NATIONAL AIR TOXICS TRENDS STATIONS
QUALITY ASSURANCE ANNUAL REPORT
CALENDAR YEARS 2011 AND 2012

FINAL
December 12,2014

Environmental Protection Agency
Office of Air Quality, Planning and Standards
Air Quality Analysis Division
109 TW Alexander Drive
Research Triangle Park, NC 27711


-------
FORWARD

This technical report was prepared by Battelle under Contract No. GS-10F-0275K, Task Order
EP-G11D-00028. This report describes the Quality Assurance (QA) data collected for the
NATTS program during calendar years (CYs) 2011 and 2012. The report was prepared for
Margaret Dougherty, Task Order Project Officer, and David Shelow, Alternate Task Order
Project Officer at the Office of Air Quality Planning and Standards (OAQPS) in Research
Triangle Park, North Carolina.

Please note that this report contains a change to the analysis that differs from previous quality
assurance annual reports. The change pertains to the analysis of the precision data. In the
previous report for 2010, all precision data records that reported a value, whether below, equal
to, or above the method detection limit (MDL), were included in the precision calculations as
described in Section 2.5. In this report, data are utilized for the precision calculations for each
site and HAP only where both results in the replicate pair are equal to or above the reported
MDL.

This report was revised to correct the MDL measurement quality objective (MQO) for
naphthalene. The naphthalene MDL MQO was reported to be 0.029 ng/m3 in the April 17, 2014
version of report, which was in error. The correct value is 29.0 ng/m3. Applicable sections of
this report have been changed to reflect the correct MDL.

ii


-------
NATIONAL AIR TOXICS TRENDS STATIONS
QUALITY ASSURANCE ANNUAL REPORT
CALENDAR YEARS 2011 AND 2012

Prepared by
Douglas J. Turner
Laura L.S. Aume
Robert T. Woodruff
Robert A. Lordo
Ian C. MacGregor
Battelle

For:

U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Analysis Division
109 TW Alexander Drive
Research Triangle Park, NC 27711

Under:

U.S. EPA Contract GS-10F-0275K

Task Order EP-G11D-00028
Battelle Contract CON00010461
Task B, Option Period III
Project # 100005824

iii


-------
ACKNOWLEDGEMENTS

This report was prepared by Battelle, under Task Order EP-G11D-00028, Contract GS-10F-
0275K.

iv


-------
TABLE OF CONTENTS

Section	Page

1.0 INTRODUCTION	1

2.0 NATTS QUALITY ASSURANCE DATA FOR CY2011 AND CY2012	3

2.1	The NATTS Network Sites in CY2011 and CY2012	3

2.2	HAPs Measured in the NATTS Network in CY2011 and CY2012	4

2.3	Measurement Quality Objectives	5

2.4	Completeness of NATTS Data	7

2.5	Precision of NATTS Data	14

2.5.1	Analytical Precision Results	20

2.5.2	Overall Precision Results	26

2.6	Laboratory Bias Data Based on Proficiency Testing (PT) Samples	46

2.7	Flow Audit Results from Instrument Performance Audits (IPAs)	58

2.8	Method Detection Limit (MDL) Data	67

3.0 SUMMARY	86

4.0 RECOMMENDATIONS	89

5.0 REFERENCES	90

v


-------
LIST OF TABLES

Table	Page

Table 1. EPA Regions, NATTS Sites, Site Type, and Air Quality Systems Site Codes	4

Table 2. The 27 NATTS Hazardous Air Pollutants and Air Quality Systems Parameter

Codes	5

Table 3. Measurement Quality Objectives for the Seven HAPs of Primary Importance

to the NATTS Program	6

Table 4. Data Sources Used to Evaluate the NATTS Data Quality Indicators	7

Table 5. Percentage Completeness Values by NATTS Site and the Seven HAPs of

Primary Importance for CY2011	9

Table 6. Percentage Completeness Values by NATTS Site and the Seven HAPs of

Primary Importance for CY2012	10

Table 7. Percentage of NATTS Sites Meeting the Completeness MQO for the Seven

HAPs of Primary Importance for CY2011 and CY2012	11

Table 8. Parameter Occurrence Codes by NATTS Site and HAP Type - CY2011	16

Table 9. Parameter Occurrence Codes by NATTS Site and HAP Type - CY2012	17

Table 10. Laboratories Performing Analyses by HAP Type for Each NATTS Site in

CY2011 and CY2012	19

Table 11. Laboratory Abbreviations and Descriptions for NATTS Laboratories	20

Table 12. Analytical Precision for Replicate Analyses > MDL - CY2011	22

Table 13. Analytical Precision for Replicate Analyses > MDL - CY2012	24

Table 14. Percentage of NATTS Sites Meeting the MQO for Overall Precision -

CY2011 and CY2012	27

Table 15. Overall Precision for Primary, Duplicate, and Collocated Samples > MDL -

CY2011	28

Table 16. Overall Precision for Primary, Duplicate, and Collocated Samples > MDL -

CY2012	30

Table 17. Percentage of NATTS Laboratories Participating in the NATTS Proficiency

Testing Program in CY2011 and CY2012	46

Table 18. Additional HAPs Contained in NATTS Proficiency Test Samples in

CY2011 and/or CY2012	47

Table 19. Non-NATTS Laboratories Analyzing Proficiency Test Samples in CY2011

and CY2012	48

Table 20. Percentages of NATTS Laboratories Meeting the Bias MQO for Proficiency

Test Samples in CY2011 and CY2012	49

vi


-------
Table 21. NATTS Proficiency Testing Bias Results (Percent Difference from Target)

for VOCs - CY2011 QTR4	50

Table 22. NATTS Proficiency Testing Bias Results (Percent Difference from Target)

for Carbonyls - CY2011 QTR4	51

Table 23. NATTS Proficiency Testing Bias Results (Percent Difference from Target)

for Metals - CY2011 QTR2	52

Table 24. NATTS Proficiency Testing Bias Results (Percent Difference from Target)

for PAHs - CY2011 QTR2	53

Table 25. NATTS Proficiency Testing Bias Results (Percent Difference from Target)

for Metals - CY2012 QTR1	54

Table 26. NATTS Proficiency Testing Bias Results (Percent Difference from Target)

for PAHs - CY2012 QTR1	55

Table 27. NATTS Proficiency Testing Bias Results (Percent Difference from Target)

for Chromium (VI) - CY2012 QTR1	56

Table 28. Mean Network Flow Bias From CY2011 and CY2012	59

Table 29. Flow Audit Results from the Instrument Performance Audits - CY2011	60

Table 30. Flow Audit Results from the Instrument Performance Audits - CY2012	62

Table 31. Percentage of Audited NATTS Sites Meeting the Flow Bias MQO - CY2011

and CY2012	64

Table 32. Percentage of Sites Meeting the MDL MQO - CY2011 and CY2012	68

Table 33. Average Method Detection Limits (MDLs) by Site and Overall for VOCs

(l^g/m3) and PAHs (ng/m3) - CY2011	69

Table 34. Average Method Detection Limits (MDLs) by Site and Overall for carbonyls

(^ig/m3), metals (ng/m3), and chromium (VI) (ng/m3) - CY2011	71

Table 35. Average Method Detection Limits (MDLs) by Site and Overall for VOCs

(l^g/m3) and PAHs (ng/m3) - CY2012	72

Table 36. Average Method Detection Limits (MDLs) by Site and Overall for carbonyls

(^ig/m3), metals (ng/m3), and chromium (VI) (ng/m3) - CY2012	74

Table 37. Summary Statistics for Method Detection Limits across All Reporting

NATTS Laboratories - CY2011	75

Table 38. Summary Statistics for Method Detection Limits across All Reporting

NATTS Laboratories - CY2012	75

Table 39. Summary of NATTS Quality Assurance Results - Percentage of Sites

Meeting Measurement Quality Objectives in CY2011 and CY2012	86

Table 40. Median Completeness for the Seven HAPs of Primary Importance for

CY2011 and CY2012	86

Table 41. Ratio of the MDL Network Geometric Mean to the MQO for the Seven

HAPs of Primary Importance for CY2011 and CY2012	88

vii


-------
,12

.13

.32

.33

.34

.35

.36

.37

.38

.39

.40

.41

.42

.43

.44

.45

.57

.58

.65

.66

.76

LIST OF FIGURES

Box and Whisker Plot of Percentage Complete Values for the Seven HAPs

of Primary Importance, for CY2011	

Box and Whisker Plot of Percentage Complete Values for the Seven HAPs

of Primary Importance, for CY2012	

Analytical and Overall Precision Summary for Acrolein > MDL at NATTS

Sample Collection Sites in CY2011 	

Analytical and Overall Precision Summary for Benzene > MDL at NATTS

Sample Collection Sites in CY2011	

Analytical and Overall Precision Summary for 1,3-Butadiene > MDL at

NATTS Sample Collection Sites in CY2011	

Analytical and Overall Precision Summary for Formaldehyde > MDL at

NATTS Sample Collection Sites in CY2011	

Analytical and Overall Precision Summary for Naphthalene > MDL at

NATTS Sample Collection Sites in CY2011	

Analytical and Overall Precision Summary for PMio Arsenic > MDL at

NATTS Sample Collection Sites in CY2011	

Analytical and Overall Precision Summary for Chromium (VI) > MDL at

NATTS Sample Collection Sites in CY2011	

Analytical and Overall Precision Summary for Acrolein > MDL at NATTS

Sample Collection Sites in CY2012	

Analytical and Overall Precision Summary for Benzene > MDL at NATTS

Sample Collection Sites in CY2012	

Analytical and Overall Precision Summary for 1,3-Butadiene > MDL at

NATTS Sample Collection Sites in CY2012	

Analytical and Overall Precision Summary for Formaldehyde > MDL at

NATTS Sample Collection Sites in CY2012	

Analytical and Overall Precision Summary for Naphthalene > MDL at

NATTS Sample Collection Sites in CY2012	

Analytical and Overall Precision Summary for PMio Arsenic > MDL at

NATTS Sample Collection Sites in CY2012	

Analytical and Overall Precision Summary for Chromium (VI) > MDL at

NATTS Sample Collection Sites in CY2012	

Distribution of Laboratory Bias by HAP for Proficiency Testing Data -

CY2011 	

Distribution of Laboratory Bias by HAP for Proficiency Testing Data -

CY2012	

Summary of Instrument Performance Flow Audit Results for NATTS Sites

CY2011	

Summary of Instrument Performance Flow Audit Results for NATTS Sites

CY2012	

Distribution of VOCs Average Method Detection Limits for NATTS Data -
CY2011 	

viii


-------
Figure 22.	Distribution of VOCs Average Method Detection Limits for NATTS Data -

CY2012	77

Figure 23.	Distribution of Carbonyls Average Method Detection Limits - CY2011	78

Figure 24.	Distribution of Carbonyls Average Method Detection Limits - CY2012	79

Figure 25.	Distribution of PAHs Average Method Detection Limits - CY2011	80

Figure 26.	Distribution of PAHs Average Method Detection Limits - CY2012	81

Figure 27.	Distribution of Metals Average Method Detection Limits - CY2011	82

Figure 28.	Distribution of Metals Average Method Detection Limits - CY2012	83

Figure 29.	Distribution of Chromium (VI) Average Method Detection Limits - CY2011	84

Figure 30.	Distribution of Chromium (VI) Average Method Detection Limits - CY2012	85

ix


-------
ABBREVIATIONS AND ACRONYMS

AQS

Air Quality System

CV

coefficient of variation

CY

calendar year

DQI(s)

data quality indicator(s)

DQO

data quality objective

HAP(s)

hazardous air pollutant(s)

IPA(s)

instrument performance audit(s)

IQR

interquartile range

LC

local conditions

MDL(s)

method detection limit(s)

MQO(s)

measurement quality objective(s)

NATTS

National Air Toxics Trends Stations

NIST

National Institute of Standards and Technology

PAH(s)

polycyclic aromatic hydrocarbon(s)

POC(s)

parameter occurrence code(s)

PTs)

proficiency test(s)

QAAR

quality assurance annual report

QTR

quarter

RD

raw data record in AQS

RP

replicate record in AQS

STP

standard temperature and pressure

TSA

technical systems audit

VOC(s)

volatile organic compound(s)

x


-------
1.0 INTRODUCTION

As mandated under the Government Performance Results Act, the U.S. Environmental
Protection Agency (EPA) is focused on reducing the risk of cancer and other serious health
effects associated with hazardous air pollutants (HAPs) by achieving a 75% reduction in air
toxics emissions chemicals, based on 1993 levels. The current inventory of HAPs includes 188
chemicals regulated under the Clean Air Act that have been linked to numerous adverse human
health and ecological effects, including cancer, neurological effects, reproductive effects, and
developmental effects. Current Agency attention is targeting risk reduction associated with
human exposure to air toxics.

The National Air Toxics Trends Station (NATTS) network was established to create a database
of air quality data to assess progress in reducing ambient concentrations of air toxics and
concomitant exposure-associated risk. During calendar years (CY) 2011 and 2012, the NATTS
network consisted of 27 stations in the contiguous 48 states. To ensure the quality of the data
collected under the NATTS network, EPA has implemented a Quality System comprising three
primary components: (1) Technical Systems Audits (TSAs) of sample analysis laboratories and
network stations, (2) Instrument Performance Audits (IPAs) of network stations, and (3)
quarterly proficiency testing (PT) of the sample analysis laboratories. These assessments ensure
that sampling and analysis techniques are consistent with required completeness, precision, bias,
and method detection limits (MDLs) as specified by the NATTS Measurement Quality
Objectives (MQOs).

This report describes and summarizes the quality assurance (QA) data generated for the NATTS
program during CY2011 and CY2012. For data retrieved from EPA's Air Quality Systems
(AQS) database, only data input prior to November 7, 2013, are considered in this assessment.
Although this report details substantive information on 27 different HAPs of interest, it focuses
primarily on results for seven pollutants: acrolein, benzene, 1,3-butadiene, formaldehyde,
naphthalene, PMio arsenic, and chromium (VI). These pollutants represent each of the five
classes of HAPs that are analyzed within the NATTS program: VOCs, carbonyls, PAHs, PMio
metals, and hexavalent chromium. At the request of EPA, these seven pollutants were selected
as being representative of their respective constituent class and were of particular interest by
virtue of their associated health risk due to inhalation exposure and the frequency of their
occurrence at measurable concentrations. Although no HAP or group of HAPs can provide
complete representation of their respective HAP class, it is presumed that if the NATTS program
can meet the measurement quality objectives (MQOs) for these seven HAPs, the additional
20 pollutants of concern will be of comparable quality by virtue of the representativeness of the
physicochemical properties and the consistency of the collection and analysis methodologies.

The information in this Quality Assurance Annual Report (QAAR) was compiled from data
acquired from numerous sources. The following general categories of information are presented:

•	Descriptive background information on the AQS sites, HAPs of interest, and MQOs;

•	Assessment of the completeness of the data available in the AQS database;

1


-------
•	Precision estimates for both analytical and overall sampling error computed for as
many of the 27 HAPs and for as many of the 27 NATTS sites as available in AQS for
CY2011 and CY2012;

•	Evaluation of analytical laboratory bias based on results of blind audit PT samples for
many of the 27 HAPs;

•	Field bias data, which are expressed as the percent difference between sampler flow
readings and a calibrated flow standard for each of four different sampler types
associated with carbonyls, PMio metals, chromium (VI), and PAHs for primary
samplers and precision (collocated or duplicate) samplers (where available) during
IPAs conducted at eight sites visited in CY2011 and five sites visited in CY2012; and

•	MDL data for each site. The AQS database, specifically the ALT_MDL variable, was
used as the source of MDLs for CY2011 and CY2012.

Where possible, all data analyses were performed in SAS, version 9.3. Graphs and plots were
prepared using STATA version 13.0. Field flow audit data were transcribed into Microsoft
Excel.

2


-------
2.0 NATTS QUALITY ASSURANCE DATA FOR CY2011 AND CY2012

2.1 The NATTS Network Sites in CY2011 and CY2012

The NATTS network included 27 sites in CY2011 and CY2012. Table 1 lists these sites along
with the EPA Region in which each site is located, the site name, whether the site is located in a
predominantly urban or rural area, and the site's unique AQS identification code [1].

Although a city and state are typically used as the site name, a county name is used for the two
Florida sites on either side of Tampa Bay, for the South Carolina site, and for the site located in
Harrison County, TX. Historical consistency has been maintained for the Grand Junction, CO
site, to which two separate AQS site identification codes were assigned, one code for VOCs,
carbonyls, PAHs, and chromium (VI) (08-077-0018), and another code for PMio metals
(08-077-0017). This convention is unique to this site and is used because the organics and
metals samplers are situated at separate physical locations at the sampling site. The Bronx, NY
site had a different AQS site code starting in July 2012, when the site location changed upon
completion of renovation construction. Prior to July 2012, this site had a site code of
36-005-0080, and upon completion of the renovation, sample collection resumed under site code
36-005-0110.

3


-------
Table 1. EPA Regions, NATTS Sites, Site Type, and Air Quality Systems Site Codes

EPA Region

Site Name

Site Type

AQS Site Identification Code

I

Boston-Roxbury, MA

Urban

25-025-0042

I

Underhill, VT

Rural

50-007-0007

I

Providence, RI

Urban

44-007-0022

II

Bronx, NY

Urban

36-005-0080a, -0110b

II

Rochester, NY

Urban

36-055-1007

III

Washington, DC

Urban

11-001-0043

III

Richmond, VA

Urban

51-087-0014

IV

Chesterfield, SC

Rural

45-025-0001

IV

Decatur, GA

Urban

13-089-0002

IV

Grayson Lake, KY

Rural

21-043-0500

IV

Hillsborough County, FL

Urban

12-057-3002

IV

Pinellas County, FL

Urban

12-103-0026

V

Dearborn, MI

Urban

26-163-0033

V

Horicon, WI

Rural

55-027-0001

V

Northbrook, IL

Urban

17-031-4201

VI

Deer Park, TX

Urban

48-201-1039

VI

Harrison County, TX

Rural

48-203-0002

VII

St. Louis, MO

Urban

29-510-0085

VIII

Bountiful, UT

Urban

49-011-0004

VIII

Grand Junction, CO

Rural

08-077-0017°, -0018d

IX

Phoenix, AZ

Urban

04-013-9997

IX

San Jose, CA

Urban

06-085-0005

IX

Rubidoux, CA

Urban

06-065-8001

IX

Los Angeles, CA

Urban

06-037-1103

X

La Grande, OR

Rural

41-061-0119

X

Portland, OR

Urban

41-051-0246

X

Seattle, WA

Urban

53-033-0080

a Discontinued July 2012
b Resumed July 2012
cPMio metals only

d VOCs, carbonyls, PAHs, and Cr(VI) only

2.2 HAPs Measured in the NATTS Network in CY2011 and CY2012

The 27 HAPs measured in the NATTS program are listed in Table 2. EPA selected these air
pollutants due to their significant health concern. These include 16 VOCs, 2 carbonyls, 2 PAHs,
6 PMio metals, and chromium (VI). Succinct abbreviations of each chemical name are also
specified in this table, as they are used to identify HAPs in subsequent tables and figures
throughout this report.

4


-------
Table 2. The 27 NATTS Hazardous Air Pollutants and
Air Quality Systems Parameter Codes

HAP

HAP
Abbreviation

AQS Label

AQS Code(s)

HAP Class

benzene

BENZa

Benzene

45201

voc

1,3-butadiene

BUTAa

1,3-Butadiene

43218

voc

carbon tetrachloride

CTET

Carbon Tetrachloride

43804

voc

chloroform

CLFRM

Chloroform

43803

voc

1,2-dibromoethane

EDB

Ethylene Dibromide

43843

voc

1,2-dichloropropane

DCP

1,2-Dichloropropane

43829

voc

1,2-dichloroethane

EDC

Ethylene Dichloride

43815

voc

dichloromethane

MECL

Dichloromethane

43802

voc

1,1,2,2-tetrachloroethane

TCE1122

1,1,2,2-T etrachloroethane

43818

voc

tetrachloroethylene

PERC

T etrachloroethylene

43817

voc

trichloroethylene

TCE

T richloroethylene

43824

voc

vinyl chloride

VC

Vinyl Chloride

43860

voc

cis -1,3 -dichloropropene

cDCPEN

Cis -1,3 -Dichloropropylene

43831

voc

trans -1,3 -dichloropropene

tDCPEN

T rans -1,3 -Dichloropropylene

43830

voc

acrolein

ACROa

Acrolein

43505b43509°

voc

acrylonitrile

ACRY

Acrylonitrile

43704

voc

naphthalene

NAPHa

Naphthalene (TSP) STP

17141d

PAH

benzo[a]pyrene

BaP

Benzo[A]Pyrene (TSP) STP

17242d

PAH

formaldehyde

FORM2

Formaldehyde

43502

Carbonyl

acetaldehyde

ACET

Acetaldehyde

43503

Carbonyl

arsenic

Asa

Arsenic PMio

82103d851036

Metal

beryllium

Be

Beryllium PMio

82105d 851056

Metal

cadmium

Cd

Cadmium PMio

82110d 85110e

Metal

lead

Pb

Lead PMio

82128d 851286

Metal

manganese

Mn

Manganese PMio

82132d 851326

Metal

nickel

Ni

Nickel PMio

82136d 851366

Metal

chromium (VI)

CrVIa

Chromium (VI) TSP

12115d 14115e

Metal

a HAP is representative of other chemicals in this class.
b unverified results
c verified results
d standard conditions (STP)
e local conditions (LC)

Note that the superscript "a" in the HAP Abbreviation column of Table 2 denotes the seven
HAPs that serve as representative of their respective constituent classes for this quality
investigation: acrolein, benzene, 1,3-butadiene, formaldehyde, naphthalene, PMio arsenic, and
chromium (VI). In this document, these seven HAPs are referred to the HAPs of "primary
importance" to the NATTS program.

2.3 Measurement Quality Objectives

MQOs applicable to the various data quality indicators (DQIs) for seven HAPs of primary
importance are summarized in Table 3. The MQOs for the DQIs of completeness, precision, and
laboratory bias, as established for the NATTS program to ensure acceptable data quality within

5


-------
the network, are documented in the Technical Assistance Document [4] dated April 1, 2009. The
DQI of sensitivity is represented as the method detection limits (MDLs), and the MDL MQOs
for CY2011 and CY2012 are documented in the National Air Toxics Trends Station Work Plan
Template revised 2/09/2011 and 4/11/2012, respectively [2, 3]. The stated Data Quality
Objective (DQO) for the NATTS program is "to be able to detect a 15 percent difference (trend)
between two consecutive 3-year annual mean concentrations within acceptable levels of decision
error" [5].

Table 3. Measurement Quality Objectives for the Seven HAPs of Primary Importance to

the NATTS Program

HAP

Data Quality Indicators a

Completeness
(Section 2.4)

Analytical and
Overall Precision
(% Coefficient of
Variation)
(Section 2.5)

Laboratory

Bias
(Section 2.6)

Method Detection Limit (MDL)
(Section 2.8)

CY2011

CY2012

acrolein

benzene

1,3-butadiene

formaldehyde

naphthalene

arsenic (PMio)

chromium(VI)

>85%

< 15%

<25%

<	0.10 (xg/m3
<0.13 (xg/m3

<	0.10 (xg/m3

<	0.98 (xg/m3

<	29 ng/m3

<	0.23 ng/m3

<	0.08 ng/m3

<	0.09 (xg/m3
<0.13 (xg/m3

<	0.10 (xg/m3

<	0.08 (xg/m3

<	29 ng/m3

<	0.23 ng/m3

<	0.08 ng/m3

a. Technical Assistance Document for the National Ambient Air Toxics Trends and Assessment Program,

Revision 2, April 2009. [4]

Additional information and requirements associated with the DQIs and MQOs in Table 3 are as
follows:

1	Completeness is measured by calculating the percentage of full sample collection that
occurred, where full sample collection denotes the collection of samples every sixth
day through the entire calendar year.

2	Precision is calculated as the percent coefficient of variation (CV) for replicate
analyses, and for duplicate and collocated samples. Two types of precision are
assessed: analytical precision, and overall precision.

3	Bias denotes the assessment of laboratory performance through analysis of blind audit
PT samples.

4	MDLs inform measurement sensitivity. Sensitivity requirements are achieved if the
reported MDLs are less than or equal to target MDLs in Table 3.

5	Comparability requirements are achieved if the methods are consistent and all of the
above MQOs are met.

The MQO for flow rate, or field, bias is < 10%. Data acquired to assess compliance with the
MQOs were derived from a variety of sources. These sources are given in Table 4.

6


-------
Table 4. Data Sources Used to Evaluate the NATTS Data Quality Indicators

Data Quality Indicator

Data Source

Representativeness/Completenes s

AQS

Analytical and Overall Precision

AQS

Bias - Laboratory/analytical

Proficiency testing results reported by Wibby Environmental and Battelle

Bias - Flow rate/sampling

Audits of sampler flow rates conducted by RTI International

Sensitivity/MDL

AQS augmented with information from the analytical laboratories

For completeness, precision (analytical and overall), and MDL metrics, Battelle retrieved from
the AQS database data records corresponding to relevant samples collected from the 27 NATTS
sites during CY2011 and CY2012. Only those data present in AQS prior to November 7, 2013
were included in this report.

Analytical bias was calculated using PT sample analysis results distributed by Wibby
Environmental (in the 2nd quarter (QTR2) of 2011) and Battelle (in the 4th quarter (QTR4) of
2011 and the 1st quarter (QTR1) of 2012). Sampling bias was estimated using results from
independent measurement of sampler flow rates with National Institute of Standards and
Technology (NIST)-traceable flow standards during on-site Instrument Performance Audits
(IPAs).

2.4 Completeness of NATTS Data

Tables 5 and 6 present the completeness of NATTS data in AQS for CY2011 and CY2012 for
the seven HAPs of primary importance to the NATTS program. Based on the specified
collection frequency MQO of l-in-6 day sample collection, 61 records for the primary parameter
occurrence code (POC) represent 100% completeness. Thus, for a given HAP and site,
percentage complete was calculated by dividing by 61 the total number of records with valid
results reported to AQS. For the purposes of this completeness calculation, nondetects were
counted as valid results, but missing values or nullified results were not.

Completeness statistics were computed using records corresponding to primary measurements
or, if the primary measurement was missing, to collocated measurement(s) collected at the same
location during the same sampling period. A record was understood to be missing if no record
existed in AQS for the expected date, the record did not include a result, or the record was
nullified. Only a single record was included for each site, date, and HAP.

Sample collection at some sites was performed more frequently than others in order to meet the
requirements of other sampling networks or for other specific purposes. Thus, an algorithm was
developed to compile the AQS data so as to allow for flexibility in handling missed and
subsequent make-up samples which may not have complied strictly with the NATTS protocol of
sampling every six days. This algorithm was designed as follows:

1. RD (raw data) records in AQS corresponding to any POC were considered valid if the
"Sample Value" was not specified as missing and "Sample Duration" was equal to 7

7


-------
(corresponding to 24-hour sample collection). This included any primary, duplicate,
or collocated data in the RD dataset.

2.	A maximum of one record was counted per given sampling day.

3.	The first record reported in a given calendar year was always counted. The date of
this record was then used to determine the elapsed time to the next record.

4.	Any record that corresponded to sampling at six or more days following the previous
record was always counted.

5.	If a record corresponded to sampling at fewer than six days after the previous record,
then that record is counted only if the time interval between the record and the
immediate prior two records is 12 days or more. (This assumes that the sample serves
as a make-up for a sample that was missed prior to the last record. It eliminates the
use of back-to-back samples to make up for weeks of missing data.)

The calculated percentage complete values are presented for each NATTS site and for each of
the seven HAPs of primary importance in Table 5 and Table 6, for CY2011 and CY2012,
respectively. These tables also include the mean and median percentage complete values across
all NATTS sites for each HAP. Percentage complete values that fall below the NATTS MQO of
85% are noted in red within these tables. The percentage of NATTS sites meeting the
completeness MQO for CY2011 and CY2012 are shown in Table 7.

8


-------
Table 5. Percentage Completeness Values by NATTS Site and the Seven HAPs of Primary

Importance for CY2011







HAP Abbreviation and Parameter Code(s)



AQS Site Identification Code

Site Name

ACRO
43505
43509

BENZ
45201

BUTA
43218

FORM
43502

NAPH
17141

As
82103
85103

CrVI
12115
14115

25-025-0042

Boston, MA

98%

98%

98%

100%

100%

97%

100%

50-007-0007

Underhill, VT

0%b

92%

92%

93%

95%

80%

93%

44-007-0022

Providence, RI

89%

89%

89%

85%

92%

92%

90%

36-005-0080

Bronx, NY

98%

98%

98%

90%

98%

95%

98%

36-055-1007

Rochester, NY

85%

85%

85%

85%

93%

89%

92%

11-001-0043

Washington, DC

98%

93%

93%

97%

93%

95%

97%

51-087-0014

Richmond, VA

97%

97%

97%

95%

95%

95%

98%

45-025-0001

Chesterfield, SC

100%

100%

100%

95%

98%

98%

97%

13-089-0002

Decatur, GA

97%

97%

97%

97%

92%

93%

93%

21-043-0500

Grayson Lake, KY

98%

100%

100%

84%

100%

98%

100%

12-057-3002

Hillsborough County, FL

98%

98%

98%

98%

97%

100%

93%

12-103-0026

Pinellas County, FL

100%

100%

100%

100%

98%

97%

98%

26-163-0033

Dearborn, MI

98%

100%

100%

100%

98%

100%

98%

55-027-0001

Horicon, WI

97%

97%

97%

100%

98%

97%

100%

17-031-4201

Northbrook, IL

82%

84%

84%

98%

98%

87%

97%

48-201-1039

Deer Park, TX

100%

93%

93%

93%

100%

97%

98%

48-203-0002

Harrison County, TX

100%

98%

98%

95%

97%

98%

100%

29-510-0085

St. Louis, MO

88%

92%

92%

95%

93%

97%

93%

49-011-0004

Bountiful, UT

87%

90%

90%

97%

100%

98%

98%

08-077-0017, 0018

Grand Junction, CO

98%

98%

98%

98%

100%

95%

90%

04-013-9997

Phoenix, AZ

93%

95%

95%

80%

85%

97%

100%

06-085-0005

San Jose, CA

98%

98%

98%

100%

98%

95%

0%a

06-065-8001

Rubidoux, CA

79%

95%

79%

74%

100%

87%

98%

06-037-1103

Los Angeles, CA

79%

97%

79%

75%

97%

97%

98%

41-061-0119

La Grande, OR

50%

97%

97%

92%

95%

98%

98%

41-051-0246

Portland, OR

50%

97%

98%

100%

93%

97%

98%

53-033-0080

Seattle, WA

90%

90%

90%

89%

89%

90%

90%



Mean

90%

95%

94%

93%

96%

95%

96%



Median

97%

97%

97%

95%

97%

97%

98%

Note: Percentage complete values below 85% are specified in red.

a.	Chromium (VI) was not collected at this site - this value was excluded from mean and median calculation.

b.	All acrolein results were invalidated by the site administrator - this value was excluded from mean and median calculation.

9


-------
Table 6. Percentage Completeness Values by NATTS Site and the Seven HAPs of Primary

Importance for CY2012







HAP Abbreviation and Parameter Code(s)



AQS Site Identification Code

Site Name

ACRO

BENZ

BUTA

FORM

NAPH

As

CrVI





43505

45201

43218

43502

17141

82103

12115





43509









85103

14115

25-025-0042

Boston, MA

95%

95%

95%

97%

90%

100%

98%

50-007-0007

Underhill, VT

0%b

95%

95%

98%

95%

97%

100%

44-007-0022

Providence, RI

90%

92%

92%

85%

95%

93%

100%

36-005-0080,0110

Bronx, NY

98%

98%

98%

80%

82%

97%

100%

36-055-1007

Rochester, NY

92%

92%

92%

87%

95%

77%

92%

11-001-0043

Washington, DC

100%

97%

97%

100%

85%

100%

97%

51-087-0014

Richmond, VA

95%

95%

95%

95%

89%

98%

97%

45-025-0001

Chesterfield, SC

97%

97%

97%

95%

85%

95%

82%

13-089-0002

Decatur, GA

100%

100%

100%

90%

95%

93%

89%

21-043-0500

Grayson Lake, KY

97%

98%

98%

100%

93%

97%

100%

12-057-3002

Hillsborough County, FL

95%

95%

95%

98%

93%

92%

95%

12-103-0026

Pinellas County, FL

93%

93%

93%

97%

98%

90%

98%

26-163-0033

Dearborn, MI

97%

97%

97%

98%

95%

100%

97%

55-027-0001

Horicon, WI

82%

82%

82%

79%

97%

93%

97%

17-031-4201

Northbrook, IL

89%

92%

92%

98%

92%

89%

100%

48-201-1039

Deer Park, TX

100%

100%

100%

97%

100%

100%

100%

48-203-0002

Harrison County, TX

100%

98%

98%

97%

92%

97%

97%

29-510-0085

St. Louis, MO

93%

95%

95%

100%

97%

100%

98%

49-011-0004

Bountiful, UT

85%

89%

89%

85%

93%

92%

95%

08-077-0017, 0018

Grand Junction, CO

89%

89%

89%

98%

98%

92%

93%

04-013-9997

Phoenix, AZ

98%

98%

98%

100%

95%

100%

95%

06-085-0005

San Jose, CA

98%

98%

98%

100%

95%

98%

0%a

06-065-8001

Rubidoux, CA

49%

49%

49%

49%

97%

98%

98%

06-037-1103

Los Angeles, CA

48%

48%

48%

48%

93%

98%

97%

41-061-0119

La Grande, OR

69%

90%

90%

90%

93%

95%

98%

41-051-0246

Portland, OR

95%

95%

95%

93%

95%

95%

98%

53-033-0080

Seattle, WA

93%

93%

93%

98%

92%

93%

93%



Mean

90%

91%

91%

91%

93%

95%

96%



Median

95%

95%

95%

97%

95%

97%

97%

Note: Percentage complete values below 85% are specified in red.

a.	Chromium (VI) was not collected at this site - this value was excluded from mean and median calculation.

b.	All acrolein results were invalidated by the site administrator - this value was excluded from mean and median calculation.

10


-------
Figures 1 and 2 present "box and whisker" plots (or "boxplots") of the percentage complete
values presented in Tables 5 and 6, respectively. Thus, they represent a summary of the
distribution of percentage complete values across the 27 NATTS sites for each of the seven
HAPs of primary importance. In these figures, the bottom and top of each "box" represents the
25th and 75th percentiles, respectively; the horizontal line inside the box represents the median
value; and the diamond symbol represents the arithmetic mean. The "whiskers" emanating from
both ends of a box extend to the largest or smallest values, up to a maximum length of 1.5 times
the inter-quartile range (IQR), the distance between the 25th and 75th percentiles of the
distribution of values (i.e., the length of the box). Any values that are more than 1.5 times the
IQR in distance from the box are denoted by open circles. (The sites having percentage
complete values represented by open circles are noted in these plots.) Within both Figure 1 and
Figure 2, the dashed reference line at 85% denotes the NATTS MQO for completeness.

Table 7. Percentage of NATTS Sites Meeting the Completeness MQO for the Seven HAPs
of Primary Importance for CY2011 and CY2012

Calendar Year

VOCs

carbonyls

PAHs

metals



Acrolein3

Benzene

1,3-Butadiene

Formaldehyde

Naphthalene

Arsenic

Chromium (VI)b

2011

77%

93%

85%

78%

96%

96%

100%

2012

81%

89%

89%

78%

89%

96%

96%

a Underhill, VT site excluded from the completeness calculation.
b San Jose, CA site excluded from the completeness calculation.

11


-------
K>

100

95

®
4-i

®
Q.

E

o

O

90

85

I 80

o
_

CD

0_

75

70

65

Portland. OR and
LaGrande. OR
50% Complete

Phoenix, AZ

I

os Angeles, CA

Rubidoui, CA



Jnderhill. VT

ACRO

BENZ

BUTA FORM NAPH

AS

CrVI

HAP

Figure 1. Box and Whisker Plot of Percentage Complete Values for the Seven HAPs of Primary Importance, for CY2011


-------
100

95

90

85

c

® 80-

75-

70

65

a

T

Deer Park, IX

©

5



O Horicon.

O Horicon. •,

'.Vashington. DC _
Chesterfield. SC '

Bronx. MY

Decatur, GA

Chesterfield SC

O Rochester, NY

La Grande, OR

Rubidoux CA
[49% Cosrolete

Rubidoux CA
49% Complete

Rubidoux OA
49% Complete

Los Angeles CA Los Angeles CA Los Angeles CA.
48% Complete 48% Complete 48% Complete

Rubidoux CA
49% Complete

Los Angeles CA
48% Complete

ACRO

BENZ

BUTA

FORM

NAPH

AS

CrVI

HAP

Figure 2. Box and Whisker Plot of Percentage Complete Values for the Seven HAPs of Primary Importance, for CY2012


-------
Data completeness across the entire NATTS network met the MQO in both CY2011 and
CY2012: both the mean and median network-wide completeness for all seven priority HAPs
was greater than 85% in both CY2011 and CY2012. Failures of sites to meet the completeness
MQO were generally more prevalent for VOCs and carbonyls than for other HAP groups for
both CY2011 and CY2012. Some key findings were as follows:

•	Los Angeles, CA, and Rubidoux, CA, did not achieve the MQO for acrolein,
1,3-butadiene, and formaldehyde, in both calendar years, and failed to meet the MQO
for benzene in CY2012. Horicon, WI, did not meet the MQO for these four HAPs in
CY2012, while it did in CY2011. While Northbrook, IL, did not achieve the MQO
for acrolein, benzene, and 1,3-butadiene in CY2011, it did meet the MQO for these
HAPs in CY2012.

•	For both CY2011 and CY2012, chromium (VI) sampling was not conducted at the
San Jose, CA site, and acrolein results were invalidated at the Underhill, VT site.

The percentage of sites meeting the completeness MQO was 85% or greater for benzene,
1,3-butadiene, naphthalene, arsenic, and chromium (VI) in both CY2011 and CY2012. Fewer
sites met the MQO for acrolein and formaldehyde with 77% and 81% meeting the MQO for
acrolein and 78% and 78% for formaldehyde in CY2011 and CY2012, respectively.

2.5 Precision of NATTS Data

Precision of NATTS data was assessed by inspection of results in AQS from replicate anlaysis
and replicate sampling.

The term "replicate sampling" refers to the collection of duplicate and collocated sample
collections, terms that are defined as follows:

Three basic sample types are collected at NATTS sites:

•	Primary sample: a single sample that represents a particular sampling event.

•	Duplicate sample: a replicate sample, collected simultaneously with the primary
sample, that represents a second measurement from the same sample stream (e.g.,
the inlet stream of an outdoor air monitor) but employs an independent sample
collection device (e.g., pump or separate channel) and collection substrate (e.g.,
filter, canister, or cartridge) from the primary sample. Duplicate samples provide
the basis for assessing the aggregate variability associated with the collection
device, sampling substrate, and sample analysis.

•	Collocated sample: a replicate sample, collected simultaneously with the primary
sample, that represents a second measurement from a completely independent (but
spatially close, usually 1 to 2 meters away from the primary sampler) sample
stream, collection device, and collection substrate from the primary sample.
Collocated samples provide the basis for assessing the total variability associated
with all components of the sample collection and analysis scheme. One may
assume that the atmosphere sampled by the primary and collocated samplers is
identical in its composition.

14


-------
The above sample types are differentiated within the AQS database by POC. Tables 8 and 9
provide the POCs encountered in the AQS database for each site by HAP class, for CY2011 and
CY2012, respectively.

Precision assessments associated with replicate sampling are distinctly different from those
associated with replicate analyses in the following way:

•	Precision assessments associated with replicate analyses are derived from a second
chemical analysis of a single sample, be that a primary, duplicate, or collocated
sample.

•	Precision assessments associated with replicate sampling are derived from
independent chemical analyses of two different sample substrates (filter, canister,
etc).

The precision for the NATTS data was assessed from both analytical (i.e., instrumental) and
overall (i.e., analytical plus sampling) perspectives:

•	Analytical precision (Section 2.5.1) measures the variability in reported results due
exclusively to differences in laboratory analytical performance and is assessed by
comparing results from two analyses of a single sample, whether that sample be a
primary, duplicate, or collocated sample.

•	Overall precision (Section 2.5.2), which accounts for the combined variability
associated with sample collection and laboratory analysis, is assessed by comparing
the results of paired primary and collocated samples or paired primary and duplicate
samples.

15


-------
Table 8. Parameter Occurrence Codes by NATTS Site and HAP Type - CY2011













Parameter Occurrence Codes (POCs)







EPA
Region

Site Name

AQS Site Code

P

VOCs
D

C

Carbonyls
P D

C

PAHs
P C

Metals
P C

Chromium (VI)
P D C

I

Boston, MA

25-025-0042

10

11



! 3

4



6



6

7

6

7



I

Underbill, VT

50-007-0007

1





[ 1





6



3

4

6

7



I

Providence, RI

44-007-0022

2





[ 5



7

6



1

2

6



7

II

Bronx, NY

36-005-0080

2





! 2





6



1

2

6



7

II

Rochester, NY

36-055-1007

2





! 2





6



1



6



7

III

Washington, DC

11-001-0043

4

1

2

1 2





1



1



1



2

III

Richmond, VA

51-087-0014

4

7



1 2



4

6



1



6



7

IV

Chesterfield, SC

45-025-0001

1



2

1 1



2

6



1

2

6

7



IV

Decatur, GA

13-089-0002

1,3



2,4,5

1 2



3

6

7

1

2

6



7

IV
IV

Grayson Lake, KY
Hillsborough Cty,
FL

21-043-0500
12-057-3002

6
1

7



1 1,6
! 6

2,7



6
6

7

1,6
5

2,7
6

6
6



7
7

IV

Pinellas Cty, FL

12-103-0026

1





! 6





6



5



6



7

V

Dearborn, MI

26-163-0033

1



2

[ i



2

1

2

1

9

1



2

V

Horicon, WI

55-027-0001

1



2

[ i



2

1

2

1

2

6



7

V

Northbrook, IL

17-031-4201

6

8



[ 6





6



6



6



7

VI

Deer Park, TX

48-201-1039

2



3

1 3





1

2,6

1



6



7

VI

Harrison Cty, TX

48-203-0002

1





1 1





1



1



6





VII

St. Louis, MO

29-510-0085

6





1 6





6



6

7

6



7

VIII

Bountiful, UT

49-011-0004

6





1 6





6



1

2

6



7

VIII

Grand Junction, CO

08-077-0017/-0018

6





1 6





6



3

4

6



7

IX

Phoenix, AZ

04-013-9997

6



7

' 30



31

3



1



6



7

IX

Los Angeles, CA

06-037-1103

4



5

i 4



5

6



2

3

4



5

IX

Rubidoux, CA

06-065-8001

4



5

i 4



5

6

7

2

4

4



5

IX

San Jose, CA

06-085-0005

3



5

i 3



1

1



1









X

La Grande, OR

41-061-0119

7





i 7





7



7



7





X

Portland, OR

41-051-0246

7



9

i 7



9

7

9

7

9

7



9

X

Seattle, WA

53-033-0080

6



7

i 6



7

6

7

6



6

7



P = primary
D = duplicate
C = collocated

16


-------
Table 9. Parameter Occurrence Codes by NATTS Site and HAP Type - CY2012

EPA
Region

Site Name

AQS Site Code

P

VOCs
D

C

Parameter Occurrence Codes (POCs)

Carbonyls PAHs Metals
P D C P C P C

Chromium
(VI)

P D C

I

Boston, MA

25-025-0042

10

11



3

4



6



6

7

i 6

7

I

Underhill, VT

50-007-0007

1





1





6



3

4

i 6

7

I

Providence, RI

44-007-0022

2





5



7

6



1

2

i 6

7

II

Bronx, NY

36-005-0080/-0110

2





2





6



1

2

i 6

7

II

Rochester, NY

36-055-1007

2





2





6



1



i 6

7

III

Washington, DC

11-001-0043

4

1

2

2





1



1



¦ 1

2

III

Richmond, VA

51-087-0014

4

7



2



4

6



1



i 6

7

IV

Chesterfield, SC

45-025-0001

1



2

1



2

6



1

2

! 6

7

IV

Decatur, GA

13-089-0002

1,3



2,4,5

2



3

6

7

1

2

! 6

7

IV

Grayson Lake, KY

21-043-0500

6

7



1,6

2,7



6



1,6

2,7

!6

7

IV

Hillsborough Cty, FL

12-057-3002

1





6





6

7

5

6

!6

7

IV

Pinellas Cty, FL

12-103-0026

1





6





6



5



!6

7

V

Dearborn, MI

26-163-0033

1



2

1



2

1

2

1

9

[ i

2

V

Horicon, WI

55-027-0001

1



2

1



2

1

2

1

2

[ 6

7

V

Northbrook, IL

17-031-4201

6

8



6





6



6



[ 6

7

VI

Deer Park, TX

48-201-1039

2



3

3





1

2,6

1



1 6

7

VI

Harrison Cty, TX

48-203-0002

1





1





1



1



1 6



VII

St. Louis, MO

29-510-0085

6





6





6



6

7

1 6

7

VIII

Bountiful, UT

49-011-0004

6





6





6



1

2

1 6

7

VIII

Grand Junction, CO

08-077-0017/-0018

6





6





6



3

4

1 6

7

IX

Phoenix, AZ

04-013-9997

6



7

30



31

3



1



' 6

7

IX

Los Angeles, CA

06-037-1103

4



5

4



5

6



2

3

i 4

5

IX

Rubidoux, CA

06-065-8001

4



5

4



5

6

7

2

4

i 4

5

IX

San Jose, CA

06-085-0005

3



5

3



1

1



1







X

La Grande, OR

41-061-0119

7





7





7



7



i 7



X

Portland, OR

41-051-0246

7



9

7



9

7

9

7

9

i 7

9

X

Seattle, WA

53-033-0080

6



7

6



7

6

7

6



i 6

7

P = primary
D = duplicate
C = collocated

17


-------
For the purposes of these precision assessments, the AQS database was queried for two distinct
record types: RP records and RD records. RP records contain data for various types of replicate
samples and analyses associated with a particular sampling date, site, and chemical parameter
(HAP). Different types of replicates are identified by the value of the precision ID variable
(PRECISID) according to the following scheme:

•	PRECISID = 1: Collocated sample data

•	PRECISID = 2: Replicate analysis of a primary sample

•	PRECISID = 3: Replicate analysis of a collocated sample

Analytical precision for this report was computed from the replicate pairs of data contained in
RP records that were coded with either Precision ID 2 or 3. Overall precision was computed
from the replicate pairs of data contained in RP records that were coded with Precision ID 1 and
from paired RD records.

In addition to the replicate records, raw data (AQS RD) transactions provide a second source of
primary and collocated data in AQS. Using the POCs shown for each NATTS site listed in
Tables 8 and 9, it is possible to distinguish among primary, duplicate, and collocated sampling
events. For example, primary samples collected at the Chesterfield, SC, NATTS site are
assigned a POC of 1 for VOCs, carbonyls, and metals, while collocated samples are assigned a
POC of 2. This results in the creation of two distinct records for each sampling event at which a
collocated sample is collected. Duplicate samples are identified with a separate POC. The
assignment of a particular POC is made at the discretion of each NATTS site, thus extensive
effort was required to ensure that the POCs for each site were correctly identified. POCs for
primary, duplicate, and collocated samples of each HAP class for CY2011 and CY2012 were
determined based on POCs at each NATTS collection site in CY2007, CY2008, CY2009, and
CY2010 and discrepancies and/or uncertainties about POC assignments were resolved by direct
contact with NATTS administrators for specific collection sites.

Prior to the beginning of CY2012, ERG contacted sites for which it performs analyses to confirm
whether POCs were being appropriately assigned as collocated or duplicate based on sample
characteristics. This resulted in a number of POC assignment changes, primarily involving
POCs previously designated as duplicate sampling updated to indicate the POC represents
collocated sampling.

Multiple POCs for a given site, HAP, and sample type reflect a number of factors unique to sites
during CY2011 and CY2012, largely assigned for reasons known only to the NATTS site
administrators. Overall precision estimates were computed by comparing primary and collocated
or primary and duplicate results for a particular site, HAP, and sample collection date. To reflect
possible differences in analytical and overall precision based on the magnitude of the
contributing measurements, precision was computed as percent coefficient of variance (CV) for
each site and HAP where both replicate values were equivalent to or exceeded the reported
MDL.

18


-------
Laboratories analyzing samples for NATTS sites in CY2011 and CY2012 are listed in Table 10,
with laboratory identification codes for each laboratory shown in Table 11. Of particular note is
that several laboratories provided analytical chemistry services for multiple NATTS sites.

Table 10. Laboratories Performing Analyses by HAP Type for Each
NATTS Site in CY2011 and CY2012

Site Name

VOCs

Carbonyls

PAHs

Metals

Chromium (VI)

Boston-Roxbury, MA

RIDOH

MADEP

ERG

ERG

ERG

Underhill, VT

ERG

VTDEC

ERG

ERG

ERG

Providence, RI

RIDOH

RIDOH

ERG

RIDOH

ERG

Bronx, NY

NYSDEC

NYSDEC

ERG

RTI

ERG

Rochester, NY

NYSDEC

NYSDEC

ERG

RTI

ERG

Washington, DC

MDE

PAMSL

ERG

WVDEP

ERG

Richmond, VA

VA DCLS

VA DCLS

VA DCLS

VA DCLS

ERG

Chesterfield, SC

SCDHEC

SCDHEC

ERG

SCDHEC

ERG

Decatur, GA

GADNR

GADNR

ERG

GADNR

ERG

Grayson Lake, KY

ERG

ERG

ERG

ERG

ERG

Hillsborough County, FL

PCDEM

ERG

ERG

EPCHC

ERG

Pinellas County, FL

PCDEM

ERG

ERG

EPCHC

ERG

Dearborn, MI

ERG

ERG

ERG

MIDEQ

ERG

Horicon, WI

WSLH

WSLH

WSLH

WSLH

ERG

Northbrook, IL

ERG

ERG

ERG

ERG

ERG

Deer Park, TX

TCEQ

TCEQ

TCEQ

TCEQ

ERG

Harrison County, TX

TCEQ

TCEQ

TCEQ

TCEQ

ERG

St. Louis, MO

ERG

ERG

ERG

ERG

ERG

Bountiful, UT

ERG

ERG

ERG

ERG

ERG

Grand Junction, CO

ERG

ERG

ERG

CDPHE

ERG

Phoenix, AZ

ERG

ERG

ERG

ERG

ERG

San Jose, CAa

BAAQMD

BAAQMD

ERG

ERG

-

Rubidoux, CA

SCAQMD

SCAQMD

ERG

SCAQMD

SCAQMD

Los Angeles, CA

SCAQMD

SCAQMD

ERG

SCAQMD

SCAQMD

La Grande, OR

ODEQ

ODEQ

ODEQ

ODEQ

CHLBNT

Portland, OR

ODEQ

ODEQ

ODEQ

ODEQ

CHLBNT

Seattle, WA

ERG

ERG

ERG

ERG

ERG

aSan Jose does not collect Chromium (VI) for the NATTS program.

19


-------
Table 11. Laboratory Abbreviations and Descriptions for NATTS Laboratories



Laboratory



Laboratory Code(s)

Abbreviation

Laboratory Description

01-01-C,M,V

RIDOH

Rhode Island Department of Health

01-02-C,V

VTDEC

Vermont Department of Environmental Conservation

01-03-C

MADEP

Massachusetts Department of Environmental Protection

02-01-C,V

NYSDEC

New York State Department of Environmental Conservation

03-01-V

MDE

Maryland Department of the Environment

03-01-C

PAMSL

Philadelphia Air Management Services Laboratory

03-01-M

WVDEP

West Virginia Department of Environmental Protection

03-02-C,M,P,R,V

VADCLS

Virginia Division of Consolidated Laboratory Services

04-01-M

EPCHC

Environmental Protection Commission of Hillsborough County

04-01-V

PCDEM

Pinellas County Department of Environmental Management

04-02-C,M,P,V

SCDHEC

South Carolina Department of Health and Environmental Control

04-03-C,M,V

KYDES

Kentucky Division of Environmental Services

04-04-C,M,V

GADNR

Georgia Department of Natural Resources

05-01-M

MIDEQ

Michigan Department of Environmental Quality

05-03-C,M,P,V

WSLH

Wisconsin State Laboratory of Hygiene

06-01-C,M,P,R,V

TCEQ

Texas Commission on Environmental Quality

08-02-M

CDPHE

Colorado Department of Public Health and Environment

09-03-C,V

BAAQMD

Bay Area Air Quality Management District

09-08-C,M,P,R,V

SCAQMD

South Coast Air Quality Management District

10-02-R

CHLBNT

Chester LabNet

10-02-C,M,P,V

ODEQ

Oregon Department of Environmental Quality

11-01-C,M,P,R,V

ERG

Eastern Research Group

11-02-M

RTI

RTI International

2.5.1 Analytical Precision Results

Analytical precision was calculated from the replicate analysis results associated with either a
primary, collocated, or duplicate sample as extracted from RP records from the AQS database.
For this calculation, the two analysis results for a given sample are distinguished by referring to
one as the "principal" result and the other as the "replicate" result. The measure for analytical
precision, expressed as the percentage coefficient of variation (%CV), is defined in Eq. 1:

%CV = 100-

(Eq. 1)

20


-------
where

pi = the principal result for sample i,
n = the replicate result for sample i, and

n = the number of samples having primary-replicate result pairs.

Analytical precision was calculated only when pi > MDL and n > MDL. For those sites that did
not report MDLs into AQS, it could not be determined if the RP records exceeded the
corresponding MDLs. As a result, such data were excluded from the analytical precision
calculation.

The analytical precision for each of the 27 NATTS HAPs is presented in Table 12 and Table 13
for CY2011 and CY2012, respectively. For the seven HAPs of primary importance, analytical
precision is summarized graphically in Figures 3 through 9 for CY2011 and Figures 10 through
16 for CY2012.

For CY2011 the network mean analytical precision met the MQO of 15% for carbonyls, PAHs,
and chromium (VI), for all metals except beryllium, and for 9 of the 16 VOCs reporting
concentrations above MDLs. Analytical precision data for VOCs show some variability with no
discernible trend noted among sites or HAPs. Records for 1,2-dibromoethane and
1,2-dichloropropane did not include replicate pairs for which both results were above their
respective MDL, and are not included in Table 12. For all sites reporting metals above MDLs,
only Boston, MA met the precision MQO for all metals.

For CY2012 the network mean analytical precision met the MQO for all HAPs except for
acrylonitrile and beryllium. Moreover, all sites met the MQO for carbonyls, PAHs, and
chromium (VI). Among VOCs, sites showed close agreement with only an occasional MQO
exceedence. Records for 1,2-dibromoethane, 1,2-dichloropropane, and 1,1,2,2-tetrachloroethane
did not include replicate pairs for which both results were above their respective MDL, and are
not included in Table 13. Replicate analysis showed similar close agreement for metals, with
only two sites having analytical precision exceeding the MQO.

Overall precision comprises analytical variability and sampling variability and more fully
characterizes network-wide precision. Network achievement of the precision MQO is discussed
in Section 2.5.2.

21


-------
Table 12. Analytical Precision for Replicate Analyses > MDL - CY2011

IO
IO

AQS Site Code

Site Name













VOCs













BENZ

BUTA

CTET

CLFRM

EDC

MECL

TCE1122

PERC

TCE

VC

cDCPEN

tDCPEN ACRO

ACRY

25-025-0042

Boston, MA

-

-

-

-

-

-

-

-

-

-

-

-

-

50-007-0007

Underbill, VT

-

-

-

-

-

-

-

-

-

-

-

-

-

44-007-0022

Providence, RI

-

-

-

-

-

-

-

-

-

-

-

-

-

36-005-0080

Bronx, NY

-

-

-

-

-

-

-

-

-

-

-

-

-

36-055-1007

Rochester, NY

-

-

-

-

-

-

-

-

-

-

-

-

-

11-001-0043

Washington, DC

-

-

-

-

-

-

-

-

-

-

-

-

-

45-025-0001

Chesterfield, SC

-

-

-

-

-

-

-

-

-

-

-

-

-

13-089-0002

Decatur, GA

-

-

-

-

-

-

-

-

-

-

-

-

-

21-043-0500

Grayson Lake, KY

6.7 (13)

6.4 (6)

6.6 (13)

6.8 (3)

6.2 (2)

4.7 (13)

-

-

-

20.2 (1)

-

4.5 (13)

4.4 (5)

12-057-3002

Hillsborough Cty, FL

4.0 (2)

21.1(1)

0.6 (2)

6.3 (2)

4.3 (2)

4.0 (2)

-

20.2 (1)

-

-

0.5 (1)

0.8 (1) 2.8 (2)

-

12-103-0026

Pinellas Cty, FL

4.4 (54)

14.1 (41)

3.5 (54)

8.5 (54)

12.7 (45)

10.3 (54)

36.8 (5)

14.3 (51)

93.4(1)

-

0(1)

22.5 (3) 10.9 (53)

9.7 (41)

26-163-0033

Dearborn, MI

6.5 (12)

7.6 (12)

6.3 (12)

6.0 (12)

5.6 (2)

4.8 (12)

-

5.2 (9)

-

-

-

6.2 (12)

-

55-027-0001

Horicon, WI

-

-

-

-

-

-

-

-

-

-

-

-

-

17-031-4201

Northbrook, IL

12.0 (24)

6.2 (12)

17.0(12)

25.4(12)

10.2 (5)

21.4(12)

-

5.0 (10)

-

-

-

21.8(12)

6.9 (4)

48-201-1039

Deer Park, TX

-

-

-

-

-

-

-

-

-

-

-

-

-

29-510-0085

St. Louis, MO

7.1 (16)

8.4 (16)

6.8 (16)

6.4 (14)

5.4 (8)

9.0(16)

-

5.5 (10)

-

-

-

26.3 (16)

-

49-011-0004

Bountiful, UT

12.6(13)

8.9 (13)

6.3 (13)

14.3 (5)

8.8 (5)

10.5 (13)

-

9.1 (4)

-

-

-

39.3 (13)

-

08-077-0017/-0018

Grand Junction, CO

4.5 (12)

5.5 (10)

17.1 (11)

6.8 (9)

2.0 (2)

16.8 (12)

-

7.2 (8)

4.3 (2)

-

-

14.4(11)

4.6 (2)

04-013-9997

Phoenix, AZ

6.6(12)

5.2 (12)

6.7 (12)

4.9 (12)

5.3 (2)

7.0(12)

-

4.8 (12)

-

-

-

5.5 (12)

2.8 (2)

06-037-1103

Los Angeles, CA

-

-

-

-

-

-

-

-

-

-

-

-

-

06-065-8001

Rubidoux, CA

-

-

-

-

-

-

-

-

-

-

-

-

-

06-085-0005

San Jose, CA

3.1 (31)

10.4(11)

7.1 (31)

15.9 (27)

-

7.9 (29)

-

2.4 (31)

0(4)

-

-

-

-

53-033-0080

Seattle, WA

3.1 (12)

7.4 (12)

4.6 (12)

8.4 (9)

2.9 (2)

3.2 (12)

-

3.8 (2)

-

-

-

3.6 (12)

8.0(1)



Network Mean

6.9 (201)

10.0(146)

8.0 (188)

11.8 (159)

10.7 (75)

10.4(187)

36.8 (5)

9.7 (138)

35.4 (7)

Ji > J . 1 i

0.3 (2)

19.5 (4) 17.3 (156)

8.8 (55)

Analytical precision is expressed as percentage coefficient of variation (%CV) with number of contributing data pairs (/;) shown in parentheses.
Values shown in red exceed the MQO of < 15% CV.


-------
Table 12. Analytical Precision for Replicate Analyses > MDL - CY2011 (continued)

IO

oo

AQS Site Code

Site Name

carbonyls

PAHs





metals







FORM

ACET

NAPH

BaP

As

Be

Cd Pb

Mn

Ni

CrVI

25-025-0042

Boston, MA

-

--

-

-

2.1 (74)

11.0(18)

6.8 (74) 1.5 (74)

1.1 (74)

1.8(72)

8.3 (10)

50-007-0007

Underhill, VT

0.7 (4)

1.7 (4)

--

-

16.4 (8)

--

5.8(9) 1.5(9)

3.4 (9)

2.9 (2)

7.7 (7)

44-007-0022

Providence, RI

-

-

-

-

-

-

-

-

-

7.2 (10)

36-005-0080

Bronx, NY

-

-

-

-

-

-

-

-

-

5.5 (12)

36-055-1007

Rochester, NY

-

-

-

-

-

-

-

-

-

9.4 (8)

11-001-0043

Washington, DC

-

--

--

-

--

-

--

-

-

8.3 (11)

45-025-0001

Chesterfield, SC

-

-

-

-

-

-

-

-

-

7.3 (4)

13-089-0002

Decatur, GA

-

-

4.1 (12)

3.8 (2)

-

-

-

-

-

5.4(11)

21-043-0500

Grayson Lake, KY

0.7 (8)

0.9 (8)

--

-

11.2 (64)

33.3 (2)

5.0(78) 1.0(80)

0.8 (80)

33.4 (4)

5.3 (9)

12-057-3002

Hillsborough Cty, FL

2.3 (14)

1.6(14)

3.3 (12)

7.4 (3)

-

--

-

--

--

5.3 (12)

12-103-0026

Pinellas Cty, FL

2.2(12)

2.0(12)

-

--

-

--

--

--

--

5.7(10)

26-163-0033

Dearborn, MI

1.3 (12)

0.3 (12)

2.9(12)

4.4 (8)

-

--

-

--

--

4.6(12)

55-027-0001

Horicon, WI

-

-

-

-

-

-

-

-

-

3.6(10)

17-031-4201

Northbrook, IL

2.0(12)

1.7(12)

-

-

-

-

-

-

-

6.4(12)

48-201-1039

Deer Park, TX

-

-

-

-

-

-

-

-

-

5.2(12)

29-510-0085

St. Louis, MO

2.1 (12)

1.0(12)

-

--

7.2 (66)

21.0 (8)

2.8 (67) 0.6 (67)

1.2 (67)

9.6(17)

6.6(11)

49-011-0004

Bountiful, UT

4.4 (14)

1.6(14)

-

--

-

--

--

--

--

13.8(12)

08-077-0017/ -0018

Grand Junction, CO

2.2(12)

2.0(12)

-

-

-

-

-

-

-

8.0(12)

04-013-9997

Phoenix, AZ

0.9 (14)

1.2(14)

-

--

-

--

--

--

--

4.9 (14)

06-037-1103

Los Angeles, CA

-

-

--

-

5.7 (5)

-

21.1(5) 4.2(5)

21.1 (5)

3.3 (5)

--

06-065-8001

Rubidoux, CA

-

--

2.6(12)

--

8.0 (9)

--

34.8 (4) 7.2 (9)

0(5)

3.1 (9)

--

06-085-0005

San Jose, CA

0.9 (9)

1.0 (9)

-

-

-

-

-

-

-

-

53-033-0080

Seattle, WA

0.8(12)

1.3 (12)

2.5 (10)

6.3 (2)

-

--

-

--

--

5.6(14)



Network Mean

2.1 (135) 1.4(135) 3.1 (58) 5.3 (15) 8.1 (226) 16.8 (28) 7.5 (237) 1.9 (244) 3.3 (240) 7.7 (109) 7.0(213)

Analytical precision is expressed as percentage coefficient of variation (%CV) with number of contributing data pairs (/;) shown in parentheses.
Values shown in red exceed the MQO of < 15% CV.


-------
Table 13. Analytical Precision for Replicate Analyses > MDL - CY2012

IO

AQS Site Code

Site Name













VOCs













BENZ

BUTA

CTET

CLFRM

EDC

MECL

PERC

TCE

VC

cDCPEN

tDCPEN

ACRO

ACRY

25-025-0042

Boston, MA

-

-

-

-

-

-

-

-

-

-

-

-

-

50-007-0007

Underbill, VT

-

-

-

-

-

-

-

-

-

-

-

-

-

44-007-0022

Providence, RI

-

-

-

-

-

-

-

-

-

-

-

-

-

36-005-0080/-0110

Bronx, NY

2.5 (9)

34.1 (9)

1.3 (9)

8.1 (9)

3.6 (9)

29.6 (9)

34.5 (9)

11.4(5)

11.2 (3)

-

-

15.8 (9)

-

36-055-1007

Rochester, NY

-

-

-

-

-

-

-

-

-

-

-

-

-

11-001-0043

Washington, DC

-

-

-

-

-

-

-

-

-

-

-

-

-

51-087-0014

Richmond, VA

-

-

-

-

-

-

-

-

-

-

-

-

-

45-025-0001

Chesterfield, SC

-

-

-

-

-

-

-

-

-

-

-

-

-

13-089-0002

Decatur, GA

-

-

-

-

-

-

-

-

-

-

-

-

-

21-043-0500

Grayson Lake, KY

5.5 (8)

13.2 (8)

6.1 (10)

18.5 (2)

8.4 (6)

25.6(10)

-

-

-

-

-

10.0(10)

73.4 (8)

12-057-3002

Hillsborough Cty, FL

4.4 (2)

6.5 (2)

1.2 (2)

2.3 (2)

0(2)

2.4 (2)

15.4(2)

-

-

-

-

9.0 (2)

-

12-103-0026

Pinellas Cty, FL

3.1 (18)

8.7 (17)

1.7 (18)

7.2 (18)

2.5 (18)

2.8 (18)

9.0(18)

-

-

1.9 (2)

2.4 (2)

7.1 (18)

14.3 (14)

26-163-0033

Dearborn, MI

7.2 (12)

5.4 (12)

5.9 (12)

20.8 (12)

5.7 (8)

5.2 (12)

4.7 (6)

-

-

-

-

10.1 (12)

-

55-027-0001

Horicon, WI

-

-

-

-

-

-

-

-

-

-

-

-

-

17-031-4201

Northbrook, IL

10.2 (28)

9.8 (15)

10.7 (14)

15.0(14)

12.2 (10)

11.1 (14)

10.6 (6)

16.0 (4)

-

-

-

30.8 (12)

-

48-201-1039

Deer Park, TX

-

-

-

-

-

-

-

-

-

-

-

-

-

48-203-0002

Harrison Cty, TX

-

-

-

-

-

-

-

-

-

-

-

-

-

29-510-0085

St. Louis, MO

7.9 (16)

7.1 (16)

15.2 (16)

8.3 (14)

12.4(10)

7.4 (16)

11.0(8)

6.2 (2)

-

-

-

11.7 (16)

-

49-011-0004

Bountiful, UT

10.2 (13)

9.3 (13)

12.0(13)

11.7 (13)

14.3 (6)

19.9(13)

7.7 (2)

-

-

-

-

8.5 (13)

-

08-077-0017/-0018

Grand Junction, CO

4.9 (12)

5.5 (12)

10.8 (12)

6.1 (8)

11.0(10)

6.5 (12)

8.3 (10)

-

-

-

-

16.9 (12)

11.1(2)

04-013-9997

Phoenix, AZ

4.0 (12)

4.1 (12)

4.5 (12)

4.2 (11)

7.1 (8)

4.0 (12)

4.3 (12)

-

-

3.8(1)

-

5.0(12)

-

06-037-1103

Los Angeles, CA

-

-

-

-

-

-

-

-

-

-

-

-

-

06-065-8001

Rubidoux, CA

-

-

-

-

-

-

-

-

-

-

-

-

-

06-085-0005

San Jose, CA

-

-

-

-

-

-

-

-

-

-

-

-

-

53-033-0080

Seattle, WA

7.6 (13)

5.6 (13)

7.5 (13)

9.0(13)

5.5 (5)

15.4(13)

3.7 (2)

—

—

—

—

15.0(13)

—



Network Mean

7.3 (130)

11.8 (116)

8.9 (118)

11.3 (103)

8.7 (87)

14.2 (118)

14.4 (73)

12.6(11)

11.2 (3)

2.7 (3)

2.4 (2)

14.4(116)

43.9 (24)

Analytical precision is expressed as percentage coefficient of variation (%CV) with number of contributing data pairs (/;) shown in parentheses.
Values shown in red exceed the MQO of < 15% CV.


-------
Table 13. Analytical Precision for Replicate Analyses > MDL - CY2012 (continued)

AQS Site Code

Site Name

carbonyls

PAHs



metals







FORM

ACET

NAPH

BaP

As Be

Cd

Pb

Mn

Ni

CrVI

25-025-0042

Boston, MA

-

-

3.5(1)

1.9(1)

2.6 (66) 8.8 (9)

5.7 (66)

1.4 (66)

1.5 (66)

3.5 (66)

5.0(12)

50-007-0007

Underhill, VT

0.6 (9)

0.8 (9)

0.7(1)

--

28.6 (6)

20.5 (11)

0.7 (6)

1.0(12)

1.2 (4)

7.3(1)

44-007-0022

Providence, RI

-

-

3.2 (3)

4.4(1)

-

-

-

-

-

7.5 (9)

36-005-0080/-0110

Bronx, NY

-

-

6.1 (4)

2.6(1)

-

-

-

-

-

8.3 (13)

36-055-1007

Rochester, NY

-

-

1.9 (3)

2.0(1)

-

-

-

-

-

4.0 (5)

11-001-0043

Washington, DC

-

-

0.6(1)

-

--

--

-

--

-

7.1 (12)

51-087-0014

Richmond, VA

-

-

0.5 (3)

10.1(1)

-

-

-

-

-

8.0(12)

45-025-0001

Chesterfield, SC

-

-

2.5(1)

-

-

-

-

-

-

8.0 (3)

13-089-0002

Decatur, GA

-

-

2.7(11)

3.1 (2)

-

-

-

-

-

5.6(10)

21-043-0500

Grayson Lake, KY

1.7(12)

1.5(12)

2.7 (4)

2.0 (2)

13.7 (40)

9.7 (46)

0.7 (49)

1.3 (49)

1.7 (6)

12.0 (7)

12-057-3002

Hillsborough Cty, FL

9.0(10)

2.9 (10)

3.4(13)

--

-

-

-

--

-

5.6 (7)

12-103-0026

Pinellas Cty, FL

4.7(12)

3.1 (12)

0.9 (4)

3.7(1)

-

-

-

--

-

9.3 (13)

26-163-0033

Dearborn, MI

1.3 (12)

0.6(12)

2.1(15)

4.1(11)

-

-

-

--

-

4.8(14)

55-027-0001

Horicon, WI

-

-

-

-

-

-

-

-

-

10.0 (8)

17-031-4201

Northbrook, IL

3.3 (16)

2.3 (16)

-

-

1.5 (6) 10.6 (5)

1.6(6)

2.4 (6)

1.3 (6)

3.0 (6)

5.8(14)

48-201-1039

Deer Park, TX

-

-

-

-

-

-

-

-

-

3.5 (12)

48-203-0002

Harrison Cty, TX

-

-

-

-

-

-

-

-

-

4.3 (3)

29-510-0085

St. Louis, MO

1.6(12)

1.2(12)

4.8 (6)

7.4 (6)

9.2(117) 18.1 (21)

4.5 (118)

0.5 (48)

0.9(118) 10.4(107)

6.1 (12)

49-011-0004

Bountiful, UT

2.8(10)

2.7 (10)

11.0(1)

-

-

-

-

-

-

8.8(12)

08-077-0017/-0018

Grand Junction, CO

0.8(12)

0.8(12)

-

-

-

-

-

-

-

10.7 (7)

04-013-9997

Phoenix, AZ

0.9(12)

0.5 (12)

1.0 (2)

--

14.3 (4) 13.0 (4)

6.8 (4)

0.3(1)

2.4 (4)

4.0 (4)

4.3 (12)

06-037-1103

Los Angeles, CA

-

-

11.4(1)

--

-

-

-

--

-

-

06-065-8001

Rubidoux, CA

-

-

12.8 (16)

4.9 (2)

-

-

-

-

-

-

06-085-0005

San Jose, CA

0.3 (9)

42.7 (10)

1.2 (2)

--

4.6 (2)

7.7 (4)

0.5 (2)

0.7 (6)

1.4 (6)

--

53-033-0080

Seattle, WA

1.7(12)

1.5(12)

2.3 (17)

3.1 (2)

1.2(3)

3.3 (3)

1.7 (3)

1.8(3)

1.2 (3)

4.7 (12)



Network Mean

3.3 (126) 11.6 (127)

5.8 (92)

4.9 (29) 9.8 (241) 15.0 (39) 7.3 (255) 1.1 (178) 1.2 (261)

7.9 (199)

7.1 (198)

Analytical precision is expressed as percentage coefficient of variation (%CV) with number of contributing data pairs (/;) shown in parentheses.
Values shown in red exceed the MQO of < 15% CV.


-------
2.5.2 Overall Precision Results

Overall precision was calculated using the (principal) results of the primary sample paired with
either the duplicate or collocated samples in the AQS database. This measure of agreement,
expressed as the % CV, is defined in Eq. 2:

pi = the result of the principal analysis performed on the primary sample within the ith
pair,

n = the result of the principal analysis performed on either the collocated or duplicate
sample within the ith pair, and

n = the number of primary-collocated and primary-duplicate sample pairs.

Overall precision was calculated only when pi > MDL and n > MDL. For those sites that did not
report MDLs into AQS, it could not be determined if the records exceeded the corresponding
MDLs. As a result, such data were excluded from the overall precision calculation.

In order to ensure all precision records were evaluated, both the RP and RD data were extracted
for precision records. The precision calculation algorithm was designed to ensure that records
that appeared both in RP and RD transactions were not represented twice in the analysis of
overall precision. Approximately half of the pairs entered into AQS for overall precision
consisted of values above the MDL for CY2011 and CY2012. Overall precision for each of the
27 NATTS HAPs is presented in Table 15 for CY2011 and in Table 16 for CY2012. For the
seven HAPs of primary importance, overall precision is presented graphically in Figures 3
through 9 for CY2011 and Figures 10 through 16 for CY2012.

As is expected given the additional variability contribution of sample collection, overall
precision for CY2011 showed much greater variability than the analytical precision: the network
mean overall precision met the MQO for carbonyls, 1 PAH, 1 metal, and 5 of 16 VOCs; the
MQO was not met for chromium (VI). Only the two carbonyl compounds met the MQO of 15%
for all sites. Those VOCs that exceeded the MQO generally showed CVs of 25% or greater.
Precision data were not available for 1,2-dichloropropane, vinyl chloride, and
cis-l,3-dichloropropene and these are not included in Table 15.

As in CY2011, CY2012 overall precision showed greater variability than CY2012 analytical
precision. The network mean overall precision met the MQO for carbonyls, 1 PAH, and 5 of 16
VOCs; the MQO was not met for any of the metals or for chromium (VI). All sites achieved the
MQO for carbonyls except for Providence, RI and LaGrande, OR; for PAHs, all sites met the
MQO except for Decatur, GA, for naphthalene, which appeared to weight the network mean
overall precision to exceed the MQO. Only Dearborn, MI, and Bountiful, UT, met the MQO for
all metals and chromium (VI). Only Pinellas County, FL, met the MQO for overall precision for

%CV = 100-"I

f (Pi-n)

(Eq. 2)

2 n

where

26


-------
all VOCs measured above the MDL. Precision data were not available for 1,2-dibromoethane
and 1,2-dichloropropane and these are not included in Table 16.

As can be seen in Figures 3 through 16, the aggregate precision associated with sample
collection and analysis varies substantially by collection site and HAP when compared to the
precision associated with analytical variability alone for both CY2011 and CY2012. Although
some of this variability may be attributable to one or more extreme values, substantial effort
would be needed to determine the extent of this impact. The fact that many sites exhibit
percentage CVs above the MQO points to a collection methodology contribution to the overall
variability, particularly for metals and VOCs.

Overall precision data analysis was limited to the number of sites reporting precision sample
pairs and corresponding MDL values into AQS. A breakdown of total sites evaluated for overall
precision is included in Table 14. The number of sites reporting precision samples with
corresponding MDLs ranged from 8 (PAHs) to 24 (chromium (VI)) in both CY2011 and
CY2012. In CY2011, all sites met the precision MQO for formaldehyde and less than 85% of
sites met the MQO for six of the remaining seven HAPs of primary importance, with less than
half of sites meeting the MQO for acrolein and arsenic. In CY2012, more than 85% of sites met
the precision MQO for benzene, formaldehyde, and naphthalene with the remaining four HAPs
of primary importance showing 84% or less of sites meeting the MQO. As in CY2011, less than
50% of sites met the precision MQO for acrolein in CY2012.

Table 14. Percentage of NATTS Sites Meeting the MQO for Overall Precision -

CY2011 and CY2012







VOCs



carbonyls

PAHs

metals











1,3-







Chromium

Metric

CY

Acrolein

Benzene

Butadiene

Formaldehyde

Naphthalene

Arsenic

(VI)

Number sites

2011

19

19

19

21

8

17

24

reporting precision
values with MDLs

2012

19

20

19

19

8

16

24

Number of sites

2011

7

16

15

21

5

8

13

meeting the
precision MQO

2012

9

18

16

18

7

9

15

Percentage of sites

2011

37%

84%

79%

100%

63%

47%

54%

meeting precision
MQO

2012

47%

90%

84%

95%

88%

56%

63%

27


-------
Table 15. Overall Precision for Primary, Duplicate, and Collocated Samples > MDL - CY2011

AQS Site Code

Site Name













VOCs











BENZ

BUTA

CTET

CLFRM

EDB

EDC

MECL

TCE1122

PERC

TCE

tDCPEN

ACRO

ACRY

25-025-0042

Boston, MA

3.4 (31)

23.7 (30)

2.9 (31)

4.9 (31)

-

7.6(19)

14.1 (31)

-

3.5 (20)

6.7(1)

-

13.3 (27)

-

50-007-0007

Underhill, VT

-

-

-

-

-

-

-

-

-

-

-

-

-

44-007-0022

Providence, RI

-

-

-

-

-

-

-

-

-

-

-

-

-

36-005-0080

Bronx, NY

-

-

-

-

-

-

-

-

-

-

-

-

-

36-055-1007

Rochester, NY

-

-

-

-

-

-

-

-

-

-

-

-

-

11-001-0043

Washington, DC

--

75.4 (3)

6.1 (27)

10.4 (25)

-

--

31.3 (27)

-

32.0(13)

--

-

-

43.9 (5)

51-087-0014

Richmond, VA

8.1 (26)

--

5.8 (26)

--

-

--

18.3 (24)

-

10.9(1)

--

-

17.8(10)

--

45-025-0001

Chesterfield, SC

8.5 (61)

--

3.5 (57)

--

0(1)

--

65.5 (57)

0(7)

14.7(16)

0(1)

-

26.5 (49)

--

13-089-0002

Decatur, GA

25.3 (23)

-

15.0(13)

-

--

-

55.6 (7)

-

12.9(1)

--

-

-

-

21-043-0500

Grayson Lake, KY

6.1 (6)

20.1 (3)

4.3 (6)

4.8 (2)

-

4.0(1)

43.5 (6)

-

-

--

-

43.1 (6)

81.9 (2)

12-057-3002

Hillsborough Cty, FL

--

--

--

--

-

--

--

--

--

-

--

--

-

12-103-0026

Pinellas Cty, FL

9.2(11)

11.9 (8)

3.5 (11)

10.2 (11)

-

18.2(10)

23.0(11)

-

20.0(11)

--

30.3 (1)

18.0(11)

52.3 (1)

26-163-0033

Dearborn, MI

4.2 (6)

6.7 (6)

7.9 (6)

34.7 (6)

--

3.6(1)

16.6 (6)

-

5.6 (4)

--

-

10.7 (6)

--

55-027-0001

Horicon, WI

0.3 (2)

-

-

-

-

-

-

-

-

-

-

12.2 (3)

-

17-031-4201

Northbrook, IL

--

--

20.7 (6)

34.5 (6)

--

5.0 (2)

30.0 (6)

-

5.2 (5)

--

-

30.5 (6)

13.6 (3)

48-201-1039

Deer Park, TX

-

-

-

-

-

-

-

-

-

-

-

-

-

29-510-0085

St. Louis, MO

5.8(8)

5.3 (8)

6.5 (8)

3.7 (6)

-

8.9 (4)

10.9 (8)

--

11.7 (5)

-

--

40.5 (8)

--

49-011-0004

Bountiful, UT

17.9 (6)

10.6 (6)

6.5 (6)

3.6 (2)

-

10.3 (3)

13.3 (6)

-

6.7 (2)

-

-

59.9 (6)

-

08-077-0017/-0018

Grand Junction, CO

5.8 (6)

5.5 (5)

27.3 (5)

7.1 (4)

--

0(1)

21.3 (6)

-

6.9 (4)

1.2(1)

-

19.2 (5)

1.9(1)

04-013-9997

Phoenix, AZ

7.0 (6)

4.1 (6)

8.1 (6)

5.1 (6)

-

6.7(1)

58.4 (6)

-

5.5 (6)

--

-

44.9 (6)

--

06-037-1103

Los Angeles, CA

--

--

--

--

-

--

--

--

--

-

--

--

-

06-065-8001

Rubidoux, CA

-

-

-

-

-

-

-

-

-

-

-

-

-

06-085-0005

San Jose, CA

14.7 (29)

46.9 (7)

15.0 (29)

29.1 (20)

--

-

31.6 (24)

-

42.8 (29)

--

-

66.6 (22)

--

41-051-0246

Portland, OR

62.1 (42)

-

14.0 (21)

-

--

-

29.3 (9)

-

70.7 (1)

--

-

36.8 (22)

--

53-033-0080

Seattle, WA

4.6 (6)

5.0 (6)

13.1 (6)

16.6 (4)

—

11.5(1)

68.5 (6)

—

0.9(1)

—

—

23.9 (6)

—



Network Mean

26.8 (263)

24.6 (82) 9.8 (258)

17.5 (119)

0(1)

11.1 (42)

41.6 (234)

0(7)

26.1 (118)

3.9 (3)

30.3 (1)

.5 (187)

46.9 (12)

Overall precision is expressed as percentage coefficient of variation (%CV) with number of contributing data pairs shown in parentheses.
Values shown in red exceed the MQO of < 15% CV.


-------
Table 15. Overall Precision for Primary, Duplicate, and Collocated Samples > MDL - CY2011 (continued)





carbonyls

PAHs





metals







AQS Site Code

Site Name

FORM

ACET

NAPH

BaP

As

Be

Cd

Pb

Mn

Ni

CrVI

25-025-0042

Boston, MA

12.5 (30)

13.2 (30)

-

-

4.1 (37)

20.4 (6)

18.1 (37)

6.4 (37)

3.7 (37)

4.9 (36)

27.6 (5)

50-007-0007

Underhill, VT

-

-

-

-

23.2 (3)

-

16.2 (4)

4.6 (4)

4.2 (4)

0(1)

15.5 (3)

44-007-0022

Providence, RI

9.6 (23)

10.3 (23)

--

-

16.0 (20)

--

--

14.0 (23)

13.9 (27)

37.2 (26)

17.1 (5)

36-005-0080

Bronx, NY

--

--

--

-

6.7 (51)

-

6.2 (47)

2.8(51)

4.4 (51)

4.7 (51)

8.3 (6)

36-055-1007

Rochester, NY

-

-

-

-

-

-

-

-

-

-

19.6 (5)

11-001-0043

Washington, DC

--

--

--

-

--

-

-

-

--

-

10.3 (5)

51-087-0014

Richmond, VA

2.5 (59)

2.6 (59)

--

-

--

--

--

-

-

--

11.5 (5)

45-025-0001

Chesterfield, SC

14.6 (58)

14.7 (58)

--

--

35.3 (92)

--

37.9 (98)

37.5 (98)

32.7 (100)

57.7 (14)

8.9 (2)

13-089-0002

Decatur, GA

--

--

6.4 (6)

5.2(1)

20.0(18)

53.3 (1)

--

14.2 (22)

14.0 (22)

23.4 (22)

16.8 (5)

21-043-0500

Grayson Lake, KY

6.6(16)

6.5 (16)

--

-

20.3 (39)

0(1)

20.6 (38)

15.1 (55)

26.7 (51)

35.6 (3)

20.4 (5)

12-057-3002

Hillsborough Cty, FL

2.8 (7)

2.1 (7)

15.3 (6)

13.8 (1)

21.9 (35)

--

8.6 (9)

9.8 (36)

10.3 (59)

24.5 (53)

6.0 (4)

12-103-0026

Pinellas Cty, FL

2.6 (6)

2.6 (6)

-

-

-

-

-

-

-

-

10.8 (5)

26-163-0033

Dearborn, MI

11.7 (5)

10.1 (5)

8.6 (6)

4.4 (4)

8.4 (59)

8.7 (29)

26.0 (58)

--

26.0 (58)

33.4 (58)

10.9 (6)

55-027-0001

Horicon, WI

10.5 (4)

10.0 (4)

11.2 (3)

5.8 (2)

24.1 (5)

16.7(1)

30.2 (5)

21.6 (5)

15.6 (5)

18.3 (5)

9.1 (5)

17-031-4201

Northbrook, IL

2.9 (6)

2.2 (6)

-

-

-

-

-

-

-

-

18.2 (6)

48-201-1039

Deer Park, TX

--

--

--

-

5.3 (28)

-

12.0 (8)

14.3 (28)

4.6 (28)

8.0 (28)

11.0 (6)

29-510-0085

St. Louis, MO

3.2 (6)

1.4 (6)

--

-

11.8(32)

19.7 (3)

8.6 (33)

5.1 (33)

5.5 (33)

43.2 (7)

7.9 (6)

49-011-0004

Bountiful, UT

5.6 (7)

2.3 (7)

--

-

11.0 (3)

11.1(2)

15.7 (4)

13.1 (4)

10.4 (4)

3.1(1)

22.3 (6)

08-077-0017/-0018

Grand Junction, CO

2.2 (6)

2.1 (6)

-

-

43.2(11)

--

20.2 (3)

6.3 (10)

6.1(11)

58.4(11)

10.9 (6)

04-013-9997

Phoenix, AZ

5.8(6)

2.6 (6)

-

-

-

-

-

-

-

-

31.2 (7)

06-037-1103

Los Angeles, CA

--

--

--

-

17.1 (6)

--

6.3 (3)

7.3 (6)

0(6)

3.8 (6)

35.3 (6)

06-065-8001

Rubidoux, CA

-

-

15.5 (6)

-

12.6 (5)

-

60.1 (2)

11.4 (5)

13.2 (5)

47.9 (5)

34.7 (3)

06-085-0005

San Jose, CA

-

-

-

-

-

-

-

-

-

-

-

41-051-0246

Portland, OR

11.0 (46)

10.8 (46)

22.7 (40)

4.4 (5)

4.3 (44)

6.6 (2)

12.2 (33)

7.0 (43)

14.0 (44)

4.4 (37)

11.9 (3)

53-033-0080

Seattle, WA

4.0 (6)

1.8(6)

6.0 (5)

12.7(1)

—

—

—

—

—

—

11.0 (7)



Network Mean

9.7 (285)

9.8 (285)

18.5 (67)

6.6(13)

20.3 (488)

14.4 (45) 24.7 (382)

19.6 (460)

19.8 (545)

26.9 (364)

18.3 (115)

Expressed as percentage coefficient of variation (%CV) with number of contributing data pairs presented in parentheses.
Values shown in red exceed the MQO of < 15% CV.


-------
Table 16. Overall Precision for Primary, Duplicate, and Collocated Samples > MDL - CY2012

AQS Site Code

Site Name













VOCs













BENZ

BUTA

CTET

CLFRM

EDC

MECL

TCE1122 PERC

TCE

VC

cDCPEN

tDCPEN

ACRO

ACRY

25-025-0042

Boston, MA

10.1 (30)

20.8 (27)

2.7 (30)

16.8 (26)

4.1 (7)

14.1 (30)

28.0(11)

5.2 (1)

-

-

-

24.2 (23)

-

50-007-0007

Underbill, VT

-

-

-

-

-

-

-

-

-

-

-

-

-

44-007-0022

Providence, RI

-

-

-

-

-

-

-

-

-

-

-

-

-

36-005-0080/-0110

Bronx, NY

3.1 (9)

35.2 (9)

1.2 (9)

9.0 (9)

4.2 (9)

29.3 (9)

32.6(1) 35.1 (9)

11.6(5)

3.9 (3)

-

-

15.8 (9)

-

36-055-1007

Rochester, NY

-

-

-

-

-

-

-

-

-

-

-

-

-

11-001-0043

Washington, DC

-

32.5 (8)

3.8 (23)

12.9 (26)

-

11.8 (22)

29.8 (12)

-

-

-

-

-

71.1 (4)

51-087-0014

Richmond, VA

7.1 (24)

10.9(1)

5.6 (25)

-

-

35.6(10)

0(1)

-

-

-

-

25.0 (3)

-

45-025-0001

Chesterfield, SC

7.6 (59)

-

5.4 (59)

-

-

81.7 (54)

3.4(8) 1.0(13)

-

-

-

-

59.7 (2)

-

13-089-0002

Decatur, GA

27.6 (45)

-

9.4 (46)

9.1 (3)

-

-

-

-

-

-

-

-

-

21-043-0500

Grayson Lake, KY

10.5 (4)

8.2 (3)

7.2 (5)

29.8(1)

7.3 (3)

40.0 (5)

-

-

-

-

-

14.8 (5)

102.5 (3)

12-057-3002

Hillsborough Cty, FL

-

-

-

-

-

-

-

-

-

-

-

-

-

12-103-0026

Pinellas Cty, FL

1.6 (4)

6.2 (4)

1.8 (4)

8.3 (4)

4.4 (4)

2.4 (4)

10.4 (4)

-

-

2.7(1)

0(1)

5.2 (4)

-

26-163-0033

Dearborn, MI

7.8 (6)

5.6 (6)

7.0 (6)

29.0 (6)

6.4 (4)

6.3 (6)

4.7 (3)

-

-

-

-

11.8 (6)

-

55-027-0001

Horicon, WI

1.9 (2)

-

-

-

-

-

-

-

-

-

-

11.1(2)

-

17-031-4201

Northbrook, IL

-

-

9.3 (7)

20.8 (7)

7.7 (5)

14.2 (7)

7.4 (3)

10.5 (2)

-

-

-

40.9 (6)

-

48-201-1039

Deer Park, TX

-

-

-

-

-

-

-

-

-

-

-

-

-

29-510-0085

St. Louis, MO

9.0 (8)

7.8 (8)

17.9 (8)

4.9 (7)

8.8 (5)

11.2 (8)

10.7 (4)

2.0(1)

-

-

-

15.2 (8)

-

49-011-0004

Bountiful, UT

9.0 (6)

7.8 (6)

10.9 (6)

9.3 (6)

8.0 (3)

27.7 (6)

8.8(1)

-

-

-

-

5.1 (6)

-

08-077-0017/-0018

Grand Junction, CO

6.0 (6)

4.2 (6)

15.5 (6)

7.4 (4)

7.6 (5)

8.7 (6)

5.9 (5)

-

-

-

-

23.6 (6)

15.7 (1)

04-013-9997

Phoenix, AZ

2.2 (6)

4.5 (6)

3.7 (6)

2.9 (5)

3.4 (4)

38.6 (6)

4.4 (6)

-

-

-

-

21.5 (6)

-

06-037-1103

Los Angeles, CA

-

-

-

-

-

-

-

-

-

-

-

-

-

06-065-8001

Rubidoux, CA

-

-

-

-

-

-

-

-

-

-

-

-

-

06-085-0005

San Jose, CA

23.6 (41)

15.8 (7)

17.0 (43)

30.1 (28)

32.4(12)

29.4 (34)

23.9 (38)

31.5 (2)

-

-

-

35.7 (39)

24.8 (8)

41-051-0246

Portland, OR

10.5 (29)

4.4 (2)

11.1 (28)

-

-

26.3 (21)

-

-

-

-

-

33.6 (30)

-

53-033-0080

Seattle, WA

6.3 (6)

3.5 (6)

8.3 (6)

14.1 (6)

0(2)

21.7 (6)

2.3 (1)

—

—

—

—

23.4 (6)

—



Network Mean

15.7 (279)

18.8 (93)

9.6 (311)

18.9 (132)

15.2 (61)

44.4 (228)

11.3(9) 21.9(110)

16.3 (11)

3.9 (3)

2.7(1)

0(1)

28.7 (155)

59.6(16)

Expressed as percentage coefficient of variation (%CV) with number of contributing data pairs presented in parentheses.
Values shown in red exceed the MQO of < 15% CV.


-------
Table 16. Overall Precision for Primary, Duplicate, and Collocated Samples > MDL - CY2012 (continued)

AQS Site Code

Site Name

carbonyls

PAHs





metals







FORM

ACET

NAPH BaP

As

Be

Cd

Pb

Mn

Ni

CrVI

25-025-0042

Boston, MA

6.6(19)

5.9(15)

-

3.0 (36)

13.2 (4)

27.9 (33)

3.7 (37)

2.4 (52)

8.6 (55)

5.8(12)

50-007-0007

Underhill, VT

-

-

-

33.9 (3)

-

19.7 (6)

7.0 (3)

6.4 (6)

6.6(1)

-

44-007-0022

Providence, RI

18.9 (17)

13.1 (17)

--

11.0 (25)

--

25.3 (22)

9.6 (26)

16.5 (21)

28.8 (26)

6.2 (4)

36-005-0080/-0110

Bronx, NY

--

-

-

9.6 (54)

9.7(1)

7.1 (46)

4.3 (55)

6.4 (43)

19.6 (55)

9.7 (6)

36-055-1007

Rochester, NY

-

-

-

-

-

-

-

-

-

13.9 (4)

11-001-0043

Washington, DC

--

-

-

--

--

--

-

-

-

33.7 (10)

51-087-0014

Richmond, VA

3.9 (61)

2.4 (61)

-

--

-

-

--

-

-

13.9 (6)

45-025-0001

Chesterfield, SC

6.7 (58)

9.3 (57)

-

19.4 (88)

54.1 (52)

31.4 (90)

34.0 (92)

34.8 (96)

65.6 (24)

9.7 (2)

13-089-0002

Decatur, GA

--

-

41.1 (21)

15.8 (7)

-

-

13.0(12)

21.8(12)

11.7(12)

27.4 (7)

21-043-0500

Grayson Lake, KY

2.1 (6)

1.6(6)

-

16.9 (20)

-

29.6 (22)

4.1 (24)

3.4 (24)

1.2 (2)

5.2 (3)

12-057-3002

Hillsborough Cty, FL

12.1 (5)

4.3 (5)

10.7 (12)

29.5 (41)

-

5.9(11)

21.2 (47)

7.5 (57)

13.2 (40)

0.7 (2)

12-103-0026

Pinellas Cty, FL

5.3 (6)

4.2 (6)

-

-

-

-

-

-

-

19.5 (9)

26-163-0033

Dearborn, MI

11.2 (8)

6.3 (7)

6.0(12) 14.9(8)

10.2 (58)

10.6 (32)

10.8 (58)

-

6.8 (59)

9.3 (59)

5.6 (9)

55-027-0001

Horicon, WI

7.9 (4)

6.3 (4)

4.9 (2)

5.6 (2)

--

61.5 (2)

21.4 (2)

1.6 (2)

3.4 (2)

16.5 (8)

17-031-4201

Northbrook, IL

3.9 (8)

2.6 (8)

-

3.6 (2)

9.4(1)

5.0(1)

0.3 (2)

0.1 (2)

16.5 (2)

26.2 (13)

48-201-1039

Deer Park, TX

--

-

-

23.4 (40)

0(1)

42.5 (11)

18.0 (40)

18.1 (40)

26.2 (39)

24.8 (9)

29-510-0085

St. Louis, MO

2.5 (6)

1.8(6)

-

11.2 (58)

17.1 (11)

11.9 (59)

4.6 (25)

5.2 (59)

17.0 (53)

6.4 (8)

49-011-0004

Bountiful, UT

3.9 (5)

3.6(5)

-

11.8 (4)

14.3 (2)

10.6 (4)

5.9 (4)

9.9 (4)

11.2 (3)

9.2 (6)

08-077-0017/-0018

Grand Junction, CO

1.0 (6)

1.4 (6)

-

32.3 (11)

-

52.9 (5)

47.8(11)

57.2 (12)

68.3 (12)

15.4 (3)

04-013-9997

Phoenix, AZ

10.6 (7)

3.5 (6)

-

-

-

-

-

-

-

10.1 (6)

06-037-1103

Los Angeles, CA

--

-

-

--

--

--

-

-

-

47.5 (6)

06-065-8001

Rubidoux, CA

-

-

14.0 (12)

-

-

-

-

-

-

20.6 (5)

06-085-0005

San Jose, CA

-

-

-

-

-

-

-

-

-

-

41-051-0246

Portland, OR

20.3 (30)

16.2 (30)

12.6(23) 10.4(5)

5.0 (41)

9.4 (8)

19.3 (39)

3.9 (41)

7.0 (41)

4.3 (39)

11.7 (13)

53-033-0080

Seattle, WA

3.1 (6)

1.7 (6)

4.4(11) 12.6(2)

—

—

—

—

—

—

8.0(10)



Network Mean

10.2 (246)

8.5 (239) 21.6 (82) 13.3 (13)

16.7 (490)

37.9 (112)

:3.4 (409)

20.4 (421)

19.1 (530)

24.9 (424)

19.4(151)

Expressed as percentage coefficient of variation (%CV) with number of contributing data pairs presented in parentheses.
Values shown in red exceed the MQO of < 15% CV.


-------
Site Name
Boston, MA
Richmond, VA
Chesterfield, SC
Grayson Lake, KY
Hillsborough Cty, FL
Pinellas Cty, FL
Dearborn, Ml
Horicon, Wl
Northbrook, IL
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Phoenix, AZ
San Jose, CA
Portland, OR
Seattle, WA

MQO

$/$ no data available

30	45

Percent CV
i Analytical Precision ¦ Overall Precision

Figure 3. Analytical and Overall Precision Summary for Acrolein > MDL at NATTS Sample Collection Sites

CY2011


-------
Site Name
Boston, MA
Richmond, VA
Chesterfield, SC
Decatur, GA
Grayson Lake, KY
Hillsborough Cty, FL
Pinellas Cty, FL
Dearborn, Ml
Horicon, Wl
Northbrook, IL
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Phoenix, AZ
San Jose, CA
Portland, OR
Seattle, WA

MQO

15

30	45

Percent CV

60

75

$/$ no data available

i Analytical Precision ¦ Overall Precision

Figure 4. Analytical and Overall Precision Summary for Benzene > MDL at NATTS Sample Collection Sites in

CY2011


-------
Site Name

Boston, MA
Washington, DC
Grayson Lake, KY
Hillsborough Cty, FL
Pinellas Cty, FL
Dearborn, Ml
Northbrook, IL
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Phoenix, AZ
San Jose, CA
Seattle, WA

15

30

45
Percent CV

60

75

90

$/$ no data available

i Analytical Precision ¦ Overall Precision

Figure 5. Analytical and Overall Precision Summary for 1,3-Butadiene > MDL at NATTS Sample Collection Sites in

CY2011


-------
Site Name
Boston, MA
Underbill, VT
Providence, Rl
Richmond, VA
Chesterfield, SC
Grayson Lake, KY
Hillsborough Cty, FL
Pinellas Cty, FL
Dearborn, Ml
Horicon, Wl
Northbrook, IL
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Phoenix, AZ
San Jose, CA
Portland, OR
Seattle, WA

$/$ no data available

i Analytical Precision ¦ Overall Precision

Figure 6. Analytical and Overall Precision Summary for Formaldehyde > MDL at NATTS Sample Collection Sites in

CY2011


-------
Site Name
Decatur, GA

Hillsborough Cty, FL
Dearborn, Ml
Horicon, Wl
Rubidoux, CA
Portland, OR
Seattle, WA

* no data available

*

0	5	10	15	20	25

Percent CV

¦ Analytical Precision ¦ Overall Precision

Figure 7. Analytical and Overall Precision Summary for Naphthalene > MDL at NATTS Sample Collection Sites

CY2011


-------
Site Name
Boston, MA
Underbill, VT
Providence, Rl
Bronx, NY
Chesterfield, SC
Decatur, GA
Grayson Lake, KY
Hillsborough Cty, FL
Dearborn, Ml
Horicon, Wl
Deer Park, TX
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Los Angeles, CA
Rubidoux, CA
Portland, OR

MQO

10

15

20 25
Percent CV

30

35

40

45

$ no data available

i Analytical Precision ¦ Overall Precision

Figure 8. Analytical and Overall Precision Summary for PMio Arsenic > MDL at NATTS Sample Collection Sites

CY2011


-------
oo
oo

Site Name

Boston, MA
Underbill, VT
Providence, Rl
Bronx, NY
Rochester, NY
Washington, DC
Richmond, VA
Chesterfield, SC
Decatur, GA
Grayson Lake, KY
Hillsborough Cty, FL
Pinellas Cty, FL
Dearborn, Ml
Horicon, Wl
Northbrook, IL
Deer Park, TX
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Phoenix, AZ
Los Angeles, CA
Rubidoux, CA
Portland, OR
Seattle, WA

MQO

* no data available

i Analytical Precision ¦ Overall Precision

Figure 9. Analytical and Overall Precision Summary for Chromium (VI) > MDL at NATTS Sample Collection Sites in

CY2011


-------
Site Name
Boston, MA
Bronx, NY
Richmond, VA
Chesterfield, SC
Grayson Lake, KY
Hillsborough Cty, FL
Pinellas Cty, FL
Dearborn, Ml
Horicon, Wl
Northbrook, IL
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Phoenix, AZ
San Jose, CA
La Grande, OR
Seattle, WA

MQO

$/$ no data available

i Analytical Precision ¦ Overall Precision

Figure 10. Analytical and Overall Precision Summary for Acrolein > MDL at NATTS Sample Collection Sites in

CY2012


-------
Site Name
Boston, MA
Bronx, NY
Richmond, VA
Chesterfield, SC
Decatur, GA
Grayson Lake, KY
Hillsborough Cty, FL
Pinellas Cty, FL
Dearborn, Ml
Horicon, Wl
Northbrook, IL
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Phoenix, AZ
San Jose, CA
La Grande, OR
Seattle, WA

MQO

10

15

Percent CV

20

25

30

$/$ no data available

i Analytical Precision ¦ Overall Precision

Figure 11. Analytical and Overall Precision Summary for Benzene > MDL at NATTS Sample Collection Sites in

CY2012


-------
Site Name
Boston, MA
Bronx, NY
Washington, DC
Richmond, VA
Grayson Lake, KY
Hillsborough Cty, FL
Pinellas Cty, FL
Dearborn, Ml
Northbrook, IL H
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Phoenix, AZ
San Jose, CA
La Grande, OR
Seattle, WA

MQO

10	15	20	25

Percent CV

30

35

40

$/$ no data available

I Analytical Precision ¦ Overall Precision

Figure 12. Analytical and Overall Precision Summary for 1,3-Butadiene > MDL at NATTS Sample Collection Sites

CY2012


-------
Site Name
Boston, MA
Underbill, VT
Providence, Rl
Richmond, VA
Chesterfield, SC
Grayson Lake, KY
Hillsborough Cty, FL
Pinellas Cty, FL
Dearborn, Ml
Horicon, Wl
Northbrook, IL
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Phoenix, AZ
San Jose, CA
La Grande, OR
Seattle, WA

MQO

10	15

Percent CV

$/$ no data available

i Analytical Precision ¦ Overall Precision

Figure 13. Analytical and Overall Precision Summary for Formaldehyde > MDL at NATTS Sample Collection Sites

CY2012


-------
Site Name

Boston, MA
Underbill, VT
Providence, Rl
Bronx, NY
Rochester, NY
Washington, DC
Richmond, VA
Chesterfield, SC
Decatur, GA
Grayson Lake, KY
Hillsborough Cty, FL
Pinellas Cty, FL
Dearborn, Ml
Horicon, Wl
St. Louis, MO
Bountiful, UT
Phoenix, AZ
Los Angeles, CA
Rubidoux, CA
San Jose, CA
Seattle, WA

*/* no data available

0	5 10 15 20 25 30 35 40 45

Percent CV

¦ Analytical Precision ¦ Overall Precision

Figure 14. Analytical and Overall Precision Summary for Naphthalene > MDL at NATTS Sample Collection Sites in

CY2012


-------
Site Name

Boston, MA
Underbill, VT
Providence, Rl
Bronx, NY
Chesterfield, SC
Decatur, GA
Grayson Lake, KY
Hillsborough Cty, FL
Dearborn, Ml
Horicon, Wl
Northbrook, IL
Deer Park, TX
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Phoenix, AZ
San Jose, CA
La Grande, OR
Seattle, WA

MQO

10	15	20	25

Percent CV

30

35

40

$/$ no data available

i Analytical Precision ¦ Overall Precision

Figure 15. Analytical and Overall Precision Summary for PMio Arsenic > MDL at NATTS Sample Collection Sites

CY2012


-------
Site Name

Boston, MA
Underhill, VT
Providence, Rl
Bronx, NY
Rochester, NY
Washington, DC
Richmond, VA
Chesterfield, SC
Decatur, GA
Grayson Lake, KY
Hillsborough Cty, FL
Pinellas Cty, FL
Dearborn, Ml
Horicon, Wl
Northbrook, IL
Deer Park, TX
Harrison Cty, TX
St. Louis, MO
Bountiful, UT
Grand Junction, CO
Phoenix, AZ
Los Angeles, CA
Rubidoux, CA
La Grande, OR
Seattle, WA

MQO

5f / no data available

I Analytical Precision ¦ Overall Precision

Figure 16. Analytical and Overall Precision Summary for Chromium (VI) > MDL at NATTS Sample Collection Sites in

CY2012


-------
2.6 Laboratory Bias Data Based on Proficiency Testing (PT) Samples

PT analyses were performed in CY2011 QTR2 and QTR4 and in CY2012 QTR 1. Blind
"spiked" PT samples were prepared for metals and PAHs for CY2011 QTR2 by Wibby
Environmental. Battelle prepared VOC and carbonyl PT samples in CY2011 QTR4 and
prepared metals, PAH, and chromium (VI) PTs in CY2012 QTR1. Participating NATTS
analysis laboratories submitted results to the respective PT provider, which were evaluated for
acceptability by the provider, Wibby Environmental or Battelle, as appropriate.

Laboratory bias is measured by the percentage difference between the laboratory's measured
value and the target value for the PT sample for a given HAP:

Measured - Target	1\

%Differenc e =	2	100	(Eq. J)

Target

Target values were typically assigned as the average of the results of one or more confirmatory
analysis (referee) samples.

The percentage of NATTS laboratories that participated in the PT program for CY2011 and
CY2012 is shown in Table 17.

Table 17. Percentage of NATTS Laboratories Participating in the NATTS Proficiency

Testing Program in CY2011 and CY2012

PT year and quarter

VOCs

carbonyls

PAHs

metals

chromium (VI)

CY2011 QTR2

-

-

83%

69%

-

CY2011 QTR4

100%

92%

-

-

-

CY2012 QTR1

-

-

100%

100%

100%

The CY2011 and CY2012 PT samples were prepared to contain the 27 HAPs listed in Table 2
(except acrylonitrile) and many of the 11 HAPs given in Table 18. These 11 additional HAPs
include two carbonyls, six PAHs, and three metals.

46


-------
Table 18. Additional HAPs Contained in NATTS Proficiency Test Samples in

CY2011 and/or CY2012

HAP

HAP
Abbreviation

HAP Class

Spiked in
CY2011 PT
Samples?

Spiked in
CY2012 PT
Samples?

benzaldehyde

BNZD

carbonyl

Yes

No

propionaldehyde

PRPD

carbonyl

Yes

No

acenaphthene

ACEN

PAH

Yes

Yes

anthracene

ANTH

PAH

Yes

Yes

fluorene

FLUR

PAH

Yes

Yes

fluoranthene

FTHN

PAH

Yes

Yes

phenanthrene

PHEN

PAH

Yes

Yes

pyrene

PYR

PAH

Yes

Yes

cobalt

Co

metal

Yes

Yes

antimony

Sb

metal

No

Yes

selenium

Se

metal

Yes

Yes

For the two PTs performed in CY2011, the PT samples were spiked with 15 VOCs, four
carbonyls, eight PAHs, and eight metals, for a total of 35 HAPs. Tables 21 through 24 present
the PT results for these 35 HAPs for the two CY2011 PTs (one table per HAP class). Tables 25
through 27 present the PT results for the CY2012 PT for the eight PAHs, nine metals, and
chromium (VI) for a total of 18 HAPs. To reflect overall bias independent of direction, the mean
of the absolute value of the percent difference, along with the minimum and maximum values,
are presented at the bottom and in the right-most columns of these tables.

Figures 17 and 18 are box and whisker plots summarizing the percent difference values for
CY2011 and CY2012, respectively, for the seven HAPs of primary importance: acrolein,
benzene, 1,3-butadiene, formaldehyde, naphthalene, arsenic, and chromium (VI). The CY2012
PT only included PAHs, metals, and chromium (VI), hence only three of the seven HAPs of
primary importance were analyzed. A laboratory's results were included in these summaries
only if the laboratory provided analysis results for a particular sample type.

The two reference lines in Figures 17 and 18 represent the MQO of 25% for laboratory bias, in
either direction of zero bias. Thus, laboratories whose percent difference values fall within the
reference lines have achieved the MQO. Those results that fall more than 1.5 times the IQR
either above the 75th percentile or below the 25th percentile are identified by their laboratory
number (Tables 10 and 11). Figures 17 and 18 present PT results for all labs participating in the
NATTS PT program, including those labs not affiliated with NATTS sites. These non-NATTS
labs are assigned identification codes similar to those of the NATTS labs; see Table 19.

47


-------
Table 19. Non-NATTS Laboratories Analyzing Proficiency Test Samples

in CY2011 and CY2012

Laboratory



Code(s)

Laboratory Description

01-04-V

US EPA Region I Laboratory

01-05-V

Maine Department of Environmental Protection Air Laboratory

03-03-M

Pennsylvania Department of Environmental Protection, Bureau of Air Quality

04-06-V

North Carolina Department of Environment and Natural Resources

05-04-C,M,V

State of Minnesota Pollution Control Agency

05-06-M,V

Indiana Department of Environmental Management

05-07-M,V

Ohio Environmental Protection Agency Division of Environmental Services Lab

07-02-C,V

State Hygenic Laboratory at The University of Iowa

09-06-C,V

Air Pollution Control District County of San Diego

09-09-V

Joint Water Pollution Control Plant of Los Angeles County

11-03-V

US EPA National Exposure Research Laboratory

As can be seen in Tables 21 through 27, and as is summarized in Figures 17 and 18, with some
exceptions for certain laboratories and HAPs, the majority of laboratories met the laboratory bias
MQO for each of the three rounds of PTs for the seven HAPs of primary importance. In Figures
17 and 18, the central tendency of the analysis bias is best characterized by the median bias
(indicated by black horizontal lines within the IQR boxes), which lessen the effect of extreme
values.

Figure 17 shows that across laboratories, PT analyses in CY2011, based on the median bias,
tended to demonstrate a marginally low analytical bias for acrolein, formaldehyde, and arsenic, a
marginally high analytical bias for benzene, a slightly high analytical bias for 1,3-butadiene, and
a very low analytical bias for naphthalene. As shown in Table 20, percentages of NATTS
laboratories meeting the MQO for acrolein, benzene and 1,3-butadiene were 77%, 85%, and
85%, respectively. All NATTS laboratories but one met the MQO for arsenic and for
formaldehyde. For the five NATTS laboratories reporting PAH results, only two laboratories
met the MQO for naphthalene. Specifically, acceptable measurement bias was difficult to obtain
for acrolein and naphthalene; the mean absolute percent bias across all participating laboratories
was 29.5% and 26.0%, respectively. The CY2011 PAH PT results should be interpreted with
caution, as it appears that the target value may have been biased high: all reported results but one
showed a negative bias.

For the three HAPs of primary importance that were spiked for the CY2012 PT (naphthalene,
arsenic, and chromium (VI)), analytical bias was slightly high for naphthalene and arsenic and
very high for chromium (VI) as seen in Figure 18. All NATTS laboratories met the MQO for
chromium (VI) and all but one laboratory met the MQO for both naphthalene and arsenic.
CY2012 PT results are shown in Tables 25 through 27.

48


-------
Table 20. Percentages of NATTS Laboratories Meeting the Bias MQO for Proficiency Test

Samples in CY2011 and CY2012





VOCs

carbonyls

PAHs

metals



CY

Acrolein

Benzene 1,3-Butadiene

Formaldehyde

Naphthalene

Arsenic

Chromium (VI)

2011

77%

85% 85%

87%

40%

90%

-

2012

-

-

-

83%

93%

100%

49


-------
Table 21. NATTS Proficiency Testing Bias Results (Percent Difference from Target) for VOCs - CY2011 QTR4

o

Lab Code

Laboratory Description

ACRO

BENZ

BUTA

CLFRM

CTET

DCP

EDB

EDC

MECL

PERC

TCE

TCE 1122

VC

c-DCPEN

t-DCPEN

Mean
Abs. Bias
(across HAPs)

Min

Max

01-01-V

RI Dept of Health

-10.0

15.4

1.9

-10.7

-3.6

-4.4

8.1

-15.1

-6.7

0.4

-11.9

-3.4

-7.0

1.8

14.2

7.6

-15.1

15.4

01-04-V

US EPA Region 1 Laba

-

-10.6

-4.2

-10.0

-2.0

-42.1

-16.2

-30.0

-8.5

-17.7

-28.7

-21.5

-7.2

-3.7

0.9

14.5

-42.1

0.9

01-05-V

Maine DEP Air Laba

-4.3

22.8

3.3

-3.4

3.8

-

22.1

-1.0

6.8

7.6

19.8

77.4

-3.2

-

39.3

16.5

-4.3

11A

02-01-V

New York State DEC

3.4

12.4

6.4

-11.8

2.6

0.4

3.4

-14.4

0.6

6.0

-3.6

-3.7

-4.1

7.5

26.3

7.1

-14.4

26.3

03-01-V

Maryland DOE

2.8

4.0

6.2

-14.0

-4.9

-2.3

-3.9

-13.3

-4.9

-0.4

-9.7

-13.0

-5.4

-5.9

28.6

7.9

-14.0

28.6

03-02-V

Virginia Division of
Consolidated Labs

45.9

23.8

17.9

-5.5

3.1

12.3

35.8

6.7

5.8

14.7

6.7

48.1

18.0

0.7

25.2

18.0

-5.5

48.1

04-01-V

Pinellas County DEM AQD

-12.3

6.0

3.1

-10.8

2.2

3.0

12.6

-14.4

-10.2

8.4

-1.3

-4.9

-2.1

22.5

40.0

10.2

-14.4

40.0

04-02-V

SCDHEC/DAQA

-42.6

-41.9

-42.1

-45.8

-38.6

-38.9

-29.8

-45.2

-45.4

-32.8

-38.7

-19.3

-37.7

-27.0

-14.3

36.0

-45.8

-14.3

04-04-V

Georgia DNR

-25.0

-11.3

-13.7

-25.8

-20.0

-17.2

-15.5

-21.2

-20.0

-20.8

-19.7

-22.2

-14.4

-1.5

17.8

17.7

-25.8

17.8

04-06-V

North Carolina DENRa

197.8

18.9

12.4

2.9

2.2

23.0

30.5

2.5

4.2

26.8

13.2

38.6

6.6

26.7

51.2

30.5

2.2

197.8

05-03-V

Wisconsin DNR

29.4

43.4

36.8

9.1

25.4

33.3

9.4

21.2

32.2

20.8

20.0

40.7

37.7

14.8

26.1

26.7

9.1

43.4

05-04-V

Minnesota PCAa

-

8.4

10.9

-19.1

2.4

-2.6

-4.8

-14.5

6.9

8.6

-4.9

-7.7

5.8

5.3

17.3

8.5

-19.1

17.3

05-06-V

Indiana DEM a

53.5

49.1

49.8

26.1

42.7

17.9

32.8

41.2

47.8

43.8

21.3

29.6

35.7

28.9

58.3

38.6

17.9

58.3

05-07-V

Ohio EPAa

-

32.1

19.3

-3.0

5.1

12.3

77.4

-6.1

8.5

-5.7

33.3

18.5

18.0

66.7

130.4

31.2

-6.1

130.4

06-01-V

Texas CEQ Air Laboratory

-11.8

13.2

5.3

-9.1

15.3

-12.3

20.8

-15.2

-11.9

5.7

0.0

3.7

-8.2

25.9

47.8

13.7

-15.2

47.8

07-02-V

State Hygenic Lab, Univ. of
Iowaa

27.4

32.8

-2.8

26.4

19.3

31.6

57.0

45.2

15.6

59.6

30.3

71.5

-3.0

34.4

82.2

35.9

-3.0

82.2

09-03-V

BAAQMD

19.1

20.8

17.5

3.0

32.2

-

37.7

4.5

20.3

84.9

6.7

-

6.6

-

-

23.0

3.0

84.9

09-06-V

San Diego APCDa

-13.3

13.1

2.2

-5.6

2.0

-5.2

7.7

-12.9

0.7

11.3

-2.2

3.8

2.8

15.0

48.4

9.7

-13.3

48.4

09-08-V

South Coast AQMD
Laboratory

-8.5

9.8

1.1

-12.4

-5.1

2.5

6.4

-13.6

-2.0

9.4

-11.7

9.6

-3.9

5.6

28.3

8.7

-13.6

28.3

09-09-V

JWPCP of Los Angeles3

-

17.0

8.8

-6.1

5.1

-

17.0

-6.1

5.1

13.2

6.7

-

4.9

-

-

9.0

-6.1

17.0

10-02-V

Oregon DEQ

14.7

-10.2

-23.2

-27.6

-13.2

-17.2

-9.8

-31.8

-11.9

-20.4

-18.7

-14.4

-17.0

1.5

8.3

16.0

-31.8

14.7

11-01-V

ERG

-20.3

10.2

10.9

-9.7

25.4

2.8

5.7

-11.8

6.4

0.8

-8.3

1.1

0.3

21.5

41.3

11.8

-20.3

41.3

11-03-V

US EPA NERLa

-18.5

6.4

0.4

-13.6

-3.4

-2.5

20.4

-15.8

0.7

7.9

-5.7

5.9

-5.2

10.0

40.4

10.5

-18.5

40.4

Mean Bias (across laboratories)

12.0

12.4

5.6

-7.7

4.3

-0.3

14.1

-7.0

1.7

10.1

-0.3

11.4

0.8

12.5

36.1







Mean Abs. Bias (across laboratories)

29.5

18.8

13.0

13.5

12.2

14.2

21.1

17.5

12.3

18.6

14.0

19.9

11.1

16.3

37.5

17.8





Median Bias (across laboratories)

-4.3

13.1

5.3

-9.7

2.4

-1.0

9.4

-13.3

0.7

7.9

-2.2

3.7

-3.0

8.7

28.6







Minimum

-42.6

-41.9

-42.1

-45.8

-38.6

-42.1

-29.8

-45.2

-45.4

-32.8

-38.7

-22.2

-37.7

-27.0

-14.3







Maximum

197.8

49.1

49.8

26.4

42.7

33.3

77.4

45.2

47.8

84.9

33.3

77.4

37.7

66.7

130.4







a Laboratories not performing analysis for NATTS sites

Values listed in red indicate absolute bias outside the MQO (>25%); values listed in rangi indicate absolute bias between 20-25%.


-------
Table 22. NATTS Proficiency Testing Bias Results (Percent Difference from Target) for Carbonyls - CY2011 QTR4













Mean





Laboratory











Abs. Bias (across





Code

Laboratory Description

ACET

BNZD

FORM

PRPD

HAPs)

Min

Max

01-01-C

Rhode Island Department of Health Air Pollution Lab

0.5

-

6.8

-

3.6

0.5

6.8

01-03-C

Massachusetts Division of Environmental Protection

-7.6

-5.6

-8.4

-16.9

9.6

-16.9

-5.6

01-04-C

US EPA Region 1 Laboratory

-1.4

2.1

1.3

-4.6

2.4

-4.6

2.1

02-01-C

New York State Department of Environmental Conser

-22.8

-19.3

-25.5

-29.4

24.2

-29.4

-19.3

03-01-C

Philadelphia Air Management Services Laboratory

-5.6

-

-5.8

-15.4

8.9

-15.4

-5.6

03-02-C

Virginia Division of Consolidated Services Laboratory

-3.7

-

-0.8

-10.0

4.8

-10.0

-0.8

04-02-C

South Carolina Division of Health and Envir Control

-1.0

-

6.6

-

3.8

-1.0

6.6

04-03-C

Kentucky Division of Environmental Services

69.5

75.8

66.9

56.2

67.1

56.2

75.8

04-04-C

Georgia Department of Natural Resources

-16.5

-5.1

-21.7

-30.5

18.5

-30.5

-5.1

05-03-C

Wisconsin State Laboratory of Hygiene

-4.6

-8.9

-5.8

-15.4

8.7

-15.4

-4.6

05-04-C

State of Minnesota Pollution Control Agency

-7.6

-10.0

-9.1

-15.4

10.5

-15.4

-7.6

06-01-C

Texas Commission on Environmental Quality Air Lab

-7.6

-0.4

-5.8

-23.1

9.2

-23.1

-0.4

07-02-C

State Hygenic Laboratory at The University of Iowa

-5.6

-3.2

-6.8

-11.5

6.8

-11.5

-3.2

09-03-C

Bay Area Air Quality Management District

-6.8

-

-6.3

-

6.5

-6.8

-6.3

09-06-C

Air Pollution Control District County of San Diego

-0.4

-

-0.6

-

0.5

-0.6

-0.4

09-08-C

South Coast Air Quality Management District

-2.8

-

-2.1

-

2.5

-2.8

-2.1

10-02-C

Oregon Division of Environmental Quality

-1.8

-1.3

-3.6

-7.7

3.6

-7.7

-1.3

11-01-C

Environmental Resource Group

3.0

8.0

-3.4

-14.6

7.3

-14.6

8.0



Mean Bias (across laboratories)

-1.3

2.9

-1.3

-10.6









Mean Abs. Bias (across laboratories)

9.4

12.7

10.4

19.3

11.0







Median Bias (across laboratories)

-4.1

-3.2

-4.7

-15.4









Minimum

-22.8

-19.3

-25.5

-30.5









Maximum

69.5

75.8

66.9

56.2







a Laboratories not performing analysis for NATTS sites

Values listed in red indicate absolute bias outside the MQO (>25%); values listed in rangi indicate absolute bias between 20-25%.


-------
Table 23. NATTS Proficiency Testing Bias Results (Percent Difference from Target) for Metals - CY2011 QTR2





















Mean

























Abs.

























Bias





Laboratory



















(across





Code

Laboratory Description

As

Be

Cd

Co

Mn

Ni

Pb

Se

HAPs)

Min

Max

01-01-M

Rhode Island Department of Health Air Pollution Laboratory

-

49

-16.1

-

5.4

1.9

-5.4

-

15.5

-16.1

49

03-02-M

Virginia Division of Consolidated Services Laboratory

-6.8

-8.2

-11.5

-

-6.5

-8.9

-7.5

-

8.2

-11.5

-6.5

04-01-M

Enviromental Protection Commission of Hillsborough County

24.7

-7

-4.6

-2.9

-1.1

-5.4

0.6

-11.1

7.2

-11.1

24.7

04-02-M

South Carolina Division of Health and Environmental Control

-15.1

-14.8

-8

10.6

12.4

3

13.7

-27.3

13.1

-27.3

13.7

04-04-M

Georgia Division of Natural Resources

-5.5

-2.5

-9.2

-2.9

-4.3

-8.1

-6.6

-14.2

6.7

-14.2

-2.5

05-03-M

Wisconsin State Laboratory of Hygiene

-2.7

3.7

-4.6

-

8.1

-5.1

-0.3

-

4.1

-5.1

8.1

06-01-M

Texas Commission on Environmental Quality Air Laboratory

2.7

10.7

5.7

-

4.3

11.4

3.3

-

6.4

2.7

11.4

09-08-M

South Coast Air Quality Management District

-15.1

-18.9

-13.8

-10.1

-14.5

-11.1

-10.4

-20.9

14.4

-20.9

-10.1

10-02-M

Oregon Division of Environmental Quality

1.4

-8.6

-5.7

-6.7

-5.4

-9.2

-3.3

-10.9

6.4

-10.9

1.4

11-01-M

Environmental Resource Group

1.4

-8.2

-5.7

-4.3

-3.8

-6.8

-6.3

-15.6

6.5

-15.6

1.4



Mean Bias (across laboratories)

-1.7

-0.5

-7.4

-2.7

-0.5

-3.8

-2.2

-16.7









Mean Abs. Bias (across laboratories)

8.4

13.2

8.5

6.3

6.6

7.1

5.7

16.7

8.8







Median Bias (across laboratories)

-2.7

-7.6

-6.9

-3.6

-2.4

-6.1

-4.3

-14.9









Minimum

-15.1

-18.9

-16.1

-10.1

-14.5

-11.1

-10.4

-27.3









Maximum



49.0

5.7

10.6

12.4

11.4

13.7

-10.9







Values listed in red indicate absolute bias outside the MQO (>25%); values listed in rangi indicate absolute bias between 20-25%.


-------
Table 24. NATTS Proficiency Testing Bias Results (Percent Difference from Target) for PAHs - CY2011 QTR2

Laboratory
Code

Laboratory Description

ACEN

ANTH

BaP

FLUR

FTHN

NAPH

PHEN

PYR

Mean
Abs.
Bias
(across
HAPs)

Min

Max

03-02-P

Virginia Division of Consolidated Services Laboratory



-30.5

-37.0

-17.5

3.9

-29.7

-16.7

-6.9



-37.0

3.9

04-02-P

South Carolina Division of Health and Environmental Control

-29.7

-25.7

-27.4

-36.1

-27.3

-30.7

-23.3

-24.6

28.1

-36.1

-23.3

06-01-P

Texas Commission on Environmental Quality Air Laboratory

-15.9

-10.7

-11.2

-10.3

-14.3

-23.7

-5.0

-14.6

13.2

-23.7

-5.0

10-02-P

Oregon Division of Environmental Quality

-26.6

-14.3

-11.1

-24.7

-14.0

-32.0

-23.3

-13.3

19.9

-32.0

-11.1

11-01-P

Environmental Resource Group

-11.9

-13.6

-2.1

-7.2

-9.4

-13.9

-8.3

-6.9

9.2

-13.9

-2.1



Mean Bias (across laboratories)

-21.6

-19.0

-17.8

-19.2

-12.2

-26.0

-15.3

-13.2









Mean Abs. Bias (across laboratories)

21.6

19.0

17.8

19.2

13.8

26.0

15.3

13.2

18.2







Median Bias (across laboratories)

-24.1

-14.3

-11.2

-17.5

-14.0

-29.7

-16.7

-13.3









Minimum

-29.7

-30.5

-37.0

-36.1

-27.3

-32.0

-23.3

-24.6









Maximum

-11.9

-10.7

-2.1

-7.2

3.9

-13.9

-5.0

-6.9







Values listed in red indicate absolute bias outside the MQO (>25%); values listed in orange indicate absolute bias between 20-25%.

Ui


-------
Table 25. NATTS Proficiency Testing Bias Results (Percent Difference from Target) for Metals - CY2012 QTR1

Mean
Abs.
Bias

Laboratory	(across

Code

Laboratory Description

As

Be

Cd

Co

Mn

Ni

Pb

Sb

Se

HAPs)

Min

Max

01-01-M

Rhode Island Department of Health Air Pollution Laboratory

-28.8



-8.2

-

-5.2

-10.8

4.0

-2.1

-

11.5

-28.8



03-01-M

West Virginia Department of Environmental Protection

-0.1

7.2

8.5

-

5.6

-7.4

11.4

-

-

6.7

-7.4

11.4

03-02-M

Virginia Division of Consolidated Services Laboratory

12.9

14.3

8.3

-

8.9

-0.3

10.2

-

-

9.1

-0.3

14.3

03-03-M

Pennsylvania Department of Environmental Protection, Bureau of
Air Quality

-11.3

-12.5

1.5

11.6

13.7

2.8

13.3

-10.7

-45.0

13.6

-45.0

13.7

04-01-M

Enviromental Protection Commission of Hillsborough County

3.1

9.3

7.0

15.5

13.8

21.9

16.9

32.0

9.5

14.3

3.1

32.0

04-02-M

South Carolina Division of Health and Environmental Control

-11.7

-4.5

3.9

14.6

15.9

-5.0

21.8

-32.1

-33.7

15.9

-33.7

21.8

04-03-M

Kentucky Division of Environmental Services

7.8

7.6

5.7

-

8.1

2.3

7.2

-19.9

-

8.4

-19.9

8.1

04-04-M

Georgia Division of Natural Resources

4.9

7.2

4.2

16.0

11.0

2.1

10.9

-8.5

-21.8

9.6

-21.8

16.0

05-01-M

Michigan Department of Environmental Quality

-7.5

-6.0

-9.4

-2.1

-2.2

-17.5

-1.0

-

-

6.5

-17.5

-1.0

05-03-M

Wisconsin State Laboratory of Hygiene

10.8

14.7

12.0

-

16.5

5.2

16.8

-

-

12.7

5.2

16.8

05-04-M

State of Minnesota Pollution Control Agency

24.7

1.8

-4.8

8.9

2.1

-9.0

10.3

8.4

-21.7

10.2

-21.7

24.7

05-06-M

Indiana Department of Environmental Management

36.3

-1.3

10.5

-

15.4

60.2

27.6

-

-

25.2

-1.3

60.2

05-07-M

Ohio Environmental Protection Agency Division of
Environmental Services Laboratory

13.7

16.9

12.2

19.8

-

0.7

11.1

-

-25.3

14.2

-25.3

19.8

06-01-M

Texas Commission on Environmental Quality Air Laboratory

13.5

17.1

9.0

-

11.0

3.4

12.8

-

-

11.1

3.4

17.1

08-02-M

Colorado Department of Public Health and Environment,
Laboratory Services Division

17.6

17.7

15.6

29.2

141.8

7.1

31.9

7.3

-4.2

30.2

-4.2

141.8

09-08-M

South Coast Air Quality Management District

-1.6

-1.9

-1.1

11.8

8.5

-8.1

8.7

-5.1

-22.6

7.7

-22.6

11.8

10-02-M

Oregon Division of Environmental Quality

10.8

8.4

8.5

12.1

7.5

-3.3

5.0

-

-21.9

9.7

-21.9

12.1

11-01-M

Environmental Resource Group

15.7

17.5

16.6

29.8

21.5

11.4

19.9

-6.0

-7.3

16.2

-7.3

29.8

11-02-M

RTI International

6.2

14.5

7.1

11.2

5.1

-7.4

7.9

-3.4

-13.3

8.5

-13.3

14.5



Mean Bias (across laboratories)

6.2

7.9

5.6

14.9

16.6

2.5

13.0

-3.6

-18.8









Mean Abs. Bias (across laboratories)

12.6

10.6

8.1

15.2

17.4

9.8

13.1

12.3

20.6

12.7







Median Bias (across laboratories)

7.8

8.4

7.1

13.3

10.0

0.7

11.1

-5.1

-21.8









Minimum

-28.8

-12.5

-9.4

-2.1

-5.2

-17.5

-1.0

-32.1

-45.0









Maximum

36.3

21.0

16.6

29.8

141.8

60.2

31.9

32.0

9.5







a Laboratories not performing analysis for NATTS sites

Values listed in red indicate absolute bias outside the MQO (>25%); values listed in rangi indicate absolute bias between 20-25%.


-------
Table 26. NATTS Proficiency Testing Bias Results (Percent Difference from Target) for PAHs - CY2012 QTR1





















Mean

























Abs.

























Bias





Laboratory



















(across





Code

Laboratory Description

ACEN

ANTH

BaP

FLUR

FTHN NAPH PHEN

PYR

HAPs)

Min

Max

03-02-P

Virginia Division of Consolidated Services Laboratory



-2.1



27.3

33.0

12.8

23.7

27.7



-2.1

33.0

04-02-P

South Carolina Division of Health and Environmental Control

-15.5

-22.4

17.5

-27.1

-13.2

-21.0

-1.1

-11.4

16.1

-27.1

17.5

05-03-P

Wisconsin State Laboratory of Hygiene

10.5

5.9

45.6

16.3

29.7

8.0

26.3

46.5

23.6

5.9

46.5

06-01-P

Texas Commission on Environmental Quality Air Laboratory

16.8

15.6

14.6

15.7

26.5

32.8

61.1

24.9

26.0

14.6

61.1

10-02-P

Oregon Division of Environmental Quality

-2.1

2.4

29.1

-7.6

11.1

-14.2

7.4

15.1

11.1

-14.2

29.1

11-01-P

Environmental Resource Group

19.2

4.1

25.2

9.6

15.1



18.4

17.9

16.4

4.1

25.2



Mean Bias (across laboratories)

8.7

0.6

25.4

5.7

17.0

6.6

22.6

20.1









Mean Abs. Bias (across laboratories)

14.6

8.7

25.4

17.3

21.4

18.4

23.0

23.9

19.1







Median Bias (across laboratories)

13.7

3.2

22.8

12.7

20.8

10.4

21.1

21.4









Minimum

-15.5

-22.4

14.6

-27.1

-13.2

-21.0

-1.1

-11.4









Maximum



15.6

45.6

27.3

33.0

32.8

61.1

46.5







a. Reported results were from a second PT sample which replaced the first. Sample storage integrity had been compromised during shipping.

Values listed in red indicate absolute bias outside the MQO (>25%); values listed in rangi indicate absolute bias between 20-25%.


-------
Table 27. NATTS Proficiency Testing Bias Results (Percent Difference from Target) for

Chromium (VI) - CY2012 QTR1

Laboratory Code

Laboratory Description

Chromium (VI)



Virginia Division of Consolidated Services



03-02-R

Laboratory

21.7

09-08-R

South Coast Air Quality Management District

18.0

10-02-R

Chester LabNet

14.6

11-01-R

Environmental Resource Group

19.5



Mean Bias (across laboratories)

18.5



Mean Abs. Bias (across laboratories)

18.5



Median Bias (across laboratories)

18.8



Minimum

14.6



Maximum



Values listed in oram indicate absolute bias between 20-25%.

56


-------
100
75
50-
25
0-
-25
-50
-75
-100-

04-06
198%
Difference

0 04-03

-&05-06

O05-06
0 05-03

.04-01

*

o 04-02

10 02

O04-02

gO«l.
02-01

i

r

10-02

Dashed lines indicate +/-25% WGO

ACRO	BENZ	BUTA	FORM NAPH

HAP

AS

Figure 17. Distribution of Laboratory Bias by HAP for Proficiency Testing Data - CY2011

57


-------
40

—06-01

20-

14	I

©

g

—i—

-20- !		!	

01-01

-40 - Dashed Inos indicate +/-25% MQO

As	NAPH	CrVI

HAP

Figure 18. Distribution of Laboratory Bias by HAP for Proficiency Testing Data - CY2012

2.7 Flow Audit Results from Instrument Performance Audits (IPAs)

Instrument performance audits (IPAs) of carbonyl, PAH, PMio, and chromium (VI) sampler
units were performed at the following NATTS field sites as follows:

•	CY2011 (eight sites): Rochester, NY; Bronx, NY; Washington, DC; Richmond, VA;
Chesterfield, SC; Decatur, GA; Hillsborough County, FL; and Pinellas County, FL

•	CY2012 (five sites): Horicon, WI; Northbrook, IL; Deer Park, TX; St Louis, MO;
and Portland, OR.

RTI performed the flow audits in CY2011 and CY2012. During each IPA, when flows were
sufficient for measurement, the flow rates on all sampler types at the NATTS site were
determined with certified, calibrated volumetric flow measurement devices and reported in
standard temperature and pressure (STP, 25°C and 1 atm) or ambient conditions (also referred to
as local conditions, LC) based on the typical reporting convention of the site operators. Field
bias was calculated by comparing the sampler flow reading (or setting) to the audit flow rate.

58


-------
Field bias is defined as the percentage difference between the site flow (Fs) and the audit flow
(Fa) under the same conditions (standard or ambient):

Fs — Fa

% Difference =	100	(Eq. 4)

Fa

The results from the flow audits conducted during CY2011 and CY2012 are indicated in
Tables 29 and 30 respectively. If present at the site, collocated samplers were also audited.
Carbonyl and chromium (VI) samplers may have multiple flow channels which allow for
duplicate sampling; the flow rates of any such flow channels were audited when used by the site
to collect duplicate samples. PMio metals and PAH samplers have only primary channels.

With few exceptions, most air samplers met the flow bias MQO of < 10%. The most frequent
exceedances occurred for PAH and chromium (VI) samplers. With the exception of VOC
samplers, which were not audited, the mean and mean absolute network flow bias met the MQO
for all HAP classes in CY2011 and CY2012, as indicated in Table 28.

Table 28. Mean Network Flow Bias From CY2011 and CY2012





HAP class





CY

Flow Bias (% difference)

VOC carbonyl

PAH

metals

Cr(VI)

2011

mean

-3.1

-3.7

1.4

-1.3

mean absolute

3.7

7.6

3.5

7.6

2012

mean

-1.8

2.5

-1.1

0.4

mean absolute

2.0

2.6

1.6

3.0

59


-------
Table 29. Flow Audit Results from the Instrument Performance Audits - CY2011

Site Identifier
and AQS ID

Method

Precision
Assignment

Channel

Sampler
Reading

Standard Reading

Units

Conditions

Percent
Difference

Rochester, NY
36-055-1007

voc

carbony
1

primary
primary

1

0.999

no flow readings recorded
0.984 L/min

ambient

1.5



PAH

primary



0.196

0.190

nrVmin

STP

3.2



metals

primary



16.66

17.12

L/min

ambient

-2.7



Cr(VI)

primary



15.0

15.02

L/min

ambient

-0.1



Cr(VI)

collocated



15.0

14.77

L/min

ambient

1.6

Bronx, NY
36-005-0080

VOC
carbony
1

primary
primary

1

0.999

no flow readings recorded
1.068 L/min

ambient

-6.5



PAH

primary



0.217

0.198

nrVmin

STP

9.6



metals

primary



16.65

16.85

L/min

ambient

-1.2



metals

collocated



16.68

16.92

L/min

ambient

-1.4



Cr(VI)

primary



15.0

14.35

L/min

ambient

4.5



Cr(VI)

collocated



15.0

15.13

L/min

ambient

-0.9

Washington, DC
11-001-0043

VOC
VOC

primary
duplicate





no flow readings recorded
no flow readings recorded



-



carbony
1

primary

1

0.500

0.561

L/min

ambient

-10.9



carbony
1

collocated

2

0.525

0.572

L/min

ambient

-8.2



PAH

primary



127

128.9

L/min

STP

-1.5



metals

primary



40

42.67

ftVmin

STP

-6.3



Cr(VI)

primary



15.0

12.81

IVmin

ambient

17.1



Cr(VI)

collocated



15.0

11.41

IVmin

ambient

31.5

Richmond, VA
51-087-0014

VOC
VOC

primary
collocated





no flow readings recorded
no flow readings recorded



-



carbony
1

primary

1

0.250

0.261

IVmin

ambient

-4.2



carbony
1

collocated

2

0.250

0.256

IVmin

ambient

-2.3



PAH

primary



135

134.4

IVmin

STP

0.4



PAH

collocated



144.6

144.2

IVmin

STP

0.3



metals

primary



42.05

39.21

ftVmin

STP

7.2



Cr(VI)

primary



15.0

15.89

L/min

ambient

-5.6



Cr(VI)

collocated



15.0

15.45

L/min

ambient

-2.9

Chesterfield, SC
45-025-0001

VOC
VOC

primary
collocated





no flow readings recorded
no flow readings recorded



-



carbony
1

primary



0.120

0.120

IVmin

ambient

0.0



carbony
1

collocated



0.121

0.125

IVmin

ambient

-3.2



PAH

primary



0.2

0.2290

nrVmin

STP

-12.7



PAH

collocated



0.2

0.2458

nrVmin

STP

-18.6



metals

primary



1.14

1.066

nrVmin

STP

6.9



metals

collocated



1.14

1.076

nrVmin

STP

5.9



Cr(VI)

primary



15.0

15.08

L/min

ambient

-0.5



Cr(VI)

collocated



15.0

16.26

L/min

ambient

-7.7

Percent difference values in red exceed the flow bias MQO of ±10%.

60


-------
Table 29. Flow Audit Results from the Instrument Performance Audits - CY2011

(continued)

Site Identifier
and AQS ID

Method

Precision
Assignment

Channel

Sampler
Reading

Standard Reading

Units

Conditions

Percent
Difference

Decatur, GA

voc

primary





no flow readings recorded



-

13-089-0002

voc

collocated





no flow readings recorded



-



carbony
1

primary



0.123

0.125

17min

STP

-1.6



carbony
1

collocated



0.124

0.121

IVmin

STP

2.5



PAH

primary



0.204

0.2102

nrVmin

STP

-2.9



PAH

collocated



0.198

0.1801

nrVmin

STP

9.9



metals

primary



1.05

1.058

nrVmin

STP

-0.8



metals

collocated



1.03

1.033

nrVmin

STP

-0.3



Cr(VI)

primary

1

2

14.75
14.75

16.72
15.68

L/min
IVmin

ambient
ambient

-11.8
-5.9



Cr(VI)

collocated

1

2

14.44
14.44

16.34
16.88

L/min
IVmin

ambient
ambient

-11.6
-14.5

Hillsborough County, FL
12-057-3002

VOC
carbony
1

primary
primary

1

0.725

no flow readings recorded
0.768 L/min

STP

-5.6



carbony
1

collocated

2

0.725

0.748

IVmin

STP

-3.1



PAH

primary



0.2

0.2279

nrVmin

STP

-12.2



PAH

collocated



0.2

0.2429

nrVmin

STP

-17.7



metals

primary



1.138

1.089

nrVmin

STP

4.5



metals

collocated



1.138

1.085

nrVmin

STP

4.9



Cr(VI)

primary



15.0

15.65

L/min

ambient

-4.2



Cr(VI)

collocated



15.0

14.72

L/min

ambient

1.9

Pinellas County, FL

VOC

primary





no flow readings recorded



-

12-103-0026

VOC

duplicate





no flow readings recorded



-



carbony
1

primary

1

0.670

0.685

IVnrin

STP

-2.2



carbony
1

collocated

2

0.620

0.620

IVnrin

STP

0.0



PAH

primary



0.177

0.1801

nrVmin

STP

-1.7



metals

primary



1.172

1.176

nrVmin

STP

-0.3



Cr(VI)

primary



15.0

15.85

L/min

ambient

-5.4



Cr(VI)

collocated



15.0

16.36

L/min

ambient

-8.3

Percent difference values in red exceed the flow bias MQO of ±10%.

61


-------
Table 30. Flow Audit Results from the Instrument Performance Audits - CY2012

Site Identifier



Precision



Sampler





Percent

and AQS ID

Method

Assignment

Channel

Reading

Standard Reading Units

Conditions

Difference

Horicon, WI

voc

primary



no

flow readings recorded

- flow too low to detect

-

55-027-0001

voc

duplicate



no

flow readings recorded

- flow too low to detect

-



carbonyl

primary

1

0.704

0.695

IVmin

ambient

1.3



carbonyl

duplicate

2

0.705

0.706

IVmin

ambient

-0.1



PAH

primary



0.236

0.226

nrVmin

STP

4.4



PAH

collocated



0.242

0.235

nrVmin

STP

3.0



metals

primary



1.153

1.155

nrVmin

ambient

-0.2



metals

collocated



1.148

1.152

nrVmin

ambient

-0.3



Cr(VI)

primary



15.0

13.99

IVmin

ambient

7.2



Cr(VI)

collocated



15.0

14.67

IVmin

ambient

2.2

Northbrook, IL

VOC

primary



no

flow readings recorded

- flow too low to detect

-

17-031-4201

VOC

duplicate



no

flow readings recorded

- flow too low to detect

-



carbonyl

primary

1

0.375

0.388

IVmin

ambient

-3.4



carbonyl

duplicate

2

0.414

0.444

IVmin

ambient

-6.8



PAH

primary



8.00

7.12

ft'/min

STP

12.4



PAH

collocated





sampler inoperable







metals

primary



40.0

41.46

ft'/min

STP

-3.5



metals

collocated



40.0

41.93

ft'/min

STP

-4.6



Cr(VI)

primary



15.00

14.68

IVmin

ambient

2.2



Cr(VI)

collocated



15.00

14.96

IVmin

ambient

0.3

Deer Park, TX

VOC

primary



no

flow readings recorded

- flow too low to detect

-

48-201-1039

VOC

collocated



no

flow readings recorded

- flow too low to detect

-



carbonyl

primary

2

1.102

1.11

IVmin

STP

-0.7



carbonyl

duplicate

3

1.106

1.11

IVmin

STP

-0.4



PAH

primary



8.49

8.50

ft'/min

STP

-0.1



PAH

collocated



7.60

7.54

ft'/min

STP

0.8



metals

primary



39.87

39.81

ft'/min

STP

0.2



metals

collocated



39.94

39.87

ft'/min

STP

0.2



Cr(VI)

primary



11.98

12.30

IVmin

STP

-2.6



Cr(VI)

collocated



11.99

11.21

IVmin

STP

7.0

Percent difference values in red exceed the flow bias MQO of ±10%.

62


-------
Table 30. Flow Audit Results from the Instrument Performance Audits - CY2012

(continued)

Site Identifier



Precision



Sampler





Percent

and AQS ID

Method

Assignment

Channel

Reading Standard Reading Units

Conditions

Difference

St Louis, MO

VOC

primary



no

flow readings recorded

- flow too low to detect

-

29-510-0085

VOC

duplicate



no

flow readings recorded

- flow too low to detect

-



carbonyl

primary

1

740

822

cc/min

ambient

-2.1 a



carbonyl

duplicate

2

750

840

cc/min

ambient

00



PAH

primary



0.231

0.229

m3/min

ambient

0.9



metals

primary



16.7

17.0

L/min

ambient

-0.3a



metals

collocated



16.7

17.0

L/min

ambient

-2.1a



Cr(VI)

primary



15.1"

14.83

L/min

ambient

1.8



Cr(VI)

collocated



14.0

14.56

L/min

ambient

-3.8



Cr(VI)

primaryc



15.0

15.82

L/min

ambient

-5.2



Cr(VI)

collocatedc



14.98

15.26

L/min

ambient

-1.8

Portland, OR

VOC

primary



no

flow readings recorded

- flow too low to detect

-

41-051-0246

VOC

duplicate



no

flow readings recorded

- flow too low to detect

-



carbonyl

primary



1.2

1.21

L/min

ambient

-0.8



carbonyl

collocated



0.96

0.98

L/min

ambient

-2.0



PAH

primary



222.6

226.1

L/min

STP

-1.5



PAH

collocated



221.4

220.5

L/min

STP

0.4



metals

primary



41.85

40.81

ft3/min

ambient

2.5



metals

collocated



37.17

38.14

ft3/min

ambient

-2.5



Cr(VI)

primary



14.97

15.01

L/min

ambient

-0.3



Cr(VI)

collocated



14.44

14.80

L/min

STP

-2.4

a Flow audits performed with two different flow standards; average of the two audit results reported.
b Average of flow range reported (15.0 - 15.2 L/min).
c Samplers installed to begin operation in July 2012.

Graphical summaries of the flow audit results (mean percent differences within a HAP class) are
presented by site in Figures 19 and 20 for CY2011 and CY2012, respectively.

Non-biased sampler flow rates for carbonyls, PAHs, PMio metals, and hexavalent chromium
samplers are critical for determining sample concentration. Flow rate verification for VOC
samplers is less important to determining concentration, but is important in demonstrating a
representative composite sample is collected over 24 hours.

In CY2011 all sites met the <10% flow bias MQO for metals, 7 of 8 sites met the MQO for
carbonyls, and 6 of 8 sites met the MQO for PAHs and chromium (VI). In CY2012 all sites met
the flow bias MQO for carbonyls, metals, and chromium (VI), and 4 of 5 sites met the MQO for
PAHs. Percent completeness for audited sites in CY2011 and CY2012 are shown in Table 31.

63


-------
Table 31. Percentage of Audited NATTS Sites Meeting the Flow Bias MQO -

CY2011 and CY2012





HAP class



CY

VOCs carbonyls

PAHs

metals

Chromium (VI)

2011

88%

75%

100%

75%

2012

100%

80%

100%

100%

64


-------
carbonyl

PAH

metals

Cr(VI)

Bronx, NY
Rochester, NY
Washington, DC
Richmond, VA
Chesterfield, SC
Decatur, GA
Hillsborough County, FL
Pinellas County, FL

Bronx, NY
Rochester, NY
Washington, DC
Richmond, VA
Chesterfield, SC
Decatur, GA
Hillsborough County, FL
Pinellas County, FL

Bronx, NY
Rochester, NY
Washington, DC
Richmond, VA
Chesterfield, SC
Decatur, GA
Hillsborough County, FL
Pinellas County, FL

Bronx, NY
Rochester, NY
Washington, DC
Richmond, VA
Chesterfield, SC
Decatur, GA
Hillsborough County, FL
Pinellas County, FL

-20

Dashed lines indicate MQO of +/-10%

	1	1	

0	10

Mean Percent Difference

20

~i—

30

Figure 19. Summary of Instrument Performance Flow Audit Results for NATTS Sites CY2011


-------
carbonyl

PAH

metals

Cr(VI)

Horicon, Wl
Northbrook, IL
Deer Park, TX
St Louis, MO
Portland, OR

Horicon, Wl
Northbrook, IL
Deer Park, TX
St Louis, MO
Portland, OR

Horicon, Wl
Northbrook, IL
Deer Park, TX
St Louis, MO
Portland, OR

Horicon, Wl
Northbrook, IL
Deer Park, TX
St Louis, MO
Portland, OR

-10

Dashed lines indicate MQO of +/-10%

	1	1	

0	5

Mean Percent Difference

10

~i—

15

Figure 20. Summary of Instrument Performance Flow Audit Results for NATTS Sites CY2012


-------
2.8 Method Detection Limit (MDL) Data

For CY2011 and CY2012 the AQS database, specifically the ALT_MDL variable within records
having an RD record type, served as the primary source of MDL data. AQS allows the posting
of MDL data in a variety of units, even within chemical classes; thus, for the purposes of this
report, all AQS-acquired MDLs were standardized to ng/m3 for PAHs, metals, and chromium
(VI), and to |ig/m3 for VOCs and carbonyls. Where necessary, conversion from mole fraction
(ppb) assumed conditions at STP.

The MDL results presented in this report are arithmetic means of the AQS-posted ALT_MDL
values. The MDL data for individual sites, in addition to the mean across all sites reporting data,
are indicated in Tables 33 and 34 for CY2011 and Tables 35 and 36 for CY2012. Summary
statistics for MDL data for CY2011 and CY2012 are indicated in Tables 37 and 38, respectively.

Box and whisker plots and complementary scatter plots, indicated in Figures 21 through 30,
illustrate the MDLs for VOCs, carbonyls, metals, chromium (VI), and PAHs, respectively, for
CY2011 and CY2012. Note the log scale of the y-axes in these figures. The MDL MQOs for
each HAP are added to the respective plots (as a red horizontal line) for reference. Laboratories
whose MDLs fell outside of a window defined by 1.5 x IQR in either direction of the box are
identified by circles on the graphical display. Only HAPs for which an MQO is established for
the MDL are included in Figures 21 through 30.

Because ERG serves as the analytical laboratory for numerous NATTS sites (Table 10) for
VOCs, carbonyls, metals, and particularly for chromium (VI) and PAHs, the MDLs summarized
in Tables 33 through 36 and in Figures 21 through 30 reflect a consistency in instrumental
detection limits associated with an analytical laboratory common to multiple sites. Values for
MDL MQOs remained the same for CY2011 compared to CY2010, however, several pollutants
had lower MDL MQOs in CY2012 including: acrolein, formaldehyde, lead, and
trichloroethylene. Most notably, in CY2012 the MDL MQOs for formaldehyde and lead were
lower by an order of magnitude or more compared to CY2011. Only the MDL MQO for carbon
tetrachloride increased from CY2011 to CY2012, an increase of 150%.

MDL values varied widely among sites and frequently exceeded the respective MQOs for
several HAPs. Network-wide, the geometric means met the MDL MQO for the seven HAPs of
primary importance in CY2011 and CY2012 except for acrolein in CY2011 and CY2012, as
evidenced by the ratios of the geometric mean to the MDL MQO being > 1.

The percentages of NATTS sites meeting the MDL MQOs for CY2011 and CY2012 are shown
in Table 32.

Of the sites reporting results in CY2011, only approximately one third met the MDL MQO for
acrolein. In general, the MDL MQOs for benzene, 1,3-butadiene, formaldehyde, and arsenic
were met by 70% or more of sites. All sites met the MDL MQO for chromium (VI) and
naphthalene.

In CY2012, less than 25% of sites met the MDL MQO for acrolein. Approximately half or less
of sites met the MQO for benzene and formaldehyde; approximately 80% of sites met the MDL

67


-------
MQO for 1,3-butadiene and arsenic; and all sites met the MDL MQO for chromium (VI) and
naphthalene.

Table 32. Percentage of Sites Meeting the MDL MQO - CY2011 and CY2012





VOCs

carbonyls

PAHs

metals



CY

Acrolein

Benzene 1,3-Butadiene

Formaldehyde

Naphthalene

Arsenic

Chromium (VI)

2011

35%

78% 70%

89%

100%

85%

100%

2012

23%

41% 81%

52%

100%

78%

100%

Percentage based on N = 27 NATTS sites except for acrolein and chromium (VI) where N = 26.

68


-------
Table 33. Average Method Detection Limits (MDLs) by Site and Overall for VOCs (jig/m3)

and PAHs (ng/m3) - CY2011

Site Description

AQS Site Code

VOCs

BENZ

BUTA

CTET

CLFRM

EDB

DCP

EDC

MECL

TCE1122

Boston, MA

25-025-0042

0.033

0.018

0.053

0.032

0.079

0.042

0.041

0.036

0.261

Underbill, VT

50-007-0007

0.090

0.020

0.150

0.040

0.140

0.110

0.040

0.030

0.160

Providence, RI

44-007-0022

0.033

0.018

0.053

0.032

0.079

0.042

0.041

0.036

0.261

Bronx, NY

36-005-0080

0.032

0.044

0.063

0.049

0.077

0.092

0.040

0.035

0.069

Rochester, NY

36-055-1007

0.032

0.044

0.063

0.049

0.077

0.092

0.040

0.035

0.069

Washington, DC

11-001-0043

0.032

0.022

0.095b

0.098

0.154

0.092

0.081

0.069

0.206

Richmond, VA

51-087-0014

0.080

0.114

0.195

0.189

0.240

0.283

0.105

0.189

0.234

Chesterfield, SC

45-025-0001

0.045

0.168

0.195

0.112

0.230

0.088

0.125

0.094

0.158

Decatur, GA

13-089-0002

0.102

0.081

0.075b

0.182

0.305

0.332

0.327

7.610

0.365

Grayson Lake, KY

21-043-0500

0.093

0.015

0.151

0.044

0.138

0.106

0.036

0.035

0.165

Hillsborough Cty, FL

12-057-3002

0.029

0.033

0.044

0.034

0.046

0.046

0.036

0.017

0.027

Pinellas Cty, FL

12-103-0026

0.029

0.033

0.044

0.034

0.046

0.046

0.036

0.017

0.027

Dearborn, MI

26-163-0033

0.104a

0.020a

0.153

0.050

0.154

0.111

0.044

0.093

0.203

Horicon, WI

55-027-0001

0.319

0.221

0.629

0.488

0.768

0.462

0.404

0.347

0.686

Northbrook, IL

17-031-4201

0.105

0.060b

0.151

0.044

0.138

0.106

0.036

0.035

0.165

Deer Park, TX

48-201-1039

-

-

-

-

-

-

-

-

-

Flarrison Cty, TX

48-203-0002

-

-

-

-

-

-

-

-

-

St. Louis, MO

29-510-0085

0.093

0.015

0.151

0.044

0.138

0.106

0.036

0.035

0.165

Bountiful, UT

49-011-0004

0.093

0.015

0.151

0.044

0.138

0.106

0.036

0.035

0.165

Grand Junction, CO

08-077-0017

0.093

0.015

0.151

0.044

0.138

0.106

0.036

0.035

0.165

Phoenix, AZ

04-013-9997

0.093

0.015

0.151

0.044

0.138

0.106

0.036

0.035

0.165

Los Angeles, CA

06-037-1103

0.160

0.088

0.126

0.098

-

-

-

0.347

-

Rubidoux, CA

06-065-8001

0.160

0.088

0.126

0.098

-

-

-

0.347

-

San Jose, CA

06-085-0005

0.097b

0.117b

0.085b

0.066

0.077

-

0.404

0.347

-

La Grande, OR

41-061-0119

0.132a

0.158b

0.197b

0.259a

-

0.245

-

0.225

-

Portland, OR

41-051-0246

0.128

0.165b

0.186b

0.244

-

0.231

-

0.216

-

Seattle, WA

53-033-0080

0.093

0.015

0.151

0.044

0.138

0.106

0.036

0.035

0.165



Arithmetic Mean

0.092

0.064

0.144

0.098

0.164

0.139

0.096

0.413

0.194



Geometric Mean

0.076

0.042

0.119

0.070

0.130

0.112

0.061

0.081

0.153



Median

0.093

0.033

0.151

0.049

0.138

0.106

0.040

0.036

0.165



Standard Deviation

0.062

0.061

0.113

0.105

0.153

0.106

0.121

1.504

0.141



MQO

0.130

0.100

0.067

0.500

-

--

-

-

-

Note: Shaded cells indicate that all reported MDLs are above the MQO and unshaded cells indicate all reported MDLs are below the MQO and
with the following exceptions indicated:

a.	Some reported MDLs (less than 5%) are above the MQO

b.	Between 33% and 75% of the MDLs are above the MQO

69


-------
Table 33. Average Method Detection Limits (MDLs) by Site and Overall for VOCs (jig/m3)

and PAHs (ng/m3) - CY2011 (continued)

Site Description

AQS Site Code

VOCs

PAHs

PERC

TCE

vc

cDCPEN

tDCPEN

ACRO ACRY

NAPH

BaP

Boston, MA

25-025-0042

0.069

0.048

0.017

0.040

0.026

0.123

0.260

0.107

0.060

Underbill, VT

50-007-0007

0.120

0.130

0.020

0.100

0.110

	

0.030

0.139

0.078

Providence, RI

44-007-0022

0.069

0.048

0.017

0.040

0.026

0.123

0.260

0.159

0.090

Bronx, NY

36-005-0080

0.068

0.054

0.026

0.045

0.045

0.069

—

0.126

0.071

Rochester, NY

36-055-1007

0.068

0.054

0.026

0.045

0.045

0.069

-

0.187

0.105

Washington, DC

11-001-0043

0.101

0.054

0.051

0.091

0.091

0.045

0.033

0.168

0.094

Richmond, VA

51-087-0014

0.191

0.107

0.136

0.175

0.068

0.428

0.131

0.185

0.104

Chesterfield, SC

45-025-0001

0.142

0.140

0.089

0.086

0.082

0.085

—

0.156

0.087

Decatur, GA

13-089-0002

0.119

0.236

0.070

0.277

0.251

0.043

—

0.142

0.080

Grayson Lake, KY

21-043-0500

0.122

0.134

0.020

0.100

0.113

0.115

0.026

0.127

0.071

Hillsborough Cty, FL

12-057-3002

0.034

0.048

0.041

0.082

0.045

0.087

0.022

0.140

0.078

Pinellas Cty, FL

12-103-0026

0.034

0.048

0.041

0.082

0.045

0.087

0.022

0.135

0.075

Dearborn, MI

26-163-0033

0.134a

0.142

0.026a

0.109

0.126

0.115

0.041

0.124

0.070

Horicon, WI

55-027-0001

0.678

0.537

0.255

0.454

0.454

0.229

—

0.133

0.136

Northbrook, IL

17-031-4201

0.122

0.134

0.020

0.100

0.113

0.115

0.026

0.147

0.072

Deer Park, TX

48-201-1039

-

-

-

-

-

-

-

0.162

0.091

Flarrison Cty, TX

48-203-0002

—

—

—

—

—

—

—

—

—

St. Louis, MO

29-510-0085

0.122

0.134

0.020

0.100

0.113

0.115

—

0.133

0.074

Bountiful, UT

49-011-0004

0.122

0.134

0.020

0.100

0.113

0.115

0.026

0.199

0.111

Grand Junction, CO

08-077-0017

0.122

0.134

0.020

0.100

0.113

0.115

0.026

0.167

0.094

Phoenix, AZ

04-013-9997

0.122

0.134

0.020

0.100

0.113

0.115

0.026

0.109

0.061

Los Angeles, CA

06-037-1103

0.068

0.107

-

0.454

0.454

0.687

-

0.130

0.072

Rubidoux, CA

06-065-8001

0.068

0.107

—

0.454

0.454

0.687

—

0.164

0.092

San Jose, CA

06-085-0005

0.046

0.072

0.255

0.454

0.454

0.403

0.217

0.140

0.078

La Grande, OR

41-061-0119

0.248b

0.285a

0.131b

-

—

0.092

-

1.035

0.262

Portland, OR

41-051-0246

0.235b

0.269

0.127b

—

—

0.092

—

2.292a

0.233

Seattle, WA

53-033-0080

0.122

0.134

0.020

0.100

0.113

0.115

0.026

0.111

0.063



Arithmetic Mean

0.134

0.137

0.064

0.160

0.155

0.178

0.078

0.262

0.096



Geometric Mean

0.106

0.111

0.041

0.117

0.107

0.129

0.047

0.172

0.089



Median

0.122

0.134

0.026

0.100

0.113

0.115

0.026

0.141

0.079



Standard Deviation

0.126

0.106

0.072

0.146

0.148

0.183

0.091

0.450

0.048



MQO

0.170

0.500

0.110

-

--

0.100

--

29.0

0.910

Note: Shaded cells indicate that all reported MDLs are above the MQO and unshaded cells indicate all reported MDLs are below the MQO and
with the following exceptions indicated:

a.	Some reported MDLs (less than 5%) are above the MQO

b.	Between 33% and 75% of the MDLs are above the MQO

70


-------
Table 34. Average Method Detection Limits (MDLs) by Site and Overall for carbonyls
(|ig/m3), metals (ng/m3), and chromium (VI) (ng/m3) - CY2011

Site Description

AQS Site Code

carbonyls

metals



FORM

ACET

As

Be

Cd

Pb

Mn

Ni

CrVI

Boston, MA

25-025-0042

0.103

0.069

0.060a

0.017

0.019

0.159

0.205

0.490a

0.004

Underbill, VT

50-007-0007

0.019

0.013

0.143

0.013

0.010

0.037

0.203

0.896

0.004

Providence, RI

44-007-0022

0.088

0.026

0.095

0.054

0.102

0.606

0.035

0.054

0.004

Bronx, NY

36-005-0080

0.018

0.018

0.191b

0.054

0.038

0.025

0.038

0.089

0.004

Rochester, NY

36-055-1007

0.018

0.018

0.191b

0.054

0.038

0.025

0.038

0.089

0.004

Washington, DC

11-001-0043

0.023

0.011

1.664

	

	

	

—

—

0.004

Richmond, VA

51-087-0014

0.097

0.180

0.023

0.013

0.017

0.064

0.029

0.143

0.004

Chesterfield, SC

45-025-0001

0.251

0.221

0.031

0.001

0.001

0.003

0.002

0.003

0.004

Decatur, GA

13-089-0002

1.165c

1.165c

0.265b

0.010

0.300

0.010

0.011

0.026

0.004

Grayson Lake, KY

21-043-0500

0.013

0.012

0.159

0.041

0.040

0.044

0.188

0.836

0.004

Hillsborough Cty, FL

12-057-3002

0.011

0.011

0.460

0.200

0.150

1.040

0.140

0.920

0.004

Pinellas Cty, FL

12-103-0026

0.011

0.011

-

-

-

-

-

-

0.004

Dearborn, MI

26-163-0033

0.010

0.009

0.039

0.014

0.028

-

0.298

0.160

0.004

Horicon, WI

55-027-0001

0.064

0.054

0.029

0.010

0.016

0.034

0.094

0.093

0.004

Northbrook, IL

17-031-4201

0.011

0.011

0.054

0.007

0.009

0.111

0.138

0.388

0.004

Deer Park, TX

48-201-1039

-

-

0.064

0.060

0.040

0.020

0.264

0.181

0.005

Flarrison Cty, TX

48-203-0002

-

-

0.069

0.060

0.040

0.020

0.272

0.188

0.005

St. Louis, MO

29-510-0085

0.012

0.011

0.124

0.008

0.009

0.045

0.421a

0.872

0.004

Bountiful, UT

49-011-0004

0.011

0.010

0.182

0.010

0.012

0.020

0.223

1.261

0.004

Grand Junction, CO

08-077-0017

0.013

0.013

0.066

0.159

0.152

0.102

0.089

0.080

0.004

Phoenix, AZ

04-013-9997

0.015

0.014

0.151

0.012

0.010

0.035

0.209

0.956

0.004

Los Angeles, CA

06-037-1103

0.123

0.180

0.100

0.100

0.100

0.100

0.100

0.100

0.020

Rubidoux, CA

06-065-8001

0.123

0.180

0.091

0.091

0.092

0.091

0.092

0.091

0.020

San Jose, CA

06-085-0005

0.068

0.075

0.160

0.010

0.010

0.020

0.200

1.130

-

La Grande, OR

41-061-0119

0.131

0.033

0.035

0.004

0.035

0.351

0.351

0.351

0.035

Portland, OR

41-051-0246

0.118

0.031

0.031

0.003

0.034

0.344

0.344

0.344

0.035

Seattle, WA

53-033-0080

0.015

0.014

0.058

0.015

0.021

0.180

0.208

0.506

0.004



Arithmetic Mean

0.101

0.096

0.174

0.041

0.053

0.145

0.168

0.410

0.008



Geometric Mean

0.038

0.031

0.096

0.020

0.028

0.059

0.109

0.213

0.005



Median

0.019

0.018

0.093

0.014

0.034

0.045

0.188

0.188

0.004



Standard Deviation

0.230

0.232

0.318

0.051

0.067

0.238

0.116

0.395

0.009



MQO

0.980

0.450

0.230

0.420

0.560

150

5.00

2.10

0.080

Note: Shaded cells indicate that all reported MDLs are above the MQO and unshaded cells indicate all reported MDLs are below the MQO and

with the following exceptions indicated:

a.	Some reported MDLs (less than 5%) are above the MQO

b.	Between 33% and 75% of the MDLs are above the MQO

c.	Over 95% of the reported MDLs are above the MQO

71


-------
Table 35. Average Method Detection Limits (MDLs) by Site and Overall for VOCs (jig/m3)

and PAHs (ng/m3) - CY2012













VOCs





























description

uitc coae























BENZ

BUTA

CTET

CLFRM

EDB

DCP

EDC

MECL

TCE1122

Boston, MA

25-025-0042

0.037

0.019

0.060

0.051

0.094

0.055

0.062

0.040

0.214

Underbill, VT

50-007-0007

0.190

0.020

0.150

0.070

0.130

0.090

0.060

0.080

0.120

Providence, RI

44-007-0022

0.043

0.024

0.105

0.080

0.149

0.076

0.074

0.055

0.178

Bronx, NY

36-005-0080

0.016

0.013

0.044

0.029

0.054

0.037

0.020

0.028

0.021

Rochester, NY

36-055-1007

0.016

0.013

0.044

0.029

0.054

0.037

0.020

0.028

0.021

Washington, DC

11-001-0043

0.061

0.022

0.126

0.049

0.077

0.092

0.081

0.104

0.137

Richmond, VA

51-087-0014

0.134

0.148

0.189

0.146

0.691

0.332

0.263

0.326

0.686

Chesterfield, SC

45-025-0001

0.115

0.221

0.226

0.205

0.200

0.162

0.133

0.132

0.185

Decatur, GA

13-089-0002

0.117

0.068

0.082

0.171

0.312

0.270

0.220

6.942

0.387

Grayson Lake, KY

21-043-0500

0.195

0.024

0.151

0.068

0.131

0.088

0.065

0.080

0.123

Hillsborough Cty, FL

12-057-3002

0.019

0.029

0.031

0.024

0.038

0.023

0.016

0.024

0.034

Pinellas Cty, FL

12-103-0026

0.019

0.029

0.031

0.024

0.038

0.023

0.016

0.024

0.034

Dearborn, MI

26-163-0033

0.195

0.024

0.151

0.068

0.131

0.088

0.065

0.080

0.123

Horicon, WI

55-027-0001

0.319

0.221

0.629

0.488

0.768

0.462

0.404

0.347

0.686

Northbrook, IL

17-031-4201

0.149b

0.066b

0.151

0.068

0.131

0.088

0.065

0.080

0.123

Deer Park, TX

48-201-1039

-

-

-

--

-

--

-

-

-

Flarrison Cty, TX

48-203-0002

-

-

-

--

-

--

-

-

-

St. Louis, MO

29-510-0085

0.195

0.024

0.151

0.068

0.131

0.088

0.065

0.080

0.123

Bountiful, UT

49-011-0004

0.195

0.024

0.151

0.068

0.131

0.088

0.065

0.101

0.123

Grand Junction, CO

08-077-0017

0.195

0.024

0.151

0.068

0.131

0.088

0.065

0.258

0.123

Phoenix, AZ

04-013-9997

0.195

0.024

0.151

0.068

0.131

0.088

0.065

0.080

0.123

Los Angeles, CA

06-037-1103

0.160

0.088

0.126

0.098

--

-

-

0.347

-

Rubidoux, CA

06-065-8001

0.160

0.088

0.126

0.098

--

-

-

0.347

-

San Jose, CA

06-085-0005

0.083b

0.042

0.046

0.071

0.038

-

0.032

0.280

--

La Grande, OR

41-061-0119

0.128

0.088

0.063

0.244

--

0.230

--

0.609

--

Portland, OR

41-051-0246

0.1343

0.093b

0.0663

0.2593

--

0.246

--

0.184

-

Seattle, WA

53-033-0080

0.195

0.024

0.151

0.068

0.131

0.088

0.065

0.080

0.123



Arithmetic Mean

0.131

0.058

0.134

0.107

0.176

0.129

0.091

0.429

0.184



Geometric Mean

0.098

0.040

0.106

0.080

0.122

0.095

0.064

0.124

0.122



Median

0.134

0.024

0.126

0.068

0.131

0.088

0.065

0.080

0.123



Standard Deviation

0.078

0.059

0.116

0.102

0.195

0.111

0.094

1.365

0.189



MQO

0.130

0.100

0.170

0.500

--

-

-

-

-

Note: Shaded cells indicate that all reported MDLs are above the MQO and unshaded cells indicate all reported MDLs are below the MQO and
with the following exceptions indicated:

a.	Some reported MDLs (less than 5%) are above the MQO

b.	Between 33% and 75% of the MDLs are above the MQO

72


-------
Table 35. Average Method Detection Limits (MDLs) by Site and Overall for VOCs (jig/m3)

and PAHs (ng/m3) - CY2012 (continued)

Site Description

AQS Site Code

VOCs

PAHs

PERC

TCE

vc

cDCPEN

tDCPEN

ACRO

ACRY

NAPH

BaP

Boston, MA

25-025-0042

0.108

0.068

0.023

0.058

0.058

0.174

0.274

0.141

0.053

Underbill, VT

50-007-0007

0.140

0.120

0.030

0.070

0.070

-

0.040

0.193

0.073

Providence, RI

44-007-0022

0.179b

0.090

0.035

0.070

0.106

0.265

0.284

0.217

0.082

Bronx, NY

36-005-0080

0.027

0.027

0.013

0.036

0.032

0.085

-

0.171

0.064

Rochester, NY

36-055-1007

0.027

0.027

0.013

0.036

0.032

0.085

-

0.183

0.069

Washington, DC

11-001-0043

0.136

0.054

0.077

0.045

0.045

0.069

0.043

0.232

0.087

Richmond, VA

51-087-0014

0.427

0.360

0.171

0.331

0.259

0.275

0.373

0.247

0.093

Chesterfield, SC

45-025-0001

0.183

0.193

0.146

0.122

0.082

0.108

-

0.180

0.068

Decatur, GA

13-089-0002

0.139

0.314

0.091

0.287

0.218

0.081b

-

0.090

0.034

Grayson Lake, KY

21-043-0500

0.136

0.118

0.028

0.068

0.073

0.137

0.043

0.153

0.058

Hillsborough Cty, FL

12-057-3002

0.027

0.021

0.033

0.023

0.068

0.066

0.013

0.197

0.074

Pinellas Cty, FL

12-103-0026

0.027

0.021

0.033

0.023

0.068

0.066

0.013

0.204

0.067

Dearborn, MI

26-163-0033

0.136

0.118

0.028

0.068

0.073

0.137

0.043

0.193

0.064

Horicon, WI

55-027-0001

0.678

0.537

0.255

0.454

0.454

0.229

-

0.140

0.280

Northbrook, IL

17-031-4201

0.136

0.118

0.028

0.068

0.073

0.137

0.043

0.182

0.059

Deer Park, TX

48-201-1039

-

--

--

--

--

--

--

0.179

0.067

Flarrison Cty, TX

48-203-0002

-

--

--

--

--

--

--

--

-

St. Louis, MO

29-510-0085

0.136

0.118

0.028

0.068

0.073

0.137

0.043

0.194

0.062

Bountiful, UT

49-011-0004

0.136

0.118

0.028

0.068

0.073

0.137

0.043

0.268

0.101

Grand Junction, CO

08-077-0017

0.136

0.118

0.028

0.068

0.073

0.137

0.043

0.300

0.088

Phoenix, AZ

04-013-9997

0.136

0.118

0.028

0.068

0.073

0.137

0.043

0.166

0.062

Los Angeles, CA

06-037-1103

0.068

0.107

-

0.454

0.454

0.687

-

0.234

0.065

Rubidoux, CA

06-065-8001

0.068

0.107

-

0.454

0.454

0.687

-

0.220

0.074

San Jose, CA

06-085-0005

0.031

0.075

0.015

0.454

0.454

0.399

0.024

0.184

0.069

La Grande, OR

41-061-0119

0.136

0.269

0.102

-

-

0.092

-

3.042

0.224

Portland, OR

41-051-0246

0.142a

0.286

0.107a

-

-

0.092

-

4.208

1.017a

Seattle, WA

53-033-0080

0.136

0.118

0.028

0.068

0.073

0.137

0.043

0.146

0.055



Arithmetic Mean

0.145

0.145

0.059

0.150

0.149

0.190

0.088

0.456

0.120



Geometric Mean

0.106

0.105

0.041

0.093

0.102

0.147

0.051

0.235

0.083



Median

0.136

0.118

0.028

0.068

0.073

0.137

0.043

0.193

0.069



Standard Deviation

0.137

0.122

0.061

0.160

0.152

0.172

0.113

0.948

0.190



MQO

0.170

0.200

0.110

-

-

0.090

-

29.0

0.910

Note: Shaded cells indicate that all reported MDLs are above the MQO and unshaded cells indicate all reported MDLs are below the MQO and
with the following exceptions indicated:

a.	Some reported MDLs (less than 5%) are above the MQO

b.	Between 33% and 75% of the MDLs are above the MQO

73


-------
Table 36. Average Method Detection Limits (MDLs) by Site and Overall for carbonyls
(|ig/m3), metals (ng/m3), and chromium (VI) (ng/m3) - CY2012

Site Description

AQS Site Code

carbonyls

metals



FORM

ACET

As

Be

Cd

Pb

Mn

Ni

CrVI

Boston, MA

25-025-0042

0.110

0.056

0.062

0.005

0.007

0.124

0.363

0.454

0.004

Underbill, VT

50-007-0007

0.030

0.017

0.170

0.020

0.010

0.070

0.320

0.400

0.004

Providence, RI

44-007-0022

0.305b

0.054

0.091

0.055

0.034

0.533

0.121

0.059

0.003

Bronx, NY

36-005-0080

-

-

0.140

0.058

0.036

0.025

0.028

0.053

0.003

Rochester, NY

36-055-1007

-

-

0.141

0.057

0.037

0.026

0.027

0.051

0.004

Washington, DC

11-001-0043

--

-

-

-

-

--

--

--

0.004

Richmond, VA

51-087-0014

0.073b

0.108

0.020

0.016

0.020

0.054

0.026

0.155

0.003

Chesterfield, SC

45-025-0001

0.240

0.280

0.030

0.001

0.001

0.003

0.002

0.003

0.003

Decatur, GA

13-089-0002

1.178

1.163c

0.324

0.015

0.071

0.019

0.017

0.027

0.003

Grayson Lake, KY

21-043-0500

0.007

0.007

0.170

0.020

0.010

0.070

0.320

0.400

0.003

Hillsborough Cty, FL

12-057-3002

0.012

0.012

0.460

0.200

0.150

1.040

0.140

0.920

0.003

Pinellas Cty, FL

12-103-0026

0.011

0.011

0.460

0.200

0.150

1.040

0.140

0.920

0.004

Dearborn, MI

26-163-0033

0.010

0.010

0.040

0.012

0.030

-

0.301

0.161

0.004

Horicon, WI

55-027-0001

0.062

0.070

0.029

0.010

0.016

0.034

0.094

0.093

0.004

Northbrook, IL

17-031-4201

0.017

0.017

0.065

0.005

0.007

0.130

0.379

0.474

0.004

Deer Park, TX

48-201-1039

-

-

0.065

0.056

0.040

0.019

0.237

0.158

0.004

Flarrison Cty, TX

48-203-0002

-

-

0.068

0.051

0.043

0.017

0.218

0.141

0.004

St. Louis, MO

29-510-0085

0.011

0.011

0.170

0.020

0.010

0.070

0.320

0.400

0.004

Bountiful, UT

49-011-0004

0.011

0.010

0.192

0.020

0.010

0.082

0.357

0.452

0.003

Grand Junction, CO

08-077-0017

0.012

0.012

0.083

0.253

0.075

0.046

0.094

0.070

0.003

Phoenix, AZ

04-013-9997

0.015

0.014

0.173

0.020

0.010

0.073

0.324

0.407

0.004

Los Angeles, CA

06-037-1103

0.123

0.180

-

-

-

-

-

-

0.007

Rubidoux, CA

06-065-8001

0.123

0.180

—

—

—

—

—

—

0.007

San Jose, CA

06-085-0005

0.067b

0.073

0.170

0.020

0.010

0.070

0.320

0.400

	

La Grande, OR

41-061-0119

0.136

0.033

0.034

0.003

0.034

0.346

0.350

0.345

0.035

Portland, OR

41-051-0246

0.124c

0.032

0.034

0.003

0.034

0.347

0.350

0.345

0.035

Seattle, WA

53-033-0080

0.016

0.014

0.065

0.005

0.007

0.129

0.377

0.472

0.004



Arithmetic Mean

0.122

0.107

0.136

0.047

0.035

0.190

0.218

0.307

0.006



Geometric Mean

0.044

0.036

0.095

0.020

0.021

0.076

0.137

0.183

0.004



Median

0.046

0.025

0.087

0.020

0.025

0.070

0.269

0.345

0.004



Standard Deviation

0.249

0.246

0.123

0.069

0.040

0.298

0.138

0.252

0.009



MQO

0.080

0.450

0.230

0.420

0.560

15

5.00

2.10

0.080

Note: Shaded cells indicate that all reported MDLs are above the MQO and unshaded cells indicate all reported MDLs are below the MQO and

with the following exceptions indicated:

a.	Some reported MDLs (less than 5%) are above the MQO

b.	Between 33% and 75% of the MDLs are above the MQO

c.	Over 95% of the reported MDLs are above the MQO

74


-------
Table 37. Summary Statistics for Method Detection Limits across All Reporting NATTS

Laboratories - CY2011





VOCs



carbonyls

PAHs

metals





Acrolein

Benzene

1,3-Butadiene

Formaldehyde

Naphthalene

Arsenic

Chromium

MDL

(jig/m3)

(jig/m3)

(jig/m3)

(jig/m3)

(ng/m3)

(ng/m3)

(VI) (ng/m3)

Geometric
Mean

0.129

0.076

0.042

0.038

0.172

0.096

0.005

Arithmetic
Mean

0.178

0.092

0.064

0.101

0.262

0.174

0.008

Standard
Deviation

0.183

0.062

0.061

0.230

0.450

0.318

0.009

Minimum

0.043

0.029

0.015

0.010

0.107

0.023

0.004

Median

0.115

0.093

0.033

0.019

0.141

0.093

0.004

Maximum

0.687

0.319

0.221

1.165

2.292

1.664

0.035

MQO

0.10

0.13

0.10

0.98

29.0

0.23

0.08

Ratio of















Geometric

1.3

0.58

0.42

0.038

0.0059

0.42

0.07

Mean to MQO















Table 38. Summary Statistics for Method Detection Limits across All Reporting NATTS

Laboratories - CY2012





VOCs



carbonyls

PAHs

metals





Acrolein

Benzene

1,3-Butadiene

Formaldehyde

Naphthalene

Arsenic

Chromium (VI)

MDL

(jtg/m3)

(ng/m3)

(ng/m3)

(ng/m3)

(ng/m3)

(ng/m3)

(ng/m3)

Geometric
Mean

0.147

0.098

0.040

0.044

0.235

0.095

0.004

Arithmetic
Mean

0.190

0.131

0.058

0.122

0.456

0.136

0.006

Standard
Deviation

0.172

0.078

0.059

0.249

0.948

0.123

0.009

Minimum

0.066

0.016

0.013

0.007

0.090

0.020

0.003

Median

0.137

0.134

0.024

0.046

0.193

0.087

0.004

Maximum

0.687

0.319

0.221

1.178

4.208

0.460

0.035

MQO

0.09

0.13

0.10

0.08

29.0

0.23

0.08

Ratio of















Geometric

1.6

0.75

0.40

0.5

0.0081

0.41

0.06

Mean to MQO















75


-------
ro

E

OJ

o
 1.5 x IQR



: B
~







TCE VC ACRO

Figure 21. Distribution of VOCs Average Method Detection Limits for NATTS Data -

CY2011

76


-------
1 -

.5-

ro

CD

3 .25 H

o

=0 .H

<1)

-t—'

CD
Q

o .05 H

 1.5 x IQR

HAP

Figure 22. Distribution of VOCs Average Method Detection Limits for NATTS Data -

CY2012

77


-------
Figure 23. Distribution of Carbonyls Average Method Detection Limits - CY2011

78


-------
Figure 24. Distribution of Carbonyls Average Method Detection Limits - CY2012

79


-------
.4

.3

co
E

c

¦¦£ 2

c

o
—'

o
d)

"S3
Q

T3
O
-C

o>

.1

MQO = 29.0

Observations at
1.035 and 2.292

MQO = 0.91

Circles indicate MDLs > 1.5 x IQR

NAPH

BaP

HAP

Figure 25. Distribution of PAHs Average Method Detection Limits - CY2011

80


-------
.75-

c

o
—'

o
d)

"S3
Q

T3
O
-C

o>

.5

co
E

c

- .25-

.1-

.05-

MQO = 29.0
Observations at
3.042 and 4

Observation
at 1.017

MQO = 0.91

NAPH

BaP

Circles indicate MDLs > 1.5 x IQR

HAP

Figure 26. Distribution of PAHs Average Method Detection Limits - CY2012

81


-------
o


2.5

n

O)

.25

.1

o> .025

.01

.0025

.001

i

o
o

As	Be

Circles indicate MDLs > 1.5 x IQR

Cd

MQO = 150 MQO = 5

O
o
0

Pb

Mn

Ni

HAP

Figure 27. Distribution of Metals Average Method Detection Limits - CY2011

82


-------
ro

a>

T3
O

 1.5 x IQR

MQO = 15
o

o
o

MQO = 5

Cd	Pb

HAP

Mn

Ni

Figure 28. Distribution of Metals Average Method Detection Limits - CY2012

83


-------
.05-

	





co
E
B)

c



o

! .025-









£=
O



o

o
 1,5 x IQR

HAP

Figure 29. Distribution of Chromium (VI) Average Method Detection Limits - CY2011

84


-------


.008-

.007-

MQO = 0.08
Two observations
at 0.035

? .006-

o
<1)
•+—>


.005-

.004-

.003

Circles indicate MDLs > 1.5 x IQR

CrVI
HAP

Figure 30. Distribution of Chromium (VI) Average Method Detection Limits - CY2012

85


-------
3.0 SUMMARY

A summary of the quality assurance results for the seven HAPs of primary importance -
acrolein, benzene, 1,3-butadiene, formaldehyde, naphthalene, arsenic, and chromium (VI) is
presented in Table 39.

Table 39. Summary of NATTS Quality Assurance Results - Percentage of Sites Meeting
Measurement Quality Objectives in CY2011 and CY2012







VOCs



carbonyls

PAHs

metals



Data Quality

Calendar





1,3-







Chromium

Indicator

Year

Acrolein

Benzene

Butadiene

Formaldehyde

Naphthalene

Arsenic

(VI)



2011

77%

93%

85%

78%

96%

96%

100%

Completeness



















2012

81%

89%

89%

78%

89%

96%

96%

Overall

2011

37%

84%

79%

100%

63%

47%

54%

Precision

2012

47%

90%

84%

95%

88%

56%

63%

Laboratory

2011

77%

85%

85%

87%

40%

90%



















Bias

2012

--

--

--

--

83%

93%

100%



2011







88%

75%

100%

75%

Field Biasa





Not applicable













2012







100%

80%

100%

100%

Method

2011

35%

78%

70%

89%

100%

85%

100%

Dection Limit

2012

23%

41%

81%

52%

100%

78%

100%

a Field bias was determined by HAPs class - carbonyls, PAHs, and metals - not for specific HAPs (except chromium (VI))

The following summary observations are provided:

1. Completeness: Data completeness across the entire NATTS network met the MQO in
both CY2011 and CY2012: both the mean and median network-wide completeness
for all seven priority HAPs was greater than 85% in both CY2011 and CY2012.
Median network-wide completeness is indicated in Table 40 for CY2011 and
CY2012.

The percentage of sites that met the completeness MQO for CY2011 and CY2012
was 85% or greater for all of the seven HAPs of primary importance except acrolein
and formaldehyde, for which approximately 80% of sites met the MQO.

Table 40. Median Completeness for the Seven HAPs of Primary Importance for CY2011

and CY2012





VOCs



carbonyls

PAHs

metals



Acrolein

Benzene

1,3-Butadiene

Formaldehyde

Naphthalene

Arsenic

Chromium (VI)

MQO

>85%

>85%

>85%

>85%

>85%

>85%

>85%

CY2011

97%

97%

97%

95%

97%

97%

98%

CY2012

95%

95%

95%

97%

95%

97%

97%

86


-------
2.	Analytical and Overall Precision: For CY2011 the network mean analytical precision
met the MQO of 15% for carbonyls, PAHs, and chromium (VI), for all metals except
beryllium, and for 9 of the 16 VOCs for reported concentrations equal to or above
MDLs.

For CY2012 the network mean analytical precision met the precision MQO for all
HAPs except for acrylonitrile and beryllium.

As is expected given the additional variability contribution of sample collection,
overall precision for CY2011 showed much greater variability than the analytical
precision: the network mean overall precision met the MQO for carbonyls, 1 PAH,
1 metal, and 5 of 16 VOCs; the MQO was not met for chromium (VI).

As in CY2011, CY2012 overall precision showed greater variability than CY2012
analytical precision. The network mean overall precision met the MQO for
carbonyls, 1 PAH, and 5 of 16 VOCs; the MQO was not met for any of the metals or
for chromium (VI).

In CY2011, all sites met the precision MQO for formaldehyde and less than 85% of
sites met the MQO for six of the seven representative HAPs, with less than half of
sites meeting the MQO for acrolein and arsenic. In CY2012, more than 85% of sites
met the precision MQO for benzene, formaldehyde, and naphthalene with the
remaining four HAPs showing 84% or less of sites meeting the MQO. As in
CY2011, less than 50% of sites met the precision MQO for acrolein in CY2012.

3.	Laboratory Bias: Percentages of NATTS laboratories meeting the bias MQO for
acrolein, benzene and 1,3-butadiene were 77%, 85%, and 85%, respectively. All
NATTS laboratories but one met the MQO for arsenic and formaldehyde. For the
five NATTS laboratories reporting PAH results, only two laboratories met the MQO
for naphthalene. Specifically, acceptable measurement bias was difficult to obtain for
acrolein and naphthalene; the mean absolute percent bias across all participating
laboratories was 29.5% and 26.0%, respectively. The CY2011 PAH PT results
should be interpreted with caution, as it appears that the target value may have been
biased high: all reported results but one showed a negative bias.

For the three HAPs of primary importance that were spiked for the CY2012 PT
(naphthalene, arsenic, and chromium (VI)), all NATTS laboratories met the MQO for
chromium (VI) and all but one laboratory met the MQO for both naphthalene and
arsenic.

4.	Field Bias: Sampler flows measured during IPAs conducted at NATTS field sites
indicated less than 10% absolute difference from the sampler settings with few
exceptions for both CY2011 and CY2012. The most frequent MQO exceedances
occurred for PAH and chromium (VI) samplers.

In CY2011 all sites met the <10% flow bias MQO for metals, 7 of 8 sites met the
MQO for carbonyls, and 6 of 8 sites met the MQO for PAHs and chromium (VI). In
CY2012 all sites met the flow bias MQO for carbonyls, metals, and chromium (VI),
and 4 of 5 sites met the MQO for PAHs.

87


-------
5. Method Detection Limits: MDL values varied widely among sites and frequently
exceeded the respective MQOs for several HAPs. For many HAPs the overall
network geometric mean value fell within the MQO threshold when all sites were
considered together, except for acrolein in CY2011 and CY2012. The ratios of the
network geometric means to the corresponding MQOs for the seven HAPs of primary
importance are indicated in Table 41.

Of the sites reporting results in CY2011, approximately one third met the MDL MQO
for acrolein. In general, the MDL MQOs for benzene, 1,3-butadiene, formaldehyde,
and arsenic were met by 70% or more of sites. All sites met the MDL MQO for
naphthalene and chromium (VI).

In CY2012, less than one-quarter of all sites met the MDL MQO for acrolein.
Approximately half or less of sites met the MQO for benzene and formaldehyde;
approximately 80% of sites met the MDL MQO for 1,3-butadiene and arsenic; and all
sites met the MDL MQO for naphthalene and chromium (VI).

Table 41. Ratio of the MDL Network Geometric Mean to the MQO for the Seven HAPs of

Primary Importance for CY2011 and CY2012





VOCs



carbonyls

PAHs

metals





Acrolein

Benzene

1,3-
Butadiene

Formaldehyde

Naphthalene

Arsenic

Chromium
(VI)

CY2011 MQO

<0.10 |xg/m3

< 0.13 |xg/m3

< 0.10 |xg/m3

< 0.98 (xg/m3

< 29 ng/m3

< 0.23 ng/m3

<0.08 ng/m3

ratio of
network
geometric mean
MDL to MQO
(CY2011)

1.3

0.58

0.42

0.038

0.0059

0.42

0.07

CY2012 MQO

< 0.09 |xg/m3

< 0.13 |xg/m3

< 0.10 (xg/m3

< 0.08 (xg/m3

< 29 ng/m3

< 0.23 ng/m3

<0.08 ng/m3

ratio of
network
geometric mean
MDL to MQO
(CY2012)

2

0.75

0.40

0.5

0.0081

0.41

0.06

88


-------
4.0 RECOMMENDATIONS

The information in the AQS database required for this report, both analytical results and site
characteristics, was acquired successfully, based on a thorough understanding of the database's
structure. Moreover, based on knowledge of POC assignments in previous years, the POCs for
the primary, duplicate, and collocated samples were assigned with greater facility than
previously. With the added AQS functionality in CY2011 that permitted MDL data to be
reported along with sample data, MDL information for CY2011 and CY2012 was taken solely
from the AQS database. Several sites still had not reported MDL data to AQS, and this report
reflects only those data in the database at the time of data extraction on November 7, 2013.
Requiring the timely reporting of MDL values to AQS would ensure the MDL results are
available so that the data user may better interpret reported results.

As stated in previous QAARs, POCs are present in the AQS database, but the associated sample
type information (e.g., primary, duplicate, or collocated) is not. There is no definitive way to
determine, from AQS alone, the relationship between specific POCs and primary, duplicate, or
collocated samples for a given site. Because POCs are assigned by either the agency monitoring
a particular NATTS site or the laboratory uploading the data to AQS, and are largely non-
standardized across NATTS sites [6, 7, 8, 9, and 10] (refer to Tables 7 and 8), the inclusion of a
field in the AQS database to specify whether a particular POC is primary, duplicate, or
collocated would be a significant benefit to the utility of the AQS data and would streamline the
analyses reported here.

Minimization of field sampler flow bias is directly correlated to improved accuracy in the
measurement of HAP concentrations. As seen in this report, relatively few samplers indicated
flow bias greater than ± 10% from the desired flow. However, flow bias for relatively few
NATTS network air monitors is assessed during any given calendar year. As most sites are
already periodically assessing flow bias, capturing this information in AQS would be beneficial.
Addition of a field in AQS to record the results of these periodic flow audits would provide a
means to minimize bias in reported results.

Lastly, AQS accepts data in a variety of units at the discretion of the agency performing the
upload. This requires careful scrutiny of the UNIT variable so that measurements can be
standardized prior to subsequent data analysis and interpretation. Standardization of the
ambiguous "ppbC" unit is particularly problematic. Implementing a requirement to report results
in mass/volume (e.g., (ig/m3) would improve the consistency of the data and facilitate
interpretation by data end-users.

To summarize, our recommendations are to:

•	Require the reporting of MDLs to AQS;

•	Include fields in AQS to specify the meaning of various POCs, and require the
populatation of these fields;

•	Include fields in AQS to capture the results of ongoing flow audits performed by the
montoring agencies, and require the population of these fields; and

•	Standardize the units of concentration used in AQS, and require that results be
uploaded in these units only.

89


-------
5.0 REFERENCES

1.	U.S. Environmental Protection Agency. (October 17, 2008). AQS Data Coding Manual
(Version 2.33). Available at:

http://www.epa.gov/ttn/airs/airsaqs/manuals/AQS%20Data%20Coding%20Manual.pdf
[last accessed 11/8/2011].

2.	U.S. Environ mental Protection Agency. National Air Toxics Trends Station Work Plan
Template, Revised February 9, 2011.

3.	U.S. Environmental Protection Agency. National Air Toxics Trends Station Work Plan
Template, Revised April 11, 2012.

4.	Eastern Research Group. Technical Assistance Document for the National Ambient Air
Toxics Trends and Assessment Program. Revision 2, April 2009.

5.	U.S. Environmental Protection Agency. (July 2004). Final Draft, July 2004, National
Monitoring Strategy, Air Toxics Component. Available at:
http://www.epa.gov/ttn/amtic/files/ambient/monitorstrat/atstrat804.pdf

6.	U.S. Environmental Protection Agency. (2007). National Air Toxics Trends Stations
Quality Assurance Annual Report - Calendar Year 2006. Prepared by Battelle Memorial
Institute.

7.	U.S. Environmental Protection Agency. (2009). National Air Toxics Trends Stations
Quality Assurance Annual Report - Calendar Year 2007. Prepared by RTI International.

8.	U.S. Environmental Protection Agency. (2010). National Air Toxics Trends Stations
Quality Assurance Annual Report - Calendar Year 2008. Prepared by RTI International.

9.	U.S. Environmental Protection Agency. (2011). National Air Toxics Trends Stations
Quality Assurance Annual Report - Calendar Year 2009. Prepared by RTI International.

10.	U.S. Environmental Protection Agency. (2012). National Air Toxics Trends Stations
Quality Assurance Annual Report - Calendar Year 2010. Prepared by RTI International.

90


-------