-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 5-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 5-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Table 5-7 and 5-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. Further, the cancer and noncancer surrogate risk approximations based on each site's
annual averages are limited to those pollutants for which each respective site sampled. As
discussed in Section 5.3, PXSS sampled for VOC, carbonyl compounds, PAH, metals (PMio),
and hexavalent chromium; SPAZ sampled for VOC only. In addition, the cancer and noncancer
surrogate risk approximations are limited to those pollutants with enough data to meet the criteria
for annual averages to be calculated. A more in-depth discussion of this analysis is provided in
Section 3.5.5.3.
5-32
-------
Table 5-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer
UREs for the Arizona Monitoring Sites
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Phoenix, Arizona (Maricopa County) - PXSS
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
POM, Group la
Propylene oxide
1,256.37
969.28
766.10
487.50
179.20
90.54
28.45
14.68
2.61
1.66
Formaldehyde
Benzene
1,3 -Butadiene
Naphthalene
Ethylbenzene
Hexavalent Chromium, PM
POM, Group 2b
Acetaldehyde
Arsenic, PM
POM, Group 5a
1.26E-02
9.80E-03
5.38E-03
3.08E-03
1.92E-03
1.70E-03
1.29E-03
1.07E-03
7.14E-04
3.83E-04
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Acrylonitrile
Naphthalene
Arsenic
£>-Dichlorobenzene
Hexavalent Chromium
Ethylbenzene
Nickel
10.76
6.22
3.97
3.22
3.03
2.40
1.72
1.57
1.50
0.59
South Phoenix, Arizona (Maricopa County) - SPAZ
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
POM, Group la
Propylene oxide
1,256.37
969.28
766.10
487.50
179.20
90.54
28.45
14.68
2.61
1.66
Formaldehyde
Benzene
1,3 -Butadiene
Naphthalene
Ethylbenzene
Hexavalent Chromium, PM
POM, Group 2b
Acetaldehyde
Arsenic, PM
POM, Group 5a
1.26E-02
9.80E-03
5.38E-03
3.08E-03
1.92E-03
1.70E-03
1.29E-03
1.07E-03
7.14E-04
3.83E-04
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
£>-Dichlorobenzene
Ethylbenzene
Trichloroethylene
Tetrachloroethylene
26.40
13.15
7.67
3.98
3.07
1.90
0.22
0.09
-------
Table 5-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with
Noncancer RfCs for the Arizona Monitoring Sites
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Phoenix, Arizona (Maricopa County) - PXSS
Toluene
Xylenes
Benzene
Formaldehyde
Hexane
Ethylbenzene
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
Glycol ethers, gas
3,931.16
2,893.73
1,256.37
969.28
963.82
766.10
487.50
240.60
179.20
98.40
Acrolein
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Lead, PM
Benzene
Naphthalene
Xylenes
Arsenic, PM
Propionaldehyde
2,430,403.47
98,905.78
89,599.85
54,166.44
42,964.38
41,879.01
30,178.52
28,937.26
11,077.09
8,611.73
Manganese
1,3 -Butadiene
Benzene
Arsenic
Naphthalene
Acrylonitrile
Lead
Trichloroethylene
Nickel
Cadmium
0.25
0.10
0.05
0.04
0.03
0.02
0.02
0.02
0.01
0.01
South Phoenix, Arizona (Maricopa County) - SPAZ
Toluene
Xylenes
Benzene
Formaldehyde
Hexane
Ethylbenzene
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
Glycol ethers, gas
3,931.16
2,893.73
1,256.37
969.28
963.82
766.10
487.50
240.60
179.20
98.40
Acrolein
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Lead, PM
Benzene
Naphthalene
Xylenes
Arsenic, PM
Propionaldehyde
2,430,403.47
98,905.78
89,599.85
54,166.44
42,964.38
41,879.01
30,178.52
28,937.26
11,077.09
8,611.73
Acrylonitrile
1,3 -Butadiene
Benzene
Trichloroethylene
Tetrachloroethylene
Carbon Tetrachloride
Chloroform
Ethylbenzene
£>-Dichlorobenzene
0.19
0.13
0.06
0.02
0.01
0.01
0.01
<0.01
<0.01
-------
Observations from Table 5-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in Maricopa County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
cancer UREs) are formaldehyde, benzene, and 1,3-butadiene.
• Seven of the highest emitted pollutants in Maricopa County also have the highest
toxicity-weighted emissions.
• Benzene, 1,3-butadiene, and carbon tetrachloride have highest cancer surrogate risk
approximations for PXSS. While benzene and 1,3-butadiene both appear on the list of
10 highest emissions and 10 highest toxicity-weighted emissions for Maricopa
County, carbon tetrachloride does not appear on either list.
• POM, Group 2b is the eighth highest emitted "pollutant" in Maricopa County and
ranks sixth for toxicity-weighted emissions. POM, Group 2b includes several PAH
sampled for at PXSS including acenaphthylene, benzo(e)pyrene, fluoranthene, and
perylene. None of the PAH included in POM, Group 2b were identified as pollutants
of interest for PXSS.
• While acrylonitrile's cancer risk approximation is the highest cancer risk
approximation for SPAZ, this pollutant appears on neither emissions-based list.
Observations from Table 5-8 include the following:
• Toluene, xylenes, and benzene are the highest emitted pollutants with noncancer
RfCs in Maricopa County.
• The pollutants with the highest toxicity-weighted emissions (of the pollutants with
noncancer RfCs) are acrolein, formaldehyde, and 1,3-butadiene.
• Five of the highest emitted pollutants also have the highest toxicity-weighted
emissions.
• Acrolein has the highest toxicity-weighted emissions (by two orders of magnitude)
for Maricopa County. Although acrolein was sampled for at both sites, this pollutant
was excluded from the pollutants of interest designation, and thus subsequent risk
screening evaluations, due to questions about the consistency and reliability of the
measurements, as discussed in Section 3.2.
• Manganese, 1,3-butadiene, and benzene have the highest noncancer risk
approximations for PXSS, although all of them are well below an HQ of 1.0. Only
benzene and 1,3-butadiene appear on all three lists. In addition to benzene and
1,3-butadiene, arsenic, lead, and naphthalene appear on both toxicity-based lists.
5-35
-------
• Acrylonitrile, 1,3-butadiene, and benzene have the highest noncancer risk
approximations for SPAZ, although all of them are well below an HQ of 1.0. Benzene
and 1,3-butadiene appear on all three lists while acrylonitrile appears on neither
emissions-based list.
5.6 Summary of the 2010 Monitoring Data for PXSS and SPAZ
Results from several of the data treatments described in this section include the
following:
»«» Twenty-two pollutants failed screens for PXSS; 14 of these are NATTSMQO Core
Analytes. Nine pollutants failed screens for SPAZ, of which four are NATTSMQO
Core Analytes.
*»* Of the site-specific pollutants of interest for the Arizona sites, benzene had the highest
annual average concentration for both sites. This was the only pollutant with an
annual average greater than 1 fj,g/m3 for either site.
»«» Concentrations of several VOC, including benzene and 1,3-butadiene, tended to be
slightly higher during the colder months of the year.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest, where they could be calculated,
were greater than their associatedMRL noncancer health risk benchmarks.
5-36
-------
6.0 Sites in California
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at three NATTS sites in California, and integrates these concentrations
with emissions, meteorological, and risk information. Data generated by sources other than ERG
are not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
6.1 Site Characterization
This section characterizes the California monitoring sites by providing geographical and
physical information about the locations of the sites and the surrounding areas. This information
is provided to give the reader insight regarding factors that may influence the air quality near the
sites and assist in the interpretation of the ambient monitoring measurements.
The California monitoring sites are located in Los Angeles, Rubidoux, and San Jose.
Figures 6-1 through 6-3 are composite satellite images retrieved from ArcGIS Explorer showing
the monitoring sites in their urban locations. Figures 6-4 through 6-6 identify point source
emissions locations by source category, as reported in the 2008 NEI for point sources. Note that
only sources within 10 miles of the sites are included in the facility counts provided in
Figures 6-4 through 6-6. Thus, sources outside the 10-mile radius have been grayed out, but are
visible on the maps to show emissions sources outside the 10-mile boundary. A 10-mile
boundary was chosen to give the reader an indication of which emissions sources and emissions
source categories could potentially have a direct effect on the air quality at the monitoring sites.
Further, this boundary provides both the proximity of emissions sources to the monitoring sites
as well as the quantity of such sources within a given distance of the sites. Table 6-1 describes
the area surrounding each monitoring site by providing supplemental geographical information
such as land, location setting, and locational coordinates.
6-1
-------
to
Figure 6-1. Los Angeles, California (CELA) Monitoring Site
__^ " V, .'- » i ' ?-!_.•«-
-------
Figure 6-2. Rubidoux, California (RUCA) Monitoring Site
-------
Figure 6-3. San Jose, California (SJJCA) Monitoring Site
-------
Figure 6-4. NEI Point Sources Located Within 10 Miles of CELA
Legend
H8"15-0-W 118*1 (WW 11SWW 118WW
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
CELA NATTS site
10 mile radius
County boundary
Source Category Group (No. of Facilities)
ljl Aerospace/Afrcraft Manufacturing (2)
Q Air-conditraning/Refngeration (1)
•f* Aircraft Operations (62)
I Asphalt Processing/Roofing Manufacturing (2)
0 Auto Body Shop/Painlers (12)
A AuEdinobiteiTruek Manufacturing (4)
$ Bakery (4)
•X Battery Manufacturing (2)
fl Building Construction (4)
B Bulk Terminal&fBulk Plants (2)
C Chemical Manufacturing (8)
• Concrete Batch Plant (1)
iXjCrematory -Animal/Human (2)
6 Electrical Equipment (2)
f Electricity Generalion via Combustion (5)
E Electrop!atir>g. Plating. Polishing, Anodizing, & Colonng (17)
® Fabricated Metal Products (11)
,^- Flexible Potyurethane Foam ProdncUon {2)
F Food Processing/Agriculture (13)
L ! Furniture Plant (20)
IT Glass Manufacturing (2)
A Grain Handling (2)
-------
Figure 6-5. NEI Point Sources Located Within 10 Miles of RUCA
117'30'Q-W 117'25'irW 117:20'O^W IWIS'fTW
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
Legend
@ RUCA NATTS site 10 mile radius Q ^] County boundary
Source Category Group (No. of Facilities)
lj* Aerospace/Aircraft Manufacturing (1)
41 Aircraft Operations (12)
I Asphalt ProcessingyRoofing Manufacturing (3)
0 Auto Body Shop/Painters (6)
£4 Automobile/Truck Manufacturing (7)
fc Ba1cery<3)
i Boat Manufacturing (1)
Brick Manufacturing & Structural Clay {2}
A Building Construction (3)
B Bulk Terminals/Bulk Plants (2)
C Chemical M anufactunng {1)
• Concrete Batch Plant (5)
6 Electrical Equipment (5)
f Electricity Generation via Combustion (4)
© Fabncaled Metal Products (10)
Flexible Polyurethane Foam Production (1}
Food Process ing.1 Agriculture {18)
Furniture Plant (1)
Gasoline/Diesel Service Station {1}
Hospital (1)
HotMix Asphalt Plant <1)
Industrial Machinery and Equipment (2)
Institutional - prison (1)
Institutional • school (5)
Landfill (4)
Lumber/sawmill (1)
Mine/Quarry (3)
Mineral Products (2)
Miscellaneous Commercial/lndusHial (10)
Miscellaneous Manufacturing (2)
* Oil and/or Gas Production (3)
• Other Solid Waste Incineration (1)
7 Portland Cement Manufacturing (2)
1 Primary Metal Production (4)
P Printing/Publishing (10)
B Pulp and Paper Ptant/Wsod Products (4)
R Rubber and Miscellaneous Plastics Products (6)
2 Secondary Metal Processing (2)
< Site RerrwdtationActivityO)
> Solid Vteste Disposal - Commercial/!nstitubonaI (1)
V Steel Mill (4)
S Surface Coaling (4)
«* Transportation Equipment (3)
•#• Transportation and Marketing of Pelrofeum Products (1)
i Wasiewater Treatment (9)
W Woodwork, Furniture. Miltwork & Wood Preserving (2)
6-6
-------
Figure 6-6. NEI Point Sources Located Within 10 Miles of SJJCA
Legend
1ZV55'CrW tZrSO'FW 121*4510"W 121 4QXJ-W
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
SJJCA NATTS site Q 10 mile radius
County boundary
Source Category Group (No. of Facilities)
:\h An (2)
HH Aircraft Operations (19)
I Asphalt Processing/Roofing Manufacturing (4)
0 Auto Body Shop/Painters (235)
H Automobile/Truck Manufacturing (17)
fe BaKery(2)
X Battery ManufBdunng (1)
Brick Manufacturing & Structural Clay (3)
ft Budding ConsUuclion (15)
5 Bulk Terminals/Bulk Plants (5)
C Chemical Manufactunr^<5)
O Clay Ceramics Manufacturing (1)
• Concrete Batch Plant (10)
XCrematory - Aiwnal/Human (3)
® Dry Cleaning Facdtty (161)
& Electncal Equipment (250)
f Electricity Generation via Combuston (2)
E Etectroplaling, Plalmg, Pobshmg. ArwdJang, SCotonng (14)
4 Ermine Testing (1)
© Fabricated Metal Producls (35)
^ Flexible Potyurethane Foam Produclidn (2)
F Food Process ing'AgricurtuTe (43)
I , Furniture Plant (37)
V Gasoline/Diesel Service Station (9)
^t Glass Manufacturing (1)
A Grain Handling (2)
// Healing Equipment Manufacturing (7)
[3 Hospital (16)
£ Hoi Mix Asphalt Plant (1)
^fr Industrial Machinery and E(^iiprnenl (19)
l|r Institutional - school (27)
I Iron and Steel Foundry (1)
^ Laboratory (7)
W Leather and Leather Products {1)
/ Lumber/sawmill (1)
A Mflftary Base/Nalional Secitnty Facility (7)
X Mine/Quarry (5)
4i Mineral Wool Manufacturing (1)
? Miscellaneous Commercial/Industrial (212)
M Miscellaneous Manufacturing (69)
• 0«1 and/or C3as Production (3)
••—I Pharmaceutical Manufacturing (9)
1 Primary Melal Production (2)
P Printing/Publishing (30)
B Pulp and Paper Plant/Wood Products (W)
R Rubber and Miscellaneous Plastics Products (3)
2 Secondary Melal Processing (3)
< Erie Remedialion Activity (13)
> Solid Waste Disposal - Commercial,'mstitut«>na] (26)
V Steel Mill (1}
S Surface Coating (32)
IT Tetecommunica1.K)ns{102)
T Textile Mill (5)
«*• Transportation Equipment (5)
•ifc Transportation and Marketing of Petroleum Products (3)
1 Wfcstewater Trea tmenl (IS)
W Woodwork. Furrature. Miflwork & Wood Preserving (2)
6-7
-------
Table 6-1. Geographical Information for the California Monitoring Sites
Site
Code
CELA
RUCA
SJJCA
AQS Code
06-037-1103
06-065-8001
06-085-0005
Location
Los
Angeles
Rubidoux
San Jose
County
Los
Angeles
Riverside
Santa
Clara
Micro- or
Metropolitan
Statistical Area
Los Angeles-Long
Beach-Santa Ana,
CAMSA
Riverside-San
Bernardino-
Ontario, CA MSA
San Jose-
Sunnyvale-Santa
Clara, CAMSA
Latitude
and
Longitude
34.06659,
-118.22688
33.99958,
-117.41601
37.3485,
-121.895
Land Use
Residential
Residential
Commercial
Location
Setting
Urban/City
Center
Suburban
Urban/City
Center
Additional Ambient Monitoring Information1
TSP, TSP Speciation, Hexavalent chromium, CO,
SO2, NO, NO2, NOX, NOy, PAMS, Carbonyl
compounds, VOC, O3, Meteorological parameters,
PM10, PM10 Speciation, PM2 5, PM25 Speciation.
Haze, TSP Speciation, Hexavalent chromium, CO,
SO2, NO, NO2, NOX, PAMS, VOC, Carbonyl
compounds, O3, Meteorological parameters, PM10,
PM10 Speciation, PM coarse, PM25,
PM25 Speciation.
TSP Speciation, Hexavalent chromium, CO, SO2,
NO, NO2, NOX, VOC, Carbonyl compounds, O3,
NMOC, Meteorological parameters, PM10, PM10
Speciation, Black carbon, PM2 5, PM2 5 Speciation.
These monitoring sites report additional pollutants to AQS (EPA, 2012h); however, these data are not generated by ERG and are therefore not included in this report.
BOLD ITALICS = EPA-designated NATTS Site.
oo
-------
CELA is located on the rooftop of a two-story building just northeast of downtown Los
Angeles, near Dodgers' Stadium. Figure 6-1 shows that CELA is surrounded by major freeways,
including 1-5 and Route 110. Highway 101 is located farther south. Although the area is
classified as residential, a freight yard is located to the south of the site. The Los Angeles River
runs north-south just east of the site. This monitoring site was originally set up as an emergency
response monitor. As Figure 6-4 shows, CELA is situated among numerous point sources. There
is a cluster of emissions sources located just to the southwest of CELA. A large number of
emissions sources within 10 miles of CELA are involved in aircraft operations, which include
airports as well as small runways, heliports, or landing pads; furniture products; electroplating,
plating, polishing, anodizing, and coloring; and printing or publishing.
RUCA is located just outside of Riverside, in a residential area of the suburban town of
Rubidoux. Highway 60 runs east-west to the north of the site. Flabob Airport is located about
three-quarters of a mile to the southeast of the site. Figure 6-2 shows that RUCA is adjacent to a
power substation near the intersection of Mission Boulevard and Riverview Drive. RUCA and
CELA are located less than 45 miles apart. Figure 6-5 shows that fewer emissions sources
surround RUCA than CELA. Most of the emissions sources are located to the northeast and
northwest of the site. The point source located closest to RUCA is Flabob Airport. The emissions
source categories with the highest number of sources near RUCA include food processing,
aircraft operations, printing and publishing, and fabricated metals products.
SJJCA is located in central San Jose. Figure 6-3 shows that SJJCA is located in a
commercial area surrounded by residential areas. A railroad is shown just east of the monitoring
site, running north-south in Figure 6-3. Guadalupe Parkway, which can be seen on the bottom
left of Figure 6-3, intersects with 1-880 approximately 1 mile northwest of the monitoring site.
San Jose International Airport is just on the other side of this intersection. Figure 6-6 shows that
the density of point sources is significantly higher near SJJCA than CELA and RUCA. The
emissions source categories with the highest number of sources are electrical equipment; auto
body/paint shops; and telecommunications.
Table 6-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the areas surrounding the
California monitoring sites. Table 6-2 also includes a vehicle registration-to-county population
6-9
-------
ratio (vehicles-per-person) for each site. In addition, the population within 10 miles of each site
is presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-
level vehicle registration-to-population ratio to the 10-mile population surrounding each
monitoring site. Table 6-2 also contains annual average daily traffic information. Finally,
Table 6-2 presents the daily VMT for each county.
Table 6-2. Population, Motor Vehicle, and Traffic Information for the California
Monitoring Sites
Site
CELA
RUCA
SJJCA
Estimated
County
Population1
9,830,420
2,203,332
1,787,694
County-level
Vehicle
Registration2
7,410,625
1,707,950
1,517,995
Vehicles per
Person
(Registration:
Population)
0.75
0.78
0.85
Population
within 10
miles3
3,679,965
990,029
1,486,476
Estimated
10-mile
Vehicle
Ownership
2,774,128
767,438
1,262,220
Annual
Average
Daily
Traffic4
235,000
145,000
103,000
County-
level Daily
VMT5
211,876,660
55,167,650
39,402,370
1 County-level population estimates reflect data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the California DMV (CA DMV, 2010)
3 10-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2010 data from the California DOT (CA DOT, 2010)
5 County-level VMT reflects 2010 data for all public roads from the California DOT (CA DOT, 2011)
BOLD ITALICS = EPA-designaled NATTS Site.
Observations from Table 6-2 include the following:
• Los Angeles County (CELA) has the highest county population and county-level
vehicle registration compared to all counties with NMP sites. CELA also had the
highest 10-mile estimated vehicle ownership. However, the 10-mile population near
this site ranks third behind BXNY and MONY, which are located in Bronx County
and part of New York City.
• Riverside and Santa Clara Counties are also in the top 10 for county population and
county-level vehicle registration among counties with NMP sites.
• Among the California sites, the vehicle-per-person ratio is lowest for the most
populous area (CELA) and highest for the least populated area (RUCA), based on
county population. In general, this trend is also true among all NMP sites.
• CELA experiences the second highest annual average daily traffic among NMP sites,
and has a substantially higher traffic volume than both RUCA and SJJCA. The traffic
count for CELA is based on data from Exit 136 off 1-5 at Main Street. The traffic
count for RUCA is based on data from Mission Boulevard at Rubidoux Boulevard.
The traffic count for SJJCA is based on the intersection of Guadalupe Parkway at
West Taylor Street.
6-10
-------
• The Los Angeles County's daily VMT was the highest among all counties with NMP
sites, where VMT was available. This VMT was an order of magnitude higher than
the next highest county-level VMT (Cook County, IL). Riverside and Santa Clara
Counties were also in the top 10 for VMT among counties with NMP sites (where
VMT data were available).
6.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in California on sample days, as well as over the course of the year.
6.2.1 Climate Summary
The climate of Los Angeles is generally mild. While the proximity to the Pacific Ocean
acts as a moderating influence on the Los Angeles area, the elevation changes between the
mountains and valleys allow the distance from the ocean to create substantial differences in
temperature, rainfall, and wind over a relatively short distance. Precipitation falls primarily in
winter months, while summers tend to be dry. Stagnant wind conditions in the summer can result
in air pollution episodes, while breezy Santa Ana winds can create hot, dusty conditions. Fog and
cloudy conditions are more prevalent near the coast than farther inland (Bair, 1992 and WRCC,
2012).
San Jose is located to the southeast of San Francisco, near the base of the San Francisco
Bay. The city is situated in the Santa Clara Valley, between the Santa Cruz Mountains to the
south and west and the Diablo Range to the east. San Jose experiences a Mediterranean climate,
with distinct wet-dry seasons. The period from November through March represents the wet
season, with cool but mild conditions prevailing. Little rainfall occurs the rest of the year and
conditions tend to be warm and sunny. San Jose is not outside the marine influences of the cold
ocean currents typically affecting the San Francisco area (Bair, 1992 and NOAA, 1999).
6.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather stations nearest these sites were
retrieved for 2010 (NCDC, 2010). The weather station nearest CELA is located at Downtown
Los Angeles/USC Campus; the nearest NWS weather station to RUCA is located at Riverside
Municipal Airport; and the nearest NWS station to SJJCA is located at San Jose International
(WBAN 93134, 03171 and 23293, respectively). Additional information about these weather
6-11
-------
stations, such as the distance between the sites and the weather stations, is provided in Table 6-3.
These data were used to determine how meteorological conditions on sample days vary from
normal conditions throughout the year.
Table 6-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 6-3 is the
95 percent confidence interval for each parameter. As shown in Table 6-3, average
meteorological conditions on sample days near these sites were representative of average
weather conditions throughout the year. Table 6-3 also shows a marked wind speed difference
between CELA and RUCA (which are located less than 50 miles apart), as alluded to in
Section 6.2.1, although wind speeds for both sites are very light. A statistically significant
difference is also shown for the average maximum temperature. As expected, conditions near
SJJCA tended to be cooler.
6-12
-------
Table 6-3. Average Meteorological Conditions near the California Monitoring Sites
Closest NWS Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar
Wind
Speed
(kt)
Los Angeles, California - CELA
Downtown L.A./USC
Campus Airport
93134
(34.03, -118.30)
4.57
miles
248°
(WSW)
Sample
Day
2010
73.6
±2.3
73.5
±0.9
64.3
±1.8
64.4
±0.7
49.5
± 1.9
49.1
±0.9
56.3
±1.4
56.2
±0.6
62.2
±3.2
61.4
±1.4
1014.3
±0.9
1014.3
±0.4
1.3
±0.2
1.3
±0.1
Rubidoux, California - RUCA
Riverside Municipal
Airport
03171
(33.95, -117.44)
3.49
miles
214°
(SW)
Sample
Day
2010
77.7
±3.2
77.5
±1.3
64.6
±2.3
64.4
±1.0
46.2
±2.3
45.9
±1.0
54.7
±1.7
54.6
±0.7
57.3
±3.9
57.5
±1.8
1013.2
±0.9
1013.2
±0.4
3.7
±0.4
3.8
±0.1
San Jose, California - SJJCA
San Jose Intl. Airport
23293
(37.36, -121.93)
1.90
miles
316°
(NW)
Sample
Day
2010
69.2
±2.6
69.2
±1.1
59.4
±1.9
59.1
±0.8
47.4
±1.2
47.1
±0.6
53.0
±1.3
52.8
±0.6
67.6
±2.8
67.6
±1.1
1015.8
±1.3
1015.7
±0.5
5.1
±0.5
5.3
±0.2
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
6.2.3 Back Trajectory Analysis
Figure 6-7 is the composite back trajectory map for days on which samples were
collected at the CELA monitoring site in 2010. Included in Figure 6-7 are four back trajectories
per sample day. Figure 6-8 is the corresponding cluster analysis for 2010. Similarly, Figure 6-9 is
the composite back trajectory map for days on which samples were collected at RUCA and
Figure 6-10 is the corresponding cluster analysis; Figure 6-11 is the composite back trajectory
map for days on which samples were collected at SJJCA and Figure 6-12 is the corresponding
cluster analysis. An in-depth description of these maps and how they were generated is presented
in Section 3.5.2.1. For the composite maps, each line represents the 24-hour trajectory along
which a parcel of air traveled toward the monitoring site on a given sample day and time. For the
cluster analyses, each line corresponds to a back trajectory representative of a given cluster of
trajectories. For all maps, each concentric circle around the sites in Figures 6-7 through 6-12
represents 100 miles.
Observations from Figures 6-7 and 6-8 for CELA include the following:
• The 24-hour air shed domain was somewhat smaller for CELA than for many other
NMP monitoring sites, based on the average distance of the trajectories. The farthest
away a trajectory originated was off the northwest coast of California, or less than
650 miles away. However, most trajectories (88 percent) originated within 300 miles
of CELA.
• Back trajectories originated from a variety of directions at CELA. However, a large
cluster of trajectories originated from the northwest. Another cluster originated from
the east-northeast. Very few trajectories originated from the east, southeast, south, or
southwest.
• The cluster analysis shows that over 50 percent of trajectories originated from the
northwest, although of varying distances. The cluster analysis also shows that
approximately 25 percent of trajectories originated from a direction within the
northeast quadrant. Another 22 percent originated off the coast and within about 300
miles of CELA. The cluster marked with 2 percent represents the five back
trajectories originating well to the south and off Baja California.
6-14
-------
Figure 6-7. 2010 Composite Back Trajectory Map for CELA
Figure 6-8. Back Trajectory Cluster Map for CELA
>
I /
/ '' ' /
/ /
6-15
-------
Figure 6-9. 2010 Composite Back Trajectory Map for RUCA
Figure 6-10. Back Trajectory Cluster Map for RUCA
6-16
-------
Figure 6-11. 2010 Composite Back Trajectory Map for SJJCA
Figure 6-12. Back Trajectory Cluster Map for SJJCA
6-17
-------
Observations from Figures 6-9 and 6-10 for RUCA include the following:
• Not surprisingly, the back trajectories for RUCA resemble the ones for CELA. The
24-hour air shed domain for RUCA is similar in size to CELA, as the farthest away a
trajectory originated was also off the northwest coast of California, or nearly 600
miles away. Like CELA, most trajectories (90 percent) originated within 300 miles of
RUCA.
• Back trajectories originated from a variety of directions at RUCA. A large cluster of
trajectories originated from the northwest of the site and a secondary cluster
originated from the northeast. Few trajectories originated from the east, southeast, or
south.
• The cluster analysis for RUCA is similar to CELA in direction, but not necessarily
the percentage of said directions. Figure 6-10 shows that 35 percent of trajectories
originated primarily from the northwest. However, some trajectories originating to the
northwest but of shorter distances were grouped with the trajectories originating
offshore (as represented by the cluster shown with 46 percent). The clustering
program uses both direction and distance to determine clusters, which is why there
are two clusters originating to the northwest, one offshore and one inland. The cluster
analysis also shows that approximately 18 percent of trajectories originated from the
northeast. The cluster marked with 2 percent represents the five back trajectories
originating well to the south and off Baja California.
Observations from Figures 6-11 and 6-12 for SJJCA include the following:
• Based on the average length of the back trajectories, the 24-hour air shed domain for
SJJCA is larger than the other two California sites but still smaller compared to other
NMP sites. The farthest away a trajectory originated was less than 600 miles away,
well off shore the Oregon coast. However, 72 percent of trajectories originated within
300 miles of SJJCA and 90 percent originated within 400 miles of the site.
• Back trajectories originated from a variety of directions at SJJCA. The map shows a
larger number of trajectories originating from the northwest to north to northeast of
the site. Fewer trajectories originated from the east, southeast, south, and southwest.
• The cluster analysis shows that 78 percent of trajectories originated from the
northwest and northeast quadrants. Most of the trajectories originating from the
southeast and southwest quadrants were shorter in length and are represented by the
19 percent cluster. Only four back trajectories, representing approximately two
percent of trajectories, originated to the southwest and well offshore.
6-18
-------
6.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather stations at the Downtown Los Angeles/USC
Campus (for CELA), Riverside Municipal Airport (for RUCA), and San Jose International
Airport (for SJJCA) were uploaded into a wind rose software program to produce customized
wind roses, as described in Section 3.5.2.2. A wind rose shows the frequency of wind directions
using "petals" positioned around a 16-point compass, and uses different colors to represent wind
speeds.
Figure 6-13 presents three different wind roses for the CELA monitoring site. First, a
historical wind rose representing 2000 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days that samples were collected in 2010 is presented. These can be used to determine if wind
observations on sample days were representative of conditions experienced over the entire year
and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figures 6-14 and 6-15
present the three wind roses and distance maps for RUCA and SJJCA, respectively.
Observations from Figure 6-13 for CELA include the following:
• The NWS weather station at the Downtown Los Angeles/USC Campus is located
approximately 4.6 miles west-southwest of CELA.
• Historically, winds were generally light near this site, with calm winds (< 2 knots)
observed for 57 percent of the wind observations. For wind speeds greater than
2 knots, westerly and west-southwesterly winds were most common. Wind speeds
greater than 11 knots were not measured at this weather station.
• The 2010 wind rose is similar to the historical wind rose in wind patterns, although
calms winds were observed more often (67 percent) in 2010. Further, the wind
patterns shown on the sample day wind rose also resemble the historical and full-year
wind patterns, indicating that conditions on sample days were representative of those
experienced over the entire year and historically.
6-19
-------
Figure 6-13. Wind Roses for the Downtown Los Angeles/USC Campus Weather Station
near CELA
2000-2009 Historical Wind Rose
2010 Wind Rose
x
'•„ """"•-—
NORTH"--.,
,._ '«, 10%
"*», 8%,
4%,
f 2% ': ; :
^3^— •; ; EAST!
,,-'" /' / WINDSPEED
(Knots)
, • ' * ^B 17
S_OUTH---"'' J 11
H 4-
-21
• 17
11
7
4
Calms: 66.91%
2010 Sample Day Wind Rose
Distance between CELA and NWS Station
.'VEST
i *VR/ &
, ~ «*• i >•<,
j^.,^j ^,
^^
* >» ^•*.
S^«9^
•^ or / •-,,
/_A';~" ; •; ;— "- i -
iffi
6-20
-------
Figure 6-14. Wind Roses for the Riverside Municipal Airport Weather Station near RUCA
1999-2009 Historical Wind Rose
2010 Wind Rose
WIND SPEED
(Knots)
^| 17 • 21
^| 11 • 17
^| 2- 4
Calms: 35.78%
2010 Sample Day Wind Rose
Distance between RUCA and NWS Station
WIND SPEED
(Knots)
cn 4.7
^1 2- 4
Calms: 37.43%
£ i C«»l.
i S
M*~
-_» >• i,s
MWSV
"""" >.,.,!, f
I I I U«M.«. /
\
"^
- .k nk*
Station
HM^
6-21
-------
Figure 6-15. Wind Roses for the San Jose International Airport Weather Station near
SJJCA
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between SJJCA and NWS Station
«^jpv** A-Sz
/ •-.. ^^" - ssr\ •
l=-/^xv ' V'v
v^ -xv
y x^> %-f
4-
^ / ./>-;.. v x- '-:.|-^T
6-22
-------
Observations from Figure 6-14 for RUCA include the following:
• The NWS weather station at the Riverside Municipal Airport is located across the
Santa Ana River and Wildlife Area, approximately 3.5 miles southwest of RUCA.
• Although calm winds were observed approximately 31 percent of the time near
RUCA, westerly and west-northwesterly winds were also frequently observed, based
on the historical wind rose.
• The 2010 wind rose exhibits a similar percentage of calm winds as the historical wind
rose. However, west-northwesterly winds were rarely observed in 2010. Westerly
winds make up almost the same percentage of wind observations in 2010 as both
westerly and west-northwesterly winds on the historical wind rose.
• The wind patterns shown on the sample day wind rose resemble the wind patterns
shown on the full-year wind rose, indicating that conditions on sample days were
representative of those experienced over the entire year.
Observations from Figure 6-15 for SJJCA include the following:
• The NWS weather station at the San Jose International Airport is located
approximately 2 miles northwest of SJJCA.
• Historically, 40 percent of winds were from the northwest to north. Another
20 percent of winds were from the southeast to south. Northeasterly, easterly, and
southwesterly winds were rarely observed. Approximately one-fifth of the winds
were calm.
• The wind patterns on the 2010 and sample day wind roses exhibit a shift in primary
wind direction, from northwest to north on the historical wind rose to west to
northwest on the 2010 wind roses. This shift is also shown in the secondary wind
directions, from southeast to south on the historical to east-southeast to southeast on
the 2010 wind roses. This shift was also shown on the 2009 sample day wind rose in
the 2008-2009 NMP report.
• The wind patterns shown on the sample day wind rose resemble the wind patterns
shown on the full-year wind rose, indicating that conditions on sample days were
representative of those experienced over the entire year.
6.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the California monitoring sites
in order to allow analysts and readers to focus on a subset of pollutants through the context of
risk. For each site, each pollutant's preprocessed daily measurement was compared to its
associated risk screening value. If the concentration was greater than the risk screening value,
then the concentration "failed the screen." Pollutants of interest are those for which the
6-23
-------
individual pollutant's total failed screens contribute to the top 95 percent of the site's total failed
screens. In addition, if any of the NATTS MQO Core Analytes measured by each monitoring site
did not meet the pollutant of interest criteria based on the preliminary risk screening, that
pollutant was added to the list of site-specific pollutants of interest. A more in-depth description
of the risk screening process is presented in Section 3.2.
Table 6-4 presents the pollutants of interest for CELA, RUCA, and SJJCA. The
pollutants that failed at least one screen and contributed to 95 percent of the total failed screens
for each monitoring site are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of
interest are shaded and/or bolded. CELA and RUCA sampled for PAH only, while SJJCA
sampled for metals (PMi0) and PAH.
Table 6-4. Risk Screening Results for the California Monitoring Sites
Pollutant
Screening
Value
(ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Los Angeles, California - CELA
Naphthalene
Acenaphthene
Fluorene
Benzo(a)pyrene
0.029
0.011
0.011
0.00057
Total
58
5
5
2
70
59
59
59
18
195
98.31
8.47
8.47
11.11
35.90
82.86
7.14
7.14
2.86
82.86
90.00
97.14
100.00
Rubidoux, California - RUCA
Naphthalene
Benzo(a)pyrene
0.029
0.00057
Total
56
1
57
60
22
82
93.33
4.55
69.51
98.25
1.75
98.25
100.00
San Jose, California - SJJCA
Naphthalene
Arsenic (PM10)
Manganese (PM10)
Nickel (PM10)
0.029
0.00023
0.005
0.0021
Total
45
37
13
1
96
59
58
58
58
233
76.27
63.79
22.41
1.72
41.20
46.88
38.54
13.54
1.04
46.88
85.42
98.96
100.00
Observations from Table 6-4 include the following:
• Naphthalene failed the majority of screens for all three California monitoring sites,
with its site-specific contribution to the total failed screens ranging from 76 percent
(SJJCA) to 98 percent (CELA).
• Four pollutants failed screens for CELA, including the two PAH NATTS MQO Core
Analytes. In addition to naphthalene, acenaphthene and fluorene were also identified
as pollutants of interest. Benzo(a)pyrene was added to CELA's pollutants of interest
6-24
-------
because it is a NATTS MQO Core Analyte, even though it did not contribute to
95 percent of CELA's total failed screens.
• Benzo(a)pyrene was the only other pollutant besides naphthalene to fail screens for
RUCA. Although this pollutant only failed one screen, it was added as a pollutant of
interest because it is a NATTS MQO Core Analyte.
• Four pollutants failed screens for SJJCA, all of which are NATTS MQO Core
Analytes. Three of these were initially identified as SJJCA's pollutants of interest.
Nickel was added as a pollutant of interest, even though it did not contribute to
95 percent of SJJCA's total failed screens, because it is a NATTS MQO Core
Analyte. Four additional NATTS MQO Core Analytes were added to SJJCA's
pollutants of interest, even though their concentrations did not fail any screens:
benzo(a)pyrene, beryllium, cadmium, and lead. These four pollutants are not shown
in Table 6-4.
6.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the California monitoring sites. Concentration averages are provided for the pollutants of
interest for each site, where applicable. Concentration averages for select pollutants are also
presented graphically for each site, where applicable, to illustrate how each site's concentrations
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at
each site, where applicable. Additional site-specific statistical summaries are provided in
Appendices M and N.
6.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each California site, as described in Section 3.1. The quarterly average of a particular
pollutant is simply the average concentration of the preprocessed daily measurements over a
given calendar quarter. Quarterly average concentrations include the substitution of zeros for all
non-detects. A site must have a minimum of 75 percent valid samples of the total number of
samples possible within a given quarter for a quarterly average to be calculated. An annual
average includes all measured detections and substituted zeros for non-detects for the entire year
of sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the California
6-25
-------
monitoring sites are presented in Table 6-5, where applicable. Note that if a pollutant was not
detected in a given calendar quarter, the quarterly average simply reflects "0" because only zeros
substituted for non-detects were factored into the quarterly average concentration.
Table 6-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest
for the California Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(ng/m3)
2nd
Quarter
Average
(ng/m3)
3rd
Quarter
Average
(ng/m3)
4th
Quarter
Average
(ng/m3)
Annual
Average
(ng/m3)
Los Angeles, California - CELA
Acenaphthene
Benzo(a)pyrene
Fluorene
Naphthalene
59/59
18/59
59/59
59/59
7.74
±3.10
0.14
±0.14
8.34
±2.92
190.70
±76.18
4.21
±1.12
0.07
±0.13
5.06
±0.90
94.79
± 15.28
6.94
±2.96
O.01
±O.01
7.87
±2.81
127.69
±23.04
5.74
±1.76
0.06
±0.06
6.31
±1.66
165.84
±56.21
6.15
±1.18
0.07
±0.05
6.90
±1.11
143.33
± 24.02
Rubidoux, California - RUCA
Benzo(a)pyrene
Naphthalene
22/60
60/60
0.04
±0.03
64.08
±24.16
0.01
±0.01
52.21
±11.24
0.01
±0.02
83.38
±15.81
0.09
±0.09
141.82
±59.37
0.04
±0.02
84.40
± 17.24
San Jose, California - SJJCA
Arsenic (PM10)
Benzo(a)pyrene
Beiy Ilium (PM10)
Cadmium (PM10)
Lead (PM10)
Manganese (PM10)
Naphthalene
Nickel (PM10)
58/58
10/59
57/58
58/58
58/58
58/58
59/59
58/58
0.37
±0.15
0.06
±0.05
O.01
±<0.01
0.09
±0.04
2.57
±1.03
3.37
±1.13
80.47
±25.88
0.79
±0.21
0.20
±0.05
0.01
±0.01
O.01
±O.01
0.04
±0.01
1.74
±1.04
2.66
±0.61
31.46
±8.30
0.64
±0.07
0.49
±0.18
0
O.01
±O.01
0.05
±0.02
1.84
±0.61
4.35
±1.64
45.94
± 17.52
0.95
±0.22
0.40
±0.12
0.03
±0.04
O.01
±O.01
0.06
±0.02
2.48
±0.90
4.81
±1.91
98.20
±38.38
0.99
±0.22
0.37
±0.07
0.02
±0.02
O.01
±O.01
0.06
±0.01
2.13
±0.44
3.76
±0.68
63.44
±13.38
0.84
±0.10
Observations for the California monitoring sites from Table 6-5 include the following:
• Naphthalene and benzo(a)pyrene are pollutants of interest for each site. The annual
average concentration of naphthalene for CELA is higher than RUCA and more than
twice that of SJJCA. A similar pattern in the annual average concentrations of
benzo(a)pyrene is also shown in Table 6-5. Benzo(a)pyrene was detected in less than
half of the samples collected at each site while the other pollutants of interest were
detected in nearly all of the samples collected at each site.
6-26
-------
• Naphthalene and benzo(a)pyrene were generally highest during the first and fourth
quarters of 2010 for each California site and have relatively large confidence intervals
associated with them compared to the other quarters, particularly benzo(a)pyrene.
These large confidence intervals make it difficult to definitively indentify a quarterly
trend in the concentrations, especially for benzo(a)pyrene, which was detected in less
than half of samples collected at these sites and therefore has many substituted zeros
within the calculations, leading to a higher level of variability within the average
concentrations, which is reflected in those confidence intervals.
• The first quarter average concentrations of all four pollutants of interest for CELA are
higher than the other quarterly averages and have relatively high confidence intervals
associated with them, indicating the possibility of outliers. With the exception of
benzo(a)pyrene, the highest concentration of each of the pollutants of interest was
measured on March 21, 2010. For naphthalene, the measurement on March 21, 2010
was 619 ng/m3, while the next two highest concentrations, which were measured
during the fourth quarter of 2010, were considerably lower (359 ng/m3 and
358 ng/m3). For acenaphthene and fluorene, the only concentrations greater than
20 ng/m3 were measured on March 21, 2010 (24.6 ng/m3 and 23.8 ng/m3,
respectively) and August 18, 2010 (22.6 ng/m3 and 21.5 ng/m3, respectively). The
two highest benzo(a)pyrene concentrations were measured on February 1, 2010
(0.987 ng/m3) and June 7, 2010 (0.981 ng/m3); these two concentrations were m<
than three times higher than the third highest concentration.
3
• At RUCA, only four measurements of benzo(a)pyrene were greater than 0.1 ng/m ,
with three of the four being measured during the fourth quarter of 2010. The highest
concentration of this pollutant was measured on December 4, 2010 (0.599 ng/m3). Of
the seven measurements of naphthalene greater than 150 ng/m3, six were measured
during the fourth quarter of 2010. The highest concentration of this pollutant was
measured on December 10, 2010 (336 ng/m3), which is almost half the highest
concentration of naphthalene measured at CELA.
• Benzo(a)pyrene was detected the least at SJJCA (10 out of 59 samples). Five of these
10 were measured during the first quarter of 2010, one was measured during the
second quarter, and four were measured during the fourth quarter of 2010. Of the
three measurements of naphthalene greater than 150 ng/m3, two were measured
during the fourth quarter of 2010 and one in the first quarter of 2010. The highest
concentration of this pollutant was measured at SJJCA on November 4, 2010 (289
ng/m3).
• Of the PMio metals measured at SJJCA, manganese and lead are the only two
pollutants with annual average concentrations greater than 1 ng/m3. Based on the
quarterly averages, the metals do not exhibit a quarterly trend like naphthalene and
benzo(a)pyrene.
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for the
California sites from those tables include the following:
6-27
-------
• CELA appears in Table 4-11 for PAHs a total of four times. CELA has the second
highest annual average concentration naphthalene among NMP sites sampling PAHs,
behind only GPCO. This site also has the third highest annual average of fluorene,
fourth highest concentration of acenaphthene, and ninth highest annual average of
benzo(a)pyrene. RUCA has the tenth highest annual average concentration of
fluorene. SJJCA does not appear in Table 4-11.
• Because only nine sites sampled PMio metals, SJJCA appears in Table 4-12 for every
program-level metal pollutant of interest. However, this site was not in the top five
for any of these pollutants.
6.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for benzo(a)pyrene and
naphthalene were created for all three California sites. Box plots were also created for arsenic
and manganese for SJJCA. Figures 6-16 through 6-19 overlay the sites' minimum, annual
average, and maximum concentrations onto the program-level minimum, first quartile, average,
median, third quartile, and maximum concentrations, as described in Section 3.5.3.
Observations from Figures 6-16 through 6-19 include the following:
• Figure 6-16 shows that the annual average arsenic (PMio) concentration for
SJJCA is less than the program-level average and median concentrations of
arsenic (PMio). There were no non-detects of arsenic measured at SJJCA.
• Figure 6-17 for benzo(a)pyrene shows the annual average concentrations of all
three California sites. The program-level maximum concentration (42.7 ng/m3) is
not shown directly on the box plot because the scale of the box plot would be too
large to readily observe data points at the lower end of the concentration range.
Thus, the scale has been reduced to 2 ng/m3. Note that the first quartile for this
pollutant is zero and is not visible on this box plot. Each of the annual average
concentrations of benzo(a)pyrene for the California sites is below the program-
level average concentration. Figure 6-17 allows the reader to easily visualize how
the California sites' annual average and maximum concentrations compare to
each other.
• Figure 6-18 shows the annual average concentration of manganese (PMio) for
SJJCA is less than the program-level average and just below the program median.
The maximum concentration measured at SJJCA is well below the program
maximum concentration. There were no non-detects of manganese measured at
SJJCA.
6-28
-------
• Figure 6-19 for naphthalene shows the annual average concentrations of all three
California sites. CELA's annual average concentration is greater than the
program-level average and median concentrations, although its maximum
concentration is well below the maximum measured across the program. RUCA's
annual average is just below the program-level average and SJJCA's annual
average is below both the program-level average and median concentrations.
There were no non-detects of naphthalene measured at CELA, RUCA, or SJJCA.
Figure 6-16. Program vs. Site-Specific Average Arsenic (PMio) Concentration
2 2.5 3
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 6-17. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
h
,
Program Max Concentration = 42.7 ng/m3 |
j Program Max Concentration = 42.7 ng/m3
I
I
]
J
k
'
| Program Max Concentration
0 0.2 0.4 0.6 0.8 1 1.2
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
6-29
-------
Figure 6-18. Program vs. Site-Specific Average Manganese (PMi0) Concentration
^
80 100 120 140 160 180 200
Concentration (ng/m3)
Program
Site:
: IstQuartile
Site Average
0
2ndQuartile SrdQuartile 4thQuartile Ave
• n
i i i i
Site Minimum/Maximum
;rage
Figure 6-19. Program vs. Site-Specific Average Naphthalene Concentration
i
lie '
II-
>
1 1 1
\
0
1
1
200 400 600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile
• D D
Site: Site Average Site Minimum/Maximum
o —
1000 1200 1400
4thQuartile Average
6.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. None of the California monitoring sites have sampled continuously for 5 years as
part of the NMP; therefore, the trends analysis was not conducted.
6-30
-------
6.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at each
California monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
6.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
California monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
for each site were compared to the acute MRL; the quarterly averages were compared to the
intermediate MRL; and the annual averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the California monitoring sites were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as a pollutant of interest for
the California monitoring sites.
6.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the California monitoring sites and where annual
average concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 6-6, where applicable.
6-31
-------
Table 6-6. Cancer and Noncancer Surrogate Risk Approximations for the California
Monitoring Sites
Pollutant
Cancer
URE
(Ug/m3)1
Noncancer
RfC
(mg/m3)
# of Measured
Detections vs.
# Samples
Annual
Average
(ng/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Los Angeles, California - CELA
Acenaphthene
Benzo(a)pyrene
Fluorene
Naphthalene
0.000088
0.00176
0.000088
0.000034
0.003
59/59
18/59
59/59
59/59
6.15
±1.18
0.07
±0.05
6.90
±1.11
143.33
± 24.02
0.54
0.12
0.61
4.87
0.05
Rubidoux, California - RUCA
Benzo(a)pyrene
Naphthalene
0.00176
0.000034
0.003
22/60
60/60
0.04
±0.02
84.40
± 17.24
0.07
2.87
0.03
San Jose, California - SJJCA
Arsenic (PM10)
Benzo(a)pyrene
Beryllium (PM10)
Cadmium (PM10)
Lead (PM10)
Manganese (PM10)
Naphthalene
Nickel (PM10)
0.0043
0.00176
0.0024
0.0018
0.000034
0.00048
0.000015
0.00002
0.00001
0.00015
0.00005
0.003
0.00009
58/58
10/59
57/58
58/58
58/58
58/58
59/59
58/58
0.37
±0.07
0.02
±0.02
O.01
±<0.01
0.06
±0.01
2.13
±0.44
3.76
±0.68
63.44
±13.38
0.84
±0.10
1.57
0.04
O.01
0.10
2.16
0.40
0.02
O.01
0.01
0.01
0.08
0.02
0.01
— = a Cancer URE or Noncancer RfC is not available.
Observations for the California sites from Table 6-6 include the following:
• Naphthalene has the highest annual average concentration for each of the California
sites, as discussed in the previous section.
• Naphthalene has the highest cancer risk approximation among the pollutants of
interest for all three California monitoring sites. The cancer risk approximations range
from 2.16 in-a-million for SJJCA to 4.87 in-a-million for CELA.
• Of the metals sampled at SJJCA, arsenic has the highest cancer risk approximation
and is the only metal for which a cancer risk approximation was greater than 1.0 in-a-
million (1.57 in-a-million).
6-32
-------
• All of the noncancer risk approximations for the pollutants of interest for the
California monitoring sites are less than 1.0, indicating no risk of noncancer health
effects.
6.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 6-7 and 6-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 6-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 6-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 6-7 and 6-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. Further, the cancer and noncancer surrogate risk approximations based on each site's
annual averages are limited to those pollutants for which each respective site sampled. As
discussed in Section 6.3, all three California monitoring sites sampled for PAH and SJJCA also
sampled PMio metals. In addition, the cancer and noncancer surrogate risk approximations are
limited to those pollutants with enough data to meet the criteria for annual averages to be
calculated. A more in-depth discussion of this analysis is provided in Section 3.5.5.3.
6-33
-------
Table 6-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the California Monitoring Sites
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Los Angeles, California (Los Angeles County) - CELA
Formaldehyde
Benzene
Dichloromethane
Acetaldehyde
Ethylbenzene
1,3 -Butadiene
Naphthalene
p-Dichlorobenzene
Tetrachloroethylene
Trichloroethylene
3,019.71
1,847.23
1,347.58
1,236.99
959.04
358.22
207.88
144.14
105.99
50.40
Formaldehyde
Hexavalent Chromium, PM
Benzene
1,3 -Butadiene
Naphthalene
Arsenic, PM
Nickel, PM
Acetaldehyde
Ethylbenzene
POM, Group la
3.93E-02
2.57E-02
1.44E-02
1.07E-02
7.07E-03
5.28E-03
5.01E-03
2.72E-03
2.40E-03
1.99E-03
Naphthalene
Fluorene
Acenaphthene
Benzo(a)pyrene
Cancer Risk
Approximation
(in-a-million)
4.87
0.61
0.54
0.12
Rubidoux, California (Riverside County) - RUCA
Formaldehyde
Benzene
Acetaldehyde
Ethylbenzene
Tetrachloroethylene
Dichloromethane
1,3 -Butadiene
Naphthalene
1 ,3 -Dichloropropene
£>-Dichlorobenzene
793.92
409.04
353.08
207.95
163.53
150.35
81.30
55.49
37.72
28.62
Hexavalent Chromium, PM
Formaldehyde
Benzene
1,3 -Butadiene
Naphthalene
Arsenic, PM
Acetaldehyde
POM, Group la
Ethylbenzene
Nickel, PM
2.13E-02
1.03E-02
3.19E-03
2.44E-03
1.89E-03
1.82E-03
7.77E-04
5.85E-04
5.20E-04
4.85E-04
Naphthalene
Benzo(a)pyrene
2.87
0.07
-------
Table 6-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the California Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
San Jose, California (Santa Clara County) - SJJCA
Formaldehyde
Benzene
Acetaldehyde
Ethylbenzene
Dichloromethane
Tetrachloroethylene
1,3 -Butadiene
Naphthalene
£>-Dichlorobenzene
POM, Group la
577.88
353.73
254.78
201.33
102.75
97.93
74.71
39.30
25.48
19.28
Formaldehyde
Hexavalent Chromium, PM
Benzene
1,3 -Butadiene
POM, Group la
Naphthalene
Arsenic, PM
Acetaldehyde
Ethylbenzene
Nickel, PM
7.51E-03
7.13E-03
2.76E-03
2.24E-03
1.70E-03
1.34E-03
9.76E-04
5.61E-04
5.03E-04
2.82E-04
Naphthalene
Arsenic
Nickel
Cadmium
Benzo(a)pyrene
Beryllium
2.16
1.57
0.40
0.10
0.04
0.01
-------
Table 6-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the California Monitoring Sites
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity- Weighted Emissions
(County-Level)
Pollutant
Noncancer
Toxicity
Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Los Angeles, California (Los Angeles County) - CELA
1,1,1 -Trichloroethane
Toluene
Formaldehyde
Benzene
Dichloromethane
Hexane
Acetaldehyde
Ethylbenzene
Xylenes
Ethylene glycol
7,345.47
5,503.09
3,019.71
1,847.23
1,347.58
1,286.59
1,236.99
959.04
873.15
655.71
Acrolein
Chlorine
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Nickel, PM
Arsenic, PM
Hexamethylene- 1 ,6-diisocyanate, gas
Naphthalene
Benzene
8,698,970.49
367,827.04
308,133.92
179,110.50
137,443.72
115,985.91
81,886.64
78,379.05
69,291.67
61,574.38
Naphthalene 0.05
Rubidoux, California (Riverside County) - RUCA
Toluene
1,1,1 -Trichloroethane
Formaldehyde
Benzene
Acetaldehyde
Hexane
Ethylbenzene
Xylenes
Tetrachloroethylene
Dichloromethane
1,289.54
807.28
793.92
409.04
353.08
300.78
207.95
178.62
163.53
150.35
Acrolein
Chlorine
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Arsenic, PM
Naphthalene
Hexavalent Chromium, PM
Bromomethane
Benzene
1,964,801.08
87,536.25
81,012.73
40,650.19
39,230.78
28,239.83
18,496.03
17,714.98
16,899.96
13,634.82
Naphthalene 0.03
Oi
OJ
Oi
-------
Table 6-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the California Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity- Weighted Emissions
(County-Level)
Pollutant
Noncancer
Toxicity
Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
San Jose, California (Santa Clara County) - SJJCA
Toluene
1,1,1 -Trichloroethane
Formaldehyde
Benzene
Hexane
Acetaldehyde
Ethylbenzene
Xylenes
Ethylene glycol
Propionaldehyde
1,394.35
1,290.78
577.88
353.73
265.48
254.78
201.33
176.68
120.59
113.19
Acrolein
Chlorine
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Arsenic, PM
Propionaldehyde
Naphthalene
Benzene
Lead, PM
1,999,590.36
109,927.46
58,967.45
37,354.73
28,308.52
15,130.81
14,149.25
13,100.12
11,790.92
10,622.90
Manganese
Arsenic
Naphthalene
Lead
Nickel
Cadmium
Beryllium
0.08
0.02
0.02
0.01
0.01
0.01
0.01
-------
Observations from Table 6-7 include the following:
• Formaldehyde and benzene are the two highest emitted pollutants with cancer UREs
in all three California counties. The quantity emitted is much higher for Los Angeles
County than Riverside and Santa Clara Counties.
• Formaldehyde and hexavalent chromium are the pollutants with the highest toxicity-
weighted emissions (of the pollutants with cancer UREs) for Los Angeles and Santa
Clara Counties, while the order is reversed for Riverside County.
• Six of the highest emitted pollutants also have the highest toxi city-weighted
emissions for Los Angeles and Riverside Counties, while there are seven in common
for Santa Clara County. While hexavalent chromium is at or near the top in toxicity-
weighted emissions for all three counties, this pollutant is not among the 10 highest
emitted pollutants. Hexavalent chromium emissions rank between 13th highest for
RUCA to 18th highest for SJJCA.
• Naphthalene is the only pollutant to appear on all three lists for all three counties.
This pollutant also has the highest cancer risk approximations for all three sites.
• Arsenic and nickel, which have the second and third highest cancer risk
approximations for SJJCA, respectively, have the seventh and tenth highest toxicity-
weighted emissions for Santa Clara County, but are not one of the 10 highest emitted
pollutants for the county. These are the only pollutants sampled by SJJCA, other than
naphthalene, to appear on either emissions-based list.
Observations from Table 6-8 include the following:
• Toluene, 1,1,1-trichloroethane, formaldehyde, and benzene are the highest emitted
pollutants with noncancer RfCs in all three California counties (although not
necessarily in that order). Consistent with pollutants having cancer UREs, emissions
are higher in Los Angeles County than Riverside and Santa Clara Counties.
• Acrolein, chlorine, and formaldehyde are the pollutants with the highest toxicity-
weighted emissions (of the pollutants with noncancer RfCs) for all three counties.
While acrolein and chlorine rank highest for toxicity-weighted emissions for each
county, neither pollutant appears among the highest emitted for any of the sites.
Conversely, formaldehyde has the third highest emissions for each county.
• Three of the highest emitted pollutants also have the highest toxicity-weighted
emissions for Los Angeles and Riverside Clara Counties, while four of the highest
emitted pollutants also have the highest toxicity-weighted emissions for Santa Clare
County.
• Naphthalene, the only pollutant for which a noncancer risk approximation could be
calculated for CELA and RUCA, has one of the 10 highest toxicity-weighted
emissions for each county, but does not appear on the list of the 10 highest total
emissions for either county. This is also true for Santa Clara County.
6-38
-------
• Arsenic and lead are the only two pollutants for which noncancer risk approximations
could be calculated for SJJCA and that also appear on the list of 10 highest toxicity-
weighted emissions totals. None of the metals appear on the list of the 10 highest total
emissions.
6.6 Summary of the 2010 Monitoring Data for CELA, RUCA, and SJJCA
Results from several of the data treatments described in this section include the
following:
»«» Four PAH, including naphthalene and benzo(a)pyrene, failed screens for CELA,
while only naphthalene and benzo(a)pyrene failed screens for RUCA. Naphthalene
and three metals failed screens for SJJCA.
»«» Naphthalene had the highest annual average concentration among all the pollutants
of interest for the California sites. The annual average concentrations of naphthalene
were higher in magnitude for CELA than for RUCA and SJJCA. CELA 's annual
average naphthalene concentration was the second highest annual average among
NMP sites sampling this pollutant.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
6-39
-------
7.0 Sites in Colorado
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the UATMP and NATTS sites in Colorado, and integrates these
concentrations with emissions, meteorological, and risk information. Data generated by sources
other than ERG are not included in the data analyses contained in this report. Readers are
encouraged to refer to Sections 1 through 4 and the glossary (Appendix P) for detailed
discussions and definitions regarding the various data analyses presented below.
7.1 Site Characterization
This section characterizes the Colorado monitoring sites by providing geographical and
physical information about the location of the sites and the surrounding areas. This information
is provided to give the reader insight regarding factors that may influence the air quality near the
sites and assist in the interpretation of the ambient monitoring measurements.
The NATTS site is located in Grand Junction (GPCO), while the other five sites are
located in Garfield County, between 35 and 55 miles northeast of Grand Junction, in the towns of
Battlement Mesa (BMCO), Silt (BRCO), Parachute (PACO), Rifle (RICO), and Rulison
(RUCO). Figures 7-1 through 7-6 are composite satellite images retrieved from ArcGIS Explorer
showing the monitoring sites in their urban and rural locations. Figures 7-7 and 7-8 identify point
source emissions locations by source category, as reported in the 2008 NEI for point sources.
Note that only sources within 10 miles of each site are included in the facility counts provided in
Figures 7-7 and 7-8. Thus, sources outside the 10-mile radius have been grayed out, but are
visible on the maps to show emissions sources outside the 10-mile boundary. A 10-mile
boundary was chosen to give the reader an indication of which emissions sources and emissions
source categories could potentially have a direct effect on the air quality at the monitoring sites.
Further, this boundary provides both the proximity of emissions sources to the monitoring sites
as well as the quantity of such sources within a given distance of the sites. Table 7-1 describes
the areas surrounding the monitoring sites by providing supplemental geographical information
such as land use, location setting, and locational coordinates.
7-1
-------
Figure 7-1. Grand Junction, Colorado (GPCO) Monitoring Site
to
-------
Figure 7-2. Battlement Mesa, Colorado (BMCO) Monitoring Site
-------
Figure 7-3. Silt, Colorado (BRCO) Monitoring Site
-------
Figure 7-4. Parachute, Colorado (PACO) Monitoring Site
-------
Figure 7-5. Rifle, Colorado (RICO) Monitoring Site
-------
Figure 7-6. Rulison, Colorado (RUCO) Monitoring Site
-------
Figure 7-7. NEI Point Sources Located Within 10 Miles of GPCO
Legend
8; 35TTW 1QS=3Q'CrW 10fl*25lD'W 10fi; SQ'ffW
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
GPCO NATTS site
10 mile radius
County boundary
Source Category Group (No. of Facilities)
41 Aircraft Operations (9)
0 Auto Body Shop/Painters (4)
iS Automobile/Truck Manufacturing (2)
Brick Manufacturing & Structural Clay (1)
B Bulk Terminals/Bulk Plants (4)
C Chemical Manufacturing (1)
• Concrete Batch Plant (6)
- Crematory -Animal/Human (4)
(D Dry Cleaning (4)
e Electrical Equipment (2)
E Electroplating. Plating. Polishing, Anodizing, and Coloring (2)
© Fabricated Metal Products (1)
IL1 Furniture Plant (1)
f Gasoline/Diesel Service Station (47)
it Hot Mix Asphalt Plant (2)
# Industrial Machinery and Equipment (1)
® Institutional - school (4)
• Landfill (1)
.-- Mine/Quarry (23)
? Miscellaneous Commercial/Industrial (3)
• Oil andtor Gas Production (4)
R Rubber and Miscellaneous Plastics Products (3)
2 Secondary Metal Processing (1)
< Site Remediation Activity (2)
S Surface Coating (2)
T Textile Mill (1)
• Wastewater Treatment (1)
7-8
-------
Figure 7-8. NEI Point Sources Located Within 10 Miles of BMCO, BRCO, PACO, RICO,
and RUCO
108-15'O-W 108-KTO-W 10S'5T)-W
Legend
@ BMCO UATMP site
* BRCO UATMP site
107'55'0'W 107-50'CrW 107^5'O'W m*WG~W 107-35'0'W HDT'M'O'W 107'25'0"W
Note Due to facility density and collocation, Ihe total facilities
displayed may not represent all facilities within the area of interest.
PACO UATMP site
RICO UATMP site
RUCO UATMP site
10 mile radius
County boundary
Source Category Group (No. of Facilities) r Gasoline/Diesel Service Station (17)
41 Aircraft Operations (7)
ft Building Construction (1)
B Bulk Terminals/Bulk Plants (1)
• Concrete Batch Plant (1)
tf> Dry Cleaning (1)
* Electricity Generation via Combustion (1)
$ Hot Mix Asphalt Plant (1)
• Landfill (1)
K Mine/Quarry {12}
? Miscellaneous Commercial/Industrial (2)
• Oil anchor Gas Production (920)
1 Pipeline Compressor Station (8)
7-9
-------
Table 7-1. Geographical Information for the Colorado Monitoring Sites
Site
Code
GPCO
BMCO
BRCO
PACO
RICO
RUCO
AQS Code
08-077-0017
08-077-0018
NA
08-045-0009
08-045-0005
08-045-0007
NA
Location
Grand
Junction
Battlement
Mesa
Silt
Parachute
Rifle
Rulison
County
Mesa
Garfield
Garfield
Garfield
Garfield
Garfield
Micro- or
Metropolitan
Statistical Area
Grand Junction,
CO MSA
Not in an MSA
Not in an MSA
Not in an MSA
Not in an MSA
Not in an MSA
Latitude
and
Longitude
39.064289,
-108.56155
39.4399898,
-108.029769
39.487755,
-107.659685
39.453654,
-108.053259
39.531813,
-107.782298
39.488744,
-107.936989
Land Use
Commercial
Residential
Agricultural
Residential
Commercial
Agricultural
Location
Setting
Urban/City
Center
Rural
Rural
Urban/City
Center
Urban/City
Center
Rural
Additional Ambient Monitoring Information1
Meteorological parameters, CO, PM10, PM10
Speciation, PM2 5, and PM2 5 Speciation.
No AQS entry.
None.
PMio, PM10 Speciation.
PMio, PM10 Speciation.
No AQS entry.
1 These monitoring sites report additional pollutants to AQS (EPA, 2012h); however, these data are not generated by ERG and are therefore not included in this report.
BOLD ITALICS = EPA-designaled NATTS Site.
NA = No AQS entry
-------
The GPCO monitoring site is comprised of two locations. The first location is a small
1-story shelter that houses the VOC and carbonyl compound samplers, with the PAH sampler
located just outside the shelter. The second location is on an adjacent 2-story building that has
the hexavalent chromium samplers on the roof. As a result, two AQS codes are provided
in Table 7-1. Figure 7-1 shows that the area surrounding GPCO is of mixed usage, with
commercial businesses to the west, northwest and north, residential areas to the northeast and
east, and industrial areas to the southeast, south and southwest. The site's location is next to one
of the major east-west roads in Grand Junction (1-70 Business). A railroad runs east-west to the
south of the GPCO monitoring site, and merges with another railroad to the southwest of the site.
As Figure 7-7 shows, GPCO is located within 10 miles of numerous emissions sources. Many of
the sources are located along a diagonal line running roughly northwest to southeast along
Highways 6 and 50 and Business 70. Many of the point sources near GPCO fall into the
gasoline/diesel service station and mine/quarry source categories. The sources closest to GPCO
are a bulk terminal/bulk plant, an automobile/truck manufacturer, and a gasoline/diesel service
station.
The BMCO monitoring site is located in Battlement Mesa, a rural community located to
the southeast of Parachute. As shown in Figure 7-2, the monitoring site is located on the roof of
the Grand Valley Fire Protection District facility, near the intersection of Stone Quarry Road and
W. Battlement Parkway. The site is surrounded primarily by residential subdivisions. A cemetery
is located to the south of the site and a church to the east.
The BRCO monitoring site is located on Bell/Melton Ranch, off Owens Drive,
approximately 4 miles south of the town of Silt. The site is both rural and agricultural in nature.
As shown in Figure 7-3, the closest major roadway is County Road 331, Dry Hollow Road.
PACO is located on the roof of the old Parachute High School building, which is
presently operating as a day care facility. This location is in the center of the town of Parachute,
as shown in Figure 7-4. The surrounding area is considered residential. Interstate-70 is less than
a quarter of a mile from the monitoring site.
7-11
-------
RICO is located on the roof of the Henry Annex Building in downtown Rifle. This
location is near the crossroads of several major roadways through town, as shown in Figure 7-5.
Highway 13 and US-6 intersect just south of the site and across the Colorado River, 1-70 is just
over a half-mile south of the monitoring site. The surrounding area is considered commercial.
RUCO is located on the Potter Ranch, in Rulison, Colorado, about halfway between the
towns of Parachute and Rifle. This location is less than 1 mile south of the 1-70, as shown in
Figure 7-6. The surrounding area is considered rural and agricultural.
The five Garfield County sites are located along a line running roughly east-west and
spanning approximately 20 miles; hence they are shown together in Figure 7-8. There are more
than 900 petroleum or natural gas wells (collectively shown as the oil and/or gas production
source category) within 10 miles of these sites. One reason Garfield County is conducting air
monitoring is to characterize the effects these wells may have on the surrounding areas (GCPH,
2010).
Table 7-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the areas surrounding the
Colorado monitoring sites. Table 7-2 also includes a vehicle registration-to-county population
ratio (vehicles-per-person) for each site. In addition, the population within 10 miles of each site
is presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-
level vehicle registration-to-population ratio to the 10-mile population surrounding each
monitoring site. Table 7-2 also contains annual average daily traffic information. Finally, Table
7-2 presents the daily VMT for Mesa and Garfield Counties. Note that the VMT presented is for
state highways only, which differs from the VMT presented in this table in other state sections.
7-12
-------
Table 7-2. Population, Motor Vehicle, and Traffic Information for the Colorado
Monitoring Sites
Site
GPCO
BMCO
BRCO
PACO
RICO
RUCO
Estimated
County
Population1
146,313
56,139
County-level
Vehicle
Registration2
180,119
74,847
Vehicles
per Person
(Registration:
Population)
1.23
1.33
Population
within
10 miles3
117,098
5,941
24,174
7,898
17,641
17,641
Estimated
10-mile
Vehicle
Ownership
144,154
7,921
32,230
10,530
23,520
23,520
Annual
Average
Daily
Traffic4
12,000
2,527
150
2,600
17,000
699
County-level
Daily VMT5
2,047,739
1,942,038
1 County-level population estimates reflect data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2009 data from the Colorado Dept of Revenue (CO DOR, 2010)
3 10-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2002 data for BMCO, BRCO, and RUCO from Garfield
County (GCRBD, 2002) and 2010 data for GPCO, PACO, and RICO from the Colorado DOT (CO DOT, 2011)
5 County-level VMT reflects 2010 data for state highways only from the Colorado DOT (CO DOT, 2010)
BOLD ITALICS = EPA-designated NATTS Site.
Observations from Table 7-2 include the following:
• Mesa County's population and vehicle ownership are considerably higher than those
for Garfield County. This is also true for its 10-mile population and vehicle
ownership. However, both counties rank in the bottom-third compared to all counties
with NMP sites.
• The vehicle-per-person ratios for all six sites are among the highest for all NMP sites.
• The traffic volumes near GPCO and RICO are considerably higher than the traffic
volumes near the other Garfield County sites. With the exception of RICO, the traffic
volumes near the Colorado sites are in the bottom-third compared to other NMP sites.
The lowest traffic volume among all NMP sites is for BRCO. The traffic estimate for
GPCO came from Business-70 near 5th Avenue; from S. Battlement Parkway for
BMCO; from the junction of County Roads 331 and 326 for BRCO; from Route 6 at
Parachute Avenue for PACO; from the junction of Highway 13 and 1-70 for RICO;
and from County Road 323 for RUCO.
• While the Mesa and Garfield County VMTs are fairly similar to each other, they are
also among the lowest for counties with NMP sites, where VMT data were available.
However, the county-level VMT available from the Colorado DOT is for state
highways only.
7-13
-------
7.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in Colorado on sample days, as well as over the course of the year.
7.2.1 Climate Summary
Grand Junction is located in a mountain valley on the west side of the Rockies. The
valley location of the city helps protect it from dramatic weather changes. The area tends to be
fairly dry and winds tend to flow out of the east-southeast on average, due to the valley breeze
effect (Bair, 1992). Valley breezes occur as the sun heats up the side of a mountain; the warm air
rises, creating a current that will move up the valley walls (Boubel, et al., 1994).
The towns of Battlement Mesa, Parachute, Rifle, Rulison, and Silt are located to the
northeast of Grand Junction, across the county line and along the 1-70 corridor. These towns are
located along a river valley running north of the Grand Mesa. Similar to Grand Junction, these
towns are shielded from drastic changes in weather by the surrounding terrain and tend to
experience fairly dry conditions for most of the year. Wind patterns in these towns are affected
by the high canyons, the Colorado River, and valley breezes (GCPH, 2010 and WRCC, 2011).
7.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather stations nearest these sites were
retrieved for 2010 (NCDC, 2010). The weather station nearest GPCO is located at Walker Field
Airport (WBAN 23066); the closest weather station to the five Garfield County sites is located at
Garfield County Regional Airport (WBAN 03016). Additional information about these weather
stations, such as the distance between the sites and the weather stations, is provided in Table 7-3.
These data were used to determine how meteorological conditions on sample days vary from
normal conditions throughout the year.
7-14
-------
Table 7-3. Average Meteorological Conditions near the Colorado Monitoring Sites
Closest NWS
Station (WBAN
and Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Grand Junction, Colorado - GPCO
Walker Field
Airport
23066
(39.13, -108.54)
4.96
miles
22°
(NNE)
Sample
Day
2010
64.4
±5.8
64.5
±2.3
52.8
±5.2
52.8
±2.1
31.0
±2.7
30.4
±1.2
41.8
±3.3
41.6
±1.4
52.6
±6.0
51.3
±2.4
1014.0
±1.9
1014.0
±0.9
5.8
±0.6
6.2
±0.3
Battlement Mesa, Colorado - BMCO
Garfield Co.
Regional Airport
03016
(39.53, -107.73)
16.41
miles
76°
(ENE)
Sample
Day
2010
55.5
±9.1
62.6
±2.3
44.7
±7.0
48.9
±2.0
29.2
±4.7
28.4
±1.2
37.5
±5.0
39.0
±1.4
60.9
±8.5
53.9
±2.1
1017.4
±3.1
1015.3
±0.8
4.0
±1.2
4.4
±0.3
Silt, Colorado - BRCO
Garfield Co.
Regional Airport
03016
(39.53, -107.73)
4.23
miles
316°
(NW)
Sample
Day
2010
62.8
±5.9
62.6
±2.3
49.4
±4.9
48.9
±2.0
28.7
±2.7
28.4
±1.2
39.3
±3.3
39.0
±1.4
53.6
±5.2
53.9
±2.1
1015.1
±1.8
1015.3
±0.8
4.3
±0.6
4.4
±0.3
Parachute, Colorado - PACO
Garfield Co.
Regional Airport
03016
(39.53, -107.73)
17.22
miles
81°
(E)
Sample
Day
2010
62.1
±5.7
62.6
±2.3
48.6
±4.8
48.9
±2.0
28.3
±2.8
28.4
±1.2
38.7
±3.3
39.0
± 1.4
54.0
±4.9
53.9
±2.1
1015.9
±1.8
1015.3
±0.8
4.2
±0.6
4.4
±0.3
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
Table 7-3. Average Meteorological Conditions near the Colorado Monitoring Sites (Continued)
Closest NWS
Station (WBAN
and Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
<°F)
Average
Dew Point
Temperature
<°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Rifle, Colorado - RICO
Garfield Co.
Regional Airport
03016
(39.53, -107.73)
2.89
miles
105°
(ESE)
Sample
Day
2010
62.5
±5.9
62.6
±2.3
49.2
±4.9
48.9
±2.0
28.8
±2.8
28.4
±1.2
39.2
±3.3
39.0
±1.4
54.1
±5.1
53.9
±2.1
1015.4
±1.8
1015.3
±0.8
4.2
±0.6
4.4
±0.3
Rulison, Colorado - RUCO
Garfield Co.
Regional Airport
03016
(39.53, -107.73)
10.94
miles
84°
(E)
Sample
Day
2010
66.2
±7.3
62.6
±2.3
51.6
±6.4
48.9
±2.0
28.9
±3.5
28.4
±1.2
40.3
±4.3
39.0
±1.4
50.5
±6.2
53.9
±2.1
1014.9
±2.3
1015.3
±0.8
4.4
±0.7
4.4
±0.3
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
Table 7-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 7-3 is the
95 percent confidence interval for each parameter. As shown in Table 7-3, average
meteorological conditions on sample days near each site were representative of average weather
conditions throughout the year, with one exception. The instruments at RUCO were moved in
September 2010 to BMCO, where sampling took place from mid-September through the end of
year; thus only sample days in the cooler months of autumn and the beginning of winter are
included in the 2010 sample day average for BMCO. This explains the difference in several of
the meteorological parameters, such as maximum and average temperatures and relative
humidity, for BMCO. The sample day averages for RUCO were also affected by the exclusion of
the latter portion of the year, but to a lesser extent.
7.2.3 Back Trajectory Analysis
Figure 7-9 is the composite back trajectory map for days on which samples were
collected at the GPCO monitoring site in 2010. Included in Figure 7-9 are four back trajectories
per sample day. Figure 7-10 is the corresponding cluster analysis for 2010. Similarly,
Figures 7-11 through 7-19 are the composite back trajectory maps and corresponding cluster
analyses for the Garfield County monitoring sites. A cluster analysis was not performed for
BMCO because this site has less than 30 sample days. An in-depth description of these maps and
how they were generated is presented in Section 3.5.2.1. For the composite maps, each line
represents the 24-hour trajectory along which a parcel of air traveled toward the monitoring site
on a given sample day and time. For the cluster analyses, each line corresponds to a back
trajectory representative of a given cluster of trajectories. For all maps, each concentric circle
around the sites in Figures 7-9 through 7-21 represents 100 miles.
7-17
-------
Figure 7-9. 2010 Composite Back Trajectory Map for GPCO
Figure 7-10. Back Trajectory Cluster Map for GPCO
7-18
-------
Figure 7-11. 2010 Composite Back Trajectory Map for BMCO
Figure 7-12. 2010 Composite Back Trajectory Map for BRCO
7-19
-------
Figure 7-13. Back Trajectory Cluster Map for BRCO
Figure 7-14. 2010 Composite Back Trajectory Map for PACO
7-20
-------
Figure 7-15. Back Trajectory Cluster Map for PACO
Figure 7-16. 2010 Composite Back Trajectory Map for RICO
7-21
-------
Figure 7-17. Back Trajectory Cluster Map for RICO
Figure 7-18. 2010 Composite Back Trajectory Map for RUCO
7-22
-------
Figure 7-19. Back Trajectory Cluster Map for RUCO
Observations for GPCO from Figures 7-9 and 7-10 include the following:
• The 24-hour air shed domain for GPCO was smaller than most other NMP
monitoring sites. The farthest away a trajectory originated was central Arizona, or just
less than 400 miles away. However, most trajectories (89 percent) originated within
300 miles of GPCO and the average trajectory length was approximately 150 miles.
• Back trajectories originated from a variety of directions at GPCO, although the
majority of them had a westerly component.
• The cluster analysis shows that back trajectories frequently originated from the
northwest, west, and southwest. Shorter back trajectories (200 miles or less)
originating from the south (labeled 22 percent) were also common. The short cluster
originating to the southeast represented several relatively short back trajectories
originating from the northeast, east, southeast, and south. Thus, air moving towards
GPCO is generally originating in Colorado, Utah, and Arizona.
Observations from Figures 7-11 through 7-19 for the Garfield County sites include the
following:
• The composite back trajectory maps for the Garfield County sites resemble the ones
for GPCO. This is expected, given the sites' close proximity to GPCO.
• The 24-hour air shed domains were among the smallest in size compared to other
NMP sites, with the longest trajectories originating over northwest Wyoming, or just
7-23
-------
less than 400 miles away. Note that this trajectory (November 29, 2010) is absent in
Figure 7-18 for RUCO because the instrumentation had been moved to BMCO and
thus this sample day is not included on this map. The average back trajectory length
ranged from 145 to 148 miles for the Garfield County sites.
• Most of the back trajectories for the Garfield County sites had a westerly component,
as confirmed by the cluster analysis maps. Those with a northeasterly to southeasterly
component were generally represented by the short cluster originating to the southeast
and represented up to a quarter of the trajectories.
• The composite back trajectory map for BMCO has fewer trajectories shown in
Figure 7-11 because this site did not begin sampling until September 2010. A cluster
analysis was not performed for BMCO because this site has less than 30 sample days.
7.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather stations at the Walker Field Airport (for GPCO)
and Garfield County Regional Airport (for BMCO, BRCO, PACO, RICO, and RUCO) were
uploaded into a wind rose software program to produce customized wind roses, as described in
Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals" positioned
around a 16-point compass, and uses different colors to represent wind speeds.
Figure 7-20 presents three different wind roses for the GPCO monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days that samples were collected in 2010 is presented. These can be used to determine if wind
observations on sample days were representative of conditions experienced over the entire year
and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figures 7-21 through 7-25
present the wind roses and distance maps for the Garfield County monitoring sites.
7-24
-------
Figure 7-20. Wind Roses for the Walker Field Airport Weather Station near GPCO
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between GPCO and NWS Station
7-25
-------
Figure 7-21. Wind Roses for the Garfield County Regional Airport near BMCO
1999-2009 Historical Wind Rose
2010 Wind Rose
WEST
2010 Sample Day Wind Rose
Distance between BMCO and NWS Station
WES r
7-26
-------
Figure 7-22. Wind Roses for the Garfield County Regional Airport near BRCO
1999-2009 Historical Wind Rose
2010 Wind Rose
WEST
2010 Sample Day Wind Rose
Distance between BRCO and NWS Station
7-27
-------
Figure 7-23. Wind Roses for the Garfield County Regional Airport near PACO
1999-2009 Historical Wind Rose
2010 Wind Rose
I'S'ES I
2010 Sample Day Wind Rose
Distance between PACO and NWS Station
7-28
-------
Figure 7-24. Wind Roses for the Garfield County Regional Airport near RICO
1999-2009 Historical Wind Rose
2010 Wind Rose
WEST
2010 Sample Day Wind Rose
Distance between RICO and NWS Station
N
+
7-29
-------
Figure 7-25. Wind Roses for the Garfield County Regional Airport near RUCO
1999-2009 Historical Wind Rose
2010 Wind Rose
WEST
2010 Sample Day Wind Rose
Distance between RUCO and NWS Station
N
-f
7-30
-------
Observations from Figure 7-20 for GPCO include the following:
• The Walker Field Airport weather station is located approximately 5 miles north-
northeast of GPCO.
• The historical wind rose shows that easterly, east-southeasterly, and southeasterly
winds were prevalent near GPCO. Calm winds (< 2 knots) were observed for less
than 15 percent of the hourly wind measurements.
• The 2010 wind rose exhibits similar wind patterns as the historical wind rose. Further,
the sample day wind patterns also resemble the historical and full-year wind patterns,
indicating that conditions on sample days were representative of those experienced
over the entire year and historically.
Observations from Figures 7-21 through 7-25 for the Garfield County sites include the
following:
• The NWS weather station at Garfield County Regional Airport is the closest weather
station to all five monitoring sites in Garfield County. The weather station is located
just east of Rifle. The distances from the weather station to the sites varies from less
than 3 miles (RICO) to just over 17 miles (PACO).
• The historical and 2010 wind roses for the Garfield County sites are identical to each
other. This is because the wind observations came from the same NWS weather
station for all five sites.
• The historical wind roses show that calm winds were prevalent (representing just less
than one-third of observations) near these five monitoring sites. Westerly and
southerly winds were also common.
• The 2010 wind roses exhibit similar wind patterns as the historical wind rose,
although there was a slightly higher percentage of calm wind observations and
slightly fewer southerly wind observations.
• With the exception of BMCO, the sample day wind patterns for each site also
resemble the historical and full-year wind patterns, indicating that conditions on
sample days were representative of those experienced over the entire year and
historically.
• BMCO's sample day wind rose is the only one that differs from the full-year and
historical wind roses. Calm winds were still prevalent but represented 40 percent of
the wind observations. Westerly winds were still frequently observed, but the
southerly to south-southwesterly wind observations were greatly reduced. Instead,
there was a higher number of west-northwesterly to northwesterly winds. Because
this wind rose includes only sample days from mid-September through the end of the
year, this wind rose is likely exhibiting a seasonal wind pattern variation.
7-31
-------
7.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Colorado monitoring sites in
order to allow analysts and readers to focus on a subset of pollutants through the context of risk.
For each site, each pollutant's preprocessed daily measurement was compared to its associated
risk screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by each monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 7-4 presents the pollutants of interest for each Colorado monitoring site. The
pollutants that failed at least one screen and contributed to 95 percent of the total failed screens
for each monitoring site are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of
interest are shaded and/or bolded. GPCO sampled for VOC, carbonyl compounds, PAH, and
hexavalent chromium; the Garfield County sites sampled for SNMOC and carbonyl compounds
only.
Table 7-4. Risk Screening Results for the Colorado Monitoring Sites
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Grand Junction, Colorado - GPCO
Acetaldehyde
Formaldehyde
Benzene
1,3-Butadiene
Naphthalene
Carbon Tetrachloride
Ethylbenzene
Acenaphthene
Acrylonitrile
1 ,2-Dichloroethane
Benzo(a)pyrene
Dichloromethane
Acenaphthylene
Bromomethane
£>-Dichlorobenzene
0.45
0.077
0.13
0.03
0.029
0.17
0.4
0.011
0.015
0.038
0.00057
7.7
0.011
0.5
0.091
61
61
59
57
56
54
33
8
8
7
5
4
2
2
2
61
61
59
58
57
59
59
57
8
7
32
59
41
52
14
100.00
100.00
100.00
98.28
98.25
91.53
55.93
14.04
100.00
100.00
15.63
6.78
4.88
3.85
14.29
14.39
14.39
13.92
13.44
13.21
12.74
7.78
1.89
1.89
1.65
1.18
0.94
0.47
0.47
0.47
14.39
28.77
42.69
56.13
69.34
82.08
89.86
91.75
93.63
95.28
96.46
97.41
97.88
98.35
98.82
7-32
-------
Table 7-4. Risk Screening Results for the Colorado Monitoring Sites (Continued)
Pollutant
Fluorene
1 ,2-Dibromoethane
Hexachloro- 1 ,3 -butadiene
Propionaldehyde
Screening
Value
(Ug/m3)
0.011
0.0017
0.045
0.8
Total
#of
Failed
Screens
2
1
1
1
424
#of
Measured
Detections
57
1
2
61
805
%of
Screens
Failed
3.51
100.00
50.00
1.64
52.67
% of Total
Failures
0.47
0.24
0.24
0.24
Cumulative
%
Contribution
99.29
99.53
99.76
100.00
Rifle, Colorado - BMCO
Benzene
1,3-Butadiene
Formaldehyde
Acetaldehyde
0.13
0.03
0.077
0.45
Total
18
8
7
6
39
18
8
7
7
40
100.00
100.00
100.00
85.71
97.50
46.15
20.51
17.95
15.38
46.15
66.67
84.62
100.00
Silt, Colorado - BRCO
Benzene
Formaldehyde
Acetaldehyde
1,3-Butadiene
Ethylbenzene
0.13
0.077
0.45
0.03
0.4
Total
60
17
12
6
5
100
61
17
17
6
61
162
98.36
100.00
70.59
100.00
8.20
61.73
60.00
17.00
12.00
6.00
5.00
60.00
77.00
89.00
95.00
100.00
Parachute, Colorado - PACO
Benzene
Formaldehyde
Acetaldehyde
1,3-Butadiene
Ethylbenzene
0.13
0.077
0.45
0.03
0.4
Total
58
28
25
22
13
146
58
28
28
22
58
194
100.00
100.00
89.29
100.00
22.41
75.26
39.73
19.18
17.12
15.07
8.90
39.73
58.90
76.03
91.10
100.00
Rifle, Colorado - RICO
Benzene
1,3-Butadiene
Ethylbenzene
Acetaldehyde
Formaldehyde
0.13
0.03
0.4
0.45
0.077
Total
60
38
30
24
24
176
60
38
60
24
24
206
100.00
100.00
50.00
100.00
100.00
85.44
34.09
21.59
17.05
13.64
13.64
34.09
55.68
72.73
86.36
100.00
Rulison, Colorado - RUCO
Benzene
Formaldehyde
Acetaldehyde
1,3-Butadiene
Ethylbenzene
0.13
0.077
0.45
0.03
0.4
Total
40
18
17
8
4
87
40
18
18
8
39
123
100.00
100.00
94.44
100.00
10.26
70.73
45.98
20.69
19.54
9.20
4.60
45.98
66.67
86.21
95.40
100.00
Observations from Table 7-4 include the following:
• Nineteen pollutants failed at least one screen for GPCO, of which seven are NATTS
MQO Core Analytes.
7-33
-------
• Ten pollutants were initially identified as pollutants of interest for GPCO based on
the risk screening process, of which six are NATTS MQO Core Analytes.
Benzo(a)pyrene was added to GPCO's pollutants of interest because it is a NATTS
MQO Core Analyte, even though it did not contribute to 95 percent of GPCO's total
failed screens. Five additional NATTS MQO Core Analytes were also added to
GPCO's pollutants of interest even though their concentrations did not fail any
screens: chloroform, hexavalent chromium, tetrachloroethylene, trichloroethylene,
and vinyl chloride. These five pollutants are not shown in Table 7-4.
• The number of pollutants failing screens for the Garfield County sites ranged from
four to five. Four pollutants (benzene, 1,3-butadiene, formaldehyde, and
acetaldehyde) failed screens for each Garfield County site. These four pollutants were
identified as pollutants of interest for all five sites. Ethylbenzene also failed screens
for four of the five Garfield County sites (BMCO being the exception). Ethylbenzene
was identified as a pollutant of interest for PACO and RICO.
• Note that carbonyl compound samples were collected on a l-in-12 day sampling
schedule at the Garfield County sites, while SNMOC were collected on a l-in-6 day
sampling schedule; thus, there were often less than half the number of samples of
carbonyl compounds collected at these sites than SNMOC.
• Formaldehyde failed 100 percent of screens for all six Colorado sites. Benzene and
1,3-butadiene failed 100 percent of screens at most of the sites.
7.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Colorado monitoring sites. Concentration averages are provided for the pollutants of
interest for each Colorado monitoring site, where applicable. Concentration averages for select
pollutants are also presented graphically for each site, where applicable, to illustrate how each
site's concentrations compare to the program-level averages. In addition, concentration averages
for select pollutants are presented from previous years of sampling in order to characterize
concentration trends at the sites, where applicable. Additional site-specific statistical summaries
are provided in Appendices J through M and Appendix O.
7.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each Colorado site, as described in Section 3.1. The quarterly average of a particular
pollutant is simply the average concentration of the preprocessed daily measurements over a
given calendar quarter. Quarterly average concentrations include the substitution of zeros for all
non-detects. A site must have a minimum of 75 percent valid samples of the total number of
7-34
-------
samples possible within a given quarter for a quarterly average to be calculated. An annual
average includes all measured detections and substituted zeros for non-detects for the entire year
of sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the Colorado
monitoring sites are presented in Table 7-5, where applicable. Note that concentrations of the
PAH and hexavalent chromium for GPCO are presented in ng/m3 for ease of viewing. Also note
that if a pollutant was not detected in a given calendar quarter, the quarterly average simply
reflects "0" because only zeros substituted for non-detects were factored into the quarterly
average concentration.
Table 7-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Colorado Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(jig/m3)
2nd
Quarter
Average
(jig/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Annual
Average
(jig/m3)
Grand Junction, Colorado - GPCO
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Ethylbenzene
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
61/61
8/59
59/59
58/59
59/59
54/59
7/59
59/59
61/61
59/59
19/59
2/59
1.95
±0.42
0.03
±0.03
1.46
±0.26
0.15
±0.04
0.56
±0.11
0.06
±0.02
0.01
±0.02
0.43
±0.10
2.41
±0.38
0.35
±0.14
0
0
1.59
±0.34
0.03
±0.04
1.09
±0.28
0.07
±0.01
0.39
±0.10
0.08
±0.01
0.02
±0.02
0.35
±0.07
2.23
±0.43
0.26
±0.11
0.02
±0.02
<0.01
±<0.01
2.26
±0.40
0.01
±0.03
1.27
±0.36
0.10
±0.03
0.59
±0.08
0.11
±0.02
0
0.58
±0.18
3.39
±0.44
0.38
±0.13
0.04
±0.02
0
2.19
±0.40
0
1.79
±0.38
0.22
±0.06
0.57
±0.10
0.09
±0.02
0
0.64
±0.15
3.04
±0.40
0.60
±0.18
0.06
±0.04
0
2.00
±0.20
0.02
±0.01
1.40
±0.16
0.14
±0.02
0.53
±0.05
0.09
±0.01
0.01
±0.01
0.50
±0.07
2.78
±0.23
0.39
±0.07
0.03
±0.01
<0.01
±<0.01
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line for GPCO are presented in ng/m3
for ease of viewing.
7-35
-------
Table 7-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Colorado Monitoring Sites (Continued)
Pollutant
Acenaphthene3
Benzo(a)pyrenea
Hexavalent Chromium3
Naphthalene3
#of
Measured
Detections
vs. # of
Samples
57/57
32/57
35/58
57/57
1st
Quarter
Average
(Ug/m3)
4.41
±1.59
0.35
±0.18
0.01
±0.01
143.02
±36.11
2nd
Quarter
Average
(Ug/m3)
7.44
±2.08
0.01
±0.01
0.01
±0.01
97.64
±21.69
3rd
Quarter
Average
(Ug/m3)
11.75
±4.50
0.01
±0.02
0.01
±0.01
158.19
± 52.99
4th
Quarter
Average
(Ug/m3)
5.37
±1.76
0.38
±0.23
0.01
±0.01
192.03
± 59.44
Annual
Average
(Ug/m3)
7.30
±1.53
0.18
±0.08
0.01
±<0.01
147.04
±22.61
Battlement Mesa, Colorado - BMCO
Acetaldehyde
Benzene
1,3 -Butadiene
Formaldehyde
111
18/18
8/18
111
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
0.90
±0.34
1.43
±0.34
0.03
±0.02
1.39
±0.46
NA
NA
NA
NA
Silt, Colorado - BRCO
Acetaldehyde
Benzene
1,3 -Butadiene
Formaldehyde
17/17
61/61
6/61
17/17
NA
1.28
±0.35
<0.01
±0.01
NA
0.96
±0.34
0.88
±0.19
0
1.10
±0.26
0.84
±0.42
1.07
±0.45
0.01
±0.01
1.27
±0.51
NA
1.15
±0.38
0.03
±0.04
NA
NA
1.10
±0.18
0.01
±0.01
NA
Parachute, Colorado - PACO
Acetaldehyde
Benzene
1,3 -Butadiene
Ethylbenzene
Formaldehyde
28/28
58/58
22/58
58/58
28/28
NA
1.77
±0.38
0.03
±0.03
0.32
±0.07
NA
0.99
±0.25
1.71
±0.42
0.01
±0.02
0.31
±0.05
1.39
±0.29
1.08
±0.37
1.67
±0.35
0.03
±0.02
2.96
±3.84
1.86
±0.42
0.89
±0.29
1.74
±0.55
0.17
±0.12
0.26
±0.07
1.46
±0.30
0.93
±0.16
1.72
±0.21
0.06
±0.04
0.98
±0.98
1.53
±0.20
Rifle, Colorado - RICO
Acetaldehyde
Benzene
1,3 -Butadiene
24/24
60/60
38/60
NA
1.73
±0.41
0.17
±0.08
1.50
±0.23
1.16
±0.21
0.03
±0.03
1.72
±0.36
1.27
±0.20
0.07
±0.03
1.21
±0.33
1.71
±0.47
0.35
±0.17
NA
1.46
±0.17
0.16
±0.06
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line for GPCO are presented in ng/m3
for ease of viewing.
7-36
-------
Table 7-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Colorado Monitoring Sites (Continued)
Pollutant
Ethylbenzene
Formaldehyde
#of
Measured
Detections
vs. # of
Samples
60/60
24/24
1st
Quarter
Average
(Ug/m3)
0.44
±0.10
NA
2nd
Quarter
Average
(Ug/m3)
0.39
±0.08
1.69
±0.23
3rd
Quarter
Average
(Ug/m3)
2.42
±3.29
2.07
±0.39
4th
Quarter
Average
(Ug/m3)
0.44
±0.16
1.65
±0.37
Annual
Average
(Ug/m3)
0.95
±0.86
NA
Rulison, Colorado - RUCO
Acetaldehyde
Benzene
1,3 -Butadiene
Formaldehyde
18/18
40/40
8/40
18/18
NA
1.86
±0.33
0.04
±0.03
NA
1.18
±0.24
1.47
±0.30
0.01
±0.01
1.35
±0.19
1.27
±0.61
1.55
±0.32
0.01
±0.01
1.57
±0.54
NA
NA
NA
NA
NA
1.62
±0.18
0.02
±0.01
NA
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line for GPCO are presented in ng/m3
for ease of viewing.
Observations for GPCO from Table 7-5 include the following:
• The pollutants with the highest annual average concentrations by mass are
formaldehyde (2.78 ± 0.23 |ig/m3), acetaldehyde (2.00 ± 0.20 |ig/m3), and benzene
(1.40 ± 0.16 |ig/m3). These are also the only pollutants with annual average
concentrations greater than 1 |ig/m3.
• The confidence intervals associated with the quarterly average concentrations of
acrylonitrile are equal to or greater than the quarterly averages themselves (except for
the fourth quarter, where it was not detected). This pollutant was detected only eight
times in 2010 and its measurements ranged from 0.087 to 0.300 |ig/m3. Thus, a large
number of zeros were substituted within the calculations, leading to a higher level of
variability within the average concentrations, which is reflected in the confidence
intervals.
• Formaldehyde concentrations were highest during third quarter of the year.
• Tetrachloroethylene's fourth quarter average concentration is the highest of the four
quarterly averages. A review of the data shows that three concentrations greater than
1 |ig/m3 were measured at GPCO, two in December and one in January. Only 13
concentrations of tetrachloroethylene above 1 |ig/m3 were measured among all NMP
sites sampling this pollutant. Further, the highest tetrachloroethylene concentration
among all sites was measured at GPCO (1.35 |ig/m3).
• Concentrations of naphthalene appear highest during the fourth quarter of the year. A
review of the data shows that of the 10 measurements of naphthalene greater than
7-37
-------
200 ng/m3, six were measured during the fourth quarter (three were measured in the
third quarter and one in the first quarter.
• Average benzo(a)pyrene concentrations for the first and fourth quarters of 2010 are
significantly higher than the average concentrations for the other two quarters and
have large confidence intervals associated with them. A review of the data shows that
the 23 concentration of this pollutant greater than 0.01 ng/m3 were all measured
during the first and fourth quarters of 2010. The remaining nine measured detections
were spread across the quarters. The three measurements of benzo(a)pyrene greater
than 1 ng/m3 were among the 10 highest concentrations measured by all NMP sites
sampling this pollutant.
• The third quarter average concentration of acenaphthene is higher than the other
quarterly averages and has a relatively large confidence interval associated with it.
The two highest concentrations of this pollutant were measured on
September 17, 2010 (38.7 ng/m3) and September 29, 2010 (22.4 ng/m3). The
September 17, 2010 measurement is the fifth highest concentration of this pollutant
among all NMP sites sampling acenaphthene.
Observations for the Garfield County sites from Table 7-5 include the following:
• Annual average concentrations for BMCO could not be calculated due to the
abbreviated sampling duration; however, fourth quarter averages are presented.
Annual average concentrations for the carbonyl compound pollutants of interest could
not be calculated for BRCO and RICO because these sites did not meet the quarterly
completeness criteria. In addition, annual average concentrations for the carbonyl
compound pollutants of interest could not be calculated for RUCO due to a
combination of an abbreviated sampling period and not meeting the quarterly
completeness criteria. However, Appendices K and L provide the pollutant-specific
average concentration for all valid samples collected over the entire sample period for
each site.
• Of the SNMOC, benzene has the highest annual average concentrations by mass for
each of the Garfield County sites. Annual average concentrations of benzene ranged
from 1.10±0.18 |ig/m3 for BRCO to 1.72 ±0.21 |ig/m3 for PACO. While PACO's
benzene concentrations were steady across 2010, the quarterly benzene
concentrations for the other sites show more variability.
• The fourth quarter average concentrations of 1,3-butadiene for PACO and RICO are
much higher than the other quarterly averages of this pollutant and have relatively
large confidence intervals associated with them, indicating the likely influence of
outliers. The highest 1,3-butadiene concentrations measured among five of the six
Colorado monitoring sites were all measured in December (the exception being
RUCO, but this site was not sampling in December). Each site's maximum
1,3-butadiene concentration was measured in the same two week stretch between
December 11, 2010 and December 23, 2010. Among other NMP sites sampling this
pollutant, three of RICO's 1,3-butadiene concentrations and one of PACO's rank
among the 10 highest 1,3-butadiene concentrations program-wide.
7-38
-------
• The third quarter average concentration of ethylbenzene for PACO and RICO are
significantly higher than the other quarterly averages of this pollutant and have large
confidence intervals associated with them, indicating the likely influence of outliers.
Both sites measured this pollutant above 25 |ig/m3 within one sample day of each
other (26.7 |ig/m3 on August 12, 2010 for PACO and 25.7 |ig/m3 on August 18, 2010
for RICO). These two concentrations represent the second and third highest
measurements of ethylbenzene among all sites sampling this pollutant, behind only
UCSD. BRCO and RUCO, for which ethylbenzene was not a pollutant of interest,
measured their maximum ethylbenzene concentrations on August 18, 2010 and
August 6, 2010, respectively. Program-wide, the 10 highest concentrations of this
pollutant were measured during the third quarter of 2010.
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for the Colorado
sites from those tables include the following:
• As shown in Tables 4-9 through 4-12, the annual average concentrations for GPCO
for 12 pollutants are among the 10 highest average concentrations for all NMP sites.
• GPCO appears frequently in Table 4-11 for the PAHs. GPCO has the highest annual
concentration of naphthalene, the second highest concentration of benzo(a)pyrene, the
third highest annual average of acenaphthene, and the fourth highest annual average
of fluorene.
• As shown in Table 4-9, the five of the six Colorado sites account for five of the 10
highest daily average concentrations of benzene. PACO and RICO rank first and
second among the highest annual average concentrations of ethylbenzene while RICO
and GPCO rank third and fifth among the highest annual average concentrations of
1,3-butaidene. GPCO also has the second highest annual concentration of
tetrachloroethylene, behind only PXSS.
7.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for benzene and 1,3-butadiene
were created for each of the Colorado sites. Box plots for acetaldehyde and formaldehyde were
created for the Colorado sites where annual averages could be calculated (GPCO and PACO).
Box plots were also created for benzo(a)pyrene, hexavalent chromium, and naphthalene for
GPCO. Figures 7-26 through 7-32 overlay the sites' minimum, annual average, and maximum
concentrations onto the program-level minimum, first quartile, average, median, third quartile,
and maximum concentrations, as described in Section 3.5.3.
7-39
-------
Figure 7-26. Program vs. Site-Specific Average Acetaldehyde Concentration
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
o
Figure 7-27a. Program vs. Site-Specific Average Benzene (Method TO-15) Concentration
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
7-40
-------
Figure 7-27b. Program vs. Site-Specific Average Benzene (SNMOC) Concentration
-o
-o
2 2.5
Concentration (|ig/m3)
Program
Site:
: IstQuartile
Site Average
o
2ndQuartile SrdQuartile 4thQuartile Aw
Site Minimum/Maximum
rage
Figure 7-28. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
/-tt \ Program Max Concentration = 42.7 ng/m3
f
0
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Ave
• • • D
Site: Site Average Site Minimum/Maximum
o —
1
'rage
.8 2
7-41
-------
Figure 7-29a. Program vs. Site-Specific Average 1,3-Butadiene (Method TO-15)
Concentration
•n-
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o —
Figure 7-29b. Program vs. Site-Specific Average 1,3-Butadiene (SNMOC) Concentration
•
1
1 1 1 1
1
?
lc
1
1 1 1 1
f ; ;
1.5
Concentration (ug/m3)
Program: IstQuartile
•
Site:
Site Average
O
2ndQuartile SrdQuartile 4thQuartile Ave
• n
Site Minimum/Maximum
?rage
7-42
-------
Figure 7-30. Program vs. Site-Specific Average Formaldehyde Concentration
E
N
1 1 1 1 1 1 1 1 1
20 25 30 35
Concentration (|ig/m3)
Program: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Average
• •GDI
Site: Site Average Site Minimum/Maximum
o —
Figure 7-31. Program vs. Site-Specific Average Hexavalent Chromium Concentration
-O - k
I
I Program Max Concentration = 3.51 ng/m3 j
0.3 0.45
Concentration (ng/m3)
Program: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 7-32. Program vs. Site-Specific Average Naphthalene Concentration
1.
Fh i i
i i i i
600 800
Concentration (ng/m3)
Program: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Average
Site: Site Average Site Minimum/Maximum
o
7-43
-------
Observations from Figures 7-26 through 7-32 include the following:
• Figure 7-26 shows that GPCO's annual average acetaldehyde concentration is just
greater than the program-level average while PACO's annual average
concentration is well below the program-level average. The maximum
acetaldehyde concentration for GPCO is more than twice the maximum
concentration measured at PACO, although both were well below the maximum
concentration measured across the program.
• Figure 7-27a presents the benzene concentration for GPCO compared to the
benzene concentrations measured across the program for NMP sites sampling
VOC with Method TO-15; Figure 7-27b presents the annual average benzene
concentrations for the Garfield County sites compared to the benzene
concentrations measured across the program for NMP sites sampling SNMOC.
The box plots are presented this way to correspond with Tables 4-1 through 4-6 in
Section 4.1, as discussed in Section 3.5.3.
• Figure 7-27a shows that the annual average benzene concentration for GPCO is
greater than the program-level average, and is greater than the 75th percentile (or
third quartile) for the program. For the Garfield County sites sampling SNMOC,
the annual average for benzene is greater than the program-level average for all
four sites for which annual average benzene concentrations could be calculated, as
shown in Figure 7-27b. Note that the maximum benzene concentration measured
by sites sampling SNMOC was measured at BRCO. A similar concentration was
also measured at PACO. Among the Garfield County sites, the highest annual
average concentration of benzene was calculated for PACO, followed by RUCO,
RICO, and BRCO. An annual average concentration could not be calculated for
BMCO.
• Figure 7-28 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
GPCO is greater than the program-level average concentration and is greater than
the 75th percentile (or third quartile) for the program. Figure 7-28 also shows that
the maximum concentration measured at GPCO is well below the maximum
concentration measured across the program. Several non-detects of
benzo(a)pyrene were measured at GPCO.
• Similar to the box plots for benzene, Figure 7-29a presents the annual average
concentration of 1,3-butadiene for GPCO compared to the 1,3-butadiene
concentrations measured across the program for NMP sites sampling VOC with
Method TO-15; Figure 7-29b presents the annual average 1,3-butadiene
concentrations for the Garfield County sites compared to the 1,3-butadiene
concentrations measured across the program for NMP sites sampling SNMOC.
7-44
-------
• Figure 7-29a shows that GPCO's annual 1,3-butadiene concentration is greater
than the program average and similar to other pollutants, this annual average is
greater than the 75th percentile (or third quartile) for the program. For the Garfield
County sites sampling SNMOC, RICO's annual average 1,3-butadiene
concentration is greater than the program-level average, as shown in
Figure 7-29b, while PACO's annual average is equal to the program-level
average, and RUCO and BRCO's are below the program-level average. Note that
the first and second quartiles for the program-level are both zero, and thus not
shown in Figure 7-29b, indicating that at least half of the 1,3-butadiene
concentrations measured by sites sampling SNMOC were non-detects. An annual
average concentration could not be calculated for BMCO.
• Figure 7-30 shows that GPCO's annual average formaldehyde concentration is
just greater than the program-level average while PACO's annual average
concentration is less than the program-level average, similar to acetaldehyde
concentrations for these sites. While it appears that the range of concentrations
measured at these sites is relatively small, the range of measurements at the
program-level is quite large.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 7-31 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 7-31 shows the annual average concentration of hexavalent chromium for
GPCO is less than half the program-level average. GPCO's annual average
concentration is also less than the program-level median (or 50th percentile).
• Figure 7-32 shows that the annual naphthalene average for GPCO is greater than
the program-level average concentration. As discussed in the previous section, the
annual average naphthalene concentration for GPCO is the highest among all
NMP sites sampling this pollutant. However, the maximum naphthalene
concentration measured at GPCO was well below the program-level maximum
concentration.
7.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. While the Garfield County sites have not sampled continuously for 5 years as part
of the NMP, GPCO has sampled carbonyl compounds and VOC since 2004 and hexavalent
chromium since 2005. Thus, Figures 7-33 through 7-37 present the 3-year rolling statistical
metrics for acetaldehyde, benzene, 1,3-butadiene, formaldehyde, and hexavalent chromium for
7-45
-------
GPCO, respectively. The statistical metrics presented for assessing trends include the
substitution of zeros for non-detects.
Observations from Figure 7-33 for acetaldehyde measurements at GPCO include the
following:
• The maximum acetaldehyde concentration was measured during the 2004-2006 time
frame, specifically 2004. The maximum concentrations measured in subsequent time
periods were significantly lower. The two highest acetaldehyde concentrations
(93 and 55 |ig/m3) were measured in 2004 and the six highest acetaldehyde
concentrations (ranging from 93 |ig/m3 to 6.35 |ig/m3) were all measured in 2004 and
2005.
• The 5th and 95th percentiles, the median, and the average concentrations exhibit
relatively little variation over time if the 2004-2006 time frame is excluded.
• The average concentrations show little variability after the 2004-2006 time frame and
ranged from 2.47 |ig/m3 (2008-2010) to 2.73 |ig/m3 (2007-2009).
• Although difficult to discern in Figure 7-33, the rolling average and median values
became more similar to each other over the periods shown. This indicates decreasing
variability in the central tendency of acetaldehyde concentrations measured over the
periods shown.
Figure 7-33. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at GPCO
o 40
2006-2008
Three-Year Period
• 5th Pei(entile - Minimum - Median - Maximum • 95th Percentile
• -- Average
7-46
-------
Figure 7-34. Three-Year Rolling Statistical Metrics for Benzene Concentrations
Measured at GPCO
l«
2004-2.006 2005-2007
2006-2008
Three- Year Period
2007-2009 2008-2010
5th Percentile — Minimum - Median - Maximum • 95th Perc en tile ...».- Averag
Figure 7-35. Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations
Measured at GPCO
2004-2006 2005-2007
2006-2008
Three-Year Period
2007-2009 2008-2010
• 5th Percentile — Minimum — Median — Maximum • 95th Pertentile ...4.. Average
7-47
-------
Figure 7-36. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at GPCO
1,
1
o
<-J
^
_
I T T
^^^^^_ ^^^^^_
HH'
i — * — i i i
2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three- Year Period
• 5th Percentile — Minimum - Median — Maximum • 95th Percentile ...+.. Average
Figure 7-37. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at GPCO
0.7
0,6
I
j
" 0.1
O.i
T
1
B^
200S 2.W/
| • JlhPetunlife
_^^_ i '
—^ ""^ H
^Otib 2008 200 /
Ttiree-V«ar Period
1 1
2005> 2008 2030
— Mitiliiiimi - Mcdi
-------
Observations from Figure 7-34 for benzene measurements at GPCO include the
following:
• The maximum benzene concentration was measured on December 11, 2004. The
maximum concentrations measured in subsequent years were much lower until
July 9, 2009, when a similar concentration was measured.
• The 5th and 95th percentiles and the median have decreased slightly over time. The
rolling average decreased as well, but increased slightly for 2007-2009 and
2008-2010, primarily as a result of the high concentration measured in 2009 (if this
concentration were removed from consideration, the average concentrations would
continue the slight, although not statistically significant, decreasing trend).
• The minimum concentration was greater than zero for all 3-year time periods,
indicating that there were no non-detects reported for benzene over the period of
sampling.
Observations from Figure 7-35 for 1,3-butadiene measurements at GPCO include the
following:
• Similar to benzene, the maximum 1,3-butadiene concentration was measured on
December 11, 2004. The maximum concentrations measured in subsequent time
periods were lower.
• The rolling average concentrations appear to have a slight decreasing trend; however,
confidence intervals calculated from the individual concentrations show that this
decrease is not statistically significant.
• In addition to the rolling average, the median and 95th percentile also exhibit a slight
decreasing trend in concentrations.
• Conversely, the 5th percentile increased from zero for 2006-2008 and beyond. The
number of non-detects, and subsequently zeros substituted for non-detects, decreased
from approximately 30 percent in 2004 and 2005, to eight percent in 2006, to none in
2007, 2008, and 2009, and three percent in 2010.
Observations from Figure 7-36 for formaldehyde measurements at GPCO include the
following:
• The trends graph for formaldehyde resembles the graph for acetaldehyde in that the
maximum formaldehyde concentration was measured in 2004. The three highest
concentrations of formaldehyde were measured on the same days as the three highest
acetaldehyde concentrations. The maximum concentrations in subsequent time
periods were significantly lower.
7-49
-------
• Even with decreasing maximum concentrations, the rolling average formaldehyde
concentrations (as well as several other statistical parameters) have a slight increasing
trend through the 2007-2009 time frame, then decrease slightly for 2008-2010.
• Although difficult to discern in Figure 7-36, the minimum concentration measured
increased for each time frame, doubling from 2007-2009 to 2008-2010.
Observations from Figure 7-37 for hexavalent chromium measurements at GPCO include
the following:
• The maximum hexavalent chromium concentration was measured on July 5, 2008
(0.685 ng/m3). Only two concentrations measured at GPCO are greater than
0.1 ng/m3, with the other being measured on August 9, 2006 (0.113 ng/m3), which is
the maximum concentration shown for the 2005-2007 time period.
• The rolling average concentrations of hexavalent chromium exhibit a slight
decreasing trend, although the confidence intervals calculated on the dataset are
relatively wide due, at least in part, to the maximum concentration. However, the
median concentrations also show a decreasing trend, and this parameter is influenced
less by outliers.
• Both the minimum concentration and 5th percentile for all 3-year periods shown are
zero, indicating the presence of non-detects. The percentage of non-detects increased
from 2006 through 2009, then decreased for 2010.
7.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at each
Colorado monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
7.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Colorado monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
for each site were compared to the acute MRL; the quarterly averages were compared to the
intermediate MRL; and the annual averages were compared to the chronic MRL. The results of
7-50
-------
this risk screening are summarized in Table 7-6. Where a quarterly or annual average exceeds the
applicable MRL, the concentration is bolded.
Observations about dichloromethane in Table 7-6 include the following:
• Dichloromethane was the only pollutant where a preprocessed daily measurement
and/or time-period average is greater than one or more of the MRL noncancer health
risk benchmarks. Although this pollutant is not a pollutant of interest for any of the
Colorado sites, its highest concentration is among the highest concentrations
measured for any of the pollutants sampled under the NMP.
• One out of 59 measured detections of dichloromethane for GPCO is greater than the
ATSDR acute MRL for this pollutant (2,000 |ig/m3). This concentration was
measured on October 17, 2010. While this measurement (5,256 |ig/m3) was the
highest concentration of this pollutant measured among all NMP sites sampling VOC,
it was not the only concentration of dichloromethane to exceed the acute MRL, as
discussed in Section 3.3.
• While none of the quarterly average concentrations of dichloromethane were greater
than the ATSDR intermediate MRL of 1,000 |ig/m3, it is easy to see from Table 7-6
during which quarter the maximum measurement of dichloromethane was measured.
The fourth quarter average dichloromethane concentration is two orders of magnitude
higher than the other quarterly averages and its confidence interval is twice as high as
the average itself. The second highest concentration of dichloromethane measured at
GPCO was 67.9 |ig/m3 and was measured during the first quarter of 2010. However,
the median dichloromethane concentration for this site is less than 1 |ig/m3.
• The annual average concentration of dichloromethane for GPCO
(91.68 ± 176.67 |ig/m3) was well below the ATSDR chronic MRL for this pollutant
(1,000 |ig/m3). Note that if the maximum concentration of dichloromethane were
removed from the calculation, the annual average concentration would be 2.64 |ig/m3.
7-51
-------
Table 7-6. Noncancer Risk Screening Summary for the Colorado Monitoring Sites
Pollutant
Acute
ATSDR
Acute
MRL1
(Hg/m3)
#of
Concentrations
>MRL
#of
Measured
Detections
Intermediate
ATSDR
Intermediate
MRL1
(Hg/m3)
1st
Quarter
Average
(jig/m3)
2nd
Quarter
Average
(jig/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Chronic
ATSDR
Chronic
MRL1
(Hg/m3)
Annual
Average
(jig/m3)
Grand Junction, Colorado - GPCO
Dichloromethane
2,000
1
59
1,000
5.07
±9.24
1.74
±2.20
1.23
±0.87
377.75
± 775.46
1,000
91.68
± 176.67
Reflects the use of one significant digit for the MRLs
to
-------
For the pollutants whose concentrations are greater than their respective ATSDR acute
MRL noncancer health risk benchmark(s), the concentrations were further examined by
developing pollution roses for these pollutants. A pollution rose is a plot of concentration vs.
wind speed and wind direction, as described in Section 3.5.5.1. Figure 7-38 is the
dichloromethane pollution rose for GPCO.
Observations from the Figure 7-38 include the following:
• There was only one measured detection that was greater than the ATSDR acute MRL
(2,000 |ig/m3) for dichloromethane (shown in orange).
• The concentration greater than the ATSDR acute MRL was measured on a day with
light winds blowing from the southeast. However, there were many measurements of
dichloromethane that were much lower and measured on days with average winds
from the southeast.
• The three next highest concentrations, ranging from 16.6 |ig/m3 to 67.9 |ig/m3, were
measured on days with winds from the north, as shown in Figure 7-38.
7-53
-------
Figure 7-38. Dichloromethane Pollution Rose for GPCO
360 0
...-
315,/' ,-""""
/ / V""
?~ft i '' £ ( i ^
"kits i 14 U2 10 8 :6 V* \2 ,--''"'
1 \ v X
v V \ \ N -x' v-
\ \ • "•• •-'
\ \ \ \ \ X\
\ \ \ \ X
\ \ \ >< ^ --
\ /x """""---,
225 \ "---,
\ 45
xs
\ /'* \
"'•*-, -*'' '*- \ \ \
,-'
NX'*' '"- *'' '• \
""""-v--"" \ \ \ \ \
>-,. A \ \ \ \ I
cV;- •> \
~-b ''. ' o
/ o °° i ' 90
*\/V / / 1 I I
,---"'' \ /o /III
x / / / /
0 ... > \ / ° / / /
.--""' ''X /
^- — ' ..-'"' \
...'-•"' /' "N\ /"
.,.---' .,-' ATSDR MEL- 2,000 iig/m3,
_,.-•••'" s which corresponds to the upper
,-.--"" ..'•'' end of the 100- ^ 000 u»iii3 (or
..--•'" vellow) concentrationrange
„ "
o 0-10 fig iu3
ISO
010-2,000 fig in3
O>2,000 iig/mJ
-------
7.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Colorado monitoring sites and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 7-7, where applicable.
Observations for GPCO from Table 7-7 include the following:
• Formaldehyde, acetaldehyde, and benzene have the highest annual average
concentrations for GPCO.
• Formaldehyde also has the highest cancer risk approximation (36.12 in-a-million) for
this site. Benzene has the second highest cancer risk approximation
(10.88 in-a-million), while naphthalene has the third highest cancer risk
approximation (5.00 in-a-million).
• None of the pollutants of interest for GPCO have noncancer risk approximations
greater than 1.0. Formaldehyde has the highest noncancer risk approximation (0.28).
Observations for the Garfield County sites from Table 7-7 include the following:
• Annual averages, and thus cancer and noncancer surrogate risk approximations, could
not be calculated for BMCO due to the short sampling duration.
• Annual averages, and thus cancer and noncancer surrogate risk approximations, could
not be calculated for acetaldehyde and formaldehyde for BRCO, RICO, and RUCO.
This is due to the completeness criteria combined with the l-in-12 day sampling
schedule for these pollutants. Formaldehyde has the highest cancer risk
approximation for P AGO among its pollutants of interest.
• For all sites except PACO, benzene's cancer risk approximation is the highest among
each site's pollutants of interest. Benzene's cancer risk approximations range from
8.55 in-a-million (BRCO) to 13.43 in-a-million (PACO). PACO's benzene cancer
risk approximation is the second highest benzene cancer risk approximation
compared to other NMP sites.
• None of the noncancer risk approximations calculated for the Garfield County sites
were greater than 1.0. The highest noncancer risk approximation was calculated for
PACO (for formaldehyde, 0.16).
7-55
-------
Table 7-7. Cancer and Noncancer Surrogate Risk Approximations for the Colorado
Monitoring Sites
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs.
# of Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Grand Junction, Colorado - GPCO
Acenaphthene3
Acetaldehyde
Acrylonitrile
Benzene
Benzo(a)pyrenea
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Ethylbenzene
Formaldehyde
Hexavalent Chromium3
Naphthalene3
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.000088
0.0000022
0.000068
0.0000078
0.00176
0.00003
0.000006
0.000026
0.0000025
0.000013
0.012
0.000034
0.00000026
0.0000048
0.0000088
0.009
0.002
0.03
0.002
0.1
0.098
2.4
1
0.0098
0.0001
0.003
0.04
0.002
0.1
57/57
61/61
8/59
59/59
32/57
58/59
59/59
54/59
7/59
59/59
61/61
35/58
57/57
59/59
19/59
2/59
0.01
±<0.01
2.00
±0.20
0.02
±0.01
1.40
±0.16
<0.01
±<0.01
0.14
±0.02
0.53
±0.05
0.09
±0.01
0.01
±0.01
0.50
±0.07
2.78
±0.23
0.01
±0.01
0.15
±0.02
0.39
±0.07
0.03
±0.01
O.01
±O.01
0.64
4.40
1.35
10.88
0.33
4.07
3.17
0.24
1.25
36.12
0.14
5.00
0.10
0.13
O.01
0.22
0.01
0.05
0.07
0.01
0.01
O.01
0.01
0.28
O.01
0.05
0.01
0.01
O.01
Battlement Mesa, Colorado - BMCO
Acetaldehyde
Benzene
1,3 -Butadiene
Formaldehyde
0.0000022
0.0000078
0.00003
0.000013
0.009
0.03
0.002
0.0098
7/7
18/18
8/18
7/7
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
— = a Cancer URE or Noncancer RfC is not available.
NA = Not available due to the criteria for calculating an annual average.
3 For the annual average concentration of this pollutant in ng/m3, refer to Table 7-5.
7-56
-------
Table 7-7. Cancer and Noncancer Surrogate Risk Approximations for the Colorado
Monitoring Sites (Continued)
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs.
# of Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Silt, Colorado - BRCO
Acetaldehyde
Benzene
1,3 -Butadiene
Formaldehyde
0.0000022
0.0000078
0.00003
0.000013
0.009
0.03
0.002
0.0098
17/17
61/61
6/61
17/17
NA
1.10
±0.18
0.01
±0.01
NA
NA
8.55
0.31
NA
NA
0.04
0.01
NA
Parachute, Colorado - PACO
Acetaldehyde
Benzene
1,3 -Butadiene
Ethylbenzene
Formaldehyde
0.0000022
0.0000078
0.00003
0.0000025
0.000013
0.009
0.03
0.002
1
0.0098
28/28
58/58
22/58
58/58
28/28
0.93
±0.16
1.72
±0.21
0.06
±0.04
0.98
±0.98
1.53
±0.20
2.04
13.43
1.91
2.46
19.95
0.10
0.06
0.03
<0.01
0.16
Rifle, Colorado - RICO
Acetaldehyde
Benzene
1,3 -Butadiene
Ethylbenzene
Formaldehyde
0.0000022
0.0000078
0.00003
0.0000025
0.000013
0.009
0.03
0.002
1
0.0098
24/24
60/60
38/60
60/60
24/24
NA
1.46
±0.17
0.16
±0.06
0.95
±0.86
NA
NA
11.38
4.71
2.38
NA
NA
0.05
0.08
0.01
NA
Rulison, Colorado - RUCO
Acetaldehyde
Benzene
1,3 -Butadiene
Formaldehyde
0.0000022
0.0000078
0.00003
0.000013
0.009
0.03
0.002
0.0098
18/18
40/40
8/40
18/18
NA
1.62
±0.18
0.02
±0.01
NA
NA
12.65
0.51
NA
NA
0.05
0.01
NA
— = a Cancer URE or Noncancer RfC is not available.
NA = Not available due to the criteria for calculating an annual average.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 7-5.
7-57
-------
7.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 7-8 and 7-9 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 7-8 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 7-9 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 7-8 and 7-9 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. Further, the cancer and noncancer surrogate risk approximations based on each site's
annual averages are limited to those pollutants for which each respective monitoring site
sampled. As discussed in Section 7.3, GPCO sampled for VOC, carbonyl compounds, PAH, and
hexavalent chromium; the Garfield County sites sampled for SNMOC and carbonyl compounds
only. In addition, the cancer and noncancer surrogate risk approximations are limited to those
pollutants with enough data to meet the criteria for annual averages to be calculated. A more in-
depth discussion of this analysis is provided in Section 3.5.5.3.
7-58
-------
Table 7-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Colorado Monitoring Sites
VO
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Grand Junction, Colorado (Mesa County) - GPCO
Benzene
Formaldehyde
Acetaldehyde
Ethylbenzene
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
Tetrachloroethylene
POM, Group 6
140.94
112.12
45.52
41.56
14.18
7.63
6.38
2.02
1.20
0.20
Formaldehyde
Benzene
1,3 -Butadiene
POM, Group 3
Naphthalene
POM, Group 2b
Arsenic, PM
Hexavalent Chromium, PM
Ethylbenzene
Acetaldehyde
1.46E-03
1.10E-03
4.25E-04
3.93E-04
2.59E-04
1.78E-04
1.58E-04
1.37E-04
1.04E-04
l.OOE-04
Formaldehyde
Benzene
Naphthalene
Acetaldehyde
1,3 -Butadiene
Carbon Tetrachloride
Acrylonitrile
Ethylbenzene
Acenaphthene
Benzo(a)pyrene
36.12
10.88
5.00
4.40
4.07
3.17
1.35
1.25
0.64
0.33
Battlement Mesa, Colorado (Garfield County) - BMCO
Formaldehyde
Benzene
Acetaldehyde
Ethylbenzene
1,3 -Butadiene
Naphthalene
POM, Group 2b
Tetrachloroethylene
Dichloromethane
POM, Group 6
324.22
269.17
79.93
27.66
16.04
6.02
2.69
1.34
0.74
0.32
Formaldehyde
Benzene
POM, Group 3
1,3 -Butadiene
POM, Group 2b
Naphthalene
Acetaldehyde
POM, Group 5a
Ethylbenzene
POM, Group 6
4.21E-03
2.10E-03
8.97E-04
4.81E-04
2.37E-04
2.05E-04
1.76E-04
l.OOE-04
6.92E-05
5.61E-05
-------
Table 7-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Colorado Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Silt, Colorado (Garfield County) - BRCO
Formaldehyde
Benzene
Acetaldehyde
Ethylbenzene
1,3 -Butadiene
Naphthalene
POM, Group 2b
Tetrachloroethylene
Dichloromethane
POM, Group 6
324.22
269.17
79.93
27.66
16.04
6.02
2.69
1.34
0.74
0.32
Formaldehyde
Benzene
POM, Group 3
1,3 -Butadiene
POM, Group 2b
Naphthalene
Acetaldehyde
POM, Group 5a
Ethylbenzene
POM, Group 6
4.21E-03
2.10E-03
8.97E-04
4.81E-04
2.37E-04
2.05E-04
1.76E-04
l.OOE-04
6.92E-05
5.61E-05
Benzene
1,3 -Butadiene
8.55
0.31
Parachute, Colorado (Garfield County) - PACO
Formaldehyde
Benzene
Acetaldehyde
Ethylbenzene
1,3 -Butadiene
Naphthalene
POM, Group 2b
Tetrachloroethylene
Dichloromethane
POM, Group 6
324.22
269.17
79.93
27.66
16.04
6.02
2.69
1.34
0.74
0.32
Formaldehyde
Benzene
POM, Group 3
1,3 -Butadiene
POM, Group 2b
Naphthalene
Acetaldehyde
POM, Group 5a
Ethylbenzene
POM, Group 6
4.21E-03
2.10E-03
8.97E-04
4.81E-04
2.37E-04
2.05E-04
1.76E-04
l.OOE-04
6.92E-05
5.61E-05
Formaldehyde
Benzene
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
19.95
13.43
2.46
2.04
1.91
-------
Table 7-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Colorado Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Cancer Risk
Approximation
Pollutant (in-a-million)
Rifle, Colorado (Garfield County) - RICO
Formaldehyde
Benzene
Acetaldehyde
Ethylbenzene
1,3 -Butadiene
Naphthalene
POM, Group 2b
Tetrachloroethylene
Dichloromethane
POM, Group 6
324.22
269.17
79.93
27.66
16.04
6.02
2.69
1.34
0.74
0.32
Formaldehyde
Benzene
POM, Group 3
1,3 -Butadiene
POM, Group 2b
Naphthalene
Acetaldehyde
POM, Group 5a
Ethylbenzene
POM, Group 6
4.21E-03
2.10E-03
8.97E-04
4.81E-04
2.37E-04
2.05E-04
1.76E-04
l.OOE-04
6.92E-05
5.61E-05
Benzene 11.38
1,3-Butadiene 4.71
Ethylbenzene 2.38
Rulison, Colorado (Garfield County) - RUCO
Formaldehyde
Benzene
Acetaldehyde
Ethylbenzene
1,3 -Butadiene
Naphthalene
POM, Group 2b
Tetrachloroethylene
Dichloromethane
POM, Group 6
324.22
269.17
79.93
27.66
16.04
6.02
2.69
1.34
0.74
0.32
Formaldehyde
Benzene
POM, Group 3
1,3 -Butadiene
POM, Group 2b
Naphthalene
Acetaldehyde
POM, Group 5a
Ethylbenzene
POM, Group 6
4.21E-03
2.10E-03
8.97E-04
4.81E-04
2.37E-04
2.05E-04
1.76E-04
l.OOE-04
6.92E-05
5.61E-05
Benzene 12.65
1,3-Butadiene 0.51
-------
Table 7-9. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Colorado Monitoring Sites
to
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Grand Junction, Colorado (Mesa County) - GPCO
Toluene
Xylenes
Benzene
Formaldehyde
Methanol
Hexane
Acetaldehyde
Ethylbenzene
Hydrochloric acid
1,3 -Butadiene
240.58
188.54
140.94
112.12
88.05
50.96
45.52
41.56
27.54
14.18
Acrolein
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Benzene
Naphthalene
Arsenic, PM
Xylenes
Hydrochloric acid
Manganese, PM
553,576.79
11,440.45
7,087.62
5,058.01
4,698.01
2,543.73
2,443.88
1,885.37
1,376.84
1,158.03
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Naphthalene
Benzene
Trichloroethylene
Acrylonitrile
Tetrachloroethylene
Carbon Tetrachloride
Chloroform
0.28
0.22
0.07
0.05
0.05
0.01
0.01
0.01
0.01
<0.01
Battlement Mesa, Colorado (Garfield County) - BMCO
Toluene
Formaldehyde
Xylenes
Benzene
Hexane
Acetaldehyde
Methanol
Acrolein
Ethylbenzene
1,3 -Butadiene
419.91
324.22
321.33
269.17
82.14
79.93
60.71
40.42
27.66
16.04
Acrolein
Formaldehyde
Benzene
Acetaldehyde
1,3 -Butadiene
Xylenes
Naphthalene
Cyanide Compounds, gas
Lead, PM
Arsenic, PM
2,021,060.94
33,083.40
8,972.23
8,880.90
8,021.37
3,213.31
2,008.12
988.69
430.70
233.88
-------
Table 7-9. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Colorado Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Silt, Colorado (Garfield County) - BRCO
Toluene
Formaldehyde
Xylenes
Benzene
Hexane
Acetaldehyde
Methanol
Acrolein
Ethylbenzene
1,3 -Butadiene
419.91
324.22
321.33
269.17
82.14
79.93
60.71
40.42
27.66
16.04
Acrolein
Formaldehyde
Benzene
Acetaldehyde
1,3 -Butadiene
Xylenes
Naphthalene
Cyanide Compounds, gas
Lead, PM
Arsenic, PM
2,021,060.94
33,083.40
8,972.23
8,880.90
8,021.37
3,213.31
2,008.12
988.69
430.70
233.88
Benzene
1,3 -Butadiene
0.04
0.01
Parachute, Colorado (Garfield County) - PACO
Toluene
Formaldehyde
Xylenes
Benzene
Hexane
Acetaldehyde
Methanol
Acrolein
Ethylbenzene
1,3 -Butadiene
419.91
324.22
321.33
269.17
82.14
79.93
60.71
40.42
27.66
16.04
Acrolein
Formaldehyde
Benzene
Acetaldehyde
1,3 -Butadiene
Xylenes
Naphthalene
Cyanide Compounds, gas
Lead, PM
Arsenic, PM
2,021,060.94
33,083.40
8,972.23
8,880.90
8,021.37
3,213.31
2,008.12
988.69
430.70
233.88
Formaldehyde
Acetaldehyde
Benzene
1,3 -Butadiene
Ethylbenzene
0.16
0.10
0.06
0.03
0.01
-------
Table 7-9. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Colorado Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Rifle, Colorado (Garfield County) - RICO
Toluene
Formaldehyde
Xylenes
Benzene
Hexane
Acetaldehyde
Methanol
Acrolein
Ethylbenzene
1,3 -Butadiene
419.91
324.22
321.33
269.17
82.14
79.93
60.71
40.42
27.66
16.04
Acrolein
Formaldehyde
Benzene
Acetaldehyde
1,3 -Butadiene
Xylenes
Naphthalene
Cyanide Compounds, gas
Lead, PM
Arsenic, PM
2,021,060.94
33,083.40
8,972.23
8,880.90
8,021.37
3,213.31
2,008.12
988.69
430.70
233.88
1,3-Butadiene 0.08
Benzene 0.05
Ethylbenzene <0.01
Rulison, Colorado (Garfield County) - RUCO
Toluene
Formaldehyde
Xylenes
Benzene
Hexane
Acetaldehyde
Methanol
Acrolein
Ethylbenzene
1,3 -Butadiene
419.91
324.22
321.33
269.17
82.14
79.93
60.71
40.42
27.66
16.04
Acrolein
Formaldehyde
Benzene
Acetaldehyde
1,3 -Butadiene
Xylenes
Naphthalene
Cyanide Compounds, gas
Lead, PM
Arsenic, PM
2,021,060.94
33,083.40
8,972.23
8,880.90
8,021.37
3,213.31
2,008.12
988.69
430.70
233.88
Benzene 0.05
1,3-Butadiene 0.01
-------
Observations from Table 7-8 include the following:
• Benzene, formaldehyde, and acetaldehyde are the highest emitted pollutants with
cancer UREs in Mesa County. These same pollutants also topped the list for Garfield
County, although not in the same order. Note that the quantity emitted for each
pollutant was roughly twice as high in Garfield County than Mesa County.
• In Mesa County, the two pollutants with the highest toxicity-weighted emissions (of
the pollutants with cancer UREs) are formaldehyde and benzene, while 1,3-butadiene
and POM, Group 3 rank third and fourth. In Garfield County, the pollutants with the
highest toxicity-weighted emissions (of the pollutants with cancer UREs) are
formaldehyde, benzene, POM, Group 3, and 1,3-butadiene.
• Seven of the highest emitted pollutants also have the highest toxicity-weighted
emissions in Mesa County while eight of the highest emitted pollutants also have the
highest toxicity-weighted emissions in Mesa County.
• For GPCO, six of the 10 pollutants with the highest cancer risk approximations also
appear on both emissions-based lists for Mesa County. For each of the Garfield
County sites, all of the pollutants of interest listed with cancer risk approximations,
where they could be calculated, appear on both emissions-based lists for Garfield
County.
• POM, Group 2b is the eighth highest emitted "pollutant" in Mesa County and ranks
sixth for toxicity-weighted emissions. POM, Group 2b includes several PAH sampled
for at GPCO including acenaphthene, fluoranthene, fluorene, and perylene.
Acenaphthene has the ninth highest cancer risk approximation for GPCO.
• Benzo(a)pyrene is included in POM, Group 5a. The cancer risk approximation for
this pollutant ranks 10th for GPCO, but POM, Group 5a does not appear on either
emissions-based list (although its emissions rank 15th and its toxicity-weighted
emissions rank 11th in Mesa County).
• POM, Groups 2b, 3, 5a, and 6 appear on Garfield County's emissions-based lists.
PAH were not sampled at the Garfield County sites.
Observations from Table 7-9 include the following:
• Toluene is the highest emitted pollutant with a noncancer RfC in both Mesa and
Garfield Counties, although the emissions are higher in Garfield County. These two
counties share an additional eight pollutants on their lists of highest emitted pollutants
with noncancer RfCs.
• The two pollutants with the highest toxicity-weighted emissions (of the pollutants
with noncancer RfCs) for both counties are acrolein and formaldehyde. Although
acrolein was sampled for at GPCO, this pollutant was excluded from the pollutants of
interest designation, and thus subsequent risk screening evaluations, due to questions
about the consistency and reliability of the measurements, as discussed in Section 3.2.
7-65
-------
• Six of the 10 highest emitted pollutants in Mesa County also have the highest
toxi city-weighted emissions. Six of the 10 highest emitted pollutants in Garfield
County (including acrolein) also have the highest toxicity-weighted emissions.
• Formaldehyde, acetaldehyde, benzene, and 1,3-butadiene appear on all three lists for
GPCO. Additionally, naphthalene appears among the pollutants with the highest
noncancer risk approximations and highest toxicity-weighted emissions list, but this
pollutant is not among the highest pollutants emitted in Mesa County.
• With the exception of ethylbenzene, all of the pollutants of interest listed on the
noncancer risk approximations lists for the Garfield County sites also appear on both
emissions-based lists for Mesa County. Although ethylbenzene is one of the highest
emitted pollutants in Garfield County, it is not among the most toxic.
7.6 Summary of the 2010 Monitoring Data for the Sites in Colorado
Results from several of the data treatments described in this section include the
following:
»«» Nineteen pollutants failed at least one screen for GPCO, while the number of
pollutants failing screens for the Garfield County sites ranged from four to five.
»«» The pollutants with the highest annual average concentrations for GPCO were
formaldehyde, acetaldehyde, and benzene. These were also the only pollutants with
annual average concentrations greater than 1 jug/m . Benzene had the highest annual
average concentrations for each of the Garfield County sites.
»«» One preprocessed daily measurement ofdichloromethanefrom GPCO was greater
than its associated acute MRL noncancer health risk benchmark. None of the
quarterly or annual average concentrations of the pollutants of interest, where they
could be calculated, were greater than their associated intermediate or chronic MRL
noncancer health risk benchmarks.
7-66
-------
8.0 Site in the District of Columbia
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in Washington, D.C., and integrates these
concentrations with emissions, meteorological, and risk information. Data generated by sources
other than ERG are not included in the data analyses contained in this report. Readers are
encouraged to refer to Sections 1 through 4 and the glossary (Appendix P) for detailed
discussions and definitions regarding the various data analyses presented below.
8.1 Site Characterization
This section characterizes the Washington, D.C. monitoring site by providing
geographical and physical information about the location of the site and the surrounding area.
This information is provided to give the reader insight regarding factors that may influence the
air quality near the site and assist in the interpretation of the ambient monitoring measurements.
Figure 8-1 is a composite satellite image retrieved from ArcGIS Explorer showing the
monitoring site in its urban location. Figure 8-2 identifies point source emissions locations by
source category, as reported in the 2008 NEI for point sources. Note that only sources within
10 miles of the site are included in the facility counts provided in Figure 8-2. Thus, sources
outside the 10-mile radius have been grayed out, but are visible on the map to show emissions
sources outside the 10-mile boundary. A 10-mile boundary was chosen to give the reader an
indication of which emissions sources and emissions source categories could potentially have a
direct effect on the air quality at the monitoring site. Further, this boundary provides both the
proximity of emissions sources to the monitoring site as well as the quantity of such sources
within a given distance of the site. Table 8-1 describes the area surrounding the monitoring site
by providing supplemental geographical information such as land use, location setting, and
locational coordinates.
3-1
-------
Figure 8-1. Washington, B.C. (WADC) Monitoring Site
oo
to
-------
Figure 8-2. NEI Point Sources Located Within 10 Miles of WADC
^.Montgomery / Prince George's \
CountV / County ,
77-20'drW 77'15'0'W 77'1CrO*Vtf TT'S'ITW 77-fl'trW 76"55T3*W
Note: Due to facility density and collocation, the total facilities
• i displayed may not represent all facilities within the area of interest.
@ WADC NATTS site 10 rnile radius | j County boundary
Source Category Group (No. of Facilities)
ts Air-conditioning/Refrigeration (2)
-f Aircraft Operations (27)
I Asphalt Processing/Roofing Manufacturing (3)
* Electricity Generation via Combustion (3)
F Food Processing/Agriculture (1)
13 Hospital (5)
^ Institutional - school (10)
A Military Base/National Security (5)
? Miscellaneous Commercial/Industrial (18)
P Printing/Publishing (5)
* Transportation and Marketing of Petroleum Products (1)
8-3
-------
Table 8-1. Geographical Information for the Washington, D.C. Monitoring Site
Site
Code
WADC
AQS Code
11-001-0043
Location
Washington
County
District
Of
Columbia
Micro- or
Metropolitan
Statistical Area
Washington-
Arlington-
Alexandria, DC-
VA-MD-WVMSA
Latitude
and
Longitude
38.921847,
-77.013178
Land Use
Commercial
Location
Setting
Urban/City
Center
Additional Ambient Monitoring Information1
Arsenic, CO, VOC, SO2, NOy, NO, NO2, NOX,
PAMS, Carbonyl compounds, O3, Meteorological
parameters, PM10, PM25, PM10 Speciation, Black
carbon, PM Coarse, PM25 Speciation.
BOLD ITALICS = EPA-designated NATTS Site.
oo
-------
Figure 8-1 shows that the WADC monitoring site is located in an open field at the
southeast of end of the McMillan Water Reservoir in Washington, D.C. It is also located near
several heavily traveled roadways. The site is located in a commercial area, and is surrounded by
a hospital, a cemetery, and a university. As Figure 8-2 shows, WADC is surrounded
predominantly by sources in the aircraft operations source category. This category includes
airports as well as small runways, heliports, or landing pads. Aside from aircraft operations,
schools is the next most numerous source category within 10 miles of the WADC monitoring
site. The closest sources to WADC are hospitals and heliports at hospitals.
Table 8-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the
Washington, D.C. monitoring site. Table 8-2 also includes a vehicle registration-to-county
population ratio (vehicles-per-person). In addition, the population within 10 miles of the site is
presented. An estimate of 10-mile vehicle registration was calculated by applying the county-
level vehicle registration-to-population ratio to the 10-mile population surrounding the
monitoring site. Table 8-2 also contains annual average daily traffic information. District-level
VMT was not readily available; thus, daily VMT for Washington, D.C. is not provided in
Table 8-2.
Table 8-2. Population, Motor Vehicle, and Traffic Information for the Washington, D.C.
Monitoring Site
Site
WADC
Estimated
County
Population1
604,453
County-level
Vehicle
Registration2
219,173
Vehicles
per Person
(Registration:
Population)
0.36
Population
within 10
miles3
1,911,152
Estimated
10-mile
Vehicle
Ownership
692,978
Annual
Average
Daily
Traffic4
7,700
County-
level
Daily
VMT5
NA
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2009 data from the Federal Highway Administration (FHWA, 2011)
3 10-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2009 data from the District DOT (DC DOT, 2011)
5 County-level VMT was not available for this site
BOLD ITALICS = EPA-designated NATTS Site
8-5
-------
Observations from Table 8-2 include the following:
• The District's population is in the middle of the range compared to other counties
with NMP sites. However, its 10-mile population ranks fifth highest.
• The District-level vehicle registration is in the bottom third compared to other
counties with NMP sites, while its 10-mile ownership is in the top third of the range.
• The vehicle-per-person ratio is among the lowest compared to other NMP sites.
• The traffic volume experienced near WADC is in the bottom third compared to other
NMP monitoring sites. The traffic estimate used came from the intersection of Bryant
Street and First Street.
8.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in Washington, D.C. on sample days, as well as over the course of the year.
8.2.1 Climate Summary
Located on the Potomac River that divides Virginia and Maryland, the capital enjoys all
four seasons, although its weather is somewhat variable. Summers are warm and often humid,
where southerly winds prevail, which can be accentuated by the urban heat island effect. Winters
are typical of the Mid-Atlantic region, where cool, blustery air masses are common followed by
a fairly quick return to mild temperatures. Precipitation is evenly distributed across the seasons
(Bair, 1992).
8.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest this site were retrieved
for 2010 (NCDC, 2010). The closest weather station to WADC is located at Ronald Reagan
Washington National Airport (WBAN 13743). Additional information about the National
Airport weather station, such as the distance between the site and the weather station, is provided
in Table 8-3. These data were used to determine how meteorological conditions on sample days
vary from normal conditions throughout the year.
8-6
-------
Table 8-3. Average Meteorological Conditions near the Washington, D.C. Monitoring Site
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Washington, D.C. - WADC
Ronald Reagan
Washington
National Airport
13743
(38.87, -77.03)
4.07
miles
183°
(S)
Sample
Day
2010
66.8
±5.0
67.7
±2.1
58.8
±4.6
59.5
± 1.9
43.2
±4.6
43.5
±1.9
51.0
±4.1
51.5
± 1.7
59.2
±3.2
58.4
±1.3
1014.9
±1.8
1015.2
±0.7
7.7
±0.8
7.6
±0.3
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
oo
-------
Table 8-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 8-3 is the
95 percent confidence interval for each parameter. As shown in Table 8-3, average
meteorological conditions on sample days were representative of average weather conditions
throughout the year.
8.2.3 Back Trajectory Analysis
Figure 8-3 is the composite back trajectory map for days on which samples were
collected at the WADC monitoring site in 2010. Included in Figure 8-3 are four back trajectories
per sample day. Figure 8-4 is the corresponding cluster analysis for 2010. An in-depth
description of these maps and how they were generated is presented in Section 3.5.2.1. For the
composite map, each line represents the 24-hour trajectory along which a parcel of air traveled
toward the monitoring site on a given sample day and time. For the cluster analysis, each line
corresponds to a back trajectory representative of a given cluster of trajectories. For all maps,
each concentric circle around the site in Figures 8-3 and 8-4 represents 100 miles.
Observations from Figures 8-3 and 8-4 include the following:
• Back trajectories originated from a variety of directions at WADC. The bulk of
trajectories appear to originate from the southwest, west, and northwest, while fewer
trajectories originated from the northeast, east, and southeast.
• The 24-hour air shed domain for WADC was comparable in size to many other NMP
monitoring sites. The farthest away a trajectory originated was over Lake Michigan,
or approximately 550 miles away. However, the average trajectory length was 237
miles and nearly 90 percent of back trajectories originated within 400 miles of the
site.
• The cluster analysis shows that 53 percent of trajectories originated from the
southwest, west, and northwest of WADC. The short cluster (26 percent) represents
trajectories originating within 200 miles of the site and generally from the northwest
to north to northeast. Another 12 percent of trajectories originated to the southeast to
southwest and also within 200 miles of the site. Trajectories originating to the north
to northeast were also common (9 percent).
-------
Figure 8-3. 2010 Composite Back Trajectory Map for WADC
Figure 8-4. Back Trajectory Cluster Map for WADC
8-9
-------
8.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at Ronald Reagan Washington National
Airport were uploaded into a wind rose software program to produce customized wind roses, as
described in Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals"
positioned around a 16-point compass, and uses different colors to represent wind speeds.
Figure 8-5 presents three different wind roses for the WADC monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at this location.
Observations from Figure 8-5 for WADC include the following:
• The NWS weather station at Washington National Airport is located approximately
4.1 miles to the south of WADC. Note that between WADC and Washington
National is the city of Washington and the Potomac River.
• Historically, southerly to south-south westerly winds account for approximately
25 percent of wind observations near WADC, while northwesterly to northerly winds
account for another 25 percent of observations. Calm winds (< 2 knots) were
observed for less than 10 percent of the hourly measurements.
• The full-year wind patterns are similar to the wind patterns shown on the historical
wind rose, indicating that wind patterns in 2010 were similar to what is expected
climatologically near this site. Further, the sample day wind patterns for 2010 are
similar to both the full-year and historical wind patterns. This indicates that
conditions on sample days were representative of conditions experienced throughout
the year and historically.
8-10
-------
Figure 8-5. Wind Roses for the Ronald Reagan Washington National Airport Weather
Station near WADC
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between WADC and NWS Station
HMIM / NtnMa R11«
a«n> | / anwr ,n,«
8-11
-------
8.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Washington, D.C.
monitoring site in order to allow analysts and readers to focus on a subset of pollutants through
the context of risk. Each pollutant's preprocessed daily measurement was compared to its
associated risk screening value. If the concentration was greater than the risk screening value,
then the concentration "failed the screen." Pollutants of interest are those for which the
individual pollutant's total failed screens contribute to the top 95 percent of the site's total failed
screens. In addition, if any of the NATTS MQO Core Analytes measured by the monitoring site
did not meet the pollutant of interest criteria based on the preliminary risk screening, that
pollutant was added to the list of site-specific pollutants of interest. A more in-depth description
of the risk screening process is presented in Section 3.2.
Table 8-4 presents WADC's pollutants of interest. The pollutants that failed at least one
screen and contributed to 95 percent of the total failed screens for the WADC monitoring site are
shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of interest are shaded and/or
bolded. WADC sampled for hexavalent chromium and PAH.
Table 8-4. Risk Screening Results for the Washington, D.C. Monitoring Site
Pollutant
Screening
Value
(Hg/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Washington, D.C. - WADC
Naphthalene
Fluorene
0.029
0.011
Total
58
2
60
58
58
116
100.00
3.45
51.72
96.67
3.33
96.67
100.00
Observations from Table 8-4 include the following:
• Naphthalene and fluorene failed screens for WADC. Naphthalene failed 100 percent
of its screens and contributed to almost 97 percent of the total failed screens for
WADC. Fluorene failed only two screens, contributing to roughly 3 percent of the
total failed screens.
• Benzo(a)pyrene and hexavalent chromium were added as pollutants of interest for
WADC because they are the other NATTS MQO Core Analytes measured by this
site. These two pollutants are not shown in Table 8-4.
8-12
-------
8.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Washington, D.C. monitoring site. Concentration averages are provided for the pollutants
of interest for WADC, where applicable. Concentration averages for select pollutants are also
presented graphically for the site, where applicable, to illustrate how the site's concentrations
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at the
site, where applicable. Additional site-specific statistical summaries are provided in Appendices
M and O.
8.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for the Washington, D.C. monitoring site, as described in Section 3.1. The quarterly average of a
particular pollutant is simply the average concentration of the preprocessed daily measurements
over a given calendar quarter. Quarterly average concentrations include the substitution of zeros
for all non-detects. A site must have a minimum of 75 percent valid samples of the total number
of samples possible within a given quarter for a quarterly average to be calculated. An annual
average includes all measured detections and substituted zeros for non-detects for the entire year
of sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for WADC are presented
in Table 8-5, where applicable. Note that if a pollutant was not detected in a given calendar
quarter, the quarterly average simply reflects "0" because only zeros substituted for non-detects
were factored into the quarterly average concentration.
8-13
-------
Table 8-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest
for the Washington, D.C. Monitoring Site
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(ng/m3)
2nd
Quarter
Average
(ng/m3)
3rd
Quarter
Average
(ng/m3)
4th
Quarter
Average
(ng/m3)
Annual
Average
(ng/m3)
Washington, D.C. - WADC
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
22/58
40/60
58/58
0.13
±0.08
0.01
±0.01
112.31
± 64.73
0.02
±0.02
0.02
±0.01
91.51
±16.18
0.01
±0.01
0.03
±0.01
114.13
±26.36
0.07
±0.05
0.02
±0.01
123.21
±38.41
0.05
±0.02
0.02
±<0.01
110.77
± 18.56
Observations for WADC from Table 8-5 include the following:
• Naphthalene was detected in every PAH sample collected at WADC while
benzo(a)pyrene was detected in less than half of the PAH samples collected.
Hexavalent chromium was detected in two-thirds of the samples collected.
• The annual average concentration of naphthalene is significantly higher than the
annual average concentrations of benzo(a)pyrene and hexavalent chromium.
WADC's annual average concentration ranks sixth highest among NMP sites
sampling this pollutant, as shown in Table 4-11.
• While the quarterly average concentrations of naphthalene did not differ significantly
across the quarters, the confidence interval for the first quarter average is much higher
than the confidence intervals for the other quarterly averages. The highest
naphthalene concentration was measured at WADC on January 14, 2010 (454 ng/m3).
The next highest measurement during the first quarter of 2010 was less than half that
measurement (215 ng/m3 on February 1, 2010) and the median concentration for this
quarter is 76.7 ng/m3.
• Benzo(a)pyrene concentrations tended to be higher during the colder months of the
year, as indicated by the quarterly averages. However, both the first and fourth
quarters have relatively high confidence intervals associated with them, indicating
higher variability within these quarters. Note that of the 14 measurements of
benzo(a)pyrene greater than or equal to 0.1 ng/m3, eight were measured during the
first quarter and five during the fourth quarter (the one additional measurement was
measured during the second quarter). In all, there were 10 measured detections of this
pollutant during the first quarter of 2010, three during the second quarter, one during
the third quarter, and eight during the fourth quarter.
8-14
-------
8.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for benzo(a)pyrene,
hexavalent chromium, and naphthalene were created for WADC. Figures 8-6 through 8-8
overlay the site's minimum, annual average, and maximum concentrations onto the program-
level minimum, first quartile, average, median, third quartile, and maximum concentrations, as
described in Section 3.5.3.
Observations from Figures 8-6 through 8-8 include the following:
• Figure 8-6 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
WADC is below the program-level average concentration. Figure 8-6 also shows
that the maximum concentration measured at WADC is well below the maximum
concentration measured across the program. There were several non-detects of
benzo(a)pyrene measured at WADC as the minimum, first quartile, and median
(second quartile) concentrations were all zero.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 8-7 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 8-7 shows that WADC's annual average concentration (0.0182 ng/m3) is
half the program-level average (0.0369 ng/m3). The maximum concentration
measured at WADC is well below the program maximum concentration. There
were several non-detects of hexavalent chromium measured at WADC as both the
minimum and first quartile concentrations are zero.
• Figure 8-8 shows that the annual naphthalene average for WADC is just greater
than the program-level average concentration. The maximum naphthalene
concentration measured at WADC is well below the program-level maximum
concentration. The minimum concentration measured at WADC is greater than
the program-level first quartile. There were no non-detects of naphthalene
measured at WADC. Compared to other sites sampling naphthalene, this site has
one of the highest minimum concentrations of this pollutant.
8-15
-------
Figure 8-6. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
— o —
Program Max Concentration = 42.7 ng/m3
) 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Ave
• D D D
Site: Site Average Site Minimum/Maximum
o —
1.8 2
'rage
Figure 8-7. Program vs. Site-Specific Average Hexavalent Chromium Concentration
E
0
Program:
Site:
0.15 0.3 0.45
Concentration (ng/m3)
Program Max Concentration = 3.51 ng/m3
|
0.6
1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Ave
• • D D
Site Average Site Minimum/Maximum
o —
;rage
0.
Figure 8-8. Program vs. Site-Specific Average Naphthalene Concentration
I b
•
) 200 400
Program: 1st Quartile
•
Site: SiteAverage
O
600 800 1000 1200 1400
Concentration (ng/m3)
2nd Quartile 3rd Quartile 4th Quartile
D D D
Site Minimum/Maximum
Average
8-16
-------
8.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. WADC has sampled hexavalent chromium under the NMP since 2005. Thus,
Figure 8-9 presents the 3-year rolling statistical metrics for hexavalent chromium for WADC.
The statistical metrics presented for assessing trends include the substitution of zeros for non-
detects.
Figure 8-9. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at WADC
Concentration
• Mh Percaitllt
- Maximum
• Sllli P«(«itlle
Hexavalent chromium sampling at WADC began in March 2005.
Observations from Figure 8-9 for hexavalent chromium measurements at WADC include
the following:
• Sampling for hexavalent chromium began in March 2005.
• The maximum hexavalent chromium concentration was measured on
August 20, 2005 (2.97 ng/m3), and is an order of magnitude higher than the next
highest measurement (0.645 ng/m3 measured on July 4, 2006). The August 20, 2005
measurement is also second highest hexavalent chromium measured at any site since
the onset of sampling for this pollutant under the NMP (only four total measurements
greater than 1 ng/m3 have been measured). Even the second-highest measurement for
WADC is an order of magnitude higher than most other concentrations measured at
this site (all but three concentrations measured at WADC are less than 0.1 ng/m3).
8-17
-------
• Hexavalent chromium concentrations appear to have decreased through the
2007-2009 time frame, with a slight uptick for 2008-2010. Because of the magnitude
of the maximum concentrations from 2005 and 2006, it is difficult to determine if the
initial changes in the rolling average concentrations are attributable to an actual
decrease in concentrations or just the shifting of the data to a 3-year period without
one of these high values. However, the maximum, median, and 95th percentile
concentrations also exhibit a decreasing trend through 2007-2009. While the most of
the statistical parameters stayed the same for 2008-2010, the rolling average and 95th
percentile increased slightly.
8.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the
WADC monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
8.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Washington, D.C monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
were compared to the acute MRL; the quarterly averages were compared to the intermediate
MRL; and the annual averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the WADC monitoring site were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
the Washington D.C. monitoring site.
8.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for WADC and where annual average concentrations could
be calculated, risk was further examined by calculating cancer and noncancer surrogate risk
approximations (refer to Section 3.5.5.2 regarding the criteria for calculating annual averages
and how cancer and noncancer surrogate risk approximations are calculated). Annual averages,
8-18
-------
cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk approximations
are presented in Table 8-6, where applicable.
Table 8-6. Cancer and Noncancer Surrogate Risk Approximations for the Washington,
D.C. Monitoring Site
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(ng/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Washington, D.C. - WADC
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
0.00176
0.012
0.000034
0.0001
0.003
22/58
40/60
58/58
0.05
± 0.02
0.02
± <0.01
110.77
± 18.56
0.10
0.22
3.77
<0.01
0.04
— = a Cancer URE or Noncancer RfC is not available.
Observations for WADC from Table 8-6 include the following:
• As discussed in Section 8.4.1, naphthalene's annual average concentration is four
magnitudes higher than the annual average concentrations for the other two
pollutants of interest.
• Naphthalene's cancer risk approximation is greater than 1.0 in-a-million
(3.77 in-a-million), while its noncancer risk approximation is well below an HQ
of 1.0(0.04).
• Benzo(a)pyrene's cancer risk approximation is much less than naphthalene's
(0.10 in-a-million). A noncancer RfC is not available for benzo(a)pyrene, thus a
noncancer risk approximation could not be calculated.
• The cancer surrogate risk approximation based on hexavalent chromium's annual
average concentration is well below 1.0 in-a-million (0.22 in-a-million). The
noncancer surrogate risk approximation is also low (<0.01).
8.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 8-7 and 8-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 8-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
8-19
-------
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 8-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 8-7 and 8-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. Further, cancer and noncancer surrogate risk approximations based on each site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 8.3, WADC sampled for PAH and hexavalent chromium. In addition, the cancer and
noncancer surrogate risk approximations are limited to those pollutants with enough data to meet
the criteria for annual averages to be calculated. A more in-depth discussion of this analysis is
provided in Section 3.5.5.3.
Observations from Table 8-7 include the following:
• Benzene and formaldehyde are the highest emitted pollutants with cancer UREs in the
District of Columbia. Formaldehyde and benzene are the pollutants with the highest
toxi city-weighted emissions (of the pollutants with cancer UREs).
• Seven of the highest emitted pollutants also have the highest toxi city-weighted
emissions.
• Naphthalene is the only pollutant sampled for at WADC that appears on both
emissions-based lists. Naphthalene is the sixth highest emitted pollutant with a cancer
URE in the District of Columbia and has the fifth highest toxicity-weighted emissions
(of the pollutants with cancer UREs).
• While hexavalent chromium is not one of the 10 highest emitted pollutants in the
District, its toxi city-weighted emissions ranked seventh highest (of the pollutants with
cancer UREs).
• Several POM Groups are among the highest emitted "pollutants" in the District
and/or rank among the highest toxicity-weighted emissions. POM, Group 5a, which
includes benzo(a)pyrene, appears on both emissions lists for the District. POM,
Group 2b includes several PAH sampled for at WADC including acenaphthylene,
fluoranthene, fluorene, and perylene. POM, Group 6 includes benzo(a)anthracene and
indeno(l,2,3-cd)pyrene. None of the PAH included in POM, Groups 2b or 6 were
identified as pollutants of interest for WADC.
8-20
-------
Table 8-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Washington, D.C. Monitoring Site
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity- Weighted Emissions
(County-Level)
Pollutant
Cancer Toxicity
Weight
Top 10 Cancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Washington, D.C. - WADC
Benzene
Formaldehyde
Acetaldehyde
Ethylbenzene
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
POM, Group la
POM, Group 6
149.75
139.99
83.08
76.80
24.93
14.40
4.38
3.56
0.39
0.35
Formaldehyde
Benzene
1,3 -Butadiene
POM, Group 3
Naphthalene
POM, Group 2b
Hexavalent Chromium, PM
Ethylbenzene
Acetaldehyde
POM, Group 5a
1.82E-03
1.17E-03
7.48E-04
6.38E-04
4.90E-04
3.13E-04
2.01E-04
1.92E-04
1.83E-04
1.32E-04
Naphthalene
Hexavalent Chromium
Benzo(a)pyrene
3.77
0.22
0.10
oo
to
-------
Table 8-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Washington, D.C. Monitoring Site
oo
to
to
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Washington, D.C. - WADC
Toluene
Methanol
Xylenes
Benzene
Formaldehyde
Hexane
Acetaldehyde
Ethylbenzene
Ethylene glycol
1,3 -Butadiene
433.60
342.72
309.21
149.75
139.99
93.23
83.08
76.80
36.01
24.93
Acrolein
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Benzene
Naphthalene
Xylenes
Nickel, PM
Arsenic, PM
Propionaldehyde
361,586.14
14,284.26
12,467.17
9,231.05
4,991.64
4,799.27
3,092.12
1,910.51
1,536.26
966.01
Naphthalene 0.04
Hexavalent Chromium <0.01
-------
Observations from Table 8-8 include the following:
• Toluene, methanol, and xylenes are the highest emitted pollutants with noncancer
RfCs in the District of Columbia.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
noncancer RfCs) are acrolein, formaldehyde, and 1,3-butadiene.
• Five of the highest emitted pollutants in the District of Columbia also have the
highest toxicity-weighted emissions.
• Naphthalene has the highest noncancer risk approximation (albeit low). Naphthalene
has the sixth highest toxicity-weighted emissions (of the pollutants with noncancer
RfCs) but is not one of the 10 highest emitted pollutants.
• Hexavalent chromium, the only other pollutant of interest for which a noncancer RfC
is available, does not appear on either emissions-based list.
8.6 Summary of the 2010 Monitoring Data for WADC
Results from several of the data treatments described in this section include the
following:
»«» Naphthalene andfluorene were the only pollutants to fail screens for WADC. While
naphthalene was the only pollutant of interest identified via the risk screening
process, hexavalent chromium and benzo(a)pyrene were added to WADC's pollutants
of interest because they are NATTSMQO Core Analytes.
*»* Of the site-specific pollutants of the interest, naphthalene had the highest annual
average concentration for WADC.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
8-23
-------
9.0 Sites in Florida
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS and UATMP sites in Florida, and integrates these
concentrations with emissions, meteorological, and risk information. Data generated by sources
other than ERG are not included in the data analyses contained in this report. Readers are
encouraged to refer to Sections 1 through 4 and the glossary (Appendix P) for detailed
discussions and definitions regarding the various data analyses presented below.
9.1 Site Characterization
This section characterizes the Florida monitoring sites by providing geographical and
physical information about the location of the sites and the surrounding areas. This information
is provided to give the reader insight regarding factors that may influence the air quality near the
sites and assist in the interpretation of the ambient monitoring measurements.
The Florida sites are located in two different urban areas. Three sites (AZFL, SKFL, and
SYFL) are located in the Tampa-St. Petersburg-Clearwater, FL MSA. ORFL and PAFL are
located in the Orlando-Kissimmee, FL MSA. Figures 9-1 through 9-5 are composite satellite
images retrieved from ArcGIS Explorer showing the monitoring sites in their urban and rural
locations. Figures 9-6 and 9-7 identify point source emissions locations by source category, as
reported in the 2008 NEI for point sources. Note that only sources within 10 miles of the sites are
included in the facility counts provided in Figures 9-6 through 9-7. Thus, sources outside the
10-mile radius have been grayed out, but are visible on the maps to show emissions sources
outside the 10-mile boundary. A 10-mile boundary was chosen to give the reader an indication of
which emissions sources and emissions source categories could potentially have a direct effect
on the air quality at the monitoring sites. Further, this boundary provides both the proximity of
emissions sources to the monitoring sites as well as the quantity of such sources within a given
distance of the sites. Table 9-1 describes the area surrounding each monitoring site by providing
supplemental geographical information such as land use, location setting, and locational
coordinates.
9-1
-------
Figure 9-1. St. Petersburg, Florida (AZFL) Monitoring Site
to
-------
Figure 9-2. Pinellas Park, Florida (SKFL) Monitoring Site
-------
Figure 9-3. Plant City, Florida (SYFL) Monitoring Site
-------
Figure 9-4. Winter Park, Florida (ORFL) Monitoring Site
-------
Figure 9-5. Orlando, Florida (PAFL) Monitoring Site
-------
Figure 9-6. NEI Point Sources Located Within 10 Miles of the
Tampa/St. Petersburg, Florida Monitoring Sites
B2-50U-W 82'15'0-W S2-401TW 82'35TJ-W 82*30trW S2'25'0-W a2'2(Rrw 02'15'irW
I I 1 . - -I-
Legend
Solid Waste Disposa! - Commercial/Institutional (1)
S Surface Coating (4)
it Telecommunications (2)
** TransportationEquipment(l)
» Wa stewate r Treatrn ent (2)
9-7
-------
Figure 9-7. NEI Point Sources Located Within 10 Miles of ORFL and PAFL
Legend
Note: Due to facility density and collocation, the total facilities
displayed may not represent ail facilities within the area of interest
ORFL UATMP site ® PAFL UATMP site 1 0 mile radius f 1 County boundary
Source Category Group (No. of Facilities)
-f Aircraft Operations (23)
I Asphalt Processing/Roofing Manufacturing (1)
0 Auto Body Shop/Painters (1)
ft Automobile/Truck Manufacturing (1)
6 Bakery (3)
± Boat Manufacturing (1)
Brick Manufacturing & Structural Clay (1)
A Cement Kiln/Dryer (1)
C Chemical Manufacturing (1)
fl> Dry Cleaning (1)
6 Electrical Equipment (2)
* Electricity Generation via Combustion (1)
E Electroplating, Plating, Polishing, Anodizing, and Coloring (1)
© Fabricated Metal Products (2)
F Food Processing/Agriculture (2)
B Furniture Plant (2)
• Gasoline/Diesel Service Station (1)
: Hospital (2)
$ Hot Mix Asphalt Plant (4)
-$• Industrial Machinery and Equipment (2)
^t Institutional - school (3)
V Mineral Products (1)
? Miscellaneous Commercial/Industrial (5)
M Miscellaneous Manufacturing (2)
• Oil and/or Gas Production (1}
P Printing/Publishing (2)
H Pulp and Paper Plant/Wood Products (3)
R Rubber and Miscellaneous Plastics Products (3)
S Surface Coating (5)
rr Telecommunications (1)
9-8
-------
Table 9-1. Geographical Information for the Florida Monitoring Sites
Site
Code
AZFL
SKFL
SYFL
ORFL
PAFL
AQS Code
12-103-0018
12-103-0026
12-057-3002
12-095-2002
12-095-1004
Location
St.
Petersburg
Pinellas
Park
Plant City
Winter
Park
Orlando
County
Pinellas
Pinellas
Hillsborough
Orange
Orange
Micro- or
Metropolitan
Statistical Area
Tampa-St.
Petersburg-
Clearwater, FL
Tampa-St.
Petersburg-
Clearwater, FL
Tampa-St.
Petersburg-
Clearwater, FL
Orlando-
Kissimmee, FL
Orlando-
Kissimmee, FL
Latitude
and
Longitude
27.785556,
-82.74
27.850041,
-82.714590
27.96565,
-82.2304
28.596444,
-81.362444
28.550833,
-81.345556
Land Use
Residential
Residential
Residential
Commercial
Commercial
Location
Setting
Suburban
Suburban
Rural
Urban/City
Center
Suburban
Additional Ambient Monitoring Information1
NO, NO2, NOX, VOC, O3, Meteorological parameters,
PM10, PM10 Speciation, PM25.
VOC, Meteorological parameters, PM10 Speciation,
Black carbon, PM25 Speciation, PM25.
CO, SO2, NOy, NO, NO2, NOX, VOC, O3,
Meteorological parameters, PM10, PM10 Speciation,
PM2 5, PM2 5 Speciation, PM Coarse.
CO, SO2, NO, NO2, NOX, VOC, O3, Meteorological
parameters, PM10, PM25.
Meteorological parameters, PM10, PM25.
VO
BOLD ITALICS = EPA-designated NATTS Site.
-------
AZFL is located at Azalea Park in St. Petersburg. Figure 9-1 shows that the area
surrounding AZFL consists of mixed land use, including residential, commercial, and industrial
properties. Heavily traveled roadways are located less than 1 mile from the monitoring site.
AZFL is just over 1 mile east of Boca Ciega Bay.
SKFL is located in Pinellas Park, north of St. Petersburg. This site is on the property of
Skyview Elementary School near 86th Avenue North. Figure 9-2 shows that SKFL is located in a
primarily residential area. However, a railroad intersects with the Pinellas Park Ditch near a
construction company in the bottom left corner of Figure 9-2. Population exposure is the purpose
behind monitoring in this location and this site is the Pinellas County NATTS site.
SYFL is located in Plant City, which is also part of the Tampa-St. Petersburg-Clearwater,
FL MSA, although it is on the eastern outskirts of the area. Unlike the other Florida sites, the
SYFL monitoring site is in a rural area although, as Figure 9-3 shows, a residential community
and country club lie just to the west of the site. Just south of the site is a tank that is part of the
local water treatment facility. This site serves as a background site, although the effect of
increased development in the area is likely being captured by the monitor. This site is the Tampa
NATTS site.
Figure 9-6 shows the location of the Tampa/St. Petersburg sites in relation to each other.
SYFL is located the furthest east and AZFL is the furthest west. A large cluster of point sources
is located just north of SKFL. Another cluster of emissions sources is located about halfway
between SYFL and the other two sites, although grayed out and not included in the facility
counts in Figure 9-6. Aircraft operations, which include airports as well as small runways,
heliports, or landing pads, printing and publishing facilities, and fabricated metal processing
facilities are the source categories with the highest number of emissions sources in the Tampa/St.
Petersburg area (based on the areas covered by the 10-mile radii).
ORFL is located in Winter Park, north of Orlando. Figure 9-4 shows that ORFL is
located near Lake Mendsen, east of Lake Killarney and south of Winter Park Village. This site
lies in a commercial area and serves as a population exposure monitor.
9-10
-------
PAFL is located in northern Orlando, on the northwestern edge of the Orlando Executive
Airport property, as shown in Figure 9-5. The area is considered commercial and experiences
heavy traffic. The airport is bordered by Colonial Drive to the north and the East-West
Expressway (Toll Road 408) to the south (although not shown in Figure 9-5). A large shopping
complex is located to the northeast of the site, just north of the airport, between Colonial Drive
and Maguire Boulevard. Interstate-4 runs north-south less than 2 miles to the west of the
monitoring site.
Figure 9-7 shows that ORFL is located a few miles north of PAFL. Most of the point
sources are located on the western side of the 10-mile radii. Although the emissions sources
surrounding ORFL and PAFL are involved in a variety of industries and processes, the aircraft
operations source category has the highest number of emissions sources within 10 miles of these
sites.
Table 9-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the areas surrounding the Florida
monitoring sites. Table 9-2 also includes a vehicle registration-to-county population ratio
(vehicles-per-person) for each site. In addition, the population within 10 miles of each site is
presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-level
vehicle registration-to-population ratio to the 10-mile population surrounding each monitoring
site. Table 9-2 also contains annual average daily traffic information. Finally, Table 9-2 presents
the daily VMT for Pinellas, Hillsborough, and Orange Counties.
Observations from Table 9-2 include the following:
• Hillsborough County, where SYFL is located, is the most populous of the Florida
counties with monitoring sites, although Orange County also has over 1 million
people. Broward County is the eleventh most populous county of counties with NMP
sites covered in this report.
• Of the five Florida monitoring sites, ORFL has the highest population within 10 miles
of all the Florida sites. ORFL's 10-mile population ranks 12th highest among NMP
sites. Note the difference between SYFL's 10-mile and county-level populations. This
is an example of a site located within a populous county that is not near the
population center.
9-11
-------
• With the exception of Pinellas County (AZFL and SKFL), the vehicle registration
counts for the Florida sites are all over 1 million, with Hillsborough County having
the most. The 10-mile ownership estimates are more variable, with SYFL having the
least number of vehicles and ORFL having the most.
• The vehicle-per-person ratios range from 0.90 (Orange County) to 0.96 (Pinellas
County).
• The traffic volume near SYFL is the lowest among the Florida sites and highest near
SKFL. Traffic volumes near the Florida monitoring sites are mid-range compared to
other NMP sites. The following list provides the roadways or intersections from
which the traffic data were obtained:
AZFL - 66th Street North, north of 9th Street
ORFL - Orlando Avenue, north of Morse Boulevard
PAFL - East Colonial Drive, between Primrose Road and Bumby Avenue
SKFL - Park Boulevard, east of 66th Street North
SYFL - Martin Luther King Jr. Boulevard (574), east of Mclntosh Road
• VMT is highest for Orange County and lowest for Pinellas County (among the
Florida sites). The Orange, Hillsborough, and Pinellas County VMTs ranked seventh,
ninth, and 12th highest among counties with NMP sites, respectively.
Table 9-2. Population, Motor Vehicle, and Traffic Information for the Florida Monitoring
Sites
Site
AZFL
SKFL
SYFL
ORFL
PAFL
Estimated
County
Population1
916,719
1,233,846
1,149,500
County-level
Vehicle
Registration2
879,317
1,125,844
1,037,369
Vehicles per
Person
(Registration:
Population)
0.96
0.91
0.90
Population
within
10 miles3
554,850
672,114
323,844
1,003,746
872,658
Estimated
10-mile
Vehicle
Ownership
532,212
644,692
295,497
905,833
787,532
Annual
Average
Daily
Traffic4
41,500
49,500
10,700
31,500
43,500
County-
level
Daily
VMT5
23,138,726
34,745,256
35,657,527
1 County-level population estimates reflect data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the FL Dept of Highway Safety & Motor
Vehicles (FL DHSMV, 2010)
3 10-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2010 data from the Florida DOT (FL DOT, 201 Oa)
5 County-level VMT reflects 2010 data for all public roads from the Florida DOT (FL DOT, 2010b)
BOLD ITALICS = EPA-designated NATTS Site.
9-12
-------
9.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in Florida on sample days, as well as over the course of the year.
9.2.1 Climate Summary
The Tampa and Orlando areas experience very mild winters and warm, humid summers.
Precipitation tends to be concentrated during the summer, as afternoon thunderstorms occur
frequently. Semi-permanent high pressure offshore over the Atlantic Ocean extends westward
towards Florida in the winter, resulting in reduced precipitation amounts. Land and sea breezes
affect coastal locations and the proximity to the Atlantic Ocean or Gulf of Mexico can have a
marked affect on the local meteorological conditions. Florida's orientation and location between
the warm waters of the Gulf of Mexico, the Atlantic Ocean, and Caribbean Sea make it
susceptible to tropical systems (Bair, 1992 and FCC, 2012).
9.2.2 Meteorological Conditions in 2010
Hourly meteorological data from NWS weather stations nearest these sites were retrieved
for 2010 (NCDC, 2010). These data were used to determine how meteorological conditions on
sample days vary from normal conditions throughout the year. The weather station closest to the
AZFL monitoring site is located at St. Petersburg/Whitted Airport (WBAN 92806); closest to
SYFL is at Plant City Municipal Airport (WBAN 92824); closest to SKFL is at
St. Petersburg/Clearwater International Airport (WBAN 12873); and closest to ORFL and PAFL
is at Orlando Executive Airport (WBAN 12841). Additional information about each of these
weather stations, such as the distance between the sites and the weather stations, is provided in
Table 9-3.
Table 9-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 9-3 is the
95 percent confidence interval for each parameter. As shown in Table 9-3, average
meteorological conditions on sample days in 2010 at the Florida monitoring sites were
representative of average weather conditions throughout the entire year.
9-13
-------
Table 9-3. Average Meteorological Conditions near the Florida Monitoring Sites
Closest NWS Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar
Wind Speed
(kt)
St. Petersburg, Florida - AZFL
St. Petersburg/
Whirled Airport
92806
(27.77, -82.63)
6.77
miles
94°
(E)
Sample
Day
2010
78.8
±3.2
78.7
± 1.3
72.3
±3.3
72.3
± 1.3
61.3
±3.5
61.1
±1.4
65.6
±3.1
65.5
± 1.2
69.7
±2.0
69.2
±0.9
1016.1
±1.1
1016.1
±0.4
7.3
±0.7
7.3
±0.3
Pinellas Park, Florida - SKFL
St Petersburg-
Clearwater Intl
Airport
12873
(27.91, -82.69)
4.46
miles
12°
(NNE)
Sample
Day
2010
79.6
±3.0
79.0
±1.3
71.3
±3.1
70.7
±1.3
60.4
±3.3
59.6
± 1.4
64.7
±2.9
64.2
±1.3
70.4
±2.0
70.0
±1.0
1016.4
±1.0
1016.5
±0.4
7.0
±0.7
6.9
±0.3
Plant City, Florida - SYFL
Plant City
Municipal Airport
92824
(28.00, -82.16)
4.56
miles
50°
(NE)
Sample
Day
2010
81.6
±3.2
81.4
±1.3
70.4
±3.4
70.3
±1.4
59.5
±3.6
59.5
±1.5
63.9
±3.2
63.9
±1.3
71.8
±2.0
72.0
±0.9
NA
NA
4.5
±0.5
4.4
±0.2
Winter Park, Florida - ORFL
Orlando Executive
Airport
12841
(28.55, -81.33)
3.95
miles
145°
(SE)
Sample
Day
2010
79.8
±3.2
79.7
± 1.3
70.2
±3.2
70.0
±1.3
59.0
±3.7
58.7
±1.6
63.7
±3.2
63.5
± 1.3
70.5
±2.5
70.4
±1.2
1016.9
±1.2
1017.0
±0.4
6.4
±0.7
6.1
±0.3
VO
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
NA= Sea level pressure was not recorded at the Plant City Municipal Airport
-------
Table 9-3. Average Meteorological Conditions near the Florida Monitoring Sites (Continued)
Closest NWS Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Orlando, Florida - PAFL
Orlando Executive
Airport
12841
(28.55, -81.33)
0.84
miles
111°
(ESE)
Sample
Day
2010
80.1
±4.4
79.7
±1.3
70.8
±4.4
70.0
± 1.3
60.9
±4.7
58.7
±1.6
64.9
±4.2
63.5
±1.3
73.0
±3.6
70.4
±1.2
1015.8
±1.8
1017.0
±0.4
6.3
±0.9
6.1
±0.3
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
NA= Sea level pressure was not recorded at the Plant City Municipal Airport
-------
9.2.3 Back Trajectory Analysis
Figure 9-8 is the composite back trajectory map for days on which samples were
collected at the AZFL monitoring site in 2010. Included in Figure 9-8 are four back trajectories
per sample day. Figure 9-9 is the corresponding cluster analysis for 2010. Similarly, Figures 9-10
through 9-17 are the composite back trajectory maps and corresponding cluster analyses for the
remaining Florida monitoring sites. An in-depth description of these maps and how they were
generated is presented in Section 3.5.2.1. For the composite maps, each line represents the
24-hour trajectory along which a parcel of air traveled toward the monitoring site on a given
sample day and time. For the cluster analyses, each line corresponds to a back trajectory
representative of a given cluster of trajectories. For all maps, each concentric circle around the
sites in Figures 9-8 through 9-17 represents 100 miles.
Figure 9-8. 2010 Composite Back Trajectory Map for AZFL
9-16
-------
Figure 9-9. Back Trajectory Cluster Map for AZFL
Figure 9-10. 2010 Composite Back Trajectory Map for SKFL
9-17
-------
Figure 9-11. Back Trajectory Cluster Map for SKFL
Figure 9-12. 2010 Composite Back Trajectory Map for SYFL
9-18
-------
Figure 9-13. Back Trajectory Cluster Map for SYFL
Figure 9-14. 2010 Composite Back Trajectory Map for ORFL
1 V
125
250
500
_
9-19
-------
Figure 9-15. Back Trajectory Cluster Map for ORFL
500
_
Figure 9-16. 2010 Composite Back Trajectory Map for PAFL
9-20
-------
Figure 9-17. Back Trajectory Cluster Map for PAFL
Observations from Figures 9-8 through 9-13 for the Tampa/St. Petersburg sites include
the following:
• The composite back trajectory maps for the Tampa/St. Petersburg sites are generally
similar to each other, even though their representative weather stations were different.
• Back trajectories originated from a variety of directions at the Tampa/St. Petersburg
sites.
• The 24-hour air shed domains for these sites were comparable in size to other NMP
monitoring sites. For all three sites, the farthest away a trajectory originated was just
greater than 500 miles away in lower Mississippi. For SYFL, back trajectories of
similar distance originated over the Atlantic Ocean.
• Most trajectories (between 88 and 92 percent for each site) originated within 400
miles of the Tampa/St. Petersburg monitoring sites. The average trajectory length
ranged from 231 miles to 237 miles for each site.
• The back trajectories for AZFL are broken into four clusters. Forty-one percent of
trajectories originated from south Florida westward over the Gulf of Mexico as well
as those originating within 100 miles of AZFL. These trajectories were grouped
together because many of them were of shorter length (originating less than 200 miles
away). Another cluster of trajectories originated over the northern Gulf and southern
U.S; these tended to originate farther away. A third cluster presents trajectories
originating over Georgia, north Florida, and just off the east coast of north Florida.
9-21
-------
The last cluster represents those originating over south Florida, the Bahamas, and
north of the Bahamas and farther over the Atlantic.
• There are six clusters presented on the cluster analysis for SKFL. The division of
trajectories for SKFL is similar to AZFL in geographical break up, but with two
differences. First, trajectories originating over south Florida, the Bahamas, north of
the Bahamas, and farther over the Atlantic are split into two clusters (16 and
9 percent). Second, the trajectory representing both shorter trajectories and ones
originating over the Gulf of Mexico to the southwest of the Tampa area for AZFL are
split into two back trajectories for SKFL (27 percent and 11 percent, respectively).
• The cluster map for SYFL is similar to AZFL in geographical breakup, but the
percentages differ somewhat, particularly for those trajectories originating along the
west coast of Florida and the Gulf of Mexico.
Observations from Figures 9-14 through 9-17 for ORFL and PAFL include the
following:
• Even though they are close in proximity to each other, the trajectory distribution for
PAFL appears different than the trajectory distribution for ORFL. This is because
sampling at PAFL occurred on a l-in-12 day schedule, yielding approximately half
the sample days as ORFL.
• The 24-hour air shed domains were similar in size compared to the other Florida
monitoring sites. The farthest away a trajectory originated was approximately 500
miles away and over the Atlantic Ocean for both sites
• Similar to the Tampa/St. Petersburg sites, 90 and 92 percent of back trajectories
originated with 400 miles of ORFL and PAFL, respectively. The average trajectory
length for ORFL and PAFL varied though, with an average trajectory length of 251
miles for ORFL and 224 miles for PAFL.
• The composite back trajectory map for ORFL shows that back trajectories originated
from a variety of directions around ORFL. The corresponding cluster map looks like
a pinwheel, confirming that trajectories originated from a variety of directions. The
short cluster representing 16 percent of trajectories includes trajectories originating
over south-central Florida as well as several shorter trajectories originating from the
east, southeast, and south of the site and within 100 miles of ORFL.
• The short cluster representing nearly 40 percent of PAFL's back trajectories includes
back trajectories originating from both west-central Florida and the Gulf of Mexico as
well as shorter trajectories originating from a variety of directions and within 100
miles of the site. The remaining clusters represent trajectories originating from the
northwest and north (9 percent), the northeast (11 percent), the east and over the
Atlantic Ocean (23 percent), and south Florida and the surrounding waters (20
percent).
9-22
-------
9.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather stations nearest the Florida sites, as presented
in Section 9.2.2, were uploaded into a wind rose software program to produce customized wind
roses, as described in Section 3.5.2.2. A wind rose shows the frequency of wind directions using
"petals" positioned around a 16-point compass, and uses different colors to represent wind
speeds.
Figure 9-18 presents three different wind roses for the AZFL monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figures 9-19 through 9-22
present the three wind roses and distance maps for SKFL, SYFL, ORFL, and PAFL,
respectively.
9-23
-------
Figure 9-18. Wind Roses for the St. Petersburg/Whitted Airport Weather Station near
AZFL
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between AZFL and NWS Station
•-„ "•• , \
• • \
9-24
-------
Figure 9-19. Wind Roses for the St. Petersburg/Clearwater International Airport Weather
Station near SKFL
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between SKFL and NWS Station
WHO SPEED
(Knots)
LJ »ZZ
^| 17 • 21
^| 11 - 17
^| 7- 11
o 4.7
Calms: 10.81%
9-25
-------
Figure 9-20. Wind Roses for the Plant City Municipal Airport Weather Station near SYFL
2008-2009 Historical Wind Rose
2010 Wind Rose
WIND SPEED
(Knots)
n -22
^| 17 • 21
^| 11 • 17
o 4.7
Calms: 26.21%
2010 Sample Day Wind Rose
Distance between SYFL and NWS Station
r—««"* *>
4-
9-26
-------
Figure 9-21. Wind Roses for the Orlando Executive Airport Weather Station near ORFL
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between ORFL and NWS Station
f'E':, T
A WM
V W-r-.-l-
\
% t odibm) *
\
i 8 ECOI*,HD, ^"'jT
I ' l.^"?5" I I'-*-
I E*7^mli I I
EKnlMHiiMSI Ufi
9-27
-------
Figure 9-22. Wind Roses for the Orlando Executive Airport Weather Station near PAFL
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between PAFL and NWS Station
,«'Bi r
*=.!
.
. ,™..., !
obM.4
,—, j
... •
9-28
-------
Observations from Figure 9-18 for AZFL include the following:
• The NWS weather station at St. Petersburg/Whitted Airport is located approximately
6.8 miles east of AZFL. Between them is most of the city of St. Petersburg. Note that
the Whitted Airport is on the Tampa Bay coast while AZFL is on the west side of the
peninsula near the Boca Ciega Bay.
• The historical wind rose shows that calm winds (< 2 knots) accounted for less than
10 percent of the hourly wind measurements from 1999 to 2009. Northerly,
northeasterly, and easterly winds were the most commonly observed wind directions
near AZFL while winds from the western quadrants were observed less frequently.
• The full-year wind rose shows that while winds from all directions were observed
near AZFL, winds from the north and east were the predominant wind directions.
There were fewer northeasterly winds observed in 2010 compared to historical
observations. The calm rate in 2010 was similar to the historical calm rate, at just
below 10 percent.
• The sample day wind patterns favor the full-year wind patterns, indicating the
conditions on sample days were representative of wind conditions experienced in
2010.
Observations from Figure 9-19 for SKFL include the following:
• The NWS weather station at St. Petersburg/Clearwater Airport is located just less
than 4.5 miles north-northeast of SKFL. Note that the St. Petersburg/Clearwater
Airport is located on Old Tampa Bay while SKFL is farther inland.
• The historical wind rose shows that winds from a variety of directions were observed
near SKFL from 1999 to 2009, although winds from the northeastern quadrant were
the most commonly observed wind directions. Calm winds accounted for
approximately 10 percent of the hourly wind measurements.
• The 2010 wind rose favors the historical wind rose, although there were more
northwesterly and north-northwesterly winds and fewer northeasterly winds observed
in 2010 than historically. The calm rate in 2010 was just over 10 percent.
• The sample day wind rose exhibits an even higher percentage of east-northeasterly,
easterly, and northwesterly winds than the 2010 wind rose. The sample day calm rate
was just below 11 percent.
Observations from Figure 9-20 for SYFL include the following:
• The NWS weather station at Plant City Municipal Airport is located 4.6 miles
northeast of SYFL. Note that this weather station has less historical data than the
other sites. This station did not begin operating until 2006 and data availability is
lacking until mid-2007.
9-29
-------
• The historical wind rose shows that calm winds (< 2 knots) account for approximately
one-quarter of the hourly wind measurements during 2008 and 2009. Similar to
SKFL, winds from the eastern quadrants were observed more often than the other
quadrants, although winds from all directions were observed near SYFL.
• Both the full-year and sample day wind patterns are similar to the historical wind
patterns, although the percentage of winds from due east was higher than the
historical wind rose. This indicates that conditions on sample days were
representative of wind conditions experienced throughout the year and historically.
Observations from Figures 9-21 and 9-22 for ORFL and PAFL include the following:
• The closest NWS station to both ORFL and PAFL is the Orlando Executive Airport.
The weather station is located just less than 4 miles southeast of ORFL and less than
1 mile east-southeast of PAFL, as PAFL is located on the edge of the Orlando
Executive Airport property. Thus, the historical and full-year wind roses for these
sites are the same.
• The historical wind rose shows that from 1999 to 2009 winds from all directions were
observed near these sites, with easterly winds being observed the most, although
winds from the due north, due south, and with an easterly component were observed
more often than winds from the remaining directions.
• The full-year wind roses also exhibit an easterly wind prominence, but winds from
the northwest quadrant were observed more frequently during 2010.
• The 2010 sample day wind rose for ORFL is similar to the full-year wind rose,
although with less southerly winds and more northerly winds. The 2010 sample day
for PAFL shows less uniformity in the wind directions than ORFL. Note, however,
that PAFL samples on a l-in-12 day sampling schedule, leading to roughly half the
sample days included in the sample day wind rose as ORFL.
9.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Florida monitoring sites in
order to allow analysts and readers to focus on a subset of pollutants through the context of risk.
For each site, each pollutant's preprocessed daily measurement was compared to its associated
risk screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by each monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
9-30
-------
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 9-4 presents the pollutants of interest for each of the Florida monitoring sites. The
pollutants that failed at least one screen and contributed to 95 percent of the total failed screens
for each monitoring site are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of
interest are shaded and/or bolded. AZFL and ORFL sampled for carbonyl compounds only.
SKFL and SYFL sampled hexavalent chromium and PAH in addition to carbonyl compounds.
PAFL sampled only PMio metals.
Table 9-4. Risk Screening Results for the Florida Monitoring Sites
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
St. Petersburg, Florida - AZFL
Acetaldehyde
Formaldehyde
0.45
0.077
Total
61
61
122
61
61
122
100.00
100.00
100.00
50.00
50.00
50.00
100.00
Pinellas Park, Florida - SKFL
Acetaldehyde
Formaldehyde
Naphthalene
0.45
0.077
0.029
Total
61
61
55
177
61
61
59
181
100.00
100.00
93.22
97.79
34.46
34.46
31.07
34.46
68.93
100.00
Plant City, Florida - SYFL
Acetaldehyde
Formaldehyde
Naphthalene
0.45
0.077
0.029
Total
61
61
36
158
61
61
60
182
100.00
100.00
60.00
86.81
38.61
38.61
22.78
38.61
77.22
100.00
Winter Park, Florida - ORFL
Acetaldehyde
Formaldehyde
0.45
0.077
Total
60
60
120
60
60
120
100.00
100.00
100.00
50.00
50.00
50.00
100.00
Orlando, Florida - PAFL
Arsenic (PM10)
Manganese (PM10)
Nickel (PM10)
Lead (PM10)
0.00023
0.005
0.0021
0.015
Total
30
3
2
1
36
30
30
30
30
120
100.00
10.00
6.67
3.33
30.00
83.33
8.33
5.56
2.78
83.33
91.67
97.22
100.00
Observations from Table 9-4 include the following:
• Acetaldehyde and formaldehyde were the only two pollutants to fail screens for
AZFL and ORFL. These two pollutants contributed equally to the total number of
failed screens for each site and failed 100 percent of screens. These two sites sampled
9-31
-------
only carbonyl compounds; among the carbonyls, only acetaldehyde, formaldehyde,
and propionaldehyde have screening values. Propionaldehyde did not fail any screens
for these sites.
• Four metals (arsenic, lead, nickel, and manganese) failed screens for PAFL; all of
these are NATTS MQO Core Analytes. Arsenic, manganese, and nickel were initially
identified as PAFL's pollutants of interest, with arsenic failing the bulk of the screens
(83 percent). Lead was added as a pollutant of interest for PAFL because it is a
NATTS MQO Core Analyte. Two additional metal NATTS MQO Core Analytes,
cadmium and beryllium, were added to PAFL's pollutants of interest, even though
they did not fail any screens. These two pollutants are not shown in Table 9-4.
• Three pollutants (acetaldehyde, formaldehyde, and naphthalene) failed screens for
SKFL and SYFL, of which all three are NATTS MQO Core Analytes and were
identified as pollutants of interest via the risk screening process. Two additional
NATTS MQO Core Analytes, hexavalent chromium and benzo(a)pyrene, were added
to SKFL and SYFL's pollutants of interest, even though they did not fail any screens.
These two pollutants are not shown in Table 9-4 for either site.
• Acetaldehyde and formaldehyde failed 100 percent of screens for all four sites
sampling carbonyl compounds. Of the PMio metals sampled at PAFL, arsenic failed
100 percent of its screens.
9.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Florida monitoring sites. Concentration averages are provided for the pollutants of interest
for each Florida site, where applicable. Concentration averages for select pollutants are also
presented graphically for each site, where applicable, to illustrate how each site's concentrations
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at
each site, where applicable. Additional site-specific statistical summaries are provided in
Appendices L, M, N, and O.
9.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each Florida site, as described in Section 3.1. The quarterly average of a particular pollutant
is simply the average concentration of the preprocessed daily measurements over a given
calendar quarter. Quarterly average concentrations include the substitution of zeros for all non-
detects. A site must have a minimum of 75 percent valid samples of the total number of samples
possible within a given quarter for a quarterly average to be calculated. An annual average
9-32
-------
includes all measured detections and substituted zeros for non-detects for the entire year of
sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the Florida monitoring
sites are presented in Table 9-5, where applicable. Note that concentrations of the PAH, metals,
and hexavalent chromium are presented in ng/m3 for ease of viewing. Also note that if a
pollutant was not detected in a given calendar quarter, the quarterly average simply reflects "0"
because only zeros substituted for non-detects were factored into the quarterly average
concentration.
Table 9-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest
for the Florida Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(jig/m3)
2nd
Quarter
Average
(jig/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Annual
Average
(jig/m3)
St. Petersburg, Florida - AZFL
Acetaldehyde
Formaldehyde
61/61
61/61
3.69
±1.05
1.28
±0.22
2.56
±0.50
1.58
±0.19
1.76
±0.31
1.74
±0.44
3.84
±0.84
2.22
±0.38
2.94
±0.40
1.71
±0.18
Pinellas Park, Florida - SKFL
Acetaldehyde
Formaldehyde
Benzo(a)pyrenea
Hexavalent Chromium3
Naphthalene3
61/61
61/61
36/59
52/61
59/59
3.94
±1.31
0.97
±0.22
0.11
±0.05
0.01
±0.01
87.45
±42.11
2.71
±0.48
1.59
±0.22
0.02
±0.01
0.02
±0.01
69.53
± 22.27
2.34
±0.24
1.34
±0.16
0.01
±0.01
0.03
±0.01
75.00
± 22.76
4.53
±1.18
1.00
±0.19
0.08
±0.05
0.03
±0.01
128.98
±45.22
3.36
±0.48
1.23
±0.11
0.05
±0.02
0.02
±<0.01
90.08
± 17.04
Plant City, Florida - SYFL
Acetaldehyde
Formaldehyde
Benzo(a)pyrene3
Hexavalent Chromium3
Naphthalene3
61/61
61/61
15/60
34/57
60/60
1.68
±0.36
1.65
±0.33
0.03
±0.02
0.01
±0.01
42.41
± 13.63
1.71
±0.33
3.50
±0.99
0.01
±0.01
0.02
±0.01
38.80
±9.98
1.41
±0.47
3.58
±1.46
0
0.01
±0.01
41.46
±11.87
1.23
±0.20
2.23
±0.43
0.03
±0.02
0.01
±<0.01
50.68
± 13.05
1.51
±0.18
2.76
±0.49
0.02
±0.01
0.01
±<0.01
43.38
±5.88
1 Average concentrations provided below the black line for this site and/or pollutant are presented in
ng/m3 for ease of viewing.
9-33
-------
Table 9-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest
for the Florida Monitoring Sites (Continued)
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Ug/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(Ug/m3)
4th
Quarter
Average
(Ug/m3)
Annual
Average
(Ug/m3)
Winter Park, Florida - ORFL
Acetaldehyde
Formaldehyde
60/60
60/60
1.79
±0.55
1.18
±0.34
1.08
±0.25
2.47
±0.40
1.09
±0.46
2.25
±0.44
1.87
±0.60
1.73
±0.54
1.45
±0.25
1.92
±0.24
Orlando, Florida - PAFLa
Arsenic (PM10)a
Bery Ilium (PM10)a
Cadmium (PM10)a
Lead (PM10)a
Manganese (PM10)a
Nickel (PM10)a
30/30
30/30
30/30
30/30
30/30
30/30
0.73
±0.26
<0.01
±<0.01
0.08
±0.03
4.67
±4.91
2.08
±0.78
0.89
±0.38
0.43
±0.13
<0.01
±<0.01
0.06
±0.02
3.41
±2.68
2.95
±0.97
1.59
±0.72
0.43
±0.13
0.01
±0.01
0.04
±0.01
1.87
±1.78
4.53
±3.46
0.78
±0.18
0.71
±0.24
<0.01
±<0.01
0.07
±0.02
2.48
±1.49
2.41
±1.29
0.91
±0.32
0.57
±0.10
<0.01
±<0.01
0.06
±0.01
3.08
±1.34
3.04
±0.97
1.05
±0.24
1 Average concentrations provided below the black line for this site and/or pollutant are presented in
ng/m3 for ease of viewing.
Observations from Table 9-5 include the following:
• SKFL's annual average concentration of acetaldehyde is the highest annual average
concentration among the Florida sites. The annual average concentrations of
acetaldehyde range from 1.45 ± 0.25 |ig/m3 (ORFL) to 3.36 ± 0.48 |ig/m3 (SKFL).
• The first and fourth quarter acetaldehyde averages for both AZFL and SKFL are
greater than the other quarterly average concentrations and have relatively large
confidence intervals associated with them. For AZFL, each of the three
concentrations (out of 61) greater than 6 |ig/m3 was measured during the first and
fourth quarters of 2010. Of the seven concentrations (also out of 61) greater than
6 |ig/m3 measured at SKFL, all were measured during the first and fourth quarters of
2010.
• Formaldehyde is the only other pollutant with annual average concentration greater
than 1 |ig/m3. The annual average concentrations of formaldehyde range from
1.23 ± 0.11 |ig/m3 (SKFL) to 2.76 ± 0.49 |ig/m3 (SYFL).
• The second and third quarter formaldehyde averages for SYFL are greater than the
other quarterly average concentrations and have relatively large confidence intervals
associated with them. For SYFL, the two highest concentrations of formaldehyde
were measured in July (13.9 |ig/m3), which is the seventh highest formaldehyde
9-34
-------
concentration among all NMP sites sampling this pollutant, and June (9.21 |ig/m3).
The next highest measurement, also measured in June, is 4.60 |ig/m3.
• As previously discussed, SKFL and SYFL both sampled hexavalent chromium and
PAH in addition to carbonyl compounds. Hexavalent chromium, naphthalene, and
benzo(a)pyrene are all pollutants of interest for these two sites. The annual average
concentrations of these three pollutants are roughly twice as high at SKFL than at
SYFL.
• Concentrations of benzo(a)pyrene appear higher during the colder months at SKFL.
Of the nine concentrations of this pollutant greater than 0.1 ng/m3, all were measured
during the first and fourth quarters of 2010 (five during the first quarter and four
during the fourth quarter). Similarly, naphthalene appears to exhibit a quarterly trend,
with higher quarterly averages for the colder months of the year. However, the
confidence intervals indicate that this difference is not statistically significant.
• For PAFL, lead and manganese have the highest annual average concentrations
among the PMio metals. The first quarter lead average for 2010 is much higher than
the other quarterly averages and also has a large confidence interval. A review of the
data shows that the highest concentration of lead was measured on March 9, 2010
(17.6 ng/m3) and is 50 percent higher than the next highest concentration measured on
April 2, 2010 (11.7 ng/m3). These are the only two measurements of lead greater than
10 ng/m3.
• For PAFL, the third quarter manganese average for 2010 is much higher than the
other quarterly averages and also has a large confidence interval. A review of the data
shows that two highest concentrations of manganese were measured during the
summer (13.9 ng/m3 on August 12, 2010 and 8.20 ng/m3 on July 19, 2010). The next
highest concentration was measured in October (5.39 ng/m3).
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for the Florida
sites from those tables include the following:
• As shown in Table 4-10, SKFL and AZFL rank second and third highest,
respectively, for acetaldehyde among all NMP sites sampling carbonyl compounds.
• PAFL's annual average concentration of beryllium is the third highest among NMP
sites sampling PMio metals, as shown in Table 4-12. PAFL also has the fourth highest
annual average concentration of lead and the fifth highest annual average
concentration of arsenic and nickel. Note however, that only nine NMP sites sampled
metals.
9-35
-------
9.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for acetaldehyde and
formaldehyde were created for AZFL, SKFL, SYFL, and ORFL. Box plots were also created for
benzo(a)pyrene, hexavalent chromium, and naphthalene for SKFL and SYFL and for arsenic and
manganese for PAFL. Figures 9-23 through 9-29 overlay the sites' minimum, annual average,
and maximum concentrations onto the program-level minimum, first quartile, average, median,
third quartile, and maximum concentrations, as described in Section 3.5.3.
Figure 9-23. Program vs. Site-Specific Average Acetaldehyde Concentration
-
—
Concentration (ng/m3)
Program: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Average
Site: Site Average Site Minimum/Maximum
o —
9-36
-------
Figure 9-24. Program vs. Site-Specific Average Arsenic (PMi0) Concentration
•-
!
0 0.5 1 1.5 2 2.5 3 3.5
Concentration (ng/m3)
4
Program: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Ave
• D D D
Site: Site Average Site Minimum/Maximum
o —
4.5 5
'rage
Figure 9-25. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
:
t
\-
i
i
3 0.2 0.4 0.6 0.8 1 1.2 1.4
Concentration (ng/m3)
Program: IstQuartile 2nc
•
Site: Site Average Sitf
O
Program Max Cnnrpntratinn = 47.7 n0/m3
i
Program Max Concentration = 42.7 ng/m3
|
1.6
Quartile 3rd Quartile 4th Quartile Avc
D D D
; Minimum/Maximum
i
'rage
8 2
9-37
-------
Figure 9-26. Program vs. Site-Specific Average Formaldehyde Concentration
1 1
1 1 1 1 1
V-
•
r r
25 30 35
Concentration (ng/m3)
50 55
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 9-27. Program vs. Site-Specific Average Hexavalent Chromium Concentration
F
Program Max Concentration = 3.51
ng/m3 j
E
Program Max Concentration = 3.51 ng/m3 j
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
9-38
-------
Figure 9-28. Program vs. Site-Specific Average Manganese (PMi0) Concentration
r
100 120
Concentraticn (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
o
Figure 9-29. Program vs. Site-Specific Average Naphthalene Concentration
,
] i
E
600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
Observations from Figures 9-23 through 9-29 include the following:
• Figure 9-23 for acetaldehyde shows that annual average concentrations for two sites
(AZFL and SKFL) are greater than the program-level average while annual averages for
the other two sites are less than the program-level average. SKFL's maximum
concentration is the second highest concentration measured among NMP sites sampling
acetaldehyde. There were no non-detects of acetaldehyde measured at the Florida sites.
• Figure 9-24 for arsenic shows that PAFL's annual average concentration is similar to the
program-level average concentration. The minimum arsenic concentration measured at
PAFL is just greater than the program-level first quartile (or 25th percentile).
9-39
-------
• Figure 9-25 is the box plot for benzo(a)pyrene. Note that the program-level maximum
concentration (42.7 ng/m3) is not shown directly on the box plot because the scale of the
box plot would be too large to readily observe data points at the lower end of the
concentration range. Thus, the scale has been reduced to 2 ng/m3. Also note that the first
quartile for this pollutant is zero and is not visible on this box plot. This box plot shows
that the annual average concentration for SKFL is greater than the annual average
concentration for SYFL and that both annual average concentrations are less than the
program-level average concentration. Figure 9-25 also shows that the maximum
concentration measured at SKFL is well below the maximum concentration measured
across the program. The maximum concentration measured at SYFL is less than the
program-level average concentration.
• Figure 9-26 for formaldehyde shows that the annual average concentrations of
formaldehyde for AZFL, SKFL, and ORFL are less than the program-level average while
SYFL's annual average is just greater than the program-level average. Note that the range
of formaldehyde concentrations measured at AZFL, SKFL, and ORFL is relatively small
compared to SYFL and at the program level.
• Figure 9-27 is the box plot for hexavalent chromium, which was measured at SKFL and
SYFL. Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 9-27 as a result of a relatively large maximum concentration. The program-level
maximum concentration (3.51 ng/m3) is not shown directly on the box plot in order to
allow for observation of data points at the lower end of the concentration range; thus, the
scale has been reduced to 0.75 ng/m3. Both sites' annual average concentrations are less
than the program-level average and SYFL's annual average concentration is also less
than the program-level median concentration. The maximum concentrations measured at
SKFL and SYFL are well below the maximum concentration measured among all NMP
sites sampling this pollutant. Note that the maximum concentration measured at SYFL is
just greater than the average concentration measured at the program level.
• Figure 9-28 for manganese shows that PAFL's annual average concentration is less than
the program-level average concentration. PAFL's annual average manganese
concentration is roughly half the program-level average. The maximum manganese
concentration measured at PAFL is well below the maximum concentration measured
among all NMP sites sampling PMio metals.
• Figure 9-29 is the box plot for naphthalene. This box plot shows that the annual average
concentration for SKFL is greater than the annual average concentration for SYFL and
just below the program-level average concentration. The range of concentrations
measured at SKFL is much larger than the range of concentrations measured at SYFL.
The maximum concentration measured at SYFL is just greater than the program-level
average concentration.
9-40
-------
9.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. AZFL, ORFL, SKFL, and SYFL have sampled carbonyl compounds as part of the
NMP for at least 5 consecutive years. Thus, Figures 9-30 through 9-37 present the 3-year rolling
statistical metrics for acetaldehyde and formaldehyde for each of these sites. In addition, SYFL
has sampled hexavalent chromium since 2005; thus, Figure 9-38 presents the 3-year rolling
statistical metrics for hexavalent chromium for SYFL. The statistical metrics presented for
assessing trends include the substitution of zeros for non-detects.
Observations from Figure 9-30 for acetaldehyde measurements at AZFL include the
following:
• Carbonyl compounds have been measured at AZFL since 2001, making this site one
of the longest running UATMP sites.
• The maximum acetaldehyde concentration was measured in 2010, but a similar
concentration was also measured in 2003.
• The rolling average and median concentrations increased through the 2003-2005 time
frame then began to decrease significantly. The rolling average began to increase for
the last two time frames, although the median did not begin to increase again until the
2008-2010 time frame. Even with these increases, the rolling average concentrations
remained below the levels from the earlier years.
• Prior to 2010, there had been 17 concentrations of acetaldehyde greater than 5 |ig/m3
measured at AZFL (six of them were measured in 2003, five in 2004, and six in 2009,
with none measured in the years in between). With the addition of 2010 data, that
number increases to 24, with seven concentrations greater than 5 |ig/m3 measured in
2010.
• With the exception of the 2001-2003 time frame, the minimum concentration for each
3-year period is greater than zero. Only two non-detects of acetaldehyde have been
reported since the onset of carbonyl compound sampling (both in 2001).
9-41
-------
Figure 9-30. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at AZFL
...*••
2001-2003 2002-2004 2003-2005
2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three- Year Period
• 5th Pel (entile - Minimum — Median — Maximum • 95th Percentile ...*.. Averag.
Figure 9-31. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at AZFL
L
B
o
i
^
T T T
r^i
ta ^m
^bi i M^B ^av
••• •••
> — ' — • — ' — » — '
-1 -y- -p i L i -«- -*-
2001-2003 2002-2004 2003-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Percentile - Minimum — Median - Maximum • 95thPeirentile ...+.. Average
9-42
-------
Figure 9-32. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at ORFL
-------
Figure 9-34. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at SKFL
2006-2008
Three-Year Period
5th F'ercentile - Minimum — Median — Maximum • 95th Percentile ...*.. Averag.
Sampling for carbonyl compounds at SKFL began in July 2004.
Figure 9-35. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at SKFL
90
80
70
«f 60
f "
3 40 -
30
20
10
0
2004-2006 1 1005-2007 2006-2008 2007-2009 2008-2010
Three- Year Period
• 5th Percentile — Minimum — Median — Maximum • 95th Pertentile .-.^.. Average
Sampling for carbonyl compounds at SKFL began in July 2004.
9-44
-------
Figure 9-36. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at SYFL
"10
c
o
1
2
^*m
.
2004-2006
• 5th Percentilt
•*•
1
2005-2007
— Minimum
••••»
1
2006-2008
Three-Year Period
— Median
-
Maximum
•
•"••»•••..
I 1
2007-2009 2008 2030
• 95thPercentile ...*.. Average
Figure 9-37. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at SYFL
25
1 M
entrationfug
c
3
5 -
•ri
1 2 1 i * ' i J.
2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Pert en tile - Minimum — Median - Maximum • 95th Pert entile ..- + .- Average
9-45
-------
Figure 9-38. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at SYFL
5th P ere en tile - Minimum - Median - Maximum • 95th Percentile ...... Average
Observations from Figure 9-31 for formaldehyde measurements at AZFL include the
following:
• The maximum formaldehyde concentration was measured in 2001, after which the
highest concentration measured decreased by nearly half. The three highest
concentrations of formaldehyde ranged from 16.1 to 9.30 |ig/m3 and were all
measured in 2001.
• The rolling average concentration decreased through the 2003-2005 time frame,
increased through 2006-2008, then began to decrease again. The median
concentrations follow a similar pattern.
• The trends for formaldehyde in Figure 9-31 are almost the opposite of the trends
shown for acetaldehyde in Figure 9-30.
• The minimum concentration for each 3-year period is greater than zero. No non-
detects of formaldehyde have been reported since the onset of carbonyl compound
sampling in 2001.
Observations from Figure 9-32 for acetaldehyde measurements at ORFL include the
following:
• Carbonyl compounds have been measured at ORFL since April 2003.
9-46
-------
• The maximum acetaldehyde concentration was measured in 2006.
• The rolling average concentrations exhibit a slight decreasing trend beginning with
the 2005-2007 time frame. The median concentrations also exhibit this trend.
• The spread of concentrations measured is fairly static, as shown by the 5th and 95th
percentiles.
• The minimum concentration for each 3-year period is greater than zero. No non-
detects of acetaldehyde have been reported since the onset of carbonyl compound
sampling in 2003.
Observations from Figure 9-33 for formaldehyde measurements at ORFL include the
following:
• The maximum formaldehyde concentration was measured in 2007, although
concentrations greater than 10 |ig/m3 have also been measured in 2005 and 2008.
• Even with the relatively high concentrations measured in the later years of sampling,
several of the statistical parameters exhibit a slight decreasing trend over the period of
sampling.
• The minimum concentration for each 3-year period is greater than zero. No non-
detects of formaldehyde have been reported since the onset of carbonyl compound
sampling in 2003.
Observations from Figure 9-34 for acetaldehyde measurements at SKFL include the
following:
• Carbonyl compounds have been measured at SKFL since July 2004.
• The maximum acetaldehyde concentration was measured in 2004 (50.7 |ig/m3) and is
more than five times higher than the next highest measurement (10.3 |ig/m3), which
was measured in 2010. Of the 19 concentrations of acetaldehyde greater than or equal
to 5 |ig/m3, 11 were measured in 2010 (with three in 2004, two in 2009, two in 2008,
and one in 2005).
• Although difficult to discern in Figure 9-34, the rolling average concentration
decreased from 2004-2006 to 2005-2007, then an increasing trend began that
continues into the 2008-2010 time frame. The median and 95th percentiles also exhibit
this pattern.
• The minimum concentration for each 3-year period is greater than zero. No non-
detects of acetaldehyde have been reported since the onset of carbonyl compound
sampling in 2004.
9-47
-------
Observations from Figure 9-35 for formaldehyde measurements at SKFL include the
following:
• Two highest formaldehyde concentrations were measured at SKFL during 2005
(91.7 |ig/m3) and 2004 (70.4 |ig/m3). Aside from these two measurements, all other
concentrations measured at this site were at least an order of magnitude lower. The
high 2004 formaldehyde concentration corresponded with the high acetaldehyde
concentration (both measured on August 31, 2004).
• Although difficult to discern in Figure 9-35, the rolling average and median
concentrations show a steady decreasing trend over the periods shown.
• The minimum concentration for each 3-year period is greater than zero. No non-
detects of formaldehyde have been reported since the onset of carbonyl compound
sampling in 2004.
Observations from Figure 9-36 for acetaldehyde measurements at SYFL include the
following:
• Carbonyl compounds have been measured at SYFL since January 2004.
• The maximum acetaldehyde concentration was measured on January 18, 2007
(15.3 |ig/m3). The next highest concentration, also measured in 2007, is roughly half
of the highest measured concentration (7.55 |ig/m3).
• The rolling average concentrations exhibit an increase from 2004-2006 to 2005-2007,
remain static through 2007-2009, after which a decrease is shown.
• With the exception of the 2004-2006 time frame, the minimum concentration for each
3-year period is greater than zero. Only one non-detect of acetaldehyde has been
reported since the onset of carbonyl compound sampling (2004).
Observations from Figure 9-37 for formaldehyde measurements at SYFL include the
following:
• The highest formaldehyde concentration measured at SKFL was measured in 2005
(32.5 |ig/m3), and was nearly twice the next highest concentration measured in 2008
(17.1 |ig/m3), although several measurements similar in magnitude to this one were
also measured in 2007.
• Both the rolling average and median concentrations show a slight increasing trend
over the periods shown.
• The minimum concentration for each 3-year period is greater than zero. No non-
detects of formaldehyde have been reported since the onset of carbonyl compound
sampling in 2004.
9-48
-------
Observations from Figure 9-38 for hexavalent chromium measurements at SYFL include
the following:
• Hexavalent chromium sampling at SYFL began in January 2005.
• The highest hexavalent chromium concentration measured at SYFL was measured on
July 3, 2005 and is similar in magnitude to the next highest concentrations, measured
on July 4, 2006 and March 17, 2005.
• Both the rolling average and median concentrations exhibit a significant decreasing
trend over the first three 3-year periods shown, as do the other statistical parameters.
Slight increases in the rolling average and 95th percentiles are noted for 2008-2010
while the median continued to decrease.
• The minimum concentrations, 5th percentiles, and median concentrations for 2007-
2009 and 2008-2010 are all zero, indicating that at least 50 percent of the
measurements during those 3-year periods are non-detects.
9.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at each
Florida monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
9.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Florida monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where available.
As described in Section 3.3, acute risk results from exposures of 1 to 14 days; intermediate risk
results from exposures of 15 to 364 days; and chronic risk results from exposures of 1 year or
greater. The preprocessed daily measurements of the pollutants of interest were compared to the
acute MRL; the quarterly averages were compared to the intermediate MRL; and the annual
averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the Florida monitoring sites were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
the Florida monitoring sites.
9-49
-------
9.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Florida sites and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 of this report regarding the
criteria for annual averages and how cancer and noncancer surrogate risk approximations are
calculated). Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer
surrogate risk approximations are presented in Table 9-6, where applicable.
Observations for the Florida sites from Table 9-6 include the following:
• Formaldehyde has the highest cancer surrogate risk approximation among the sites
sampling carbonyl compounds, ranging from 15.93 in-a-million (SKFL) to 35.82 in-
a-million (for SYFL). The cancer surrogate risk approximations for formaldehyde are
an order of magnitude higher than any of the other cancer surrogate risk
approximations for AZFL, ORFL, SKFL, and SYFL.
• For PAFL, arsenic has the highest cancer risk approximation (2.44 in-a-million). The
cancer surrogate risk approximations are less than 1.0 in-a-million for the remaining
pollutants, where a cancer URE is available.
• For the two sites sampling PAH and hexavalent chromium in addition to carbonyl
compounds, naphthalene has the third highest cancer risk approximations for each
site, behind formaldehyde and acetaldehyde. Cancer risk approximations for
hexavalent chromium and benzo(a)pyrene are less than 1.0 in-a-million for both sites.
• All of the noncancer risk approximations for the site-specific pollutants of interest are
less than 1.0, indicating no risk of noncancer health effects.
9-50
-------
Table 9-6. Cancer and Noncancer Surrogate Risk Approximations for the Florida
Monitoring Sites
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(Ug/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
St. Petersburg, Florida - AZFL
Acetaldehyde
Formaldehyde
0.0000022
0.000013
0.009
0.0098
61/61
61/61
2.94
±0.40
1.71
±0.18
6.47
22.19
0.33
0.17
Pinellas Park, Florida - SKFL |
Acetaldehyde
Benzo(a)pyrene
Formaldehyde
Hexavalent Chromium
Naphthalene
0.0000022
0.00176
0.000013
0.012
0.000034
0.009
0.0098
0.0001
0.003
61/61
36/59
61/61
52/61
59/59
3.36
±0.48
O.01
±<0.01
1.23
±0.11
O.01
±<0.01
0.09
±0.02
7.40
0.09
15.93
0.28
3.06
0.37
0.13
O.01
0.03
Plant City, Florida - SYFL
Acetaldehyde
Benzo(a)pyrene
Formaldehyde
Hexavalent Chromium
Naphthalene
0.0000022
0.00176
0.000013
0.012
0.000034
0.009
0.0098
0.0001
0.003
61/61
15/60
61/61
34/57
60/60
1.51
±0.18
O.01
±<0.01
2.76
±0.49
O.01
±<0.01
0.04
±0.01
3.31
0.03
35.82
0.14
1.47
0.17
0.28
O.01
0.01
Winter Park, Florida - ORFL
Acetaldehyde
Formaldehyde
0.0000022
0.000013
0.009
0.0098
60/60
60/60
1.45
±0.25
1.92
±0.24
3.18
25.02
0.16
0.20
Orlando, Florida - PAFL
Arsenic (PM10)
Beryllium (PM10)
Cadmium (PM10)
Lead (PM10)
Manganese (PM10)
Nickel (PM10)
0.0043
0.0024
0.0018
0.00048
0.000015
0.00002
0.00001
0.00015
0.00005
0.00009
30/30
30/30
30/30
30/30
30/30
30/30
O.01
±0.01
O.01
±<0.01
0.01
±0.01
O.01
±0.01
0.01
±0.01
O.01
±0.01
2.44
0.01
0.11
0.51
0.04
O.01
0.01
0.02
0.06
0.01
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 9-
9-51
-------
9.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 9-7 and 9-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 9-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 9-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 9-7 and 9-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. Further, the cancer and noncancer surrogate risk approximations based on each site's
annual averages are limited to those pollutants for which each respective site sampled. As
discussed in Section 9.3, AZFL and ORFL sampled for carbonyl compounds only; SKFL and
SYFL sampled hexavalent chromium and PAH in addition to carbonyl compounds; and PAFL
sampled only PMio metals. In addition, the cancer and noncancer surrogate risk approximations
are limited to those pollutants with enough data to meet the criteria for annual averages to be
calculated. A more in-depth discussion of this analysis is provided in Section 3.5.5.3.
9-52
-------
Table 9-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Florida Monitoring Sites
Top 10 Total Emissions for Pollutants
with Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
St. Petersburg, Florida (Pinellas County) - AZFL
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
Nickel, PM
Tetrachloroethylene
430.92
261.69
214.09
118.05
59.55
26.34
10.63
3.58
2.82
0.84
Benzene
Formaldehyde
1,3 -Butadiene
Nickel, PM
Hexavalent Chromium, PM
Naphthalene
Ethylbenzene
Arsenic, PM
POM, Group 2b
Acetaldehyde
3.36E-03
2.78E-03
1.79E-03
1.35E-03
9.45E-04
8.96E-04
6.54E-04
4.13E-04
3.15E-04
2.60E-04
Formaldehyde
Acetaldehyde
22.19
6.47
Pinellas Park, Florida (Pinellas County) - SKFL
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
Nickel, PM
Tetrachloroethylene
430.92
261.69
214.09
118.05
59.55
26.34
10.63
3.58
2.82
0.84
Benzene
Formaldehyde
1,3 -Butadiene
Nickel, PM
Hexavalent Chromium, PM
Naphthalene
Ethylbenzene
Arsenic, PM
POM, Group 2b
Acetaldehyde
3.36E-03
2.78E-03
1.79E-03
1.35E-03
9.45E-04
8.96E-04
6.54E-04
4.13E-04
3.15E-04
2.60E-04
Formaldehyde
Acetaldehyde
Naphthalene
Hexavalent Chromium
Benzo(a)pyrene
15.93
7.40
3.06
0.28
0.09
-------
Table 9-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Florida Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants
with Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Plant City, Florida (Hillsborough County) - SYFL
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
Methyl tert butyl ether
POM, Group 2b
Nickel, PM
545.24
339.12
333.41
174.74
77.10
37.70
11.51
8.62
5.02
2.88
Formaldehyde
Benzene
1,3 -Butadiene
Hexavalent Chromium, PM
Nickel, PM
Naphthalene
Arsenic, PM
Ethylbenzene
POM, Group 2b
Acetaldehyde
4.33E-03
4.25E-03
2.31E-03
1.44E-03
1.38E-03
1.28E-03
1.01E-03
8.48E-04
4.42E-04
3.84E-04
Formaldehyde
Acetaldehyde
Naphthalene
Hexavalent Chromium
Benzo(a)pyrene
35.82
3.31
1.47
0.14
0.03
Winter Park, Florida (Orange County) - ORFL
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
Tetrachloroethylene
Propylene oxide
581.25
350.51
334.02
169.72
80.64
34.81
10.25
6.17
2.34
1.17
Formaldehyde
Benzene
1,3 -Butadiene
Naphthalene
Ethylbenzene
Hexavalent Chromium, PM
POM, Group 2b
Acetaldehyde
POM, Group 3
Arsenic, PM
4.56E-03
4.53E-03
2.42E-03
1.18E-03
8.35E-04
6.86E-04
5.43E-04
3.73E-04
3.57E-04
2.84E-04
Formaldehyde
Acetaldehyde
25.02
3.18
-------
Table 9-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Florida Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants
with Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Orlando, Florida (Orange County) - PAFL
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
Tetrachloroethylene
Propylene oxide
581.25
350.51
334.02
169.72
80.64
34.81
10.25
6.17
2.34
1.17
Formaldehyde
Benzene
1,3 -Butadiene
Naphthalene
Ethylbenzene
Hexavalent Chromium, PM
POM, Group 2b
Acetaldehyde
POM, Group 3
Arsenic, PM
4.56E-03
4.53E-03
2.42E-03
1.18E-03
8.35E-04
6.86E-04
5.43E-04
3.73E-04
3.57E-04
2.84E-04
Arsenic
Nickel
Cadmium
Beryllium
2.44
0.51
0.11
0.01
-------
Table 9-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Florida Monitoring Sites
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
St. Petersburg, Florida (Pinellas County) - AZFL
Toluene
Xylenes
Methanol
Benzene
Hexane
Ethylbenzene
Formaldehyde
Acetaldehyde
Styrene
Ethylene glycol
1,374.54
1,035.91
540.89
430.92
353.10
261.69
214.09
118.05
113.79
73.16
Acrolein
Nickel, PM
1,3 -Butadiene
Formaldehyde
Benzene
Acetaldehyde
Xylenes
Manganese, PM
Naphthalene
Lead, PM
625,802.58
31,311.86
29,777.05
21,845.80
14,364.15
13,116.58
10,359.11
10,118.05
8,780.20
6,986.71
Acetaldehyde 0.33
Formaldehyde 0.17
Pinellas Park, Florida (Pinellas County) - SKFL
Toluene
Xylenes
Methanol
Benzene
Hexane
Ethylbenzene
Formaldehyde
Acetaldehyde
Styrene
Ethylene glycol
1,374.54
1,035.91
540.89
430.92
353.10
261.69
214.09
118.05
113.79
73.16
Acrolein
Nickel, PM
1,3 -Butadiene
Formaldehyde
Benzene
Acetaldehyde
Xylenes
Manganese, PM
Naphthalene
Lead, PM
625,802.58
31,311.86
29,777.05
21,845.80
14,364.15
13,116.58
10,359.11
10,118.05
8,780.20
6,986.71
Acetaldehyde 0.37
Formaldehyde 0.13
Naphthalene 0.03
Hexavalent Chromium <0.01
-------
Table 9-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Florida Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Plant City, Florida (Hillsborough County) - SYFL
Toluene
Xylenes
Methanol
Benzene
Hexane
Cyanide Compounds, gas
Hydrochloric acid
Ethylbenzene
Formaldehyde
Acetaldehyde
1,643.63
1,310.35
688.28
545.24
438.64
428.19
350.01
339.12
333.41
174.74
Acrolein
Cyanide Compounds, gas
1,3 -Butadiene
Formaldehyde
Nickel, PM
Manganese, PM
Acetaldehyde
Benzene
Hydrochloric acid
Arsenic, PM
851,462.75
535,232.13
38,551.07
34,021.47
31,976.42
23,688.87
19,415.44
18,174.77
17,500.47
15,600.20
Formaldehyde 0.28
Acetaldehyde 0.17
Naphthalene 0.01
Hexavalent Chromium O.01
Winter Park, Florida (Orange County) - ORFL
Toluene
Xylenes
Methanol
Benzene
Hexane
Formaldehyde
Ethylbenzene
Acetaldehyde
Styrene
Ethylene glycol
1,693.63
1,316.88
629.52
581.25
434.13
350.51
334.02
169.72
164.64
81.80
Acrolein
1,3 -Butadiene
Formaldehyde
Benzene
Acetaldehyde
Xylenes
Cyanide Compounds, gas
Naphthalene
Arsenic, PM
Lead, PM
870,786.72
40,319.80
35,765.86
19,374.99
18,857.75
13,168.77
12,178.13
11,604.18
4,404.23
3,678.02
Formaldehyde 0.20
Acetaldehyde 0.16
-------
Table 9-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Florida Monitoring Sites (Continued)
oo
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity- Weighted
Emissions
(County-Level)
Pollutant
Orlando, Florida (Oran
Toluene
Xylenes
Methanol
Benzene
Hexane
Formaldehyde
Ethylbenzene
Acetaldehyde
Styrene
Ethylene glycol
1,693.63
1,316.88
629.52
581.25
434.13
350.51
334.02
169.72
164.64
81.80
Acrolein
1,3 -Butadiene
Formaldehyde
Benzene
Acetaldehyde
Xylenes
Cyanide Compounds, gas
Naphthalene
Arsenic, PM
Lead, PM
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
?e County) - PAFL
870,786.72
40,319.80
35,765.86
19,374.99
18,857.75
13,168.77
12,178.13
11,604.18
4,404.23
3,678.02
Manganese
Arsenic
Lead
Nickel
Cadmium
Beryllium
0.06
0.04
0.02
0.01
0.01
O.01
-------
Observations from Table 9-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in Pinellas, Hillsborough, and Orange Counties, although not
necessarily in that order.
• Benzene, formaldehyde, and 1,3-butadiene have the highest toxi city-weighted
emissions for all three counties.
• For Pinellas and Orange Counties, seven of the highest emitted pollutants also have
the highest toxicity-weighted emissions. Eight pollutants also have the highest
toxicity-weighted emissions for Hillsborough County.
• Hexavalent chromium and arsenic are among the pollutants with the highest cancer
toxicity-weighted emissions for each county, yet are not among the highest emitted
pollutants in any of the counties.
• Formaldehyde, which has the highest cancer risk approximations for all sites
sampling carbonyl compounds, is one of the highest emitted pollutants and has one of
the highest toxicity-weighted emissions for each county.
• PAFL sampled only PMio metals; arsenic and nickel have the highest cancer risk
approximations for this site. Arsenic appears on the list of 10 highest toxicity-
weighted emissions for Orange County, but does not appear on the list of highest
pollutants emitted, indicating the relative toxicity of a low quantity of emissions.
Observations from Table 9-8 include the following:
• Toluene, xylenes, and methanol are the highest emitted pollutants with noncancer
RfCs in all three Florida counties.
• Acrolein has the highest toxicity-weighted emissions of the pollutants with noncancer
RfCs for each county, but does not appear in any county's list of 10 highest emitted
pollutants.
• For Pinellas and Orange Counties, four of the highest emitted pollutants also have the
highest toxicity-weighted emissions. Five pollutants also have the highest toxicity-
weighted emissions for Hillsborough County.
• Formaldehyde and acetaldehyde appear on all three lists for each site/county, with the
exception of PAFL, where carbonyl compounds were not sampled. For PAFL, arsenic
and lead appear on two of the three lists. No metals are among the highest emitted
pollutants in Orange County.
9-59
-------
9.6 Summary of the 2010 Monitoring Data for the Florida Sites
Results from several of the data treatments described in this section include the
following:
»«» Acetaldehyde and formaldehyde failed screens for every site sampling carbonyl
compounds (AZFL, SKFL, SYFL, and ORFL). In addition to acetaldehyde and
formaldehyde, naphthalene also failed screens for SKFL and SYFL, the only two
Florida sites at which PAH were sampled. Four metals failed screens for PAFL.
»«» Acetaldehyde had the highest annual average concentration of any of the pollutants
of interest among the Florida sites, which was calculated for SKFL. Formaldehyde
was the only other pollutant of interest with an annual average concentration greater
than 1 jug/m .
»«» The annual average concentrations of acetaldehyde for SKFL and AZFL ranked
second and third highest among allNMP sites sampling carbonyl compounds;
PAFL's annual average concentration of beryllium was the third highest among NMP
sites sampling PM 10 metals.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest for the Florida sites were greater
than their associatedMRL noncancer health risk benchmarks.
9-60
-------
10.0 Site in Georgia
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in Georgia, and integrates these concentrations with
emissions, meteorological, and risk information. Data generated by sources other than ERG are
not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding on the various data analyses presented below.
10.1 Site Characterization
This section characterizes the SDGA monitoring site by providing geographical and
physical information about the location of the site and the surrounding area. This information is
provided to give the reader insight regarding factors that may influence the air quality near the
site and assist in the interpretation of the ambient monitoring measurements.
The SDGA monitoring is located in Decatur, Georgia, a suburb of Atlanta. Figure 10-1 is
a composite satellite image retrieved from ArcGIS Explorer showing the monitoring site in its
urban location. Figure 10-2 identifies point source emissions locations by source category, as
reported in the 2008 NEI for point sources. Note that only sources within 10 miles of the site are
included in the facility counts provided below the map in Figure 10-2. Thus, sources outside the
10-mile radius have been grayed out, but are visible on the map to show emissions sources
outside the 10-mile boundary. A 10-mile boundary was chosen to give the reader an indication of
which emissions sources and emissions source categories could potentially have a direct effect
on the air quality at the monitoring site. Further, this boundary provides both the proximity of
emissions sources to the monitoring site as well as the quantity of such sources within a given
distance of the site. Table 10-1 describes the area surrounding the monitoring site by providing
supplemental geographical information such as land use, location setting, and locational
coordinates.
10-1
-------
Figure 10-1. Decatur, Georgia (SDGA) Monitoring Site
-------
Figure 10-2. NEI Point Sources Located Within 10 Miles of SDGA
Legend
•^ SDGA NATTS site
64'10'0*W
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
10 mile radius
County boundary
Source Category Group (No. of Facilities)
+ Aircraft Operations (22)
6 Bakery (1)
© Fabricated Metal Products (1)
--- Flexible Polyurethane Foam Production (2)
F Food Processing/Agriculture (1)
V Glass Manufacturing (1)
® Institutional-school (1)
• Landfill (2)
? Miscellaneous Commercial/lndustrial (3)
• Oil and/or Gas Production (1)
S Surface Coating (1)
10-3
-------
Table 10-1. Geographical Information for the Georgia Monitoring Site
Site
Code
SDGA
AQS Code
13-089-0002
Location
Decatur
County
DeKalb
Micro- or
Metropolitan
Statistical Area
Atlanta-Sandy
Springs-Marietta,
CrA
Latitude
and
Longitude
33.688007,
-84.290325
Land Use
Residential
Location
Setting
Suburban
Additional Ambient Monitoring Information1
CO, SO2, NOy, NO, NO2, NOX, PAMS, Carbonyl
compounds, VOC, O3, Meteorological parameters,
PMio, PM Coarse, PM10 Speciation, Black carbon,
PM2 5, and PM2 5 Speciation, Haze.
BOLD ITALICS = EPA-designated NATTS Site.
-------
SDGA is located on the DeKalb County Schools Environmental Education property off
Wildcat Road and is the South DeKalb NATTS site. Figure 10-1 shows that residential
subdivisions, a greenhouse and horse barn, an athletic field, and a high school surround the
monitoring site. A golf course backs up against the school property. Interstate-285 is located less
than 1 mile north of the site. As Figure 10-2 shows, only one point source (a bakery) is located in
close proximity to SDGA. Additional sources are located primarily on the west side of the
10-mile radius. The aircraft operations source category (which includes airports as well as small
runways, heliports, or landing pads) is the source category with the highest number of emissions
sources within 10 miles of SDGA.
Table 10-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the Georgia
monitoring site. Table 10-2 also includes a vehicle registration-to-county population ratio
(vehicles-per-person). In addition, the population within 10 miles of the site is presented. An
estimate of 10-mile vehicle ownership was calculated by applying the county-level vehicle
registration-to-population ratio to the 10-mile population surrounding the monitoring site.
Table 10-2 also contains annual average daily traffic information. Finally, Table 10-2 presents
the daily VMT for DeKalb County.
Table 10-2. Population, Motor Vehicle, and Traffic Information for the Georgia
Monitoring Site
Site
SDGA
Estimated
County
Population1
692,902
County-level
Vehicle
Registration2
472,535
Vehicles
per Person
(Registration:
Population)
0.68
Population
within 10
miles3
793,817
Estimated
10-mile
Vehicle
Ownership
541,355
Annual
Average
Daily
Traffic4
145,890
County-
level
Daily
VMT5
21,057,000
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2011 data from the GA Dept of Revenue (GA DOR, 2011)
310-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2010 data from the Georgia DOT (GA DOT, 20 lOa)
5 County-level VMT reflects 2010 data for all public roads from the Georgia DOT (GA DOT, 2010b)
BOLD ITALICS = EPA-designated NATTS Site.
10-5
-------
Observations from Table 10-2 include the following:
• SDGA's county-level population and vehicle registration are in the middle of the
range compared to other counties with NMP sites. The same is also true for its
10-mile population and estimated vehicle ownership.
• The vehicle-per-person ratio is among the lower ratios compared to other NMP sites.
• The traffic volume experienced near SDGA ranks sixth highest compared to other
NMP monitoring sites. The traffic estimate used came from 1-285, north of Clifton
Spring Road. This is a change in location from the 2008-2009 NMP report.
• The daily VMT for DeKalb County is in the middle of the range among counties with
NMP sites (where VMT data were available).
10.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in Georgia on sample days, as well as over the course of the year.
10.2.1 Climate Summary
Atlanta is the largest city in Georgia, and is located at the base of the Blue Ridge
Mountains. The Gulf of Mexico to the south is the major moisture source for weather systems
that move across the region. Both topographical features, in addition to the Atlantic Ocean to the
east, exert moderating influences on the area's climate, tempering cold air outbreaks from the
north as well as summer heat waves. Summers are warm and humid while winters are relatively
mild, although snow is not uncommon. The semi-permanent Bermuda High Pressure offshore
over the Atlantic Ocean is a dominant weather feature affecting the Atlanta area, which pulls
warm, moist air into the region. Precipitation is ample, although autumn is the driest season
(Bair, 1992 and GSCO, 1998).
10.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest this site were retrieved
for 2010 (NCDC, 2010). The closest weather station to SDGA is located at W. B.
Hartsfield/Atlanta International Airport (WBAN 13874). Additional information about the
Hartsfield weather station, such as the distance between the site and the weather station, is
10-6
-------
provided in Table 10-3. These data were used to determine how meteorological conditions on
sample days vary from normal conditions throughout the year.
Table 10-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 10-3 is the 95
percent confidence interval for each parameter. As shown in Table 10-3, average meteorological
conditions on sample days were representative of average weather conditions throughout the
year.
10.2.3 Back Trajectory Analysis
Figure 10-3 is the composite back trajectory map for days on which samples were
collected at the SDGA monitoring site in 2010. Included in Figure 10-3 are four back trajectories
per sample day. Figure 10-4 is the corresponding cluster analysis for 2010. An in-depth
description of these maps and how they were generated is presented in Section 3.5.2.1. For the
composite map, each line represents the 24-hour trajectory along which a parcel of air traveled
toward the monitoring site on a given sample day and time. For the cluster analysis, each line
corresponds to a back trajectory representative of a given cluster of trajectories. For both maps,
each concentric circle around the site in Figures 10-3 and 10-4 represents 100 miles.
Observations from Figures 10-3 and 10-4 include the following:
• Back trajectories originated from a variety of directions at SDGA.
• The 24-hour air shed domain for SDGA was somewhat smaller in size compared to
other NMP monitoring sites. While the farthest away a trajectory originated was
central Illinois, or just greater than 550 miles away, the average back trajectory length
was 190 miles. Nearly 85 percent of back trajectories originated within 300 miles of
the site. The longest trajectories tended to originate from the northwest, over Indiana,
Illinois, and Missouri.
• The cluster analysis shows that trajectories originating from the southwest (and
within a relatively short distance) were most common. Trajectories also commonly
originated from the northwest to north, as well as the southeast and northeast
quadrants.
10-7
-------
Table 10-3. Average Meteorological Conditions near the Georgia Monitoring Site
Closest NWS Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Decatur, Georgia - SDGA
W.B.
Hartsfield/ Atlanta
Intl Airport
13874
(33.64, -84.43)
8.18
miles
239°
(WSW)
Sample
Day
2010
71.0
±4.6
71.6
± 1.9
61.5
±4.5
62.1
±1.8
47.7
±4.6
48.5
± 1.9
54.1
±4.1
54.8
±1.7
63.2
±2.8
64.2
± 1.4
1016.6
±1.4
1016.7
±0.5
7.2
±0.8
6.8
±0.3
1 Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
o
oo
-------
Figure 10-3. 2010 Composite Back Trajectory Map for SDGA
Figure 10-4. Back Trajectory Cluster Map for SDGA
10-9
-------
10.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at Hartsfield International Airport near
SDGA were uploaded into a wind rose software program to produce customized wind roses, as
described in Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals"
positioned around a 16-point compass, and uses different colors to represent wind speeds.
Figure 10-5 presents three different wind roses for the SDGA monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at this location.
Observations from Figure 10-5 for SDGA include the following:
• The NWS weather station at Hartsfield International Airport is the closest weather
station to SDGA. The weather station is located approximately 8.2 miles west-
southwest of SDGA.
• The historical wind rose shows that winds from the west to north-northwest account
for just greater than one-third of wind observations. Easterly winds were also
common. Winds from the northeast quadrant were rarely observed. Calm
winds (< 2 knots) were observed for nearly 10 percent of the hourly wind
measurements.
• The wind patterns on both full-year and sample day wind roses exhibit the same wind
patterns as those of the historical wind rose, although northwesterly and north-
northwesterly winds were observed for a higher percentage of observations and
easterly winds for a lower percentage than shown on the historical wind rose.
• The wind patterns on the full-year wind rose resemble the sample day wind rose,
indicating the wind conditions on sample days were representative of wind conditions
throughout 2010.
10-10
-------
Figure 10-5. Wind Roses for the Hartsfield International Airport Weather Station near
SDGA
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between SDGA and NWS Station
10-11
-------
10.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for SDGA in order to allow
analysts and readers to focus on a subset of pollutants through the context of risk. Each
pollutant's preprocessed daily measurement was compared to its associated risk screening value.
If the concentration was greater than the risk screening value, then the concentration "failed the
screen." Pollutants of interest are those for which the individual pollutant's total failed screens
contribute to the top 95 percent of the site's total failed screens. In addition, if any of the NATTS
MQO Core Analytes measured by the monitoring site did not meet the pollutant of interest
criteria based on the preliminary risk screening, that pollutant was added to the list of site-
specific pollutants of interest. A more in-depth description of the risk screening process is
presented in Section 3.2.
Table 10-4 presents SDGA's pollutants of interest. The pollutants that failed at least one
screen and contributed to 95 percent of the total failed screens for the monitoring site are shaded.
NATTS MQO Core Analytes are bolded. Thus, pollutants of interest are shaded and/or bolded.
SDGA sampled for PAH and hexavalent chromium only.
Table 10-4. Risk Screening Results for the Georgia Monitoring Site
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Decatur, Georgia - SDGA
Naphthalene
Acenaphthene
Benzo(a)pyrene
0.029
0.011
0.00057
Total
56
1
1
58
59
59
26
144
94.92
1.69
3.85
40.28
96.55
1.72
1.72
96.55
98.28
100.00
Observations from Table 10-4 for SDGA include the following:
• Naphthalene, acenaphthene, and benzo(a)pyrene failed screens. Naphthalene failed
the majority of the screens (roughly 97 percent), accounting for 56 of the 58 total
failed screens; the other two pollutants failed only one screen each.
• Naphthalene was the only pollutant initially identified as a pollutant of interest based
on the risk screening process. Benzo(a)pyrene was added as a pollutant of interest for
SDGA because it is a NATTS MQO Core Analyte. Hexavalent chromium was also
added as a pollutant of interest for SDGA because it is a NATTS MQO Core Analyte,
even though it did not fail any screens. This pollutant is not shown in Table 10-4.
10-12
-------
10.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Georgia monitoring site. Concentration averages are provided for the pollutants of interest
for the SDGA monitoring site, where applicable. Concentration averages for select pollutants are
also presented graphically for the site, where applicable, to illustrate how the site's
concentrations compare to the program-level averages. In addition, concentration averages for
select pollutants are presented from previous years of sampling in order to characterize
concentration trends at the site, where applicable. Additional site-specific statistical summaries
are provided in Appendices M and O.
10.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for SDGA, as described in Section 3.1. The quarterly average of a particular pollutant is simply
the average concentration of the preprocessed daily measurements over a given calendar quarter.
Quarterly average concentrations include the substitution of zeros for all non-detects. A site must
have a minimum of 75 percent valid samples of the total number of samples possible within a
given quarter for a quarterly average to be calculated. An annual average includes all measured
detections and substituted zeros for non-detects for the entire year of sampling. Annual averages
were calculated for pollutants where three valid quarterly averages could be calculated and
where method completeness was greater than or equal to 85 percent, as presented in Section 2.4.
Quarterly and annual average concentrations for the Georgia monitoring site are presented in
Table 10-5, where applicable. Note that if a pollutant was not detected in a given calendar
quarter, the quarterly average simply reflects "0" because only zeros substituted for non-detects
were factored into the quarterly average concentration.
10-13
-------
Table 10-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Georgia Monitoring Site
Pollutant
#of
Measured
Detections vs.
# of Samples
1st
Quarter
Average
(ng/m3)
2nd
Quarter
Average
(ng/m3)
3rd
Quarter
Average
(ng/m3)
4th
Quarter
Average
(ng/m3)
Annual
Average
(ng/m3)
Decatur, Georgia - SDGA
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
26/59
41/60
59/59
0.08
±0.04
0.01
±0.01
99.73
±53.83
0
0.02
±0.01
136.55
±48.93
0.01
±0.01
0.02
±0.01
117.88
±24.51
0.16
±0.14
0.02
±0.01
155.34
± 59.84
0.06
±0.04
0.02
±<0.01
127.84
±23.35
Observations for SDGA from Table 10-5 include the following:
• The annual average concentration of naphthalene is significantly higher than the
annual average concentrations of benzo(a)pyrene and hexavalent chromium. SDGA's
annual average concentration of naphthalene ranks fifth highest among NMP sites
sampling this pollutant (as shown in Table 4-11).
• The confidence intervals associated with naphthalene's quarterly averages indicate
the relative variability in the naphthalene measurements (note how the lowest
quarterly average concentration has one of the highest confidence intervals).
Concentrations of naphthalene measured at SDGA ranged from 19.1 ng/m3 to
322 ng/m3.
• The first and fourth quarter averages of benzo(a)pyrene are much higher than the
other quarterly averages and have relatively large confidence intervals associated with
them, which is indicative of the inclusion of potential outliers. The highest
concentration of this pollutant (1.01 ng/m3) was measured on December 16, 2010 and
is three times greater than the next highest measurement (0.361 ng/m3 measured on
December 10, 2010). The measurement on December 16, 2010 is one of 11
concentrations greater than 1 ng/m3 among all NMP sites sampling this pollutant. At
SDGA, nine concentrations were greater than 0.1 ng/m3; of these, four were measured
during the first quarter of 2010 and five were measured during the fourth quarter of
2010. This pollutant was not detected at all during the second quarter of 2010.
10.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for benzo(a)pyrene,
hexavalent chromium, and naphthalene were created for SDGA. Figures 10-6 through 10-8
overlay the site's minimum, annual average, and maximum concentrations onto the program-
level minimum, first quartile, average, median, third quartile, and maximum concentrations, as
described in Section 3.5.3.
10-14
-------
Figure 10-6. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
1
1 1 1
1 1 !
| Program Max Concentration = 42.7 ng/m3
1 1 1
0.8 1 1.2
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 10-7. Program vs. Site-Specific Average Hexavalent Chromium Concentration
E
1
Program Max Concentration = 3.51 ng/m3
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 10-8. Program vs. Site-Specific Average Naphthalene Concentration
fir
600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
10-15
-------
Observations from Figures 10-6 through 10-8 include the following:
• Figure 10-6 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
SDGA is less than the program-level average concentration. Figure 10-6 also
shows that the maximum concentration measured at SDGA is well below the
maximum concentration measured across the program. However, the maximum
concentration measured at SDGA is the 11th highest measurement of this pollutant
among all NMP sites. However, most of the "high" concentrations of
benzo(a)pyrene across the program ranged from 1-2 ng/m3, with the exception of
the two that exceed the scale in Figure 10-6. Several non-detects of
benzo(a)pyrene were measured at SDGA.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 10-7 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 10-7 shows the annual average concentration of hexavalent chromium for
SDGA is less than the program-level average (by more than half). SDGA's annual
average concentration is also less than the program-level median concentration.
The maximum concentration measured at SDGA is significantly less than the
program-level maximum concentration, but greater than the program-level
average concentration. Several non-detects of hexavalent chromium were
measured at SDGA.
• Figure 10-8 shows that the annual naphthalene average for SDGA is greater than
the program-level average concentration. As discussed previously, the annual
average naphthalene concentration is the fifth highest annual average among
NMP sites sampling this pollutant. However, the maximum naphthalene
concentration measured at SDGA is well below the program-level maximum
concentration. There were no non-detects of naphthalene measured at SDGA.
10.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. SDGA has sampled hexavalent chromium under the NMP since 2005. Thus,
Figure 10-9 presents the 3-year rolling statistical metrics for hexavalent chromium for SDGA.
The statistical metrics presented for assessing trends include the substitution of zeros for non-
detects.
10-16
-------
Figure 10-9. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at SDGA
1
.
(
r
|
:
3
t
^^m
2005-2007 i
• 5th Percentile
2
— Minimum
T T
^^^ -jgj^^
2006-2008 2 2007-2009 2 2008-2010 2
Three- Year Period
- Median - Maximum • 95th Percentile .-#.. Average
Sampling for hexavalent chromium at SDGA began in February 2005.
2Samples were not collected between September 2007 and May 2008.
Observations from Figure 10-9 for hexavalent chromium measurements at SDGA include
the following:
• Sampling for hexavalent chromium began in February 2005 at SDGA.
• The maximum hexavalent chromium concentration was measured on
November 25, 2006 (0.300 ng/m3), and thus appears as the maximum concentration
for the first two 3-year periods. Only five concentrations measured at SDGA were
greater than 0.1 ng/m3 and four of the five were measured in 2006 (and the other in
2005).
• The rolling average concentration exhibits a slight decrease from 2005-2007 to
2006-2008, and a significant decrease is shown from 2006-2008 to 2007-2009,
followed by another slight decrease for 2008-2010. The median concentrations and
95th percentiles exhibit similar trends.
• As denoted in Figure 10-9, there was a gap in sampling from September 2007 to
May 2008.
10-17
-------
10.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the
SDGA monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations regarding
the various risk factors, time frames, and calculations associated with these risk screenings.
10.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
SDGA monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where available.
As described in Section 3.3, acute risk results from exposures of 1 to 14 days; intermediate risk
results from exposures of 15 to 364 days; and chronic risk results from exposures of 1 year or
greater. The preprocessed daily measurements of the pollutants of interest were compared to the
acute MRL; quarterly averages were compared to the intermediate MRL; and annual averages
were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the SDGA monitoring site were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
SDGA.
10.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the SDGA monitoring site and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 10-6, where applicable.
Observations for SDGA from Table 10-6 include the following:
• Naphthalene was the only pollutant of interest for which the cancer risk
approximation was greater than 1.0 in-a-million (4.35 in-a-million).
• Both noncancer risk approximations for naphthalene and hexavalent chromium were
well below 1.0. Benzo(a)pyrene does not have a noncancer RfC.
10-18
-------
Table 10-6. Cancer and Noncancer Surrogate Risk Approximations for the Georgia
Monitoring Site
Pollutant
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
Cancer
URE
(Hg/m3)1
0.00176
0.012
0.000034
Noncancer
RfC
(mg/m3)
—
0.0001
0.003
#of
Measured
Detections
vs. # of
Samples
26/59
41/60
59/59
Annual
Average
(ng/m3)
0.06
± 0.04
0.02
± <0.01
127.84
± 23.35
Cancer Risk
Approximation
(in-a-million)
0.11
0.18
4.35
Noncancer
Risk
Approximation
(HQ)
—
<0.01
0.04
- = A Cancer URE or Noncancer RfC is not available.
10.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 10-7 and 10-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 10-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 10-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 10-7 and 10-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. Further, the cancer and noncancer surrogate risk approximations based on each site's
annual average are limited to those pollutants for which each respective site sampled. As
discussed in Section 10.3, SDGA sampled for PAH and hexavalent chromium. In addition, the
cancer and noncancer surrogate risk approximations are limited to those pollutants with enough
data to meet the criteria for annual averages to be calculated. A more in-depth discussion of this
analysis is provided in Section 3.5.5.3.
10-19
-------
Table 10-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Georgia Monitoring Site
to
o
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Decatur, Georgia (DeKalb County) - SDGA
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
POM, Group la
Methyl tert butyl ether
378.41
213.44
191.48
102.73
46.29
24.10
6.82
3.49
0.49
0.29
Benzene
Formaldehyde
1,3 -Butadiene
Naphthalene
Ethylbenzene
Hexavalent Chromium, PM
POM, Group 2b
Acetaldehyde
Arsenic, PM
POM, Group 5a
2.95E-03
2.49E-03
1.39E-03
8.19E-04
5.34E-04
3.82E-04
3.07E-04
2.26E-04
1.31E-04
1.01E-04
Naphthalene
Hexavalent Chromium
Benzo(a)pyrene
Cancer Risk
Approximation
(in-a-million)
4.35
0.18
0.11
-------
Table 10-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Georgia Monitoring Site
o
to
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Decatur, Georgia (DeKalb County) - SDGA
Toluene
Xylenes
Methanol
Benzene
Hexane
Ethylbenzene
Formaldehyde
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
1,018.74
794.36
430.90
378.41
223.04
213.44
191.48
102.73
56.41
46.29
Acrolein
1,3 -Butadiene
Formaldehyde
Benzene
Acetaldehyde
Naphthalene
Xylenes
Lead, PM
2,4-Toluene diisocyanate
Arsenic, PM
576,341.15
23,142.89
19,539.04
12,613.63
11,414.67
8,033.73
7,943.60
4,023.67
2,140.67
2,029.50
Naphthalene 0.04
Hexavalent Chromium O.01
-------
Observations from Table 10-7 include the following:
• Benzene, ethylbenzene, and formaldehyde are the highest emitted pollutants with
cancer UREs in DeKalb County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
cancer UREs) are benzene, formaldehyde, and 1,3-butadiene.
• Seven of the highest emitted pollutants also have the highest toxi city-weighted
emissions for DeKalb County.
• Naphthalene, which has the highest cancer risk approximation for SDGA, has the
fourth highest toxicity-weighted emissions and sixth highest emissions for DeKalb
County.
• Hexavalent chromium ranks sixth highest for toxicity-based emissions, but is not
among one of the highest emitted pollutants in DeKalb County.
• POM, Group 2b is the eight highest emitted "pollutant" in DeKalb County and ranks
seventh for toxicity-weighted emissions. POM, Group 2b includes several PAH
sampled for at SDGA including acenaphthene, benzo(e)pyrene, fluoranthene, and
perylene. None of the PAH included in POM, Group 2b were identified as pollutants
of interest for SDGA.
• Benzo(a)pyrene is part of POM, Group 5a. POM, Group 5a ranks tenth highest for
toxicity-based emissions, but is not among one of the highest emitted pollutants in
DeKalb County.
Observations from Table 10-8 include the following:
• Toluene, xylenes, and methanol are the highest emitted pollutants with noncancer
RfCs in DeKalb County.
• The pollutants with the highest toxicity-weighted emissions (of the pollutants with
noncancer RfCs) are acrolein, 1,3-butadiene, and formaldehyde.
• Four of the highest emitted pollutants in DeKalb County also have the highest
toxicity-weighted emissions.
• While naphthalene is not one of the 10 highest emitted pollutants with a noncancer
toxicity factor in DeKalb County, its toxicity-weighted emissions rank sixth. Neither
hexavalent chromium nor POM, Group 5a (benzo(a)pyrene) appear on either
emissions-based list.
10-22
-------
10.6 Summary of the 2010 Monitoring Data for SDGA
Results from several of the data treatments described in this section include the
following:
»«» Naphthalene, acenaphthene, and benzo(a)pyrene failed screens for SDGA, although
naphthalene accounted for the bulk of failed screens.
*»* Of the site-specific pollutants of the interest, naphthalene had the highest annual
average concentration for SDGA. SDGA 's annual average concentration of
naphthalene was the fifth highest among NMP sites sampling this pollutant.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
10-23
-------
11.0 Sites in Illinois
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS and UATMP sites in Illinois, and integrates these
concentrations with emissions, meteorological, and risk information. Data generated by sources
other than ERG are not included in the data analyses contained in this report. Readers are
encouraged to refer to Sections 1 through 4 and the glossary (Appendix P) for detailed
discussions and definitions regarding the various data analyses presented below.
11.1 Site Characterization
This section characterizes the Illinois monitoring sites by providing geographical and
physical information about the location of the sites and the surrounding areas. This information
is provided to give the reader insight regarding factors that may influence the air quality near the
sites and assist in the interpretation of the ambient monitoring measurements.
Both Illinois sites are located in northwestern suburbs of Greater Chicago. More
specifically, NBIL is located in Northbrook and SPIL is located in Schiller Park. Figures 11-1
and 11-2 are composite satellite images retrieved from ArcGIS Explorer showing the monitoring
sites in their urban locations. Figure 11-3 identifies point source emissions locations by source
category, as reported in the 2008 NEI for point sources. Note that only sources within 10 miles
of the sites are included in the facility counts provided in Figure 11-3. Thus, sources outside each
10-mile radius have been grayed out, but are visible on the map to show emissions sources
outside the 10-mile boundary. A 10-mile boundary was chosen to give the reader an indication of
which emissions sources and emissions source categories could potentially have a direct effect
on the air quality at the monitoring sites. Further, this boundary provides both the proximity of
emissions sources to the monitoring sites as well as the quantity of such sources within a given
distance of the sites. Table 11-1 describes the area surrounding each monitoring site by providing
supplemental geographical information such as land use, location setting, and locational
coordinates.
11-1
-------
Figure 11-1. Northbrook, Illinois (NBIL) Monitoring Site
i^Wromoll ;W?
$r-?&!|> ;£v, 4\v « ?M5 r^-'-';, qjfoif-
63
-------
Figure 11-2. Schiller Park, Illinois (SPIL) Monitoring Site
-------
Figure 11-3. NEI Point Sources Located Within 10 Miles of NBIL and SPIL
tO'Q-W 86'5'Q-W 88 Q'Q-W $7J55'0~W S7-SO'D~W &J~4$'QrW a7'4Q"Q"W W'KQ'W 87'30'CrW
Legend
-,- - - *s
88'OTTW 87"55"CTW 87"50'D"W 87'45'CTW 97^0'0'W B7"35XTVV
Note: Due lo facility density and collocation, the total facilities
displayed may not represent all: facilities within the area of interest,
NBIL NATTS site
SPIL UATMP Site
Source Category Group (No. of Facilities)
::: AtwastvE Product Manufacturing (2)
•;?-" Air-eondiliomngi'Rer'rfcjjeration (2)
Hh Aircraft Operations (32)
I Asphalt Pf Qcessingi'Roolmg Manufacturing (3)
0 Aulo Body Shop/Painters (3)
ffl Au1amobile>Tfuck Manufacturing (9)
0 Bakery <9)
r Brow&ry/DistilleryArVinery (1)
. : Bnck Manufacturing i Structural Clay (3)
f Building Construction (6)
B Bulk Tertninals/BulJc Plants {10}
C Chemical Manufacturing {20)
• Concrete Batch Plant (20)
[XJ Crematory •AnimaVHuman<12)
0 Dry Cleaning Facility (51)
6 Electrical Equipment (18)
f Electricity Generation via Combustion (6)
E Electroplating, Plating, Polishing Anodizing. & Coloring (58)
4 Engine Testing (2)
10 mile radius |_ J County boundary
M Mss-eeHaneous Manufacturing (86)
• Oil andtorGas Production (1)
CD Pharmaceutical Manufacturing (8)
i Pipeline Compressor Station {1)
1 Primary Metal Production (24)
^« Printing. Coaling & Dyeing of Fabnc <1)
P Printino/Puatehmg [90)
B Pulp and Paper PlantWood Products (18)
R Rubber and Miscelianeaus Plashes Products (25)
2 Secondary Metal Processing <5)
> Solid Waste Disposal - Commercial/Institutional i' i
V S1eel Mill (9)
S Surface Coaling (41)
TT TBiecominunDcalwns (t9)
J U.-Ki so Mi i-j!
^^ Transportalion Equipment (9)
^ Transportalion and Marketing of Petroleum Products (€)
I Wa-5tewate r Treatment {Q)
W Woodworh, Furniture. MillwonV & Wood Preserving (3)
<*> Fabricated Meial Products (53)
Cv> Flexible Polyurethane Foam Production (5|i
F Food Processifia'AgrKuhure (3S'|
[ _j Fumilure Planl (10)
jf Gaaoline;Diesel S?rvHe Stalcort iSi
V G'ass Manufacturing (3)
fV Heating Ewpment Msnufsetunng (2)
[J] hi>-.ji-.=il if.;.
^ Hot Mix A&phalt Plant (&>
-%: Industrial Macnmery and Equipment (33)
^ tnsbtulional - school (30)
I Iron and Steel Foundry (2)
^ Laboratory (1)
A Landfill*?)
|_ Latga Appliance Manufactufing 41)
X Mine/Ouarry (10>
^T Mineral Produced)
5 Miscellaneous Coating ManuFacluring (1)
? Miscellaneous Cwnmercia'JInduslnal [571
11-4
-------
Table 11-1. Geographical Information for the Illinois Monitoring Sites
Site
Code
NBIL
SPIL
AQS Code
17-031-4201
17-031-3103
Location
Northbrook
Schiller
Park
County
Cook
County
Cook
County
Micro- or
Metropolitan
Statistical Area
Chicago-Joliet-
Naperville, IL-IN-
WIMSA
(Chicago Div)
Chicago-Joliet-
Naperville, IL-IN-
WIMSA
(Chicago Div)
Latitude
and
Longitude
42.139996,
-87.799227
41.965193,
-87.876265
Land Use
Residential
Mobile
Location
Setting
Suburban
Suburban
Additional Ambient Monitoring Information1
TSP, TSP Metals, CO, Hg, SO2, NO, NO2, NOX, NH3,
PAMS, O3, Meteorological parameters, PM10, PM2.5,
PM2s Speciation.
TSP, TSP Metals, CO, NO, NO2, NOX, Meteorological
parameters, PM2 5.
BOLD ITALICS = EPA-designated NATTS Site.
-------
NBIL is located on the property of the Northbrook Water Filtration Station. Figure 11-1
shows that NBIL is located off State Highway 68, Dundee Road, near Exit 30 on 1-94 (the clover
leaf of which is located on the lower right hand side of Figure 11-1. A railway intersects Dundee
Road close to the monitoring site. The surrounding area is classified as suburban and residential.
Commercial, residential, and forested areas are nearby, as well as a country club and golf course.
The NBIL monitoring site is the Chicago NATTS site.
SPIL is located on the eastern edge of the Chicago-O'Hare International Airport between
Mannheim Road and 1-294, just north of the toll plaza. The nearest runway is less than 1/2 mile
from the site. The surrounding area is classified as suburban and mobile. Commercial and
residential areas are nearby with a railyard located on the east side of 1-294.
Figure 11-3 shows that NBIL and SPIL are located within approximately 12 miles of
each other. Each site is located within 10 miles of numerous point sources. The source categories
with the largest number of sources within 10 miles of the Illinois monitoring sites are printing
and publishing; fabricated metal products; electroplating, plating, polishing, anodizing, and
coloring; and dry cleaning. Few point sources are located within 2 miles of NBIL, with most of
the sources located farther west or south. The closest source to NBIL is plotted under the symbol
for the site in Figure 11-3; this source is a dry cleaning facility. Numerous sources are located in
close proximity of SPIL. Besides the airport, the closest point source to SPIL is involved in
electroplating, plating, polishing, anodizing, and coloring.
Table 11-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the areas surrounding the Illinois
monitoring sites. Table 11-2 also includes a vehicle registration-to-county population ratio
(vehicles-per-person) for each site. In addition, the population within 10 miles of each site is
presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-level
vehicle registration-to-population ratio to the 10-mile population surrounding each monitoring
site. Table 11-2 also contains annual average daily traffic information. Finally, Table 11-2
presents the daily VMT for Cook County.
11-6
-------
Table 11-2. Population, Motor Vehicle, and Traffic Information for the Illinois Monitoring
Sites
Site
NBIL
SPIL
Estimated
County
Population1
s onn cKn
County-level
Vehicle
Registration2
o n«7 i AI
Vehicles per
Person
(Registration:
Population)
n An
Population
within 10
miles3
859,738
2,046,549
Estimated
10-mile
Vehicle
Ownership
344,352
819,706
Annual
Average
Daily
Traffic4
34,100
170,700
County-
level Daily
VMT5
SO A01 77 A
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the IL Secretary of State (IL SOS, 2010)
310-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2009 data from the Illinois DOT (IL DOT, 2009)
5 County-level VMT reflects 2010 data from the Illinois DOT (IL DOT, 2010)
BOLD ITALICS = EPA-designaled NATTS Site.
Observations from Table 11-2 include the following:
• Cook County has the second highest county-level population (behind Los Angeles
County) and fourth highest county-level vehicle registration (behind Los Angeles
County, CA; Maricopa County, AZ; and Harris County, TX) compared to other
counties with NMP sites.
• The vehicle-per-person ratio for these sites is among the lowest compared to other
NMP sites.
• The 10-mile radius population and estimated vehicle ownership are much higher near
SPIL than NBIL.
• SPIL experiences a higher annual average daily traffic volume than NBIL. SPIL's
traffic volume is the fifth highest among all NMP sites, behind ELNJ, CELA, SEW A,
and PXSS, while the traffic volume for NBIL is in the middle of the range among
NMP sites. Traffic data for SPIL is from 1-294 at Lawrence Avenue; traffic data for
NBIL is for Dundee Road near the railroad crossing.
• The Cook County daily VMT ranks second among counties with NMP sites, behind
only Los Angeles County (where VMT data were available).
11.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in Illinois on sample days, as well as over the course of the year.
11-7
-------
11.2.1 Climate Summary
Daily weather fluctuations are common for the Chicago area. The proximity of Chicago
to Lake Michigan offers moderating effects from the continental climate of the region. In the
summertime, afternoon lake breezes can cool the city when winds from the south and southwest
push temperatures upward. In the winter, the origin of an air mass determines the amount and
type of precipitation. The largest snowfalls tend to occur when cold air masses flow southward
over Lake Michigan, most of which does not freeze in winter. Wind speeds average around
10 miles per hour, but can be greater due to winds channeling between tall buildings downtown,
giving the city its nickname, "The Windy City" (Bair, 1992).
11.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather stations nearest these sites were
retrieved for 2010 (NCDC, 2010). The two closest weather stations are located at Palwaukee
Municipal Airport (near NBIL) and O'Hare International Airport (near SPIL), WBAN 04838 and
94846, respectively. Additional information about the Palwaukee and O'Hare weather stations,
such as the distance between the sites and the weather stations, is provided in Table 11-3. These
data were used to determine how meteorological conditions on sample days vary from normal
conditions throughout the year.
Table 11-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 11-3 is the
95 percent confidence interval for each parameter. As shown in Table 11-3, average
meteorological conditions on sample days were representative of average weather conditions
throughout the year for both sites.
11-8
-------
Table 11-3. Average Meteorological Conditions near the Illinois Monitoring Sites
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Northbrook, Illinois - NBIL
Palwaukee
Municipal
Airport
04838
(42.12, -87.91)
5.27
miles
250°
(WSW)
Sample
Day
2010
60.3
±5.3
59.7
±2.3
52.0
±4.9
51.5
±2.1
40.5
±4.7
40.2
±2.0
46.3
±4.4
45.9
±1.9
67.0
±2.1
67.8
±1.1
1016.8
±1.5
1016.3
±0.7
5.8
±0.8
6.3
±0.3
Schiller Park, Illinois - SPIL
O'Hare
International
Airport
94846
(41.99, -87.91)
2.32
miles
303°
(WNW)
Sample
Day
2010
59.2
±5.6
59.7
±2.3
51.3
±5.1
51.7
±2.1
40.0
±4.7
40.5
±2.0
45.7
±4.6
46.1
±1.9
67.7
±2.5
68.5
±1.3
1016.0
±1.6
1015.7
±0.7
7.5
±0.8
8.0
±0.3
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
11.2.3 Back Trajectory Analysis
Figure 11-4 is the composite back trajectory map for days on which samples were
collected at the NBIL monitoring site in 2010. Included in Figure 11-4 are four back trajectories
per sample day. Figure 11-5 is the corresponding cluster analysis for 2010. Similarly,
Figure 11-6 is the composite back trajectory map for days on which samples were collected at
SPIL and Figure 11-7 is the corresponding cluster analysis. An in-depth description of these
maps and how they were generated is presented in Section 3.5.2.1. For the composite maps, each
line represents the 24-hour trajectory along which a parcel of air traveled toward the monitoring
site on a given sample day and time. For the cluster analyses, each line corresponds to a back
trajectory representative of a given cluster of trajectories. For all maps, each concentric circle
around the sites in Figures 11-4 through 11-7 represents 100 miles.
Figure 11-4. 2010 Composite Back Trajectory Map for NBIL
11-10
-------
Figure 11-5. Back Trajectory Cluster Map for NBIL
Figure 11-6. 2010 Composite Back Trajectory Map for SPIL
11-11
-------
Figure 11-7. Back Trajectory Cluster Map for SPIL
Observations from Figures 11-4 through 11-7 include the following:
• The composite back trajectory maps for NBIL and SPIL are similar to each other in
back trajectory distribution. This is expected given their proximity to each other.
• Back trajectories originated from a variety of directions at the sites, although less
frequently from a direction with an easterly component. The predominant direction of
trajectory origin appears to be from the northwest, north, and south.
• The 24-hour air shed domains for NBIL and SPIL were similar in size compared to
other NMP sites. The longest trajectories originated to the northwest, over the
Dakotas, generally between 600-700 miles away. However, the average trajectory
length for these sites was approximately 250 miles and most (approximately
84 percent) trajectories originated within 400 miles of the sites.
• The cluster map for NBIL is similar to the cluster map for SPIL in geographical
distribution of the clusters as well as the percentage of trajectories representing each
cluster.
• Nearly one-third of back trajectories for both sites originated to the north of the sites,
over Wisconsin, Michigan, and/or Lake Michigan. Another one-fourth of back
trajectories originated from the east, southeast, and south over Ohio, Indiana, and
southern Illinois. One-fifth of back trajectories originated from the northwest and
southwest quadrants and generally within 300 miles of NBIL and SPIL. The longest
trajectories originated from the northwest towards Minnesota and the Dakotas, from
11-12
-------
the south over the Mississippi Valley region, and to the north of the Great Lakes and
into Canada.
11.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather stations at Palwaukee Municipal Airport (for
NBIL) and O'Hare International Airport (for SPIL) were uploaded into a wind rose software
program to produce customized wind roses, as described in Section 3.5.2.2. A wind rose shows
the frequency of wind directions using "petals" positioned around a 16-point compass, and uses
different colors to represent wind speeds.
Figure 11-8 presents three different wind roses for the NBIL monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figure 11-9 presents the
three wind roses and distance map for SPIL.
Observations from Figure 11-8 for NBIL include the following:
• The Palwaukee Municipal Airport weather station is located approximately 5.3 miles
west-southwest of NBIL.
• The historical wind rose shows that winds from a variety of directions were observed
near NBIL, although winds from the south, south-southwest, and west accounted for
nearly one-quarter of all observations. Winds from the east to east-southeast were
observed the least often. Calm winds (<2 knots) were observed for approximately
15 percent of the hourly measurements.
• The 2010 wind rose exhibits similar patterns in wind directions as the historical wind
rose, although northwesterly and north-northwesterly winds were observed slightly
more often and southerly and south-southwesterly winds less often.
• The 2010 sample day wind patterns generally resemble the 2010 full-year wind
patterns, although the reduced percentages of some wind directions (such as west-
southwest, northeast, and east-northeast) may be reflected in the increased percentage
of calm winds.
11-13
-------
Figure 11-8. Wind Roses for the Palwaukee Municipal Airport Weather Station near NBIL
1999-2009 Historical Wind Rose
2010 Wind Rose
IWESTi
2010 Sample Day Wind Rose
Distance between NBIL and NWS Station
,«'Bi r
I I
«... - / ....
\ I
»,
;H Em DtLim JiAvhig ypc--! (K*nrjgi ^
11-14
-------
Figure 11-9. Wind Roses for the O'Hare International Airport Weather Station near SPIL
1999-2009 Historical Wind Rose
2010 Wind Rose
WEST
2010 Sample Day Wind Rose
Distance between SPIL and NWS Station
11-15
-------
Observations from Figure 11-9 for SPIL include the following:
• The O'Hare International Airport weather station is located 2.3 miles west-northwest
of SPIL. The bulk of the airport property lies between the weather station and the
monitoring site.
• The historical wind rose for SPIL shows that winds from a variety of directions were
observed, although winds from the south to southwest to west account for the highest
percentage of observations (greater than 40 percent). Winds from these directions also
tended to be the strongest. Winds from the southeast quadrant were observed the least
often. Calm winds (< 2 knots) were observed for less than 10 percent of the hourly
measurements.
• The 2010 wind rose exhibits similar patterns in wind directions as the historical wind
rose. The 2010 sample day wind patterns resemble the full-year wind patterns,
although with a slightly higher percentage of easterly winds and fewer southerly
winds. This indicates that conditions on sample days were representative of
conditions experienced throughout the year.
11.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Illinois monitoring sites in
order to allow analysts and readers to focus on a subset of pollutants through the context of risk.
For each site, each pollutant's preprocessed daily measurement was compared to its associated
risk screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by each monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 11-4 presents NBIL's and SPIL's pollutants of interest. The pollutants that failed at
least one screen and contributed to 95 percent of the total failed screens for each monitoring site
are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of interest are shaded
and/or bolded. NBIL sampled for VOC, carbonyl compounds, SNMOC, metals (PMio), PAH,
and hexavalent chromium, and is one of two NMP sites sampling for all six pollutant groups.
SPIL sampled for VOC and carbonyl compounds only.
11-16
-------
Table 11-4. Risk Screening Results for the Illinois Monitoring Sites
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
%of
Total
Failures
Cumulative
%
Contribution
Northbrook, Illinois - NBIL
Benzene
Carbon Tetrachloride
Formaldehyde
Naphthalene
Arsenic (PM10)
Acet aldehyde
Manganese (PM10)
1,3-Butadiene
Fluorene
Acenaphthene
£>-Dichlorobenzene
Fluoranthene
1,2-Dichloroethane
Ethylbenzene
Trichloroethylene
Dichloromethane
Nickel (PM10)
Acrylonitrile
Chloroform
Hexavalent Chromium
0.13
0.17
0.077
0.029
0.00023
0.45
0.005
0.03
0.011
0.011
0.091
0.011
0.038
0.4
0.2
7.7
0.0021
0.015
9.8
0.000083
Total
55
55
55
52
51
47
32
31
23
22
11
10
9
4
4
2
2
1
1
1
468
55
55
55
59
61
55
61
42
59
59
26
59
9
54
32
55
61
1
55
48
961
100.00
100.00
100.00
88.14
83.61
85.45
52.46
73.81
38.98
37.29
42.31
16.95
100.00
7.41
12.50
3.64
3.28
100.00
1.82
2.08
48.70
11.75
11.75
11.75
11.11
10.90
10.04
6.84
6.62
4.91
4.70
2.35
2.14
1.92
0.85
0.85
0.43
0.43
0.21
0.21
0.21
11.75
23.50
35.26
46.37
57.26
67.31
74.15
80.77
85.68
90.38
92.74
94.87
96.79
97.65
98.50
98.93
99.36
99.57
99.79
100.00
Schiller Park, Illinois - SPIL
Benzene
Carbon Tetrachloride
1,3-Butadiene
Acet aldehyde
Formaldehyde
Trichloroethylene
1 ,2-Dichloroethane
Ethylbenzene
£>-Dichlorobenzene
Acrylonitrile
Propionaldehyde
Bromomethane
Chloroprene
Dichloromethane
0.13
0.17
0.03
0.45
0.077
0.2
0.038
0.4
0.091
0.015
0.8
0.5
0.0021
7.7
Total
60
60
59
58
58
33
12
11
7
6
4
2
2
1
373
60
60
59
58
58
48
12
60
27
6
58
50
2
60
618
100.00
100.00
100.00
100.00
100.00
68.75
100.00
18.33
25.93
100.00
6.90
4.00
100.00
1.67
60.36
16.09
16.09
15.82
15.55
15.55
8.85
3.22
2.95
1.88
1.61
1.07
0.54
0.54
0.27
16.09
32.17
47.99
63.54
79.09
87.94
91.15
94.10
95.98
97.59
98.66
99.20
99.73
100.00
11-17
-------
Observations from Table 11-4 include the following:
• Twenty pollutants, including 12 NATTS MQO Core Analytes, failed screens for
NBIL. Approximately 49 percent of the measured detections of these pollutants failed
screens.
• Based on the risk screening process, 13 pollutants, of which eight are NATTS MQO
Core Analytes, were identified as pollutants of interest for NBIL. Four additional
NATTS MQO Core Analytes (trichloroethylene, nickel, chloroform, and hexavalent
chromium) were added to NBIL's list of pollutants of interest, even though they did
not contribute to 95 percent of the failed screens for NBIL. In addition, six more
NATTS MQO Core Analytes were added to the pollutants of interest for NBIL, even
though they did not fail any screens (benzo(a)pyrene, beryllium, cadmium, lead,
tetrachloroethylene, and vinyl chloride). These six pollutants are not shown in
Table 11-4.
• Benzene, carbon tetrachloride, and formaldehyde were detected in every VOC or
carbonyl compound sample collected at NBIL and failed 100 percent of screens.
While acrylonitrile and 1,2-dichloroethane also failed 100 percent of screens for
NBIL, these pollutants were detected in few of the 55 valid samples collected.
• Fourteen pollutants, including six NATTS MQO Core Analytes, failed screens for
SPIL. Note that NBIL sampled four additional methods than SPIL but only failed six
additional screens.
• Based on the risk screening process, nine pollutants, of which six are NATTS MQO
Core Analytes, were identified as pollutants of interest for SPIL. Three additional
NATTS MQO Core Analytes were added to SPIL's list of pollutants of interest, even
though they did not fail any screens (chloroform, tetrachloroethylene, and vinyl
chloride). These pollutants are not shown in Table 11-4.
• Acetaldehyde, benzene, carbon tetrachloride, and formaldehyde were detected in
every VOC and carbonyl compound sample collected at SPIL and failed 100 percent
of their screens. Four additional pollutants also failed 100 percent of screens, but the
detection rate was lower.
• Recall from Section 3.2 that if a pollutant was measured by both the TO-15 and
SNMOC methods at the same site, the TO-15 results were used for the risk screening
process. As NBIL sampled both VOC (TO-15) and SNMOC, the TO-15 results were
used for the 12 pollutants these methods have in common.
11.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Illinois monitoring sites. Concentration averages are provided for the pollutants of interest
for each Illinois site, where applicable. Concentration averages for select pollutants are also
presented graphically for each site, where applicable, to illustrate how each site's concentrations
11-18
-------
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at
each site, where applicable. Additional site-specific statistical summaries are provided in
Appendices J through O.
11.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each Illinois site, as described in Section 3.1. The quarterly average of a particular pollutant
is simply the average concentration of the preprocessed daily measurements over a given
calendar quarter. Quarterly average concentrations include the substitution of zeros for all non-
detects. A site must have a minimum of 75 percent valid samples of the total number of samples
possible within a given quarter for a quarterly average to be calculated. An annual average
includes all measured detections and substituted zeros for non-detects for the entire year of
sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the Illinois monitoring
sites are presented in Table 11-5, where applicable. Note that concentrations of the PAH, metals,
and hexavalent chromium for NBIL are presented in ng/m3 for ease of viewing. Also note that if
a pollutant was not detected in a given calendar quarter, the quarterly average simply reflects "0"
because only zeros substituted for non-detects were factored into the quarterly average
concentration.
Table 11-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Illinois Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(jig/m3)
2nd
Quarter
Average
(jig/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Annual
Average
(jig/m3)
Northbrook, Illinois - NBIL
Acetaldehyde
Benzene
1,3 -Butadiene
55/55
55/55
42/55
0.71
±0.18
0.56
±0.09
0.02
±0.01
0.85
±0.21
0.58
±0.15
0.04
±0.02
1.01
±0.19
0.79
±0.18
0.05
±0.02
NA
NA
NA
1.02
±0.16
0.70
±0.10
0.05
±0.01
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of
viewing.
11-19
-------
Table 11-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Illinois Monitoring Sites (Continued)
Pollutant
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Acenaphthene3
Arsenic (PM10)a
Benzo(a)pyrenea
Bery Ilium (PM10)a
Cadmium (PM10)a
Fluoranthene3
Fluorene3
Hexavalent Chromium3
Lead(PM10)a
Manganese (PM10) a
Naphthalene3
Nickel (PM10)a
#of
Measured
Detections
vs. # of
Samples
55/55
55/55
26/55
9/55
55/55
50/55
32/55
1/55
59/59
61/61
53/59
56/61
61/61
59/59
59/59
48/61
61/61
61/61
59/59
61/61
1st
Quarter
Average
(Ug/m3)
0.76
±0.06
0.20
±0.07
0.01
±0.01
0.03
±0.02
0.95
±0.23
0.11
±0.05
0.04
±0.03
0
1.90
±0.72
0.38
±0.12
0.18
±0.07
0.01
±0.01
0.12
±0.04
1.88
±0.61
2.57
±0.85
0.01
±0.01
2.19
±0.56
3.56
±1.30
46.23
± 14.21
0.71
±0.12
2nd
Quarter
Average
(Ug/m3)
0.73
±0.05
0.95
±0.64
0.03
±0.02
0.02
±0.02
1.35
±0.33
0.17
±0.06
0.05
±0.03
O.01
±O.01
10.28
±4.49
0.77
±0.38
0.09
±0.04
0.01
±0.01
0.11
±0.04
8.05
±4.10
11.56
±5.42
0.03
±0.01
3.15
±1.20
7.62
±3.16
79.31
±28.65
1.61
±0.69
3rd
Quarter
Average
(Ug/m3)
0.72
±0.06
1.92
±1.65
0.07
±0.04
0
1.74
±0.33
0.27
±0.08
0.11
±0.04
0
22.91
±8.83
0.91
±0.41
0.07
±0.04
0.01
±0.01
0.11
±0.03
13.05
±3.54
22.43
±7.62
0.02
±0.01
2.89
±0.81
6.58
±1.94
183.29
±114.27
1.04
±0.19
4th
Quarter
Average
(Ug/m3)
NA
NA
NA
NA
NA
NA
NA
NA
7.58
±4.42
0.95
±0.36
0.12
±0.04
0.01
±0.01
0.19
±0.08
2.46
±1.01
6.98
±3.57
0.03
±0.01
4.25
±1.48
9.19
±5.00
118.51
±49.53
0.89
±0.15
Annual
Average
(Ug/m3)
0.72
±0.03
1.06
±0.53
0.06
±0.03
0.01
±0.01
3.59
±2.18
0.21
±0.05
0.07
±0.02
O.01
±O.01
10.46
±3.16
0.75
±0.17
0.11
±0.03
0.01
±0.01
0.13
±0.02
6.25
±1.74
10.69
±2.98
0.02
±O.01
3.11
±0.53
6.74
±1.57
105.54
±31.79
1.06
±0.19
Schiller Park, Illinois - SPIL
Acetaldehyde
Benzene
58/58
60/60
1.37
±0.32
0.80
±0.14
1.38
±0.36
0.96
±0.32
1.64
±0.15
0.99
±0.21
2.05
±0.46
1.03
±0.19
1.62
±0.18
0.94
±0.11
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of
viewing.
11-20
-------
Table 11-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Illinois Monitoring Sites (Continued)
Pollutant
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Ethylbenzene
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
#of
Measured
Detections
vs. # of
Samples
59/60
60/60
55/60
27/60
12/60
60/60
58/58
58/60
48/60
2/60
1st
Quarter
Average
(Ug/m3)
0.13
±0.04
0.76
±0.07
0.06
±0.02
0.01
±0.01
0.02
±0.02
0.18
±0.06
1.65
±0.37
0.26
±0.15
0.20
±0.15
<0.01
±<0.01
2nd
Quarter
Average
(Ug/m3)
0.14
±0.05
0.70
±0.07
0.14
±0.06
0.07
±0.05
0.03
±0.02
0.29
±0.15
2.12
±0.54
0.34
±0.13
0.98
±0.87
<0.01
±<0.01
3rd
Quarter
Average
(Ug/m3)
0.12
±0.03
0.71
±0.06
0.14
±0.02
0.06
±0.03
0.01
0.01
0.32
±0.08
2.77
±0.31
0.39
±0.13
1.57
±1.30
0
4th
Quarter
Average
(Ug/m3)
0.17
±0.05
0.68
±0.06
0.09
±0.02
0.03
±0.02
0.01
±0.02
0.30
±0.09
3.52
±1.42
0.31
±0.13
0.42
±0.21
0
Annual
Average
(Ug/m3)
0.14
±0.02
0.71
±0.03
0.11
±0.02
0.04
±0.02
0.02
±0.01
0.27
±0.05
2.53
±0.42
0.32
±0.06
0.79
±0.40
O.01
±O.01
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of
viewing.
Observations for NBIL from Table 11-5 include the following:
• The pollutants with the highest annual average concentrations by mass are
formaldehyde (3.59 ±2.18 |ig/m3), chloroform (1.06 ± 0.53 |ig/m3), and acetaldehyde
(1.02 ± 0.16 |ig/m3). None of the annual average concentrations for any of the
pollutants of interest were greater than 1.00 |ig/m3 in 2008 or 2009.
• Fourth quarter average concentrations could not be calculated for the VOC or
carbonyl compounds because there were fewer than 75 percent of samples were valid
during this quarter.
• Note how much higher the annual average concentration of formaldehyde is
compared to the first, second, and third quarter averages. This indicates that the
measurements driving the annual average were measured during the fourth quarter. A
review of the data shows that the seven highest concentrations of this pollutant were
measured between October and December 2010. The highest concentration was
measured on December 4, 2010 (53.5 |ig/m3). The next highest formaldehyde
concentration was roughly half that (24.2 |ig/m3) and was measured on
November 4, 2010. Three additional formaldehyde concentrations between 10 and
20 |ig/m3 were also measured during this quarter. Of the 10 formaldehyde
11-21
-------
concentrations greater than 10 |ig/m3 measured across the NMP sites sampling
carbonyl compounds, five were measured at NBIL, including the top four.
• Chloroform's third quarter average is significantly higher than the other available
quarterly average concentrations and has a relatively large confidence interval
associated with it, indicating the likely presence of outliers. The highest concentration
of chloroform was measured on September 17, 2010 (12.1 |ig/m3). This is also the
highest measurement of this pollutant among all NMP sites sampling VOC. The
second highest concentration of chloroform was measured on the previous sample
day (7.14 |ig/m3 on September 11, 2010) and is also the third highest chloroform
measurement among all NMP sites. Of the 10 highest concentrations of chloroform
among NMP sites, eight were measured at NBIL. Similar findings have been found in
previous NMP reports.
• Concentrations of most of the pollutants of interest for NBIL did not vary
significantly from quarter to quarter. However, fourth quarter average concentrations
could not be calculated for the VOC and carbonyl compounds, making a quarterly
trend harder to determine. Several of the PAH exhibited higher concentrations during
the warmer months of the year. For example, fluoranthene's second and third
quarterly average concentrations are greater than the first and fourth quarterly
averages. Acenaphthene, flourene, and naphthalene exhibit this trend too, although
the fourth quarter averages of these pollutants are also relatively high. Many of the
quarterly averages of the PAHs also have large confidence intervals associated with
them, indicating a relatively high level of variability in the measurements.
• Naphthalene's third quarter average is significantly higher than the other available
quarterly average concentrations and has a relatively large confidence interval
associated with it, indicating the likely presence of outliers. The highest concentration
of naphthalene was measured on September 23, 2010 (869 ng/m3). This concentration
is among the 10 highest measurements of this pollutant among NMP sites sampling
PAH. The second highest concentration of naphthalene was measured two weeks later
but is significantly less, by more than half (363 ng/m3 on October 11, 2010). The
median naphthalene concentration for NBIL is 84.0 ng/m3.
• Several of the quarterly averages for the PMio metals are highest for the fourth
quarter, although the difference is not statistically significant among the quarters.
This is most notable for manganese. The highest concentration of manganese was
measured on November 10, 2010 (40.2 ng/m3). The next highest concentration of
manganese measured during this quarter was much less (13.7 ng/m3 on
October 23, 2010), indicating that this high measurement is driving the quarterly
average. The median manganese concentration for this quarter is 6.90 ng/m3.
Observations for SPIL from Table 11-5 include the following:
• The pollutants with the highest annual average concentrations by mass are
formaldehyde (2.53 ± 0.42 |ig/m3) and acetaldehyde (1.62 ±0.18 |ig/m3). These are
the only pollutants with annual average concentrations greater than 1 |ig/m3.
11-22
-------
• Concentrations of most of the pollutants of interest for SPIL did not vary significantly
across calendar quarters. However, a few quarterly averages do stand out, as
described below.
• The second and third quarterly average concentrations of trichloroethylene are higher
than the other quarters and have rather large confidence intervals associated with
them, particularly the third quarter of 2010, indicating that outliers are likely
influencing these averages. The highest trichloroethylene concentration was measured
on September 17, 2010 (9.64 |ig/m3) and is the highest trichloroethylene
concentration measured among NMP sites sampling VOC. Of the 18 concentrations
of trichloroethylene greater than 1 |ig/m3across the program, 15 of these were
measured at SPIL. The bulk of these were measured during the third quarter of 2010.
• The fourth quarter average concentration of formaldehyde also has a high confidence
interval associated with it. The five highest concentrations of formaldehyde were all
measured during the fourth quarter of 2010, and ranged from 10.8 |ig/m3 (measured
on December 22, 2010) to 3.98 |ig/m3 (measured on October 11, 2010).
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for NBIL and
SPIL from those tables include the following:
• NBIL and SPIL appear in Tables 4-9 through 4-12 a total of 22 times.
• As shown in Table 4-9, NBIL's annual average concentration of chloroform is the
highest among all NMP sites sampling this pollutant. SPIL has the highest annual
average concentration of trichloroethylene, which is more than eight times greater
than the next highest annual average of this pollutant. NBIL also has the fourth
highest annual average concentration of trichloroethylene. NBIL and SPIL have the
second and third highest annual average concentrations of carbon tetrachloride,
respectively. SPIL has the fourth highest annual average concentrations of
1,3-butadiene and tetrachloroethylene.
• NBIL has the third highest annual average concentration of formaldehyde among
sites sampling carbonyl compounds, as shown in Table 4-10. However, the high
confidence interval indicates that there are outliers driving this average, as opposed to
the site measuring higher concentrations on a regular basis, which, based on the
confidence intervals, would be expected for the sites ranking first and second highest
(ELNJ and BTUT) for formaldehyde.
• NBIL has the second highest annual average concentrations of acenaphthene and
fluorene and the fifth highest annual average concentration of benzo(a)pyrene among
NMP sites sampling PAH, as shown in Table 4-11.
• As shown in Table 4-12, the annual average concentrations for NBIL were among the
top four for all of the program-level PMi0 metals pollutants of interest. However, it is
11-23
-------
important to note that only nine sites sampled PMio metals and have enough data for
annual averages to be calculated.
11.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for acetaldehyde, benzene,
1,3-butadiene, and formaldehyde were created for both NBIL and SPIL. Box plots were also
created for arsenic, benzo(a)pyrene, hexavalent chromium, manganese, and naphthalene for
NBIL. Figures 11-10 through 11-18 overlay the sites' minimum, annual average, and maximum
concentrations onto the program-level minimum, first quartile, average, median, third quartile,
and maximum concentrations, as described in Section 3.5.3.
Figure 11-10. Program vs. Site-Specific Average Acetaldehyde Concentration
—
Concentration (ng/m3)
Program: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Average
Site:
Site Average Site Minimum/Maximum
o
11-24
-------
Figure 11-11. Program vs. Site-Specific Average Arsenic (PMi0) Concentration
1 -
0 0.5 1 1.5
Program: IstQuartile
•
Site: Site Average
o
2 2.5 3 3.5 4 4.5 5
Concentration (ng/m3)
2ndQuartile 3rdQuartile 4thQuartile Average
D D D
Site Minimum/Maximum
Figure 11-12. Program vs. Site-Specific Average Benzene Concentration
3 4
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 11-13. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
n
1 1 1
| Program Max Concentration = 42.7 ng/m3
0 0.2 0.4 0.6 0.8 1 1.2
Concentration (ng/m3)
1.4 1.6
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
11-25
-------
Figure 11-14. Program vs. Site-Specific Average 1,3-Butadiene Concentration
0.4 0.5 0.6
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
o
Figure 11-15. Program vs. Site-Specific Average Formaldehyde Concentration
^ 1
1
10 15 20 25 30 35
Concentration (ng/m3)
45 50
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
O
11-26
-------
Figure 11-16. Program vs. Site-Specific Average Hexavalent Chromium Concentration
E
0 0.15
Program Max Concentration = 3.51 ng/m3
0.3 0.45 0.6
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
• DID
Site: Site Average Site Minimum/Maximum
o —
0.7
Figure 11-17. Program vs. Site-Specific Average Manganese (PMio) Concentration
1
r
3 20 40 60
| |
80 100 120 140 160 180
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
• • D D
Site: Site Average Site Minimum/Maximum
o —
20
Figure 11-18. Program vs. Site-Specific Average Naphthalene Concentration
L
3 200 400
|
600 800 1000 1200 1400
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
• ODD
Site: Site Average Site Minimum/Maximum
o —
11-27
-------
Observations from Figures 11-10 through 11-18 include the following:
• Figure 11-10 shows that while SPIL's annual average acetaldehyde concentration
is greater than NBIL's annual average acetaldehyde concentration, both annual
averages are less than the program-level average. NBIL's annual average is less
than the program-level first quartile while SPIL's annual average is equivalent to
the program-level median (or second quartile). There were no non-detects of
acetaldehyde measured at either site.
• Figure 11-11 is the box plot for arsenic, which was measured at NBIL but not at
SPIL. The box plot shows that the annual average concentration for NBIL is
greater than the program-level average concentration. The annual average for
NBIL is also greater than the program-level third quartile. While the maximum
concentration measured at NBIL is not the maximum measured across the
program, it is the second highest arsenic concentration measured among NMP
sites sampling PMio metals.
• Figure 11-12 shows that SPIL's annual average benzene concentration is greater
than NBIL's annual average benzene concentration. Although both annual
averages are less than the program-level average, the difference is minimal for
SPIL. The maximum benzene concentrations measured at the Illinois sites are
well below the program-level maximum concentration. There were no non-detects
of benzene measured at either site.
• Figure 11-13 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average benzo(a)pyrene
concentration for NBIL is just less than the program-level average concentration.
Figure 11-13 also shows that the maximum concentration measured at NBIL is
well below the maximum concentration measured across the program. Several
non-detects of benzo(a)pyrene were measured at NBIL.
• Figure 11-14 for 1,3-butadiene also shows both sites. Figure 11-14 shows that
NBIL's annual average 1,3-butadiene concentration is less than both the program-
level average and median concentrations. Conversely, SPIL's annual average
1,3-butadiene concentration is nearly twice the program-level average for
1,3-butadiene. Non-detects of 1,3-butadiene were measured at both sites.
• Figure 11-15 presents the box plots for formaldehyde. The box plots show that
while NBIL's annual average formaldehyde concentration is greater than the
program-level average, SPIL's annual average formaldehyde concentration is
roughly equal to program-level average. However, what is most prominent in
Figure 11-15 is that NBIL's maximum concentration of formaldehyde is the
maximum concentration measured across the program. As discussed in the
previous section, the four highest concentrations of this pollutant measured
among all NMP sites sampling carbonyl compounds were all measured at NBIL.
11-28
-------
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 11-16 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for the observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 11-16 shows that the annual average concentration of hexavalent
chromium for NBIL is less than the program-level average. The maximum
concentration measured at NBIL is well below the program-level maximum
concentration. There were several non-detects of hexavalent chromium measured
at NBIL.
• Figure 11-17 is the box plot for manganese, which was measured at NBIL. The
box plot shows that the annual average concentration for NBIL is roughly
equivalent to the program-level average concentration. While the maximum
concentration measured at NBIL is not the maximum measured across the
program, it is among the higher manganese concentrations measured among NMP
sites sampling PMi0 metals. There were no non-detects of manganese measured at
NBIL.
• Figure 11-18 shows that the annual naphthalene average for NBIL is greater than
the program-level average concentration. While the maximum naphthalene
concentration measured at NBIL is well below the program-level maximum
concentration, NBIL's maximum concentration is among the highest
concentrations measured at the program-level. There were no non-detects of
naphthalene measured at NBIL.
11.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. NBIL and SPIL have sampled VOC under the NMP since 2003. Both sites have
also sampled carbonyl compounds since 2005. Additionally, NBIL has also sampled PMio metals
and hexavalent chromium since 2005. Figures 11-19 through 11-25 present the 3-year rolling
statistical metrics for acetaldehyde, arsenic, benzene, 1,3-butadiene, formaldehyde, hexavalent
chromium, and manganese for NBIL, respectively. Figures 11-26 through 11-29 present the
3-year rolling statistical metrics for acetaldehyde, benzene, 1,3-butadiene, and formaldehyde for
SPIL, respectively. The statistical metrics presented for assessing trends include the substitution
of zeros for non-detects.
11-29
-------
Figure 11-19. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at NBIL
I 1
1
1
11
• sill
T T
^^
i
2005-2007 '
'ctuiillk-
^i^m
i
2006-200B
- Met
^^
1
2007-2009
Three-Year Period
1,111 - (l.l.MMMIII
• '.IMIllMtlllMk-
••
—
[
2008-2010
« -•'',.! .*'
^arbonyl compound sampling at NBIL began in March 2005.
Figure 11-20. Three-Year Rolling Statistical Metrics for Arsenic (PMi0) Concentrations
Measured at NBIL
4
"it*
H
S ,
Concintrat
M
* U-i »
Oi
*
4
^
^
^
1
*
^™
1
^
1
<
^
>
M
i
ZOOi 2tl(J/ 2006 2008 £UU/ iOliy 2008 2030
Thr«*-Y*tr Ptrlod
• "5tli Pert Hitil«* — Minimum - Median — M axirmtm • <) 'ith Per* HtTite ••-»-• Average
11-30
-------
Figure 11-21. Three-Year Rolling Statistical Metrics for Benzene Concentrations
Measured at NBIL
J.S
— Miriii
file * • •''. fl,i;:r
1 VOC sampling at NBIL began in April 2003.
2No VOC samples were collected from November to December 2004.
11-31
-------
Figure 11-23. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at NBIL
— Minimum
- Md*iimim • 95
-------
Figure 11-25. Three-Year Rolling Statistical Metrics for Manganese (PMi0) Concentrations
Measured at NBIL
•3 vt.tm
•••*•
•
-r
Threa-Yt«r Period
- JlhPefiaidk - l.lMiniiimi - Mi'dijri - MdKinnnii • HARtKBMIIe •••*••
Figure 11-26. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at SPIL
Three-Year Period
— MiitiiiKiin - Motlidii — • l.i iinui, k
!Carbonyl compound sampling at SPIL began in February 2005.
11-33
-------
Figure 11-27. Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured
at SPIL
200* 2()W>
iuui umt iuuh mm
Thru-Year Period
• ithPercentlle - Minimum - Median - Maximum • 9itliPm«mle
1VOC sampling at SPIL began in April 2003.
Figure 11-28. Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations
Measured at SPIL
:
\
[«
|
j
0.4
•i
• '
!•
^MH
£00$JOO&1 20042001
• Jth Pertenol*
••
n ^ ^ 1
i — — i — —
•*— I' •*•
t T r
2005 l«c«inle •••»•• Average
JVOC sampling at SPIL began in April 2003.
11-34
-------
Figure 11-29. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at SPIL
• Mti mwntlle - Minimum - Median - Mixlmuin • 9MliP«<«itllc •••*•• Average
Carbonyl compound sampling at SPIL began in February 2005.
Observations from Figure 11-19 for acetaldehyde measurements at NBIL include the
following:
• Carbonyl compound sampling at NBIL began in March 2005, as denoted in
Figure 11-19.
• The maximum acetaldehyde concentration was measured in 2010, as was the second
highest concentration.
• The rolling average and median concentrations, as well as the other statistical
parameters, have a slight decreasing trend through 2007-2009, after which most of the
statistical parameters exhibit an increase.
• The minimum concentration for each 3-year period is greater than zero, indicating
that there were no non-detects of acetaldehyde reported since the onset of carbonyl
compound sampling.
Observations from Figure 11-20 for arsenic (PMio) measurements at NBIL include the
following:
• Metals sampling at NBIL began in January 2005.
• The maximum arsenic concentration was measured on July 12, 2009, although a
similar concentration was also measured in 2010.
11-35
-------
• The rolling average concentrations exhibit a slight decreasing trend from 2006-2008
through 2008-2010, although the confidence intervals indicate that the decrease is not
statistically significant.
• Note that the minimum concentration for each 3-year period is greater than zero,
indicating that there were no non-detects of arsenic reported since the onset of metals
sampling.
Observations from Figure 11-21 for benzene measurements at NBIL include the
following:
• VOC sampling at NBIL began in April 2003, as denoted in Figure 11-21. However,
VOC samples were not collected during November and December 2004.
• The maximum benzene concentration was measured on September 18, 2004, although
similar measurements were also measured in 2005.
• The rolling average and median concentrations, as well as the other statistical
parameters, have a decreasing trend through the 2007-2009 time frame, after which
an increase is shown.
• Note that the minimum concentration for each 3-year period is greater than zero,
indicating that there were no non-detects of benzene reported since the onset of VOC
sampling.
Observations from Figure 11-22 for 1,3-butadiene measurements at NBIL include the
following:
• The maximum 1,3-butadiene concentration was measured on December 16, 2010; this
concentration is more than three times greater than the next highest concentration
measured at NBIL (2005).
• Although fluctuations in the rolling average concentrations are shown in
Figure 11-22, confidence intervals calculated on these averages indicate that the
differences are not statistically significant.
• The minimum, 5th percentile, and median concentrations were all zero for the first
two 3-year periods, indicating the presence of non-detects (at least 50 percent). The
number of non-detects reported has fluctuated from year to year, from as high as
87 percent (2004) to as low as seven percent (2007). From the 2005-2007 time frame
forward, the median is greater than zero and exhibits a similar trend as the rolling
average.
11-36
-------
Observations from Figure 11-23 for formaldehyde measurements at NBIL include the
following:
• The maximum formaldehyde concentration was measured on January 5, 2006
(91.7 |ig/m3). However, the next five highest concentrations, ranging from 53.5 |ig/m3
to 14.4 |ig/m3, were all measured in 2010.
• Although difficult to discern in Figure 11-23, the rolling average concentrations
exhibited a decreasing trend through 2007-2009, after which the rolling average
concentration increased. However, the high variability associated with three of the
four 3-year periods make it difficult to determine if any of these changes are
statistically significant.
• Although difficult to discern in Figure 11-23, the minimum concentration for each
3-year period is greater than zero, indicating that there were no non-detects of
formaldehyde reported since the onset of carbonyl compound sampling.
Observations from Figure 11-24 for hexavalent chromium measurements at NBIL include
the following:
• Hexavalent chromium sampling at NBIL began in January 2005.
The maximum hexavalent chromium concentration was measured on July 5, 2007
•om NBIL are greater than
to 0.108 ng/m3 (of which four of
(0.307 ng/m3). Only five additional measurements from NBIL are greater than
0.1 ng/m3, with the others ranging from 0.235 ng/m3
the five were measured in 2006).
• The rolling average concentrations of hexavalent chromium exhibit a decreasing trend
through the 2007-2009 time frame, as do the medians and 95th percentiles. A slight
increase is shown for these parameters for the 2008-2010 time frame, even though the
maximum concentration shown for this time period decreased substantially.
• Both the minimum concentration and 5th percentile for all 3-year periods shown are
zero, indicating the presence of non-detects.
Observations from Figure 11-25 for manganese (PMio) measurements at NBIL include
the following:
• Metals sampling at NBIL began in January 2005.
• The maximum manganese concentration was measured on August 26, 2005
(54.6 ng/m3). However, concentrations in the 40-45 ng/m3 range have been measured
in 2005, 2008, and 2010.
• The rolling average exhibits a significant decrease from 2005-2007 to 2006-2008,
with more subtle decreases afterward. The 95th percentile exhibits a similar trend over
11-37
-------
the period of sampling. The median decreased from 2005-2007 to 2006-2008,
increased slightly for 2007-2009, then returned to previous levels for 2008-2010.
• The rolling average and median concentrations became more similar to each other
over time, indicating less variability in the central tendency of manganese
measurements at NBIL.
• The minimum concentration for each 3-year period is greater than zero, indicating
that there were no non-detects of manganese reported since the onset of metals
sampling.
Observations from Figure 11-26 for acetaldehyde measurements at SPIL include the
following:
• Carbonyl compound sampling at SPIL began in February 2005, as denoted in
Figure 11-26.
• The maximum acetaldehyde concentration was measured on May 29, 2006. Of the
eight acetaldehyde concentrations greater than 4.0 |ig/m3, all but one was measured in
2006.
• The rolling average concentrations for the 2005-2007 and 2006-2008 periods were
similar to each other, the 2007-2009 rolling average exhibited a decrease from the
previous 3-year periods, then the rolling average increased slightly for 2008-2010.
Although difficult to discern in Figure 11-26, the median concentrations exhibit a
similar trend across the periods.
• Note that the minimum concentration for each 3-year period is greater than zero,
indicating that there were no non-detects of acetaldehyde reported since the onset of
carbonyl compound sampling at SPIL.
Observations from Figure 11-27 for benzene measurements at SPIL include the
following:
• VOC sampling at SPIL began in April 2003, as denoted in Figure 11-27.
• The maximum benzene concentration was measured on October 13, 2005, although a
similar concentration was also measured in February 2005.
• Similar to NBIL, the median and average rolling concentrations have a decreasing
trend over the time periods shown, although a slight increase is shown for the final
time frame.
• The differences between the 5th and 95th percentiles and the rolling average and
median concentrations have generally decreased over time, both indicators of
decreasing variability in the central tendency.
11-38
-------
• The minimum concentration for each 3-year period is greater than zero, indicating
that no non-detects of benzene have been reported since the onset of VOC sampling.
Observations from Figure 11-28 for 1,3-butadiene measurements at SPIL include the
following:
• The maximum concentration of 1,3-butadiene (1.29 |ig/m3) was measured on
February 3, 2005. Only three concentrations greater than 0.5 |ig/m3 have been
measured at SPIL, one in 2004 and two in 2005. This explains the large decrease in
the maximum concentration from the 2005-2007 to 2006-2008.
• Although there have been fluctuations in the rolling average and median
concentrations of 1,3-butadiene, the changes are not statistically significant.
• 1,3-Butadiene's detection rate has increased over time, ranging from approximately
45 percent non-detects in 2003 and 2004 to zero or one non-detects during the 2007-
2010 years of sampling.
Observations from Figure 11-29 for formaldehyde measurements at SPIL include the
following:
• The maximum formaldehyde concentration was measured on May 29, 2006, which is
the same day the highest acetaldehyde concentration was measured. Three additional
formaldehyde concentrations greater than 100 |ig/m3 were measured in 2005. The 35
highest formaldehyde concentrations (those greater than 11 |ig/m3), all were
measured in 2005 and 2006.
• The rolling average concentrations exhibit a dramatic decreasing trend. Although
difficult to discern in Figure 11-29, the median concentration decreased as well.
11.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at each
Illinois monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
11.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Illinois monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where available.
As described in Section 3.3, acute risk results from exposures of 1 to 14 days; intermediate risk
results from exposures of 15 to 364 days; and chronic risk results from exposures of 1 year or
11-39
-------
greater. The preprocessed daily measurements of the pollutants of interest were compared to the
acute MRL; the quarterly averages were compared to the intermediate MRL; and the annual
averages were compared to the chronic MRL. The results of this risk screening are summarized
in Table 11-6. Where a quarterly or annual average exceeds the applicable MRL, the
concentration is bolded.
Observations from Table 11-6 include the following:
• Formaldehyde was the only pollutant for NBIL where a preprocessed daily
measurement and/or time-period average was greater than one or more of the MRL
noncancer health risk benchmarks.
• One out of 55 measured detections of formaldehyde is greater than the ATSDR acute
MRL for this pollutant (50 |ig/m3). As discussed in Section 11.4.1, this concentration
was measured on December 4, 2010. This measurement (53.5 |ig/m3) is the highest
concentration of formaldehyde measured among all NMP sites sampling this pollutant
and is the only formaldehyde concentration to exceed the acute MRL, as discussed in
Section 3.3.
• Although none of the quarterly average concentrations of formaldehyde are greater
than the ATSDR intermediate MRL of 40 |ig/m3 (where they could be calculated), the
highest concentrations of formaldehyde were measured during the fourth quarter of
2010, for which no fourth quarter average could be calculated. There were not enough
valid samples during the fourth quarter to meet the completeness requirements for a
quarterly average to be calculated.
• The annual average concentration of formaldehyde for NBIL (3.59 ±2.18 |ig/m3) is
less than the ATSDR chronic MRL for this pollutant (10 |ig/m3).
For the pollutants whose concentrations are greater than their respective ATSDR acute
MRL noncancer health risk benchmark(s), the concentrations are further examined by
developing pollution roses for these pollutants. A pollution rose is a plot of concentration vs.
wind speed and wind direction, as described in Section 3.5.5.1. Figure 11-30 is the formaldehyde
pollution rose for NBIL.
11-40
-------
Table 11-6. Noncancer Risk Screening Summary for the Illinois Monitoring Sites
Pollutant
Acute
ATSDR
Acute
MRL1
(Hg/m3)
#of
Concentrations
>MRL
#of
Measured
Detections
Intermediate
ATSDR
Intermediate
MRL1
(Hg/m3)
1st
Quarter
Average
(jig/m3)
2nd
Quarter
Average
(jig/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Chronic
ATSDR
Chronic
MRL1
(Hg/m3)
Annual
Average
(jig/m3)
Northbrook, Illinois - NBIL
Formaldehyde || 50
1
55
40
0.95
±0.23
1.35
±0.23
1.74
±0.33
NA
10
3.59
±2.18
Reflects the use of one significant digit for the MRLs
-------
270
to
Figure 11-30. Formaldehyde Pollution Rose for NBIL
360/0
225
0-10ng/m3
180
O10-50|ig/m3
, 90
18 kts
ATSDRMRL = 50 ng/m3,
which corresponds to the upper
end of the 10-50 |ig/m3 (or
yellow) concentration range
O>50 ng/m3
-------
Observations from the Figure 11-30 include the following:
• There was only one measured detection that is greater than the ATSDR acute MRL
(50 |ig/m3) for formaldehyde (shown in orange).
• The concentration greater than the ATSDR acute MRL was measured on a day with
light winds blowing from the north-northwest, as was NBIL's second highest
concentration (24.2 |ig/m3, shown in yellow). However, there were other
measurements of formaldehyde that were much less and measured on days with
average winds out of the north-northwest.
• The three next highest concentrations, ranging from 14.4 |ig/m3 to 16.9 |ig/m3, were
measured on days with winds from the southwest. However, other measurements of
formaldehyde were less and were measured on days with average winds out of the
southwest.
• Figure 11-3 shows that there are many point sources located to the north-northwest
and southwest of NBIL.
11.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Illinois monitoring sites and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 11-7, where applicable.
Observations for NBIL from Table 11-7 include the following:
• Formaldehyde, chloroform, and acetaldehyde are the pollutants with the highest
annual average concentrations for NBIL.
• Formaldehyde, benzene, and carbon tetrachloride have the highest cancer surrogate
risk approximations. NBIL's cancer risk approximation for formaldehyde
(46.66 in-a-million) is the third highest cancer risk approximation calculated among
all site-specific pollutants of interest for NMP sites (the other two were also for
formaldehyde).
• None of NBIL's pollutants of interest have noncancer surrogate risk approximations
greater than 1.0. The pollutant with the highest noncancer surrogate risk
approximation is formaldehyde (0.37).
11-43
-------
Table 11-7. Cancer and Noncancer Surrogate Risk Approximations for the Illinois
Monitoring Sites
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs.
# of Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Northbrook, Illinois - NBIL
Acenaphthene
Acetaldehyde
Arsenic (PM10)a
Benzene
Benzo(a)pyrene a
Bery Ilium (PM10)a
1,3 -Butadiene
Cadmium (PM10)a
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1,2-Dichloroethane
Fluoranthene a
Fluorene a
Formaldehyde
Hexavalent Chromium3
Lead(PM10)a
Manganese (PM10) a
Naphthalene a
Nickel (PM10)a
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.000088
0.0000022
0.0043
0.0000078
0.00176
0.0024
0.00003
0.0018
0.000006
0.000011
0.000026
0.000088
0.000088
0.000013
0.012
0.000034
0.00048
2.6E-07
0.0000048
0.0000088
0.009
0.000015
0.03
0.00002
0.002
0.00001
0.1
0.098
0.8
2.4
_
0.0098
0.0001
0.00015
0.00005
0.003
0.00009
0.04
0.002
0.1
59/59
55/55
61/61
55/55
53/59
56/61
42/55
61/61
55/55
55/55
26/55
9/55
59/59
59/59
55/55
48/61
61/61
61/61
59/59
61/61
50/55
32/55
1/55
0.01
±<0.01
1.02
±0.16
<0.01
±<0.01
0.70
±0.10
<0.01
±<0.01
<0.01
±0.01
0.05
±0.01
0.01
±0.01
0.72
±0.03
1.06
±0.53
0.06
±0.03
0.01
±0.01
0.01
±0.01
0.01
±0.01
3.59
±2.18
O.01
±O.01
0.01
±0.01
0.01
±O.01
0.11
±0.03
O.01
±O.01
0.21
±0.05
0.07
±0.02
0.01
±0.01
0.92
2.24
3.24
5.45
0.20
0.01
1.41
0.23
4.32
0.62
0.31
0.55
0.94
46.66
0.27
3.59
0.51
0.05
0.35
0.01
0.11
0.05
0.02
0.01
0.02
0.01
0.01
0.01
O.01
O.01
_
0.37
O.01
0.02
0.13
0.04
0.01
0.01
0.04
0.01
- = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 11-
11-44
-------
Table 11-7. Cancer and Noncancer Surrogate Risk Approximations for the Illinois
Monitoring Sites (Continued)
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs.
# of Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Schiller Park, Illinois - SPIL
Acetaldehyde
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Ethylbenzene
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.0000022
0.0000078
0.00003
0.000006
0.000011
0.000026
0.0000025
0.000013
2.6E-07
0.0000048
0.0000088
0.009
0.03
0.002
0.1
0.098
0.8
2.4
1
0.0098
0.04
0.002
0.1
58/58
60/60
59/60
60/60
55/60
27/60
12/60
60/60
58/58
58/60
48/60
2/60
1.62
±0.18
0.94
±0.11
0.14
±0.02
0.71
±0.03
0.11
±0.02
0.04
±0.02
0.02
±0.01
0.27
±0.05
2.53
±0.42
0.32
±0.06
0.79
±0.40
0.01
±O.01
3.56
7.36
4.25
4.28
0.45
0.42
0.68
32.86
0.08
3.79
O.01
0.18
0.03
0.07
0.01
O.01
0.01
O.01
0.01
0.26
0.01
0.39
O.01
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 11-5.
Observations for SPIL from Table 11-7 include the following:
• Formaldehyde, acetaldehyde, and benzene are the pollutants with the highest annual
average concentrations for SPIL.
• Formaldehyde has the highest cancer surrogate risk approximation for SPIL
(32.86 in-a-million), followed by benzene and carbon tetrachloride.
• None of SPIL's pollutants of interest have noncancer surrogate risk approximations
greater than 1.0. The pollutant with the highest noncancer surrogate risk
approximation is trichloroethylene (0.39), which is the fourth highest noncancer
surrogate risk approximation calculated among all site-specific pollutants of interest
for NMP sites.
11-45
-------
11.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 11-8 and 11-9 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 11-8 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from annual averages.
Table 11-9 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 11-8 and 11-9 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer surrogate risk approximations based on each site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 11.3, SPIL sampled for VOC and carbonyl compounds. NBIL sampled for these
pollutants as well, but also sampled for SNMOC, PMio metals, PAH, and hexavalent chromium.
In addition, the cancer and noncancer risk approximations are limited to those pollutants with
enough data to meet the criteria for annual averages to be calculated. A more in-depth discussion
of this analysis is provided in Section 3.5.5.3.
11-46
-------
Table 11-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Illinois Monitoring Sites
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Northbrook, Illinois (Cook County) - NBIL
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Tetrachloroethylene
Trichloroethylene
Naphthalene
Dichloromethane
POM, Group 2b
1,445.06
1,055.89
791.28
689.38
209.30
190.73
134.75
126.41
93.18
20.55
Formaldehyde
Benzene
Hexavalent Chromium, PM
1,3 -Butadiene
Naphthalene
Cadmium, PM
Nickel, PM
Arsenic, PM
Ethylbenzene
POM, Group 2b
1.37E-02
1.13E-02
6.92E-03
6.28E-03
4.30E-03
2.67E-03
2.44E-03
2.17E-03
1.98E-03
1.81E-03
Formaldehyde
Benzene
Carbon Tetrachloride
Naphthalene
Arsenic
Acetaldehyde
1,3 -Butadiene
Fluorene
Acenaphthene
£>-Dichlorobenzene
46.66
5.45
4.32
3.59
3.24
2.24
1.41
0.94
0.92
0.62
Schiller Park, Illinois (Cook County) - SPIL
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Tetrachloroethylene
Trichloroethylene
Naphthalene
Dichloromethane
POM, Group 2b
1,445.06
1,055.89
791.28
689.38
209.30
190.73
134.75
126.41
93.18
20.55
Formaldehyde
Benzene
Hexavalent Chromium, PM
1,3 -Butadiene
Naphthalene
Cadmium, PM
Nickel, PM
Arsenic, PM
Ethylbenzene
POM, Group 2b
1.37E-02
1.13E-02
6.92E-03
6.28E-03
4.30E-03
2.67E-03
2.44E-03
2.17E-03
1.98E-03
1.81E-03
Formaldehyde
Benzene
Carbon Tetrachloride
1,3 -Butadiene
Trichloroethylene
Acetaldehyde
Ethylbenzene
£>-Dichlorobenzene
1,2-Dichloroethane
Tetrachloroethylene
32.86
7.36
4.28
4.25
3.79
3.56
0.68
0.45
0.42
0.08
-------
Table 11-9. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Illinois Monitoring Sites
oo
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Noncancer Toxicity
Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Northbrook, Illinois (Cook County) - NBIL
Toluene
Xylenes
Methanol
1,1,1 -Trichloroethane
Methyl isobutyl ketone
Benzene
Formaldehyde
Hexane
Ethylbenzene
Acetaldehyde
5,432.03
3,804.27
3,800.86
3,014.08
2,290.29
1,445.06
1,055.89
1,006.50
791.28
689.38
Acrolein
Manganese, PM
Cadmium, PM
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Trichloroethylene
Nickel, PM
Benzene
Naphthalene
3,705,826.42
161,905.82
148,156.47
107,743.48
104,651.38
76,598.28
67,372.78
56,590.58
48,168.61
42,137.52
Formaldehyde
Manganese
Acetaldehyde
Arsenic
Trichloroethylene
Naphthalene
1,3 -Butadiene
Benzene
Lead
Cadmium
0.37
0.13
0.11
0.05
0.04
0.04
0.02
0.02
0.02
0.01
Schiller Park, Illinois (Cook County) - SPIL
Toluene
Xylenes
Methanol
1,1,1 -Trichloroethane
Methyl isobutyl ketone
Benzene
Formaldehyde
Hexane
Ethylbenzene
Acetaldehyde
5,432.03
3,804.27
3,800.86
3,014.08
2,290.29
1,445.06
1,055.89
1,006.50
791.28
689.38
Acrolein
Manganese, PM
Cadmium, PM
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Trichloroethylene
Nickel, PM
Benzene
Naphthalene
3,705,826.42
161,905.82
148,156.47
107,743.48
104,651.38
76,598.28
67,372.78
56,590.58
48,168.61
42,137.52
Trichloroethylene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Benzene
Tetrachloroethylene
Carbon Tetrachloride
Chloroform
Ethylbenzene
/>-Dichlorobenzene
0.39
0.26
0.18
0.07
0.03
0.01
0.01
0.01
0.01
0.01
-------
Observations from Table 11-8 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in Cook County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
cancer UREs) for Cook County are formaldehyde, benzene, and hexavalent
chromium.
• Six of the highest emitted pollutants in Cook County also have the highest toxicity-
weighted emissions.
• For both monitoring sites, formaldehyde has the highest cancer surrogate risk
approximation. This pollutant ranks high on all three lists shown in Table 11-8. For
NBIL, benzene, naphthalene, and 1,3-butadiene also appear on all three lists. For
SPIL, benzene, 1,3-butadiene, and ethylbenzene appear on all three lists.
• Carbon tetrachloride, which appears among the highest cancer risk approximations
for both sites, did not appear on either emissions-based list.
• Several metals appear among the pollutants with the highest toxicity-weighted
emissions, including arsenic, which has the fifth highest cancer risk approximation
for NBIL. SPIL did not sample metals.
• NBIL is one of two NMP sites that sampled pollutants from all six methods. At least
one pollutant from each of the six methods appears on the list of highest toxi city-
weighted emissions.
Observations from Table 11-9 include the following:
• Toluene, xylenes, and methanol are the highest emitted pollutants with noncancer
RfCs in Cook County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
noncancer RfCs) for Cook County are acrolein, manganese, and cadmium. Although
acrolein was sampled for at both NBIL and SPIL, this pollutant was excluded from
the pollutants of interest designation, and thus subsequent risk screening evaluations,
due to questions about the consistency and reliability of the measurements, as
discussed in Section 3.2.
• Three of the highest emitted pollutants also have the highest toxi city-weighted
emissions (benzene, formaldehyde, and acetaldehyde).
• Formaldehyde, manganese, and acetaldehyde have the highest noncancer risk
approximations for NBIL (albeit well below an HQ of 1.0). Formaldehyde and
acetaldehyde appear on both emissions-based lists while manganese has the second
highest toxi city-weighted emissions, but does appear on the list of highest emitted
pollutants in Cook County.
11-49
-------
• Trichloroethylene has the highest noncancer risk approximation for SPIL (albeit well
below an HQ of 1.0). This pollutant has the seventh highest toxicity-weighted
emissions, but does appear on the list of highest emitted pollutants in Cook County.
11.6 Summary of the 2010 Monitoring Data for NBIL and SPIL
Results from several of the data treatments described in this section include the
following:
»«» Twenty pollutants, including 12 NA TTS MQO Core Analytes, failed screens for NBIL.
Fourteen pollutants, including six NA TTS MQO Core Analytes, failed screens for
SPIL.
»«» The pollutant with the highest annual average concentration among the pollutants of
interest for both sites was formaldehyde. The four highest concentrations of
formaldehyde measured among allNMP sites were measured at NBIL.
»«» One preprocessed daily measurement of formaldehyde from NBIL was greater than
its associated acute MRL noncancer health risk benchmark. None of the quarterly or
annual average concentrations of the pollutants of interest, where they could be
calculated, were greater than their associated intermediate or chronic MRL
noncancer health risk benchmarks.
11-50
-------
12.0 Sites in Indiana
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the UATMP and CSATAM sites in Indiana, and integrates these
concentrations with emissions, meteorological, and risk information. Data generated by sources
other than ERG are not included in the data analyses contained in this report. Readers are
encouraged to refer to Sections 1 through 4 and the glossary (Appendix P) for detailed
discussions and definitions regarding the various data analyses presented below.
12.1 Site Characterization
This section characterizes the Indiana monitoring sites by providing geographical and
physical information about the location of the sites and the surrounding areas. This information
is provided to give the reader insight regarding factors that may influence the air quality near the
sites and assist in the interpretation of the ambient monitoring measurements.
One Indiana site (INDEM) is located in the Chicago-Naperville-Joliet, IL-IN-WI MSA,
while a second site (WPIN) is located in the Indianapolis-Carmel, IN MSA. Figures 12-1 and
12-2 are composite satellite images retrieved from ArcGIS Explorer showing the monitoring
sites in their urban locations. Figures 12-3 and 12-4 identify point source emissions locations by
source category for INDEM and WPIN, respectively, as reported in the 2008 NEI for point
sources. Note that only sources within 10 miles of the sites are included in the facility counts
provided in Figures 12-3 and 12-4. Thus, sources outside the 10-mile radius have been grayed
out, but are visible on the maps to show emissions sources outside the 10-mile boundary. A
10-mile boundary was chosen to give the reader an indication of which emissions sources and
emissions source categories could potentially have a direct effect on the air quality at the
monitoring sites. Further, this boundary provides both the proximity of emissions sources to the
monitoring sites as well as the quantity of such sources within a given distance of the sites.
Table 12-1 describes the area surrounding each monitoring site by providing supplemental
geographical information such as land use, location setting, and locational coordinates.
12-1
-------
Figure 12-1. Gary, Indiana (INDEM) Monitoring Site
to
to
-------
Figure 12-2. Indianapolis, Indiana (WPIN) Monitoring Site
-------
Figure 12-3. NEI Point Sources Located Within 10 Miles of INDEM
Legend
@ INDEM UATMP site 10 mile radius [
Source Category Group (No. of Facilities)
-f Aircraft Operations (14)
• Asphalt Processing/Roofing Manufacturing (1)
B Bulk Terminals/Bulk Plants (7)
c Chemical Manufacturing (5)
N Coke Battery (1)
* Electricity Generation via Combustion (6)
© Fabricated Metal Products (3)
F Food Processing/Agriculture (1)
+ Gypsum Manufacturing (1)
tf Hot Mix Asphalt Plant (1)
• Landfill (1)
>• Lime Manufacturing (1)
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
County boundary
•* Mine/Quarry (3)
V Mineral Products (7)
? Miscellaneous Commercial/Industrial (24)
M Miscellaneous Manufacturing (1)
• Oil and/or Gas Production (3)
^ Petroleum Refinery (1)
1 Primary Metal Production (1)
H Pulp and Paper Plant/Wood Products (1)
R Rubber and Miscellaneous Plastics Products (1)
> Solid Waste Disposal - Commercial/Institutional (1)
V Steel Mill (14)
s Surface Coating (1)
* Transportation and Marketing of Petroleum Products (6)
12-4
-------
Figure 12-4. NEI Point Sources Located Within 10 Miles of WPIN
Legend
BB'lStTW 8S'10'0-W WSXnW WQncrW SS-'SSXrW
Note: Due to facility density and collocation, [he total facilities
displayed may not represent all facilities, within the area of interest
WPIN UATMP site
10 mile radius
County boundary
Source Category Group (No. of Facilities)
•i« Aerospace/Aircraft Manufacturing (2)
-f Aircraft Operations (26)
B Bulk Terminals/Bulk Plants (1)
c Chemical Manufacturing (1)
* Electricity Generation via Combustion (3)
© Fabricated Metal Products (2)
fc» Flexible Polyurethane Foam Production (2)
A Grain Handling (2)
•® Institutional -school (1)
•
Iron and Steel Foundry (1)
Landfill (1)
Miscellaneous Commercial/Industrial (1)
Municipal Waste Combustor (1)
Oil and/or Gas Production (1)
Pharmaceutical Manufacturing (2)
Steel Mill(1)
Surface Coating (4)
Transportation Equipment (1)
Wastewater Treatment (1)
12-5
-------
Table 12-1. Geographical Information for the Indiana Monitoring Sites
Site
Code
INDEM
WPIN
AQS Code
18-089-0022
18-097-0078
Location
Gary
Indianapolis
County
Lake
Marion
Micro- or
Metropolitan
Statistical Area
Chicago-Joliet-
Naperville, IL-IN-
WIMSA
(Gary Div)
Indianapolis-
Carmel, IN MSA
Latitude
and
Longitude
41.606667,
-87.304722
39.811097,
-86.114469
Land Use
Industrial
Residential
Location
Setting
Urban/City
Center
Suburban
Additional Ambient Monitoring Information1
VOC, SO2, NO, NO2, NOX, PAMS, O3,
Meteorological parameters, PM10, Black carbon,
UV Carbon, PM2 5, and PM2 5 Speciation,
Tetrahydrofuran, 1,4-Dioxane.
TSP Metals, CO, VOC, SNMOC, SO2, NOy, NO, O3,
Meteorological parameters, PM10, Black carbon,
UV Carbon, PM2 5, and PM2 5 Speciation,
Tetrahydrofuran, 1,4-Dioxane, PM Coarse.
These monitoring sites report additional pollutants to AQS (EPA, 2012h); however, these data are not generated by ERG and are therefore not included in this report.
to
-------
INDEM is located in Gary, Indiana, a few miles east of the Indiana-Illinois border and
southeast of Chicago. Gary is located on the southernmost bank of Lake Michigan. The site is
just north of 1-90, the edge of which can be seen in the bottom left portion of Figure 12-1, and
1-65. Although INDEM resides on the Indiana Dunes National Lakeshore, the surrounding area
is highly industrialized, as shown in Figure 12-1, and several railroads transverse the area.
Figure 12-3 shows that the majority of point sources are located to the west of INDEM. The
sources closest to INDEM are a mine/quarry, a steel mill, and a facility that falls into the
miscellaneous commercial/industrial category. The source categories with the highest number of
sources within 10 miles of INDEM include steel mills, aircraft operations, mineral products, and
bulk terminals and plants.
WPIN is located in the parking lot of George Washington Park, near East 30th Street in
northeast Indianapolis. Figure 12-2 shows that the area surrounding WPIN is suburban and
residential, with little industry in close proximity. A church and a charitable organization are
located across the street from Washington Park, as is Oscar Charleston Park. Figure 12-4 shows
that the majority of point sources are located to the south and southwest of WPIN, towards the
center of Marion County. The source category with the highest number of sources near WPIN is
the aircraft operations source category, which include airports as well as small runways,
heliports, or landing pads. The sources closest to WPIN are an aircraft operations facility and a
fabricated metal products facility.
Table 12-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the areas surrounding the Indiana
monitoring sites. Table 12-2 also includes a vehicle registration-to-county population ratio
(vehicles-per-person) for each site. In addition, the population within 10 miles of each site is
presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-level
vehicle registration-to-population ratio to the 10-mile population surrounding each monitoring
site. Table 12-2 also contains annual average daily traffic information. Finally, Table 12-2
presents the daily VMT for Lake and Marion Counties.
12-7
-------
Table 12-2. Population, Motor Vehicle, and Traffic Information for the Indiana Monitoring
Sites
Site
INDEM
WPIN
Estimated
County
Population1
495,981
904,878
County-level
Vehicle
Registration2
182,989
204,908
Vehicles per
Person
(Registration:
Population)
0.37
0.23
Population
within 10
miles3
406,979
787,003
Estimated
10-mile
Vehicle
Ownership
150,152
178,215
Annual
Average
Daily
Traffic4
52,440
143,410
County-
level Daily
VMT5
11,801,000
35,081,000
1 County-level population estimates reflect data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the IN Bureau of Motor Vehicles (IN BMV, 2011)
3 10-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2009 data from the Indiana DOT (IN DOT, 2009)
5 County-level VMT reflects 2010 data from the Indiana DOT (IN DOT, 2010)
Observations from Table 12-2 include the following:
• Marion County has almost twice the county population as Lake County. The
difference between the two counties decreases significantly for the county-level
vehicle registration. The 10-mile population and estimated vehicle registration follow
a similar pattern as the county-level values.
• The county-level and 10-mile populations are in the middle third of populations
among NMP sites. The county-level and 10-mile vehicle registrations are in the
bottom third among the NMP sites.
• The vehicle-per-person ratios for both Indiana sites are among the lowest ratios
among the NMP sites.
• WPIN experiences a significantly higher traffic volume than INDEM. The traffic
estimate for WPIN is based on data from 1-70 between exits 85 and 87 while the
traffic volume for INDEM is based on data from 1-90 at 12/20. The traffic volume
near WPIN is the eighth highest among NMP sites.
• The VMT for Marion County is more than three times higher than the VMT for Lake
County. Marion County VMT ranked eighth among counties with NMP sites, while
the VMT for Lake County is in the middle of the range, at 17th.
12.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in Indiana on sample days, as well as over the course of the year.
12-8
-------
12.2.1 Climate Summary
The city of Gary is located to the southeast of Chicago, and at the southern-most tip of
Lake Michigan. Gary's proximity to Lake Michigan is an important factor controlling the
weather of the area. In the summer, warm temperatures can be suppressed, while cold winter
temperatures are often moderated. Winds that blow across Lake Michigan and over Gary in the
winter can provide abundant amounts of lake-effect snow while lake breezes can bring relief
from summer heat (Bair, 1992; Gary, 2012; and ISCO, 2002).
The city of Indianapolis is located in the center of Indiana, and experiences a temperate
continental climate and frequently changing weather patterns. Summers are warm and often
humid, as moist air flows northward out of the Gulf of Mexico. Winters are chilly with
occasional Arctic outbreaks. Precipitation is spread rather evenly throughout the year, with much
of the spring and summer precipitation resulting from showers and thunderstorms. The
prevailing wind direction is southwesterly (Bair, 1992 and ISCO, 2002).
12.2.2 Meteorological Conditions in 2010
Hourly meteorological data from NWS weather stations nearest these sites were retrieved
for 2010 (NCDC, 2010). The two closest weather stations are located at Lansing Municipal
Airport (near INDEM) and, Eagle Creek Airpark (near WPIN), WBAN 04879 and 53842,
respectively. Additional information about these weather stations, such as the distance between
the sites and the weather stations, is provided in Table 12-3. These data were used to determine
how meteorological conditions on sample days vary from normal conditions throughout the year.
Table 12-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 12-3 is the
95 percent confidence interval for each parameter. As shown in Table 12-3, average
meteorological conditions on sample days were representative of average weather conditions
throughout the year for INDEM and WPIN.
12-9
-------
Table 12-3. Average Meteorological Conditions near the Indiana Monitoring Sites
Closest NWS Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Gary, Indiana - INDEM
Lansing Municipal
Airport
04879
(41.54, -87.52)
11.36
miles
241°
(WSW)
Sample
Day
2010
60.7
±5.5
60.9
±2.3
51.8
±5.1
52.1
±2.1
39.6
±4.7
39.9
±2.0
45.8
±4.5
46.1
±1.9
66.2
±2.6
66.6
±1.2
NA
NA
5.5
±0.9
6.0
±0.4
Indianapolis, Indiana - WPIN
Eagle Creek Airpark
53842
(39.83, -86.30)
9.13
miles
270°
(W)
Sample
Day
2010
62.2
±5.6
62.9
±2.3
53.5
±5.3
54.1
±2.2
41.4
±4.7
42.1
±2.0
47.3
±4.7
47.9
±1.9
66.7
±2.8
67.4
±1.2
1016.2
±1.6
1016.0
±0.7
5.3
±0.8
5.4
±0.3
to
o
1 Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
NA= Sea level pressure was not recorded at the Lansing Municipal Airport
-------
12.2.3 Back Trajectory Analysis
Figure 12-5 is the composite back trajectory map for days on which samples were
collected at the INDEM monitoring site in 2010. Included in Figure 12-5 are four back
trajectories per sample day. Figure 12-6 is the corresponding cluster analysis for 2010. Similarly,
Figure 12-7 is the composite back trajectory map for days on which samples were collected at
WPIN and Figure 12-8 is the corresponding cluster analysis. An in-depth description of these
maps and how they were generated is presented in Section 3.5.2.1. For the composite maps, each
line represents the 24-hour trajectory along which a parcel of air traveled toward the monitoring
site on a given sample day and time. For the cluster analyses, each line corresponds to a back
trajectory representative of a given cluster of trajectories. For all maps, each concentric circle
around the sites in Figures 12-5 through 12-8 represents 100 miles.
Figure 12-5. 2010 Composite Back Trajectory Map for INDEM
12-11
-------
Figure 12-6. Back Trajectory Cluster Map for INDEM
Figure 12-7. 2010 Composite Back Trajectory Map for WPIN
12-12
-------
Figure 12-8. Back Trajectory Cluster Map for WPIN
Observations from Figures 12-5 and 12-6 for INDEM include the following:
• Back trajectories originated from a variety of directions at INDEM, although less
frequently from the southeast.
• The 24-hour air shed domain was similar in size to other NMP sites, as the farthest
away a trajectory originated was over central North Dakota, or greater than 700 miles
away. However, the average trajectory length was 246 miles, and most trajectories
(90 percent) originated within 450 miles of INDEM.
• The longest trajectories originated to the northwest, as represented by the trajectory
originating over Minnesota (9 percent), with additional trajectories also originating
from this direction, but of shorter length (23 percent). Longer trajectories also
originated to north over the Great Lakes and into Canada. Back trajectories from the
south and southwest are combined into one cluster (25 percent). Back trajectories
from the northeast and southeast are represented by the short cluster (11 percent).
Trajectories with an easterly component were often shorter than trajectories
originating from other directions.
Observations from Figures 12-7 and 12-8 for WPIN include the following:
• The composite back trajectory map for WPIN shows that back trajectories originated
from a variety of directions, although less frequently from the southeast, similar to
INDEM.
12-13
-------
• The 24-hour air shed domain was similar in size compared to many other NMP
monitoring sites as the farthest away a trajectory originated was over southeast South
Dakota, or greater than 600 miles away. The trajectories originating to the northwest
tended to be the longest. The average trajectory length was 234 miles, while most
trajectories (88 percent) originated within 400 miles of WPIN.
• The cluster analysis for WPIN confirms that the longest trajectories originated to the
northwest, as represented by the trajectory originating over Iowa (9 percent).
Additional trajectories also originated from this direction, but were of shorter length
(10 percent). Trajectories originating from the south and southwest account for
another 35 percent of trajectories. Trajectories with an easterly component were often
shorter than trajectories originating from other directions.
12.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather stations near the Indiana sites, as presented in
Section 12.2.2, were uploaded into a wind rose software program to produce customized wind
roses, as described in Section 3.5.2.2. A wind rose shows the frequency of wind directions using
"petals" positioned around a 16-point compass, and uses different colors to represent wind
speeds.
Figure 12-9 presents three different wind roses for the INDEM monitoring site. First, a
historical wind rose representing 2003 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figure 12-10 presents the
three wind roses and distance map for WPIN.
12-14
-------
Figure 12-9. Wind Roses for the Lansing Municipal Airport Weather Station near INDEM
2003-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between INDEM and NWS Station
f'E':, T
12-15
-------
Figure 12-10. Wind Roses for the Indianapolis International Airport Weather Station near
WPIN
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between WPIN and NWS Station
12-16
-------
Observations from Figure 12-9 for INDEM include the following:
• The NWS weather station at Lancing Municipal Airport is the closest weather station
to INDEM, although it is located approximately 11.4 miles west-southwest of
INDEM. The location of the weather station is just east of the Illinois-Indiana state
line and farther inland than INDEM and thus, farther away from the influences of
Lake Michigan than INDEM.
• The historical wind rose for INDEM shows that winds from the south to south-
southwest and west are the predominant wind directions over the 2003-2009 time
frame. Northerly to northeasterly winds off Lake Michigan accounted for just less
than 20 percent of the wind measurements, as did calm winds. The strongest winds
blew from the south, southwest, and west of INDEM.
• The wind patterns shown on the 2010 wind rose resemble the wind patterns shown on
the historical wind rose, although the calm rate was just over 24 percent in 2010.
• The sample day wind rose also has predominant southerly, south-southwesterly, and
westerly winds, although there were fewer wind observations from the north-
northeast and northeast. In addition, the calm rate is higher on the sample day wind
rose, with greater than 27 percent of winds speeds less than 2 knots.
Observations from Figure 12-10 for WPIN include the following:
• The NWS weather station at Eagle Creek Airpark is the closest weather station to
WPIN and is located approximately 9.1 miles west of WPIN.
• Winds from the south, from the western quadrants, and from the north account for the
majority (greater than 55 percent) of wind observations from 1999-2009, while winds
from the eastern quadrants were observed less than one-third of the time. Calm winds
(< 2 knots) were observed for nearly 17 percent of observations. The strongest winds
tended to flow from the northwest.
• The wind patterns on the 2010 wind rose resemble the historical wind patterns,
although the calm rate was higher, at greater than 21 percent, and there were fewer
southerly to southwesterly winds.
• The sample day wind rose has a higher calm rate (24 percent) than the historical wind
rose, as the full-year wind rose does, but there were also fewer wind observations
from the southwest quadrant and more wind observations from the northwest on
sample days.
12-17
-------
12.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Indiana monitoring sites in
order to allow analysts and readers to focus on a subset of pollutants through the context of risk.
For each site, each pollutant's preprocessed daily measurement was compared to its associated
risk screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by each monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 12-4 presents the pollutants of interest for the Indiana monitoring sites. The
pollutants that failed at least one screen and contributed to 95 percent of the total failed screens
for each monitoring site are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of
interest are shaded and/or bolded. INDEM and WPIN sampled for carbonyl compounds only.
Table 12-4. Risk Screening Results for the Indiana Monitoring Sites
Pollutant
Screening
Value
(Hg/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Gary, Indiana - INDEM
Acetaldehyde
Formaldehyde
0.45
0.077
Total
61
61
122
61
61
122
100.00
100.00
100.00
50.00
50.00
50.00
100.00
Indianapolis, Indiana - WPIN
Acetaldehyde
Formaldehyde
Propionaldehyde
0.45
0.077
0.8
Total
56
56
1
113
56
56
56
168
100.00
100.00
1.79
67.26
49.56
49.56
0.88
49.56
99.12
100.00
Observations from Table 12-4 include the following:
• Formaldehyde, acetaldehyde, and propionaldehyde are the only carbonyl compounds
with risk screening values.
• Acetaldehyde and formaldehyde failed screens for INDEM. They contributed equally
to the total number of failed screens. Both pollutants failed 100 percent of the total
failed screens.
12-18
-------
• All three carbonyl compounds with risk screening values failed screens for WPIN.
Acetaldehyde and formaldehyde failed 100 percent of the total failed screens while
propionaldehyde failed only one screen.
12.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Indiana monitoring sites. Concentration averages are provided for the pollutants of interest
for each Indiana site, where applicable. Concentration averages for select pollutants are also
presented graphically for each site, where applicable, to illustrate how each site's concentrations
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at
each site, where applicable. Additional site-specific statistical summaries are provided in
Appendix L.
12.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each Indiana site, as described in Section 3.1. The quarterly average of a particular pollutant
is simply the average concentration of the preprocessed daily measurements over a given
calendar quarter. Quarterly average concentrations include the substitution of zeros for all non-
detects. A site must have a minimum of 75 percent valid samples of the total number of samples
possible within a given quarter for a quarterly average to be calculated. An annual average
includes all measured detections and substituted zeros for non-detects for the entire year of
sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the Indiana monitoring
sites are presented in Table 12-5, where applicable. Note that note that if a pollutant was not
detected in a given calendar quarter, the quarterly average simply reflects "0" because only zeros
substituted for non-detects were factored into the quarterly average concentration.
12-19
-------
Table 12-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest
for the Indiana Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Ug/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(Ug/m3)
4th
Quarter
Average
(Ug/m3)
Annual
Average
(Ug/m3)
Gary, Indiana - INDEM
Acetaldehyde
Formaldehyde
61/61
61/61
1.23
±0.29
1.63
±0.33
1.17
±0.24
2.08
±0.48
1.61
±0.22
3.10
±0.41
1.53
±0.34
2.76
±0.54
1.39
±0.14
2.41
±0.26
Indianapolis, Indiana - WPIN
Acetaldehyde
Formaldehyde
56/56
56/56
1.76
±0.32
2.28
±0.36
2.39
±0.46
3.54
±0.69
2.66
±0.78
4.09
±1.13
3.46
±0.76
4.51
±0.97
2.56
±0.33
3.58
±0.45
Observations for the Indiana sites from Table 12-5 include the following:
• Formaldehyde has the highest annual average concentration by mass of the pollutants
of interest for both INDEM and WPIN.
• The annual average concentrations of acetaldehyde and formaldehyde are higher for
WPIN than INDEM.
• Concentrations of both acetaldehyde and formaldehyde tended to be highest during
the third and fourth quarters of 2010 for both sites; however, the difference is not
statistically significant.
• The confidence interval for WPIN's third and fourth quarter formaldehyde averages
are rather large, indicating the potential influence of outliers. A review of the data
shows that the highest concentration for WPIN was measured on October 11, 2010
(5.21 |ig/m3). The four concentrations of formaldehyde greater than 4 |ig/m3 were
measured in July, August, October, and November.
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for the Indiana
sites from those tables include the following:
• As shown in Table 4-10, WPIN's annual average concentration of formaldehyde is
the fourth highest average among NMP sites sampling this pollutant. WPIN also has
the sixth highest annual average concentration of acetaldehyde.
• INDEM does not appear in Table 4-10. Its annual average concentration of
formaldehyde ranks 15th and its annual average concentration of acetaldehyde ranks
19th among NMP sites sampling carbonyl compounds.
12-20
-------
12.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for acetaldehyde and
formaldehyde were created for both INDEM and WPIN. Figures 12-11 and 12-12 overlay the
sites' minimum, annual average, and maximum concentrations onto the program-level minimum,
first quartile, average, median, third quartile, and maximum concentrations, as described in
Section 3.5.3.
Figure 12-11. Program vs. Site-Specific Average Acetaldehyde Concentration
0
1
Concentration (ng/m3)
Program: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Average
Site:
Site Average Site Minimum/Maximum
o
12-21
-------
Figure 12-12. Program vs. Site-Specific Average Formaldehyde Concentration
1 1 1 1 1 1 1 1
• lo
r
25 30
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Observations from Figures 12-11 and 12-12 include the following:
• Figure 12-11 shows that WPIN's annual average acetaldehyde concentration is
greater than the annual average acetaldehyde concentration for INDEM. WPIN's
annual average concentration is greater than program-level average for
acetaldehyde as well as the third quartile for the program. Conversely, INDEM's
annual average concentration is less than program-level average as well as the
median for the program. There were no non-detects of acetaldehyde measured at
either site.
• Figure 12-12 shows that WPIN's annual average formaldehyde concentration is
greater than the program average formaldehyde concentration while INDEM's
annual average concentration is very similar to the program-level average. The
maximum concentrations for both sites are well below the maximum
concentration measured across the program. There were no non-detects of
formaldehyde measured at either site.
12.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. INDEM has sampled carbonyl compounds since 2004; thus, Figures 12-13 and
12-4 present the 3-year rolling statistical metrics for acetaldehyde and formaldehyde for INDEM,
respectively. The statistical metrics presented for assessing trends include the substitution of
zeros for non-detects.
12-22
-------
Figure 12-13. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at INDEM
»
2006-2008
Three-Year Period
• 5th Percentile — Minimum - Median — Maximum • 95thPer
-------
Observations from Figure 12-13 for acetaldehyde measurements at INDEM include the
following:
• The maximum acetaldehyde concentration (13.8 |ig/m3) was measured on
June 14, 2004. An additional four concentrations measured at INDEM were greater
than 10 |ig/m3 (one in 2006 and three in 2008).
• Most of the statistical parameters show a slight increasing trend through the
2006-2008 time frame, but show fairly substantial decreases thereafter. The average
and median concentrations for the last two 3-year periods decreased below 2004-2006
levels.
• The carbonyl compound samplers were switched out in 2009, which seems to have
had a significant impact on the concentrations measured, particularly with respect to
formaldehyde, which is discussed in more detail below.
• There was a 3-month gap in sampling between September and November 2005 at the
INDEM site, which is denoted in Figure 12-13.
Observations from Figure 12-14 for formaldehyde measurements at INDEM include the
following:
• Five formaldehyde concentrations greater than 400 |ig/m3 were measured in the
summer of 2008 (ranging from 414 to 499 |ig/m3). While these are extremely high
values of formaldehyde, concentrations of formaldehyde have been historically high
at this site, as shown by the statistics in Figure 12-14. There have been 38
measurements of formaldehyde greater than 100 |ig/m3, ranging from one in 2007 up
to 13 in 2005 (and none measured during 2009 or 2010).
• The rolling average and the median concentrations are not similar to each other; the
median is roughly half of the average for each time period. This reflects the influence
of the outliers on the average concentrations compared to the median concentrations,
which are influenced less by outliers.
• The rolling average and median concentrations changed little through the 2006-2008
time frame, but exhibit fairly substantial decreases afterward. Note, however, that the
confidence intervals are wide for each of the averages shown due to the large range of
concentrations measured at this site. The addition of future years not including years
with the higher concentrations will be more telling.
• The rolling averages shown for INDEM are the highest of any rolling averages
calculated for any other NMP site measuring formaldehyde. INDEM's formaldehyde
concentrations have historically been higher than any other NMP site sampling
carbonyl compounds. During the summer PAMS season, which begins on June 1, a
state-owned multi-channel collection system is used at INDEM to collect multiple
samples per day. At the end of each PAMS season, sample collection goes back to a
state-owned single-channel collection system. The multi-channel sampler used at
INDEM during the PAMS season was replaced in 2009 and their formaldehyde
12-24
-------
concentrations decreased substantially (as did their acetaldehyde concentrations, but
the difference is less dramatic). Given that the elevated concentrations of
formaldehyde were typically measured during the summer, this sampler change could
account for the differences in the concentrations for 2009 and 2010 compared to
previous years. Thus, the elevated concentrations from previous years were likely
related to the multi-channel collection equipment and may not reflect the actual levels
in the ambient air. The annual average concentrations for 2010 for both acetaldehyde
and formaldehyde were similar in magnitude to those reported for 2009 in the 2008-
2009 NMP report. The addition of future years not including years with the higher
concentrations will be more telling.
12.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at each
Indiana monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
12.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Indiana monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where available.
As described in Section 3.3, acute risk results from exposures of 1 to 14 days; intermediate risk
results from exposures of 15 to 364 days; and chronic risk results from exposures of 1 year or
greater. The preprocessed daily measurements of the pollutants of interest for each site were
compared to the acute MRLs; the quarterly averages were compared to the intermediate MRLs;
and the annual averages were compared to the chronic MRLs.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the Indiana monitoring sites were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
the Indiana monitoring sites.
12.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Indiana monitoring sites and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
12-25
-------
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 12-6, where applicable.
Table 12-6. Cancer and Noncancer Surrogate Risk Approximations for the Indiana
Monitoring Sites
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs.
# of Samples
Annual
Average
(jig/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Gary, Indiana - INDEM
Acetaldehyde
Formaldehyde
0.0000022
0.000013
0.009
0.0098
61/61
61/61
1.39
±0.14
2.41
±0.26
3.05
31.27
0.15
0.25
Indianapolis, Indiana - WPIN
Acetaldehyde
Formaldehyde
0.0000022
0.000013
0.009
0.0098
56/56
56/56
2.56
±0.33
3.58
±0.45
5.63
46.56
0.28
0.37
Observations for the Indiana sites from Table 12-6 include the following:
• For both sites, the annual average concentration of formaldehyde is greater than the
annual average concentrations of acetaldehyde.
• For each site, the cancer risk approximation for formaldehyde is an order of
magnitude higher than the cancer risk approximation for acetaldehyde.
• The cancer risk approximation for formaldehyde for WPIN (46.56 in-a-million) is the
fourth highest calculated cancer risk approximation among all site-specific pollutants
of interest (INDEM's ranked 15th).
• None of the noncancer risk approximations are greater than 1.0 for the pollutants of
interest for INDEM and WPIN.
12.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 12-7 and 12-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 12-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 12-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
12-26
-------
Table 12-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Indiana Monitoring Sites
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Cancer Risk
Approximation
Pollutant (in-a-million)
Gary, Indiana (Lake County) - INDEM
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
Coke Oven Emissions, PM
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
Trichloroethylene
231.53
157.04
107.50
100.19
33.43
29.96
29.83
4.72
2.39
2.23
Coke Oven Emissions, PM
Formaldehyde
Hexavalent Chromium, PM
Benzene
Naphthalene
Arsenic, PM
1,3 -Butadiene
Nickel, PM
Ethylbenzene
Cadmium, PM
3.31E-02
2.04E-03
1.81E-03
1.81E-03
1.01E-03
9.93E-04
8.99E-04
6.66E-04
2.69E-04
2.63E-04
Formaldehyde 31.27
Acetaldehyde 3.05
Indianapolis, Indiana (Marion County) - WPIN
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Tetrachloroethylene
Dichloromethane
Trichloroethylene
POM, Group 2b
489.79
356.47
217.73
205.36
67.18
34.96
14.44
7.84
6.85
6.52
Formaldehyde
Benzene
1,3 -Butadiene
Hexavalent Chromium, PM
Naphthalene
Arsenic, PM
POM, Group 3
POM, Group 2b
Ethylbenzene
Nickel, PM
4.63E-03
3.82E-03
2.02E-03
1.23E-03
1.19E-03
1.10E-03
6.97E-04
5.74E-04
5.44E-04
4.59E-04
Formaldehyde 46.56
Acetaldehyde 5.63
to
to
-------
Table 12-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Indiana Monitoring Sites
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer Toxicity
Weight
Top 10 Noncancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Gary, Indiana (Lake County) - INDEM
Toluene
Xylenes
Methanol
Ethylene glycol
Hexane
Benzene
Formaldehyde
Hydrochloric acid
Ethylbenzene
Acetaldehyde
676.97
528.24
315.13
271.89
231.90
231.53
157.04
153.19
107.50
100.19
Acrolein
Manganese, PM
Lead, PM
Cyanide Compounds, gas
Formaldehyde
Nickel, PM
Arsenic, PM
1,3 -Butadiene
Cadmium, PM
Acetaldehyde
503,495.00
468,665.08
109,768.53
36,721.25
16,024.72
15,405.79
15,399.37
14,981.21
14,626.32
11,131.89
Formaldehyde 0.25
Acetaldehyde 0.15
Indianapolis, Indiana (Marion County) - WPIN
Toluene
Xylenes
Methanol
Ethylene glycol
Benzene
Hydrochloric acid
Formaldehyde
Hexane
Ethylbenzene
Acetaldehyde
1,373.22
935.63
684.51
504.44
489.79
460.45
356.47
269.35
217.73
205.36
Acrolein
Formaldehyde
1,3 -Butadiene
Hydrochloric acid
Acetaldehyde
Arsenic, PM
Benzene
Naphthalene
Lead, PM
Nickel, PM
1,220,725.02
36,374.58
33,589.47
23,022.60
22,818.28
17,017.69
16,326.44
11,654.54
11,113.23
10,619.54
Formaldehyde 0.37
Acetaldehyde 0.28
to
to
oo
-------
The pollutants listed in Tables 12-7 and 12-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. Further, the cancer and noncancer surrogate risk approximations based on each site's
annual averages are limited to those pollutants for which each respective site sampled. As
discussed in Section 12.3, INDEM and WPIN sampled for carbonyl compounds only. In
addition, the cancer and noncancer surrogate risk approximations are limited to those pollutants
with enough data to meet the criteria for annual averages to be calculated. A more in-depth
discussion of this analysis is provided in Section 3.5.5.3.
Observations from Table 12-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the three highest emitted pollutants
with cancer UREs in both Marion and Lake County.
• Coke oven emissions, formaldehyde, and hexavalent chromium are the pollutants
with the highest toxicity-weighted emissions (of the pollutants with cancer UREs) for
Lake County while formaldehyde, benzene, and 1,3-butadiene are the pollutants with
the highest toxicity-weighted emissions for Marion County.
• Six of the highest emitted pollutants in Lake County also have the highest toxicity-
weighted emissions. For Marion County, five of the highest emitted pollutants also
have the highest toxicity-weighted emissions.
• While several metals (arsenic, cadmium, hexavalent chromium, and nickel) are
among the pollutants with the highest toxicity-weighted emissions for both counties,
none of these appear on the list of highest emitted pollutants for either county. This
demonstrates that a pollutant does not have to be emitted in large quantities to be
toxic.
• Acetaldehyde and formaldehyde are the only pollutants of interest for the Indiana
monitoring sites. Acetaldehyde and formaldehyde appear on both emissions-based
lists for INDEM and WPIN, with formaldehyde ranking relatively high on both lists.
Observations from Table 12-8 include the following:
• While toluene is the highest emitted pollutant with a noncancer RfC in both counties,
the toluene emissions in Marion County are nearly twice that of Lake County.
Xylenes and methanol are the second and third highest emitted pollutants in both
counties, with a similar pattern in quantities of emissions.
• Acrolein is the pollutant with the highest toxicity-weighted emissions (of the
pollutants with noncancer RfCs) for both counties. Manganese and lead rank second
12-29
-------
and third for Lake County, while formaldehyde and 1,3-butadiene rank second and
third for Marion County.
• Only two of the highest emitted pollutants in Lake County also have the highest
toxicity-weighted emissions (formaldehyde and acetaldehyde). Several metals
(manganese, lead, nickel, arsenic, and cadmium) are among the pollutants with the
highest toxicity-weighted emissions for Lake County, although none of these appear
on the list of highest emitted pollutants.
• Four of the highest emitted pollutants in Marion County also have the highest
toxicity-weighted emissions (formaldehyde, acetaldehyde, hydrochloric acid, and
benzene).
• Acetaldehyde and formaldehyde appear on both emissions-based lists for INDEM and
WPIN.
12.6 Summary of the 2010 Monitoring Data for INDEM and WPIN
Results from several of the data treatments described in this section include the
following:
»«» Two carbonyl compounds failed screens for INDEM and three failed screens for
WPIN.
»«» Formaldehyde had the highest annual average concentration for each of the Indiana
monitoring sites. The annual averages concentrations for WPIN were higher than the
annual average concentrations for INDEM.
»«» Concentrations of formaldehyde and acetaldehyde exhibit a decreasing trend at
INDEM.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
12-30
-------
13.0 Site in Kentucky
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in Kentucky, and integrates these concentrations
with emissions, meteorological, and risk information. Data generated by sources other than ERG
are not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
13.1 Site Characterization
This section characterizes the Kentucky monitoring site by providing geographical and
physical information about the location of the site and the surrounding area. This information is
provided to give the reader insight regarding factors that may influence the air quality near the
site and assist in the interpretation of the ambient monitoring measurements.
The Kentucky monitoring site is located near Grayson Lake in northeast Kentucky.
Figure 13-1 is a composite satellite image retrieved from ArcGIS Explorer showing the
monitoring site in its rural location. Figure 13-2 identifies point source emissions locations by
source category, as reported in the 2008 NEI for point sources. Note that only sources within
10 miles of the site are included in the facility counts provided in Figure 13-2. Thus, sources
outside the 10-mile radius have been grayed out, but are visible on the map to show emissions
sources outside the 10-mile boundary. A 10-mile boundary was chosen to give the reader an
indication of which emissions sources and emissions source categories could potentially have a
direct effect on the air quality at the monitoring site. Further, this boundary provides both the
proximity of emissions sources to the monitoring site as well as the quantity of such sources
within a given distance of the site. Table 13-1 describes the area surrounding the monitoring site
by providing supplemental geographical information such as land use, location setting, and
locational coordinates.
13-1
-------
Figure 13-1. Grayson, Kentucky (GLKY) Monitoring Site
to
-------
Figure 13-2. NEI Point Sources Located Within 10 Miles of GLKY
Mote; Due to facility density and education, the total facilities
displayed may not represent all facilities within the area of interest,
Legend
© GLKY NATTS site
10 mile radius
County boundary
Source Category Group (No. of Facilities)
•f1 Aircraft Operations (1)
Brick Manufacturing & Structural Clay (2)
B Bulk Terminals/Bulk Plants (1)
F Food Processing/Agriculture (1)
* Hot Mix Asphalt Plant (1)
x Mine/Quarry (1)
? Miscellaneous Commercial/Industrial (1)
13-3
-------
Table 13-1. Geographical Information for the Kentucky Monitoring Site
Site
Code
GLKY
AQS Code
21-043-005
Location
Gray son
County
Carter
Micro- or
Metropolitan
Statistical Area
Not in an MSA
Latitude
and
Longitude
38.238333,
-82-988333
Land Use
Residential
Location
Setting
Rural
Additional Ambient Monitoring Information1
Carbonyl compounds, O3, Meteorological parameters,
PMio, PM10 Speciation, PM25, andPM25 Speciation
BOLD ITALICS = EPA-designated NATTS Site.
-------
Grayson Lake is located in northeast Kentucky, south of the town of Grayson, and west
of the Huntington-Ashland, WV-KY MSA. The Little Sandy River feeds into Grayson Lake,
which is a U.S. Army Corps of Engineers-managed project, and part of the Kentucky State Parks
system. The lake is narrow and winding, as shown in Figure 13-1, with sandstone cliffs rising to
up to 200 feet above the lake surface (KY, 2012 and ACE, 2012). The closest road to the
monitoring site is a service road feeding into Camp Grayson. This site serves as the Grayson
Lake NATTS site. Figure 13-2 shows that few point sources surround GLKY and that most of
them are on the outer periphery of the 10-mile radius around GLKY. This is not surprising given
the rural nature of the area and that Grayson Lake is located roughly in the center of the 10-mile
radii in Figure 13-2, oriented from northeast to southwest. Sources within 10 miles of GLKY are
involved in aircraft operations, brick and structural clay manufacturing, food processing, and
mining, among others.
Table 13-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the
Kentucky monitoring site. Table 13-2 also includes a vehicle registration-to-county population
ratio (vehicles-per-person) for GLKY. In addition, the population within 10 miles of the site is
presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-level
vehicle registration-to-population ratio to the 10-mile population surrounding the monitoring
site. Table 13-2 also contains annual average daily traffic information. Finally, Table 13-2
presents the daily VMT for Carter County.
Table 13-2. Population, Motor Vehicle, and Traffic Information for the Kentucky
Monitoring Site
Site
GLKY
Estimated
County
Population1
27,675
County-level
Vehicle
Registration2
36,031
Vehicles per
Person
(Registration:
Population)
1.30
Population
within 10
miles3
16,880
Estimated
10-mile
Vehicle
Ownership
21,977
Annual
Average
Daily
Traffic4
428
County-
level
Daily
VMT5
1,164,000
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the Kentucky Transportation Cabinet (KYTC, 201 la)
310-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2009 data from the Kentucky Transportation Cabinet (KYTC, 2009)
5 County-level VMT reflects 2010 data from the Kentucky Transportation Cabinet (KYTC, 201 Ib)
BOLD ITALICS = EPA-designated NATTS Site.
13-5
-------
Observations from Table 13-2 include the following:
• The Carter County population is the second lowest compared to counties with NMP
sites (behind only UCSD). The 10-mile population for GLKY is also on the low side
compared to other sites. The corresponding vehicle ownership data mimicked these
rankings. The rather low population and vehicle ownership compared to other NMP
sites is not surprising given the rural nature of the surrounding area.
• The vehicle-per-person ratio is among the higher ratios compared to other NMP sites.
• The traffic data for GLKY came from the intersection of State Road 1496 with Camp
Webb Road, one of several secondary roads leading to Grayson Lake. This site has
the third lowest traffic volume among NMP sites.
• The Carter County daily VMT is the second lowest compared to other counties with
NMP sites (where VMT data were available), behind only Union County, South
Dakota.
13.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in Kentucky on sample days, as well as over the course of the year.
13.2.1 Climate Summary
Kentucky experiences a continental climate, where conditions tend to be slightly cooler
and drier in the northeast portion of the state and warmer and wetter in the southwest portion.
Kentucky's mid-latitude location ensures an active weather pattern, in a convergence zone
between cooler air from the north and warm, moist air from the south. The state enjoys all four
seasons. Summers are persistently warm and humid; winters are cloudy but not harsh; and spring
and fall are pleasant. Precipitation is well distributed throughout the year, although fall tends to
be driest and spring wettest (NCDC, 2012).
13.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest this site were retrieved
for 2010 (NCDC, 2010). The closest weather station to GLKY is located at Tri-State/Ml.
Ferguson Field Airport (WBAN 03860). Additional information about this weather station, such
as the distance between the site and the weather station, is provided in Table 13-3. These data
were used to determine how meteorological conditions on sample days vary from normal
conditions throughout the year.
13-6
-------
Table 13-3. Average Meteorological Conditions near the Kentucky Monitoring Site
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Grayson, Kentucky - GLKY
Tri-St/MJ.
Ferguson Field
Airport
03860
(38.38, -82.56)
24.27
miles
58°
(ENE)
Sample
Day
2010
64.4
±5.4
64.9
±2.2
54.9
±4.8
55.4
±1.9
43.9
±4.8
44.2
±1.9
49.3
±4.4
49.7
±1.8
70.0
±3.2
69.5
±1.2
1016.4
± 1.6
1016.3
±0.6
3.9
±0.5
4.0
±0.2
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
Table 13-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 13-3 is the
95 percent confidence interval for each parameter. As shown in Table 13-3, average
meteorological conditions on sample days near GLKY were representative of average weather
conditions throughout the year.
13.2.3 Back Trajectory Analysis
Figure 13-3 is the composite back trajectory map for days on which samples were
collected at the GLKY monitoring site in 2010. Included in Figure 13-3 are four back trajectories
per sample day. Figure 13-4 is the corresponding cluster analysis for 2010. An in-depth
description of these maps and how they were generated is presented in Section 3.5.2.1. For the
composite map, each line represents the 24-hour trajectory along which a parcel of air traveled
toward the monitoring site on a given sample day and time. For the cluster analysis, each line
corresponds to a back trajectory representative of a given cluster of trajectories. For both maps,
each concentric circle around the site in Figures 13-3 and 13-4 represents 100 miles.
Observations from Figures 13-3 and 13-4 for GLKY include the following:
• Back trajectories originated from a variety of directions at GLKY, the majority of
which originated to the south, west and northwest, and north.
• The farthest away a back trajectory originated was north-central Missouri, or nearly
500 miles away; however, the average trajectory length was 194 miles and 88 percent
of trajectories originated within 350 miles of the monitoring site.
• The cluster program grouped the trajectories originating from the northwest to
northeast to southeast and within 200 miles of the site together, which are represented
by the short cluster originating over West Virginia (29 percent). The trajectories
originated over southern Ohio, West Virginia, southwest Virginia, and northeast
Kentucky. Another 21 percent of back trajectories originated from the west, although
these trajectories are represented by two cluster trajectories (6 percent and 15 percent)
to represent the different lengths of the trajectories. One quarter of trajectories
originated to the south of GLKY.
13-8
-------
Figure 13-3. 2010 Composite Back Trajectory Map for GLKY
Figure 13-4. Back Trajectory Cluster Map for GLKY
13-9
-------
13.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at the Tri-State/MJ. Ferguson Field
Airport near GLKY were uploaded into a wind rose software program to produce customized
wind roses, as described in Section 3.5.2.2. A wind rose shows the frequency of wind directions
using "petals" around a 16-point compass, and uses different colors to represent wind speeds.
Figure 13-5 presents three different wind roses for the GLKY monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at this location.
Observations from Figure 13-5 for GLKY include the following:
• The Tri-State/MJ. Ferguson Field weather station is located over 24 miles to the
east-northeast of GLKY and just across the state border in West Virginia.
• The historical wind rose shows that calm winds were observed for more than
22 percent of the hourly measurements near GLKY. Winds from the south to
southwest to west make up the majority of observations near GLKY, particularly
those from south-southwest.
• The wind patterns on the 2010 wind rose are similar to those on the historical wind
rose, but calm winds accounted for a higher percentage of the wind observations in
2010 (28 percent).
• The sample day wind rose has an even higher percentage of calm winds (30 percent)
than the historical and full-year wind roses. Although winds from the south-southwest
were still observed the most, the percentage of wind observations is more evenly
distributed between the northwest and southwest quadrants, as well as a northeasterly
direction.
13-10
-------
Figure 13-5. Wind Roses for the Tri-State/M. J. Ferguson Field Airport Weather Station
near GLKY
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between GLKY and NWS Station
A ----„,.
" ' ' ...
• • .
13-11
-------
13.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for GLKY in order to allow
analysts and readers to focus on a subset of pollutants through the context of risk. Each
pollutant's preprocessed daily measurement was compared to its associated risk screening value.
If the concentration was greater than the risk screening value, then the concentration "failed the
screen." Pollutants of interest are those for which the individual pollutant's total failed screens
contribute to the top 95 percent of the site's total failed screens. In addition, if any of the NATTS
MQO Core Analytes measured by the monitoring site did not meet the pollutant of interest
criteria based on the preliminary risk screening, that pollutant was added to the list of site-
specific pollutants of interest. A more in-depth description of the risk screening process is
presented in Section 3.2.
Table 13-4 presents GLKY's pollutants of interest. The pollutants that failed at least one
screen and contributed to 95 percent of the total failed screens for the monitoring site are shaded.
NATTS MQO Core Analytes are bolded. Thus, pollutants of interest are shaded and/or bolded.
GLKY sampled for hexavalent chromium, PAH, and VOC.
Table 13-4. Risk Screening Results for the Kentucky Monitoring Site
Pollutant
Screening
Value
(Hg/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Grayson, Kentucky - GLKY
Benzene
Carbon Tetrachloride
1,3-Butadiene
Naphthalene
£>-Dichlorobenzene
1 ,2-Dichloroethane
0.13
0.17
0.03
0.029
0.091
0.038
Total
35
35
19
12
1
1
103
35
35
25
60
7
1
163
100.00
100.00
76.00
20.00
14.29
100.00
63.19
33.98
33.98
18.45
11.65
0.97
0.97
33.98
67.96
86.41
98.06
99.03
100.00
Observations from Table 13-4 include the following:
• GLKY sampled hexavalent chromium and PAH throughout 2010, but did not
begin sampling VOC through the NMP until June. Even though VOC samples
were only collected half the year in 2010, VOC make up the majority of failed
screens in Table 13-4.
• Six pollutants failed screens for GLKY, including four NATTS MQO Core
Analytes.
13-12
-------
• Four pollutants were initially identified as pollutants of interest via the risk
screening process. All four of these pollutants are NATTS MQO Core Analytes.
Hexavalent chromium, benzo(a)pyrene, chloroform, and tetrachloroethylene were
added to GLKY's pollutants of interest because they are NATTS MQO Core
Analytes, even though they did not fail any screens. These pollutants are not
shown in Table 13-4. Trichloroethylene and vinyl chloride are also NATTS MQO
Core Analytes, but were not detected at this site and were therefore not added to
the list of pollutants of interest.
• Benzene, carbon tetrachloride, and 1,2-dichloroethane each failed 100 percent of
screens at GLKY. While benzene and carbon tetrachloride were detected in every
valid VOC sample collected at GLKY, 1,2-dichloroethane was detected only
once.
13.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Kentucky monitoring site. Concentration averages are provided for the pollutants of
interest for GLKY, where applicable. Concentration averages for select pollutants are also
presented graphically for the site, where applicable, to illustrate how the site's concentrations
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at the
site, where applicable. Additional site-specific statistical summaries are provided in Appendices
J, M, and O.
13.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for the Kentucky site, as described in Section 3.1. The quarterly average of a particular pollutant
is simply the average concentration of the preprocessed daily measurements over a given
calendar quarter. Quarterly average concentrations include the substitution of zeros for all non-
detects. A site must have a minimum of 75 percent valid samples of the total number of samples
possible within a given quarter for a quarterly average to be calculated. An annual average
includes all measured detections and substituted zeros for non-detects for the entire year of
sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the Kentucky
monitoring site are presented in Table 13-5, where applicable. Note that concentrations of the
PAH and hexavalent chromium for GLKY are presented in ng/m3 for ease of viewing. Also note
13-13
-------
that if a pollutant was not detected in a given calendar quarter, the quarterly average simply
reflects "0" because only zeros substituted for non-detects were factored into the quarterly
average concentration.
Table 13-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Kentucky Monitoring Site
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(jig/m3)
2nd
Quarter
Average
(jig/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Annual
Average
(jig/m3)
Grayson, Kentucky - GLKY
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
Tetrachloroethylene
Benzo(a)pyrenea
Hexavalent Chromium3
Naphthalene3
35/35
25/35
35/35
28/35
15/35
30/60
34/61
60/60
NA
NA
NA
NA
NA
0.13
±0.08
0.01
±0.01
28.61
±9.44
NA
NA
NA
NA
NA
0.03
±0.03
0.01
±0.01
17.00
±5.35
0.69
±0.27
0.02
±0.01
0.71
±0.07
0.10
±0.02
0.02
±0.02
<0.01
±<0.01
0.02
±0.01
17.84
±4.57
0.66
±0.15
0.04
±0.01
0.57
±0.06
0.05
±0.02
0.03
±0.02
0.06
±0.03
0.01
±0.01
27.39
±9.49
NA
NA
NA
NA
NA
0.06
±0.02
0.01
±0.01
22.71
±3.80
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of
viewing.
Observations for GLKY from Table 13-5 include the following:
• Annual average concentrations could not be calculated for the VOC because sampling
did not begin until June 2010. However, Appendix J provides the pollutant-specific
average concentration for all valid VOC samples collected over the entire sample
period.
• The annual average concentration of naphthalene is significantly higher than the
annual average concentrations of hexavalent chromium and benzo(a)pyrene.
• Concentrations of benzo(a)pyrene were highest during the first and fourth quarters of
2010. These quarterly averages also have relatively high confidence intervals
associated with them (as does the second quarter average). The three concentrations
greater than 0.25 ng/m3 were all measured in January and February 2010. Of the 30
measured detections of benzo(a)pyrene, 14 were measured during the first quarter, six
were measured during the second quarter, one was measured during the third quarter,
and nine were measured during the fourth quarter.
13-14
-------
• Similar to benzo(a)pyrene, naphthalene concentrations tended to be lower during the
warmer months and higher during the colder months. However, the confidence
intervals indicate that the difference is not statistically significant.
• The annual average concentrations of naphthalene and hexavalent chromium are
among the lowest compared to NMP sites sampling these pollutants. The annual
average concentration of benzo(a)pyrene for GLKY is in the middle of the range
compared to other sites sampling PAH.
13.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for benzo(a)pyrene,
hexavalent chromium, and naphthalene were created for GLKY. Figures 13-6 through 13-8
overlay the site's minimum, annual average, and maximum concentrations onto the program-
level minimum, first quartile, average, median, third quartile, and maximum concentrations, as
described in Section 3.5.3.
Figure 13-6. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
o
i Program Max Concentration = 42.7 ng/m3
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
13-15
-------
Figure 13-7. Program vs. Site-Specific Average Hexavalent Chromium Concentration
Program Max Concentration = 3.51 ng/m3
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o —
Figure 13-8. Program vs. Site-Specific Average Naphthalene Concentration
600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o —
Observations from Figures 13-6 through 13-8 include the following:
• Figure 13-6 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
GLKY is less than the program-level average concentration as well as the
program-level median concentration. Several non-detects of benzo(a)pyrene were
measured at GLKY.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 13-7 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 13-7 shows the annual average concentration of hexavalent chromium for
13-16
-------
GLKY is well below the program-level average and median concentrations. The
maximum hexavalent chromium concentration measured at GLKY is just greater
than the program-level average concentration. Several non-detects of hexavalent
chromium were measured at GLKY.
• Figure 13-8 shows that the annual naphthalene average for GLKY is well below
the program-level average concentration. The maximum naphthalene
concentration measured at GLKY is also less than the program-level average
concentration, but just greater than the median concentration for the program.
There were no non-detects of naphthalene at GLKY.
• Recall that annual averages for GLKY could not be calculated for the VOC, as
discussed in Section 13.4.1.
13.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. Sampling at GLKY under the NMP began in 2008; therefore, a trends analysis was
not conducted for this site.
13.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the
Kentucky monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
13.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Kentucky monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
were compared to the acute MRL; the quarterly averages were compared to the intermediate
MRL; and the annual averages were compared to the chronic MRL.
13-17
-------
None of the measured detections or time-period average concentrations of the pollutants
of interest for the Kentucky monitoring site were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
the Kentucky monitoring site.
13.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Kentucky monitoring site and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations for GLKY are presented in Table 13-6, where applicable.
Table 13-6. Cancer and Noncancer Surrogate Risk Approximations for the Kentucky
Monitoring Site
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections vs.
# of Samples
Annual
Average
(ng/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Grayson, Kentucky - GLKY
Benzene
Benzo(a)pyrene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
Hexavalent Chromium
Naphthalene
Tetrachloroethylene
0.000007
8
0.00176
0.00003
0.000006
0.012
0.000034
0.000000
26
0.03
0.002
0.1
0.098
0.0001
0.003
0.04
35/35
30/60
25/35
35/35
28/35
34/61
60/60
15/35
NA
0.06
±0.02
NA
NA
NA
0.01
±<0.01
22.71
±3.80
NA
NA
0.10
NA
NA
NA
0.12
0.77
NA
NA
NA
NA
NA
<0.01
0.01
NA
NA = Not available due to the criteria for calculating an annual average.
- = a Cancer URE or Noncancer RfC is not available.
Observations for GLKY from Table 13-6 include the following:
• The cancer risk approximations for the pollutants of interest for GLKY are all less
than 1.0 in-a-million (ranging from 0.10 in-a-million for benzo(a)pyrene to
0.77 in-a-million for naphthalene).
13-18
-------
• The noncancer risk approximations for naphthalene and hexavalent chromium are
well below an HQ of 1.0 (0.01 or less). A noncancer risk approximation for
benzo(a)pyrene could not be calculated because there is not a noncancer RfC for this
pollutant.
• The cancer and noncancer risk approximations for hexavalent chromium and
naphthalene for GLKY are among the lowest calculated for these pollutants of
interest across the NMP.
• Annual averages, and therefore cancer and noncancer risk approximations, could not
be calculated for the VOC pollutants of interest.
13.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 13-7 and 13-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 13-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 13-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 13-7 and 13-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. Further, the cancer and noncancer risk approximations based on each site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 13.3, GLKY sampled for hexavalent chromium, PAH, and VOC. In addition, the cancer
and noncancer surrogate risk approximations are limited to those pollutants with enough data to
meet the criteria for annual averages to be calculated. As mentioned in Section 13.5.2, because
annual averages could not be calculated for the VOC, cancer and noncancer surrogate risk
approximations were also not calculated. A more in-depth discussion of this analysis is provided
in Section 3.5.5.3.
13-19
-------
Table 13-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Kentucky Monitoring Site
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Pollutant
Grayson, Kentucky (Carter County) - GLKY
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
POM, Group 2b
Dichloromethane
POM, Group 6
POM, Group la
26.56
15.19
11.80
9.53
3.03
2.06
0.40
0.25
0.04
0.02
Benzene
Formaldehyde
1,3 -Butadiene
Naphthalene
POM, Group 2b
Hexavalent Chromium, PM
Ethylbenzene
POM, Group 3
Acetaldehyde
POM, Group 5a
2.07E-04
1.97E-04
9.09E-05
7.00E-05
3.56E-05
3.33E-05
2.95E-05
2.47E-05
2.10E-05
1.78E-05
Naphthalene
Hexavalent Chromium
Benzo(a)pyrene
Cancer Risk
Approximation
(in-a-million)
0.77
0.12
0.10
to
o
-------
Table 13-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Kentucky Monitoring Site
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Grayson, Kentucky (Carter County) - GLKY
Toluene
Xylenes
Benzene
Methanol
Hexane
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Styrene
59.99
44.41
26.56
15.98
15.43
15.19
11.80
9.53
3.03
2.50
Acrolein
Formaldehyde
1,3 -Butadiene
Cyanide Compounds, gas
Acetaldehyde
Benzene
Naphthalene
Xylenes
Arsenic, PM
Propionaldehyde
52,572.66
1,549.95
1,514.80
1,335.64
1,058.92
885.42
686.00
444.15
138.10
111.83
Naphthalene 0.01
Hexavalent Chromium <0 . 0 1
to
-------
Observations from Table 13-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in Carter County. The emissions for this county are low compared to
other counties with NMP sites.
• Benzene, formaldehyde, and 1,3-butadiene are the pollutants with the highest
toxi city-weighted emissions (of the pollutants with cancer UREs) for Carter County.
• Seven of the highest emitted pollutants also have the highest toxi city-weighted
emissions for Carter County. Note that benzene and formaldehyde top both lists.
• Naphthalene appears on both emissions-based lists for Carter County and has the
highest cancer risk approximation for this county. Hexavalent chromium ranks sixth
for toxi city-weighted emissions but is not one of the highest emitted pollutants in
Carter County.
• Three POM Groups appear among the highest emitted pollutants (POM, Groups la,
2b, and 6). Three POM Groups also appear among the pollutants with the highest
toxicity-weighted emissions (POM, Groups 2b, 3, and 5a). Benzo(a)pyrene, a
pollutant of interest for GLKY, is part of POM Group 5a. Several pollutants
measured with Method TO-13 are part of POM, Group 2b, which appears on both
emissions-based lists.
Observations from Table 13-8 include the following:
• Toluene, xylenes, and benzene are the highest emitted pollutants with noncancer
RfCs in Carter County.
• The pollutant with the highest toxicity-weighted emissions (of the pollutants with
noncancer RfCs) is acrolein. Acrolein did not appear on Carter County's list of
highest emitted pollutants.
• Five of the highest emitted pollutants also have the highest toxicity-weighted
emissions for Carter County.
• While naphthalene does not appear among the pollutants with the highest emissions
(of the pollutants with noncancer RfCs), it ranks seventh on the list of highest
toxicity-weighted emissions for Carter County. Hexavalent chromium does not
appear on either emissions-based list.
13-22
-------
13.6 Summary of the 2010 Monitoring Data for GLKY
Results from several of the data treatments described in this section include the
following:
»«» Six pollutants, five VOC and one PAH, failed screens for GLKY, including four
NATTSMQO Core Analytes.
»«» Naphthalene had the highest annual average concentration among the pollutants of
interest for GLKY.
»«» Because VOC sampling did not begin untilJune 2010, annual average concentrations
could not be calculated for these pollutants.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest, where they could be calculated,
were greater than their associatedMRL noncancer health risk benchmarks.
13-23
-------
14.0 Site in Massachusetts
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in Massachusetts, and integrates these concentrations
with emissions, meteorological, and risk information. Data generated by sources other than ERG
are not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
14.1 Site Characterization
This section characterizes the BOMA monitoring site by providing geographical and
physical information about the location of the site and the surrounding area. This information is
provided to give the reader insight regarding factors that may influence the air quality near the
site and assist in the interpretation of the ambient monitoring measurements.
The BOMA monitoring site is located in Boston. Figure 14-1 is a composite satellite
image retrieved from ArcGIS Explorer showing the monitoring site in its urban location.
Figure 14-2 identifies point source emissions locations by source category, as reported in the
2008 NEI for point sources. Note that only sources within 10 miles of the site are included in the
facility counts provided in Figure 14-2. Thus, sources outside the 10-mile radius have been
grayed out, but are visible on the map to show emissions sources outside the 10-mile boundary.
A 10-mile boundary was chosen to give the reader an indication of which emissions sources and
emissions source categories could potentially have a direct effect on the air quality at the
monitoring site. Further, this boundary provides both the proximity of emissions sources to the
monitoring site as well as the quantity of such sources within a given distance of the site.
Table 14-1 describes the area surrounding the monitoring site by providing supplemental
geographical information such as land use, location setting, and locational coordinates.
14-1
-------
Figure 14-1. Boston, Massachusetts (BOMA) Monitoring Site
-------
Figure 14-2. NEI Point Sources Located Within 10 Miles of BOMA
?nd
SOMANATTSsite
10 mile radius
/V 7riQ'0"W
Note: Due to facility density and collocation, the total facilities
displayed may not represent a II facilities within the area of interest.
County boundary
e
B
C
•
e
e
I
E
I
*
Source Category Group (No. of Facilities)
•$1 Aerospace/Aircraft Manufacturing (1)
v Air-conditioning/Refrigeration (3)
-t< Aircraft Operations (18)
0 Auto Body Shop/Painters (1)
A Automobile/Truck Manufacturing (1)
Bakery (4)
Bulk Terminals/Bulk Plants (7)
Chemical Manufacturing (1)
Concrete Batch Plant (1)
Dry Cleaning (2)
Electrical Equipment (3)
Electricity Generation via Combustion (8) ^
Electroplating. Plating, Polishing, Anodizing, and Coloring (4) p
© Fabricated Metal Products (5) A^
F Food Processing/Agriculture (1) >
n Furniture Plant (1)
f Gasoline/Diesel Service Station (2)
ffl Hospital (4)
Hot Mix Asphalt Plant (2)
Institutional - school (39)
Iron and Steel Foundry (1)
Laboratory (2)
• Landfill (1)
L Large Appliance Manufacturing (1)
V Mineral Products (1)
? Miscellaneous Commercial/Industrial (38)
M Miscellaneous Manufacturing (6)
H Municipal Waste Combustor(l)
— Pharmaceutical Manufacturing (1)
Printing/Publishing (1)
Pulp and Paper PlanlWood Products (1)
Rubber and Miscellaneous Plastics Products (2)
ii. Ship Building and Repairing (2)
> Solid Waste Disposal - Commercial/Institutional (2)
S Surface Coating (3)
Tt Telecommunications (4)
1 Wastewater Treatment {2)
14-3
-------
Table 14-1. Geographical Information for the Massachusetts Monitoring Site
Site
Code
BOMA
AQS Code
25-025-0042
Location
Boston
County
Suffolk
Micro- or
Metropolitan
Statistical Area
Boston-
Cambridge-
Quincy, MA-NH
MSA (Boston Div)
Latitude
and
Longitude
42.32944,
-71.0825
Land Use
Commercial
Location
Setting
Urban/City
Center
Additional Ambient Monitoring Information1
Lead (TSP), CO, VOC, SO2, NO, NO2, NOX,
PAMS/NMOC, Carbonyl compounds, O3,
Meteorological parameters, PM10, Black carbon,
PM2 5, PM2 5 Speciation.
BOLD ITALICS = EPA-designated NATTS Site.
-------
The BOMA monitoring site is located at Dudley Square in Roxbury, a southwest
neighborhood of Boston and is the Roxbury NATTS site. The surrounding area is commercial as
well as residential, as shown in Figure 14-1. The monitoring site is approximately 1.25 miles
south of 1-90 and 1 mile west of 1-93. The original purpose for the location of this site was to
measure population exposure to a city bus terminal located across the street from the monitoring
site. In recent years, the buses servicing the area were converted to compressed natural gas
(CNG). As Figure 14-2 shows, BOMA is located near a large number of point sources, with a
high density of sources located within a few miles to the west, northwest, and north of the site.
The source categories with the highest number of emissions sources surrounding BOMA include
institutional facilities (schools), aircraft operations, which includes airports as well as small
runways, heliports, or landing pads, and electricity generating units (via combustion).
Table 14-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the
Massachusetts monitoring site. Table 14-2 also includes a vehicle registration-to-county
population ratio (vehicles-per-person). In addition, the population within 10 miles of the site is
presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-level
vehicle registration-to-population ratio to the 10-mile population surrounding the monitoring
site. Table 14-2 also contains annual average daily traffic information. County-level VMT was
not readily available; thus, daily VMT for Suffolk County is not shown in Table 14-2.
Table 14-2. Population, Motor Vehicle, and Traffic Information for the Massachusetts
Monitoring Site
Site
BOMA
Estimated
County
Population1
723,323
County-level
Vehicle
Registration2
501,587
Vehicles per
Person
(Registration:
Population)
0.69
Population
within 10
miles3
1,670,959
Estimated
10-mile
Vehicle
Ownership
1,158,723
Annual
Average
Daily
Traffic4
31,400
County-level
Daily VMT5
NA
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the Massachusetts Registry of Motor Vehicles (MA
RMV, 2011)
310-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2007 data from the Massachusetts DOT (MA DOT, 2007)
5 County-level VMT was not available for this site
BOLD ITALICS = EPA-designated NATTS Site.
14-5
-------
Observations from Table 14-2 include the following:
• The Suffolk County population is in the middle of the range compared to other
counties with NMP sites, while BOMA's 10-mile population is among the higher
10-mile populations.
• Similar to the populations, the Suffolk County vehicle registration is in the middle of
the range compared to other counties with NMP sites, while its 10-mile estimated
ownership is among the higher estimates.
• The vehicle-per-person ratio is among the bottom-third of ratios when compared to
other NMP sites.
• The traffic volume experienced near BOMA ranks in the middle of the range
compared to other NMP sites. The traffic estimate used came from Melnea Cass
Boulevard between Washington Street and Harrison Avenue.
14.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in Massachusetts on sample days, as well as over the course of the year.
14.2.1 Climate Summary
Boston's New England location ensures that the city experiences a fairly active weather
pattern. Storm systems frequently track across the region, bringing ample precipitation to the
area. The proximity to the Atlantic Ocean helps moderate temperatures, both in the summer and
the winter, while at the same time allowing winds to gust higher than they would farther inland.
Winds generally flow from the northwest in the winter and southwest in the summer. Coastal
storm systems called "Nor'easters," strong low pressure systems that produce heavy rain or snow
and winds up to hurricane strength along the Mid-Atlantic and northeast coastal states, often
produce the heaviest snowfalls for the area (Bair, 1992 and NOAA, 2012a).
14.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest this site were retrieved
for 2010 (NCDC, 2010). The closest weather station to BOMA is located at Logan International
Airport (WBAN 14739). Additional information about the Logan Airport weather station, such
as the distance between the site and the weather station, is provided in Table 14-3. These data
were used to determine how meteorological conditions on sample days vary from normal
conditions throughout the year.
14-6
-------
Table 14-3. Average Meteorological Conditions near the Massachusetts Monitoring Site
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Boston, Massachusetts - BOMA
Logan
International
Airport
14739
(42.36, -71.01)
4.05
miles
42°
(NE)
Sample
Day
2010
59.8
±4.8
60.6
±1.9
53.0
±4.4
53.6
±1.8
40.0
±4.6
40.0
±1.9
47.0
±4.0
47.3
±1.6
64.9
±4.2
63.6
±1.7
1012.1
±2.0
1012.2
±0.8
9.9
±0.8
9.8
±0.4
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
Table 14-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 14-3 is the 95
percent confidence interval for each parameter. As shown in Table 14-3, average meteorological
conditions on sample days were representative of average weather conditions throughout the
year.
14.2.3 Back Trajectory Analysis
Figure 14-3 is the composite back trajectory map for days on which samples were
collected at the BOMA monitoring site in 2010. Included in Figure 14-3 are four back
trajectories per sample day. Figure 14-4 is the corresponding cluster analysis for 2010. An in-
depth description of these maps and how they were generated is presented in Section 3.5.2.1. For
the composite map, each line represents the 24-hour trajectory along which a parcel of air
traveled toward the monitoring site on a given sample day and time. For the cluster analysis,
each line corresponds to a back trajectory representative of a given cluster of trajectories. For
both maps, each concentric circle around the site in Figures 14-3 and 14-4 represents 100 miles.
Figure 14-3. 2010 Composite Back Trajectory Map for BOMA
14-8
-------
Figure 14-4. Back Trajectory Cluster Map for BOMA
/ » \ ll ' \
rm ' \ ' \ \ '
, ^ \ ' \ '. • '
Observations from Figures 14-3 and 14-4 include the following:
• The composite back trajectory map shows that back trajectories originated from a
variety of directions at BOMA. The longest trajectories originated from the south-
southeast and are associated with a strong low pressure system that moved through
the region January 25-26, 2010.
• The 24-hour air shed domain for BOMA is larger in size compared to other NMP
monitoring sites. The farthest away a trajectory originated was nearly 800 miles, off
the North Carolina coast and over the Atlantic Ocean. However, the average
trajectory length was 310 miles. Most trajectories (85 percent) originated within 500
miles of the monitoring site.
• Nearly half of back trajectories originated to the west, northwest, and north of
BOMA, as shown by the 32 percent and 14 percent trajectories on the cluster
analysis. Another 32 percent of trajectories originated within approximately 250
miles of BOMA, as represented by the short cluster approximately 50 miles long.
This cluster represents several trajectories originating from a variety of directions but
within 200 or so miles of the site. It is important to recall that the HYSPLIT model
includes both distance and direction when determining clusters. Another 13 percent
originated from south of BOMA, and includes the longest trajectories in Figure 14-3.
Finally, 10 percent of trajectories originated to the northeast over Quebec, New
Brunswick, or Nova Scotia, Canada, or over the Atlantic Ocean.
14-9
-------
14.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at Logan International Airport near
BOMA were uploaded into a wind rose software program to produce customized wind roses, as
described in Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals"
positioned around a 16-point compass, and uses different colors to represent wind speeds.
Figure 14-5 presents three different wind roses for the BOMA monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at this location.
Observations from Figure 14-5 for BOMA include the following:
• The Logan International Airport weather station is located approximately 4 miles
northeast of BOMA. Note that the airport sits on a peninsula on Boston Harbor with
downtown Boston to the west, Chelsea to the north, and Winthrop to the east, while
the BOMA monitoring site is located west of South Boston, farther inland.
• The historical wind rose shows that calm winds (< 2 knots) account for only three
percent of the wind observations. Winds with a westerly component (south-southwest
to north-northwest) make up the bulk (nearly 60 percent) of winds greater than
2 knots.
• The wind patterns shown on the 2010 wind rose resemble the historical wind patterns,
although there was a higher percentage of westerly to northwesterly winds, indicating
that wind conditions during 2010 were typical of conditions normally experienced.
• The sample day wind patterns resemble the full-year wind patterns, with an even
higher percentage of west-northwesterly and northwesterly winds, indicating that
conditions on sample days were representative of those experienced over the entire
year.
14-10
-------
Figure 14-5. Wind Roses for the Logan International Airport Weather Station near BOMA
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between BOMA and NWS Station
- N
-..r_ s/
j
14-11
-------
14.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Massachusetts monitoring
site in order to allow analysts and readers to focus on a subset of pollutants through the context
of risk. Each pollutant's preprocessed daily measurement was compared to its associated risk
screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by the monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 14-4 presents BOMA's pollutants of interest. The pollutants that failed at least one
screen and contributed to 95 percent of the total failed screens for the monitoring site are shaded.
NATTS MQO Core Analytes are bolded. Thus, pollutants of interest are shaded and/or bolded.
BOMA sampled for metals (PMio), PAH, and hexavalent chromium.
Table 14-4. Risk Screening Results for the Massachusetts Monitoring Site
Pollutant
Screening
Value
(Hg/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Boston, Massachusetts - BOMA
Naphthalene
Arsenic (PM10)
Manganese (PM10)
Nickel (PM10)
Hexavalent Chromium
0.029
0.00023
0.005
0.0021
0.000083
Total
57
43
6
5
3
114
60
61
61
61
43
286
95.00
70.49
9.84
8.20
6.98
39.86
50.00
37.72
5.26
4.39
2.63
50.00
87.72
92.98
97.37
100.00
Observations from Table 14-4 include the following:
• Five pollutants failed at least one screen for BOMA; all five are NATTS MQO Core
Analytes.
• Nearly 40 percent of the measured detections (of the pollutants that failed at least one
screen) failed screens for BOMA. Naphthalene accounted for half of the total failed
screens for BOMA.
• Four of the five pollutants failing screens were initially identified as pollutants of
interest for BOMA based on the risk screening process. Hexavalent chromium was
14-12
-------
added to BOMA's pollutants of interest because it is a NATTS MQO Core Analyte,
even though they did not contribute to 95 percent of the failed screens.
Benzo(a)pyrene, beryllium, cadmium, and lead were also added to BOMA's
pollutants of interest because they are NATTS MQO Core Analytes, even though
they did not fail any screens. These four pollutants are not shown in Table 14-4.
14.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Massachusetts monitoring site. Concentration averages are provided for the pollutants of
interest for BOMA, where applicable. Concentration averages for select pollutants are also
presented graphically for the site, where applicable, to illustrate how the site's concentrations
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at the
site, where applicable. Additional site-specific statistical summaries are provided in Appendices
M through O.
14.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for BOMA, as described in Section 3.1. The quarterly average of a particular pollutant is simply
the average concentration of the preprocessed daily measurements over a given calendar quarter.
Quarterly average concentrations include the substitution of zeros for all non-detects. A site must
have a minimum of 75 percent valid samples of the total number of samples possible within a
given quarter for a quarterly average to be calculated. An annual average includes all measured
detections and substituted zeros for non-detects for the entire year of sampling. Annual averages
were calculated for pollutants where three valid quarterly averages could be calculated and
where method completeness was greater than or equal to 85 percent, as presented in Section 2.4.
Quarterly and annual average concentrations for BOMA are presented in Table 14-5, where
applicable. Note that if a pollutant was not detected in a given calendar quarter, the quarterly
average simply reflects "0" because only zeros substituted for non-detects were factored into the
quarterly average concentration.
14-13
-------
Table 14-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Massachusetts Monitoring Site
Pollutant
#of
Measured
Detections vs.
# of Samples
1st
Quarter
Average
(ng/m3)
2nd
Quarter
Average
(ng/m3)
3rd
Quarter
Average
(ng/m3)
4th
Quarter
Average
(ng/m3)
Annual
Average
(ng/m3)
Boston, Massachusetts - BOMA
Arsenic (PM10)
Benzo(a)pyrene
Beryllium (PM10)
Cadmium (PM10)
Hexavalent Chromium
Lead (PM10)
Manganese (PM10)
Naphthalene
Nickel (PM10)
61/61
52/60
55/61
61/61
43/61
61/61
61/61
60/60
61/61
0.27
±0.06
0.17
±0.05
O.01
±<0.01
0.24
±0.04
0.01
±0.01
2.57
±0.75
2.52
±0.61
57.43
± 13.04
1.46
±0.39
0.48
±0.14
0.07
±0.02
O.01
±<0.01
0.20
±0.03
0.02
±0.01
3.14
±0.98
4.18
±0.89
67.38
±11.04
1.31
±0.21
0.40
±0.09
0.06
±0.03
O.01
±<0.01
0.17
±0.03
0.04
±0.02
2.47
±0.61
3.69
±1.37
82.35
± 14.24
1.04
±0.21
0.28
±0.09
0.10
±0.04
O.01
±<0.01
0.15
±0.03
0.02
±0.01
1.96
±0.39
2.31
±0.48
64.94
±19.91
1.20
±0.33
0.36
±0.05
0.10
±0.02
O.01
±<0.01
0.19
±0.02
0.02
±0.01
2.53
±0.35
3.18
±0.48
68.31
±7.30
1.25
±0.14
Observations for BOMA from Table 14-5 include the following:
• Naphthalene is the pollutant with the highest annual average concentration by mass
(68.31 ± 7.30 ng/m3). The annual average concentrations for the remaining pollutants
of interest are at least an order of magnitude lower.
• Benzo(a)pyrene concentrations appear to be highest during the colder months of the
year, as indicated by the first and fourth quarter average concentrations. Of the 12
concentrations greater than 0.15 ng/m3, only one was measured outside the first or
fourth quarter of 2010 and the six highest concentrations measured at BOMA were
from samples collected in January, February, or December.
• The second and third quarter manganese average concentrations are higher than the
other quarterly averages and the third quarter has a relatively large confidence
interval associated with it. Of the six manganese concentrations greater than 5 ng/m3,
four were measured during the second quarter of 2010 and two were measured during
the third quarter of 2010, including the highest concentration of manganese measured
on July 7, 2010 (12.3 ng/m3).
• The hexavalent chromium average for the third quarter of 2010 is twice the average
concentration of the other quarterly averages. A review of the data shows that the two
highest concentrations of hexavalent chromium were measured during this quarter
(0.138 ng/m3 on July 7, 2010 and 0.103 ng/m3 on August 24, 2010). These two
concentrations are the only concentrations greater than 0.10 ng/m3 measured at
14-14
-------
BOMA. Note that the highest concentration of hexavalent chromium was measured
on the same day as the highest concentration of manganese.
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for BOMA from
those tables include the following:
• BOMA's annual average concentration of benzo(a)pyrene ranks sixth highest among
sites sampling PAH.
• BOMA's annual average concentrations of cadmium and nickel rank second highest
among other sites sampling PMio metals.
• BOMA's annual average concentration of hexavalent chromium ranks tenth highest
among sites sampling this pollutant.
14.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for arsenic, benzo(a)pyrene,
hexavalent chromium, manganese, and naphthalene were created for BOMA. Figures 14-6
through 14-10 overlay the site's minimum, annual average, and maximum concentrations onto
the program-level minimum, first quartile, average, median, third quartile, and maximum
concentrations, as described in Section 3.5.3.
Figure 14-6. Program vs. Site-Specific Average Arsenic (PMio) Concentration
,
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
• D D D
Site: Site Average Site Minimum/Maximum
o —
14-15
-------
Figure 14-7. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
Program Max Concentration = 42.7 ng/m3
1 1
0.8 1 1.2
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 14-8. Program vs. Site-Specific Average Hexavalent Chromium Concentration
i
r
! Progr
1
m Max Concentration = 3.51 ng/m3 j
1
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 14-9. Program vs. Site-Specific Average Manganese (PMio) Concentration
• I '
1
D 20 40 60 80 100 120 140 160 180 20
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile
• • n
Average
i i i i
Site: Site Average Site Minimum/Maximum
o —
14-16
-------
Figure 14-10. Program vs. Site-Specific Average Naphthalene Concentration
600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rd Quartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o —
Observations from Figures 14-6 through 14-10 include the following:
• Figure 14-6 shows that BOMA's annual average arsenic (PMio) concentration is
below both the program-level average and median concentrations for arsenic
(PMio). There were no non-detects of arsenic measured at BOMA.
• Figure 14-7 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
BOMA is below the program-level average concentration of benzo(a)pyrene and
roughly equivalent to the program-level third quartile. Figure 14-7 also shows that
the maximum concentration measured at BOMA is well below the maximum
concentration measured across the program. Several non-detects of
benzo(a)pyrene were measured at BOMA.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 14-8 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for the observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 14-8 shows the annual average concentration of hexavalent chromium for
BOMA is below the program-level average concentration. The maximum
concentration measured at BOMA is well below the maximum concentration
measured at the program level. Several non-detects of hexavalent chromium were
measured at BOMA.
• Figure 14-9 shows that BOMA's annual average manganese (PMio) concentration
is below the program-level average concentration, as well as the median
concentration, for manganese (PMio). There were no non-detects of manganese at
BOMA.
14-17
-------
• Similar to BOMA's other pollutants of interest, Figure 14-10 shows that the
annual naphthalene average for BOMA is less than the program-level average
concentration. There were no non-detects of naphthalene measured at BOMA.
14.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. BOMA has been sampling metals since 2003 and hexavalent chromium since
2005. Thus, Figures 14-11 through 14-13 present the 3-year rolling statistical metrics for arsenic,
hexavalent chromium, and manganese for BOMA, respectively. The statistical metrics presented
for calculating trends include the substitution of zeros for non-detects.
Figure 14-11. Three-Year Rolling Statistical Metrics for Arsenic (PMio) Concentrations
Measured at BOMA
m
1
|a
01
3
—
£ £
2004-2006
1
2005-2007
• 5th Percentile — Minimum
•*•
2006-2008
Three-Year Period
- Median
-
Maximum
^g^
2007-2009
• 95tli Percentile
—
ta
± '
2008-2010
••*•• Average
Samples were not collected between April 3 and May 21 and September 24 through November 6 in 2004.
14-18
-------
Figure 14-12. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at BOMA
I
!
•
••
2005-2007
• 5th Percentlle
— Minimum
^mm
2006-2008
Three-Year Period
*
2007-2009
— Median — Maximum
• 95thPercentile
<
•
^^
2008-2010
•••*•• Average
Figure 14-13. Three-Year Rolling Statistical Metrics for Manganese (PMi0) Concentrations
Measured at BOMA
1
° r
4
4
M
1
"II
•
— m*m ^ s^
I i -I- -r-
2004-2006 l 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Percen tile — Minimum — Median — Maximum • 95th Percentile ... + ,. Average
1 Samples were not collected between April 3 and May 21 and September 24 through November 6 in 2004.
14-19
-------
Observations from Figure 14-11 for arsenic measurements at BOMA include the
following:
• While PMio metals sampling began in 2003, data from that year were excluded from
this analysis because sampling did not begin until October. In addition, samples were
not collected in parts of April, May, September, and October 2004, which is denoted
in Figure 14-11.
• The maximum arsenic concentration shown was measured on July 5, 2008. The next
highest concentration measured is approximately half as high and was measured on
July 4, 2006 (and is shown as the maximum concentration for the first two 3-year
periods).
• The rolling average concentrations exhibit very little change over the years of
sampling, which is also true for most of the other statistical parameters.
• The minimum concentration measured for each 3-year period is greater than zero,
indicating that there were no non-detects of arsenic measured at BOMA since the
onset of sampling.
Observations from Figure 14-12 for hexavalent chromium measurements at BOMA
include the following:
• The maximum hexavalent chromium concentration was measured in 2008
(0.525 ng/m3). Less than 10 percent of hexavalent chromium concentrations measured
were greater than 0.1 ng/m3; of these, at least two have been measured in each year
since the onset of sampling.
• While the rolling average concentration has been decreasing slightly since the onset
of sampling, this decrease is not statistically significant. The medians and 95th
percentiles also show slight decreases.
• The minimum and 5th percentile are both zero for each 3-year period of sampling,
indicating the presence of non-detects.
Observations from Figure 14-13 for manganese measurements at BOMA include the
following:
• The maximum manganese concentration was measured in 2004. Of the six
manganese measurements greater than 10 ng/m3, two were measured in 2004, two in
2005, and one each in 2008 and 2010. Consequently, the second highest manganese
concentration was measured on July 7, 2010, as discussed in the previous section.
• The rolling average and median concentrations exhibit a steady decreasing trend over
the years of sampling. Other statistical measures, such as the median and
95th percentile, also show a downward trend.
14-20
-------
• The minimum concentration measured for each 3-year period is greater than zero,
indicating that there were no non-detects of manganese measured at BOMA since the
onset of sampling.
14.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the
BOMA monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
14.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Massachusetts monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
were compared to the acute MRL; the quarterly averages were compared to the intermediate
MRL; and the annual averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for BOMA were greater than their respective MRL noncancer health risk benchmarks.
This is also true for pollutants not identified as pollutants of interest for the Massachusetts
monitoring site.
14.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Massachusetts monitoring site and where annual
average concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 14-6, where applicable.
14-21
-------
Table 14-6. Cancer and Noncancer Surrogate Risk Approximations for the Massachusetts
Monitoring Site
Pollutant
Cancer
URE
(Hg/m3)-1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections vs.
# of Samples
Annual
Average
(ng/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Boston, Massachusetts - BOMA
Arsenic (PM10)
Benzo(a)pyrene
Bery Ilium (PM10)
Cadmium (PM10)
Hexavalent Chromium
Lead (PM10)
Manganese (PM10)
Naphthalene
Nickel (PM10)
0.0043
0.00176
0.0024
0.0018
0.012
0.000034
0.00048
0.000015
0.00002
0.00001
0.0001
0.00015
0.00005
0.003
0.00009
61/61
52/60
55/61
61/61
43/61
61/61
61/61
60/60
61/61
0.36
±0.05
0.10
±0.02
O.01
±<0.01
0.19
±0.02
0.02
±0.01
2.53
±0.35
3.18
±0.48
68.31
±7.30
1.25
±0.14
1.54
0.17
0.01
0.34
0.25
2.32
0.60
0.02
O.01
0.02
O.01
0.02
0.06
0.02
0.01
— = a Cancer URE or Noncancer RfC is not available.
Observations for BOMA from Table 14-6 include the following:
• Naphthalene has the highest annual average concentration for BOMA. Manganese,
lead, and nickel also have annual average concentrations greater than 1.0 ng/m3.
• Naphthalene and arsenic are the only pollutants of interest with cancer surrogate risk
approximations greater than 1.0 in-a-million (2.32 in-a-million and 1.54 in-a-million,
respectively).
• None of BOMA's pollutants of interest have noncancer risk approximations greater
than 1.0, indicating little risk of noncancer effects due to these pollutants.
14.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 14-7 and 14-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 14-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 14-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
14-22
-------
Table 14-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Massachusetts Monitoring Site
-^
to
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Boston, Massachusetts (Suffolk County) - BOMA
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
Methyl tert butyl ether
Nickel, PM
168.18
164.08
83.84
79.23
26.64
14.54
5.42
3.92
3.63
1.87
Formaldehyde
Benzene
Nickel, PM
1,3 -Butadiene
POM, Group 3
Naphthalene
Hexavalent Chromium, PM
POM, Group 2b
Arsenic, PM
Ethylbenzene
2.13E-03
1.31E-03
8.99E-04
7.99E-04
6.36E-04
4.94E-04
4.31E-04
3.45E-04
2.86E-04
2.10E-04
Naphthalene
Arsenic
Nickel
Cadmium
Hexavalent Chromium
Benzo(a)pyrene
Beryllium
2.32
1.54
0.60
0.34
0.25
0.17
0.01
-------
Table 14-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Massachusetts Monitoring Site
-^
to
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted
Emissions (County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Boston, Massachusetts (Suffolk County) - BOMA
Toluene
Xylenes
Benzene
Formaldehyde
Hexane
Ethylbenzene
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
Glycol ethers, gas
456.02
372.68
168.18
164.08
97.07
83.84
79.23
44.58
26.64
18.42
Acrolein
Nickel, PM
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Benzene
Naphthalene
Arsenic, PM
Xylenes
Cadmium, PM
313,612.93
20,816.71
16,743.22
13,322.42
8,803.49
5,606.08
4,847.85
4,440.73
3,726.77
2,204.94
Manganese
Arsenic
Naphthalene
Cadmium
Lead
Nickel
Hexavalent Chromium
Beryllium
0.06
0.02
0.02
0.02
0.02
0.01
<0.01
<0.01
-------
The pollutants listed in Tables 14-7 and 14-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer risk approximations based on each site's annual averages are
limited to those pollutants for which each respective site sampled. As discussed in Section 14.3,
BOMA sampled for PAH, PMio metals, and hexavalent chromium. In addition, the cancer and
noncancer risk approximations are limited to those pollutants with enough data to meet the
criteria for an annual average to be calculated. A more in-depth discussion of this analysis is
provided in Section 3.5.5.3.
Observations from Table 14-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in Suffolk County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
cancer UREs) are formaldehyde, benzene, and nickel.
• Seven of the highest emitted pollutants also have the highest toxi city-weighted
emissions.
• Naphthalene and arsenic are the pollutants with the highest cancer surrogate risk
approximations for BOMA. Naphthalene ranks sixth on the list of highest emitted
pollutants and sixth for toxicity-weighted emissions. Arsenic ranks ninth on the list of
highest toxicity-weighted emissions but is not among the highest emitted.
• Nickel also appears on both emissions-based lists and has the third highest cancer risk
approximation (albeit low) for BOMA. Hexavalent chromium ranks seventh on the
list of highest toxicity-weighted emissions but is not among the highest emitted.
• POM, Group 2b is among the 10 highest emitted "pollutants" in Suffolk County and
also ranks among the 10 highest for toxicity-weighted emissions. POM, Group 2b
includes several PAH sampled for at BOMA including acenaphthylene, fluoranthene,
and perylene. None of the PAH included in POM, Group 2b were identified as
pollutants of interest for BOMA. Benzo(a)pyrene is part of POM, Group 5a, which is
not listed on either emissions-based list.
Observations from Table 14-8 include the following:
• Toluene, xylenes, and benzene are the highest emitted pollutants with noncancer
RfCs in Suffolk County.
• The pollutants with the highest toxicity-weighted emissions (of the pollutants with
noncancer RfCs) are acrolein, nickel, and formaldehyde.
14-25
-------
• Five of the highest emitted pollutants also have the highest toxi city-weighted
emissions.
• While several metals are among the pollutants with the highest toxicity-weighted
emissions, no metals appear among the highest emitted pollutants. Nickel, which has
the second highest toxicity-weighted emissions for Suffolk County, has a negligible
noncancer risk approximation, as do the remaining pollutants of interest for BOMA.
14.6 Summary of the 2010 Monitoring Data for BOMA
Results from several of the data treatments described in this section include the
following:
»«» Five pollutants failed screens for BOMA, of which all are NA TTS MQO Core
Analytes.
»«» Naphthalene had the highest annual average concentration for 2010 among the
pollutants of interest for BOMA.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
14-26
-------
15.0 Site in Michigan
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in Michigan, and integrates these concentrations
with emissions, meteorological, and risk information. Data generated by sources other than ERG
are not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
15.1 Site Characterization
This section characterizes the monitoring site by providing geographical and physical
information about the location of the site and the surrounding area. This information is provided
to give the reader insight regarding factors that may influence the air quality near the site and
assist in the interpretation of the ambient monitoring measurements.
The DEMI monitoring site is located in the Detroit-Warren-Livonia, MI MSA.
Figure 15-1 is the composite satellite image retrieved from ArcGIS Explorer showing the
monitoring site in its urban location. Figure 15-2 identifies point source emissions locations by
source category, as reported in the 2008 NEI for point sources. Note that only sources within
10 miles of the site are included in the facility counts provided in Figure 15-2. Thus, sources
outside the 10-mile radius have been grayed out, but are visible on the map to show emissions
sources outside the 10-mile boundary. A 10-mile boundary was chosen to give the reader an
indication of which emissions sources and emissions source categories could potentially have a
direct effect on the air quality at the monitoring site. Further, this boundary provides both the
proximity of emissions sources to the monitoring site as well as the quantity of such sources
within a given distance of the site. Table 15-1 describes the area surrounding the monitoring site
by providing supplemental geographical information such as land use, location setting, and
locational coordinates.
15-1
-------
Figure 15-1. Dearborn, Michigan (DEMI) Monitoring Site
-------
Figure 15-2. NEI Point Sources Located Within 10 Miles of DEMI
Legend
Note: Due to facility density and collocation, the total facilities
displayed may nol represent all facilities within the area of interest.
DEMI NATTS site
10 mile radius
Source Category Group (No. of Facilities)
8 Air-conditioning/Refrigeration (5)
4f Aircraft Operations (9)
I Asphalt Processing/Roofing Manufacturing (2)
& Automobile/Truck Manufacturing (4)
B Bulk Terminals/Bulk Plants (8)
A Cement Kiln/Dryer (1)
C Chemical Manufacturing (2)
6 Electrical Equipment (1)
f Electricity Generation via Combustion (7)
E Electroplating. Plating. Polishing, Anodizing, and Coloring (6)
4 Engine Testing (1)
© Fabricated Metal Products (6)
**~ Flexible Polyurethane Foam Production (1)
F Food Processing/Agriculture (2)
-•- Gypsum Manufacturing (1)
G3 Hospital (2)
'i Hot Mix Asphalt Plant (2)
•*- Industrial Machinery and Equipment (1)
^ Institutional - school (7)
County boundary
I Iron and Steel Foundry (1)
v Laboratory {4)
>• Lime Manufacturing (1)
X Mine/Quarry (3)
T Mineral Products (4)
? Miscellaneous Commercial/lndustrial (12)
[H] Municipal Waste Combustor (1)
• Oil and/or Gas Production (2)
A Petroleum Refinery (1)
P PrinlingJPublishing(l)
R Rubber and Miscellaneous Plastics Products (2)
2 Secondary Metal Processing (1)
> Solid Waste Disposal - Commercial/Institutional (4)
V Steel Mill (2)
S Surface Coating (7)
TT Telecommunications (1)
•» Transportation Equipment (3)
^ Transportation and Marketing of Petroleum Products (2)
' Wastewater Treatment (1)
W Woodwork. Furniture, Millwork & Wood Preserving (1)
15-3
-------
Table 15-1. Geographical Information for the Michigan Monitoring Site
Site
Code
DEMI
AQS Code
26-163-0033
Location
Dearborn
County
Wayne
Micro- or
Metropolitan
Statistical Area
Detroit-Warren-
Livonia, MI MSA
(Detroit Div)
Latitude
and
Longitude
42.30754,
-83.14961
Land Use
Industrial
Location
Setting
Suburban
Additional Ambient Monitoring Information1
TSP Metals, Meteorological parameters, PM10, PM10
Speciation, PM2 5, and PM2 5 Speciation.
BOLD ITALICS = EPA-designated NATTS Site.
-------
DEMI is located at Paul Costea Park in Dearborn, just southwest of Detroit, and is the
Detroit NATTS site. The surrounding area is both suburban and industrial in nature. Figure 15-1
shows that a freight yard is located to the west of the site and a residential neighborhood is
located to the east. Industrial sources such as automobile and steel manufacturing facilities are
also located in the vicinity. The monitoring site lies between two heavily traveled roadways, 1-75
and 1-94. As Figure 15-2 shows, numerous point sources surround DEMI, a cluster of which is
located just southwest of the site. The source categories with the most point sources within 10
miles of DEMI include the aircraft operations source category, which includes airports as well as
small runways, heliports, or landing pads; bulk terminals and bulk plants; electricity generation
via combustion; institutional facilities (schools); and surface coating facilities.
Table 15-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the
Michigan monitoring site. Table 15-2 also includes a vehicle registration-to-county population
ratio (vehicles-per-person). In addition, the population within 10 miles of the site is presented.
An estimate of 10-mile vehicle ownership was calculated by applying the county-level vehicle
registration-to-population ratio to the 10-mile population surrounding the monitoring site.
Table 15-2 also contains annual average daily traffic information. Finally, Table 15-2 presents
the daily VMT for Wayne County.
Table 15-2. Population, Motor Vehicle, and Traffic Information for the Michigan
Monitoring Site
Site
DEMI
Estimated
County
Population1
1,815,734
County-level
Vehicle
Registration2
1,336,940
Vehicles per
Person
(Registration:
Population)
0.74
Population
within 10
miles3
1,082,362
Estimated
10-mile
Vehicle
Ownership
796,952
Annual
Average
Daily
Traffic4
106,900
County-
level Daily
VMT5
47,115,093
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the Michigan Dept of State (MDS, 2011)
3 10-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2010 data from the Michigan DOT (MI DOT, 2010)
5 County-level VMT reflects 2010 data for all public roads from the Michigan DOT (MI DOT, 2012)
BOLD ITALICS = EPA-designated NATTS Site.
Observations from Table 15-2 include the following:
• Wayne County's population ranks seventh highest and its vehicle registration ranks
eighth among counties with NMP sites.
15-5
-------
• The vehicle-per-person ratio for DEMI is in the bottom third among NMP sites.
• The 10-mile population and estimated 10-mile vehicle ownership for DEMI are in the
top third among NMP sites.
• Similar to several other characterizing statistics, the traffic volume near DEMI is in
the top third among NMP sites. Traffic for DEMI was obtained from 1-94, between
Ford Plant Road and Rotunda Drive.
• The Wayne County daily VMT is the fifth highest compared to other counties with
NMP sites (where VMT data were available).
15.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in Michigan on sample days, as well as over the course of the year.
15.2.1 Climate Summary
Detroit is located in a region of active weather. Winters tend to be cold and wet, while
summers are generally mild, although temperatures exceeding 90°F are not uncommon. Two
major influences on the city's weather include the urbanization of the area and Lake St. Clair to
the east. The lake tends to keep the Detroit area warmer in the winter and cooler in the summer
than more inland areas. The urban heat island keeps the city warmer than outlying areas. Winds
are often breezy and flow from the southwest on average. Precipitation is fairly well distributed
throughout the year, with summer precipitation coming primarily in the form of showers and
thunderstorms. Approximately 30 inches of snow falls on average during winter (Bair, 1992 and
MSU, 2012).
15.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest this site were retrieved
for 2010 (NCDC, 2010). The closest weather station to DEMI is located at Detroit City Airport
(WBAN 14822). Additional information about this weather station, such as the distance between
the site and the weather station, is provided in Table 15-3. These data were used to determine
how meteorological conditions on sample days vary from normal conditions throughout the year.
15-6
-------
Table 15-3. Average Meteorological Conditions near the Michigan Monitoring Site
Closest
NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
Average
Temperature
Average
Dew Point
Temperature
Average
Wet Bulb
Temperature
Average
Relative
Humidity
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Dearborn, Michigan - DEMI
Detroit City
Airport
14822
(42.41,
-83.01)
9.70
miles
54°
(NE)
Sample
Day
2010
59.3
±5.2
59.2
±2.2
52.0
±4.9
51.8
±2.0
40.2
±4.4
40.2
±1.9
46.1
±4.3
46.1
±1.8
66.8
±2.7
67.3
±1.3
1016.0
± 1.6
1015.7
±0.7
6.3
±0.6
6.6
±0.3
1 Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
Table 15-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 15-3 is the
95 percent confidence interval for each parameter. As shown in Table 15-3, average
meteorological conditions on sample days at DEMI were representative of average weather
conditions throughout the year.
15.2.3 Back Trajectory Analysis
Figure 15-3 is the composite back trajectory map for days on which samples were
collected at the DEMI monitoring site in 2010. Included in Figure 15-3 are four back trajectories
per sample day. Figure 15-4 is the corresponding cluster analysis for 2010. An in-depth
description of these maps and how they were generated is presented in Section 3.5.2.1. For the
composite map, each line represents the 24-hour trajectory along which a parcel of air traveled
toward the monitoring site on a given sample day and time. For the cluster analysis, each line
corresponds to a back trajectory representative of a given cluster of trajectories. For both maps,
each concentric circle around the site in Figures 15-3 and 15-4 represents 100 miles.
Observations from Figures 15-3 and 15-4 for DEMI include the following:
• Back trajectories originated from a variety of directions at the DEMI site, although
fewer trajectories originated from the east and southeast.
• The 24-hour air shed domain for DEMI was similar in size compared to other NMP
monitoring sites. The farthest away a trajectory originated was west-central
Minnesota, or less than 650 miles away. However, the average trajectory length was
250 miles and 88 percent of trajectories originated within 400 miles of the site.
• The cluster analysis shows that nearly 40 percent of back trajectories originated
northward over the Great Lakes and parts of Ontario and Quebec, Canada. Another
30 percent originated from a direction with a westerly component, from Minnesota
and Wisconsin to Illinois and Missouri. Nearly 20 percent of trajectories originated
from the south of DEMI, over Indiana, Ohio, Kentucky, and Tennessee. The
remaining back trajectories originated eastward over Lakes Ontario and Erie,
southern Ontario, Canada, western New York and Pennsylvania, and northeast Ohio.
15-8
-------
Figure 15-3. 2010 Composite Back Trajectory Map for DEMI
Figure 15-4. Back Trajectory Cluster Map for DEMI
15-9
-------
15.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at the Detroit City Airport were
uploaded into a wind rose software program to produce customized wind roses, as described in
Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals" positioned
around a 16-point compass, and uses different colors to represent wind speeds.
Figure 15-5 presents three different wind roses for the DEMI monitoring site. First, a
historical wind rose representing 2001 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location.
Observations from Figures 15-5 for DEMI include the following:
• The NWS weather station at Detroit City Airport is located 9.7 miles northeast of
DEMI. Downtown Detroit lies between the weather station and the site.
• The historical wind rose for DEMI shows that winds from a variety of directions were
observed near DEMI, although winds from the southeastern quadrant were observed
less frequently than winds from other directions. Calm winds (< 2 knots) were
observed for approximately 10 percent of the hourly measurements. The strongest
winds were observed with southwesterly and westerly winds.
• The wind patterns on the 2010 wind rose resemble the historical wind patterns,
indicating that conditions during 2010 were consistent with those experienced
historically.
• The sample day wind rose generally resembles the full-year wind rose, although there
were fewer west-northwesterly to northwesterly winds and more northerly to north-
northeasterly winds on sample days.
15-10
-------
Figure 15-5. Wind Roses for the Detroit City Airport Weather Station near DEMI
2001-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between DEMI and NWS Station
15-11
-------
15.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Michigan monitoring site in
order to allow analysts and readers to focus on a subset of pollutants through the context of risk.
Each pollutant's preprocessed daily measurement was compared to its associated risk screening
value. If the concentration was greater than the risk screening value, then the concentration
"failed the screen." Pollutants of interest are those for which the individual pollutant's total
failed screens contribute to the top 95 percent of the site's total failed screens. In addition, if any
of the NATTS MQO Core Analytes measured by the monitoring site did not meet the pollutant
of interest criteria based on the preliminary risk screening, that pollutant was added to the list of
site-specific pollutants of interest. A more in-depth description of the risk screening process is
presented in Section 3.2.
Table 15-4 presents the pollutants of interest for DEMI. The pollutants that failed at least
one screen and contributed to 95 percent of the total failed screens for the monitoring site are
shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of interest are shaded and/or
bolded. DEMI sampled for VOC, PAH, carbonyl compounds, and hexavalent chromium.
Table 15-4. Risk Screening Results for the Michigan Monitoring Site
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Dearborn, Michigan - DEMI
Benzene
Carbon Tetrachloride
Acetaldehyde
Formaldehyde
Naphthalene
1,3-Butadiene
Ethylbenzene
Acenaphthene
Fluorene
1 ,2-Dichloroethane
£>-Dichlorobenzene
Hexavalent Chromium
Fluoranthene
Acrylonitrile
Benzo(a)pyrene
Trichloroethylene
0.13
0.17
0.45
0.077
0.029
0.03
0.4
0.011
0.011
0.038
0.091
0.000083
0.011
0.015
0.00057
0.2
Total
61
61
60
60
59
58
33
18
17
12
8
7
4
3
3
2
466
61
61
60
60
59
60
61
59
59
12
29
53
59
3
55
15
766
100.00
100.00
100.00
100.00
100.00
96.67
54.10
30.51
28.81
100.00
27.59
13.21
6.78
100.00
5.45
13.33
60.84
13.09
13.09
12.88
12.88
12.66
12.45
7.08
3.86
3.65
2.58
1.72
1.50
0.86
0.64
0.64
0.43
13.09
26.18
39.06
51.93
64.59
77.04
84.12
87.98
91.63
94.21
95.92
97.42
98.28
98.93
99.57
100.00
15-12
-------
Observations from Table 15-4 for DEMI include the following:
• Sixteen pollutants, of which nine are NATTS MQO Core Analytes, failed at least one
screen for DEMI.
• Eleven pollutants contributed to 95 percent of all failed screens for DEMI; of these
six are NATTS MQO Core Analytes. Hexavalent chromium, benzo(a)pyrene, and
trichloroethylene were added to DEMFs pollutants of interest because they are
NATTS MQO Core Analytes, even though they did not contribute to 95 percent of
the total failed screens. Three additional pollutants (chloroform, tetrachloroethylene,
and vinyl chloride) were also added to the list, even though they did not fail any
screens. These three pollutants are not shown in Table 15-4.
• Of the pollutants failing screens, nearly 60 percent of their measured detections failed
screens. Seven pollutants failed 100 percent of their screens.
• The six pollutants failing the most screens contributed to over 75 percent of the total
failed screens, are all NATTS MQO Core Analytes, and, with the exception of
1,3-butadiene, failed 100 percent of their screens.
15.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Michigan monitoring site. Concentration averages are provided for the pollutants of
interest for DEMI, where applicable. Concentration averages for select pollutants are also
presented graphically for the site, where applicable, to illustrate how the site's concentrations
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at the
site, where applicable. Additional site-specific statistical summaries are provided in Appendices
J, L, M, and O.
15.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for the Michigan site, as described in Section 3.1. The quarterly average of a particular pollutant
is simply the average concentration of the preprocessed daily measurements over a given
calendar quarter. Quarterly average concentrations include the substitution of zeros for all non-
detects. A site must have a minimum of 75 percent valid samples of the total number of samples
possible within a given quarter for a quarterly average to be calculated. An annual average
includes all measured detections and substituted zeros for non-detects for the entire year of
sampling. Annual averages were calculated for pollutants where three valid quarterly averages
15-13
-------
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for DEMI are presented in
Table 15-5, where applicable. Note that concentrations of the PAH and hexavalent chromium are
presented in ng/m3 for ease of viewing. Also note that if a pollutant was not detected in a given
calendar quarter, the quarterly average simply reflects "0" because only zeros substituted for
non-detects were factored into the quarterly average concentration.
Table 15-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Michigan Monitoring Site
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(jig/m3)
Dearborn, Michi
Acetaldehyde
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Ethylbenzene
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Acenaphthene3
Benzo(a)pyrenea
Fluorene3
Hexavalent Chromium3
Naphthalene3
60/60
61/61
60/61
61/61
61/61
29/61
12/61
61/61
60/60
56/61
15/61
7/61
59/59
55/59
59/59
53/59
59/59
1.25
±0.28
0.82
±0.27
0.08
±0.04
0.67
±0.07
0.39
±0.08
0.01
±0.01
0.03
±0.02
0.33
±0.19
2.34
±0.40
0.16
±0.06
<0.01
±<0.01
0.01
±0.01
3.05
±2.20
0.23
±0.10
3.69
±1.77
0.03
±0.02
108.88
±41.82
2nd
Quarter
Average
(jig/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Annual
Average
(jig/m3)
gan - DEMI
1.57
±0.32
0.82
±0.19
0.06
±0.02
0.70
±0.06
0.89
±0.17
0.05
±0.03
0.03
±0.02
0.63
±0.34
3.26
±0.86
0.17
±0.06
0.01
±0.01
0.01
±0.01
13.29
±6.75
0.08
±0.05
11.78
±5.55
0.05
±0.02
120.78
±32.56
1.77
±0.25
1.04
±0.18
0.08
±0.02
0.71
±0.08
0.79
±0.17
0.04
±0.02
0
0.59
±0.16
3.36
±0.63
0.21
±0.04
0.02
±0.01
0
32.42
±27.93
0.22
±0.14
29.47
± 24.73
0.05
±0.01
179.91
±43.40
1.64
±0.36
1.06
±0.26
0.12
±0.05
0.69
±0.06
0.43
±0.06
0.03
±0.03
0.01
±0.01
0.45
±0.14
2.26
±0.41
0.32
±0.17
0.06
±0.05
0.01
±0.01
7.43
±2.90
0.16
±0.03
6.66
±2.21
0.05
±0.02
144.59
± 48.05
1.56
±0.15
0.94
±0.11
0.09
±0.02
0.69
±0.03
0.63
±0.08
0.03
±0.01
0.01
±0.01
0.50
±0.11
2.80
±0.31
0.21
±0.05
0.02
±0.01
0.01
±0.01
13.74
±7.06
0.17
±0.04
12.62
±6.23
0.04
±0.01
137.84
± 20.80
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for
ease of viewing.
15-14
-------
Observations for DEMI from Table 15-5 include the following:
• The pollutants with the highest annual average concentrations by mass are
formaldehyde and acetaldehyde; all other annual average concentrations are less than
1.0|ig/m3.
• Although the averages for formaldehyde appear higher during the warmer months of
the year, the confidence intervals indicate that the difference is not significant.
• The second and third quarter averages of chloroform are significantly higher than the
first and fourth quarter averages for this pollutant. A review of the data shows that the
eight highest concentrations (those greater than 1 |ig/m3) were measured between
June and August 2010.
• The second quarter average of ethylbenzene has a relatively high confidence interval
associated with it. A review of the data shows that the highest concentration of this
pollutant was measured on June 19, 2010 (2.52 |ig/m3) and was nearly twice the next
highest concentration (1.39 |ig/m3 measured on August 18, 2010).
• The fourth quarter average of tetrachloroethylene also has a relatively high
confidence interval associated with it. A review of the data shows that the highest
concentration of this pollutant was measured on November 10, 2010 (1.09 |ig/m3) and
was the only concentration of this pollutant greater than 1.0 |ig/m3 measured at
DEMI. In addition, it is one of only 13 concentrations of tetrachloroethylene
measured among NMP sites sampling VOC that was greater than 1.0 |ig/m3 (and
ranked 10th highest among these 13).
• The fourth quarter average of trichloroethylene is higher than the other quarterly
averages and has a relatively high confidence interval associated with it. The four
highest concentrations of this pollutant were all measured during the fourth quarter of
2010 and were the only measurements greater than 0.1 |ig/m3 measured. Note that this
pollutant was detected at DEMI in only 15 out of 61 valid samples.
• Fluorene and acenaphthene have relatively high confidence intervals for the third
quarter of 2010. The highest concentrations of these pollutants were measured on the
same days, August 18, 2010 (152 and 175 ng/m3, respectively) and August 30, 2010
(114 and 121 ng/m3, respectively). These concentrations are two and three times
higher than the next highest concentrations measured for these pollutants (which were
also measured on the same day, June 19, 2010) and are the top two concentrations of
these pollutants among NMP sites sampling PAH. The third and fourth highest
concentrations of naphthalene measured at DEMI were also measured on these two
sample days in August.
15-15
-------
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for DEMI from
those tables include the following:
• DEMI has the highest concentration of vinyl chloride among the NMP sites sampling
VOC. However, it should be noted that this pollutant was detected in only seven out
of 61 valid samples. Yet this was also the highest number of measured detections of
vinyl chloride among the sites sampling VOC.
• DEMI has the second highest annual average concentration of chloroform, behind
NBIL, and the fourth highest annual average concentration of carbon tetrachloride.
• DEMI's annual average concentrations of acenaphthene and fluorene are the highest
among NMP sites sampling PAH. DEMI's annual average concentrations of
benzo(a)pyrene and naphthalene both rank third highest.
• The annual average concentration of hexavalent chromium for DEMI ranks fourth
highest among sites sampling this pollutant.
15.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for acetaldehyde, benzene,
benzo(a)pyrene, 1,3-butadiene, formaldehyde, hexavalent chromium, and naphthalene were
created for DEMI. Figures 15-6 through 15-12 overlay the site's minimum, annual average, and
maximum concentrations onto the program-level minimum, first quartile, average, median, third
quartile, and maximum concentrations, as described in Section 3.5.3.
Figure 15-6. Program vs. Site-Specific Average Acetaldehyde Concentration
Concentration (|ig/m3)
Program
Site:
: 1st Quartile
Site Average
0
2nd Quartile 3rd Quartile 4th Quartile Avt
Site Minimum/Maximum
;rage
15-16
-------
Figure 15-7. Program vs. Site-Specific Average Benzene Concentration
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 15-8. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
Program Max Concentration =42.7 ng/m3
0.4 0.6
0.8 1 1.2
Concentration (ng/m3)
1.4 1.6
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 15-9. Program vs. Site-Specific Average 1,3-Butadiene Concentration
H
0 0.1
0.2 0.3
0.4
0.5
0.6
0.7 0.8
0.9 1
Concentration (ng/m3)
Program:
Site:
IstQuartile
Site Average
0
2ndQuartile SrdQuartile
Site Minimum/Maximum
4thQuartile Ave
;rage
15-17
-------
Figure 15-10. Program vs. Site-Specific Average Formaldehyde Concentration
I b
1 p
15 20
25 30 35
Concentration (ng/m3)
45 50
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 15-11. Program vs. Site-Specific Average Hexavalent Chromium Concentration
Program Max Concentration =3.51 ng/m3
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
Figure 15-12. Program vs. Site-Specific Average Naphthalene Concentration
fir
600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
O
15-18
-------
Observations from Figures 15-6 through 15-12 include the following:
• Figure 15-6 shows that DEMI's annual average acetaldehyde concentration is
below the program-level average and median concentration. The maximum
concentration measured at DEMI is well below the maximum concentration
measured at the program level. There were no non-detects of acetaldehyde
measured at DEMI.
• Figure 15-7 shows that DEMI's annual average benzene concentration is just
below the program-level average concentration. The maximum concentration of
benzene measured at DEMI is well below the maximum concentration measured
at the program level. There were no non-detects of benzene measured at DEMI.
• Figure 15-8 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
DEMI is greater than the program-level average concentration. Although
Figure 15-8 shows that the maximum concentration measured at DEMI is well
below the maximum concentration measured across the program, this
concentration is the ninth highest concentration of benzo(a)pyrene measured
among NMP sites sampling this pollutant.
• Figure 15-9 for 1,3-butadiene shows that the annual average concentration for
DEMI is just greater than the program-level average concentration. The maximum
concentration of 1,3-butadiene measured at DEMI is well below the maximum
concentration measured at the program level. One non-detect of 1,3-butadiene
was measured at DEMI.
• Figure 15-10 shows that DEMI's annual average formaldehyde concentration is
greater than the program-level average and median concentrations. The maximum
concentration measured at DEMI is well below the maximum concentration
measured at the program level. The minimum concentration of formaldehyde
measured at DEMI is greater than the program-level first quartile (25th percentile).
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 15-11 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot. The
box plot shows that the annual average concentration of hexavalent chromium for
DEMI is greater than the program-level average concentration. The maximum
concentration measured at DEMI is well below the maximum concentration
measured across the program. There were a few non-detects of this pollutant
measured at DEMI.
15-19
-------
• Figure 15-12 shows that the annual naphthalene average for DEMI is greater than
the program-level average concentration. Recall from the previous section that
DEMI's annual average concentration is the third highest annual average among
NMP sites sampling this pollutant. The maximum naphthalene concentration
measured at DEMI is well below the program-level maximum concentration. The
minimum concentration of naphthalene measured at DEMI is greater than the
program-level first quartile (25th percentile).
15.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. DEMI has sampled VOC and carbonyl compounds under the NMP since 2003 and
hexavalent chromium since 2005. However, a trends analysis was not conducted for the carbonyl
compounds. Carbonyl compound samples from the primary sampler were invalidated from
March 13, 2007 through March 25, 2008 by the state of Michigan due to a leak in the sample
line. Thus, Figures 15-13 through 15-15 present the 3-year rolling statistical metrics for benzene,
1,3-butadiene, and hexavalent chromium for DEMI. The statistical metrics presented for
assessing trends include the substitution of zeros for non-detects.
Observations from Figure 15-13 for benzene measurements at DEMI include the
following:
• The three highest benzene concentrations were measured in 2004, and ranged from
7.62 |ig/m3to 5.44 |ig/m3.
• Both the median and rolling average concentrations exhibit a decreasing trend over
the time periods shown. The difference between these two parameters decreased over
the years, indicating a decrease in the variability of concentrations measured.
• The 5th and 95th percentiles show a steady decreasing trend as does the difference
between the two percentiles, although the decrease slowed in later years.
• Sampling increased from a l-in-12 day sampling schedule in 2003 to a l-in-6 day
sampling schedule in 2004, which continues into 2010.
• The minimum concentration is greater than zero for all 3-year periods, indicating that
this pollutant has been detected in every sample collected at DEMI.
15-20
-------
Figure 15-13. Three-Year Rolling Statistical Metrics for Benzene Concentrations
Measured at DEMI
2003-2005 2004-2006
2005-2007 2006-2008
Three-Year Period
2007-2009 2008-2010
• 5th Percentile — Minimum — Median — Maximum • 95th Percentile ...^.. Average
Figure 15-14. Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations
Measured at DEMI
200J-2005 2004-2006
2005-2007 2006-2008
Three-Year Period
2007-2009 2008-2010
5th Percentile - Minimum - Median — Maximum • 95th Percentile -..^.- Average
15-21
-------
Figure 15-15. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at DEMI
I
!
«
t-
2005-2007
• 5th Percentile — Minimum
^™
2006-2008
Three-Year Period
^H
2007-2009
— Median — Maximum
• 95th Percentile
^^
2008-2010
•"»•• Average
Observations from Figure 15-14 for 1,3-butadiene measurements at DEMI include the
following:
• The maximum concentration was measured in 2004, although similar concentrations
were also measured in 2006.
• The rolling average concentrations have been fairly steady over the time periods
shown, although a slight decrease is shown for the 2007-2009 period. A review of the
confidence intervals calculated for these averages indicates that this decrease is not
statistically significant, although the confidence intervals calculated for the first two
3-year periods are relatively large.
• The maximum concentrations and 95th percentiles exhibit a decreasing trend
throughout the period of sampling, indicating a decrease in the range of
concentrations measured at DEMI.
• The minimum and 5th percentile are both zero for the first three 3-year periods shown,
indicating the presence of non-detects. However, the number of non-detects has been
decreasing since 2005, with only one to two non-detects reported per year since 2007.
• Sampling increased from a l-in-12 day sampling schedule in 2003 to a l-in-6 day
sampling schedule in 2004, which continues into 2010.
15-22
-------
Observations from Figure 15-15 for hexavalent chromium measurements at DEMI
include the following:
• The maximum hexavalent chromium concentration was measured in 2006. The two
highest hexavalent chromium concentrations for this site were both measured around
July 4th - 0.496 ng/m3 on July 4, 2006 and 0.392 ng/m3 on July 5, 2008.
• A slight decrease in the rolling average concentrations is shown through the 2007-
2009 period. However, the confidence intervals calculated indicate that these changes
are not statistically significant.
• The minimum concentrations and 5th percentiles for all 3-year periods are zero,
indicating the presence of non-detects.
15.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the
Michigan monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
15.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Michigan monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
for DEMI were compared to the acute MRL; the quarterly averages were compared to the
intermediate MRL; and the annual averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the Michigan monitoring site were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
DEMI.
15.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Michigan monitoring site and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
15-23
-------
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 15-6, where applicable.
Table 15-6. Cancer and Noncancer Surrogate Risk Approximations for the Michigan
Monitoring Site
t Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(jig/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Dearborn, Michigan - DEMI
Vcenaphthene3
Acetaldehyde
Benzene
Benzo(a)pyrenea
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Ethylbenzene
Fluorene3
Formaldehyde
Hexavalent Chromium3
Naphthalene3
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.000088
0.0000022
0.0000078
0.00176
0.00003
0.000006
0.000011
0.000026
0.0000025
0.000088
0.000013
0.012
0.000034
2.6E-07
0.0000048
0.0000088
_
0.009
0.03
0.002
0.1
0.098
0.8
2.4
1
0.0098
0.0001
0.003
0.04
0.002
0.1
59/59
60/60
61/61
55/59
60/61
61/61
61/61
29/61
12/61
61/61
59/59
60/60
53/59
59/59
56/61
15/61
7/61
0.01
±0.01
1.56
±0.15
0.94
±0.11
O.01
±O.01
0.09
±0.02
0.69
±0.03
0.63
±0.08
0.03
±0.01
0.01
±0.01
0.50
±0.11
0.01
±0.01
2.80
±0.31
0.01
± 0.01
0.14
±0.02
0.21
±0.05
0.02
±0.01
0.01
±0.01
1.21
3.42
7.30
0.30
2.65
4.17
0.37
0.38
1.25
1.11
36.45
0.53
4.69
0.06
0.11
0.02
_
0.17
0.03
0.04
0.01
0.01
<0.01
0.01
<0.01
0.29
0.01
0.05
0.01
0.01
0.01
— = a Cancer URE or Noncancer RfC is not available.
3 For the annual average concentration of this pollutant in ng/m3, refer to Table 15-5.
15-24
-------
Observations from Table 15-6 include the following:
• The pollutants with the highest annual average concentrations for DEMI are
formaldehyde, acetaldehyde, and benzene.
• The pollutants with the highest cancer surrogate risk approximations for DEMI are
formaldehyde, benzene, and naphthalene (36.45, 7.30, and 4.69 in-a-million,
respectively). The cancer risk approximation for formaldehyde is an order of
magnitude higher than the other cancer risk approximations.
• None of the pollutants of interest have associated noncancer risk approximations
greater than 1.0 for DEMI. The pollutant with the highest noncancer risk
approximation for DEMI was formaldehyde (0.29).
15.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 15-7 and 15-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 15-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 15-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Table 15-7 and 15-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer table.
Further, the cancer and noncancer surrogate risk approximations based on the site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 15.3, DEMI sampled for VOC, PAH, carbonyl compounds, and hexavalent chromium. In
addition, the cancer and noncancer risk approximations are limited to those pollutants with
enough data to meet the criteria for annual averages to be calculated. A more in-depth discussion
of this analysis is provided in Section 3.5.5.3.
15-25
-------
Table 15-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Michigan Monitoring Site
to
Top 10 Total Emissions for Pollutants
with Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Dearborn, Michigan (Wayne County) - DEMI
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
Tetrachloroethylene
Coke Oven Emissions, PM
Trichloroethylene
1,134.11
612.43
500.03
334.62
134.38
74.43
52.69
47.69
35.72
13.67
Coke Oven Emissions, PM
Benzene
Formaldehyde
POM, Group 5a
Hexavalent Chromium, PM
1,3 -Butadiene
Arsenic, PM
Naphthalene
Nickel, PM
Ethylbenzene
3.54E-02
8.85E-03
7.96E-03
7.57E-03
6.69E-03
4.03E-03
3.81E-03
2.53E-03
1.42E-03
1.25E-03
Formaldehyde
Benzene
Naphthalene
Carbon Tetrachloride
Acetaldehyde
1,3 -Butadiene
Ethylbenzene
Acenaphthene
Fluorene
Hexavalent Chromium
36.45
7.30
4.69
4.17
3.42
2.65
1.25
1.21
1.11
0.53
-------
Table 15-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Michigan Monitoring Site
to
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer Toxicity
Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Dearborn, Michigan (Wayne County) - DEMI
Hydrochloric acid
Toluene
Xylenes
Methanol
Benzene
Formaldehyde
Hexane
Ethylbenzene
Acetaldehyde
Methyl isobutyl ketone
3,765.69
2,480.93
1,912.74
1,227.86
1,134.11
612.43
551.94
500.03
334.62
277.33
Acrolein
Hydrochloric acid
Manganese, PM
1,3 -Butadiene
Cyanide Compounds, gas
Formaldehyde
Arsenic, PM
Benzene
Acetaldehyde
Nickel, PM
1,866,893.10
188,284.52
98,186.30
67,191.86
64,535.63
62,493.26
58,998.69
37,803.81
37,180.09
32,767.48
Formaldehyde
Acetaldehyde
Naphthalene
1,3 -Butadiene
Benzene
Trichloroethylene
Carbon Tetrachloride
Chloroform
Tetrachloroethylene
Ethylbenzene
0.29
0.17
0.05
0.04
0.03
0.01
0.01
0.01
0.01
0.01
-------
Observations from Table 15-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in Wayne County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
cancer UREs) for Wayne County are coke oven emissions, benzene, and
formaldehyde.
• Six of the highest emitted pollutants in Wayne County also have the highest toxicity-
weighted emissions.
• Formaldehyde, benzene, and naphthalene have the highest cancer surrogate risk
approximations. For DEMI, these three pollutants, as well as ethylbenzene and
1,3-butadiene, appear on both emissions-based lists. Acetaldehyde is one of the
highest emitted pollutants but does not appear among those with the highest toxicity-
weighted emissions. Hexavalent chromium has the fifth highest toxicity-weighted
emissions but does not appear among the highest emitted. Carbon tetrachloride does
not appear on either emissions-based list.
• POM, Group 5a ranks fourth for toxicity-weighted emissions in Wayne County.
POM, Group 5a includes benzo(a)pyrene, which has one of the lowest cancer risk
approximations for DEMI.
Observations from Table 15-8 include the following:
• Hydrochloric acid, toluene, and xylenes are the highest emitted pollutants with
noncancer RfCs in Wayne County. Wayne County has the highest hydrochloric acid
emissions of any county with an NMP site.
• The pollutants with the highest toxicity-weighted emissions (of the pollutants with
noncancer RfCs) for Wayne County are acrolein, hydrochloric acid, and manganese.
Although acrolein was sampled for at DEMI, this pollutant was excluded from the
pollutants of interest designation and thus subsequent risk screening evaluations due
to questions about the consistency and reliability of the measurements, as discussed in
Section 3.2.
• Four of the highest emitted pollutants in Wayne County also have the highest
toxicity-weighted emissions.
• The pollutant with the highest noncancer risk approximation for DEMI is
formaldehyde (0.29), although none of the pollutants of interest have associated
noncancer risk approximations greater than 1.0. Formaldehyde emissions rank sixth
for Wayne County and sixth for toxicity-weighted emissions.
15-28
-------
15.6 Summary of the 2010 Monitoring Data for DEMI
Results from several of the data treatments described in this section include the
following:
»«» Sixteen pollutants, of which nine are NA TTS MQO Core Analytes, failed screens for
DEMI.
*»* Of the site-specific pollutants of interest, formaldehyde had the highest annual
average concentration for DEMI, followed by acetaldehyde and benzene.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than any of the
associated MRL noncancer health risk benchmarks.
15-29
-------
16.0 Site in Missouri
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in Missouri, and integrates these concentrations with
emissions, meteorological, and risk information. Data generated by sources other than ERG are
not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
16.1 Site Characterization
This section characterizes the S4MO monitoring site by providing geographical and
physical information about the location of the site and the surrounding area. This information is
provided to give the reader insight regarding factors that may influence the air quality near the
site and assist in the interpretation of the ambient monitoring measurements.
The S4MO monitoring site is located in the St. Louis, MO-IL MSA. Figure 16-1 is a
composite satellite image retrieved from ArcGIS Explorer showing the monitoring site in its
urban location. Figure 16-2 identifies point source emissions locations by source category, as
reported in the 2008 NEI for point sources. Note that only sources within 10 miles of the site are
included in the facility counts provided in Figure 16-2. Thus, sources outside the 10-mile radius
have been grayed out, but are visible on the map to show emissions sources outside the 10-mile
boundary. A 10-mile boundary was chosen to give the reader an indication of which emissions
sources and emissions source categories could potentially have a direct effect on the air quality at
the monitoring site. Further, this boundary provides both the proximity of emissions sources to
the monitoring site as well as the quantity of such sources within a given distance of the site.
Table 16-1 describes the area surrounding the monitoring site by providing supplemental
geographical information such as land use, location setting, and locational coordinates.
16-1
-------
Figure 16-1. St. Louis, Missouri (S4MO) Monitoring Site
to
-------
Figure 16-2. NEI Point Sources Located Within 10 Miles of S4MO
arm'trw so 5'trw
Note: Due la facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
S4MO NATTS site
Source Category Group (No. of Facilities)
i= Air-con diiioning/Ref figuration (1)
-fi Aircraft Operations (21)
I Aspnatt Processing/Roofing Manufacturing (1)
H Automobile/Truck Manufacturing (1)
$} Bakery (1)
y BreweryDistilleryMAnery (1)
ft Buiktiny Construction i'2)
B Bulk Termmaf&'Bulk Plants (6)
C Chem.cal Manufacturing (9)
• Concrete Batcti Plant (11)
E> ] Crematory -Animal/Human (1)
f£) Dry Cteamng Facility (4)
Electrical Equipment 42)
Electricity Generation via Combustion (4)
Electroplating, Plating. Polishing, Anodizing, & Coloring (2)
<*) Fabricated Metal Products (2)
;;^ Flexible Polyurethane Foam Production (1)
e
*
E
F
IT
A
it
ffl
*
*
o
10 mile radius I
Toed Proce&&inaYArjricultLir.e (1C)
Gasoline/Diesel Service Station 42}
Grain Handling (?)
Heating Equipms-nt Manufacturing (T)
HospilaUl)
Hat Mix Asphuii Plan! i-i-
IrtrJustnal Macriinery and Equipmenl (1)
Inslitulional - prison (1)
Inslilulional - school <7J
Iron and Sleel Foundry (1)
Landfill (3)
Leather and Leather Products 41)
Marine Port 4 5)
Military Base/National Security Facilrty (1)
Mine/Quarry (6)
Mineral Products (1)
Miscellaneous CommerclaVlndustfral (25)
County boundary
M Miscellaneous Manufacturing (12)
• Oil and/or Gas Production (3)
_ Pharmaceutical Manufacturing (2)
1 Pnmary Metal Production (4)
f Prinlltg. Coaling SDyeng of Fabric ID
P Pnnting/Purjlisning (5)
H Pulp and Paper PlanlMttod ProOucts (2)
R Rubber and Miscellaneous Plastics Products (2)
2 Secondary Metal Processing (2)
< Srte Remediation Actwlty (1)
> Solid Waste Disposal • Commerctal/lnsritulional (2)
Y Steel Mill (3)
S Surface Coabng (8)
TT Telecommunications (1)
T Textile Mill (2)
$ Transportation and Marketing of Petroleum Products (3)
> V&slewater Treatment (5)
16-3
-------
Table 16-1. Geographical Information for the Missouri Monitoring Site
Site
Code
S4MO
AQS Code
29-510-0085
Location
St. Louis
County
St. Louis
City
Micro- or
Metropolitan
Statistical Area
St. Louis, MO-IL
MSA
Latitude
and
Longitude
38.656436,
-90.198661
Land Use
Residential
Location
Setting
Urban/City
Center
Additional Ambient Monitoring Information1
CO, O3, Meteorological parameters, PM10, Black
carbon, PM2 5, PM25 Speciation, SO2, NOy, NO.
BOL D ITALICS = EPA-designated NATTS Site.
-------
S4MO is located in central St. Louis. Figure 16-1 shows that the S4MO monitoring site is
located less than 1/4 mile west of 1-70. The Mississippi River, which separates Missouri from
Illinois, is less than 1 mile east of the site. Although the area directly around the monitoring site
is primarily residential, industrial facilities are located just on the other side of 1-70. Figure 16-2
shows that a large number of point sources are located within 10 miles of S4MO. The source
categories with the highest number of point sources surrounding S4MO include aircraft
operations, which include airports as well as small runways, heliports, or landing pads; food
processing facilities; and concrete batch plants. In the immediate vicinity of S4MO are a
pharmaceutical manufacturing facility and a printing and publishing facility.
Table 16-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the Missouri
monitoring site. Table 16-2 also includes a vehicle registration-to-county population ratio
(vehicles-per-person). In addition, the population within 10 miles of the site is presented. An
estimate of the 10-mile vehicle ownership was calculated by applying the county-level vehicle
registration-to-population ratio to the 10-mile population surrounding the monitoring site.
Table 16-2 also contains annual average daily traffic information. Finally, Table 16-2 presents
the daily VMT for St. Louis City and County.
Table 16-2. Population, Motor Vehicle, and Traffic Information for the Missouri
Monitoring Site
Site
S4MO
Estimated
County
Population1
1,318,037
County-level
Vehicle
Registration2
1,121,528
Vehicles per
Person
(Registration:
Population)
0.85
Population
within 10
miles3
811,927
Estimated
10-mile
Vehicle
Ownership
690,875
Annual
Average
Daily
Traffic4
81,174
County-
level Daily
VMT5
23,385,327
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the Missouri Dept of Revenue (MO DOR, 2011)
3 10-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2009 data from the Missouri DOT (MO DOT, 2009)
5 County-level VMT reflects 2010 data for all public roads from the Missouri DOT (MO DOT, 2012)
BOLD ITALICS = EPA-designated NATTS Site.
Observations from Table 16-2 include the following:
• S4MO's county population and vehicle registration are in the upper third of the range
compared to other counties with NMP sites. This is also true for its 10-mile vehicle
ownership, but its 10-mile population ranked 18th compared to other sites.
16-5
-------
• The vehicle-per-person ratio is in the middle of the range compared to other NMP
sites.
• The traffic volume experienced near S4MO ranks 18th among other NMP monitoring
sites. The traffic estimate used came from 1-70 near Exit 250.
• The St. Louis City and County daily VMT ranks 11th among counties with NMP sites
(where VMT data were available).
16.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in Missouri on sample days, as well as over the course of the year.
16.2.1 Climate Summary
The City of St. Louis is located along the Mississippi River, which acts as Missouri's
eastern border. St. Louis has a climate that is continental in nature, with cold, dry winters; warm,
somewhat wetter summers; and significant seasonal variability. Warm, moist air flowing
northward from the Gulf of Mexico alternating with cold, dry air marching southward from
Canada and the northern U.S. result in weather patterns that do not persist for very long. The
City of St. Louis does experience the urban heat island effect, retaining more heat within the city
than outlying areas (Bair, 1992 and MCC, 2012).
16.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest this site were retrieved
for 2010 (NCDC, 2010). The closest weather station is located at St. Louis Downtown Airport
(WBAN 03960). Additional information about this weather station, such as the distance between
the site and the weather station, is provided in Table 16-3. These data were used to determine
how meteorological conditions on sample days vary from normal conditions throughout the year.
16-6
-------
Table 16-3. Average Meteorological Conditions near the Missouri Monitoring Site
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
From
Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
St. Louis, Missouri - S4MO
Ct T niiis
Downtown
Airport
03960
(38.57, -90.16)
6 27
miles
156°
(SSE)
Sample
Day
2010
65.6
±5.0
65.7
±2.2
55.5
±4.7
55.8
±2.1
45.2
±4.6
45.1
±2.0
50.2
±4.3
50.3
±1.9
70.9
±2.5
70.5
±1.1
1016.7
± 1.6
1016.2
±0.7
5.6
±0.8
5.6
±0.3
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
Table 16-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 16-3 is the
95 percent confidence interval for each parameter. As shown in Table 16-3, average
meteorological conditions on sample days were representative of average weather conditions
throughout the year.
16.2.3 Back Trajectory Analysis
Figure 16-3 is the composite back trajectory map for days on which samples were
collected at the S4MO monitoring site in 2010. Included in Figure 16-3 are four back trajectories
per sample day. Figure 16-4 is the corresponding cluster analysis for 2010. An in-depth
description of these maps and how they were generated is presented in Section 3.5.2.1. For the
composite map, each line represents the 24-hour trajectory along which a parcel of air traveled
toward the monitoring site on a given sample day and time. For the cluster analysis, each line
corresponds to a back trajectory representative of a given cluster of trajectories. For both maps,
each concentric circle around the site in Figures 16-3 and 16-4 represents 100 miles.
Figure 16-3. 2010 Composite Back Trajectory Map for S4MO
600
Miles
16-8
-------
Figure 16-4. Back Trajectory Cluster Map for S4MO
\
\ \
Observations from Figures 16-3 and 16-4 for S4MO include the following:
• Back trajectories originated from a variety of directions at S4MO, although
trajectories from the northwest and south-southeast to south-south west were most
common.
• The 24-hour air shed domain for S4MO was among the larger in size compared to
other NMP sites. The farthest away a trajectory originated was greater than 800 miles,
over southwest North Dakota. However, the average trajectory length was 235 miles.
Most trajectories (85 percent) originated within 400 miles of the monitoring site.
• The cluster analysis shows that many (27 percent) trajectories originated to the
northwest of S4MO, although of varying lengths. The cluster trajectory originating
from the southwest of S4MO (24 percent) represents trajectories originating over
western Illinois, Missouri, and northern Arkansas and within 250 miles of the site.
The cluster trajectory originating over southern Illinois (19 percent) represents
trajectories originating from the east, southeast, and south of the monitoring site and
within 200 miles of S4MO. Trajectories also originated from the south and northeast
of S4MO.
16-9
-------
16.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at St. Louis Downtown Airport near
S4MO were uploaded into a wind rose software program to produce customized wind roses, as
described in Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals"
positioned around a 16-point compass, and uses different colors to represent wind speeds.
Figure 16-5 presents three different wind roses for the S4MO monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location.
Observations from Figure 16-5 for S4MO include the following:
• The St. Louis Downtown Airport weather station is located approximately 6.3 miles
south-southeast of S4MO. The weather station location is across the Mississippi River
and state border in Illinois.
• The historical wind rose shows that winds from the southeast, south-southeast, and
south were frequently observed near S4MO. Winds from these directions accounted
for approximately 28 percent of observations. Calm winds (<2 knots) were observed
for approximately 18 percent of the hourly wind measurements. Winds from the west
to northwest to north account for the bulk of the remaining wind observations. The
strongest winds were from the west-northwest and northwest.
• The wind patterns shown on the 2010 wind rose generally resemble those shown on
the historical wind rose, although there were fewer southeasterly winds and more
northerly winds. The calms rate in 2010 was nearly 23 percent.
• The sample day wind patterns also resemble the historical and full-year wind patterns,
although the calm rate on sample days was greater than 25 percent.
16-10
-------
Figure 16-5. Wind Roses for the St. Louis Downtown Airport Weather Station near S4MO
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between S4MO and NWS Station
«£/ \ "-• ?
"rfi * ;••->.. .
v
\
Cjhokn
16-11
-------
16.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for S4MO in order to allow analysts
and readers to focus on a subset of pollutants through the context of risk. Each pollutant's
preprocessed daily measurement was compared to its associated risk screening value. If the
concentration was greater than the risk screening value, then the concentration "failed the
screen." Pollutants of interest are those for which the individual pollutant's total failed screens
contribute to the top 95 percent of the site's total failed screens. In addition, if any of the NATTS
MQO Core Analytes measured by the monitoring site did not meet the pollutant of interest
criteria based on the preliminary risk screening, that pollutant was added to the list of site-
specific pollutants of interest. A more in-depth description of the risk screening process is
presented in Section 3.2.
Table 16-4 presents S4MO's pollutants of interest. The pollutants that failed at least one
screen and contributed to 95 percent of the total failed screens for the monitoring site are shaded.
NATTS MQO Core Analytes are bolded. Thus, pollutants of interest are shaded and/or bolded.
S4MO sampled for VOC, PAH, carbonyl compounds, metals (PMio), and hexavalent chromium.
Observations from Table 16-4 include the following:
• Twenty-four pollutants, of which 14 are NATTS MQO Core Analytes, failed at least
one screen for S4MO.
• S4MO failed the highest number of screens among all NMP sites. More than 50
percent of measured detections failed screens (of the pollutants that failed at least one
screen) for S4MO.
• Seven pollutants failed 100 percent of screens for S4MO: acetaldehyde,
formaldehyde, benzene, acrylonitrile, 1,2-dichloroethane, 1,2-dibromoethane and
1,1,2,2-tetrachloroethane. The last three pollutants were detected in only a few
samples.
• Sixteen pollutants were identified as pollutants of interest for S4MO based on the risk
screening process; of these, 10 are NATTS MQO Core Analytes. Four additional
pollutants (nickel, hexavalent chromium, trichloroethylene, and benzo(a)pyrene) were
added to S4MO's pollutants of interest because they are NATTS MQO Core
Analytes, even though they did not contribute to 95 percent of S4MO's failed screens.
Four more pollutants (beryllium, chloroform, tetrachloroethylene, and vinyl chloride)
were also added to S4MO's pollutants of interest because they are NATTS MQO
Core Analytes, even though they did not fail any screens. These four pollutants are
not shown in Table 16-4.
16-12
-------
Table 16-4. Risk Screening Results for the Missouri Monitoring Site
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
St. Louis, Missouri - S4MO
Arsenic (PM10)
Naphthalene
Acet aldehyde
Formaldehyde
Benzene
1,3-Butadiene
Carbon Tetrachloride
Manganese (PM10)
Cadmium (PM10)
£>-Dichlorobenzene
Ethylbenzene
Acrylonitrile
Fluorene
Lead (PM10)
Acenaphthene
1 ,2-Dichloroethane
Hexavalent Chromium
Trichloroethylene
Nickel (PM10)
Benzo(a)pyrene
Fluoranthene
Propionaldehyde
1 ,2-Dibromoethane
1 , 1 ,2,2-Tetrachloroethane
0.00023
0.029
0.45
0.077
0.13
0.03
0.17
0.005
0.00056
0.091
0.4
0.015
0.011
0.015
0.011
0.038
0.000083
0.2
0.0021
0.00057
0.011
0.8
0.0017
0.017
Total
58
57
54
54
53
52
52
48
26
23
16
14
14
12
11
10
5
4
3
2
2
2
1
1
574
60
58
54
54
53
53
53
60
60
45
53
14
58
60
58
10
46
20
60
50
58
54
1
1
1093
96.67
98.28
100.00
100.00
100.00
98.11
98.11
80.00
43.33
51.11
30.19
100.00
24.14
20.00
18.97
100.00
10.87
20.00
5.00
4.00
3.45
3.70
100.00
100.00
52.52
10.10
9.93
9.41
9.41
9.23
9.06
9.06
8.36
4.53
4.01
2.79
2.44
2.44
2.09
1.92
1.74
0.87
0.70
0.52
0.35
0.35
0.35
0.17
0.17
10.10
20.03
29.44
38.85
48.08
57.14
66.20
74.56
79.09
83.10
85.89
88.33
90.77
92.86
94.77
96.52
97.39
98.08
98.61
98.95
99.30
99.65
99.83
100.00
16.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Missouri monitoring site. Concentration averages are provided for the pollutants of interest
for S4MO, where applicable. Concentration averages for select pollutants are also presented
graphically for the site, where applicable, to illustrate how the site's concentrations compare to
the program-level averages. In addition, concentration averages for select pollutants are
presented from previous years of sampling in order to characterize concentration trends at the
site, where applicable. Additional site-specific statistical summaries are provided in
Appendices J, L, M, N, and O.
16-13
-------
16.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for the Missouri site, as described in Section 3.1. The quarterly average of a particular pollutant
is simply the average concentration of the preprocessed daily measurements over a given
calendar quarter. Quarterly average concentrations include the substitution of zeros for all non-
detects. A site must have a minimum of 75 percent valid samples of the total number of samples
possible within a given quarter for a quarterly average to be calculated. An annual average
includes all measured detections and substituted zeros for non-detects for the entire year of
sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for S4MO are presented in
Table 16-5, where applicable. Note that concentrations of the PAH, metals, and hexavalent
chromium are presented in ng/m3 for ease of viewing. Also note that if a pollutant was not
detected in a given calendar quarter, the quarterly average simply reflects "0" because only zeros
substituted for non-detects were factored into the quarterly average concentration.
Table 16-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Missouri Monitoring Site
Pollutant
#of
Measured
Detections vs.
# of Samples
1st
Quarter
Average
(Hg/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Annual
Average
(jig/m3)
St. Louis, Missouri - S4MO
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
54/54
14/53
53/53
53/53
53/53
52/53
45/53
10/53
4.10
±1.26
0.04
±0.03
0.94
±0.24
0.10
±0.05
0.54
±0.10
0.13
±0.04
0.09
±0.14
0.02
±0.02
4.13
±1.23
0.01
±0.02
0.90
±0.23
0.10
±0.04
0.60
±0.07
0.20
±0.08
0.27
±0.26
0.03
±0.02
4.07
±1.29
NA
NA
NA
NA
NA
NA
NA
4.09
±1.26
0.37
±0.29
1.26
±0.38
0.16
±0.08
0.59
±0.08
0.19
±0.08
0.42
±0.41
0.01
±0.02
4.10
±0.59
0.17
±0.14
1.03
±0.15
0.12
±0.03
0.58
±0.05
0.19
±0.04
0.35
±0.18
0.02
±0.01
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of
viewing.
16-14
-------
Table 16-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Missouri Monitoring Site (Continued)
Pollutant
Ethylbenzene
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Acenaphthene3
Arsenic (PM10)a
Benzo(a)pyrenea
Bery Ilium (PM10)a
Cadmium (PM10)a
Fluorene3
Hexavalent Chromium3
Lead(PM10)a
Manganese (PM10) a
Naphthalene3
Nickel (PM10)a
#of
Measured
Detections vs.
# of Samples
53/53
54/54
51/53
20/53
4/53
58/58
60/60
50/58
59/60
60/60
58/58
46/57
60/60
60/60
58/58
60/60
1st
Quarter
Average
(Hg/m3)
0.30
±0.13
1.69
±0.39
0.17
±0.10
0.02
±0.04
0
2.06
±0.78
0.81
±0.17
0.25
±0.14
0.01
±0.01
0.64
±0.22
2.81
±0.93
0.02
±0.01
9.60
±4.22
11.33
±6.62
133.57
± 102.60
0.93
±0.22
2nd
Quarter
Average
(Ug/m3)
0.41
±0.19
2.97
±0.83
0.21
±0.09
0.04
±0.03
O.01
±O.01
5.77
±2.13
0.89
±0.25
0.07
±0.03
0.01
±0.01
0.48
±0.17
7.06
±2.56
0.03
±0.01
8.59
±4.35
13.06
±6.07
106.80
±30.37
0.96
±0.16
3rd
Quarter
Average
(jig/m3)
NA
3.42
±0.57
NA
NA
NA
10.55
±3.00
1.12
±0.58
0.10
±0.05
0.01
±0.01
0.58
±0.18
11.17
±3.18
0.05
±0.02
12.84
±6.39
11.63
±3.69
151.96
± 60.47
0.92
±0.16
4th
Quarter
Average
(jig/m3)
0.51
±0.25
2.99
±0.46
0.30
±0.16
0.07
±0.05
O.01
±0.01
4.92
±2.91
1.26
±0.51
0.22
±0.09
0.01
±0.01
0.78
±0.31
5.47
±2.65
0.04
±0.02
15.63
±5.96
32.58
± 26.78
150.34
± 82.45
1.35
±0.48
Annual
Average
(jig/m3)
0.44
±0.11
2.74
±0.33
0.23
±0.06
0.05
±0.02
O.01
±O.01
5.76
±1.35
1.02
±0.20
0.16
±0.05
0.01
±0.01
0.62
±0.11
6.57
±1.39
0.03
±0.01
11.66
±2.60
17.15
±7.07
135.13
±35.06
1.04
±0.14
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of
viewing.
Observations for S4MO from Table 16-5 include the following:
• The pollutants with the highest annual average concentrations by mass are
acetaldehyde (4.10 ± 0.59 |ig/m3), formaldehyde (2.74 ± 0.33 |ig/m3), and benzene
(1.03 ±0.15 |ig/m3).
• The annual average acetaldehyde concentration for S4MO is the third highest annual
average concentration for any site-specific pollutant of interest (not just carbonyl
compounds) among all NMP sites. Fifteen concentrations of acetaldehyde measured
at S4MO were greater than 5 |ig/m3, which is the highest among NMP sites sampling
16-15
-------
this pollutant (SKFL has the second highest at 11). The median acetaldehyde
concentration for S4MO is 3.81 |ig/m3.
• Third quarter averages could not be calculated for the VOC pollutants of interest
because there were not enough valid samples to meet the quarterly completeness
criteria. Several canisters returned to the laboratory under excess vacuum, including
several duplicate pairs. There was also an invalid sample attributed to a sampling
error.
• Several of the VOC have relatively large fourth quarter average concentrations as
well as large associated confidence intervals, including acrylonitrile, benzene,
1,3-butadiene,/7-dichlorobenzene, ethylbenzene, tetrachloroethylene, and
trichloroethylene. The data for these pollutants are discussed below.
• The highest concentration of acrylonitrile was measured during the third quarter on
September 29, 2010 (2.94 |ig/m ). This is the second highest concentration of this
pollutant among all NMP sites sampling VOC. As discussed above, third quarter
average concentrations could not be calculated for S4MO. However, five of the six
highest concentrations of acrylonitrile were measured at S4MO during December
2010. This pollutant was detected in only 14 of 53 samples collected at S4MO, with
most measured during the first quarter and fourth quarter of 2010.
• The five highest concentrations of the remaining VOC discussed above were
reviewed to see if correlations exist among the sample dates. Two dates appear in the
top five for each pollutant: September 29, 2010 and October 5, 2010. The
September 29, 2010 sample has the highest concentrations of/>-dichlorobenzene and
ethylbenzene measured at this site, the second highest concentrations of benzene,
1,3-butadiene, and trichloroethylene, and the third highest concentrations of
tetrachloroethylene. Because third quarter average concentrations could not be
calculated for S4MO, how the third quarter averages may have been affected by these
results cannot be determined. The October 5, 2010 sample has the highest
concentrations of benzene and 1,3-butadiene measured at this site, the second highest
concentrations of ethylbenzene and tetrachloroethylene, and the fifth highest
concentrations ofp-dichlorobenzene and trichloroethylene. Benzene, 1,3-butadiene,
and tetrachloroethylene also have relatively high concentrations on
December 28, 2010 while/>-dichlorobenzene, ethylbenzene, and trichloroethylene
have relatively high concentrations on October 11, 2010.
• The highest hexavalent chromium concentration for S4MO was measured on
July 4, 2010. There were only four concentrations of hexavalent chromium greater
than 0.1 ng/m3 measured at S4MO, two of which were measured in July 2010 (with
the others measured in October and November).
16-16
-------
• The first quarter naphthalene average has a very large confidence interval associated
with it. The highest concentration of naphthalene was measured on January 20, 2010
(784 ng/m3), which was the 11th highest naphthalene concentration measured among
NMP sites sampling PAH. However, the third and fourth quarter 2010 averages of
naphthalene are higher than the other quarterly averages and also have large
confidence intervals associated with them. The second and third highest
concentrations of this pollutant (554 ng/m3 and 483 ng/m3) were measured on
October 5, 2010 and September 29, 2010, respectively, corresponding to the dates of
some of the highest VOC measurements.
• The third quarter average concentrations of fluorene and acenaphthene are higher
than the other quarterly average and have relatively large confidence intervals
associated with them. The highest concentrations of both pollutants were measured on
August 12, 2010. Of the six highest concentrations of each of these pollutants, five
occurred on the same sample days, of which most were collected during the third
quarter.
• The first and fourth quarter averages of benzo(a)pyrene also exhibit large confidence
intervals. The highest concentration of this pollutant (1.03 ng/m3) was the ninth
highest concentration of this pollutant measured among all NMP sites sampling PAH.
The two highest concentrations of this pollutant were both measured in February
2010 at S4MO. Of the 15 highest concentrations of benzo(a)pyrene for S4MO, five
were measured during the first quarter of 2010, none were measured during the
second quarter, three were measured during the third quarter, and seven were
measured during the fourth quarter.
• Several of the quarterly averages for the PMi0 metals are highest for the third or
fourth quarters of 2010. Some of these have rather large confidence intervals
associated with outliers, as described in the next several bullets.
• The fourth quarter average of manganese for S4MO is almost three times higher than
the other quarterly averages and has a large confidence interval associated with it. On
November 16, 2010, the concentration of manganese measured at S4MOs was
200 ng/m3, which is also the highest manganese measurement among all NMP sites
sampling metals. The next highest manganese concentration measured at S4MO was
half as high (84.5 ng/m3 on December 10, 2010), but is still the second highest
manganese concentration among all sites sampling metals. Of the 13 concentrations
of manganese greater than 20 ng/m3, the majority were measured during the fourth
quarter.
• The third and fourth quarter averages of lead for 2010 are higher than the other
quarterly averages. The two highest concentrations of lead were measured at S4MO
on August 6, 2010 (47.8 ng/m3) and October 6, 2010 (42.6 ng/m3). Of the 12 highest
concentrations measured at S4MO, three were measured during the first half of 2010
and nine were measured during the second. Twelve of the 13 highest concentrations
of lead among all NMP sites were measured at S4MO.
16-17
-------
• The highest concentration of arsenic was also measured at S4MO on August 6, 2010
(4.77 ng/m3). This is also the highest concentration of arsenic measured among all
NMP sites sampling metals. The other three concentrations arsenic greater than
2 ng/m3 were measured during the fourth quarter of 2010, including October 6, 2010,
which is the day the second highest lead measurement was collected.
• The fourth quarter average of nickel for S4MO is higher than the other quarterly
averages and has a relatively large confidence interval associated with it. The highest
concentration of nickel (3.59 ng/m3) was measured on October 11, 2010, and is the
11th highest concentration of nickel measured among all NMP sites sampling metals.
Four of the five highest concentrations were measured in October and November.
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for S4MO from
those tables include the following:
• S4MO appears in Tables 4-9 through 4-12 a total of 21 times, the most of any NMP
site.
• S4MO has the second highest annual average concentrations of acrylonitrile and
/>-dichlorobenzene (although both have relatively large confidence intervals), and the
fifth highest annual average concentrations of tetrachloroethylene and
trichl oroethy 1 ene.
• S4MO has the highest annual average concentration of acetaldehyde and the tenth
highest concentration of formaldehyde.
• S4MO's annual average concentrations of the four PAHs ranked among the top five
compared to other sites sampling the program-level PAH pollutants of interest.
• S4MO has the highest annual average concentrations of arsenic, beryllium, cadmium,
lead, and manganese among NMP sites sampling PMio metals (and in some cases,
those sampling TSP metals).
16.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for acetaldehyde, arsenic,
benzene, benzo(a)pyrene, 1,3-butadiene, formaldehyde, hexavalent chromium, manganese, and
naphthalene were created for S4MO. Figures 16-6 through 16-14 overlay the site's minimum,
annual average, and maximum concentrations onto the program-level minimum, first quartile,
average, median, third quartile, and maximum concentrations, as described in Section 3.5.3.
16-18
-------
Figure 16-6. Program vs. Site-Specific Average Acetaldehyde Concentration
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
o
Figure 16-7. Program vs. Site-Specific Average Arsenic (PMio) Concentration
-
—o
2 2.5 3
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 16-8. Program vs. Site-Specific Average Benzene Concentration
) 1 2
Program: IstQuartile
•
Site: Site Average
O
3456
Concentration (ng/m3)
2ndQuartile 3rdQuartile 4thQuartile Average
n n n
Site Minimum/Maximum
7
16-19
-------
Figure 16-9. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
0
1 ,
) 0.2 0.4 0.6
Program: IstQuartile
•
Site: Site Average
O
Figure 16-10. Program vs
• 1
) 0.1 0.2 0.3
Program: IstQuartile
•
Site: SiteAverage
0
Figure 16-11. Program vs.
Vr~
) 5 10 15
Program: IstQuartile
•
Site: SiteAverage
O
1 1 1
Program Max Concentration = 42.7 ng/m3
| |
0.8 1 1.2 1.4 1.6 1.8 2
Concentration (ng/m3)
2nd Quartile 3rd Quartile 4th Quartile Average
D D D
Site Minimum/Maximum
Site-Specific Average 1,3-Butadiene Concentration
| |
0.4 0.5 0.6 0.7 0.8 0.9 1
Concentration (ng/m3)
2nd Quartile SrdQuartile 4th Quartile Average
Site Minimum/Maximum
Site-Specific Average Formaldehyde Concentration
| |
20 25 30 35 40 45 50 55
Concentration (ng/m3)
2nd Quartile SrdQuartile 4th Quartile Average
• D D
Site Minimum/Maximum
16-20
-------
Figure 16-12. Program vs. Site-Specific Average Hexavalent Chromium Concentration
1
j Program Max Concentration = 3.51 ng/m3
1
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 16-13. Program vs. Site-Specific Average Manganese (PMio) Concentration
PH
80 100 120 140 160 180 200
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
Figure 16-14. Program vs. Site-Specific Average Naphthalene Concentration
'
|.
1 1 1 1
3 200 400 600 800 1000 1200
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile
• DM
Site: Site Average Site Minimum/Maximum
o —
4thQuartile
Ave
1400
;rage
16-21
-------
Observations from Figures 16-6 through 16-14 include the following:
• Figure 16-6 shows that S4MO's annual average acetaldehyde concentration is
more than twice the program-level average. Recall from the previous section that
this site has the highest annual average concentration of acetaldehyde among sites
sampling carbonyl compounds. The maximum concentration measured at S4MO
is the highest concentration measured across the program. The minimum
acetaldehyde concentration measured at S4MO is greater than the program-level
first quartile (25th percentile).
• Figure 16-7 shows that S4MO's annual average arsenic (PMio) concentration is
greater than the program-level average for arsenic (PMio). Recall from the
previous section that this site has the highest annual average arsenic concentration
among sites sampling metals. The maximum concentration measured at S4MO is
the highest concentration measured across the program. There were no non-
detects of arsenic measured at S4MO.
• Figure 16-8 for benzene shows that the annual average benzene concentration for
S4MO is just greater than the program-level average concentration. The
maximum benzene concentration measured at S4MO is well below the maximum
concentration measured at the program level. There were no non-detects of
benzene measured at S4MO.
• Figure 16-9 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
S4MO is greater than the program-level average concentration. Figure 16-9 also
shows that the maximum concentration measured at S4MO is well below the
maximum concentration measured across the program. Several non-detects of
benzo(a)pyrene were measured at S4MO.
• Figure 16-10 for 1,3-butadiene shows that the annual average concentration for
S4MO is greater than the program-level average concentration. Although the
maximum concentration measured at S4MO is less than the maximum
concentration measured across the program, it is among the highest 1,3-butadiene
concentrations measured across the program. There were no non-detects of
1,3-butadiene measured at S4MO.
• Figure 16-11 for formaldehyde shows that the annual average concentration for
S4MO is just greater than the program-level average concentration. The
maximum formaldehyde concentration measured at S4MO is well below the
maximum concentration measured across the program. There were no non-detects
of formaldehyde measured at S4MO.
16-22
-------
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 16-12 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 16-12 shows the annual average concentration of hexavalent chromium for
S4MO is just less than the program-level average concentration and that both are
greater than the program-level third quartile. The maximum concentration
measured at S4MO is well below the program-level maximum concentration.
There were several non-detects of hexavalent chromium measured at S4MO.
• Figure 16-13 shows that S4MO's annual average manganese (PMio)
concentration is greater than the program-level average for manganese (PMio).
Recall from the previous section that this site has the highest annual average
manganese concentration among sites sampling metals. The maximum
concentration measured at S4MO is the highest concentration measured across the
program. There were no non-detects of manganese measured at S4MO.
• Figure 16-14 shows that the annual naphthalene average for S4MO is greater than
both the program-level average concentration and third quartile. Although the
maximum naphthalene concentration measured at S4MO is well below the
program-level maximum concentration, this concentration is among the top
1 percent of concentrations measured across the program. There were no non-
detects of naphthalene measured at S4MO.
16.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. S4MO has sampled VOC and carbonyl compounds under the NMP since 2002,
PMio metals since 2003, and hexavalent chromium since 2005. Thus, Figures 16-15 through
16-21 present the 3-year rolling statistical metrics for acetaldehyde, arsenic, benzene,
1,3-butadiene, formaldehyde, hexavalent chromium, and manganese (respectively) for S4MO.
The statistical metrics presented for assessing trends include the substitution of zeros for non-
detects.
16-23
-------
Figure 16-15. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at S4MO
~m
.entration ;i.; r,
H
1
5
^H
1 Z i ' ± ' ' ± '
2003-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three- Year Period
• 5th Percentile — Minimum - Median — Maximum • 95th Percentile ...*.. Average
Figure 16-16. Three-Year Rolling Statistical Metrics for Arsenic (PMio) Concentrations
Measured at S4MO
r
° 20
2003-2005 2004-2006 2005-2007 2006-2008
Three-Year Period
SthPercentile — Minimum - Median — Maximum • 95th Percentile
• Average
'Metals sampling at S4MO began in July 2003.
16-24
-------
Figure 16-17. Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured
at S4MO
m"
1,
s 4
I
4
•
T _
^ ^»—
1 l i i —t-1 i
2003-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Percentile — Minimum - Median — Maximum • 95th Percentile ...*.. Average
Figure 16-18. Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations
Measured at S4MO
2003-2005 2004-2006 2005-2007 2006-2008
Three-Year Period
— Minimum — Median
— Maximum
95th Percentile
16-25
-------
Figure 16-19. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at S4MO
1
I
s
5.000
1 I I
~^~ -^ I
1 ' I*' I*' A ' *
2003-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Percentile — Minimum - Median — Maximum • 95th Percentile .-.*.. Average
Figure 16-20. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at S4MO
Concentration (ng/m3)
0.05
7005-2007
• 5th Pertenttle
— Minimum
2006-2008
Three-Year Period
•Ufa*
2007-2009
— Median — Maximum
• 95th Percentile
to
2008-2010
•••*•• Average
16-26
-------
Figure 16-21. Three-Year Rolling Statistical Metrics for Manganese (PMi0) Concentrations
Measured at S4MO
~ 400
2005-2007 2006-2008
Three-Year Period
5th Percentile — Minimum - Median - Maximum • 95thPercentile ...4.. Average
Metals sampling at S4MO began in July 2003.
Observations from Figure 16-15 for acetaldehyde include the following:
• Because carbonyl compound sampling did not begin until December 2002, 2002 data
were excluded from this analysis.
• The maximum acetaldehyde concentration was measured in 2004 and is more than
twice the next highest concentration (measured in 2007).
• The rolling average concentration has fluctuated between approximately 2.70 |ig/m3
and 3.25 |ig/m3 across the period of sampling and corresponding confidence intervals
confirm that a significant increasing or decreasing trend is not apparent.
• Even though the maximum concentration has decreased over the sampling period, the
variability of the concentrations measured has increased, as indicated by the
increasing spread between the 5th and 95th percentiles.
• Note that the minimum concentration measured is greater than zero for all 3-year
periods, indicating that there were no non-detects reported for acetaldehyde since the
onset of sampling.
16-27
-------
Observations from Figure 16-16 for arsenic include the following:
• S4MO began sampling metals in July 2003, as denoted in Figure 16-16.
• The maximum arsenic concentration was measured on December 26, 2007, and
therefore affects the 2005-2007, 2006-2008, and 2007-2009 time frames. The
maximum arsenic concentration has varied significantly by year, ranging from
3.39 ng/m3 in 2006 to 44.1 ng/m3 in 2007.
• Several of the statistical parameters exhibit a slight decreasing trend since the onset of
sampling. However, confidence intervals calculated for the rolling average
concentrations show that the decrease in the rolling averages is not statistically
significant.
Observations from Figure 16-17 for benzene measurements include the following:
• Because VOC sampling did not begin until December 2002, 2002 data was excluded
from this analysis.
• All four benzene concentrations greater than 5 ug/m3 were measured in 2003.
• The rolling average and median concentrations exhibit a decreasing trend over time.
Although both statistical parameters increase slightly for the 2008-2010 time frame,
the increase is not statistically significant.
• The minimum concentration measured is greater than zero for all 3-year periods,
indicating that there were no non-detects reported for benzene since the onset of
sampling.
Observations from Figure 16-18 for 1,3-butadiene include the following:
• The maximum 1,3-butadiene concentration was measured at S4MO in 2003, although
a similar concentration was also measured in 2008. These are the only two
1,3-butaidene concentrations greater than 1.0 ug/m3 that have been measured at
S4MO.
• The rolling average concentrations have fluctuated from approximately 0.8 ug/m3 to
0.9 ug/m3 over the years of sampling.
• The median concentration has remained relatively unchanged since sampling began
for 1,3-butadiene at S4MO, with the exception of the increase following the first
3-year period. During the 2003-2005 sample period, the median concentration was
zero, indicating that at least half of the measurements were non-detects. The
percentage of non-detects has been decreasing over the years of sampling, from as
high as 66 percent in 2004 to zero percent in 2010.
16-28
-------
Observations from Figure 16-19 for formaldehyde include the following:
• The maximum formaldehyde concentration was measured in 2004 and is more than
three times the next highest concentration (also measured in 2004).
• Although difficult to discern in Figure 16-19 due to the relatively high concentration
measured in 2004, both the median and average concentrations exhibit a decreasing
trend. The 95th percentile also exhibits a decrea:
increased for the last three periods of sampling.
trend. The 95th percentile also exhibits a decreasing trend, while the 5th percentile has
• The minimum concentration measured for all 3-year periods is greater than zero,
indicating that there were no non-detects of formaldehyde reported since the onset of
sampling.
Observations from Figure 16-20 for hexavalent chromium include the following:
• The maximum hexavalent chromium concentration was measured on July 5, 2008;
the second and third highest hexavalent chromium concentrations were measured on
July 4, 2006 and July 4, 2010, respectively. These three concentrations support the
potential correlation between hexavalent chromium concentrations and fireworks
discussed in Section 4.1.2. However, the maximum concentration measured in 2010
(0.188 ng/m3) is less than the maximum concentrations measured in 2008
(0.460 ng/m3) and 2006 (0.422 ng/m3).
• The rolling average concentration exhibits a decreasing trend. However, confidence
intervals calculated for the rolling averages indicate that this decrease is not
statistically significant. The confidence intervals, though, are relatively large due to
the rather high maximum concentrations factored into them.
• For each 3-year period shown, both the minimums and 5th percentiles are zero,
indicating the presence of non-detects. The percentage of non-detects has ranged from
16 percent (2007) to 43 percent (2009).
Observations from Figure 16-21 for manganese include the following:
• The maximum manganese concentration was measured on November 26, 2008 and is
nearly twice the next highest concentration (measured in 2004).
• No significant increase or decrease in the rolling average concentrations is shown in
Figure 16-21. Yet, the medians and 5th and 95th percentiles exhibit decreases for
several periods, indicating a general decrease in the majority of concentrations
measured since sampling began in 2003.
16.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the
S4MO monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations regarding
the various risk factors, time frames, and calculations associated with these risk screenings.
16-29
-------
16.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
S4MO monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where available.
As described in Section 3.3, acute risk results from exposures of 1 to 14 days; intermediate risk
results from exposures of 15 to 364 days; and chronic risk results from exposures of 1 year or
greater. The preprocessed daily measurements of the pollutants of interest were compared to the
acute MRL; the quarterly averages were compared to the intermediate MRL; and the annual
averages were compared to the chronic MRL.
Even with all of the relatively high concentrations of the pollutants of interest measured
at S4MO, as discussed in Section 16.4.1, none of the measured detections or time-period average
concentrations of the pollutants of interest for the S4MO monitoring site were greater than their
respective MRL noncancer health risk benchmarks. This is also true for pollutants not identified
as pollutants of interest for S4MO.
16.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the S4MO monitoring site and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 16-6, where applicable.
Table 16-6. Cancer and Noncancer Surrogate Risk Approximations for the Missouri
Monitoring Site
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(jig/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
St. Louis, Missouri - S4MO
Acenaphthene3
Acetaldehyde
0.000088
0.0000022
—
0.009
58/58
54/54
0.01
±<0.01
4.10
±0.59
0.51
9.02
~
0.46
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 16-5.
16-30
-------
Table 16-6. Cancer and Noncancer Surrogate Risk Approximations for the Missouri
Monitoring Site (Continued)
Pollutant
Acrylonitrile
Arsenic (PM10)a
Benzene
Benzo(a)pyrenea
Bery Ilium (PM10)a
1,3 -Butadiene
Cadmium (PM10) a
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Ethylbenzene
Fluorene3
Formaldehyde
Hexavalent Chromium3
Lead(PM10)a
Manganese (PM10) a
Naphthalene3
Nickel (PM10)a
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Cancer
URE
(Hg/m3)1
0.000068
0.0043
0.0000078
0.00176
0.0024
0.00003
0.0018
0.000006
0.000011
0.000026
0.0000025
0.000088
0.000013
0.012
0.000034
0.00048
2.6E-07
0.0000048
0.0000088
Noncancer
RfC
(mg/m3)
0.002
0.000015
0.03
0.00002
0.002
0.00001
0.1
0.098
0.8
2.4
1
0.0098
0.0001
0.00015
0.00005
0.003
0.00009
0.04
0.002
0.1
#of
Measured
Detections
vs. # of
Samples
14/53
60/60
53/53
50/58
59/60
53/53
60/60
53/53
52/53
45/53
10/53
53/53
58/58
54/54
46/57
60/60
60/60
58/58
60/60
51/53
20/53
4/53
Annual
Average
(Hg/m3)
0.17
±0.14
O.01
±<0.01
1.03
±0.15
O.01
±<0.01
0.01
±0.01
0.12
±0.03
0.01
±0.01
0.58
±0.05
0.19
±0.04
0.35
±0.18
0.02
±0.01
0.44
±0.11
0.01
±O.01
2.74
±0.33
O.01
±0.01
0.01
±0.01
0.02
±0.01
0.14
±0.04
O.01
±O.01
0.23
±0.06
0.05
±0.02
0.01
±0.01
Cancer Risk
Approximation
(in-a-million)
11.42
4.39
8.05
0.28
0.02
3.72
1.12
3.48
3.82
0.42
1.10
0.58
35.66
0.40
4.59
0.50
0.06
0.22
0.01
Noncancer
Risk
Approximation
(HQ)
0.08
0.07
0.03
0.01
0.06
0.06
0.01
0.01
O.01
O.01
0.01
0.28
O.01
0.08
0.34
0.05
0.01
0.01
0.02
0.01
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 16-5.
16-31
-------
Observations for S4MO from Table 16-6 include the following:
• The pollutants with the highest annual average concentrations for S4MO are
acetaldehyde, formaldehyde, and benzene.
• Formaldehyde, acrylonitrile, and acetaldehyde have the highest cancer risk
approximations for 2010. The cancer risk approximation for formaldehyde
(35.66 in-a-million) is more than three times higher than the pollutant with the next
highest cancer risk approximation (acrylonitrile at 11.42 in-a-million).
• Note how low the annual average concentration of acrylonitrile is compared to the
annual average concentrations of many of the other pollutants of interest. Yet
acrylonitrile has the second highest cancer risk approximation, indicating the relative
toxicity of this pollutant.
• None of the pollutants of interest for S4MO have noncancer risk approximations
greater than 1.0. The pollutant with the highest noncancer risk approximation is
acetaldehyde (0.46), which is the second highest noncancer risk approximation
calculated for a site-specific pollutant interest among NMP sites.
16.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 16-7 and 16-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 16-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer risk approximations (in-a-million), as calculated from the annual averages. Table 16-8
presents similar information, but identifies the 10 pollutants with the highest noncancer risk
approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 16-7 and 16-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer surrogate risk approximations based on the site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 16.3, S4MO sampled for VOC, PAH, carbonyl compounds, metals (PMio), and
hexavalent chromium. In addition, the cancer and noncancer surrogate risk approximations are
limited to those pollutants with enough data to meet the criteria for annual averages to be
calculated. A more in-depth discussion of this analysis is provided in Section 3.5.5.3.
16-32
-------
Table 16-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Missouri Monitoring Site
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
St. Louis, Missouri (St. Louis City) - S4MO
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Trichloroethylene
Naphthalene
Dichloromethane
POM, Group 2b
Methyl tert butyl ether
148.92
118.03
78.65
74.93
20.98
15.79
13.16
7.23
2.53
0.86
Hexavalent Chromium, PM
Formaldehyde
Arsenic, PM
Benzene
1,3 -Butadiene
Naphthalene
Nickel, PM
POM, Group 3
POM, Group 2b
Ethylbenzene
2.11E-03
1.53E-03
1.49E-03
1.16E-03
6.29E-04
4.47E-04
3.92E-04
3.24E-04
2.22E-04
1.97E-04
Formaldehyde
Acrylonitrile
Acetaldehyde
Benzene
Naphthalene
Arsenic
£>-Dichlorobenzene
1,3 -Butadiene
Carbon Tetrachloride
Cadmium
35.66
11.42
9.02
8.05
4.59
4.39
3.82
3.72
3.48
1.12
-------
Table 16-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Missouri Monitoring Site
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
St. Louis, Missouri (St. Louis City) - S4MO
Toluene
Xylenes
Methanol
Benzene
Formaldehyde
Hexane
Hydrochloric acid
Ethylbenzene
Acetaldehyde
Methyl isobutyl ketone
522.43
358.32
289.83
148.92
118.03
116.82
109.87
78.65
74.93
73.11
Acrolein
Manganese, PM
Arsenic, PM
Formaldehyde
1,3 -Butadiene
Chlorine
Nickel, PM
Acetaldehyde
Trichloroethylene
Lead, PM
316,721.93
34,593.25
23,029.98
12,043.75
10,489.20
9,452.80
9,073.56
8,325.31
7,895.27
6,587.56
Acetaldehyde
Manganese
Formaldehyde
Acrylonitrile
Lead
Arsenic
Cadmium
1,3 -Butadiene
Naphthalene
Benzene
0.46
0.34
0.28
0.08
0.08
0.07
0.06
0.06
0.05
0.03
-------
Observations from Table 16-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in the city of St. Louis.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
cancer UREs) are hexavalent chromium, formaldehyde, and arsenic.
• Six of the highest emitted pollutants also have the highest toxi city-weighted
emissions.
• Four of the pollutants with the highest cancer risk approximations for S4MO also
appear on both emissions-based lists (formaldehyde, benzene, 1,3-butadiene, and
naphthalene). While arsenic is not one of the highest emitted pollutants, it does
appear on the list of highest toxi city-weighted emissions. While acetaldehyde does
not appear on the list of highest toxi city-weighted emissions, it is one of the highest
emitted pollutants in the city of St. Louis. Acrylonitrile, which has the second highest
cancer surrogate risk approximation for S4MO, appears on neither emissions-based
list.
• POM, Group 2b is the ninth highest emitted "pollutant" in St. Louis and ranks ninth
for toxicity-weighted emissions. POM, Group 2b includes several PAH sampled for
at S4MO including acenaphthene and fluorene, which are pollutants of interest for
S4MO. These pollutants are not among those with the 10 highest cancer risk
approximations for S4MO.
Observations from Table 16-8 include the following:
• Toluene, xylenes, and methanol are the highest emitted pollutants with noncancer
RfCs in the city of St. Louis.
• The pollutants with the highest toxicity-weighted emissions (of the pollutants with
noncancer RfCs) are acrolein, manganese, and arsenic. Although acrolein was
sampled for at S4MO, this pollutant was excluded from the pollutants of interest
designation, and thus subsequent risk screening evaluations, due to questions about
the consistency and reliability of the measurements, as discussed in Section 3.2.
• Only two of the highest emitted pollutants in the city of St. Louis also have the
highest toxicity-weighted emissions.
• Acetaldehyde and formaldehyde are the pollutants with the highest and third highest
noncancer risk approximations for S4MO, respectively, and are the only two
pollutants of interest to appear on both emissions-based lists. Manganese, the
pollutant with the second highest noncancer risk approximation, is the pollutant with
the second highest toxicity-weighted emissions but is not one of the highest emitted.
16-35
-------
16.6 Summary of the 2010 Monitoring Data for S4MO
Results from several of the data treatments described in this section include the
following:
»«» Twenty-four pollutants, of which 14 are NA TTS MQO Core Analytes, failed screens
for S4MO.
»«» Acetaldehyde and formaldehyde had the highest annual average concentrations for
S4MO. S4MO had the highest annual average concentration ofacetaldehyde,
arsenic, beryllium, cadmium, lead, and manganese among all NMP sites sampling
these pollutants.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest for S4MO, where they could be
calculated, were greater than their associatedMRL noncancer health risk
benchmarks.
16-36
-------
17.0 Sites in New Jersey
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at UATMP and CSATAM sites in New Jersey, and integrates these
concentrations with emissions, meteorological, and risk information. Data generated by sources
other than ERG are not included in the data analyses contained in this report. Readers are
encouraged to refer to Sections 1 through 4 and the glossary (Appendix P) for detailed
discussions and definitions regarding the various data analyses presented below.
17.1 Site Characterization
This section characterizes the New Jersey monitoring sites by providing geographical and
physical information about the location of the sites and the surrounding areas. This information
is provided to give the reader insight regarding factors that may influence the air quality near the
sites and assist in the interpretation of the ambient monitoring data.
The New Jersey sites are all located within the New York-Northern New Jersey-Long
Island, NY-NJ-PA MSA, although within different divisions. Figures 17-1 through 17-4 are
composite satellite images retrieved from ArcGIS Explorer showing the monitoring sites in their
urban and rural locations. Figures 17-5 through 17-7 identify point source emissions locations by
source category, as reported in the 2008 NEI for point sources. Note that only sources within
10 miles of the sites are included in the facility counts provided in Figures 17-5 through 17-7.
Thus, sources outside the 10-mile radius have been grayed out, but are visible on the maps to
show emissions sources outside the 10-mile boundary. A 10-mile boundary was chosen to give
the reader an indication of which emissions sources and emissions source categories could
potentially have a direct effect on the air quality at the monitoring sites. Further, this boundary
provides both the proximity of emissions sources to the monitoring sites as well as the quantity
of such sources within a given distance of the sites. Table 17-1 describes the area surrounding
each monitoring site by providing supplemental geographical information such as land use,
location setting, and locational coordinates.
17-1
-------
Figure 17-1. Chester, New Jersey (CHNJ) Monitoring Site
to
-------
Figure 17-2. Elizabeth, New Jersey (ELNJ) Monitoring Site
-------
Figure 17-3. New Brunswick, New Jersey (NBNJ) Monitoring Site
-------
Figure 17-4. Paterson, New Jersey (PANJ) Monitoring Site
-------
Figure 17-5. NEI Point Sources Located Within 10 Miles of CHNJ
Note: Due- to facility donsity and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
Legend
@ CHNJ UATMP site 10 mile radius Q ^] County boundary
Source Category Group (No. of Facilities) F Food Processing/Agriculture (1)
+ Aircraft Operations (12) ffl Hospital (1)
i Asphalt Processing/Roofing Manufacturing (1) x Mine/Quarry (1)
o Clay Ceramics Manufacturing (3) M Miscellaneous Manufacturing (2)
• Concrete Batch Plant (1) — Pharmaceutical Manufacturing (1)
© Fabricated Metal Products (1) B Pulp and Paper Plant/Wood Products (2)
17-6
-------
Figure 17-6. NEI Point Sources Located Within 10 Miles of ELNJ and NBNJ
Legend
74*25'(TW 7^20'CTW 74"15"0'W 74'1fl'0"W 7J'5"0"W
Note: Due to facility density and collocation, the total facilities
displayed may nol represent all facilities within the area of Interest.
ELNJ UATMP site
NBNJ UATMP site
10 mile radius
County boundary
Source Category Group (No. of Facilities)
-f< Aircraft Operations (44)
• Asphalt Processing/Roofing Manufacturing (3)
0 Auto Body Shop/Painters (1)
|J Bakery (2)
Brewery /Distillery /Winery (1)
Building Construction (1)
Bulk Terminats/BulX Plants (13)
Chemical Manufacturing (24)
Clay Ceramics Manufacturing (3)
Electricity Generation via Combustion (21)
Electroplating Plating. Polishing. Anodizing, and Coloring (2)
Fabricated Metal Products (12)
Rex.bte PolyureChane foam Production (2)
Food Pracessm ^Agriculture (6)
FumilurePiant{2)
r
ft
B
c
o
*
E
F
I
jf Gasoline/Diesel Service Station 11>
-^- Gypsum Manufacturing {1)
(V Heating Equipment Manufacturing <1}
(3 Hospital (2)
^ Hoi Mix Asptiatt Plant (2)
*- industrial Machinery and Equipment [5]
^ [nstifuiionai - scnoo) i i-D
® Labocatoryd)
A Landfill 42}
/ Lumber/sawmill {1}
V Marine Port n>
> Marine Vessel Loading Rack (t>
X Mine^Quarry (1)
? Miscellaneous Commercial/lrvdustrtal (24)
M Miscellaneous Manufactunng {13)
gg Municipal Vtfaslo Com buster (2)
Oil and/or Gas Production (1)
Petroteum Refinery (3)
Priarmaceulical Manufaclunng (7)
Pnmary Melal Produclran <1)
Printing, Coatirtg & Dyeing of Fabrics (1)
PnntingrPublishingd",
Pulp and Paper Plant/Wood Products (13)
Rubber and Miscellaneous Piastres Products (12)
Secondary Metal P recessing 41)
Solid Waste Disposal - CommerciaL/lnstitutiona) (4)
Steel Mr i (?)
Surface Coaling (16)
Tire Manufacture (1)
Transportation and Marketing o( Petroleum Products (2)
Wastewater Treatment (8}
W Wbodwark. Furniture. MKIwork & Wood Preserving (2)
17-7
-------
Figure 17-7. NEI Point Sources Located Within 10 Miles of PANJ
Legend
74--1 STO'W 74' 10'0"W 74" 5'0"W 74'0'0'W
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
PANJ CSATAM site
10 mile radius
County boundary
Source Category Group (No. of Facilities) ?
•4< Aircraft Operations (21) M
B Bakery (1) •
B Bulk Terminals/Bulk Plants (1)
C Chemical Manufacturing (8) 1
O Clay Ceramics Manufacturing (1) -t
6 Electrical Equipment (4) P
f Electricity Generation via Combustion (4) ffi
E Electroplating. Plating. Polishing, Anodizing, & Coloring (i) R
© Fabricated Metal Products <3) <
^ Flexible Polyurethane Foam Production (3) V
F Food Processing/Agriculture (3) S
GJ Hospital (3) T
$ Hot Mix Asphalt Plant (3) —
•$• industrial Machinery and Equipment (2) '
^ Institutional - school (2) W
Miscellaneous CornmerciaVlrtdustrial (3)
Miscellaneous Manufacturing (20)
Oil and/or Gas Production (1)
Pharmaceutical Manufacturing {4)
Primary Metal Production (3)
Printing, Coaling & Dyeing of Fabric f 1)
Prinling/Publishirtg (t3)
Pulp and Pa pe r PlanLWood Products (10)
Rubber and Miscellaneous Plastics Products (4)
Site Remediation Activity (1)
Steel Mill {1)
Surface Coating (S}
Textile Mill (4)
Transportation Equipment (1)
Wastewater Treatment (1)
Woodwork, Furniture, MiHwork & Wood Preserving (1J
17-8
-------
Table 17-1. Geographical Information for the New Jersey Monitoring Sites
Site
Code
CHNJ
ELNJ
NBNJ
PANJ
AQS Code
34-027-3001
34-039-0004
34-023-0006
34-031-0005
Location
Chester
Elizabeth
New
Brunswick
Paterson
County
Morris
Union
Middlesex
Passaic
Micro- or
Metropolitan
Statistical Area
New York-
Northern New
Jersey-Long
Island, NY-NJ-PA
MSA
(Newark Div)
New York-
Northern New
Jersey-Long
Island, NY-NJ-PA
MSA
(Newark Div)
New York-
Northern New
Jersey-Long
Island, NY-NJ-PA
MSA (Edison Div)
New York-
Northern New
Jersey-Long
Island, NY-NJ-PA
MSA
(New York Div)
Latitude
and
Longitude
40.78763,
-74.6763
40.64144,
-74.20836
40.472786,
-74.42251
40.918381,
-74.168092
Land Use
Agricultural
Industrial
Agricultural
Commercial
Location
Setting
Rural
Suburban
Rural
Urban/City
Center
Additional Ambient Monitoring Information1
SO2, NO, NO2, O3, Meteorological parameters,
PM2 5, PM2 5 Speciation
CO, SO2, NO2, NOX, Meteorological parameters,
PM2 5, PM2 5 Speciation
Meteorological parameters, PM2 5, PM2 5 Speciation
Meteorological Parameters, PM2 5
1 These monitoring sites report additional pollutants to AQS (EPA, 2012h); however, these data are not generated by ERG and are therefore not included in this report.
-------
CHNJ is located in northern New Jersey, west of the New York City metropolitan area.
Figure 17-1 shows that CHNJ is located in an open area near Building 1 on the property of Bell
Labs, which is owned by Alcatel-Lucent. The surrounding area is rural and agricultural with a
rolling topography, but surrounded by small neighborhoods. Although the location is considered
part of the New York City MSA, the site's location is outside most of the urbanized areas.
Figure 17-5 shows that few sources are close to CFINJ and that the source category with the
highest number of emissions sources surrounding CFINJ is the aircraft operations category,
which includes airports as well as small runways, heliports, or landing pads.
ELNJ is located in the city of Elizabeth, which lies just south of Newark and west of
Newark Bay and Staten Island, New York. As Figure 17-2 shows, the monitoring site is located
just off Exit 13 of the New Jersey Turnpike (1-95), near the toll plaza. Interstate-278 intersects
the Turnpike here as well. The surrounding area is highly industrialized, with an oil refinery
located just southwest of the site. Additional industry is located to the west and southwest, while
residential neighborhoods are located to the northwest and north of the site.
NBNJ is located in New Brunswick, less than 20 miles southwest of Elizabeth. The
monitoring site is located on the property of Rutgers University's Cook-Douglass campus, on a
horticultural farm. The surrounding area is agricultural and rural, although residential
neighborhoods are located to the east, across a branch of the Raritan River, as shown in
Figure 17-3. County Road 617 (Ryders Lane) and US-1 intersect just west of the site and 1-95
runs northeast-southwest about 1 mile east of the site.
Figure 17-6 shows that the outer portions of NBNJ and ELNJ's 10-mile radii intersect
and that many emissions sources surround these two sites. The bulk of the emissions sources are
located in northern Middlesex County and northeastward toward New York City and northern
New Jersey. The source categories with the highest number of emissions sources in the vicinity
of these sites include aircraft operations, chemical manufacturing, electricity generation via
combustion, and surface coating. The emissions sources in closest proximity to the ELNJ
monitoring site are in the miscellaneous manufacturing, wastewater treatment, chemical
manufacturing, electricity generation via combustion, and petroleum refining categories. The
emissions sources in closest proximity to the NBNJ monitoring site are involved in aircraft
operations and pharmaceutical manufacturing.
17-10
-------
PANJ is located in northern New Jersey, in the town of Paterson, north of Newark and
between Clifton and Hackensack. The monitoring site is located at the local health department
with residential areas to the east and commercial areas to the west, as shown in Figure 17-4. The
Passaic River runs northeast-southwest just north of PANJ and is shown in the upper left corner
of Figure 17-4. Interstate-80 runs east-west less than 1 mile south of PANJ. Figure 17-7 shows
that the majority of point sources within 10 miles of PANJ are located to the southwest of the
site. Although the majority of sources near PANJ are involved in aircraft operations, printing and
publishing, or pulp and paper products, the source closest to PANJ falls in the miscellaneous
industries category.
Table 17-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the areas surrounding the New
Jersey monitoring sites. County-level vehicle registration data for Union, Morris, Passaic, and
Middlesex Counties were not available from the State of New Jersey. Thus, state-level vehicle
registration, which was obtained from the Federal Highway Administration (FFIWA, 2011), was
allocated to the county level using the county-level proportion of the state population. State-level
and county-level population information for these counties was obtained from the U.S. Census
Bureau. Table 17-2 also includes a vehicle registration-to-county population ratio (vehicles-per-
person) for each site. In addition, the population within 10 miles of each site is presented. An
estimate of 10-mile vehicle ownership was calculated by applying the county-level vehicle
registration-to-population ratio to the 10-mile population surrounding each monitoring site.
Table 17-2 also contains annual average daily traffic information. Finally, Table 17-2 presents
the daily VMT for Middlesex, Morris, Passaic, and Union Counties.
Observations from Table 17-2 include the following:
• Middlesex County, where NBNJ is located, has the highest county-level population of
the New Jersey sites. But ELNJ has the highest 10-mile population among the four
New Jersey sites. The 10-mile populations for both ELNJ and PANJ are greater than
1 million people.
• Compared to NMP monitoring sites in other locations, the county-level populations
are in the middle of the range. However, ELNJ has one of the highest 10-mile
populations, ranking fourth among NMP sites while PANJ rounds out the top 10.
While NBNJ's 10-mile population is in the middle of the range, CHNJ's 10-mile
population is in the bottom third compared to other NMP sites.
17-11
-------
Table 17-2. Population, Motor Vehicle, and Traffic Information for the New Jersey
Monitoring Sites
Site
CHNJ
ELNJ
NBNJ
PANJ
Estimated
County
Population1
492,694
537,661
810,986
501,860
County-level
Vehicle
Registration2
389,359
424,894
640,893
396,602
Vehicles per
Person
(Registration:
Population)
0.79
0.79
0.79
0.79
Population
within 10
miles3
244,577
2,180,662
783,724
1,332,800
Estimated
10-mile
Vehicle
Ownership
193,281
1,723,298
619,349
1,053,264
Annual
Average
Daily
Traffic4
12,917
250,000
114,322
22,272
County-
level Daily
VMT5
14,256,044
12,485,902
20,415,685
8,178,167
1 County-level population estimates reflect data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects ratios based on 2010 state-level vehicle registration data from the
FHWA and the county-level proportion of the state population data (FHWA, 2011)
310-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2006 data from for ELNJ from NJ Department of Treasury and 2010 data
from the New Jersey DOT for the other sites (Steer, 2008 and NJ DOT, 2010a)
5 County-level VMT reflects 2010 data from the New Jersey DOT (NJ DOT, 2010b)
• The estimated county-level vehicle ownership is highest for NBNJ while the vehicle
ownerships across the remaining New Jersey sites are fairly similar to each other. The
county-level registration estimates are in the middle of the range compared to other
NMP sites. ELNJ and PANJ have two of the highest 10-mile vehicle ownership
estimates compared to other NMP sites.
• ELNJ experiences a significantly higher average traffic volume than other New
Jersey sites, while CHNJ experiences the least. Traffic data for ELNJ were obtained
for 1-95, between Exit 13 and 13 A; this is the highest traffic volume among all NMP
sites. Traffic data for CHNJ were obtained for Main Street (County Road 513) near
Highway 206 in downtown Chester; traffic data for NBNJ were obtained for US-1
near State Road 617 (Ryders Lane); and traffic data for PANJ were obtained for
Memorial Drive between Ellison Street and College Boulevard.
• Among the New Jersey counties with monitoring sites, VMT for Middlesex County is
highest while Passaic County is the lowest. However, county-level VMT for the New
Jersey counties are in the middle of the range compared to other counties with NMP
sites (where VMT data were available).
17.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in New Jersey on sample days, as well as over the course of the year.
17-12
-------
17.2.1 Climate Summary
Frontal systems push across the state of New Jersey regularly, producing variable
weather. The state's proximity to the Atlantic Ocean has a moderating effect on temperature.
Summers along the coast tend to be cooler than areas farther inland, while winters tend to be
warmer. Large urban areas within the state experience the urban heat island effect, in which
urban areas retain more heat than outlying areas. New Jersey's mid-Atlantic location also allows
for ample annual precipitation and relatively high humidity. A southwesterly wind is most
common in the summer and a northwesterly wind is typical in the winter. Winds from the west
and northwest result in air masses that dry out, stabilize, and warm as they move eastward from
higher elevations to sea level (Bair, 1992 and Rutgers, 2012).
17.2.2 Meteorological Conditions in 2010
Hourly meteorological data from NWS weather stations nearest these sites were retrieved
for 2010 (NCDC, 2010). The three closest weather stations are located at Somerville-Somerset
Airport (near CHNJ and NBNJ), Newark International Airport (near ELNJ), and Essex County
Airport (near PANJ), WBAN 54785, 14734, and 54743, respectively. Additional information
about these weather stations, such as the distance between the sites and the weather stations, is
provided in Table 17-3. These data were used to determine how meteorological conditions on
sample days vary from normal conditions throughout the year.
Table 17-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 17-3 is the
95 percent confidence interval for each parameter. As shown in Table 17-3, average
meteorological conditions on sample days were representative of average weather conditions
throughout the year for CHNJ, ELNJ, and NBNJ. It appears that sample days at PANJ were
warmer and wetter than for the entire year as a whole. However, sampling did not begin at PANJ
until April 2010, thereby missing the coldest months of the year.
17-13
-------
Table 17-3. Average Meteorological Conditions near the New Jersey Monitoring Sites
Closest NWS Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Chester, New Jersey - CHNJ
Somerville, New
Jersey/Somerset
Airport
54785
(40.62, -74.67)
11.30
miles
165°
(SSE)
Sample
Day
2010
64.1
±5.2
64.0
±2.1
53.6
±4.6
53.5
±1.8
42.5
±4.9
42.1
±2.0
48.2
±4.3
48.0
±1.7
69.7
±3.3
69.3
±1.3
1012.9
±1.9
1013.3
±0.7
3.3
±0.5
3.6
±0.3
Elizabeth, New Jersey - ELNJ
Newark International
Airport
14734
(40.68, -74.17)
3.45
miles
20°
(NNE)
Sample
Day
2010
64.9
±5.1
65.0
±2.0
57.2
±4.7
57.2
±1.9
40.8
±4.8
40.2
± 1.9
49.2
±4.2
49.0
±1.7
57.6
±3.6
56.6
± 1.5
1013.1
±1.9
1013.4
±0.7
8.4
±0.8
8.5
±0.4
New Brunswick, New Jersey - NBNJ
Somerville, New
Jersey/Somerset
Airport
54785
(40.62, -74.67)
16.06
miles
297°
(WNW)
Sample
Day
2010
64.1
±5.2
64.0
±2.1
53.6
±4.6
53.5
±1.8
42.5
±4.9
42.1
±2.0
48.2
±4.3
48.0
±1.7
69.7
±3.3
69.3
±1.3
1012.9
± 1.9
1013.3
±0.7
o o
J.J
±0.5
3.6
±0.3
Passaic, New Jersey - PANJ
Essex County
Airport
54743
(40.88, -74.28)
6.39
miles
229°
(SW)
Sample
Day
2010
70.3
±6.5
63.3
±2.0
61.5
±5.8
54.3
± 1.9
49.6
±5.7
40.9
±2.0
55.0
±5.2
47.9
± 1.7
68.3
±5.0
64.2
±1.5
1013.6
±2.7
1014.2
±0.7
3.4
±0.7
4.0
±0.3
1 Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
17.2.3 Back Trajectory Analysis
Figure 17-8 is the composite back trajectory map for days on which samples were
collected at the CHNJ monitoring site in 2010. Included in Figure 17-8 are four back trajectories
per sample day. Figure 17-9 is the corresponding cluster analysis for 2010. Similarly,
Figures 17-10 through 17-15 are the composite back trajectory maps and corresponding cluster
analyses for the remaining New Jersey monitoring sites. An in-depth description of these maps
and how they were generated is presented in Section 3.5.2.1. For the composite maps, each line
represents the 24-hour trajectory along which a parcel of air traveled toward the monitoring site
on a given sample day and time. For the cluster analyses, each line corresponds to a back
trajectory representative of a given cluster of trajectories. For all maps, each concentric circle
around the sites in Figures 17-8 through 17-15 represents 100 miles.
Observations from Figures 17-8 through 17-15 include the following:
• Due to their relatively close proximity to each other and the standardization of sample
days, the back trajectories shown on each composite back trajectory map for the New
Jersey sites are similar to each other. The composite back trajectory map for PANJ
includes one-third fewer back trajectories as this site did not begin sampling until
April 2010.
• Back trajectories originated from a variety of directions at the sites, although fewer
from the east and southeast. In general, trajectories originating from the south, west,
or north were longer than trajectories originating from the east.
• The 24-hour air shed domains for the New Jersey sites were similar in size to each
other. Back trajectories greater than 600 miles in length originated near Lake
Michigan, off the coast of North Carolina and South Carolina, or over the Atlantic
Ocean south of Newfoundland, Canada. The average trajectory length for these sites
ranged from 267 miles (PANJ) to 279 miles (ELNJ).
• The cluster trajectories for the New Jersey sites are similar to each other, although the
percentages vary. The cluster maps show a propensity for trajectories to originate
from the northwest quadrant, including west and north, at these sites, although of
varying lengths. Trajectories also originated from the south and east, although the
trajectories with an easterly component tended to be relatively short.
17-15
-------
Figure 17-8. 2010 Composite Back Trajectory Map for CHNJ
Figure 17-9. Back Trajectory Cluster Map for CHNJ
17-16
-------
Figure 17-10. 2010 Composite Back Trajectory Map for ELNJ
/ iii
1 '
/ ill
Figure 17-11. Back Trajectory Cluster Map for ELNJ
17-17
-------
Figure 17-12. 2010 Composite Back Trajectory Map for NBNJ
\f^ \ \
'' \ \
\ \ \ \
\ \ ', \ \ >
Figure 17-13. Back Trajectory Cluster Map for NBNJ
17-18
-------
Figure 17-14. 2010 Composite Back Trajectory Map for PANJ
Figure 17-15. Back Trajectory Cluster Map for PANJ
17-19
-------
17.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather stations nearest the New Jersey sites, as
presented in Section 17.2.2, were uploaded into a wind rose software program to produce
customized wind roses, as described in Section 3.5.2.2. A wind rose shows the frequency of wind
directions using "petals" positioned around a 16-point compass, and uses different colors to
represent wind speeds.
Figure 17-16 presents three different wind roses for the CHNJ monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figures 17-17 through
17-19 present the three wind roses and distance maps for ELNJ, NBNJ, and PANJ, respectively.
17-20
-------
Figure 17-16. Wind Roses for the Summerville-Somerset Airport Weather Station near
CHNJ
1999-2009 Historical Wind Rose
2010 Wind Rose
WIND SPEED
(Knots)
^| 17-21
^| 11 - 17
• IT
IH 2- 4
Calms: 41.31%
2010 Sample Day Wind Rose
Distance between CHNJ and NWS Station
m~\
17-21
-------
Figure 17-17. Wind Roses for the Newark International Airport Weather Station near
ELNJ
1999-2009 Historical Wind Rose
2010 Wind Rose
,'VEST
2010 Sample Day Wind Rose
Distance between ELNJ and NWS Station
«& r
„••""
":
Eliz«t«ltl
/ ,
•v * ' . 1
• / \\
*""« *
A
'//
,- % ^
=»...,::
,J.,.
[^ f
17-22
-------
Figure 17-18. Wind Roses for the Summerville-Somerset Airport Weather Station near
NBNJ
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between NBNJ and NWS Station
17-23
-------
Figure 17-19. Wind Roses for the Essex County Airport Weather Station near PANJ
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between PANJ and NWS Station
X
.-.. .. .,
/
--. / ,
- / ,
17-24
-------
Observations from Figures 17-16 and 17-18 for CHNJ and NBNJ include the following:
• The NWS weather station at Somerville/Somerset Airport is the closest weather
station to both CHNJ and NBNJ. The Somerville/Somerset Airport weather station is
located approximately 11.3 miles south-southeast of CHNJ and 16.1 miles
west-northwest of NBNJ.
• The wind data for the historical and full-year wind roses for CHNJ and NBNJ are the
same because they are from the same weather station; thus, the wind roses are
identical.
• The historical wind roses for these sites show that calm winds accounted for greater
than 40 percent of observations. For wind speeds greater than 2 knots, northerly
winds were observed most frequently, while winds from the southwest quadrant were
rarely observed.
• Calm winds account for more than 40 percent of the wind observations throughout
2010 and on sample days in 2010. Winds from the northwest quadrant, including
northerly and westerly winds, account for another one-third of wind observations
throughout 2010 and on sample days. Winds on sample days resemble the full-year
wind patterns, indicating that conditions in 2010 were similar to conditions
experienced near these sites over the course of 2010.
• While the 2010 wind roses and 2010 sample day wind roses do exhibit the same
prevalence for calm winds as the historical wind roses, they do not exhibit the same
northerly predominance for wind speeds greater than 2 knots. Instead, there was an
increase in winds from the northwest quadrant. A similar observation is made for
2009 in the 2008-2009 NMP report.
Observations from Figure 17-17 for ELNJ include the following:
• The Newark International Airport weather station is located approximately 3.5 miles
north-northeast of ELNJ.
• The historical wind rose shows that winds from a variety of directions were observed
near ELNJ, although easterly and southeasterly winds were observed less frequently.
Calm winds were observed for just less than six percent of observations. The
strongest winds were associated with westerly and northwesterly winds.
• The wind patterns shown on the 2010 wind rose generally resemble the historical
wind patterns, as do the sample day wind patterns for 2010, although the percentages
vary slightly across the wind directions. This indicates that conditions on sample days
were representative of those experienced over the entire year and historically.
Observations from Figure 17-19 for PANJ include the following:
• The Essex County Airport weather station is located approximately 6.4 miles
southwest of PANJ.
17-25
-------
• The historical wind rose shows that calm winds account for approximately one-third
of the wind observations near PANJ. Winds from the western quadrants account for
the bulk of winds greater than 2 knots, particularly winds from the west-northwest to
northwest. The strongest winds were associated with westerly to northwesterly winds.
• The 2010 wind rose shows that calm winds accounted for nearly 40 percent of wind
observations in 2010 and that west-northwesterly to north-northwesterly winds
account for the bulk of wind observations greater than 2 knots. This represents a
northward shift in the predominant wind direction from the historical wind patterns
near PANJ.
• The sample day wind rose for PANJ exhibits several differences from the historical
and full-year wind roses. The sample day wind rose has an even higher percentage of
calm winds. There is also a higher percentage of winds from the northeast to east and
fewer from the west and northwest. This wind rose likely reflects a seasonal pattern
and is the result of the exclusion of wind observations from the first quarter (and part
of April) 2010 to correspond with the sample period.
17.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the New Jersey monitoring sites
in order to allow analysts and readers to focus on a subset of pollutants through the context of
risk. For each site, each pollutant's preprocessed daily measurement was compared to its
associated risk screening value. If the concentration was greater than the risk screening value,
then the concentration "failed the screen." Pollutants of interest are those for which the
individual pollutant's total failed screens contribute to the top 95 percent of the site's total failed
screens. In addition, if any of the NATTS MQO Core Analytes measured by each monitoring site
did not meet the pollutant of interest criteria based on the preliminary risk screening, that
pollutant was added to the list of site-specific pollutants of interest. A more in-depth description
of the risk screening process is presented in Section 3.2.
Table 17-4 presents the pollutants of interest for the New Jersey sites. The pollutants that
failed at least one screen and contributed to 95 percent of the total failed screens for each
monitoring site are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of interest
are shaded and/or bolded. All three UATMP sites sampled for VOC and carbonyl compounds
while PANJ sampled for VOC only.
17-26
-------
Table 17-4. Risk Screening Results for the New Jersey Monitoring Sites
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Chester, New Jersey - CHNJ
Formaldehyde
Acetaldehyde
Benzene
Carbon Tetrachloride
Acrylonitrile
1 ,2-Dichloroethane
1,3-Butadiene
1 ,2-Dibromoethane
£>-Dichlorobenzene
Chloromethylbenzene
Hexachloro- 1 ,3 -butadiene
1 , 1 ,2,2-Tetrachloroethane
0.077
0.45
0.13
0.17
0.015
0.038
0.03
0.0017
0.091
0.02
0.045
0.017
Total
59
58
57
57
16
14
8
2
2
1
1
1
276
59
59
57
57
16
14
17
2
12
1
1
1
296
100.00
98.31
100.00
100.00
100.00
100.00
47.06
100.00
16.67
100.00
100.00
100.00
93.24
21.38
21.01
20.65
20.65
5.80
5.07
2.90
0.72
0.72
0.36
0.36
0.36
21.38
42.39
63.04
83.70
89.49
94.57
97.46
98.19
98.91
99.28
99.64
100.00
Elizabeth, New Jersey - ELNJ
Acetaldehyde
Benzene
1,3-Butadiene
Formaldehyde
Carbon Tetrachloride
Ethylbenzene
£>-Dichlorobenzene
1 ,2-Dichloroethane
Acrylonitrile
Propionaldehyde
1 ,2-Dibromoethane
Chloroprene
1 , 1 ,2,2-Tetrachloroethane
Trichloroethylene
0.45
0.13
0.03
0.077
0.17
0.4
0.091
0.038
0.015
0.8
0.0017
0.0021
0.017
0.2
Total
59
59
59
59
58
32
25
11
10
10
2
1
1
1
387
59
59
59
59
58
59
44
11
10
59
2
1
1
22
503
100.00
100.00
100.00
100.00
100.00
54.24
56.82
100.00
100.00
16.95
100.00
100.00
100.00
4.55
76.94
15.25
15.25
15.25
15.25
14.99
8.27
6.46
2.84
2.58
2.58
0.52
0.26
0.26
0.26
15.25
30.49
45.74
60.98
75.97
84.24
90.70
93.54
96.12
98.71
99.22
99.48
99.74
100.00
New Brunswick, New Jersey - NBNJ
Acetaldehyde
Formaldehyde
Benzene
Carbon Tetrachloride
1,3-Butadiene
Acrylonitrile
1 ,2-Dichloroethane
£>-Dichlorobenzene
Ethylbenzene
1 ,2-Dibromoethane
Hexachloro- 1 ,3 -butadiene
1 , 1 ,2,2-Tetrachloroethane
0.45
0.077
0.13
0.17
0.03
0.015
0.038
0.091
0.4
0.0017
0.045
0.017
Total
58
58
55
54
46
23
11
9
3
1
1
1
320
58
58
55
54
53
23
11
33
55
1
1
1
403
100.00
100.00
100.00
100.00
86.79
100.00
100.00
27.27
5.45
100.00
100.00
100.00
79.40
18.13
18.13
17.19
16.88
14.38
7.19
3.44
2.81
0.94
0.31
0.31
0.31
18.13
36.25
53.44
70.31
84.69
91.88
95.31
98.13
99.06
99.38
99.69
100.00
17-27
-------
Table 17-4. Risk Screening Results for the New Jersey Monitoring Sites (Continued)
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Paterson, New Jersey - PANJ
Benzene
1,3-Butadiene
Carbon Tetrachloride
p-Dichlorobenzene
Ethylbenzene
1 ,2-Dichloroethane
1 ,2-Dibromoethane
Trichloroethylene
0.13
0.03
0.17
0.091
0.4
0.038
0.0017
0.2
Total
21
21
21
21
17
2
1
1
105
21
21
21
21
21
2
1
8
116
100.00
100.00
100.00
100.00
80.95
100.00
100.00
12.50
90.52
20.00
20.00
20.00
20.00
16.19
1.90
0.95
0.95
20.00
40.00
60.00
80.00
96.19
98.10
99.05
100.00
Observations from Table 17-4 include the following:
• Twelve pollutants failed at least one screen for CHNJ, of which five are NATTS
MQO Core Analytes; 14 failed screens for ELNJ (six are NATTS MQO Core
Analytes); 12 failed screens for NBNJ (five are NATTS MQO Core Analytes); and
eight failed screens for PANJ (four are NATTS MQO Core Analytes).
• The risk screening process identified seven pollutants of interest for CHNJ (of which
five are NATTS MQO Core Analytes). Chloroform, tetrachloroethylene, and
trichloroethylene were added as pollutants of interest because they are also NATTS
MQO Core Analytes, even though they did not fail any screens. These three
pollutants are not shown in Table 17-4. Vinyl chloride is also a NATTS MQO Core
Analyte, but was not added as a pollutant of interest because it was not detected at
CHNJ.
• The risk screening process identified 10 pollutants of interest for ELNJ (of which five
are NATTS MQO Core Analytes). Trichloroethylene was added as a pollutant of
interest because it is a NATTS MQO Core Analyte, even though it did not contribute
to 95 percent of failed screens. Chloroform, tetrachloroethylene, and vinyl chloride
were also added because they are NATTS MQO Core Analytes, even though they did
not fail any screens. These three pollutants are not shown in Table 17-4.
• The risk screening process identified seven pollutants of interest for NBNJ (of which
five are NATTS MQO Core Analytes). Chloroform, tetrachloroethylene,
trichloroethylene, and vinyl chloride were added as pollutants of interest because they
are also NATTS MQO Core Analytes, even though they did not fail any screens.
These four pollutants are not shown in Table 17-4.
• The risk screening process identified five pollutants of interest for PANJ (of which
three are NATTS MQO Core Analytes). Trichloroethylene was added as a pollutant
of interest because it is a NATTS MQO Core Analyte, even though it did not
contribute to 95 percent of failed screens. Chloroform, tetrachloroethylene, and vinyl
chloride were also added as pollutants of interest because they are also NATTS MQO
17-28
-------
Core Analytes, even though they did not fail any screens. These three pollutants are
not shown in Table 17-4.
• The total failure rate ranged from 76.94 percent for ELNJ to 93.24 percent for CHNJ
(of the pollutants with at least one failed screen).
17.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the New Jersey monitoring sites. Concentration averages are provided for the pollutants of
interest for each New Jersey site, where applicable. Concentration averages for select pollutants
are also presented graphically for each site, where applicable, to illustrate how each site's
concentrations compare to the program-level averages. In addition, concentration averages for
select pollutants are presented from previous years of sampling in order to characterize
concentration trends at each site, where applicable. Additional site-specific statistical summaries
are provided in Appendices J and L.
17.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each New Jersey site, as described in Section 3.1. The quarterly average of a particular
pollutant is simply the average concentration of the preprocessed daily measurements over a
given calendar quarter. Quarterly average concentrations include the substitution of zeros for all
non-detects. A site must have a minimum of 75 percent valid samples of the total number of
samples possible within a given quarter for a quarterly average to be calculated. An annual
average includes all measured detections and substituted zeros for non-detects for the entire year
of sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the New Jersey
monitoring sites are presented in Table 17-5, where applicable. Note that if a pollutant was not
detected in a given calendar quarter, the quarterly average simply reflects "0" because only zeros
substituted for non-detects were factored into the quarterly average concentration.
17-29
-------
Table 17-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the New Jersey Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Ug/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(Ug/m3)
4th
Quarter
Average
(Ug/m3)
Annual
Average
(Ug/m3)
Chester, New Jersey - CHNJ
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
Trichloroethylene
59/59
16/57
57/57
17/57
57/57
48/57
14/57
59/59
39/57
4/57
1.72
±0.43
0.01
±0.02
0.51
±0.09
<0.01
±0.01
0.66
±0.08
0.04
±0.02
0.04
±0.02
0.93
±0.32
0.03
±0.02
0
1.76
±0.36
0.03
±0.03
0.48
±0.1
0.01
±0.01
0.66
±0.07
0.11
±0.01
0.03
±0.02
2.20
±0.71
0.09
±0.02
<0.01
±0.01
0.90
±0.21
0.06
±0.05
0.41
±0.05
0.01
±0.01
0.65
±0.06
0.13
±0.02
0
2.29
±0.76
0.10
±0.05
0.01
±0.01
0.83
±0.21
0.04
±0.03
0.50
±0.07
0.02
±0.01
0.60
±0.05
0.07
±0.02
0
1.19
±0.32
0.06
±0.03
0
1.31
±0.19
0.04
±0.02
0.48
±0.04
0.01
±<0.01
0.64
±0.03
0.08
±0.01
0.02
±0.01
1.64
±0.31
0.07
±0.02
<0.01
±<0.01
Elizabeth, New Jersey - ELNJ
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
£>-Dichlorobenzene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Ethylbenzene
Formaldehyde
Propionaldehyde
Tetrachloroethylene
59/59
10/59
59/59
59/59
44/59
58/59
51/59
11/59
59/59
59/59
59/59
58/59
2.27
±0.97
0.03
±0.03
0.87
±0.14
0.11
±0.03
0.04
±0.03
0.60
±0.12
0.06
±0.04
0.03
±0.02
0.28
±0.07
3.83
±1.73
0.40
±0.21
0.15
±0.05
3.62
±0.87
0.01
±0.03
1.18
±0.28
0.14
±0.02
0.11
±0.03
0.59
±0.06
0.16
±0.02
0.03
±0.02
0.51
±0.08
5.64
±1.35
0.78
±0.19
0.22
±0.05
2.75
±0.53
0.02
±0.03
1.02
±0.38
0.10
±0.01
0.13
±0.03
0.61
±0.06
0.19
±0.03
0
0.49
±0.07
4.82
±1.06
0.59
±0.13
0.25
±0.07
2.24
±0.61
0.02
±0.02
1.01
±0.15
0.14
±0.03
0.06
±0.04
0.60
±0.09
0.10
±0.03
0
0.39
±0.07
3.47
±0.71
0.38
±0.09
0.18
±0.05
2.73
±0.39
0.02
±0.01
1.02
±0.13
0.12
±0.01
0.09
±0.02
0.60
±0.04
0.13
±0.02
0.02
±0.01
0.42
±0.04
4.46
±0.64
0.54
±0.09
0.20
±0.03
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
17-30
-------
Table 17-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the New Jersey Monitoring Sites (Continued)
Pollutant
Trichloroethylene
Vinyl Chloride
#of
Measured
Detections
vs. # of
Samples
22/59
2/59
1st
Quarter
Average
(Ug/m3)
0.02
±0.03
<0.01
±<0.01
2nd
Quarter
Average
(Ug/m3)
0.04
±0.02
<0.01
±<0.01
3rd
Quarter
Average
(Ug/m3)
0.05
±0.03
0
4th
Quarter
Average
(Ug/m3)
0.01
±0.01
0
Annual
Average
(Ug/m3)
0.03
±0.01
O.01
±O.01
New Brunswick, New Jersey - NBNJ
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
58/58
23/55
55/55
53/55
54/55
49/55
11/55
58/58
52/55
18/55
6/55
2.15
±0.45
0.04
±0.04
0.72
±0.10
0.05
±0.02
0.56
±0.17
0.07
±0.06
0.03
±0.02
1.86
±0.34
0.10
±0.03
<0.01
±0.01
0
3.65
±0.77
0.11
±0.07
0.55
±0.07
0.04
±0.01
0.54
±0.06
0.12
±0.01
0.03
±0.02
2.36
±0.54
0.12
±0.02
0.02
±0.02
0.01
±0.01
3.42
±0.80
0.35
±0.39
0.59
±0.08
0.04
±0.01
0.61
±0.06
0.18
±0.03
0.01
±0.01
1.30
±0.32
0.14
±0.03
0.03
±0.02
0.01
±0.01
2.47
±0.77
0.05
±0.04
0.75
±0.12
0.07
±0.02
0.53
±0.10
0.12
±0.02
0
1.01
±0.14
0.13
±0.04
0.02
±0.02
0.01
±0.01
2.92
±0.37
0.14
±0.11
0.65
±0.05
0.05
±0.01
0.56
±0.05
0.12
±0.02
0.02
±0.01
1.63
±0.22
0.12
±0.01
0.02
±0.01
0.01
±0.01
Paterson, New Jersey - PANJ
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
Ethylbenzene
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
21/21
21/21
21/21
18/21
21/21
21/21
20/21
8/21
2/21
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
1.15
±0.27
0.17
±0.05
0.63
±0.16
0.21
±0.09
0.33
±0.14
0.75
±0.16
0.26
±0.19
0.04
±0.04
0
1.66
±0.33
0.32
±0.09
0.61
±0.09
0.20
±0.12
0.30
±0.10
0.65
±0.20
0.41
±0.16
0.04
±0.06
O.01
±0.01
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
17-31
-------
Observations for CHNJ from Table 17-5 include the following:
• The pollutants of interest with the highest annual average concentrations by mass are
formaldehyde, acetaldehyde, and carbon tetrachloride. Note that only the two
carbonyl compounds have annual average concentrations greater than 1 |ig/m3.
• Some of the pollutants of interest for CHNJ exhibit a quarterly trend, such as
chloroform, which tended to be higher during the warmer months. Formaldehyde and
tetrachloroethylene exhibit similar tendencies, although the confidence intervals
indicate the differences are not statistically significant.
• The first and second quarter average concentrations for acetaldehyde are higher than
the third and fourth quarter average concentrations. Of the nine concentrations of
acetaldehyde greater than 2 |ig/m3, all but one was measured during the first or
second quarter of 2010 (five in the first quarter and three in the second).
• 1,2-Dichloroethane exhibits a quarterly tendency similar to acetaldehyde, although it
is reflected in Table 17-5 in a different way. 1,2-Dichloroethane was detected at total
of 14 times in 2010, of which all were measured during the first and second quarters
of 2010; thus, this pollutant was not detected at all during the second half of the year.
Observations for ELNJ from Table 17-5 include the following:
• The pollutants of interest with the highest annual average concentrations by mass are
formaldehyde, acetaldehyde, and benzene. These are the only pollutants with annual
average concentrations greater than 1 |ig/m3. The annual average concentration of
formaldehyde for ELNJ is nearly three times higher than the annual average
concentrations of formaldehyde for CHNJ and NBNJ.
• The concentrations of many of the pollutants of interest for ELNJ appear to be higher
during the warmer months of the year, as illustrated by the second and third quarter
average concentrations. However, most of the differences are not statistically
significant. Chloroform is the only pollutant for which the differences are deemed
significant.
• Concentrations of the three carbonyl compound pollutants of interest are greatest for
the second quarter of 2010. A review of the data shows that the two highest
concentrations of each of these pollutants were measured on March 21, 2010 and
May 20, 2010. ELNJ's March 21, 2010 concentration (14.7 |ig/m3) is the fifth highest
formaldehyde concentration among all NMP sites sampling formaldehyde (and its
May 20, 2010 concentration, 12.6 |ig/m3 ranks ninth). ELNJ's acetaldehyde
concentrations for these two dates rank similarly among the program-wide
measurements of this pollutant. The propionaldehyde concentrations rank third and
fourth highest among NMP sites sampling carbonyl compounds.
• 1,2-Dichloroethane exhibited a quarterly tendency for ELNJ as it did for CHNJ; this
pollutant was detected at total of 11 times at ELNJ, of which all were measured
during the first and second quarters of 2010; thus, this pollutant was not detected at
17-32
-------
all during the second half of the year. Nine of the 11 measured detections of
1,2-dichloroethane at ELNJ were measured between February and April 2010.
Observations for NBNJ from Table 17-5 include the following:
• The pollutants of interest with the highest annual average concentrations by mass are
acetaldehyde, formaldehyde, and benzene. Acetaldehyde and formaldehyde are the
only pollutants with annual average concentrations greater than 1 |ig/m3. The annual
average concentration of acetaldehyde for NBNJ is the highest among the three New
Jersey sites sampling this pollutant.
• The third quarter average concentration of acrylonitrile is significantly higher than the
other quarterly averages and has a large confidence interval associated with it,
indicating that the concentration average is likely influenced by outliers. The
maximum concentration of acrylonitrile is 2.96 |ig/m3 and was measured on
July 7, 2010. This concentration is seven times the next highest concentration
measured at NBNJ (0.422 |ig/m3 measured on August 6, 2010). The July 7, 2010
acrylonitrile concentration for NBNJ is the highest concentration of this pollutant
measured among NMP sites sampling VOC. The highest concentration of
acetaldehyde was also measured at NBNJ on this date.
• Chloroform concentrations also appear higher during the summer months at NBNJ
compared to the rest of the year, although the relatively high confidence interval for
the first quarter 2010 average indicates that there is more variability in the chloroform
measurements at NBNJ than the other New Jersey sites. A review of the data shows
that the highest concentration of chloroform was measured on February 7, 2010
(0.391 |ig/m3). However, of the eight concentrations greater than 0.175 |ig/m3
measured at NBNJ, most were measured during the third quarter of 2010.
• 1,2-Dichloroethane exhibited the same quarterly tendency at NBNJ as it did for
CFINJ and ELNJ, although there was one measured detection in the third quarter of
2010. The bulk of the measured detections of 1,2-dichloroethane at NBNJ were
measured between March and April 2010.
Observations for PANJ from Table 17-5 include the following:
• VOC sampling at PANJ began at the end of April 2010. Thus, first quarter average
concentrations could not be calculated. Second quarter concentrations were also not
calculated because there were not enough samples collected to meet the completeness
criteria. Although enough make-up samples were collected during the second half of
2010 for third and fourth quarter average concentrations to be calculated, annual
average concentrations were not calculated for PANJ because there were not at least
three quarterly averages available. However, Appendix J provides the pollutant-
specific average concentration for all valid samples collected at PANJ over the entire
sample period.
17-33
-------
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for the New
Jersey sites from those tables include the following:
• The New Jersey sites appear in Table 4-9 for VOC a total of 13 times (CHNJ - 3;
ELNJ - 6; and NBNJ - 4). However, the highest rankings for the New Jersey sites are
for NBNJ, which has the third highest annual average concentration of acrylonitrile
among NMP sites sampling VOC and CJrDSTJ, which has the third highest annual
average concentration of 1,2-dichloroethane among NMP sites sampling VOC.
• The New Jersey sites appear in Table 4-10 for carbonyl compounds three times.
NBNJ and ELNJ have the fourth and fifth highest annual average concentrations of
acetaldehyde, respectively, while ELNJ has the highest annual average concentration
of formaldehyde.
17.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for acetaldehyde, benzene,
1,3-butadiene, and formaldehyde were created for CJrtNJ, ELNJ, and NBNJ. Box plots were not
created for PANJ because annual averages could not be calculated for this site. Figures 17-20
through 17-23 overlay the sites' minimum, annual average, and maximum concentrations onto
the program-level minimum, first quartile, average, median, third quartile, and maximum
concentrations, as described in Section 3.5.3.
17-34
-------
Figure 17-20. Program vs. Site-Specific Average Acetaldehyde Concentration
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 17-21. Program vs. Site-Specific Average Benzene Concentration
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
17-35
-------
Figure 17-22. Program vs. Site-Specific Average 1,3-Butadiene Concentration
i
•
^
0
1
1
1
^
0
-1
• 1 0
H-
0.1
Program:
Site:
Figure 17-23.
V
Dl
1°
1
E
5 1
Program:
Site:
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Concentration (ng/m3)
IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
• DID
Site Average Site Minimum/Maximum
o —
Program vs. Site-Specific Average Formaldehyde Concentration
1 1 II
1 1 II
1 1
1 1 1
0 15 20 25 30 35 40 45 50 55
Concentration (ng/m3)
IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
• DID
Site Average Site Minimum/Maximum
o —
17-36
-------
Observations from Figures 17-20 through 17-23 include the following:
• Figure 17-20 shows that while CFENJ's annual average acetaldehyde
concentration is below the program-level average for acetaldehyde, the annual
averages for ELNJ and NBNJ are greater than the program-level average
concentration. The range of acetaldehyde measurements is greatest for ELNJ and
least for CFENJ. There were no non-detects of acetaldehyde measured at any of
these sites.
• Figure 17-21 for benzene shows that both the annual average and maximum
benzene concentration for CFDSTJ are less than the program-level average. This site
has the second lowest annual average benzene concentration among sites
sampling benzene. NBNJ's annual average benzene concentration is also below
the program-level average while ELNJ's annual average is just greater than the
program-level average concentration. There were no non-detects of benzene
measured at any of these sites.
• Figure 17-22 for 1,3-butadiene resembles Figure 17-21 for benzene. Both the
annual average and maximum 1,3-butadiene concentrations for CFDSTJ are less
than the program-level average concentration. Further, the maximum
1,3-butadiene concentration for CHNJ is just greater than the program-level
median (or 50th percentile). This site has the third lowest annual average
1,3-butadiene concentration among sites sampling this pollutant. NBNJ's annual
average 1,3-butadiene concentration is also less than the program-level average.
Although ELNJ's annual average concentration is greater than the program-level
average, the maximum benzene concentration measured at ELNJ is well below
the program-level maximum concentration. Several non-detects of 1,3-butaidene
were measured at CFtNJ and NBNJ, but there were no non-detects of
1,3-butadiene measured at ELNJ.
• Figure 17-23 for formaldehyde shows that while the annual average
concentrations of formaldehyde for CFINJ and NBNJ are below the program-level
average, the annual average for ELNJ is greater than the program-level average
concentration. Although ELNJ has the highest annual average formaldehyde
concentration among NMP sites sampling carbonyl compounds, the maximum
concentration of formaldehyde was not measured at ELNJ. There were no
non-detects of formaldehyde measured at any of these sites.
17-37
-------
17.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. CHNJ, ELNJ, and NBNJ have sampled VOC and carbonyl compounds under the
NMP for many years. ELNJ has sampled under the NMP since 2000 and CHNJ and NBNJ since
2001. Thus, Figures 17-24 through 17-35 present the 3-year rolling statistical metrics for
acetaldehyde, benzene, 1,3-butadiene, and formaldehyde for CHNJ, ELNJ, and NBNJ,
respectively. The statistical metrics presented for assessing trends include the substitution of
zeros for non-detects.
Figure 17-24. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at CHNJ
"m
£ >fl -
a.
|
I
r1
^.
^p
~^~
i
••
2001-2005 i 2002-2004 2003-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Percentite
Minimum
— Median — Maximum • 95th Percentile ...... Average
Carbonyl compound sampling began in May 2001.
17-38
-------
Figure 17-25. Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured
at CHNJ
2001-20031 2002-2004
2003-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
5th Percentile — Minimum - Median — Maximum • 95th Percentile •••*•• Average
1VOC sampling began in May 2001.
Figure 17-26. Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations
Measured at CHNJ
'
r
E01
i
|
>
5
0
2
^J
>01-20<
T —
_u_ »A>. ...+...
^*^ m
^ •••»•• ^^— ^^ m^m
31 2002-2004 2005-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Pei tentile
— Minimum — Median — Maximum • 95th Percentile ...... Average
'VOC sampling began in May 2001.
17-39
-------
Figure 17-27. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at CHNJ
entration (|ig
0
T
|
^ U]
T T "
3 : b cL
a ' ^^^ I a ' ^^™ * i i ' 2 i A '
2001-20031 2002-2004 2003-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Percentile - Minimum - Median - Maximum • 9 5 tli Pert en tile ...*.- Average
Carbonyl compound sampling began in May 2001.
Figure 17-28. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at ELNJ
I 1
2000-2002 2001-20031 2002-2004' 2003-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
« 5th Pertentile — Mininiuni — Median — Maximum • 95th Pertentile
. Average
'Carbonyl compound samples were not collected in January 2003.
17-40
-------
Figure 17-29. Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured
at ELNJ
2000-2002 2001-2003 L 2002-2004 J 2003-2005 * 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
5thPercentile — Minimum - Median — Maximum • 95th Percentile
'VOC samples were not collected in January 2003.
Figure 17-30. Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations
Measured at ELNJ
2000-2002 2001-2003 2002-2004x 2003-2005 ' 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
5thPercenttle - Minimum - Median — Maximum • 95th Pertentile •••*•• Average
VOC samples were not collected in January 2003.
17-41
-------
Figure 17-31. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at ELNJ
V T
T T
2000-2002 2001-2003* 2002-2004 ' 2003-2005 i 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
5th Percentile — Minimum - Median — Maximum • 95th Percentile •••*•• Average
Carbonyl compound samples were not collected in January 2003.
Figure 17-32. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at NBNJ
I
I «„
Concent rat
i C
5 C
0
ij -
ta rfa a
H T
i_^
2001-2003 i 2002-2004 2003-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three- Year Per od
• 5th Percentile — Mininiuni — Median — Maximum • 95th Pert en tile ...^.. Average
'Carbonyl compound sampling began May 2001.
17-42
-------
Figure 17-33. Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured
at NBN J
2001-2005 x 2002-2004
2003-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
5th Percentile — Minimum - Median — Maximum • 95th Percentile •••*•• Average
'VOC sampling began May 2001.
Figure 17-34. Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations
Measured at NBNJ
3- 0.25
|
8
J 0.2
2001-2003 ± 2002-2004 2003-2005 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Perteutile — Minimum — Median — Maximum • 95th Pertentile
. Average
'VOC sampling began May 2001.
17-43
-------
Figure 17-35. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at NBNJ
I
2001-2003 2002-2004
2005-2005 2004-2006 2005-2007
Three-Year Period
5th Percentfle - Minimum - Median — Maximum • 95th Percentile ...4.. Average
Carbonyl compound sampling began May 2001.
Observations from Figure 17-24 for acetaldehyde measurements at CHNJ include the
following:
• Carbonyl compound sampling at CHNJ began in May 2001.
• The maximum acetaldehyde concentration was measured in 2004. The second and
third highest concentrations were measured in 2004 and 2005; excluding these three
concentrations, all other acetaldehyde concentrations measured at CHNJ were less
than 5 |ig/m3.
• The rolling average and the median values were similar to each other for each time
period after 2004-2006. This indicates decreasing variability in the central tendency
of acetaldehyde concentrations measured at CHNJ over the periods shown.
• Although difficult to discern in Figure 17-24, a decreasing trend in the average and
median acetaldehyde concentrations is shown since the onset of sampling, although
both the median and average concentrations leveled out across the last few periods.
17-44
-------
Observations from Figure 17-25 for benzene measurements at CHNJ include the
following:
• Similar to carbonyl compounds, VOC sampling at CHNJ began in May 2001.
• The five highest benzene concentrations were measured in 2008 and 2009, although
no benzene measurement at CHNJ was greater than 2.5 |ig/m3.
• The average and median concentrations exhibit a decreasing trend that levels out over
the last few 3-year periods.
Observations from Figure 17-26 for 1,3-butadiene measurements at CHNJ include the
following:
• The maximum 1,3-butadiene concentration was measured in 2003 and was nearly
twice the next highest concentration, which was measured in 2008.
• The rolling average and median concentrations have an increasing trend through the
2006-2008 time frame and then begin a slight decreasing trend for the final time
frames.
• The minimum, 5th percentile, and median concentrations were all zero through the
2004-2006 time frame, indicating the presence of non-detects (at least 50 percent).
The number of non-detects reported has decreased through the years as the MDL has
improved, from as high as 97 percent in 2001 to as low as 17 percent in 2008. The
number of non-detects for 2010 is 70 percent.
Observations from Figure 17-27 for formaldehyde measurements at CHNJ include the
following:
• The statistical metrics presented for formaldehyde are similar to those for
acetaldehyde in Figure 17-24.
• The maximum formaldehyde concentration was measured in 2004. This concentration
of formaldehyde is nearly four times the maximum concentrations shown for other
periods not including 2004. The second highest concentration was also measured in
2004, but was nearly half the magnitude. These two maximum concentrations were
measured on the same days as the acetaldehyde maximum concentrations.
• Although difficult to discern in Figure 17-27, a decreasing trend in the average
formaldehyde concentrations is shown through 2005-2007, after which the average
concentrations leveled out.
17-45
-------
Observations from Figure 17-28 for acetaldehyde measurements at ELNJ include the
following:
• Carbonyl compound sampling at ELNJ began in January 2000. A 1-month period
when samples were not collected occurred in January 2003, as denoted in
Figure 17-28.
• The maximum acetaldehyde concentration was measured in 2007, although
concentrations of similar magnitude were also measured in 2005 and 2006.
• The rolling average and the median concentrations have steadily increased through
the 2005-2007 time frame, after which a decreasing trend begins and continues
through the 2008-2010 time frame.
• The difference between the rolling average and the median values decreased
significantly for the 2008-2010 period. The decreasing difference between these
statistical parameters indicates decreasing variability in the central tendency.
Observations from Figure 17-29 for benzene measurements at ELNJ include the
following:
• VOC sampling at ELNJ also began in January 2000. A 1-month period when samples
were not collected occurred in January 2003, as denoted in Figure 17-29.
• The maximum benzene concentration was measured in 2008 and is more than four
times higher than the next highest concentration (measured in 2009).
• Although difficult to discern in Figure 17-29, a decreasing trend in the rolling average
and median concentrations is shown across all time frames through 2005-2007. Even
with the higher concentrations measured in 2008 and 2009, the average
concentrations for the 2006-2008 through 2008-2010 time frames were similar to the
average concentration for the 2005-2007 time frame and the median concentration
continued its decreasing trend through these periods. If the maximum concentration
from 2008 were removed from the calculations, the rolling average would continue its
downward trend through 2006-2008, then hold steady for the final two 3-year periods.
• The difference between the rolling average and the median concentrations for the
2006-2008 through 2008-2010 time frames illustrates the effect of outliers on the
average concentration that is not apparent in the median concentration.
Observations from Figure 17-30 for 1,3-butadiene measurements at ELNJ include the
following:
• The maximum concentration of 1,3-butadiene was measured in 2009 and is nearly
two and a half times the next highest concentration (measured in 2001). These two
concentrations are the only measurements of 1,3-butadiene from ELNJ greater than
1 |ig/m3.
17-46
-------
• Figure 17-30 shows a decreasing trend in the earlier years of sampling, then a
leveling off of average concentrations that continues through the 2008-2010 time
frame. Even with the higher concentration measured in 2009, the average
concentration for the 2007-2009 and 2008-2010 time frames were similar to the
average concentrations for the previous four 3-year periods. The median
concentrations exhibit a similar pattern.
Even with the maximum concentration measured in 2009, the difference between the
5th and 95th percentiles has been decreasing since the onset of sampling, indicating an
overall decrease in the range of concentrations measured at ELNJ.
Observations from Figure 17-31 for formaldehyde measurements at ELNJ include the
following:
• The maximum formaldehyde concentration was measured in 2010, although a similar
concentration was also measured in 2000.
• Similar to acetaldehyde, the rolling average and the median concentrations of
formaldehyde have steadily increased over much of the sampling period, although a
decreasing trend begins with 2006-2008 and continues through the 2008-2010 time
frame, even with the maximum concentration measured in 2010.
Observations from Figure 17-32 for acetaldehyde measurements atNBNJ include the
following:
• Carbonyl compound sampling at NBNJ began in May 2001.
• Similar to CFINJ, the maximum acetaldehyde concentration was measured in 2004.
This concentration of acetaldehyde (111 |ig/m3) is nearly seven times higher, and an
order of magnitude higher, than the next highest concentration (16.2 |ig/m3 measured
in 2005). Of the 29 concentrations greater than 8 |ig/m3, 28 were measured in 2004 or
2005.
• The rolling average concentration appears to increase beginning with the inclusion of
2004 data then decreases after. However, the median concentrations follow a similar
increasing then decreasing pattern as the rolling average.
Observations from Figure 17-33 for benzene measurements at NBNJ include the
following:
• VOC sampling at NBNJ also began in May 2001.
• The maximum benzene concentration was measured in 2002, but similar
concentrations were also measured in 2005 and 2009.
17-47
-------
• The rolling averages and the medians are similar to each other for each time period
shown. The difference between them was less than 0.15 |ig/m3 for each 3-year period.
This indicates relatively little variability in the central tendency.
• A decreasing trend in the rolling average and median concentrations is shown across
much of the sampling period, although the concentrations leveled out for the last three
3-year periods.
Observations from Figure 17-34 for 1,3-butadiene measurements atNBNJ include the
following:
• The maximum 1,3-butadiene concentration was measured in 2005.
• The rolling average concentrations of 1,3-butadiene at NBNJ have fluctuated over the
years of sampling, ranging from 0.030 |ig/m3 (2002-2004) to 0.055 |ig/m3 (2006-
2008). The increase shown in Figure 17-34 is likely a result of fewer non-detects, and
thus zeros, included in the calculation, as discussed below. The rolling average
concentration of 1,3-butadiene leveled out for the last several time frames shown.
• The minimum, 5th percentile, and median concentrations were all zero through the
2004-2006 time frame, indicating the presence of non-detects (at least 50 percent).
The number of non-detects reported has decreased through the later years, from as
high as 93 percent in 2004 to as low as two percent in 2008. Between one and three
non-detects has been reported each year between 2008 and 2010.
Observations from Figure 17-35 for formaldehyde measurements atNBNJ include the
following:
• The statistical metrics presented in Figure 17-35 for formaldehyde are similar to those
presented in Figure 17-32 for acetaldehyde.
• The maximum formaldehyde concentration was measured on the same day in 2004
that the highest acetaldehyde concentration was measured. This concentration of
formaldehyde is more than four times the next highest concentration (measured in
2006). Note that at least one concentration of about 20 jig/m3 was measured in 2001,
2003, 2006, and 2009.
• The rolling average concentration appears to increase beginning with the 2002-2004
time frame then decreases after the 2004-2006 time frame. The decrease from the
2005-2007 to the 2006-2008 time frame is significant, although this is difficult to
discern in Figure 17-35 because of the outlier. The rolling average concentrations for
the 2007-2009 and 2008-2010 time frames are similar to the average concentration
for the 2006-2008 period. The median concentrations over the period of sampling
follow the same trend as the rolling averages.
17-48
-------
17.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at each New
Jersey monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations regarding
the various risk factors, time frames, and calculations associated with these risk screenings.
17.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
New Jersey monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
for each site were compared to the acute MRL; the quarterly averages were compared to the
intermediate MRL; and the annual averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the New Jersey monitoring sites were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
the New Jersey monitoring sites.
17.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the New Jersey monitoring sites and where annual
average concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 17-6, where applicable.
17-49
-------
Table 17-6. Cancer and Noncancer Surrogate Risk Approximations for the New Jersey
Monitoring Sites
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
# of Measured
Detections vs.
# of Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Chester, New Jersey - CHNJ
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
Trichloroethylene
0.0000022
0.000068
0.0000078
0.00003
0.000006
0.000026
0.000013
2.6E-07
0.0000048
0.009
0.002
0.03
0.002
0.1
0.098
2.4
0.0098
0.04
0.002
59/59
16/57
57/57
17/57
57/57
48/57
14/57
59/59
39/57
4/57
1.31
±0.19
0.04
±0.02
0.48
±0.04
0.01
±0.01
0.64
±0.03
0.08
±0.01
0.02
±0.01
1.64
±0.31
0.07
±0.02
0.01
±0.01
2.88
2.51
3.72
0.29
3.85
0.47
21.32
0.02
0.01
0.15
0.02
0.02
0.01
0.01
0.01
O.01
0.17
O.01
0.01
Elizabeth, New Jersey - ELNJ
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Ethylbenzene
Formaldehyde
Propionaldehyde
Tetrachloroethylene
0.0000022
0.000068
0.0000078
0.00003
0.000006
_
0.000011
0.000026
0.0000025
0.000013
2.6E-07
0.009
0.002
0.03
0.002
0.1
0.098
0.8
2.4
1
0.0098
0.008
0.04
59/59
10/59
59/59
59/59
58/59
51/59
44/59
11/59
59/59
59/59
59/59
58/59
2.73
±0.39
0.02
±0.01
1.02
±0.13
0.12
±0.01
0.60
±0.04
0.13
±0.02
0.09
±0.02
0.02
±0.01
0.42
±0.04
4.46
±0.64
0.54
±0.09
0.20
±0.03
6.01
1.34
7.97
3.57
3.60
_
0.94
0.39
1.05
57.93
0.05
0.30
0.01
0.03
0.06
0.01
0.01
O.01
0.01
O.01
0.45
0.07
0.01
NA = Not available due to the criteria for calculating an annual average.
- = a Cancer URE or Noncancer RfC is not available.
17-50
-------
Table 17-6. Cancer and Noncancer Surrogate Risk Approximations for the New Jersey
Monitoring Sites (Continued)
Pollutant
Trichloroethylene
Vinyl Chloride
Cancer
URE
(Hg/m3)1
0.0000048
0.0000088
Noncancer
RfC
(mg/m3)
0.002
0.1
# of Measured
Detections vs.
# of Samples
22/59
2/59
Annual
Average
(Hg/m3)
0.03
±0.01
<0.01
±<0.01
Cancer Risk
Approximation
(in-a-million)
0.14
<0.01
Noncancer
Risk
Approximation
(HQ)
0.02
<0.01
New Brunswick, New Jersey - NBNJ
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.0000022
0.000068
0.0000078
0.00003
0.000006
0.000026
0.000013
2.6E-07
0.0000048
0.0000088
0.009
0.002
0.03
0.002
0.1
0.098
2.4
0.0098
0.04
0.002
0.1
58/58
23/55
55/55
53/55
54/55
49/55
11/55
58/58
52/55
18/55
6/55
2.92
±0.37
0.14
±0.11
0.65
±0.05
0.05
±0.01
0.56
±0.05
0.12
±0.02
0.02
±0.01
1.63
±0.22
0.12
±0.01
0.02
±0.01
<0.01
±0.01
6.42
9.60
5.07
1.53
3.37
0.40
21.24
0.03
0.08
0.02
0.32
0.07
0.02
0.03
0.01
0.01
O.01
0.17
O.01
0.01
0.01
Paterson, New Jersey - PANJ
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
Ethylbenzene
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.0000078
0.00003
0.000006
0.000011
0.0000025
2.6E-07
0.0000048
0.0000088
0.03
0.002
0.1
0.098
0.8
1
0.04
0.002
0.1
21/21
21/21
21/21
18/21
21/21
21/21
20/21
8/21
2/21
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA = Not available due to the criteria for calculating an annual average.
— = a Cancer URE or Noncancer RfC is not available.
17-51
-------
Observations from Table 17-6 include the following:
• For CHNJ, the pollutants with the highest annual averages are formaldehyde,
acetaldehyde, and carbon tetrachloride. Formaldehyde has the highest cancer risk
approximation for this site, followed by carbon tetrachloride and benzene. The cancer
risk approximations for formaldehyde are at least an order of magnitude higher than
the approximations for the other pollutants of interest. None of the pollutants of
interest for CFDSTJ have noncancer risk approximations greater than 1.0.
• For ELNJ, the pollutants with the highest annual averages are formaldehyde,
acetaldehyde, and benzene. These three pollutants also have the highest cancer risk
approximations for this site, although the cancer risk approximation for benzene is
greater than the cancer risk approximation for acetaldehyde. The cancer risk
approximation for formaldehyde for ELNJ (57.93 in-a-million) is the highest
calculated cancer risk approximation among NMP sites. None of the pollutants of
interest for ELNJ have noncancer risk approximations greater than 1.0, although the
noncancer risk approximation for formaldehyde for ELNJ (0.45) is the third highest
among NMP sites.
• For NBNJ, the pollutants with the highest annual averages are acetaldehyde,
formaldehyde, and benzene. Formaldehyde has the highest cancer risk approximation
for NBNJ, followed by acrylonitrile and acetaldehyde. None of the pollutants of
interest for NBNJ have noncancer risk approximations greater than 1.0.
• Because annual averages could not be calculated for PANJ, cancer and noncancer risk
approximations could not be calculated either.
17.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 17-7 and 17-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 17-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer risk approximations (in-a-million), as calculated from the annual averages. Table 17-8
presents similar information, but identifies the 10 pollutants with the highest noncancer risk
approximations (HQ), also calculated from annual averages.
17-52
-------
Table 17-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the New Jersey Monitoring Sites
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Chester, New Jersey (Morris County) - CHNJ
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
POM, Group la
POM, Group 6
207.44
118.39
111.67
75.87
32.41
12.97
9.08
1.98
0.32
0.16
Benzene
Formaldehyde
1,3 -Butadiene
Naphthalene
Hexavalent Chromium, PM
Ethylbenzene
POM, Group 2b
Acetaldehyde
Arsenic, PM
POM, Group 3
1.62E-03
1.45E-03
9.72E-04
4.41E-04
3.11E-04
2.96E-04
1.74E-04
1.67E-04
1.31E-04
1.24E-04
Formaldehyde
Carbon Tetrachloride
Benzene
Acetaldehyde
Acrylonitrile
1 ,2-Dichloroethane
1,3 -Butadiene
Tetrachloroethylene
Trichloroethylene
21.32
3.85
3.72
2.88
2.51
0.47
0.29
0.02
0.01
Elizabeth, New Jersey (Union County) - ELNJ
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
Dichloromethane
1,3 -Butadiene
Naphthalene
POM, Group 2b
Nickel, PM
Propylene oxide
160.93
111.28
89.90
67.31
41.31
23.48
11.41
1.57
1.36
0.70
Formaldehyde
Benzene
1,3 -Butadiene
Nickel, PM
Hexavalent Chromium, PM
Naphthalene
Arsenic, PM
Ethylbenzene
Acetaldehyde
POM, Group 2b
1.45E-03
1.26E-03
7.04E-04
6.52E-04
6.03E-04
3.88E-04
2.37E-04
2.25E-04
1.48E-04
1.38E-04
Formaldehyde
Benzene
Acetaldehyde
Carbon Tetrachloride
1,3 -Butadiene
Acrylonitrile
Ethylbenzene
/>-Dichlorobenzene
1 ,2-Dichloroethane
Trichloroethylene
57.93
7.97
6.01
3.60
3.57
1.34
1.05
0.94
0.39
0.14
-------
Table 17-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the New Jersey Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
New Brunswick, New Jersey (Middlesex County) - NBNJ
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
Ethylene oxide
H^ 'rachloroethylene
263.33
175.06
146.78
107.72
40.00
20.62
7.26
2.82
1.05
0.96
Formaldehyde
Benzene
1,3 -Butadiene
Naphthalene
Hexavalent Chromium, PM
Ethylbenzene
Nickel, PM
POM, Group 2b
Acetaldehyde
Arsenic, PM
2.28E-03
2.05E-03
1.20E-03
7.01E-04
4.61E-04
3.67E-04
2.67E-04
2.48E-04
2.37E-04
1.84E-04
Formaldehyde
Acrylonitrile
Acetaldehyde
Benzene
Carbon Tetrachloride
1,3 -Butadiene
1 ,2-Dichloroethane
Trichloroethylene
Tetrachloroethylene
Vinyl Chloride
21.24
9.60
6.42
5.07
3.37
1.53
0.40
0.08
0.03
0.02
Paterson, New Jersey (Passaic County) - PANJ
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
POM, Group la
POM, Group 6
115.84
63.19
62.14
41.52
17.51
7.26
3.63
1.09
0.32
0.09
Benzene
Formaldehyde
1,3 -Butadiene
Naphthalene
Hexavalent Chromium, PM
Ethylbenzene
POM, Group 2b
Acetaldehyde
POM, Group 3
Arsenic, PM
9.04E-04
8.08E-04
5.25E-04
2.47E-04
1.63E-04
1.58E-04
9.60E-05
9.14E-05
8.90E-05
7.09E-05
-------
Table 17-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the New Jersey Monitoring Sites
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer Toxicity
Weight
Top 10 Noncancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Chester, New Jersey (Morris County) - CHNJ
Toluene
Xylenes
Benzene
Hexane
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Ethylene glycol
Naphthalene
624.27
493.82
207.44
131.84
118.39
111.67
75.87
32.41
29.66
12.97
Acrolein
1,3 -Butadiene
Formaldehyde
Acetaldehyde
Benzene
Xylenes
Lead, PM
Naphthalene
Arsenic, PM
Propionaldehyde
302,596.87
16,206.30
11,395.19
8,429.94
6,914.65
4,938.24
4,349.55
4,323.69
2,026.03
1,006.84
Formaldehyde
Acetaldehyde
Acrylonitrile
Benzene
Carbon Tetrachloride
1,3 -Butadiene
Tetrachloroethylene
Trichloroethylene
Chloroform
1 ,2-Dichloroethane
0.17
0.15
0.02
0.02
0.01
<0.01
0.01
0.01
0.01
0.01
Elizabeth, New Jersey (Union County) - ELNJ
Toluene
Xylenes
Benzene
Formaldehyde
Hexane
Ethylbenzene
Acetaldehyde
Dichloromethane
Ethylene glycol
Hydrochloric acid
471.64
352.06
160.93
111.28
109.81
89.90
67.31
41.31
36.52
24.34
Acrolein
Nickel, PM
1,3 -Butadiene
Formaldehyde
Acetaldehyde
Benzene
Naphthalene
Manganese, PM
Arsenic, PM
Xylenes
267,278.15
15,081.94
11,739.58
11,354.92
7,478.86
5,364.23
3,803.12
3,755.20
3,671.47
3,520.63
Formaldehyde
Acetaldehyde
Propionaldehyde
1,3 -Butadiene
Benzene
Trichloroethylene
Acrylonitrile
Carbon Tetrachloride
Tetrachloroethylene
Chloroform
0.45
0.30
0.07
0.06
0.03
0.02
0.01
0.01
0.01
0.01
-------
Table 17-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the New Jersey Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer Toxicity
Weight
Top 10 Noncancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
New Brunswick, New Jersey (Middlesex County) - NBNJ
Toluene
Xylenes
Benzene
Hexane
Formaldehyde
Ethylbenzene
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
Glycol ethers, gas
776.52
593.79
263.33
227.61
175.06
146.78
107.72
49.05
40.00
37.20
Acrolein
1,3 -Butadiene
Formaldehyde
Manganese, PM
Acetaldehyde
Benzene
Naphthalene
Titanium tetrachloride
Nickel, PM
Xylenes
445,148.01
20,000.36
17,862.97
12,571.41
11,968.98
8,777.75
6,872.93
6,385.00
6,175.26
5,937.95
Acetaldehyde
Formaldehyde
Acrylonitrile
1,3 -Butadiene
Benzene
Trichloroethylene
Carbon Tetrachloride
Tetrachloroethylene
Chloroform
Vinyl Chloride
0.32
0.17
0.07
0.03
0.02
0.01
0.01
0.01
0.01
O.01
Paterson, New Jersey (Passaic County) - PANJ
Toluene
Xylenes
Benzene
Hexane
Ethylbenzene
Formaldehyde
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
Glycol ethers, gas
358.91
286.81
115.84
71.05
63.19
62.14
41.52
29.87
17.51
13.40
Acrolein
1,3 -Butadiene
Formaldehyde
Acetaldehyde
Benzene
Xylenes
Naphthalene
Arsenic, PM
Lead, PM
Glycol ethers, gas
154,681.77
8,754.09
6,341.18
4,613.84
3,861.24
2,868.11
2,419.97
1,099.65
797.37
670.22
-------
The pollutants listed in Tables 17-7 and 17-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer risk approximations based on each site's annual averages are
limited to those pollutants for which each respective site sampled. As discussed in Section 17.3,
CHNJ, ELNJ, and NBNJ sampled for VOC and carbonyl compounds. In addition, the cancer and
noncancer surrogate risk approximations are limited to those pollutants with enough data to meet
the criteria for annual averages to be calculated. The completeness criteria were not met by
PANJ; as a result, annual averages, and thus cancer and noncancer risk approximations, were not
calculated for this site. A more in-depth discussion of this analysis is provided in Section 3.5.5.3.
Observations from Table 17-7 include the following:
• Benzene is the highest emitted pollutant with a cancer URE in all four New Jersey
counties, followed by ethylbenzene, formaldehyde, and acetaldehyde (although not
necessarily in that order).
• Benzene, formaldehyde, and 1,3-butadiene are the pollutants with the highest
toxi city-weighted emissions (of the pollutants with cancer UREs) for all four New
Jersey counties, although not necessarily in that order.
• Seven of the 10 highest emitted pollutants in Morris, Middlesex, and Passaic Counties
also have the highest toxi city-weighted emissions while eight of the highest emitted
pollutants in Union County also have the highest toxicity-weighted emissions.
• Formaldehyde is the pollutant with the highest cancer risk approximations for CJrDSTJ,
ELNJ, and NBNJ. This pollutant also appeared at or near the top of both emissions-
based lists. Acetaldehyde, benzene, and 1,3-butadiene also appear on all three lists for
these sites. Conversely, carbon tetrachloride and acrylonitrile appear on neither
emissions-based list for these three New Jersey sites but appeared among the
pollutants with the highest cancer risk approximations for each site.
Observations from Table 17-8 include the following:
• Toluene, xylenes, and benzene are the highest emitted pollutants with noncancer
RfCs for all four New Jersey counties. Toluene did not appear on any county's list of
highest toxicity-weighted emissions and is not a pollutant of interest for any site.
• Acrolein is the pollutant with the highest toxicity-weighted emissions (of the
pollutants with noncancer RfCs) for all four counties but is not among the highest
emitted pollutants for any of the New Jersey counties. Although acrolein was sampled
for at all four sites, this pollutant was excluded from the pollutant of interest
17-57
-------
designation, and thus subsequent risk screening evaluations, due to questions about
the consistency and reliability of the measurements, as discussed in Section 3.2.
• The number of pollutants in common between the highest emitted pollutants and
those with the highest toxicity-weighted emissions ranged from four (Union County)
to six (Passaic and Morris Counties).
• Formaldehyde and acetaldehyde are among the pollutants with the highest noncancer
risk approximations for CHNJ, ELNJ, and NBNJ (although all were less than an HQ
of 1.0). These pollutants also appear among the pollutants with the highest emissions
and toxicity-weighted emissions for all counties. Benzene also appears on all three
lists.
17.6 Summary of the 2010 Monitoring Data for the New Jersey Monitoring Sites
Results from several of the data treatments described in this section include the
following:
»«» Twelve pollutants failed at least one screen for CHNJ; 14 failed screens for ELNJ; 12
failed screens for NBNJ; and eight failed screens for PANJ.
»«» Formaldehyde had the highest annual average concentration for CHNJ and ELNJ
while acetaldehyde had the highest annual average concentration for NBNJ. Annual
average concentrations could not be calculated for PANJ.
»«» The annual average formaldehyde concentration for ELNJ is the highest annual
average among NMP sites sampling this pollutant.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest, where they could be calculated,
were greater than their associatedMRL noncancer health risk benchmarks.
17-58
-------
18.0 Sites in New York
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS and CSATAM sites in New York, and integrates these
concentrations with emissions, meteorological, and risk information. Data generated by sources
other than ERG are not included in the data analyses contained in this report. Readers are
encouraged to refer to Sections 1 through 4 and the glossary (Appendix P) for detailed
discussions and definitions regarding the various data analyses presented below.
18.1 Site Characterization
This section characterizes the New York monitoring sites by providing geographical and
physical information about the locations of the sites and the surrounding areas. This information
is provided to give the reader insight regarding factors that may influence the air quality near the
sites and assist in the interpretation of the ambient monitoring measurements.
The New York monitoring sites are located in New York City (BXNY and MONY),
Rochester (ROCH), and Tonawanda (TONY). Figures 18-1 through 18-4 are composite satellite
images retrieved from ArcGIS Explorer showing the monitoring sites in their urban locations.
Figures 18-5 through 18-7 identify point source emissions locations by source category, as
reported in the 2008 NEI for point sources. Note that only sources within 10 miles of the sites are
included in the facility counts provided in Figures 18-5 through 18-7. Thus, sources outside the
10-mile radius have been grayed out, but are visible on the maps to show emissions sources
outside the 10-mile boundary. A 10-mile boundary was chosen to give the reader an indication of
which emissions sources and emissions source categories could potentially have a direct effect
on the air quality at the monitoring sites. Further, this boundary provides both the proximity of
emissions sources to the monitoring sites as well as the quantity of such sources within a given
distance of the sites. Table 18-1 describes the area surrounding each monitoring site by providing
supplemental geographical information such as land use, location setting, and locational
coordinates.
18-1
-------
Figure 18-1. Public School 52, New York City, New York (BXNY) Monitoring Site
oo
to
-------
Figure 18-2. Morrisania, New York City, New York (MONY) Monitoring Site
oo
-------
Figure 18-3. Rochester, New York (ROCH) Monitoring Site
oo
-------
Figure 18-4. Tonawanda, New York (TONY) Monitoring Site
oo
-------
Figure 18-5. NEI Point Sources Located Within 10 Miles of BXNY and MONY
\
, Wtestc Hester
County \
~^~-^^ \
Legend
7410'fTW 73'5S'0"W 73' 5D'0*W
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
BXNY NATTS site
MONY NATTS site O 10 mile radius
County boundary
Source Category Group (No. of Facilities) *
"'r Abrasive Product Manufacturing (1) Q
•0 Air-conditioning/Refngeraiion (3) ^
•41 Aircraft Operations (25) ?
JpQ Automobile/Tnjck Manufacturing (1) ^
£ Bakery (2) ,
B Bulk Terminals/Bulk Plants (2)
C Chemical Manufacturing (5) ^
6 Electrical Equipment (2) p
f Electricity Generation via Combustion (16) • ,
E Electroplating. Plating, Polishing, Anodizing, & Coloring (1) p
© Fabricated Metal Products (2) A
£S- Flexible Polyurethane Foam Production (2) >
F Food Processing/AgricuHure (3) g
H* Gasoline/Diesel Service Station (3) j
(V Heating Equipment Manufacturing (2) i
H Hospital (5)
Induslfial Machinery and Equipment (t)
Institutional - prison (1)
Institutional - school (21J
Miscellaneous Commercial/Industrial (26)
Miscellaneous Manufacturing (11)
Oil and/or Gas Production (1)
Pharmaceutical Manufacturing (2)
Printing, Coating & Dyeing of Fabric (1)
Printing/Publishing (10)
Pulp and Paper Plant/Wood Products (5)
Rubber and Miscellaneous Piastres Products (2)
Ship Building and Repairing (1)
Solid Waste Disposal -CommerciaUlnstilutional (1)
Surface Coaling (5)
Textile Mill (2)
Wastewater Treatment {6}
18-6
-------
Figure 18-6. NEI Point Sources Located Within 10 Miles of ROCH
7r45irw 7740'0-W 77-351TW 77'3fl'0"W 77'25'tTW
Legend
ROCH NATTS site
T7*35tnV TTWWW TT'25'ffW
Nole Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest,
10 mile radius | | County boundary
Source Category Group (No. of Facilities)
Air-conditioning/Refrigeration (1)
•+• Aircraft Operations (6)
c Chemical Manufacturing (2)
* Electricity Generation via Combustion (1}
tt Heating Equipment Manufacturing (2)
• Landfill (2)
P Printing/Publishing (2)
R Rubber and Miscellaneous Plastics Products (1)
1 Wastewater Treatment (1)
18-7
-------
Figure 18-7. NEI Point Sources Located Within 10 Miles of TONY
75 15'OTV TB'IOXTW
Legend
79-'Q'Q~W 7Er55'£TW 78:50'CTW 78'45'OTfV
Nolc: DUD to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
3$ TONY CSATAM site 10 mile radius
Source Category Group (No. of Facilities) ®
s Abrasive Product Manufacturing (1) ?
Air-conditioning/Refrigeration (1) H
-H Aircraft Operations (13) 1
8 Automobile/Truck Manufacturing (1) ffl
c Chemical Manufacturing (4) R
K. Coke Battery (1) ©
* Electricity Generation via Combustion (4) •»
F Food Processing/Agriculture (1) «
-------
Table 18-1. Geographical Information for the New York Monitoring Sites
Site
Code
BXNY
MONY
ROCH
TONY
AQS Code
36-005-0110
36-005-0080
36-055-1007
36-029-1013
Location
New York
New York
Rochester
Tonawanda
County
Bronx
Bronx
Monroe
Erie
Micro- or
Metropolitan
Statistical Area
New York-
Northern New
Jersey-Long
Island, NY-NJ-PA
MSA
(New York Div)
New York-
Northern New
Jersey-Long
Island, NY-NJ-PA
MSA
(New York Div)
Rochester, NY
MSA
Buffalo-Niagara
Falls, NY MSA
Latitude
and
Longitude
40.81616,
-73.90207
40.83606,
-73.92009
43.146198,
-77.54813
42.988443,
-78.918589
Land Use
Residential
Residential
Residential
Industrial
Location
Setting
Urban/City
Center
Urban/City
Center
Urban/City
Center
Urban/City
Center
Additional Ambient Monitoring Information1
Haze, SO2, NO, NO2, NOX, O3, VOC, Carbonyl
compounds, Meteorological parameters, PM Coarse,
Black Carbon, PM10, PM10 Speciation, PM25, and
PM2 5 Speciation.
Carbonyl Compounds, VOC, Meteorological
Parameters, Black carbon, PM10 Speciation, PM25.
CO, SO2, VOC, Carbonyl compounds, O3,
Meteorological parameters, Black Carbon, PM10,
PM10 Speciation, PM25, and PM25 Speciation.
VOC, PM2 5, Carbonyl compounds.
These monitoring sites report additional pollutants to AQS (EPA, 2012h); however, these data are not generated by ERG and are therefore not included in this report.
BOLD ITALICS = EPA-designated NATTS Site.
oo
-------
BXNY is located on the property of Public School 52 (PS 52) in the Bronx Borough of
New York City, northeast of Manhattan. The site was established in 1999 and is considered one
of the premier particulate sampling sites in New York City and is the Bronx (#1) NATTS site.
The surrounding area is urban and residential, as shown in Figure 18-1. The Bruckner
Expressway (1-278) is located a few blocks east of the monitoring site and other heavily traveled
roadways are also located within a few miles of the site. A freight yard and other industries lie on
the southeast and south side of 1-278, part of which can be seen in the lower right-hand side of
Figure 18-1. BXNY is less than 1/2 mile from the East River.
In June 2010, the monitoring instruments at BXNY were moved to a new location nearby
due to roofing construction at the previous location. The new location (MONY) is located at the
Morrisania Neighborhood Family Care center and is 1.65 miles east of the old location. This is
considered the Bronx (#2) NATTS site. MONY is located less than three-quarters of a mile south
of 1-95, one-half mile east of 1-87 and east of the Harlem River, which separates the island of
Manhattan from the Bronx. Part of the Harlem River can be seen in the upper left-hand corner of
Figure 18-2. The Hudson River is just a few blocks farther west. The area surrounding MONY is
primarily residential, although commercial areas are located along Jerome Avenue and East 167th
Street.
Figure 18-5 shows the proximity of BXNY to MONY, as well as the numerous point
sources that are located within 10 miles of the sites. The bulk of the emissions sources are
located to the south and west of the sites, with another cluster to the north. The source categories
with the highest number of emissions sources surrounding BXNY include aircraft operations,
which include airports as well as small runways, heliports, or landing pads; electricity generation
via combustion; schools; and printing and publishing. The point source closest to BXNY is a
wastewater treatment facility while the source closest to MONY is a medical school.
ROCH is located on the east side of Rochester, in western New York, at a power
substation. Rochester is approximately halfway between Syracuse and Buffalo, and Lake
Ontario lies to the north. Although the area north and west of the site is primarily residential, as
shown in Figure 18-3, a railroad transverses the area just south of the site, and 1-590 and 1-490
intersect farther south with commercial areas adjacent to this corridor. The site is used by
researchers from several universities for short-term air monitoring studies and is the Rochester
18-10
-------
NATTS site. As Figure 18-6 shows, the relatively few point sources within a 10-mile radius of
ROCH are located primarily on the west side of the 10-mile radius. The aircraft operations
source category is the category with the highest number of emissions sources surrounding
ROCH, although there are also landfills, chemical manufacturers, printing and publishing
facilities, and heating equipment manufacturers nearby, to name a few.
TONY is located in Tonawanda, New York, north of Buffalo, along the eastern branch of
the Niagara River. The area is wedged between Lake Erie to the south and Lake Ontario to the
north, with the river flowing in-between the two. The monitoring site is located off Grand Island
Boulevard (324), which parallels 1-190, and is less than 1/2 mile from the 1-190 and 1-290
interchange. The surrounding area is industrial and the site itself resides under high power
transmission lines. There are 45 companies regulated by the state of New York within close
proximity of this monitoring site (NYS DEC, 2009), including chemical manufacturers, bulk
terminals/plants, landfills, facilities generating electricity via combustion, a concrete batch plant,
an iron and steel foundry, and a steel mill. Figure 18-7 shows this cluster of point sources
immediately south of TONY. Although the source category with the most sources within
10 miles of TONY is the aircraft operations category, other nearby source categories include a
coke battery, chemical manufacturers, a heating equipment manufacturer, and a rubber and
miscellaneous plastics manufacturer. Note that any possible emissions sources located in Canada
are not provided in Figure 18-7.
Table 18-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the areas surrounding the New
York monitoring sites. Table 18-2 also includes a vehicle registration-to-county population ratio
(vehicles-per-person) for each site. In addition, the population within 10 miles of each site is
presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-level
vehicle registration-to-population ratio to the 10-mile population surrounding each monitoring
site. Table 18-2 also contains annual average daily traffic information. County-level VMT was
not readily available for these sites; thus, daily VMT is not provided in Table 18-2.
18-11
-------
Table 18-2. Population, Motor Vehicle, and Traffic Information for the New York
Monitoring Sites
Site
BXNY
MONY
ROCH
TONY
Estimated
County
Population1
1,386,657
744,389
918,652
County-level
Vehicle
Registration2
248,600
552,184
669,746
Vehicles per
Person
(Registration:
Population)
0.18
0.74
0.73
Population
within 10
miles3'4
6,590,357
639,090
598,180
Estimated
10-mile
Vehicle
Ownership
1,181,520
474,074
436,105
Annual
Average
Daily
Traffic5
100,230
134,421
116,725
74,406
County-
level
Daily
VMT6
NA
NA
NA
1 County-level population estimates reflect data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the New York State Department of Motor Vehicles
(NYSDMV, 2010)
310-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 The 10-mile population estimate for BXNY was used as a surrogate for MONY.
5 Annual Average Daily Traffic reflects 2008 data from the New York State DOT (NYS DOT, 2008)
6 County-level VMT was not available for these sites
BOLD ITALICS = EPA-designaled NATTS Site.
Observations from Table 18-2 include the following:
• Bronx County has the ninth highest county population among counties with NMP
sites, but the 10-mile radius populations for BXNY and MONY are the highest
among all NMP sites. Note that the 10-mile radius for BXNY is used as a surrogate
for MONY. These sites are located 1.65 miles apart.
• County-level vehicle ownership for Bronx County is in the middle of the range
among NMP sites. Although the 10-mile ownership estimate is among the highest for
all NMP sites, given the large population living within 10 miles, the vehicle-per-
person ratio is very low (0.18), which is the lowest vehicle-per-person ratio
calculated. This might seem surprising given the high population, but may be
explained by the use of mass transportation systems.
• The populations surrounding ROCH and TONY are lower than BXNY and MONY.
However, the county-level vehicle ownership is much higher near these sites. The
same is not true of the 10-mile ownership estimate. The population and vehicle
ownership data for ROCH and TONY are in the middle of the range compared to
NMP other sites.
• Among the New York sites, the traffic volume near MONY is the highest and near
TONY is the lowest. Compared to other NMP sites, the traffic volumes near BXNY,
MONY, and ROCH are in the top third while the traffic volume for TONY is just
outside the top third. The traffic data for BXNY were obtained from 1-278 between
1-87 and 1-895; the traffic data for MONY were obtained from 1-87 between the
Bronx Expressway and Macombs Bridge; the traffic data for ROCH were obtained
from 1-490 at 1-590; and the traffic data for TONY were obtained from 1-190 between
Exits 16 and 17.
18-12
-------
18.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in New York on sample days, as well as over the course of the year.
18.2.1 Climate Summary
Weather is somewhat variable in New York City as frontal systems frequently affect the
area. Precipitation is spread fairly evenly throughout the year, with thunderstorms in the summer
and fall and more significant rain or snow events in the winter and spring. The proximity to the
Atlantic Ocean offers a moderating influence from cold outbreaks as well as the summertime
heat. In addition the urban heat island effect tends to keep the city warmer than outlying areas.
Both influences result in a relatively small diurnal range of temperatures. In addition, air sinking
down from the mountains to the west can help drive temperatures higher during warm spells
(Bair, 1992).
Rochester is located in western New York and borders Lake Ontario's south side.
Elevation increases significantly from the shore to the southern-most parts of the city, rising over
800 feet. While the lake acts as a moderating influence on the city's temperatures, both in the
summer and the winter, it also plays a major factor in the city's precipitation patterns. Lake
effect snow enhances the area's snowfall totals, although snowfall rates tend to be higher near
Lake Ontario than farther inland. Spring and summer tend to be sunny while cloudy conditions
are prevalent in the fall and winter (Bair, 1992 and NOAA, 2012c).
Cloudy conditions prevail over the Buffalo area from late autumn through early spring,
and snowy conditions are common. Lake-effect snow events may lead to heavy snowfall, with
heavier snowfalls to the south of Buffalo and closer to the shore than towards the Tonawanda
area. Lake-effect snows tend to diminish after Lake Erie freezes. Because Lake Erie is so cold
(and eventually frozen) during the winter, areas immediately near the shore may be much colder
than farther inland, especially during the spring and summer. Due to the stabilizing effects of the
Lake, the Buffalo area experiences one of the sunniest and driest summers along the northeast
coast. Cooler air passes over the warmer Lake with the arrival of autumn, increasing cloud cover.
Southwesterly winds prevail over the area, but winds off Lake Erie tend to be stronger than
farther inland. Wind direction in Tonawanda can be altered by its proximity to the Niagara River.
18-13
-------
Summer temperature extremes are tempered by the area's location between Lake Erie and Lake
Ontario (Bair, 1992 andNOAA, 2012d).
18.2.2 Meteorological Conditions in 2010
Hourly meteorological data from NWS weather stations nearest these sites were retrieved
for 2010 (NCDC, 2010). The closest weather stations are located at La Guardia International
Airport (near BXNY), Central Park (near MONY), Greater Rochester International Airport (near
ROCH), and Niagara Falls International Airport (near TONY), WBAN 14732, 94728, 14768,
and 04724, respectively. Additional information about these weather stations, such as the
distance between the sites and the weather stations, is provided in Table 18-3. These data were
used to determine how meteorological conditions on sample days vary from normal conditions
throughout the year.
Table 18-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 18-3 is the
95 percent confidence interval for each parameter. As shown in Table 18-3, the sample day
averages for both BXNY and MONY exhibit differences in the average meteorological
parameters from the full-year averages. Note that the monitoring instruments at BXNY were
moved to MONY in June 2010. Thus, conditions on sample days at BXNY appear cooler while
conditions at MONY appear warmer than conditions experienced throughout 2010. While
sampling was discontinued at TONY in July, the differences between the sample day and full-
year averages are less noticeable than for the New York City sites, with the dew point exhibiting
the largest difference. Table 18-3 also shows that meteorological conditions near ROCH, the
only New York monitoring site to sample year-round, were representative of average weather
conditions throughout the year.
18-14
-------
Table 18-3. Average Meteorological Conditions near the New York Monitoring Sites
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Public School 52, New York City, New York - BXNY
La Guardia
Airport
14732
(40.78, -73.88)
2.77
miles
144° (SE)
Central Park
94728
(40.78, 73.97)
4.35
miles
199°
(SSW)
Sample
Day
2010
Sample
Day
2010
57.7
±7.6
60.8
+ 1.8
50.5
±6.5
54.5
+ 1.7
35.5
±6.5
40.5
+ 1.9
43.8
±5.7
48.0
+ 1.6
59.4
±5.6
61.7
+ 1.5
1010.8
±3.5
1016.4
+ 0.8
10.0
±1.4
9.2
+ 0.4
Morrisania, New York City, New York - MONY
66.9
±6.8
60.8
+ 1.8
60.3
±6.4
54.5
+ 1.7
46.2
±6.5
40.5
+ 1.9
53.0
±5.8
48.0
+ 1.6
62.1
±5.3
61.7
+ 1.5
1014.3
±2.1
1016.4
+ 0.8
5.1
±0.6
9.2
+ 0.4
Rochester, New York - ROCH
Greater
Rochester Intl
Airport
14768
(43.12, -77.68)
6.46
miles
240°
(WSW)
Sample
Day
2010
56.7
±5.2
55.4
+ 1.9
49.1
±4.7
47.3
+ 1.7
40.0
±4.5
37.7
+ 1.9
44.7
±4.3
43.0
+ 1.7
73.2
±2.4
71.9
+ 1.2
1014.4
±1.7
1016.5
+ 0.8
7.2
±0.8
7.0
+ 0.3
Tonawanda, New York - TONY
Niagara Falls
Intl Airport
04724
(43.11, -78.95)
8.28
miles
340°
(NNW)
Sample
Day
2010
54.2
±7.8
55.6
+ 2.0
45.9
±7.0
47.8
+ 1.8
35.4
±6.3
38.3
+ 1.8
41.0
±6.2
43.4
+ 1.7
70.1
±3.8
72.1
+ 1.1
1014.0
±2.9
1016.7
+ 0.8
8.4
±1.5
7.9
+ 0.4
oo
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
18.2.3 Back Trajectory Analysis
Figure 18-8 is the composite back trajectory map for days on which samples were
collected at the BXNY monitoring site in 2010. Included in Figure 18-8 are four back trajectories
per sample day. A cluster analysis for BXNY was not performed because there were fewer than
30 sample days. Figure 18-9 is the composite back trajectory map for days on which samples
were collected at MONY and Figure 18-10 is the corresponding cluster analysis. Similarly,
Figures 18-11 through 18-14 are the composite back trajectory maps for days on which samples
were collected at ROCH and TONY and the corresponding cluster analyses. An in-depth
description of these maps and how they were generated is presented in Section 3.5.2.1. For the
composite maps, each line represents the 24-hour trajectory along which a parcel of air traveled
toward the monitoring site on a given sample day and time. For the cluster analyses, each line
corresponds to a back trajectory representative of a given cluster of trajectories. For all maps,
each concentric circle around the sites in Figures 18-8 through 18-14 represents 100 miles.
Figure 18-8. 2010 Composite Back Trajectory Map for BXNY
1 . '.,!.,
: : :
: :
18-16
-------
Figure 18-9. 2010 Composite Back Trajectory Map for MONY
Figure 18-10. Back Trajectory Cluster Map for MONY
18-17
-------
Figure 18-11. 2010 Composite Back Trajectory Map for ROCH
Figure 18-12. Back Trajectory Cluster Map for ROCH
18-18
-------
Figure 18-13. 2010 Composite Back Trajectory Map for TONY
Figure 18-14. Back Trajectory Cluster Map for TONY
18-19
-------
Observations from Figure 18-8 for BXNY include the following:
• Back trajectories originated from a variety of directions at BXNY, although less
frequently from the east and southeast. A cluster of trajectories originated from the
west to northwest and a second cluster originated from the south. Note that sampling
at BXNY was discontinued in June 2010.
• The 24-hour air shed domain for BXNY was somewhat larger in size compared to
other NMP sites, as the farthest away a trajectory originated was nearly 700 miles to
the southeast, off the North Carolina coast and over the Atlantic Ocean. The longest
trajectory originated from the south-southeast and is associated with a strong low
pressure system that moved through the region January 25-26, 2010. However, the
average trajectory length was 263 miles and more than 81 percent of trajectories
originated within 400 miles of the site.
Observations from Figures 18-9 and 18-10 for MONY include the following:
• Back trajectories originated from a variety of directions at MONY.
• The 24-hour air shed domain for MONY is among the largest in size compared to
other NMP sites. Although the farthest away a trajectory originated was over
northwest Indiana, or less than 650 miles away, the average trajectory length was 300
miles and 86 percent of trajectories originated within 500 miles of the site.
• The cluster analysis shows that nearly 40 percent of back trajectories originated to the
west of MONY over western New York and Pennsylvania and generally within
300 miles of the site. Longer trajectories originating to the west of MONY and over
the Great Lakes, Michigan, and northern Ohio account for another 14 percent of
trajectories. Trajectories originating from the northwest, north, and east were also
common.
• Figures 18-9 and 18-10 include back trajectories from June to December 2010 only,
based on the start date of the sampling effort at MONY.
Observations from Figures 18-11 and 18-12 for ROCH include the following:
• Back trajectories originated from a variety of directions at ROCH, although very few
originated from the southeast of ROCH.
• The 24-hour air shed domain for ROCH was comparable in size to other NMP sites.
The farthest away a trajectory originated was over western Lake Superior, or less than
700 miles away. However, the average trajectory length was 264 miles and 86 percent
of trajectories originated within 400 miles of the site.
• The cluster analysis shows that the bulk (43 percent) of trajectories originated to the
west of ROCH. These include shorter trajectories originating over the Great Lakes
and within 300 miles of ROCH (as shown by the 30 percent cluster) and longer
trajectories originating over Lake Superior, Michigan (including the Upper
18-20
-------
Peninsula), and Lake Michigan (as shown by the 13 percent cluster). Back trajectories
originating from the southwest of ROCH accounted for another 19 percent of
trajectories, with shorter trajectories (15 percent) originating over western
Pennsylvania, northeastern Ohio, and northern West Virginia, and longer trajectories
(4 percent) originating over western Ohio and Indiana. Back trajectories also
originated over Ontario and Quebec, Canada and the eastern half of New York.
Observations from Figures 18-13 and 18-14 for TONY include the following:
• Similar to ROCH, back trajectories originated from a variety of directions at TONY,
although infrequently from the east and southeast.
• The 24-hour air shed domain for TONY was comparable in size to ROCH as well as
other NMP sites. The farthest away a trajectory originated was over central Illinois, or
less than 600 miles away. However, the average trajectory length was 251 miles and
88 percent of trajectories originated within 400 miles of the site.
• The cluster analysis shows that trajectories originating from the southwest, west,
northwest of TONY account for the bulk of back trajectories. Back trajectories also
originated from the north and northeast over Ontario and Quebec, Canada, as well as
a few originating over western New York State, which are included in the cluster
trajectory accounting for 33 percent of trajectories.
• Figures 18-13 and 18-14 include back trajectories from January to July 2010 only,
based on the stop date of the sampling effort at TONY.
18.2.4 Wind Rose Comparison
Hourly wind data from the weather stations at La Guardia International Airport (for
BXNY), Central Park (for MONY), Greater Rochester International Airport (for ROCH), and
Niagara Falls International Airport (for TONY) were uploaded into a wind rose software
program to produce customized wind roses, as described in Section 3.5.2.2. A wind rose shows
the frequency of wind directions using "petals" positioned around a 16-point compass, and uses
different colors to represent wind speeds.
Figure 18-15 presents three different wind roses for the BXNY monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
18-21
-------
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figures 18-16 through
18-18 presents the three wind roses and distance maps for MONY, ROCH, and TONY.
Observations from Figure 18-15 for BXNY include the following:
• The La Guardia International Airport weather station is located across the East River
from BXNY, approximately 2.75 miles southeast of the monitoring site.
• The historical wind rose shows that winds from the southwest, northwest, and
northeast quadrants were frequently observed, while winds from the southeast
quadrant were rarely observed. Among these wind directions, northwesterly and
southerly winds were observed the most. Calm winds (<2 knots) were observed for
less than four percent of the hourly measurements near BXNY, while the strongest
winds were most frequently observed with westerly to northwesterly winds.
• Although west-northwesterly and northwesterly winds accounted for a higher
percentage of wind observations in 2010 and east-northeasterly winds were hardly
observed at all, these differences do not detract from the many similarities in the wind
patterns between the 2010 wind rose and the historical wind rose.
• An even higher percentage of west-northwesterly and northwesterly winds were
observed on sample days in 2010 compared to throughout 2010 and historically, as
were northeasterly winds. However, the sample day wind rose includes sample days
from January to June 2010 only, and may reflect a seasonal pattern.
Observations from Figure 18-16 for MONY include the following:
• The weather station at Central Park is located 4.35 miles south-south west of MONY.
• The historical wind rose shows that winds from the west and west-northwest account
for 25 percent of wind observations. Winds from the northeast and east-northeast
account for another 15 percent of observations. Calm winds (<2 knots) were observed
for greater than 12 percent of the hourly measurements near MONY. Note the wind
speed differences between the wind roses for BXNY and MONY. The Central Park
weather station is more protected by the city than the La Guardia weather station,
which is located right on the East River bank.
• The 2010 full-year wind rose shares many similarities with the historical wind rose,
although westerly and west-northwesterly winds accounted for approximately
36 percent of wind observations in 2010, compared to 25 percent historically. There
were hardly any northwesterly or southerly wind observations in 2010 compared to
the historical wind rose.
• While westerly and west-northwesterly winds accounted for the majority of wind
observations on sample days in 2010, the percentage (30 percent) is a less than for the
18-22
-------
full-year wind rose (36 percent). The number of northeasterly wind observations is
higher on samples, which may be a seasonal pattern as the sample day wind rose
includes sample days from June to December 2010 only.
Observations from Figure 18-17 for ROCH include the following:
• The Rochester International Airport weather station is located approximately
6.5 miles west-southwest of ROCH, with much of the southern half of the city of
Rochester between them.
• The historical wind rose shows that winds from the south-southwest to west were
frequently observed, while winds from other directions were infrequently observed.
Calm winds were observed for less than 10 percent of the hourly measurements near
ROCH, while the strongest winds were most frequently observed with west-
southwesterly and westerly winds.
• The wind patterns shown on the 2010 wind rose are similar to the historical wind
patterns for ROCH, although westerly and west-northwesterly winds were observed
more frequently and south-southwesterly winds were observed less often. A slightly
higher percentage of calm winds were observed in 2010.
• The sample day wind patterns are similar to those shown on the full-year wind rose,
indicating that conditions on sample days were representative of those experienced
over the entire year in 2010.
Observations from Figure 18-18 for TONY include the following:
• The Niagara Falls weather station is located 8.3 miles north-northwest of TONY.
Grand Island and the Niagara River lie between the site and the weather station.
• The wind patterns for TONY resemble the wind patterns for ROCH.
• The historical wind rose shows that winds from the south to southwest to west were
the most frequently observed wind directions. Calm winds account for approximately
10 percent of the hourly measurements near TONY. The strongest winds were most
frequently observed with southwesterly, west-southwesterly, and westerly winds,
those generally flowing off Lake Erie.
• The wind patterns shown on the 2010 wind rose are similar to the historical wind
patterns for TONY, although there were fewer southerly and south-southwesterly
winds and a higher percentage of calms. This indicates that conditions in 2010 were
similar to those experienced historically.
• While southwest was also the prevalent wind direction on sample days in 2010, winds
from the northwest quadrant accounted for roughly the same percentage of wind
observations as the southwest quadrant. Recall the sampling at TONY was
discontinued in July 2010, thereby missing half of the year and perhaps revealing a
seasonal pattern.
18-23
-------
Figure 18-15. Wind Roses for the LaGuardia International Airport Weather Station near
BXNY
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between BXNY and NWS Station
20%
...
*
18-24
-------
Figure 18-16. Wind Roses for the Central Park Weather Station near MONY
1999-2009 Historical Wind Rose
2010 Wind Rose
WIND SPEED
(Knots)
O -22
JH 17-21
^| 11 • 17
HI 7- 11
EH 4-7
!• 2- 4
Calms: 12.08%
".;•-
25%
**\ 20%
15%
M 0%
WIND SPEED
(Knots)
n -22
H 17 - 21
^| 11 • 17
|H 7- 11
EH 4-7
!• 2- 4
Calms: 12.88%
2010 Sample Day Wind Rose
Distance between MONY and NWS Station
18-25
-------
Figure 18-17. Wind Roses for the Greater Rochester International Airport Weather Station
near ROCH
1999-2009 Historical Wind Rose
2010 Wind Rose
Calms; 8.51%
2010 Sample Day Wind Rose
Distance between ROCH and NWS Station
18-26
-------
Figure 18-18. Wind Roses for the Niagara Falls International Airport Weather Station
near TONY
2002-2009 Historical Wind Rose
2010 Wind Rose
IWEST
2010 Sample Day Wind Rose
Distance between TONY and NWS Station
18-27
-------
18.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the New York monitoring sites
in order to allow analysts and readers to focus on a subset of pollutants through the context of
risk. For each site, each pollutant's preprocessed daily measurement was compared to its
associated risk screening value. If the concentration was greater than the risk screening value,
then the concentration "failed the screen." Pollutants of interest are those for which the
individual pollutant's total failed screens contribute to the top 95 percent of the site's total failed
screens. In addition, if any of the NATTS MQO Core Analytes measured by each monitoring site
did not meet the pollutant of interest criteria based on the preliminary risk screening, that
pollutant was added to the list of site-specific pollutants of interest. A more in-depth description
of the risk screening process is presented in Section 3.2.
Table 18-4 presents the pollutants of interest for the New York monitoring sites. The
pollutants that failed at least one screen and contributed to 95 percent of the total failed screens
for each monitoring site are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of
interest are shaded and/or bolded. BXNY, MONY, and ROCH sampled for hexavalent
chromium and PAH while TONY sampled only for PAH.
Observations from Table 18-4 include the following:
• The number of failed screens is significantly higher for the two New York City sites
than for ROCH and TONY. However, many of the pollutants that failed screens for
the New York City sites only failed one screen. A review of the data shows that the
majority of these failed screens were from a single sample. For BXNY, the January
14, 2010 sample resulted in many failed screens; for MONY, it was the
December 28, 2010 sample. These two samples are discussed in further detail in the
next section.
• Naphthalene failed the most screens for each site. Naphthalene accounts for between
40 percent (MONY) and 100 percent (ROCH) of each site's total failed screens.
• For BXNY and MONY, 14 pollutants, of which three are NATTS MQO Core
Analytes, failed screens. For these sites, the risk screening process identified all of the
pollutants that failed a screen as pollutants of interest. This is because so many
pollutants contributed equally to the total number of failed screens (by failing only
one screen, in this instance) leading up to the 95 percent criteria that they are all
considered pollutants of interest. This criterion is discussed in more detail in
Section 3.2.
18-28
-------
Table 18-4. Risk Screening Results for the New York Monitoring Sites
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
PS 52, New York City, New York - BXNY
Naphthalene
Fluorene
Acenaphthene
Benzo(a)pyrene
Fluoranthene
Acenaphthylene
Benzo(a)anthracene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Dibenz(a,h)anthracene
Hexavalent Chromium
Indeno( 1,2,3 -cd)pyrene
0.029
0.011
0.011
0.00057
0.011
0.011
0.0057
0.0057
0.011
0.011
0.0057
0.00052
0.000083
0.0057
Total
28
6
5
3
3
1
1
1
1
1
1
1
1
1
54
28
28
28
26
28
20
25
28
28
28
24
10
25
28
354
100.00
21.43
17.86
11.54
10.71
5.00
4.00
3.57
3.57
3.57
4.17
10.00
4.00
3.57
15.25
51.85
11.11
9.26
5.56
5.56
.85
.85
.85
.85
.85
.85
.85
.85
.85
51.85
62.96
72.22
77.78
83.33
85.19
87.04
88.89
90.74
92.59
94.44
96.30
98.15
100.00
Morrisania, New York City, New York - MONY
Naphthalene
Fluorene
Acenaphthene
Fluoranthene
Acenaphthylene
Benzo(a)pyrene
Benzo(a)anthracene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Dibenz(a,h)anthracene
Hexavalent Chromium
Indeno( 1,2,3 -cd)pyrene
0.029
0.011
0.011
0.011
0.011
0.00057
0.0057
0.0057
0.011
0.011
0.0057
0.00052
0.000083
0.0057
Total
29
13
12
7
2
2
1
1
1
1
1
1
1
1
73
29
29
29
29
20
25
15
29
29
29
24
6
27
28
348
100.00
44.83
41.38
24.14
10.00
8.00
6.67
3.45
3.45
3.45
4.17
16.67
3.70
3.57
20.98
39.73
17.81
16.44
9.59
2.74
2.74
.37
.37
.37
.37
.37
.37
.37
.37
39.73
57.53
73.97
83.56
86.30
89.04
90.41
91.78
93.15
94.52
95.89
97.26
98.63
100.00
Rochester, New York - ROCH
Naphthalene
0.029
Total
2
2
3
3
66.67
66.67
100.00
100.00
Tonawanda, New York - TONY
Naphthalene
Fluorene
Fluoranthene
Acenaphthene
Benzo(a)pyrene
Acenaphthylene
0.029
0.011
0.011
0.011
0.00057
0.011
Total
26
13
9
7
4
2
61
29
29
29
29
29
29
174
89.66
44.83
31.03
24.14
13.79
6.90
35.06
42.62
21.31
14.75
11.48
6.56
3.28
42.62
63.93
78.69
90.16
96.72
100.00
18-29
-------
• For ROCH, only naphthalene failed screens. Hexavalent chromium and
benzo(a)pyrene were added as pollutants of interest for ROCH because they are
NATTS MQO Core Analytes, even though they did not fail any screens. These two
pollutants are not shown in Table 18-4.
• Note the relatively low number of measured detections shown in Table 18-4 for
ROCH compared to the other sites. Recall from Section 2.4 that problems with the
PAH sampler at ROCH led to the invalidation of nearly all of ROCH's PAH data for
2010. The sampler was re-certified at the end of 2010 and the final three samples
from 2010 were retained.
• Six PAH, of which two are NATTS MQO Core Analytes, failed screens for TONY.
Five of the six pollutants were identified as pollutants of interest by the risk screening
process, including the two NATTS MQO Core Analytes, naphthalene and
benzo(a)pyrene.
18.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the New York monitoring sites. Concentration averages are provided for the pollutants of
interest for each New York site, where applicable. Concentration averages for select pollutants
are also presented graphically for each site, where applicable, to illustrate how each site's
concentrations compare to the program-level averages. In addition, concentration averages for
select pollutants are presented from previous years of sampling in order to characterize
concentration trends at each site, where applicable. Additional site-specific statistical summaries
are provided in Appendices M and O.
18.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each New York site, as described in Section 3.1. The quarterly average of a particular
pollutant is simply the average concentration of the preprocessed daily measurements over a
given calendar quarter. Quarterly average concentrations include the substitution of zeros for all
non-detects. A site must have a minimum of 75 percent valid samples of the total number of
samples possible within a given quarter for a quarterly average to be calculated. An annual
average includes all measured detections and substituted zeros for non-detects for the entire year
of sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the New York
18-30
-------
monitoring sites are presented in Table 18-5, where applicable. Note that if a pollutant was not
detected in a given calendar quarter, the quarterly average simply reflects "0" because only zeros
substituted for non-detects were factored into the quarterly average concentration.
Table 18-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the New York Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(ng/m3)
2nd
Quarter
Average
(ng/m3)
3rd
Quarter
Average
(ng/m3)
4th
Quarter
Average
(ng/m3)
Annual
Average
(ng/m3)
PS 52, New York City, New York - BXNY
Acenaphthene
Acenaphthylene
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Dibenz(a,h)anthracene
Fluoranthene
Fluorene
Hexavalent Chromium
Indeno( 1,2,3 -cd)pyrene
Naphthalene
28/28
20/28
25/28
26/28
28/28
28/28
28/28
24/28
10/28
28/28
28/28
25/28
28/28
28/28
3.69
±2.21
12.37
± 19.84
1.69
±2.82
1.82
±3.04
2.20
±3.12
1.14
±1.53
1.89
±2.41
0.75
±1.12
0.12
±0.19
11.67
± 14.69
11.83
± 15.07
0.04
±0.01
1.58
±2.31
174.96
± 148.30
9.85
±3.21
0.51
±0.46
0.16
±0.08
0.16
±0.08
0.38
±0.12
0.22
±0.07
0.31
±0.10
0.08
±0.03
0.01
±0.01
5.61
±1.49
11.14
±3.28
0.04
±0.02
0.23
±0.08
141.95
±32.05
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
18-31
-------
Table 18-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the New York Monitoring Sites (Continued)
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(ng/m3)
2nd
Quarter
Average
(ng/m3)
3rd
Quarter
Average
(ng/m3)
4th
Quarter
Average
(ng/m3)
Annual
Average
(ng/m3)
Morrisania, New York, New York - MONY
Acenaphthene
Acenaphthylene
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Dibenz(a,h)anthracene
Fluoranthene
Fluorene
Hexavalent Chromium
Indeno( 1,2,3 -cd)pyrene
Naphthalene
29/29
20/29
15/29
25/29
29/29
29/29
29/29
24/29
6/29
29/29
29/29
27/27
28/29
29/29
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
16.52
±3.48
0.37
±0.31
0.02
±0.03
0.05
±0.02
0.21
±0.04
0.12
±0.03
0.19
±0.05
0.04
±0.02
0
9.66
±1.94
16.79
±3.87
0.03
±0.01
0.12
±0.04
146.07
± 18.90
5.95
±2.00
14.87
±23.63
2.60
±4.88
3.06
±5.83
3.11
±5.24
1.80
±3.00
2.51
±3.93
1.04
±1.83
0.25
±0.48
9.88
±11.20
11.93
± 12.35
0.05
±0.01
2.41
±4.16
198.66
± 155.29
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Rochester, New York - ROCH
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
2/3
35/59
3/3
NA
<0.01
±<0.01
NA
NA
0.01
±0.01
NA
NA
0.01
±0.01
NA
NA
0.01
±0.01
NA
NA
0.01
±<0.01
NA
Tonawanda, New York - TONY
Acenaphthene
Benzo(a)pyrene
Fluoranthene
Fluorene
Naphthalene
29/29
29/29
29/29
29/29
29/29
1.31
±0.52
0.12
±0.07
4.46
±3.29
8.56
±6.11
354.79
±266.39
12.02
±5.16
0.40
±0.18
10.91
±4.46
17.39
±6.13
519.06
± 197.95
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
18-32
-------
Observations from Table 18-5 include the following:
• Annual average concentrations for the pollutants of interest for BXNY, MONY, and
TONY could not be calculated due to the abbreviated sampling period at each site.
Recall that the sampling equipment was moved from BXNY to MONY in mid-2010
while sampling was discontinued at TONY in July 2010. In addition, annual average
concentrations for the PAH for ROCH could not be calculated due to problems with
the sampler. Thus, the only pollutant for which an annual average concentration could
be calculated is hexavalent chromium for ROCH. However, Appendices M and O
provide the pollutant-specific average concentration for all valid samples collected
over the entire sample period for each site.
• Both the annual average and the quarterly average concentrations of hexavalent
chromium for ROCH were approximately 0.01 ng/m3. Among NMP sites sampling
hexavalent chromium, the annual average concentration for ROCH is among the
lowest, ranking 16th.
• With the exception of acenaphthene, fluorene, and naphthalene, the first quarter
average concentrations of the PAH for BXNY are an order of magnitude higher than
the second quarter averages and the confidence intervals indicate the inclusion of
outliers. The concentrations of the PAH for the sample collected January 14, 2010 are
the maximum concentrations for most of these pollutants and in many cases is an
order of magnitude higher than any other measurement and, in many cases, is one of
the highest concentrations among all NMP sites sampling PAH. For example, the
concentration of benzo(a)pyrene on January 14, 2010 was 22.4 ng/m3, and is more
than 12 times the next highest concentration of this pollutant measured at BXNY
(1.85 ng/m3 measured on February 1, 2010). This is also the second highest
measurement of this pollutant among all NMP sites sampling PAH (behind only
MONY, discussed below). The concentration of acenaphthyelene on
January 14, 2010 was 147 ng/m3, and is more than 18 times the next highest
concentration of this pollutant measured at BXNY (8.04 ng/m3 measured on
February 25, 2010). This is also the second highest measurement of this pollutant
among all NMP sites sampling PAH (also behind MONY).
• Although the first and second quarter average concentrations of naphthalene and
fluorene for BXNY were not significantly different, the highest concentrations of
these pollutants were also measured on January 14, 2010. The concentration of
naphthalene on January 14, 2010 was 1,170 ng/m3, and is more than four times the
next highest concentration of this pollutant (255 ng/m3 measured on April 8, 2010).
This is the fourth highest measurement of this pollutant among all NMP sites
sampling PAH (behind two concentrations from TONY and one from MONY). The
concentration of fluorene on January 14, 2010 was 114 ng/m3, and is more than five
times the next highest concentration of this pollutant (22.2 ng/m3 measured on
May 26, 2010). This is the third highest measurement of this pollutant among all
NMP sites sampling PAH (behind two concentrations from DEMI).
• The only pollutant for which the maximum concentration was not measured on
January 14, 2010 at BXNY was acenaphthene. The difference is evident in the
quarterly averages, where the second quarter average is nearly three times the first
18-33
-------
quarter average. The maximum concentration of acenaphthene was measured on
3
May 26, 2010 and was 20.9 ng/m , although a similar concentration was also
measured on April 8, 2010 (19.3 ng/m3). Note that these are the same days that the
second highest concentrations of naphthalene and fluorene were measured,
respectively. The third highest measurement of this pollutant was measured on
January 14, 2010 (16.2 ng/m3).
• In most cases, the fourth quarter average concentrations of the PAH for MONY are
an order of magnitude higher than the third quarter averages and the confidence
intervals indicate the inclusion of outliers. The concentrations of the PAH for the
sample collected December 28, 2010 are the maximum concentrations for most of
these pollutants and in many cases are an order of magnitude higher than any other
measurements and are among the highest concentrations measured at NMP sites
sampling PAH. For example, the concentration of benzo(a)pyrene on
December 28, 2010 was 42.7 ng/m3, and is two orders of magnitude higher than the
next highest concentration of this pollutant (0.597 ng/m3 measured on October 5,
2010). This is also the highest measurement of this pollutant among all NMP sites
sampling PAH . The concentration of acenaphthyelene on December 28, 2010 was
175 ng/m3, and is nearly 13 times the next highest concentration of this pollutant
(13.7 ng/m3 measured on December 10, 2010). This is also the highest measurement
of this pollutant among all NMP sites sampling PAH.
• Although the third and fourth quarter average concentrations of naphthalene,
fluoranthene, and fluorene for MONY were not as drastically different, the highest
concentrations of these pollutants were also measured on December 28, 2010. The
concentration of naphthalene on December 28, 2010 was 1,240 ng/m3, and is more
than six times the next highest concentration of this pollutant (200 ng/m3 measured on
August 18, 2010). This is the third highest measurement of this pollutant among all
NMP sites sampling PAH (behind two concentrations from TONY). The
concentration of fluorene on December 28, 2010 was 95.2 ng/m3, and is more than
three times the next highest concentration of this pollutant (29.7 ng/m3 measured on
July 19, 2010). This is the fourth highest measurement of this pollutant among all
NMP sites sampling PAH (behind two concentrations from DEMI and one from
BXNY). The concentration of fluoranthene on December 28, 2010 was 85.3 ng/m3,
and is more than five times the next highest concentration of this pollutant
(15.6 ng/m3 also measured on July 19, 2010). This is the second highest measurement
of this pollutant among all NMP sites sampling PAH (behind only BXNY).
• The only pollutant for which the maximum concentration was not measured on
December 28, 2010 at MONY was acenaphthene. The difference is evident in the
quarterly averages, where the third quarter average is nearly three times the fourth
quarter average. The maximum concentration of acenaphthene was measured on
July 29, 2010 and was 29.3 ng/m3. Note that this is the same day that the second
highest concentration of fluorene and fluoranthene were measured.
• Even with the outliers of naphthalene measured at BXNY and MONY, the quarterly
average naphthalene concentrations for TONY are significantly higher than for
BXNY and MONY, although very large confidence intervals are associated with
18-34
-------
them. The highest concentration of naphthalene was measured at TONY on
January 26, 2010 (1,490 ng/m3), although the next highest concentration of this
pollutant was of a similar magnitude (1,390 ng/m3 measured on May 14, 2010). These
are the two highest measurements of this pollutant among all NMP sites sampling
PAH. Of the 15 concentrations of naphthalene greater than 500 ng/m3, nine were
measured at TONY.
• The second quarter average concentrations of several pollutants of interest for TONY
were higher than the first quarter averages and have relatively large confidence
intervals associated with them. The highest concentration of acenaphthene was
measured on May 26, 2010 (38.3 ng/m3). Of the 12 concentrations of acenaphthene
greater than 5 ng/m3 measured at TONY, 11 were measured during the second quarter
of 2010. Of the 14 concentrations of fluorene greater than 10 ng/m3 measured at
TONY, all were measured during the second quarter of 2010. Similar patterns exist
for the remaining pollutants of interest.
18.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, a box plot for hexavalent chromium was
created for ROCH. Recall that annual average concentrations for the other sites' pollutants of
interest could not be calculated. Figure 18-19 overlays the site's minimum, annual average, and
maximum concentrations onto the program-level minimum, first quartile, average, median, third
quartile, and maximum concentrations, as described in Section 3.5.3.
Figure 18-19. Program vs. Site-Specific Average Hexavalent Chromium Concentration
Program Max Concentration =3.51 ng/m3
0.3 0.45
Concentration (ng/m3)
Program: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Average
Site: Site Average Site Minimum/Maximum
o
18-35
-------
Observations from Figure 18-19 include the following:
• Figure 18-19 is the box plot for hexavalent chromium. The scale has been
adjusted as a result of a relatively large maximum concentration. The program-
level maximum concentration (3.51 ng/m3) is not shown directly on the box plot
in order to allow for the observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on the box plot.
• The annual average concentration of hexavalent chromium for ROCH is well
below the program-level average. The maximum concentration measured at
ROCH is less than the average concentration at the program level and is just
greater than the program-level 75th percentile. Several non-detects of hexavalent
chromium were measured at ROCH.
18.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. The New York monitoring sites have not sampled continuously for 5 years as part
of the NMP; therefore, a trends analysis was not conducted.
18.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at each New
York monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations regarding
the various risk factors, time frames, and calculations associated with these risk screenings.
18.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
New York monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
were compared to the acute MRL; the quarterly averages were compared to the intermediate
MRL; and the annual averages were compared to the chronic MRL.
18-36
-------
Of the pollutants sampled for at the New York sites, only naphthalene and hexavalent
chromium have ATSDR MRLs. None of the measured detections or time-period average
concentrations of naphthalene and hexavalent chromium, where they could be calculated for the
New York monitoring sites, were greater than their respective MRL noncancer health risk
benchmarks.
18.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the New York monitoring sites and where annual
average concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 18-6, where applicable.
Observations for New York sites from Table 18-6 include the following:
• As discussed in previous sections, annual average concentrations, and thus cancer and
noncancer risk approximations, could not be calculated for BXNY, MONY, or
TONY.
• In addition, annual average concentrations, and thus cancer and noncancer risk
approximations, could not be calculated for the PAH for ROCH.
• Based on the annual average hexavalent chromium concentration for ROCH, the
cancer and noncancer surrogate risk approximations are well below the levels of
concern.
18-37
-------
Table 18-6. Cancer and Noncancer Surrogate Risk Approximations for the New York
Monitoring Sites
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
PS 52, New York City, New York - BXNY
Acenaphthene
Acenaphthylene
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Dibenz(a,h)anthracene
Fluoranthene
Fluorene
Hexavalent Chromium
Indeno( 1,2,3 -cd)pyrene
Naphthalene
0.000088
0.000088
0.000176
0.00176
0.000176
0.000088
0.000088
0.000176
0.0019184
0.000088
0.000088
0.012
0.000176
0.000034
0.0001
_
0.003
28/28
20/28
25/28
26/28
28/28
28/28
28/28
24/28
10/28
28/28
28/28
25/28
28/28
28/28
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Morrisania, New York City, New York - MONY
Acenaphthene
Acenaphthylene
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
0.000088
0.000088
0.000176
0.00176
0.000176
0.000088
0.000088
0.000176
29/29
20/29
15/29
25/29
29/29
29/29
29/29
24/29
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA = Not available due to the duration criteria for calculating an annual average.
- = a Cancer URE or Noncancer RfC is not available.
18-38
-------
Table 18-6. Cancer and Noncancer Surrogate Risk Approximations for the New York
Monitoring Sites (Continued)
Pollutant
Dibenz(a,h)anthracene
Fluoranthene
Fluorene
Hexavalent Chromium
Indeno( 1,2,3 -cd)pyrene
Naphthalene
Cancer
URE
(Hg/m3)1
0.0019184
0.000088
0.000088
0.012
0.000176
0.000034
Noncancer
RfC
(mg/m3)
0.0001
0.003
#of
Measured
Detections
vs. # of
Samples
6/29
29/29
29/29
27/27
28/29
29/29
Annual
Average
(Hg/m3)
NA
NA
NA
NA
NA
NA
Cancer Risk
Approximation
(in-a-million)
NA
NA
NA
NA
NA
NA
Noncancer Risk
Approximation
(HQ)
NA
NA
NA
NA
NA
NA
Rochester, New York - ROCH
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
0.00176
0.012
0.000034
0.0001
0.003
2/3
35/59
3/3
NA
0.01
±0.01
NA
NA
0.14
NA
NA
0.01
NA
Tonawanda, New York - TONY
Acenaphthene
Benzo(a)pyrene
Fluoranthene
Fluorene
Naphthalene
0.000088
0.00176
0.000088
0.000088
0.000034
_
0.003
29/29
29/29
29/29
29/29
29/29
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA = Not available due to the duration criteria for calculating an annual average.
— = a Cancer URE or Noncancer RfC is not available.
18.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 18-7 and 18-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 18-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer risk approximations (in-a-million), as calculated from the annual averages. Table 18-8
presents similar information, but identifies the 10 pollutants with the highest noncancer risk
approximations (HQ), also calculated from annual averages.
18-39
-------
Table 18-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the New York Monitoring Sites
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Cancer Risk
Approximation
Pollutant (in-a-million)
PS 52, New York City, New York (Bronx County) - BXNY
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
Naphthalene
1,3 -Butadiene
Dichloromethane
POM, Group 2b
Tetrachloroethylene
POM, Group la
225.51
171.25
104.61
62.07
23.44
22.86
12.53
2.17
1.28
0.93
Benzene
Formaldehyde
Naphthalene
1,3 -Butadiene
Ethylbenzene
POM, Group 3
Hexavalent Chromium, PM
Arsenic, PM
POM, Group 2b
Acetaldehyde
1.76E-03
1.36E-03
7.97E-04
6.86E-04
4.28E-04
3.35E-04
3.18E-04
2.23E-04
1.91E-04
1.37E-04
Morrisania, New York City, New York (Bronx County) - MONY
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
Naphthalene
1,3 -Butadiene
Dichloromethane
POM, Group 2b
Tetrachloroethylene
POM, Group la
225.51
171.25
104.61
62.07
23.44
22.86
12.53
2.17
1.28
0.93
Benzene
Formaldehyde
Naphthalene
1,3 -Butadiene
Ethylbenzene
POM, Group 3
Hexavalent Chromium, PM
Arsenic, PM
POM, Group 2b
Acetaldehyde
1.76E-03
1.36E-03
7.97E-04
6.86E-04
4.28E-04
3.35E-04
3.18E-04
2.23E-04
1.91E-04
1.37E-04
oo
-k
o
-------
Table 18-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the New York Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Cancer Risk
Approximation
Pollutant (in-a-million)
Rochester, New York (Monroe County) - ROCH
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Dichloromethane
Naphthalene
POM, Group 2b
Tetrachloroethylene
POM, Group 6
397.57
198.79
197.58
106.73
53.16
47.06
29.93
6.47
1.62
0.67
Benzene
Formaldehyde
1,3 -Butadiene
POM, Group 3
Naphthalene
Hexavalent Chromium, PM
Arsenic, PM
POM, Group 2b
Ethylbenzene
POM, Group 5a
3.10E-03
2.57E-03
1.59E-03
1.36E-03
1.02E-03
7.52E-04
5.79E-04
5.70E-04
4.97E-04
3.51E-04
Hexavalent Chromium 0.14
Tonawanda, New York (Erie County) - TONY
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
Coke Oven Emissions, PM
POM, Group 2b
Tetrachloroethylene
556.67
260.72
240.93
131.01
62.38
38.68
18.34
8.35
7.77
2.20
Coke Oven Emissions, PM
Benzene
Formaldehyde
1,3 -Butadiene
Hexavalent Chromium, PM
POM, Group 3
Naphthalene
POM, Group 2b
Ethylbenzene
Arsenic, PM
8.27E-03
4.34E-03
3.39E-03
1.87E-03
1.59E-03
1.56E-03
1.32E-03
6.83E-04
6.02E-04
4.26E-04
oo
-------
Table 18-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the New York Monitoring Sites
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer Toxicity
Weight
Top 10 Noncancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
PS 52, New York City, New York (Bronx County) - BXNY
Methanol
Toluene
Xylenes
Benzene
Hexane
Ethylene glycol
Ethylbenzene
Formaldehyde
Acetaldehyde
Hydrochloric acid
808.80
801.90
456.53
225.51
197.82
194.12
171.25
104.61
62.07
55.88
Acrolein
1,3 -Butadiene
Formaldehyde
Naphthalene
Benzene
Acetaldehyde
Xylenes
Cadmium, PM
Arsenic, PM
Hydrochloric acid
999,876.74
11,427.60
10,674.18
7,813.40
7,517.03
6,896.87
4,565.34
3,703.33
3,455.69
2,793.99
Morrisania, New York City, New York (Bronx County) - MONY
Methanol
Toluene
Xylenes
Benzene
Hexane
Ethylene glycol
Ethylbenzene
Formaldehyde
Acetaldehyde
Hydrochloric acid
808.80
801.90
456.53
225.51
197.82
194.12
171.25
104.61
62.07
55.88
Acrolein
1,3 -Butadiene
Formaldehyde
Naphthalene
Benzene
Acetaldehyde
Xylenes
Cadmium, PM
Arsenic, PM
Hydrochloric acid
999,876.74
11,427.60
10,674.18
7,813.40
7,517.03
6,896.87
4,565.34
3,703.33
3,455.69
2,793.99
oo
-k
to
-------
Table 18-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the New York Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer Toxicity
Weight
Top 10 Noncancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Rochester, New York (Monroe County) - ROCH
Toluene
Xylenes
Methanol
Benzene
Hexane
Hydrochloric acid
Ethylbenzene
Formaldehyde
Acetaldehyde
Ethylene glycol
944.15
669.05
479.68
397.57
220.52
200.68
198.79
197.58
106.73
104.78
Acrolein
1,3 -Butadiene
Formaldehyde
Benzene
Acetaldehyde
Hydrochloric acid
Naphthalene
Arsenic, PM
Chlorine
Manganese, PM
641,820.42
26,578.28
20,160.87
13,252.21
11,858.72
10,033.82
9,976.12
8,971.06
7,563.24
7,454.48
Hexavalent Chromium <0 . 0 1
Tonawanda, New York (Erie County) - TONY
Toluene
Xylenes
Benzene
Methanol
Hexane
Formaldehyde
Ethylbenzene
Carbon disulfide
Acetaldehyde
Ethylene glycol
1,219.83
794.76
556.67
532.33
275.55
260.72
240.93
175.99
131.01
127.73
Acrolein
1,3 -Butadiene
Formaldehyde
Benzene
Acetaldehyde
Naphthalene
Cadmium, PM
Manganese, PM
Lead, PM
Xylenes
858,961.34
31,188.33
26,603.88
18,555.79
14,556.63
12,893.95
12,029.93
10,377.91
10,244.59
7,947.63
oo
-------
The pollutants in these tables are limited to those that have cancer and noncancer risk
factors, respectively. As a result, although the actual value of the emissions is the same, the
highest emitted pollutants in the cancer table may be different from the noncancer table. The
cancer and noncancer surrogate risk approximations based on each site's annual averages are
limited to those pollutants for which each respective site sampled. As discussed in Section 18.3,
all four New York sites sampled PAH; BXNY, MONY, and ROCH also sampled hexavalent
chromium. In addition, the cancer and noncancer risk approximations are limited to those
pollutants with enough data to meet the criteria for annual averages to be calculated. Because
annual average concentrations could not be calculated for BXNY, MONY, and TONY, cancer
and noncancer risk approximations were not calculated. This is also true for the PAH for ROCH;
thus, cancer and noncancer risk approximations are presented only for hexavalent chromium for
ROCH.
Observations from Table 18-7 include the following:
• Benzene, ethylbenzene, and formaldehyde are the highest emitted pollutants with
cancer UREs in all three New York counties. The magnitudes of the emissions are
highest in Erie County and lowest in Bronx County.
• Benzene and formaldehyde are the pollutants with the highest toxicity-weighted
emissions (of the pollutants with cancer UREs) for Bronx and Monroe Counties.
Coke oven emissions have the highest toxicity-weighted emissions for Erie County,
followed by benzene and formaldehyde.
• Seven of the highest emitted pollutants also have the highest toxicity-weighted
emissions for Bronx County and Erie County; six of the highest emitted pollutants
also have the highest toxicity-weighted emissions for Monroe County.
• Hexavalent chromium, which was sampled for at three of the four New York sites,
appears among the pollutants with the highest toxicity-weighted emissions for all
three counties, but is not among the highest emitted.
• Emissions of several POM Groups rank among the 10 highest emitted pollutant as
well as the highest toxicity-weighted emissions for all three New York counties.
POM, Group 2b includes several PAH sampled for at these sites, including
acenaphthylene, fluoranthene, fluorene, and perylene. POM, Group 5a includes
benzo(a)pyrene. POM, Group 6 includes benzo(a)anthracene, benzo(b)fluoranthene,
benzo(k)fluoranthene, and indeno(l,2,3-cd)pyrene.
18-44
-------
Observations from Table 18-8 include the following:
• Methanol, toluene, and xylenes are the highest emitted pollutants with noncancer
RfCs in both Bronx and Monroe Counties, although not necessarily in that order;
toluene, xylenes, and benzene are the highest emitted pollutants with noncancer RfCs
in Erie County. The magnitudes of the emissions are highest in Erie County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
noncancer RfCs) are acrolein, 1,3-butadiene, and formaldehyde for all three counties.
• Between four and five of the highest emitted pollutants in Bronx, Monroe, and Erie
Counties are also among the pollutants with the highest toxicity-weighted emissions
for each county.
• Naphthalene, which was sampled for at all four New York sites, is among the
pollutants with the highest toxicity-weighted emissions, but not among the highest
emitted pollutants. Hexavalent chromium, which was sampled for at three of the four
sites, does not appear on either emissions-based list for any of the New York
Counties.
18.6 Summary of the 2010 Monitoring Data for the New York Monitoring Sites
Results from several of the data treatments described in this section include the
following:
»«» Fourteen pollutants failed screens for BXNY andMONY, although a single sample
for each site contributed to the bulk of these failed screens. Only naphthalene failed
screens for ROCH. Six pollutants failed screens for TONY.
»«» The sampling equipment was moved from BXNY to MONY in mid-2010 due to roofing
construction while sampling was discontinued at TONY in July 2010.
»«» Problems with the PAH sampler at ROCH led to the invalidation of a majority of
PAH samples for ROCH.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest, where they could be calculated,
were greater than their associatedMRL noncancer health risk benchmarks.
18-45
-------
19.0 Sites in Oklahoma
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the UATMP sites in Oklahoma, and integrates these concentrations
with emissions, meteorological, and risk information. Data generated by sources other than ERG
are not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
19.1 Site Characterization
This section characterizes the Oklahoma monitoring sites by providing geographical and
physical information about the location of the sites and the surrounding areas. This information
is provided to give the reader insight regarding factors that may influence the air quality near the
sites and assist in the interpretation of the ambient monitoring measurements.
Two Oklahoma sites (TOOK and TMOK) are located in the Tulsa, OK MSA. Another
site, PROK, is located east of the Tulsa area in Pry or Creek, Oklahoma. There are also two sites
in the Oklahoma City, OK MSA; one site is located in Oklahoma City (OCOK) and another is
located just outside Oklahoma City in Midwest City (MWOK).
Figures 19-1 through 19-5 are composite satellite images retrieved from ArcGIS Explorer
showing the monitoring sites in their urban and rural locations. Figures 19-6 through 19-8
identify point source emissions locations by source category, as reported in the 2008 NEI for
point sources. Note that only sources within 10 miles of each site are included in the facility
counts provided in Figures 19-6 through 19-8. Thus, sources outside the
10-mile radius have been grayed out, but are visible on the maps to show emissions sources
outside the 10-mile boundary. A 10-mile boundary was chosen to give the reader an indication of
which emissions sources and emissions source categories could potentially have a direct effect
on the air quality at the monitoring sites. Further, this boundary provides both the proximity of
emissions sources to the monitoring sites as well as the quantity of such sources within a given
distance of the sites. Table 19-1 describes the area surrounding each monitoring site by providing
supplemental geographical information such as land use, location setting, and locational
coordinates.
19-1
-------
Figure 19-1. Tulsa, Oklahoma (TOOK) Monitoring Site
to
-------
Figure 19-2. Tulsa, Oklahoma (TMOK) Monitoring Site
-------
Figure 19-3. Pryor Creek, Oklahoma (PROK) Monitoring Site
-------
Figure 19-4. Midwest City, Oklahoma (MWOK) Monitoring Site
V » >^ -y I 'LfjA—m
u '//
-------
Figure 19-5. Oklahoma City, Oklahoma (OCOK) Monitoring Site
.A
-------
Figure 19-6. NEI Point Sources Located Within 10 Miles of TMOK and TOOK
Legend
@ TMOK UATMP site
96:5tTW WO'O'W Sfr'SStnrV 95'50'DT/V
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
TOOK UATMP site
10 mile radius
County boundary
Source Category Group (No. of Facilities)
•i" Aerospace/Aircraft Manufacturing (1)
-f Aircraft Operations (14)
8 Automobile/Truck Manufacturing (1)
* Electricity Generation via Combustion (2)
ft* Glass Manufacturing (2)
? Miscellaneous Commercial/Industrial (1)
H Municipal Waste Combustor (1)
* Petroleum Refinery (2)
7 Portland Cement Manufacturing (1)
V Steel Mill (1)
•*• Transportation and Marketing of Petroleum Products (1)
19-7
-------
Figure 19-7. NEI Point Sources Located Within 10 Miles of PROK
Legend
•§• PROK UATMP site
e5-20'0"W SS'lS'trW QSMQtrW SS'S'CTW
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
10 mile radius
County boundary
Source Category Group (No. of Facilities)
41 Aircraft Operations (5)
c Chemical Manufacturing (1)
* Electricity Generation via Combustion (2)
F Food Processing/Agriculture (1)
7 Portland Cement Manufacturing (1)
19-8
-------
Figure 19-8. NEI Point Sources Located Within 10 Miles of MWOK and OCOK
97-45'0-W 97"40'0*W 97'35'Q*W S7i3C<10"W S/'JS'O'W 97'20'0"W
Note: Due to facility density and collocation. th9 total facilities
Legend
Sfe MWOK UATMP site
displayed may not represent all facilities within the area of interest.
OCOK UATMP site 10 mile radius | | County boundary
Source Category Group (No. of Facilities)
41 Aircraft Operations (19)
IB Bakery (1)
F Food Processing/Agriculture (1)
A Military Base/National Security Facility (1)
• Oil and/or Gas Production (2)
P Printing/Publishing (1)
19-9
-------
Table 19-1. Geographical Information for the Oklahoma Monitoring Sites
Site
Code
TOOK
TMOK
PROK
MWOK
OCOK
AQS Code
40-143-0235
40-143-1127
40-097-0187
40-109-0041
40-109-1037
Location
Tulsa
Tulsa
Pryor
Creek
Midwest
City
Oklahoma
City
County
Tulsa
Tulsa
Mayes
Oklahoma
Oklahoma
Micro- or
Metropolitan
Statistical Area
Tulsa, OK MSA
Tulsa, OK MSA
Not in an MSA
Oklahoma City,
OK MSA
Oklahoma City,
OK MSA
Latitude
and
Longitude
36.126945,
-95.998941
36.204902,
-95.976537
36.292941,
-95.303409
35.437641,
-97.387254
35.614131,
-97.475083
Land Use
Industrial
Residential
Industrial
Commercial
Residential
Location
Setting
Urban/City
Center
Urban/City
Center
Suburban
Urban/City
Center
Suburban
Additional Ambient Monitoring Information1
SO2 and H2S.
CO, SO2, NOy, NO, NO2, NOX, O3, Meteorological
parameters, PM10, PM Coarse, PM25, and PM25
Speciation.
None.
None.
SO2, NO, NO2, NOX, O3, Meteorological parameters,
PMio, PM2 5, and PM2 5 Speciation.
These monitoring sites report additional pollutants to AQS (EPA, 2012h); however, these data are not generated by ERG and are therefore not included in this report.
VO
o
-------
TOOK is located in West Tulsa, on the southwest side of the Arkansas River. The site is
located in the parking lot of the Public Works building. The monitoring site is positioned
between the Arkansas River and 1-244, which runs parallel to Southwest Boulevard. The
surrounding area is primarily industrial. As shown in Figure 19-1, an oil refinery is located just
south of the site. Another refinery is located to the northwest of the site, on the other side of
1-244. A rail yard is located on the opposite side of 1-244.
TMOK is located in north Tulsa on the property of Fire Station Number 24. As shown in
Figure 19-2, the intersection of North Peoria Avenue (Highway 11) and East 36th Street North
lies just to the northeast of the site. The surrounding area is primarily residential, with wooded
areas just to the east, an early childhood education facility and an elementary school to the south,
and a park to the west.
Figure 19-6 shows that the Tulsa sites are located approximately 5 miles apart, with the
TMOK site to the north and TOOK to the south. Most of the emissions sources are clustered
around TOOK, while there are no point sources within a couple miles of TMOK. The source
category with the highest number of sources surrounding the Tulsa sites is the aircraft operations
source category, which includes airports as well as small runways, heliports, or landing pads.
Point sources closest to TOOK include petroleum refineries, a municipal waste combustor, and a
facility generating electricity via combustion.
PROK is located on the eastern edge of the town of Pry or Creek, on the property of Pry or
Creek High School. Residential areas are located to the northwest, west, and south of the site,
while agricultural areas are located to the east, as shown in Figure 19-3. The monitoring site is
located due north (and downwind) of an industrial park located a few miles to the south.
Figure 19-7 shows that there are relatively few emissions sources surrounding PROK and that
the aircraft operations source category has the highest number of emissions sources near the site.
An aircraft operations facility is located a quarter mile north of PROK but is located under the
site symbol in Figure 19-7. The aforementioned industrial park is represented in Figure 19-7 by
the chemical manufacturing and food processing/agriculture facilities located to the south of
PROK.
19-11
-------
The MWOK monitoring site is located in Midwest City, southeast of Oklahoma City. The
site is located in a commercial area on South Midwest Boulevard just north of 1-40, although
residential areas lie to the west. This site is located at a school enrollment center just north of
Tinker Air Force Base, the northern portion of which can be seen just south of 1-40 in
Figure 19-4. Residential areas are located to the northwest and north, while an extension of
Tinker AFB is located to the east.
OCOK is located in northern Oklahoma City, on the property of Oklahoma Christian
University of Science and Arts. The site is located in the northwest corner of the University, near
the athletic fields. The areas surrounding the university are primarily residential. Heavily
traveled roadways such as 1-35 and 1-44 to the east and John Kilpatrick Turnpike to the south are
within a few miles of the site, although outside the boundaries of Figure 19-5.
Figure 19-8 shows that MWOK and OCOK are approximately 13 miles apart and that
most of the point sources located within 10 miles of them are located between the sites in the
center of Oklahoma City (west and northwest of MWOK and south of OCOK). The source
category with the highest number of sources surrounding the two sites is the aircraft operations
source category. The source closest to MWOK is the military base; the source closest to OCOK
is a heliport.
Table 19-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the areas surrounding the
Oklahoma monitoring sites. Table 19-2 also includes a vehicle registration-to-county population
ratio (vehicles-per-person) for each site. In addition, the population within 10 miles of each site
is presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-
level vehicle registration-to-population ratio to the 10-mile population surrounding each
monitoring site. Table 19-2 also contains annual average daily traffic information. County-level
VMT was not readily available; thus, daily VMT for the Oklahoma sites is not shown in
Table 19-2.
19-12
-------
Table 19-2. Population, Motor Vehicle, and Traffic Information for the Oklahoma
Monitoring Sites
Site
TOOK
TMOK
PROK
MWOK
OCOK
Estimated
County
Population1
605,418
41,283
721,178
County-level
Vehicle
Registration2
604,284
40,832
809,783
Vehicles per
Person
(Registration:
Population)
1.00
0.99
1.12
Population
within 10
miles3
456,229
320,319
26,739
361,698
380,090
Estimated
10-mile
Vehicle
Ownership
455,374
319,719
26,447
406,137
426,788
Annual
Average
Daily
Traffic4
62,566
12,700
15,900
41,200
41,600
County-
level
Daily
VMT5
NA
NA
NA
1 County-level population estimates reflect data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the Oklahoma Tax Commission (OKTC, 2010)
3 10-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2010 data from the Oklahoma DOT (OK DOT, 2010)
5 County-level VMT was not available for these sites
NA = Data unavailable.
Observations from Table 19-2 include the following:
• The Mayes County population is significantly lower than the Tulsa and Oklahoma
County populations. This is also true of the 10-mile populations. Compared to other
NMP monitoring sites, the Tulsa and Oklahoma City populations are in the middle of
the range, while Pry or Creek's populations are on the low end.
• The Mayes County vehicle registration is also significantly lower than vehicle
registration for Tulsa and Oklahoma Counties. Similar observations can be made for
the 10-mile vehicle registration estimates. These observations are expected given the
rural nature of the area surrounding PROK compared to the urban location of the
Tulsa and Oklahoma City sites. Compared to other NMP monitoring sites, the
ownership estimates follow a similar pattern as the populations.
• The average daily traffic volume passing the TMOK site is the lowest among the
Oklahoma monitoring sites and is similar to the traffic passing the PROK site, while
the traffic passing by TOOK is the highest. The traffic data for TMOK and PROK are
in the bottom third among NMP sites.
19.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in Oklahoma on sample days, as well as over the course of the year.
19.2.1 Climate Summary
Tulsa is located in northeast Oklahoma, just southeast of the Osage Indian Reservation,
and along the Arkansas River. Pry or Creek is also in northeast Oklahoma, approximately
19-13
-------
30 miles east of Tulsa. Oklahoma City is located in the center of the state. These areas are
characterized by a continental climate, with very warm summers and cool winters. Precipitation
is generally concentrated in the spring and summer months, with spring as the wettest season,
although precipitation amounts generally decrease across the state from east to west. Spring and
summer precipitation usually results from showers and thunderstorms, while fall and winter
precipitation accompanies frontal systems. A southerly wind prevails for much of the year,
bringing warm, moist air northward from the Gulf of Mexico. Oklahoma is part of "Tornado
Alley," where severe thunderstorms capable of producing strong winds, hail, and tornadoes
occur more frequently than other areas around the county; tornadoes are more prevalent here
than any other region in the U.S. (Bair, 1992; NCDC, 2012; and NOAA, 2012e).
19.2.2 Meteorological Conditions in 2010
Hourly meteorological data from NWS weather stations nearest these sites were retrieved
for 2010 (NCDC, 2010). The closest weather stations to the Tulsa sites are located at Richard
Lloyd Jones Jr. Airport (near TOOK) and Tulsa International Airport (near TMOK), WBAN
53908 and 13968, respectively. The closest weather station to the Pryor Creek site is located at
Claremore Regional Airport, WBAN 53940. The two closest weather stations to the Oklahoma
City sites are located at Tinker Air Force Base Airport (near MWOK) and Wiley Post Airport
(near OCOK), WBAN 13919 and 03954, respectively. Additional information about these
weather stations, such as the distance between the sites and the weather stations, is provided in
Table 19-3. These data were used to determine how meteorological conditions on sample days
vary from normal conditions throughout the year.
Table 19-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 19-3 is the 95
percent confidence interval for each parameter. As shown in Table 19-3, average meteorological
conditions on sample days were representative of average weather conditions throughout the year
near TOOK, TMOK, and PROK. Sample days at MWOK and OCOK appear slightly cooler than
conditions experienced throughout the year, but the difference is not statistically significant. A
few extra samples were collected during first part of the year at MWOK and OCOK in order to
make up a few invalid samples, which may result in these subtle differences.
19-14
-------
Table 19-3. Average Meteorological Conditions near the Oklahoma Monitoring Sites
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Tulsa, Oklahoma - TOOK
Richard Lloyd
Jones Jr.
Airport
53908
(36.04, -95.98)
6.12
miles
172°
(S)
Sample
Day
2010
71.8
±5.0
72.0
+ 2.1
61.4
±4.8
61.2
+ 2.0
49.0
±4.8
48.8
+ 2.0
54.7
±4.4
54.5
+ 1.8
67.0
±2.8
67.2
+ 1.2
1016.3
±1.7
1016.3
±0.7
5.2
±0.7
5.3
+ 0.3
Tulsa, Oklahoma - TMOK
Tulsa
International
Airport
13968
(36.20, -95.89)
4.81
miles
96°
(E)
Sample
Day
2010
70.9
±4.9
71.1
+ 2.1
61.4
±4.7
61.3
+ 2.0
48.5
±4.8
48.1
+ 2.0
54.5
±4.3
54.2
+ 1.8
65.6
±2.9
65.3
+ 1.3
1015.2
±1.8
1015.3
+ 0.7
7.8
±0.9
7.9
+ 0.4
Pryor Creek, Oklahoma - PROK
Claremore
Regional
Airport
53940
(36.29, -95.47)
8.66
miles
270°
(W)
Sample
Day
2010
68.6
±5.0
69.0
+ 2.0
58.8
±4.8
58.7
+ 2.0
48.7
±4.9
49.0
+ 2.0
53.3
±4.5
53.5
+ 1.8
73.0
±3.0
73.3
+ 1.3
NA
NA
6.4
±0.9
6.4
+ 0.3
Midwest City, Oklahoma - MWOK
Tinker
AFB/Airport
13919
(35.42, -97.39)
1.57
miles
178°
(S)
Sample
Day
2010
68.8
±4.8
70.8
+ 2.0
59.6
±4.5
60.7
+ 1.9
47.9
±4.6
48.5
+ 1.9
53.4
±4.1
54.1
+ 1.7
68.7
±3.6
68.0
+ 1.6
1015.3
±1.7
1015.2
±0.7
9.2
±0.9
9.5
+ 0.4
Oklahoma City, Oklahoma - OCOK
Wiley Post
Airport
03954
(35.53, -97.65)
10.68
miles
240°
(WSW)
Sample
Day
2010
69.6
±4.9
71.3
+ 2.0
60.2
±4.7
61.2
+ 2.0
46.3
±4.4
46.9
+ 1.8
52.7
±4.1
53.4
+ 1.7
63.3
±3.0
62.9
+ 1.4
1015.1
± 1.7
1015.0
±0.8
9.7
±1.0
10.2
+ 0.4
VO
1 Sample day averages are hi;
NA = Sea level pressure was
;hlighted to help differentiate the sample day averages from the full-year averages.
not recorded at the Claremore Regional Airport.
-------
19.2.3 Back Trajectory Analysis
Figure 19-9 is the composite back trajectory map for days on which samples were
collected at the TOOK monitoring site in 2010. Included in Figure 19-9 are four back trajectories
per sample day. Figure 19-10 is the corresponding cluster analysis for 2010. Similarly,
Figures 19-11 through 19-18 are the composite back trajectory maps for days on which samples
were collected at the remaining Oklahoma sites and the corresponding cluster analyses. An in-
depth description of these maps and how they were generated is presented in Section 3.5.2.1. For
the composite maps, each line represents the 24-hour trajectory along which a parcel of air
traveled toward the monitoring site on a given sample day and time. For the cluster analyses,
each line corresponds to a back trajectory representative of a given cluster of trajectories. For all
maps, each concentric circle around the sites in Figures 19-9 through 19-18 represents 100 miles.
Figure 19-9. 2010 Composite Back Trajectory Map for TOOK
19-16
-------
Figure 19-10. Back Trajectory Cluster Map for TOOK
Figure 19-11. 2010 Composite Back Trajectory Map for TMOK
I, '
19-17
-------
Figure 19-12. Back Trajectory Cluster Map for TMOK
Figure 19-13. 2010 Composite Back Trajectory Map for PROK
19-18
-------
Figure 19-14. Back Trajectory Cluster Map for PROK
Figure 19-15. 2010 Composite Back Trajectory Map for MWOK
19-19
-------
Figure 19-16. Back Trajectory Cluster Map for MWOK
Figure 19-17. 2010 Composite Back Trajectory Map for OCOK
19-20
-------
Figure 19-18. Back Trajectory Cluster Map for OCOK
Observations from Figures 19-9 through 19-18 include the following:
• The back trajectory maps for the Tulsa sites, the Pry or Creek site, and the Oklahoma
City sites are similar to each other in trajectory distribution. This is somewhat
expected, given their relatively close proximity to each other and the similarity in
sample days, although not all sites sampled on the exact same days over the period.
• The air shed domains for the Tulsa and Pryor Creek sites were among the largest
compared to other NMP sites. The farthest away a trajectory originated was over
central Montana, or greater than 900 miles away and ranking second, third, and fourth
longest for PROK, TMOK, and TOOK, respectively, among NMP sites. However,
the average trajectory length for these sites ranged from 268 to 272 miles, which is
still in the top third among NMP sites.
• For the Oklahoma City sites, the farthest away a trajectory originated was also over
Montana, or nearly 850 miles away. The average trajectory length for these sites
ranged from 286 to 290 miles, ranking fourth and sixth highest among NMP sites for
average trajectory length.
• Each of the sites show a strong tendency for trajectories to originate from the south-
southeast to south-southwest of the sites, and from the northwest to north of the sites.
Back trajectories also originated from the east to southeast, but they infrequently
originated from the west.
19-21
-------
• For the Tulsa and Pry or Creeks sites, approximately one-third of back trajectories
originated from the southeast to southwest over Texas. Another one-third of
trajectories originated generally from the east to southeast and within 200 miles of the
sites, but also includes shorter trajectories originating from the south and west. The
remaining back trajectories originated from the northwest to northeast, with the
trajectories originating from a northwesterly direction being longer than those
originating from the northeast.
• The cluster analysis maps for the Oklahoma City sites are similar to the cluster maps
for the Tulsa and Pryor Creeks sites in cluster distribution patterns, although there
were more trajectories originating from west Texas for MWOK and OCOK, thus
there is an additional cluster trajectory in Figures 19-16 and 19-18 to represent these
back trajectories.
19.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather stations at Richard Lloyd Jones Junior Airport
(for TOOK), Tulsa International Airport (for TMOK), Claremore Regional Airport (for PROK),
Wiley Post Airport (for OCOK), and Tinker Air Force Base (for MWOK) were uploaded into a
wind rose software program to produce customized wind roses, as described in Section 3.5.2.2.
A wind rose shows the frequency of wind directions using "petals" positioned around a 16-point
compass, and uses different colors to represent wind speeds.
Figure 19-19 presents three different wind roses for the TOOK monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figures 19-20 through
19-23 presents the three wind roses and distance maps for the remaining Oklahoma monitoring
sites.
19-22
-------
Figure 19-19. Wind Roses for the Richard Lloyd Jones Jr. Airport Weather Station near
TOOK
1999-2009 Historical Wind Rose
2010 Wind Rose
Calms: 24.12%
2010 Sample Day Wind Rose
Distance between TOOK and NWS Station
VWNDSPEED
(Knots)
•
Calms: 25.14H
19-23
-------
Figure 19-20. Wind Roses for the Tulsa International Airport Weather Station near
TMOK
1999-2009 Historical Wind Rose
2010 Wind Rose
Calms: 8.23%
2010 Sample Day Wind Rose
Distance between TMOK and NWS Station
19-24
-------
Figure 19-21. Wind Roses for the Claremore Regional Airport Weather Station near
PROK
2003-2009 Historical Wind Rose
2010 Wind Rose
Calms; 15.69%
2010 Sample Day Wind Rose
Distance between PROK and NWS Station
19-25
-------
Figure 19-22. Wind Roses for the Tinker Air Force Base Airport Weather Station near
MWOK
2006-2009 Historical Wind Rose
2010 Wind Rose
Calms: 2.93%
2010 Sample Day Wind Rose
Distance between MWOK and NWS Station
J EE-^f-D, | " I /
** <««,«. n z • «
19-26
-------
Figure 19-23. Wind Roses for the Wiley Post Airport Weather Station near OCOK
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between OCOK and NWS Station
™ P
W Hm
-------
Observations from Figures 19-19 through 19-23 include the following:
• The distance maps show that the distances between the sites and the weather stations
varies from 1.6 miles between Tinker Air Force Base and MWOK to 10.7 miles
between OCOK and the Wiley Post Airport.
• Even though the historical data shown are from five different weather stations, the
wind patterns shown on wind roses for the Oklahoma sites are similar to each other.
Each of the historical wind roses shows that southerly winds prevailed near each
Oklahoma monitoring site, accounting for one-fifth to one-quarter of the observations
among the historical time periods. The historical wind roses varied in the percentage
of calm winds (<2 knots) observed, ranging from as little as three percent at the
Tinker Air Force Base (MWOK) to as high as 24 percent at the Richard Lloyd Jones
Jr. Airport (TOOK). Further, calm winds, winds from the south-southeast through
south-southwest, and winds from the north-northwest to north-northeast account for
almost all observations at these sites; winds from the west or east are rarely observed.
• For TOOK, the 2010 wind patterns are very similar to the historical wind patterns, as
are the sample day wind patterns. This indicates that conditions on sample days were
representative of those experienced over the entire year and historically.
• For TMOK, the 2010 wind patterns resemble the historical wind patterns, although
there is a slightly higher percentage of south-southeasterly and northwesterly to
north-northwesterly winds observed in 2010.
• For PROK, the 2010 wind rose shows a significantly higher percentage of
southeasterly to south-southeasterly wind observations and fewer northeasterly and
south-southwesterly winds than the historical wind rose. The sample day wind
patterns are similar to the full-year wind patterns, indicating that conditions on
sample days were representative of conditions experienced throughout the year.
• For MWOK, the historical wind rose includes only four years worth of data. The
2010 wind patterns resemble the historical wind patterns, although there were slightly
fewer south-southwesterly wind observations and more southeasterly to south-
southeasterly winds observations. The sample day wind rose wind patterns resemble
the historical and the full-year wind patterns rose.
• For OCOK, the wind patterns shown on the 2010 wind rose are similar to the
historical wind patterns. The sample day wind rose for OCOK is similar to historical
and full-year wind roses, indicating that conditions on sample days were
representative of those experienced over the entire year and historically.
19.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Oklahoma monitoring sites
in order to allow analysts and readers to focus on a subset of pollutants through the context of
risk. For each site, each pollutant's preprocessed daily measurement was compared to its
19-28
-------
associated risk screening value. If the concentration was greater than the risk screening value,
then the concentration "failed the screen." Pollutants of interest are those for which the
individual pollutant's total failed screens contribute to the top 95 percent of the site's total failed
screens. In addition, if any of the NATTS MQO Core Analytes measured by each monitoring site
did not meet the pollutant of interest criteria based on the preliminary risk screening, that
pollutant was added to the list of site-specific pollutants of interest. A more in-depth description
of the risk screening process is presented in Section 3.2.
Table 19-4 presents the pollutants of interest for each Oklahoma monitoring site. The
pollutants that failed at least one screen and contributed to 95 percent of the total failed screens
for each monitoring site are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of
interest are shaded and/or bolded. The five Oklahoma sites sampled for VOC, carbonyl
compounds, and metals (TSP).
Table 19-4. Risk Screening Results for the Oklahoma Monitoring Sites
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Tulsa, Oklahoma - TOOK
Benzene
Carbon Tetrachloride
Acet aldehyde
Formaldehyde
Manganese (TSP)
Arsenic (TSP)
1,3-Butadiene
£>-Dichlorobenzene
Ethylbenzene
Propionaldehyde
Cadmium (TSP)
1 ,2-Dichloroethane
Acrylonitrile
Nickel (TSP)
Chloromethylbenzene
1 ,2-Dibromoethane
Trichloroethylene
0.13
0.17
0.45
0.077
0.005
0.00023
0.03
0.091
0.4
0.8
0.00056
0.038
0.015
0.0021
0.02
0.0017
0.2
Total
61
61
60
60
60
58
50
33
31
8
5
5
2
2
1
1
1
499
61
61
60
60
61
61
56
55
61
58
61
5
2
61
1
1
11
736
100.00
100.00
100.00
100.00
98.36
95.08
89.29
60.00
50.82
13.79
8.20
100.00
100.00
3.28
100.00
100.00
9.09
67.80
12.22
12.22
12.02
12.02
12.02
11.62
10.02
6.61
6.21
1.60
1.00
1.00
0.40
0.40
0.20
0.20
0.20
12.22
24.45
36.47
48.50
60.52
72.14
82.16
88.78
94.99
96.59
97.60
98.60
99.00
99.40
99.60
99.80
100.00
19-29
-------
Table 19-4. Risk Screening Results for the Oklahoma Monitoring Sites (Continued)
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Tulsa, Oklahoma - TMOK
Benzene
Carbon Tetrachloride
Acetaldehyde
Formaldehyde
Arsenic (TSP)
Manganese (TSP)
1,3-Butadiene
Ethylbenzene
£>-Dichlorobenzene
1 ,2-Dichloroethane
Acrylonitrile
Propionaldehyde
Cadmium (TSP)
Chloroprene
1 ,2-Dibromoethane
Nickel (TSP)
0.13
0.17
0.45
0.077
0.00023
0.005
0.03
0.4
0.091
0.038
0.015
0.8
0.00056
0.0021
0.0017
0.0021
Total
61
61
60
60
58
56
55
35
31
10
8
3
2
1
1
1
503
61
61
60
60
60
60
57
61
57
10
8
59
60
1
1
60
736
100.00
100.00
100.00
100.00
96.67
93.33
96.49
57.38
54.39
100.00
100.00
5.08
3.33
100.00
100.00
1.67
68.34
12.13
12.13
11.93
11.93
11.53
11.13
10.93
6.96
6.16
1.99
1.59
0.60
0.40
0.20
0.20
0.20
12.13
24.25
36.18
48.11
59.64
70.78
81.71
88.67
94.83
96.82
98.41
99.01
99.40
99.60
99.80
100.00
Pryor Creek, Oklahoma - PROK
Benzene
Carbon Tetrachloride
Acetaldehyde
Formaldehyde
Arsenic (TSP)
Manganese (TSP)
1,3-Butadiene
£>-Dichlorobenzene
1 ,2-Dichloroethane
Nickel (TSP)
Cadmium (TSP)
Hexachloro- 1 ,3 -butadiene
1 , 1 ,2,2-Tetrachloroethane
0.13
0.17
0.45
0.077
0.00023
0.005
0.03
0.091
0.038
0.0021
0.00056
0.045
0.017
Total
61
61
60
60
56
52
40
22
15
3
2
1
1
434
61
61
60
60
61
61
50
48
15
61
61
1
1
601
100.00
100.00
100.00
100.00
91.80
85.25
80.00
45.83
100.00
4.92
3.28
100.00
100.00
72.21
14.06
14.06
13.82
13.82
12.90
11.98
9.22
5.07
3.46
0.69
0.46
0.23
0.23
14.06
28.11
41.94
55.76
68.66
80.65
89.86
94.93
98.39
99.08
99.54
99.77
100.00
19-30
-------
Table 19-4. Risk Screening Results for the Oklahoma Monitoring Sites (Continued)
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Midwest City, Oklahoma - MWOK
Benzene
Carbon Tetrachloride
Formaldehyde
Acetaldehyde
Arsenic (TSP)
Manganese (TSP)
1,3-Butadiene
£>-Dichlorobenzene
1 ,2-Dichloroethane
Ethylbenzene
Nickel (TSP)
Acrylonitrile
Chloromethylbenzene
1 ,2-Dibromoethane
Hexachloro- 1 ,3 -butadiene
Lead (TSP)
1 , 1 ,2,2-Tetrachloroethane
Vinyl chloride
0.13
0.17
0.077
0.45
0.00023
0.005
0.03
0.091
0.038
0.4
0.0021
0.015
0.02
0.0017
0.045
0.015
0.017
0.11
Total
61
60
60
59
58
47
45
42
13
8
5
2
1
1
1
1
1
1
466
61
61
60
60
61
61
52
61
13
61
61
2
1
1
2
61
1
3
683
100.00
98.36
100.00
98.33
95.08
77.05
86.54
68.85
100.00
13.11
8.20
100.00
100.00
100.00
50.00
1.64
100.00
33.33
68.23
13.09
12.88
12.88
12.66
12.45
10.09
9.66
9.01
2.79
1.72
1.07
0.43
0.21
0.21
0.21
0.21
0.21
0.21
13.09
25.97
38.84
51.50
63.95
74.03
83.69
92.70
95.49
97.21
98.28
98.71
98.93
99.14
99.36
99.57
99.79
100.00
Oklahoma City, Oklahoma - OCOK
Benzene
Acetaldehyde
Carbon Tetrachloride
Formaldehyde
Arsenic (TSP)
Manganese (TSP)
£>-Dichlorobenzene
1,3-Butadiene
1 ,2-Dichloroethane
Acrylonitrile
Ethylbenzene
Chloromethylbenzene
Hexachloro- 1 ,3 -butadiene
1 , 1 ,2,2-Tetrachloroethane
Bromomethane
Chloroprene
1 ,2-Dibromoethane
1 , 1 ,2-Trichloroethane
Trichloroethylene
0.13
0.45
0.17
0.077
0.00023
0.005
0.091
0.03
0.038
0.015
0.4
0.02
0.045
0.017
0.5
0.0021
0.0017
0.0625
0.2
Total
61
60
60
60
54
51
42
41
11
10
6
2
2
2
1
1
1
1
1
467
61
60
61
60
61
61
54
48
11
10
61
2
2
2
52
1
1
3
4
615
100.00
100.00
98.36
100.00
88.52
83.61
77.78
85.42
100.00
100.00
9.84
100.00
100.00
100.00
1.92
100.00
100.00
33.33
25.00
75.93
13.06
12.85
12.85
12.85
11.56
10.92
8.99
8.78
2.36
2.14
1.28
0.43
0.43
0.43
0.21
0.21
0.21
0.21
0.21
13.06
25.91
38.76
51.61
63.17
74.09
83.08
91.86
94.22
96.36
97.64
98.07
98.50
98.93
99.14
99.36
99.57
99.79
100.00
19-31
-------
Observations from Table 19-4 include the following:
• Seventeen pollutants failed at least one screen for TOOK; 16 pollutants failed screens
for TMOK; 13 pollutants failed screens for PROK; 18 pollutants failed screens for
MWOK; and 19 pollutants failed screens for OCOK.
• The risk screening process identified 10 pollutants of interest for TOOK, of which
seven are NATTS MQO Core Analytes. Cadmium, nickel, and trichloroethylene were
added to TOOK's pollutants of interest because they are NATTS MQO Core
Analytes, even though they did not contribute to 95 percent of the total failed screens.
Five additional pollutants (beryllium, chloroform, lead, tetrachloroethylene, and vinyl
chloride) were added to TOOK's pollutants of interest because they are NATTS
MQO Core Analytes, even though they did not fail any screens. These five pollutants
do not appear in Table 19-4.
• The risk screening process identified 10 pollutants of interest for TMOK, of which
seven are NATTS MQO Core Analytes. Cadmium and nickel were added to TMOK's
pollutants of interest because they are NATTS MQO Core Analytes, even though
they did not contribute to 95 percent of the total failed screens. Six additional
pollutants (beryllium, chloroform, lead, tetrachloroethylene, trichloroethylene, and
vinyl chloride) were added to TOOK's pollutants of interest because they are NATTS
MQO Core Analytes, even though they did not fail any screens. These six pollutants
do not appear in Table 19-4.
• The risk screening process identified nine pollutants of interest for PROK, of which
seven are NATTS MQO Core Analytes. Nickel and cadmium were added to PROK's
pollutants of interest because they are NATTS MQO Core Analytes, even though
they did not contribute to 95 percent of the total failed screens. An additional six
pollutants (beryllium, chloroform, lead, tetrachloroethylene, trichloroethylene, and
vinyl chloride) were added to PROK's pollutants of interest because they are NATTS
MQO Core Analytes, even though they did not fail any screens. These six pollutants
do not appear in Table 19-4.
• The risk screening process identified nine pollutants of interest for MWOK, of which
seven are NATTS MQO Core Analytes. Nickel, lead, and vinyl chloride were added
to MWOK's pollutants of interest because they are NATTS MQO Core Analytes,
even though they did not contribute to 95 percent of the total failed screens. This was
the only site for which there was a failed screen of vinyl chloride among all NMP
sites sampling VOC. An additional five pollutants (beryllium, cadmium, chloroform,
tetrachloroethylene, and trichloroethylene) were added to MWOK's pollutants of
interest because they are NATTS MQO Core Analytes, even though they did not fail
any screens. These five pollutants do not appear in Table 19-4.
• The risk screening process identified 10 pollutants of interest for OCOK, of which
seven are NATTS MQO Core Analytes. Trichloroethylene was added to OCOK's
pollutants of interest because it is a NATTS MQO Core Analyte, even though it did
not contribute to 95 percent of the total failed screens. Seven additional pollutants
19-32
-------
(four metals and three VOC) were added to OCOK's pollutants of interest because
they are NATTS MQO Core Analytes, even though they did not fail any screens.
These pollutants do not appear in Table 19-4.
• Benzene and formaldehyde each failed 100 percent of screens for each site.
• The percentage of measured detections failing screens (of the pollutants that failed at
least one screen) ranged from 68 percent (TOOK) to 76 percent (OCOK). TMOK and
TOOK failed the third and fourth highest number of screens among all NMP sites,
although the other Oklahoma sites ranked seventh (OCOK), ninth (MWOK), and
eleventh (PROK).
19.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Oklahoma monitoring sites. Concentration averages are provided for the pollutants of
interest for each Oklahoma site, where applicable. Concentration averages for select pollutants
are also presented graphically for each site, where applicable, to illustrate how each site's
concentrations compare to the program-level averages. In addition, concentration averages for
select pollutants are presented from previous years of sampling in order to characterize
concentration trends at each site, where applicable. Additional site-specific statistical summaries
are provided in Appendices J, L, and N.
19.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each Oklahoma site, as described in Section 3.1. The quarterly average of a particular
pollutant is simply the average concentration of the preprocessed daily measurements over a
given calendar quarter. Quarterly average concentrations include the substitution of zeros for all
non-detects. A site must have a minimum of 75 percent valid samples of the total number of
samples possible within a given quarter for a quarterly average to be calculated. An annual
average includes all measured detections and substituted zeros for non-detects for the entire year
of sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the Oklahoma
monitoring sites are presented in Table 19-5, where applicable. Note that concentrations of the
TSP metals are presented in ng/m3 for ease of viewing. Also note that if a pollutant was not
19-33
-------
detected in a given calendar quarter, the quarterly average simply reflects "0" because only zeros
substituted for non-detects were factored into the quarterly average concentration.
Table 19-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Oklahoma Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(jig/m3)
2nd
Quarter
Average
(jig/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Annual
Average
(jig/m3)
Tulsa, Oklahoma - TOOK
Acetaldehyde
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
Ethylbenzene
Formaldehyde
Propionaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Arsenic (TSP)a
Bery Ilium (TSP)a
Cadmium (TSP)a
Lead(TSP)3
Manganese (TSP)a
Nickel (TSP)a
60/60
61/61
56/61
61/61
38/61
55/61
61/61
60/60
58/60
57/61
11/61
3/61
61/61
61/61
61/61
61/61
61/61
61/61
1.19
±0.25
1.45
±0.46
0.07
±0.04
0.67
±0.07
0.05
±0.03
0.04
±0.02
0.30
±0.12
1.48
±0.25
0.22
±0.05
0.09
±0.04
0.02
±0.03
<0.01
±<0.01
0.46
±0.09
0.010
±0.01
0.30
±0.16
3.85
±1.14
16.06
±5.60
0.88
±0.20
2.19
±0.47
1.93
±0.49
0.06
±0.02
0.63
±0.07
0.10
±0.03
0.12
±0.08
0.50
±0.17
3.46
±0.95
0.43
±0.13
0.12
±0.05
0.02
±0.02
<0.01
±<0.01
0.70
±0.16
0.02
±0.01
0.26
±0.07
4.28
±1.12
21.66
±5.51
1.04
±0.19
3.15
±0.56
2.72
±0.47
0.06
±0.02
0.64
±0.05
0.12
±0.03
0.27
±0.05
0.49
±0.13
4.86
±0.82
0.60
±0.14
0.17
±0.06
0.02
±0.02
0
0.77
±0.14
0.03
±0.01
0.22
±0.06
4.84
±0.87
28.75
±5.93
1.24
±0.28
2.12
±0.54
3.23
±1.09
0.11
±0.04
0.56
±0.06
0.02
±0.02
0.16
±0.04
0.60
±0.20
2.52
±0.36
0.44
±0.14
0.12
±0.04
0.01
±0.02
0
0.71
±0.24
0.02
±0.01
0.36
±0.20
4.83
±1.13
27.62
±7.26
1.23
±0.25
2.20
±0.29
2.34
±0.36
0.07
±0.01
0.62
±0.03
0.07
±0.02
0.15
±0.03
0.47
±0.08
3.14
±0.45
0.43
±0.07
0.13
±0.02
0.02
±0.01
O.01
±O.01
0.66
±0.08
0.02
±0.01
0.28
±0.06
4.46
±0.51
23.61
±3.15
1.10
±0.12
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for
ease of viewing.
19-34
-------
Table 19-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Oklahoma Monitoring Sites (Continued)
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Ug/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(Ug/m3)
4th
Quarter
Average
(Ug/m3)
Annual
Average
(Ug/m3)
Tulsa, Oklahoma - TMOK
Acetaldehyde
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Ethylbenzene
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Arsenic (TSP)a
Bery Ilium (TSP)a
Cadmium (TSP) a
Lead (TSP) a
Manganese (TSP) a
Nickel (TSP) a
60/60
61/61
57/61
61/61
39/61
57/61
10/61
61/61
60/60
49/61
10/61
2/61
60/60
59/60
60/60
60/60
60/60
60/60
1.45
±0.22
1.27
±0.28
0.09
±0.03
0.58
±0.09
0.04
±0.02
0.06
±0.02
0.03
±0.02
0.59
±0.46
2.63
±0.30
0.06
±0.03
0.01
±0.01
<0.01
±<0.01
0.42
±0.09
0.01
±0.01
0.19
±0.05
3.04
±0.54
11.57
±3.36
0.75
±0.20
2.17
±0.33
1.32
±0.31
0.08
±0.02
0.65
±0.08
0.08
±0.03
0.12
±0.07
0.03
±0.03
0.43
±0.11
4.20
±0.85
0.09
±0.03
0.01
±0.02
0
0.60
±0.16
0.01
±0.01
0.23
±0.06
3.59
±0.88
15.29
±3.90
0.81
±0.12
2.66
±0.41
1.73
±0.34
0.10
±0.03
0.62
±0.06
0.11
±0.04
0.26
±0.04
0
0.59
±0.15
4.42
±0.88
0.10
±0.03
0.02
±0.02
0
0.80
±0.16
0.02
±0.01
0.27
±0.15
4.03
±0.69
18.53
±3.77
1.02
±0.22
1.73
±0.38
1.93
±0.55
0.14
±0.05
0.58
±0.05
0.06
±0.04
0.12
±0.05
0
0.59
±0.20
2.16
±0.39
0.09
±0.05
0
0
0.66
±0.27
0.01
±0.01
0.22
±0.06
4.23
±1.11
18.11
±4.41
1.00
±0.30
2.00
±0.20
1.57
±0.20
0.10
±0.02
0.61
±0.03
0.07
±0.02
0.14
±0.03
0.01
±0.01
0.55
±0.13
3.35
±0.40
0.09
±0.02
0.01
±0.01
O.01
±O.01
0.62
±0.09
0.01
±0.01
0.23
±0.04
3.72
±0.40
15.88
±1.95
0.89
±0.11
a Average concentrations provided for the pollutants below the black line are presented in ng/m for
ease of viewing.
19-35
-------
Table 19-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Oklahoma Monitoring Sites (Continued)
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Ug/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(Ug/m3)
4th
Quarter
Average
(Ug/m3)
Annual
Average
(Ug/m3)
Pryor Creek, Oklahoma - PROK
Acetaldehyde
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Arsenic (TSP)a
Bery Ilium (TSP)a
Cadmium (TSP) a
Lead (TSP) a
Manganese (TSP) a
Nickel (TSP) a
60/60
61/61
50/61
61/61
44/61
48/61
15/61
60/60
29/61
3/61
1/61
61/61
61/61
61/61
61/61
61/61
61/61
0.91
±0.18
0.68
±0.11
0.04
±0.01
0.60
±0.07
0.04
±0.03
0.02
±0.01
0.04
±0.02
1.14
±0.21
0.02
±0.02
<0.01
±<0.01
0.01
±0.01
0.43
±0.13
0.01
±0.01
0.24
±0.16
2.59
±1.58
8.63
±4.36
1.06
±1.01
1.31
±0.21
0.84
±0.20
0.05
±0.01
0.68
±0.08
0.10
±0.02
0.12
±0.11
0.04
±0.03
2.89
±0.97
0.04
±0.02
O.01
±0.01
0
0.56
±0.14
0.02
±0.01
0.15
±0.04
2.80
±1.07
11.46
±2.98
0.66
±0.11
1.77
±0.29
0.61
±0.06
0.03
±0.01
0.68
±0.06
0.31
±0.41
0.19
±0.02
0
4.10
±0.72
0.03
±0.02
0.01
±0.01
0
0.68
±0.16
0.03
±0.01
0.14
±0.03
4.06
±1.26
17.98
±4.43
0.99
±0.41
1.45
±0.25
0.72
±0.12
0.03
±0.02
0.64
±0.03
0.09
±0.07
0.06
±0.02
0
2.30
±0.37
0.03
±0.02
0
0
0.53
±0.15
0.03
±0.01
0.14
±0.04
2.72
±0.65
12.94
±3.23
0.56
±0.06
1.37
±0.14
0.71
±0.06
0.04
±0.01
0.65
±0.03
0.14
±0.11
0.10
±0.03
0.02
±0.01
2.64
±0.41
0.03
±0.01
O.01
±O.01
0.01
±0.01
0.55
±0.07
0.02
±0.01
0.17
±0.04
3.06
±0.58
12.84
±2.00
0.82
±0.26
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for
ease of viewing.
19-36
-------
Table 19-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Oklahoma Monitoring Sites (Continued)
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Ug/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(Ug/m3)
4th
Quarter
Average
(Ug/m3)
Annual
Average
(Ug/m3)
Midwest City, Oklahoma - MWOK
Acetaldehyde
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Arsenic (TSP)a
Bery Ilium (TSP)a
Cadmium (TSP) a
Lead (TSP) a
Manganese (TSP) a
Nickel (TSP) a
60/60
61/61
52/61
61/61
45/61
61/61
13/61
60/60
53/61
4/61
3/61
61/61
61/61
61/61
61/61
61/61
61/61
0.90
±0.19
0.87
±0.18
0.06
±0.03
0.62
±0.10
0.05
±0.02
0.09
±0.02
0.04
±0.02
1.09
±0.25
0.17
±0.06
0
0.01
±0.01
0.35
±0.07
0.01
±0.01
0.11
±0.02
2.06
±0.28
7.97
±4.30
1.04
±0.51
1.25
±0.22
0.83
±0.18
0.07
±0.02
0.69
±0.07
0.07
±0.02
0.21
±0.11
0.03
±0.03
2.47
±0.84
0.29
±0.17
0.01
±0.01
0.01
±0.01
0.52
±0.17
0.01
±O.01
0.09
±0.01
2.20
±0.46
7.89
±1.82
1.01
±0.41
1.89
±0.22
0.83
±0.13
0.06
±0.03
0.66
±0.05
0.08
±0.02
0.28
±0.08
0.01
±0.01
4.17
±0.55
0.27
±0.18
0.01
±0.02
0.01
±0.02
0.48
±0.11
0.01
±O.01
0.08
±0.02
2.23
±0.43
10.97
±2.11
0.84
±0.15
1.55
±0.26
1.00
±0.29
0.07
±0.04
0.60
±0.07
0.06
±0.03
0.14
±0.04
0
2.40
±0.40
0.13
±0.08
0
0
0.42
±0.09
0.01
±O.01
0.11
±0.02
4.29
±3.24
11.29
±2.67
1.05
±0.30
1.41
±0.14
0.88
±0.10
0.06
±0.01
0.64
±0.04
0.07
±0.01
0.18
±0.04
0.02
±0.01
2.56
±0.38
0.22
±0.06
O.01
±O.01
0.01
±0.01
0.44
±0.06
0.01
±O.01
0.10
±0.01
2.69
±0.80
9.55
±1.41
0.98
±0.17
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for
ease of viewing.
19-37
-------
Table 19-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Oklahoma Monitoring Sites (Continued)
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Ug/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(Ug/m3)
4th
Quarter
Average
(Ug/m3)
Annual
Average
(Ug/m3)
Oklahoma City, Oklahoma - OCOK
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Arsenic (TSP)a
Bery Ilium (TSP)a
Cadmium (TSP) a
Lead (TSP) a
Manganese (TSP) a
60/60
10/61
61/61
48/61
61/61
40/61
54/61
11/61
60/60
53/61
4/61
4/61
61/61
61/61
61/61
61/61
61/61
1.01
±0.17
0.04
±0.03
0.82
±0.16
0.04
±0.01
0.65
±0.08
0.04
±0.02
0.36
±0.22
0.03
±0.02
1.34
±0.19
0.07
±0.02
0.01
±0.01
O.01
±O.01
0.36
±0.07
0.01
±0.01
0.11
±0.04
1.85
±0.19
8.67
±4.47
1.66
±0.30
0.05
±0.10
1.05
±0.26
0.05
±0.02
0.71
±0.06
0.09
±0.02
0.22
±0.07
0.04
±0.03
2.83
±0.81
0.07
±0.03
0
O.01
±0.01
0.56
±0.29
0.01
±0.01
0.09
±0.02
1.94
±0.44
11.83
±2.99
2.21
±0.31
0.02
±0.04
1.17
±0.30
0.04
±0.01
0.66
±0.05
0.09
±0.03
0.27
±0.17
0
4.00
±0.57
0.10
±0.05
0.01
±0.01
0
0.46
±0.07
0.02
±0.01
0.09
±0.03
2.14
±0.37
14.83
±2.99
1.45
±0.27
0
1.00
±0.30
0.04
±0.03
0.53
±0.10
0.03
±0.02
0.11
±0.10
0
1.76
±0.34
0.12
±0.06
0.05
±0.08
0
0.39
±0.09
0.01
±0.01
0.10
±0.04
2.40
±0.71
12.39
±2.81
1.59
±0.17
0.03
±0.03
1.01
±0.13
0.04
±0.01
0.64
±0.04
0.06
±0.01
0.24
±0.07
0.02
±0.01
2.50
±0.36
0.09
±0.02
0.01
±0.02
O.01
±O.01
0.44
±0.08
0.01
±0.01
0.10
±0.01
2.08
±0.22
11.98
±1.68
a Average concentrations provided for the pollutants below the black line are presented in ng/m for
ease of viewing.
19-38
-------
Observations for all five Oklahoma sites from Table 19-5 include the following:
• Formaldehyde has the highest annual average concentration by mass for each site,
followed by acetaldehyde and benzene, with one exception. The annual average
concentration for benzene is greater than the annual average concentration of
acetaldehyde for TOOK. The annual average concentrations of these three pollutants
are highest for TOOK and TMOK among the Oklahoma sites.
• The annual average concentrations of manganese are the highest among the TSP
metals for each site, and are also highest for TOOK and TMOK among the Oklahoma
sites.
• Concentrations of the carbonyl compounds tended to be highest in the summer
months and lowest in the winter months, especially for formaldehyde.
• 1,2-Dichloroethane is a pollutant of interest for four of the five Oklahoma sites. For
each of these sites, the majority of the measured detections were measured during the
first and second quarters of 2010. This pollutant was not detected at all at TMOK,
PROK, and OCOK during the third and fourth quarters of the year. For MWOK, there
was a single measured detection in July, after which there were no others.
• Concentrations of the TSP metals tended to be higher at TOOK, TMOK, and PROK
than OCOK and MWOK.
Observations for the Tulsa sites from Table 19-5 include the following:
• The fourth quarter average concentration of benzene for TOOK has a fairly large
confidence interval associated with it. The highest benzene concentration was
measured on December 10, 2010 (6.95 jig/m3), although two additional
concentrations greater than 6 |ig/m3 were measured in October. These three
concentrations are the three highest benzene measurements among NMP sites
sampling this pollutant. Of the 19 concentrations of benzene that are greater than
3.5 |ig/m3 (among all NMP sites), 11 of these were measured at TOOK. Of these 11,
one was measured in the first quarter of 2010, one in the second quarter, four were
measured in the third quarter, and five in the fourth quarter. While benzene
concentrations greater than 3.5 |ig/m3 were not measured at TMOK, this site exhibits
a similar quarterly trend as six of the nine highest concentrations of benzene (those
greater than 2.5 |ig/m3) were measured during the fourth quarter of 2010.
• Concentrations of />-dichlorobenzene were highest in the third quarter and lowest in
the first quarter of 2010 at TOOK. A review of the data shows that the highest
concentrations of this pollutant were measured in June and July while five of the six
non-detects were reported for January and February. A similar trend is realized at
TMOK.
• The quarterly average concentrations of manganese are higher for the third and fourth
quarters than the first and second quarters of 2010 for TOOK. A review of the data
shows that six of the eight concentrations greater than 40 ng/m3 were measured
19-39
-------
during in August, September, and October. Conversely, seven of the eight lowest
concentrations of manganese were measured in January, February, and March 2010.
Observations for PROK from Table 19-5 include the following:
• The annual average concentration of chloroform for PROK is roughly twice the
annual average concentrations of this pollutant for the other Oklahoma sites. The
third quarter average concentration of chloroform is significantly higher than the
other quarterly averages and the confidence interval associated with this average is
large. A review of the data shows that the maximum concentration of chloroform was
measured on August 18, 2010 (3.28 |ig/m3) and was an order of magnitude higher
than the next highest concentration (0.391 |ig/m3 measured on October 17, 2010). The
concentration measured on August 18, 2010 is the sixth highest concentration of this
pollutant among sites sampling VOC.
• Concentrations of />-dichlorobenzene were highest in the third quarter and lowest in
the first quarter of 2010 at PROK, similar to TOOK and TMOK. A review of the data
shows that the highest concentrations of this pollutant were measured in June and
July while over half of the non-detects were reported in the first quarter of 2010.
Observations for the Oklahoma City sites from Table 19-5 include the following:
• The second quarter acrylonitrile average for OCOK has a relatively large confidence
interval associated with it, indicating the potential presence of outliers. The
concentration of acrylonitrile on April 26, 2010 (0.744 |ig/m3) is more than twice the
next highest concentration (0.311 |ig/m3 measured on August 10, 2010). Only four of
the 10 measured detections of this pollutant at OCOK were greater than 0.1 |ig/m3.
• While />-dichlorobenzene's trend of being lowest during the first quarter and highest
during the third quarter continues at MWOK, the first quarter average concentration
is actually the highest among the quarterly averages for OCOK. However, this
average has a relatively large confidence interval associated with it. A review of the
data shows that the two highest concentrations of this pollutant were measured at
OCOK on February 19, 2010 and March 2, 2010 (1.27 |ig/m3 and 1.19 |ig/m3,
respectively), although another concentration greater than 1 jig/m3 was also measured
on September 29, 2010. These three concentrations are among the 10 concentrations
greater than 1 |ig/m3 measured among NMP sites sampling VOC.
• The highest annual average concentration of tetrachloroethylene among the
Oklahoma sites was calculated for MWOK. The second and third quarter average
concentrations are greater than the other quarterly average and have relatively large
confidence intervals associated with them. The only two concentrations of
tetrachloroethylene greater than 1 |ig/m3 were measured at MWOK on
August 30, 2010 and June 25, 2010 (1.13 and 1.09 |ig/m3, respectively). These are the
seventh and tenth highest concentrations of tetrachloroethylene measured among
NMP sites sampling VOC.
19-40
-------
• The fourth quarter average concentration of lead for MWOK has a large confidence
interval associated with it, indicating the likely presence of outliers. The maximum
concentration of lead was measured on October 29, 2010 (26.0 ng/m3) and is nearly
five times the next highest concentration measured at MWOK (5.47 ng/m3), also
measured in October. This concentration is the highest lead concentration measured
among the Oklahoma sites.
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for the
Oklahoma sites include the following:
• TOOK has the highest annual average of concentration of benzene among all NMP
sites sampling this pollutant. The annual average for TMOK ranks fifth.
• The annual average concentrations for all five Oklahoma sites ranked among the 10
highest annual average concentrations ofp-dichlorobenzene, with OCOK ranking
fourth.
• PROK and MWOK rank highest and second highest, respectively, for their annual
average concentrations of 1,2-dichloroethane, with OCOK ranking fifth.
• TMOK and TOOK rank fifth and sixth highest, respectively, for their annual average
concentrations of formaldehyde, among sites sampling carbonyl compounds. TOOK
also ranks eighth for acetaldehyde while TMOK ranks tenth.
• The Oklahoma sites were the only NMP sites to monitor for TSP metals; thus, they
are the only sites that appear in Table 4-12 under TSP metals.
• For five of the six TSP metals shown in Table 4-12, TOOK has the highest annual
average concentration among the Oklahoma sites. The only exception is for
beryllium, where PROK ranks highest and TOOK ranks second.
19.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for acetaldehyde, arsenic,
benzene, 1,3-butadiene, formaldehyde, and manganese were created for the Oklahoma sites.
Figures 19-24 through 19-29 overlay the sites' minimum, annual average, and maximum
concentrations onto the program-level minimum, first quartile, average, median, third quartile,
and maximum concentrations, as described in Section 3.5.3.
19-41
-------
Figure 19-24. Program vs. Site-Specific Average Acetaldehyde Concentration
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
o
19-42
-------
Figure 19-25. Program vs. Site-Specific Average Arsenic (TSP) Concentration
m
4t
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
19-43
-------
Figure 19-26. Program vs. Site-Specific Average Benzene Concentration
-o-
•
0
12345
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Ave
• n n n
Site: Site Average Site Minimum/Maximum
6
'rage
7
19-44
-------
Figure 19-27. Program vs. Site-Specific Average 1,3-Butadiene Concentration
t+
1 1 1 1 1
o|
1
1 1 1 1 1 1 1
I
1
1 1
1 1 1 1 1 1 1
0.4 0.5 0.6
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
o
19-45
-------
Figure 19-28. Program vs. Site-Specific Average Formaldehyde Concentration
If
K-
!t
F
10 15 20 25 30 35
Concentration (|ig/m3)
45 50
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
19-46
-------
Figure 19-29. Program vs. Site-Specific Average Manganese (TSP) Concentration
30 35
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Observations from Figures 19-24 through 19-29 include the following:
• Figure 19-24 shows that the annual average acetaldehyde concentrations for
TOOK and TMOK are greater than the program-level average for acetaldehyde.
The range of acetaldehyde concentrations is widest for TOOK and smallest for
MWOK. There were no non-detects of acetaldehyde reported for the Oklahoma
sites.
• Because the Oklahoma sites are the only sites sampling TSP metals, Figure 19-25
compares the individual Oklahoma site data against the combined Oklahoma data.
Figure 19-25 shows that the annual average arsenic (TSP) concentration is highest
for TOOK and lowest for OCOK. This Figure also shows that the range of
measurements of arsenic is widest for TOOK and TMOK. The minimum arsenic
19-47
-------
concentration measured at PROK is the minimum concentration measured among
the five sites sampling TSP metals.
• Figure 19-26 for benzene shows that the annual average concentrations are less
than the program-level average for MWOK and PROK, similar to the program
level average for OCOK, and greater than the program-level average for TMOK
and TOOK. The maximum concentration measured at TOOK is the maximum
benzene concentration measured across the program. There were no non-detects
of benzene measured at any of the Oklahoma sites.
• Figure 19-27 for 1,3-butadiene shows that the annual average concentrations for
the Oklahoma sites are less than the program-level average concentration of
1,3-butadiene, with the exception of TMOK. This site also has the widest range of
1,3-butadiene measurements among the sites, although the maximum
concentration measured at TMOK is well below the program-level maximum
concentration. Several non-detects of 1,3-butadiene were measured at the
Oklahoma sites.
• Figure 19-28 shows that the maximum formaldehyde concentration measured at
each Oklahoma site is well below the maximum concentration measured across
the program. However, the annual average for each Oklahoma site is greater than
the program-level average concentration of formaldehyde, although the annual
average for OCOK is just greater than the program-level average concentration.
There were no non-detects of formaldehyde measured at the Oklahoma sites.
• Because the OK sites are the only sites sampling TSP metals, Figure 19-29
compares the individual Oklahoma site data against the combined Oklahoma data.
Figure 19-29 shows that the annual average manganese (TSP) concentration is
highest for TOOK and lowest for MWOK. This Figure also shows that the range
of manganese measurements was greatest for TOOK and smallest for TMOK. The
maximum manganese concentration measured at TOOK is nearly twice the
maximum concentration for the other sites.
19.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. TOOK has sampled carbonyl compounds and VOC since 2006; thus, Figures
19-30 through 19-33 present the 3-year rolling statistical metrics for acetaldehyde, benzene,
1,3-butadiene, and formaldehyde, respectively. The statistical metrics presented for assessing
trends include the substitution of zeros for non-detects. Although TOOK has also sampled TSP
metals since 2006, sampling for these pollutants did not begin until October 2006. Because three
months of data is not considered representative of a year, a trends analysis was not performed for
the metals.
19-48
-------
Figure 19-30. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at TOOK
6
1
1,
**
2
i
<
•
1
>
^
I
•••*-
1
—
J
>
•
200b 200B 100 f 1009 2UQB 2U1Q
Three-Year Period
1
Figure 19-31. Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured
at TOOK
9
7
i
2
• '
*•••
^
J
M
.
1
•••<
a
1
•
I
2001)2006 2110 /20TO iOUB JUKI
Three-Year Period
I
• Srhprt-frtitile — Minimum - Median — Ma
-------
Figure 19-32. Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations
Measured at TOOK
I
2000 2008 200/201)11 iOUS 2010
Thre«.Year F>* — Minimum - Median — Maximum * 'ir,tl)prt.rtiti|p •••*••
Figure 19-33. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at TOOK
- Minimum - M.,|L.«, - Maximum
19-50
-------
Observations from Figure 19-30 for acetaldehyde measurements at TOOK include the
following:
• Although the maximum concentration of acetaldehyde was measured in 2010
(5.07 |ig/m3), similar concentrations were also measured in 2006 and 2009.
• The rolling average and median concentrations exhibit a slight decrease between
2006-2008 and 2007-2009 followed by a slight increase for 2008-2010. However,
these changes are not statistically significant.
• The 95th percentile increased for the 2008-2010 period. This is not surprising given
that of the 10 concentrations greater than 4 |ig/m3 measured at TOOK, five were
measured in 2010 (plus one each in 2008 and 2009).
Observations from Figure 19-31 for benzene measurements at TOOK include the
following:
• The maximum concentration of benzene was measured in 2008 (8.26 |ig/m3). Since
this year's data is included in all three 3-year periods, this measurement is shown as
the maximum concentration for all three periods. However, a similar concentration
was also measured in 2007.
• The rolling average and median concentrations exhibit a slight decrease between
2006-2008 and 2007-2009 followed by a slight increase for 2008-2010. However,
these changes are not statistically significant. This trend is reflected in the median and
95th percentiles as well.
Observations from Figure 19-32 for 1,3-butadiene measurements at TOOK include the
following:
• Although the maximum concentration of 1,3-butadiene was measured in 2007
(0.326 |ig/m3), a similar concentration was also measured in 2008.
• While the rolling average and median concentrations exhibit a decreasing trend, the
change is not statistically significant.
• The minimum concentration is zero for all three 3-year periods, indicating the
presence of non-detects.
Observations from Figure 19-33 for formaldehyde measurements at TOOK include the
following:
• Although the maximum concentration of formaldehyde was measured in 2007
(9.21 |ig/m3), similar concentrations were also measured in 2006 and 2008.
19-51
-------
• The rolling average concentration exhibits a slight decrease from 2006-2008 to 2007-
2009 but the change is not statistically significant. Little change is shown between
2007-2009 and 2008-2010. The median however, has a slight decreasing trend
throughout the sampling period.
19.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at each
Oklahoma monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
19.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Oklahoma monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
were compared to the acute MRL; the quarterly averages were compared to the intermediate
MRL; and the annual averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the Oklahoma monitoring sites were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
the Oklahoma monitoring sites.
19.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Oklahoma monitoring sites and where annual
average concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 19-6, where applicable.
19-52
-------
Table 19-6. Cancer and Noncancer Surrogate Risk Approximations for the Oklahoma
Monitoring Sites
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
# of Measured
Detections vs.
# of Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Tulsa, Oklahoma - TOOK
Acetaldehyde
Arsenic (TSP) a
Benzene
Beryllium (TSP) a
1,3 -Butadiene
Cadmium (TSP) a
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
Ethylbenzene
Formaldehyde
Lead (TSP) a
Manganese (TSP) a
Nickel (TSP) a
Propionaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.0000022
0.0043
0.0000078
0.0024
0.00003
0.0018
0.000006
0.000011
0.0000025
0.000013
0.00048
_
2.6E-07
0.0000048
0.0000088
0.009
0.000015
0.03
0.00002
0.002
0.00001
0.1
0.098
0.8
1
0.0098
0.00015
0.00005
0.00009
0.008
0.04
0.002
0.1
60/60
61/61
61/61
61/61
56/61
61/61
61/61
38/61
55/61
61/61
60/60
61/61
61/61
61/61
58/60
57/61
11/61
3/61
2.20
±0.29
0.01
±0.01
2.34
±0.36
0.01
±0.01
0.07
±0.01
0.01
±0.01
0.62
±0.03
0.07
±0.02
0.15
±0.03
0.47
±0.08
3.14
±0.45
0.01
±0.01
0.02
±O.01
0.01
±0.01
0.43
±0.07
0.13
±0.02
0.02
±0.01
O.01
±O.01
4.83
2.85
18.24
0.05
2.15
0.51
3.73
1.65
1.19
40.78
0.53
_
0.03
0.09
0.01
0.24
0.04
0.08
0.01
0.04
0.03
0.01
0.01
O.01
0.01
0.32
0.03
0.47
0.01
0.05
O.01
0.01
O.01
- = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 19-5.
19-53
-------
Table 19-6. Cancer and Noncancer Surrogate Risk Approximations for the Oklahoma
Monitoring Sites (Continued)
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
# of Measured
Detections vs.
# of Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Tulsa, Oklahoma - TMOK
Acetaldehyde
Arsenic (TSP) a
Benzene
Beryllium (TSP) a
1,3 -Butadiene
Cadmium (TSP) a
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Ethylbenzene
Formaldehyde
Lead (TSP) a
Manganese (TSP) a
Nickel (TSP) a
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.0000022
0.0043
0.0000078
0.0024
0.00003
0.0018
0.000006
0.000011
0.000026
0.0000025
0.000013
0.00048
2.6E-07
0.0000048
0.0000088
0.009
0.000015
0.03
0.00002
0.002
0.00001
0.1
0.098
0.8
2.4
1
0.0098
0.00015
0.00005
0.00009
0.04
0.002
0.1
60/60
60/60
61/61
59/60
57/61
60/60
61/61
39/61
57/61
10/61
61/61
60/60
60/60
60/60
60/60
49/61
10/61
2/61
2.00
±0.20
0.01
±0.01
1.57
±0.20
0.01
±0.01
0.10
±0.02
0.01
±0.01
0.61
±0.03
0.07
±0.02
0.14
±0.03
0.01
±0.01
0.55
±0.13
3.35
±0.40
O.01
±O.01
0.02
±0.01
O.01
±0.01
0.09
±0.02
0.01
±0.01
O.01
±O.01
4.40
2.67
12.25
0.03
3.00
0.41
3.64
1.54
0.36
1.38
43.61
0.43
0.02
0.05
O.01
0.22
0.04
0.05
0.01
0.05
0.02
0.01
0.01
O.01
0.01
O.01
0.34
0.02
0.32
0.01
O.01
0.01
O.01
- = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 19-5.
19-54
-------
Table 19-6. Cancer and Noncancer Surrogate Risk Approximations for the Oklahoma
Monitoring Sites (Continued)
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
# of Measured
Detections vs.
# of Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Pryor Creek, Oklahoma - PROK
Acetaldehyde
Arsenic (TSP) a
Benzene
Beryllium (TSP) a
1,3 -Butadiene
Cadmium (TSP) a
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Formaldehyde
Lead (TSP) a
Manganese (TSP) a
Nickel (TSP) a
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.0000022
0.0043
0.0000078
0.0024
0.00003
0.0018
0.000006
0.000011
0.000026
0.000013
0.00048
2.6E-07
0.0000048
0.0000088
0.009
0.000015
0.03
0.00002
0.002
0.00001
0.1
0.098
0.8
2.4
0.0098
0.00015
0.00005
0.00009
0.04
0.002
0.1
60/60
61/61
61/61
61/61
50/61
61/61
61/61
44/61
48/61
15/61
60/60
61/61
61/61
61/61
29/61
3/61
1/61
1.37
±0.14
0.01
±0.01
0.71
±0.06
0.01
±0.01
0.04
±0.01
0.01
±0.01
0.65
±0.03
0.14
±0.11
0.10
±0.03
0.02
±0.01
2.64
±0.41
0.01
±0.01
0.01
±O.01
0.01
±0.01
0.03
±0.01
0.01
±O.01
0.01
±0.01
3.01
2.37
5.54
0.05
1.11
0.30
3.89
1.12
0.54
34.31
0.39
0.01
0.01
0.01
0.15
0.04
0.02
0.01
0.02
0.02
0.01
0.01
O.01
0.01
0.27
0.02
0.26
0.01
0.01
O.01
0.01
- = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 19-5.
19-55
-------
Table 19-6. Cancer and Noncancer Surrogate Risk Approximations for the Oklahoma
Monitoring Sites (Continued)
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
# of Measured
Detections vs.
# of Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Midwest City, Oklahoma - MWOK
Acetaldehyde
Arsenic (TSP) a
Benzene
Beryllium (TSP) a
1,3 -Butadiene
Cadmium (TSP) a
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Formaldehyde
Lead (TSP) a
Manganese (TSP) a
Nickel (TSP) a
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.0000022
0.0043
0.0000078
0.0024
0.00003
0.0018
0.000006
0.000011
0.000026
0.000013
0.00048
2.6E-07
0.0000048
0.0000088
0.009
0.000015
0.03
0.00002
0.002
0.00001
0.1
0.098
0.8
2.4
0.0098
0.00015
0.00005
0.00009
0.04
0.002
0.1
60/60
61/61
61/61
61/61
52/61
61/61
61/61
45/61
61/61
13/61
60/60
61/61
61/61
61/61
53/61
4/61
3/61
1.41
±0.14
0.01
±0.01
0.88
±0.10
0.01
±0.01
0.06
±0.01
0.01
±0.01
0.64
±0.04
0.07
±0.01
0.18
±0.04
0.02
±0.01
2.56
±0.38
0.01
±0.01
0.01
±O.01
0.01
±0.01
0.22
±0.06
0.01
±O.01
0.01
±0.01
3.10
1.90
6.88
0.03
1.94
0.18
3.86
1.98
0.50
33.30
0.47
0.06
0.02
0.02
0.16
0.03
0.03
0.01
0.03
0.01
0.01
0.01
O.01
0.01
0.26
0.02
0.19
0.01
0.01
O.01
0.01
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 19-5.
19-56
-------
Table 19-6. Cancer and Noncancer Surrogate Risk Approximations for the Oklahoma
Monitoring Sites (Continued)
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
# of Measured
Detections vs.
# of Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer Risk
Approximation
(HQ)
Oklahoma City, Oklahoma - OCOK
Acetaldehyde
Acrylonitrile
Arsenic (TSP)a
Benzene
Beryllium (TSP) a
1,3 -Butadiene
Cadmium (TSP) a
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Formaldehyde
Lead (TSP)a
Manganese (TSP) a
Nickel (TSP) a
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.0000022
0.000068
0.0043
0.0000078
0.0024
0.00003
0.0018
0.000006
0.000011
0.000026
0.000013
0.00048
2.6E-07
0.0000048
0.0000088
0.009
0.002
0.000015
0.03
0.00002
0.002
0.00001
0.1
0.098
0.8
2.4
0.0098
0.00015
0.00005
0.00009
0.04
0.002
0.1
60/60
10/61
61/61
61/61
61/61
48/61
61/61
61/61
40/61
54/61
11/61
60/60
61/61
61/61
61/61
53/61
4/61
4/61
1.59
±0.17
0.03
±0.03
<0.01
±<0.01
1.01
±0.13
<0.01
±<0.01
0.04
±0.01
<0.01
±<0.01
0.64
±0.04
0.06
±0.01
0.24
±0.07
0.02
±0.01
2.50
±0.36
<0.01
±<0.01
0.01
±0.01
<0.01
±<0.01
0.09
±0.02
0.01
±0.02
<0.01
±<0.01
3.51
1.90
1.90
7.91
0.03
1.31
0.18
3.84
2.65
0.45
32.54
0.28
0.02
0.06
0.01
0.18
0.01
0.03
0.03
<0.01
0.02
0.01
0.01
<0.01
<0.01
<0.01
0.26
0.01
0.24
0.01
<0.01
0.01
<0.01
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 19-5.
19-57
-------
Observations from Table 19-6 include the following:
• Formaldehyde has the highest annual average concentration by mass for each site.
Among the TSP metals, the annual average concentrations of manganese are the
highest for each site.
• Formaldehyde and benzene have the highest cancer risk approximations for all of the
Oklahoma monitoring sites. Formaldehyde cancer risk approximations range from
32.54 in-a-million for OCOK to 43.61 in-a-million for TMOK. Benzene cancer risk
approximations range from 5.54 in-a-million for PROK to 18.24 in-a-million for
TOOK.
• Among the metals, arsenic has the highest cancer risk approximations for all of the
Oklahoma monitoring sites, ranging from 1.90 in-a-million for MWOK and OCOK to
2.85 in-a-million for TOOK.
• None of the pollutants of interest have noncancer risk approximations greater than
1.0, the HQ level of concern for noncancer. Among the noncancer risk
approximations for the Oklahoma sites, formaldehyde, manganese, and acetaldehyde
have the highest noncancer risk approximations for each site.
19.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 19-7 and 19-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 19-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer risk approximations (in-a-million), as calculated from the annual averages. Table 19-8
presents similar information, but identifies the 10 pollutants with the highest noncancer risk
approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 19-7 and 19-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer surrogate risk approximations based on each site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 19.3, the Oklahoma sites sampled VOC, carbonyl compounds, and TSP metals. In
addition, the cancer and noncancer risk approximations are limited to those pollutants with
enough data to meet the criteria for annual averages to be calculated, as discussed in previous
sections. A more in-depth discussion of this analysis is provided in Section 3.5.5.3.
19-58
-------
Table 19-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Oklahoma Monitoring Sites
Top 10 Total Emissions for Pollutants with Cancer
Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Tulsa, Oklahoma (Tulsa County) - TOOK
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Tetrachloroethylene
Naphthalene
Dichloromethane
POM, Group 2b
Nickel, PM
398.15
236.92
198.13
105.13
52.47
41.33
23.57
5.51
3.59
0.40
Benzene
Formaldehyde
1,3 -Butadiene
Hexavalent Chromium, PM
Naphthalene
Ethylbenzene
POM, Group 2b
Acetaldehyde
POM, Group 3
Nickel, PM
3.11E-03
2.58E-03
1.57E-03
1.11E-03
8.02E-04
5.92E-04
3.16E-04
2.31E-04
2.30E-04
1.90E-04
Formaldehyde
Benzene
Acetaldehyde
Carbon Tetrachloride
Arsenic
1,3 -Butadiene
£>-Dichlorobenzene
Ethylbenzene
Nickel
Cadmium
40.78
18.24
4.83
3.73
2.85
2.15
1.65
1.19
0.53
0.51
Tulsa, Oklahoma (Tulsa County) - TMOK
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Tetrachloroethylene
Naphthalene
Dichloromethane
POM, Group 2b
Nickel, PM
398.15
236.92
198.13
105.13
52.47
41.33
23.57
5.51
3.59
0.40
Benzene
Formaldehyde
1,3 -Butadiene
Hexavalent Chromium, PM
Naphthalene
Ethylbenzene
POM, Group 2b
Acetaldehyde
POM, Group 3
Nickel, PM
3.11E-03
2.58E-03
1.57E-03
1.11E-03
8.02E-04
5.92E-04
3.16E-04
2.31E-04
2.30E-04
1.90E-04
Formaldehyde
Benzene
Acetaldehyde
Carbon Tetrachloride
1,3 -Butadiene
Arsenic
£>-Dichlorobenzene
Ethylbenzene
Nickel
Cadmium
43.61
12.25
4.40
3.64
3.00
2.67
1.54
1.38
0.43
0.41
VO
(!/i
VO
-------
Table 19-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Oklahoma Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with Cancer
Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Pryor Creek, Oklahoma (Mayes County) - PROK
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Chloromethylbenzene
Nickel, PM
Dichloromethane
Arsenic, PM
37.28
24.25
17.46
12.40
3.72
2.04
1.60
1.17
1.03
0.53
Arsenic, PM
Hexavalent Chromium, PM
Nickel, PM
Formaldehyde
Benzene
1,3 -Butadiene
Beryllium, PM
Cadmium, PM
Chloromethylbenzene
Naphthalene
2.29E-03
9.45E-04
5.62E-04
3.15E-04
2.91E-04
1.11E-04
8.46E-05
8.35E-05
7.85E-05
6.93E-05
Formaldehyde
Benzene
Carbon Tetrachloride
Acetaldehyde
Arsenic
£>-Dichlorobenzene
1,3 -Butadiene
1 ,2-Dichloroethane
Nickel
Cadmium
34.31
5.54
3.89
3.01
2.37
1.12
1.11
0.54
0.39
0.30
Midwest City, Oklahoma (Oklahoma County) - MWOK
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
Tetrachloroethylene
POM, Group 2b
Bis(2-ethylhexyl)phthalate (DEHP), gas
447.79
275.02
250.63
137.04
59.70
29.28
15.08
10.79
4.51
0.49
Benzene
Formaldehyde
1,3 -Butadiene
Naphthalene
Ethylbenzene
Hexavalent Chromium, PM
POM, Group 2b
Acetaldehyde
POM, Group 3
Arsenic, PM
3.49E-03
3.26E-03
1.79E-03
9.96E-04
6.88E-04
6.16E-04
3.97E-04
3.01E-04
2.64E-04
1.85E-04
Formaldehyde
Benzene
Carbon Tetrachloride
Acetaldehyde
£>-Dichlorobenzene
1,3 -Butadiene
Arsenic
1 ,2-Dichloroethane
Nickel
Cadmium
33.30
6.88
3.86
3.10
1.98
1.94
1.90
0.50
0.47
0.18
-------
Table 19-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Oklahoma Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with Cancer
Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Oklahoma City, Oklahoma (Oklahoma County) - OCOK
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
Tetrachloroethylene
POM, Group 2b
Bis(2-ethylhexyl)phthalate (DEHP), gas
447.79
275.02
250.63
137.04
59.70
29.28
15.08
10.79
4.51
0.49
Benzene
Formaldehyde
1,3 -Butadiene
Naphthalene
Ethylbenzene
Hexavalent Chromium, PM
POM, Group 2b
Acetaldehyde
POM, Group 3
Arsenic, PM
3.49E-03
3.26E-03
1.79E-03
9.96E-04
6.88E-04
6.16E-04
3.97E-04
3.01E-04
2.64E-04
1.85E-04
Formaldehyde
Benzene
Carbon Tetrachloride
Acetaldehyde
£>-Dichlorobenzene
Acrylonitrile
Arsenic
1,3 -Butadiene
1 ,2-Dichloroethane
Nickel
32.54
7.91
3.84
3.51
2.65
1.90
1.90
1.31
0.45
0.28
VO
-------
Table 19-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Oklahoma Monitoring Sites
to
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted
Emissions (County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Tulsa, Oklahoma (Tulsa County) - TOOK
Toluene
Xylenes
Methanol
Benzene
Hexane
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Ethylene glycol
1,180.06
902.52
402.45
398.15
293.67
236.92
198.13
105.13
52.47
50.77
Acrolein
1,3 -Butadiene
Formaldehyde
Manganese, PM
Benzene
Lead, PM
Acetaldehyde
Xylenes
Naphthalene
Cobalt , PM
577,473.50
26,235.74
20,217.32
18,832.62
13,271.68
13,200.35
11,681.56
9,025.21
7,857.87
5,351.77
Manganese
Formaldehyde
Acetaldehyde
Benzene
Propionaldehyde
Arsenic
1,3 -Butadiene
Lead
Cadmium
Nickel
0.47
0.32
0.24
0.08
0.05
0.04
0.04
0.03
0.03
0.01
Tulsa, Oklahoma (Tulsa County) - TMOK
Toluene
Xylenes
Methanol
Benzene
Hexane
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Ethylene glycol
1,180.06
902.52
402.45
398.15
293.67
236.92
198.13
105.13
52.47
50.77
Acrolein
1,3 -Butadiene
Formaldehyde
Manganese, PM
Benzene
Lead, PM
Acetaldehyde
Xylenes
Naphthalene
Cobalt , PM
577,473.50
26,235.74
20,217.32
18,832.62
13,271.68
13,200.35
11,681.56
9,025.21
7,857.87
5,351.77
Formaldehyde
Manganese
Acetaldehyde
Benzene
1,3 -Butadiene
Arsenic
Lead
Cadmium
Nickel
Carbon Tetrachloride
0.34
0.32
0.22
0.05
0.05
0.04
0.02
0.02
0.01
0.01
-------
Table 19-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Oklahoma Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted
Emissions (County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Pryor Creek, Oklahoma (Mayes County) - PROK
Hydrochloric acid
Toluene
Xylenes
Benzene
Cyanide Compounds, gas
Hydrofluoric acid
Formaldehyde
Methanol
Hexane
Ethylbenzene
145.18
89.08
72.70
37.28
34.90
26.35
24.25
23.55
22.30
17.46
Acrolein
Chlorine
Cyanide Compounds, gas
Arsenic, PM
Nickel, PM
Manganese, PM
Hydrochloric acid
Cyanide Compounds, PM
Cadmium, PM
Lead, PM
91,374.65
61,006.67
43,619.17
35,508.85
13,015.34
8,604.28
7,259.15
7,154.96
4,636.70
4,369.13
Formaldehyde
Manganese
Acetaldehyde
Arsenic
Benzene
Lead
1,3 -Butadiene
Cadmium
Nickel
Carbon Tetrachloride
0.27
0.26
0.15
0.04
0.02
0.02
0.02
0.02
0.01
0.01
Midwest City, Oklahoma (Oklahoma County) - MWOK
Toluene
Xylenes
Methanol
Benzene
Hexane
Ethylbenzene
Formaldehyde
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
1,373.20
1,058.17
468.16
447.79
337.57
275.02
250.63
137.04
61.51
59.70
Acrolein
1,3 -Butadiene
Formaldehyde
Acetaldehyde
Benzene
Xylenes
Naphthalene
Arsenic, PM
Lead, PM
Propionaldehyde
875,997.95
29,851.09
25,574.80
15,226.70
14,926.42
10,581.67
9,759.95
2,874.57
2,417.07
2,079.24
Formaldehyde
Manganese
Acetaldehyde
1,3 -Butadiene
Arsenic
Benzene
Lead
Nickel
Cadmium
Carbon Tetrachloride
0.26
0.19
0.16
0.03
0.03
0.03
0.02
0.01
0.01
0.01
-------
Table 19-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Oklahoma Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted
Emissions (County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Oklahoma City, Oklahoma (Oklahoma County) - OCOK
Toluene
Xylenes
Methanol
Benzene
Hexane
Ethylbenzene
Formaldehyde
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
1,373.20
1,058.17
468.16
447.79
337.57
275.02
250.63
137.04
61.51
59.70
Acrolein
1,3 -Butadiene
Formaldehyde
Acetaldehyde
Benzene
Xylenes
Naphthalene
Arsenic, PM
Lead, PM
Propionaldehyde
875,997.95
29,851.09
25,574.80
15,226.70
14,926.42
10,581.67
9,759.95
2,874.57
2,417.07
2,079.24
Formaldehyde
Manganese
Acetaldehyde
Benzene
Arsenic
1,3 -Butadiene
Acrylonitrile
Lead
Cadmium
Nickel
0.26
0.24
0.18
0.03
0.03
0.02
0.01
0.01
0.01
0.01
-------
Observations from Table 19-7 include the following:
• Benzene is the highest emitted pollutant with a cancer URE in Mayes, Oklahoma, and
Tulsa Counties, followed by ethylbenzene and formaldehyde in Oklahoma and Tulsa
Counties and formaldehyde and ethylbenzene in Mayes County. The emissions of
these pollutants in Mayes County are an order of magnitude lower than the emissions
for Oklahoma and Tulsa Counties.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
cancer UREs) for Oklahoma and Tulsa Counties are benzene, formaldehyde, and
1,3-butadiene. The pollutants with the highest toxicity-weighted emissions for Mayes
County are arsenic, hexavalent chromium, and nickel.
• Eight of the highest emitted pollutants in Tulsa County also have the highest toxicity-
weighted emissions. Seven of the highest emitted pollutants in Mayes County also
have the highest toxicity-weighted emissions. Seven of the highest emitted pollutants
in Oklahoma County also have the highest toxicity-weighted emissions.
• Benzene and formaldehyde have the highest cancer risk approximations among the
Oklahoma sites' pollutants of interest. These pollutants appear on both emissions-
based lists for all five sites. Conversely, carbon tetrachloride, another pollutant with
relatively high cancer risk approximations, does not appear on either emissions-based
list.
• While hexavalent chromium is among the pollutants with the highest toxicity-
weighted emissions for each county, it is not among the highest emitted pollutants.
This indicates that lower emissions can translate to higher risk levels.
• The toxicity-weighted pollutants listed for Mayes County are considerably different
than for the other two counties. There are five metals listed for Mayes County while
the other counties only have two each. In addition, there are no POM Groups listed
for Mayes County, while POM, Groups 2b and 3 appear for Oklahoma and Tulsa
Counties.
Observations from Table 19-8 include the following:
• Toluene, xylenes, and methanol are the highest emitted pollutants with noncancer
RfCs in Oklahoma and Tulsa Counties. Hydrochloric acid, toluene, and xylenes are
the highest emitted pollutants with noncancer RfCs in Mayes County. Note that the
magnitude of the emissions is much higher in Tulsa and Oklahoma Counties than in
Mayes County.
• Acrolein is the pollutant with the highest toxicity-weighted emissions (of the
pollutants with noncancer RfCs) for all three counties. Yet, this pollutant is not
among the highest emitted pollutants for any of the three counties. This indicates that
lower emissions can translate to higher risk levels. Acrolein was sampled for at all of
the Oklahoma sites, but this pollutant was excluded from the pollutants of interest
designation, and thus subsequent risk screening evaluations, due to questions about
the consistency and reliability of the measurements, as discussed in Section 3.2.
19-65
-------
• Three of the highest emitted pollutants in Mayes County also have the highest
toxi city-weighted emissions; five of the highest emitted pollutants in Tulsa and
Oklahoma Counties also have the highest toxicity-weighted emissions.
• Five of the 10 pollutants with the highest noncancer toxicity-weighted emissions in
Mayes County were metals. Cyanide compounds, gaseous and particulate, account for
two more.
• Formaldehyde and manganese have the highest noncancer risk approximations among
the Oklahoma sites. Formaldehyde appears on both emissions-based lists for Tulsa
and Oklahoma Counties but ranks 11th for toxicity-emissions for Mayes County and
therefore does not appear in Table 19-8 in that column. Manganese appears among
the pollutants with the highest toxicity-weighted emissions for Tulsa and Mayes
Counties but not Oklahoma County. There are no metals listed among the highest
emitted pollutants.
• It is important to note that for the metals, the emissions-based lists are PMi0 while the
Oklahoma sites sampled TSP metals.
19.6 Summary of the 2010 Monitoring Data for the Oklahoma Monitoring Sites
Results from several of the data treatments described in this section include the
following:
»«» Seventeen pollutants failed at least one screen for TOOK; 16 pollutants failed screens
for TMOK; 13 pollutants failed screens for PROK; 18 pollutants failed screens for
MWOK; and 19 pollutants failed screens for OCOK.
»«» Formaldehyde had the highest annual average concentration by mass for each site.
Among the TSP metals, the annual average concentrations of manganese were the
highest for each site.
»«» TOOK had the highest annual average of concentration of benzene among all NMP
sites sampling this pollutant.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
19-66
-------
20.0 Site in Rhode Island
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in Rhode Island, and integrates these concentrations
with emissions, meteorological, and risk information. Data generated by sources other than ERG
are not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
20.1 Site Characterization
This section characterizes the PRRI monitoring site by providing geographical and
physical information about the location of the site and the surrounding area. This information is
provided to give the reader insight regarding factors that may influence the air quality near the
site and assist in the interpretation of the ambient monitoring measurements.
The PRRI monitoring site is located in south Providence. Figure 20-1 is a composite
satellite image retrieved from ArcGIS Explorer showing the monitoring site in its urban location.
Figure 20-2 identifies point source emissions locations by source category, as reported in the
2008 NEI for point sources. Note that only sources within 10 miles of the site are included in the
facility counts provided in Figure 20-2. Thus, sources outside the 10-mile radius have been
grayed out, but are visible on the map to show emissions sources outside the 10-mile boundary.
A 10-mile boundary was chosen to give the reader an indication of which emissions sources and
emissions source categories could potentially have a direct effect on the air quality at the
monitoring site. Further, this boundary provides both the proximity of emissions sources to the
monitoring site as well as the quantity of such sources within a given distance of the site.
Table 20-1 describes the area surrounding the monitoring site by providing supplemental
geographical information such as land use, location setting, and locational coordinates.
20-1
-------
Figure 20-1. Providence, Rhode Island (PRRI) Monitoring Site
to
o
-------
Figure 20-2. NEI Point Sources Located Within 10 Miles of PRRI
Legend
7V SO'fTW 71 "KWVf 71
Note: Due to faculty density and collocation, the total facilities
displayed may nol ropjcsont all facilities within the area of interest
PRRI NATTS site 10 mile radius
County boundary
Source Category Group (No. of Facilities)
H^ Aircran Operations (13)
$3 Automobile/Truck Manufacturing (2)
£ Battery (2)
£ Boat Manufacturing (1)
C Chemical Manufacturing {7}
• Concrete Batch Plant (2)
X Crematory -Animal/Human (1)
(D Dry Cleaning Facility (37)
6 Eleclncal Equipment (4)
£ Eleclncity Gerseralion via Combustion (4)
E Electroplating. Plating, Polishing, Anodizing & Coloring (24)
<•> Fabricated Metal Products (25)
id> Flexible Polyurethane Foam Production (I >
F Food Processing/Agncullure {4)
[T] Furniture Plani (2)
jf Gasolinej'Diesel Service Station (4)
fV Glass Manufacturing (1)
(3 Hospital (3)
J| Hot Mix Asphalt Plant (4)
-^ Industnal Machinery and Equipment (5)
^ In stilutional - school (11)
A LandMI (1i
? Miscellaneous Commercial/Industrial {21}
M Miscellaneous Manufacturing (45)
4 Oil and/or Gas Production 11)
•-—.• Pharmaceutical Manufacturing (1)
1 Primary Metal Production (6)
P Printing/Publishing (11)
El PulP and Paper Plant/Wood Products (4)
R Rubber and Miscellaneous Plastics Products (10)
< Site Remediation Activity (1 >
S Surface Coaling (3)
TT Telecommuntcatjons (4)
T Textile Mril (12)
I V&stewater Treatment (1)
W Waochvork Furmlure Milhvork & Wood Preserving (1)
20-3
-------
Table 20-1. Geographical Information for the Rhode Island Monitoring Site
Site
Code
PRRI
AQS Code
44-007-0022
Location
Providence
County
Providence
Micro- or
Metropolitan
Statistical Area
Providence-New
Bedford-Fall
River, RI-MA
MSA
Latitude
and
Longitude
41.807949,
-71.415
Land Use
Residential
Location
Setting
Urban/City
Center
Additional Ambient Monitoring Information1
PAMS, VOC, Carbonyl Compounds, Meteorological
parameters, PM10, PM10 Speciation, Black Carbon,
PM2 5, and PM2 5 Speciation, Germanium.
1 This monitoring site reports additional pollutants to AQS (EPA, 2012h); however, these data are not generated by ERG and are therefore not included in this report.
BOLD ITALICS = EPA-designaled NATTS Site.
to
o
-------
Figure 20-1 shows that the areas to the west and south of PRRI are residential, but areas
to the north and east are commercial. A hospital lies to the northeast of the site, just north of
Dudley Street. About 1/2 mile to the east 1-95 runs north-south, then turns northwestward,
entering downtown Providence. Narragansett Bay and the Port of Providence are a few tenths of
a mile farther to the east, just on the other side of 1-95. Figure 20-2 shows that a large number of
point sources are located within 10 miles of PRRI, especially to the north of the site. Many of
these sources seem to parallel 1-95. The source categories with the largest number of point
sources include dry cleaners; fabricated metals products facilities; electroplating, plating,
polishing, anodizing, and coloring facilities; and aircraft operations, which includes airports as
well as small runways, heliports, or landing pads.
Table 20-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the Rhode
Island monitoring site. County-level vehicle registration data for Providence County was not
available from the State of Rhode Island. Thus, state-level registration, which was obtained from
the Federal Highway Administration (FHWA, 2011), was allocated to the county level using the
county-level portion of the state population. State-level and county-level population information
was obtained from the U.S. Census Bureau. Table 20-2 also includes a vehicle registration-to-
county population ratio (vehicles-per-person). In addition, the population within 10 miles of the
site is presented. An estimate of 10-mile vehicle ownership was calculated by applying the
county-level vehicle registration-to-population ratio to the 10-mile population surrounding the
monitoring site. Table 20-2 also contains annual average daily traffic information. County-level
VMT was not readily available; thus, daily VMT for PRRI is not shown in Table 20-2.
Table 20-2. Population, Motor Vehicle, and Traffic Information for the Rhode Island
Monitoring Site
Site
PRRI
Estimated
County
Population1
627,070
County-level
Vehicle
Registration2
485,837
Vehicles per Person
(Registration:
Population)
0.77
Population
within 10
miles3
660,225
Estimated
10-mile
Vehicle
Ownership
511,525
Annual
Average
Daily
Traffic4
136,800
County-
level
Daily
VMT5
NA
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects a ratio based on 2010 state-level vehicle registration data from the
FHWA and the county-level proportion of the state population data (FHWA, 2011)
3 10-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2009 data from the Rhode Island DOT (RI DOT, 2009)
5 County-level VMT was not available for this site
BOLD ITALICS = EPA-designaled NATTS Site.
20-5
-------
Observations from Table 20-2 include the following:
• Providence County's population is in the middle of the range compared to other
counties with NMP sites, as is the 10-mile population.
• The county-level vehicle registration is in the middle of the range compared to other
counties with NMP sites, as is the 10-mile ownership estimate.
• The vehicle-per-person ratio is in the bottom third compared to other NMP sites.
• The traffic volume experienced near PRRI is the ninth highest compared to other
monitoring sites. The traffic estimate used came from 1-95 near the 1-195 interchange.
20.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in Rhode Island on sample days, as well as over the course of the year.
20.2.1 Climate Summary
Providence is a coastal city on the Narragansett Bay, which opens to the Rhode Island
Sound and the Atlantic Ocean. The city's proximity to the Sound and the Atlantic Ocean temper
cold air outbreaks, and breezes off the ocean moderate summertime heat. On average, southerly
and southwesterly winds in the summer become northwesterly in the winter and precipitation in
Providence is well distributed throughout the year. Weather is fairly variable as frequent storm
systems affect the New England region (Bair, 1992).
20.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest PRRI were retrieved
for 2010 (NCDC, 2010). The closest weather station is located at Theodore F. Green State
Airport (WBAN 14765). Additional information about the T.F. Green weather station, such as
the distance between the site and the weather station, is provided in Table 20-3. These data were
used to determine how meteorological conditions on sample days vary from normal conditions
throughout the year.
20-6
-------
Table 20-3. Average Meteorological Conditions near the Rhode Island Monitoring Site
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Providence, Rhode Island - PRRI
Theodore F.
Green State
Airport
14765
(41.72, -71.43)
6.01
miles
173°
(S)
Sample
Day
2010
61.5
±4.5
61.8
+ 1.9
53.3
±4.2
53.7
+ 1.8
41.0
±4.5
41.1
+ 1.9
47.6
±3.9
47.8
+ 1.7
65.9
±3.6
65.7
+ 1.5
1012.8
±2.0
1012.7
+ 0.8
7.4
±0.7
7.7
+ 0.3
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
to
o
-------
Table 20-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 20-3 is the
95 percent confidence interval for each parameter. As shown in Table 20-3, average
meteorological conditions on sample days were representative of average weather conditions
throughout the year.
20.2.3 Back Trajectory Analysis
Figure 20-3 is the composite back trajectory map for days on which samples were
collected at the PRRI monitoring site in 2010. Included in Figure 20-3 are four back trajectories
per sample day. Figure 20-4 is the corresponding cluster analysis for 2010. An in-depth
description of these maps and how they were generated is presented in Section 3.5.2.1. For the
composite map, each line represents the 24-hour trajectory along which a parcel of air traveled
toward the monitoring site on a given sample day and time. For the cluster analysis, each line
corresponds to a back trajectory representative of a given cluster of trajectories. For both maps,
each concentric circle around the site in Figures 20-3 and 20-4 represents 100 miles.
Observations from Figures 20-3 and 20-4 for PRRI include the following:
• Back trajectories originated from a variety of directions at PRRI, although the bulk of
trajectories originated from the west, northwest, and north.
• The airshed domain for PRRI was among the largest in size compared to other NMP
sites. The farthest away a trajectory originated was off the South Carolina coast and
over the Atlantic Ocean, or nearly 800 miles away. The two longest trajectories are
associated with a strong low pressure system that moved through the region
January 25-26, 2010. While the average trajectory length was 306 miles long,
87 percent of back trajectories originated within 500 miles of the site.
• The cluster analysis shows that 55 percent of back trajectories originated from the
west, northwest, and north, although of differing lengths, as represented by the two
clusters originating over Canada (18 and 19 percent) and the short cluster originating
over New York (18 percent). The short cluster originating offshore and northeast of
Boston (23 percent) represents back trajectories of shorter length, generally less than
250 miles, originating to the north, east, and south over Maine, New Hampshire,
Massachusetts and the offshore waters adjacent to those states, including Rhode
Island. Sixteen percent of trajectories originated to the south, over New Jersey,
Delaware, Maryland, and the offshore waters of those states and Virginia. Finally, six
20-8
-------
percent of back trajectories originated over Newfoundland and New Brunswick,
Canada and farther offshore.
Figure 20-3. 2010 Composite Back Trajectory Map for PRRI
Figure 20-4. Back Trajectory Cluster Map for PRRI
20-9
-------
20.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at T.F. Green Airport near PRRI were
uploaded into a wind rose software program to produce customized wind roses, as described in
Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals" positioned
around a 16-point compass, and uses different colors to represent wind speeds.
Figure 20-5 presents three different wind roses for the PRRI monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location.
Observations from Figure 20-5 for PRRI include the following:
• The NWS weather station at T.F. Green Airport is located approximately 6 miles
south of PRRI.
• The historical wind rose for PRRI shows that while westerly winds were observed the
most (11 percent of observations), wind directions from the western quadrants and
due north and due south are common near PRRI. Calm winds (< 2 knots) account for
less than nine percent of the hourly measurements.
• The wind patterns shown on the 2010 wind rose for PRRI are similar to the historical
wind patterns, although there were slightly more westerly and west-northwesterly
wind observations and somewhat fewer southwesterly wind observations in 2010.
These similarities indicate that winds in 2010 near PRRI were similar to what is
expected climatologically.
• The wind patterns shown on the sample day wind rose are similar to the full-year and
historical wind patterns, although there were more north-northeasterly winds on
sample days. This indicates that conditions on sample days were generally
representative of conditions experienced throughout the year.
20-10
-------
Figure 20-5. Wind Roses for the T.F. Green State Airport Weather Station near PRRI
1999-2009 Historical Wind Rose
2010 Wind Rose
.'VEST
2010 Sample Day Wind Rose
Distance between PRRI and NWS Station
/VEST
20-11
-------
20.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Rhode Island monitoring
site in order to allow analysts and readers to focus on a subset of pollutants through the context
of risk. Each pollutant's preprocessed daily measurement was compared to its associated risk
screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by the monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 20-4 presents PRRI's pollutants of interest. The pollutants that failed at least one
screen and contributed to 95 percent of the total failed screens for PRRI are shaded. NATTS
MQO Core Analytes are bolded. Thus, pollutants of interest are shaded and/or bolded. PRRI
sampled for PAH and hexavalent chromium.
Table 20-4. Risk Screening Results for the Rhode Island Monitoring Site
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Providence, Rhode Island - PRRI
Naphthalene
Benzo(a)pyrene
Fluorene
Hexavalent Chromium
0.029
0.00057
0.011
0.000083
Total
56
6
2
1
65
58
49
58
44
209
96.55
12.24
3.45
2.27
31.10
86.15
9.23
3.08
1.54
86.15
95.38
98.46
100.00
Observations from Table 20-4 include the following:
• Four pollutants (naphthalene, benzo(a)pyrene, fluorene, and hexavalent chromium)
failed screens for PRRI. Naphthalene accounted for 86 percent of PRRI's failed
screens (56 out of 65 total failed screens).
• Naphthalene and benzo(a)pyrene were identified as the only pollutants of interest for
PRRI based on the risk screening process. Hexavalent chromium were added to
PRRI's pollutants of interest because is it a NATTS MQO Core Analyte, even though
it did not contribute to 95 percent of failed screens.
20-12
-------
20.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Rhode Island monitoring site. Concentration averages are provided for the pollutants of
interest for the PRRI monitoring site, where applicable. Concentration averages for select
pollutants are also presented graphically for the site, where applicable, to illustrate how the site's
concentrations compare to the program-level averages. In addition, concentration averages for
select pollutants are presented from previous years of sampling in order to characterize
concentration trends at the site, where applicable. Additional site-specific statistical summaries
are provided in Appendices M and O.
20.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for the Rhode Island site, as described in Section 3.1. The quarterly average of a particular
pollutant is simply the average concentration of the preprocessed daily measurements over a
given calendar quarter. Quarterly average concentrations include the substitution of zeros for all
non-detects. A site must have a minimum of 75 percent valid samples of the total number of
samples possible within a given quarter for a quarterly average to be calculated. An annual
average includes all measured detections and substituted zeros for non-detects for the entire year
of sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for PRRI are presented in
Table 20-5, where applicable. Note that if a pollutant was not detected in a given calendar
quarter, the quarterly average simply reflects "0" because only zeros substituted for non-detects
were factored into the quarterly average concentration.
20-13
-------
Table 20-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Rhode Island Monitoring Site
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(ng/m3)
2nd
Quarter
Average
(ng/m3)
3rd
Quarter
Average
(ng/m3)
4th
Quarter
Average
(ng/m3)
Annual
Average
(ng/m3)
Providence, Rhode Island - PRRI
Benzo(a) pyrene
Hexavalent Chromium
Naphthalene
49/58
44/60
58/58
0.39
±0.14
0.01
±0.02
102.29
±37.14
0.19
±0.18
0.02
±<0.01
70.36
±20.01
0.04
±0.02
0.03
±0.01
70.29
± 15.60
0.21
±0.10
0.01
±0.01
115.14
±38.90
0.20
±0.07
0.02
±<0.01
89.63
± 14.87
Observations for PRRI from Table 20-5 include the following:
• The annual average concentration of naphthalene is significantly higher than the
annual averages of the other two pollutants of interest.
• The first and fourth quarter averages of naphthalene are higher than the second and
third quarter averages. Of the 10 concentrations of naphthalene greater than
125 ng/m3, most were measured during the first and fourth quarters of 2010 at PRRI.
• The first quarter average concentration of hexavalent chromium has a relatively high
confidence interval associated with it. The maximum hexavalent chromium
concentration was measured at PRRI on February 25, 2010 (0.114 ng/m3) and is the
only concentration greater than 0.1 ng/m3 measured at this site.
• The quarterly average concentrations of benzo(a)pyrene have relatively high levels of
variability associated with them. The first quarter average is higher than the other
quarterly averages, particularly the third quarter, but both the first and second quarter
average concentrations have relatively high confidence intervals associated with
them. The maximum benzo(a)pyrene concentration was measured at PRRI on
April 5, 2010 (1.07 ng/m3) and is the only concentration greater than 1 ng/m3
measured at this site. This measurement is also the eighth highest benzo(a)pyrene
concentration measured among NMP sites sampling PAH. Of the seven
concentrations of benzo(a)pyrene greater than 0.5 ng/m3, more than half were
measured during the first quarter of 2010.
• Eight of the nine non-detects of benzo(a)pyrene reported for PRRI were measured
between June and September 2010, the exception was one measured at the beginning
of October. The median concentration is 0.081 ng/m3, which is considerably less than
the annual average concentration, indicating that the higher measurements are likely
driving some of the quarterly and annual averages. The annual average
benzo(a)pyrene concentration for PRRI is the highest annual average benzo(a)pyrene
concentration among sites sampling PAH.
20-14
-------
20.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots were created for
benzo(a)pyrene, hexavalent chromium, and naphthalene for PRRI. Figures 20-6 through 20-8
overlay the site's minimum, annual average, and maximum concentrations onto the program-
level minimum, first quartile, average, median, third quartile, and maximum concentrations, as
described in Section 3.5.3.
Observations from Figures 20-6 through 20-8 include the following:
• Figure 20-6 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
PRRI is greater than the program-level average concentration. Although the
maximum concentration measured at PRRI is well below the maximum
concentration measured across the program, this site measured one of the few
concentrations of benzo(a)pyrene greater than 1 ng/m3. Several non-detects of
benzo(a)pyrene were measured at PRRI.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 20-7 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for the observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 20-7 shows the annual average concentration of hexavalent chromium for
PRRI is less than the program-level average and roughly equivalent to the
program-level median concentration. The maximum concentration measured at
PRRI is well below the program-level maximum concentration. There were no
non-detects of hexavalent chromium measured at PRRI.
• Figure 20-8 shows that the annual average naphthalene concentration for PRRI is
similar to the program-level average concentration. The maximum naphthalene
concentration measured at PRRI is well below the program-level maximum
concentration. There were no non-detects of naphthalene at PRRI.
20-15
-------
Figure 20-6. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
1
| Program Max Concentration = 42.7 ng/m3
0.4 0.6 0.8 1 1.2
Concentration (ng/m3)
1.4 1.6
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 20-7. Program vs. Site-Specific Average Hexavalent Chromium Concentration
Program Max Concentration = 3.51 ng/m3
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 20-8. Program vs. Site-Specific Average Naphthalene Concentration
600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
20-16
-------
20.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. PRRI has sampled hexavalent chromium under the NMP since 2005. Thus,
Figure 20-9 presents the 3-year rolling statistical metrics for hexavalent chromium for PRRI. The
statistical metrics presented for assessing trends include the substitution of zeros for non-detects.
Figure 20-9. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at PRRI
h H«'(«ittle — Minimum — Median
r Period
— Maximum
Observations from Figure 20-9 for hexavalent chromium measurements at PRRI include
the following:
• Sampling for hexavalent chromium at PRRI began in January 2005.
• The maximum hexavalent chromium concentration was measured on
August 28, 2007 (0.193 ng/m3), although a similar concentration was also measured
on July 4, 2006 (0.192 ng/m3).
• The rolling average concentrations are very similar in magnitude for 2005-2007 and
2006-2008, but exhibit a decrease for 2007-2009 and a very slight increase for 2008-
2010. Confidence intervals calculated for these averages show that the changes over
the period of sampling are not statistically significant. A similar trend is shown for
the median concentrations.
20-17
-------
• For each 3-year period shown, the minimum and 5th percentile are zero, indicating the
presence of non-detects. The number of non-detects reported has varied by year, from
as low as 18 percent in 2006 to as high as 65 percent in 2009.
20.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the PRRI
monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations regarding the
various risk factors, time frames, and calculations associated with these risk screenings.
20.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Rhode Island monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
were compared to the acute MRL; the quarterly averages were compared to the intermediate
MRL; and the annual averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for PRRI were greater than their respective MRL noncancer health risk benchmarks.
This is also true for pollutants not identified as pollutants of interest for the Rhode Island
monitoring site.
20.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Rhode Island monitoring site and where annual
average concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 20-6, where applicable.
20-18
-------
Table 20-6. Cancer and Noncancer Surrogate Risk Approximations for the Rhode Island
Monitoring Site
Pollutant
Cancer
URE
(Ug/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(ng/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Providence, Rhode Island - PRRI
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
0.00176
0.012
0.000034
0.0001
0.003
49/58
44/60
58/58
0.20
±0.07
0.02
±0.01
89.63
± 14.87
0.36
0.22
3.05
0.01
0.03
— = a Cancer URE or Noncancer RfC is not available.
Observations for PRRI from Table 20-6 include the following:
• The cancer surrogate risk approximation for naphthalene (3.05 in-a-million) is the
highest cancer surrogate risk approximation for PRRI's pollutants of interest. The
noncancer risk approximations for benzo(a)pyrene and hexavalent chromium are
considerably lower (0.36 and 0.22 in-a-million, respectively).
• None of PRRI's pollutants of interest have noncancer risk approximations greater
than 1.0. The pollutant with the highest noncancer risk approximation is naphthalene
(0.03).
20.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 20-7 and 20-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 20-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 20-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 20-7 and 20-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer surrogate risk approximations based on the site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 20.3, PRRI sampled for PAH and hexavalent chromium only. In addition, the cancer and
20-19
-------
noncancer surrogate risk approximations are limited to those pollutants with enough data to meet
the criteria for annual averages to be calculated. A more in-depth discussion of this analysis is
provided in Section 3.5.5.3.
Observations from Table 20-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in Providence County.
• Formaldehyde is also the pollutant with the highest toxicity-weighted emissions (of
the pollutants with cancer UREs), followed by benzene and POM, Group 3.
• Six of the highest emitted pollutants also have the highest toxicity-weighted
emissions for Providence County.
• Naphthalene, which has the highest cancer risk approximation among the pollutants
of interest for PRRI, has the seventh highest emissions and the fifth highest toxicity-
weighted emissions.
• POM, Group 2b is the tenth highest emitted "pollutant" in Providence County and
ranks sixth for toxicity-weighted emissions. POM, Group 2b includes several PAH
sampled for at PRRI including acenaphthylene, fluoranthene, fluorene, and perylene.
None of the PAH included in POM, Group 2b were identified as pollutants of interest
for PRRI.
• POM, Group 5a ranks eighth for toxicity-weighted emissions. POM, Group 5a
includes benzo(a)pyrene, one of PRRI's pollutants of interest. POM, Group 5a is not
among the highest emitted "pollutants" in Providence County.
• Hexavalent chromium, which is also one of PRRI's pollutants of interest, has the
seventh highest toxicity-weighted emissions for Providence County, but does not
appear on the list of highest emitted pollutants.
20-20
-------
Table 20-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer
UREs for the Rhode Island Monitoring Site
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Providence, Rhode Island (Providence County) - PRRI
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Tetrachloroethylene
Naphthalene
Dichloromethane
Trichloroethylene
POM, Group 2b
204.49
155.91
96.49
84.05
30.18
27.44
17.46
8.03
6.19
4.97
Formaldehyde
Benzene
POM, Group 3
1,3 -Butadiene
Naphthalene
POM, Group 2b
Hexavalent Chromium, PM
POM, Group 5a
Ethylbenzene
Arsenic, PM
2.03E-03
1.60E-03
1.18E-03
9.05E-04
5.93E-04
4.37E-04
3.47E-04
2.51E-04
2.41E-04
1.98E-04
Naphthalene
Benzo(a)pyrene
Hexavalent Chromium
3.05
0.36
0.22
to
o
to
-------
Table 20-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with
Noncancer RfCs for the Rhode Island Monitoring Site
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer Toxicity
Weight
Top 10 Noncancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Providence, Rhode Island (Providence County) - PRRI
Toluene
Xylenes
Methanol
Benzene
Formaldehyde
Hexane
Ethylbenzene
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
576.48
398.70
368.54
204.49
155.91
128.27
96.49
84.05
38.49
30.18
Acrolein
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Benzene
Naphthalene
Xylenes
Nickel, PM
Trichloroethylene
Arsenic, PM
335,397.31
15,908.87
15,090.50
9,339.31
6,816.26
5,818.50
3,986.98
3,977.28
3,096.76
3,072.81
Naphthalene 0.03
Hexavalent Chromium <0.01
to
o
to
to
-------
Observations from Table 20-8 include the following:
• Toluene, xylenes, and methanol are the highest emitted pollutants with noncancer
RfCs in Providence County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
noncancer RfCs) are acrolein, formaldehyde, and 1,3-butadiene.
• Five of the highest emitted pollutants in Providence County also have the highest
toxicity-weighted emissions.
• While naphthalene ranks sixth on the list of pollutants with the highest toxicity-
weighted emissions, it is not one of the highest emitted pollutants (with a noncancer
toxicity factor) in Providence County. Hexavalent chromium does not appear on the
list of highest emitted pollutants or the list of highest toxicity-weighted emissions for
Providence County.
20.6 Summary of the 2010 Monitoring Data for PRRI
Results from several of the data treatments described in this section include the
following:
»«» Naphthalene, hexavalent chromium, fluorene, and benzo(a)pyrene failed at least one
screen for PRRI.
*»* Of the site-specific pollutants of the interest, naphthalene had the highest annual
average concentration for PRRI. However, the annual average benzo(a)pyrene
concentration for PRRI is the highest annual average among NMP sites sampling this
pollutant.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
20-23
-------
21.0 Site in South Carolina
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in South Carolina, and integrates these
concentrations with emissions, meteorological, and risk information. Data generated by sources
other than ERG are not included in the data analyses contained in this report. Readers are
encouraged to refer to Sections 1 through 4 and the glossary (Appendix P) for detailed
discussions and definitions regarding the various data analyses presented below.
21.1 Site Characterization
This section characterizes the South Carolina monitoring site by providing geographical
and physical information about the location of the site and the surrounding area. This
information is provided to give the reader insight regarding factors that may influence the air
quality near the site and assist in the interpretation of the ambient monitoring measurements.
CHSC is located in central Chesterfield County, South Carolina. Figure 21-1 is a
composite satellite image retrieved from ArcGIS Explorer showing the monitoring site in its
rural location. Figure 21-2 identifies point source emissions locations by source category, as
reported in the 2008 NEI for point sources. Note that only sources within 10 miles of the site are
included in the facility counts provided in Figure 21-2. Thus, sources outside the 10-mile radius
have been grayed out, but are visible on the map to show emissions sources outside the 10-mile
boundary. A 10-mile boundary was chosen to give the reader an indication of which emissions
sources and emissions source categories could potentially have a direct effect on the air quality at
the monitoring site. Further, this boundary provides both the proximity of emissions sources to
the monitoring site as well as the quantity of such sources within a given distance of the site.
Table 21-1 describes the area surrounding the monitoring site by providing supplemental
geographical information such as land use, location setting, and locational coordinates.
21-1
-------
Figure 21-1. Chesterfield, South Carolina (CHSC) Monitoring Site
to
-------
Figure 21-2. NEI Point Sources Located Within 10 Miles of CHSC
Legend
eo"2ffo-w eg115'crw s>' itrtrw srs'trw
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
CHSC NATTS site 10 mile radius [ | County boundary
Source Category Group (No. of Facilities)
•f Aircraft Operations (1)
-$• Industrial Machinery and Equipment (1)
21-3
-------
Table 21-1. Geographical Information for the South Carolina Monitoring Site
Site
Code
CHSC
AQS Code
45-021-0001
Location
Not in a
city
County
Chesterfield
Micro- or
Metropolitan
Statistical Area
Not in an MSA
Latitude
and
Longitude
34.615367,
-80.198789
Land Use
Forest
Location
Setting
Rural
Additional Ambient Monitoring Information1
TSP, TSP Metals, VOC, O3, Meteorological
parameters, PM10, PM10 Speciation, PM25, and
PM2 5 Speciation, Carbonyl Compounds,
Hexachlorobutadiene.
BOLD ITALICS = EPA-designated NATTS Site.
to
-------
CHSC is located about 15 miles south of the North and South Carolina border, between
the towns of McBee and Chesterfield. The monitoring site is located near the Ruby fire tower
and, as Figure 21-1 shows, is located just off State Highway 145. The surrounding area is rural in
nature and is part of the Carolina Sandhills Wildlife Refuge. Figure 21-2 shows that few point
sources are located within 10 miles of CHSC, the closest of which is the Wild Irish Rose Airport.
Table 21-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the South
Carolina monitoring site. Table 21-2 also includes a vehicle registration-to-county population
ratio (vehicles-per-person). In addition, the population within 10 miles of the site is presented.
An estimate of 10-mile vehicle ownership was calculated by applying the county-level vehicle
registration-to-population ratio to the 10-mile population surrounding the monitoring site.
Table 21-2 also contains annual average daily traffic information. Finally, Table 21-2 presents
the daily VMT for Chesterfield County.
Table 21-2. Population, Motor Vehicle, and Traffic Information for the South Carolina
Monitoring Site
Site
CHSC
Estimated
County
Population1
46,665
County-level
Vehicle
Registration2
40,431
Vehicles per
Person
(Registration:
Population)
0.87
Population
within 10
miles3
5,605
Estimated
10-mile
Vehicle
Ownership
4,856
Annual
Average
Daily
Traffic4
550
County-
level Daily
VMT5
1,302,685
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2009 data from the South Carolina Dept of Public Safety (SC DPS, 2009)
3 10-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2010 data from the South Carolina DOT (SC DOT, 2011)
5 County-level VMT reflects 2010 data from the South Carolina DOT (SC DOT, 2012)
BOLD ITALICS = EPA-designated NATTS Site.
Observations from Table 21-2 include the following:
• Chesterfield County's population is among lowest compared to other counties with
NMP sites. This site's 10-mile population is the second lowest among NMP sites,
behind only CAMS 85 (in Texas). Similar rankings were found for both the county-
level and 10-mile vehicle ownerships.
• The vehicle-per-person ratio is in the middle of the range among NMP sites.
• The traffic volume experienced near CHSC ranks among the lowest compared to
other NMP monitoring sites. The traffic estimate used came from State Highway 145
between State Highway 109 and US-1.
21-5
-------
• The Chesterfield County daily VMT is the third lowest VMT compared to other
counties with NMP sites (where VMT data were available).
21.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in South Carolina on sample days, as well as over the course of the year.
21.2.1 Climate Summary
The town of Chesterfield is located just south of the North Carolina/South Carolina
border, about 35 miles northwest of the city of Florence. Although the area experiences all four
seasons, South Carolina's southeastern location ensures mild winters and long, hot summers.
Summers are dominated by the Bermuda high pressure system over the Atlantic, which allows
southwesterly winds to prevail, bringing in warm, moist air out of the Gulf of Mexico. During
winter, winds out of the southwest shift northeasterly after frontal systems move across the area.
Chesterfield County leads the state in average number of sleet and freezing rain events per year
(Bair, 1992 and SC SCO, 2012).
21.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest this site were retrieved
for 2010 (NCDC, 2010). The closest weather station with adequate data is located at the Monroe
Airport in Monroe, North Carolina (WBAN 53872). Additional information about the Monroe
Airport weather station, such as the distance between the site and the weather station, is provided
in Table 21-3. These data were used to determine how meteorological conditions on sample days
vary from normal conditions throughout the year.
Table 21-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 21-3 is the
95 percent confidence interval for each parameter. As shown in Table 21-3, average
meteorological conditions on sample days were representative of average weather conditions
throughout the year.
21-6
-------
Table 21-3. Average Meteorological Conditions near the South Carolina Monitoring Site
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Chesterfield, South Carolina - CHSC
Monroe Airport
53872
(35.02, -80.62)
35.81
miles
311°
(NW)
Sample
Day
2010
69.4
±4.7
70.9
+ 1.9
59.3
±4.6
60.3
+ 1.8
46.6
±5.0
47.6
+ 2.0
52.9
±4.3
53.8
+ 1.7
66.2
±3.1
66.5
+ 1.4
1016.8
± 1.6
1016.9
+ 0.6
4.8
±0.6
4.5
+ 0.2
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
to
-------
21.2.3 Back Trajectory Analysis
Figure 21-3 is the composite back trajectory map for days on which samples were
collected at the CHSC monitoring site in 2010. Included in Figure 21-3 are four back trajectories
per sample day. Figure 21-4 is the corresponding cluster analysis for 2010. An in-depth
description of these maps and how they were generated is presented in Section 3.5.2.1. For the
composite map, each line represents the 24-hour trajectory along which a parcel of air traveled
toward the monitoring site on a given sample day and time. For the cluster analysis, each line
corresponds to a back trajectory representative of a given cluster of trajectories. For both maps,
each concentric circle around the site in Figures 21-3 and 21-4 represents 100 miles.
Observations from Figures 21-3 and 21-4 for CHSC include the following:
• Back trajectories originated from a variety of directions at CHSC.
• The 24-hour air shed domain for CHSC was similar in size to other NMP monitoring
sites. The farthest away a trajectory originated was over Lake Michigan, or greater
than 600 miles away. However, the average trajectory length was 194 miles and most
trajectories (85 percent) originated within 300 miles of the site.
• The cluster analysis shows that nearly half of back trajectories originated from the
southeast to south to southwest of CHSC over South Carolina and Georgia and less
than 300 miles from the site, as represented by the cluster trajectory originating near
the South Carolina-Georgia border (41 percent). Another 30 percent originated from
the north to east to southeast of CHSC, over North Carolina and Virginia, and
generally less than 200 miles from the site. Thus, the majority of back of trajectories
originated within 300 miles of CHSC. Another 25 percent of trajectories originated
from the northwest and north of CHSC, although of varying lengths. The longest
trajectories originated from this direction. Another 4 percent originated from the west,
generally over Tennessee and northern Alabama and Georgia.
21-8
-------
Figure 21-3. 2010 Composite Back Trajectory Map for CHSC
Figure 21-4. Back Trajectory Cluster Map for CHSC
21-9
-------
21.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at Monroe Airport near CHSC were
uploaded into a wind rose software program to produce customized wind roses, as described in
Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals" positioned
around a 16-point compass, and uses different colors to represent wind speeds.
Figure 21-5 presents three different wind roses for the CHSC monitoring site. First, a
historical wind rose representing 2000 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at this location.
Observations from Figure 21-5 for CHSC include the following:
• The Monroe Airport weather station is located across the North Carolina/South
Carolina border, approximately 36 miles northwest of CHSC.
• The historical wind rose for CHSC shows that calm winds (< 2 knots) account for
22 percent of the hourly measurements. Winds from the south-southwest to west-
southwest account for just slightly more observations than winds from the north-
northeast to east-northeast. Winds from the southeast quadrant are generally not
observed.
• The wind patterns shown on the 2010 wind rose for CHSC are similar to the historical
wind patterns, although there were more calm observations and fewer winds
observations from the northeast quadrant. This indicates that wind conditions in 2010
were similar to what is expected climatologically near this site.
• The sample day wind patterns for 2010 also resemble the historical and full-year wind
roses in the calm rate and southwest wind prevalence. However, there were fewer
wind observations from the northeast quadrant, similar to the 2010 full-year wind
rose, and an increased number of observations from the northwest quadrant.
21-10
-------
Figure 21-5. Wind Roses for the Monroe Airport Weather Station near CHSC
2000-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between CHSC and NWS Station
21-11
-------
21.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the South Carolina monitoring
site in order to allow analysts and readers to focus on a subset of pollutants through the context
of risk. Each pollutant's preprocessed daily measurement was compared to its associated risk
screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by the monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 21-4 presents CHSC's pollutants of interest. The pollutants that failed at least one
screen and contributed to 95 percent of the total failed screens for the monitoring site are shaded.
NATTS MQO Core Analytes are bolded. Thus, pollutants of interest are shaded and/or bolded.
CHSC sampled hexavalent chromium and PAH.
Table 21-4. Risk Screening Results for the South Carolina Monitoring Site
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Chesterfield, South Carolina - CHSC
Naphthalene
0.029
Total
6
6
58
58
10.34
10.34
100.00
100.00
Observations from Table 21-4 include the following:
• Naphthalene was detected in all 58 valid samples collected at CHSC and failed six
screens, or approximately 10 percent of screens.
• This site has the second lowest number of failed screens (6) among all NMP sites.
• Benzo(a)pyrene and hexavalent chromium were added to CHSC's pollutants of
interest because they are NATTS MQO Core Analytes, even though they did not fail
any screens. These pollutants are not shown in Table 21-4.
21-12
-------
21.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the South Carolina monitoring site. Concentration averages are provided for the pollutants of
interest for the CHSC monitoring site, where applicable. Concentration averages for select
pollutants are also presented graphically for the site, where applicable, to illustrate how the site's
concentrations compare to the program-level averages. In addition, concentration averages for
select pollutants are presented from previous years of sampling in order to characterize
concentration trends at the site, where applicable. Additional site-specific statistical summaries
are provided in Appendices M and O.
21.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for the South Carolina site, as described in Section 3.1. The quarterly average of a particular
pollutant is simply the average concentration of the preprocessed daily measurements over a
given calendar quarter. Quarterly average concentrations include the substitution of zeros for all
non-detects. A site must have a minimum of 75 percent valid samples of the total number of
samples possible within a given quarter for a quarterly average to be calculated. An annual
average includes all measured detections and substituted zeros for non-detects for the entire year
of sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for CHSC are presented in
Table 21-5, where applicable. Note that if a pollutant was not detected in a given calendar
quarter, the quarterly average simply reflects "0" because only zeros substituted for non-detects
were factored into the quarterly average concentration.
21-13
-------
Table 21-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the South Carolina Monitoring Site
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(ng/m3)
2nd
Quarter
Average
(ng/m3)
3rd
Quarter
Average
(ng/m3)
4th
Quarter
Average
(ng/m3)
Annual
Average
(ng/m3)
Chesterfield, South Carolina - CHSC
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
12/58
27/62
58/58
0.04
±0.04
<0.01
±<0.01
19.27
±6.00
0
0.01
±<0.01
13.61
±2.86
0
0.01
±0.01
14.99
±2.10
0.02
±0.01
0.01
±<0.01
28.21
± 12.80
0.01
±0.01
0.01
±<0.01
19.16
±3.84
Observations for CHSC from Table 21-5 include the following:
• The annual average concentration of naphthalene is significantly higher than the
annual average concentrations of hexavalent chromium and benzo(a)pyrene.
Compared to other NMP sites, CHSC has some of the lowest annual average
concentrations of these three pollutants.
• Benzo(a)pyrene was not detected in the second or third quarters of 2010. This
pollutant was detected in less than half of the samples collected during the first and
fourth quarters of 2010.
• Although hexavalent chromium was detected in all four quarters of 2010, it was
detected in fewer than half of the samples collected (27 out of 62). The measurements
ranged from 0.0057 ng/m3 to 0.0407 ng/m3, with two-thirds of the concentrations
measured during the third (10) and fourth (8) quarters of 2010.
• Naphthalene was detected in every sampled collected at CHSC. The fourth quarter
average concentration is higher than the other quarterly averages, although the
difference is not statistically significant. This quarterly average also has a relatively
large confidence interval associated with it. The maximum naphthalene concentration
was measured on October 5, 2010 (106 ng/m3) and is nearly twice the next highest
measurement (54.2 ng/m3, measured on March 9, 2010). The concentrations
measured at CHSC ranged from 6.90 ng/m3 to 106 ng/m3, with a median
concentration of 15.6 ng/m3.
21.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for benzo(a)pyrene,
hexavalent chromium, and naphthalene were created for CHSC. Figures 21-6 through 21-8
overlay the site's minimum, annual average, and maximum concentrations onto the program-
21-14
-------
level minimum, first quartile, average, median, third quartile, and maximum concentrations, as
described in Section 3.5.3.
Figure 21-6. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
o
1
| Program Max Concentration
= 42.7 ng/m3
0.4 0.6
0.8 1 1.2
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 21-7. Program vs. Site-Specific Average Hexavalent Chromium Concentration
H
3 0.15
0.3 0.45
Concentration (ng/m3)
Program Max Concentration = 3.51 ng/m3
0.6 0.
Program: IstQuartile 2nd Quartile 3rd Quartile 4th Quartile Average
• • D D
Site: Site Average Site Minimum/Maximum
o —
Figure 21-8. Program vs. Site-Specific Average Naphthalene Concentration
a
600 800
Concentration (ng/m3)
Program: IstQuartile 2nd Quartile 3rd Quartile 4th Quartile Average
Site: Site Average Site Minimum/Maximum
O
21-15
-------
Observations from Figures 21-6 through 21-8 include the following:
• Figure 21-6 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
CHSC is well below the program-level average concentration and just greater
than the program median concentration. Figure 21-6 also shows that the
maximum concentration measured at CHSC is well below the maximum
concentration measured across the program. Several non-detects of
benzo(a)pyrene were measured at CHSC.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 21-7 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for the observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 21-7 shows the annual average concentration of hexavalent chromium for
CHSC is well below the program-level average concentration and the program-
level median concentration. The maximum concentration measured at CHSC is
not only well below the maximum concentration measured across the program, it
is just greater than the program-level average concentration of hexavalent
chromium. More than half of the measurements of hexavalent chromium for
CHSC were non-detects.
• Figure 21-8 shows that the annual naphthalene average for CHSC is less than the
program-level average concentration as well as the program-level median
concentration. The maximum naphthalene concentration measured at CHSC is
less than the program-level third quartile (75th percentile).
21.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. CHSC has sampled hexavalent chromium under the NMP since 2005. Thus,
Figure 21-9 presents the 3-year rolling statistical metrics for hexavalent chromium for CHSC.
The statistical metrics presented for assessing trends include the substitution of zeros for non-
detects.
21-16
-------
Figure 21-9. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at CHSC
0.14
0.12
"in
I
0.0?
0 -
T
T
[
A
mm BBBHI ". 1 j^^j ; BB^M i
200i 200/ 2DUti 2WOS 2UO/ 200!f 200W 2010
Ihree-Tea. Ptrind
• Slli Per cviilik - Minimum - Mrdijn - II.. i.niiin • 95lliPitu.-iil.ik- * -"
-------
21.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
South Carolina monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
were compared to the acute MRL; the quarterly averages were compared to the intermediate
MRL; and the annual averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for CHSC were greater than their respective MRL noncancer health risk benchmarks.
This is also true for pollutants not identified as pollutants of interest for the South Carolina
monitoring site.
21.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the South Carolina monitoring site and where annual
average concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 21-6, where applicable.
Table 21-6. Cancer and Noncancer Surrogate Risk Approximations for the South Carolina
Monitoring Site
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(ng/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Chesterfield, South Carolina - CHSC
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
0.00176
0.012
0.000034
0.0001
0.003
12/58
27/62
58/58
0.01
±0.01
0.01
± 0.01
19.16
±3.84
0.03
0.09
0.65
0.01
0.01
— = a Cancer URE or Noncancer RfC is not available.
21-18
-------
Observations for CHSC from Table 21-6 include the following:
• The cancer risk approximations for the pollutants of interest for CHSC are all less
than 1 in-a-million, with the highest cancer risk approximation calculated for
naphthalene (0.65 in-a-million).
• The noncancer risk approximations for the three pollutants of interest are very low
(0.01 or less). Because benzo(a)pyrene has no RfC, a noncancer risk approximation
could not be calculated.
21.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 21-7 and 21-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 21-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from annual averages.
Table 21-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from the annual averages.
The pollutants listed in Tables 21-7 and 21-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer risk surrogate approximations based on the site's annual
averages are limited to those pollutants for which the site sampled. As discussed in Section 21.3,
CHSC sampled for PAH and hexavalent chromium only. In addition, the cancer and noncancer
surrogate risk approximations are limited to those pollutants with enough data to meet the criteria
for annual averages to be calculated. A more in-depth discussion of this analysis is provided in
Section 3.5.5.3.
21-19
-------
Table 21-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the South Carolina Monitoring Site
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Chesterfield, South Carolina (Chesterfield County) - CHSC
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Naphthalene
Trichloroethylene
Dichloromethane
POM, Group 2b
Ethylene oxide
28.92
14.89
13.15
7.77
3.17
1.50
0.47
0.40
0.32
0.07
Benzene
Formaldehyde
1,3 -Butadiene
Hexavalent Chromium, PM
Naphthalene
Ethylbenzene
POM, Group 2b
POM, Group 3
POM, Group 5a
Acetaldehyde
2.26E-04
1.71E-04
9.52E-05
7.65E-05
5.11E-05
3.72E-05
2.81E-05
2.20E-05
1.92E-05
1.71E-05
Naphthalene
Hexavalent Chromium
Benzo(a)pyrene
0.65
0.09
0.03
to
o
-------
Table 21-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the South Carolina Monitoring Site
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Chesterfield, South Carolina (Chesterfield County) - CHSC
Toluene
Xylenes
Benzene
Methanol
Hexane
Ethylbenzene
Formaldehyde
Acetaldehyde
Methyl isobutyl ketone
Ethylene glycol
78.42
60.74
28.92
28.81
19.60
14.89
13.15
7.77
6.32
3.86
Acrolein
Cyanide Compounds, gas
1,3 -Butadiene
Formaldehyde
Benzene
Acetaldehyde
Xylenes
Naphthalene
Lead, PM
Manganese, PM
42,691.89
1,852.73
1,586.49
1,341.80
963.86
863.64
607.44
500.57
432.31
287.58
Naphthalene 0.01
Hexavalent Chromium O.01
to
to
-------
Observations from Table 21-7 include the following:
• Benzene, ethylbenzene, and formaldehyde are the highest emitted pollutants with
cancer UREs in Chesterfield County.
• Benzene, formaldehyde, and 1,3-butadiene are the pollutants with the highest
toxi city-weighted emissions (of the pollutants with cancer UREs) for Chesterfield
County.
• Seven of the highest emitted pollutants also have the highest toxi city-weighted
emissions for Chesterfield County.
• Naphthalene appears on all three lists, with the sixth highest emissions, the fifth
highest toxicity-weighted emissions, and highest cancer risk approximation among
CHSC's pollutants of interest.
• Hexavalent chromium ranks fourth for its toxicity-weighted emissions, but is not
among the highest emitted pollutants.
• Several POM Groups appear among the pollutants with the highest emissions and
toxicity-weighted emissions. POM, Group 2b includes several PAH sampled for at
CHSC including acenaphthylene, fluoranthene, and perylene. POM, Group 5a
includes benzo(a)pyrene, which is one of CHSC's pollutants of interest. POM, Group
3 does not include any pollutants sampled for at CHSC.
Observations from Table 21-8 include the following:
• Toluene, xylenes, and benzene are the highest emitted pollutants with noncancer
RfCs in Chesterfield County.
• The pollutants with the highest toxicity-weighted emissions (of the pollutants with
noncancer RfCs) are acrolein, cyanide compounds (gaseous), and 1,3-butadiene.
• Four of the highest emitted pollutants in Chesterfield County also have the highest
toxicity-weighted emissions.
• Naphthalene does not appear among the highest emitted pollutants, but ranks eighth
among the pollutants with the 10 highest toxicity-weighted emissions. Hexavalent
chromium does not appear on either emissions-based list.
21.6 Summary of the 2010 Monitoring Data for CHSC
Results from several of the data treatments described in this section include the
following:
»«» Naphthalene was the only pollutant to fail screens for CHSC.
21-22
-------
Of the site-specific pollutants of the interest, naphthalene had the highest annual
average concentration for CHSC; however, it was quite low compared to other NMP
sites sampling naphthalene.
None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
21-23
-------
22.0 Sites in South Dakota
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the UATMP sites in South Dakota, and integrates these
concentrations with emissions, meteorological, and risk information. Data generated by sources
other than ERG are not included in the data analyses contained in this report. Readers are
encouraged to refer to Sections 1 through 4 and the glossary (Appendix P) for detailed
discussions and definitions regarding the various data analyses presented below.
22.1 Site Characterization
This section characterizes the monitoring sites by providing geographical and physical
information about the location of the sites and the surrounding areas. This information is
provided to give the reader insight regarding factors that may influence the air quality near the
sites and assist in the interpretation of the ambient monitoring measurements.
There are two South Dakota monitoring sites. One monitoring site is located in Sioux
Falls, South Dakota (SSSD) while the other is located in Union County (UCSD). Figures 22-1
and 22-2 are composite satellite images retrieved from ArcGIS Explorer showing the monitoring
sites in their rural and urban locations. Figures 22-3 and 22-4 identify point source emissions
locations by source category, as reported in the 2008 NEI for point sources. Note that only
sources within 10 miles of the sites are included in the facility counts provided in Figures 22-3
and 22-4. Thus, sources outside the 10-mile radius have been grayed out, but are visible on the
maps to show emissions sources outside the 10-mile boundary. A 10-mile boundary was chosen
to give the reader an indication of which emissions sources and emissions source categories
could potentially have a direct effect on the air quality at the monitoring sites. Further, this
boundary provides both the proximity of emissions sources to the monitoring sites as well as the
quantity of such sources within a given distance of the sites. Table 22-1 describes the area
surrounding each monitoring site by providing supplemental geographical information such as
land use, location setting, and locational coordinates.
22-1
-------
Figure 22-1. Sioux Falls, South Dakota (SSSD) Monitoring Site
to
to
-------
Figure 22-2. Union County, South Dakota (UCSD) Monitoring Site
to
to
-------
Figure 22-3. NEI Point Sources Located Within 10 Miles of SSSD
67-otrw
Legend
S'dS'fTW S6:40'u"W 96-'35'G"W 96'30'CTW
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
SSSD UATMP site
10 mile radius
County boundary
Source Category Group (No. of Facilities)
41 Aircraft Operations (7)
•» Transportation Equipment (1)
22-4
-------
Figure 22-4. NEI Point Sources Located Within 10 Miles of UCSD
Legend
96-J5X1-W 96'4WW 96'35'0-W
Note: Due to facility density and collocation the total facilities
displayed may nol represent ali facilities within the area of interest.
UCSD UATM P site
10 mile radius
County boundary
Source Category Group (No. of Facilities)
-f1 Aircraft Operations (1)
22-5
-------
Table 22-1. Geographical Information for the South Dakota Monitoring Sites
Site
Code
SSSD
UCSD
AQS Code
46-099-0008
46-127-0001
Location
Sioux Falls
Not in a
City
County
Minnehaha
Union
Micro- or
Metropolitan
Statistical
Area
Sioux Falls, SD
MSA
Sioux City, IA-
NE-SD MSA
Latitude and
Longitude
43.54792,
-96.700769
42.751518,
-96.707208
Land Use
Commercial
Agricultural
Location
Setting
Urban/City
Center
Rural
Additional Ambient Monitoring Information1
SO2, NO, NO2, NOX, O3, Meteorological
parameters, PM10, PM2 5, and PM25 Speciation.
CO, SO2, NO, NO2, NOX, Meteorological
parameters, PM10, and PM25.
These monitoring sites report additional pollutants to AQS (EPA, 2012h); however, these data are not generated by ERG and are therefore not included in this report.
to
to
-------
SSSD is located on the east side of Sioux Falls, in eastern South Dakota. The monitoring
site is located at the South Dakota School for the Deaf. The surrounding area is mixed usage,
with both commercial and residential areas surrounding the site. SSSD is less than 1/2 mile from
the intersection of Highway 42 (East 10th Street) and 1-229, as shown in Figure 22-1. As
Figure 22-3 shows, few emissions sources are located within 10 miles of SSSD. There are only
two source categories shown in Figure 22-3, the aircraft operations category and the
transportation equipment category. The emissions source closest to SSSD is a hospital heliport.
UCSD is located in Union County, the southeastern-most county of the state, where the
South Dakota state border follows the Missouri River and comes to a point near Sioux City, Iowa
at the Nebraska and Iowa borders. The UCSD monitoring site is located in a rural and
agricultural area in the town of Brule, north of Elk Point and west of Vermillion. As shown in
Figure 22-2, the monitoring site is located on a residential property surrounded by agricultural
fields. Interstate-29 runs northwest-southeast through the center of Union County and lies less
than 1.5 miles west of UCSD. Figure 22-4 shows that there is a single point source located
within 10 miles of the site. However, UCSD is south of a proposed power plant and oil refinery.
The purpose of the monitoring at UCSD is to collect air quality data before, during, and after the
construction of the proposed power plant and oil refinery (SD DENR, 2010).
Table 22-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the areas surrounding the South
Dakota monitoring sites. Table 22-2 also includes a vehicle registration-to-county population
ratio (vehicles-per-person) for each site. In addition, the population within 10 miles of each site
is presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-
level vehicle registration-to-population ratio to the 10-mile population surrounding each
monitoring site. Table 22-2 also contains annual average daily traffic information. Finally,
Table 22-2 presents the daily VMT for Minnehaha and Union Counties.
22-7
-------
Table 22-2. Population, Motor Vehicle, and Traffic Information for the South Dakota
Monitoring Sites
Site
SSSD
UCSD
Estimated
County
Population1
169,987
14,501
County-level
Vehicle
Registration2
208,911
25,051
Vehicles per
Person
(Registration:
Population)
1.23
1.73
Population
within 10
miles3
190,685
6,153
Estimated
10-mile
Vehicle
Ownership
234,348
10,630
Annual
Average
Daily
Traffic4
21,340
156
County-
level Daily
VMT5
3,716,475
790,541
1 County-level population estimates reflect data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the South Dakota Department of Revenue (SD DOR,
2010)
3 10-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2010 data for SSSD and 2007 data for UCSD from the South Dakota DOT
(SD DOT, 2007 and 2011)
5 County-level VMT reflects 2010 data from the South Dakota DOT (SD DOT, 2011)
Observations from Table 22-2 include the following:
• Although SSSD's county-level population is significantly higher than the county-
level population for UCSD, both county-level populations are in the bottom third
compared to other counties with NMP sites, with UCSD ranking last. The 10-mile
populations for each site are also on the low side compared to other NMP sites,
particularly for UCSD.
• SSSD's county-level vehicle registration is an order of magnitude higher than
UCSD's, but both of the county-level vehicle registrations are on the low side
compared to other counties with NMP sites. Union County's vehicle registration is
the lowest of all NMP counties, while Minnehaha County is in the bottom third. The
10-mile vehicle ownership estimates for SSSD and UCSD rank slightly higher among
NMP sites than county-level vehicle ownerships.
• The vehicle-per-person ratios for these sites are among the highest, indicating that
residents likely own multiple vehicles. The ratio for UCSD is the second highest
among all NMP sites.
• The traffic volume for SSSD is two orders of magnitude higher than the traffic
volume for UCSD. The traffic near UCSD is the second lowest among all NMP sites,
behind only BRCO. Traffic data for SSSD were obtained for East 10th Avenue
between South Mable Avenue and South Highland Avenue; traffic data for UCSD
were obtained for 475th Avenue near 317th Street.
• The Union County VMT is the lowest among NMP sites (where VMT was available).
The Minnehaha County VMT ranks eighth lowest among counties with NMP sites
(where VMT was available).
22-8
-------
22.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in South Dakota on sample days, as well as over the course of the year.
22.2.1 Climate Summary
The Sioux Falls area has a continental climate, with cold winters, warm summers, and
often drastic day-to-day variations. Precipitation varies throughout the year, with the spring and
summer seasons receiving more than half of the annual rainfall. On average, a south wind blows
in the summer and fall and a northwest wind blows in the winter and spring. Flooding is often a
concern in the area during springtime when snow begins to melt, although a flood control
system, including levees and a diversion channel, was constructed to reduce the flood threat
within the city limits and to divert water from the Big Sioux River and Skunk Creek around the
city (Bair, 1992).
Sioux City is located just north of the Missouri River where the Iowa border meets the
Nebraska and South Dakota borders. The climate near Sioux City is generally continental in
nature, with warm summers and cold, relatively dry winters. Precipitation is concentrated in the
spring and summer months. Wind direction varies with season, with southeasterly to southerly
winds in the spring and summer, and northwesterly winds in the autumn and winter (Bair, 1992).
22.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather stations nearest these sites were
retrieved for 2010 (NCDC, 2010). The closest weather stations are located at Joe Foss Field
Airport (near SSSD) and Sioux Gateway Airport (near UCSD), WBAN 14944 and 14943,
respectively. Additional information about these weather stations, such as the distance between
the sites and the weather stations, is provided in Table 22-3. These data were used to determine
how meteorological conditions on sample days vary from normal conditions throughout the year.
22-9
-------
Table 22-3. Average Meteorological Conditions near the South Dakota Monitoring Sites
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Sioux Falls, South Dakota - SSSD
Joe Foss Field
Airport
14944
(43.58, -96.75)
3.20
miles
309°
(NW)
Sample
Day
2010
55.4
±6.4
55.5
+ 2.5
46.3
±6.0
46.4
+ 2.4
37.6
±5.6
37.2
+ 2.2
42.1
±5.5
41.9
+ 2.2
74.5
±3.0
73.4
+ 1.2
1015.6
±2.1
1015.6
+ 0.9
7.7
±0.9
7.9
+ 0.4
Union County, South Dakota - UCSD
Sioux
Gateway/Col.
Bud Day Field
Airport
14943
(42.39, -96.38)
29.45
miles
148°
(SSE)
Sample
Day
2010
60.0
±6.1
59.3
+ 2.6
50.4
±5.6
49.0
+ 2.4
40.2
±5.2
38.5
+ 2.2
45.3
±5.1
43.8
+ 2.2
71.5
±3.2
70.5
+ 1.2
1015.1
±2.1
1015.7
+ 0.9
7.9
±0.9
7.9
+ 0.4
to
to
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
Table 22-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 22-3 is the
95 percent confidence interval for each parameter. As shown in Table 22-3, average
meteorological conditions on sample days were representative of average weather conditions
throughout the year for both sites.
22.2.3 Back Trajectory Analysis
Figure 22-5 is the composite back trajectory map for days on which samples were
collected at the SSSD monitoring site in 2010. Included in Figure 22-5 are four back trajectories
per sample day. Figure 22-6 is the corresponding cluster analysis for 2010. Similarly,
Figure 22-7 is the composite back trajectory map for days on which samples were collected at
UCSD and Figure 22-8 is the corresponding cluster analysis. An in-depth description of these
maps and how they were generated is presented in Section 3.5.2.1. For the composite maps, each
line represents the 24-hour trajectory along which a parcel of air traveled toward the monitoring
site on a given sample day and time. For the cluster analyses, each line corresponds to a back
trajectory representative of a given cluster of trajectories. For all maps, each concentric circle
around the sites in Figures 22-5 through 22-8 represents 100 miles.
Observations from Figures 22-5 and 22-6 for SSSD include the following:
• Back trajectories originated from a variety of directions at the SSSD site, although
primarily from the northwest and south.
• The 24-hour air shed domain for SSSD is among the larger air sheds compared to the
other NMP monitoring sites. The farthest away a trajectory originated was southern
Alberta, Canada, or greater than 800 miles away. However, the average back
trajectory length was nearly 280 miles and 88 percent of back trajectories originated
within 500 miles of the site.
• The cluster analysis shows that back trajectories originating from the west to
northwest to north accounted for 50 percent of the back trajectories for SSSD,
although of varying lengths. The shorter cluster (36 percent) originating to the
southeast of SSSD represents trajectories originating from a variety of directions but
within 100-300 miles or so of the site. Another 14 percent of trajectories originated
from the south of SSSD.
22-11
-------
Figure 22-5. 2010 Composite Back Trajectory Map for SSSD
Figure 22-6. Back Trajectory Cluster Map for SSSD
22-12
-------
Figure 22-7. 2010 Composite Back Trajectory Map for UCSD
Figure 22-8. Back Trajectory Cluster Map for UCSD
22-13
-------
Observations from Figures 22-7 and 22-8 for UCSD include the following:
• Back trajectories originated from a variety of directions at the UCSD monitoring site.
The composite map for UCSD shares similarities in the trajectory distribution for
SSSD.
• The 24-hour air shed domain for UCSD was similar in size compared to SSSD. Two
back trajectories originated approximately 800 miles away, one in north-central
Montana and one in central Saskatchewan, Canada. However, the average trajectory
length was nearly 280 miles and 90 percent of the trajectories originated within
500 miles of the site.
• The cluster analysis for UCSD shows that 22 percent of trajectories originated from
the west, northwest, and north-northwest of the site. Another cluster (33 percent)
represents shorter trajectories originating primarily to the north over southeast North
Dakota and much of South Dakota. Back trajectories originating from the south of
UCSD accounted for 36 percent of the back trajectories, although of varying lengths.
Another 10 percent of trajectories originated to east over Iowa.
22.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather stations at Joe Foss Field (for SSSD) and Sioux
Gateway (for UCSD) Airports were uploaded into a wind rose software program to produce
customized wind roses, as described in Section 3.5.2.2. A wind rose shows the frequency of wind
directions using "petals" positioned around a 16-point compass, and uses different colors to
represent wind speeds.
Figure 22-9 presents three different wind roses for the SSSD monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS stations and the
monitoring sites is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figure 22-10 presents the
three wind roses and distance map for the UCSD monitoring site.
22-14
-------
Figure 22-9. Wind Roses for the Joe Foss Field Airport Weather Station near SSSD
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between SSSD and NWS Station
i ,.i-"-T!: ?H
i ii'iHd*;
i • j *™" "»" "t**^
f!r^ifft.!! fir\{,;;
rji— .»,___, •;=:; |
"*V«.» .)-.,;
-..:,•"• '""" i ! I.
si 1 !• |== j |' *j
• IMAM « F "-.«' *
22-15
-------
Figure 22-10. Wind Roses for the Sioux Gateway Airport Weather Station near UCSD
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between UCSD and NWS Station
22-16
-------
Observations from Figure 22-9 for SSSD include the following:
• The Joe Foss Field Airport weather station is located approximately 3.2 miles
northwest of SSSD.
• The historical wind rose shows that winds from a variety of directions were observed
near SSSD, although winds from the south were observed the most (13 percent), and
southwesterly and west-southwesterly winds observed the least (less than 3 percent).
Calm winds were observed for approximately 11 percent of the observations. The
strongest winds tend to be from the south or the northwest quadrant.
• The 2010 wind patterns are very similar to the historical wind patterns, although a
slightly higher percentage of calm winds (14 percent) was observed during 2010.
• The sample day wind rose resembles the historical and full-year wind roses, but do
exhibit some differences. The sample day wind rose has a higher percentage of
northwesterly winds. In addition, the strongest winds, those greater than 22 knots,
were not captured on the sample day wind rose.
Observations from Figure 22-10 for UCSD include the following:
• The Sioux Gateway Airport weather station is located approximately 29 miles
south-southeast of UCSD, across the border in Sioux City, Iowa. The weather station
is located less than a mile from the Missouri River.
• The historical wind rose shows that winds from the southeast and northwest quadrants
were observed the most near UCSD. Calm winds were observed for less than eight
percent of the observations. The strongest winds tend to be from the south or the
northwest quadrant.
• The 2010 wind patterns are similar to the historical wind patterns, although the calm
rate is higher for 2010 (11 percent).
• The sample day wind patterns resemble the full-year wind patterns, but have a higher
percentage of northwesterly and north-northwesterly wind observations and less
northerly wind observations.
22.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the South Dakota monitoring
sites in order to allow analysts and readers to focus on a subset of pollutants through the context
of risk. For each site, each pollutant's preprocessed daily measurement was compared to its
associated risk screening value. If the concentration was greater than the risk screening value,
then the concentration "failed the screen." Pollutants of interest are those for which the
individual pollutant's total failed screens contribute to the top 95 percent of the site's total failed
screens. In addition, if any of the NATTS MQO Core Analytes measured by each monitoring site
22-17
-------
did not meet the pollutant of interest criteria based on the preliminary risk screening, that
pollutant was added to the list of site-specific pollutants of interest. A more in-depth description
of the risk screening process is presented in Section 3.2.
Table 22-4 presents the pollutants of interest for the South Dakota monitoring sites. The
pollutants that failed at least one screen and contributed to 95 percent of the total failed screens
for each monitoring site are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of
interest are shaded and/or bolded. SSSD and UCSD sampled for VOC, SNMOC, and carbonyl
compounds.
Observations from Table 22-4 include the following:
• Thirteen pollutants failed at least one screen for SSSD; of these, six are NATTS
MQO Core Analytes. Fifteen pollutants failed screens for UCSD, of which six are
also NATTS MQO Core Analytes. Of the pollutants failing screens, the sites share 11
pollutants in common in Table 22-4.
• For SSSD, six pollutants (of which five are NATTS MQO Core Analytes) were
identified as pollutants of interest by the risk screening process. Trichloroethylene
was added to SSSD's pollutants of interest because it's a NATTS MQO Core
Analyte, even though it did not contribute to 95 percent of the total failed screens.
Chloroform, tetrachloroethylene, and vinyl chloride were added to SSSD's pollutants
of interest because they are NATTS MQO Core Analytes, even though they did not
fail any screens. These three pollutants are not shown in Table 22-4.
• For UCSD, nine pollutants (of which six are NATTS MQO Core Analytes) were
identified as pollutants of interest by the risk screening process. Chloroform,
tetrachloroethylene, and vinyl chloride were added to UCSD's pollutants of interest
because they are NATTS MQO Core Analytes, even though they did not fail any
screens. These three pollutants are not shown in Table 22-4.
• Formaldehyde, benzene, and acetaldehyde were detected in every sample collected at
UCSD and SSSD and failed 100 percent of screens. Other pollutants, such as
acrylonitrile, 1,2-dichloroethane, and 1,2-dibromoethane also failed 100 percent
screens for each site but were detected infrequently.
• Recall from Section 3.2 that if a pollutant was measured by both the TO-15 and
SNMOC methods at the same site, the TO-15 results were used for the risk screening
process. As the South Dakota sites sampled both VOC (TO-15) and SNMOC, the
TO-15 results were used for the 12 pollutants these methods have in common.
22-18
-------
Table 22-4. Risk Screening Results for the South Dakota Monitoring Sites
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Sioux Falls, South Dakota - SSSD
Acetaldehyde
Formaldehyde
Benzene
Carbon Tetrachloride
1,3-Butadiene
1 ,2-Dichloroethane
1 ,2-Dibromoethane
p-Dichlorobenzene
Ethylbenzene
Acrylonitrile
Chloromethylbenzene
Hexachloro- 1 ,3 -butadiene
Trichloroethylene
0.45
0.077
0.13
0.17
0.03
0.038
0.0017
0.091
0.4
0.015
0.02
0.045
0.2
Total
61
61
60
58
42
9
4
2
2
1
1
1
1
303
61
61
60
60
47
9
4
17
60
1
1
2
5
388
100.00
100.00
100.00
96.67
89.36
100.00
100.00
11.76
3.33
100.00
100.00
50.00
20.00
78.09
20.13
20.13
19.80
19.14
13.86
2.97
1.32
0.66
0.66
0.33
0.33
0.33
0.33
20.13
40.26
60.07
79.21
93.07
96.04
97.36
98.02
98.68
99.01
99.34
99.67
100.00
Union County, South Dakota - UCSD
Benzene
Acetaldehyde
Formaldehyde
Carbon Tetrachloride
Acrylonitrile
1 ,2-Dichloroethane
Ethylbenzene
1,3-Butadiene
Trichloroethylene
1 ,2-Dibromoethane
£>-Dichlorobenzene
Hexachloro- 1 , 3 -butadiene
Propionaldehyde
1 , 1 ,2,2-Tetrachloroethane
1 , 1 ,2-Trichloroethane
0.13
0.45
0.077
0.17
0.015
0.038
0.4
0.03
0.2
0.0017
0.091
0.045
0.8
0.017
0.0625
Total
59
58
58
56
14
11
8
4
3
2
2
2
2
2
2
283
59
58
58
57
14
11
59
14
15
2
9
2
58
2
3
421
100.00
100.00
100.00
98.25
100.00
100.00
13.56
28.57
20.00
100.00
22.22
100.00
3.45
100.00
66.67
67.22
20.85
20.49
20.49
19.79
4.95
3.89
2.83
1.41
1.06
0.71
0.71
0.71
0.71
0.71
0.71
20.85
41.34
61.84
81.63
86.57
90.46
93.29
94.70
95.76
96.47
97.17
97.88
98.59
99.29
100.00
22.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the South Dakota monitoring sites. Concentration averages are provided for the pollutants of
interest for each site, where applicable. Concentration averages for select pollutants are also
presented graphically for each site, where applicable, to illustrate how each site's concentrations
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at the
22-19
-------
sites, where applicable. Additional site-specific statistical summaries are provided in Appendices
J through L.
22.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each South Dakota site, as described in Section 3.1. The quarterly average of a particular
pollutant is simply the average concentration of the preprocessed daily measurements over a
given calendar quarter. Quarterly average concentrations include the substitution of zeros for all
non-detects. A site must have a minimum of 75 percent valid samples of the total number of
samples possible within a given quarter for a quarterly average to be calculated. An annual
average includes all measured detections and substituted zeros for non-detects for the entire year
of sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the South Dakota
monitoring sites are presented in Table 22-5, where applicable. Note that if a pollutant was not
detected in a given calendar quarter, the quarterly average simply reflects "0" because only zeros
substituted for non-detects were factored into the quarterly average concentration.
Table 22-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the South Dakota Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Hg/m3)
2nd
Quarter
Average
(Hg/m3)
3rd
Quarter
Average
(Hg/m3)
4th
Quarter
Average
(Hg/m3)
Annual
Average
(Hg/m3)
Sioux Falls, South Dakota - SSSD
Acetaldehyde
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
61/61
60/60
47/60
60/60
43/60
9/60
61/61
47/60
2.21
±0.43
0.67
±0.08
0.03
±0.02
0.48
±0.12
0.05
±0.02
0.03
±0.02
2.20
±0.29
0.06
±0.03
2.14
±0.87
0.74
±0.12
0.04
±0.01
0.60
±0.10
0.09
±0.01
0.01
±0.01
2.45
±0.54
0.09
±0.02
1.47
±0.18
0.82
±0.21
0.04
±0.02
0.66
±0.07
0.09
±0.03
0
2.02
±0.31
0.11
±0.05
1.73
±0.36
0.74
±0.18
0.04
±0.02
0.59
±0.10
0.04
±0.03
0
2.06
±0.36
0.10
±0.06
1.88
±0.26
0.74
±0.07
0.04
±0.01
0.58
±0.05
0.07
±0.01
0.01
±0.01
2.18
±0.19
0.09
±0.02
22-20
-------
Table 22-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the South Dakota Monitoring Sites (Continued)
Pollutant
Trichloroethylene
Vinyl Chloride
#of
Measured
Detections
vs. # of
Samples
5/60
2/60
1st
Quarter
Average
(Ug/m3)
0.07
±0.14
0
2nd
Quarter
Average
(Ug/m3)
0
0
3rd
Quarter
Average
(Ug/m3)
0.01
±0.01
<0.01
±<0.01
4th
Quarter
Average
(Ug/m3)
0.01
±0.02
0
Annual
Average
(Ug/m3)
0.02
±0.03
<0.01
±<0.01
Union County, South Dakota - UCSD
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Ethylbenzene
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
58/58
14/59
59/59
14/59
57/59
36/59
11/59
59/59
58/58
39/59
15/59
2/59
2.28
±0.89
0.03
±0.03
0.49
±0.06
<0.01
±<0.01
0.49
±0.14
0.04
±0.02
0.04
±0.02
0.11
±0.04
2.61
±1.29
0.03
±0.02
0.01
±0.02
0
2.12
±0.80
0.01
±0.01
0.53
±0.19
0.01
±0.01
0.53
±0.12
0.08
±0.01
0.02
±0.02
0.11
±0.02
3.64
±1.01
0.04
±0.01
0.01
±0.01
0
1.65
±0.25
0.10
±0.09
0.57
±0.17
0.01
±0.01
0.65
±0.07
0.06
±0.02
0
2.15
±3.76
2.01
±0.43
0.08
±0.04
0.21
±0.35
<0.01
±<0.01
1.58
±0.37
0.06
±0.06
0.44
±0.05
<0.01
±0.01
0.53
±0.12
0.02
±0.02
0
0.31
±0.29
1.45
±0.33
0.03
±0.02
0.04
±0.06
0
1.88
±0.28
0.05
±0.03
0.51
±0.07
0.01
±<0.01
0.56
±0.05
0.05
±0.01
0.01
±0.01
0.75
±1.05
2.38
±0.42
0.05
±0.01
0.08
±0.10
<0.01
±<0.01
Observations for SSSD from Table 22-5 include the following:
• The pollutants with the highest annual average concentrations by mass are
formaldehyde (2.18 ± 0.19 |ig/m3), acetaldehyde (1.88 ± 0.26 |ig/m3), and benzene
(0.74 ± 0.07 |ig/m3).
• Acetaldehyde concentrations appear highest during the first half of the year. The
maximum acetaldehyde concentration was measured on May 26, 2009 (7.13 |ig/m3).
This concentration is nearly twice the next highest concentration (4.03 |ig/m3
measured on April 20, 2010), although similar concentrations were also measured in
February, March, and October. Note that of the 17 acetaldehyde concentrations
greater than 2 |ig/m3 measured at SSSD, eight were measured during the first quarter
of the year, four in the second quarter, one in the third quarter, and four in the fourth
quarter (all in October).
22-21
-------
• For 1,2-dichloroethane, all of the measured detections were measured during the first
and second quarters of 2010. In fact, five of the nine measured detections were
measured in February, and no measured detections were measured after May 8, 2010.
• The first quarter average concentration of trichloroethylene is significantly higher
than the other quarterly averages and the confidence interval is high, indicating that
this average is influenced by outliers. The highest concentration of trichloroethylene
was measured on January 8, 2010 (1.02 |ig/m3) and is an order of magnitude higher
than the next highest concentration (0.118 |ig/m3 measured on November 22, 2010).
The January 8th concentration is the only measured detection of trichloroethylene
measured during the first quarter of 2010. There were only five measured detections
of this pollutant for SSSD, with the other four being measured in the third and fourth
quarters (three in the third and one in the fourth).
Observations for UCSD from Table 22-5 include the following:
• The pollutants with the highest annual average concentrations by mass are
formaldehyde (2.38 ± 0.42 |ig/m3), acetaldehyde (1.88 ± 0.28 |ig/m3), and
ethylbenzene (0.75 ± 1.05 |ig/m3).
• For formaldehyde, the second quarter 2010 average is relatively high compared to
other quarterly averages and has a relatively high large confidence interval associated
with it, as does the first quarter average concentration. A review of the data shows
that the highest concentration of formaldehyde was measured on May 8, 2010
(9.14 |ig/m3). A similar concentration was also measured on March 15, 2010
(8.76 |ig/m3). The five highest concentrations of formaldehyde (those greater than
4 |ig/m3) were all measured between March and May 2010. Of the 12 formaldehyde
concentrations greater than 3 |ig/m3, two were measured during the first quarter of
2010, seven were measured during the second quarter, and three during the third
quarter (and none in the fourth quarter).
• A similar quarterly trend is shown for acetaldehyde measurements in Table 22-5,
although the trend is less pronounced. The two highest concentrations of acetaldehyde
were also measured on May 8th and March 15th.
• The third quarter ethylbenzene concentration is significantly higher than the other
quarterly averages and has a high confidence interval associated with it, indicating the
presence of outliers. The maximum concentration of ethylbenzene was measured at
UCSD on July 7, 2010 (31.5 |ig/m3) and was 15 times higher than the next highest
concentration (2.08 |ig/m3 measured on October 5, 2010). The July 7, 2010
concentration is the maximum ethylbenzene concentration measured among NMP
sites sampling this pollutant (SNMOC or TO-15). Only three ethylbenzene
concentrations were greater than 1 |ig/m3 at UCSD, with the third being measured in
December.
• The three highest concentrations of trichloroethylene were measured on the same
dates as the three highest concentrations of ethylbenzene, which explains the third
22-22
-------
quarter average concentration of this pollutant for UCSD, although this pollutant was
detected less frequently than ethylbenzene.
• Concentrations of tetrachloroethylene also appear higher during the third quarter of
2010. The maximum concentration of tetrachloroethylene was measured at UCSD on
July 2, 2010 (0.353 |ig/m3) and was twice the next highest concentration
(0.143 |ig/m3 measured on September 5, 2010). Eight of the 10 highest
tetrachloroethylene concentrations were measured at UCSD in July, August, and
September.
• Concentrations of chloroform appear higher during the warmer months of 2010. The
three highest concentrations of chloroform (those greater than 0.1 |ig/m3) were
measured in June, July, and August. The bulk of the measured detections (36) were
measured during the second (13) and third (11) quarters of 2010.
• Concentrations of acrylonitrile appear higher during the third and fourth quarters of
2010 and these quarterly averages have relatively high confidence intervals associated
with them. The four highest concentrations of acrylonitrile (those greater than 0.3
|ig/m3) were measured in August and September. Of the 14 measured detections of
acrylonitrile, four were measured during the first quarter of 2010, one was measured
during the second quarter, five during the third quarter, and four during the fourth
quarter.
• Similar to SSSD, all of the measured detections of 1,2-dichloroethane were measured
during the first and second quarters of 2010. Over half of the 11 measured detections
were measured in the period between January 26, 2010 through February 25, 2010,
and no measured detections were measured after June 19, 2010.
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for SSSD and
UCSD from those tables include the following:
• None of the annual average concentrations of the pollutants of interest for SSSD
appear in Tables 4-9 through 4-12.
• UCSD has the third highest concentration of trichloroethylene among NMP sites
sampling VOC, as shown in Table 4-9. UCSD has the fourth highest concentrations
of acrylonitrile and ethylbenzene among NMP sites sampling these pollutants.
22.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for acetaldehyde, benzene,
1,3-butadiene, and formaldehyde were created for both SSSD and UCSD. Figures 22-11 through
22-23
-------
22-14 overlay the sites' minimum, annual average, and maximum concentrations onto the
program-level minimum, first quartile, average, median, third quartile, and maximum
concentrations, as described in Section 3.5.3.
Figure 22-11. Program vs. Site-Specific Average Acetaldehyde Concentration
Concentration (ng/m3)
Program: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 22-12. Program vs. Site-Specific Average Benzene Concentration
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
22-24
-------
Figure 22-13. Program vs. Site-Specific Average 1,3-Butadiene Concentration
•o 1 ' ' '
III
0__
^^\ \ \ \
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
• • D D
Site: SiteAverage Site Minimum/Maximum
o —
Figure 22-14. Program vs. Site-Specific Average Formaldehyde Concentration
E
I
r
0
5 10 15 20
| |
25 30 35 40 45 50 55
Concentration (|ig/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
• n n n
Site: SiteAverage Site Minimum/Maximum
o —
Observations from Figures 22-11 through 22-14 include the following:
• Figure 22-11 shows that the annual average acetaldehyde concentrations for
SSSD and UCSD are nearly identical to each other and both are just slightly less
than the program-level average concentration. The range of concentrations
measured is slightly higher at SSSD than at UCSD, although the maximum
concentration measured at both sites is below the maximum concentration
measured across the program. There were no non-detects of acetaldehyde
measured at either site.
22-25
-------
• Figure 22-12 shows that the annual average benzene concentrations for both sites
are below both the program-level average and median concentrations of benzene.
Further, the annual average for UCSD is less than the program-level first quartile
(25th percentile). UCSD has the third lowest benzene concentration among sites
sampling this pollutant. There were no non-detects of benzene measured at either
site.
• Figure 22-13 shows that the annual average 1,3-butadiene concentrations for both
sites are below both the program-level average and median concentrations of
1,3-butadiene. Further, the annual average for UCSD is less than the program-
level first quartile (25th percentile). UCSD had the second lowest 1,3-butadiene
concentration among sites sampling this pollutant. More than 75 percent of the
measurements of 1,3-butadiene at UCSD were non-detects.
• Figure 22-14 shows that the annual average formaldehyde concentrations are both
just slightly less than the program-level average concentration of formaldehyde.
The range of concentrations measured at UCSD is twice than at SSSD, although
the maximum concentration measured at both sites is well below the maximum
concentration measured across the program. There were no non-detects of
formaldehyde measured at either site.
22.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. Sampling at SSSD began in 2008 and UCSD in 2009; thus, a trends analysis was
not conducted for these sites.
22.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the South
Dakota monitoring sites. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
22.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
South Dakota monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
22-26
-------
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
for each site were compared to the acute MRL; the quarterly averages were compared to the
intermediate MRL; and the annual averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the South Dakota monitoring sites were greater than their respective MRL
noncancer health risk benchmarks. This is also true for pollutants not identified as pollutants of
interest for the South Dakota monitoring sites.
22.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the South Dakota monitoring sites and where annual
average concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 22-6, where applicable.
Observations from Table 22-6 for SSSD include the following:
• The pollutants with the highest annual average concentrations for SSSD are
formaldehyde, acetaldehyde, and benzene.
• These same pollutants also have the highest cancer risk approximations among this
site's pollutants of interest, although formaldehyde's cancer risk approximation is an
order of magnitude higher than the cancer risk approximations for the other
pollutants.
• None of the noncancer surrogate risk approximations were greater than an HQ of 1.0.
22-27
-------
Table 22-6. Cancer and Noncancer Surrogate Risk Approximations for the South Dakota
Monitoring Sites
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Sioux Falls, South Dakota - SSSD
Acetaldehyde
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.0000022
0.0000078
0.00003
0.000006
0.000026
0.000013
2.6E-07
0.0000048
0.0000088
0.009
0.03
0.002
0.1
0.098
2.4
0.0098
0.04
0.002
0.1
61/61
60/60
47/60
60/60
43/60
9/60
61/61
47/60
5/60
2/60
1.88
±0.26
0.74
±0.07
0.04
±0.01
0.58
±0.05
0.07
±0.01
0.01
±0.01
2.18
±0.19
0.09
±0.02
0.02
±0.03
0.01
±0.01
4.13
5.78
1.19
3.50
0.27
28.38
0.02
0.10
0.01
0.21
0.02
0.02
0.01
O.01
0.01
0.22
0.01
0.01
0.01
Union County, South Dakota - UCSD
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Ethylbenzene
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.0000022
0.000068
0.0000078
0.00003
0.000006
0.000026
0.0000025
0.000013
2.6E-07
0.0000048
0.0000088
0.009
0.002
0.03
0.002
0.1
0.098
2.4
1
0.0098
0.04
0.002
0.1
58/58
14/59
59/59
14/59
57/59
36/59
11/59
59/59
58/58
39/59
15/59
2/59
1.88
±0.28
0.05
±0.03
0.51
±0.07
0.01
±0.01
0.56
±0.05
0.05
±0.01
0.01
±0.01
0.75
±1.05
2.38
±0.42
0.05
±0.01
0.08
±0.10
0.01
±0.01
4.13
3.62
4.00
0.16
3.34
0.35
1.87
30.97
0.01
0.36
0.01
0.21
0.03
0.02
0.01
0.01
0.01
O.01
0.01
0.24
0.01
0.04
0.01
— = a Cancer URE or Noncancer RfC is not available.
22-28
-------
Observations from Table 22-6 for UCSD include the following:
• The pollutants with the highest annual average concentrations for UCSD are
formaldehyde, acetaldehyde, and ethylbenzene.
• Formaldehyde has the highest cancer risk approximation for UCSD, followed by
acetaldehyde and benzene. The fourth highest cancer risk approximation was
calculated for acrylonitrile, which has a much lower annual average concentration
than the other aforementioned pollutants, indicating the relative toxicity of this
pollutant.
• None of the noncancer surrogate risk approximations for UCSD's pollutants of
interest were greater than an HQ of 1.0.
22.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 22-7 and 22-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 22-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer risk approximations (in-a-million), as calculated from the annual averages. Table 22-8
presents similar information, but identifies the 10 pollutants with the highest noncancer risk
approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 22-7 and 22-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer surrogate risk approximations based on each site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 22.3, SSSD and UCSD sampled for VOC, SNMOC, and carbonyl compounds. In
addition, the cancer and noncancer risk approximations are limited to those pollutants with
enough data to meet the criteria for annual averages to be calculated. A more in-depth discussion
of this analysis is provided in Section 3.5.5.3.
22-29
-------
Table 22-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer
UREs for the South Dakota Monitoring Sites
Top 10 Total Emissions for Pollutants
with Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Sioux Falls, South Dakota (Minnehaha County) - SSSD
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
POM, Group 6
POM, Group la
84.66
56.95
54.47
36.84
12.38
6.21
1.65
1.34
0.12
0.12
Formaldehyde
Benzene
1,3 -Butadiene
Naphthalene
POM, Group 3
Hexavalent Chromium, PM
Ethylbenzene
POM, Group 2b
Acetaldehyde
POM, Group 5a
7.40E-04
6.60E-04
3.72E-04
2.11E-04
1.94E-04
1.48E-04
1.36E-04
1.18E-04
8.10E-05
5.09E-05
Formaldehyde
Benzene
Acetaldehyde
Carbon Tetrachloride
1,3 -Butadiene
1 ,2-Dichloroethane
Trichloroethylene
Tetrachloroethylene
Vinyl Chloride
28.38
5.78
4.13
3.50
1.19
0.27
0.10
0.02
0.01
Union County, South Dakota (Union County) - UCSD
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
POM, Group 2b
Dichloromethane
POM, Group 6
POM, Group la
15.98
12.81
10.98
8.37
2.12
1.15
0.23
0.13
0.02
0.01
Formaldehyde
Benzene
1,3 -Butadiene
Naphthalene
Ethylbenzene
POM, Group 2b
Acetaldehyde
POM, Group 3
Hexavalent Chromium, PM
Arsenic, PM
1.67E-04
1.25E-04
6.36E-05
3.90E-05
2.74E-05
2.04E-05
1.84E-05
1.71E-05
1.56E-05
1.05E-05
Formaldehyde
Acetaldehyde
Benzene
Acrylonitrile
Carbon Tetrachloride
Ethylbenzene
Trichloroethylene
1 ,2-Dichloroethane
1,3 -Butadiene
Tetrachloroethylene
30.97
4.13
4.00
3.62
3.34
1.87
0.36
0.35
0.16
0.01
to
to
-------
Table 22-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with
Noncancer RfCs for the South Dakota Monitoring Sites
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Sioux Falls, South Dakota (Minnehaha County) - SSSD
Toluene
Xylenes
Methanol
Benzene
Formaldehyde
Ethylbenzene
Hexane
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
247.74
232.25
104.42
84.66
56.95
54.47
49.47
36.84
13.66
12.38
Acrolein
1,3 -Butadiene
Formaldehyde
Acetaldehyde
Benzene
Xylenes
Naphthalene
Lead, PM
Arsenic, PM
Propionaldehyde
148,988.96
6,191.87
5,811.72
4,093.12
2,821.95
2,322.48
2,068.40
871.54
550.67
461.12
Formaldehyde
Acetaldehyde
Benzene
1,3 -Butadiene
Trichloroethylene
Carbon Tetrachloride
Tetrachloroethylene
Chloroform
Vinyl Chloride
1 ,2-Dichloroethane
0.22
0.21
0.02
0.02
0.01
0.01
0.01
0.01
0.01
O.01
Union County, South Dakota (Union County) - UCSD
Toluene
Xylenes
Benzene
Formaldehyde
Ethylbenzene
Hexane
Acetaldehyde
Methanol
1,3 -Butadiene
1,1,1 -Trichloroethane
49.76
48.01
15.98
12.81
10.98
8.85
8.37
8.23
2.12
1.31
Acrolein
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Cyanide Compounds, gas
Benzene
Xylenes
Naphthalene
Arsenic, PM
Propionaldehyde
33,649.12
1,307.04
1,060.53
930.04
603.39
532.80
480.09
382.14
162.43
119.46
Formaldehyde
Acetaldehyde
Trichloroethylene
Acrylonitrile
Benzene
Carbon Tetrachloride
1,3 -Butadiene
Tetrachloroethylene
Ethylbenzene
Chloroform
0.24
0.21
0.04
0.03
0.02
0.01
O.01
O.01
O.01
O.01
to
to
-------
Observations from Table 22-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in Minnehaha and Union Counties. The emissions are higher in
Minnehaha County than in Union County. UCSD has the lowest emissions of these
three pollutants among all counties with NMP sites.
• Formaldehyde, benzene, and 1,3-butaidene are the pollutants with the highest
toxi city-weighted emissions (of the pollutants with cancer UREs) for both counties.
• Seven of the highest emitted pollutants also have the highest toxi city-weighted
emissions for Minnehaha County. The same seven pollutants appear on both
emissions-based lists for Union County.
• Formaldehyde is the pollutant with the highest cancer surrogate risk approximation
for SSSD; this pollutant also appeared on both emissions-based lists. This is also true
for acetaldehyde, benzene, and 1,3-butadiene. Conversely, carbon tetrachloride
appears on neither emissions-based list but is among the pollutants with the highest
cancer risk approximations for SSSD.
• Formaldehyde, acetaldehyde, benzene, and ethylbenzene are among the pollutants
with the highest cancer surrogate risk approximations for UCSD and appear on both
emissions-based lists. Conversely, acrylonitrile and carbon tetrachloride appear on
neither emissions-based list but were among the pollutants with the highest cancer
risk approximations for UCSD.
Observations from Table 22-8 include the following:
• Toluene and xylenes are the highest emitted pollutants with noncancer RfCs in
Minnehaha and Union Counties. The emissions of these pollutants were an order of
magnitude higher in Minnehaha County than in Union County.
• Acrolein is the pollutant with the highest toxi city-weighted emissions (of the
pollutants with noncancer RfCs) for both counties. Although acrolein was sampled
for at SSSD and UCSD, this pollutant was excluded from the pollutants of interest
designation, and thus subsequent risk screening evaluations, due to questions about
the consistency and reliability of the measurements, as discussed in Section 3.2.
Acrolein is not one of the highest emitted pollutants in Minnehaha or Union Counties.
• Five of the highest emitted pollutants also have the highest toxi city-weighted
emissions for Minnehaha County. The same five pollutants appear on both emissions-
based lists for Union County.
• Formaldehyde and acetaldehyde, which have the highest noncancer risk
approximations for SSSD and UCSD, appear on both emissions-based lists. Benzene
and 1,3-butadiene also appear on all three lists for each South Dakota monitoring site.
22-32
-------
22.6 Summary of the 2010 Monitoring Data for SSSD and UCSD
Results from several of the data treatments described in this section include the
following:
»«» Thirteen pollutants failed at least one screen for SSSD and 15 pollutants failed at
least one screen for UCSD.
»«» Formaldehyde and acetaldehyde had the highest annual average concentrations for
both SSSD and UCSD.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
22-33
-------
23.0 Sites in Texas
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS sites in Texas, and integrates these concentrations with
emissions, meteorological, and risk information. Data generated by sources other than ERG are
not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
23.1 Site Characterization
This section characterizes the CAMS 35 and CAMS 85 monitoring sites by providing
geographical and physical information about the location of the sites and the surrounding areas.
This information is provided to give the reader insight regarding factors that may influence the
air quality near the sites and assist in the interpretation of the ambient monitoring measurements.
The CAMS 35 monitoring site is located in the Houston-Sugarland-Baytown, Texas
MSA and CAMS 85 is part of the Marshall, Texas MSA. Figures 23-1 and 23-2 are composite
satellite images retrieved from ArcGIS Explorer showing the monitoring sites in their urban and
rural locations. Figures 23-3 and 23-4 identify point source emissions locations by source
category for each site, as reported in the 2008 NEI for point sources. Note that only sources
within 10 miles of the sites are included in the facility counts provided in Figures 23-3 and 23-4.
Thus, sources outside the 10-mile radius have been grayed out, but are visible on the maps to
show emissions sources outside the 10-mile boundary. A 10-mile boundary was chosen to give
the reader an indication of which emissions sources and emissions source categories could
potentially have a direct effect on the air quality at the monitoring sites. Further, this boundary
provides both the proximity of emissions sources to the monitoring sites as well as the quantity
of such sources within a given distance of the sites. Table 23-1 describes the area surrounding
each monitoring site by providing supplemental geographical information such as land use,
location setting, and locational coordinates.
23-1
-------
Figure 23-1. Deer Park, Texas (CAMS 35) Monitoring Site
to
-------
Figure 23-2. Karnack, Texas (CAMS 85) Monitoring Site
to
-------
Figure 23-3. NEI Point Sources Located Within 10 Miles of CAMS 35
Legend
© CAMS 35 NATTS site
&&:5'0"W 95 0-D-W
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest
10 mile radius
County boundaries
Source Category Group (No. of Facilities)
+ Aircraft Operations (24)
I Asphalt Processing/Roofing Manufacturing (1)
B Bulk Terminals/Bulk Plants (6)
C Chemical Manufacturing (59)
f Electricity Generation via Combustion (8)
E Electroplating, Plating, Polishing. Anodizing, & Coloring (2)
<•> Fabricated Metal Products (8)
It Glass Manufacturing (1)
.'.' Heating Equipment Manufacturing (1)
• Landfill (2)
V Marine Port (4)
? Miscellaneous Commercial/Industrial (22)
M Miscellaneous Manufacturing (2)
• Oil and/or Gas Production (11)
4 Petroleum Refinery (5)
B Pulp and Paper Plant/Wood Products (3)
R Rubber and Miscellaneous Plastics Products (19)
A Ship Building and Repairing (4)
'•' Stationary Combustion Turbines (1)
* Transportation and Marketing of Petroleum Products (6)
' Wastewater Treatment (2)
23-4
-------
Figure 23-4. NEI Point Sources Located Within 10 Miles of CAMS 85
Legend
94"15t>"W W10WV W5'B"W 94rO'0"W
Note: Due to facilny density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
CAMS 85 NATTS site Q 10 mile radius
County boundaries
Source Category Group (No. of Facilities)
-f Aircraft Operations (3)
• Oil and/or Gas Production (2)
23-5
-------
Table 23-1. Geographical Information for the Texas Monitoring Sites
Site Code
CAMS 35
CAMS 85
AQS Code
48-201-1039
48-203-0002
Location
Deer Park
Karnack
County
Harris
Harrison
Micro- or
Metropolitan
Statistical Area
Houston-Sugar
Land-Baytown,
TXMSA
Marshall, TX
MSA
Latitude
and
Longitude
29.670046,
-95.128485
32.669004,
-94.167449
Land Use
Residential
Agricultural
Location
Setting
Suburban
Rural
Additional Ambient Monitoring Information1
Haze, CO, NOy, NO, NO2, NOX, PAMS, NMOC,
VOC, Carbonyl compounds, O3, Meteorological
parameters, PM10, PM Coarse, PM10 Speciation,
PM2 5, and PM2 5 Speciation, SO2, SVOC.
SVOC, NO2, NO, NOX, PAMS, NMOC, Carbonyl
Compounds, VOC, O3, Meteorological parameters,
PMio, PM10 Speciation, PM25, PM25 Speciation.
These monitoring sites report additional pollutants to AQS (EPA, 2012h); however, these data are not generated by ERG and are therefore not included in this report.
BOLD ITALICS = EPA-designated NATTS Site.
to
-------
The CAMS 35 monitoring site is located in Deer Park, southeast of Houston, in east
Texas. This site serves as the Houston NATTS Site. The site is located at Brown Memorial Park,
in a primarily residential area, as shown in Figure 23-1. Major thoroughfares are near the site,
including Beltway 8 (1.5 miles to the west) and Highway 225 (nearly 3 miles to the north).
Galveston Bay is located to the east and southeast and the Houston Ship Channel, which runs
from the Bay westward towards downtown Houston, is located to the north on the other side of
Highway 225. The east side of Houston has significant industry, including several oil refineries.
As Figure 23-3 shows, the point source located closest to the CAMS 35 monitoring site is a
heliport at San Jacinto College in Pasadena. However, a large number of emissions sources are
located roughly along a line that runs east to west just north of the site (or along the Houston
Ship Channel). A second cluster of emissions sources is located to the southeast of the
monitoring site. The source category with the largest number of sources (59) surrounding
CAMS 35 is chemical manufacturing. Other source categories with a number of sources around
CAMS 35 include aircraft operations, which include airports as well as small runways, heliports,
or landing pads; rubber and miscellaneous plastics products; and oil and gas production.
The CAMS 85 NATTS site is located in Karnack, in northeast Texas. The monitoring is
about 10 miles northeast of Marshall, Texas and about 7 miles from the Texas-Louisiana border.
This site is located on the property of the Longhorn Army Ammunition Plant near the
intersection of FM Road 134 and Spur Road 449 (Taylor Avenue), as shown in Figure 23-2. The
surrounding area is rural and agricultural. As Figure 23-4 shows, there are few point sources
within 10 miles of CAMS 85 and these sources fall into two source categories: aircraft
operations and oil and gas production. The closest source to CAMS 85 is the Shreveport
Regional Airport.
Table 23-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the areas surrounding the Texas
monitoring sites. Table 23-2 also includes a vehicle registration-to-county population ratio
(vehicles-per-person). An estimate of 10-mile vehicle ownership was calculated by applying the
county-level vehicle registration-to-population ratio to the 10-mile population surrounding the
monitoring sites. Table 23-2 also contains annual average daily traffic information. County-level
VMT was not readily available for these sites; thus, daily VMT for CAMS 35 and CAMS 85 is
not provided in Table 23-2.
23-7
-------
Table 23-2. Population, Motor Vehicle, and Traffic Information for the Texas
Monitoring Sites
Site
CAMS 35
CAMS 85
Estimated
County
Population1
4,110,771
65,766
County-level
Vehicle
Registration2
3,115,974
69,883
Vehicles per
Person
(Registration:
Population)
0.76
1.06
Population
within 10
miles3
715,640
3,034
Estimated
10-mile
Vehicle
Ownership
542,457
3,224
Annual
Average
Daily
Traffic4
31,043
1,400
County-
level Daily
VMT5
NA
NA
1 County-level population estimates reflect data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the Texas Department of Motor Vehicles (TX DMV, 2011)
310-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2010 data for CAMS 85 from the Texas DOT and 2004 data for CAMS 35 from
Harris County Public Infrastructure Department (TX DOT, 2010 and HCPID, 2010)
5 County-level VMT was not readily available for these sites
BOLD ITALICS = EPA-designaled NATTS Site
Observations from Table 23-2 include the following:
• The population and vehicle ownership counts are significantly higher at CAMS 35
than CAMS 85. Compared to other counties with NMP monitoring sites, Harris
County is the third highest for both county-level population and county-level vehicle
ownership. Conversely, Harrison County is among the lowest for both county-level
population and county-level vehicle ownership.
• The 10-mile populations for both CAMS 35 and CAMS 85 do not reflect the
magnitude of the county-level populations, indicating that these sites are not located
near the centers of highest population density. The 10-mile population for CAMS 35
is in the middle of the range compared to other NMP sites, while the 10-mile
population for CAMS 85 is the lowest among all NMP sites.
• The vehicle-per-person ratio for CAMS 85 is higher than for CAMS 35. Compared to
other sites, the ratio for CAMS 85 is in the top third while the ratio for CAMS 35 is in
the bottom third.
• The traffic volume passing CAMS 35 is significantly higher than the traffic volume
passing CAMS 85. The traffic volume for CAMS 35 is in the middle of the range
compared to other NMP sites while the traffic volume near CAMS 85 is among the
lower traffic volumes for NMP sites. Traffic data for CAMS 35 were obtained for
Spencer Highway between Red Bluff Road and Underwood Road; the traffic data for
CAMS 85 were obtained for FM Road 134.
23.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in Texas on sample days, as well as over the course of the year.
23-8
-------
23.2.1 Climate Summary
The eastern third of Texas is characterized by a subtropical humid climate, with the
climate becoming more continental in nature farther north and west. The proximity to the Gulf of
Mexico acts as a moderating influence as temperatures soar in the summer or dip in the winter.
Areas closer to the coast, such as Houston, remain slightly cooler in the summer than
neighboring areas to the north. The reverse is also true, as coastal areas are warmer in the winter
than areas farther inland, although East Texas winters are relatively mild. The onshore flow from
the Gulf of Mexico also allows humidity levels to remain high in East Texas, particularly near
the coast. The winds flow out of the Gulf of Mexico a majority of the year, with the winter
months being the exception, as frontal systems allow colder air to filter in from the north.
Abundant rainfall is also typical of the region, again due in part to the nearness to the Gulf of
Mexico. Severe weather is most common in spring, particularly in May, and tropical systems can
be a threat to the state during the summer and fall. Snowfall is rare in East Texas but ice storms
are more common in northeast Texas than in other parts of the state (Bair, 1992 and TAMU,
2012).
23.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather stations nearest these sites were
retrieved for 2010 (NCDC, 2010). The closest weather station to CAMS 35 is located at William
P. Hobby Airport, WBAN 12918; the closest weather station to CAMS 85 is located at
Shreveport Regional Airport, WBAN 13957. Additional information about the Hobby and
Shreveport Regional Airport weather stations, such as the distance between the sites and the
weather stations, is provided in Table 23-3. These data were used to determine how
meteorological conditions on sample days vary from normal conditions throughout the year.
23-9
-------
Table 23-3. Average Meteorological Conditions near the Texas Monitoring Sites
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
Average
Temperature
Average
Dew Point
Temperature
Average
Wet Bulb
Temperature
Average
Relative
Humidity
Average
Sea Level
Pressure
(mb)
Average
Scalar
Wind
Speed
(kt)
Deer Park, Texas - CAMS 35
William P.
Hobby Airport
12918
(29.65, -95.28)
8.85
miles
258°
(WSW)
Sample
Day
2010
77.6
±3.6
77.9
+ 1.4
69.3
±3.5
69.3
+ 1.4
57.7
±4.2
57.9
+ 1.6
62.7
±3.4
62.8
+ 1.4
69.9
±3.6
70.2
+ 1.3
1016.3
±1.3
1016.3
+ 0.6
6.3
±0.7
6.4
+ 0.3
Karnack, Texas - CAMS 85
Shreveport
Regional
Airport
13957
(32.45, -93.82)
24.46
miles
127°
(SE)
Sample
Day
2010
79.3
±4.4
77.1
+ 1.8
68.3
±4.3
66.0
+ 1.7
54.7
±4.4
52.8
+ 1.8
60.5
±3.9
58.6
+ 1.6
65.9
±3.1
66.2
+ 1.2
1015.0
± 1.4
1015.9
+ 0.6
6.0
±0.8
5.9
+ 0.3
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
to
o
-------
Table 23-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 23-3 is the 95
percent confidence interval for each parameter. As shown in Table 23-3, average meteorological
conditions on sample days at CAMS 35 were fairly representative of average weather conditions
throughout the year. Sample days at CAMS 85 appear slightly warmer and more humid.
Sampling at CAMS 85 did not begin until February 2010, thereby missing the coldest month of
the year, which may account for these slight differences.
23.2.3 Back Trajectory Analysis
Figure 23-5 is the composite back trajectory map for days on which samples were
collected at the CAMS 35 monitoring site in 2010. Included in Figure 23-5 are four back
trajectories per sample day. Figure 23-6 is the corresponding cluster analysis for 2010. Similarly,
Figure 23-7 is the composite back trajectory map for days on which samples were collected at
CAMS 85 and Figure 23-8 is the corresponding cluster analysis. An in-depth description of these
maps and how they were generated is presented in Section 3.5.2.1. For the composite maps, each
line represents the 24-hour trajectory along which a parcel of air traveled toward the monitoring
site on a given sample day and time. For the cluster analyses, each line corresponds to a back
trajectory representative of a given cluster of trajectories. For all maps, each concentric circle
around the sites in Figures 23-5 through 23-8 represents 100 miles.
Observations from Figures 23-5 and 23-6 for CAMS 35 include the following:
• Back trajectories originated from a variety of directions at the CAMS 35 monitoring
site, although the majority of trajectories originated over the Gulf of Mexico.
• The 24-hour air shed domain for CAMS 35 is the largest in size compared to other
NMP monitoring sites. Two trajectories originated nearly 1,000 miles away, one over
northwest Nebraska and one over south-central South Dakota. These two trajectories
are for December 12, 2010, a day during which a strong cold front pushed across the
state. Nearly 80 percent of trajectories originated within 400 miles of the site. The
average trajectory length was 284 miles, which is among the longest average
trajectory lengths for NMP sites.
• The cluster analysis for CAMS 35 shows that that the majority (67 percent) of
trajectories originated over the Gulf of Mexico, although the position over the Gulf
and the trajectory length varies. Recall that both direction and distance from the
23-11
-------
monitoring site factor into the cluster analysis. Another common trajectory origin is
from the northwest to north (10 percent). The short cluster trajectory originating to
the north of CAMS 35 (23 percent) represents relatively short back trajectories
originating to the northwest, north, and northeast, and generally within 300 miles of
CAMS 35, over East Texas and Louisiana.
Observations from Figures 23-7 and 23-8 for CAMS 85 include the following:
• Back trajectories originated from a variety of directions at the CAMS 85 monitoring
site, although back trajectories originating to the east and west are rare.
• The 24-hour air shed domain for CAMS 35 is comparable in size to other NMP
monitoring sites. The farthest away a trajectory originated was nearly 800 miles
away, over southeast South Dakota. However, the average trajectory length is 237
miles and most trajectories (85 percent) originated less than 350 miles from
CAMS 85.
• The cluster analysis for CAMS 85 shows that that 44 percent of back trajectories
originated to the south of the site, as indicated by the short cluster (33 percent)
representing trajectories originating over East Texas and the longer cluster
(11 percent) originating over the Gulf of Mexico. Another common trajectory origin
is from the southeast over Louisiana (22 percent). Another 21 percent of back
trajectories originated to the northwest to north of the site, as indicated by the short
cluster (13 percent) representing trajectories originating over Oklahoma and the
longer cluster (8 percent) originating over the central Plains. Lastly, the cluster
trajectory originating from the northeast represents relatively short back trajectories
originating over Arkansas.
23-12
-------
Figure 23-5. 2010 Composite Back Trajectory Map for CAMS 35
Figure 23-6. Back Trajectory Cluster Map for CAMS 35
23-13
-------
Figure 23-7. 2010 Composite Back Trajectory Map for CAMS 85
Figure 23-8. Back Trajectory Cluster Map for CAMS 85
23-14
-------
23.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather stations at Hobby Airport near CAMS 35 and
Shreveport Regional near CAMS 85 were uploaded into a wind rose software program to
produce customized wind roses, as described in Section 3.5.2.2. A wind rose shows the
frequency of wind directions using "petals" positioned around a 16-point compass, and uses
different colors to represent wind speeds.
Figure 23-9 presents three different wind roses for the CAMS 35 monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figure 23-10 presents the
three wind roses and distance map for the CAMS 85 monitoring site.
Observations from Figure 23-9 for CAMS 35 include the following:
• The Hobby Airport weather station is located approximately 8.9 miles west-southwest
of CAMS 35.
• The historical wind rose shows that winds from the southeast quadrant, including
both easterly and southerly winds, prevailed near the CAMS 35 site. Northerly winds
were also observed often. Calm winds (<2 knots) were observed for approximately
13 percent of the wind measurements.
• The wind patterns shown on the 2010 wind rose are very similar to the historical wind
patterns, indicating that conditions during 2010 were similar to conditions observed in
past years.
• The 2010 sample day wind patterns generally resemble the full-year and historical
wind patterns with a few exceptions. The sample day wind rose has fewer northerly
and southerly wind observations and more easterly wind observations.
23-15
-------
Figure 23-9. Wind Roses for the William P. Hobby Airport Weather Station near CAMS 35
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between CAMS 35 and NWS Station
23-16
-------
Figure 23-10. Wind Roses for the Shreveport Regional Airport Weather Station near
CAMS 85
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between CAMS 85 and NWS Station
23-17
-------
Observations from Figure 23-10 for CAMS 85 include the following:
• The Shreveport Regional Airport weather station is located across the Texas-
Louisiana border, approximately 24.5 miles southeast of CAMS 85.
• The wind patterns on the wind roses for CAMS 85 resemble those on the wind roses
for CAMS 35.
• The historical wind rose shows that winds from the southeast to south account for just
less than one-third of the wind observations near the CAMS 85. Northerly winds were
also observed often. Calm winds were observed for approximately 16 percent of the
wind measurements.
• The wind patterns shown on the 2010 wind rose are very similar to the historical wind
patterns, indicating that conditions during 2010 were similar to conditions observed in
past years.
• The 2010 sample day wind patterns resemble the full-year and historical wind
patterns, indicating that wind conditions on sample days were representative of those
experienced throughout 2010 and historically near CAMS 85.
23.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Texas monitoring sites in
order to allow analysts and readers to focus on a subset of pollutants through the context of risk.
For each site, each pollutant's preprocessed daily measurement was compared to its associated
risk screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by each monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 23-4 presents the pollutants of interest for CAMS 35 and CAMS 85. The pollutants
that failed at least one screen and contributed to 95 percent of the total failed screens for each
monitoring site are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of interest
are shaded and/or bolded. CAMS 35 sampled for hexavalent chromium and PAH while
CAMS 85 sampled for hexavalent chromium only. Note that hexavalent chromium sampling at
these sites through the NMP did not begin until February 2010.
23-18
-------
Table 23-4. Risk Screening Results for the Texas Monitoring Sites
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Deer Park, Texas - CAMS 35
Naphthalene
Hexavalent Chromium
Acenaphthene
Fluorene
0.029
0.000083
0.011
0.011
Total
55
2
1
1
59
57
52
57
57
223
96.49
3.85
1.75
1.75
26.46
93.22
3.39
1.69
1.69
93.22
96.61
98.31
100.00
Karnack, Texas - CAMS 85
Hexavalent Chromium
0.000083
Total
33
33
51
51
64.71
64.71
100.00
100.00
Observations from Table 23-4 include the following:
• Four pollutants (three PAH and hexavalent chromium) failed at least one screen for
CAMS 35.
• Naphthalene contributed to 93 percent of the total number of failed screens for
CAMS 35. Hexavalent chromium, acenaphthene, and fluorene contribute to the other
seven percent of failed screens.
• Naphthalene and hexavalent chromium were initially identified as pollutants of
interest for CAMS 35. Benzo(a)pyrene was added to CAMS 35's pollutants of
interest because it is a NATTS MQO Core Analyte, even though it did not fail any
screens. Benzo(a)pyrene is not shown in Table 23-4.
• Hexavalent chromium is the only pollutant of interest for CAMS 85.
• Hexavalent chromium failed nearly 65 percent of screens for CAMS 85. This is a
much higher percentage than for CAMS 35, even though the number of samples
collected at each site is similar.
23.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Texas monitoring sites. Concentration averages are provided for the pollutants of interest
for each Texas site, where applicable. Concentration averages for select pollutants are also
presented graphically for each site, where applicable, to illustrate how each site's concentrations
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at the
sites, where applicable. Additional site-specific statistical summaries are provided in Appendices
M and O.
23-19
-------
23.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each Texas site, as described in Section 3.1. The quarterly average of a particular pollutant is
simply the average concentration of the preprocessed daily measurements over a given calendar
quarter. Quarterly average concentrations include the substitution of zeros for all non-detects. A
site must have a minimum of 75 percent valid samples of the total number of samples possible
within a given quarter for a quarterly average to be calculated. An annual average includes all
measured detections and substituted zeros for non-detects for the entire year of sampling. Annual
averages were calculated for pollutants where three valid quarterly averages could be calculated
and where method completeness was greater than or equal to 85 percent, as presented in
Section 2.4. Quarterly and annual average concentrations for the Texas monitoring sites are
presented in Table 23-5, where applicable. Note that if a pollutant was not detected in a given
calendar quarter, the quarterly average simply reflects "0" because only zeros substituted for
non-detects were factored into the quarterly average concentration.
Table 23-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Texas Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(ng/m3)
2nd
Quarter
Average
(ng/m3)
3rd
Quarter
Average
(ng/m3)
4th
Quarter
Average
(ng/m3)
Annual
Average
(ng/m3)
Deer Park, Texas - CAMS 35
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
16/57
52/52
57/57
0.03
±0.03
NA
71.27
± 22.97
0.01
±0.01
0.06
±0.01
84.59
± 19.97
0.01
±0.01
0.06
±0.01
111.18
± 46.77
0.02
±0.03
0.04
±0.01
108.14
±27.83
0.02
±0.01
0.05
±0.01
92.93
±14.64 |
Karnack, Texas - CAMS 85 |
Hexavalent Chromium
51/51
NA
0.19
±0.06
0.72
±0.44
0.06
±0.04
0.31 1
±0.16
Observations from Table 23-5 include the following:
• Naphthalene's annual average concentration is significantly higher than the annual
averages for benzo(a)pyrene and hexavalent chromium for CAMS 35.
• Although naphthalene concentrations appear to be higher during the second half of
the year, the confidence intervals indicate that concentrations of this pollutant have a
lot of variability associated with them. Although the highest concentration of
naphthalene was measured at CAMS 35 on August 24, 2010 (302 ng/m3), a review of
23-20
-------
the data shows that concentrations greater than 100 ng/m3 were measured several
times in each quarter of 2010.
• Benzo(a)pyrene was detected in less than half the PAH samples collected at
CAMS 35. The confidence intervals for each of the quarterly averages are equal to or
greater than their associated average concentrations, indicating that outliers may be
affecting these averages. Three concentrations greater than 0.1 ng/m3 were measured
at CAMS 35 (one in February and two in November), while the magnitude of the
measured detections ranged from 0.00959 ng/m3 to 0.211 ng/m3. Although the
number of measured detections ranged from as few as two for the second quarter to
seven for the first quarter of 2010, the number of zeros substituted into the quarterly
averages may be a bigger factor in variability of the quarterly averages, as opposed to
potential outliers.
• Because hexavalent chromium sampling through the NMP did not begin until
February 2010, first quarter average concentrations could not be calculated for either
Texas site.
• The second, third, and fourth quarter average concentrations for CAMS 85 are greater
than the same quarterly concentrations for CAMS 35, particularly for the third
quarter. The maximum concentration of hexavalent chromium for CAMS 85
(3.51 ng/m3) is significantly higher than the maximum concentration of hexavalent
chromium for CAMS 35 (0.108 ng/m3). This measurement is the highest hexavalent
chromium concentration among NMP sites sampling this pollutant in 2010 and the
highest hexavalent chromium concentration measured among NMP sites since this
method was added to the program in 2005. The seven July measurements of
hexavalent chromium for CAMS 85 are the seven highest measurements among all
NMP sites sampling this pollutant in 2010. Of the 25 measurements of hexavalent
chromium greater than 0.25 ng/m3 across the program in 2010, 17 were measured at
CAMS 85 (and eight were measured at PXSS).
• As shown in Table 4-12, the annual average hexavalent chromium concentration for
CAMS 85 is more than twice the next highest annual average hexavalent chromium
concentration (PXSS). Although six times lower, the annual average hexavalent
chromium concentration for CAMS 35 ranks third highest among NMP sites.
23.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for benzo(a)pyrene,
hexavalent chromium, and naphthalene for created for CAMS 35. A box plot for hexavalent
chromium for CAMS 85 was also created. Figures 23-11 through 23-13 overlay the sites'
minimum, annual average, and maximum concentrations onto the program-level minimum, first
23-21
-------
quartile, average, median, third quartile, and maximum concentrations, as described in
Section 3.5.3.
Figure 23-11. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
k
F
1
Program Max Concentration = 42.7 ng/m3 j
1
0.6 0.8 1 1.2
Concentration (ng/m3)
1.6 1.8
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 23-12. Program vs. Site-Specific Average Hexavalent Chromium Concentration
||_ j Program Max Concentration =3.51ng/m3
i i i
P
i i i
r^ i Program Max Concentration = 3.51 ng/m3
1
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2nd Quartile 3rd Quartile 4th Quartile Average
Site: Site Average Site Minimum/Maximum
o
23-22
-------
Figure 23-13. Program vs. Site-Specific Average Naphthalene Concentration
A
IT
1 1 1 1 1 1
600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rd Quartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o —
Observations from Figures 23-11 through 23-13 include the following:
• Figure 23-11 is the box plot for benzo(a)pyrene for CAMS 35. Note that the
program-level maximum concentration (42.7 ng/m3) is not shown directly on the
box plot because the scale of the box plot would be too large to readily observe
data points at the lower end of the concentration range. Thus, the scale has been
reduced to 2 ng/m3. Also note that the first quartile for this pollutant is zero and is
not visible on this box plot. This box plot shows that the annual average
concentration for CAMS 35 is well below the program-level average
concentration and just less than the program-level median. Figure 23-11 also
shows that the maximum concentration measured at CAMS 35 is well below the
maximum concentration measured across the program. Many non-detects of
benzo(a)pyrene were measured at CAMS 35, as discussed in the previous section.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 23-12 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 23-12 shows the annual average concentrations of hexavalent chromium
for both sites are greater than the program-level average, although the annual
average concentration for CAMS 85 is six times higher than the annual average
concentration for CAMS 35. While the maximum concentration measured at
CAMS 35 is well below the maximum concentration measured across the
program, CAMS 85 has the highest concentration among NMP sites sampling
hexavalent chromium, as discussed in the previous section. The minimum
concentration of hexavalent chromium measured at both CAMS 35 and CAMS 85
is greater than the first quartile (25th percentile) across the program.
• Figure 23-13 shows that the annual average naphthalene concentration for
CAMS 35 is similar to the program-level average concentration. The maximum
23-23
-------
naphthalene concentration measured at CAMS 35 is well below the program-level
maximum concentration. There were no non-detects of naphthalene measured at
CAMS 35.
23.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. CAMS 35 has not sampled PAH continuously for 5 years as part of the NMP and
both sites began sampling hexavalent chromium under the NMP in 2010. Therefore, a trends
analysis was not conducted.
23.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the Texas
monitoring sites. Refer to Sections 3.3 and 3.5.5 for definitions and explanations regarding the
various risk factors, time frames, and calculations associated with these risk screenings.
23.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Texas monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where available.
As described in Section 3.3, acute risk results from exposures of 1 to 14 days; intermediate risk
results from exposures of 15 to 364 days; and chronic risk results from exposures of 1 year or
greater. The preprocessed daily measurements of the pollutants of interest were compared to the
acute MRL; the quarterly averages were compared to the intermediate MRL; and the annual
averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the Texas monitoring sites were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
the Texas monitoring sites.
23.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Texas monitoring sites and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
23-24
-------
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 23-6, where applicable.
Table 23-6. Cancer and Noncancer Surrogate Risk Approximations for the Texas
Monitoring Sites
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(ng/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Deer Park, Texas - CAMS 35
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
0.00176
0.012
0.000034
0.0001
0.003
16/57
52/52
57/57
0.02
±0.01
0.05
±0.01
92.93
± 14.64
0.03
0.61
3.16
0.01
0.03
Karnack, Texas - CAMS 85
Hexavalent Chromium
0.012
0.0001
51/51
0.31
±0.16
3.70
<0.01
— = a Cancer URE or Noncancer RfC is not available.
Observations from Table 23-6 include the following:
• The cancer risk approximation for naphthalene for CAMS 35 is 3.16 in-a-million,
based on the annual average. This is the only cancer risk approximation greater than
1.0 in-a-million for CAMS 35.
• The cancer risk approximation for hexavalent chromium for CAMS 85 is
3.70 in-a-million, based on the annual average. This cancer risk approximation is
more than five times the cancer risk approximation for hexavalent chromium for
CAMS 35.
• The noncancer risk approximations for CAMS 35 and CAMS 85, where they could be
calculated, are well below the level of concern, an HQ of 1.0.
23.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 23-7 and 23-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 23-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer risk approximations (in-a-million), as calculated from the annual averages. Table 23-8
presents similar information, but identifies the 10 pollutants with the highest noncancer risk
approximations (HQ), also calculated from annual averages.
23-25
-------
Table 23-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Texas Monitoring Sites
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Cancer Risk
Approximation
Pollutant (in-a-million)
Deer Park, Texas (Harris County) - CAMS 35
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Methyl tert butyl ether
Naphthalene
Propylene oxide
Dichloromethane
Tetrachloroethylene
1,418.35
1,260.11
827.79
699.55
446.24
168.21
142.12
85.80
77.80
23.32
Formaldehyde
1,3 -Butadiene
Benzene
Hexavalent Chromium, PM
Naphthalene
Nickel, PM
Arsenic, PM
Ethylbenzene
Acetaldehyde
POM, Group 2b
1.64E-02
1.34E-02
1.11E-02
1.04E-02
4.83E-03
4.62E-03
2.26E-03
2.07E-03
1.54E-03
1.44E-03
Naphthalene 3.16
Hexavalent Chromium 0.61
Benzo(a)pyrene 0.03
Karnack, Texas (Harrison County) - CAMS 85
Formaldehyde
Benzene
Acetaldehyde
Ethylbenzene
1,3 -Butadiene
Naphthalene
Ethylene oxide
Dichloromethane
Chloromethylbenzene
Carbon tetrachloride
108.38
64.37
55.52
37.76
15.22
13.37
9.90
3.67
1.31
1.08
Hexavalent Chromium, PM
Formaldehyde
Ethylene oxide
Benzene
1,3 -Butadiene
Naphthalene
Nickel, PM
Arsenic, PM
Acetaldehyde
Ethylbenzene
6.35E-03
1.41E-03
8.72E-04
5.02E-04
4.57E-04
4.54E-04
3.29E-04
1.34E-04
1.22E-04
9.44E-05
Hexavalent Chromium 3.70
to
to
-------
Table 23-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Texas Monitoring Sites
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer
Toxicity
Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Deer Park, Texas (Harris County) - CAMS 35
Toluene
Xylenes
Methanol
Hexane
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Styrene
4,428.76
3,249.06
2,707.58
1,559.23
1,418.35
1,260.11
827.79
699.55
446.24
359.89
Acrolein
1,3 -Butadiene
Formaldehyde
Nickel, PM
Acetaldehyde
Titanium tetrachloride
Hexamethylene-l,6-diisocyanate, gas
Chlorine
Naphthalene
Benzene
4,253,301.87
223,117.95
128,583.14
106,919.25
77,728.16
77,037.49
62,970.00
57,589.67
47,373.42
47,278.41
Naphthalene 0.03
Hexavalent Chromium <0.01
Karnack, Texas (Harrison County) - CAMS 85
Toluene
Xylenes
Formaldehyde
Benzene
Ethylene glycol
Acetaldehyde
Hexane
Methanol
Chloromethane
Ethylbenzene
155.11
142.12
108.38
64.37
63.56
55.52
53.09
48.73
40.86
37.76
Acrolein
Hexamethylene-l,6-diisocyanate, gas
Manganese, PM
Chlorine
Formaldehyde
Cyanide Compounds, PM
Nickel, PM
1,3 -Butadiene
Acetaldehyde
Hexavalent Chromium, PM
647,128.44
31,490.00
23,210.90
22,445.25
11,059.67
9,776.20
7,623.51
7,611.02
6,168.54
5,294.84
Hexavalent Chromium O.01
to
to
-------
The pollutants listed in Tables 23-7 and 23-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer surrogate risk approximations based on each site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 23.3, both Texas monitoring sites sampled hexavalent chromium; in addition, CAMS 35
also sampled for PAH. In addition, the cancer and noncancer surrogate risk approximations are
limited to those pollutants with enough data to meet the criteria for annual averages to be
calculated. A more in-depth discussion of this analysis is provided in Section 3.5.5.3.
Observations from Table 23-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in Harris County. Formaldehyde, benzene, and acetaldehyde are the
highest emitted pollutants with cancer UREs in Harrison County. The magnitude of
the emissions is significantly higher in Harris County than in Harrison County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
cancer UREs) for Harris County are formaldehyde, 1,3-butadiene, and benzene. The
pollutants with the highest toxicity-weighted emissions for Harrison County are
hexavalent chromium, formaldehyde, and ethylene oxide.
• Six of the highest emitted pollutants in Harris County also have the highest toxicity-
weighted emissions while seven of the highest emitted pollutants in Harrison County
also have the highest toxicity-weighted emissions.
• Naphthalene is the only pollutant of interest that appears on both emissions-based
lists for CAMS 35. Although hexavalent chromium, which has the second highest
cancer risk approximation for CAMS 35, appears among the pollutants with the
highest toxicity-weighted emissions, this pollutant is not one of the highest emitted in
Harris County.
• POM, Group 2b ranks tenth for toxicity-weighted emissions in Harris County. POM,
Group 2b includes several PAH sampled for at CAMS 35 including acenaphthylene,
benzo(e)pyrene, fluoranthene, and perylene. Benzo(a)pyrene, another pollutant of
interest for CAMS 35, is part of POM, Group 5a, which does not appear on either
emissions-based list for Harris County.
• Hexavalent chromium, the only pollutant of interest for CAMS 85, is the pollutant
with the highest toxicity-weighted emissions for Harrison County, but is not among
the highest emitted.
23-28
-------
Observations from Table 23-8 include the following:
• Toluene, xylenes, and methanol are the highest emitted pollutants with noncancer
RfCs in Harris County. Toluene, xylenes, and formaldehyde are the highest emitted
pollutants in Harrison County. The magnitude of the emissions is significantly higher
in Harris County than in Harrison County.
• The pollutant with the highest toxicity-weighted emissions (of the pollutants with
noncancer RfCs) for both counties is acrolein.
• Four of the highest emitted pollutants also have the highest toxicity-weighted
emissions for Harris County while only two of the highest emitted pollutants also
have the highest toxicity-weighted emissions for Harrison County.
• Naphthalene ranks ninth for toxicity-weighted emissions in Harris County but is not
one of the highest emitted. Hexavalent chromium does not appear on either
emissions-based list for Harris County.
• Hexavalent chromium ranks tenth for toxicity-weighted emissions in Harrison County
but is not one of the highest emitted.
23.6 Summary of the 2010 Monitoring Data for CAMS 35 and CAMS 85
Results from several of the data treatments described in this section include the
following:
»«» Four pollutants failed at least one screen for CAMS 35, with naphthalene accounting
for 93 percent of the total failed screens. Hexavalent chromium failed screens for
CAMS 85, although it was the only pollutant sampled for at this site.
*»* Of the site-specific pollutants of the interest, naphthalene had the highest annual
average concentration for CAMS 35.
»«» The maximum concentration ofhexavalent chromium for CAMS 85 was the highest
hexavalent chromium concentration among allNMP sites sampling this pollutant in
2010 and the highest hexavalent chromium concentration measured among NMP
sites since this method was added to the program in 2005.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
23-29
-------
24.0 Site in Utah
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in Utah, and integrates these concentrations with
emissions, meteorological, and risk information. Data generated by sources other than ERG are
not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
24.1 Site Characterization
This section characterizes the BTUT monitoring site by providing geographical and
physical information about the location of the site and the surrounding area. This information is
provided to give the reader insight regarding factors that may influence the air quality near the
site and assist in the interpretation of the ambient monitoring measurements.
BTUT is located in Bountiful, in northern Utah. Figure 24-1 is a composite satellite
image retrieved from ArcGIS Explorer showing the monitoring site in its urban location.
Figure 24-2 identifies point source emissions locations by source category, as reported in the
2008 NEI for point sources. Note that only sources within 10 miles of the site are included in the
facility counts provided in Figure 24-2. Thus, sources outside the 10-mile radius have been
grayed out, but are visible on the map to show emissions sources outside the 10-mile boundary.
A 10-mile boundary was chosen to give the reader an indication of which emissions sources and
emissions source categories could potentially have a direct effect on the air quality at the
monitoring site. Further, this boundary provides both the proximity of emissions sources to the
monitoring site as well as the quantity of such sources within a given distance of the site.
Table 24-1 describes the area surrounding the monitoring site by providing supplemental
geographical information such as land use, location setting, and locational coordinates.
24-1
-------
Figure 24-1. Bountiful, Utah (BTUT) Monitoring Site
Tl
to
-------
Figure 24-2. NEI Point Sources Located Within 10 Miles of BTUT
112'IQ'CrW 112'5'trW
trssxrw nt-so'trw nv^xrw
111 55'fJ-W lirSCNTW 111'45'CrW 111'40'Q"W
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
BTUT NATTS site O 10 mile radius | | County boundary
Source Category Group (No. of Facilities)
Air-conditioning/Refrigeration (1 )
-f Aircraft Operations (8)
* Electricity Generation via Combustion (2)
® Institutional - school (1 )
? Miscellaneous Commercial/Industrial (2)
M Miscellaneous Manufacturing (1)
^ Petroleum Refinery (5)
24-3
-------
Table 24-1. Geographical Information for the Utah Monitoring Site
Site
Code
BTUT
AQS Code
49-011-0004
Location
Bountiful
County
Davis
Micro- or
Metropolitan
Statistical Area
Ogden-Clearfield,
UTMSA
Latitude
and
Longitude
40.902967,
-111.884467
Land Use
Residential
Location
Setting
Suburban
Additional Ambient Monitoring Information1
SO2, NO, NO2, NOX, PAMS, O3, Meteorological
parameters, PM10, PM2 5, and PM2 5 Speciation.
BOLD ITALICS = EPA-designated NATTS Site.
to
-k
-------
Bountiful is north of Salt Lake City, and is situated in a valley between the Great Salt
Lake to the west and the Wasatch Mountains to the east. Figure 24-1 shows that BTUT is located
on the property of Viewmont High School, in a primarily residential area. The site is located
about one-third of a mile from 1-15, which runs north-south through most of the surrounding
urban area including Salt Lake City, Clearfield, and Ogden. Figure 24-2 shows that all of the
point sources near BTUT are located to the south of the site. The facilities surrounding BTUT
are involved in a variety of industries, although the source categories with the highest number of
point sources surrounding BTUT include aircraft operations, which include airports as well as
small runways, heliports, or landing pads, and petroleum refineries. The source closest to BTUT
generates electricity via combustion.
Table 24-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the Utah
monitoring site. Table 24-2 also includes a vehicle registration-to-county population ratio
(vehicles-per-person). In addition, the population within 10 miles of the site is presented. An
estimate of 10-mile vehicle ownership was calculated by applying the county-level vehicle
registration-to-population ratio to the 10-mile population surrounding the monitoring site.
Table 24-2 also contains annual average daily traffic information. Finally, Table 24-2 presents
the daily VMT for Davis County.
Table 24-2. Population, Motor Vehicle, and Traffic Information for the Utah Monitoring
Site
Site
BTUT
Estimated
County
Population1
307,856
County-level
Vehicle
Registration2
239,754
Vehicles per
Person
(Registration:
Population)
0.78
Population
within 10
miles3
259,066
Estimated
10-mile
Vehicle
Ownership
201,757
Annual
Average
Daily
Traffic4
113,955
County-
level
Daily
VMT5
7,360,752
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the Utah Tax Commission (UT TC, 2010)
3 10-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2010 data from the Utah DOT (UT DOT, 2010)
5 County-level VMT reflects 2010 data from the Utah DOT (UT DOT, 2011)
BOLD ITALICS = EPA-designated NATTS Site.
Observations from Table 24-2 include the following:
• Davis County's population is in the mid-to-low end of the range, as is its 10-mile
population, compared to counties with NMP sites. The county-level vehicle
24-5
-------
registration and 10-mile ownership estimate rankings are similar to the population
rankings.
• The vehicle-per-person ratio (0.78) is in the bottom third of the range compared to
other NMP sites.
• The traffic volume experienced near BTUT is in the top third compared to other NMP
monitoring sites. The traffic estimate used came from the intersection of 1-15 with
US-89, just west of the site.
• The Davis County VMT is on the low end compared to counties with NMP sites
(where VMT was available).
24.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in Utah on sample days, as well as over the course of the year.
24.2.1 Climate Summary
The Salt Lake City area's climate can be classified as semi-arid and continental in nature,
and experiences large seasonal variations. Summers are hot and dry while winters are cold and
snow is common. The area is generally dry, with spring as the wettest season, and sunshine
prevails across the area during much of the year. Precipitation that does fall can be enhanced
over the eastern parts of the valley as storm systems move up the side of the Wasatch Mountains,
located to the east. Surrounding mountains protect the valley from winter storm systems moving
in from the southwest or north, preventing cold air outbreaks. The Great Salt Lake tends to have
a moderating influence on the area's temperature. Moderate winds flow out of the southeast on
average, although there is a valley breeze/lake breeze system that affects the area. High pressure
systems that occasionally settle over the area can result in stagnation episodes (Bair, 1992 and
WRCC, 2012).
24.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest BTUT were retrieved
for 2010 (NCDC, 2010). The closest weather station is located at Salt Lake City International
Airport (WBAN 24127). Additional information about the Salt Lake City International Airport
weather station, such as the distance between the site and the weather station, is provided in
24-6
-------
Table 24-3. These data were used to determine how meteorological conditions on sample days
vary from normal conditions throughout the year.
Table 24-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 24-3 is the
95 percent confidence interval for each parameter. As shown in Table 24-3, average
meteorological conditions on sample days were representative of average weather conditions
throughout the year.
24.2.3 Back Trajectory Analysis
Figure 24-3 is the composite back trajectory map for days on which samples were
collected at the BTUT monitoring site in 2010. Included in Figure 24-3 are four back trajectories
per sample day. Figure 24-4 is the corresponding cluster analysis for 2010. An in-depth
description of these maps and how they were generated is presented in Section 3.5.2.1. For the
composite map, each line represents the 24-hour trajectory along which a parcel of air traveled
toward the monitoring site on a given sample day and time. For the cluster analysis, each line
corresponds to a back trajectory representative of a given cluster of trajectories. For both maps,
each concentric circle around the site in Figures 24-3 and 24-4 represents 100 miles.
Observations from Figures 24-3 and 24-4 include the following:
• Back trajectories originated from a variety of directions at BTUT.
• Similar to other sites located in the inter-mountain west, the 24-hour air shed domain
for BTUT is smaller in size compared to other NMP monitoring sites. The farthest
away a trajectory originated was over northeast Oregon, less than 450 miles away. A
trajectory of similar length also originated over west-central Arizona. However, the
average trajectory length was 177 miles and nearly 90 percent of back trajectories
originated within 300 miles of the site.
• The cluster analysis shows that one-third of back trajectories originated from the
south of BTUT, although of varying distances, as represented by the shorter trajectory
(13 percent) and the longer trajectory (21 percent). Trajectories also originated from
the west of BTUT, generally over northwest Utah and northeast Nevada, and from the
northwest, over Idaho. The short cluster trajectory originating from the east of BTUT
represents short trajectories originating to the northeast, east, and southeast of the site.
24-7
-------
to
J^.
oo
Table 24-3. Average Meteorological Conditions near the Utah Monitoring Site
Closest NWS Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Bountiful, Utah - BTUT
Salt Lake City
International
24127
(40.79, -111.97)
8.98
miles
217°
(SW)
Sample
Day
2010
62.1
±5.2
62.6
+ 2.1
52.6
±4.5
52.6
+ 1.9
33.7
±2.2
33.2
+ 1.0
43.0
±2.8
42.8
+ 1.2
56.9
±5.6
55.9
+ 2.2
1014.2
±1.8
1014.2
+ 0.8
6.7
±0.7
6.8
+ 0.4
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
-------
Figure 24-3. 2010 Composite Back Trajectory Map for BTUT
Figure 24-4. Back Trajectory Cluster Map for BTUT
24-9
-------
24.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at Salt Lake City International Airport
near BTUT were uploaded into a wind rose software program to produce customized wind roses,
as described in Section 3.5.2.2. A wind rose shows the frequency of wind directions using
"petals" positioned around a 16-point compass, and uses different colors to represent wind
speeds.
Figure 24-5 presents three different wind roses for the BTUT monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at this location.
Observations from Figure 24-5 for BTUT include the following:
• The Salt Lake City International weather station is located approximately 9 miles
southwest of BTUT.
• The historical wind rose shows that southeasterly, south-southeasterly, and southerly
winds were prevalent near BTUT. Winds from the north-northwest to north were also
common. Calm winds (<2 knots) were observed for approximately 11 percent of the
hourly measurements from 1999-2009. The strongest wind speeds were observed
with south-southeasterly and southerly winds.
• The wind patterns shown on the 2010 wind rose are similar to the historical wind
patterns, although there were slightly more calm winds and fewer southeasterly to
south-southeasterly winds. This indicates that wind conditions in 2010 were similar to
conditions experienced historically near BTUT.
• The wind patterns shown on the sample day wind rose resemble the 2010 wind
patterns, indicating that conditions on sample days were representative of those
experienced over the entire year (and historically).
24-10
-------
Figure 24-5. Wind Roses for the Salt Lake City International Airport Weather Station near
BTUT
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between BTUT and NWS Station
24-11
-------
24.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the BTUT monitoring site in
order to allow analysts and readers to focus on a subset of pollutants through the context of risk.
Each pollutant's preprocessed daily measurement was compared to its associated risk screening
value. If the concentration was greater than the risk screening value, then the concentration
"failed the screen." Pollutants of interest are those for which the individual pollutant's total
failed screens contribute to the top 95 percent of the site's total failed screens. In addition, if any
of the NATTS MQO Core Analytes measured by the BTUT monitoring site did not meet the
pollutant of interest criteria based on the preliminary risk screening, that pollutant was added to
the list of site-specific pollutants of interest. A more in-depth description of the risk screening
process is presented in Section 3.2.
Table 24-4 presents BTUT's pollutants of interest. The pollutants that failed at least one
screen and contributed to 95 percent of the total failed screens for the monitoring site are shaded.
NATTS MQO Core Analytes are bolded. Thus, pollutants of interest are shaded and/or bolded.
BTUT sampled for VOC, carbonyl compounds, SNMOC, PAH, metals (PMio), and hexavalent
chromium and is one of only two sites sampling the entire suite of pollutants under the NMP
(NBIL is the other).
Observations from Table 24-4 include the following:
• Twenty-six pollutants, of which 14 are NATTS MQO Core Analytes, failed at least
one screen for BTUT.
• The risk screening process identified 14 pollutants of interest for BTUT, of which
eight are NATTS MQO Core Analytes. Six additional pollutants (benzo(a)pyrene,
cadmium, chloroform, hexavalent chromium, nickel, and trichloroethylene) were
added to BTUT's pollutants of interest because they are NATTS MQO Core
Analytes, even though they did not contribute to 95 percent of the total failed screens.
Four more pollutants were added to BTUT's pollutants of interest because they are
also NATTS MQO Core Analytes, even though they did not fail any screens:
beryllium, lead, tetrachloroethylene, and vinyl chloride. These four pollutants are not
shown in Table 24-4.
• The pollutants not identified as pollutants of interest via the risk screening process
failed two or less screens for BTUT.
• Nearly 50 percent of measured detections failed screens (of the pollutants that failed
at least one screen) for BTUT.
24-12
-------
• Recall from Section 3.2 that if a pollutant was measured by both the TO-15 and
SNMOC methods at the same site, the TO-15 results were used for the risk screening
process. As BTUT sampled both VOC (TO-15) and SNMOC, the TO-15 results were
used for the 12 pollutants these methods have in common.
Table 24-4. Risk Screening Results for the Utah Monitoring Site
Pollutant
Screening
Value
(Hg/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Bountiful, Utah - BTUT
Benzene
Carbon Tetrachloride
Acetaldehyde
Formaldehyde
1,3-Butadiene
Arsenic (PM10)
Naphthalene
Manganese (PM10)
Dichloromethane
Ethylbenzene
/>-Dichlorobenzene
Propionaldehyde
Acrylonitrile
1 ,2-Dichloroethane
Trichloroethylene
Acenaphthylene
Benzo(a)pyrene
Cadmium (PM10)
Chloroform
Chloromethylbenzene
1 ,2-Dibromoethane
Hexachloro- 1 ,3 -butadiene
Hexavalent Chromium
Nickel (PM10)
1, 1,2,2-Tetrachloroethane
Xylenes
0.13
0.17
0.45
0.077
0.03
0.00023
0.029
0.005
7.7
0.4
0.091
0.8
0.015
0.038
0.2
0.011
0.00057
0.00056
9.8
0.02
0.0017
0.045
0.000083
0.0021
0.017
10
Total
57
56
53
53
51
49
43
31
20
20
15
8
7
7
2
1
1
1
1
1
1
1
1
1
1
1
483
57
57
53
53
53
59
57
59
57
57
42
53
7
7
16
30
10
59
43
1
1
1
53
59
1
57
1,002
100.00
98.25
100.00
100.00
96.23
83.05
75.44
52.54
35.09
35.09
35.71
15.09
100.00
100.00
12.50
3.33
10.00
1.69
2.33
100.00
100.00
100.00
1.89
1.69
100.00
1.75
48.20
11.80
11.59
10.97
10.97
10.56
10.14
8.90
6.42
4.14
4.14
3.11
1.66
1.45
1.45
0.41
0.21
0.21
0.21
0.21
0.21
0.21
0.21
0.21
0.21
0.21
0.21
11.80
23.40
34.37
45.34
55.90
66.05
74.95
81.37
85.51
89.65
92.75
94.41
95.86
97.31
97.72
97.93
98.14
98.34
98.55
98.76
98.96
99.17
99.38
99.59
99.79
100.00
24.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Utah monitoring site. Concentration averages are provided for the pollutants of interest for
BTUT, where applicable. Concentration averages for select pollutants are also presented
graphically for the site, where applicable, to illustrate how the site's concentrations compare to
the program-level averages. In addition, concentration averages for select pollutants are
24-13
-------
presented from previous years of sampling in order to characterize concentration trends at the
site, where applicable. Additional site-specific statistical summaries are provided in Appendix J
through Appendix O.
24.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for BTUT, as described in Section 3.1. The quarterly average of a particular pollutant is simply
the average concentration of the preprocessed daily measurements over a given calendar quarter.
Quarterly average concentrations include the substitution of zeros for all non-detects. A site must
have a minimum of 75 percent valid samples of the total number of samples possible within a
given quarter for a quarterly average to be calculated. An annual average includes all measured
detections and substituted zeros for non-detects for the entire year of sampling. Annual averages
were calculated for pollutants where three valid quarterly averages could be calculated and
where method completeness was greater than or equal to 85 percent, as presented in Section 2.4.
Quarterly and annual average concentrations for the Utah monitoring site are presented in
Table 24-5, where applicable. Note that concentrations of the PAH, metals, and hexavalent
chromium are presented in ng/m3 for ease of viewing. Also note that if a pollutant was not
detected in a given calendar quarter, the quarterly average simply reflects "0" because only zeros
substituted for non-detects were factored into the quarterly average concentration.
Table 24-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Utah Monitoring Site
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Hg/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Annual
Average
(Ug/m3)
Bountiful, Utah - BTUT
Acetaldehyde
Acrylonitrile
Benzene
1,3 -Butadiene
Carbon Tetrachloride
53/53
7/57
57/57
53/57
57/57
NA
0.04
±0.04
1.46
±0.38
0.17
±0.07
0.55
±0.06
2.13
±0.27
0
0.94
±0.15
0.06
±0.02
0.60
±0.06
3.11
±0.67
0.03
±0.04
1.08
±0.29
0.05
±0.02
0.57
±0.08
1.50
±0.43
0.01
±0.02
1.43
±0.40
0.13
±0.04
0.59
±0.09
2.25
±0.27
0.02
±0.02
1.22
±0.16
0.10
±0.02
0.58
±0.03
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of viewing.
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
24-14
-------
Table 24-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Utah Monitoring Site (Continued)
Pollutant
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Dichloromethane
Ethylbenzene
Formaldehyde
Propionaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Arsenic (PM10)a
Benzo(a)pyrene a
Bery Ilium (PM10)a
Cadmium (PM10)a
Hexavalent Chromium3
Lead(PM10)a
Manganese (PM10) a
Naphthalene a
Nickel (PM10)a
#of
Measured
Detections
vs. # of
Samples
43/57
42/57
7/57
57/57
57/57
53/53
53/53
53/57
16/57
4/57
59/59
10/57
30/59
59/59
53/59
59/59
59/59
57/57
59/59
1st
Quarter
Average
(Ug/m3)
0.08
±0.02
0.27
±0.32
0.07
±0.05
105.67
± 157.03
0.48
±0.15
NA
NA
0.16
±0.05
0.03
±0.03
0.01
±0.01
0.93
±0.51
0.11
±0.11
0
0.17
±0.08
0.03
±0.01
3.28
±1.62
4.15
±1.29
76.42
± 29.05
0.92
±0.17
2nd
Quarter
Average
(Ug/m3)
0.09
±0.02
0.24
±0.42
0.01
±0.01
121.14
±245.91
0.30
±0.06
3.50
±0.56
0.55
±0.07
0.10
±0.02
0.07
±0.13
0
0.33
±0.10
0.01
±0.01
O.01
±0.01
0.05
±0.01
0.01
±0.01
1.81
±0.52
5.50
±1.71
36.98
±7.60
0.96
±0.29
3rd
Quarter
Average
(Ug/m3)
0.86
±1.57
1.48
±2.90
0
6.53
±5.65
0.60
±0.57
6.10
±1.53
0.80
±0.19
0.18
±0.12
0.22
±0.41
0
0.44
±0.10
0
0.01
±0.01
0.07
±0.02
0.03
±0.03
2.25
±0.71
7.69
±1.53
47.66
± 12.32
0.88
±0.19
4th
Quarter
Average
(Ug/m3)
0.05
±0.04
0.12
±0.13
0
274.95
±354.08
0.54
±0.17
1.79
±0.23
0.31
±0.09
0.18
±0.07
0.03
±0.03
0
0.74
±0.28
0.02
±0.03
O.01
±0.01
0.12
±0.04
0.02
±0.01
3.45
±1.40
5.06
±1.51
73.79
± 27.42
1.01
±0.36
Annual
Average
(Ug/m3)
0.28
±0.40
0.54
±0.74
0.02
±0.01
125.23
± 109.76
0.48
±0.15
3.66
±0.63
0.55
±0.08
0.15
±0.04
0.09
±0.11
O.01
±O.01
0.61
±0.15
0.03
±0.03
O.01
±0.01
0.10
±0.03
0.03
±0.01
2.68
±0.57
5.61
±0.79
57.83
± 10.49
0.94
±0.12
a Average concentrations provided for the pollutants below the black line are presented in ng/m for ease of viewing.
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
Observations for BTUT from Table 24-5 include the following:
• The pollutants with the highest annual average concentrations by mass are
dichloromethane, formaldehyde, acetaldehyde, and benzene, consistent with the last
few years of sampling. The annual average for dichloromethane is significantly
higher than the annual averages of the other pollutants.
24-15
-------
• Dichloromethane has the highest annual average for BTUT, but also has a very large
confidence interval associated it, as do the quarterly averages. This indicates the
likely presence of outliers. The concentrations of dichloromethane at BTUT range
from 0.251 to 2,430 |ig/m3. Three measurements of this pollutant are greater than
1,000 |ig/m3, nine are greater than 100 |ig/m3, and a total of 20 are greater than
10.0 |ig/m3. Although the maximum concentration of dichloromethane measured
across the program was not measured at BTUT, this site has the highest number of
dichloromethane measurements greater than 10.0 |ig/m3 (20 out of a program total of
29). Only one other NMP site has a dichloromethane measurement greater than
100 |ig/m3 (GPCO, which also has the maximum measurement program-wide).
• Although some pollutants of interest exhibit quarterly trends, such as 1,3-butadiene,
which was highest during the colder months (first and fourth quarters), most of the
pollutants exhibit high variability overall and/or may be affected by potential outliers
(such as/>-dichlorobenzene), as illustrated by the quarterly averages and their
associated confidence intervals.
• Several pollutants appear higher in one quarter or another, but have very large
confidence intervals associated them. For example, chloroform is highest during the
third quarter of 2010, but the confidence interval for this quarterly average is twice
the average itself, indicating the presence of outliers. A review of the data shows that
the highest concentration of chloroform was measured on August 12, 2010
(11.5 |ig/m3) and is the second highest chloroform concentration measured across the
program. The August 12, 2010 concentration is two orders of magnitude higher than
the next highest concentration (0.28 |ig/m3 confirm measured at BTUT on
July 25, 2010).
• The maximum concentration of several of the VOC was measured on July 25, 2010,
including /?-dichlorobenzene, ethylbenzene, tetrachloroethylene, and
trichl oroethy 1 ene.
• The confidence intervals for each of the quarterly average concentrations of
/>-dichlorobenzene are higher than the averages themselves, indicating a high level of
variability within each quarterly average and/or the presence of outliers. A review of
the data shows that the highest concentration of />-dichlorobenzene was measured on
July 25, 2010 (21.2 |ig/m3) and is nearly seven times higher than the next highest
concentration (3.11 |ig/m3 measured on June 13, 2010). These are the two highest
/>-dichlorobenzene concentrations measured across the program. BTUT also has the
highest/>-dichlorobenzene concentration in 2009 and the second highest in 2008. Of
the 57 valid samples collected at BTUT in 2010, three measurements were greater
than 1 |ig/m3 and 15 were greater than 0.1 |ig/m3. The median/>-dichlorobenzene
concentration is 0.06 |ig/m3.
• For 1,2-dichloroethane, all seven of the measured detections were measured during
the first and second quarters of 2010. Six of the seven measured detections were
measured in January and February and one was measured in April, with no measured
detections after April 2, 2010.
24-16
-------
• First quarter 2010 average concentrations could not be calculated for the carbonyl
compounds because they did not meet the completeness criteria for calculating a
quarterly average.
• Formaldehyde concentrations appear highest during the warmer months of the year,
although without a first quarter average concentration, this is a difficult assessment to
make. However, 11 of the 12 concentrations of formaldehyde greater than 5 |ig/m3,
ranging from 5.02 |ig/m3to 12.7 |ig/m3, were measured in July, August, and
September, while eight of the nine concentrations less than 2 |ig/m3 were measured in
October, November, and December.
• Concentrations of naphthalene appear highest in the colder months, although the
confidence intervals indicate that the difference is not statistically significant. Seven
of the eight concentrations greater than 100 ng/m3 were measured in the first and
fourth quarters of 2010.
• Concentrations of benzo(a)pyrene also appear highest in the colder months. Although
this pollutant was detected in only 10 samples, six of these were measured during the
first quarter of 2010, two in the second quarter and two in the fourth quarter.
• Of the PMio metals, manganese has the highest annual average concentration and the
highest quarterly averages.
• Arsenic, cadmium, and lead appear to be higher during the colder months of the year
(first and fourth quarters). For each of these pollutants, the highest concentrations
were measured during the first and second quarters of 2010.
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for BTUT from
those tables include the following:
• BTUT has highest annual average concentration of />-dichlorobenzene, as shown in
Table 4-9. BTUT also has the second highest annual average concentration of
trichloroethylene, the third highest annual average of vinyl chloride, and the fourth
highest annual average concentrations of chloroform and 1,2-dichloroethane.
• BTUT has the second highest annual average concentration of formaldehyde (behind
only ELNJ) and seventh highest annual average concentration of acetaldehyde, as
shown in Table 4-10.
• BTUT does not appear in Table 4-11 for PAH.
• BTUT has the third highest annual average concentration of arsenic (behind only
S4MO and NBIL). With the exception of arsenic, annual average concentrations for
the metals for BTUT are in the middle of the range compared to other NMP sites
sampling PMio metals, as shown in Table 4-12.
24-17
-------
24.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for acetaldehyde, arsenic,
benzene, benzo(a)pyrene, 1,3-butadiene, formaldehyde, hexavalent chromium, manganese, and
naphthalene were created for BTUT. Figures 24-6 through 24-14 overlay the site's minimum,
annual average, and maximum concentrations onto the program-level minimum, first quartile,
average, median, third quartile, and maximum concentrations, as described in Section 3.5.3.
Figure 24-6. Program vs. Site-Specific Average Acetaldehyde Concentration
Concentration (|ig/m3)
Program
Site:
: 1st Quartile
Site Average
o
2nd Quartile 3rd Quartile 4th Quartile Avt
Site Minimum/Maximum
^^^^^~
;rage
Figure 24-7. Program vs. Site-Specific Average Arsenic (PMio) Concentration
1°
1
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Concentration (ng/m3)
Program
Site:
: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Average
• D D D
Site Average Site Minimum/Maximum
o —
24-18
-------
Figure 24-8. Program vs. Site-Specific Average Benzene Concentration
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 24-9. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
I
• Mf t f 47 7 / 3
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 24-10. Program vs. Site-Specific Average 1,3-Butadiene Concentration
n-
^
1 1 1 1 1 1 1
0.3 0.4 0.5 0.6
Concentration (ng/m3)
0.8 0.9
Program: IstQuartile
•
Site:
Site Average
0
2ndQuartile SrdQuartile 4thQuartile Ave
Site Minimum/Maximum
^^^^~
;rage
24-19
-------
Figure 24-11. Program vs. Site-Specific Average Formaldehyde Concentration
15 20
25 30 35
Concentration (ng/m3)
45 50
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 24-12. Program vs. Site-Specific Average Hexavalent Chromium Concentration
I
j Program Max Concentration = 3.51 ng/m3 j
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
Figure 24-13. Program vs. Site-Specific Average Manganese (PMi0) Concentration
•J
80 100 120
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
0
24-20
-------
Figure 24-14. Program vs. Site-Specific Average Naphthalene Concentration
•H ; ; ; ;
600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rd Quartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o —
Observations from Figures 24-6 through 24-14 include the following:
• Figure 24-6 shows that BTUT's annual average acetaldehyde concentration is
greater than the program-level average and just less than the program-level third
quartile (75th percentile). There were no non-detects of acetaldehyde measured at
BTUT.
• Figure 24-7 shows that BTUT's annual average arsenic (PMio) concentration is
just greater than the program-level average for arsenic (PMio). Although the
maximum concentration of arsenic at the program level was not measured at
BTUT, the maximum concentration of arsenic for BTUT is the third highest
among sites sampling PMio metals. There were no non-detects of arsenic
measured at BTUT.
• Figure 24-8 for benzene shows the annual average concentration for BTUT is
greater than the program-level average and just greater than the program-level
third quartile (75th percentile). The maximum concentration of benzene measured
at BTUT is well below the maximum concentration measured across the program.
There were no non-detects of benzene measured at BTUT.
• Figure 24-9 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
BTUT is well below the program-level average concentration. Figure 24-9 also
shows that the maximum concentration measured at BTUT is well below the
maximum concentration measured across the program. A number of non-detects
of benzo(a)pyrene were measured at BTUT.
• Figure 24-10 for 1,3-butadiene shows the annual average concentration for BTUT
is greater than the program-level average and just less than the program-level
third quartile (75th percentile). The maximum concentration of 1,3-butadiene
24-21
-------
measured at BTUT is below the maximum concentration measured across the
program. A few non-detects of 1,3-butadiene were measured at BTUT.
• Figure 24-11 shows that BTUT's annual average formaldehyde concentration is
greater than the program-level average and greater than the program-level third
quartile (75th percentile). The minimum concentration of formaldehyde measured
at BTUT is just less than the program-level first quartile (25th percentile).
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 24-12 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for the observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 24-12 shows the annual average concentration of hexavalent chromium for
BTUT is less than the program-level average. The maximum concentration
measured at BTUT is well below the program-level maximum concentration.
There were a few non-detects of hexavalent chromium at BTUT.
• Figure 24-13 shows that the annual average concentration of manganese (PMio)
for BTUT is less than the program-level average. The maximum concentration
measured at BTUT is well below the program-level maximum concentration.
There were no non-detects of manganese measured at BTUT.
• Figure 24-14 shows that the annual average naphthalene concentration for BTUT
is less than both the program-level average and median concentrations. The
maximum naphthalene concentration measured at BTUT is well below the
program-level maximum concentration. There were no non-detects of naphthalene
measured at BTUT.
24.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. BTUT has sampled carbonyl compounds, VOC, metals, and SNMOC as part of
the NMP since July 2003. BTUT has also sampled hexavalent chromium since 2005. Thus,
Figures 24-15 through 24-21 present the 3-year rolling statistical metrics for acetaldehyde,
arsenic, benzene, 1,3-butadiene, formaldehyde, hexavalent chromium, and manganese for
BTUT, respectively. The statistical metrics presented for assessing trends include the substitution
of zeros for non-detects.
24-22
-------
Figure 24-15. Three-Year Rolling Statistical Metrics for Acetaldehyde Concentrations
Measured at BTUT
!15
2005-2007 2006-2008
Three-Year Period
5th Percentile — Minimum - Median — Maximum • 95th Percentile •••*•• Average
Sampling for carbonyl compounds began in July 2003.
Figure 24-16. Three-Year Rolling Statistical Metrics for Arsenic (PMio) Concentrations
Measured at BTUT
1.20
1
c
.2
0 -
^^ ^
2003-2005 L 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Percentile - Minimum - Median - Maximum * 95th Percentile .-.*.. Average
Sampling for PM10 metals began in July 2003.
24-23
-------
Figure 24-17. Three-Year Rolling Statistical Metrics for Benzene Concentrations Measured
at BTUT
"10
c
o
4
2
4
1 i * « •
2003-2005 x 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Percentile - Minimum - Median - Maximum • 95th Pert en tile ...*.- Average
'Sampling for VOC began in July 2003.
Figure 24-18. Three-Year Rolling Statistical Metrics for 1,3-Butadiene Concentrations
Measured at BTUT
Concentration (ng/m3)
= P
k CT>
0 -
<
M
.003
1
T
- ^ ' 1 ' 1 f
2005 L 2004-1006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Percentile - Minimum - Median - Maximum * 95th Percentile ...+.. Average
'Sampling for VOC began in July 2003.
24-24
-------
Figure 24-19. Three-Year Rolling Statistical Metrics for Formaldehyde Concentrations
Measured at BTUT
40
1
>
5 -
-,111
I
>•— *
— ^_ ^jgjjj, ^^ ^^
2003-2005 i 2004-2006 2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
• 5th Percentile -
Minimum — Median — Maximum • 95th Percentile ...^.. Average
Sampling for carbonyl compounds began in July 2003.
Figure 24-20. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at BTUT
I
1
u.z
•
••
mm MM mmmm ^^H
2005-2007 2006-2008 2007-2009 2008-2010
Three-Year Period
* 5th Percentile — Minimum - Median - Maximum « 95th Percentile ...^.. Average
24-25
-------
Figure 24-21. Three-Year Rolling Statistical Metrics for Manganese (PMi0) Concentrations
Measured at BTUT
2005-2007 2006-2008
Three-Year Period
5th Percentfle - Minimum - Median — Maximum • 95th Percentile ...4.. Average
Sampling for PM10 metals began in July 2003.
Observations from Figure 24-15 for acetaldehyde measurements include the following:
• The maximum acetaldehyde concentration was measured in 2004 (32.7 |ig/m3). The
second highest concentration of acetaldehyde measured at BTUT is the maximum
shown for 2008 (20.0 |ig/m3).
• Both the rolling average and median concentrations exhibit a steady decrease through
2007-2009, after which the average concentration held steady and the median
increased slightly.
• The range of the majority of concentrations measured has also decreased, as indicated
by the decreasing spread between the 5th and 95th percentiles.
Observations from Figure 24-16 for arsenic measurements include the following:
• The maximum arsenic concentration was measured in 2004. The maximum
concentration measured (33.0 ng/m3) is nearly twice the next highest concentration
(16.8 ng/m3), also measured in 2004. The three highest measurements since sampling
began in 2003 were all measured in 2004; further, eight of the 12 highest
concentrations of arsenic (those greater than 5 ng/m3) were measured in 2004. Of
these 12, eight were measured in the first quarter of the calendar year and four were
measured during the fourth quarter of the calendar year, supporting the tendency
discussed in Section 24.4.1.
24-26
-------
Although difficult to discern in Figure 24-16, the rolling average concentrations of
arsenic decreased through the 2006-2008 time period, increased slightly for the 2007-
2009 time frame, then decreased slightly for 2008-2010. The median decreased as
well, but was static from 2006-2008 to 2007-2009, then decreased again for 2008-
2010.
Observations from Figures 24-17 for benzene include the following:
• The maximum concentration of benzene was measured in 2003 (15.8 |ig/m3). The
next highest concentration (9.44 |ig/m3) was also measured in 2003.
• The rolling average and median concentrations have a decreasing trend through the
2006-2008 time frame, after which a slight increase is shown for 2007-2009. These
metrics hold steady for the 2008-2010 time frame.
• Non-detects of benzene have not been measured at BTUT since the onset of VOC
sampling.
Observations from Figures 24-18 for 1,3-butadiene include the following:
• The maximum concentration of 1,3-butadiene was measured in October 2003. This is
the only concentration of 1,3-butaidene greater than 1 |ig/m3.
• The minimum, 5th percentile, and median concentrations are all zero for the 2003-
2005 time frame, indicating that at least 50 percent of the measurements were non-
detects. The detection rate of 1,3-butadiene has increased over the period sampling,
up to a 100 percent detection rate for 2008 and 2009, although a few non-detects were
reported in 2010.
• Figure 24-18 shows that the rolling average concentration has changed little over the
years of sampling.
Observations from Figure 24-19 for formaldehyde measurements include the following:
• The maximum formaldehyde concentration was measured in 2004 (45.4 |ig/m3), on
the same day as the highest acetaldehyde concentration, August 31, 2004. This
measurement is more than twice the next highest concentration (18.2 |ig/m3),
measured in 2007. Concentrations of similar magnitude were also measured on
additional days in 2004 and 2007.
• The rolling average concentrations increased slightly from 2003-2005 to 2004-2006,
then decreased each period through 2007-2009, and then held steady for 2008-2010.
This is also true of the median concentrations.
• The difference between the median and the average concentrations decreased over
most of the periods shown, indicating decreasing variability in the central tendency of
formaldehyde measurements.
24-27
-------
Observations from Figure 24-20 for hexavalent chromium measurements include the
following:
• The maximum hexavalent chromium concentration was measured on July 4, 2006.
The next highest concentration was measured on July 25, 2010 and is roughly half as
high.
• Both the rolling average and median concentrations increased slightly during the
second 3-year period then returned to 2005-2007 levels for the third and fourth 3-year
periods. These changes, however, are not statistically significant.
• The minimum and 5th percentile are both zero for each time frame, indicating the
presence of non-detects. The number of non-detects has varied over the years, ranging
from eight percent (2006) to 38 percent (2009).
Observations from Figure 24-21 for manganese measurements include the following:
• The maximum manganese concentration (40.4 ng/m3) was measured in 2004,
although the next highest concentration, measured in 2007, is similar in magnitude
(36.0 ng/m3). The second, third, and fourth highest concentrations were measured in
2007.
• The rolling average concentrations exhibit an increasing trend through the 2006-2008
time frame, but decrease for the last two 3-year periods, particularly for the 2008-
2010 time frame. However, the calculation of confidence intervals shows that these
changes are not statistically significant. The median follows a similar trend, with a
more significant decrease for the final time frame.
• The difference between the 5th and 95th percentiles increased over each period except
for the final time frame, indicating an increasing spread in the measurements of
manganese since the onset of sampling.
24.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the
BTUT monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations regarding
the various risk factors, time frames, and calculations associated with these risk screenings.
24.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Utah monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where available. As
described in Section 3.3, acute risk results from exposures of 1 to 14 days; intermediate risk
results from exposures of 15 to 364 days; and chronic risk results from exposures of 1 year or
greater. The preprocessed daily measurements of the pollutants of interest were compared to the
24-28
-------
acute MRL; the quarterly averages were compared to the intermediate MRL; and the annual
averages were compared to the chronic MRL. The results of this risk screening are summarized
in Table 24-6. Where a quarterly or annual average exceeds the applicable MRL, the
concentration is bolded.
Observations from Table 24-6 include the following:
• Dichloromethane is the only pollutant for BTUT where a preprocessed daily
measurement and/or time-period average was greater than one or more of the MRL
noncancer health risk benchmarks.
• One out of 59 measured detections of dichloromethane is greater than the ATSDR
acute MRL for this pollutant (2,000 |ig/m3). This concentration was measured on
October 11, 2010 (2,430 |ig/m3) and is the second highest concentration of
dichloromethane measured among NMP sites sampling this pollutant (behind only
GPCO) and is one of two dichloromethane concentrations to exceed the acute MRL,
as discussed in Section 4.2.2.
• Although none of the quarterly average concentrations of dichloromethane are greater
than the ATSDR intermediate MRL of 1,000 |ig/m3, each of the quarterly averages
reflects the inclusion of outliers, based on the confidence intervals, which are all
greater than the associated averages, with the exception of the third quarter average
concentration. These four quarterly averages are the second, third, fourth, and fifth
highest quarterly averages calculated for dichloromethane among NMP sites (the
highest ranking quarterly average of dichloromethane was calculated for the fourth
quarter average for GPCO).
• Although the annual average concentration of dichloromethane for BTUT
(125.23 ± 109.76 |ig/m3) is less than the ATSDR chronic MRL for this pollutant
(1,000 |ig/m3), this is the highest annual average concentration calculated among any
site-specific pollutants of interest.
For the pollutants whose concentrations are greater than their respective ATSDR acute
MRL noncancer health risk benchmark(s), the concentrations were further examined by
developing pollution roses for these pollutants. A pollution rose is a plot of concentration vs.
wind speed and wind direction, as described in Section 3.5.5.1. Figure 24-22 is the
dichloromethane pollution rose for BTUT.
24-29
-------
Table 24-6. Noncancer Risk Screening Summary for the Utah Monitoring Site
Pollutant
Acute
ATSDR
Acute
MRL1
(Hg/m3)
#of
Concentrations
>MRL
#of
Measured
Detections
Intermediate
ATSDR
Intermediate
MRL1
(Hg/m3)
1st
Quarter
Average
(jig/m3)
2nd
Quarter
Average
(jig/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Chronic
ATSDR
Chronic
MRL1
(Hg/m3)
Annual
Average
(jig/m3)
Bountiful, Utah - BTUT
Dichloromethane
2,000
1
59
1,000
105.67
± 157.03
121.14
±245.91
6.53
±5.65
274.95
±354.08
1,000
125.23
± 109.76
Reflects the use of one significant digit for the MRLs
to
-j^
OJ
o
-------
Figure 24-22. Dichloromethane Pollution Rose for BTUT
360/0
2^0
/\, ••-•'" ,--'""'
>' / "x* ,-•"**
/' /' '•' "x •*""'
/ /' / /' Xt j.''*
/ / /'""^, ••"-"*" *
III// .o-p"
/ / ; ; / / oX °° «
! f i ' / X .-©^
f / ' O VO
| I j 4 W-S
\il\\\ vx-
\ \ \ \ \ \X '^---
;
""""•" ^
--,,^ N^
*'""*'-., '*N, .''' \
— , '"'-, '*'»*''' %'
"""— , ***» ,''" %** "'\
'""-^
.'* '*
""'--, .••>'.
,-• '•-.
"--x S\ \ \ \ \
-. ,"
X \ \ \ \ \
---. -•• -.,
v \ \ \ \ \ \
-,./\ \ \ \ 1 1
/' 'f i 'i i 90
3>® /2 4 |6 8 10 12 14 .16 kts
® V / ;'
""©' \, ,/ / .. ' /
3--"®"X'x / /III]
a V\ / /' / / /
.—---"'' /-, X /' / / /
o O .--' \ /
°o
* -" .--' X /
x / /
>-- ,.--' ,x.
...» "" .---" „>' ATSDRR1EL = 2,000 iia/m3.
.,.--'' ,..-'' which corresponds to the upper
.--- .,.-•'' end of the 100-2,000 jigin3 (or
,.--•"'' yellow) concerttrationrange
__._ - ""
> 0-100 fig/m3
180
0100-2,000 iig,in3
O>2,000)iig/ni3
-------
Observations from the Figure 24-22 include the following:
• There is only one measured detection that is greater than the AT SDR acute MRL
(2,000 |ig/m3) for dichloromethane (shown in orange).
• The concentration greater than the ATSDR acute MRL was measured on a day with
winds blowing from the southeast (on average). A review of the hourly wind data
shows that southeasterly winds in the morning shifted northwesterly in the afternoon,
reflecting the valley breeze effect often experienced in the area.
• The bulk of the dichloromethane concentrations greater than 100 |ig/m3 were
measured on days with a northwesterly average wind direction. The average wind
speed for these days is fairly light. Of the nine concentrations greater than 100 |ig/m3,
eight of them were measured in the first and fourth quarters of 2010, although the one
in June is the second highest dichloromethane concentration measured at BTUT
(1,793 |ig/m3).
• Note that nearly all of the concentrations were measured on days with northwesterly
or southeasterly winds, reflecting the same wind direction distribution on the wind
roses shown in Section 24.2.4.
• Figure 24-2 shows that there are no point sources located to the northwest and
southeast of BTUT, as reported in the 2008 NEI.
24.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Utah monitoring site and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 24-7, where applicable.
Observations for BTUT from Table 24-7 include the following:
• The pollutants with the highest annual average concentrations are dichloromethane,
formaldehyde, acetaldehyde, and benzene.
• The pollutants with the highest cancer risk approximations are formaldehyde,
dichloromethane, benzene, and/?-dichlorobenzene.
• There were no pollutants of interest with a noncancer risk approximation greater than
1.0. The highest noncancer risk approximation was calculated for formaldehyde
(0.37).
24-32
-------
Table 24-7. Cancer and Noncancer Surrogate Risk Approximations for the Utah
Monitoring Site
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Bountiful, Utah - BTUT
Acetaldehyde
Acrylonitrile
Arsenic (PM10)a
Benzene
Benzo(a)pyrene a
Bery Ilium (PM10)a
1,3 -Butadiene
Cadmium (PM10) a
Carbon Tetrachloride
Chloroform
£>-Dichlorobenzene
1 ,2-Dichloroethane
Dichloromethane
Ethylbenzene
Formaldehyde
Hexavalent Chromium3
Lead(PM10)a
Manganese (PM10) a
Naphthalene a
Nickel (PM10)a
Propionaldehyde
Tetrachloroethylene
0.0000022
0.000068
0.0043
0.0000078
0.00176
0.0024
0.00003
0.0018
0.000006
0.000011
0.000026
1.3E-07
0.0000025
0.000013
0.012
0.000034
0.00048
2.6E-07
0.009
0.002
0.000015
0.03
0.00002
0.002
0.00001
0.1
0.098
0.8
2.4
0.6
1
0.0098
0.0001
0.00015
0.00005
0.003
0.00009
0.008
0.04
53/53
7/57
59/59
57/57
10/57
30/59
53/57
59/59
57/57
43/57
42/57
7/57
57/57
57/57
53/53
53/59
59/59
59/59
57/57
59/59
53/53
53/57
2.25
±0.27
0.02
±0.02
O.01
±<0.01
1.22
±0.16
O.01
±<0.01
0.01
±0.01
0.10
±0.02
0.01
±0.01
0.58
±0.03
0.28
±0.40
0.54
±0.74
0.02
±0.01
125.23
± 109.76
0.48
±0.15
3.66
±0.63
O.01
±O.01
0.01
±0.01
0.01
±O.01
0.06
±0.01
O.01
±O.01
0.55
±0.08
0.15
±0.04
4.94
1.29
2.61
9.50
0.05
0.01
3.00
0.18
3.45
5.97
0.46
16.28
1.20
47.55
0.31
1.97
0.45
0.04
0.25
0.01
0.04
0.04
0.01
0.05
0.01
0.01
0.01
O.01
O.01
0.21
O.01
0.37
O.01
0.02
0.11
0.02
0.01
0.07
O.01
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 24-5.
24-33
-------
Table 24-7. Cancer and Noncancer Surrogate Risk Approximations for the Utah
Monitoring Site (Continued)
Pollutant
Trichloroethylene
Vinyl Chloride
Cancer
URE
(jig/m3)1
0.0000048
0.0000088
Noncancer
RfC
(mg/m3)
0.002
0.1
#of
Measured
Detections
vs. # of
Samples
16/57
4/57
Annual
Average
(jig/m3)
0.09
±0.11
<0.01
±<0.01
Cancer Risk
Approximation
(in-a-million)
0.42
0.02
Noncancer
Risk
Approximation
(HQ)
0.04
<0.01
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 24-5.
24.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 24-8 and 24-9 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 24-8 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer risk approximations (in-a-million), as calculated from the annual averages. Table 24-9
presents similar information, but identifies the 10 pollutants with the highest noncancer risk
approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 24-8 and 24-9 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
tables. The cancer and noncancer surrogate risk approximations based on each site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 24.3, BTUT sampled for VOC, carbonyl compounds, SNMOC, metals (PMio), PAH, and
hexavalent chromium. In addition, the cancer and noncancer surrogate risk approximations are
limited to those pollutants with enough data to meet the criteria for annual averages to be
calculated. A more in-depth discussion of this analysis is provided in Section 3.5.5.3.
24-34
-------
Table 24-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Utah Monitoring Site
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Bountiful, Utah (Davis County) - BTUT
Benzene
Ethylbenzene
Formaldehyde
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
POM, Group la
POM, Group 6
174.72
80.37
76.71
44.32
20.55
10.19
2.66
1.59
0.20
0.14
Benzene
Formaldehyde
1,3 -Butadiene
Naphthalene
Ethylbenzene
POM, Group 3
POM, Group 2b
Hexavalent Chromium, PM
Acetaldehyde
POM, Group 5a
1.36E-03
9.97E-04
6.17E-04
3.46E-04
2.01E-04
1.87E-04
1.40E-04
1.31E-04
9.75E-05
7.17E-05
Formaldehyde
Dichloromethane
Benzene
£>-Dichlorobenzene
Acetaldehyde
Carbon Tetrachloride
1,3 -Butadiene
Arsenic
Naphthalene
Acrylonitrile
47.55
16.28
9.50
5.97
4.94
3.45
3.00
2.61
1.97
1.29
to
-------
Table 24-9. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Utah Monitoring Site
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Bountiful, Utah (Davis County) - BTUT
Toluene
Xylenes
Benzene
Methanol
Hexane
Ethylbenzene
Formaldehyde
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
403.75
329.01
174.72
171.92
100.56
80.37
76.71
44.32
22.52
20.55
Acrolein
1,3 -Butadiene
Formaldehyde
Benzene
Acetaldehyde
Naphthalene
Xylenes
Lead, PM
Arsenic, PM
Propionaldehyde
215,538.97
10,275.13
7,827.57
5,823.93
4,924.23
3,395.03
3,290.06
1,295.06
736.79
619.63
Formaldehyde
Acetaldehyde
Dichloromethane
Manganese
Propionaldehyde
1,3 -Butadiene
Trichloroethylene
Benzene
Arsenic
Naphthalene
0.37
0.25
0.21
0.11
0.07
0.05
0.04
0.04
0.04
0.02
to
-------
Observations from Table 24-8 include the following:
• Benzene, ethylbenzene, and formaldehyde are the highest emitted pollutants with
cancer UREs in Davis County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
cancer UREs) are benzene, formaldehyde, 1,3-butadiene.
• Seven of the highest emitted pollutants also have the highest toxi city-weighted
emissions.
• Formaldehyde and benzene, which have the highest and third highest cancer risk
approximations for BTUT, appear near the top of both emissions-based lists.
Dichloromethane, which has the second highest cancer risk approximation for BTUT,
ranks seventh for emissions in Davis County but does not have one of the highest
toxicity-weighted emissions (it ranks 21st). />-Dichlorobenzene, which has the fourth
highest cancer risk approximation for BTUT, appears on neither emissions-based list.
• POM, Group 2b is the eighth highest emitted "pollutant" in Davis County and ranks
seventh for toxicity-weighted emissions. POM, Group 2b includes several PAH
sampled for at BTUT including acenaphthylene, fluoranthene, and perylene. None of
the PAH included in POM, Group 2b were identified as pollutants of interest for
BTUT.
• POM, Group 5a ranks tenth for toxicity-weighted emissions in Davis County. POM,
Group 5a includes benzo(a)pyrene, which is one of BTUT's pollutants of interest.
POM, Group 5a is not one of the highest emissions pollutants and is not among the
pollutants with the highest cancer risk approximations for BTUT.
Observations from Table 24-9 include the following:
• Toluene, xylenes, and benzene are the highest emitted pollutants with noncancer
RfCs in Davis County.
• The pollutants with the highest toxicity-weighted emissions (of the pollutants with
noncancer RfCs) are acrolein, 1,3-butadiene, and formaldehyde. Although acrolein
was sampled for at BTUT, this pollutant was excluded from the pollutants of interest
designation, and thus subsequent risk screening evaluations, due to questions about
the consistency and reliability of the measurements, as discussed in Section 3.2.
• Five of the highest emitted pollutants also have the highest toxicity-weighted
emissions.
• Although less than the level of concern, formaldehyde, acetaldehyde, and
dichloromethane have the highest noncancer risk approximations for BTUT.
Formaldehyde and acetaldehyde rank third and fifth (respectively) for toxicity-
weighted emissions and seventh and eighth (respectively) for total emissions.
Dichloromethane appears on neither emissions-based list.
24-37
-------
24.6 Summary of the 2010 Monitoring Data for BTUT
Results from several of the data treatments described in this section include the
following:
»«» Twenty-six pollutants failed at least one screen for BTUT; of these, 14 were NATTS
MQO Core Analytes.
»«» Dichloromethane had the highest annual average concentration among the pollutants
of interest for BTUT, followed by formaldehyde and acetaldehyde.
»«» One preprocessed daily measurement of dichloromethane from BTUT was greater
than its associated acute MRL noncancer health risk benchmark. None of the
quarterly or annual average concentrations of the pollutants of interest were greater
than their associated intermediate or chronic MRL noncancer health risk
benchmarks.
24-38
-------
25.0 Sites in Vermont
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the UATMP and NATTS sites in Vermont, and integrates these
concentrations with emissions, meteorological, and risk information. Data generated by sources
other than ERG are not included in the data analyses contained in this report. Readers are
encouraged to refer to Sections 1 through 4 and the glossary (Appendix P) for detailed
discussions and definitions regarding the various data analyses presented below.
25.1 Site Characterization
This section characterizes the Vermont monitoring sites by providing geographical and
physical information about the location of the sites and the surrounding areas. This information
is provided to give the reader insight regarding factors that may influence the air quality near the
sites and assist in the interpretation of the ambient monitoring measurements.
The Vermont NATTS site (UNVT) and one of the UATMP sites (BURVT) are located in
northwest Vermont in the Burlington-South Burlington, VT MSA. The third site is located
farther south in Rutland, Vermont. Figures 25-1 through 25-3 are the composite satellite images
retrieved from ArcGIS Explorer showing the monitoring sites in their urban and rural locations.
Figures 25-4 and 25-5 identify point source emissions locations by source category, as reported
in the 2008 NEI for point sources. Note that only sources within 10 miles of the sites are
included in the facility counts provided in Figures 25-4 and 25-5. Thus, sources outside the 10-
mile radius have been grayed out, but are visible on the maps to show emissions sources outside
the 10-mile boundary. A 10-mile boundary was chosen to give the reader an indication of which
emissions sources and emissions source categories could potentially have a direct effect on the
air quality at the monitoring sites. Further, this boundary provides both the proximity of
emissions sources to the monitoring sites as well as the quantity of such sources within a given
distance of the sites. Table 25-1 describes the area surrounding the monitoring sites by providing
supplemental geographical information such as land use, location setting, and locational
coordinates.
25-1
-------
Figure 25-1. Burlington, Vermont (BURVT) Monitoring Site
to
-------
Figure 25-2. Underbill, Vermont (UNVT) Monitoring Site
to
I
OJ
: USG5
Source: NASA, NGA. USGS
-------
Figure 25-3. Rutland, Vermont (RUVT) Monitoring Site
to
-------
Figure 25-4. NEI Point Sources Located Within 10 Miles of BURVT and UNVT
WtS'OTnf 73 -
73'trO-W 72-55'CrW 72-WO-W 72'45'0-W 72'40'0-W 72'35'0-W
7^30'0-W 73"25'0-W 73'20'0'W 73*1 PCTW 73MOVW 73'5-Q-W 73'0'0*W 72*55-0^ TrSOmiV
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest-
Legend
@ BURVT UATMP site $ UNVT NATTS site 10 mile radius f H County boundary
Source Category Group (No. of Facilities)
-t< Aircraft Operations (13)
e Electrical Equipment (2)
f$l Institutional - school (1)
® Laboratory (1)
A Military Base/National Security Facility (1)
* Electricity Generation via Combustion (2) M Miscellaneous Manufacturing (1)
•" Gasoline/Diesel Service Station (1)
t HotMixAsphaltPlant(l)
1 Primary Metal Production (1)
P Printing/Publishing (2)
* Industrial Machinery and Equipment (2) n Telecommunications (1)
25-5
-------
Figure 25-5. NEI Point Sources Located Within 10 Miles of RUVT
73'10'OTW 73'5'0'W
73"0'0-W 7Z'55'0"W 72:50'Q-W 72'45YTW
Legend
73-S'O-w nww TJ-WO-W irxrtrvt
Note: Due to facility density and collocation, the total facilities
displayed may not represent all facilities within the area of interest.
RUVT UATMP site 10 mile radius | | County boundary
Source Category Group (No. of Facilities)
•J< Aerospace/Aircraft Manufacturing (2)
41 Aircraft Operations (4)
o Clay Ceramics Manufacturing (1)
tf Hot Mix Asphalt Plant (1)
5 Miscellaneous Coating Manufacturing (1)
Plywood, Particleboard, OSB (1)
B Pulp and Paper Plant/Wood Products (1)
w Woodwork, Furniture, Millwork & Wood Preserving (1)
25-6
-------
Table 25-1. Geographical Information for the Vermont Monitoring Sites
Site
Code
BURVT
RUVT
UNVT
AQS Code
50-007-0014
50-021-0002
50-007-0007
Location
Burlington
Rutland
Underbill
County
Chittenden
Rutland
Chittenden
Micro- or
Metropolitan
Statistical Area
Burlington-South
Burlington, VT
MSA
Rutland, VT MSA
Burlington-South
Burlington, VT
MSA
Latitude
and
Longitude
44.476202,
-73.210383
43.608056,
-72.982778
44.52839,
-72.86884
Land Use
Commercial
Commercial
Forest
Location
Setting
Urban/City
Center
Urban/City
Center
Rural
Additional Ambient Monitoring Information1
CO, NO, NO2, NOX, Carbonyl compounds,
Meteorological parameters, PM10, PM25.
CO, SO2, NO, NO2 ,NOX, Carbonyl compounds,
Meteorological parameters, PM10, PM25.
Haze, Sulfate, SO2, O3, Meteorological parameters,
PM10, PM Coarse, PM25, and PM25 Speciation.
These monitoring sites report additional pollutants to AQS (EPA, 2012h); however, these data are not generated by ERG and are therefore not included in this report.
BOLD ITALICS = EPA-designated NATTS Site.
to
-------
BURVT is located in a municipal parking lot in downtown Burlington near the
intersection of Main Street and South Winooski Avenue. This location is less than 1 mile east of
Burlington Bay on Lake Champlain. The areas to the west are commercial while the areas to the
east are residential, as shown in Figure 25-1. Route 2 (Main Street) and Route 7 (South Willard
Street) intersect two blocks east of the monitoring site and 1-89 runs north-south just over 1 mile
east of the site. Between the two roadways and the interstate lies the University of Vermont.
The UNVT monitoring site is located on the Proctor Maple Research Farm in Underhill,
Vermont, east of the Burlington area. Mount Mansfield, the highest peak in Vermont, lies to the
east in Underhill State Park, less than 3 miles away. The Underhill Artillery Range is a few miles
to the south. Figure 25-2 shows that the area surrounding the site is rural in nature and heavily
forested. This site is intended to serve as a background site for the region for trends assessment,
standards compliance, and long-range transport assessment.
As Figure 25-4 shows, UNVT and BURVT are located approximately 16 miles apart.
Most of the emissions sources are located between these two sites, although closer to BURVT.
The source category with the highest number of emissions sources surrounding these sites is the
aircraft operations source category, which includes airports as well as small runways, heliports,
or landing pads. The sources closest to BURVT are a medical school/hospital, an airport, and
two facilities generating electricity via combustion. The sources closest to UNVT are private
airports.
The RUVT monitoring site is located in Rutland, in central Vermont. The city of Rutland
is in a valley between the Green Mountains to the east and Taconic Mountains to the west. The
monitoring site is located in the courthouse parking lot in downtown Rutland, just north of West
Street. Commercial areas are located to the east and south, while residential areas are located to
the north and west, as shown in Figure 25-3. A railway parallels Route 4 coming into Rutland
from the west, crosses under Route 4, then meanders around a shopping plaza just south of
Route 4. The north junction of Route 4 and Route 7 is approximately 2 miles east of the site.
Figure 25-5 shows that most of the emissions sources near RUVT are located along Route 4 and
Route 7, just south of the monitoring site. The source categories with the highest number of
sources include aircraft operations and aerospace/aircraft manufacturing. The source closest to
RUVT is an aerospace/aircraft manufacturer.
25-8
-------
Table 25-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the Vermont
monitoring sites. Table 25-2 also includes a vehicle registration-to-county population ratio
(vehicles-per-person) for each site. In addition, the population within 10 miles of each site is
presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-level
vehicle registration-to-population ratio to the 10-mile population surrounding each monitoring
site. Table 25-2 also contains annual average daily traffic information. Finally, Table 25-2
presents the daily VMT for Chittenden and Rutland Counties.
Table 25-2. Population, Motor Vehicle, and Traffic Information for the Vermont
Monitoring Sites
Site
BURVT
RUVT
UNVT
Estimated
County
Population1
156,705
61,566
156,705
County-level
Vehicle
Registration2
223,316
118,002
223,316
Vehicles per
Person
(Registration:
Population)
1.43
1.92
1.43
Population
within 10
miles3
116,261
34,336
35,228
Estimated
10-mile
Vehicle
Ownership
165,680
65,811
50,202
Annual
Average
Daily
Traffic4
4,000
7,200
1,200
County-
level Daily
VMT5
4,027,945
1,766,027
4,027,945
1 County-level population estimates reflect data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registrations reflect 2010 data from the Vermont DMV Commissioner's Office (VT DMV,
2010)
310-mile population estimates reflect 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2010 and 2005 data for BURVT and UNVT, respectively, from the
Chittenden County Regional Planning Commission (CCRPC, 2005 and 2010) and 2010 data for RUVT from
Vermont Agency of Transportation (VTrans, 2011)
5 County-level VMT reflects 2010 data from the Vermont Agency of Transportation (VTrans, 2010)
BOLD ITALICS = EPA-designated NATTS Site.
Observations from Table 25-2 include the following:
• The population for Chittenden County is more than twice the population of Rutland
County. The populations for both counties are in the bottom third compared to other
counties with NMP sites. RUVT's 10-mile population is the lowest among the three
Vermont sites, although it is similar to the 10-mile population surrounding UNVT.
• Although similar patterns are shown in the vehicle ownership data, the number of
vehicles registered in each county is significantly higher than the population counts,
leading to rather large vehicle-per-person ratios, including the largest among all NMP
sites (1.92 for RUVT). This indicates that many people own more than one vehicle.
• The traffic volume experienced near RUVT is highest and lowest near UNVT among
the Vermont sites. The traffic estimates near these sites are among the lower traffic
counts for NMP sites. The traffic estimate for BURVT was obtained for South
Winooski Avenue between Main Street and Maple Street; Pleasant Valley Road,
north of Harvey Road for UNVT; and US-4 Business between Merchants Row and
Grove Street for RUVT.
25-9
-------
• VMT for the Vermont counties rank among the lowest compared to other counties
with NMP monitoring sites, with Rutland County fourth lowest and Chittenden
County ninth lowest (where VMT data were available).
25.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
sites in Vermont on sample days, as well as over the course of the year.
25.2.1 Climate Summary
The city of Burlington resides just to the east of Lake Champlain in northwest Vermont.
Lake Champlain has a moderating affect on the city, keeping the city slightly warmer in winter
than it could be given its New England location. The town of Underhill is located to the east of
Burlington but still within the Burlington MSA. The city of Rutland is located 60 miles south of
the Burlington area. Rutland resides within the same climatic division of Vermont as Burlington,
but misses the moderating influences of Lake Champlain. The state of Vermont is affected by
most storm systems that track across the country, producing variable weather and often cloudy
skies. Summers in Vermont are pleasant, with warm days and cool nights, escaping much of the
heat and humidity much of the East Coast experiences. Winters are warmer in the Champlain
Valley region than in other portions of the state but snow is common state-wide. Precipitation is
evenly distributed throughout the year. Average annual winds parallel the valleys, generally from
the south ahead of advancing weather systems, or from the north behind these systems. These
storm systems tend to be moderated somewhat due to the Adirondacks to the west and Green
Mountains to the east (Bair, 1992; NCDC, 2012; NOAA, 2012f).
25.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather stations nearest the Vermont
monitoring sites were retrieved for 2010 (NCDC, 2010). The closest weather station to BURVT
is located at Burlington International Airport; nearest RUVT is Rutland State Airport; and
nearest UNVT is Morrisville-Stowe State Airport (WBANs 14742, 94737, and 54771,
respectively). Additional information about these weather stations, such as the distance between
the sites and the weather stations, is provided in Table 25-3. These data were used to determine
how meteorological conditions on sample days vary from normal conditions throughout the year.
25-10
-------
Table 25-3. Average Meteorological Conditions near the Vermont Monitoring Sites
Closest NWS Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar
Wind
Speed
(kt)
Burlington, Vermont - BURVT
Burlington Intl
Airport
14742
(44.48, -73.16)
2.38
miles
87°
(E)
Sample
Day
2010
57.0
±7.4
56.2
±2.2
48.6
±6.9
48.3
±2.0
37.7
±6.3
38.2
± 1.9
43.4
±6.1
43.5
±1.8
69.1
±3.5
71.1
± 1.2
1013.2
±2.7
1013.0
±0.8
6.5
±1.0
6.0
±0.3
Rutland, Vermont - RUVT
Rutland State Airport
94737
(43.53, -72.95)
5.60
miles
150°
(SSE)
Sample
Day
2010
55.0
±7.7
55.6
±2.1
46.5
±6.9
47.3
±1.9
36.7
±6.5
36.8
±1.9
41.9
±6.2
42.5
±1.8
71.0
±3.1
69.6
±1.2
NA
NA
5.4
±0.8
5.7
±0.3
Underbill, Vermont - UNVT
Morrisville-Stowe
State Airport
54771
(44.53, -72.61)
11.84
miles
78°
(E)
Sample
Day
2010
54.8
±5.4
55.3
+ 2.1
45.7
±4.7
46.1
+ 1.9
36.4
±4.6
36.8
+ 1.9
41.4
±4.3
41.9
+ 1.8
73.0
±2.5
73.0
+ 1.1
1013.5
±1.8
1013.6
+ 0.8
3.1
±0.5
3.3
+ 0.2
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
to
-------
Table 25-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 25-3 is the 95
percent confidence interval for each parameter. As shown in Table 25-3, average meteorological
conditions on sample days at the Vermont monitoring sites were fairly representative of average
weather conditions throughout the year.
25.2.3 Back Trajectory Analysis
Figure 25-6 is the composite back trajectory map for days on which samples were
collected at the BURVT monitoring site in 2010. Included in Figure 25-6 are four back
trajectories per sample day. Figure 25-7 is the corresponding cluster analysis for 2010. Similarly,
Figure 25-8 and 25-10 are the composite back trajectory maps for days on which samples were
collected at UNVT and RUVT and Figures 25-9 and 25-11 are the corresponding cluster
analyses for these two sites. An in-depth description of these maps and how they were generated
is presented in Section 3.5.2.1. For the composite maps, each line represents the 24-hour
trajectory along which a parcel of air traveled toward the monitoring site on a given sample day
and time. For the cluster analyses, each line corresponds to a back trajectory representative of a
given cluster of trajectories. For all maps, each concentric circle around the sites in Figures 25-6
through 25-11 represents 100 miles.
25-12
-------
Figure 25-6. 2010 Composite Back Trajectory Map for BURVT
Figure 25-7. Back Trajectory Cluster Map for BURVT
25-13
-------
Figure 25-8. 2010 Composite Back Trajectory Map for RUVT
Figure 25-9. Back Trajectory Cluster Map for RUVT
25-14
-------
Figure 25-10. 2010 Composite Back Trajectory Map for UNVT
Figure 25-11. Back Trajectory Cluster Map for UNVT
\ '' \
>,
S ;
\
25-15
-------
Observations from Figures 25-6 through 25-11 for the Vermont monitoring sites include
the following:
• The composite back trajectories maps for the Vermont sites are fairly similar to each
other, which is not unexpected given their relatively close proximity to each other.
• The composite back traj ectory maps show that the maj ority of back traj ectories
originated from the southwest, west, northwest, and north of the Vermont monitoring
sites. Note that the BURVT and RUVT monitoring sites sampled on a l-in-12
schedule, yielding roughly half as many sample days for these sites as UNVT.
• For each of the Vermont sites, the farthest away a trajectory originated was off the
North Carolina coast, greater than 700 miles away. However, this is the only
trajectory of similar origin for BURVT and RUVT, while a few originate off the
Virginia coast for UNVT. This trajectory is for the evening of January 25, 2010,
when a strong frontal system was pushing through New England. The average
trajectory length varied from 266 miles for UNVT, 282 miles for BURVT, and 289
miles for RUVT and most trajectories (roughly 80 percent) originated within 400
miles of each site.
• The cluster analyses for the Vermont sites are fairly similar to each other
directionally, although percentage-wise there are some differences. For example, the
cluster analyses for BURVT and RUVT show that roughly one quarter of back
trajectories originated north to northeastward over Maine and eastern Quebec,
Canada. For UNVT, this is also true, but the cluster trajectory is divided into two,
with longer trajectories accounting for 5 percent of the trajectories and shorter
trajectories accounting for 16 percent. The cluster analyses for BURVT and UNVT
show that 28 percent of back trajectories originated northwest to northward over
western Quebec, Canada. For RUVT, several of the back trajectories originating from
the west-northwest and northwest are included with the westward originating cluster
trajectory (44 percent) instead. Thus, trajectories originating from the southwest,
west, northwest, north, and northeast account for 78-79 percent of back trajectories
for UNVT and BURVT and 85 percent of back trajectories for RUVT. Each cluster
analysis also has one southward originating trajectory. For each site, this trajectory
represents southward originating trajectories as well as a sampling of trajectories
originating from other directions and within 100-200 miles of the sites.
25.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather stations at Burlington International Airport (for
BURVT), Morrisville-Stowe State Airport (for UNVT), and Rutland State Airport (for RUVT)
were uploaded into a wind rose software program to produce customized wind roses, as
described in Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals"
positioned around a 16-point compass, and uses different colors to represent wind speeds.
25-16
-------
Figure 25-12 presents three different wind roses for the BURVT monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location. Figures 25-13 and 25-14
present the three wind roses and distance maps for the UNVT and RUVT monitoring sites.
Observations from Figure 25-12 for BURVT include the following:
• The Burlington International Airport weather station is located approximately
2.8 miles east of BURVT.
• The historical wind rose shows that southerly winds are prevalent near BURVT,
accounting for nearly 25 percent of the hourly measurements. Calm winds (< 2 knots)
account for another 18 percent of measurements. Winds from the northwest quadrant,
including north, account for another 25 percent of the wind observations. Winds from
the eastern quadrants are rarely measured.
• The wind patterns shown on the 2010 wind roses are similar to the historical wind
patterns, although slightly fewer southerly winds and slightly more northwesterly to
northerly winds were measured in 2010 and slightly more calm winds (21 percent)
were observed.
• The sample day wind rose shows that wind conditions on sample days were similar to
those experienced throughout 2010, although northwesterly winds accounted for an
even higher percentage of the hourly wind measurements.
25-17
-------
Figure 25-12. Wind Roses for the Burlington International Airport Weather Station
near BURVT
1999-2009 Historical Wind Rose
2010 Wind Rose
Calms: 13.22%
2010 Sample Day Wind Rose
Distance between BURVT and NWS Station
25-18
-------
Figure 25-13. Wind Roses for the Rutland State Airport Weather Station near RUVT
2003-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between RUVT and NWS Station
25-19
-------
Figure 25-14. Wind Roses for the Morrisville-Stowe State Airport Weather Station near
UNVT
1999-2009 Historical Wind Rose
2010 Wind Rose
Calms; 43.13W
2010 Sample Day Wind Rose
Distance between UNVT and NWS Station
NWS
N
4-
25-20
-------
Observations from Figure 25-13 for RUVT include the following:
• The Rutland State Airport weather station is located approximately 5.6 miles south-
southeast of RUVT.
• The historical wind rose shows that east-southeasterly and southeasterly winds were
prevalent near RUVT, as these directions account for over a quarter of the hourly
measurements. Winds from the southwest and northwest quadrants were observed
frequently while winds from the northeast quadrant were almost never observed.
Calm winds were observed for over 17 percent of the hourly measurements.
• The wind patterns shown on the 2010 wind rose are similar to the historical wind
patterns, although a higher percentage of winds from the northwest quadrant and
fewer east-southeasterly winds were observed in 2010. The 2010 calm rate is nearly
identical to the historical calm rate.
• The sample day wind rose shows that wind conditions on sample days were similar to
those experienced throughout 2010.
Observations from Figure 25-14 for UNVT include the following:
• The Morrisville-Stowe Airport weather station is located approximately 12 miles east
of UNVT. Between the site and the weather station lie the Green Mountains.
• The historical wind rose shows that calm winds were prevalent near UNVT, as calm
winds were observed for over 40 percent of the hourly measurements. Winds from the
northwest to north account for another 20 percent of the wind observations greater
than two knots and winds from the south to south-southwest account for another
15 percent of observations.
• The wind patterns shown on the 2010 wind rose are similar to the historical wind
patterns, although a higher percentage of northwesterly to northerly winds and fewer
southerly to south-southwesterly winds were observed in 2010. The calm rate is
slightly higher in 2010.
• The sample day wind rose shows that wind conditions on sample days were similar to
those experienced throughout 2010, although calm winds account for nearly 49
percent of wind measurements.
25.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Vermont monitoring sites in
order to allow analysts and readers to focus on a subset of pollutants through the context of risk.
For each site, each pollutant's preprocessed daily measurement was compared to its associated
risk screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
25-21
-------
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by each monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 25-4 presents the pollutants of interest for the Vermont monitoring sites. The
pollutants that failed at least one screen and contributed to 95 percent of the total failed screens
for each monitoring site are shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of
interest are shaded and/or bolded. BURVT and RUVT sampled for VOC only while UNVT
sampled for VOC, carbonyl compounds, hexavalent chromium, PAH, and metals (PMi0). Note,
however, that carbonyl compounds sampling at UNVT was discontinued in June 2010.
Observations from Table 25-4 include the following:
• A total of eight pollutants, including three NATTS MQO Core Analytes, failed
screens for BURVT. Six pollutants, including the same three NATTS MQO Core
Analytes, failed screens for RUVT. Five pollutants failing screens were the same for
BURVT and RUVT.
• Thirteen pollutants, of which nine are NATTS MQO Core Analytes, failed screens for
UNVT.
• The preliminary risk screening process identified five pollutants of interest for
BURVT (benzene, carbon tetrachloride, 1,3-butadiene, 1,2-dichloroethane, and
ethylbenzene). Three additional pollutants (chloroform, tetrachloroethylene, and
trichloroethylene) were added as pollutants of interest because they are NATTS MQO
Core Analytes, even though these pollutants did not fail any screens. Although vinyl
chloride is also a NATTS MQO Core Analyte, this pollutant was not added because it
was not detected at this site. These pollutants are not shown in Table 25-4.
• The preliminary risk screening process identified four pollutants of interest for RUVT
(benzene, carbon tetrachloride, 1,3-butadiene, and ethylbenzene). Four additional
pollutants (chloroform, tetrachloroethylene, trichloroethylene, and vinyl chloride)
were added as pollutants of interest because they are NATTS MQO Core Analytes,
even though these pollutants did not fail any screens. These pollutants are not shown
in Table 25-4.
• The preliminary risk screening process identified eight pollutants of interest for
UNVT (three VOC, two carbonyl compounds, two metals, and one PAH).
Benzo(a)pyrene and 1,3-butadiene were added to UNVT's pollutants of interest
because they are NATTS MQO Core Analytes, even though they did not contribute to
25-22
-------
95 percent of UNVT's total failed screens. Nine additional pollutants (four metals,
four VOC, and hexavalent chromium) were added as pollutants of interest for UNVT
because they are NATTS MQO Core Analytes, even though these pollutants did not
fail any screens. These pollutants are not shown in Table 25-4.
Benzene, carbon tetrachloride, and 1,3-butadiene were identified as pollutants of
interest for each of the Vermont monitoring sites.
Table 25-4. Risk Screening Results for the Vermont Monitoring Sites
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Burlington, Vermont - BURVT
Benzene
Carbon Tetrachloride
1,3-Butadiene
1 ,2-Dichloroethane
Ethylbenzene
Acrylonitrile
Hexachloro- 1 ,3 -butadiene
Chloroprene
0.13
0.17
0.03
0.038
0.4
0.015
0.045
0.0021
Total
31
30
29
6
4
2
2
1
105
31
30
29
6
31
2
2
1
132
100.00
100.00
100.00
100.00
12.90
100.00
100.00
100.00
79.55
29.52
28.57
27.62
5.71
3.81
1.90
1.90
0.95
29.52
58.10
85.71
91.43
95.24
97.14
99.05
100.00
Rutland, Vermont - RUVT
Benzene
Carbon Tetrachloride
1,3-Butadiene
Ethylbenzene
1 ,2-Dichloroethane
£>-Dichlorobenzene
0.13
0.17
0.03
0.4
0.038
0.091
Total
28
28
27
8
3
1
95
28
28
27
28
o
J
12
126
100.00
100.00
100.00
28.57
100.00
8.33
75.40
29.47
29.47
28.42
8.42
3.16
1.05
29.47
58.95
87.37
95.79
98.95
100.00
Underbill, Vermont - UNVT
Benzene
Carbon Tetrachloride
Formaldehyde
Acet aldehyde
Arsenic (PM10)
1 ,2-Dichloroethane
Manganese (PM10)
Naphthalene
Acrylonitrile
Benzo(a)pyrene
1,3-Butadiene
1 ,2-Dibromoethane
Dichloromethane
0.13
0.17
0.077
0.45
0.00023
0.038
0.005
0.029
0.015
0.00057
0.03
0.0017
7.7
Total
60
57
30
28
23
10
5
5
1
1
1
1
1
223
60
59
30
30
58
10
61
60
1
10
12
1
59
451
100.00
96.61
100.00
93.33
39.66
100.00
8.20
8.33
100.00
10.00
8.33
100.00
1.69
49.45
26.91
25.56
13.45
12.56
10.31
4.48
2.24
2.24
0.45
0.45
0.45
0.45
0.45
26.91
52.47
65.92
78.48
88.79
93.27
95.52
97.76
98.21
98.65
99.10
99.55
100.00
25-23
-------
25.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Vermont monitoring sites. Concentration averages are provided for the pollutants of
interest, where applicable. Concentration averages for select pollutants are also presented
graphically for each site, where applicable, to illustrate how each site's concentrations compare
to the program-level averages. In addition, concentration averages for select pollutants are
presented from previous years of sampling in order to characterize concentration trends at the
site, where applicable. Additional site-specific statistical summaries are provided in Appendices
J, L, M, N, and O.
25.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for each Vermont site, as described in Section 3.1. The quarterly average of a particular
pollutant is simply the average concentration of the preprocessed daily measurements over a
given calendar quarter. Quarterly average concentrations include the substitution of zeros for all
non-detects. A site must have a minimum of 75 percent valid samples of the total number of
samples possible within a given quarter for a quarterly average to be calculated. An annual
average includes all measured detections and substituted zeros for non-detects for the entire year
of sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for the Vermont
monitoring sites are presented in Table 25-5, where applicable. Note that concentrations of the
PAH, metals, and hexavalent chromium for UNVT are presented in ng/m3 for ease of viewing.
Also note that if a pollutant was not detected in a given calendar quarter, the quarterly average
simply reflects "0" because only zeros substituted for non-detects were factored into the
quarterly average concentration.
25-24
-------
Table 25-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Vermont Monitoring Sites
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(Ug/m3)
2nd
Quarter
Average
(Ug/m3)
3rd
Quarter
Average
(Ug/m3)
4th
Quarter
Average
(Ug/m3)
Annual
Average
(Ug/m3)
Burlington, Vermont - BURVT
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1,2-Dichloroethane
Ethylbenzene
Tetrachloroethylene
Trichloroethylene
31/31
29/31
30/31
21/31
6/31
31/31
27/31
1/31
0.79
±0.13
0.07
±0.03
0.47
±0.18
0.06
±0.04
0.02
±0.03
0.18
±0.04
0.07
±0.05
0
0.94
±0.19
0.11
±0.03
0.59
±0.16
0.09
±0.04
0.03
±0.03
0.33
±0.11
0.12
±0.03
0
0.78
±0.22
0.08
±0.03
0.60
±0.10
0.12
±0.04
0
0.33
±0.11
0.10
±0.05
0.01
±0.03
0.89
±0.34
0.09
±0.02
0.52
±0.10
0.04
±0.04
0
0.22
±0.06
0.10
±0.08
0
0.85
±0.11
0.09
±0.01
0.54
±0.06
0.08
±0.02
0.01
±0.01
0.26
±0.04
0.09
±0.03
O.01
±0.01
Rutland, Vermont - RUVT |
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
Ethylbenzene
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
28/28
27/28
28/28
18/28
28/28
26/28
2/28
1/28
0.96
±0.45
0.12
±0.13
0.66
±0.16
0.05
±0.03
0.20
±0.11
0.20
±0.15
0
0.01
±0.02
0.86
±0.24
0.09
±0.04
0.63
±0.09
0.09
±0.06
0.35
±0.13
0.19
±0.05
0.01
±0.01
0
0.83
±0.19
0.08
±0.02
0.62
±0.10
0.10
±0.06
0.39
±0.10
0.17
±0.04
0.01
±0.02
0
1.27
±0.71
0.18
±0.15
0.58
±0.13
0.05
±0.05
0.28
±0.15
0.15
±0.09
0
0
0.98 1
±0.20
0.12
±0.05
0.62
±0.05
0.07
±0.02
0.31
±0.06
0.18
±0.04
0.01
±0.01
O.01
±O.01
Underbill, Vermont - UNVT
Acetaldehyde
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
30/30
60/60
12/60
59/60
35/60
0.66
±0.09
0.51
±0.08
0.01
±0.01
0.60
±0.13
0.05
±0.02
0.98
±0.22
0.35
±0.06
0.01
±0.01
0.58
±0.10
0.07
±0.02
NA
0.38
±0.15
0
0.64
±0.05
0.06
±0.02
NA
0.44
±0.10
0.01
±0.01
0.54
±0.09
0.01
±0.01
NA
0.42
±0.05
0.01
±0.01
0.59
±0.05
0.05
±0.01
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of
viewing.
25-25
-------
Table 25-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Vermont Monitoring Sites (Continued)
Pollutant
1,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Arsenic (PM10)a
Benzo(a)pyrene a
Beryllium (PM10)a
Cadmium (PM10)a
Hexavalent Chromium3
Lead(PM10)a
Manganese (PM10) a
Naphthalene a
Nickel (PM10)a
#of
Measured
Detections
vs. # of
Samples
10/60
30/30
25/60
1/60
4/60
58/61
10/60
23/61
61/61
12/58
61/61
61/61
60/60
61/61
1st
Quarter
Average
(Ug/m3)
0.02
±0.02
1.35
±0.15
0.02
±0.01
0
0.01
±0.01
0.18
±0.07
0.14
±0.18
0.01
±0.01
0.07
±0.02
0.01
±0.01
1.47
±0.36
1.55
±0.71
22.38
±7.37
0.26
±0.10
2nd
Quarter
Average
(Ug/m3)
0.03
±0.02
2.23
±0.51
0.04
±0.02
O.01
±O.01
0.01
±0.01
0.28
±0.10
0
0.01
±0.01
0.06
±0.01
0.01
±0.01
1.46
±0.37
3.12
±0.95
7.10
±1.98
0.26
±0.06
3rd
Quarter
Average
(Ug/m3)
0
NA
0.02
±0.01
0
0
0.20
±0.10
0
0.01
±0.01
0.05
±0.02
0.01
±0.01
1.28
±0.53
1.96
±0.59
7.64
±1.75
0.24
±0.05
4th
Quarter
Average
(Ug/m3)
0
NA
0.02
±0.02
0
0
0.19
±0.07
O.01
±O.01
0.01
±0.01
0.05
±0.01
0.01
±0.01
1.59
±1.17
1.11
±0.48
15.03
±3.56
0.29
±0.10
Annual
Average
(Ug/m3)
0.01
±0.01
NA
0.02
±0.01
O.01
±O.01
0.01
±0.01
0.21
±0.04
0.04
±0.04
0.01
±0.01
0.06
±0.01
0.01
±0.01
1.45
±0.32
1.94
±0.38
13.17
±2.59
0.26
±0.04
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of
viewing.
Observations for BURVT and RUVT from Table 25-5 include the following:
• BURVT and RUVT sampled VOC on a l-in-12 day schedule.
• For both sites, the pollutants with the highest annual average concentrations are
benzene, carbon tetrachloride, and ethylbenzene, although all of the annual average
concentrations for the pollutants of interest for both sites are less than 1 |ig/m3.
• The fourth quarter benzene average for RUVT is higher than the other quarterly
averages and has a relatively large confidence interval associated with it.
Concentrations of benzene measured at RUVT ranged from 0.455 |ig/m3 to
2.91 |ig/m3, with the maximum benzene concentration measured on
December 28, 2010. The next highest concentration was measured on
January 14, 2010 (2.18 |ig/m3). These were the only two concentrations greater than
25-26
-------
2 |ig/m3 measured at this site. The third highest concentration (1.83 |ig/m3) was
measured on November 22, 2010, also in the fourth quarter of 2010. The three highest
1,3-butadiene concentrations were also measured on these three days.
• Chloroform concentrations measured at BURVT and RUVT appear higher during the
warmer months of the year, although the confidence intervals indicate that the
differences are not statistically significant.
• All of the measured detections of 1,2-dichloroethane for BURVT were measured
during the first and second quarters of 2010 with no measured detections after
May 14, 2010. This is similar to other NMP sites sampling VOC (including RUVT,
for which this pollutant is not a pollutant of interest).
• Trichloroethylene was detected only once at BURVT and twice at RUVT. Vinyl
chloride was detected once at RUVT.
Observations for UNVT from Table 25-5 include the following:
• UNVT sampled VOC, carbonyl compounds, PAH, PMi0 metals, and hexavalent
chromium on a l-in-6 day schedule.
• Carbonyl compound sampling was discontinued at UNVT at the end of the June
2010; thus, annual average concentrations were not calculated for these pollutants.
However, Appendix L provides the pollutant-specific average concentration for all
valid samples collected at UNVT over the entire sample period.
• For the pollutants of interest for which annual average concentrations could be
calculated, carbon tetrachloride, benzene, and chloroform are the pollutants with the
highest annual average concentrations. Similar to BURVT and RUVT, all of the
annual average concentrations for the pollutants of interest for UNVT are less than
1 |ig/m3.
• Of the metals, manganese has the highest annual average concentration, followed by
lead and nickel.
• The first quarter benzene average for UNVT is higher than the other quarterly
averages, although the differences are not statistically significant. While the
maximum concentration of benzene was measured on July 7, 2010 (1.46 |ig/m3) and
is the only benzene measurement greater than 1 |ig/m3, most of the higher benzene
measurements were collected during the first and fourth quarters of 2010. Of the 13
measurements greater than 0.5 |ig/m3, six were measured during the first quarter, one
each in the second and third quarters, and five were measured during the fourth
quarter.
• Chloroform exhibits the same tendency at UNVT as it did at BURVT and UNVT,
with the higher quarterly averages calculated for the second and third quarters of the
year.
25-27
-------
• Similar to BURVT and RUVT, as well as other NMP sites sampling VOC, there were
no measured detections of 1,2-dichloroethane measured at UNVT after May 14, 2010.
• Trichloroethylene was detected only once and vinyl chloride was detected only four
times at UNVT (three times in the first quarter and once in the second).
• The first quarter benzo(a)pyrene average for UNVT is higher than the other quarterly
averages and has a relatively large confidence interval associated with it. The
maximum concentration of this pollutant (1.26 ng/m3) was measured on
January 14, 2010 and is nearly three times higher than the next highest concentration
(0.451 ng/m3, measured on March 27, 2010). Of the 10 measured detections of
benzo(a)pyrene, nine were measured during the first quarter and one was measured
during the fourth quarter, with no measured detections in the second and third
quarters of 2010.
• The fourth quarter lead average for UNVT has a relatively large confidence interval
associated with it, compared to the other quarterly averages. The maximum
concentration of this pollutant (9.20 ng/m3) was measured on October 11, 2010 and is
nearly three times higher than the next highest concentration (3.36 ng/m3, measured
on August 30, 2010). These were the only two measurements greater than 3 ng/m3
among the 61 measured detections for UNVT. The second highest lead concentration
measured during the fourth quarter was measured on December 10, 2010
(2.41 ng/m3).
• The second quarter manganese average for UNVT is higher than the other quarterly
averages (although not statistically so). All five concentrations of manganese greater
than 5 ng/m3 were measured during the second quarter of 2010.
• Concentrations of naphthalene at UNVT tended to be higher during the colder months
of the year. The three highest concentrations were all measured in January 2010 and
of the 17 highest concentrations (those greater than 15 ng/m3), eight were measured
during the first quarter of the year, one each during the second and third quarter, and
seven were measured during the fourth quarter of 2010.
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for the Vermont
monitoring sites from those tables include the following:
• RUVT appears three times in Table 4-9. RUVT has the fourth, eighth, and tenth
highest annual average concentrations of vinyl chloride, 1,3-butaidene, and
tetrachloroethylene (respectively) among NMP sites sampling VOC.
• UNVT has the eighth highest annual average concentration of vinyl chloride.
However, the annual average concentrations of this pollutant for both RUVT and
UNVT, as well as all NMP sites sampling this pollutant, are less than 0.01 |ig/m3.
25-28
-------
• Because only nine NMP sites sampled PMio metals, all nine sites appear in
Table 4-12. UNVT ranks ninth for all six program-wide metal pollutants of interest
and last for hexavalent chromium.
• Compared to other NMP sites, UNVT has some of the lowest annual average
concentrations for each of the program-wide pollutants of interest.
25.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for benzene and 1,3-butadiene
were created for BURVT, RUVT, and UNVT. Box plots were also created for arsenic,
benzo(a)pyrene, hexavalent chromium, manganese, and naphthalene for UNVT. Figures 25-15
through 25-21 overlay the sites' minimum, annual average, and maximum concentrations onto
the program-level minimum, first quartile, average, median, third quartile, and maximum
concentrations, as described in Section 3.5.3.
Figure 25-15. Program vs. Site-Specific Average Arsenic (PMio) Concentration
mt
2 2.5 3
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
25-29
-------
Figure 25-16. Program vs. Site-Specific Average Benzene Concentration
3 4
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 25-17. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
E
| Program Max Coi
incentration = 42.7 ng/m3 j
0.8 1 1.2
Concentration (ng/m3)
1.4 1.6
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
25-30
-------
Figure 25-18. Program vs. Site-Specific Average 1,3-Butadiene Concentration
I I I I I I I I
0.4 0.5 0.6
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 25-19. Program vs. Site-Specific Average Hexavalent Chromium Concentration
I
| Program Max Concentration = 3.51 ng/m3 j
k
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
25-31
-------
Figure 25-20. Program vs. Site-Specific Average Manganese (PMi0) Concentration
1
20 40 60 80 100 120 140 160 180 200
Concentration (ng/m3)
Program
Site:
: IstQuartile
Site Average
0
2ndQuartile SrdQuartile 4thQuartile Ave
• n
i i i i
Site Minimum/Maximum
>rage
Figure 25-21. Program vs. Site-Specific Average Naphthalene Concentration
600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
o
Observations from Figures 25-15 through 25-21 include the following:
• Figure 25-15 shows that UNVT's annual average arsenic (PMio) concentration is
well below the program-level average and median concentrations for arsenic
(PMio) as well as the program-level first quartile (25th percentile). The annual
average concentration of arsenic for UNVT is the lowest annual average
concentration among NMP sites sampling this pollutant. A few non-detects of
arsenic were measured at UNVT.
• Figure 25-16 for benzene shows all three Vermont sites. This figure shows that
the annual average concentration of benzene is highest for RUVT and lowest for
UNVT and that all three annual averages are less than the program-level average
benzene concentration. Figure 25-16 also shows that UNVT's annual average
benzene concentration is below the program-level average, median, and first
quartile (25th percentile) concentrations (and the lowest among all NMP sites
sampling benzene). The range of benzene measurements is smallest for UNVT
and largest for RUVT, although there were no non-detects of benzene measured at
the Vermont sites.
25-32
-------
• Figure 25-17 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
UNVT is less than the program-level average concentration. Although the
maximum concentration measured at UNVT is well below the maximum
concentration measured across the program, the January 14, 2010 measurement
(1.26 ng/m3) is the fifth highest concentration of this pollutant measured among
NMP sites sampling PAH. However, more than 80 percent of the measurements at
UNVT were non-detects.
• Figure 25-18 for 1,3-butadiene also shows all three sites. The annual average
concentration is highest for RUVT and lowest for UNVT, among the Vermont
sites. The annual averages for BURVT and RUVT are greater than the program-
level average concentration while the annual average for UNVT is the lowest of
all NMP sites sampling this pollutant. Although the maximum concentration
measured at RUVT is well below the maximum concentration across the program,
the December 28, 2010 measurement (0.514 |ig/m3) is the eleventh highest
1,3-butadiene concentration across the program (for Method TO-15). A single
non-detects of 1,3-butadiene was measured at RUVT, two were measured at
BURVT, and 80 percent of the measurements were non-detects for UNVT.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 25-19 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 25-19 shows the annual average concentration of hexavalent chromium for
UNVT is less than both the program-level average and median concentrations.
Similar to 1,3-butadiene, the annual average concentration for UNVT is the
lowest annual average hexavalent chromium concentration among NMP sites
sampling this pollutant. Nearly 80 percent of the measurements of hexavalent
chromium were non-detects for UNVT.
• Figure 25-20 shows the annual average concentration of manganese (PMio) for
UNVT is less than the program-level average, median, and first quartile
(25th percentile) concentrations for this pollutant. The annual average
concentration of manganese for UNVT is the lowest annual average concentration
among NMP sites sampling this pollutant, even though there were no non-detects
of manganese measured at UNVT.
• Figure 25-21 shows that the annual naphthalene average for UNVT is also less
than the program-level average, median, and first quartile (25th percentile)
concentrations. The maximum naphthalene concentration measured at UNVT is
less than the program-level median concentration. UNVT's annual average
25-33
-------
naphthalene concentration is the lowest annual average concentration of this
pollutant among NMP sites sampling PAH (even though there were no non-
detects of naphthalene measured at UNVT).
• Recall that annual averages could not be calculated for formaldehyde and
acetaldehyde, as discussed in Section 25.4.1.
25.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. UNVT has sampled hexavalent chromium under the NMP since 2005. Thus,
Figure 25-22 presents the 3-year rolling statistical metrics for hexavalent chromium for UNVT.
The statistical metrics presented for calculating trends include the substitution of zeros for non-
detects.
Observations from Figure 25-22 for hexavalent chromium measurements at UNVT
include the following:
• The maximum hexavalent chromium concentration was measured at UNVT on
June 16, 2006 (0.399 ng/m3). The next highest hexavalent chromium concentration
was measured on April 22, 2005 (0.101 ng/m3). All other measurements of this
pollutant are less than 0.1 ng/m3.
• The rolling average concentration has decreased since the onset of sampling.
However, the confidence intervals calculated for the first two 3-year periods are very
large due to the presence of outliers. The 95th percentile exhibits a similar decreasing
trend as the rolling average.
• For all time frames shown, the minimum, 5th percentile, and median concentrations
are zero, indicating that at least 50 percent of the measurements are non-detects. The
number of non-detects has varied over the years of sampling, from as low as 63
percent in 2006 to as high as 95 percent in 2009.
25-34
-------
Figure 25-22. Three-Year Rolling Statistical Metrics for Hexavalent Chromium
Concentrations Measured at UNVT
— Minimum
— Maximum
• 95thPercentile
25.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the
Vermont monitoring sites. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
25.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Vermont monitoring sites to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
for each site were compared to the acute MRL; the quarterly averages were compared to the
intermediate MRL; and the annual averages were compared to the chronic MRL.
25-35
-------
None of the measured detections or time-period average concentrations of the pollutants
of interest for the Vermont monitoring sites were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
the Vermont monitoring sites.
25.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Vermont monitoring sites and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 25-6, where applicable.
Table 25-6. Cancer and Noncancer Surrogate Risk Approximations for the Vermont
Monitoring Sites
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(jig/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Burlington, Vermont - BURVT
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Ethylbenzene
Tetrachloroethylene
Trichloroethylene
0.0000078
0.00003
0.000006
0.000026
0.0000025
2.6E-07
0.0000048
0.03
0.002
0.1
0.098
2.4
1
0.04
0.002
31/31
29/31
30/31
21/31
6/31
31/31
27/31
1/31
0.85
±0.11
0.09
±0.01
0.54
±0.06
0.08
±0.02
0.01
±0.01
0.26
±0.04
0.09
±0.03
<0.01
±0.01
6.60
2.64
3.26
0.36
0.66
0.02
0.02
0.03
0.04
0.01
0.01
<0.01
0.01
O.01
0.01
NA = Not available due to the criteria for calculating an annual average.
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 25-5.
25-36
-------
Table 25-6. Cancer and Noncancer Surrogate Risk Approximations for the Vermont
Monitoring Sites (Continued)
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Rutland, Vermont - RUVT
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
Ethylbenzene
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
0.0000078
0.00003
0.000006
0.0000025
2.6E-07
0.0000048
0.0000088
0.03
0.002
0.1
0.098
1
0.04
0.002
0.1
28/28
27/28
28/28
18/28
28/28
26/28
2/28
1/28
0.98
±0.20
0.12
±0.05
0.62
±0.05
0.07
±0.02
0.31
±0.06
0.18
±0.04
<0.01
±0.01
0.01
±0.01
7.63
3.54
3.73
0.76
0.05
0.02
0.02
0.03
0.06
0.01
O.01
0.01
O.01
0.01
0.01
Underbill, Vermont - UNVT
Acetaldehyde
Arsenic (PM10)a
Benzene
Benzo(a)pyrene a
Beryllium (PM10) a
1,3 -Butadiene
Cadmium (PM10)a
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Formaldehyde
Hexavalent Chromium3
0.0000022
0.0043
0.0000078
0.00176
0.0024
0.00003
0.0018
0.000006
0.000026
0.000013
0.012
0.009
0.000015
0.03
0.00002
0.002
0.00001
0.1
0.098
2.4
0.0098
0.0001
30/30
58/61
60/60
10/60
23/61
12/60
61/61
59/60
35/60
10/60
30/30
12/58
NA
O.01
±0.01
0.42
±0.05
O.01
±O.01
0.01
±0.01
O.01
±O.01
0.01
±0.01
0.59
±0.05
0.05
±0.01
0.01
±0.01
NA
O.01
±O.01
NA
0.91
3.25
0.06
0.01
0.12
0.10
3.55
0.31
NA
0.04
NA
0.01
0.01
0.01
O.01
0.01
0.01
0.01
O.01
NA
O.01
NA = Not available due to the criteria for calculating an annual average.
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 25-5.
25-37
-------
Table 25-6. Cancer and Noncancer Surrogate Risk Approximations for the Vermont
Monitoring Sites (Continued)
Pollutant
Lead(PM10)a
Manganese (PM10)a
Naphthalene a
Nickel (PM10)a
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
Cancer
URE
(Hg/m3)1
0.000034
0.00048
2.6E-07
0.0000048
0.0000088
Noncancer
RfC
(mg/m3)
0.00015
0.00005
0.003
0.00009
0.04
0.002
0.1
#of
Measured
Detections
vs. # of
Samples
61/61
61/61
60/60
61/61
25/60
1/60
4/60
Annual
Average
(Hg/m3)
0.01
±0.01
O.01
±0.01
0.01
±0.01
O.01
±0.01
0.02
±0.01
O.01
±0.01
0.01
±0.01
Cancer Risk
Approximation
(in-a-million)
0.45
0.13
0.01
O.01
0.01
Noncancer
Risk
Approximation
(HQ)
0.01
0.04
0.01
O.01
0.01
O.01
0.01
NA = Not available due to the criteria for calculating an annual average.
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 25-5.
Observations from Table 25-6 include the following:
• For BURVT, benzene and carbon tetrachloride have the highest annual average
concentrations. These two pollutants also have the highest cancer risk approximations
for BURVT (6.60 in-a-million and 3.26 in-a-million, respectively).
• Similar to BURVT, benzene and carbon tetrachloride have the highest annual average
concentrations for RUVT. These two pollutants also have the highest cancer risk
approximations for RUVT (7.63 in-a-million and 3.73 in-a-million, respectively).
• Carbon tetrachloride and benzene have the highest annual average concentrations for
UNVT. These two pollutants also have the highest cancer risk approximations for
UNVT (3.55 in-a-million and 3.25 in-a-million, respectively).
• The noncancer risk approximations for the pollutants of interest for all three Vermont
sites are well below the level of concern, indicating virtually no noncancer health
risks attributable to these pollutants.
25-38
-------
25.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 25-7 and 25-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 25-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer risk approximations (in-a-million) for the Vermont monitoring sites, as calculated from
the annual averages. Table 25-8 presents similar information, but identifies the 10 pollutants with
the highest noncancer risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 25-7 and 25-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer surrogate risk approximations based on each site's annual
averages are limited to those pollutants for which each respective site sampled, as discussed in
Section 25.3. As discussed in Section 25.3, UNVT sampled for VOC, carbonyl compounds,
PAH, metals (PMio), and hexavalent chromium; BURVT and RUVT sampled for VOC only. In
addition, the cancer and noncancer surrogate risk approximations are limited to those pollutants
with enough data to meet the criteria for annual averages to be calculated. A more in-depth
discussion of this analysis is provided in Section 3.5.5.3.
25-39
-------
Table 25-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Vermont Monitoring Sites
Top 10 Total Emissions for Pollutants with Cancer
Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Burlington, Vermont (Chittenden County) - BURVT
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Dichloromethane
Naphthalene
POM, Group 2b
POM, Group 6
POM, Group la
117.39
57.01
42.71
30.61
12.99
7.10
6.54
1.69
0.20
0.12
Benzene
Formaldehyde
1,3 -Butadiene
Hexavalent Chromium, PM
POM, Group 3
Arsenic, PM
Naphthalene
POM, Group 2b
POM, Group 5a
Ethylbenzene
9.16E-04
7.41E-04
3.90E-04
2.57E-04
2.46E-04
2.44E-04
2.22E-04
1.49E-04
1.24E-04
1.07E-04
Benzene
Carbon Tetrachloride
1,3 -Butadiene
Ethylbenzene
1 ,2-Dichloroethane
Tetrachloroethylene
Trichloroethylene
6.60
3.26
2.64
0.66
0.36
0.02
0.02
Underbill, Vermont (Chittenden County) - UNVT
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Dichloromethane
Naphthalene
POM, Group 2b
POM, Group 6
POM, Group la
117.39
57.01
42.71
30.61
12.99
7.10
6.54
1.69
0.20
0.12
Benzene
Formaldehyde
1,3 -Butadiene
Hexavalent Chromium, PM
POM, Group 3
Arsenic, PM
Naphthalene
POM, Group 2b
POM, Group 5a
Ethylbenzene
9.16E-04
7.41E-04
3.90E-04
2.57E-04
2.46E-04
2.44E-04
2.22E-04
1.49E-04
1.24E-04
1.07E-04
Carbon Tetrachloride
Benzene
Arsenic
Naphthalene
1 ,2-Dichloroethane
Nickel
1,3 -Butadiene
Cadmium
Benzo(a)pyrene
Hexavalent Chromium
3.55
3.25
0.91
0.45
0.31
0.13
0.12
0.10
0.06
0.04
to
v\
J±
o
-------
Table 25-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Vermont Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with Cancer
Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Rutland, Vermont (Rutland County) - RUVT
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
POM, Group 2b
Dichloromethane
POM, Group 6
POM, Group la
54.68
24.42
20.01
14.97
6.47
3.14
0.93
0.47
0.12
0.06
Benzene
Formaldehyde
1,3 -Butadiene
POM, Group 3
Hexavalent Chromium, PM
Naphthalene
POM, Group 2b
POM, Group 5a
Ethylbenzene
Acetaldehyde
4.26E-04
3.17E-04
1.94E-04
1.38E-04
1.24E-04
1.07E-04
8.17E-05
6.57E-05
5.00E-05
3.29E-05
Benzene
Carbon Tetrachloride
1,3 -Butadiene
Ethylbenzene
Tetrachloroethylene
Trichloroethylene
Vinyl Chloride
7.63
3.73
3.54
0.76
0.05
0.02
0.02
to
-k
-------
Table 25-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Vermont Monitoring Sites
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Noncancer
Pollutant Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Burlington, Vermont (Chittenden County) - BURVT
Toluene
Xylenes
Benzene
Methanol
Formaldehyde
Ethylbenzene
Hexane
Hydrochloric acid
Acetaldehyde
1,3 -Butadiene
221.14
170.08
117.39
88.67
57.01
42.71
41.71
41.61
30.61
12.99
Acrolein
Manganese, PM
Chlorine
1,3 -Butadiene
Formaldehyde
Benzene
Arsenic, PM
Acetaldehyde
Cyanide Compounds, gas
Naphthalene
552,694.88
67,632.70
13,450.30
6,493.30
5,817.27
3,912.93
3,782.88
3,400.90
2,550.52
2,180.85
1,3 -Butadiene
Benzene
Carbon Tetrachloride
Tetrachloroethylene
Trichloroethylene
Chloroform
Ethylbenzene
1 ,2-Dichloroethane
0.04
0.03
0.01
0.01
<0.01
<0.01
<0.01
<0.01
Underbill, Vermont (Chittenden County) - UNVT
Toluene
Xylenes
Benzene
Methanol
Formaldehyde
Ethylbenzene
Hexane
Hydrochloric acid
Acetaldehyde
1,3 -Butadiene
221.14
170.08
117.39
88.67
57.01
42.71
41.71
41.61
30.61
12.99
Acrolein
Manganese, PM
Chlorine
1,3 -Butadiene
Formaldehyde
Benzene
Arsenic, PM
Acetaldehyde
Cyanide Compounds, gas
Naphthalene
552,694.88
67,632.70
13,450.30
6,493.30
5,817.27
3,912.93
3,782.88
3,400.90
2,550.52
2,180.85
Manganese
Arsenic
Benzene
Lead
Carbon Tetrachloride
Cadmium
Naphthalene
Nickel
1,3 -Butadiene
Tetrachloroethylene
0.04
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
to
v\
J±
to
-------
Table 25-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Vermont Monitoring Sites (Continued)
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Noncancer
Pollutant Toxicity Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Rutland, Vermont (Rutland County) - RUVT
Toluene
Xylenes
Benzene
Methanol
Formaldehyde
Ethylbenzene
Hexane
Acetaldehyde
1,3 -Butadiene
Styrene
121.10
77.74
54.68
37.29
24.42
20.01
17.83
14.97
6.47
5.25
Acrolein
1,3 -Butadiene
Formaldehyde
Cyanide Compounds, gas
Benzene
Acetaldehyde
Naphthalene
Xylenes
Arsenic, PM
Lead, PM
74,401.74
3,234.62
2,492.15
2,311.36
1,822.62
1,663.70
1,045.42
777.44
462.98
399.93
1,3 -Butadiene
Benzene
Carbon Tetrachloride
Tetrachloroethylene
Trichloroethylene
Chloroform
Ethylbenzene
Vinyl Chloride
0.06
0.03
0.01
0.01
<0.01
<0.01
<0.01
<0.01
to
-------
Observations from Table 25-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in Chittenden and Rutland Counties, although the emissions were nearly
twice as high in Chittenden County than in Rutland County.
• Benzene and formaldehyde are also the pollutants with the highest toxicity-weighted
emissions (of the pollutants with cancer UREs) for both counties, followed by
1,3-butadiene.
• Six of the highest emitted pollutants also have the highest toxicity-weighted
emissions for Chittenden County while seven of the highest emitted pollutants also
have the highest toxicity-weighted emissions for Rutland County.
• Benzene and carbon tetrachloride have the highest cancer risk approximations for all
three sites. Benzene topped both emissions-based lists for both counties, while carbon
tetrachloride appeared on neither emissions-based list for either county. Ethylbenzene
and 1,3-butadiene also appear on all three lists for BURVT and RUVT. While
1,3-butadiene appears on all three lists for UNVT, ethylbenzene is not a pollutant of
interest for this site.
• Among UNVT's non-VOC pollutants of interest, naphthalene appears on all three
lists. Three additional pollutants, hexavalent chromium, arsenic, and benzo(a)pyrene,
are among the pollutants with the highest cancer risk approximations for UNVT and
are among the pollutants with the highest toxicity-weighted emissions. Note that
benzo(a)pyrene is part of POM, Group 5a. None of these three pollutants are among
the highest emitted in Chittenden County.
• POM, Group 2b is the eighth highest emitted "pollutant" in Chittenden County and
ranks eighth for toxicity-weighted emissions. POM, Group 2b includes several PAH
sampled for at UNVT including acenaphthylene, fluoranthene, and perylene. None of
the PAH included in POM, Group 2b failed screens for UNVT. POM, Groups la, 3,
and 6 also appear in Table 25-7, but only POM, Group 6 includes PAH sampled for at
UNVT (benzo(a)anthracene, for example), but none of these pollutants failed screens.
Observations from Table 25-8 include the following:
• Toluene, xylenes, and benzene are the highest emitted pollutants with noncancer
RfCs in Chittenden and Rutland Counties.
• Acrolein is the pollutant with the highest toxicity-weighted emissions (of the
pollutants with noncancer RfCs) for Chittenden and Rutland Counties. Although
acrolein was sampled for at all three sites, this pollutant was excluded from the
pollutants of interest designation, and thus subsequent risk screening evaluations, due
to questions about the consistency and reliability of the measurements, as discussed in
Section 3.2.
25-44
-------
• Four of the highest emitted pollutants for Chittenden County also have the highest
toxi city-weighted emissions while four of the highest emitted pollutants for Rutland
County also have the highest toxicity-weighted emissions.
• Although very low, 1,3-butadiene, benzene, and carbon tetrachloride have the highest
noncancer risk approximations for BURVT and RUVT. While benzene and
1,3-butadiene appear on both emissions-based lists, carbon tetrachloride appears on
neither emissions-based list.
• Although very low, manganese and arsenic have the highest noncancer risk
approximations for UNVT. While these pollutants rank second and seventh among
the toxicity-weighted emissions for Chittenden County, respectively, neither pollutant
appears among the highest emitted.
25.6 Summary of the 2010 Monitoring Data for the Vermont Monitoring Sites
Results from several of the data treatments described in this section include the
following:
»«» A total of eight pollutants failed screens for BURVT; six pollutants failed screens for
RUVT; and 13 pollutants failed screens for UNVT.
»«» Benzene and carbon tetrachloride have the highest annual average concentrations
among the pollutants of interest for BURVT and RUVT while carbon tetrachloride 's
annual average concentration for UNVT was greater than benzene's annual average
concentration. None of the annual average concentrations for the pollutants of
interest for the Vermont sites were greater than 1 jug/m3.
»«» The annual average concentrations for several of UNVT's pollutants of interest were
the lowest annual averages among allNMP sites sampling those pollutants.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest, where they could be calculated,
were greater than their associatedMRL noncancer health risk benchmarks.
25-45
-------
26.0 Site in Virginia
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in Virginia, and integrates these concentrations with
emissions, meteorological, and risk information. Data generated by sources other than ERG are
not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
26.1 Site Characterization
This section characterizes the Virginia monitoring site by providing geographical and
physical information about the location of the site and the surrounding area. This information is
provided to give the reader insight regarding factors that may influence the air quality near the
site and assist in the interpretation of the ambient monitoring measurements.
The RIVA monitoring site is located just outside the Richmond, Virginia city limits.
Figure 26-1 is a composite satellite image retrieved from ArcGIS Explorer showing the
monitoring site in its urban location. Figure 26-2 identifies point source emissions locations by
source category, as reported in the 2008 NEI for point sources. Note that only sources within
10 miles of the site are included in the facility counts provided in Figure 26-2. Thus, sources
outside the 10-mile radius have been grayed out, but are visible on the map to show emissions
sources outside the 10-mile boundary. A 10-mile boundary was chosen to give the reader an
indication of which emissions sources and emissions source categories could potentially have a
direct effect on the air quality at the monitoring site. Further, this boundary provides both the
proximity of emissions sources to the monitoring site as well as the quantity of such sources
within a given distance of the site. Table 26-1 describes the area surrounding the monitoring site
by providing supplemental geographical information such as land use, location setting, and
locational coordinates.
26-1
-------
Figure 26-1. Richmond, Virginia (RIVA) Monitoring Site
to
ON
to
-------
Figure 26-2. NEI Point Sources Located Within 10 Miles of RIVA
Legend
RIVA NATTS site
7r3Q'fl"W 77c25'
-------
Table 26-1. Geographical Information for the Virginia Monitoring Site
Site
Code
RIVA
AQS Code
51-087-0014
Location
Richmond
County
Henrico
Micro- or
Metropolitan
Statistical Area
Richmond, VA
MSA
Latitude
and
Longitude
37.55655,
-77.400411
Land Use
Residential
Location
Setting
Suburban
Additional Ambient Monitoring Information1
TSP Metals, SO2, NOy, NO, NO2, NOX, PAMS,
NMOC, VOC, Carbonyl compounds, O3,
Meteorological parameters, PM10, PM10 Metals, PM
Coarse, PM2 5, and PM2 5 Speciation, CO,
Tetrahydrofuran.
BOLD ITALICS = EPA-designated NATTS Site.
to
-k
-------
The RIVA monitoring site is located just northeast of the capital city of Richmond, in
east-central Virginia. The site is located at the MathScience Innovation Center in a residential
area less than 1/4 mile from 1-64. The 1-64 interchange with Mechanicsville Turnpike (360) is
less than 1/2 mile southwest of the site, as shown in Figure 26-1. Beyond the residential areas
surrounding the school property are a golf course to the southeast, a high school to the south (on
the southside of 1-64), and commercial areas to the west. As Figure 26-2 shows, RIVA is located
near several point sources, most of which are located to the southeast and south and within the
city of Richmond. The sources closest to RIVA are a fabricated metal products facility and a
heliport at the Medical College of Virginia. The source categories with the highest number of
emissions sources within 10 miles of RIVA are aircraft operations, which include airports as well
as small runways, heliports, or landing pads; printing and publishing facilities; bulk terminals
and bulk plants; and facilities generating electricity via combustion.
Table 26-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the Virginia
monitoring site. Table 26-2 also includes a vehicle registration-to-county population ratio
(vehicles-per-person). In addition, the population within 10 miles of the site is presented. An
estimate of 10-mile vehicle ownership was calculated by applying the county-level vehicle
registration-to-population ratio to the 10-mile population surrounding the monitoring site.
Table 26-2 also contains annual average daily traffic information. Finally, Table 26-2 presents
the daily VMT for Henrico County.
Table 26-2. Population, Motor Vehicle, and Traffic Information for the Virginia
Monitoring Site
Site
RIVA
Estimated
County
Population1
307,435
County-level
Vehicle
Registration2
347,790
Vehicles per
Person
(Registration:
Population)
1.13
Population
within 10
miles3
460,195
Estimated
10 mile
Vehicle
Ownership
520,602
Annual
Average
Daily
Traffic4
74,000
County-
level
Daily
VMT5
8,260,273
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the Revenue Division of the County of Henrico
(Henrico County, 2011)
310-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2009 data from the Virginia DOT (VA DOT, 2009)
5 County-level VMT reflects 2010 data from the Virginia DOT (VA DOT, 2011)
BOLD ITALICS = EPA-designated NATTS Site.
26-5
-------
Observations from Table 26-2 include the following:
• RIVA's county-level population is in the lower third compared to other counties with
NMP sites. The 10-mile population is in the middle of the range among NMP sites.
• The county-level vehicle ownership and 10-mile vehicle ownership are in the middle
of the ranged compared to other NMP sites.
• The vehicle-per-person ratio is among the higher ratios compared to other NMP sites.
• The traffic volume experienced near RIVA is in the middle of the range compared to
other NMP monitoring sites. The traffic estimate used came from the interchange of
US-360 (Mechanicsville Turnpike) and 1-64.
• The VMT for Henrico County is in the middle of the range compared to other county
with NMP sites.
26.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in Virginia on sample days, as well as over the course of the year.
26.2.1 Climate Summary
The city of Richmond is located in east-central Virginia, east of the Blue Ridge
Mountains and west of the Chesapeake Bay. The James River flows through the west, center, and
south parts of town. Richmond has a modified continental climate. Winters tend to be mild, as
the mountains act as a barrier to cold air and the proximity to the Atlantic Ocean prevents
temperatures from plummeting too low. Summers are warm and humid, also due to these
influences. Precipitation is well distributed throughout the year (Bair, 1992).
26.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest this site were retrieved
for 2010 (NCDC, 2010). The closest weather station is located at Richmond International Airport
(WBAN 13740). Additional information about the Richmond International Airport weather
station, such as the distance between the site and the weather station, is provided in Table 26-3.
These data were used to determine how meteorological conditions on sample days vary from
normal conditions throughout the year.
26-6
-------
Table 26-3. Average Meteorological Conditions near the Virginia Monitoring Site
Closest NWS Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Richmond, Virginia - RTVA
Richmond
International Airport
13740
(37.51, -77.32)
5.16
miles
118°
(ESE)
Sample
Day
2010
69.5
±5.3
70.0
+ 2.1
59.3
±4.9
59.8
+ 1.9
44.8
±4.9
45.1
+ 2.0
51.9
±4.4
52.3
+ 1.7
62.5
±3.3
62.2
+ 1.4
1015.5
±1.8
1015.6
+ 0.7
6.3
±0.8
6.3
+ 0.3
1 Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
to
-------
Table 26-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 26-3 is the
95 percent confidence interval for each parameter. As shown in Table 26-3, average
meteorological conditions on sample days were representative of average weather conditions
throughout the year.
26.2.3 Back Trajectory Analysis
Figure 26-3 is the composite back trajectory map for days on which samples were
collected at the RIVA monitoring site in 2010. Included in Figure 26-3 are four back trajectories
per sample day. Figure 26-4 is the corresponding cluster analysis for 2010. An in-depth
description of these maps and how they were generated is presented in Section 3.5.2.1. For the
composite map, each line represents the 24-hour trajectory along which a parcel of air traveled
toward the monitoring site on a given sample day and time. For the cluster analysis, each line
corresponds to a back trajectory representative of a given cluster of trajectories. For both maps,
each concentric circle around the site in Figures 26-3 and 26-4 represents 100 miles.
Figure 26-3. 2010 Composite Back Trajectory Map for RIVA
26-8
-------
Figure 26-4. Back Trajectory Cluster Map for RIVA
Observations from Figures 26-3 and Figure 26-4 for RIVA include the following:
• Back trajectories originated from a variety of directions near RIVA, although
primarily to the southwest, west, and northwest of the site.
• The 24-hour air shed domain for RIVA was similar in size to many other NMP
monitoring sites. The farthest away a trajectory originated was over Georgian Bay,
northwest of Toronto, Canada, or over 550 miles away. However, the average
trajectory distance is 230 miles and most (91 percent) trajectories originated within
400 miles of the site.
• The cluster analysis shows that the majority of trajectories originated from the
southwest, west, and northwest. Twenty-four percent originated from the southwest to
west over Virginia, North Carolina, and South Carolina. Thirty-five percent of
trajectories originated from the west and northwest, but of different lengths, which is
why they are represented by two different cluster trajectories (12 and 23 percent). The
short trajectory originating due south of RIVA (20 percent) represents trajectories
originating from the southeast and east, but also several trajectories originating within
approximately 100 miles of the site over central Virginia. Another 21 percent of
trajectories originated from the north to northeast of the site.
26-9
-------
26.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at Richmond International Airport near
RIVA were uploaded into a wind rose software program to produce customized wind roses, as
described in Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals"
positioned around a 16-point compass, and uses different colors to represent wind speeds.
Figure 26-5 presents three different wind roses for the RIVA monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location.
Observations from Figure 26-5 for RIVA include the following:
• The Richmond International weather station is located approximately 5.1 miles east-
southeast of RIVA.
• The historical wind rose shows that the most commonly observed wind direction is
north, although winds from the north-northeast, south, south-southwest, and
southwest were also frequently observed. Winds from the southeast quadrant were
observed the least. Calm winds (< 2 knots) were observed for approximately
14 percent of the hourly wind measurements.
• The 2010 wind rose resembles the historical wind rose in some ways but exhibits
deviations as well. While the northerly prominence is the same for both, the 2010
wind rose exhibits a shift from southerly to southwesterly winds to a more even
distribution of winds from the southwest to west to northwest.
• The sample day wind patterns are similar to the full-year wind patterns, indicating
that winds conditions on sample days were representative of those experienced
throughout 2010.
26-10
-------
Figure 26-5. Wind Roses for the Richmond International Airport Weather Station near
RIVA
1999-2009 Historical Wind Rose
2010 Wind Rose
2010 Sample Day Wind Rose
Distance between RIVA and NWS Station
26-11
-------
26.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Virginia monitoring site in
order to allow analysts and readers to focus on a subset of pollutants through the context of risk.
Each pollutant's preprocessed daily measurement was compared to its associated risk screening
value. If the concentration was greater than the risk screening value, then the concentration
"failed the screen." Pollutants of interest are those for which the individual pollutant's total
failed screens contribute to the top 95 percent of the site's total failed screens. In addition, if any
of the NATTS MQO Core Analytes measured by the monitoring site did not meet the pollutant
of interest criteria based on the preliminary risk screening, that pollutant was added to the list of
site-specific pollutants of interest. A more in-depth description of the risk screening process is
presented in Section 3.2.
Table 26-4 presents the pollutants of interest for RIVA. The pollutants that failed at least
one screen and contributed to 95 percent of the total failed screens for the monitoring site are
shaded. NATTS MQO Core Analytes are bolded. Thus, pollutants of interest are shaded and/or
bolded. RIVA sampled for PAH and hexavalent chromium.
Table 26-4. Risk Screening Results for the Virginia Monitoring Site
Pollutant
Screening
Value
(Hg/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Richmond, Virginia - RIVA
Naphthalene
Acenaphthene
Benzo(a)pyrene
Fluorene
0.029
0.011
0.00057
0.011
Total
58
1
1
1
61
60
60
23
60
203
96.67
1.67
4.35
1.67
30.05
95.08
1.64
1.64
1.64
95.08
96.72
98.36
100.00
Observations from Table 26-4 include the following:
• Although four PAH failed screens for RIVA, naphthalene contributed to more than 95
percent of the total failed screens, while the other pollutants accounted for one failed
screen each.
• The maximum concentrations of naphthalene, acenaphthene, and fluorene were all
measured on August 30, 2010 for RIVA. For acenaphthene and fluorene, the
August 30th concentrations were the only ones to fail screens.
26-12
-------
• The risk screening process identified naphthalene as RIVA's only pollutant of
interest. Benzo(a)pyrene was added to this site's pollutants of interest because it is a
NATTS MQO Core Analyte, although it did not contribute to 95 percent of the failed
screens. Hexavalent chromium was also added to the pollutants of interest for RIVA
because it too is a NATTS MQO Core Analyte, even though it did not fail any
screens. This pollutant is not shown in Table 26-4.
26.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Virginia monitoring site. Concentration averages are provided for the pollutants of interest
for the RIVA monitoring site, where applicable. Concentration averages for select pollutants are
also presented graphically for the site, where applicable, to illustrate how the site's
concentrations compare to the program-level averages. In addition, concentration averages for
select pollutants are presented from previous years of sampling in order to characterize
concentration trends at the site, where applicable. Additional site-specific statistical summaries
are provided in Appendices M and O.
26.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for RIVA, as described in Section 3.1. The quarterly average of a particular pollutant is simply
the average concentration of the preprocessed daily measurements over a given calendar quarter.
Quarterly average concentrations include the substitution of zeros for all non-detects. A site must
have a minimum of 75 percent valid samples of the total number of samples possible within a
given quarter for a quarterly average to be calculated. An annual average includes all measured
detections and substituted zeros for non-detects for the entire year of sampling. Annual averages
were calculated for pollutants where three valid quarterly averages could be calculated and
where method completeness was greater than or equal to 85 percent, as presented in Section 2.4.
Quarterly and annual average concentrations for the Virginia monitoring site are presented in
Table 26-5, where applicable. Note that if a pollutant was not detected in a given calendar
quarter, the quarterly average simply reflects "0" because only zeros substituted for non-detects
were factored into the quarterly average concentration.
26-13
-------
Table 26-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Virginia Monitoring Site
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(ng/m3)
2nd
Quarter
Average
(ng/m3)
3rd
Quarter
Average
(ng/m3)
4th
Quarter
Average
(ng/m3)
Annual
Average
(ng/m3)
Richmond, Virginia - RJVA
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
23/60
34/61
60/60
0.16
±0.09
<0.01
±<0.01
108.51
±44.85
0.01
±0.01
0.01
±0.01
95.05
±29.50
0
0.01
±0.01
93.64
±32.17
0.13
±0.09
0.01
±<0.01
127.48
±32.29
0.08
±0.04
0.01
±<0.01
106.17
± 16.89
Observations for RIVA from Table 26-5 include the following:
• The annual average concentration of naphthalene is significantly higher than the
annual average concentrations of hexavalent chromium and benzo(a)pyrene.
• The quarterly averages of benzo(a)pyrene are higher in the colder months of the year
and lower in the warmer months of the year. In fact, this pollutant was not detected in
the third quarter of 2010, and was only detected twice during the second quarter of
2010. Thus, 21 of the 23 measured detections were measured during the first and
fourth quarters of the year.
• The quarterly averages of hexavalent chromium did not vary significantly from
quarter to quarter.
• Naphthalene appears to exhibit a similar quarterly trend as benzo(a)pyrene, but the
confidence intervals indicate that the difference is not statistically significant.
26.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for benzo(a)pyrene,
hexavalent chromium, and naphthalene were created for RIVA. Figures 26-6 through 26-8
overlay the site's minimum, annual average, and maximum concentrations onto the program-
level minimum, first quartile, average, median, third quartile, and maximum concentrations, as
described in Section 3.5.3.
26-14
-------
Figure 26-6. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
Program Max Concentration =42.7 ng/m3
0.2 0.4 0.6 0.8 1 1.2
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
Figure 26-7. Program vs. Site-Specific Average Hexavalent Chromium Concentration
E
1
Program Max Concentration =3.51 ng/m3
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 26-8. Program vs. Site-Specific Average Naphthalene Concentration
•
600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
26-15
-------
Observations from Figures 26-6 through 26-8 include the following:
• Figure 26-6 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
RIVA is less than the program-level average concentration. Figure 26-6 also
shows that the maximum concentration measured at RIVA is well below the
maximum concentration measured across the program. Several non-detects of
benzo(a)pyrene were measured at RIVA.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 26-7 as a result of a relatively large maximum concentration. The
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for the observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 26-7 shows the annual average concentration of hexavalent chromium for
RIVA is less than the program-level average concentration as well as the
program-level median concentration. The maximum concentration measured at
RIVA is just greater than the program-level average concentration. Several non-
detects of hexavalent chromium were measured at RIVA.
• Figure 26-8 shows that the annual naphthalene average for RIVA is greater than
the program-level average concentration. The maximum naphthalene
concentration measured at RIVA is well below the program-level maximum
concentration. There were no non-detects of naphthalene measured at RIVA.
26.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.3. RIVA did not begin sampling PAH or hexavalent chromium under the NMP until
October 2008; therefore, the trends analysis was not conducted.
26.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the RIVA
monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations regarding the
various risk factors, time frames, and calculations associated with these risk screenings.
26-16
-------
26.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Virginia monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where available.
As described in Section 3.3, acute risk results from exposures of 1 to 14 days; intermediate risk
results from exposures of 15 to 364 days; and chronic risk results from exposures of 1 year or
greater. The preprocessed daily measurements of the pollutants of interest were compared to the
acute MRL; the quarterly averages were compared to the intermediate MRL; and the annual
averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the Virginia monitoring site were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
RIVA.
26.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Virginia monitoring site and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 26-6, where applicable.
Table 26-6. Cancer and Noncancer Surrogate Risk Approximations for the Virginia
Monitoring Site
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(ng/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Richmond, Virginia - RIVA
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
0.00176
0.012
0.000034
—
0.0001
0.003
23/60
34/61
60/60
0.08
±0.04
0.01
±<0.01
106.17
± 16.89
0.13
0.13
3.61
—
<0.01
0.04
— = a Cancer URE or Noncancer RfC is not available.
26-17
-------
Observations for RIVA from Table 26-6 include the following:
• The annual average concentration of naphthalene is significantly higher than the
annual average concentrations of benzo(a)pyrene and hexavalent chromium.
• The cancer surrogate risk approximation for naphthalene is 3.61 in-a-million. The
cancer risk approximations for benzo(a)pyrene and hexavalent chromium are well
below 1.0 in-a-million.
• The noncancer risk approximations for hexavalent chromium and naphthalene are
well below than the level of concern for noncancer, which is an HQ of 1.0. There is
not a noncancer RfC for benzo(a)pyrene.
26.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 26-7 and 26-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 26-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 26-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Tables 26-7 and 26-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer surrogate risk approximations based on each site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 26.3, RIVA sampled for PAH and hexavalent chromium. In addition, the cancer and
noncancer surrogate risk approximations are limited to those pollutants with enough data to meet
the criteria for annual averages to be calculated. A more in-depth discussion of this analysis is
provided in Section 3.5.5.3.
26-18
-------
Table 26-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Virginia Monitoring Site
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Richmond, Virginia (Henrico County) - RTVA
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
Dichloromethane
POM, Group 2b
Tetrachloroethylene
POM, Group la
115.26
90.51
64.92
60.31
19.67
9.77
2.81
1.91
0.38
0.19
Formaldehyde
Benzene
1,3 -Butadiene
Naphthalene
POM, Group 3
Hexavalent Chromium, PM
POM, Group 2b
Ethylbenzene
Acetaldehyde
Arsenic, PM
1.18E-03
8.99E-04
5.90E-04
3.32E-04
2.60E-04
2.50E-04
1.68E-04
1.62E-04
1.33E-04
6.94E-05
Naphthalene
Benzo(a)pyrene
Hexavalent Chromium
Cancer Risk
Approximation
(in-a-million)
3.61
0.13
0.13
to
-------
Table 26-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Virginia Monitoring Site
Top 10 Total Emissions for Pollutants with
Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer Toxicity
Weight
Top 10 Noncancer Risk Approximations
Based on Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Richmond, Virginia (Henrico County) - RTVA
Toluene
Xylenes
Methanol
Benzene
Formaldehyde
Hexane
Ethylbenzene
Acetaldehyde
Ethylene glycol
1,3 -Butadiene
752.59
248.81
175.23
115.26
90.51
75.32
64.92
60.31
21.10
19.67
Acrolein
1,3 -Butadiene
Formaldehyde
Acetaldehyde
Benzene
Naphthalene
Xylenes
Arsenic, PM
Lead, PM
Propionaldehyde
216,262.42
9,835.00
9,235.99
6,700.91
3,842.01
3,257.58
2,488.08
1,076.66
792.69
689.88
Naphthalene 0.04
Hexavalent Chromium O.01
to
to
o
-------
Observations from Table 26-7 include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in Henrico County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
cancer UREs) are formaldehyde, benzene, and 1,3-butadiene.
• Seven of the highest emitted pollutants also have the highest toxi city-weighted
emissions for Henrico County.
• Naphthalene, which is the pollutant with the highest cancer risk approximation for
RIVA, has the sixth highest emissions and the fourth highest toxicity-weighted
emissions for Henrico County.
• Hexavalent chromium does not appear among the highest emitted pollutants, but
ranks sixth for the toxicity-weighted emissions.
• POM, Group 2b is the eighth highest emitted "pollutant" in Henrico County and ranks
seventh for toxicity-weighted emissions. POM, Group 2b includes several PAH
sampled for at RIVA including acenaphthylene, fluoranthene, and perylene. None of
the PAH included in POM, Group 2b were identified as pollutants of interest for
RIVA.
Observations from Table 26-8 include the following:
• Toluene, xylenes, and methanol are the highest emitted pollutants with noncancer
RfCs in Henrico County.
• The pollutants with the highest toxicity-weighted emissions (of the pollutants with
noncancer RfCs) are acrolein, 1,3-butadiene, and formaldehyde.
• Five of the highest emitted pollutants in Henrico County also have the highest
toxicity-weighted emissions.
• Naphthalene has the sixth highest toxicity-weighted emissions for Henrico County
but is not among the highest emitted pollutants with a noncancer toxicity factor in
Henrico County.
• Hexvalent chromium appears on neither emissions-based list.
26.6 Summary of the 2010 Monitoring Data for RIVA
Results from several of the data treatments described in this section include the
following:
»«» Although four pollutants failed screens for RIVA, naphthalene failed the majority of
screens.
26-21
-------
»«» The annual average concentration of naphthalene was significantly higher than the
annual average concentrations of the other two pollutants of interest.
»«» Benzo(a)pyrene concentrations appear to be higher during the colder months of the
year.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
26-22
-------
27.0 Site in Washington
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in Washington, and integrates these concentrations
with emissions, meteorological, and risk information. Data generated by sources other than ERG
are not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
27.1 Site Characterization
This section characterizes the monitoring site by providing geographical and physical
information about the location of the site and the surrounding area. This information is provided
to give the reader insight regarding factors that may influence the air quality near the site and
assist in the interpretation of the ambient monitoring measurements.
The NATTS site in Washington is located in Seattle. Figure 27-1 is a composite satellite
images retrieved from ArcGIS Explorer showing the monitoring site in its urban location. Figure
27-2 identifies point source emissions locations by source category, as reported in the 2008 NEI
for point sources. Note that only sources within 10 miles of the site are included in the facility
counts provided in Figure 27-2. Thus, sources outside the 10-mile radius have been grayed out,
but are visible on the map to show emissions sources outside the 10-mile boundary. A 10-mile
boundary was chosen to give the reader an indication of which emissions sources and emissions
source categories could potentially have a direct effect on the air quality at the monitoring site.
Further, this boundary provides both the proximity of emissions sources to the monitoring site as
well as the quantity of such sources within a given distance of the site. Table 27-1 describes the
area surrounding the monitoring site by providing supplemental geographical information such
as land use, location setting, and locational coordinates.
27-1
-------
Figure 27-1. Seattle, Washington (SEWA) Monitoring Site
to
-------
Figure 27-2. NEI Point Sources Located Within 10 Miles of SEWA
122-4OT-W 122-35-0-W 122'30irYV 122'25'0-W
122'15'CTW 122'10'0"W
122-3CTO-W I22-2S-0-W 122'20'0-W
Legend
12T1DWV I22'5'0"W 122"0'Crw
Note: Due- to facility density and collocation the total facilities
displayed may not represent all facilities within the area of interest.
@ SEWA NATTS site 10 mile radius | | County boundary
Source Category Group (No. of Facilities) v Glass Manufacturing (1)
"i< Aerospace/Aircraft Manufacturing (2)
«> Air-conditioning/Refrigeration (1)
•+1 Aircraft Operations (25)
H Automobile/Truck Manufacturing (1)
IB Bakery (2)
Brick Manufacturing & Structural Clay (1)
^ Institutional - school (1)
? Miscellaneous Commercial/Industrial (2)
7 Portland Cement Manufacturing (2)
A Ship Building and Repairing (1)
V Steel Mill (1)
' Wastewater Treatment (1)
27-3
-------
Table 27-1. Geographical Information for the Washington Monitoring Site
Site
Code
SEWA
AQS Code
53-033-0080
Location
Seattle
County
King
Micro- or
Metropolitan
Statistical Area
Seattle-Tacoma-
Bellevue, WA
MSA (Seattle Div)
Latitude
and
Longitude
47.568333,
-122.308056
Land Use
Industrial
Location
Setting
Suburban
Additional Ambient Monitoring Information1
Haze, CO, SO2, NOy, NO, O3, Meteorological
parameters, PM Coarse, PM10, Black Carbon, PM2 5,
PM2 5 Speciation.
BOLD ITALICS = EPA-designated NATTS Site.
to
-k
-------
The SEWA monitoring site is located in Seattle, at the southeast corner of the Beacon
Hill Reservoir. The reservoir and the Jefferson Park Golf Course to the east are separated by
Beacon Avenue. To the south of the site a middle school and a hospital can be seen in the
bottom-most portion of Figure 27-1. The site is surrounded by residential neighborhoods to the
west, north, and east. Interstate-5, which runs north-south through Seattle, is less than 1 mile to
the west and intersects with 1-90 farther north. Interstate-90 runs east-west across Seattle, a
couple of miles to the north of the site. The area to the west of 1-5 is industrial while the area to
the east is primarily residential.
Table 27-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the
Washington monitoring site. Table 27-2 also includes a vehicle registration-to-county population
ratio (vehicles-per-person) for the site. In addition, the population within 10 miles of the site is
presented. An estimate of 10-mile vehicle ownership was calculated by applying the county-level
vehicle registration-to-population ratio to the 10-mile population surrounding the monitoring
site. Table 27-2 also contains annual average daily traffic information. Finally, Table 27-2
presents the daily VMT for King County.
Table 27-2. Population, Motor Vehicle, and Traffic Information for the Washington
Monitoring Site
Site
SEWA
Estimated
County
Population1
1,937,961
County-level
Vehicle
Registration2
1,763,504
Vehicles per
Person
(Registration:
Population)
0.91
Population
within 10
miles3
952,319
Estimated
10-mile
Vehicle
Ownership
866,590
Annual
Average
Daily
Traffic4
234,000
County-
level Daily
VMT5
23,454,115
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the Washington Department of Licensing (WA DOL,
2010)
310-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2010 data from the Washington DOT (WA DOT, 2010a)
5 County-level VMT reflects 2010 data from the Washington DOT (WA DOT, 2010b)
BOLD ITALICS = EPA-designated NATTS Site.
Observations from Table 27-2 include the following:
• King County has the sixth highest county-level population among counties with NMP
sites. The 10-mile population estimate for SEWA ranks in the top third among NMP
sites.
27-5
-------
• The county-level and 10-mile vehicle registration counts for SEWA mimicked the
rankings of the county-level and 10-mile populations.
• The vehicle-per-person ratio for SEWA was in the middle of the range compared to
other NMP sites.
• The traffic volume experienced near SEWA was the third highest compared to other
NMP monitoring sites. The traffic estimate used came from 1-5 near Spokane Street.
• The King County VMT was in the top third compared to other counties with NMP
sites (where VMT data were available).
27.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in Washington on sample days, as well as over the course of the year.
27.2.1 Climate Summary
The city of Seattle is located between Puget Sound and Lake Washington. The entire
urban area is situated between the Olympic Mountains to the west and the Cascades to the east.
The area experiences a mild climate as the mountains moderate storm systems that move into the
Pacific Northwest and both the mountains and the Sound shield the city from temperature
extremes. Although the city is known for its cloudy, rainy conditions, actual precipitation totals
tend to be lower compared to many locations east of the Rocky Mountains. The winter months
are the wettest and the summer months the driest. Prevailing winds are out of the southwest
(Bair, 1992).
27.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest SEWA were retrieved
for 2010 (NCDC, 2010). The closest weather station to SEWA is located at Boeing Field/King
County International Airport (WBAN 24234). Additional information about this weather station,
such as the distance between the site and the weather station, is provided in Table 27-3. These
data were used to determine how meteorological conditions on sample days vary from normal
conditions throughout the year.
27-6
-------
Table 27-3. Average Meteorological Conditions near the Washington Monitoring Site
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Seattle, Washington - SEWA
Boeing Field/
King County
Intl Airport
24234
(47.53, -122.30)
2.66
miles
189°
(S)
Sample
Day
2010
60.4
±2.6
60.6
+ 1.1
53.9
±2.2
53.8
+ 0.9
43.6
±1.8
43.8
+ 0.8
48.7
±1.7
48.8
+ 0.8
70.5
±2.6
71.1
+ 1.1
1015.7
±1.7
1015.2
+ 0.7
4.3
±0.6
4.3
+ 0.2
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
to
-------
Table 27-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire year for 2010. Also included in Table 27-3 is the
95 percent confidence interval for each parameter. As shown in Table 27-3, average
meteorological conditions on sample days near SEWA were representative of average weather
conditions throughout the year.
27.2.3 Back Trajectory Analysis
Figure 27-3 is the composite back trajectory map for days on which samples were
collected at the SEWA monitoring site in 2010. Included in Figure 27-3 are four back trajectories
per sample day. Figure 27-4 is the corresponding cluster analysis for 2010. An in-depth
description of these maps and how they were generated is presented in Section 3.5.2.1. For the
composite map, each line represents the 24-hour trajectory along which a parcel of air traveled
toward the monitoring site on a given sample day and time. For the cluster analysis, each line
corresponds to a back trajectory representative of a given cluster of trajectories. For both maps,
each concentric circle around the site in Figures 27-3 and 27-4 represents 100 miles.
Observations from Figures 27-3 and 27-4 for SEWA include the following:
• Back trajectories originated from a variety of directions from SEWA.
• The 24-hour air shed domain for SEWA is somewhat smaller than for other NMP
sites. Although the longest trajectory originated greater than 800 miles away over the
Pacific Ocean, the average trajectory length was less than 200 miles long and
80 percent of trajectories originated within 300 miles of the site.
• The cluster analysis shows that 35 percent of back trajectories originated to the
southeast or southwest of SEWA, over the southwestern portion of Washington, and
generally less than 200 miles from the site. Another 24 percent of back trajectories
originated to the northeast or northwest of SEWA, over northwestern Washington and
southern British Columbia, Canada, and also less than 200 miles from the site. Thus,
nearly 70 percent of trajectories originated with 200 miles of the site. Another 17
percent of trajectories originated to the northwest of SEWA, over the Pacific Ocean,
with the bulk of these originating within 300 miles of the site. Fifteen percent of back
trajectories originated from the southeast over central Oregon, while another 10
percent originated off the coast of Oregon and northern California.
27-8
-------
Figure 27-3. 2010 Composite Back Trajectory Map for SEWA
Figure 27-4. Back Trajectory Cluster Map for SEWA
' «. \ - -s
» N X ^
V % \ ^
\ \ N S 1~~-
\ )
27-9
-------
27.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at Boeing Field/King County
International Airport were uploaded into a wind rose software program to produce customized
wind roses, as described in Section 3.5.2.2. A wind rose shows the frequency of wind directions
using "petals" positioned around a 16-point compass, and uses different colors to represent wind
speeds.
Figure 27-5 presents three different wind roses for the SEWA monitoring site. First, a
historical wind rose representing 1999 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose for 2010
representing wind observations for the entire year is presented. Next, a wind rose representing
days on which samples were collected in 2010 is presented. These can be used to determine if
wind observations on sample days were representative of conditions experienced over the entire
year and historically. Finally, a map showing the distance between the NWS station and the
monitoring site is presented, which may be useful for identifying topographical influences that
may affect the meteorological patterns experienced at each location.
Observations from Figure 27-5 for SEWA include the following:
• The Boeing Field/King County Airport weather station is located approximately
2.7 miles south of SEWA.
• The historical wind rose shows that southeasterly, south-southeasterly, and southerly
winds were frequently observed, accounting for more than one-third of observations.
Calm winds (< 2 knots) accounted for 23 percent of wind observations near SEWA.
• The wind patterns shown on the 2010 wind rose are similar to the historical wind
patterns, although the percentage of calms winds is higher (nearly 28 percent) in
2010.
• The wind patterns shown on the sample day wind rose also resemble the historical
wind patterns, indicating that conditions on sample days were representative of those
experienced over the entire year and historically.
27-10
-------
Figure 27-5. Wind Roses for the Boeing Field/King County International Airport Weather
Station near SEWA
1999-2009 Historical Wind Rose
2010 Wind Rose
Calms: 22.95%
2010 Sample Day Wind Rose
Distance between SEWA and NWS Station
litVn1
IMMB \ 1 I «
'SiT1 s \ •,^*..,,j:
I •• II ".-..ill
'-.
\ \ ">-•
"ar J *» "'«-
.; «i . !•'-
><«„,.. r:
.'—•" I g I rr
! •
T. \ I iio.«i«,ii
k>. \ 1*4*1 I ; I mi,!
«...» |- '""•
I
I ^ j
^ .trc.
S,.9.
••>so.h«i.«i:
^ WJfelttr f
•l Oirvctoilt
\l •.
27-11
-------
27.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Washington monitoring site
in order to allow analysts and readers to focus on a subset of pollutants through the context of
risk. Each pollutant's preprocessed daily measurement was compared to its associated risk
screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by the monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 27-4 presents the pollutants of interest for SEWA. The pollutants that failed at least
one screen and contributed to 95 percent of the total failed screens are shaded. NATTS MQO
Core Analytes are bolded. Thus, pollutants of interest are shaded and/or bolded. SEWA sampled
for PMio metals, VOC, PAH, carbonyl compounds, and hexavalent chromium.
Table 27-4. Risk Screening Results for the Washington Monitoring Site
Pollutant
Screening
Value
(Hg/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Seattle, Washington - SEWA
Benzene
Carbon Tetrachloride
Formaldehyde
Arsenic (PM10)
1,3-Butadiene
Naphthalene
Acet aldehyde
Manganese (PM10)
Nickel (PM10)
1 ,2-Dichloroethane
Ethylbenzene
Dichloromethane
Acenaphthene
Acrylonitrile
Fluorene
Hexavalent Chromium
0.13
0.17
0.077
0.00023
0.03
0.029
0.45
0.005
0.0021
0.038
0.4
7.7
0.011
0.015
0.011
0.000083
Total
59
59
59
53
50
50
49
20
20
12
5
4
1
1
1
1
444
59
59
59
58
53
58
59
58
58
12
59
59
57
1
57
50
816
100.00
100.00
100.00
91.38
94.34
86.21
83.05
34.48
34.48
100.00
8.47
6.78
1.75
100.00
1.75
2.00
54.41
13.29
13.29
13.29
11.94
11.26
11.26
11.04
4.50
4.50
2.70
1.13
0.90
0.23
0.23
0.23
0.23
13.29
26.58
39.86
51.80
63.06
74.32
85.36
89.86
94.37
97.07
98.20
99.10
99.32
99.55
99.77
100.00
27-12
-------
Observations from Table 27-4 for SEWA include the following:
• Sixteen pollutants failed at least one screen for SEWA, of which 10 are NATTS
MQO Core Analytes.
• The risk screening process identified 10 pollutants of interest, of which all but one
(1,2-dichloroethane) are NATTS MQO Core Analytes. Hexavalent chromium was
added to SEWA's pollutants of interest because it is a NATTS MQO Core Analyte,
even though it did not contribute to 95 percent of SEWA's total failed screens. Seven
additional pollutants were added to SEWA's pollutants of interest because they are
NATTS MQO Core Analytes, even though they did not fail any screens
(benzo(a)pyrene, beryllium, cadmium, chloroform, lead, tetrachloroethylene, and
trichloroethylene). These seven pollutants are not shown in Table 27-4 but are shown
in subsequent tables in the following sections. While vinyl chloride is also a NATTS
MQO Core Analyte, it was not detected at SEWA, and therefore not added to the list
of pollutants of interest.
• The percentage of measured detections failing screens (of the pollutants with at least
one failed screen) for SEWA is greater than 50 percent.
• Benzene, carbon tetrachloride, and formaldehyde failed 100 percent of screens for
SEWA. Acrylonitrile and 1,2-dichloroethane also failed 100 percent of screens for
SEWA, but these two pollutants were detected in only a few of the total sampled
collected, while benzene, carbon tetrachloride, and formaldehyde were detected in all
59 samples collected.
27.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Washington monitoring site. Concentration averages are provided for the pollutants of
interest for the site, where applicable. Concentration averages for select pollutants are also
presented graphically for the site, where applicable, to illustrate how the site's concentrations
compare to the program-level averages. In addition, concentration averages for select pollutants
are presented from previous years of sampling in order to characterize concentration trends at the
site, where applicable. Additional site-specific statistical summaries are provided in Appendices
J, L, M, N, and O.
27.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for SEWA, as described in Section 3.1. The quarterly average of a particular pollutant is simply
the average concentration of the preprocessed daily measurements over a given calendar quarter.
Quarterly average concentrations include the substitution of zeros for all non-detects. A site must
27-13
-------
have a minimum of 75 percent valid samples of the total number of samples possible within a
given quarter for a quarterly average to be calculated. An annual average includes all measured
detections and substituted zeros for non-detects for the entire year of sampling. Annual averages
were calculated for pollutants where three valid quarterly averages could be calculated and
where method completeness was greater than or equal to 85 percent, as presented in Section 2.4.
Quarterly and annual average concentrations for the Washington monitoring site are presented in
Table 27-5, where applicable. Note that concentrations of the PAH, metals, and hexavalent
chromium are presented in ng/m3 for ease of viewing. Also note that if a pollutant was not
detected in a given calendar quarter, the quarterly average simply reflects "0" because only zeros
substituted for non-detects were factored into the quarterly average concentration.
Table 27-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Washington Monitoring Site
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(jig/m3)
2nd
Quarter
Average
(jig/m3)
3rd
Quarter
Average
(jig/m3)
4th
Quarter
Average
(jig/m3)
Annual
Average
(jig/m3)
Seattle, Washington - SEWA
Acetaldehyde
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Formaldehyde
Tetrachloroethylene
Trichloroethylene
Arsenic (PM10)a
Benzo(a)pyrene a
Bery Ilium (PM10)a
Cadmium (PM10)a
59/59
59/59
53/59
59/59
59/59
12/59
59/59
52/59
5/59
58/58
20/58
41/58
57/58
0.74
±0.20
0.82
±0.14
0.08
±0.03
0.78
±0.07
0.12
±0.01
0.03
±0.02
0.59
±0.17
0.11
±0.05
0
0.64
±0.22
0.05
±0.03
<0.01
±0.01
0.09
±0.03
0.62
±0.11
0.58
±0.11
0.05
±0.01
0.71
±0.10
0.16
±0.04
0.03
±0.02
0.41
±0.09
0.10
±0.02
0.01
±0.01
0.52
±0.13
0.01
±0.01
O.01
±0.01
0.08
±0.02
1.09
±0.20
0.57
±0.11
0.05
±0.02
0.69
±0.06
0.16
±0.01
0
0.90
±0.16
0.12
±0.04
0.01
±0.01
0.55
±0.09
0.01
±0.01
O.01
±0.01
0.07
±0.05
0.75
±0.20
0.81
±0.11
0.09
±0.02
0.71
±0.06
0.12
±0.01
0
0.63
±0.17
0.16
±0.05
0
0.63
±0.28
0.05
±0.03
O.01
±0.01
0.11
±0.05
0.81
±0.10
0.69
±0.06
0.07
±0.01
0.72
±0.04
0.14
±0.01
0.02
±0.01
0.64
±0.08
0.12
±0.02
O.01
±0.01
0.58
±0.09
0.03
±0.01
O.01
±0.01
0.09
±0.02
a Average concentrations provided for the pollutants below the black line are presented in ng/m for ease of
viewing.
27-14
-------
Table 27-5. Quarterly and Annual Average Concentrations of the Pollutants of Interest for
the Washington Monitoring Site (Continued)
Pollutant
Hexavalent Chromium3
Lead(PM10)a
Manganese (PM10) a
Naphthalene a
Nickel (PM10)a
#of
Measured
Detections
vs. # of
Samples
50/59
58/58
58/58
58/58
58/58
1st
Quarter
Average
(Ug/m3)
0.03
±0.01
2.88
±0.92
5.95
±3.57
68.87
± 17.00
1.59
±0.89
2nd
Quarter
Average
(Ug/m3)
0.02
±0.01
2.13
±0.47
5.59
±2.89
38.64
± 12.03
1.95
±0.78
3rd
Quarter
Average
(Ug/m3)
0.03
±0.01
2.73
±0.55
8.04
±5.10
78.34
± 26.48
2.76
±1.30
4th
Quarter
Average
(Ug/m3)
0.03
±0.01
2.75
±0.65
2.86
±0.84
59.67
± 12.29
1.17
±0.53
Annual
Average
(Ug/m3)
0.03
±0.01
2.63
±0.32
5.75
±1.79
61.44
±9.43
1.91
±0.48
a Average concentrations provided for the pollutants below the black line are presented in ng/m3 for ease of
viewing.
Observations from Table 27-5 include the following:
• The annual average concentrations for all of SEWA's pollutants of interest are less
than 1.0 |ig/m3. The pollutants with the highest annual average concentrations by
mass are acetaldehyde (0.81 ± 0.10 |ig/m3), carbon tetrachloride (0.72 ± 0.04 |ig/m3),
benzene (0.69 ± 0.06 |ig/m3), and formaldehyde (0.64 ± 0.08 |ig/m3).
• Acetaldehyde and formaldehyde's third quarter concentration averages are higher
than the other quarterly averages. While the confidence intervals indicate that these
differences are not statistically significant, a review of the data shows that seven of
the 13 acetaldehyde concentrations greater than 1 |ig/m3 and five of the eight
formaldehyde concentrations greater than 1 |ig/m3 were measured during the third
quarter of 2010 at SEWA.
• Benzene, 1,3-butadiene, arsenic, and benzo(a)pyrene appear to be higher during the
colder months of year. However, the confidence intervals indicate that the differences
are statistically significant for benzo(a)pyrene only. All six of the benzo(a)pyrene
concentrations greater than 0.1 ng/m3 were measured during the first (3) or fourth (3)
quarters. This pollutant was detected in about one-third of PAH samples collected at
SEWA (20/58), with eight measured during the first quarter, two each during the
second and third quarters, and eight during the fourth quarter.
• All of the measured detections of 1,2-dichloroethane were measured during the first
and second quarters of 2010 with no measured detections after May 14, 2010. This is
similar to other NMP sites sampling VOC.
• The third quarter average concentration of manganese is greater than the other
quarterly averages and has a relatively high confidence interval associated with it. A
review of the data shows that the maximum manganese concentration was measured
on August 30, 2010 (44.0 ng/m3) and is four times higher than the next highest
27-15
-------
concentration measured during the third quarter (9.60 ng/m3 measured on August 12,
2010). The concentration measured on August 30 is the fifth highest manganese
concentration measured among sites sampling PMio metals. Note that the first and
second quarter average concentrations of manganese also have relatively high
confidence intervals, indicating a considerable amount of variability associated with
manganese concentrations measured at SEWA.
• The third quarter average concentration of nickel is also higher than the other
quarterly averages and has a relatively high confidence interval associated with it. A
review of the data shows that the maximum nickel concentration was measured on
July 25, 2010 (10.6 ng/m3) and is roughly 50 percent higher than the next highest
concentration measured during the third quarter (6.65 ng/m3 measured on August 24,
2010). The concentration measured on July 25 is the highest nickel concentration
measured among sites sampling metals and the August 24 concentration ranks third.
Of the 22 nickel concentrations greater than 3 ng/m3 measured among all NMP sites
sampling nickel, 13 were measured at SEWA.
Tables 4-9 through 4-12 present the sites with the 10 highest annual average
concentrations for each of the program-level pollutants of interest. Observations for SEWA from
those tables include the following:
• As shown in Table 4-9, SEWA has the highest annual average concentration of
carbon tetrachloride among NMP sites sampling VOC. However, the annual average
concentrations of this pollutant do not vary significantly among the sites.
• As shown in Table 4-12, SEWA has the highest annual average concentration of
nickel among all sites sampling metals (PMio and TSP).
• SEWA does not appear in Table 4-10 for carbonyl compounds or Table 4-11 for
PAH.
27.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, box plots for acetaldehyde, arsenic,
benzene, benzo(a)pyrene, 1,3-butadiene, formaldehyde, hexavalent chromium, manganese, and
naphthalene were created for SEWA. Figures 27-6 through 27-14 overlay the site's minimum,
annual average, and maximum concentrations onto the program-level minimum, first quartile,
average, median, third quartile, and maximum concentrations for each pollutant, as described in
Section 3.5.3.
27-16
-------
Figure 27-6. Program vs. Site-Specific Average Acetaldehyde Concentration
Concentration (|ig/m3)
Program
Site:
: IstQuartile
Site Average
0
2ndQuartile SrdQuartile 4thQuartile Avt
• n
i — i i — i
Site Minimum/Maximum
;rage
Figure 27-7. Program vs. Site-Specific Average Arsenic (PMio) Concentration
-
)
]
•
)
I
H
0.5 1 1
5 2 2.5 3 3.5 4
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Ave
• D D D
Site: Site Average Site Minimum/Maximum
o —
4.5 5
'rage
figure 27-8. Program vs. Site-Specific Average Benzene Concentration
0
1 2
345
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile 3rdQuartile 4thQuartile Ave
• D D D
Site: Site Average Site Minimum/Maximum
o —
6
'rage
7
27-17
-------
Figure 27-9. Program vs. Site-Specific Average Benzo(a)pyrene Concentration
>•%
3
Program Max Concentration = 42.7 ng/m3 j
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Concentration (ng/m3)
Program: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Ave
• D D D
Site: Site Average Site Minimum/Maximum
o —
1.8 2
'rage
Figure 27-10. Program vs. Site-Specific Average 1,3-Butadiene Concentration
o|
•
0 0.1
Program:
Site:
0.2 0.3
1st Quartile
Site Average
0
0.4 0.5 0.6 0.7 0.8 0.9 1
Concentration (ng/m3)
2nd Quartile
3rd Quartile 4th Quartile Average
n n
Site Minimum/Maximum
Figure 27-11. Program vs. Site-Specific Average Formaldehyde Concentration
I
10 15
25 30
Concentration (ng/m3)
35 40 45 50
Program: 1st Quartile 2nd Quartile 3rd Quartile 4th Quartile Average
Site: Site Average Site Minimum/Maximum
O
27-18
-------
Figure 27-12. Program vs. Site-Specific Average Hexavalent Chromium Concentration
fez
1
Program Max Concentration = 3.51 ng/m3
0.3 0.45
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
o
Figure 27-13. Program vs. Site-Specific Average Manganese (PMio) Concentration
1 1
F
80 100 120 140 160 180 200
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site: Site Average Site Minimum/Maximum
O
Figure 27-14. Program vs. Site-Specific Average Naphthalene Concentration
• o
™
600 800
Concentration (ng/m3)
Program: IstQuartile 2ndQuartile SrdQuartile 4thQuartile Average
Site:
Site Average Site Minimum/Maximum
O
27-19
-------
Observations from Figures 27-6 through 27-14 include the following:
• Figure 27-6 shows that SEWA's annual average acetaldehyde concentration is
well below the program-level average for acetaldehyde. SEWA's annual average
acetaldehyde concentration is also below the program-level median and first
quartile. Even the maximum acetaldehyde concentration measured at SEWA is
below the program-level average concentration. This site has the lowest annual
average concentration of acetaldehyde among NMP sites sampling carbonyl
compounds.
• Figure 27-7 shows that SEWA's annual average arsenic (PMio) concentration is
just greater than the program-level average for arsenic (PMio). There were no
non-detects of arsenic measured at SEWA.
• Figure 27-8 for benzene shows that the annual average concentration for SEWA is
less than both the program-level average and median concentration. The
maximum benzene concentration measured at SEWA is well below the maximum
benzene concentration measured across the program. There were no non-detects
of benzene measured at SEWA.
• Figure 27-9 is the box plot for benzo(a)pyrene. Note that the program-level
maximum concentration (42.7 ng/m3) is not shown directly on the box plot
because the scale of the box plot would be too large to readily observe data points
at the lower end of the concentration range. Thus, the scale has been reduced to
2 ng/m3. Also note that the first quartile for this pollutant is zero and is not visible
on this box plot. This box plot shows that the annual average concentration for
SEWA is well below the program-level average concentration and just greater
than the program-level median. Figure 27-9 also shows that the maximum
concentration measured at SEWA is well below the maximum concentration
measured across the program. Sixty-six percent of the benzo(a)pyrene
measurements at SEWA were non-detects.
• Figure 27-10 for 1,3-butadiene shows that the annual average concentration for
SEWA is below the program-level average concentration but greater than the
program-level median concentration. Figure 27-10 also shows that the maximum
1,3-butadiene concentration measured at SEWA is well below the maximum
concentration measured across the program. A few non-detects of 1,3-butadiene
were measured at SEWA.
• Figure 27-11 shows that SEWA's annual average formaldehyde concentration is
below not only the program-level average but also the median and first quartile
concentrations at the program level. The maximum formaldehyde concentration
measured at SEWA is less than both the program-level average and median
concentrations. Similar to its acetaldehyde concentration, this site has the lowest
annual average concentration of formaldehyde among NMP sites sampling
carbonyl compounds.
• Similar to benzo(a)pyrene, the scale for hexavalent chromium has been adjusted
in Figure 27-12 as a result of a relatively large maximum concentration. The
27-20
-------
program-level maximum concentration (3.51 ng/m3) is not shown directly on the
box plot in order to allow for the observation of data points at the lower end of the
concentration range; thus, the scale has been reduced to 0.75 ng/m3. Also note that
the first quartile for this pollutant is zero and is not visible on this box plot.
Figure 27-12 shows the annual average concentration of hexavalent chromium for
SEWA is less than the program-level average but greater than the program-level
median concentration. A few non-detects of hexavalent chromium were measured
at SEWA.
• Figure 27-13 shows the annual average concentration of manganese (PMio) for
SEWA is less than the program-level average. While the maximum concentration
measured at SEWA is well below the program maximum concentration, the
maximum concentration measured at SEWA is among one of the highest
concentrations among the NMP sites sampling manganese (PMio), as discussed in
the previous section. Although difficult to discern in Figure 27-13, there were no
non-detects of manganese measured at SEWA.
• Figure 27-14 shows that the annual naphthalene average for SEWA is less than
the program-level average concentration and just less than the program-level
median concentration. The maximum naphthalene concentration measured at
SEWA is well below the program-level maximum concentration. There were no
non-detects of naphthalene measured at SEWA.
27.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.4. Although SEWA has sampled hexavalent chromium since 2005, sampling was
discontinued for an eight-month period in 2006 from March through October. Because four
months is not considered enough to be representative of an entire year, and this year would factor
into two of the three 3-year periods, the trends analysis was not conducted. In addition, sampling
for PMio metals, VOC, and carbonyl compounds did not begin until January 2007 and PAH
sampling did not begin until March 2008.
27.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the
Washington monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
27-21
-------
27.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data for the
Washington monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
for SEWA were compared to the acute MRL; the quarterly averages were compared to the
intermediate MRL; and the annual and/or study averages were compared to the chronic MRL.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the Washington monitoring site were greater than their respective MRL noncancer
health risk benchmarks. This is also true for pollutants not identified as pollutants of interest for
SEWA.
27.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Washington monitoring site and where annual
average concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 27-6, where applicable.
Observations from Table 27-6 for SEWA include the following:
• The pollutants with the highest annual averages for SEWA are acetaldehyde, carbon
tetrachloride, benzene, and formaldehyde.
• The pollutants with the highest cancer surrogate risk approximations are
formaldehyde, benzene, carbon tetrachloride, and arsenic. Although the cancer risk
approximation for formaldehyde is the highest for SEWA, it is the lowest cancer risk
approximation for this pollutant among NMP sites sampling carbonyl compounds.
• The noncancer surrogate risk approximations for SEWA are all less than 1.0, with the
highest calculated for manganese (0.11).
27-22
-------
Table 27-6. Cancer and Noncancer Surrogate Risk Approximations for the Washington
Monitoring Site
Pollutant
Cancer
URE
(Hg/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(Hg/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Seattle, Washington - SEWA
Acetaldehyde
Arsenic (PM10)a
Benzene
Benzo (a) pyrene a
Bery Ilium (PM10)a
1,3 -Butadiene
Cadmium (PM10) a
Carbon Tetrachloride
Chloroform
1 ,2-Dichloroethane
Formaldehyde
Hexavalent Chromium3
Lead(PM10)a
Manganese (PM10) a
Naphthalene a
Nickel (PM10)a
Tetrachloroethylene
Trichloroethylene
0.0000022
0.0043
0.0000078
0.00176
0.0024
0.00003
0.0018
0.000006
0.000026
0.000013
0.012
_
0.000034
0.00048
2.6E-07
0.0000048
0.009
0.000015
0.03
0.00002
0.002
0.00001
0.1
0.098
2.4
0.0098
0.0001
0.00015
0.00005
0.003
0.00009
0.04
0.002
59/59
58/58
59/59
20/58
41/58
53/59
57/58
59/59
59/59
12/59
59/59
50/59
58/58
58/58
58/58
58/58
52/59
5/59
0.81
±0.10
0.01
±0.01
0.69
±0.06
0.01
±0.01
O.01
± O.01
0.07
±0.01
O.01
±O.01
0.72
±0.04
0.14
±0.01
0.02
±0.01
0.64
±0.08
0.01
± O.01
0.01
±0.01
0.01
±O.01
0.06
±0.01
O.01
±O.01
0.12
±0.02
O.01
±O.01
1.77
2.51
5.38
0.04
O.01
1.98
0.16
4.35
0.42
8.26
0.34
_
2.09
0.91
0.03
0.02
0.09
0.04
0.02
O.01
0.03
0.01
0.01
O.01
0.01
0.06
O.01
0.02
0.11
0.02
0.02
0.01
O.01
— = a Cancer URE or Noncancer RfC is not available.
a For the annual average concentration of this pollutant in ng/m3, refer to Table 27-5.
27-23
-------
27.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 27-7 and 27-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 27-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer surrogate risk approximations (in-a-million), as calculated from the annual averages.
Table 27-8 presents similar information, but identifies the 10 pollutants with the highest
noncancer surrogate risk approximations (HQ), also calculated from annual averages.
The pollutants listed in Table 27-7 and 27-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. Further, the cancer and noncancer surrogate risk approximations based on each site's
annual averages are limited to those pollutants for which each respective site sampled. As
discussed in Section 5.3, SEWA sampled for VOC, carbonyl compounds, PAH, metals (PMio),
and hexavalent chromium. In addition, the cancer and noncancer surrogate risk approximations
are limited to those pollutants with enough data to meet the criteria for annual averages to be
calculated. A more in-depth discussion of this analysis is provided in Section 3.5.5.3.
27-24
-------
Table 27-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Washington Monitoring Site
Top 10 Total Emissions for Pollutants with
Cancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted
Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Pollutant
Cancer Risk
Approximation
(in-a-million)
Seattle, Washington (King County) - SEWA
Benzene
Formaldehyde
Ethylbenzene
Acetaldehyde
1,3 -Butadiene
Naphthalene
POM, Group 2b
Dichloromethane
POM, Group 6
Nickel, PM
1,542.30
849.03
654.82
430.54
190.09
99.14
28.31
17.29
2.99
2.83
Benzene
Formaldehyde
POM, Group 3
1,3 -Butadiene
Naphthalene
POM, Group 2b
Ethylbenzene
Nickel, PM
Hexavalent Chromium, PM
POM, Group 5a
1.20E-02
1.10E-02
8.19E-03
5.70E-03
3.37E-03
2.49E-03
1.64E-03
1.36E-03
1.34E-03
1.11E-03
Formaldehyde
Benzene
Carbon Tetrachloride
Arsenic
Naphthalene
1,3 -Butadiene
Acetaldehyde
Nickel
1 ,2-Dichloroethane
Hexavalent Chromium
8.26
5.38
4.35
2.51
2.09
1.98
1.77
0.91
0.42
0.34
to
^1
to
-------
Table 27-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Washington Monitoring Site
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer
Toxicity Weight
Top 10 Noncancer Risk Approximations Based
on Annual Average Concentrations
(Site-Specific)
Pollutant
Noncancer Risk
Approximation
(HQ)
Seattle, Washington (King County) - SEWA
Toluene
Xylenes
Benzene
Methanol
Formaldehyde
Ethylbenzene
Hexane
Acetaldehyde
1,3 -Butadiene
Ethylene glycol
3,282.31
2,629.14
1,542.30
1,092.81
849.03
654.82
642.62
430.54
190.09
142.99
Acrolein
1,3 -Butadiene
Formaldehyde
Benzene
Acetaldehyde
Naphthalene
Nickel, PM
Xylenes
Lead, PM
Arsenic, PM
2,237,728.52
95,046.24
86,635.26
51,409.86
47,837.23
33,047.76
31,499.47
26,291.40
15,384.07
10,669.99
Manganese
Acetaldehyde
Formaldehyde
Arsenic
1,3 -Butadiene
Benzene
Nickel
Naphthalene
Lead
Cadmium
0.11
0.09
0.06
0.04
0.03
0.02
0.02
0.02
0.02
0.01
to
^1
to
-------
Observations from Table 27-7 for SEWA include the following:
• Benzene, formaldehyde, and ethylbenzene are the highest emitted pollutants with
cancer UREs in King County.
• The pollutants with the highest toxi city-weighted emissions (of the pollutants with
cancer UREs) for King County are benzene, formaldehyde, and POM, Group 3.
• Seven of the highest emitted pollutants also have the highest toxi city-weighted
emissions for King County.
• Formaldehyde and benzene topped SEWA's highest cancer risk approximations list.
These two pollutants top both emissions-based lists as well. Naphthalene,
1,3-butadiene, and nickel also appear on all three lists.
• Carbon tetrachloride, which is the third-ranked pollutant for SEWA's cancer risk
approximations, does not appear on either emissions-based list. Acetaldehyde has the
seventh highest cancer risk approximation and ranks fourth for total emissions, but
ranks 11th for toxicity-weighted emissions and thus, does not appear in this column in
Table 27-7. Hexavalent chromium has the tenth highest cancer risk approximation
and ranks ninth for toxicity-weighted emissions, but ranks 19th for total emissions and
thus, does not appear in this column in Table 27-7.
• POM, Group 2b is the seventh highest emitted "pollutant" in King County and ranks
sixth for toxicity-weighted emissions. POM, Group 2b includes several PAH sampled
for at SEWA including acenaphthene, fluorene, and perylene. Although none of the
PAH included in POM, Group 2b were identified as pollutants of interest for SEWA,
acenaphthene and fluorene did fail screens for SEWA.
• POM, Group 3 ranks third for toxicity-weighted emissions for King County. POM,
Group 3 does not include any pollutants sampled for at SEWA.
Observations from Table 27-8 for SEWA include the following:
• Toluene, xylenes, and benzene are the highest emitted pollutants with noncancer
RfCs in King County.
• Acrolein is the pollutant with the highest toxicity-weighted emissions (of the
pollutants with noncancer RfCs) for King County, followed by 1,3-butadiene and
formaldehyde. Although acrolein was sampled for at SEWA, this pollutant was
excluded from the pollutants of interest designation, and thus subsequent risk
screening evaluations, due to questions about the consistency and reliability of the
measurements, as discussed in Section 3.2.
• Five of the highest emitted pollutants also have the highest toxicity-weighted
emissions for King County.
27-27
-------
• Manganese, which has the highest noncancer risk approximation for SEW A, does not
appear on either emissions-based list for King County. Acetaldehyde, formaldehyde,
1,3-butadiene, and benzene appear on all three lists in Table 27-8.
• Three of the four metals that appear among the highest noncancer risk approximations
for SEWA are also among the pollutants with the highest toxicity-weighted emissions
(arsenic, nickel, and lead). However, none of these metals are among the highest
emitted pollutants in King County.
27.6 Summary of the 2010 Monitoring Data for SEWA
Results from several of the data treatments described in this section include the
following:
• Sixteen pollutants failed at least one screen for SEWA, of which 10 are NATTSMQO
Core Analytes.
»«» Acetaldehyde had the highest annual average concentration for SEWA. The annual
average concentrations of carbon tetrachloride and nickel for SEWA were the highest
among NMP sites sampling these pollutants.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations of the pollutants of interest were greater than their associated
MRL noncancer health risk benchmarks.
27-28
-------
28.0 Site in Wisconsin
This section examines the spatial and temporal characteristics of the ambient monitoring
concentrations measured at the NATTS site in Wisconsin, and integrates these concentrations
with emissions, meteorological, and risk information. Data generated by sources other than ERG
are not included in the data analyses contained in this report. Readers are encouraged to refer to
Sections 1 through 4 and the glossary (Appendix P) for detailed discussions and definitions
regarding the various data analyses presented below.
28.1 Site Characterization
This section characterizes the monitoring site by providing geographical and physical
information about the location of the site and the surrounding area. This information is provided
to give the reader insight regarding factors that may influence the air quality near the site and
assist in the interpretation of the ambient monitoring measurements.
The NATTS site located in Mayville, Wisconsin (MVWI) was moved to a new location
in Horicon, Wisconsin (HOWI) in mid-December 2009 and promptly began sampling. In order
to capture all of the data generated by the new site, the data results in this section include the two
sample days in December 2009 (December 21 and December 27) as well as all of 2010.
Figure 28-1 is a composite satellite image retrieved from ArcGIS Explorer showing the
monitoring site in its rural location. Figure 28-2 identifies point source emissions locations by
source category, as reported in the 2008 NEI for point sources. Note that only sources within
10 miles of the site are included in the facility counts provided in Figure 28-2. Thus, sources
outside the 10-mile radius have been grayed out, but are visible on the map to show emissions
sources outside the 10-mile boundary. A 10-mile boundary was chosen to give the reader an
indication of which emissions sources and emissions source categories could potentially have a
direct effect on the air quality at the monitoring site. Further, this boundary provides both the
proximity of emissions sources to the monitoring site as well as the quantity of such sources
within a given distance of the site. Table 28-1 describes the area surrounding the monitoring site
by providing supplemental geographical information such as land use, location setting, and
locational coordinates.
28-1
-------
Figure 28-1. Horicon, Wisconsin (HOWI) Monitoring Site
to
oo
to
-------
Figure 28-2. NEI Point Sources Located Within 10 Miles of HOWI
Legend
HOWI NATTS site
Source Category Group (No. of Facilities)
-f Aircraft Operations (3)
B Automobile/Truck Manufacturing (1)
E Electroplating, Plating, Polishing, Anodizing, & Coloring (1}
0 Fabricated Metal Products (5)
F Food Processing/Agriculture (2)
• Lime Manufacturing (1)
5 Miscellaneous Coating Manufacturing (1)
M Miscellaneous Manufacturing (1)
P Printing/Publishing (1)
28-3
-------
Table 28-1. Geographical Information for the Wisconsin Monitoring Site
Site
Code
HOW
AQS Code
55-027-0007
Location
Horicon
County
Dodge
Micro- or
Metropolitan
Statistical Area
Beaver Dam, WI
MSA
Latitude
and
Longitude
43.466111,
-88.621111
Land Use
Agricultural
Location
Setting
Rural
Additional Ambient Monitoring Information1
CO, SO2, NOy, NO, VOC, Carbonyl compounds, O3,
Meteorological parameters, PM10, PM10 Metals,
PM25, and PM25 Speciation, SVOC, PM Coarse.
BOLD ITALICS = EPA-designated NATTS Site.
to
oo
-------
The new NATTS site is located just north of the town of Horicon, in southwest
Wisconsin, within the boundaries of the Horicon Marsh Wildlife Area. The new location is
approximately 5 miles northwest of the old location. HOWI is located about 46 miles northwest
of Milwaukee and roughly 48 miles northeast of Madison. The surrounding area is rural and
agricultural in nature, although a residential subdivision is located just south of the site. The
HOWI monitoring site serves as a rural background site. However, the area is impacted by
nearby urban areas, and thus, could show the impacts on the wildlife sanctuary. State
Highway 28, which can be seen on the lower right-hand side of Figure 28-1, is the closest major
roadway. Figure 28-2 shows that a cluster of the point sources is located just south and west of
HOWI, in the town of Horicon. The closest point source near HOWI is an automobile and truck
manufacturing facility. The source categories with the most emissions sources are fabricated
metal products facilities; aircraft operations, which include airports as well as small runways,
heliports, and landing pads; and food processing/agricultural facilities.
Table 28-2 presents information related to mobile source activity, such as population,
traffic, VMT, and estimated vehicle ownership information for the area surrounding the
Wisconsin monitoring site. Table 28-2 also includes a vehicle registration-to-county population
ratio (vehicles-per-person). In addition, the population within 10 miles of the site is presented.
An estimate of 10-mile vehicle ownership was calculated by applying the county-level vehicle
registration-to-population ratio to the 10-mile population surrounding the monitoring site.
Table 28-2 also contains annual average daily traffic information. Finally, Table 28-2 presents
the daily VMT for Dodge County.
Table 28-2. Population, Motor Vehicle, and Traffic Information for the Wisconsin
Monitoring Site
Site
HOWI
Estimated
County
Population1
88,748
County-level
Vehicle
Registration2
98,211
Vehicles per
Person
(Registration:
Population)
1.11
Population
within 10
miles3
21,539
Estimated
10-mile
Vehicle
Ownership
23,836
Annual
Average
Daily
Traffic4
5,000
County-
level
Daily
VMT5
2,659,643
1 County-level population estimate reflects data from the U.S. Census Bureau (Census Bureau, 2011)
2 County-level vehicle registration reflects 2010 data from the Wisconsin DOT (WI DOT, 2010)
310-mile population estimate reflects 2011 data from Xionetic (Xionetic, 2011)
4 Annual Average Daily Traffic reflects 2008 data from the Wisconsin DOT (WI DOT, 2008)
5 County-level VMT reflects 2010 data from the Wisconsin DOT (WI DOT, 2011)
BOLD ITALICS = EPA-designated NATTS Site.
28-5
-------
Observations from Table 28-2 include the following:
• Dodge County's population is in the bottom-third compared to other counties with
NMP sites. This is also true of its 10-mile population.
• The county-level vehicle registration is also on the low end compared to other
counties with NMP sites. This is also true of its estimated 10-mile vehicle ownership.
• The vehicle-per-person ratio is slightly greater than one vehicle per person. This ratio
ranks among the higher ratios for NMP sites.
• The traffic volume experienced near HOWI is also on the low end compared to other
NMP monitoring sites. The traffic estimate used was for State Road 28 near State
Road 33 on the east side of Horicon.
• The Dodge County daily VMT is on the low side compared to other counties with
NMP sites (where VMT data were available).
28.2 Meteorological Characterization
The following sections characterize the meteorological conditions near the monitoring
site in Wisconsin on sample days, as well as over the course of the year.
28.2.1 Climate Summary
The town of Horicon is located in southeast Wisconsin, between the towns of West Bend
and Beaver Dam, and about 40 miles west of Lake Michigan. This area is far enough inland to
limit some of the moderating influences of Lake Michigan on the area's climate. This area
experiences a highly variable, continental climate as weather systems frequently track across the
region. Precipitation falls predominantly in the spring and summer months. Winters are cold and
predominantly dry, although snowfall is common. Lake effect snows can occur with winds with
a northeasterly and easterly component, although lake effect snows are often reduced this far
inland. Summers tend to be mild, although southerly winds out of the Gulf of Mexico can
occasionally advect warm, humid air into the area (Bair, 1992).
28-6
-------
28.2.2 Meteorological Conditions in 2010
Hourly meteorological data from the NWS weather station nearest this site were retrieved
for December 2009 and all of 2010 (NCDC, 2009 and 2010) to correspond with the sampling
period covered in this report. The closest weather station is located at Dodge County Airport
(WBAN 04898). Additional information about the Dodge County weather station, such as the
distance between the site and the weather station, is provided in Table 28-3. These data were
used to determine how meteorological conditions on sample days vary from normal conditions
throughout the sample period.
Table 28-3 presents average temperature (average maximum and average daily), moisture
(average dew point temperature, average wet bulb temperature, and average relative humidity),
pressure (average sea level pressure), and wind (average scalar wind speed) information for days
samples were collected and for the entire sample period. Also included in Table 28-3 is the
95 percent confidence interval for each parameter. As shown in Table 28-3, average
meteorological conditions on sample days were representative of average weather conditions
throughout the sample period.
28.2.3 Back Trajectory Analysis
Figure 28-3 is the composite back trajectory map for days on which samples were
collected at the HOWI monitoring site. Included in Figure 28-3 are four back trajectories per
sample day. Figure 28-4 is the corresponding cluster analysis for 2010. An in-depth description
of these maps and how they were generated is presented in Section 3.5.2.1. For the composite
map, each line represents the 24-hour trajectory along which a parcel of air traveled toward the
monitoring site on a given sample day and time. For the cluster analysis, each line corresponds to
a back trajectory representative of a given cluster of trajectories. For both maps, each concentric
circle around the site in Figures 28-3 and 28-4 represents 100 miles.
28-7
-------
to
oo
oo
Table 28-3. Average Meteorological Conditions near the Wisconsin Monitoring Site
Closest NWS
Station
(WBAN and
Coordinates)
Distance
and
Direction
from Site
Average
Type1
Average
Maximum
Temperature
(°F)
Average
Temperature
(°F)
Average
Dew Point
Temperature
(°F)
Average
Wet Bulb
Temperature
(°F)
Average
Relative
Humidity
(%)
Average
Sea Level
Pressure
(mb)
Average
Scalar Wind
Speed
(kt)
Horicon, Wisconsin - HOWI
Dodge County
Airport
04898
(43.43, -88.70)
4.64
miles
229°
(SW)
Sample
Day
Sample
Period
56.2
±5.5
55.5
+ 2.2
47.3
±5.1
47.2
+ 2.1
36.6
±4.7
36.7
+ 1.9
42.2
±4.6
42.3
+ 1.9
69.7
±2.9
70.6
+ 1.3
NA
NA
6.1
±0.7
6.7
+ 0.4
Sample day averages are highlighted to help differentiate the sample day averages from the full-year averages.
NA = Sea level pressure was not recorded at the Dodge County Airport.
-------
Figure 28-3. Composite Back Trajectory Map for HOWI
Figure 28-4. Back Trajectory Cluster Map for HOWI
28-9
-------
Observations from Figures 28-3 and 28-4 for HOWI include the following:
• Back trajectories originated from a variety of directions at HOWI, although less
frequently from the east.
• The 24-hour air shed domain for HOWI is similar in size to many other NMP
monitoring sites. The farthest away a trajectory originated was Manitoba, Canada, or
approximately 650 miles away. However, the average trajectory length was 256 miles
and most trajectories (81 percent) originated within 400 miles of the site.
• The cluster analysis shows that 32 percent of the back trajectories originated from the
north of HOWI, although of varying distances. Another 28 percent of back
trajectories are represented by the short trajectory originating to the west of the site.
The individual back trajectories represented by this cluster trajectory originated from
within 200 miles of the site and originated from the northwest, west, and southwest.
Longer back trajectories originating from the west to northwest of the site account for
13 percent of trajectories; back trajectories originating from the south to southwest
account for 15 percent; and back trajectories originating from the southeast of the site
account for another 13 percent of the back trajectories.
28.2.4 Wind Rose Comparison
Hourly wind data from the NWS weather station at Dodge County Airport near HOWI
were uploaded into a wind rose software program to produce customized wind roses, as
described in Section 3.5.2.2. A wind rose shows the frequency of wind directions using "petals"
positioned around a 16-point compass, and uses different colors to represent wind speeds.
Figure 28-5 presents three different wind roses for the HOWI monitoring site. First, a
historical wind rose representing 2003 to 2009 is presented, which shows the predominant
surface wind speed and direction over an extended period of time. Second, a wind rose
representing wind observations for the entire sampling period is presented. Next, a wind rose
representing days on which samples were collected during the sample period is presented. These
can be used to determine if wind observations on sample days were representative of conditions
experienced over the entire year and historically. Finally, a map showing the distance between
the NWS station and the monitoring site is presented, which may be useful for identifying
topographical influences that may affect the meteorological patterns experienced at each
location.
28-10
-------
Figure 28-5. Wind Roses for the Dodge County Airport Weather Station near HOWI
2003-2009 Historical Wind Rose
Sample Period Wind Rose
WEST
Sample Day Wind Rose
Distance between HOWI and NWS Station
28-11
-------
Observations from Figure 28-5 for MVWI include the following:
• The Dodge County Airport weather station is located approximately 4.6 miles
southwest of HOWL
• The historical wind rose shows that winds from a variety of directions were observed
near HOWL Winds from the south, southwest quadrant, and west account for the
most wind observations. The strongest wind speeds were associated with southerly to
west-southwesterly winds. Calm winds (<2 knots) were observed for nearly 15
percent of the hourly measurements.
• The wind patterns shown on the 2010 wind rose resemble the historical wind patterns,
although winds from the north were observed more frequently.
• The sample day wind rose resembles both the historical and sample period wind
roses, although winds from the northeast quadrant were observed even less frequently
and the calm rate was slightly higher.
28.3 Pollutants of Interest
Site-specific "pollutants of interest" were determined for the Wisconsin monitoring site
in order to allow analysts and readers to focus on a subset of pollutants through the context of
risk. Each pollutant's preprocessed daily measurement was compared to its associated risk
screening value. If the concentration was greater than the risk screening value, then the
concentration "failed the screen." Pollutants of interest are those for which the individual
pollutant's total failed screens contribute to the top 95 percent of the site's total failed screens. In
addition, if any of the NATTS MQO Core Analytes measured by the monitoring site did not
meet the pollutant of interest criteria based on the preliminary risk screening, that pollutant was
added to the list of site-specific pollutants of interest. A more in-depth description of the risk
screening process is presented in Section 3.2.
Table 28-4 presents HOWI's pollutants of interest. The pollutants that failed at least one
screen and contributed to 95 percent of the total failed screens for the monitoring site are shaded.
NATTS MQO Core Analytes are bolded. Thus, pollutants of interest are shaded and/or bolded.
HOWI sampled for PAH and hexavalent chromium beginning in December 2009, but stopped
sampling PAH in June 2010.
28-12
-------
Table 28-4. Risk Screening Results for the Wisconsin Monitoring Site
Pollutant
Screening
Value
(Ug/m3)
#of
Failed
Screens
#of
Measured
Detections
%of
Screens
Failed
% of Total
Failures
Cumulative
%
Contribution
Horicon, Wisconsin - HOWI
Naphthalene
0.029
Total
10
10
31
31
32.26
32.26
100.00
100.00
Observations from Table 28-4 include the following:
• Naphthalene was the only pollutant to fail screens for HOWI.
• Naphthalene was detected in all 31 valid samples collected at HOWI and failed
roughly one-third of screens.
• Naphthalene was identified as a pollutant of interest for HOWI, based on the risk
screening process. However, hexavalent chromium and benzo(a)pyrene were added to
HOWFs pollutants of interest because they are also NATTS MQO Core Analytes,
even though they did not fail any screens. These two pollutants are not shown in
Table 28-4.
28.4 Concentrations
This section presents various concentration averages used to characterize pollution levels
at the Wisconsin monitoring site. Concentration averages are provided for the pollutants of
interest for the HOWI monitoring site, where applicable. Concentration averages for select
pollutants are also presented graphically for the site, where applicable, to illustrate how the site's
concentrations compare to the program-level averages. In addition, concentration averages for
select pollutants are presented from previous years of sampling in order to characterize
concentration trends at the site, where applicable. Additional site-specific statistical summaries
are provided in Appendices M and O.
28.4.1 2010 Concentration Averages
Quarterly and annual concentration averages were calculated for the pollutants of interest
for the Wisconsin site, as described in Section 3.1. The quarterly average of a particular
pollutant is simply the average concentration of the preprocessed daily measurements over a
given calendar quarter. Quarterly average concentrations include the substitution of zeros for all
non-detects. A site must have a minimum of 75 percent valid samples of the total number of
samples possible within a given quarter for a quarterly average to be calculated. An annual
average includes all measured detections and substituted zeros for non-detects for the entire year
28-13
-------
of sampling. Annual averages were calculated for pollutants where three valid quarterly averages
could be calculated and where method completeness was greater than or equal to 85 percent, as
presented in Section 2.4. Quarterly and annual average concentrations for HOWI are presented in
Table 28-5, where applicable. Note that if a pollutant was not detected in a given calendar
quarter, the quarterly average simply reflects "0" because only zeros substituted for non-detects
were factored into the quarterly average concentration.
Table 28-5. Quarterly and Annual Average Concentrations of the Pollutants of
Interest for the Wisconsin Monitoring Site
Pollutant
#of
Measured
Detections
vs. # of
Samples
1st
Quarter
Average
(ng/m3)
2nd
Quarter
Average
(ng/m3)
3rd
Quarter
Average
(ng/m3)
4th
Quarter
Average
(ng/m3)
Annual
Average
(ng/m3)
Horicon, Wisconsin - HOWI
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
19/31
40/63
31/31
0.14
±0.05
0.01
±0.01
34.54
±9.35
0.01
±0.01
0.02
±0.01
17.89
±5.63
NA
0.02
±<0.01
NA
NA
0.01
±<0.01
NA
NA
0.01
±<0.01
NA
NA = Not available due to the criteria for calculating a quarterly and/or annual average.
Observations for HOWI from Table 28-5 include the following:
• The two December 2009 samples were factored into the first quarter 2010 averages.
• Because PAH sampling was discontinued in June 2010, annual averages
concentrations could not be calculated for naphthalene and benzo(a)pyrene. However,
Appendix M provides the pollutant-specific average concentration for all valid PAH
samples collected over the entire sample period.
• The first quarter benzo(a)pyrene average is significantly higher than the second
quarter average concentration. The three highest concentrations of this pollutant were
measured in February. Of the 19 measure detections of this pollutant at HOWI, 16
were measured during the first quarter of 2010 (including December 2009).
• A similar trend is shown in the quarterly averages of naphthalene. Concentrations of
naphthalene ranged from 7.18 ng/m3 to 80.2 ng/m3, with the five highest
concentrations measured in December 2009 and February 2010.
• Concentrations of hexavalent chromium ranged from 0 ng/m3 to 0.0667 ng/m3, with
the maximum concentration measured on June 25, 2010. The three highest
concentrations of this pollutant were measured during the second quarter of 2010.
28-14
-------
28.4.2 Concentration Comparison
In order to better illustrate how a site's annual concentration averages compare to the
program-level averages, a site-specific box plot was created for the selected NATTS MQO Core
Analytes listed in Section 3.5.3, where applicable. Thus, a box plot for hexavalent chromium was
created for HOWL Figure 28-6 overlays the site's minimum, annual average, and maximum
concentrations onto the program-level minimum, first quartile, average, median, third quartile,
and maximum concentrations, as described in Section 3.5.3.
Figure 28-6. Program vs. Site-Specific Average Hexavalent Chromium Concentration
Program Max Concentration = 3.51 ng/m3 j
0.3 0.45
Concentration (ng/m3)
Program: 1st Quartile
•
Site:
Site Average
0
2nd Quartile 3rd Quartile 4th Quartile Ave
Site Minimum/Maximum
^^^^~
;rage
Observations from Figure 28-6 include the following:
• The scale for hexavalent chromium has been adjusted in Figure 28-6 as a result of
a relatively large maximum concentration. The program-level maximum
concentration (3.51 ng/m3) is not shown directly on the box plot in order to allow
for the observation of data points at the lower end of the concentration range;
thus, the scale has been reduced to 0.75 ng/m3. Also note that the first quartile for
this pollutant is zero and is not visible on this box plot. Figure 28-6 shows that the
annual average concentration of hexavalent chromium for HOWI is less than the
program-level average and median concentrations. The maximum concentration
measured at HOWI is well below the program maximum concentration. There
were several non-detects of hexavalent chromium measured at HOWI.
28.4.3 Concentration Trends
A site-specific trends evaluation was completed for sites that have sampled one or more
of the selected NATTS MQO Core Analytes for 5 consecutive years or longer, as described in
Section 3.5.3. Because HOWI did not begin sampling under the NMP until December 2009, a
trends analysis was not conducted for this site.
28-15
-------
28.5 Additional Risk Screening Evaluations
The following risk screening evaluations were conducted to characterize risk at the
Wisconsin monitoring site. Refer to Sections 3.3 and 3.5.5 for definitions and explanations
regarding the various risk factors, time frames, and calculations associated with these risk
screenings.
28.5.1 Risk Screening Assessment Using MRLs
A noncancer risk screening was conducted by comparing the concentration data from the
Wisconsin monitoring site to the ATSDR acute, intermediate, and chronic MRLs, where
available. As described in Section 3.3, acute risk results from exposures of 1 to 14 days;
intermediate risk results from exposures of 15 to 364 days; and chronic risk results from
exposures of 1 year or greater. The preprocessed daily measurements of the pollutants of interest
were compared to the acute MRL; the quarterly averages were compared to the intermediate
MRL; and the annual averages were compared to the chronic MRL, where applicable.
None of the measured detections or time-period average concentrations of the pollutants
of interest for the Wisconsin monitoring site, where they could be calculated, were greater than
their respective MRL noncancer health risk benchmarks. This is also true for pollutants not
identified as pollutants of interest for HOWL
28.5.2 Cancer and Noncancer Surrogate Risk Approximations
For the pollutants of interest for the Wisconsin monitoring site and where annual average
concentrations could be calculated, risk was further examined by calculating cancer and
noncancer surrogate risk approximations (refer to Section 3.5.5.2 regarding the criteria for
annual averages and how cancer and noncancer surrogate risk approximations are calculated).
Annual averages, cancer UREs and/or noncancer RfCs, and cancer and noncancer surrogate risk
approximations are presented in Table 28-6, where applicable.
28-16
-------
Table 28-6. Cancer and Noncancer Surrogate Risk Approximations for the Wisconsin
Monitoring Site
Pollutant
Cancer
URE
(Ug/m3)1
Noncancer
RfC
(mg/m3)
#of
Measured
Detections
vs. # of
Samples
Annual
Average
(ng/m3)
Cancer Risk
Approximation
(in-a-million)
Noncancer
Risk
Approximation
(HQ)
Horicon, Wisconsin - HOWI
Benzo(a)pyrene
Hexavalent Chromium
Naphthalene
0.00176
0.012
0.000034
0.0001
0.003
19/31
40/63
31/31
NA
0.01
±0.01
NA
NA
0.16
NA
NA
0.01
NA
NA = Not available due to the criteria for calculating an annual average.
— = a Cancer URE or Noncancer RfC is not available.
Observations for HOWI from Table 28-6 include the following:
• The cancer risk approximation for hexavalent chromium is less than 1.0 in-a-million
(0.16 in-a-million).
• The noncancer risk approximation for hexavalent chromium (<0.01) is well below the
level of concern (an HQ of 1.0).
• Annual averages, and therefore cancer and noncancer risk approximations, could not
be calculated for benzo(a)pyrene and naphthalene.
28.5.3 Risk-Based Emissions Assessment
In addition to the risk screenings discussed above, Tables 28-7 and 28-8 present a risk-
based evaluation of county-level emissions based on cancer and noncancer toxicity, respectively.
Table 28-7 presents the 10 pollutants with the highest emissions from the 2008 NEI, the 10
pollutants with the highest toxicity-weighted emissions, and the 10 pollutants with the highest
cancer risk approximations (in-a-million), as calculated from the annual averages. Table 28-8
presents similar information, but identifies the 10 pollutants with the highest noncancer risk
approximations (HQ), also calculated from annual averages.
28-17
-------
Table 28-7. Top 10 Emissions, Toxicity-Weighted Emissions, and Cancer Risk Approximations for Pollutants with Cancer UREs for
the Wisconsin Monitoring Site
Top 10 Total Emissions for Pollutants with Cancer
Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Cancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Cancer
Toxicity
Weight
Top 10 Cancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Cancer Risk
Approximation
Pollutant (in-a-million)
Horicon, Wisconsin (Dodge County) - HOWI
Benzene
Formaldehyde
Acetaldehyde
Ethylbenzene
1,3 -Butadiene
Naphthalene
POM, Group 2b
Tetrachloroethylene
Trichloroethylene
Dichloromethane
90.74
47.98
31.44
26.35
9.67
6.11
1.31
1.00
0.85
0.82
Benzene
Formaldehyde
Hexavalent Chromium, PM
1,3 -Butadiene
Naphthalene
POM, Group 3
POM, Group 2b
POM, Group 5a
Acetaldehyde
Ethylbenzene
7.08E-04
6.24E-04
3.70E-04
2.90E-04
2.08E-04
2.07E-04
1.15E-04
1.12E-04
6.92E-05
6.59E-05
Hexavalent Chromium 0.16
to
oo
oo
-------
Table 28-8. Top 10 Emissions, Toxicity-Weighted Emissions, and Noncancer Risk Approximations for Pollutants with Noncancer
RfCs for the Wisconsin Monitoring Site
Top 10 Total Emissions for Pollutants
with Noncancer Risk Factors
(County-Level)
Pollutant
Emissions
(tpy)
Top 10 Noncancer Toxicity-Weighted Emissions
(County-Level)
Pollutant
Noncancer Toxicity
Weight
Top 10 Noncancer Risk Approximations Based on
Annual Average Concentrations
(Site-Specific)
Noncancer Risk
Approximation
Pollutant (HQ)
Horicon, Wisconsin (Dodge County) - HOWI
Toluene
Xylenes
Methanol
Benzene
Formaldehyde
Methyl isobutyl ketone
Acetaldehyde
Hexane
Hydrochloric acid
Ethylbenzene
329.03
146.74
98.40
90.74
47.98
33.47
31.44
31.15
26.97
26.35
Acrolein
Manganese, PM
Cyanide Compounds, gas
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Benzene
Chlorine
Naphthalene
Xylenes
163,416.51
6,223.90
5,737.27
4,895.94
4,837.37
3,493.32
3,024.81
2,097.57
2,036.14
1,467.40
Hexavalent Chromium O.01
to
oo
-------
The pollutants listed in Tables 28-7 and 28-8 are limited to those that have cancer and
noncancer risk factors, respectively. As a result, although the actual value of the emissions is the
same, the highest emitted pollutants in the cancer table may be different from the noncancer
table. The cancer and noncancer surrogate risk approximations based on each site's annual
averages are limited to those pollutants for which each respective site sampled. As discussed in
Section 28.3, HOWI sampled for PAH and hexavalent chromium. In addition, the cancer and
noncancer surrogate risk approximations are limited to those pollutants with enough data to meet
the criteria for annual averages to be calculated. A more in-depth discussion of this analysis is
provided in Section 3.5.5.3.
Observations from Table 28-7 include the following:
• Benzene, formaldehyde, and acetaldehyde are the highest emitted pollutants with
cancer UREs in Dodge County.
• Benzene is the pollutant with the highest toxi city-weighted emissions (of the
pollutants with cancer UREs), followed by formaldehyde and hexavalent chromium.
• Seven of the highest emitted pollutants in Dodge County also have the highest
toxicity-weighted emissions.
• Hexavalent chromium, which is the only pollutant for which a cancer risk
approximation could be calculated, has the third highest toxicity-weighted emissions
for Dodge County, but is not among the highest emitted.
• Naphthalene, one of HOWI's pollutants of interest, appears on both emissions-based
lists. Benzo(a)pyrene is part of POM, Group 5a, which ranks eighth for toxicity-
weighted emissions but is not among the highest emitted pollutants.
• POM, Group 2b ranks seventh for both total emissions and toxicity-weighted
emissions. POM, Group 2b includes several PAH sampled for at HOWI including
acenaphthylene, fluoranthene, and perylene. None of the PAH included in POM,
Group 2b were identified as pollutants of interest for HOWI. POM Group 3 ranks
sixth for toxicity-weighted emissions. POM Group 3 does not include any pollutants
sampled at HOWI.
Observations from Table 28-8 include the following:
• Toluene, xylenes, and methanol are the highest emitted pollutants with noncancer
RfCs in Dodge County.
• The pollutants with the highest toxicity-weighted emissions (of the pollutants with
noncancer RfCs) are acrolein, manganese, and cyanide compounds (gaseous).
28-20
-------
• Four of the highest emitted pollutants in Dodge County also have the highest toxicity-
weighted emissions.
• None of HOWFs pollutants of interest appear among the highest emitted pollutants
(with noncancer RfCs) in Dodge County. Naphthalene, however, ranks ninth for
toxicity-weighted emissions.
28.6 Summary of the 2010 Monitoring Data for HOWI
Results from several of the data treatments described in this section include the
following:
»«» Sampling for hexavalent chromium and PAH began in December 2009 at HOWI.
However, PAH sampling was discontinued in June 2010.
»«» Naphthalene was the only pollutant to fail at least one screen. Benzo(a)pyrene and
hexavalent chromium were added to the pollutants of interest for HOWI because they
are also NATTSMQO Core Analytes.
»«» None of the preprocessed daily measurements and none of the quarterly or annual
average concentrations for the pollutants of interest, where they could be calculated,
were greater than their associatedMRL noncancer health risk benchmarks.
28-21
-------
29.0 Data Quality
This section discusses the data quality of the ambient air measurements that constitute the
2010 NMP dataset. In accordance with the Data Quality Objectives (DQOs) presented in ERG's
EPA-approved QAPP (ERG, 2009), the following data quality indicators were assessed:
completeness, precision, and accuracy (also called bias).
The quality assessments presented in this section show that the 2010 monitoring data are
of a known and high quality. As indicators of the reliability and representativeness of
experimental measurements, both precision and accuracy are considered when interpreting
ambient air monitoring data. The method precision for collocated and duplicate analyses met the
precision DQO of 15 percent coefficient of variation (CV) for all methods except SNMOC,
which was 15.29 percent CV. The analytical precision level for replicate analyses met the DQOs.
Audit samples show that ERG is meeting the accuracy requirements of the NATTS TAD (EPA,
2009b).
29.1 Completeness
Completeness refers to the number of valid samples actually collected and analyzed
compared to the number of total samples scheduled to be collected and analyzed. The DQO for
completeness based on the EPA-approved QAPP specifies that at least 85 percent of samples
collected at a given monitoring site must be analyzed successfully to be considered sufficient for
data trends analysis (ERG, 2009). Completeness statistics are presented in Section 2.4. The DQO
of 85 percent completeness was met by all but five out 126 site-method combinations.
29.2 Method Precision
Precision defines the level of agreement between independent measurements performed
according to identical protocols and procedures. Method precision, which includes sampling and
analytical precision, quantifies random errors associated with collecting ambient air samples and
analyzing the samples in the laboratory. Method precision is evaluated by comparing
concentrations measured in duplicate or collocated samples. A duplicate sample is a sample
collected simultaneously with a primary sample using the same sampling system (i.e., two
separate samples through the same sampling system at the same time). This simultaneous
collection is typically achieved by teeing the line from the sampler to two canisters and doubling
the flow rate applied to achieve integration over the 24-hour collection period. Collocated
29-1
-------
samples are samples collected simultaneously using two independent collection systems at the
same location at the same time.
Both approaches provide valuable, but different, assessments of method precision:
• Analysis of duplicate samples provides information on the potential for variability (or
precision) expected from a single collection system (intra-system assessment).
• Analysis of collocated samples provides information on the potential for variability
(or precision) expected between different collection systems (inter-system
assessment).
During the 2010 sampling year, duplicate and collocated samples were collected on at
least 10 percent of the scheduled sample days, as outlined in the QAPP. These samples were then
analyzed in replicate. Replicate measurements are repeated analyses performed on a duplicate or
collocated pair of samples. Collocated systems were not provided under the national contract for
sites sampling PAH and were the responsibility of the participating agency. Thus, collocated
samples were not collected for most PAH sites because few sites had collocated samplers.
Therefore, the method precision data for PAH is based on only five sites for 2010.
Method precision was calculated by comparing the concentrations of the
duplicates/collocates for each compound. The CV for duplicate or collocated samples was
calculated for each pollutant and each site. The following approach was employed to estimate
how closely the collected and analyzed samples agree with one another:
Coefficient of Variation (CV) provides a relative measure of data dispersion compared to the
mean. A coefficient of variation of one percent would indicate that the analytical results
could vary slightly due to sampling error, while a variation of 50 percent means that the
results are more imprecise.
(p-r) ^2
CV = 100 x
>J 2n
Where:
p = the primary result from a duplicate or collocated pair;
r = the secondary result from a duplicate or collocated pair;
n = the number of valid data pairs (the 2 adjusts for the fact that there are two
values with error).
29-2
-------
Coefficients of variation were calculated from every pair of duplicate or collocated
samples collected during the program year. However, only results at or above the MDL were
used in these calculations. This is a change in procedure compared to NMP reports in previous
years, where /^ MDL was substituted for non-detects. To make an overall estimate of method
precision, program-average CVs were calculated for each pollutant by averaging the values from
the individual duplicate or collocated analyses. The expression "average variability" for a given
dataset refers to the average CV.
Table 29-1 presents the 2010 NMP method precision for VOC, SNMOC, carbonyl
compounds, metals, hexavalent chromium, and PAH, presented as average CV (expressed as a
percentage). Each method met the program DQO for method precision, which is 15 percent CV,
except SNMOC, which was slightly above the program DQO (15.29 percent and bolded in
Table 29-1). The average CV for the SNMOC method may be driven largely by the individual
pollutant concentration differences. Differences in very small concentrations may yield relatively
large CVs (i.e., the percent difference between 0.001 ppbC and 0.002 ppbC is 100 percent).
Table 29-1. Method Precision by Analytical Method
Method/Pollutant
Group
VOC
(TO-15)
SNMOC
Carbonyl Compounds
(TO-11A)
Metals Analysis
(Method IO-3. 5)
Hexavalent Chromium
(EPA-approved method)
PAH
(TO-13)
DQO
Average
Coefficient of
Variation
(%)
14.11
15.29
6.17
12.97
14.89
13.18
Number of
Pairs
2,935
1,191
1,209
1,178
112
378
15.00 percent CV
Tables 29-2 through 29-7 present method precision for VOC, SNMOC, carbonyl
compounds, metals, hexavalent chromium, and PAH, respectively, as the average CVs per
pollutant per site, per site, and per method. Also included in these tables is the number of
duplicated and/or collocated pairs included in the CV calculations. CVs exceeding the 15 percent
29-3
-------
DQO are bolded in each table. The shaded rows in each table identify the NATTS MQO Core
Analytes for each method.
29.2.1 VOC Method Precision
Table 29-2 presents the method precision for all duplicate and collocated VOC samples
as the average CV per pollutant per site, the average CV per site, and the overall average CV for
NMP sites sampling VOC. The average pollutant-specific CV ranged from 0 percent for a few
pollutants for several sites to 99.69 percent (carbon disulfide for PROK). The site-specific
average CV ranged from 6.66 percent for S4MO to 34.13 percent for GPCO. The overall method
precision for VOC was 14.11 percent.
Table 29-2. VOC Method Precision: Average Coefficient of Variation Based on Duplicate
and Collocated Samples by Site
Pollutant
Acetylene
Acrylonitrile
fer/-Amyl Methyl Ether
Benzene
Bromochloromethane
Bromodichloromethane
Bromoform
Bromomethane
1,3 -Butadiene
Carbon Disulfide
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
Chloromethane
Chloromethylbenzene
Chloroprene
Dibromochloromethane
1 ,2-Dibromoethane
/w-Dichlorobenzene
o-Dichlorobenzene
£>-Dichlorobenzene
Dichlorodifluoromethane
1 , 1 -Dichloroethane
1,2-Dichloroethane
1 , 1 -Dichloroethene
#of
Pairs
131
3
0
131
0
4
0
28
105
86
126
0
12
76
131
0
0
2
0
2
0
40
131
0
9
0
BTUT
12.39
28.28
NA
15.46
NA
NA
NA
59.75
9.14
15.26
33.19
NA
NA
7.16
5.59
NA
NA
NA
NA
28.02
NA
14.66
5.91
NA
NA
NA
BURVT
20.04
NA
NA
17.83
NA
NA
NA
NA
14.41
43.45
24.73
NA
16.61
9.08
7.61
NA
NA
NA
NA
NA
NA
3.85
6.23
NA
5.04
NA
CHNJ
10.35
NA
NA
11.04
NA
NA
NA
14.89
5.12
10.09
18.19
NA
NA
8.95
9.68
NA
NA
NA
NA
NA
NA
NA
9.75
NA
NA
NA
DEMI
2.36
NA
NA
5.25
NA
NA
NA
5.24
2.01
15.10
6.13
NA
10.88
45.76
8.75
NA
NA
NA
NA
NA
NA
NA
1.45
NA
12.12
NA
ELNJ
4.26
39.12
NA
15.21
NA
NA
NA
4.13
6.28
4.51
11.39
NA
4.56
5.69
5.53
NA
NA
NA
NA
NA
NA
7.53
4.17
NA
NA
NA
GLKY
4.53
NA
NA
66.70
NA
NA
NA
25.38
16.76
23.33
33.91
NA
NA
0
8.22
NA
NA
NA
NA
NA
NA
NA
6.45
NA
NA
NA
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-4
-------
Table 29-2. VOC Method Precision: Average Coefficient of Variation Based on Duplicate
and Collocated Samples by Site (Continued)
Pollutant
cis- 1 ,2-Dichloroethy lene
trans- 1 ,2-Dichloroethy lene
Dichloromethane
1 ,2-Dichloropropane
cis- 1 ,3 -Dichloropropene
trans- 1 ,3 -Dichloropropene
Dichlorotetrafluoroethane
Ethyl Aery late
Ethyl ferMSutyl Ether
Ethylbenzene
Hexachloro- 1 ,3 -butadiene
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Methyl Methacrylate
Methyl ferMSutyl Ether
w-Octane
Propylene
Styrene
1 , 1 ,2,2-Tetrachloroethane
Tetrachloroethylene
Toluene
1 ,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1 , 1 ,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Trichlorotrifluoroethane
1 ,2,4-Trimethylbenzene
1 ,3 ,5-Trimethylbenzene
Vinyl chloride
7w,£>-Xylene
o-Xylene
Average by Site
#of
Pairs
0
1
131
0
0
0
124
0
0
129
0
131
104
0
0
103
131
103
0
68
131
0
2
0
7
131
131
125
108
0
130
128
14.11
BTUT
NA
NA
21.46
NA
NA
NA
2.86
NA
NA
15.15
NA
20.74
16.53
NA
NA
11.63
8.10
19.46
NA
15.46
11.28
NA
17.07
NA
24.36
5.46
6.66
15.31
16.17
NA
16.08
15.37
16.47
BURVT
NA
NA
28.95
NA
NA
NA
7.77
NA
NA
12.40
NA
27.95
33.08
NA
NA
8.46
11.91
22.38
NA
5.68
18.45
NA
NA
NA
NA
6.14
6.45
17.84
22.47
NA
9.95
15.63
75.72
CHNJ
NA
NA
7.99
NA
NA
NA
12.11
NA
NA
17.82
NA
17.48
13.19
NA
NA
18.20
9.54
9.68
NA
7.44
20.45
NA
NA
NA
NA
9.07
10.78
17.78
11.63
NA
24.97
20.21
13.06
DEMI
NA
NA
33.85
NA
NA
NA
1.75
NA
NA
2.77
NA
10.46
14.25
NA
NA
3.57
4.92
17.62
NA
1.95
2.93
NA
NA
NA
5.48
1.26
4.77
2.61
2.26
NA
3.32
4.21
8.32
ELNJ
NA
7.62
6.45
NA
NA
NA
7.19
NA
NA
7.02
NA
27.55
27.70
NA
NA
10.60
5.83
12.36
NA
7.20
8.79
NA
NA
NA
NA
3.86
4.58
7.13
10.92
NA
6.25
6.62
9.66
GLKY
NA
NA
33.36
NA
NA
NA
4.88
NA
NA
15.22
NA
23.29
41.87
NA
NA
12.41
14.39
28.59
NA
NA
7.66
NA
0
NA
NA
4.40
4.63
8.96
11.17
NA
9.61
7.78
16.54
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
29-5
-------
Table 29-2. VOC Method Precision: Average Coefficient of Variation Based on Duplicate
and Collocated Samples by Site (Continued)
Pollutant
Acetylene
Acrylonitrile
tert-Amyl Methyl Ether
Benzene
Bromochloromethane
Bromodichloromethane
Bromoform
Bromomethane
1,3 -Butadiene
Carbon Bisulfide
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
Chloromethane
Chloromethylbenzene
Chloroprene
Dibromochloromethane
1 ,2-Dibromoethane
/w-Dichlorobenzene
o-Dichlorobenzene
/>-Dichlorobenzene
Dichlorodifluoromethane
1 , 1 -Dichloroethane
1 ,2-Dichloroethane
1 , 1 -Dichloroethene
cis- 1 ,2-Dichloroethylene
trans- 1 ,2-Dichloroethylene
Dichloromethane
1 ,2-Dichloropropane
cis- 1 ,3 -Dichloropropene
trans- 1 ,3 -Dichloropropene
Dichlorotetrafluoroethane
Ethyl Acrylate
Ethyl tert-Butyl Ether
Ethylbenzene
Hexachloro- 1 ,3 -butadiene
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Methyl Methacrylate
Methyl fer/-Butyl Ether
w-Octane
Propylene
#of
Pairs
131
3
0
131
0
4
0
28
105
86
126
0
12
76
131
0
0
2
0
2
0
40
131
0
9
0
0
1
131
0
0
0
124
0
0
129
0
131
104
0
0
103
131
GPCO
37.95
NA
NA
46.73
NA
NA
NA
14.76
16.53
43.09
28.60
NA
NA
4.56
7.23
NA
NA
NA
NA
NA
NA
NA
7.00
NA
34.57
NA
NA
NA
55.48
NA
NA
NA
4.95
NA
NA
52.78
NA
52.39
2.56
NA
NA
51.89
49.65
MWOK
7.04
NA
NA
7.29
NA
NA
NA
58.74
4.31
69.71
12.83
NA
NA
0
3.21
NA
NA
NA
NA
NA
NA
38.07
3.54
NA
4.04
NA
NA
NA
8.74
NA
NA
NA
2.27
NA
NA
7.27
NA
13.55
22.39
NA
NA
11.36
14.97
NBIL
14.64
NA
NA
5.92
NA
NA
NA
79.02
25.07
37.74
3.49
NA
NA
71.41
6.04
NA
NA
20.00
NA
NA
NA
44.56
3.19
NA
0
NA
NA
NA
55.84
NA
NA
NA
4.40
NA
NA
11.92
NA
27.09
49.27
NA
NA
16.06
10.39
NBNJ
7.76
NA
NA
13.01
NA
NA
NA
5.06
5.19
30.37
19.86
NA
5.24
3.13
8.95
NA
NA
NA
NA
NA
NA
14.63
8.93
NA
NA
NA
NA
NA
12.42
NA
NA
NA
3.91
NA
NA
9.97
NA
41.68
42.29
NA
NA
19.31
15.02
OCOK
14.59
NA
NA
8.83
NA
NA
NA
3.70
13.53
25.54
16.86
NA
66.99
0
8.55
NA
NA
NA
NA
NA
NA
36.53
7.76
NA
NA
NA
NA
NA
20.99
NA
NA
NA
8.31
NA
NA
8.48
NA
43.47
37.24
NA
NA
7.79
34.08
PANJ
5.30
NA
NA
6.57
NA
NA
NA
NA
6.97
34.05
3.29
NA
3.95
4.10
5.78
NA
NA
NA
NA
NA
NA
11.47
4.54
NA
NA
NA
NA
NA
10.20
NA
NA
NA
3.82
NA
NA
4.23
NA
10.07
12.24
NA
NA
9.31
6.09
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
29-6
-------
Table 29-2. VOC Method Precision: Average Coefficient of Variation Based on Duplicate
and Collocated Samples by Site (Continued)
Pollutant
Styrene
1 , 1 ,2,2-Tetrachloroethane
ITetrachloroethylene
Toluene
1 ,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1 , 1 ,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Trichlorotrifluoroethane
1 ,2,4-Trimethylbenzene
1,3,5 -Trimethy Ibenzene
Vinyl chloride
m,p-Xylene
o-Xylene
Average by Site
#of
Pairs
103
0
68
131
0
2
0
7
131
131
125
108
0
130
128
14.11
GPCO
56.80
NA
51.04
57.69
NA
NA
NA
NA
4.07
7.38
58.83
27.41
NA
57.62
55.89
34.13
MWOK
5.25
NA
1.47
7.78
NA
NA
NA
NA
3.24
3.64
17.62
7.96
NA
7.69
6.42
12. 98
NBIL
13.91
NA
12.30
19.71
NA
NA
NA
2.02
14.02
3.87
14.46
12.52
NA
15.34
13.83
22. 54
NBNJ
8.23
NA
4.67
45.68
NA
NA
NA
NA
9.93
8.68
20.43
14.64
NA
19.45
13.51
75.26
OCOK
12.93
NA
24.73
8.96
NA
NA
NA
NA
8.30
8.98
20.05
11.69
NA
10.27
11.68
17.81
PANJ
10.32
NA
5.59
3.36
NA
NA
NA
NA
8.31
5.76
6.99
9.26
NA
4.35
4.26
7.70
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
Table 29-2. VOC Method Precision: Average Coefficient of Variation Based on Duplicate
and Collocated Samples by Site (Continued)
Pollutant
Acetylene
Acrylonitrile
fer/-Amyl Methyl Ether
Benzene
Bromochloromethane
Bromodichloromethane
Bromoform
Bromomethane
1,3 -Butadiene
Carbon Bisulfide
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
Chloromethane
Chloromethy Ibenzene
Chloroprene
Dibromochloromethane
#of
Pairs
131
3
0
131
0
4
0
28
105
86
126
0
12
76
131
0
0
2
PROK
11.43
NA
NA
15.31
NA
NA
NA
14.99
15.69
99.69
8.46
NA
33.95
20.93
8.10
NA
NA
NA
PXSS
21.11
NA
NA
5.09
NA
NA
NA
NA
6.57
37.46
8.65
NA
NA
8.50
4.23
NA
NA
NA
S4MO
4.70
NA
NA
3.49
NA
NA
NA
13.69
4.86
10.43
6.12
NA
NA
3.89
5.68
NA
NA
NA
SEWA
2.75
NA
NA
5.29
NA
NA
NA
NA
10.02
20.61
6.91
NA
NA
17.69
3.97
NA
NA
NA
SPIL
7.51
NA
NA
16.32
NA
NA
NA
NA
6.70
32.59
11.49
NA
NA
5.26
3.18
NA
NA
NA
SSSD
7.26
NA
NA
5.38
NA
NA
NA
12.33
11.28
86.91
15.12
NA
NA
4.97
5.83
NA
NA
NA
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyle
29-7
-------
Table 29-2. VOC Method Precision: Average Coefficient of Variation Based on Duplicate
and Collocated Samples by Site (Continued)
Pollutant
1 ,2-Dibromoethane
/w-Dichlorobenzene
o-Dichlorobenzene
£>-Dichlorobenzene
Dichlorodifluoromethane
1 , 1 -Dichloroethane
1 ,2-Dichloroethane
1 , 1 -Dichloroethene
cis- 1 ,2-Dichloroethylene
trans- 1 ,2-Dichloroethylene
Dichloromethane
1 ,2-Dichloropropane
cis- 1 ,3 -Dichloropropene
tr ans- 1 , 3 -D ichloropropene
Dichlorotetrafluoroethane
Ethyl Aery late
Ethyl fer/-Butyl Ether
Ethylbenzene
Hexachloro- 1 ,3 -butadiene
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Methyl Methacrylate
Methyl ferMSutyl Ether
w-Octane
Propylene
Styrene
El , 1 ,2,2-Tetrachloroethane
Tetrachloroethylene
Toluene
1 ,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1 , 1 ,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Trichlorotrifluoroethane
1 ,2,4-Trimethylbenzene
1,3,5 -Trimethy Ibenzene
Vinyl chloride
7w,£>-Xylene
o-Xylene
Average by Site
#of
Pairs
0
2
0
40
131
0
9
0
0
1
131
0
0
0
124
0
0
129
0
131
104
0
0
103
131
103
0
68
131
0
2
0
7
131
131
125
108
0
130
128
14.11
PROK
NA
NA
NA
14.24
7.62
NA
NA
NA
NA
NA
18.82
NA
NA
NA
7.30
NA
NA
5.69
NA
25.86
10.44
NA
NA
7.20
7.50
4.33
NA
NA
7.45
NA
NA
NA
NA
7.75
6.47
8.91
11.70
NA
5.80
4.26
15.00
PXSS
NA
NA
NA
3.02
5.31
NA
NA
NA
NA
NA
25.66
NA
NA
NA
1.62
NA
NA
6.72
NA
18.01
15.61
NA
NA
15.29
11.47
28.05
NA
8.40
12.45
NA
NA
NA
NA
3.67
7.10
13.64
16.46
NA
7.53
7.00
11.94
S4MO
NA
NA
NA
3.29
4.88
NA
NA
NA
NA
NA
12.13
NA
NA
NA
6.31
NA
NA
2.21
NA
12.97
18.70
NA
NA
9.29
6.64
6.86
NA
1.10
2.29
NA
NA
NA
NA
4.50
4.96
9.94
9.35
NA
3.50
1.35
6.66
SEWA
NA
NA
NA
NA
2.38
NA
7.07
NA
NA
NA
42.04
NA
NA
NA
35.66
NA
NA
7.49
NA
48.12
31.08
NA
NA
26.27
15.88
31.83
NA
5.91
7.88
NA
NA
NA
NA
2.17
4.54
7.52
7.43
NA
5.58
5.63
14.47
SPIL
NA
NA
NA
12.20
3.72
NA
NA
NA
NA
NA
5.94
NA
NA
NA
3.93
NA
NA
13.32
NA
15.65
14.06
NA
NA
15.21
2.38
13.85
NA
7.56
12.34
NA
NA
NA
10.74
6.79
4.81
19.48
21.73
NA
14.96
16.08
11.45
SSSD
NA
NA
NA
7.07
3.48
NA
NA
NA
NA
NA
18.28
NA
NA
NA
2.36
NA
NA
20.26
NA
7.20
2.91
NA
NA
15.71
8.43
19.17
NA
4.99
25.24
NA
NA
NA
NA
2.83
2.71
28.48
13.41
NA
21.53
23.91
14.50
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-8
-------
Table 29-2. VOC Method Precision: Average Coefficient of Variation Based on Duplicate
and Collocated Samples by Site (Continued)
Pollutant
Acetylene
Acrylonitrile
tert-Amyl Methyl Ether
Benzene
Bromochloromethane
Bromodichloromethane
Bromoform
Bromomethane
1,3 -Butadiene
Carbon Bisulfide
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
Chloromethane
Chloromethylbenzene
Chloroprene
Dibromochloromethane
1 ,2-Dibromoethane
/w-Dichlorobenzene
o-Dichlorobenzene
/>-Dichlorobenzene
Dichlorodifluoromethane
1 , 1 -Dichloroethane
1 ,2-Dichloroethane
1 , 1 -Dichloroethene
cis- 1 ,2-Dichloroethylene
/ra«5-l,2-Dichloroethylene
Dichloromethane
1 ,2-Dichloropropane
cis- 1 , 3 -D ichloropropene
trans- 1 , 3 -Dichloropropene
Dichlorotetrafluoroethane
Ethyl Aery late
Ethyl tert-Butyl Ether
Ethylbenzene
Hexachloro- 1 , 3 -butadiene
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Methyl Methacrylate
Methyl tert-Butyl Ether
w-Octane
#of
Pairs
131
o
6
0
131
0
4
0
28
105
86
126
0
12
76
131
0
0
2
0
2
0
40
131
0
9
0
0
1
131
0
0
0
124
0
0
129
0
131
104
0
0
103
TMOK
6.08
15.71
NA
12.75
NA
NA
NA
5.06
8.64
49.86
26.30
NA
NA
0
6.09
NA
NA
NA
NA
4.88
NA
56.91
5.99
NA
7.86
NA
NA
NA
20.96
NA
NA
NA
7.58
NA
NA
7.16
NA
13.03
52.87
NA
NA
11.77
TOOK
5.70
NA
NA
5.97
NA
NA
NA
14.14
9.22
NA
14.62
NA
NA
0
9.78
NA
NA
NA
NA
NA
NA
39.80
4.85
NA
NA
NA
NA
NA
31.60
NA
NA
NA
3.76
NA
NA
4.41
NA
6.00
24.36
NA
NA
5.06
UCSD
5.83
NA
NA
6.61
NA
NA
NA
0
6.15
4.88
27.83
NA
14.63
6.43
6.33
NA
NA
NA
NA
NA
NA
NA
6.03
NA
NA
NA
NA
NA
7.87
NA
NA
NA
5.26
NA
NA
6.64
NA
23.26
27.78
NA
NA
6.61
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-9
-------
Table 29-2. VOC Method Precision: Average Coefficient of Variation Based on Duplicate
and Collocated Samples by Site (Continued)
Pollutant
Propylene
Styrene
1 , 1 ,2,2-Tetrachloroethane
Tetrachloroethylene
Toluene
1 ,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1 , 1 ,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Trichlorotrifluoroethane
1 ,2,4-Trimethylbenzene
1 ,3 ,5-Trimethylbenzene
Vinyl chloride
m,p-Xylene
o-Xylene
Average by Site
#of
Pairs
131
103
0
68
131
0
2
0
7
131
131
125
108
0
130
128
14.11
TMOK
3.48
4.63
NA
5.34
9.51
NA
NA
NA
NA
6.07
6.55
18.43
3.02
NA
10.11
14.31
13.83
TOOK
9.24
2.33
NA
3.70
3.61
NA
NA
NA
NA
3.44
4.80
8.06
3.03
NA
3.64
5.91
9.08
UCSD
10.93
14.24
NA
NA
10.11
NA
NA
NA
NA
7.09
11.17
4.68
3.03
NA
4.33
3.49
9.25
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
29.2.2 SNMOC Method Precision
The SNMOC method precision for duplicate and collocated samples is presented in
Table 29-3 as the average CV per pollutant per site, the average CV per site, and the overall CV
for NMP sites sampling SNMOC. The results from duplicate and collocated samples show low-
to high-level variability among pollutants per sites, ranging from an average CV of 0 percent
(2-methylheptane and 2,2,3-trimethylpentane for NBIL) to 96.61 percent (ethylbenzene for
BRCO). The site-specific average CV ranged from 11.14 percent for BRCO to 19.32 percent for
PACO, with an overall method average of 15.29 percent.
29-10
-------
Table 29-3. SNMOC Method Precision: Average Coefficient of Variation
Based on Duplicate and Collocated Samples by Site
Pollutant
Acetylene
Benzene
1,3 -Butadiene
w-Butane
c/s-2-Butene
/ra«s-2-Butene
Cyclohexane
Cyclopentane
Cyclopentene
w-Decane
1-Decene
Tw-Diethylbenzene
£>-Diethylbenzene
2,2-Dimethylbutane
2,3 -Dimethylbutane
2,3 -Dimethylpentane
2,4-Dimethylpentane
w-Dodecane
1-Dodecene
Ethane
2-Ethyl-l-butene
Ethylbenzene
Ethylene
/w-Ethyltoluene
o-Ethyltoluene
/>-Ethyltoluene
w-Heptane
1-Heptene
w-Hexane
1-Hexene
c/s-2-Hexene
/raws-2-Hexene
Isobutane
Isobutene/1 -Butene
Isopentane
Isoprene
Isopropylbenzene
2-Methyl-l-butene
3-Methyl-l-butene
2-Methyl- 1 -pentene
4-Methyl- 1 -pentene
2-Methyl-2-butene
#of
Pairs
31
31
3
27
11
13
21
28
0
17
0
2
2
28
26
27
19
6
6
31
0
25
31
18
11
10
25
8
30
1
1
0
31
29
31
18
1
5
4
0
0
0
BRCO
8.81
20.57
NA
1.65
NA
3.36
2.48
1.85
NA
13.94
NA
NA
NA
5.80
4.26
10.37
9.21
30.22
NA
1.91
NA
96.61
11.98
0.58
NA
0.47
6.12
10.60
3.83
NA
NA
NA
0.64
36.02
2.87
10.21
NA
NA
NA
NA
NA
NA
BTUT
8.76
11.14
6.91
12.73
12.03
22.67
12.48
14.00
NA
13.15
NA
17.44
21.45
27.03
9.77
15.07
13.00
8.12
24.15
8.83
NA
12.81
4.87
13.95
66.74
13.60
11.54
20.20
52.61
10.80
13.50
NA
11.36
10.66
11.79
31.34
28.34
16.59
NA
NA
NA
12.71
NBIL
14.43
10.58
NA
11.00
NA
NA
16.58
13.27
NA
44.71
NA
NA
NA
16.27
19.46
25.95
26.29
NA
29.74
18.83
NA
8.96
6.13
13.49
20.29
6.48
17.44
NA
14.99
NA
NA
NA
16.12
13.06
17.99
13.42
NA
NA
NA
NA
NA
NA
PACO
5.61
26.59
NA
20.63
6.26
11.56
6.96
11.68
NA
32.48
NA
NA
NA
5.82
4.23
10.71
6.32
73.36
NA
6.90
NA
30.70
30.31
24.51
21.49
23.41
11.07
5.98
8.47
NA
NA
NA
11.37
24.00
13.52
55.21
NA
NA
NA
NA
NA
NA
RICO
3.42
10.01
6.46
6.12
3.90
13.25
5.89
7.44
NA
8.38
NA
NA
NA
19.58
4.67
26.36
20.62
7.46
NA
6.23
NA
77.72
9.76
7.08
10.55
3.21
7.50
19.52
7.41
NA
NA
NA
6.21
63.87
3.37
13.27
NA
3.88
NA
NA
NA
10.16
SSSD
5.18
5.28
NA
1.21
8.41
3.03
28.87
27.67
NA
68.53
NA
NA
NA
33.03
11.78
19.22
4.51
8.86
7.77
1.04
NA
8.85
4.21
6.06
NA
NA
32.09
NA
23.31
NA
NA
NA
12.41
6.79
28.75
14.89
NA
6.37
NA
NA
NA
NA
UCSD
5.63
27.55
NA
1.38
NA
12.86
5.52
11.71
NA
NA
NA
NA
NA
6.36
2.55
9.50
NA
NA
4.66
1.55
NA
8.10
5.48
NA
NA
NA
2.22
NA
17.11
NA
NA
NA
5.10
4.53
13.42
44.00
NA
NA
NA
NA
NA
NA
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-11
-------
Table 29-3. SNMOC Method Precision:
Based on Duplicate and Collocated
Average Coefficient of Variation
Samples by Site (Continued)
Pollutant
Methylcyclohexane
Methylcyclopentane
2-Methylheptane
3-Methylheptane
2-Methylhexane
3-Methylhexane
2-Methylpentane
3-Methylpentane
w-Nonane
1-Nonene
w-Octane
1-Octene
w-Pentane
1-Pentene
c/s-2-Pentene
/raws-2-Pentene
a-Pinene
6-Pinene
Propane
w-Propylbenzene
Propylene
Propyne
Styrene
Toluene
w-Tridecane
1-Tridecene
1 ,2,3 -Trimethylbenzene
1 ,2,4-Trimethylbenzene
1 ,3 ,5-Trimethylbenzene
2,2,3 -Trimethylpentane
2,2,4-Trimethylpentane
2,3 ,4-Trimethylpentane
w-Undecane
1-Undecene
7w-Xylene/£>-Xylene
o-Xylene
Average by Site
#of
Pairs
24
31
14
17
23
27
28
31
17
2
18
1
31
29
7
19
8
1
31
8
31
0
3
31
0
0
5
21
15
8
21
15
11
9
25
21
75.29
BRCO
1.14
6.17
3.56
8.25
2.67
16.97
15.07
1.48
4.77
NA
5.27
NA
3.32
13.06
NA
19.36
12.00
NA
1.38
0.57
13.91
NA
NA
8.46
NA
NA
21.54
19.53
3.17
2.49
20.53
14.71
27.88
NA
13.93
12.81
11.14
BTUT
13.19
39.99
7.40
14.01
13.86
12.91
18.70
26.55
11.19
11.74
7.41
NA
15.84
10.65
16.37
16.02
16.43
NA
10.18
16.99
7.87
NA
23.28
7.13
NA
NA
17.95
8.39
13.32
18.58
13.40
10.89
35.14
78.78
12.41
10.65
77.26
NBIL
16.69
18.18
0
23.36
17.15
25.49
5.29
14.75
31.16
NA
2.66
NA
21.72
46.52
NA
35.19
13.08
NA
10.14
NA
8.76
NA
NA
17.20
NA
NA
NA
9.53
13.08
0
21.73
20.00
1.75
48.28
13.15
7.89
77.77
PACO
7.76
4.56
5.89
8.95
6.22
14.60
3.08
7.25
8.20
NA
10.73
12.07
59.82
30.18
NA
25.92
NA
NA
7.54
26.86
15.08
NA
NA
11.97
NA
NA
NA
15.13
9.60
37.43
NA
NA
65.58
72.67
9.16
20.15
79.32
RICO
11.65
15.84
7.81
9.43
5.31
24.69
19.86
2.77
10.15
NA
10.22
NA
5.28
54.34
2.33
28.15
15.66
NA
7.10
10.71
9.47
NA
NA
6.05
NA
NA
8.32
11.04
10.19
59.82
29.53
22.62
6.53
NA
5.34
5.69
14.20
SSSD
23.04
12.17
NA
6.36
11.36
19.44
13.67
6.60
62.94
NA
10.09
NA
33.07
13.26
NA
5.45
NA
NA
2.96
NA
2.59
NA
NA
23.50
NA
NA
NA
7.63
NA
NA
7.20
5.92
39.51
19.50
15.65
22.18
75.79
UCSD
2.26
30.33
NA
NA
8.52
42.19
48.17
5.68
NA
NA
NA
NA
22.38
9.92
NA
8.98
NA
5.15
1.09
NA
5.71
NA
15.29
9.62
NA
NA
NA
2.14
NA
NA
15.78
NA
NA
24.22
4.02
NA
12.18
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-12
-------
29.2.3 Carbonyl Compound Method Precision
Table 29-4 presents the method precision for duplicate and collocated carbonyl
compound samples as the average CV per pollutant per site, the average CV per site, and the
overall average CV for NMP sites sampling carbonyl compounds. The duplicate and collocated
sample results show low- to mid-level variability among the sites, ranging from an average CV
of 0 percent (valeraldehyde for SPIL) to 32.37 percent (tolualdehydes for ELNJ). The site-
specific average CV ranged from 3.42 percent for TOOK to 11.16 percent for ELNJ. Note that
the average CV for every site was less than the program DQO of 15 percent. The overall method
precision was 6.17 percent for carbonyl compounds.
Table 29-4. Carbonyl Compound Method Precision:
Based on Duplicate and Collocated
Average Coefficient of Variation
Samples by Site
Pollutant
Acetaldehyde
Acetone
Benzaldehyde
Butyraldehyde
ICrotonaldehyde
2,5-Dimethylbenzaldehyde
Formaldehyde
Hexaldehyde
Isovaleraldehyde
Propionaldehyde
Tolualdehydes
Valeraldehyde
Average by Site
#of
Pairs
127
127
124
127
126
0
127
125
2
127
74
123
6.17
AZFL
0.44
6.27
10.26
8.71
2.94
NA
2.20
7.72
NA
2.92
10.31
7.08
5.89
BTUT
3.45
1.10
2.15
3.95
5.06
NA
3.22
4.02
NA
4.33
3.80
9.13
4.02
CHNJ
6.30
9.71
6.06
7.02
7.71
NA
6.10
5.08
NA
4.65
4.41
5.71
6.28
DEMI
7.81
3.61
7.13
21.37
4.84
NA
5.76
11.05
NA
3.23
7.92
13.03
8.58
ELNJ
5.80
16.11
9.22
9.18
9.33
NA
3.89
8.82
NA
7.15
32.37
9.70
11.16
GPCO
1.45
8.97
5.58
3.81
4.56
NA
2.09
7.08
NA
3.53
11.44
5.66
5. 42
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
29-13
-------
Table 29-4. Carbonyl Compound Method Precision: Average Coefficient of Variation
Based on Duplicate and Collocated Samples by Site (Continued)
Pollutant
Acetaldehyde
Acetone
Benzaldehyde
Butyraldehyde
Crotonaldehyde
2,5-Dimethylbenzaldehyde
Formaldehyde
Hexaldehyde
Isovaleraldehyde
Propionaldehyde
Tolualdehydes
Valeraldehyde
Average by Site
#of
Pairs
127
127
124
127
126
0
127
125
2
127
74
123
6.17
INDEM
6.32
6.01
9.29
7.73
8.58
NA
7.48
7.53
NA
6.81
7.94
8.90
7.66
MWOK
2.01
2.70
5.41
6.44
4.10
NA
1.72
5.50
NA
2.68
12.47
4.68
4.77
NBIL
6.95
12.31
8.27
6.64
6.18
NA
7.69
8.64
NA
9.68
6.79
4.76
7.79
NBNJ
2.86
16.20
10.29
9.16
4.03
NA
3.71
7.78
NA
4.77
8.33
8.53
7.57
OCOK
1.08
2.05
2.96
3.48
2.75
NA
2.05
6.66
6.15
4.73
5.25
3.72
3.72
ORFL
2.54
7.86
7.17
8.67
5.30
NA
2.37
6.57
NA
5.67
6.98
5.23
5.84
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
Table 29-4. Carbonyl Compound Method Precision: Average Coefficient of Variation
Based on Duplicate and Collocated Samples by Site (Continued)
Pollutant
Acetaldehyde
Acetone
Benzaldehyde
Butyraldehyde
Crotonaldehyde
2,5-Dimethylbenzaldehyde
Formaldehyde
Hexaldehyde
Isovaleraldehyde
Propionaldehyde
Tolualdehydes
Valeraldehyde
Average by Site
#of
Pairs
127
127
124
127
126
0
127
125
2
127
74
123
6.17
PROK
0.89
4.18
8.56
5.10
3.81
NA
1.59
6.60
NA
2.61
5.49
8.13
4.70
PXSS
2.69
1.18
7.56
5.53
6.67
NA
3.44
9.07
NA
7.83
3.19
12.86
6.00
S4MO
2.45
4.65
8.81
4.44
3.37
NA
2.81
4.11
NA
3.32
12.12
6.98
5.31
SEWA
2.39
4.06
12.28
1.60
3.23
NA
5.23
9.56
NA
2.84
10.09
5.62
5.69
SKFL
3.02
14.05
6.78
8.50
3.17
NA
1.85
6.83
NA
5.05
9.01
5.88
6.41
SPIL
9.48
5.00
11.57
10.03
9.38
NA
12.91
13.51
NA
7.51
5.11
0
8.45
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-14
-------
Table 29-4. Carbonyl Compound Method Precision: Average Coefficient of Variation
Based on Duplicate and Collocated Samples by Site (Continued)
Pollutant
Acetaldehyde
Acetone
Benzaldehyde
Butyraldehyde
Crotonaldehyde
2,5-Dimethylbenzaldehyde
Formaldehyde
Hexaldehyde
Isovaleraldehyde
Propionaldehyde
Tolualdehydes
Valeraldehyde
Average by Site
#of
Pairs
127
127
124
127
126
0
127
125
2
127
74
123
6.17
SSSD
2.81
7.25
3.28
6.09
2.66
NA
3.66
3.22
NA
3.75
6.08
4.77
4.36
SYFL
3.60
9.73
6.84
3.52
8.21
NA
6.10
6.37
NA
3.61
12.07
5.32
6.54
TMOK
1.39
4.53
4.76
2.65
2.53
NA
2.02
5.20
NA
1.80
7.90
7.44
4.02
TOOK
1.60
1.44
6.42
1.99
3.12
NA
1.72
5.82
NA
2.01
7.56
2.54
3. 42
UCSD
5.37
15.62
8.65
7.76
7.47
NA
4.78
9.69
NA
8.50
8.08
11.77
8.77
WPIN
4.94
5.07
4.57
5.49
6.47
NA
4.19
5.76
4.99
7.80
7.55
6.55
5.76
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29.2.4 Metals Method Precision
The method precision for all collocated metals samples are presented in Table 29-5 as the
average CV per pollutant per site, the average CV per site, and the overall average CV for NMP
sites sampling metals. All samples evaluated in this section are collocated samples. The results
from collocated samples show low- to high-level variability among sites, ranging from 0 percent
(cobalt for UNVT) to 72.53 percent (mercury for NBIL). The site-specific average CV ranged
from 8.27 percent for BOMA to 27.11 percent for NBIL. The overall method precision for
metals was 12.97 percent.
29-15
-------
Table 29-5. Metals Method Precision: Average Coefficient of Variation
Based on Collocated Samples by Site
Pollutant
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Cobalt
Lead
Manganese
Mercury
Nickel
Selenium
Average by Site
#of
Pairs
140
139
104
128
15
137
137
140
83
19
136
12.97
BOMA
3.69
3.04
24.23
23.32
NA
5.60
5.03
2.42
6.37
3.47
5.55
8.27
BTUT
17.77
6.52
NA
11.55
NA
12.37
3.95
6.22
NA
31.44
7.73
12.19
NBIL
15.59
18.35
27.55
28.84
NA
35.33
25.87
22.10
72.53
14.26
10.72
27.11
S4MO
3.83
2.78
10.74
17.15
0.84
15.07
4.82
16.68
14.75
NA
3.92
9. 06
TOOK
28.57
6.55
8.80
24.77
4.68
18.21
7.18
8.76
20.21
4.50
7.34
12. 69
UNVT
5.87
9.52
NA
10.99
NA
0
3.44
4.15
NA
16.72
17.39
8.51
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
29.2.5 Hexavalent Chromium Method Precision
Table 29-6 presents the method precision results from collocated hexavalent chromium
samples as the average CV per site and the overall average CV for NMP sites sampling
hexavalent chromium. Hexavalent chromium is a NATTS MQO Core Analyte and the sites
shown are collocated NATTS sites. The site-specific average CV ranged from 2.69 percent for
BOMA to 33.84 percent for CHSC, with an overall method precision of 14.89 percent.
29-16
-------
Table 29-6. Hexavalent Chromium Method Precision: Average Coefficient of Variation
Based on Collocated Samples by Site
Site
BOMA
BTUT
BXNY
CAMS 35
CHSC
DEMI
GLKY
GPCO
HOW
MONY
NBIL
PRRI
PXSS
RIVA
ROCH
S4MO
SDGA
SEWA
SKFL
SYFL
UNVT
WADC
Average by Site
# of Pairs
Average
CV
2.69
14.03
23.36
15.41
33.84
11.28
7.13
10.86
21.12
3.63
19.50
22.44
9.59
7.81
12.94
5.31
21.97
9.99
15.93
17.75
17.53
23.45
14.89
112
BOLD ITALICS = EPA-designated
NATTS Site.
29.2.6 PAH Method Precision
The method precision results for the collocated PAH samples are shown in Table 29-7 as
the average CV per pollutant per site, the average CV per site, and the overall average CV for
NMP sites sampling PAH. The results from collocated samples show low- to high-level average
variability among sites, ranging from 0.23 percent (benzo(e)pyrene for SEWA) to 62.73 percent
(acenaphthylene for RUCA). The site-specific average CV ranged from 9.17 percent for DEMI
to 21.08 percent for RUCA. The overall method precision was 13.18 percent.
29-17
-------
Table 29-7. PAH Method Precision: Average Coefficient of Variation Based on Collocated
Samples by Site
Pollutant
Acenaphthene
Acenaphthylene
Anthracene
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Chrysene
Coronene
Cyclopenta[cd]pyrene
Dibenz(a,h)anthracene
Fluoranthene
Fluorene
9-Fluorenone
Indeno( 1,2,3 -cd)pyrene
Naphthalene
Perylene
Phenanthrene
Pyrene
Retene
Average by Site
#of
Pairs
29
11
13
6
8
13
11
20
6
29
11
1
2
29
29
29
15
29
8
29
29
21
13.18
DEMI
2.95
6.54
8.45
19.92
7.86
8.06
7.21
5.41
12.97
8.75
5.53
NA
9.12
4.39
4.80
2.82
8.34
2.98
9.28
2.48
9.17
45.58
9.17
RUCA
10.01
62.73
28.21
28.25
32.79
25.71
20.23
17.83
28.04
16.89
15.98
NA
NA
14.93
10.84
9.26
25.49
10.21
14.72
13.95
12.50
23.06
21.08
SDGA
6.61
9.80
8.86
7.66
10.89
11.01
0.52
13.15
NA
11.45
18.60
NA
NA
6.63
4.42
5.20
13.09
9.15
6.64
6.27
23.58
8.99
9.61
SEWA
15.11
6.60
5.62
4.33
1.73
28.14
0.23
26.54
1.01
7.66
1.82
1.70
10.08
9.23
16.79
24.37
35.38
30.90
9.92
16.01
11.61
43.62
14.02
SYFL
9.95
NA
NA
NA
NA
2.29
7.30
7.82
NA
12.98
26.60
NA
NA
7.42
9.37
8.63
9.89
8.01
22.64
5.54
21.36
20.34
12.01
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
29.3 Analytical Precision
Analytical precision is a measurement of random errors associated with the process of
analyzing environmental samples. These errors may result from various factors, but typically
originate from random "noise" inherent to analytical instruments. Laboratories can easily
evaluate the analytical precision of ambient air samples by comparing concentrations measured
during multiple analyses of a single sample (i.e., replicate samples). CVs were calculated for
every replicate analysis run on duplicate or collocated samples collected during the program
year. However, only results at or above the MDL were used in these calculations, similar to the
calculation of method precision discussed in Section 29.2.
29-18
-------
Table 29-8 presents the 2010 NMP analytical precision for VOC, SNMOC, carbonyl
compounds, metals, hexavalent chromium, and PAH, presented as average CV (expressed as a
percentage). The analytical precision averaged across all sites collecting duplicate or collocated
samples met the program DQO, which is 15 percent CV. The analytical precision for all six
methods was less than 8 percent.
Table 29-8. Analytical Precision by Analytical Method
Method/Pollutant
Group
VOC
(TO- 15)
SNMOC
Carbonyl Compounds
(TO- 11 A)
Metals Analysis
(Method IO-3. 5)
Hexavalent Chromium
(EPA-approved method)
PAH
(TO-13)
DQO
Average
Coefficient of
Variation
(%)
5.90
7.11
2.36
5.44
6.30
3.67
Number of
Pairs
6,090
2,417
2,674
2,409
226
769
15. 00 percent CV
Tables 29-9 through 29-14 present analytical precision for VOC, SNMOC, carbonyl
compounds, metals, hexavalent chromium, and PAH, respectively, as the average CVs per
pollutant per site, per site, and per method. Pollutants exceeding the 15 percent DQO for CV are
bolded in each table. In Tables 29-9 through 29-14, the number of pairs in comparison to the
respective tables listed for duplicate or collocated analyses in Tables 29-2 through 29-7, is
approximately twice as high because each sample produces a replicate for each duplicate (or
collocated) sample. The replicate analyses of both duplicate and collocated samples indicate that
the analytical precision level is within the program DQOs. The shaded rows in each table
identify the NATTS MQO Core Analytes for each method.
29-19
-------
29.3.1 VOC Analytical Precision
Table 29-9 presents analytical precision results from replicate analyses of all duplicate
and collocated VOC samples as the average CV per pollutant per site, the average CV per site,
and the overall average CV for NMP sites sampling VOC. The analytical precision results from
replicate analyses of all duplicate and collocated samples show that for most of the pollutants,
the VOC analytical precision was within the program DQO of 15 percent. The average CV
ranged from 0 percent for several pollutants and several sites to 47.14 percent
(p-dichlorobenzene for SEW A). The site-specific average CV ranged from 4.01 percent for SPIL
to 7.79 percent for UCSD. The overall analytical precision was 5.90 percent.
Table 29-9. VOC Analytical Precision: Average Coefficient of Variation Based on
Replicate Analyses by Site
Pollutant
Acetylene
Acrylonitrile
tert-Amyl Methyl Ether
Benzene
Bromochloromethane
Bromodichloromethane
Bromoform
Bromomethane
1,3 -Butadiene
Carbon Bisulfide
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
Chloromethane
Chloromethylbenzene
Chloroprene
Dibromochloromethane
1 ,2-Dibromoethane
/w-Dichlorobenzene
o-Dichlorobenzene
p-Dichlorobenzene
Dichlorodifluoromethane
#of
Pairs
267
6
0
267
0
9
1
77
219
194
262
0
34
168
267
0
0
5
0
4
0
95
267
BTUT
5.56
NA
NA
7.99
NA
NA
NA
5.58
6.83
5.27
9.28
NA
NA
5.04
5.31
NA
NA
NA
NA
4.64
NA
4.42
5.43
BURVT
4.00
NA
NA
5.07
NA
NA
NA
8.25
5.84
10.66
6.70
NA
7.47
6.53
5.14
NA
NA
NA
NA
NA
NA
6.30
5.11
CHNJ
6.81
NA
NA
6.56
NA
NA
6.15
5.00
3.87
6.10
4.82
NA
NA
5.79
6.33
NA
NA
NA
NA
NA
NA
6.73
6.09
DEMI
3.62
NA
NA
4.41
NA
NA
NA
5.12
12.11
3.54
3.65
NA
7.69
4.67
2.74
NA
NA
NA
NA
NA
NA
NA
2.81
ELNJ
5.92
2.44
NA
5.28
NA
NA
NA
7.51
5.44
5.64
5.79
NA
8.58
8.51
5.76
NA
NA
NA
NA
NA
NA
8.19
5.72
GLKY
4.85
NA
NA
4.19
NA
NA
NA
7.21
7.97
4.32
3.42
NA
NA
2.56
4.16
NA
NA
NA
NA
NA
NA
NA
4.13
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-20
-------
Table 29-9. VOC Analytical Precision: Average Coefficient of Variation Based
Replicate Analyses by Site (Continued)
on
Pollutant
1 , 1 -Dichloroethane
1 ,2-Dichloroethane
1 , 1 -Dichloroethene
cis- 1 ,2-Dichloroethylene
/ra«s-l,2-Dichloroethylene
Dichloromethane
1 ,2-Dichloropropane
cis- 1 ,3 -Dichloropropene
trans- 1 , 3 -Dichloropropene
Dichlorotetrafluoroethane
Ethyl Acrylate
Ethyl tort-Butyl Ether
Ethylbenzene
Hexachloro- 1 , 3 -butadiene
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Methyl Methacrylate
Methyl tert-Butyl Ether
w-Octane
Propylene
Styrene
1 , 1 ,2,2-Tetrachloroethane
Tetrachloroethylene
Toluene
1 ,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1 , 1 ,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Trichlorotrifluoroethane
1 ,2,4-Trimethylbenzene
1 ,3 ,5-Trimethylbenzene
Vinyl chloride
7w,£>-Xylene
o-Xylene
Average by Site
#of
Pairs
0
18
0
0
2
267
0
0
0
255
0
0
263
0
267
214
1
0
212
267
211
0
145
267
0
6
0
16
267
267
255
225
0
264
259
5.90
BTUT
NA
NA
NA
NA
NA
4.77
NA
NA
NA
3.99
NA
NA
7.82
NA
6.38
7.69
NA
NA
8.88
7.05
8.71
NA
11.13
8.40
NA
1.64
NA
3.20
5.34
4.84
7.72
7.61
NA
8.07
7.95
6.43
BURVT
NA
NA
NA
NA
NA
5.74
NA
NA
NA
7.87
NA
NA
5.05
NA
5.71
8.29
NA
NA
8.13
4.58
9.37
NA
6.86
4.97
NA
NA
NA
0
5.13
5.08
6.25
13.81
NA
5.01
5.53
6.53
CHNJ
NA
NA
NA
NA
NA
5.07
NA
NA
NA
8.19
NA
NA
7.72
NA
7.45
8.52
NA
NA
8.90
6.91
10.13
NA
7.89
5.85
NA
0
NA
NA
5.38
8.13
10.37
8.74
NA
8.23
7.71
6.77
DEMI
NA
NA
NA
NA
NA
4.07
NA
NA
NA
4.41
NA
NA
3.77
NA
4.28
10.16
NA
NA
7.70
4.23
7.82
NA
4.12
3.76
NA
NA
NA
6.54
3.22
4.28
4.99
5.09
NA
4.30
5.73
5.26
ELNJ
NA
NA
NA
NA
5.66
5.62
NA
NA
NA
9.42
NA
NA
7.06
NA
5.57
11.40
NA
NA
8.14
5.65
11.04
NA
7.24
5.69
NA
NA
NA
NA
5.32
5.66
6.60
11.27
NA
5.58
6.05
6.82
GLKY
NA
NA
NA
NA
NA
3.98
NA
NA
NA
5.81
NA
NA
6.94
NA
4.16
6.33
NA
NA
3.80
5.05
4.47
NA
NA
4.25
NA
5.66
NA
NA
3.42
3.71
7.43
9.88
NA
4.52
6.24
5.14
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-21
-------
Table 29-9. VOC Analytical Precision: Average Coefficient of Variation Based
Replicate Analyses by Site (Continued)
on
Pollutant
Acetylene
Acrylonitrile
tert-Amyl Methyl Ether
Benzene
Bromochloromethane
Bromodichloromethane
Bromoform
Bromomethane
1,3 -Butadiene
Carbon Bisulfide
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
Chloromethane
Chloromethylbenzene
Chloroprene
Dibromochloromethane
1 ,2-Dibromoethane
/w-Dichlorobenzene
o-Dichlorobenzene
£>-Dichlorobenzene
Dichlorodifluoromethane
1 , 1 -Dichloroethane
1 ,2-Dichloroethane
1 , 1 -Dichloroethene
cis- 1 ,2-Dichloroethylene
trans- 1 ,2-Dichloroethylene
Dichloromethane
1 ,2-Dichloropropane
cis- 1 ,3 -Dichloropropene
trans- 1 , 3 -Dichloropropene
Dichlorotetrafluoroethane
Ethyl Aery late
Ethyl tert-Butyl Ether
Ethylbenzene
#of
Pairs
267
6
0
267
0
9
1
77
219
194
262
0
34
168
267
0
0
5
0
4
0
95
267
0
18
0
0
2
267
0
0
0
255
0
0
263
GPCO
4.73
NA
NA
5.11
NA
NA
NA
7.65
6.30
4.88
5.36
NA
1.55
5.41
5.46
NA
NA
NA
NA
NA
NA
0
5.44
NA
NA
NA
NA
NA
5.16
NA
NA
NA
4.93
NA
NA
7.49
MWOK
4.38
NA
NA
5.04
NA
NA
NA
5.82
6.89
4.75
5.35
NA
5.24
5.70
4.53
NA
NA
NA
NA
NA
NA
5.19
4.40
NA
NA
NA
NA
NA
4.06
NA
NA
NA
6.78
NA
NA
3.79
NBIL
3.41
NA
NA
3.62
NA
NA
NA
6.00
5.06
2.11
4.94
NA
NA
6.48
3.93
NA
NA
4.61
NA
NA
NA
3.65
3.90
NA
NA
NA
NA
NA
3.40
NA
NA
NA
3.62
NA
NA
4.03
NBNJ
4.54
7.62
NA
4.97
NA
NA
NA
6.05
8.80
4.98
5.64
NA
3.70
3.89
4.98
NA
NA
NA
NA
NA
NA
6.83
5.05
NA
NA
NA
NA
NA
4.29
NA
NA
NA
6.35
NA
NA
5.24
OCOK
8.06
0
NA
6.08
NA
NA
NA
3.87
10.96
7.17
4.93
NA
2.13
4.04
4.91
NA
NA
NA
NA
NA
NA
5.69
4.65
NA
11.47
NA
NA
NA
5.33
NA
NA
NA
7.60
NA
NA
6.63
PANJ
3.81
NA
NA
7.92
NA
NA
NA
NA
4.97
6.15
3.88
NA
2.71
6.48
3.33
NA
NA
NA
NA
NA
NA
8.11
3.75
NA
NA
NA
NA
NA
4.52
NA
NA
NA
3.31
NA
NA
6.43
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
29-22
-------
Table 29-9. VOC Analytical Precision: Average Coefficient of Variation Based
Replicate Analyses by Site (Continued)
on
Pollutant
Hexachloro- 1 ,3 -butadiene
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Methyl Methacrylate
Methyl fer/-Butyl Ether
w-Octane
Propylene
Styrene
1 , 1 ,2,2-Tetrachloroethane
Tetrachloroethylene
Toluene
1 ,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1 , 1 ,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Trichlorotrifluoroethane
1 ,2,4-Trimethylbenzene
1 , 3 ,5 -Trimethy Ibenzene
Vinyl chloride
m,p-Xylene
o-Xylene
Average by Site
#of
Pairs
0
267
214
1
0
212
267
211
0
145
267
0
6
0
16
267
267
255
225
0
264
259
5.90
GPCO
NA
7.69
4.65
0
NA
9.01
4.74
8.73
NA
5.13
6.89
NA
23.70
NA
NA
5.22
4.88
9.54
5.24
NA
7.56
7.55
6.19
MWOK
NA
4.26
7.14
NA
NA
5.36
7.09
6.85
NA
8.41
3.23
NA
NA
NA
NA
4.09
4.13
4.23
7.81
NA
4.30
3.43
5.08
NBIL
NA
3.69
6.25
NA
NA
7.75
2.80
9.70
NA
2.04
2.50
NA
NA
NA
4.84
3.39
3.30
3.26
6.14
NA
4.50
3.41
4.20
NBNJ
NA
4.67
15.24
NA
NA
6.65
5.02
8.83
NA
6.61
4.10
NA
NA
NA
NA
4.84
5.59
6.56
18.34
NA
5.80
5.19
6.44
OCOK
NA
5.06
13.85
NA
NA
8.27
6.68
12.33
NA
11.89
6.39
NA
NA
NA
NA
5.00
5.38
10.43
23.10
NA
6.27
8.72
7.48
PANJ
NA
6.25
7.74
NA
NA
6.12
3.62
12.09
NA
4.72
6.02
NA
NA
NA
NA
3.51
3.31
8.25
9.01
NA
6.65
6.63
5.74
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
29-23
-------
Table 29-9. VOC Analytical Precision: Average Coefficient of Variation Based
Replicate Analyses by Site (Continued)
on
Pollutant
Acetylene
Acrylonitrile
tert-Amyl Methyl Ether
Benzene
Bromochloromethane
Bromodichloromethane
Bromoform
Bromomethane
1,3 -Butadiene
Carbon Bisulfide
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
Chloromethane
Chloromethylbenzene
Chloroprene
Dibromochloromethane
1 ,2-Dibromoethane
m -Dichlorobenzene
o-Dichlorobenzene
^-Dichlorobenzene
Dichlorodifluoromethane
1 , 1 -Dichloroethane
1 ,2-Dichloroethane
1 , 1 -Dichloroethene
cis- 1 ,2-Dichloroethylene
trans- 1 ,2-Dichloroethylene
Dichloromethane
1 ,2-Dichloropropane
cis- 1 ,3 -Dichloropropene
trans- 1 ,3 -Dichloropropene
Dichlorotetrafluoroethane
Ethyl Aery late
Ethyl tert-Butyl Ether
Ethylbenzene
#of
Pairs
267
6
0
267
0
9
1
77
219
194
262
0
34
168
267
0
0
5
0
4
0
95
267
0
18
0
0
2
267
0
0
0
255
0
0
263
PROK
3.92
NA
NA
4.78
NA
NA
NA
6.29
5.75
4.98
5.04
NA
5.08
11.63
3.75
NA
NA
NA
NA
NA
NA
8.18
3.84
NA
NA
NA
NA
NA
5.22
NA
NA
NA
5.13
NA
NA
7.48
PXSS
14.42
16.64
NA
3.82
NA
NA
NA
12.17
5.61
5.20
5.46
NA
4.29
7.66
4.46
NA
NA
NA
NA
NA
NA
4.36
4.56
NA
NA
NA
NA
NA
4.49
NA
NA
NA
8.59
NA
NA
4.47
S4MO
3.61
NA
NA
2.83
NA
NA
NA
5.34
10.07
3.62
3.55
NA
NA
4.13
3.73
NA
NA
NA
NA
NA
NA
2.22
3.60
NA
NA
NA
NA
NA
3.95
NA
NA
NA
4.84
NA
NA
4.55
SEWA
2.63
NA
NA
2.54
NA
NA
NA
4.29
3.46
4.32
3.31
NA
0
3.23
3.00
NA
NA
NA
NA
NA
NA
47.14
2.85
NA
2.44
NA
NA
NA
3.80
NA
NA
NA
2.78
NA
NA
5.62
SPIL
3.78
NA
NA
3.52
NA
NA
NA
0
5.38
4.59
3.98
NA
4.56
2.51
3.34
NA
NA
NA
NA
NA
NA
4.56
3.16
NA
NA
NA
NA
NA
3.97
NA
NA
NA
3.23
NA
NA
4.26
SSSD
6.38
NA
NA
4.14
NA
NA
NA
1.91
11.15
10.98
6.53
NA
NA
3.22
7.16
NA
NA
NA
NA
NA
NA
4.69
6.16
NA
NA
NA
NA
NA
5.02
NA
NA
NA
3.45
NA
NA
9.93
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
29-24
-------
Table 29-9. VOC Analytical Precision: Average Coefficient of Variation Based
Replicate Analyses by Site (Continued)
on
Pollutant
Hexachloro- 1 ,3 -butadiene
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Methyl Methacrylate
Methyl fer/-Butyl Ether
w-Octane
Propylene
Styrene
1,1 ,2,2-Tetrachloroethane
Tetrachloroethylene
Toluene
1 ,2,4-Trichlorobenzene
1 , 1 , 1 -Trichloroethane
1,1,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Trichlorotrifluoroethane
1 ,2,4-Trimethylbenzene
1,3,5 -Trimethy Ibenzene
Vinyl chloride
m,p-Xylene
o-Xylene
Average by Site
#of
Pairs
0
267
214
1
0
212
267
211
0
145
267
0
6
0
16
267
267
255
225
0
264
259
5.90
PROK
NA
3.93
9.28
NA
NA
7.30
9.03
7.24
NA
NA
3.41
NA
NA
NA
NA
3.68
3.97
5.70
8.43
NA
5.17
4.77
5.88
PXSS
NA
5.02
5.85
NA
NA
4.35
3.98
14.55
NA
5.74
3.79
NA
NA
NA
0
4.28
6.81
5.76
8.76
NA
4.59
4.55
6.35
S4MO
NA
6.20
8.31
NA
NA
4.88
3.95
6.17
NA
4.27
3.48
NA
NA
NA
NA
3.31
4.90
4.70
6.20
NA
3.55
3.74
4.60
SEWA
NA
3.19
5.55
NA
NA
8.55
2.58
12.04
NA
1.52
3.64
NA
NA
NA
NA
1.94
2.84
5.46
9.13
NA
3.63
4.91
5.59
SPIL
NA
3.87
2.32
NA
NA
6.20
4.27
9.54
NA
3.86
3.73
NA
NA
NA
3.72
3.01
2.70
4.31
5.42
NA
3.89
4.51
4.01
SSSD
NA
3.53
5.44
NA
NA
9.37
6.92
9.74
NA
5.59
3.93
NA
NA
NA
NA
5.35
5.97
6.87
8.03
NA
5.06
4.69
6.20
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
29-25
-------
Table 29-9. VOC Analytical Precision: Average Coefficient of Variation Based on
Replicate Analyses by Site (Continued)
Pollutant
Acetylene
Acrylonitrile
tert-Amyl Methyl Ether
Benzene
Bromochloromethane
Bromodichloromethane
Bromoform
Bromomethane
1,3 -Butadiene
Carbon Bisulfide
Carbon Tetrachloride
Chlorobenzene
Chloroethane
Chloroform
Chloromethane
Chloromethylbenzene
Chloroprene
Dibromochloromethane
1 ,2-Dibromoethane
/w-Dichlorobenzene
o-Dichlorobenzene
£>-Dichlorobenzene
Dichlorodifluoromethane
1 , 1 -Dichloroethane
1 ,2-Dichloroethane
1 , 1 -Dichloroethene
cis- 1 ,2-Dichloroethylene
/ra«5-l,2-Dichloroethylene
Dichloromethane
1 ,2-Dichloropropane
cis- 1 ,3 -Dichloropropene
trans- 1 , 3 -Dichloropropene
Dichlorotetrafluoroethane
Ethyl Aery late
Ethyl tert-Butyl Ether
Ethylbenzene
#of
Pairs
267
6
0
267
0
9
1
77
219
194
262
0
34
168
267
0
0
5
0
4
0
95
267
0
18
0
0
2
267
0
0
0
255
0
0
263
TMOK
3.44
0
NA
4.16
NA
NA
NA
10.68
6.43
5.40
3.29
NA
4.83
10.53
3.08
NA
NA
NA
NA
3.23
NA
4.28
2.93
NA
5.56
NA
NA
NA
7.39
NA
NA
NA
5.88
NA
NA
2.93
TOOK
6.10
NA
NA
5.39
NA
NA
NA
11.45
7.67
5.49
5.13
NA
NA
2.98
4.36
NA
NA
NA
NA
NA
NA
7.96
4.39
NA
NA
NA
NA
NA
5.30
NA
NA
NA
8.15
NA
NA
5.57
UCSD
17.13
NA
NA
5.70
NA
NA
NA
0
7.69
7.36
12.97
NA
4.56
1.90
5.96
NA
NA
NA
NA
NA
NA
NA
6.83
NA
NA
NA
NA
NA
5.32
NA
NA
NA
4.90
NA
NA
7.73
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-26
-------
Table 29-9. VOC Analytical Precision: Average Coefficient of Variation Based on
Replicate Analyses by Site (Continued)
Pollutant
Hexachloro- 1 , 3 -butadiene
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Methyl Methacrylate
Methyl fer/-Butyl Ether
w-Octane
Propylene
Styrene
1 , 1 ,2,2-Tetrachloroethane
Tetrachloroethylene
Toluene
1 ,2,4-Trichlorobenzene
1,1,1 -Trichloroethane
1 , 1 ,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Trichlorotrifluoroethane
1 ,2,4-Trimethylbenzene
1 ,3 ,5-Trimethylbenzene
Vinyl chloride
m,p-Xylene
o-Xylene
Average by Site
#of
Pairs
0
267
214
1
0
212
267
211
0
145
267
0
6
0
16
267
267
255
225
0
264
259
5.90
TMOK
NA
3.54
5.47
NA
NA
5.11
7.48
10.98
NA
3.27
4.38
NA
NA
NA
NA
3.08
2.29
6.53
2.76
NA
5.41
5.17
4.98
TOOK
NA
5.60
6.92
NA
NA
11.72
5.92
9.60
NA
7.79
5.45
NA
NA
NA
NA
4.12
4.39
5.77
6.02
NA
5.81
6.10
6.35
UCSD
NA
8.55
3.92
NA
NA
10.18
12.31
7.08
NA
NA
6.05
NA
NA
NA
NA
8.72
6.06
10.53
8.26
NA
13.77
11.25
7.79
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29.3.2 SNMOC Analytical Precision
Table 29-10 presents analytical precision results from replicate analyses of all duplicate
and collocated SNMOC samples as the average CV per pollutant per site, the average CV per
site, and the overall average CV for NMP sites sampling SNMOC. The average CV ranged from
0.18 percent (isobutane for BRCO) to 69.73 percent (1-undecene for PACO). The site-specific
average CV ranged from 4.72 percent for RICO to 9.24 percent for PACO. The overall program
average CV was 7.11 percent.
29-27
-------
Table 29-10. SNMOC Analytical Precision: Average Coefficient of Variation Based on
Replicate Analyses by Site
Pollutant
Acetylene
Benzene
1,3 -Butadiene
w-Butane
c/s-2-Butene
/raws-2-Butene
Cyclohexane
Cyclopentane
Cyclopentene
w-Decane
1-Decene
/w-Diethylbenzene
£>-Diethylbenzene
2,2-Dimethylbutane
2,3-Dimethylbutane
2,3-Dimethylpentane
2,4-Dimethylpentane
w-Dodecane
1-Dodecene
Ethane
2-Ethyl-l-butene
Ethylbenzene
Ethylene
/w-Ethyltoluene
o-Ethyltoluene
£>-Ethyltoluene
w-Heptane
1-Heptene
w-Hexane
1-Hexene
c/s-2-Hexene
/ra«s-2-Hexene
Isobutane
Isobutene/ 1 -Butene
Isopentane
Isoprene
#of
Pairs
61
61
7
55
23
23
42
54
1
35
0
7
5
55
52
52
39
16
16
61
0
52
61
37
25
21
49
16
59
4
2
0
61
57
61
36
BRCO
4.93
5.38
NA
1.66
NA
NA
0.63
7.24
NA
7.43
NA
3.28
NA
4.84
2.02
8.68
6.84
12.93
5.52
0.43
NA
9.34
5.63
6.20
NA
10.13
3.61
3.98
3.93
3.45
NA
NA
0.18
4.09
1.69
2.13
BTUT
3.40
8.04
6.14
1.44
7.42
12.66
3.92
5.48
NA
6.43
NA
11.55
30.45
13.78
4.71
4.41
4.04
6.03
24.32
1.09
NA
7.38
2.21
12.42
50.19
8.40
5.90
4.59
6.40
8.90
2.97
NA
1.17
3.36
2.08
6.43
NBIL
4.02
6.62
NA
2.53
21.04
3.49
5.41
13.51
20.65
5.92
NA
4.90
NA
13.93
7.48
9.81
9.99
NA
6.60
1.02
NA
10.55
3.12
7.98
11.12
5.35
6.70
NA
6.58
NA
NA
NA
2.75
5.50
1.67
1.66
PACO
3.29
7.98
NA
0.89
7.22
2.03
1.68
3.05
NA
6.33
NA
NA
NA
2.56
1.57
3.77
3.85
39.54
18.93
0.91
NA
12.70
17.64
7.65
23.08
7.26
6.82
5.17
4.56
NA
NA
NA
1.51
26.78
1.49
5.90
RICO
6.30
3.35
3.96
0.65
4.24
3.24
1.29
3.22
NA
2.78
NA
NA
NA
2.62
1.65
5.50
2.58
4.58
NA
0.28
NA
4.95
7.77
4.17
9.47
6.18
2.57
7.42
3.15
0.36
NA
NA
0.55
3.20
0.56
4.28
SSSD
5.43
9.13
1.82
1.93
14.53
12.33
6.48
8.37
NA
3.59
NA
NA
9.67
13.19
13.59
17.85
9.02
15.48
10.26
1.04
NA
12.77
2.72
10.98
3.91
NA
7.97
NA
10.11
NA
NA
NA
3.12
6.25
2.15
11.36
UCSD
7.03
13.20
NA
3.07
4.90
4.45
4.28
9.11
NA
NA
NA
NA
NA
8.24
9.26
15.25
NA
NA
8.92
1.83
NA
6.77
2.74
NA
NA
NA
7.30
NA
10.09
NA
NA
NA
6.51
6.21
2.00
2.08
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-28
-------
Table 29-10. SNMOC Analytical Precision: Average Coefficient of Variation Based on
Replicate Analyses by Site (Continued)
Pollutant
Isopropylbenzene
2-Methyl-l-butene
3 -Methyl- 1 -butene
2-Methyl- 1 -pentene
4-Methyl-l-pentene
2-Methyl-2-butene
Methylcyclohexane
Methylcyclopentane
2-Methylheptane
3-Methylheptane
2-Methylhexane
3 -Methy Ihexane
2-Methylpentane
3-Methylpentane
w-Nonane
1-Nonene
w-Octane
1-Octene
w-Pentane
1 -Pentene
c/s-2-Pentene
/raws-2-Pentene
a-Pinene
6-Pinene
Propane
«-Propy Ibenzene
Propylene
Propyne
Styrene
Toluene
w-Tridecane
1-Tridecene
1,2,3 -Trimethy Ibenzene
1 ,2,4-Trimethy Ibenzene
1 , 3 ,5 -Trimethy Ibenzene
2,2,3-Trimethylpentane
#of
Pairs
2
11
0
2
0
10
49
61
27
36
46
54
57
59
35
9
35
2
61
55
15
34
22
4
61
16
61
0
7
61
0
0
14
42
29
14
BRCO
NA
NA
NA
NA
NA
NA
1.93
2.24
3.57
8.15
4.18
3.40
1.78
1.51
4.13
2.07
2.55
NA
2.32
7.50
NA
8.97
2.54
NA
0.37
9.10
5.57
NA
NA
4.33
NA
NA
9.70
9.30
4.54
5.67
BTUT
10.12
6.39
NA
NA
1.67
7.42
4.14
6.79
7.27
9.77
5.67
6.56
4.87
3.93
12.92
10.01
5.44
NA
3.35
5.42
8.67
9.87
5.78
5.27
1.02
7.61
2.23
NA
7.68
8.02
NA
NA
15.94
7.19
8.06
5.76
NBIL
NA
NA
NA
NA
NA
6.45
13.25
6.12
7.09
11.48
7.70
5.85
1.19
4.59
5.83
NA
9.59
NA
1.81
15.90
NA
21.01
7.65
NA
0.77
NA
2.43
NA
NA
5.45
NA
NA
NA
6.13
13.76
4.20
PACO
NA
NA
NA
NA
NA
NA
3.58
2.06
8.86
6.11
4.31
5.48
3.87
2.26
6.62
7.60
6.65
11.34
3.57
28.36
NA
14.92
6.28
NA
0.75
16.03
6.07
NA
NA
8.36
NA
NA
7.14
6.64
5.73
23.75
RICO
NA
5.38
NA
NA
NA
7.76
2.73
2.30
1.56
5.85
3.20
2.62
1.46
1.61
4.23
10.21
3.26
NA
1.34
4.00
8.91
5.99
3.51
30.36
0.33
13.42
1.42
NA
NA
2.93
NA
NA
6.32
3.80
8.12
7.25
SSSD
NA
6.33
NA
NA
NA
NA
10.88
7.58
7.17
12.05
9.67
12.89
5.48
8.75
6.36
NA
6.06
NA
3.60
14.84
0.66
14.39
8.53
NA
1.01
NA
1.75
NA
10.06
6.06
NA
NA
0.36
12.14
NA
NA
UCSD
NA
NA
NA
NA
15.53
NA
7.45
10.15
NA
NA
12.12
7.12
7.83
9.02
NA
NA
NA
NA
5.31
8.90
NA
14.86
9.57
14.04
1.45
NA
3.11
NA
2.53
7.72
NA
NA
NA
7.17
NA
NA
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-29
-------
Table 29-10. SNMOC Analytical Precision: Average Coefficient of Variation Based on
Replicate Analyses by Site (Continued)
Pollutant
2,2,4-Trimethylpentane
2,3 ,4-Trimethylpentane
w-Undecane
1-Undecene
7w-Xylene//?-Xylene
o-Xylene
Average by Site
#of
Pairs
46
33
30
19
50
42
7.11
BRCO
21.48
1.87
8.93
8.00
6.23
9.48
5.24
BTUT
5.93
3.90
8.96
6.62
8.17
8.48
7.71
NBIL
7.40
5.85
7.16
4.81
7.21
4.47
7.32
PACO
1.35
3.36
17.53
69.73
6.09
8.15
9.24
RICO
3.39
14.10
8.24
NA
3.14
3.40
4.72
SSSD
5.29
11.31
4.16
12.04
8.09
11.94
8.08
UCSD
8.83
NA
NA
8.41
4.61
NA
7.47
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29.3.3 Carbonyl Compound Analytical Precision
Table 29-11 presents the analytical precision results from replicate analyses of duplicate
and collocated carbonyl compound samples as the average CV per pollutant per site, the average
CV per site, and the overall average CV for NMP sites sampling carbonyl compounds. The
overall average variability was 2.36 percent, which is well within the program DQO of
15 percent CV. The analytical precision results replicate analyses from duplicate and collocated
samples range from 0 percent (hexaldehyde for RICO) to 5.77 percent (tolualdehydes for
PROK). The site-specific average CV ranged from 1.66 percent for RICO to 2.93 percent for
INDEM. Note that RICO is included in Table 29-11 for analytical precision, whereas it was not
included in Table 29-4 for method precision table. This is due to the site having only one valid
collocate that was analyzed in replicate.
29-30
-------
Table 29-11. Carbonyl Compound Analytical Precision:
Based on Replicate Analyses
Average Coefficient of Variation
by Site
Pollutant
Acetaldehyde
Acetone
Benzaldehyde
Butyraldehyde
Crotonaldehyde
2,5-Dimethylbenzaldehyde
Formaldehyde
Hexaldehyde
Isovaleraldehyde
Propionaldehyde
Tolualdehydes
Valeraldehyde
Average by Site
#of
Pairs
281
281
275
281
277
0
281
276
4
281
165
272
2.36
AZFL
0.41
0.73
4.38
4.20
2.83
NA
0.61
4.33
NA
4.12
3.42
3.01
2.80
BTUT
0.16
0.39
2.90
0.95
2.31
NA
1.33
3.43
NA
2.11
5.49
3.28
2.23
CHNJ
0.41
0.68
3.91
2.65
2.77
NA
0.65
4.19
NA
2.84
2.62
5.02
2.57
DEMI
0.41
0.39
3.80
3.45
3.16
NA
0.47
3.17
NA
1.61
5.58
5.04
2.71
ELNJ
0.44
0.23
2.64
2.34
1.37
NA
0.77
3.62
NA
2.14
4.73
1.52
1.98
GPCO
0.53
0.32
2.08
1.40
1.99
NA
0.53
3.35
NA
2.33
4.23
2.80
1.95
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
Table 29-11. Carbonyl Compound Analytical Precision: Average Coefficient of Variation
Based on Replicate Analyses by Site (Continued)
Pollutant
Acetaldehyde
Acetone
Benzaldehyde
Butyraldehyde
Crotonaldehyde
2,5-Dimethylbenzaldehyde
Formaldehyde
Hexaldehyde
Isovaleraldehyde
Propionaldehyde
Tolualdehydes
Valeraldehyde
Average by Site
#of
Pairs
281
281
275
281
277
0
281
276
4
281
165
272
2.36
INDEM
0.85
0.40
5.20
4.13
3.86
NA
0.56
4.02
NA
2.32
4.38
3.61
2.93
MWOK
0.34
0.41
4.76
2.95
1.04
NA
0.72
3.42
NA
1.93
4.50
2.24
2.23
NBIL
0.71
0.70
3.42
2.78
3.73
NA
0.89
2.99
NA
2.89
2.88
2.81
2.38
NBNJ
0.20
0.17
4.68
2.88
3.30
NA
0.57
5.60
NA
2.50
2.28
2.31
2. 45
OCOK
0.33
0.34
2.57
1.62
2.51
NA
0.56
4.42
4.35
1.81
3.80
4.54
2.44
ORFL
0.50
1.28
4.76
3.56
2.02
NA
0.77
2.22
NA
3.31
4.64
3.41
2.64-
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-31
-------
Table 29-11. Carbonyl Compound Analytical Precision: Average Coefficient of Variation
Based on Replicate Analyses by Site (Continued)
Pollutant
Acetaldehyde
Acetone
Benzaldehyde
Butyraldehyde
Crotonaldehyde
2,5-Dimethylbenzaldehyde
Formaldehyde
Hexaldehyde
Isovaleraldehyde
Propionaldehyde
Tolualdehydes
Valeraldehyde
Average by Site
#of
Pairs
281
281
275
281
277
0
281
276
4
281
165
272
2.36
PROK
0.22
0.27
3.59
2.38
2.30
NA
0.58
4.41
NA
1.03
5.77
4.41
2.49
PXSS
1.17
0.30
2.92
1.09
3.62
NA
0.41
1.79
NA
3.16
5.18
5.01
2.47
RICO
0.77
0.35
2.32
2.97
1.94
NA
0.45
0
NA
2.50
NA
3.63
1.66
S4MO
0.18
0.80
2.94
3.08
2.05
NA
0.61
4.39
NA
2.70
4.21
2.04
2.30
SEWA
0.40
0.47
2.84
1.87
3.55
NA
0.46
3.40
NA
2.06
2.92
3.63
2. 16
SKFL
0.41
0.43
4.34
4.03
1.31
NA
0.74
4.62
NA
2.05
3.82
3.66
2.54
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
Table 29-11. Carbonyl Compound Analytical Precision: Average Coefficient of Variation
Based on Replicate Analyses by Site (Continued)
Pollutant
Acetaldehyde
Acetone
Benzaldehyde
Butyraldehyde
Crotonaldehyde
2,5-Dimethylbenzaldehyde
Formaldehyde
Hexaldehyde
Isovaleraldehyde
Propionaldehyde
Tolualdehydes
Valeraldehyde
Average by Site
#of
Pairs
281
281
275
281
277
0
281
276
4
281
165
272
2.36
SPIL
0.65
0.42
2.47
1.81
3.30
NA
0.94
3.81
NA
2.41
4.49
4.43
2.47
SSSD
0.30
0.26
2.05
1.90
3.50
NA
0.52
3.89
NA
1.98
4.26
3.70
2.24
SYFL
0.63
0.82
3.24
1.67
1.78
NA
0.49
2.37
NA
2.91
4.79
2.78
2. 15
TMOK
0.37
0.34
3.36
1.24
2.36
NA
0.90
3.37
NA
1.22
5.04
2.84
2.11
TOOK
0.43
0.38
3.24
2.01
1.95
NA
0.57
4.28
NA
1.85
3.92
2.31
2.09
UCSD
0.47
0.41
3.47
1.77
3.09
NA
0.88
4.11
NA
2.35
4.19
2.54
2.33
WPIN
0.63
0.63
3.33
2.87
3.34
NA
0.60
3.41
2.63
3.15
4.69
3.48
2.61_
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29-32
-------
29.3.4 Metals Analytical Precision
Table 29-12 presents analytical precision results from replicate analyses of all collocated
metals samples as the average CV per pollutant per site, the average CV per site, and the overall
average CV for NMP sites sampling metals. The results from collocated samples show low- to
high-level variability among sites, ranging from an average CV of 0 percent (cadmium for
UNVT) to 20.64 percent (selenium for BTUT). The site-specific average CV ranged from
3.33 percent for TOOK to 9.06 percent for BTUT. The overall analytical precision was
5.44 percent.
Table 29-12. Metals Analytical Precision: Average Coefficient of Variation Based on
Replicate Analyses by Site
Pollutant
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Cobalt
Lead
Manganese
Mercury
Nickel
Selenium
Average by Site
#of
Pairs
283
281
216
257
33
275
277
283
178
51
275
5.44
BOMA
2.04
2.46
16.83
7.14
NA
5.26
1.48
1.20
12.71
0.88
3.87
5.39
BTUT
13.04
5.51
NA
7.28
NA
7.56
2.18
5.69
NA
10.57
20.64
9.06
NBIL
1.09
1.84
14.89
1.62
NA
3.83
4.02
3.68
4.63
2.20
2.25
4.00
S4MO
1.27
1.14
8.51
5.00
0.92
8.05
1.46
2.21
11.63
NA
2.15
4.23
TOOK
4.09
1.87
5.94
2.41
1.24
3.46
2.13
2.28
9.05
1.05
3.10
3.33
UNVT
9.88
9.18
NA
0
NA
7.86
3.02
3.21
NA
5.75
14.03
6.62
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designated NATTS Site
Shading indicates NATTS MQO Core Analyte
29.3.5 Hexavalent Chromium Analytical Precision
Table 29-13 presents analytical precision results from replicate analyses of all collocated
hexavalent chromium samples as the average CV per site and the overall average CV for NMP
sites sampling hexavalent chromium. Hexavalent chromium is a NATTS MQO Core Analyte
and the sites shown are NATTS sites. The range of variability for hexavalent chromium was
2.04 percent (for MONY) to 13.28 percent (for SKFL), with an overall analytical precision of
6.30 percent.
29-33
-------
Table 29-13. Hexavalent Chromium Analytical Precision: Average Coefficient of Variation
Based on Replicate Analyses by Site
Site
BOMA
BTUT
BXNY
CAMS 35
CHSC
DEMI
GLKY
GPCO
HOW
MONY
NBIL
PRRI
PXSS
RIVA
ROCH
S4MO
SDGA
SEWA
SKFL
SYFL
UNVT
WADC
Average CV
# of Pairs
Average
CV
5.42
4.52
6.46
5.48
9.81
6.09
7.66
4.42
6.56
2.04
5.97
4.58
5.55
6.16
7.58
4.78
8.82
6.18
13.28
4.72
2.96
9.47
6.30
226
BOLD ITALICS = EPA-designated
NATTS Site.
29.3.6 PAH Analytical Precision
Table 29-14 presents analytical precision results from replicate analyses of all collocated
PAH samples as the average CV per pollutant per site, the average CV per site, and the overall
average CV for NMP sites sampling PAH. The analytical precision results from replicate
analysis of collocated samples show low-level variability among sites, ranging from 0.68 percent
(anthracene for SYFL) to 15.49 percent (acenaphthylene for RUCA). The site-specific average
CV ranged from 2.95 percent for DEMI to 4.37 percent for SDGA. The overall average CV for
all sites was 3.67 percent.
29-34
-------
Table 29-14. PAH Analytical Precision: Average Coefficient of Variation Based on
Replicate Analyses by Site
Pollutant
Acenaphthene
Acenaphthylene
Anthracene
Benzo(a)anthracene
Benzo(a)pyrene
Benzo(b)fluoranthene
Benzo(e)pyrene
Benzo(g,h,i)perylene
Benzo(k)fluoranthene
Chrysene
Coronene
Cyclopenta[cd]pyrene
Dibenz(a,h)anthracene
Fluoranthene
Fluorene
9-Fluorenone
Indeno( 1,2,3 -cd)pyrene
Naphthalene
Perylene
Phenanthrene
Pyrene
Retene
Average by Site
#of
Pairs
57
24
31
13
17
28
26
37
14
55
25
4
4
57
57
57
30
57
18
57
57
44
3.67
DEMI
2.66
7.13
2.31
1.70
3.98
2.45
2.82
1.90
7.33
2.11
1.63
3.40
1.23
2.22
2.87
2.50
3.68
1.29
2.43
1.37
2.58
5.41
2.95
RUCA
6.89
15.49
4.79
5.16
1.61
4.74
3.84
2.32
3.45
2.65
5.00
1.94
NA
1.99
2.20
2.86
2.96
3.87
1.52
1.89
2.17
5.15
3.93
SDGA
3.43
4.27
10.97
5.35
6.90
7.64
3.17
4.22
NA
5.95
5.88
NA
NA
2.23
2.26
2.76
4.30
2.55
3.52
1.13
2.90
3.59
4.37
SEWA
4.34
1.93
2.01
3.18
3.03
2.40
6.86
4.87
11.64
2.41
3.47
4.68
6.22
2.14
2.38
2.86
4.24
2.21
4.92
1.45
2.32
2.34
3.72
SYFL
2.98
3.12
0.68
NA
2.86
1.57
5.58
2.26
NA
4.49
9.45
NA
NA
2.27
6.12
2.18
3.42
3.72
3.80
1.72
1.87
2.50
3.37
NA=No pairs with concentrations greater than or equal to the MDL
BOLD ITALICS = EPA-designaled NATTS Site
Shading indicates NATTS MQO Core Analyte
29.4 Accuracy
Laboratories typically evaluate their accuracy (or bias) by analyzing audit samples that
are prepared by an external source. The pollutants and the respective concentrations of the audit
samples are unknown to the laboratory. The laboratory analyzes the samples and the external
source compares the measured concentrations to the reference concentrations of those audit
samples and calculates a percent difference. Accuracy, or bias, indicates the extent to which
experimental measurements represent their corresponding "true" or "actual" values.
Laboratories participating in the NATTS program are provided with proficiency test (PT)
audit samples for VOC, carbonyl compounds, metals, hexavalent chromium, and PAH which are
used to quantitatively measure analytical accuracy. Tables 29-15 through 29-19 present ERG's
results from the 2010 NATTS PT audit samples for VOC, carbonyl compounds, metals,
29-35
-------
hexavalent chromium, and PAH, respectively. The program DQO for the percent difference from
the true value is ± 25 percent, and the values exceeding this criterion are bolded in the tables.
Shaded rows present results for NATTS MQO Core Analytes.
Table 29-15. VOC NATTS PT Audit Samples: Percent Difference from True Value
Pollutant
Acrolein
Benzene
1,3 -Butadiene
Carbon Tetrachloride
Chloroform
1 ,2-Dibromoethane
1 ,2-Dichloroethane
Dichloromethane
1 ,2-Dichloropropane
cis- 1 ,3 -Dichloropropene
trans- 1 ,3 -Dichloropropene
1 , 1 ,2,2-tetrachloroethane
Tetrachloroethylene
rrichloroethylene
Vinyl Chloride
March, 2010
-9.5
-13.2
-3.7
31.3
1.0
-9.0
-2.8
4.5
-15.2
-18.2
-20.2
-14.3
-16.1
-6.4
-14.1
Shading indicates NATTS MQO Core Analyte
Table 29-16. Carbonyl Compound NATTS PT Audit Samples: Percent Difference from
True Value
Pollutant
Formaldehyde
Acetaldehyde
May, 2010
-2.8
0.7
Shading indicates NATTS MQO Core Analyte
Table 29-17. Metals NATTS PT Audit Samples: Percent Difference from True Value
Pollutant
Arsenic
Beryllium
Cadmium
Lead
Manganese
Nickel
February, 2010
7.3
11.2
4.9
-3.5
0.6
4.7
Shading indicates NATTS MQO Core Analyte
29-36
-------
Table 29-18. Hexavalent Chromium PT Audit Samples: Percent Difference from True
Value
Pollutant
[Hexavalent Chromium
January, 2010
10.5
Shading indicates NATTS MQO Core Analyte
Table 29-19. PAH NATTS PT Audit Samples: Percent Difference from True Value
Pollutant
Acenaphthene
Anthracene
Benzo(a)pyrene
Fluoranthene
Fluorene
Naphthalene
Phenanthrene
Pyrene
February, 2010
-15.4
-9.2
-2.3
-9.3
-13.4
-17.1
-13.5
-9.4
Shading indicates NATTS MQO Core Analyte
The accuracy of the 2010 monitoring data can also be assessed qualitatively by reviewing
the accuracy of the monitoring methods and how they were implemented:
• The sampling and analytical methods used in the 2010 monitoring effort have
been approved by EPA for accurately measuring ambient levels of various
pollutants - an approval that is based on many years of research into the
development of ambient air monitoring methodologies.
• When collecting and analyzing ambient air samples, all field sampling staff and
laboratory analysts are required to strictly adhere to quality control and quality
assurance guidelines detailed in the respective monitoring methods. This strict
adherence to the well-documented sampling and analytical methods suggests that
the 2010 monitoring data accurately represent ambient air quality.
29-37
-------
30.0 Results, Conclusions, and Recommendations
The following discussion summarizes the results of the data analyses contained in this
report and presents recommendations applicable to future air monitoring efforts. As
demonstrated by the results of the data analyses discussed throughout this report, NMP
monitoring data offer a wealth of information for assessing air quality by evaluating trends,
patterns, correlations, and the potential for health risk and should ultimately assist a wide range
of audiences understand the complex nature of air pollution.
30.1 Summary of Results
Analyses of the 2010 monitoring data identified the following notable results,
observations, trends, and patterns in the program-level and state- and site-specific air pollution
data.
30.1.1 National-level Summary
• Number of participating NATTS sites. Twenty-six of the 52 sites are EPA-designated
NATTS sites (BOMA, BTUT, BXNY, CAMS 35, CAMS 85, CELA, CHSC, DEMI,
GLKY, GPCO, HOWI, MONY, NBIL, PRRI, PXSS, RIVA, ROCH, RUCA, S4MO,
SDGA, SEW A, SJJCA, SKFL, SYFL, UNVT, and WADC).
• Total number of samples collected and analyzed. Over 8,500 samples were collected
yielding over 214,900 valid measurements of air toxics.
• Detects. The detection of a given pollutant is subject to the analytical methods used
and the limitations of the instruments. Simply stated, a method detection limit is the
lowest concentration of a substance that can be measured and reported with 99
percent confidence that the pollutant concentration is greater than zero.
Approximately 55 percent of the reported measurements were above the associated
MDLs. Of the 187 pollutants monitored, only three pollutants were not detected over
the course of the 2010 monitoring effort: frvms-l^-dichloropropene,
2,5-dimethylbenzaldehyde, and propyne.
• Program-level Pollutants of Interest. The pollutants of interest at the program-level
are based on the number of exceedances, or "failures," of the preliminary risk
screening values. In addition, 18 NATTS MQO Core Analytes (excluding acrolein)
are classified as pollutants of interest. Only two NATTS MQO Core Analytes
(beryllium and tetrachloroethylene) did not fail any screens.
• Noncancer Risk Screening using A TSDR MRLs. One preprocessed daily measurement
of formaldehyde (measured at NBIL) and two preprocessed daily measurements of
dichloromethane (measured at BTUT and GPCO) were higher than the associated
30-1
-------
ATSDR acute MRLs. None of the quarterly or annual average concentrations of any
pollutants were higher than the associated ATSDR intermediate or chronic MRLs.
• Cancer Surrogate Risk Approximations. The cancer surrogate risk approximation
calculated for ELNJ for formaldehyde's annual average (57.93 in-a-million) was the
highest of all annual average-based cancer risk approximations. No other sites
exhibited cancer risk approximations greater than 50 in-a-million.
• Noncancer Surrogate Risk Approximations. The noncancer surrogate risk
approximation calculated for TOOK's annual average of manganese (an HQ of 0.47)
was the highest of all annual average-based noncancer risk approximations. No site
had noncancer risk approximations greater than 1.0.
• Emissions and Toxicity Weighted Emissions. Benzene, formaldehyde, and
ethylbenzene often had the highest county-level emissions for the participating
counties (of those with a cancer URE). Both benzene and formaldehyde tended to
have the highest toxicity-weighted emissions. Toluene and xylenes were often the
highest emitted pollutants with noncancer risk factors, although they rarely had top 10
toxicity-weighted emissions. Acrolein tended to have the highest toxicity-weighted
emissions of pollutants with noncancer RfCs, although it was rarely emitted in high
enough quantities to rank in the top 10 emissions for the participating counties.
30.1.2 State-level Summary
Arizona.
• The Arizona monitoring sites are located in Phoenix. PXSS is a NATTS site; SPAZ is
a UATMP site.
• Back trajectories originated from a variety of directions at PXSS and SPAZ, though
many are from the southwest and west. Their air shed domains were smaller in size
compared to other NMP monitoring sites, as nearly all trajectories originated within
250 miles of the sites.
• The wind roses show that calm, easterly, westerly, and east-southeasterly winds were
prevalent near PXSS and SPAZ.
• PXSS sampled for VOC, carbonyl compounds, PAH, metals (PMi0), and hexavalent
chromium. SPAZ sampled for VOC only.
• Twenty-two pollutants, of which 14 are NATTS MQO Core Analytes, failed screens
for PXSS. PXSS failed the second highest number of screens among all NMP sites.
• Nine pollutants failed screens for SPAZ, of which four are NATTS MQO Core
Analytes.
30-2
-------
• Of the pollutants of interest for PXSS, benzene had the highest annual average
concentration. Benzene also had the highest annual average concentration for SPAZ.
This was the only pollutant for either site with an annual average concentration
greater than l|Jg/m3.
• PXSS had the highest annual average concentration of tetrachloroethylene and the
second highest annual average concentrations of 1,3-butadiene, hexavalent
chromium, beryllium, lead, and manganese among all NMP sites sampling these
pollutants.
• SPAZ had the highest annual average concentrations of acrylonitrile and
1,3-butadiene compared to all NMP sites sampling these pollutants.
• Annual average concentrations could not be calculated for carbonyl compounds for
PXSS due to a sampler problem that led to invalidation of all 2010 carbonyl samples
from mid-February to the end of the year.
• Metals and hexavalent chromium sampling for has occurred at PXSS for at least five
consecutive years; thus, a trends analysis was conducted for arsenic, hexavalent
chromium, and manganese. Average rolling concentrations of arsenic and hexavalent
chromium exhibit little change over the years of sampling. Manganese exhibits a
slightly decreasing trend in the rolling average concentrations over the sampling
period.
• None of the measured detections or time-period average concentrations of the
pollutants of interest, where they could be calculated, were greater than their
respective ATSDR MRL noncancer health risk benchmarks for either of the Arizona
monitoring sites.
• Benzene and 1,3-butadiene had the highest cancer risk approximations for PXSS
while acrylonitrile and benzene had the highest cancer risk approximations for SPAZ.
None of the pollutants of interest for either site had a noncancer risk approximation
greater than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Maricopa
County, while toluene was the highest emitted pollutant with a noncancer risk factor.
Formaldehyde had the highest cancer toxicity-weighted emissions, while acrolein had
the highest noncancer toxicity-weighted emissions for Maricopa County.
California.
• The three California monitoring sites are located in Los Angeles (CELA), Rubidoux
(RUCA), and San Jose (SJJCA). All three are NATTS sites.
• Back trajectories for CELA and RUCA primarily originated from the northwest, with
a secondary cluster originating from the northeast. Their air shed domains were
30-3
-------
smaller in size compared to other NMP monitoring sites as nearly all trajectories
originated within 300 miles of the sites. The back trajectories for SJJCA primarily
originated from the northwest to north to northeast directions. The air shed domain
for SJJCA is larger than the other two California sites; nearly all trajectories
originated within 400 miles of the site.
• CELA experienced primarily calm winds, although those greater than 2 knots were
predominantly from the west. Westerly winds were prevalent near RUCA. SJJCA
experienced predominantly westerly to northwesterly winds.
• CELA and RUCA sampled for PAH only. SJJCA sampled for PAH and metals
(PMio).
• Four pollutants failed screens for CELA, of which two (naphthalene and
benzo(a)pyrene) are NATTS MQO Core Analytes. Two pollutants failed screens for
RUCA, both of which are NATTS MQO Core Analytes. Four pollutants (arsenic,
naphthalene, manganese, and nickel) failed screens for SJJCA, all of which are
NATTS MQO Core Analytes.
• Naphthalene had the highest annual average concentration for each site. The annual
average concentration of naphthalene for CELA was significantly higher than the
annual average for RUCA and more than double the annual average for SJJCA, and
second highest compared to all NMP sites sampling naphthalene.
• None of the measured detections or time-period average concentrations of the
pollutants of interest were greater than their respective ATSDR MRL noncancer
health risk benchmarks.
• Of the pollutants of interest for each site, naphthalene exhibited the highest cancer
risk approximation for all three California sites. The noncancer surrogate risk
approximations for each pollutant of interest were less than 1.0 for all three sites.
• Formaldehyde was the highest emitted pollutant with a cancer risk factor in Los
Angeles, Riverside, and Santa Clara Counties; formaldehyde also had the highest
cancer toxicity-weighted emissions for Los Angeles and Santa Clara Counties while
hexavalent chromium had the highest cancer toxicity-weighted emissions for
Riverside County.
• 1,1,1 -Trichloroethane was the highest emitted pollutant with a noncancer risk factor
in Los Angeles County, while toluene was the highest emitted pollutant with a
noncancer risk factor in Riverside and Santa Clara Counties. Acrolein had the highest
noncancer toxicity-weighted emissions for all three counties.
30-4
-------
Colorado.
• The NATTS site in Colorado is located in Grand Junction (GPCO). There are also
five CSATAM sites located northeast of Grand Junction in Garfield County. The sites
are located in the towns of Battlement Mesa (BMCO), Silt (BRCO), Parachute
(PACO), Rifle (RICO), and Rulison (RUCO). The instruments at RUCO were
moved to the new location at Battlement Mesa in September 2010.
• Back trajectories originated from a variety of directions at GPCO, though almost all
had a westerly component. The 24-hour air shed domain GPCO was smaller in size
than other NMP monitoring sites, with most back trajectories originating less than
300 miles from the site. The Garfield County sites had air shed domains of similar
size to GPCO, which is expected given the close proximity of these sites to GPCO.
• The wind roses for GPCO show that easterly, east-southeasterly, and southeasterly
winds were prevalent near the site. Westerly and southerly winds were prevalent for
the Garfield County sites.
• GPCO sampled for VOC, carbonyl compounds, PAH, and hexavalent chromium. The
Garfield County sites sampled for SNMOC and carbonyl compounds.
• Nineteen pollutants failed at least one screen for GPCO, of which seven are NATTS
MQO Core Analytes. The number of pollutants that failed screens for the Garfield
County sites ranged from four (BMCO) to five (BRCO, PACO, RICO, and RUCO).
• Of the pollutants of interest for GPCO, formaldehyde had the highest annual average
concentration, followed by acetaldehyde and benzene.
• Benzene had the highest annual average concentration for each of the Garfield
County sites (where an annual average concentration could be calculated).
• GPCO had the highest annual average concentration of naphthalene and the second
highest annual average concentrations of benzo(a)pyrene and tetrachloroethylene
among all NMP sites sampling these pollutants. Annual average benzene
concentrations for the Colorado sites account for five of the 10 highest annual
average concentrations for sites that sampled benzene. In addition, PACO and RICO
have the highest and second highest annual average concentrations, respectively, of
ethylbenzene.
• VOC, carbonyl compound, and hexavalent chromium sampling has occurred at
GPCO for at least five consecutive years; thus, a trends analysis was conducted for
acetaldehyde, benzene, 1,3-butadiene, formaldehyde, and hexavalent chromium. The
rolling average concentration for acetaldehyde shows little change after the
2004-2006 timeframe. Benzene, hexavalent chromium and 1,3-butadiene exhibited a
slight decreasing trend. Formaldehyde exhibited an increasing trend through the
30-5
-------
2007-2009 timeframe, after which the rolling average concentration decreased
slightly.
• Of the 59 measured detections, one dichloromethane measurement was greater than
the ATSDR acute MRL for GPCO. This was also the highest measured detection
among all NMP sites sampling dichloromethane.
• For sites where annual averages of formaldehyde could be calculated (GPCO and
PACO), formaldehyde had the highest cancer risk approximations. The cancer risk
approximations for benzene were greater than 10 in-a-million for all sites except
BMCO (where an annual average could not be calculated) and BRCO. All noncancer
risk approximations were less than 1.0 for all six Colorado sites.
• Benzene was the highest emitted pollutant with a cancer risk factor in Mesa County,
while formaldehyde was the highest emitted pollutant with a cancer risk factor in
Garfield County. Formaldehyde had the highest cancer toxicity-weighted emissions
for both counties.
• While toluene was the highest emitted pollutant with a noncancer risk factor for both
counties, acrolein had the highest noncancer toxicity-emissions.
District of Columbia
• The Washington, D.C. monitoring site (WADC) is a NATTS site.
• Back trajectories originated from a variety of directions at WADC, with most of them
originating from the southwest, west, and northwest.
• The wind roses show that southerly and south-southwesterly winds were prevalent
near WADC.
• WADC sampled for hexavalent chromium and PAH. The only pollutants to fail
screens for WADC were naphthalene and fluorene. Naphthalene accounted for nearly
97 percent of the failed screens for the site.
• The pollutant with the highest annual average concentrations for WADC was
naphthalene, which ranked the sixth highest annual average concentration among
NMP sites sampling PAH.
• Hexavalent chromium sampling has occurred at WADC for at least five consecutive
years; thus, a trends analysis was conducted for hexavalent chromium. Hexavalent
chromium exhibited a slight decreasing trend through the 2007-2009 timeframe, after
which a slight increase in rolling average concentration was exhibited.
30-6
-------
• None of the measured detections or time-period average concentrations of the
pollutants of interest for WADC were greater than their respective ATSDR MRL
noncancer health risk benchmarks.
• Naphthalene had the only cancer risk approximation for WADC greater than
1.0 in-a-million, while none of the pollutants of interest had a noncancer risk
approximation greater than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in the District of
Columbia, while toluene was the highest emitted pollutant with a noncancer risk
factor. Formaldehyde had the highest cancer toxicity-weighted emissions, while
acrolein had the highest noncancer toxicity-weighted emissions in the District.
Florida.
• Three of the Florida monitoring sites are located in the Tampa-St. Petersburg-
Clearwater MSA (SYFL, AZFL, and SKFL) and two are located in the Orlando-
Kissimmee MSA (ORFL and PAFL. SKFL and SYFL are NATTS sites.
• Back trajectory maps were similar for the Tampa/St. Petersburg sites, where back
trajectories originated from a variety of directions for each of the Florida sites. Back
trajectories also originated from a variety of directions for ORFL and PAFL.
• Winds from a variety of directions were observed near the Tampa/St. Petersburg sites,
however winds from the southwest quadrant were observed the least. Winds from a
variety of directions were observed near the Orlando sites.
• AZFL and ORFL sampled for carbonyl compounds only. SKFL and SYFL sampled
for hexavalent chromium and PAH in addition to carbonyl compounds. PAFL
sampled only PMi0 metals.
• Acetaldehyde and formaldehyde were the only pollutants to fail screens for AZFL
and ORFL, where only carbonyl compounds were sampled. Naphthalene, in addition
to acetaldehyde and formaldehyde, failed screens for SKFL and SYFL. Arsenic, lead,
nickel, and manganese failed screens for PAFL, where only metals were sampled.
• Acetaldehyde had the highest annual average concentration for AZFL and SKFL,
while formaldehyde had the highest annual average concentration for SYFL and
ORFL. SKFL and AZFL had the second and third highest concentrations of
acetaldehyde, respectively, among all NMP sites sampling carbonyl compounds. Lead
and manganese had the highest annual average concentrations for PAFL.
• Carbonyl compound sampling has been conducted at AZFL, ORFL, SKFL, and
SYFL for at least five consecutive years; thus a trends analysis was conducted for
acetaldehyde and formaldehyde. In the later years of sampling, the rolling average
acetaldehyde concentrations have increased at AZFL and SKFL and decreased at
30-7
-------
ORFL and SYFL while the rolling average formaldehyde concentrations have
decreased at AZFL, ORFL and SKFL and increased at SYFL.
• Hevavalent chromium sampling has occurred at SYFL since 2005; thus a trends
analysis was conducted. Rolling average and median concentrations exhibit a
decreasing trend since the onset of hexavalent chromium sampling, with a slight
increase in rolling average concentration for the 2008-2010 time period.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for any of the Florida monitoring sites were greater than their
respective ATSDR MRL noncancer health risk benchmarks.
• For the four Florida sites sampling carbonyl compounds, formaldehyde had the
highest cancer surrogate risk approximations. Arsenic had the highest cancer risk
approximation for the site sampling metals (PAFL). All noncancer risk
approximations for the Florida sites' pollutants of interest were less than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in all three Florida
counties. Benzene also had the highest cancer toxicity-weighted emissions for
Pinellas County, while formaldehyde had the highest cancer toxicity-weighted
emissions for Hillsborough and Orange Counties.
• Toluene was the highest emitted pollutant with a noncancer risk factor in all three
Florida counties. Acrolein had the highest noncancer toxicity-weighted emissions for
all three counties.
Georgia.
• The SDGA monitoring site located in Decatur, south of Atlanta, is a NATTS site.
• Back trajectories originated from a variety of directions at SDGA, though trajectories
from the southwest were most common.
• The wind roses show that winds from the west to north-northwest were prevalent near
SDGA. Easterly winds were also common.
• SDGA sampled for PAH and hexavalent chromium. Naphthalene, acenaphthene, and
benzo(a)pyrene failed screens for SDGA, with naphthalene accounting for nearly 97
percent of the total failed screens.
• Of the pollutants of interest for SDGA, naphthalene had the highest annual average
concentration, ranking fifth among NMP sites sampling PAH.
• Hexavalent chromium sampling has occurred at SDGA for at least five consecutive
years; thus, a trends analysis was conducted. Hexavalent chromium exhibited a
decreasing trend over the period of sampling.
30-8
-------
• None of the measured detections or time-period average concentrations of the
pollutants of interest for SDGA were greater than their respective ATSDR MRL
noncancer health risk benchmarks.
• Naphthalene was the only pollutant with a cancer risk approximation greater than
1.0 in-a-million. None of SDGA's pollutants of interest had a noncancer risk
approximation greater than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in DeKalb
County, while toluene was the highest emitted pollutant with a noncancer risk factor.
Benzene also had the highest cancer toxicity-weighted emissions, while acrolein had
the highest noncancer toxicity-weighted emissions in DeKalb County.
Illinois.
• The Illinois monitoring sites are located near Chicago. NBIL is a NATTS site located
in Northbrook and SPIL is a UATMP site located in Schiller Park.
• Back trajectories originated from a variety of directions at the sites, although back
trajectories primarily originated from the south, north, and northwest.
• The wind roses show that winds from a variety of directions were observed near the
monitoring sites, although winds from the southeast quadrant were infrequently
observed.
• Both Illinois sites sampled for VOC and carbonyl compounds. NBIL also sampled for
SNMOC, PAH, hexavalent chromium and metals (PMio).
• Twenty pollutants failed screens for NBIL, of which 12 are NATTS MQO Core
Analytes. Fourteen pollutants failed screens for SPIL, of which six are NATTS MQO
Core Analytes.
• Of the pollutants of interest for NBIL and SPIL, formaldehyde had the highest annual
average concentrations. NBIL had the highest annual average concentration of
chloroform measured among NMP sites sampling this pollutant, and SPIL had the
highest annual concentration of trichloroethylene.
• VOC and carbonyl compound sampling have been conducted at NBIL and SPIL for
at least five consecutive years; thus, a trends analysis was conducted for
acetaldehyde, benzene, 1,3 -butadiene, and formaldehyde for both sites. Most recently,
rolling average concentrations of acetaldehyde and benzene have increased at both
sites, while 1,3 -butadiene exhibited little change. The rolling average concentration of
formaldehyde has increased at NBIL and decreased at SPIL.
metals and hexavalent chromium sampling have been conducted at NBIL for at
least five consecutive years; thus, a trends analysis was conducted for arsenic,
30-9
-------
hexavalent chromium, and manganese. Most recently, rolling average concentrations
have decreased slightly for arsenic and manganese, while an increase in rolling
average concentration is shown for hexavalent chromium in the final time frame.
• Of 55 measured detections of formaldehyde at NBIL, one measurement was greater
than the ATSDR acute MRL for this pollutant. Yet none of the time-period average
concentrations of formaldehyde were greater than their respective ATSDR MRL
noncancer health risk benchmarks. None of the measured detections or time-period
average concentrations of the pollutants of interest at SPIL were greater than their
respective ATSDR MRL noncancer health risk benchmarks.
• Formaldehyde had the highest cancer risk approximations for both sites. All
noncancer risk approximations for the Illinois sites' pollutants of interest were less
than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Cook County,
while formaldehyde had the highest cancer toxicity-weighted emissions. Toluene was
the highest emitted pollutant with a noncancer risk factor, while acrolein had the
highest noncancer toxicity-weighted emissions for Cook County.
Indiana.
• There are two Indiana monitoring sites, one located in Indianapolis (WPIN), and a
second located in Gary, near Chicago (INDEM). Both are UATMP sites.
• Back trajectories originated from a variety of directions at the Indiana sites, although
less frequently from the east and southeast. Trajectories originating from the east
were generally shorter than trajectories originating from other directions.
• The wind roses show that winds from the southwest and northwest quadrants,
including due south and due north, were observed most frequently near WPIN. Winds
from the south, south-southwest, and west were observed most frequently near
INDEM.
• WPIN and INDEM sampled for carbonyl compounds only.
• Formaldehyde and acetaldehyde failed screens for both INDEM and WPIN;
propionaldehyde also failed screen a single screen for INDEM.
• Of the pollutants of interest, formaldehyde had the highest annual average
concentrations for both sites. WPIN had the fourth and sixth highest annual average
concentrations of formaldehyde and acetaldehyde, respectively, among NMP sites
sampling carbonyl compounds.
• Carbonyl compound sampling has been conducted at INDEM for at least five
consecutive years; thus, a trends analysis was conducted for acetaldehyde and
30-10
-------
formaldehyde. The average rolling concentrations of both acetaldehyde and
formaldehyde have decreased at INDEM, beginning with the 2007-2009 time period.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for either Indiana site were greater than their respective ATSDR
MRL noncancer health risk benchmarks.
• Formaldehyde had the highest cancer risk approximations for both Indiana sites.
Neither site's pollutants of interest had a noncancer risk approximation greater than
1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Marion and
Lake Counties, while coke oven emissions (PM) had the highest cancer toxicity-
weighted emissions for Lake County and formaldehyde had the highest cancer
toxicity-weighted emissions for Marion County.
• Toluene was the highest emitted pollutant with a noncancer risk factor in both Lake
and Marion Counties. Acrolein had the highest noncancer toxicity-weighted
emissions for both counties.
Kentucky.
• The Kentucky monitoring site is located near Grayson Lake, south of Grayson,
Kentucky (GLKY). GLKY is a NATTS site.
• Back trajectories originated from a variety of directions for both sites, with the
majority originating to the south, west and northwest, and north of the site. .
• Although calm winds were prevalent near GLKY, winds from the southwest were the
most frequently observed wind directions for winds greater than 2 knots.
• GLKY sampled for hexavalent chromium, PAH, and VOC. Six pollutants failed
screens for GLKY, of which four are NATTS MQO Core Analytes.
• Annual averages could not be calculated for VOC compounds because sampling did
not begin until June. Of the pollutants of interest for which annual averages could be
calculated, naphthalene had the highest annual average concentration for GLKY.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for GLKY, where they could be calculated, were greater than
their respective ATSDR MRL noncancer health risk benchmarks.
• None of the pollutants of interest for GLKY had cancer surrogate risk approximations
greater than 1 in-a-million; similarly, none of the pollutants of interest for GLKY had
noncancer surrogate risk approximations greater than 1.0.
30-11
-------
• Benzene was the highest emitted pollutant with a cancer risk factor in Carter County
and had the highest cancer toxicity-weighted emissions. Toluene was the highest
emitted pollutant with a noncancer risk factor, while acrolein had the highest
noncancer toxicity-weighted emissions in Carter County.
Massachusetts.
• The Massachusetts monitoring site (BOMA) is a NATTS site in Boston.
• Back trajectories originated from a variety of directions at BOMA. Although the bulk
of trajectories originated from the west to northwest to north of the site.
• The wind roses show that winds from the west, west-northwest, and northwest were
prevalent near BOMA.
• BOMA sampled for metals (PMi0), PAH, and hexavalent chromium.
• Five pollutants failed screens for BOMA, all of which are NATTS MQO Core
Analytes. Naphthalene accounted for half of the site's failed screens.
• Of the pollutants of interest, naphthalene had the highest annual average
concentration. BOMA's concentrations of cadmium and nickel ranked second highest
among sites sampling PMio metals.
• Metals and hexavalent chromium sampling has been conducted at BOMA for at least
five consecutive years; thus, a trends analysis was conducted for arsenic, hexavalent
chromium, and manganese. The rolling average concentrations of arsenic have
changed little; hexavalent chromium concentrations have decreased slightly; and
manganese concentrations have decreased over the period of sampling at BOMA.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for BOMA were greater than their respective ATSDR MRL
noncancer health risk benchmarks.
• The only pollutants with cancer risk approximations greater than 1.0 in-a-million
were arsenic and naphthalene. None of the pollutants of interest for BOMA had
noncancer risk approximations greater than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Suffolk
County, while formaldehyde had the highest cancer toxicity-weighted emissions.
Toluene was the highest emitted pollutant with a noncancer risk factor in Suffolk
County, while acrolein had the highest noncancer toxicity-weighted emissions.
Michigan.
• DEMI is a NATTS site located in Dearborn, Michigan, near Detroit.
30-12
-------
• Back trajectories originated from a variety of directions at DEMI, although less
frequently from the east and southeast.
• The wind roses for DEMI show that winds from a variety of directions were observed
near the monitoring site, although winds from the southeast quadrant were observed
the least.
• DEMI sampled for VOC, carbonyl compounds, PAH, and hexavalent chromium.
• Sixteen pollutants failed screens for DEMI, of which nine are NATTS MQO Core
Analytes.
• Formaldehyde and acetaldehyde had the highest annual average concentrations for
DEMI. Compared to other NMP sites, DEMI had the second highest annual average
concentration of chloroform among sites sampling VOC. DEMI also had the highest
annual average concentrations of acenaphthene and fluorene among sites sampling
PAH, and the fourth highest annual average concentration of hexavalent chromium.
• Hexavalent chromium and VOC sampling has been conducted at DEMI for at least
five consecutive years; thus, a trends analysis was conducted for benzene,
1,3-butadiene, and hexavalent chromium. A decreasing trend in concentrations is
exhibited for benzene. A decrease in concentrations is also shown for 1,3-butadiene
and hexavalent chromium, but neither decrease is statistically significant. A trends
analysis was not performed for carbonyl compounds because a large number of 2007-
2008 carbonyl samples were invalidated.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for DEMI were greater than their respective ATSDR MRL
noncancer health risk benchmarks.
• Formaldehyde had the highest cancer surrogate risk approximation for DEMI. None
of the pollutants of interest for DEMI had a noncancer risk approximation greater
than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Wayne County,
while coke oven emissions had the highest cancer toxicity-weighted emissions.
Hydrochloric acid was the highest emitted pollutant with a noncancer risk factor in
Wayne County, while acrolein had the highest noncancer toxicity-weighted
emissions.
Missouri.
• The NATTS site in Missouri (S4MO) is located in St. Louis.
• Back trajectories originated from a variety of directions at S4MO, with trajectories
highest percentage of trajectories originating from the northwest.
30-13
-------
• The wind roses for S4MO show that south-southeasterly, southerly, and northerly
winds were the most often most frequently near this site.
• S4MO sampled for VOC, carbonyl compounds, PAH, metals (PMio), and hexavalent
chromium.
• Twenty-four pollutants failed at least one screen for S4MO, of which 14 are NATTS
MQO Core Analytes. S4MO failed the second highest number of screens among all
NMP sites.
• Of the pollutants of interest, acetaldehyde and formaldehyde had the highest annual
average concentrations for S4MO. S4MO had the highest annual average
concentration of acetaldehyde, arsenic, beryllium, cadmium, lead, and manganese
among all NMP sites sampling those pollutants.
• Carbonyl compounds, VOC, metals, and hexavalent chromium sampling have been
conducted at S4MO for at least five consecutive years; thus, a trends analysis was
conducted for acetaldehyde, arsenic, benzene, 1,3-butadiene, formaldehyde,
hexavalent chromium, and manganese. No significant change in concentrations is
shown for acetaldehyde, 1,3-butadiene, and manganese. Arsenic, formaldehyde and
hexavalent chromium have shown a slight decreasing trend, as does benzene,
although there is an increase shown in the most recent time frame.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for S4MO were greater than their respective ATSDR MRL
noncancer health risk benchmarks.
• Formaldehyde had the highest cancer risk approximation for S4MO. None of the
pollutants of interest for S4MO had a noncancer risk approximation greater than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in St. Louis (city),
while toluene was the highest emitted pollutant with a noncancer risk factor.
Hexavalent chromium had the highest cancer toxicity-weighted emissions, while
acrolein had the highest noncancer toxicity-weighted emissions in St. Louis (city).
New Jersey.
• The four UATMP sites in New Jersey are located in Chester (CHNJ), Elizabeth
(ELNJ), New Brunswick (NBNJ), and Paterson (PANJ).
• Due to the close proximity of the New Jersey sites, the composite back trajectories
exhibit similar patterns across the four sites. Back trajectories originated from a
variety of directions, though less frequently from the east and southeast.
• Calm winds were observed for a majority of the wind observations near CHNJ,
NBNJ, and PANJ. Although winds from the north are prevalent near CHNJ and
30-14
-------
NBNJ from a historical standpoint, winds from the northwest to north-northwest were
prevalent during 2010 (for winds greater than 2 knots). Winds from the northwest
quadrant were most common at PANJ. Winds from a variety of directions were
observed near ELNJ, although few easterly and southeasterly wind observations were
observed near this site.
• CHNJ, ELNJ, and NBNJ sampled for VOC and carbonyl compounds, while PANJ
sampled for VOC only.
• Twelve pollutants failed at least one screen For NBNJ and CHNJ; 14 pollutants failed
screens for ELNJ; and 8 failed screens for PANJ.
• Of the pollutants of interest, formaldehyde had the highest annual average
concentrations for CFtNJ, and ELNJ, while acetaldehyde had the highest annual
average concentration for NBNJ. Annual average concentrations could not be
calculated for PANJ due to a combination of a shortened sampling duration (sampling
did not begin until the end of April) and a l-in-12 day sampling schedule.
• Compared to other NMP sites, ELNJ had the highest annual average concentration of
formaldehyde among sites sampling carbonyl compounds.
• Carbonyl compound and VOC sampling has been conducted at CFINJ, ELNJ, and
NBNJ for at least five consecutive years; thus, a trends analysis was conducted for
acetaldehyde, benzene, 1,3-butadiene, and formaldehyde. The rolling average
concentrations of acetaldehyde showed a decreasing trend for all three sites in recent
years. Although significant changes are shown over the years of sampling, in the most
recent years, the rolling average concentrations of benzene and 1,3-butadiene exhibit
little change at CFINJ, ELNJ, and NBNJ. Formaldehyde exhibited little change in
rolling average concentrations in recent years at CFINJ and NBNJ, but exhibited a
decreasing trend at ELNJ.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for the New Jersey sites were greater than their respective
AT SDR MRL noncancer health risk benchmarks.
• Formaldehyde had the highest cancer risk approximations for CFINJ, ELNJ, and
NBNJ. None of the pollutants of interest for any of the New Jersey sites had
noncancer risk approximations greater than 1.0. Cancer and noncancer risk
approximations were not available for PANJ because annual average concentrations
could not be calculated.
• Benzene was the highest emitted pollutant with a cancer URE in Union, Middlesex,
Morris, and Passaic Counties. Benzene had the highest toxicity-weighted emissions
for Morris and Passaic Counties, while formaldehyde had the highest toxicity-
weighted emissions for Union and Middlesex Counties.
30-15
-------
• Toluene was the highest emitted pollutant with a noncancer risk factor in all four
counties, while acrolein had the highest noncancer toxicity-weighted emissions for
each county.
New York.
• Two New York monitoring sites are located in the Bronx Borough of New York City
(BXNY and MONY). A third site is located in Rochester (ROCH) and the fourth
monitoring site is located north of Buffalo in Tonawanda (TONY). The BXNY
NATTS site was relocated to the MONY location in mid-2010. The ROCH site is
also a NATTS monitoring site.
• Back trajectories originated from a variety of directions at each of the New York
sites, though less frequently from the east, particularly for ROCH and TONY.
• Winds from the northwest quadrant were observed most frequently near BXNY,
while winds from the west and west-northwest were most frequently observed near
MONY. South-southwesterly to westerly winds were most common near ROCH and
TONY.
• All three New York sites sampled PAH. BXNY, MONY, and ROCH also sampled
hexavalent chromium.
• Fourteen pollutants failed screens for BXNY and MONY. Only naphthalene failed
screens for ROCH. Six PAH failed screens for TONY.
• Due to abbreviated sampling durations, annual average concentrations could not be
calculated for the pollutants of interest for BXNY and MONY. Sampling was
discontinued at TONY in mid-2010. Additionally, sampler problems at ROCH led to
the invalidation of the majority of PAH samples for 2010. Therefore, annual average
concentrations could only be calculated for hexavalent chromium for ROCH.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for the New York sites, where they could be calculated were
greater than their respective ATSDR MRL noncancer health risk benchmarks.
• The cancer surrogate risk approximation for hexavalent chromium (for ROCH) was
well below 1 in-a-million; similarly, the noncancer surrogate risk approximation for
hexavalent chromium (for ROCH) was well below 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor for all three New
York counties. It also had the highest cancer toxicity-weighted emissions for Bronx
County and Monroe County. Coke oven emissions (PM) had the highest cancer
toxicity-weighted emissions for Erie County.
30-16
-------
• Methanol was the highest emitted pollutant with a noncancer risk factor in Bronx
County, while toluene was the highest emitted pollutant with a noncancer risk factor
in Monroe and Erie Counties. Acrolein had the highest noncancer toxicity-weighted
emissions for all three counties.
Oklahoma.
• There are five Oklahoma UATMP monitoring sites: two located in Tulsa (TOOK and
TMOK), one in Pryor Creek (PROK), and in Oklahoma City (OCOK) and one in the
Oklahoma City suburb of Midwest City (MWOK).
• The back trajectory maps for the Tulsa, Pryor Creek and Oklahoma City sites are
similar in trajectory distribution, with a strong tendency for back trajectories to
originate from the south and the northwest to north of the sites.
• The wind roses show that southerly winds prevailed near each monitoring site,
accounting for one-fifth to one-quarter of the observations at each site..
• Each of Oklahoma sites sampled for VOC, carbonyls compounds, and metals (TSP).
• Seventeen pollutants failed screens for TOOK, 16 failed screens for TMOK; 13 failed
screens for PROK; 18 failed screens for MWOK; and 19 failed screens for OCOK.
• Of the pollutants of interest, formaldehyde had the highest annual average
concentrations for each Oklahoma site.
• TOOK had the highest annual average for benzene among NMP sites sampling this
pollutant. PROK and MWOK had the highest and second highest annual average
concentrations of 1,2-dichloroethane, respectively, among NMP sites sampling this
pollutant.
• TOOK has sampled carbonyl and VOC compounds for at least five years, therefore a
trends analysis was conducted for acetaldehyde, benzene, 1,3-butadiene, and
formaldehyde. Any changes in the rolling average concentrations of these pollutants
were not statistically significant.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for the Oklahoma sites were greater than their respective
AT SDR MRL noncancer health risk benchmarks.
• Formaldehyde and benzene had the highest cancer risk approximations for all of the
Oklahoma monitoring sites. Arsenic had the highest cancer risk approximations
among the metals. None of the pollutants of interest for the Oklahoma sites had a
noncancer risk approximation greater than 1.0.
30-17
-------
• Benzene was the highest emitted pollutant with a cancer risk factor in Mayes,
Oklahoma, and Tulsa Counties. Arsenic had the highest cancer toxicity-weighted
emissions for Mayes County and benzene had the highest cancer toxicity-weighted
emissions for Oklahoma and Tulsa Counties.
• Toluene was the highest emitted pollutant with a noncancer risk factor in Tulsa and
Oklahoma Counties, while hydrochloric acid was the highest emitted pollutant with a
noncancer risk factor in Mayes County. Acrolein had the highest noncancer toxicity-
weighted emissions for all three counties.
Rhode Island.
• The Rhode Island monitoring site (PRRI) is located in Providence and is a NATTS
site.
• Back trajectories originated from a variety of directions at PRRI, although more
frequently from the west, northwest, and north.
• The wind roses show that winds from the north, south, or with a westerly component
were prevalent near PRRI.
• PRRI sampled for PAH and hexavalent chromium.
• Naphthalene, benzo(a)pyrene, flourene, and hexavalent chromium failed screens for
PRRI, although 86 percent of failed screens are attributed to naphthalene.
• The annual average concentration of naphthalene was significantly higher than that of
the other pollutants of interest. The annual average benzo(a)pyrene concentration for
PRRI is the highest among all sites sampling the pollutant.
• Hexavalent chromium sampling has been conducted at PRRI for at least five
consecutive years; thus, a trends analysis was conducted. The rolling average
concentrations of hexavalent chromium have fluctuated over the period of sampling,
though confidence intervals indicate that any changes are not statistically significant.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for PRRI were greater than their respective ATSDR MRL
noncancer health risk benchmarks.
• Naphthalene had the highest cancer risk approximation for PRRI, and the only one
greater than 1.0 in-a-million; all noncancer risk approximations for PRRI were less
than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Providence
County, while formaldehyde had the highest cancer toxicity-weighted emissions.
30-18
-------
Toluene was the highest emitted pollutant with a noncancer risk factor, while acrolein
had the highest noncancer toxicity-weighted emissions for Providence County.
South Carolina.
• The South Carolina monitoring site (CHSC) is located near Chesterfield and is a
NATTS site.
• Back trajectories originated from a variety of directions at CHSC.
• The wind roses show that calm winds, south-south westerly to west-southwesterly
winds were prevalent near CHSC.
• CHSC sampled for hexavalent chromium and PAH.
• Naphthalene was the only pollutant to fail screens for CHSC. Naphthalene failed six
screens out of 58 measured detections).
• The annual average concentration of naphthalene was significantly higher than the
annual average concentrations of the other two pollutants of interest. Compared to
other program sites sampling PAH and hexavalent chromium, CHSC had some of the
lowest daily average concentrations.
• Hexavalent chromium sampling has been conducted at CHSC for at least five
consecutive years; thus, a trends analysis was conducted. Initial decreases in
hexavalent chromium concentrations were followed by a slight increase over the most
recent three-year period.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for CHSC were greater than their respective ATSDR MRL
noncancer health risk benchmarks.
• The cancer surrogate risk approximations for the pollutants of interest were below 1
in-a-million; the noncancer surrogate risk approximations were well below 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Chesterfield
County and had the highest cancer toxicity-weighted emissions. Toluene was the
highest emitted pollutant with a noncancer risk factor, while acrolein had the highest
noncancer toxicity-weighted emissions.
South Dakota.
• The UATMP sites in South Dakota are located in Sioux Falls (SSSD) and Union
County (UCSD).
• Back trajectories originated from a variety of directions at the South Dakota sites,
although primarily from the west to northwest to north.
30-19
-------
• Winds from a variety of directions were observed near SSSD, although southerly
winds were the most common wind direction. Winds from the southeast and
northwest quadrants were the most frequently observed wind directions near UCSD.
• Both South Dakota sites sampled for VOC, SNMOC, and carbonyl compounds.
• Thirteen pollutants failed screens for SSSD, of which five are NATTS MQO Core
Analytes. Fifteen pollutants failed screens for UCSD, of which six are also NATTS
MQO Core Analytes.
• Formaldehyde had the highest annual average concentration for both UCSD and
SSSD, with the annual average concentrations of acetaldehyde being second highest.
UCSD had the highest concentration of trichloroethylene among all NMP sites
sampling VOC and the fourth highest concentrations of acrylonitrile and
ethylbenzene.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for the South Dakota sites were greater than their respective
AT SDR MRL noncancer health risk benchmarks.
• Formaldehyde had the highest cancer risk approximations for both sites. None of the
pollutants of interest for any of the South Dakota sites had a noncancer risk
approximation greater than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Minnehaha and
Union Counties, while formaldehyde had the highest toxicity-weighted emissions for
both counties.
• Toluene was the highest emitted pollutant with a noncancer risk factor in Minnehaha
and Union Counties, while acrolein had the highest noncancer toxicity-weighted
emissions for all three counties.
Texas.
• There are two NATTS sites in Texas: one in Deer Park (CAMS 35) and one in
Karnack (CAMS 85).
• Back trajectories originated from a variety of directions at CAMS 35, although most
trajectories originated from the east to southeast to south over the Gulf of Mexico.
Back trajectories also originated from a variety of directions at CAMS 85, although
there is a north-south distribution pattern of trajectories similar to CAMS 35.
• The wind roses show that winds from the southeast quadrant (including easterly and
southerly winds) were the most commonly observed wind directions near CAMS 35.
A similar wind pattern is exhibited near CAMS 85.
30-20
-------
• The CAMS 35 site sampled for PAH. Both sites began sampling for hexavalent
chromium in February 2010.
• Four pollutants failed screens for CAMS 35, with naphthalene, contributing to nearly
97 percent of the total failed screens. Hexavalent chromium failed nearly 65 percent
of screens for CAMS 85.
• Of the pollutants of interest, naphthalene had the highest annual average
concentration for CAMS 35, and is significantly higher than the annual averages for
the other pollutants of interest. The annual average concentration of hexavalent
chromium for CAMS 85 was six times higher than the annual average concentration
for CAMS 35 and the highest among NMP sites sampling hexavalent chromium.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for the Texas sites were greater than their respective ATSDR
MRL noncancer health risk benchmarks.
• Naphthalene had the highest cancer risk approximation among the pollutants of
interest for CAMS 35. The cancer risk approximation for hexavalent chromium for
CAMS 85 is one of only two hexavalent chromium risk approximations greater than 1
in-a-million. None of the pollutants of interest for either Texas site had a noncancer
risk approximation greater than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Harris County,
while formaldehyde had the highest cancer toxicity-weighted emissions. In Harrison
County, formaldehyde was the highest emitted pollutant with a cancer risk factor,
while hexavalent chromium had the highest cancer toxicity-weighted emissions.
• Toluene was the highest emitted pollutant with a noncancer risk factor in both
counties, while acrolein had the highest noncancer toxicity-weighted emissions.
Utah.
• The NATTS site in Utah is located in Bountiful (BTUT), north of Salt Lake City.
• Back trajectories originated from a variety of directions at BTUT. Back trajectories
originating from the northeast, east, and southeast tended to be shorter in length than
trajectories from other directions.
• The wind roses show that southeasterly, south-southeasterly, and southerly winds
were prevalent near BTUT.
• BTUT sampled for VOC, carbonyl compounds, SNMOC, PAH, metals (PMW\ and
hexavalent chromium.
30-21
-------
• Twenty-six pollutants failed screens for BTUT, of which 14 are NATTS MQO Core
Analytes.
• Of the pollutants of interest, dichloromethane had the highest annual average
concentration for BTUT, followed by formaldehyde, acetaldehyde and benzene.
BTUT had the highest annual average concentration of /?-dichlorobenzene, and the
second highest annual average concentrations of trichloroethylene and formaldehyde
among NMP sites sampling these pollutants. Among sites sampling PMio metals,
BTUT had the third highest annual average concentration of arsenic.
• Carbonyl compounds, VOC, SNMOC, metals (PMio), and hexavalent chromium
sampling have been conducted at BTUT for at least five consecutive years; thus, a
trends analysis was conducted for acetaldehyde, arsenic, benzene, 1,3-butadiene,
formaldehyde, hexavalent chromium, and manganese. After initial decreases,
concentrations of acetaldehyde, benzene, formaldehyde, and hexavalent chromium
have remained steady over recent years. Concentrations of 1,3-butadiene have also
changed little. Concentrations of manganese have started to decrease over the more
recent time frames.
• Of 59 measured detections of dichloromethane, one was greater than the ATSDR
acute MRL. This was the only pollutant with a measured detections or time-period
average greater than an MRL noncancer health risk benchmark for BTUT.
• The pollutant with the highest cancer surrogate risk approximation for BTUT was
formaldehyde. None of the pollutants of interest had a noncancer risk approximations
greater than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Davis County
and had the highest cancer toxicity-weighted emissions. Toluene was the highest
emitted pollutant with a noncancer risk factor, while acrolein had the highest
noncancer toxicity-weighted emissions for Davis County.
Vermont.
• Two Vermont monitoring sites are located in or near Burlington (UNVT and
BURVT); a third monitoring site is located in Rutland (RUVT). UNVT is a NATTS
monitoring site.
• The back trajectory maps for the Vermont sites are similar to each other, with most
back trajectories originating from the southwest, west, northwest, and north of the
sites.
• The wind roses for the Vermont sites show that southerly winds were prevalent near
BURVT, although northwesterly to northerly winds were also common; east-
southeasterly and southeasterly winds were prevalent near RUVT, although winds
30-22
-------
from the northwest quadrant were also common; and calm winds were prevalent near
UNVT.
• UNVT sampled for VOC, carbonyl compounds, hexavalent chromium, PAH, and
metals (PMio). BURVT and RUVT sampled for VOC only.
• Eight pollutants failed screens for BURVT and six failed screens for RUVT. Thirteen
pollutants failed screens for UNVT, of which nine are NATTS MQO Core Analytes.
• Benzene had the highest annual average concentration for BURVT and RUVT, while
carbon tetrachloride had the highest annual average concentration for UNVT.
Because carbonyl compound sampling ended in June 2010 at UNVT, annual average
concentrations could not be calculated for these compounds. Annual average
concentrations of the pollutants of interest for UNVT were among the lowest
compared to NMP sites sampling the same pollutants.
• UNVT has sampled hexavalent chromium for at least five consecutive years; thus, a
trends analysis was conducted for hexavalent chromium. A decreasing trend is shown
for hexavalent chromium over the period of sampling. At least 50 percent of the
measurements were non-detects for each 3-year period.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for the Vermont sites were higher than their respective ATSDR
MRL noncancer health risk benchmarks.
• Benzene had the highest cancer risk approximation for BURVT and RUVT. Carbon
tetrachloride had the highest cancer risk approximation for UNVT, followed by the
cancer risk approximation for benzene. None of the noncancer risk approximations,
where they could be calculated, were greater than an HQ of 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Chittenden and
Rutland Counties and also had the highest cancer toxicity-weighted emissions.
Toluene was the highest emitted pollutant with a noncancer risk factor in both
counties, while acrolein had the highest noncancer toxicity-weighted emissions.
Virginia.
• The NATTS site in Virginia is located near Richmond (RIVA).
• Back trajectories originated from a variety of directions at RIVA, although primarily
from the southwest, west, and northwest of the site.
• The wind rose shows that northerly winds were observed the most and winds from the
southeast quadrant were observed the least.
• RIVA sampled for PAH and hexavalent chromium.
30-23
-------
• Four PAH pollutants failed screens for RIVA, though naphthalene contributed to
nearly 97 percent of the total failed screens.
• Of the pollutants of interest, naphthalene had the highest annual average
concentrations.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for the Virginia site were greater than their respective ATSDR
MRL noncancer health risk benchmarks.
• Naphthalene had the highest cancer risk approximation for RIVA. None of the
pollutants of interest for RIVA had a noncancer risk approximation greater than 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in Henrico
County, while formaldehyde had the highest cancer toxicity-weighted emissions.
Toluene was the highest emitted pollutant with a noncancer risk factor in Henrico
County, while acrolein had the highest noncancer toxicity-weighted emissions.
Washington.
• The NATTS site in Washington is located in Seattle (SEWA).
• Back trajectories for SEWA originated from a variety of directions. Back trajectories
originating to the south and northwest tended to be longer than trajectories originated
to the north and east.
• The wind roses show that southeasterly to southerly winds were prevalent near the
SEWA.
• SEWA sampled for VOC, carbonyl compounds, PAH, PMio metals, and hexavalent
chromium.
• Sixteen pollutants failed screens for SEWA, of which 10 are NATTS MQO Core
Analytes.
• Of the pollutants of interest for SEWA, acetaldehyde had the highest annual average
concentration. SEWA had the highest annual average concentration for carbon
tetrachloride among NMP sites sampling VOC and the highest annual average
concentration of nickel among NMO sites sampling metals.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for SEWA were greater than their respective ATSDR MRL
noncancer health risk benchmarks.
30-24
-------
• Formaldehyde had the highest cancer surrogate risk approximation for SEWA. All of
the noncancer risk approximations for the pollutants of interest for SEWA sites were
less than an HQ of 1.0.
• Benzene was the highest emitted pollutant with a cancer risk factor in King County
and had the highest cancer toxicity-weighted emissions. Toluene was the highest
emitted pollutant with a noncancer risk factor in King County, while acrolein had the
highest noncancer toxicity-weighted emissions.
Wisconsin.
• The Wisconsin monitoring site is located in Horicon (HOWI) and is a NATTS site.
Because the site started in December 2009, the two samples from that month are also
included in this report.
• Back trajectories originated from a variety of directions at HOWI, although less
frequently from the east.
• The wind roses show that winds from a variety of directions are observed at HOWI,
though winds from the south and north were observed most frequently.
• HOWI sampled for PAH and hexavalent chromium although PAH sampling ended in
June 2010.
• Naphthalene was the only pollutant to fail screens for HOWI.
• Because PAH sampling ended in June 2010, annual average concentrations could not
be calculated for the PAH pollutants.
• None of the measured detections or time-period average concentrations of the
pollutants of interest for the Wisconsin site were greater than their respective ATSDR
MRL noncancer health risk benchmarks.
• Because annual average concentrations are not available, cancer and noncancer risk
approximations could not be calculated for naphthalene and benzo(a)pyrene. Both the
cancer and noncancer risk approximations for hexavalent chromium were low.
• Benzene was the highest emitted pollutant with a cancer risk factor in Dodge County
and had the highest cancer toxicity-weighted emissions. Toluene was the highest
emitted pollutant with a noncancer risk factor in Dodge County, while acrolein had
the highest noncancer toxicity-weighted emissions.
30.1.3 Composite Site-level Summary
• Thirty-one pollutants were identified as site-specific pollutants of interest, based on
the risk screening process. Acetaldehyde and formaldehyde were the two most
30-25
-------
common pollutants of interest among the monitoring sites. All sites (30) that sampled
carbonyl compounds had acetaldehyde and formaldehyde as pollutants of interest.
Benzene, 1,3-butadiene, and carbon tetrachloride were the most common VOC
pollutants of interest. Every site that sampled benzene (29) had this as a pollutant of
interest. Every site that sampled PAH (26) had naphthalene as a pollutant of interest.
• Among the site-specific pollutants of interest, formaldehyde frequently had the
highest annual average concentration among the monitoring sites; formaldehyde had
the highest annual average concentration for 17 sites. Naphthalene had the next
highest at 11.
• Benzo(a)pyrene was detected most frequently during the first and fourth quarters of
the calendar year. There were 269 measured detections during the first quarter of
2010, 129 during the second quarter, 87 during the third quarter, and 200 during the
fourth quarter of 2010.
• Of the 214 measured detections of 1,2-dichloroethane, 117 were measured during the
first quarter of 2010, 87 during the second quarter, 3 during the third quarter, and 5
during the fourth quarter of 2010. Virtually all (203 out of 214) were measured prior
to May 14,2010.
• The toxicity factor for formaldehyde used in the preliminary risk screening process,
the cancer risk approximation calculations, and the toxicity-weighting of emissions
decreased substantially since the 2007 report. This translated to a much higher
toxicity potential for formaldehyde, leading to more failed screens, higher cancer risk
approximations, and relatively higher toxicity-weighted emissions values for the
2008-2009 and the 2010 report than in previous reports.
• Formaldehyde and naphthalene tended to have the highest cancer risk
approximations. This is also true for the noncancer risk approximations, although
there were no noncancer risk approximations greater than an HQ of 1.0 among any of
the site-specific pollutants of interest.
• Carbon tetrachloride often had relatively high cancer risk approximations based on
annual averages among the monitoring sites, but tended to have relatively low
emissions and toxicity-weighted emissions according to the NEI. This suggests that
this pollutant is present in "background" levels of ambient air; that is, it is
consistently present at similar levels at any given location. Although production of
this pollutant has declined sharply over the last 30 years due to its role as an ozone
depleting substance, it has a relatively long atmospheric lifetime.
• Benzene, formaldehyde, and ethylbenzene tended to have the highest county-level
emissions of the pollutants with cancer risk factors. Ethylbenzene did not appear as
frequently among the highest emitted pollutants in the 2008-2009 report (based on the
2005 NEI) as it does in the 2010 report (based on the 2008 NEI). Formaldehyde,
30-26
-------
benzene, and 1,3-butadiene tended to have the highest toxicity-weighted emissions
among the pollutants with cancer risk factors.
• While toluene, xylenes, and methanol tended to have the highest county-level
emissions of the pollutants with noncancer risk factors, acrolein, formaldehyde, and
1,3-butadiene tended to have the highest toxicity-weighted emissions among the
pollutants with noncancer risk factors
• Acrolein emissions were relatively low when compared to other pollutants. However,
due to the high toxicity of this pollutant, even low emissions translated into high
noncancer toxicity-weighted emissions; the toxicity-weighted value was often several
orders of magnitude higher than other pollutants. Acrolein is a national noncancer
risk driver according to NATA.
30.1.4 Data Quality Summary
Method precision and analytical precision was determined for the 2010 NMP monitoring
efforts using CV calculations based on duplicate, collocated, and replicate samples. The overall
method precision for most methods was well within data quality objective specifications and
monitoring method guidelines, while one method exceeded the data quality objective
specifications (SNMOC). The method precision presented in this report is based on analytical
results greater than or equal to the sample- and pollutant-specific MDL, which represents a
change from previous reports (where/^MDL substitutions were performed).
Sampling and analytical method accuracy is ensured by using proven methods, as
demonstrated by third-party analysis of proficiency test audit samples, and following strict
quality control and quality assurance guidelines.
Ambient air concentration data sets generally met data quality objectives for
completeness. Completeness, or the number of valid samples collected compared to the number
expected from a l-in-6 or l-in-12 day sampling schedule, measures the reliability of the
sampling and analytical equipment as well as the efficiency of the program. Typically, a
completeness of 85-100 percent is desired for a complete data set. Only five out of 126 data sets
failed to comply with the data quality objective of 85 percent completeness. Thirty-eight data
sets achieved 100 percent completeness.
30-27
-------
NATTS data generated under the NMP for 2010 were included in EPA's NATTS
Network Assessment (EPA 2012J), which assessed the NATTS program trends DQO and
determined if the data are suitable for trends analysis. The DQO is satisfied if the data generated
for the NATTS program meets the MQOs for completeness, sensitivity, bias, and precision.
According to the NATTS Network Assessment, the NATTS data generated under the NMP for
2010 met the MQOs and were determined to be suitable for trends analysis, with the exception of
two data sets where sampling issues in the field resulted in the invalidation of a large subset of
data.
30.2 Conclusions
Conclusions resulting from the data analyses of the data generated from the 2010 NMP
monitoring efforts are presented below.
• There are a large number of concentrations that are greater than their respective
preliminary risk screening values, particularly for many of the NATTS MQO Core
Analytes. However, there are few instances where the preprocessed daily
measurements or time-period average concentrations were greater than the ATSDR
MRL noncancer health risk benchmarks.
• Where annual averages could be calculated and for those pollutants with available
cancer UREs, none of the cancer surrogate risk approximations were greater than
100-in-a-million; 34 were greater than 10-in-a-million (23 for formaldehyde, one for
dichloromethane; eight for benzene, and two for acrylonitrile and naphthalene); and
approximately half were greater than 1.0 in-a-million.
• Where annual averages could be calculated and for those pollutants with available
noncancer RfCs, none of the noncancer surrogate risk approximations was greater
than an HQ of 1.0.
• When comparing the highest emitted pollutants for a specific county to the pollutants
with the highest toxicity-weighted emissions, the listed pollutants were more similar
for the pollutants with cancer UREs than for pollutants with noncancer RfCs. This
indicates that pollutants with cancer UREs that are emitted in higher quantities are
often more toxic than pollutants emitted in lower quantities; conversely, the highest
emitted pollutants with noncancer RfCs are not necessarily the most toxic. For
example, toluene is the noncancer pollutant that was emitted in the highest quantities
for many NMP counties, yet was rarely one of the pollutants with highest toxicity-
weighted emissions. Further, while acrolein had the highest noncancer toxicity-
weighted emissions for every NMP county, it was rarely among the highest emitted
pollutants.
30-28
-------
• The number of states and sites participating in the NMP changes from year to year.
Yet, many of the data analyses utilized here require data from year-round (or nearly
year-round) sampling. Of the 52 sites whose data are included in the 2010 report,
seven sites sampled for an abbreviated duration (due to site initialization and/or site
closure/relocation). Of the 126 site-method combinations, 16 site-method
combinations did not cover the entire year As a result, time-period averages and
subsequent risk-based analyses could not be calculated for nearly 13 percent of
participating sites (and site-method combinations), although this is an improvement
from the 2008-2009 monitoring effort. While these gaps have ramifications for the
results contained in this report, they also inhibit the potential determination of trends.
• Of the 52 monitoring sites participating in the 2010 NMP, only two sampled for all
six available analytical methods under the national program (BTUT and NBIL).
Another four sites sampled all five methods required for NATTS sites through the
national program. The wide range of methods/pollutants sampled among the
sites makes it difficult to draw definitive conclusions regarding air toxics in ambient
air in a global manner.
• This report strives to utilize the best laboratory and data analysis techniques available
(which includes the improvement of MDLs and the incorporation of updated values
for various risk factors, for example). This often leads to adjusting the focus of the
report to concentrate on the air quality issues of highest concern. Thus, the NMP
report is dynamic in nature and scope; yet this approach may prevent the direct
comparison of the current report to past reports. There are two major differences
between the 2008-2009 NMP report and the 2010 report. First, all statistical
calculations include zero substitution for non-detect results (rather than just those
calculations related to risk). Second, the detect criteria applied to quarterly averages
was removed for the 2010 report, allowing for the calculation of quarterly average
concentrations for those pollutants detected less frequently than others.
30.3 Recommendations
Based on the conclusions from the 2010 NMP, a number of recommendations for future
ambient air monitoring efforts are presented below.
• Continue participation in the National Monitoring Programs. Ongoing ambient air
monitoring at fixed locations can provide insight into long-term trends in air quality
and the potential for air pollution to cause adverse health effects among the general
population. Therefore, state and local agencies should be encouraged to either 1)
develop and implement their own ambient air monitoring programs based on proven,
consistent sampling and analysis methods and EPA technical and quality assurance
guidance, or 2) consider participation in the NMP.
• Participate in the National Monitoring Programs year-round. Many of the analyses
presented in the 2010 report require a full year of data to be most useful and
representative of conditions experienced at each specified location. Therefore, state
30-29
-------
and local agencies should be encouraged to implement year-long ambient air
monitoring programs in addition to participating in future monitoring efforts.
• Monitor for additional pollutant groups based on the results of data analyses in the
annual report. The risk-based analysis where county-level emissions are weighted
based on toxicity identifies those pollutants whose emissions may result in adverse
health effects in a specific area. If a site is not sampling for a pollutant or pollutant
group identified as particularly hazardous in a given area, the agency responsible for
that site should consider sampling for those compounds.
• Strive to develop standard conventions for interpreting air monitoring data. The lack
of consistent approaches to present and summarize ambient air monitoring data
complicates direct comparisons between different studies. Thought should be given to
the feasibility of establishing standard approaches for analyzing and reporting air
monitoring data for programs with similar objectives.
• Continue to identify and implement improvements to the sampling and analytical
methods. The improvements made to the analytical methods prior to the 1999-2000
UATMP allowed for the measurement of ambient air concentrations of 11 pollutants
that were not measured during previous programs. This improvement provides
sponsoring agencies and a variety of interested parties with important information
about air quality within their area. Further research is encouraged to identify other
method improvements that would allow for the characterization of an even wider
range of components in air pollution and enhance the ability of the methods to
quantify all cancer and noncancer pollutants to at least their levels of concern (risk
screening concentrations).
• Require consistency in sampling and analytical methods. The development of the
NATTS program has shown that there are inconsistencies in collection and analytical
methods that make data comparison difficult across agencies. Requiring agencies to
use specified and accepted measurement methods is integral to the identification of
trends and the impacts of regulation.
• Perform case studies based on findings from the annual report. Often, the annual
report identifies an interesting tendency or trend, or highlights an event at a particular
site(s). For example, the 2006 annual report included observations of high hexavalent
chromium concentrations on July 4, 2006. Further examination of the data in
conjunction with meteorological phenomena and potential emissions events or
incidents, or further site characterization may help identify state and local agencies
pinpoint issues affecting air quality in their area.
• Consider more rigorous study of the impact of automobile emissions on ambient air
quality using multiple years of data. Because many NMP sites have generated years
of continuous data, a real opportunity exists to evaluate the importance and impact of
automobile emissions on ambient air quality. Suggested areas of study include
additional signature compound assessments and parking lot characterizations.
30-30
-------
• Develop and/or verify HAP and VOC emissions inventories. State/local/tribal
agencies should use the data collected from the NMP sites to develop and validate
emissions inventories, or at the very least, identify and/or verify emissions sources of
concern. Ideally, state/local/tribal agencies would compare the ambient monitoring
results with an emissions inventory for source category completeness. The emissions
inventory could then be used to develop modeled concentrations useful to compare
against ambient monitoring data.
• Promulgate ambient air standards for HAPs. Several of the pollutants sampled during
the 2010 program years were higher than risk screening values developed by various
government agencies. One way to reduce the risk to human health would be to
develop standards similar to the NAAQS for pollutants that frequently exceed
published risk screening levels.
• Incorporate/Update Risk in State Implementation Plans (SIPs). Use risk calculations
to design State Implementation Plans to implement policies that reduce the potential
for human health risk. This would be easier to enforce if ambient standards for certain
HAPs were developed (refer to above recommendation).
30-31
-------
31.0 References
ACE, 2012. U.S. Army Corps of Engineers. Huntington District, Grayson Lake website.
http://www.lrh.usace.army.mil/Missions/Recreation/Kentucky/GraysonLake.aspxDate
last accessed: 8/10/2012.
ASTM, 2004. ASTM, International. 2004. ASTM D6209-98(2004) Standard Test Method for
Determination of Gaseous and Particulate Polycyclic Aromatic Hydrocarbons in Ambient
Air (Collection on Sorbent-Backed Filters with Gas Chromatographic/Mass
Spectrometric Analysis), http://www.astm.org/Standards/D6209.htm
ATSDR, 2010. Agency for Toxic Substances and Disease Registry. December 2010. Minimal
Risk Levels (MRLs). http://www.atsdr.cdc.gov/mrls/pdfs/atsdr mrls december 2010.pdf
ATSDR, 2012. Agency for Toxic Substances and Disease Registry. Toxic Substance Portal
website, Minimal Risk Levels (MRLs). http://www.atsdr.cdc.gov/mrls/index.asp Date
last accessed: 1/2/2012.
AZ DOT, 2009. Arizona Department of Transportation. State Highway Traffic Log, AADT
2009. http://mpd.azdot.gov/mpd/data/aadt.asp
AZ DOT, 2010. Arizona Department of Transportation, Motor Vehicle Division. June 30, 2010.
Currently Registered Vehicles by County Listing.
http ://www. azdot. gov/mvd/statistics/registeredVehicles. asp
AZ DOT, 2011. Arizona Department of Transportation, Motor Vehicle Division. August 2011.
Daily Vehicle-Miles of Travel (DVMT) by County with Population Estimates for the
Year ended December 31, 2010. http://www.azdot.gov/mpd/data/hpms.asp Date last
accessed: 3/26/2012.
Bair, 1992. Bair, Frank E. 1992. Weather of U.S. Cities, 4th ed. Gale Research Inc. Detroit, MI.
Boubel, et al., 1994. Boubel, R.W., D.L. Fox, D.B. Turner, and A.C. Stern. 1994. Fundamentals
of Air Pollution, 3rd ed. San Diego, CA: Academic Press, pp 263.
CA DMV, 2010. California Department of Motor Vehicles. 2010. Estimated Vehicles Registered
by County for the Period of January 1 through December 31, 2010.
http://www.dmv.ca.gov/about/profile/est fees_pd by county.pdf
CA DOT, 2010. State of California Department of Transportation. 2010 Traffic Volumes on the
California State Highway System. Sacramento, CA.
http: //www. dot. ca. gov/hq/traffop s/safere sr/trafdata/index. htm
CA DOT, 2011. State of California Department of Transportation. October 2011. 2010
California Public Road Data, http://www.dot.ca.gov/hq/tsip/hpms/index.php
31-1
-------
CCRPC, 2005. Chittenden County Regional Planning Commission. Transportation Data-Annual
Average Daily Traffic website. 2005 Data, http://www.ccmpo.us/data/traffic data.php
Date last accessed: 12/22/2011.
CCRPC, 2010. Chittenden County Regional Planning Commission. Transportation Data-Annual
Average Daily Traffic website. 2010 Data, http://www.ccmpo.us/data/traffic data.php
Date last accessed: 12/22/2011.
Census Bureau, 2009. U.S. Census Bureau. December 2009. Metropolitan and Micropolitan
Statistical Areas and Components, December 2009, With Codes.
http://www.census.gov/population/metro/files/lists/2009/Listl.txt Date last accessed:
12/22/2011.
Census Bureau, 2011. U.S. Census Bureau. March 2011. 2010 Census - Census Tract Reference
Maps website. http://www.census.gov/geo/www/maps/pllO map suite/tract.html Date
last accessed: 10/14/2011.
Census Bureau, 2011. U.S. Census Bureau. Intercensal Estimates of the Resident Population for
Counties: April 1, 2000 to July 1, 2010. Released September 2011.
http://www.census.gov/popest/data/intercensal/county/CO-ESTOOINT-01.html Date last
accessed: 3/26/2012.
Census Bureau, 2012. U.S. Census Bureau. Metropolitan and Micropolitan Statistical Areas
website, http://www.census.gov/population/metro/ Date last accessed: 1/9/2012.
CO DOR, 2010. Colorado Department of Revenue. December 22, 2010. Colorado Department of
Revenue 2010 Annual Report. Denver, CO.
http://www.colorado.gov/cs/Satellite? c=Page&cid=1213867975035&pagename=Revenu
e-Main%2FXRMLavout
CO DOT, 2010. Colorado Department of Transportation. 2010 State Highway Statistics - Daily
Vehicle Miles of Travel (DVMT) for All Vehicles by County. 2010.
http://apps.coloradodot.info/dataaccess/Statistics/index.cfm?fuseaction=StatisticsMain&
MenuType=Statistics Date last accessed: 3/27/2012.
CO DOT, 2011. Colorado Department of Transportation. May 2011. Colorado Highway System
Traffic Volume Map 2010 (Grand Junction, Garfield County).
http://apps.coloradodot.info/dataaccess/Traffi c/index.cfm?fuseaction=TrafficMain&Men
uType=Traffic
DC DOT, 2011. District Department of Transportation. February 2011. 2009 Traffic Volumes.
Washington, D.C.
http ://ddot. dc. gov/DC/DDOT/About+DDOT/Maps/Traffic+Volume+Maps
31-2
-------
Draxler, R.R. and Rolph, G.D., 1997. Draxler, R.R. and Rolph, G.D. 1997. Description of the
HYSPLIT_4 modeling system. NOAA Air Resources Laboratory. NOAA Tech. Memo.
ERL ARL-224. Silver Spring, MD. 24 pp. http://www.arl.noaa.gov/HYSPLIT pubs.php
Draxler, R.R. and Rolph, G.D., 1998. Draxler, R.R. and Rolph, G.D. 1998. An overview of the
HYSPLIT_4 modeling system of trajectories, dispersion, and deposition. Aust. Meteor.
Mag. 47, 295-308. http://www.arl.noaa.gov/HYSPLIT pubs.php
Draxler, R.R., 1999. Draxler, R.R. 1999. HYSPLIT4 user's guide. NOAA Air Resources
Laboratory. NOAA Tech. Memo. ERL ARL-230. Silver Spring, MD.
http://www.arl.noaa.gov/HYSPLIT_pubs.php
Draxler, R.R., et al, 2009. Draxler, R.R., et al. 2009. HYSPLIT 4 User's Guide, Version 4.9.
NOAA Air Resources Laboratory. Silver Spring, MD.
http://www.arl.noaa.gov/documents/reports/hysplit user guide.pdf
EPA, 1998. U.S. EPA. September 1998. Technical Assistance Document for Sampling and
Analysis of Ozone Precursors. EPA/600-R-98/161. Research Triangle Park, NC.
http://www.epa.gov/ttnamtil/files/ambient/pams/newtad.pdfDate last accessed:
1/10/2012.
EPA, 1999a. U.S. EPA. January 1999. Compendium Method TO-15: Determination of Volatile
Organic Compounds (VOC) in Air Collected in Specially-Prepared Canisters and
Analyzed by Gas Chromatography/Mass Spectrometry (GC/MS). EPA/625/R-96/010b.
Cincinnati, OH. http://www.epa.gov/ttn/amtic/airtox.html Date last accessed: 1/10/2012.
EPA, 1999b. U.S. EPA. January 1999. Compendium Method TO-11A: Determination of
Formaldehyde in Ambient Air Using Adsorbent Cartridge Followed by High
Performance Liquid Chromatography (HPLC). EPA/625/R-96/010b. Cincinnati, OH.
http://www.epa.gov/ttn/amtic/airtox.html Date last accessed: 1/10/2012.
EPA, 1999c. U.S. EPA. January 1999. Compendium Method TO-13A: Determination of
Polycyclic Aromatic Hydrocarbons (PAHs) in Ambient Air Using Gas
Chromatography/Mass Spectrometry (GC/MS). EPA/625/R-96/010b. Cincinnati, OH.
http://www.epa.gov/ttn/amtic/airtox.html Date last accessed: 1/10/2012.
EPA, 1999d. U.S. EPA. June 1999. Compendium Method IO-3.5: Determination of Metals in
Ambient Particulate Matter Using Inductively Coupled Plasma/Mass Spectrometry
(ICP/MS). EPA/625/R-96/010a. Cincinnati, OH.
http://www.epa.gov/ttn/amtic/files/ambient/inorganic/mthd-3-5.pdf Date last accessed:
1/10/2012.
EPA, 2006a. U.S. EPA. December 2006. Standard Operating Procedure for the Determination of
Hexavalent Chromium in Ambient Air Analyzed by Ion Chromatography (1C). EPA 68-
D-00-264. Research Triangle Park, NC. http://www.epa.gov/ttn/amtic/airtox.html Date
last accessed: 1/10/2012.
31-2
-------
EPA, 2007. U.S. EPA. May 2007. 1990-2002 NEI HAP Trends: Success of CAA Air Toxic
Programs in Reducing HAP Emissions and Risk. Paper presented by Anne Pope at the
16th Annual International Emission Inventory Conference. Raleigh, NC.
http://www.epa.gOv/ttn/chief/conference/eil6/session6/a.pope.pdfDate last accessed:
10/10/2012.
EPA, 2009a. U.S. EPA. April 2009. UATMP, NATTS, CSATAM, PAMS, and NMOC Support,
Contract No. EP-D-09-48. Pages 1-2 through 1-3.
EPA, 2009b. U.S. EPA. April 1, 2009. Technical Assistance Document for the National Air
Toxics Trends Stations Program, Revision 2. Research Triangle Park, NC.
http ://www. epa. gov/ttn/amtic/airtox.html
EPA, 2009c. U.S. EPA. September 10, 2009. Schools Air Toxics Monitoring Activity (2009)
Uses of Health Effects Information in Evaluating Sample Results.
http ://www. epa. gov/schoolair/pdfs/UsesOfHealthEffectsInfoinEval SampleResults.pdf
EPA, 2010a. U.S. EPA. September 2010. Air Toxics-National Air Toxics Trends Stations
website, Listing of NATTS sites, http ://www. epa. gov/ttn/amtic/natts.html
EPA, 2010b. U.S. EPA. October 2010. A Preliminary Risk-based Screening Approach for Air
Toxics Monitoring Data Sets, version 2. EPA-904-B-06-001. Atlanta, GA.
http://www.epa.gov/region4/air/airtoxic/Screening 111610 KMEL.pdf Date last
accessed: 10/10/2012.
EPA, 2010c. U.S. EPA. April 2010. Status of Acrolein Monitoring at Schools website.
http://www.epa.gov/schoolair/acrolein.html Date last accessed: 11/29/2011.
EPA, 201 la. U.S. EPA. June 2011. Technical Report for School: Assessing Outdoor Air Near
Schools: Riggins School, North Birmingham Elementary School, and Lewis Elementary
School (Birmingham, Alabama). Definition of 95% confidence interval, pp 5.
http://www.epa.gov/schoolair/NorthBirmi.html Date last accessed: 1/12/2012.
EPA, 201 Ib. U.S EPA. March 2011. The 2005 National-Scale Air Toxics Assessment (NATA)
glossary website, http://www.epa.gov/ttn/atw/natamain/glossl.html Date last accessed:
6/27/2012.
EPA, 201 Ic. U.S. EPA. Technology Transfer Network, Air Toxics, Pollutants and Sources
website, http://www.epa.gov/ttn/atw/pollsour.html Date last accessed: 9/7/2011.
EPA, 201 Id. U.S. EPA. January 2011. An Overview of Methods for EPAs National-Scale Air
Toxics Assessment, Appendix I. http://www.epa.gov/ttn/atw/nata2005/methods.html
Date last accessed: 6/27/2012.
EPA, 2012a. U.S. EPA. Air Toxics - National Air Toxics Trends Stations website.
http://www.epa.gov/ttn/amtic/natts.html Date last accessed: 1/3/2012.
31-4
-------
EPA, 2012b. U.S. EPA. April 10, 2012. 2008 National Emissions Inventory, Version 2.
http://www.epa.gov/ttn/chief/net/2008inventory.html Date last accessed: 10/10/2012.
EPA, 2012c. U.S. EPA. 1986. 40 CFR Part 136, Appendix B. Definition and Procedure for the
Determination of the Method Detection Limit, Revision 1.11.
http://ecfr.gpoaccess.gOv/cgi/t/text/text:-
idx?c=ecfr&tpl=/ecfrbrowse/Title40/40cfrl36 main O2.tpl Date last accessed:
1/10/2012.
EPA, 2012d. Email from Ted Palma, U.S. EPA, OAQPS. July 2, 2012. Dose Response Library
spreadsheet. Date last accessed: 7/2/2012.
EPA, 2012e. U.S. EPA. Mobile Source Emissions-Past, Present, and Future, What are Mobile
Sources website, http://www.epa.gov/otaq/invntory/overview/examples.htm Date last
accessed: 6/25/2012.
EPA, 2012f U.S. EPA. Climate Change Basics website, http://epa.gov/climatechange/basics/
Date last accessed: 6/25/2012.
EPA, 2012g. U.S. EPA. High Global Warming Potential (GWP) Gases and Climate Change
website, http://epa.gov/highgwp/scientific.html Date last accessed: 6/25/2012.
EPA, 2012h. U.S. EPA. Technology Transfer Network, Air Quality System (AQS) website.
http://www.epa.gov/ttn/airs/airsaqs/ Date last accessed: 6/13/2012.
EPA, 2012L U.S. EPA. Mobile Source Emissions-Past, Present, and Future, Hydrocarbons
website, http://www.epa.gov/otaq/invntorv/overview/pollutants/hydrocarbons.htm Date
last accessed: 6/25/2012.
EPA, 2012J. U.S. EPA. September 2012. National Air Toxics Trends Stations (NATTS)
Network Assessment, Revised Draft. Prepared by Eastern Research Group, Inc.
ERG, 2009. Eastern Research Group, Inc. 2009. Support for the EPA National Monitoring
Programs (NMOC, UATMP, PAMS, HAPs, and NATTS), Quality Assurance Project
Plan, Category 1. Contract No. EP-D-03-049. Morrisville, NC.
FAC, 2007. Report of the Federal Advisory Committee on Detection and Quantitation
Approaches and Uses In Clean Water Act Programs, Appendix D: DQ FAC Single
Laboratory Procedure v2.4. December 28, 2007. Submitted to the U.S. EPA.
http://water.epa.gov/scitech/methods/cwa/det/upload/fmal-report-200712.pdfDate last
accessed: 1/10/2012.
FCC, 2012. Florida Climate Center, Florida State University. Special Topics website, Climate of
Florida, http://coaps.fsu.edu/climate center/specials.shtml Date last accessed: 8/2/2012.
31-5
-------
FHWA, 2011. Federal Highway Administration. December 2011. Highway Statistics 2010. State
Motor Vehicle Registrations-2010 (MV-1).
http://www.fhwa.dot.gov/policyinformation/statistics/2010/mvl.cfm
FL DHSMV, 2010. Florida Department of Highway Safety and Motor Vehicles. Registered
Vehicles by County as of November 1, 2010.
http://www.flhsmv.gov/html/reports and statistics/CVR/10-11/CVR-l l-2010.pdf
FL DOT, 2007. Florida Department of Transportation. October 2007. Florida Traffic Monitoring
Handbook, Chapter 4. http://www.dot.state.fl.us/planning/statistics/tmh/ Date last
accessed: 6/25/2012.
FL DOT, 2010a. Florida Department of Transportation. Florida Traffic Online Mapping
Application, 2010 AADTs. http://www.dot.state.fl.us/planning/statistics/trafficdata/Date
last accessed: 12/16/2011.
FL DOT, 201 Ob. Florida Department of Transportation. Public Road Mileage and Miles
Traveled, 2010. http://www.dot.state.fl.us/planning/statistics/mileage-rpts/public.shtm
Date last accessed: 3/26/2012.
GA DOR, 2011. Georgia Department of Revenue, Motor Vehicle Division. Statistics-Vehicle
Registration webpage. Registrations as of November 30, 2011.
http://motor.etax.dor.ga.gov/stats/renewalsstats.aspx Date last accessed: 12/5/2011.
GA DOT, 2010a. Georgia Department of Transportation. Georgia's State Traffic and Report
Statistics (STARS) mapping application. 2010 Published Data.
http://www.dot.state.ga.us/statistics/stars/Pages/default.aspx Date last accessed:
12/16/2011.
GA DOT, 2010b. Georgia Department of Transportation. Mileage by Route and Road System
Report 445 for 2010. 2010 Published Data.
http://www.dot.state.ga.us/statistics/RoadData/Pages/400Series.aspx Date last accessed:
3/26/2012.
Gary, 2012. Gary Chamber of Commerce. About Gary - Geography and Climate website.
http://www.garychamber.com/ Date last accessed: 8/10/2012.
GCPH, 2010. Garfield County Public Health Department. June 2010. Garfield County 2010 Air
Quality Monitoring Summary. Rifle, CO. http://www.garfieldcountyaq.net/ Date last
accessed: 7/27/2012.
GCRBD, 2002. Garfield County Road and Bridge Department. 2002 Average Daily Traffic,
County Road System, Garfield County, Colorado, http://www.garf! eld-county.com/road-
bridge/maps, aspx
31-6
-------
GSCO, 1998. Georgia State Climate Office, University of Georgia. 1998. Climatology of the
Georgia Piedmont. Athens, Georgia.
http://www.rivercenter.uga.edu/education/summit/general/climate/piedmont.pdfDate last
accessed: 8/6/2012.
HCPID, 2010. Harris County Public Infrastructure Department, Architecture and Engineering
Department, Traffic and Transportation Group. Traffic Volumes Report, Update 2010.
Data from December 2004. Houston, Texas, http://www.eng.hctx.net/traffic/traffic.htm
Henrico County, 2011. Email from Deborah Oxenreider, County of Henrico, Revenue Division.
December 12, 2011. 2010 Data.
IL DOT, 2009. Illinois Department of Transportation. Getting Around Illinois website. Average
Annual Daily Traffic map viewer, 2009.
http://www.gettingaroundillinois.com/default.htm?aspxerrorpath=/default.aspx Date last
accessed: 12/16/2011.
IL DOT, 2010. Illinois Department of Transportation. Illinois Travel Statistics 2010.
http://www.dot.il.gov/adttravelstats.html Date last accessed: 12/16/2011.
IL SOS, 2010. Illinois Secretary of State, Vehicle Services Department. 2010 Registration
Counts by County, http://www.sos.state.il.us/departments/vehicles/countycounts.html
IN BMV, 2011. Email from Marc Sirkin, Indiana Bureau of Motor Vehicles Communications
Department. December 8, 2011.
IN DOT, 2009. Indiana Department of Transportation. Average Daily Traffic and Commercial
Vehicles Interactive map. 2009 Data, http://www.in.gov/indot/2350.htm Date last
accessed: 12/21/2011.
IN DOT, 2010. Indiana Department of Transportation. 2010 Indiana Daily Vehicle Miles
Traveled (DVMT) in Thousands, http://www.in.gov/indot/2469.htm Date last accessed:
7/2/2012.
IPCC, 2012. Intergovernmental Panel on Climate Change. 2012. Errata for the Working Group I
Fourth Assessment Report. May 7, 2012.
http://www.ipcc.ch/publications_and_data/publications_ipcc_fourth_assessment_report_
wg 1 report the_physical science basis.htm Date last accessed: 6/26/2012.
ISCO, 2002. Indiana State Climate Office, Purdue University. About Indiana Climate website.
December 2002. http://climate.agry.purdue.edu/climate/narrative.asp Date last accessed:
8/10/2012.
KY, 2012. Kentucky State Parks. Grayson Lake History website.
http://parks.ky.gov/parks/recreationparks/grayson-lake/history.aspx Date last accessed:
8/10/2012.
31-7
-------
KYTC, 2009. Kentucky Transportation Cabinet. Traffic Counts mapping application. 2009 Data.
http://kvtcgis.ky.gov/TrafficCounts/viewer.htm Date last accessed: 12/21/2011.
KYTC, 201 la. Email from Glenda Gordon, Kentucky Transportation Cabinet, Department of
Vehicle Registration. December 8, 2011.
KYTC, 201 Ib. Kentucky Transportation Cabinet. July 2011. 2010 DVMT (in thousands) and
Mileage by County and Functional Class. 2010 Data.
http://transportation.ky.gov/Planning/Pages/Roadway-Information-and-Data.aspx Date
last accessed: 7/2/2012.
Lakes, 2011. Lakes Environmental. June 2011. WRPLOT View 7.0.0.
http://www.weblakes.com/products/wrplot/index.html
MA DOT, 2007. Massachusetts Department of Transportation, Highway Division. Traffic
Counts for Boston website. 2007 Data.
http://www.mhd.state.ma.us/traffic.asp?f=l&C=BOSTON Date last accessed:
12/21/2011.
MA RMV, 2011. Fax from Michael Hooton, Massachusetts Registry of Motor Vehicles.
December 7, 2011.
MCC, 2012. Missouri Climate Center, University of Missouri. Climate of Missouri website.
http://climate.missouri.edu/climate.php Date last accessed: 8/15/2012.
MDS, 2011. Email from Paul T. Kelly, Michigan Department of State. December 7, 2011.
Contains file Michigan Department of State Non-Expired Total Registrations on File as
of 11/13/2010.
MI DOT, 2010. Michigan Department of Transportation, Bureau of Transportation Planning.
Traffic Monitoring Information System (TMIS) website. Annual Average Daily Traffic
Report (2010). http://mdotnetpublic.state.mi.us/tmispublic/Search.aspx Date last
accessed: 12/21/2011.
MI DOT, 2012. Email from Don Howe, Michigan Department of Transportation, Bureau of
Transportation Planning. March 26, 2012. 2010 Data.
MO DOR, 2011. Missouri Department of Revenue. Total Motor Vehicles by Kind within County
(DM45080-01), 2010. Run Date 2/17/11. http://dor.mo.gov/publicreports/index.php
MO DOT, 2009. Missouri Department of Transportation, Transportation Planning. 2009 District
6 Traffic Volume and Commercial Vehicle Count Map (St. Louis-Chesterfield).
http://www.modot.org/safetv/trafficvolumemaps2009.htm Date last accessed:
12/21/2011.
31-8
-------
MO DOT, 2012. Email from Tina Schmidtz, Missouri Department of Transportation. March 29,
2012. 2010 Data.
MSU, 2012. Michigan State Climatologisf s Office, Michigan State University. Climate of
Dearborn Narrative, http://climate.geo.msu.edu/stations/2015/ Date last accessed:
8/14/2012.
NCDC, 2009. National Climatic Data Center. Subscription to Quality Controlled Local
Climatological Data. 2009 Data. https://nes.ncdc.noaa.gov/subscriptions.html Date last
accessed: 9/23/2011.
NCDC, 2010. National Climatic Data Center. Subscription to Quality Controlled Local
Climatological Data. 2010 Data. https://nes.ncdc.noaa.gov/subscriptions.html Date last
accessed: 3/19/2012.
NCDC, 2012. National Climatic Data Center. U.S. Climate Normals website, Climate of the
States, Climatography of the United States No. 60. http://hurricane.ncdc.noaa.gov/cgi-
bin/climatenormals/climatenormals.pl?directive=prod select2&prodtype=CLIM60&subr
num Date last accessed: 8/10/2012.
NIST, 2012. National Institute of Standards and Technology. NIST/SEMATECH e-Handbook of
Statistical Methods website.
http://www.itl.nist.gov/div898/handbook/pmc/section4/pmc42.htm Date last accessed:
6/27/2012.
NJ DOT, 2010a. New Jersey Department of Transportation. Roadway Information and Traffic
Counts mapping application. 2010 Data.
http://www.state.nj.us/transportation/refdata/roadway/traffic_counts/ Date last accessed:
12/22/2011.
NJ DOT, 2010b. New Jersey Department of Transportation. New Jersey's Roadway Mileage and
Daily VMT by Functional Classification Distributed by County. 2010 Data.
http://www.state.nj.us/transportation/refdata/roadway/vmt.shtm Date last accessed:
7/2/2012.
NLM, 2012. National Library of Medicine. Tox Town website for Chromium.
http://toxtown.nlm.nih.gov/text_version/chemicals.php?id= 10 Date last accessed:
1/13/2012.
NOAA, 1999. NOAA National Weather Service. April 1999. NOAA Technical Memorandum,
NWS-WR259, Climate of San Jose.
http://www.wrh.noaa.gov/mtr/sfd sic climate/sic/SJC CLIMATE3.php Date last
accessed: 7/25/2012.
NOAA, 2012a. NOAA, National Weather Service. NWS glossary website.
http://weather.gov/glossary/ Date last accessed: 8/13/2012.
31-9
-------
NOAA, 2012b. NOAA Earth System Research Laboratory. Carbon Cycle Science website.
http://www.esrl.noaa.gov/research/themes/carbon/ Date last accessed: 9/7/2011.
NOAA, 2012c. NOAA, National Weather Service, Buffalo, NY. Rochester Climate Narrative.
http://www.nws.noaa.gov/climate/local_data.php?wfo=buf Date last accessed: 8/27/2012.
NOAA, 2012d. NOAA, National Weather Service, Buffalo, NY. Buffalo Climate Narrative.
http://www.nws.noaa.gov/climate/local_data.php?wfo=buf Date last accessed: 8/27/2012.
NOAA, 2012e. NOAA, National Severe Storms Laboratory. A Severe Weather Primer:
Questions and Answers about Tornadoes website.
http://www.nssl.noaa.gov/primer/tornado/tor_climatology.html Date last accessed:
8/29/2012.
NOAA, 2012f. NOAA, National Weather Service, Climate Prediction Center. Climate Divisions
with Counties website.
http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/regional_monitoring/CLIM
DIVS/states counties climate-divisions.shtml Date last accessed: 9/17/2012.
NYS DEC, 2009. Email from Paul Sierzenga, New York State Department of Environmental
Conservation. January 29, 2009. Map of Tonawanda Toxics Grant, NYSDEC Regulated
Sites.
NYS DMV, 2010. New York State Department of Motor Vehicles. Statistics-Vehicle
Registrations in Force-2010. http://www.nydmv.state.ny.us/Statistics/reginlO.htm Date
last accessed: 12/6/2011.
NYS DOT, 2008. New York State Department of Transportation. Traffic Data Viewer mapping
application. 2008 Data. https://www.nysdot.gov/tdv Date last accessed: 12/22/2011.
OK DOT, 2010. Oklahoma Department of Transportation, Planning and Research Division. 2010
Annual Average Daily Traffic Oklahoma Highway System maps.
http://www.okladot.state.ok.us/maps/aadt/index.htm Date last accessed: 12/22/2011.
OKTC, 2010. Oklahoma Tax Commission, Motor Vehicle Division. Annual Vehicle
Registration Report, Fiscal Year July 1, 2009-June 30, 2010. Oklahoma City, Oklahoma.
http://www.tax.ok.gov/annrpts.html Date last accessed: 12/6/2011.
OR DOT, 2012. Oregon Department of Transportation. Transportation Data website, Definition
of VMT. http://www.oregon.gov/ODOT/TD/TDATA/tsm/vmtpage.shtml Date last
accessed: 6/25/2012.
Pagano, P. and Gauvreau, K., 2000. Pagano, P. and Gauvreau, K. 2000. Principles of
Biostatistics, 2nd edition. Pacific Grove, CA: Duxbury.
31-10
-------
Peltier, 2012. Peltier Techinical Services, Inc. Box and Whisker Plot Utility. Downloaded April
5, 2012. http://peltiertech.com/Utilitv/ Date last accessed: 4/5/2012.
RI DOT, 2009. Rhode Island Department of Transportation. December 2009. State Highway
Map of Rhode Island, Traffic Flow Map 2009.
http://www.dot.ri.gov/engineering/gis/maps.asp
Rutgers, 2012. Office of the New Jersey State Climatologist, Rutgers University. New Jersey
Climate Overview website. http://climate.rutgers.edu/stateclim_vl/njclimoverview.html
Date last accessed: 8/20/2012.
SC DOT, 2011. South Carolina Department of Transportation. April 20, 2011. Average Daily
Traffic. 2010 Data, http://www.dot.state.sc.us/getting/aadt.asp Date last accessed:
12/22/2011.
SC DOT, 2012. Email from Angela Hance, South Carolina Department of Transportation. April
3, 2012. 2010 Data.
SC DPS, 2009. South Carolina Department of Public Safety, Office of Highway Safety. South
Carolina Traffic Collision Fact Book 2009. http://www.scdps.org/ohs/stat_services.asp
Date last accessed: 12/22/2011.
SC SCO, 2012. South Carolina State Climate Office, South Carolina Climate website.
http://www.dnr.se.gov/climate/sco/ClimateData/cli_sc_climate.php Date last accessed:
9/4/2012.
SD DENR, 2010. South Dakota Department of Environment and Natural Resources. South
Dakota Annual Ambient Air Monitoring Network Plan 2010.
http://denr.sd.gov/des/aq/monitoring/state-mo.aspx Date last accessed: 9/3/2012.
SD DOR, 2010. South Dakota Department of Revenue, Motor Vehicle Division. Number of
Titles Awarded by Year by County website. 2010 Data.
http://www.state.sd.us/drr2/motorvehicle/titie/history_of_titlesl2.htm Date last accessed:
12/6/2011.
SD DOT, 2007. South Dakota Department of Transportation, Transportation Inventory
Management. Traffic Map website. 2007 Data.
http://www.sddot.com/pe/data/traf maps.asp Date last accessed: 12/22/2011.
SD DOT, 2010. South Dakota Department of Transportation, Transportation Inventory
Management. Traffic Map website. 2010 Data.
http://www.sddot.com/pe/data/traf maps.asp Date last accessed: 12/22/2011.
31-11
-------
SD DOT, 2011. South Dakota Department of Transportation. June 2011. South Dakota Vehicle
Miles of Travel Report by County - All Vehicles. 2010 Data.
http://www.sddot.com/transportation/highways/traffic/Default.aspx Date last accessed:
3/26/2012.
Steer, 2008. Steer Davies Gleave. January 2008. New Jersey Traffic and Revenue Study, New
Jersey Turnpike & Route 440 Asset Appraisal. Final Report prepared for State of New
Jersey Department of Treasury by Steer Davies Gleave in association with CRA
International and EDR Group.
http://slic.nistatelib.org/slic files/digidocs/f491/f4912008d.pdf
TAMU, 2012. Office of the Texas State Climatologist, Texas A&M University. Climate of
Texas. http://atmo.tamu.edu/osc/library/osc_pubs/Date last accessed: 9/5/2012.
TX DMV, 2011. Email from Diane Emrick-Dodson, Texas Department of Motor Vehicles,
Vehicle Titles and Registration Division. December 6, 2011. Vehicle Registration and
License Fees by County Fiscal Year September 1, 2009 to August 31, 2010.
TX DOT, 2010. Texas Department of Transportation. District Traffic Maps 2010.
http://www.txdot.gov/travel/traffic maps/2010.htm Date last accessed: 12/22/2011.
UT DOT, 2010. Utah Department of Transportation. Traffic Maps for 2010.
http://www.udot.utah.gov/main/f?p=100:pg:0::::V.T:.528 Date last accessed: 12/22/2011.
UT DOT, 2011. Utah Department of Transportation. October 2011. 2010 Vehicle Miles of
Travel (VMT) by County by Ownership.
http://www.udot.utah.gov/main/f?p=100:pg:0::::V.T:.530 Date last accessed: 7/2/2012.
UT TC, 2010. Utah Tax Commission. February 2010. 2010 On-road Registrations by County
and Vehicle Type, http://tax.utah.gov/esu/mv-registrations
VA DOT, 2009. Virginia Department of Transportation, Traffic Engineering Division. Average
Daily Traffic Volumes with Vehicle Classification Data on Interstate, Arterial, and
Primary Routes 2009. http://www.virginiadot.org/info/ct-TrafficCounts.asp Date last
accessed: 12/22/2011.
VA DOT, 2011. Virginia Department of Transportation. October 2011. Daily Vehicle Miles
Traveled (DVMT) by Physical Jurisdiction, with Towns Combined into Counties
(VMT1206). 2010 data.
http://www.virginiadot.org/info/2010_traffic_data_daily_vehicle_miles_traveled.asp
Date last accessed: 7/3/2012.
VT DMV, 2010. Email from Shannon Fassett, Vermont Department of Motor Vehicles. August
2, 2010. 2010 Data.
31-12
-------
Vtrans, 2010. Vermont Agency of Transportation, Highway Research. 2010 Miles and Annual
Vehicle Miles Traveled (AVMT) by County and Functional Class.
http ://www. aot. state, vt.us/planning/documents/highresearch/publications/pub .htm Date
last accessed: 7/3/2012.
Vtrans, 2011. Vermont Agency of Transportation, Planning, Outreach, and Community Affairs
Division, Traffic Research Unit. May 2011. 2010 (Route Log) AADTs State Highways.
http://www.aot.state.vt.us/Planning/Documents/TrafResearch/Publications/pub.htm Date
last accessed: 12/22/2011.
WADOL, 2010. Washington State Department of Licensing. Vehicle/Vessel Fee Collection and
Vehicle Counts Program website. Motor Vehicle Registration by Class and County.
Calendar Year 2010. http://www.dol.wa.gov/about/vehvesselreports.html Date last
accessed: 12/6/2011.
WADOT, 2010a. Washington State Department of Transportation. 2010 Annual Traffic Report.
http ://www. wsdot. wa. gov/mapsdata/travel/annual trafficreport.htm
WADOT, 201 Ob. Washington State Department of Transportation. State Highway (Interstate,
State Route and U.S. Route) Centerline Miles and Annual Vehicle Miles of Travel
(AVMT) in Thousands by Year, County, State Functional Class of Roadway and Urban
Versus Rural.xls. 2010 data.
http ://www. wsdot. wa. gov/mapsdata/travel/annual trafficreport.htm Date last accessed:
7/3/2012.
WI DOT, 2008. Wisconsin Department of Transportation. 2008 City of Horicon Dodge County
Annual Average Daily Traffic map.
http://www.dot.wisconsin.gov/travel/counts/dodge.htm Date last accessed: 12/21/2011.
WI DOT, 2010. Wisconsin Department of Transportation, Division of Motor Vehicles. Facts and
Figures 2010, A Reference Guide. Madison, WI.
http://www.dot.wisconsin.gov/drivers/facts.htm
WI DOT, 2011. Wisconsin Department of Transportation. September 2011. 2010 Vehicle Miles
of Travel (VMT) by County, http://www.dot.wisconsin.gov/travel/counts/vmt.htm Date
last accessed: 7/3/2012.
WRCC, 2012. Western Region Climate Center. Climate Narrative of the States: Climate of
Alaska, Arizona, California, Colorado, Utah.
http ://www. wrcc. dri. edu/CLIMATED AT A. html Date last accessed: 7/20/2012.
Xionetic, 2011. Xionetic Technologies. July 2011. Products website, ZipFind Demo.
http://xionetic.com/ZipFindDeluxe.aspx Date last accessed: 10/6/2011.
31-13
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-454/R-12-006a
Environmental Protection Emissions, Monitoring and Analysis Division November 2012
Agency Research Triangle Park, NC
------- |