&EPA
United States
EnviroimBiU Protection
Agoncy
Policy Assessment for the Review of the Lead
National Ambient Air Quality Standards


External Review Draft

-------
This page is intentionally blank.

-------
                                           EPA-452/P-13-001
                                               January 2013
             Policy Assessment
        for the Review of the Lead
National Ambient Air Quality Standards
          External Review Draft
         U.S. Environmental Protection Agency
       Office of Air Quality Planning and Standards
       Health and Environmental Impacts Division
         Research Triangle Park, North Carolina

-------
                                   DISCLAIMER

       This draft document has been reviewed by the Office of Air Quality Planning and
Standards (OAQPS), U.S. Environmental Protection Agency (EPA), and approved for
publication. This OAQPS Policy Assessment contains preliminary conclusions of the staff of the
OAQPS and does not necessarily reflect the views of the Agency. Mention of trade names or
commercial products is not intended to constitute endorsement or recommendation for use.

-------
                               ACKNOWLEDGMENTS

       This draft Policy Assessment is the product of the Office of Air Quality Planning and
Standards. It has been developed as part of the Environmental Protection Agency's (EPA)
ongoing review of the national ambient air quality standards (NAAQS) for lead (Pb). The Pb
NAAQS review team has been led by Dr. Deirdre Murphy.  Dr. Karen Martin has managed the
project. For the chapter on the health effects evidence and exposure/risk information, the
principal authors include Dr. Deirdre Murphy and Dr. Zach Pekar, and Dr. Murphy served as the
principal author for the chapter on the primary Pb standard. For the chapters on welfare effects
evidence and exposure/risk information and the secondary Pb standard, the principal author is
Ms. Ginger Tennant.  The principal author on the section discussing ambient air monitoring is
Mr. Kevin Cavender. Contributors of emissions information and air quality analyses are Mr.
Marc Houyoux, Mr. Josh Drukenbrod, Dr. Halil Cakir and Mr. Mark Schmidt. Staff from other
EPA offices, including the Office of Research and Development and the Office of General
Counsel, also provided valuable comments and contributions.

-------
This page is intentionally blank

-------
                               TABLE OF CONTENTS


List of Figures	iv
List of Tables	v
EXECUTIVE SUMMARY	ES-1
1  INTRODUCTION	1-1
   1.1  PURPOSE	1-1
   1.2  BACKGROUND	1-3
      1.2.1  Legislative Requirements	1-3
      1.2.2  History of Lead NAAQS Reviews	1-5
      1.2.3  Current Lead NAAQS Review	1-7
   1.3  SCOPE OF CURRENT REVIEW: FATE AND MULTIMEDIA PATHWAYS
        OF AMBIENT AIR LEAD	1-9
      1.3.1  Environmental Distribution and Exposure Pathways	1-10
          1.3.1.1  Human Exposure Pathways	1-11
          1.3.1.2  Ecosystem Exposure Pathways	1-12
      1.3.2  Considerations Related to Historically Emitted Lead	1-13
   1.4  GENERAL ORGANIZATION OF THE DOCUMENT	1-15
   1.5  REFERENCES	1-16

2  AMBIENT AIR LEAD	2-1
   2.1  SOURCES AND EMISSIONS TO AMBIENT AIR	2-1
      2.1.1  Temporal Trends on aNational  Scale	2-2
      2.1.2  Sources and Emissions on National Scale - 2008	2-3
          2.1.2.1  Stationary Sources	2-5
          2.1.2.2  Mobile Sources	2-6
          2.1.2.3  Natural Sources and Long-range Transport	2-7
          2.1.2.4  Previously Deposited or Released Lead	2-7
      2.1.3  Sources and Emissions on Local Scale	2-9
   2.2  AMBIENT AIR QUALITY	2-12
      2.2.1  Air Monitoring	2-12
          2.2.1.1  Lead NAAQS Surveillance Network	2-12
          2.2.1.2  Other Lead Monitoring Networks	2-16
          2.2.1.3  NAAQS Surveillance Monitoring Considerations	2-19
            2.2.1.3.1 Sampling Considerations	2-19
            2.2.1.3.2 Analysis Considerations	2-20
            2.2.1.3.3 Network Design Considerations	2-21
      2.2.2  Ambient Concentrations	2-24
          2.2.2.1  Temporal Trends	2-24
          2.2.2.2  Current Concentrations	2-26
   2.3  AMBIENT AIR LEAD IN OTHER MEDIA	2-32

          January 2013                   i             Draft - Do Not Quote or Cite

-------
      2.3.1  Atmospheric Deposition	2-32
      2.3.2  Terrestrial Media	2-33
         2.3.2.1  Indoor Household Dust	2-34
         2.3.2.2  Outdoor Dust in Areas of Human Activity	2-34
         2.3.2.3  Soil	2-36
         2.3.2.4  Biota	2-38
      2.3.3  Aquatic Media	2-38
         2.3.3.1  Surface Waters	2-38
         2.3.3.2  Sediments	2-39
   2.4  REFERENCES	2-40

3  HEALTH EFFECTS AND EXPOSURE/RISK INFORMATION	3-1
   3.1  INTERNAL DISPOSITION AND BIOMARKERS OF EXPOSURE AND DOSE.. 3-1
   3.2  NATURE OF EFFECTS	3-14
   3.3  PUBLIC HEALTH IMPLICATIONS AND AT-RISK POPULATIONS	3-27
   3.4  EXPOSURE AND RISK	3-36
      3.4.1  Conceptual Model for Air-Related Lead Exposure and Risk	3-37
      3.4.2  Case Studies	3-39
      3.4.3  Analysis Approach	3-41
         3.4.3.1  Estimating Exposure	3-43
         3.4.3.2  Air Quality Scenarios Included in 2007 Assessment	3-46
         3.4.3.3  Methods for Deriving Risk Estimates	3-48
            3.4.3.3.1 Full Risk Model in 2007 REA	3-49
            3.4.3.3.2 Air Quality Scenarios Reflecting the Current Standard	3-51
      3.4.4  Challenges in Characterizing Air-related Exposure and Risk	3-53
      3.4.5  Risk Estimates	3-55
      3.4.6  Treatment of Variability	3-60
      3.4.7  Characterizing Uncertainty	3-61
      3.4.8  Updated Interpretation of Risk Estimates	3-64
   3.5  REFERENCES	3-66

4  REVIEW OF THE PRIMARY STANDARD FOR LEAD	4-1
   4.1  APPROACH	4-1
      4.1.1  Approach Used in the Last Review	4-2
         4.1.1.1  Conclusion Regarding the Need for Revision	4-3
         4.1.1.2  Conclusions on Elements of Revised Standard	4-5
      4.1.2  Approach for the Current Review	4-10
   4.2  ADEQUACY OF THE CURRENT STANDARD	4-13
      4.2.1  Evidence-based Considerations	4-13
      4.2.2  Exposure/Risk-based Considerations	4-21
   4.3  PRELIMINARY STAFF CONCLUSIONS ON THE PRIMARY STANDARD	4-25
   4.4  KEY UNCERTAINTIES AND AREAS FOR FUTURE RESEARCH AND
        DATA COLLECTION	4-33
   4.5  REFERENCES	4-36

         January 2013                   ii            Draft - Do Not Quote or Cite

-------
5  WELFARE EFFECTS AND RISK/RISK INFORMATION	5-1
   5.1   WELFARE EFFECTS INFORMATION	5-1
   5.2   EXPOSURE AND RISK INFORMATION	5-13
      5.2.1  Screening Assessment from Last Review	5-14
      5.2.2  Screening Assessment Results and Interpretation	5-17
   5.3   REFERENCES	5-22
6  REVIEW OF THE SECONDARY STANDARD FOR LEAD	6-1
   6.1   APPROACH	6-1
      6.1.1  Approach Used in the Last Review	6-2
      6.1.2  Approach for the Current Review	6-3
   6.2   ADEQUACY OF THE CURRENT STANDARD	6-5
      6.2.1  Evidence-based Considerations	6-5
      6.2.2  Exposure/Risk-based Considerations	6-9
   6.3   PRELIMINARY STAFF CONCLUSIONS ON THE SECONDARY
        STANDARD	6-10
   6.4   KEY UNCERTAINTIES AND AREAS FOR FUTURE RESEARCH AND
        DATA COLLECTION	6-12
   6.5   REFERENCES	6-14
CHAPTER APPENDICES
Appendix 2A.  The 2008 NEI: Data Sources, Limitations and Confidence	2A-1
Appendix 2B. Recent Regulatory Actions on Stationary Sources of Lead	2B-1
Appendix 2C. Criteria for Air Quality Data Analysis	2C-1
Appendix 2D. Air Quality Data Analysis Summary	2D-1
Appendix 3 A. Interpolated Risk Estimates for the Generalized (Local) Urban Case Study.. 3A-1
Appendix 5 A. Additional Detail on 2006 Ecological Screening Assessment	5 A-1
         January 2013                  iii           Draft - Do Not Quote or Cite

-------
                                     LIST OF FIGURES

Figure 1 -1. Pathways of human and ecosystem exposure to lead from ambient air	1-11
Figure 2-1. Temporal trend in air emissions: 1970-2010	2-3
Figure 2-2. Geographic distribution of facilities estimated to emit at least 0.50 tpy in 2008.... 2-11
Figure 2-3. Map of Pb-TSP Monitoring Sites in Current Pb NAAQS Monitoring Network. ...2-15
Figure 2-4. Map of sites near airports for which one year of monitoring is required	2-16
Figure 2-5.Pb-PMio monitoring sites	2-17
Figure 2-6. Pb-PM2.5 monitoring sites in CSN and IMPROVE networks (2012)	2-18
Figure 2-7. Temporal trend in Pb-TSP concentrations: 1980-2010(31 sites)	2-24
Figure 2-8. Temporal trend in Pb-TSP concentrations: 2000-2010 (92 sites)	2-24
Figure 2-9. Airborne Pb -TSP concentrations (3-month average) at five sites near
          roadways: 1979-2010	2-25
Figure 2-10. Pb-TSP maximum 3-month means (215 sites), 2009-2011	2-27
Figure 2-11. Distribution of maximum 3-month mean concentrations of Pb-TSP, Pb-PMio
           and Pb-PM2.5 at different site types, 2009-2011	2-29
Figure 2-12. Distribution of annual mean concentrations of Pb-TSP, Pb-PMi0 and Pb-PM25
           at different site types, 2009-2011	2-30
Figure 2-13. Distribution of maximum monthly mean concentrations of Pb-TSP, Pb-PMi0
           and Pb-PM2.5 at different site types, 2009-2011	2-31
Figure 2-14. Temporal trend in sediment concentration from core samples in 12 lakes at
           eight National Parks or Preserves	2-41
Figure 3-1. Temporal trend in mean blood Pb levels for NHANES cohorts	3-4
Figure 3-2. Human exposure pathways for air-related lead	3-38
Figure 3-3. Overview of analysis approach	3-42
Figure 3-4. Comparison of four concentration-response functions used in risk assessment	3-50
Figure 3-5. Parsing of air-related risk estimates	3-53
Figure 4-1. Overview of approach for review of current primary standard	4-12
Figure 5-1. Analytical approach for screening-level assessment in the last review
           (2006 REA, Exhibit 2-6)	5-16
Figure 6-1. Overview of approach for review of current secondary standard	6-4
          January 2013                    iv            Draft - Do Not Quote or Cite

-------
                                      LIST OF TABLES

Table 2-1. U.S. lead emissions by source categories estimated to emit at least 4 tpy	2-4
Table 2-2. Facilities estimated to emit at least 0.50 tpy in 2008	2-10
Table 2-3. Dry deposition of lead in large metropolitan areas and in a smelter town	2-36

Table 3-1. Empirically derived air-to-blood ratios for populations inclusive of children	3-12
Table 3-2. Associations with neurocognitive function measures in analyses with child study
          group blood lead levels <5  |ig/dL	3-21
Table 3-3. Summary of quantitative relationships of IQ and blood Pb for analyses with
          blood Pb levels closest to those of young children in the U.S. today	3-26
Table 3-4. Number of children aged 5  and under in areas of elevated ambient Pb
          concentrations relative to theNAAQS	3-34
Table 3-5. Population size near larger  sources of lead emissions	3-35
Table 3-6. Types of population exposures assessed	3-40
Table 3-7. Summary of approaches used to estimate case study  media concentrations	3-44
Table 3-8. Air quality scenarios assessed	3-48
Table 3-9. Comparison of total and incremental IQ loss estimates for blood Pb below
          10 |ig/dL based on the four concentration-response functions	3-50
Table 3-10.Estimates of air-related risk from 2007 risk assessment	3-57
Table 3-11.Estimates of air-related risk for the generalized (local) urban case study,
          including interpolated estimates for current standard	3-58
          January 2013                     v             Draft - Do Not Quote or Cite

-------
This page is intentionally blank

-------
 1                                   EXECUTIVE SUMMARY
 2
 3          This draft Policy Assessment (PA) has been prepared by staff in the Environmental
 4    Protection Agency's (EPA) Office of Air Quality Planning and Standards (OAQPS) as part of
 5    the Agency's ongoing review of the primary (health-based) and secondary (welfare-based)
 6    national ambient air quality standards (NAAQS) for lead (Pb).  It presents analyses and
 7    preliminary staff conclusions regarding the policy implications of the key scientific and technical
 8    information that informs this review.  The final PA is intended to "bridge the gap" between the
 9    relevant scientific evidence and technical information and the judgments required of the EPA
10    Administrator in determining whether to retain or revise the current standards. Development of
11    the PA is  also intended to facilitate advice and recommendations on the standards to the
12    Administrator from an independent scientific review committee, the Clean Air Scientific
13    Advisory  Committee (CASAC), as provided for in the Clean Air Act (CAA).
14          Staff analyses in this draft PA are based on the scientific assessment presented in the
15    third draft Integrated Science Assessment for Pb (ISA) prepared for this review by the EPA's
16    Office of  Research and Development (ORD) as well as scientific and technical assessments from
17    prior Pb NAAQS reviews.  Such assessments include quantitative human health and ecological
18    risk and exposure assessments (REAs) developed in the last review as new health and ecological
19    REAs were not warranted based on staffs and CASAC's consideration of the evidence newly
20    available  in this review with regard to risk and exposure assessment. In considering the
21    scientific  evidence and other technical information available in this review, emphasis is given to
22    consideration of the extent to which the evidence newly available since the last review alters
23    conclusions drawn in the last review with regard to health and welfare effects of Pb, the exposure
24    levels at which they occur and the associated at-risk populations and ecological receptors or
25    ecosystems.
26          The overarching questions in this review, as in all NAAQS reviews, regard the support
27    provided by the currently available scientific evidence and exposure/risk-based information for
28    the adequacy of the current standards and the extent to which the scientific evidence and
29    technical  information provides support for concluding that consideration of alternative standards
30    may be appropriate.  The analyses presented in this draft PA to address such questions lead to
31    preliminary staff conclusions that it is appropriate to consider retaining the current primary and
32    secondary standards without revision; accordingly, no potential alternative standards have been
33    identified by staff for consideration in this review.  Comments and recommendations from
34    CASAC,  and public comments, based on review of this draft PA, will inform final staff
35    conclusions and the presentation of information in the final PA.

            January 2013                   ES-1             Draft - Do Not Quote or Cite

-------
 1    Current Lead NAAQS and Scope of Review
 2          The NAAQS for Pb was initially set in 1978. Review of the 1978 NAAQS for Pb,
 3    completed in October 2008, resulted in substantial revision based on the large body of evidence
 4    accumulated over the intervening three decades. In terms of the basic elements of the NAAQS,
 5    the level of the primary standard was lowered by an order of magnitude from 1.5 |ig/m3 to 0.15
 6    Hg/m3 and the averaging time was revised to a rolling three-month period (from a period based
 7    on calendar quarters) with a maximum (not-to-be exceeded) form, evaluated over a 3-year
 8    period. The indicator of Pb in total suspended particles (Pb-TSP) was retained, reflecting the
 9    evidence that Pb particles of all  sizes pose health risks.  The secondary standard was revised to
10    be identical in all respects to the revised primary standard.
11          The multimedia and persistent nature of Pb contributes complexities to the review of the
12    Pb NAAQS unlike issues addressed in other NAAQS reviews.  Air-related Pb distributes from
13    air to other media, including indoor and outdoor dusts, soil, food, drinking water, as well as
14    surface water and sediments.  As a result, review of the Pb NAAQS considers the protection
15    provided against the health and environmental effects of air-related Pb associated both with
16    exposures to Pb in ambient air and with exposures to Pb that makes its way from ambient air into
17    other media. Additional complexity derives from the recognition that exposure to Pb also results
18    from nonair sources, including Pb in paint, tap water affected by plumbing containing Pb, lead-
19    tainted products, as  well as surface water discharges and runoff from industrial sites. Such
20    nonair sources contribute to the  total burden of Pb in the human body and in the environment,
21    making it much more difficult to assess independently the health and welfare effects attributable
22    to air-related Pb that are the focus of the NAAQS. Further, the persistence of Pb in the human
23    body and the environment is another important consideration in assessing the adequacy  of the
24    current Pb standards. In so doing, staff is mindful of the history of the greater and more
25    widespread atmospheric emissions that occurred in previous years (e.g., under the previous Pb
26    standard, and prior to establishment of any Pb NAAQS) and that contributed to the Pb that  exists
27    in human populations and ecosystems today. Likewise, staff also recognizes the role of nonair
28    sources of Pb, now  and in the past, that also contribute to the Pb that exists in human populations
29    and ecosystems today. As in the last Pb NAAQS review, this backdrop of environmental Pb
30    exposure, and its impact on the populations and ecosystems which may be the subjects of the
31    currently available scientific evidence, complicates our consideration of the health and welfare
32    protection  afforded  by the current NAAQS.
            January 2013                   ES-2            Draft - Do Not Quote or Cite

-------
 1    Characterization of Ambient Air Lead
 2          Emissions to ambient air and associated air Pb concentrations have declined substantially
 3    over the past several decades. The most dramatic reductions in Pb emissions occurred prior to
 4    1990 in the transportation sector due to the removal of Pb from gasoline. Lead emissions were
 5    further reduced substantially between 1990 and 2008, with significant reductions occurring in the
 6    metals industries at least in part as a result of national emissions standards for hazardous air
 7    pollutants. Additional reductions in stationary source emissions are also anticipated from
 8    regulations which have been promulgated since 2008 under Section 112 of the CAA.
 9          As at the time of the last review, the majority of Pb emissions nationally is associated
10    with combustion of leaded aviation gasoline by piston-driven aircraft. The largest sources on a
11    local scale are generally associated with metals industries. As a result of revisions to monitoring
12    regulations stemming from the last Pb NAAQS review, Pb NAAQS monitors are required near
13    the largest Pb emissions sources, as well as in sites distant from such sources in large population
14    areas. Ambient air Pb monitoring data available thus far from this expanded network continue to
15    illustrate the source-related aspect of airborne Pb, with highest concentrations near large sources
16    and lowest in areas removed from sources. In addition, Pb monitoring data are also being
17    collected over at least a one-year period near a set of airports identified as most likely to have
18    elevated Pb concentrations due to leaded  aviation gasoline usage. These data will inform future
19    airport monitoring activities, as well as an ongoing investigation into Pb emissions  from piston-
20    engine aircraft  under Section 231 of the Clean Air Act, separate from this Pb NAAQS review.
21          Lead occurs in ambient air in particulate form and, with characteristics and  spatial
22    patterns influenced by a number of factors, deposits from air to surfaces in natural and human-
23    made environments. By this deposition process and subsequent transfer processes, ambient air
24    Pb is distributed into multiple human exposure pathways and environmental media in aquatic
25    and terrestrial ecosystems. In areas removed from large air emissions sources, currently
26    available information on Pb concentrations in nonair media includes numerous examples of
27    declines in surface concentrations reflective  of the reductions in deposition over the past several
28    decades. In areas near large air sources where emissions reductions have occurred, only very
29    limited information is available, such as for reductions in air and surface dust concentrations,
30    with even less information available on trends for other media such as surface soils.

31    Health Effects and  Review of the Primary Standard
32          Lead has long been recognized to exert a broad array of deleterious effects on multiple
33    organ systems as described in the ISA for this review and consistent with conclusions of prior
34    scientific assessments. Over the three decades from the time the standard was initially set in
35    1978 through its revision with the NAAQS review completed in 2008, the evidence base
            January 2013                    ES-3            Draft - Do Not Quote or Cite

-------
 1    expanded considerably in a number of areas, including with regard to effects on neurocognitive
 2    function in young children at increasingly lower blood Pb levels.  These effects formed the
 3    primary basis for the 2008 revisions to the primary standard.  The current standard was set most
 4    specifically to provide appropriate public health protection from the effects of air-related Pb on
 5    cognitive function (e.g., IQ loss) in young children. In so doing, the  standard was judged to
 6    provide the requisite public health protection from the full array of health effects of Pb,
 7    consistent with the CAA requirement that the primary standard, in the judgment of the
 8    Administrator, based on the latest scientific knowledge, is requisite to protect public health with
 9    an adequate margin of safety.
10           The health effects evidence newly available in this review, as critically assessed in the
11    ISA in conjunction with the full body of evidence, reaffirms conclusions on the broad array of
12    effects recognized for Pb in the last review. Further, staff observes the general consistency of the
13    current evidence with the evidence available in the last review, particularly with regard to key
14    aspects of the evidence on which the current standard is based.  These key aspects include those
15    regarding the relationships between air Pb concentrations and the associated Pb in the blood of
16    young children (i.e., air-to-blood ratios) as well as between total blood Pb levels and effects on
17    neurocognitive function (i.e., concentration-response (C-R) functions for IQ loss). Factors
18    characterizing these two relationships are the key inputs to the framework developed in the last
19    review to translate the available evidence into  a basis  for considering a primary Pb standard that
20    would be requisite to protect against this and other Pb-related health endpoints. This framework
21    is again considered in light of the current available evidence.  This Pb NAAQS review, like any
22    NAAQS review, requires public health policy judgments. The public health policy judgments
23    for this review include the public health significance of a given magnitude of IQ loss in a small
24    subset of highly exposed children (i.e., those likely to experience air-related Pb exposures at the
25    level of the standard), as well as how to consider the nature and magnitude  of the array of
26    uncertainties that are inherent in the evidence and in the application of this  specific framework.
27           In also considering the quantitative risk estimates associated with the current standard,
28    based on the risk assessment conducted in the last review, staff observes that these estimates
29    indicate a level of risk that is roughly consistent with and generally supportive of conclusions
30    drawn from the evidence using the evidence-based air-related IQ loss framework. Staff
31    additionally recognizes the complexity of the modeling done as part of that assessment and the
32    substantial limitations and uncertainties in the  resulting risk estimates.
33           Based on the above considerations, staff preliminarily concludes that the currently
34    available information supports  a primary standard as protective as the current standard and that it
35    is appropriate to consider retaining the current standard without revision. In so doing, staff
36    additionally notes that the final decision on the adequacy of the current standard is largely a
             January 2013                    ES-4             Draft - Do Not Quote or Cite

-------
 1    public health policy judgment to be made by the Administrator, drawing upon the scientific
 2    information as well as judgments about how to consider the range and magnitude of uncertainties
 3    that are inherent in the scientific evidence and technical analyses. In this context, staff
 4    recognizes that the uncertainties and limitations associated with the many aspects of the
 5    relationship between air Pb concentrations and blood Pb levels and associated health effects are
 6    amplified with consideration of increasingly lower air concentrations. In staffs view, based on
 7    the current evidence there is appreciable uncertainty associated with drawing conclusions
 8    regarding whether there would be reductions in risk to public health from alternative lower levels
 9    as compared to the level of the current standard. Thus, staff concludes that the basis for any
10    consideration of alternative lower standard levels would reflect different public health policy
11    judgments as to the appropriate approach for weighing uncertainties in the evidence and for
12    providing requisite protection of public health with an adequate margin of safety. Accordingly,
13    and in light of the preliminary staff conclusion that it is appropriate to consider the current
14    standard to be adequate, this document does not identify potential alternative standards for
15    consideration in this review.

16    Welfare Effects and Review of the Secondary Standard
17          Consideration of the welfare effects evidence and screening-level risk information in the
18    last review (completed in 2008) led to the conclusion that there was a potential for adverse
19    welfare effects occurring  under the then-current Pb standard (set in 1978), although there were
20    insufficient data to provide a quantitative basis for setting a secondary standard different from
21    the primary standard. Accordingly, the secondary standard was substantially revised to be
22    identical in all respects to the newly revised primary standard.
23          In assessing the currently available scientific evidence and the exposure/risk information
24    with regard to support for the adequacy of the protection afforded by the  current  standard, staff
25    observes the general consistency of the current evidence with that available in the last review,
26    including the substantial limitations in the current evidence that complicate conclusions
27    regarding the potential for Pb emissions under the current, much lower standard to contribute to
28    welfare effects. Thus, based on the staff analysis, framed by key policy-relevant questions for
29    the review, staff preliminarily concludes that consideration should be given to retaining the
30    current standard, without revision, and this document does  not identify potential  alternative
31    standards for consideration in this review.
32
             January 2013                    ES-5             Draft - Do Not Quote or Cite

-------
 1                                       1   INTRODUCTION

 2    1.1   PURPOSE
 3           The U.S. Environmental Protection Agency (EPA) is presently conducting a review of
 4    the primary (health-based) and secondary (welfare-based) national ambient air quality standards
 5    (NAAQS) for lead (Pb). The overall plan and schedule for this review were presented in the
 6    Integrated Review Plan for the National Ambient Air Quality Standards for Lead (IRP; USEP A,
 7    201 la). The IRP also identified key policy-relevant issues to be addressed in this review and
 8    discussed the key documents that generally inform NAAQS reviews, including an Integrated
 9    Science Assessment (ISA), Risk and Exposure Assessments (REAs), and a Policy Assessment
10    (PA). The PA presents a staff evaluation of the policy implications of the key scientific and
11    technical information in the ISA and REAs for EPA's consideration.1 The PA generally
12    provides a transparent evaluation and staff conclusions regarding policy considerations related to
13    reaching judgments about the adequacy of the current standards, and, if revision is considered,
14    what revisions may be appropriate to consider.
15           When final, the PA is intended to help "bridge the gap" between the Agency's scientific
16    assessments presented in the ISA and REAs, and the judgments required of the EPA
17    Administrator in determining whether it is appropriate to retain or revise the NAAQS. In
18    evaluating the adequacy of the  current standard and whether it is appropriate to consider
19    alternative standards, the PA focuses on information that is most pertinent to evaluating the basic
20    elements of the NAAQS: indicator,2 averaging time, form,3 and level.  These  elements, which
21    together serve to define each standard, must be considered collectively in evaluating the health
22    and welfare protection afforded by the Pb standards. The PA integrates and interprets the
23    information from the ISA and REAs to frame policy options for consideration by the
24    Administrator. In so doing, the PA recognizes that the selection of a specific approach to
25    reaching final decisions  on primary and secondary NAAQS will reflect the judgments of the
26    Administrator.
             1 The terms "staff and "we" throughout this document refer to staff in the EPA's Office of Air Quality
      Planning and Standards (OAQPS). In past NAAQS reviews, this document was referred to as the OAQPS Staff
      Paper.
             2 The "indicator" of a standard defines the chemical species or mixture that is to be measured in
      determining whether an area attains the standard. The indicator for the Pb NAAQS is lead in total suspended
      particles.
             3 The "form" of a standard defines the air quality statistic that is to be compared to the level of the standard
      in determining whether an area attains the standard. For example, the form of the annual PM25 NAAQS is the 3-
      year average of the weighted annual mean PM2 5 concentrations, while the form of the 8-hour CO NAAQS is the
      second-highest 8-hour average in a year.

                January 2013                    1-1            Draft - Do Not Quote  or Cite

-------
 1          The development of the PA is also intended to facilitate advice to the Agency and
 2    recommendations to the Administrator from an independent scientific review committee, the
 3    Clean Air Scientific Advisory Committee (CASAC), as provided for in the Clean Air Act. As
 4    discussed below in section 1.2.1, the CASAC is to advise not only on the Agency's assessment
 5    of the relevant scientific information, but also on the adequacy of the existing standards, and to
 6    make recommendations as to any revisions of the standards that may be appropriate. The EPA
 7    facilitates CASAC advice and recommendations, as well as public input and comment, by
 8    requesting CASAC review and public comment on one or more drafts of the PA.
 9          The decision whether to prepare one or more drafts of the PA is influenced by
10    preliminary staff conclusions and associated CASAC advice and public comment, among other
11    factors. Typically, a second draft PA has been prepared in cases where the available information
12    calls into question the adequacy of the current standard and analyses of potential alternative
13    standards are developed taking into consideration CASAC advice and public comment. In such
14    cases, a second draft PA includes preliminary staff conclusions regarding potential alternative
15    standards and undergoes CASAC review and public comment prior to preparation of the final
16    PA. When such analyses are not undertaken, a second draft PA may not be warranted.
17          In this draft PA for this review of the Pb NAAQS, we consider the scientific and
18    technical information available in this review as assessed in the third draft Integrated Science
19    Assessment for Lead (henceforth referred to as the ISA [USEPA,  2012a]),  prepared by EPA's
20    National Center for Environmental Assessment (NCEA),  and the quantitative human exposure
21    and health risk and screening-level ecological risk assessments performed  in the last review. As
22    discussed below in section 1.2.3, upon consideration of the evidence newly available in this
23    review with regard to risk and exposure assessment, staff concluded that new health and welfare
24    REAs were not warranted. Accordingly, the quantitative risk information  considered in this PA
25    is drawn from the quantitative human exposure and health risk and screening-level ecological
26    risk assessments performed in the last review (the 2007 Health Risk Assessment Report or 2007
27    REA [USEPA, 2007a] and the 2006 screening-level Ecological Risk Assessment or 2006 REA
28    [ICF, 2006]), and is interpreted in the context of newly available evidence in this review.
29          The evaluation and preliminary staff conclusions presented in this draft PA for the Pb
30    NAAQS have been informed by comments and advice received from CASAC in their reviews of
31    the other draft Agency documents  prepared thus far in this NAAQS review.  Review and
32    comments from CASAC, and public comment,  on this draft PA will inform the final evaluation
33    and staff conclusions in the final PA.
34          Beyond informing the EPA Administrator and facilitating the advice and
35    recommendations of CASAC and the public, the PA is also intended to be a useful reference to
36    all parties interested in the Pb NAAQS review.  In these roles, it is intended to serve as a single
               January 2013                    1-2            Draft - Do Not Quote or Cite

-------
 1    source of the most policy-relevant information that informs the Agency's review of the lead
 2    NAAQS, and it is written to be understandable to a broad audience.

 3    1.2   BACKGROUND
 4          1.2.1   Legislative Requirements
 5           Two sections of the Clean Air Act (CAA or the Act) govern the establishment and
 6    revision of the NAAQS. Section 108 (42 U.S.C. section 7408) directs the Administrator to
 7    identify and list certain air pollutants and then to issue air quality criteria for those pollutants.
 8    The Administrator is to list those air pollutants that in her "judgment, cause or contribute to air
 9    pollution which may reasonably be anticipated to endanger public health or welfare;" "the
10    presence of which in the ambient air results from numerous or diverse mobile or stationary
11    sources;" and "for which . . .  [the Administrator] plans to issue air quality criteria..."  Air quality
12    criteria are intended to "accurately reflect the latest scientific knowledge useful in indicating the
13    kind and extent of all identifiable effects on public health or welfare which may be expected
14    from the presence of [a]  pollutant in the ambient air . . ." 42 U.S.C. § 7408(b).  Section 109 (42
15    U.S.C. 7409) directs the Administrator to propose and promulgate "primary" and "secondary"
16    NAAQS for pollutants for which air quality criteria are issued. Section  109(b)(l) defines a
17    primary standard as one  "the attainment and maintenance of which in the judgment of the
18    Administrator,  based on such criteria and allowing an adequate margin of safety, are requisite to
19    protect the public health."4 A secondary standard, as defined in section  109(b)(2), must "specify
20    a level of air quality the  attainment and maintenance of which, in the judgment of the
21    Administrator,  based on such criteria, is requisite to protect the public welfare from any known
22    or anticipated adverse effects associated with the presence of [the] pollutant in the ambient air."5
23           The requirement that  primary standards provide an adequate margin of safety was
24    intended to address uncertainties associated with inconclusive scientific  and technical
25    information available at the time of standard setting.  It was also intended to provide a reasonable
26    degree of protection against hazards that research has not yet identified.  See Lead Industries
27    Association v. EPA, 647 F.2d 1130, 1154 (D.C. Cir 1980), cert, denied, 449 U.S. 1042 (1980);
28    American Petroleum Institute v. Costle, 665 F.2d 1176, 1186 (D.C. Cir.  1981), cert, denied, 455
             4 The legislative history of section 109 indicates that a primary standard is to be set at "the maximum
      permissible ambient air level. . . which will protect the health of any [sensitive] group of the population," and that
      for this purpose "reference should be made to a representative sample of persons comprising the sensitive group
      rather than to a single person in such a group" S. Rep. No. 91-1196, 91st Cong., 2d Sess. 10 (1970).
             5 Welfare effects as defined in section 302(h) (42 U.S.C. § 7602(h)) include, but are not limited to, "effects
      on soils, water, crops, vegetation, man-made materials, animals, wildlife, weather, visibility and climate, damage to
      and deterioration of property, and hazards to transportation, as well as effects on economic values and on personal
      comfort and well-being."

                January 2013                      1-3             Draft - Do Not Quote or Cite

-------
 1    U.S. 1034 (1982); American Farm Bureau Federation v. EPA, 559 F. 3d 512, 533 (D.C. Cir.
 2    2009); Association of Battery Recyclers v. EPA, 604 F. 3d 613, 617-18 (D.C. Cir. 2010). Both
 3    kinds of uncertainties are components of the risk associated with pollution at levels below those
 4    at which human health effects can be said to occur with reasonable scientific certainty.  Thus, in
 5    selecting primary standards that provide an adequate margin of safety, the Administrator is
 6    seeking not only to prevent pollution levels that have been demonstrated to be harmful but also
 7    to prevent lower pollutant levels that may pose an unacceptable risk of harm, even if the risk is
 8    not precisely identified as to nature or degree. The CAA does not require the Administrator to
 9    establish a primary NAAQS at a zero-risk level  or at background concentration levels, see Lead
10    Industries v. EPA, 647 F.2d at 1156 n.51, but rather at a level that reduces risk sufficiently so as
11    to protect public health with an adequate margin of safety.
12           In addressing the requirement for an adequate margin of safety, the EPA considers such
13    factors as the nature and severity of the health effects involved, the size of sensitive population(s)
14    at risk, and the kind  and degree of the uncertainties that must be addressed. The  selection of any
15    particular approach to providing an adequate margin of safety is a policy choice left specifically
16    to the Administrator's judgment.  See Lead Industries Association v. EPA, 647 F.2d at 1161-62;
17    Whitman v. American Trucking Associations, 531 U.S. 457, 495 (2001).
18           In setting primary and secondary standards that are "requisite" to protect public health
19    and welfare, respectively, as provided in section 109(b),  EPA's task is to establish standards that
20    are neither more nor less stringent than necessary for these purposes.  In so doing, the EPA may
21    not consider the costs of implementing the standards.  See generally, Whitman v.  American
22    Trucking Associations, 531 U.S. 457, 465-472, 475-76 (2001).  Likewise, "[a]ttainability and
23    technological feasibility are not relevant considerations in the promulgation of national ambient
24    air quality standards." American Petroleum Institute v. Costle, 665 F. 2d at 1185.
25           Section 109(d)(l) requires that "not later than December 31, 1980, and at 5-year
26    intervals thereafter, the Administrator shall complete a thorough review of the criteria
27    published under section 108 and the national ambient air quality standards . . . and shall make
28    such revisions in such criteria and standards and promulgate such new standards  as may be
29    appropriate . . . ."  Section  109(d)(2) requires that an independent scientific review committee
30    "shall  complete a review of the criteria . . . and the national primary and secondary ambient air
31    quality standards . .  . and shall recommend to the Administrator any new . . . standards and
32    revisions of existing criteria and standards as may be appropriate  . . . ."  Since the early 1980's,
                January 2013                    1-4             Draft - Do Not Quote or Cite

-------
 1   this independent review function has been performed by the Clean Air Scientific Advisory
 2   Committee (CASAC).6

 3         1.2.2  History of Lead NAAQS Reviews
 4          Unlike pollutants such as particulate matter and carbon monoxide, air quality criteria had
 5   not been issued for Pb as of the enactment of the Clean Air Act of 1970, which first set forth the
 6   requirement to set national ambient air quality standards based on air quality criteria.  In the
 7   years just after enactment of the CAA, the EPA did not intend to issue air quality criteria for Pb,
 8   and accordingly had not listed Pb under Section  108 of the Act.  The EPA had determined to
 9   control Pb air pollution through regulations to phase-out use of Pb additives in gasoline and the
10   EPA viewed those controls, and possibly additional federal controls, as the best approach to
11   controlling Pb emissions  (See 41 FR 14921 (April 8, 1976).  However, the decision not to list Pb
12   under Section 108 was challenged by environmental and public health groups and the U.S.
13   District Court for the Southern District of New York concluded that the EPA was required to list
14   Pb under Section 108. (Natural Resources Defense Council v. EPA, 411 F. Supp. 864 21  [S.D.
15   N.Y. 1976], aff d, 545 F.2d 320 [2d Cir. 1978]).
16          Accordingly, on April 8, 1976, the EPA published a notice in the Federal Register that
17   Pb had been listed under Section 108 as a criteria pollutant (41 FR 14921) and on October 5,
18   1978, the EPA promulgated primary  and secondary NAAQS for Pb under Section 109 of the Act
19   (43 FR 46246). Both primary and secondary standards were set at a level of 1.5 micrograms per
20   cubic meter (ug/m3), measured as Pb in total suspended particles (Pb-TSP),  not to be exceeded
21   by the maximum arithmetic mean concentration averaged over a calendar quarter. These
22   standards were based on the 1977 Air Quality Criteria for Lead (USEPA, 1977).
23          The first review of the Pb  standards was initiated in the mid-1980s. The  scientific
24   assessment for that review is  described in the 1986 Air Quality Criteria for Lead (USEPA,
25   1986a), the associated Addendum (USEPA, 1986b) and the 1990 Supplement (USEPA, 1990a).
26   As part of the review, the Agency designed and performed human exposure and health risk
27   analyses (USEPA, 1989), the results  of which were presented in a 1990 Staff Paper (USEPA,
28   1990b).  Based on the scientific assessment and the human exposure and  health  risk analyses, the
29   1990 Staff Paper presented recommendations for consideration by the Administrator (USEPA,
30   1990b).  After consideration of the documents developed during the review and the significantly
31   changed circumstances since Pb was listed in 1976, the Agency did not propose any revisions to
32   the 1978 Pb NAAQS.  In  a parallel effort, the Agency developed the broad, multi-program,
33   multimedia, integrated U.S. Strategy for Reducing Lead Exposure (USEPA,  1991).  As part of
            6 Lists of CASAC members and of members of the CASAC Pb Review Panel are available at:
      http://vosemite.epa.gov/sab/sabproduct.nsfAVebCASAC/CommitteesandMembership7OpenDocument.
                January 2013                   1-5            Draft - Do Not Quote or Cite

-------
 1    implementing this strategy, the Agency focused efforts primarily on regulatory and remedial
 2    clean-up actions aimed at reducing Pb exposures from a variety of nonair sources judged to pose
 3    more extensive public health risks to U.S. populations, as well as on actions to reduce Pb
 4    emissions to air, such as bringing more areas into compliance with the existing Pb NAAQS
 5    (USEPA, 1991).
 6          The most recent review of the Pb air quality criteria and standards was initiated in
 7    November 2004 (69 FR 64926) and the Agency's plans for preparation of the Air Quality
 8    Criteria Document and conduct of the NAAQS review were presented in documents completed
 9    in 2005 and early 2006 (USEPA, 2005; USEPA 2006a).7 The schedule for completion of the
10    review was governed by a judicial order in Missouri Coalition for the Environment v. EPA (No.
11    4:04CV00660 ERW, Sept. 14, 2005; and amended on April 29, 2008 and July 1, 2008), which
12    specified a schedule for the review of duration substantially shorter than five years.
13          The scientific assessment for the review is described in the 2006 Air Quality Criteria for
14    Lead (USEPA,  2006b; henceforth referred to as the 2006 CD), multiple drafts of which received
15    review by CASAC and the public. The EPA also conducted human exposure and health risk
16    assessments and a pilot ecological risk assessment for the review, after consultation with
17    CASAC and receiving public comment on a draft analysis plan (USEPA, 2006c).  Drafts of these
18    quantitative assessments were reviewed by CASAC and the public. The pilot ecological risk
19    assessment was released in December 2006 (ICF, 2006) and the final health risk assessment
20    report was released in November 2007 (USEPA, 2007a). The policy assessment based on both
21    of these assessments, air quality  analyses and key evidence from the AQCD was presented in the
22    Staff Paper (USEPA, 2007b), a draft of which also received CASAC and public review. The
23    final  Staff Paper presented OAQPS staffs evaluation of the public health and welfare policy
24    implications of the key studies and scientific information contained in the Criteria Document  and
25    presented and interpreted  results from the quantitative risk/exposure analyses conducted for this
26    review. Based  on this evaluation, the Staff Paper presented OAQPS staff recommendations that
27    the Administrator give consideration to substantially revising the primary and secondary
28    standards to a range of levels at or below 0.2 |ig/m3.
29          Immediately subsequent to completion of the Staff Paper, the EPA issued an advance
30    notice of proposed rulemaking (ANPR) that was signed by the Administrator on December 5,
31    2007 (72 FR 71488).8 CASAC provided advice and recommendations to the Administrator with
32    regard to the Pb NAAQS  based on its review of the ANPR and the previously released final Staff
            7 In the current review, these two documents have been combined in the IRP.
            8 The ANPR was one of the features of the revised NAAQS review process that EPA instituted in 2006.  In
      2009 (Jackson, 2009), this component of the process was replaced by reinstatement of policy assessment prepared
      by OAQPS staff (previously termed the OAQPS Staff Paper).
               January 2013                    1-6            Draft - Do Not Quote or Cite

-------
 1    Paper and risk assessment reports.  The proposed decision on revisions to the Pb NAAQS was
 2    signed on May 1, 2008 and published in the Federal Register on May 20, 2008 (73 FR 29184).
 3    Members of the public provided both written and, at two public hearings,  oral comments and the
 4    CASAC Pb Panel also provided advice and recommendations to the Administrator based on its
 5    review of the proposal notice.  The final decision on revisions to the Pb NAAQS was signed on
 6    October 15, 2008 and published in the Federal Register on November 12,  2008 (73 FR 66964).
 7          The November 2008 notice described EPA's decision to revise the primary and
 8    secondary NAAQS for Pb, as discussed more fully in section 4.1.1 below. In consideration of
 9    the much-expanded health effects evidence on neurocognitive effects of Pb in children, the EPA
10    substantially revised the primary standard from a level of 1.5 |ig/m3 to a level of 0.15 |ig/m3.
11    The averaging time was revised to a rolling three-month period with a maximum (not-to-be-
12    exceeded) form, evaluated over a three-year period. The indicator of Pb-TSP was retained,
13    reflecting the evidence that Pb particles of all sizes pose health risks.  The secondary standard
14    was revised to be identical in all respects to the revised primary standards.9  Revisions to the
15    NAAQS were accompanied by revisions to the data handling procedures,  the treatment of
16    exceptional events and the ambient air monitoring and reporting requirements, as well as
17    emissions inventory reporting requirements.10 One aspect of the new data handling requirements
18    is the allowance for the use of Pb-PMio monitoring for Pb NAAQS attainment purposes in
19    certain limited circumstances at non-source-oriented sites.  Subsequent to the 2008 rulemaking,
20    additional revisions were made to the monitoring network requirements as described in chapter 2
21    below.

22         1.2.3   Current Lead NAAQS Review
23          On February 26, 2010, the EPA formally initiated its current review of the air quality
24    criteria and standards for Pb, requesting the submission of recent scientific information on
25    specified topics (75 FR 8934).  Soon after this, the EPA held a science policy workshop to
26    discuss the policy-relevant science, which informed identification of key policy issues and
27    questions to frame the review of the Pb NAAQS (75 FR 20843). Drawing from the workshop
28    discussions, the EPA developed the draft IRP (USEPA, 201 Ic). The draft IRP was made
29    available in late March 2011 for consultation with the CASAC Pb Review Panel and for public
30    comment (76 FR 20347).  This document was discussed by the Panel via a publicly accessible
            9 The current NAAQS for Pb are specified at 40 CFR 50.16.
            10 The current federal regulatory measurement methods for Pb are specified in 40 CFR 50, Appendix G and
      40 CFR part 53. Consideration of ambient air measurements with regard to judging attainment of the standards is
      specified in 40 CFR 50, Appendix R. The Pb monitoring network requirements are specified in 40 CFR 58,
      Appendix D, section 4.5. Guidance on the approach for implementation of the new standards was described in the
      Federal Register notices for the proposed and final rules (73 FR 29184; 73 FR 66964).

                January 2013                    1-7            Draft - Do Not Quote or Cite

-------
 1   teleconference consultation on May 5, 2011 (76 FR 21346; Frey, 201 la). The final IRP,
 2   developed in consideration of CAS AC advice and public comment, was released in November,
 3   2011 (USEPA, 201 la; 76 FR 76972).
 4          In developing the ISA for this review, the EPA held a workshop in December 2010 to
 5   discuss with invited scientific experts preliminary draft materials and released the first external
 6   review draft of the document for CAS AC review and public comment on May 6, 2011 (USEPA,
 7   201 Id; 76 FR 26284; 76 FR 36120).  The CASAC Pb Review Panel met at a public meeting on
 8   July 20, 2011 to review the draft ISA (76 FR 36120). The CASAC provided comments in a
 9   December 9, 2011 letter to the EPA Administrator (Frey and Samet, 2011).  The second external
10   review draft ISA was released for CASAC review and public comment in February 2012
11   (USEPA, 2012b, 77 FR 5247) and was the subject of a public meeting on April 10-11, 2012 (77
12   FR 14783).  The CASAC provided comments in a July 20, 2012 letter (Samet and Frey, 2012).
13   The third external review draft was released for CASAC review and public comment in
14   November 2012 (USEPA, 2012a, 77 FR 70776).  That document, together with this draft Policy
15   Assessment, will be the subject of a public meeting on February 5-6, 2013 (78 FR 938). The
16   final ISA is targeted for release in late spring 2013.
17          As described in the IRP, the EPA developed and released for CASAC review and public
18   comment in June 2011 an REA Planning Document (USEPA, 201 Ib; 76 FR 58509). This
19   document presented a critical evaluation of the information related to Pb human and ecological
20   exposure and risk (e.g., data, modeling approaches) newly available in this review, with a focus
21   on consideration of the extent to which new or substantially revised REAs for health and
22   ecological risk are warranted by the newly available evidence.  Evaluation of the newly available
23   information with regard to designing and implementing health and ecological REAs for this
24   review led us to conclude that the currently available information did not provide a basis for
25   developing new quantitative risk and exposure assessments that would have substantially
26   improved utility for informing the Agency's consideration of health and welfare effects and
27   evaluation of the adequacy of the current primary and secondary standards, respectively (REA
28   Planning Document, sections 2.3 and 3.3, respectively).  The CASAC Pb Panel provided
29   consultative advice on that document and its conclusions at a public meeting on July 21, 2011
30   (76 FR 36120; Frey, 201 Ib). Based on their consideration of the REA Planning Document
31   analysis, the CASAC Pb Review Panel generally concurred with the conclusion that a new REA
32   was not warranted in this review (Frey, 201 Ib). In consideration of the conclusions reached in
33   the REA Planning Document and CASAC's consultative advice, the EPA has not developed
34   REAs for health and ecological risk for this review.  Accordingly, this Policy Assessment
35   considers the risk assessment findings from the last review with regard to any appropriate further
36   interpretation in light of the evidence  newly available in this review.
               January 2013                    1-8            Draft - Do Not Quote or Cite

-------
 1          This draft Policy Assessment is being released for public comment and review by
 2    CASAC. Advice from CASAC and public comments received on this draft will be considered in
 3    preparing the final PA.

 4    1.3   SCOPE OF CURRENT REVIEW: FATE AND MULTIMEDIA PATHWAYS OF
 5         AMBIENT AIR LEAD
 6          The multimedia  and persistent nature of Pb contributes complexities to the review of the
 7    Pb NAAQS unlike issues addressed in other NAAQS reviews.11  As described in section 1.1,
 8    NAAQS are  established to protect public health with an adequate margin of safety, and public
 9    welfare from known or anticipated adverse effects, from air pollutants (substances emitted to
10    ambient air). Since Pb distributes from air to other media and is  persistent, our review of the
11    NAAQS for  Pb considers the protection provided against such effects associated both with
12    exposures to Pb in ambient air and with exposures to Pb that makes its way into other media
13    from ambient air. Additionally, in assessing the adequacy of protection afforded by the current
14    NAAQS, we are mindful of the history of the greater and more widespread atmospheric
15    emissions that occurred in previous years (e.g., under 1978 NAAQS, and prior to establishment
16    of any Pb NAAQS) and that contributed to the Pb that exists in human populations and
17    ecosystems today. Likewise, we also recognize the role of other, nonair sources of Pb now and
18    in the past that also contribute to the Pb that exists in human populations and ecosystems today.
19    As in the last Pb NAAQS review, this backdrop of environmental Pb exposure, and its impact on
20    the populations and ecosystems which may be the subjects  of the currently available scientific
21    evidence, complicates our consideration of the health and welfare protection afforded by the
22    current NAAQS.  In the first section below, we summarize the environmental pathways of
23    human and ecosystem exposures to Pb emitted to ambient air and associated complexities.  The
24    subsequent section briefly discusses the role of historically emitted Pb in our consideration of the
25    adequacy of  the current NAAQ S for Pb.

26         1.3.1   Environmental Distribution and Exposure  Pathways
27          Lead emitted to ambient air is transported through the air and is also distributed to other
28    media through the process of deposition, which may occur  in dry conditions or in association
29    with precipitation, as summarized further in section 2.3 below (ISA, section 3.7.2). Once
30    deposited, the fate of Pb is influenced by the type of surface onto which the particles deposit and
31    by the type and activity  level of transport processes in that location. Precipitation and other
            1: Some aspects of the review of the secondary standard for oxides of nitrogen and sulfur, which involved
      consideration of pollutant transport and fate in nonair media with a focus on impacts to aquatic ecosystems, have
      some similarity to considerations for Pb, while the Pb review also differs in other important aspects.

                January 2013                    1-9             Draft - Do Not Quote or Cite

-------
 1    natural, as well as human-influenced, processes contribute to the fate of such particles, which
 2    affects the likelihood of subsequent human and ecological exposures, e.g., tracking into nearby
 3    houses or transport with surface runoff into nearby water bodies (ISA, sections 3.3 and 4.1). For
 4    example, Pb particles deposited onto impervious surfaces, such as roadway,  sidewalk or other
 5    urban surfaces, may be more available for human contact while they remain  on such surfaces or
 6    are transferred to other human environments, such as on clothing or through  resuspension and
 7    infiltration (ISA, section 4.1.1.1).  Deposited Pb can also be transported (by direct deposition or
 8    stormwater runoff) to water bodies and into associated sediments, which may provide a storage
 9    function for Pb in aquatic ecosystems (ISA, sections 3.3.2 and 7.2.1).  Lead deposited in
10    terrestrial ecosystems can also be incorporated into soil matrices (ISA, sections 3.3.3 and 7.2.1).
11           Figure 1-1 illustrates, in summary fashion, the pathways by which Pb emitted into
12    ambient air can be distributed in the environment and contribute to human and ecosystem
13    exposures.  As shown in this figure,  the multimedia distribution of Pb emitted into ambient air
14    (air-related Pb) contributes to multiple air-related pathways of human and ecosystem exposure
15    (ISA, sections 4.1.1 and 4.7.1).12 As illustrated in Figure 1-1, air-related pathways may involve
16    media other than air. Additionally, as recognized by the figure and discussed more completely in
17    the subsections below, Pb that has not passed through ambient air (nonair Pb) may complicate
18    our consideration of ambient air Pb exposures. Further, the persistence of Pb poses an additional
19    complication, also discussed below,  with regard to consideration of exposures associated with
20    current Pb emissions.
21
             12 The exposure assessment for children performed for the review completed in 2008 employed available
      data and methods to develop estimates intended to inform a characterization of these pathways, as described in the
      rulemaking notices for that review (73 FR 29184; 73 FR 66964) and the final Risk Assessment Report (USEPA,
      2007).

                January 2013                    1-10            Draft - Do Not Quote or Cite

-------
                                                'Current Pb Emissions
                                                    to Ambient Air
                                                                                 Ecosystem
                                                                                 Exposures
            Human Exposures
            (Inhalation and Ingestion)
I (iiniuiuLiuii uuu iiiyesuunj
                      /
                                                   Wildlife, Livestock, Crops
             Human Population
             (body burden)*
                             * Compartments marked by asterisk may also include Pb emitted to air in the past
                             and Pb from nonair sources (released now and in the past). Depending on the
                             compartment, nonair sources may include wastewater releases to surface water or
                             Pb usage in paint, drinking water distribution or food processing/packaging.
            Note: Arrows indicate general pathways by which Pb distributes in environment and human populations. Individual
 1          oathwav sianificance varies with location- and receptor-specific factors.

 2    Figure 1-1.  Pathways of human and ecosystem exposure to lead from ambient air.
 3

 4          1.3.1.1  Human Exposure Pathways
 5           Air-related Pb exposure pathways for humans include inhalation of ambient air or
 6    ingestion of food, water or other materials, including dust and soil, that have been contaminated
 7    through a pathway involving Pb deposition from ambient air (ISA, section 4.1.1.1). Ambient air
 8    inhalation pathways include both inhalation of air outdoors and inhalation of ambient air that has
 9    infiltrated into indoor environments. The air-related ingestion pathways occur as a result of Pb
10    passing through the ambient air, being distributed to other environmental  media and contributing
11    to human exposures via contact with and ingestion of indoor and outdoor  dusts,  outdoor soil,
12    food and drinking water.
13           Lead currently occurring in nonair media may also derive from sources other than
14    ambient air (nonair Pb sources), as summarized further in section 2.3 below (ISA, section 4.7.1).
15    For example, Pb in dust inside some houses or outdoors in some urban areas may derive from the
                January 2013
                                       1-11
Draft - Do Not Quote or Cite

-------
 1    common past usage of leaded paint, while Pb in drinking water may derive from the use of
 2    leaded pipe or solder in drinking water distribution systems (ISA, section 4.1.3.3). We also
 3    recognize the history of much greater air emissions of Pb in the past, such as that associated with
 4    leaded gasoline usage and higher industrial emissions (as summarized in section 2.1.1 below)
 5    which have also contributed to Pb currently occurring in other (nonair) media.
 6          The relative importance of different pathways of human exposure to Pb, as well as the
 7    relative contributions from Pb resulting from recent and historic air emissions and from nonair
 8    sources, vary across the U.S. population as a result of both extrinsic factors, such as  a home's
 9    proximity to industrial Pb sources or its history of leaded paint usage, and intrinsic factors, such
10    as a person's age and nutritional status (ISA, sections 6.1, 6.2, 6.2.1,  6.2.5 and 6.2.6). For
11    example, a predominant Pb exposure pathway for very young children is the incidental  ingestion
12    of indoor dust by hand-to-mouth activity (ISA, section 4.1.1.1). For  adults, diet may be the
13    primary Pb  exposure pathway (2006 CD, section 3.4).  Similarly, the relative importance of air-
14    related and nonair-related Pb also varies with the relative magnitudes of exposure by those
15    pathways, which may vary with different circumstances. For example, relative contributions to a
16    child's total Pb exposure from air-related exposure pathways compared to other (nonair) Pb
17    exposures depend on many factors, including ambient air concentrations and air deposition in the
18    area where the child resides (as well as in the area from which the child's food derives), as well
19    as access to other sources of Pb exposure such as Pb paint, tap water affected by plumbing
20    containing Pb, and lead-tainted products. Studies indicate that in the absence of paint-related
21    exposures, Pb from other sources such as nearby stationary sources of Pb emissions  may
22    dominate a child's Pb exposures (ISA, sections 4.1 and 4.1.3.2; 2006 CD,  section 3.2.3). In
23    other cases, such as children living in older housing with peeling paint or where renovations have
24    occurred, the dominant source of Pb exposure may be dust from lead paint used in the house in
25    the past. Depending on Pb levels in a home's tap water, drinking water can sometimes be a
26    significant source. Lead exposure may also be the result of a mixture of contributions from
27    multiple sources, with no one source dominating.  Our understanding of the relative  contribution
28    of air-related Pb to ingestion exposure pathways is limited by the paucity of studies that parse
29    ingestion exposure pathways with regard to air-related and nonair Pb. Our understanding of the
30    relative contribution of air-related Pb associated with historical emissions  and that from recent
31    emissions is similarly limited.

32         1.3.1.2 Ecosystem Exposure Pathways
33          The distribution of Pb from ambient air to other environmental media also influences the
34    exposure pathways in terrestrial and aquatic ecosystems. Exposure of terrestrial animals and
35    vegetation to air-related Pb can occur by contact with ambient air or by contact with soil, water

                January 2013                   1-12            Draft - Do Not Quote or Cite

-------
 1    or food items that have been contaminated by Pb from ambient air (ISA, section 7.2). Transport
 2    of Pb into aquatic systems similarly provides for exposure of biota in those systems, and
 3    exposures may vary among systems as a result of differences in sources and levels of
 4    contamination, as well as characteristics of the systems themselves.  In addition to Pb contributed
 5    by current atmospheric deposition, Pb may occur in aquatic systems as a result of nonair sources
 6    such as industrial discharges or mine-related drainage, of historical air Pb emissions (e.g.,
 7    contributing to deposition to a water body  or via runoff from soils near historical air sources) or
 8    combinations of different types of sources  (e.g.,  resuspension of sediments contaminated by
 9    urban runoff and surface water discharges).
10          The persistence of Pb contributes an important temporal aspect to lead's environmental
11    pathways, and the time (or lag) associated  with realization of the impact of air Pb concentrations
12    on concentrations in other media can vary with the media (e.g., ISA, section 7.2.2). For example,
13    human exposure pathways most directly involving Pb in ambient air and exchanges of ambient
14    air with indoor air can respond more quickly while pathways involving exposure to Pb deposited
15    from ambient air into the environment (e.g., diet) generally respond more slowly.  An additional
16    influence on the response time for nonair media  is the environmental presence of Pb associated
17    with past, generally higher, air concentrations. For example, after a reduction in air Pb
18    concentrations, the time needed for sediment or  surface soil concentrations to indicate a response
19    to reduced air Pb concentrations might be expected to be longer in areas of more substantial past
20    contamination than in areas with lesser past contamination. Thus, considering the Pb
21    concentrations  occurring in nonair media as a result of air quality conditions that meet the
22    current NAAQS is a complexity of this review, as it also was, although to a lesser degree, with
23    regard to the prior standard in the last review.

24         1.3.2   Considerations Related to Historically Emitted Lead
25          In reviews of NAAQS, the overarching consideration of each review is first focused on
26    the general question as to whether the currently available information supports or calls into
27    question the adequacy of the current standard(s). In addressing that consideration for the
28    NAAQS for Pb, our focus is on Pb emitted to ambient air under conditions meeting the current
29    standard and its potential to cause health or welfare effects as a result of exposures to Pb in air or
30    in other media. We frame the focus in this  way, which differs from that for NAAQS reviews
31    involving other pollutants, in consideration of the persistence of Pb in the environment over time,
32    a characteristic which does not affect reviews of other NAAQS.  In considering the case for Pb,
33    however, we recognize that, because of its persistence, both recent and past Pb emissions to
34    ambient air contribute to current Pb exposures (via multiple exposure pathways, as summarized
35    in section 1.4.1 above).  And we recognize that past Pb emissions in many situations were well in

                January 2013                    1-13            Draft - Do Not Quote or Cite

-------
 1    excess of the current Pb standard. Yet our task in this and every NAAQS review is focused on
 2    assessing the adequacy of the current standard.
 3          Lead emissions, and air concentrations, were higher in the past, and most substantially so
 4    during the time of leaded gasoline usage, as summarized in sections 2.1 and 2.2 below. Nonair
 5    sources of Pb to human exposure and the environment, an additional complicating feature of Pb
 6    NAAQS reviews, were also much higher in the past.  These air and nonair sources contributed
 7    to historical human and environmental exposures. Because of the persistence of Pb, historical
 8    exposures  associated both with air-related and nonair-related sources, have contributed Pb that is
 9    now stored within older humans and ecosystems. For example, concentrations of Pb in the bone
10    and blood  of older members of the U.S. population, who lived during the time of widespread air
11    emissions  associated with leaded gasoline usage (as well as higher industrial emissions) under
12    the previous Pb NAAQS or prior to establishment of any Pb NAAQS, are greater than what
13    would result from air quality conditions allowed by the current, more restrictive NAAQS.
14    Epidemiological  studies of these populations, in which this exposure history is represented by
15    current bone or blood Pb concentrations,  contribute to the overall evidence base regarding lead-
16    related health effects (as discussed in chapter 3 below). Such studies of these historically
17    exposed populations, however, are generally less informative in judging the adequacy of the
18    current primary standard (as discussed in chapter 3  below). This is in contrast to epidemiological
19    studies of very young populations with much shorter and more recent exposure histories (also
20    discussed in chapter 3).
21          The current distribution of Pb in U.S. ecosystems also reflects the widespread and greater
22    air emissions that occurred in the past, under the prior Pb NAAQS and prior to establishment of
23    any Pb NAAQS.13  We lack information on whether adverse effects could be anticipated from
24    the Pb in terrestrial and aquatic ecosystems that results from air quality conditions allowed by the
25    current, more restrictive NAAQS. As is discussed  in section 2.3 below, media in ecosystems
26    across the  U.S. are still recovering from the past period of higher atmospheric emissions, and
27    their responses (e.g., in terms of temporal change in media concentrations) differ with the extent
28    of contamination in individual systems as well as with ecosystem-specific characteristics. Thus,
29    the time required for these ecosystem media to "equilibrate" or come to a "steady-state" that
30    reflects the influence of the current NAAQS (established in 2008) will also vary, and the
31    resulting "steady-state" media concentrations for the range of U.S. ecosystems are unknown.
32    Evidence of effects pertaining to the concentrations associated with the past, higher emissions (as
33    well as from nonair sources), while informative to our understanding of welfare effects
34    associated with environmental Pb generally, does not directly inform our consideration of
             13 The current distribution of Pb in U.S. ecosystems also reflects historical nonair releases.
                January 2013                   1-14            Draft - Do Not Quote or Cite

-------
 1    welfare effects that might be anticipated under the current secondary standard and thus may be
 2    generally less informative in judging the adequacy of the current standard.

 3    1.4   GENERAL ORGANIZATION OF THE DOCUMENT
 4          Following this introductory chapter, this document is organized into three main parts: the
 5    characterization of ambient Pb; lead-related health effects and the primary Pb NAAQS; and lead-
 6    related welfare effects and the secondary Pb NAAQS. The characterization of ambient Pb is
 7    presented in Chapter 2 and includes information on Pb properties in ambient air, current Pb
 8    emissions and air quality patterns, historic trends, and background levels.  Chapter 2 also
 9    describes the Pb NAAQS surveillance and other Pb monitoring networks.  In recognition of the
10    multimedia nature of Pb and the distribution into other media of Pb emitted into the air, Chapter
11    2 also includes information on Pb in media other than air including outdoor dust, soil, surface
12    water and sediment. This chapter provides a frame of reference for exposure and risk-related
13    considerations and subsequent discussion of the Pb NAAQS.
14          Chapters 3 and 4 comprise the second main part of this document, dealing with human
15    health and the primary standard.  These chapters are organized around a series of questions,
16    drawn from the IRP, that address the key policy-relevant issues related to the primary standard.
17    Chapter 3 presents an overview of key policy-relevant health effects evidence and major health-
18    related conclusions from the ISA; an examination of issues related to the quantitative assessment
19    of health risks; key results from quantitative assessments together with a discussion of
20    uncertainty and variability in the results; and discussion of the public health implications of the
21    evidence and  exposure/risk information. In this draft PA, chapter 4 includes staffs preliminary
22    consideration of the scientific evidence and exposure/risk information related to the primary
23    standard and associated preliminary conclusions on the adequacy of the current primary
24    standard.
25          Chapters 5 and 6 comprise the third main part of this document.  These chapters are
26    similarly organized around a series of questions, drawn from the IRP, that address the key
27    policy-relevant issues related to the secondary standard.  Chapter 5 presents an overview of
28    welfare effects evidence related to these key policy-relevant issues and major welfare effects
29    related conclusions from the ISA; an examination of issues related to the screening level
30    ecological risk assessment; and key results from the risk assessment together with a discussion of
31    uncertainty and variability in the results; and discussion of the public welfare implications  of the
32    quantitative assessment with regard to the  current standard.  The final chapter, chapter 6,
33    includes staffs preliminary conclusions related to the adequacy of the current secondary
34    standard.
                January 2013                   1-15            Draft - Do Not Quote or Cite

-------
  1     1.5   REFERENCES

  2     Frey, H.C. (201 la) Letter from Dr. H. Christopher Frey, Chair, Clean Air Scientific Advisory Committee Lead
  3            Review Panel, to Administrator Lisa P. Jackson. Re:  Consultation on EPA's Draft Integrated Review Plan
  4            for the National Ambient Air Quality Standards for Lead. May 25, 2011.

  5     Frey, H.C. (201 Ib) Letter from Dr. H. Christopher Frey, Chair, Clean Air Scientific Advisory Committee Lead
  6            Review Panel, to Administrator Lisa P. Jackson. Re:  Consultation on EPA's Review of the National
  7            Ambient Air Quality Standards for Lead: Risk and Exposure Assessment Planning Document. October 14,
  8            2011.

  9     Frey, H.C. and Samet, J.M. (2011) Letter from Drs. H. Christopher Frey, Chair, Clean Air Scientific Advisory
10            Committee Lead Review Panel, and Jonathan M. Samet, Chair, Clean Air Scientific Advisory Committee,
11            to Administrator Lisa P. Jackson. Re: CAS AC Review of the EPA's Integrated Science Assessment for
12            Lead (First External Review Draft - May 2011). December 9, 2011.

13     ICF International. (2006) Lead Human Exposure and Health Risk Assessments and Ecological Risk Assessment for
14            Selected Areas.  Pilot Phase. Draft Technical Report. Prepared for the U.S. EPA's Office of Air Quality
15            Planning and Standards, Research Triangle Park, NC. December.

16     Jackson, L. (2009) Memorandum from Administrator  Lisa Jackson, Subject: Development of regulations and
17            policies. September 11, 2009. Available at:
18            http://www.epa.gov/ttn/naaqs/pdfs/NAAQSReviewProcessMemo52109.pdf

19     Samet, J.M. and Frey, H.C. (2012) Letter from Drs. Jonathan M. Samet, Chair, Clean Air Scientific Advisory
20            Committee and H. Christopher Frey, Chair, Clean Air Scientific Advisory Committee Lead Review Panel,
21            to Administrator Lisa P. Jackson. Re: CAS AC Review of the EPA's Integrated Science Assessment for
22            Lead (Second External Review Draft - February 2012). July 20, 2012.

23     U.S. Environmental Protection Agency. (1977) Air quality criteria for lead. Research Triangle Park, NC: Health
24            Effects Research Laboratory, Criteria and Special Studies Office; EPA report no. EPA-600/8-77-017.
25            Available from: NTIS, Springfield, VA; PB-280411.

26     U.S. Environmental Protection Agency. (1986a) Air quality criteria for lead. Research Triangle Park, NC: Office of
27            Health and Environmental Assessment, Environmental Criteria and Assessment Office; EPA report no.
28            EPA-600/8-83/028aF-dF. 4v. Available from: NTIS, Springfield, VA; PB87-142378.

29     U.S. Environmental Protection Agency. (1986b) Lead effects on cardiovascular function, early development, and
30            stature: an addendum to U.S. EPA Air Quality Criteria for Lead (1986). In: Air quality criteria for lead, v.
31            1. Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria
32            and Assessment Office;  pp.  A1-A67; EPA report no. EPA-600/8-83/028aF. Available from:  NTIS,
33            Springfield, VA; PB87-142378.

34     U.S. Environmental Protection Agency. (1989) Review of the national ambient air quality standards for lead:
3 5            Exposure analysis methodology and validation: OAQPS staff report. Research Triangle Park, NC: Office of
36            Air Quality Planning and Standards; report no. EPA-450/2-89/011. Available on the web:
37            http://www.epa.gov/ttn/naaqs/standards/pb/data/rnaaqsl  eamv.pdf

38     U.S. Environmental Protection Agency. (1990a) Air quality criteria for lead: supplement to the 1986 addendum.
39            Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and
40            Assessment Office; report no. EPA/600/8-89/049F. Available from: NTIS, Springfield,  VA; PB91-138420.

41     U.S. Environmental Protection Agency. (1990b) Review of the national ambient air quality standards for lead:
42            assessment of scientific and technical information:  OAQPS staff paper. Research Triangle Park, NC: Office

                  January 2013                       1-16              Draft - Do Not Quote or Cite

-------
  1            of Air Quality Planning and Standards; report no. EPA-450/2-89/022. Available from: NTIS, Springfield,
  2            VA; PB91 -206185.  Available on the web: http://www.epa.gov/ttn/naaqs/standards/pb/data/rnaaqsl asti.pdf

  3     U.S. Environmental Protection Agency. (1991) U.S. EPA Strategy for Reducing Lead Exposure. Available from
  4            U.S. EPA Headquarters Library/Washington, D.C. (Library Code EJBD; Item Call Number: EAP
  5            100/1991.6; OCLC Number 2346675). http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_pr.html

  6     U.S. Environmental Protection Agency. (2005) Project Work Plan for Revised Air Quality Criteria for Lead.
  7            CASAC Review Draft. National Center for Environmental Assessment, Research Triangle Park, NC.
  8            NCEA-R-1465. Available at: http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_cr_pd.html

  9     U.S. Environmental Protection Agency. (2006a) Plan for Review of the National Ambient Air Quality Standards for
10            Lead. Office of Air Quality Planning and Standards, Research Triangle Park, NC. Available at:
11            http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_cr_pd.html

12     U.S. Environmental Protection Agency. (2006b) Air Quality Criteria for Lead. Washington, DC, EPA/600/R-
13            5/144aF. Available online at: http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_cr.html

14     U.S. Environmental Protection Agency. (2006c) Analysis Plan for Human Health and Ecological Risk Assessment
15            for the Review of the Lead National Ambient Air Quality Standards. Office of Air Quality Planning and
16            Standards, Research Triangle Park, NC. Available at:
17            http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_cr_pd.html

18     U.S. Environmental Protection Agency. (2007a) Lead: Human Exposure and Health Risk Assessments for  Selected
19            Case Studies, Volume I. Human Exposure and Health Risk Assessments - Full-Scale and Volume II.
20            Appendices. Office of Air Quality Planning and Standards, Research Triangle Park, NC. EPA-452/R-07-
21            014a and EPA-452/R-07-014b. http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_cr_td.html

22     U.S. Environmental Protection Agency. (2007b) Review of the National Ambient Air Quality Standards for Lead:
23            Policy Assessment of Scientific and Technical Information, OAQPS Staff Paper. Office of Air Quality
24            Planning and Standards, Research Triangle Park, NC. EPA-452/R-07-013. Available at:
25            http://www.epa.gOv/ttn/naaqs/standards/pb/s jb_cr_sp.html

26     U.S. Environmental Protection Agency. (201 la) Integrated Review Plan for the National Ambient Air Quality
27            Standards for Lead. Research Triangle, NC. EPA-452/R-11-008.  Available online at:
28            http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_index.html

29     U.S. Environmental Protection Agency. (201 Ib) Review of the National Ambient Air Quality Standards for Lead:
3 0            Risk and Exposure Assessment Planning Document. Office of Air Quality Planning and Standards,
31            Research Triangle Park, NC. EPA/452/P-11 -003. Available at:
32            http://www.epa.gOv/ttn/naaqs/standards/pb/s jb_2010_pd.html

33     U.S. Environmental Protection Agency. (201 Ic) Integrated Review Plan for the National Ambient Air Quality
3 4            Standards for Lead. External Review Draft. Research Triangle, NC. EPA-452/D-11 -001. Available online
35            at: http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_index.html

36     U.S. Environmental Protection Agency. (201 Id) Integrated Science Assessment for Lead  (First External Review
37            Draft). Washington, DC, EPA/600/R-10/075A. Available online  at:
3 8            http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_index.html

39     U.S. Environmental Protection Agency. (2012a) Integrated Science Assessment for Lead (Third External Review
40            Draft). Washington, DC, EPA/600/R-10/075C. Available online at:
41            http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_2010_isa.html
                  January 2013                       1-17              Draft - Do Not Quote or Cite

-------
1    U.S. Environmental Protection Agency. (2012b) Integrated Science Assessment for Lead (Second External Review
2            Draft). Washington, DC, EPA/600/R-10/075B. Available online at:
3            http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_index.html
                 January 2013                      1-18             Draft - Do Not Quote or Cite

-------
 1                                   2  AMBIENT AIR LEAD

 2          The focus for this Pb NAAQS review is on Pb derived from those sources emitting Pb to
 3    ambient air.  As noted in section 1.3, air emissions contribute to concentrations in multiple
 4    environmental media and the role of the nonair media is enhanced by the persistent nature of Pb.
 5    Consequently this chapter discusses our current understanding of Pb in ambient air and of
 6    ambient air-related Pb in other media.
 7          Lead emitted to the air is predominantly in particulate form, with the particles occurring
 8    in various sizes (ISA, section 3.3.1).  While in  some limited circumstances associated with
 9    incomplete combustion of leaded gasoline, Pb may be emitted in gaseous form, atmospheric
10    conditions readily contribute to condensation into particles (ISA, section 3.2). Once emitted, Pb
11    can be transported long or short distances depending on their size, which influences the amount
12    of time spent in the aerosol phase.  Consistent with previous evidence, recent research on
13    particulate matter with mass median diameter of 2.5 and of 10 micrometers (PM2.5 and PMio)
14    confirms the transport of airborne Pb appreciable distances from its sources. For example,
15    samples collected at altitude over the Pacific Ocean, as well as the seasonal pattern of Pb-PM2.s
16    at rural sites in the western U.S. indicate transport of Pb from sources in Asia, although such
17    sources have been estimated to  contribute less than 1 ng/m3 to western U.S. Pb concentrations
18    (ISA, section 3.3.1; Murphy et al., 2007). As a generality, larger particles tend to deposit more
19    quickly, within shorter distances from emissions points, while smaller particles remain in aerosol
20    phase and travel longer distances before depositing (ISA, section 2.2.1). As summarized in
21    section 2.2.2 below, airborne concentrations of Pb near sources are much higher, and the
22    representation of larger particles generally greater, than at sites  not directly influenced by
23    sources.
24          In this chapter, we discuss the current information on ambient Pb regarding sources and
25    emissions (section 2.1), current ambient air monitoring methods and networks and associated
26    measurements (section 2.2) and the contribution of ambient air Pb to Pb in other media (section
27    2.3).

28    2.1   SOURCES AND EMISSIONS TO AMBIENT AIR
29          In this section we describe the most recently available information on sources and
30    emissions of Pb  into the ambient air. The section does not provide a comprehensive list of all
31    sources of Pb, nor does it provide estimates of emission rates or emission factors for all source
32    categories.  Rather, the discussion here is intended to identify the larger source categories, either
33    on a national or local scale, and generally describe their emissions and distribution within  the
34    U.S.

                January 2013                     2-1             Draft - Do Not Quote or Cite

-------
 1          The primary data source for this discussion is the National Emissions Inventory (NEI) for
 2    20081. The NEI is a comprehensive and detailed estimate of air emissions of both criteria and
 3    hazardous air pollutants from air emissions sources. The NEI is generally prepared every three
 4    years by the EPA based primarily upon emission estimates and emission model inputs provided
 5    by state, local, and tribal air agencies for sources in their jurisdictions, and supplemented by data
 6    developed by the EPA.  Some of these estimates are required by regulation while some are
 7    voluntarily reported. For example, states are required to report Pb emissions from facilities
 8    emitting more than 5 tons of Pb per year (tpy) and from facilities emitting greater than threshold
 9    amounts for other  criteria pollutants (e.g., 100 tpy of particulate matter or volatile organic
10    compounds;  CFR 51, subpart A). Estimates of Pb emissions presented in this document (and in
11    the ISA) are drawn from the 2008 NEI version 3.2  As a result of various Clean Air Act
12    requirements, emissions standards implemented since 2008 for a number of source categories
13    represented in the NEI are projected to result in considerably lower emissions at the current time
14    or in the near future.
15          The following sections present information relative to 2008 Pb emissions on a national
16    and local scale. Lead is emitted from a wide variety of source  types, some of which are small
17    individually but for which the cumulative emissions are large,  and some for which the opposite is
18    true. For example, a source category may be composed of many small (i.e., low-emitting)
19    sources or of just a few very large (high-emitting) sources. Temporal trends in the national totals
20    of Pb emissions are presented in Section 2.1.1. Information about the emissions source types or
21    categories that are large on a national scale as of 2008 is presented in Section 2.1.2, while
22    information on the sources that are large at the local scale  is presented in Section 2.1.3.
23    Additional information on data sources for, limitations of and our confidence in the information
24    summarized here is described in Appendix 2A.

25         2.1.1  Temporal Trends on a National Scale
26          Figure 2-1  shows the substantial downward trend in Pb emissions that has occurred over
27    the past several decades. The most dramatic reductions in Pb emissions occurred prior to 1990
28    in the transportation sector due to the removal of Pb from gasoline.  Lead emissions were further
29    reduced substantially between 1990 and 2008, with significant reductions occurring in the metals
30    industries at least in part as a result of national emissions standards for hazardous air pollutants.
31
            1 http://www.epa.gov/ttn/chief/net/2008inventory.html
            2 With regard to Pb emissions, the 2008 NEI, version 3 (January 2013) has been augmented with sources
      not included in the 2008 NEI, version 2.

                January 2013                    2-2            Draft - Do Not Quote or Cite

-------
 1
 2
 3
 4
 5
 6
 7
          250 i
          200 -
        1- 150
        o
        o
        o
        E 100 -
                               Highway Vehicles
                               Metal Workingand Mining
                               Fuel Combustion
                               Piston Engine Aircraft
                               Miscellaneous
               1970    1975    1980    1985   1990   1995    1999    2002    2005   2008
 9    Figure 2-1.  Temporal trend in air emissions: 1970-2010.
10
11         2.1.2   Sources and Emissions on National Scale - 2008
12           As indicated in Figure 2-1, the largest source sector emitting Pb into the atmosphere on a
13    national scale is aviation gasoline usage by piston engine driven aircraft.  The next largest
14    nationally is metal working and mining. Considering the national estimates at a more detailed
15    scale, the largest source categories emitting Pb into the atmosphere  on a national scale, after
16    emissions from aircraft operating on leaded fuel, are boilers and process heaters (fuel
17    combustion) which, while individually are generally small sources,  are large when aggregated
18    nationally (Table 2-1). The next largest categories are various metals industries, including lead-
19    specific industries  (Table 2-1).  Together these and other sources were estimated to emit just
20    under a thousand tpy of Pb in the U.S. in 2008.
                January 2013
2-3
Draft - Do Not Quote or Cite

-------
1   Table 2-1.  U.S. lead emissions by source categories estimated to emit at least 4 tpy.
Source Category Description
ALLCATEGORIESA
Aircraft operating on leaded fuel B
Industrial/Commercial/lnstitutional Boilers & Process Heaters
Utility Boilers
Iron and Steel Foundries
Integrated Iron and Steel Manufacturing
Steel Manufacturing: Electric Arc Furnaces
Secondary Lead Smelting
Primary Lead Smelting
Primary Copper Smelting
Mining
Military Base
Cement Production
Glass Manufacturing
Battery Manufacturing
Secondary Non-ferrous Metals (other)
Primary Non-ferrous Metals (other)
Carbon Black
Pulp and Paper Production
Secondary Copper Smelting
Fabricated Metal Products Manufacturing
Residential Heating
Municipal Waste Incineration
Commercial Marine Vessels
Sewage Sludge Incineration
Mineral Products Manufacturing
2008
Emissions (tons)
950
550
64
51
30
27
22
20
19 =
17
15
13
8
8
7
7
7
6
6
5
5
5
5
5
4
4
A - Emissions estimate totals from 2008 National Emissions Inventory, version 3 (January 2013) for point sources and
2008 NEI version 2 for nonpoint sources (residential heating).
B - This category includes Pb emitted at or near airports as well as Pb emitted in-flight. Lead emissions at or near
airports comprise 46% of the total aircraft lead emissions inventory. Emissions value based on EPA estimates.
C - There is some uncertainty regarding the total emissions estimate for this source category in which there is one
operational smelter, which is planning to cease the existing smelter operations at this site by April 2014 (DRRC, 2010).
              January 2013
2-4
Draft - Do Not Quote or Cite

-------
         Explanation of aggregation approaches used for Tables 2-1 and 2-2:

         Facilities have numerous processes that can fall into different source categories and the NEI includes process-specific
         emissions estimates. Source categories are groups of facilities that can be considered as the same type of emissions
         source. In order to present the emissions for source categories (e.g., secondary copper smelting) rather than for processes
         (e.g., Secondary Metal Production, Copper or Rotary Furnace) in Table 2-1, we aggregated processes for each facility and
         then present national estimates for source categories. The source categories used were assigned using a three-tiered
         approach. First, processes known to be affected by sector-specific rules were set to the source categories. This was done
         for Utility Boilers,  Portland Cement plants,  Electric Arc Furnaces, Municipal Waste Combustors, and Taconite Ore facilities
         (mapped to Integrated Iron and Steel Manufacturing). Other source categories did not use this first tier because the
         processes in the inventory have not yet been mapped to other rules.  Second, for processes that clearly map to source
         categories, the inventory process descriptions (Source Classification Codes) were used to assign the  source category.  A
         good example of this is for Industrial, Commercial, and Institutional Boilers and Process Heaters. For all remaining
         processes, the Facility Type inventory field was mapped to a source category.  Facility Types are the basis for aggregation
         used in Table 2-2. Facility Types in the NEI were set manually by EPA staff for facilities greater than 0.5 tons of Pb and
         using the North American Industrial Classification System (NAICS) codes for smaller facilities.  This (setting of the facility
         type) was done with consideration of the primary activity identified for the facility, which usually confirmed the NAICS code.

         A facility only has a single Facility Type but can have multiple processes and source categories. For example, some
         facilities are secondary metal processing plants for copper, aluminum, and non-ferrous metals, which  divides their
         emissions into Secondary Copper and  Secondary Non-Ferrous metal source categories.  In these cases, the facility
         website was reviewed to try to assess the predominant activity and the NAICS code was considered as well, and a Facility
         Type was set using the best judgement of  EPA staff. However, the emissions for these facilities are split across multiple
         processes as summed in Table 2-1. To prevent double counting of facility and state counts in Table 2-2, the Facility Type
         was used so that each facility shows up only once in this table.
 2           2.1.2.1  Stationary Sources

 3            Since the last review of the Pb NAAQS, the EPA has completed a number of regulations

 4    which will result in reduced Pb emissions from stationary sources regulated under the Clean Air

 5    Act sections 112 and 129.  For example, in January 2012, the EPA updated the National

 6    Emission Standards for Hazardous Air Pollutants (NESHAP) for Secondary Lead Smelting (77

 7    FR 555).  These amendments to the original maximum achievable control technology standards

 8    apply to facilities nationwide that use furnaces to recover lead from lead-bearing scrap, mainly

 9    from automobile batteries (15 existing  facilities, one under construction).  By the effective date

10    in 2014, this action is estimated to result in a lead emissions reduction of 13.6 tpy across the

11    category (a 68% reduction). Also, the NESHAP for Primary Lead Smelting was revised in 2011

12    (76 FR 70834) and more than a dozen additional EPA actions taken in the past 5 years, which

13    would not be reflected in the 2008 NEI estimates, will result in Pb emissions reductions

14    (Appendix 2B).

15
                  January 2013                       2-5              Draft - Do Not Quote or Cite

-------
 1         2.1.2.2  Mobile Sources
 2           Forty years ago, combustion of leaded gasoline was the main contributor of Pb to the air.
 3    In the early 1970s, the EPA set national regulations to gradually reduce the Pb content in
 4    gasoline. In 1975, unleaded gasoline was introduced for motor vehicles equipped with catalytic
 5    converters. The EPA banned the use of leaded gasoline in highway vehicles after December
 6    1995.
 7           Lead emissions from piston-engine aircraft operating on leaded fuel  are currently the
 8    largest source of Pb air emissions on a national scale.  Lead is added to aviation gasoline
 9    (commonly referred to as "avgas") used in most piston-engine aircraft in order to boost octane
10    and prevent engine knock.3 The most commonly used avgas, 100 Octane Low Lead, contains up
11    to 2.12 grams Pb per gallon (ASTM D 910). The Federal Aviation Administration estimates that
12    in 2008, 248 million gallons of avgas were consumed in the U.S.4 contributing an estimated 550
13    tons of Pb to the air that comprise 57% of the national Pb inventory.5 Leaded avgas is used at
14    approximately 20,000 airport facilities in the U.S.
15           The EPA is currently collecting and evaluating information regarding emissions and air
16    concentrations of lead resulting from avgas combustion by piston-engine aircraft. This is part of
17    an ongoing investigation under section 231  of the Clean Air Act into the potential for these
18    emissions to cause or contribute to air pollution that may reasonably be anticipated to endanger
19    public health or welfare.  This evaluation by the EPA is occurring separate from the NAAQS
20    review.  The EPA's investigation includes substantial analytical work. The  timeline for
21    completion of this investigation and possible issuance of a final endangerment determination
22    includes completion  of necessary modeling and monitoring information and other data,
23    development of a proposal which will be published for public comment, review and analysis of
24    comments received and issuance of the final determination. If the EPA issues a positive
25    determination that Pb emissions from aircraft engines cause or contribute to air pollution that
26    may reasonably be anticipated to endanger public health or welfare, the EPA would then be
27    required to propose and promulgate emissions standards to control  aircraft engine Pb emissions,
28    and the Federal Aviation Administration would be  required to promulgate regulations addressing
             3 Lead is not added to jet fuel used in commercial aircraft, military aircraft, or other turbine engine aircraft.
             4 U.S. Department of Transportation Federal Aviation Administration Aviation Policy and Plans. FAA
      Aerospace Forecast Fiscal Years 2010-2030. p.99. Available at:
      http://www.faa.gov/about/office org/headquartersoffices/apl/aviation forecasts/aerospace forecasts/2010-
      2030/media/2010%20Forecast%20Doc.pdf This document provides historical data for 2000-2008 as well as
      forecast data.
             5 EPA (2010) Calculating Piston-Engine Aircraft Airport Inventories for Lead for the 2008 National
      Emissions Inventory. EPA-420-B-10-044. Available at:
      http://www.epa.gov/otaq/regs/nonroad/aviation/420bl0044.pdf

                January 2013                    2-6             Draft - Do Not Quote or Cite

-------
 1    the fuel used by those aircraft. More information about EPA's actions is available at
 2    www.epa.gov/otaq/aviation.htm.
 3          Vehicles used in racing are not regulated by the EPA under the Clean Air Act and can
 4    therefore use alkyl-Pb additives to boost octane. The National Association for Stock Car Auto
 5    Racing (NASCAR) formed a voluntary partnership with the EPA with the goal of permanently
 6    removing alkyl-Pb from racing fuels used in the racing series now known as the Sprint Cup, the
 7    Nationwide Series and the Camping World Truck Series.  The major NASCAR race series now
 8    use unleaded fuels.
 9          Due to the presence of Pb as a trace contaminant in gasoline, diesel fuel and lubricating
10    oil, cars, trucks, and engines operating in nonroad equipment, marine engines and jet aircraft
11    emit small amounts of Pb (ISA, Section 3.2.2.6). Additional mobile sources of Pb include brake
12    wear, tire wear, and loss of Pb wheel weights (ISA, Section 3.2.2.6).

13         2.1.2.3  Natural Sources and Long-range Transport
14          Some amount of Pb in the air derives from natural sources, such as volcanoes,  sea salt,
15    and windborne soil particles from areas free of anthropogenic activity and some may also derive
16    from anthropogenic sources of airborne Pb located outside of the U.S. (ISA, section 3.5.5).
17    Emissions estimates for these sources, as well as wild forest fires and biogenic sources have not
18    been developed for the NEI and quantitative estimates for these processes remain an area of
19    significant uncertainty. Based on several different approaches, the ISA identifies several
20    estimates of the concentration of airborne Pb derived from natural sources. All of the  estimates
21    are no higher than 1 nanogram per cubic meter (ng/m3), and range down as low as 0.02 ng/m3
22    (ISA, section 3.5.5). The data available to derive such an estimate are limited and such a value
23    might be expected to vary geographically with the natural distribution of Pb.
24          Another contribution to U.S. airborne Pb concentrations is long-range transport such as
25    that associated with air masses carrying Pb from sources in Asia, where controls on Pb emissions
26    have lagged those in the U.S. and Canada (ISA,  section 3.5.5; Osterberg et al., 2008).  The most
27    recent estimates of contributions from Asia, however, conclude that the Asian contribution to
28    U.S. airborne Pb concentrations is generally less than  1 ng/m3 (ISA,  section 3.5.5; Murphy,
29    2007; Ewing et al., 2010).

30         2.1.2.4  Previously Deposited or Released Lead
31          Lead-bearing particles that occur on surface soils or built surfaces as a result of previous
32    or historic deposition can be a source of airborne Pb as a result of particle resuspension and may
33    be a significant source of airborne Pb in areas near major sources of Pb emissions.  Outdoor dust
34    may be resuspended into the air by wind or human-induced mechanical forces, such that the
35    main drivers of particle resuspension are typically mechanical stressors such as vehicular traffic,

                January 2013                    2-7            Draft - Do Not Quote or Cite

-------
 1    construction and agricultural operations, and, generally to a lesser extent, the wind (ISA, section
 2    3.3.1.3; 2006 CD, section 2.3.3). Wind resuspension is often defined in terms of a resuspension
 3    rate, which is the fraction of a surface contaminant released per unit time.  Resuspension rates
 4    are dependent on many factors, including wind speed, soil/surface moisture, particle sizes,
 5    presence of saltating particles and presence of vegetation; typical values range over several
 6    orders of magnitude (ISA, section 3.3.1.3; 2006 CD, section 2.3.3).  Vehicular resuspension
 7    results from shearing stress of tires or turbulence generated by a passing vehicle and can be
 8    affected by a number of factors including vehicle size, vehicle speed, moisture and particle size
 9    (ISA, section 3.3.1.3; 2006 CD, section 2.3.3).  Additionally, the amount of material available
10    for resuspension can be influenced by removal processes such as surface runoff associated with
11    rainfall (ISA, sections 2.2.1, 3.3.2 and 3.3.2.4). Rather than a continuous process, resuspension
12    may occur as a series of events.  Short episodes of conditions or factors conducive to
13    resuspension may dominate annual averages of upward flux (2006 CD, p. 2-65).  Variability and
14    uncertainty in these factors, and with regard to surface soil/dust composition, affect quantitative
15    emissions estimates for these processes (2006 CD, section 2.3.3).6
16           Consideration of Pb concentrations in air and surface soil/dust in different types of
17    locations also informs our understanding of the relative importance of resuspension of previously
18    deposited dust particles as an influence on airborne Pb concentrations.  The relative importance
19    would be expected to vary with site-specific circumstances, such as the magnitude of Pb
20    concentrations in the surface dust and air Pb contributions from nearby sources of new Pb
21    emissions, as well as with variation in the forces that influence particle resuspension.  For
22    example, the contribution of resuspension to airborne concentrations appears to be greatest in
23    areas of highly contaminated surface dust such as at  historically active industrial  sites, which
24    may or may not be currently active (ISA, sections 3.3.1.3  and 3.5.3).
25           Emissions from currently active metals industries include those related to current
26    industrial activity  and to resuspension of previously  deposited material (often a component of
27    "fugitive" emissions estimates7). Accordingly, the relative magnitude of air Pb concentrations
28    associated with no longer active industries would be expected to generally be lower than that for
             6 Quantitative estimates of resuspension-related emissions associated with many active industrial sources
      (particularly metals sources) are included within the NEI, although such emissions associated with previously and
      no longer active Pb sources are not as generally included in the NEI.
             7 For example, emissions factors have been established to estimate fugitive emissions from resuspension of
      previously deposited material as a result of vehicular traffic on facility roadways (USEPA, 1996-2011). Where used,
      these estimates are combined with estimates for "process fugitives" (emissions that escape capture by control
      devices) to estimate total fugitive emissions from a facility. Accordingly, control of resuspension resulting from
      facility roadways, buildings or other property may be part of a strategy to meet regulatory emissions requirements
      (e.g., national emissions standards for hazardous air pollutants for secondary lead smelting, 77 FR 556).

                 January 2013                     2-8             Draft - Do Not Quote or Cite

-------
 1    active industries yet greater than that in locations somewhat removed from industrial sources.
 2    The limited available data for comparison generally confirms this (section 2.2.2.2 below).
 3           Lead-bearing particulate matter in other, non-industrial, locations of appreciable historic
 4    lead contamination, such as in soils or on surfaces in older urban areas or near older
 5    transportation corridors may, if disturbed, also become suspended into the air and contribute to
 6    air Pb concentrations. In older transportation corridors or other locations not influenced by
 7    active industries, the significance of resuspension (e.g., in terms of resultant air Pb
 8    concentrations) appears to be much less that that associated with active industries or now closed
 9    industries with substantial emissions in the past.8 For example, the available data indicate that
10    current Pb concentrations near roadways are  substantially lower than those near large, currently
11    active industrial sources (see ISA, section 3.3.1.3 and Figures 2-9 and 2-11 below).  In general,
12    air Pb concentrations at sites described as not influenced by an active industry are much lower
13    than those near active sources (see ISA, section 3.5.1.2 and Figure 2-11 below).

14         2.1.3   Sources and Emissions on Local Scale
15           Based on the 2008 estimates, the highest emissions in specific (local) situations are
16    different types of metals industries, 23 of which had 2008 estimates greater than or equal to 1.0
17    tpy Pb (Table 2-2).  The geographic distribution of the facilities summarized in Table 2-2 is
18    presented in Figure 2-2.
19
              Mass-balance analyses of emissions in southern California newly available in the last review suggested
      that Pb in resuspended road dust may represent between 40% and 90% of Pb emissions in some areas (2006 CD, p.
      2-65; ISA, section 3.2.2.7). Air Pb monitoring data near roadways, however, including those in California, indicate
      Pb concentrations well below those near significant industrial sources and below the current Pb NAAQS (e.g., ISA,
      section 3.5.1.2).

                January 2013                     2-9             Draft - Do Not Quote or Cite

-------
1   Table 2-2.  Facilities estimated to emit at least 0.50 tpy in 2008.
Facility Type
Primary Lead Smelting Plant
Steel Mill
Primary Copper Smelting/Refining Plant
Military Base
Secondary Lead Smelting Plant
Carbon Black Plant
Primary Non-ferrous Metal Smelting/Refining Plant
(not Lead, Aluminum, or Copper)
Mine or Quarry
Ethanol Biorefinery
Portland Cement Plant
Pulp and Paper Plant
Calcined Pet Coke Plant
Battery Plant
Wood Products Plant
Foundry, Iron and Steel
Secondary Copper Smelting/Refining Plant
Foundry, Non-ferrous
Secondary Non-ferrous Metal Smelting/Refining Plant
(not Lead, Aluminum, or Copper)
Coke Battery
Electricity Generation via Combustion
Chemical Plant
Airport
Automobile/Truck or Parts Plant
Wastewater Treatment Facility
Munition or Explosives Plant
Fabricated Metal Products Plant
Municipal Waste Combustor
Glass Plant
Lumber/Sawmill
Plastic, Resin, or Rubber Products Plant
Taconite Processing Plant
Petrochemical Plant
Bulk Terminal or Bulk Station
Facilitk
No.
Facilities
1
14
3
4
5
1
1
5
2
1
2
1
3
1
3
1
3
2
2
2
1
6
1
1
1








;s emitting
No.
States
1
8
2
3
5
1
1
2
2
1
2
1
3
1
3
1
2
2
2
2
1
4
1
1
1








>1.0tpyA.B
Facility
Emissions
19.2°
1.0-9.8
3.1-8.8
1.4-7.2
1.4-7.1
6.3
5.4
1.6-3.8
1.5-3.8
3.3
2.5-3.0
2.5
1.5-2.5
2.3
1.1-2.3
2.3
1.2-2.0
1.2-2.0
1.0-1.8
1.0-1.7
1.7
1.0-1.3
1.2
1.0
1.0








Facilities c
< 1.0 and >
No.
Facilities

10

3
3





2

1

10

1
1

7
1
52
3
2
2
6
3
2
2
1
1
1
1
imitting
:0.50A.C
No.
States

9

3
2





2

1

8

1
1

5
1
17
2
2
2
5
3
2
1
1
1
1
1
A - Emissions totals from 2008 National Emissions Inventory, version 3 (January 2013), except in the case of airports for which EPA specific
estimates were used.
B - This category includes facilities with total emissions estimates greater than or equal to 0.95 tpy.
C - This category includes facilities with total emissions estimates greater than or equal to 0.495 and less than 0.95 tpy.
D - There is some uncertainty regarding this total emissions estimate for this facility, the only remaining operational primary Pb smelter in the
U.S., which is planning to cease the existing smelter operations at this site by April, 2014 (DRRC, 2010).
              January 2013
2-10
Draft - Do Not Quote or Cite

-------
                                                                                                 Airports with estimates <1.0 and =>0.50 tpy"
                                                                                                 Airports with estimates <5.0 and =>1.0 tpy"
                                                                                                 Study Airports
                                                                                                 Facilities with estimates <10 and =>0.50 tpy"
                                                                                                 Facilities with estimates <5.0 and =H.0 tpy*
                                                                                             ^P  Facilities with estimates =* 5.0 tpy
                  * Includes facilities with emissions greater than or equal to 1.0 tpy (after rounding to 1 decimal place).
                  ** Includes facilities with emissions greater than or equal to 0.50 tpy (after rounding to 2 decimal places) and less than 1.0 (after rounding to 1 decimal place).
1                 Study airports are described in section 2.2 below.

2     Figure 2-2.    Geographic distribution of facilities  estimated to emit at least 0.50 tpy in  2008.
                                           January 2013
2-11
Draft - Do Not Quote or Cite

-------
 1    2.2   AMBIENT AIR QUALITY
 2          The EPA and state and local agencies have been measuring Pb in the atmosphere since
 3    the 1970s. In response to reduced emissions (see section 2.1), Pb concentrations have decreased
 4    dramatically over that period.  Currently, the highest concentrations occur near some metals
 5    industries where some individual locations have concentrations that exceed the NAAQS. This
 6    section describes the ambient Pb measurement methods, the sites and networks where these
 7    measurements are made, and how the ambient Pb concentrations vary geographically and
 8    temporally.

 9         2.2.1   Air Monitoring
10          Ambient air Pb concentrations are measured by five national monitoring networks.  The
11    networks include the State and Local Air Monitoring Sites (SLAMS) intended for Pb NAAQS
12    surveillance, the PM2.5 Chemical Speciation Network (CSN), the Interagency Monitoring of
13    Protected Visual Environments (IMPROVE) network, the National Air Toxics Trends Stations
14    (NATTS) network, and the Urban Air Toxics Monitoring program.  All of the data from these
15    networks are accessible via EPA's Air Quality System (AQS):
16    http://www.epa.gov/ttn/airs/airsaqs/. In addition to these networks, various environmental
17    organizations have operated other sampling sites yielding data (which may or may not be
18    accessible via AQS) on ambient air concentrations of Pb, often for limited periods and/or for
19    primary purposes other than quantification of Pb itself. The subsections below describe each
20    network and the Pb measurements made at these sites.

21         2.2.1.1  Lead NAAQS Surveillance Network
22          This section describes sample collection, analysis and network aspects for the Pb SLAMS
23    network, the main purpose of which is surveillance of the Pb NAAQS.  As such, the EPA
24    regulates how this monitoring is conducted in  order to ensure accurate and comparable data for
25    determining compliance with the NAAQS.  The code of federal regulations (CFR) at parts 50, 53
26    and 58 specifies required aspects of the ambient monitoring program for NAAQS pollutants.9 In
27    order to be used in NAAQS attainment designations, ambient Pb concentration data must be
28    obtained using either the federal reference method (FRM) or a federal equivalent method (FEM).
29    The indicator for the current Pb NAAQS is Pb-TSP.  However, in some situations,10 ambient Pb-
            9 The federal reference methods (FRMs) for sample collection and analysis are specified in 40 CFR part 50,
      the procedures for approval of FRMs and federal equivalent methods are specified in 40 CFR part 53 and the rules
      specifying requirements for the planning and operations of the ambient monitoring network are specified in 40 CFR
      part 58.
            10 The Pb-PM10 measurements may be used for NAAQS monitoring as an alternative to Pb-TSP
      measurements in certain conditions defined in 40 CFR part 58 Appendix D paragraph 2.10.1.2. These conditions

                January 2013                   2-12            Draft - Do Not Quote or Cite

-------
           concentrations may be used in judging nonattainment. Accordingly, FRMs have been
 2    established for Pb-TSP and for Pb-PMi0.  The current FRM for the measurement of Pb-TSP is
 3    provided in 40 CFR part 50 Appendix G.  This FRM includes sampling using a high-volume
 4    TSP sampler that meets the design criteria identified in 40 CFR part 50 Appendix B and sample
 5    analysis for Pb content using flame atomic absorption. There are 24 FEMs currently approved
 6    for Pb-TSP.11 All 24 FEMs are based on the use of high-volume TSP samplers and a variety of
 7    approved equivalent analysis methods.
 8           A new FRM for Pb-PMio was promulgated as part of the 2008 review.  This FRM is
 9    based on the PMio sampler defined in 40 CFR part 50 Appendix J coupled with x-ray
10    fluorescence (XRF) analysis.  In addition, one FEM for Pb-PMio has been finalized for the
11    analysis of Pb-PMio based on inductively coupled plasma mass spectroscopy (TCP-MS).
12           The current Pb monitoring network design requirements for NAAQS compliance
13    purposes (40 CFR part 58, Appendix D, paragraph 4.5) include two types of monitoring sites -
14    source-oriented monitoring sites and non-source-oriented monitoring sites. Source-oriented
15    monitoring sites are required near sources of air Pb emissions which are expected to or have been
16    shown to contribute to ambient air Pb concentrations in excess of the NAAQS. At  a minimum,
17    there must be one source-oriented site located to measure the maximum Pb concentration in
18    ambient air resulting from each non-airport Pb source estimated to emit 0.50 or more tons of Pb
19    per year and from each airport estimated to emit 1.0 or more tons of Pb per year.12 The EPA
20    Regional Administrators may require additional monitoring beyond the minimum requirements
21    where the likelihood of Pb air quality violations is significant.  Such locations may  include those
22    near additional industrial Pb sources, recently closed industrial sources and other sources of
23    resuspended Pb dust,  as well as airports where piston-engine aircraft emit Pb (40 CFR, part 58,
24    Appendix D, section 4.5(c)).
25           Monitoring agencies are also required, under 40 CFR, part 58, Appendix D, to conduct
26    non-source-oriented Pb monitoring at the NCore sites required in Core metropolitan with a
27    population of 500,000 or more.13 NCore is a new network of multipollutant monitoring  stations
      include where Pb concentrations are not expected to equal or exceed 0.10 micrograms per cubic meter as an
      arithmetic three-month mean and where the source of Pb emissions is expected to emit a substantial majority of its
      Pb in the size fraction captured by PM10 monitors.
            11 A complete list of FEM can be found at the following webpage -
      http://www.epa.gov/ttn/amtic/files/ambient/criteria/reference-equivalent-methods-list.pdf
            12 The Regional Administrator may waive the requirement in paragraph 4.5(a) for monitoring near Pb
      sources if the State or, where appropriate, local agency can demonstrate the Pb source will not contribute to a
      maximum three-month average Pb concentration in ambient air in excess of 50 percent of the NAAQS level based
      on historical monitoring data, modeling, or other means (40 CFR, part 58, Appendix D, section 4.5(a)(ii)).
            13 Defined by the US Census Bureau - http://www.census.gov/population/www/metroareas/metroarea.html

                January 2013                   2-13            Draft - Do Not Quote or Cite

-------
 1    intended to meet multiple monitoring objectives. The NCore stations are a subset of the state and
 2    local air monitoring stations network and are intended to support long-term trends analysis,
 3    model evaluation, health and ecosystem studies, as well as NAAQS compliance. The complete
 4    NCore network consists of approximately 60 urban and 20 rural stations, including some existing
 5    SLAMS sites that have been modified for additional measurements. Each state will contain at
 6    least one NCore station, and 46 of the  states plus Washington, DC, will have at least one urban
 7    station.
 8          Either Pb-TSP or Pb-PMio monitoring may be performed at these sites.  Of 24 NCore
 9    sites measuring Pb concentrations,  17  are measuring Pb in TSP and 7 are measuring Pb in PMi0.
10    While non-source-oriented monitoring data can be used for purposes of NAAQS attainment
11    designations, a main objective for non-source-oriented monitoring is to gather information on
12    neighborhood-scale lead concentrations that are typical in urban areas in order to better
13    understand ambient air-related Pb exposures for populations in these areas.
14          Source-oriented monitors near sources estimated to emit 1.0 tpy Pb were required to be
15    operational by January 1, 2010, and the remainder of the newly required source-oriented
16    monitors were required to be operational by December 27, 2011 (75 FR 81126). Currently,
17    approximately 260 Pb-TSP monitors are in operation; these are a mixture of source- and non-
18    source-oriented monitors. Figure 2-3  shows the geographic distribution of these monitors (in
19    addition to the airport study monitors described below) in the current14 Pb NAAQS surveillance
20    network, with the Pb-TSP monitors existing at the time of the 2008 rulemaking indicated
21    separately from the newly sited Pb-TSP monitors.
            14 This figure reflects Pb-TSP monitors in AQS as of September 2012.

                January 2013                    2-14           Draft - Do Not Quote or Cite

-------
                 Guam
Alaska

-O


 Hawaii
o
                                           Start Date
                                             •    2008 and earlier
                                             •    2009 and later
                                                                               Puerto Rico
 2   Figure 2-3.  Map of Pb-TSP Monitoring Sites in Current Pb NAAQS Monitoring
 3                Network.15
 4          The current regulations also required one year of Pb-TSP monitoring (using FRM or
 5   FEM methods) near 15 specific airports in order to gather additional information on the
 6   likelihood of NAAQS exceedances near airports due to the combustion of leaded aviation
 7   gasoline (75 FR 81126).  These airports were selected based on three criteria: annual Pb
 8   inventory between 0.5 ton/year and 1.0 ton/year, ambient air within 150 meters of the location of
 9   maximum emissions (e.g., the end of the runway or run-up location), and airport configuration
10   and meteorological scenario that leads to a greater frequency of operations from one runway.
11   These characteristics were selected because they are expected, collectively, to identify airports
12   with the highest potential to have ambient Pb concentrations approaching or exceeding the Pb
13   NAAQS. Data from this monitoring study will be used to assess the need for additional Pb
            15 Estimates for source-oriented monitors are based on Pb emissions estimates in the 2008 National
     Emissions Inventory.
                January 2013
                      2-15
                            Draft - Do Not Quote or Cite

-------
 1    monitoring at airports.  These 15 sites (Figure 2-4) were intended to be operational no later than
 2    December 27, 2011, although delays in monitor siting extended the installation of some of these
 3    monitors into late 2012.
 5    Figure 2-4.   Map of sites near airports for which one year of monitoring is required.16
 6
 7         2.2.1.2  Other Lead Monitoring Networks
 8           The NATTS network, designed to monitor concentrations of hazardous air pollutants
 9    including lead compounds, included 25 sites measuring Pb-PMi0 as of June 2012. Most of these
10    sites (20) are in urban areas (Figure 2-5).  In addition to the NATTS network, states collect Pb-
11    PMio at an additional 28 sites (most as part of the Urban Air Toxics Monitoring program). All
12    collect particulate matter as PMio for toxic metals analysis using either a high-volume PMio
13    sampler or a low volume PMio sampler, typically on a 1 in 6 day sampling schedule. Most of
14    these monitoring locations are not measuring using FRM/FEM methods at this time. Lead in the
              The 15 Airports are: Merrill Field (Anchorage, AK), Pryor Field Regional (Limestone, AL), Palo Alto
      Airport of Santa Clara County and Reid-Hillview (both in Santa Clara, CA), McClellan-Palomar and Gillespie Field
      (both in San Diego, CA), San Carlos (San Mateo, CA), Nantucket Memorial (Nantucket, MA), Oakland County
      International (Oakland, MI), Republic and Brookhaven (both in Suffolk, NY), Stinson Municipal (Bexar, TX),
      Northwest Regional (Denton, TX), Harvey Field (Snohomish, WA), and Auburn Municipal (King, WA).
                January 2013
2-16
Draft - Do Not Quote or Cite

-------
 1   collected sample is generally quantified via an ICP/MS method. The standard operating
 2   procedure for metals by ICP/MS is available at: http://www.epa.gov/ttn/amtic/airtox.html.
                 Guam
Alaska

,0


 Hawaii
NETWORK
  c   NATTS
  A   NCORE
  n   Other
                                                                              Puerto Rico
 4   Figure 2-5.  Pb-PMio monitoring sites.
 5          Two networks measure Pb in PM2.5, the EPA CSN and the IMPROVE network. The
 6   CSN consists of 53 long-term trends sites (commonly referred to as the Speciation Trends
 7   Network or STN sites) and approximately 150 supplemental sites, all operated by state and local
 8   monitoring agencies. Most STN sites operate on a 1 in 3 day sampling schedule, while most
 9   supplemental sites operate on a 1 in 6 day sampling schedule.  All sites in the CSN network
10   determine Pb concentrations in PM2.5 samples. Lead is quantified via the XRF method.17  Data
11   are accessible through AQS. The locations of the CSN are shown in Figure 2-6. Nearly all of the
12   CSN sites are in urban areas, often at the location of highest known PM2 5 concentrations.  The
13   first CSN sites began operation around 2000.
            17 The standard operating procedure for metals by XRF is available at:
     http://www.epa.gov/ttnamtil/files/ambient/pm25/spec/xrfsop.pdf.
                January 2013
   2-17
                                      Draft - Do Not Quote or Cite

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
            The IMPROVE network is administered by the National Park Service, largely with
     funding by the EPA, on behalf of federal land management agencies and state air agencies that
     use the data to track trends in rural visibility. Lead in PM2 5 is quantified via the XRF method, as
     in the CSN. Data are managed and made accessible mainly through the VIEWS website
     (http://vista.cira.colostate.edu/views/) but are also available via AQS.  Samplers are operated by
     several different federal, state, and tribal host agencies on the same 1 in 3  day schedule as the
     STN. In the IMPROVE network, PM2 5 monitors are placed in "Class I" areas (including
     National Parks and wilderness areas) and are mostly in rural locations  (Figure 2-6). The oldest
     of these sites began operation in 1988, while many others began in the mid 1990s. There are 110
     formally designated IMPROVE sites, which are located in or near national parks and other Class
     I visibility areas, virtually all of these being rural.  Approximately 80 additional sites at various
     urban and rural locations, requested and funded by various  parties, are also informally treated as
     part of the network.
           Alaska
          0  250 500    1.000 Miles   0 70 140   280 Miles  0
                                               245     490
                                                                                   CSN
                                                                                   IMPROVE
                                                                                Puerto Rico &
                                                                                Virgin Islands
                                                                                 D 25 50  1DD Miles
15   Figure 2-6.  Pb-PM2.5 monitoring sites in CSN and IMPROVE networks (2012).
16
                January 2013
                                                2-18
Draft - Do Not Quote or Cite

-------
 1         2.2.1.3  NAAQS Surveillance Monitoring Considerations
 2           In this section aspects of the methods for sampling and analysis of the Pb are reviewed,
 3    and the current NAAQS surveillance network of monitoring locations is considered. The
 4    methods for sampling and analysis are considered in light of the indicator for the Pb NAAQS,
 5    and conclusions regarding the current NAAQS and associated indicator appear in chapter 4.
 6    Consideration of the ambient air monitoring network generally informs the interpretation of
 7    current data on ambient air concentrations and helps identify whether the monitoring network is
 8    adequate to determine compliance with the NAAQS.  This section discusses considerations
 9    related to these aspects of the ambient air monitoring program for Pb.18
10         2.2.1.3.1  Sampling Considerations
11           As described in section 2.2.1.1 above, consistent with the Pb NAAQS indicator being Pb-
12    TSP, the FRM for Pb, which is required at all source-oriented sites, is measurement of Pb-TSP
13    using a high-volume TSP sampler meeting the design criteria specified in 40 CFR part 50
14    Appendix B. During the review of the Pb NAAQS completed in 2008, CASAC noted the
15    variability in high-volume TSP sample measurements associated with the effects of wind speed
16    and wind direction on collection efficiency in their comments regarding the indicator.  However,
17    at the time of the 2008 review, no alternative TSP sampler designs were identified that had an
18    adequate characterization of their collection  efficiency over a wide range of particle sizes. The
19    existing high-volume sampler was retained as the sampling  approach for the Pb-TSP FRM and
20    FEMs.
21           Since promulgation of the 2008 Pb NAAQS, the EPA has initiated an effort to review
22    alternative sampler designs in an effort to develop a new sampler to replace the current high-
23    volume Pb-TSP sampler. Efficient collection of particles much larger  than 10 |im is
24    considerably more challenging because the greater inertia and higher settling velocities of Pb
25    particles hinder their efficient intake by samplers. The sampling difficulties and the long history
26    of research to develop adequate sampling technology for large particles have been thoroughly
27    reviewed (Garland and Nicholson, 1991).  Some existing commercially available sampler inlets
28    are designed to collect particles larger than 10 jim with greater than 50% efficiency (Kenny et
29    al., 2005), and these inlets can be tested as potential replacements for TSP sampling. However,
30    no alternatives to the FRM TSP sampler have been identified that have been adequately
31    characterized.
             18 The code of federal regulations (CFR) at parts 50, 53 and 58 specifies required aspects of the ambient
      monitoring program for NAAQS pollutants. The federal reference methods (FRMs) for sample collection and
      analysis are specified in 40 CFR part 50, the procedures for approval of FRMs and federal equivalent methods
      (FEMs) are specified in 40 CFR part 53 and the rules specifying requirements for the planning and operations of the
      ambient monitoring network are specified in 40 CFR part 58.

                January 2013                   2-19            Draft - Do Not Quote or Cite

-------
 1          The EPA has initiated efforts to characterize the collection efficiency of alternative
 2    sampler designs through wind tunnel testing as a necessary step towards the development of a
 3    new sampler capable of sampling particles larger than PMi0 without the noted wind speed and
 4    wind direction biases.  In addition to the difficulties in collection of particles much larger than  10
 5    jim, physical limitations in the ability to generate and transport ultra-coarse particles limit the
 6    ability to adequately characterize the new sampler. These limitations ultimately limit the upper
 7    cut-point of potential samplers to the range of 18-20 micrometers.  Following characterization in
 8    a wind tunnel, field testing of promising candidates would be required to evaluate performance
 9    and to make comparisons to the existing Pb-TSP samplers. This effort is expected to take
10    several years to complete, and, as such, it is unlikely that this new sampler will be available in
11    time for consideration  during this NAAQS review due to the activities which will need to be
12    completed to adequately characterize its performance both in the laboratory and the field. We
13    expect the new sampler to be available for consideration in a future review and consequently do
14    not expect to consider  new alternatives for sampling methods for Pb-TSP as part  of this review.
15          In addition to the FRM for Pb-TSP, there is also, as noted in section 2.2.1.1 above, a
16    FRM for Pb-PMio, based on the PMio sampler defined in 40 CFR part 50 Appendix J coupled
17    with XRF analysis.  The Pb-PMio measurements may be used as an alternative to Pb-TSP
18    measurements in certain conditions defined in 40 CFR part 58 Appendix D paragraph 2.10.1.2.
19    These conditions include where Pb concentrations are not expected to equal or exceed 0.10
20    micrograms per cubic  meter on an arithmetic 3-month mean and where the source of Pb
21    emissions is expected  to emit a substantial majority of its Pb in the PMio size fraction.  At this
22    time, we believe the low-volume FRM sampler for Pb-PMi0 to be  adequate for this application.
23    Hence, we do not expect to consider new sampling methods for Pb-PMio as part of this review.
24         2.2.1.3.2  Analysis Considerations
25          Due to reduced availability of laboratories capable of performing flame atomic
26    adsorption analyses and general advances in analysis methods, the EPA has initiated an effort to
27    expand FRM analysis  methods beyond atomic adsorption to include the more modern analysis
28    method, ICP-MS. A consultation with the CASAC Ambient Air Monitoring and Methods
29    Subcommittee was held on September 15, 2010 (Russell and Samet, 2010), and the EPA plans to
30    propose a new FRM for Pb-TSP based on this more modern analysis method in 2013. In
31    addition, the EPA has  approved several new FEMs (for ICP-MS and other analysis methods)
32    since the last Pb NAAQS review was completed in 2008.
33          With regard to  Pb-PMi0 samples, in addition to the FRM analysis method (XRF), two
34    FEMs have been accepted for Pb-PMio analysis since the 2008 Pb NAAQS rulemaking.  These
35    methods are based on  ICP-MS and are consistent with analysis methods used for the NATTS


                January 2013                   2-20            Draft - Do Not Quote or Cite

-------
 1    network.  The EPA will continue to consider new FEMs for analysis of Pb-PMio and Pb-TSP as
 2    applications are received, although no new FRMs (beyond the Pb-TSP FRM for ICP-MS
 3    discussed above) are expected during this NAAQS review.
 4         2.2.1.3.3 Network Design Considerations
 5          Significant revisions to the Pb network design requirements (40 CFR part 58, Appendix
 6    D) were made as part of the 2008 Pb NAAQS review and an associated revision to the
 7    requirements in 2010. As summarized in section 2.1.1.1 above, the current Pb monitoring
 8    network design requirements (40 CFR part 58, Appendix D, paragraph 4.5) include two types of
 9    monitoring sites - source-oriented monitoring sites and non-source-oriented monitoring sites - as
10    well as the collection of a year of Pb-TSP measurements at 15 specific airports. This section
11    describes the design considerations for the Pb NAAQS surveillance network.
12          Source-Oriented Monitoring. Since the phase out of Pb in on-road gasoline, Pb is
13    widely recognized as a source-oriented air pollutant.  As summarized in the ISA, variability in
14    air Pb concentrations is highest in areas including a Pb source, "with high concentrations
15    downwind of the sources and low concentration at areas far from sources" (ISA,  p. 3-88).
16    Recent data summarized in section 2.2.2.2 below indicates that the highest ambient Pb
17    concentrations are found near large Pb sources, usually metals industries  (see, for example,
18    Figures 2-11 through 2-13 below).  Analysis of the monitoring network during the last Pb
19    NAAQS review with regard to adequacy of monitoring near such sources found that monitors
20    were lacking near many of the larger Pb emissions sources, leading the EPA to conclude that the
21    monitoring network existing at that time was inadequate to determine compliance with the
22    revised Pb NAAQS (73  FR 29262). Findings and conclusions of that analysis led to revisions of
23    network design requirements for source-oriented monitoring, begun with the 2008 Pb NAAQS
24    rulemaking (73 FR 66964). Additional revisions were completed as part  of a reconsideration of
25    the monitoring requirements in December 2010 (75 FR 81126).
26          The current requirements for source-oriented  monitoring include placement of monitor
27    sites near sources of air Pb emissions  which are expected to or have been shown  to contribute to
28    ambient air Pb concentrations in excess of the NAAQS.  At a minimum, there must be one
29    source-oriented site located to measure the maximum Pb concentration in ambient air resulting
30    from each non-airport Pb source which emits 0.50 or more tons of Pb per year and from each
31    airport which emits 1.0 or more  tons of Pb per year.19 The expansion of the network, including
32    these source-oriented sites, is shown in Figure 2-3. Comparison of Figure 2-3 (monitors) with
            19 The Regional Administrator may waive the requirement in paragraph 4.5(a) for monitoring near Pb
      sources if the State or, where appropriate, local agency can demonstrate the Pb source will not contribute to a
      maximum Pb concentration in ambient air in excess of 50 percent of the NAAQS (based on historical monitoring
      data, modeling, or other means).

               January 2013                   2-21            Draft - Do Not Quote or Cite

-------
 1    Figure 2-2 (sources) illustrates the coverage which the current monitoring network now provides
 2    of large Pb emissions sources.
 3          The emissions threshold for source-oriented monitoring sites, 0.50 tpy, was developed
 4    based on an analysis intended to estimate the lowest emission rate that under reasonable worst-
 5    case conditions (e.g., meteorological and emission release conditions that lead to poor dispersion
 6    and associated elevations in Pb concentrations) could lead to Pb concentrations exceeding the Pb
 7    NAAQS (Cavender, 2008). This analysis included three approaches. The first two of the three
 8    approaches included a simple scaling of the historic 5 tpy emission threshold applied to the old
 9    1.5 |ig/m3 Pb NAAQS, and a simplified modeling effort using a screening model.  The third
10    approach relied on design values based on Pb monitoring data surrounding large sources (1 tpy
11    or greater) of Pb. At the time of the 2008 review, complete 3-year design values were only
12    available for 7 source-monitor pairs
13          As more recent data become available, analysis of the updated and expanded dataset will
14    inform evaluation of the appropriateness of the current threshold.  Since the analysis performed
15    during the 2008  review, over 150 additional source-oriented monitors have been installed that
16    may be useful in such an evaluation.  At this time, the new monitors have not collected a
17    complete 3-year dataset, which is needed for the development of complete design values for the
18    new source-oriented monitors. For monitors installed in response to the 2008 revisions (near
19    sources estimated to emit 1.0 tpy or more of Pb), 3-years of certified data will be available in
20    spring 2013.  Three-years of certified data for the monitors required in the 2010 reconsideration
21    will be available in spring 2015.  Accordingly, analysis of the emissions threshold for source-
22    oriented monitors may be an appropriate issue for consideration in future reviews.
23          One-year Airport Monitoring. In addition to the above source-oriented monitoring
24    requirement, one year of monitoring was required near 15 specific airports in order to gather
25    additional information on the likelihood of NAAQS exceedances near airports due to the
26    combustion of leaded aviation gasoline (75 FR 81126). This airport monitoring study may
27    provide useful information to assess the need for additional Pb monitoring at airports, and
28    appropriate requirements, such as consideration of the 1.0 tpy emissions threshold and regional
29    administrator authority for siting of source-oriented monitors at airports (75 FR 81131-81132).
30    This airport monitoring study is currently underway.  Given delays in monitor siting, monitor
31    installation at some of these sites extended into late 2012. Accordingly, the timing for
32    completion of the year of monitoring will vary across the 15 locations, with all of them projected
33    to finish by late 2013. The completed datasets, when available and certified, will inform EPA's
34    considerations on next steps with regard to monitoring at airports using leaded aviation gasoline.
35    Also, as described further in section 2.1.2.2above, the airport monitoring data will also inform
36    EPA's ongoing investigation  into the potential for lead emissions from piston-engine aircraft to

                January 2013                   2-22            Draft - Do Not Quote or Cite

-------
 1    cause or contribute to air pollution that may reasonably be anticipated to endanger public health
 2    or welfare.  This investigation is occurring under Section 231 of the Clean Air Act (CAA),
 3    separate from the lead NAAQS review.
 4          Non-Source-Oriented Monitoring.  Monitoring agencies are also required to conduct
 5    non-source-oriented monitoring at NCore sites with a population of 500,000 or more, as noted
 6    above.20 Currently, there are 24 NCore sites measuring Pb concentrations, with 17 measuring Pb
 7    in TSP and  7 measuring Pb in PMi0. While non-source-oriented monitoring data can be used for
 8    designation purposes, an alternative objective stated for these sites is the collection of data on
 9    neighborhood-scale Pb concentrations that are typical in urban areas to inform our understanding
10    of ambient air-related Pb exposures for the general population. At this still early stage  in data
11    collection at these sites, we have not identified specific gaps or inadequacies. Thus, at  this stage,
12    the current requirement for non-source-oriented Pb monitoring appears to be adequate for these
13    purposes.
14
             1 Defined by the US Census Bureau - http://www.census.gov/population/www/metroareas/metroarea.html

                January 2013                    2-23            Draft - Do Not Quote or Cite

-------
1
2
3
4
5
6
7
 9
10
12
13
           2.2.2  Ambient Concentrations
           2.2.2.1   Temporal Trends
             Ambient air concentrations of Pb in the U.S. have declined substantially over the past 30
      years. Figure 2-7 illustrates this decline in terms of site-specific maximum 3-month average
      concentrations at the set of 31 monitoring sites that have been operating across this period. The
      median of this dataset has declined by more than 90% over the 30-year period, and the average
      by 89%.  Over the past 10 years, a larger dataset of 92 sites operating across that period also
      indicates a decline, which is on the order of 60% for the average of that dataset (Figure 2-8).21
                     5.0
                     4.5
                     0.0
                       1111111111111111111122222222222
                       9999999999999999999900000000000
                       8888888888999999999900000000001
                       012345678901 234567S901 234567890
                              Note: Based on annual maximum 3-month average at 31 sites.
11    Figure 2-7.   Temporal trend in Pb -TSP concentrations: 1980-2010 (31 sites).
                              .Nalioral .Standard..
                                   i      i      i     i      i      i      i     i
                      2000   2001  2002  2003   2004   2005  2006  2007   2008  2009  2010
                              Note: Based on annual maximum 3-month average at 92 sites.
14    Figure 2-8.   Temporal trend in Pb-TSP concentrations: 2000-2010 (92 sites).
              http://www.epa.gov/air/airtrends/lead.html
                January 2013
                                                2-24
Draft - Do Not Quote or Cite

-------
1
2
3
4
5
              The role of the phase-out of leaded gasoline on declining concentrations is evident from
      temporal trends in air Pb concentrations near roadways,  as illustrated in Figure 2-9 which
      presents data for five monitors sited near roadways.22 These sites additionally indicate the
      concentrations currently common at such sites, with the  maximum 3-month average
      concentrations at all five sites falling below 0.03  |ig/m3 in the most recent years.
 6
 7
 9
10
                                                                                                -060374002
                                                                                                 060651003
                                                                                                 170310052
                                                                                                 170316003
                                                                                                 250250002
                                                9 •€ '5
                                                         9 •€ '5
                                                                 9 •€ '5
                                                                          9 •€ '5
                                                                                  9 -C '5
                                                                                           9 -C '5
     Monitor locations: Los Angeles, CA (06-037-4002), Riverside, CA (06-065-1003), Cook, IL (17-031-0052,17-031-6003), Suffolk, MA (25-025-0002).

     Figure 2-9.   Airborne Pb -TSP concentrations (3-month average) at five sites near
                   roadways:  1979-2010.
              22 In selecting these sites, the objective was to identify sites near roadways that do not appear to be near
      other (stationary) sources of Pb emissions.  In addition to consideration of information in national emissions
      inventories, the areas around the sites were examined using satellite pictures (in Google Maps) for signs of current
      or historical industrial activity.
                 January 2013
                                                    2-25
Draft - Do Not Quote or Cite

-------
 1         2.2.2.2  Current Concentrations
 2          As a result of revisions to the Pb NAAQS surveillance monitoring requirements
 3    (described in section 2.2.1 above), Pb monitoring sites have been in transition over the last few
 4    years as indicated by Figure 2-3. For presentation in this document, we have focused on the
 5    most recent period for which adequately complete data are available, 2009-2011, recognizing
 6    that the dataset developed for this period includes many but not all of the monitors newly
 7    required by the December 2010 regulations.
 8          Lead concentrations, in terms of maximum 3-month average Pb-TSP concentration, at
 9    monitoring sites active across the U.S. during the period 2009-2011, and for which sufficient
10    data are available to meet completeness criteria described in Appendix 2C, are presented in
11    Figure 2-10.23 Highest concentrations occur in the vicinity of large metals industries, as
12    discussed in section 2.1.2.1,24 As can be seen through comparison of Figure 2-10 with Figure 2-
13    3, only limited data (not meeting completeness criteria for the data analysis) are available to date
14    for some of the monitors newly required by the Pb NAAQS monitoring network revisions
15    completed in 2010 (see section 2.2.1.1).
            23 Criteria for development of the 2009-2011 air Pb-TSP, Pb-PM10 and Pb-PM2 5 datasets discussed in this
      section are described in Appendix 2C. Data summaries are included in Appendix 2D.
            24 Information regarding areas of U.S. designated nonattainment with the Pb NAAQS is available at:
      http://www.epa.gov/air/oaqps/greenbk/mindex.html.

                January 2013                    2-26           Draft - Do Not Quote or Cite

-------
1
2
                                                                    2009-2011 Pb-TSP
                                                                     •   maximum 3-month mean < 0.05 ug/m3
                                                                         maximum 3-month mean > 0.05 and < 0.15 ug/m^
                                                                     O   maximum 3-month mean > 0.15 and < 0.5 ug/m3
                                                                     •   maximum 3-month mean > 0.5 ug/m3
Figure 2-10. Pb-TSP maximum 3-month means (215 sites), 2009-2011.
                                January 2013
                                                            2-27
Draft - Do Not Quote or Cite

-------
 1          Figures 2-11 through 2-13 illustrate differences in Pb-TSP concentrations, in terms of
 2   three different metrics (maximum 3-month mean, annual mean, and maximum monthly mean),
 3   among sites near and more distant from emissions sources. The Pb-TSP sites indicate the much
 4   greater site and temporal variability in concentrations at source-oriented and previous source-
 5   oriented sites as compared to non-source-oriented sites.  Across the Pb-TSP sites, as would be
 6   expected, highest concentrations are observed at the source-oriented sites, followed by the
 7   previous source-oriented sites. This is the case for all three metrics analyzed (maximum 3-
 8   month mean, maximum monthly mean and annual mean).
 9          Figures 2-11 through 2-13 additional indicate distributions of Pb-PMio and Pb-PM2.5
10   concentrations in urban and rural locations where they are monitored. The Pb-PMio and Pb-
11   PM2 5 networks are described in section 2.2.1.2 and shown in Figures 2-5 and 2-6, respectively. It
12   is important to note that there are few sites in these recent datasets with colocated monitors for
13   the different size fractions.  As described in the ISA,  at 18 urban sites (without specification as to
14   proximity to Pb sources) with at least 30 co-located Pb-TSP and Pb-PMio samples collected
15   during various time periods (nearly all between 1990 and 2000), approximately 80% of the Pb
16   mass, on average, is captured by  the Pb-PMio measurements (ISA, sections 3.5.3.1 and 3.8.6).
17   The data distributions in Figures  2-11 through 2-13 indicate reduced variability in concentration
18   for non-source-oriented sites and for particulate Pb of smaller size fractions.
                January 2013                   2-28            Draft - Do Not Quote or Cite

-------
             co

              E
              C
              o

              "ro




              I
              o
              O
1.2-
1.1-
1.0-
0.9-
0.8-
0.7-
0.6-
0.5-
0.4-
0.3-
0.2-
0.1-
0.0-


















>







c

•























*



	 l_
All Pb-TSP Pb-TSP
sites source-
(n=215) oriented
sites
(n=112)








^ =
I I
Pb-TSP Pb-TSP
previous not source-
source- oriented sites
oriented sites (n=91)
(n=12)









I
Pb-PM10
urban
sites
(n=35)








=
I i
Pb-PM25 Pb-PM25
urban non-urban
CSN sites IMPROVE
(n=198) sites
(n=106)
l



2   Figure 2-11. Distribution of maximum 3-month mean concentrations of Pb-TSP, Pb-PMio and Pb-PM2.s at different site types,

3               2009-2011.
                              January 2013
2-29
Draft - Do Not Quote or Cite

-------
1
2
0.7-
0.6-
rT 0.5-
E
~D)
3.
^ 0.4-
c
o
Concentrat
pop
-». Ko w
i i i
0.0-








:


e

• •*







>)e


1




^





a n~

i
AIIPb-TSP Pb-TSP Pb-TSP Pb-TSP
sites source- previous not source-
(n=215) oriented source- oriented sites
sites oriented sites (n=91)
(n=112) (n=12)




i
Pb-PM10
urban
sites
(n=35)



~^t^
i i
Pb-PM25 Pb-PM25
urban non-urban
CSN sites IMPROVE
(n=198) sites
(n=106)
Figure 2-12. Distribution of annual mean concentrations of Pb-TSP, Pb-PMi0 and Pb-PM2.5 at different site types, 2009-2011.
                              January 2013
                                                       2-30
Draft - Do Not Quote or Cite

-------
                  co

                  E
                  C
                  o

                  I



                  I
                  o
                  O
1.8-
1.7-
1.6-
1.5-
1.4-
1.3-
1.2-
1.1-
1.0-
0.9-
0.8-
0.7-
0.6-
0.5-
0.4-
0.3-
0.2-
0.1-
0.0-





























e


































*


I












i












£— I — i
^ C2












^ — I

All Pb-TSP Pb-TSP Pb-TSP Pb-TSP
sites source- previous not source-
(n=215) oriented source- oriented sites
sites oriented sites (n=91)
(n=112) (n=12)











I
Pb-PM10
urban
sites
(n=35)











i i
Pb-PM25 Pb-PM25
urban non-urban
CSN sites IMPROVE
(n=198) sites
(n=106)
2   Figure 2-13. Distribution of maximum monthly mean concentrations of Pb-TSP, Pb-PMi0 and Pb-PM2.s at different site types,

3               2009-2011.
                              January 2013
2-31
Draft - Do Not Quote or Cite

-------
 1    2.3   AMBIENT AIR LEAD IN OTHER MEDIA
 2          Lead emitted into the ambient air, depending on its chemical and physical characteristics,
 3    deposits out of air onto surfaces in the environment. The environmental fate of atmospherically
 4    emitted Pb once deposited, is influenced by the type of surface onto which the particles deposit
 5    and by the type and activity level of transport forces in that location.  Deposited Pb can be
 6    incorporated into soil matrices in terrestrial ecosystems or transported (by direct deposition or
 7    stormwater runoff) to water bodies and into aquatic sediments, which may serve as aquatic
 8    ecosystem sinks. Particles deposited onto impervious surfaces, such as roadway or urban
 9    surfaces, are more available for human contact while they remain on such surfaces. For
10    example, they may be available for adherence onto skin or they may be resuspended into the air
11    and inhaled. Precipitation and other natural, as well as human-influenced, processes contribute
12    to the fate of such particles and their potential transport to areas or environments of lesser or
13    greater likelihood for human contact, e.g., transport with surface runoff into nearby water bodies
14    or tracking into nearby houses (ISA, section 3.3).
15          Lead from current and historical air emissions sources, in addition to a range of nonair
16    sources (e.g., land disposal of wastes as well as surface runoff and releases to surface waters),
17    contribute to Pb in outdoor dust, soil and  aquatic systems. Because of its persistence, Pb from all
18    of these  sources can contribute to media concentrations into the future. The pattern of resulting
19    Pb concentrations changes over time in response to changes in new Pb contributions to the
20    systems  (e.g., in types and rates of addition), as well as environmental processes (chemical,
21    biological and physical) particular to each type of media and ecosystem.  Accordingly, media
22    and ecosystems differ with regard to Pb concentrations and changes in those concentrations  in
23    response to airborne Pb.  These differences relate to ecosystem processes as well as Pb source or
24    emissions characteristics and proximity.
25          The initial section below (section  2.3.1) summarizes salient information regarding
26    atmospheric deposition and how it is monitored.  Subsequent sections describe the current
27    information related to the presence of ambient air Pb in nonair media and what it indicates
28    regarding impacts of current patterns of Pb in and being emitted to ambient air, including
29    relationships to impacts in the past and to other Pb sources.  Section 2.3.2 focuses on terrestrial
30    media, while section 2.3.3 discusses aquatic media.

31         2.3.1   Atmospheric Deposition
32          Deposition is the pathway by which Pb particles are removed from the atmosphere and
33    transferred to other environmental media. There are several approaches by which atmospheric
34    deposition,  or the transfer of Pb from the  atmosphere to soil or water bodies, can be assessed.
35    These include measurements of Pb in rainfall (wet deposition) and on collection surfaces during

                January 2013                    2-32            Draft - Do Not Quote or Cite

-------
 1    dry periods (dry deposition); dry deposition is also estimated from measurements of airborne Pb
 2    particles coupled with estimates of deposition velocity (see 2006 CD, Section 2.3.2).   Less direct
 3    approaches include monitoring changes in Pb concentration in various biological or physical
 4    media such as lichen or mosses, snowpack, soil and aquatic sediments (ISA, sections 3.6.2, 3.6.4
 5    and 3.6.6).  Approaches for the latter three media may involve repeated measurements of surface
 6    samples over time or core samples coupled with isotope dating.  To gain information on
 7    atmospheric deposition, unaffected by contributions from direct nonair environmental releases,
 8    such  studies are generally conducted in somewhat remote areas. As there are currently no
 9    nationwide Pb atmospheric deposition monitoring programs, the discussion of atmospheric
10    deposition in this and subsequent sections is drawn from this range of approaches (2006 CD,
11    sections 2.3, 8.2.2 and AX7.1.2.3; ISA, sections 3.3.1.2 and 3.6  ).  Geographic differences in
12    deposition generally relate to differences in the  amount and size distribution of airborne Pb
13    particles in that location, as well as meteorology and other site-specific factors. For example, the
14    size of particles, as well as solubility in rainwater, can also influence wet deposition rates (ISA,
15    section 3.3.1.2; 2006 CD, p. 2-59). Factors that particularly influence dry deposition are the
16    level of atmospheric turbulence, especially in the layer nearest the ground, particle size
17    distributions and density, and the nature of the surface itself, such as smooth or rough (2006 CD,
18    pp. 2-55 to 2-57).
19          Evidence for the temporal pattern of U.S. lead deposition generally indicates a peak
20    during the middle twentieth century (e.g., 1940s through 1970s) followed by substantial declines
21    in the latter part of the century (ISA, section 3.3.1; 2006 CD section 2.3.1; Jackson et al., 2004).
22    This  pattern reflects the increased use of Pb in the industrial age (including its use in  gasoline in
23    the twentieth century) and declines in response to environmental controls on gasoline and metals
24    industries in the latter part of the twentieth century.  Studies continue to document declines in
25    atmospheric deposition  since the  1980s in areas remote from industrial areas (e.g., Watmough
26    and Dillon, 2007).  As would be expected, stationary sources of Pb continue to contribute to
27    relatively higher deposition rates in industrial areas as compared to other areas, although rates in
28    those areas appear to have also declined since the 1970s-80s (ISA, section 3.3.1.2;  Sabin and
29    Schiff, 2008).

30         2.3.2  Terrestrial Media
31          Lead in the terrestrial media discussed here may originate from current or historical air or
32    other sources. Other sources to these media include historical use of indoor and outdoor leaded
33    paint, more prevalent in areas with older buildings, and the processing of lead-containing
34    materials.  Further, the concentrations of Pb in each media, as well as the relative contributions
35    from various types of sources to those concentrations, vary among media and among locations


                January 2013                    2-33            Draft - Do Not Quote or Cite

-------
 1    for any one media. Differences among locations in Oair-related Pb are generally related to the
 2    current and historical magnitude of Pb emissions and deposition at that location, as well as
 3    location- and media-specific removal factors.

 4         2.3.2.1  Indoor Household Dust
 5           Household dust arises from paniculate matter generated indoors and outdoors and Pb in
 6    household dust can reflect a combination of these sources. Airborne particles can be transported
 7    indoors with ambient air, while particles deposited outdoors from ambient air may be carried in
 8    on humans and their clothing or other items transported indoors. Depending on factors such as
 9    the proximity of the residence to current or historical metals industries, leaded paint usage, and
10    age of the  residence, Pb in indoor residential dust can reflect current or historical atmospheric Pb
11    or nonair-related sources, such as leaded paint. Another indoor source is tobacco smoking (ISA,
12    section 4.1.3.2; Gaitens et al., 2009; 2006 CD, p. 3-15; Mannino et al., 2003). The age of the
13    residence may be an indicator of the potential for the presence of leaded paint and, in areas of
14    significant levels of airborne Pb in the past, such as near mining or smelting industries, may also
15    indicate the potential  for the residual presence of historically emitted Pb (ISA, sections 4.1.1 and
16    4.1.3.2; 2006 CD section 3.2.3). A study of households in the Baltimore metropolitan area (in
17    which there are no areas designated non-attainment for the current Pb NAAQS)25 indicated
18    indoor air Pb concentrations to be significantly associated with outdoor air Pb concentrations but
19    did not show a statistically significant relationship of outdoor air Pb with indoor dust Pb, perhaps
20    indicating  the greater role of sources other than current airborne concentrations in those
21    households (ISA, 4.1.3.2; Egeghy et al., 2005).
22           As discussed in chapter 4 below, indoor dust is a major pathway for air-related Pb
23    exposure for pre-school children, largely related to prevalent hand-to-mouth behavior at that age.

24         2.3.2.2  Outdoor Dust in Areas of Human Activity
25           Lead in particulate matter occurring on outdoor surfaces (outdoor dust)  may reflect
26    current or historical Pb emissions,  as well as the historical uses of lead in products on buildings
27    and infrastructure in surrounding area. The concentrations of air-related Pb in outdoor dust and
28    the relative contribution of air-related Pb to the total Pb concentrations in outdoor dust varies
29    depending on the location-specific characteristics of air-related (and other) Pb sources.  The role
30    of air-related sources in outdoor dust Pb has been documented in areas near industrial sources,
31    such as smelters, where Pb concentrations in outdoor dust have been documented to decline in
32    response to reduced emissions. For example, outdoor dust Pb concentrations in a long-time
33    Canadian smelter town were found to track smelter Pb emissions, as a new smelting technology
              1 http ://www. epa. gov/air/oaqps/greenbk/mnc .html

                January 2013                   2-34            Draft - Do Not Quote or Cite

-------
 1    which reduced airborne Pb concentrations by 75% resulted in a 50% reduction in outdoor dust
 2    Pb loading rate and concentrations (2006 CD, p. 3-23; Hilts, 2003).
 3          Rates of dry deposition of Pb in large metropolitan areas during the past decade are much
 4    lower than those reported for the 1970s (ISA, sections 3.3.1 and 3.6.1; Table 2-4, below). For
 5    example, dry deposition of Pb into Los Angeles harbor during 2002-2006 was more than an
 6    order of magnitude lower than rates reported for the same location in 1975 (ISA section 3.3.1.2;
 7    Sabin and Schiff, 2008; Lim et al., 2006).  Across the Los Angeles metropolitan area, average
 8    rates generally ranged from 10 to 32 micrograms per square meter per day (|ig/m2/day) and down
 9    to 0.3 |ig/m2/day for the non-urban site of Malibu during 2002-2003 (Lim et al., 2006). Rates
10    within this range are reported from studies in Manhattan and a non-industrial area of New Jersey,
11    with a somewhat higher rate reported for a major industrial location within the New York City
12    metropolitan area. Although rates have not been reported recently for smelter locations in the
13    U.S., the outdoor dustfall deposition rate reported in a Canadian smelter town after the
14    installation of new technology that reduced average airborne Pb-TSP concentrations to a level
15    still at least twice the current U. S. NAAQS (0.3 |ig/m3 from prior average of 1.1 |ig/m3) was
16    several orders of magnitude higher than those reported for New York, Los Angeles and Chicago
17    urban areas (Table 2-4).
18          From the few studies reporting levels of Pb loading or concentration in outdoor dust
19    within the past decade or so, loading to surfaces of pedestrian traffic signals in New York City
20    was on average approximately 15% the loading reported on sidewalks in older, residential areas
21    of downtown Baltimore (e.g., approximately 250 ug/ft2 as compared to 1500 ug/ft2), an order of
22    magnitude or more below that reported in the long-time Canadian smelter town (Caravanos et al.,
23    2006b; Farfel et al 2005; Hilts, 2003).
                January 2013                   2-35            Draft - Do Not Quote or Cite

-------
 1   Table 2-3.  Dry deposition of lead in large metropolitan areas and in a smelter town.
Description
Los Angeles Harbor, June-Nov 2006
Los Angeles Harbor, 2002-2003"
Los Angeles Harbor, 1975
Los Angeles, metropolitan area urban sites, 2002-2003 B
Los Angeles, metropolitan area, non-urban site Malibu, 2002- 2003 B
Los Angeles, I-405, near Westwood (downwind side), spring 2003 c
Los Angeles, I-405, near Westwood (upwind side), spring 2003 c
Jersey City, NJ, 2001-2002
New Brunswick, NJ, 2001- 2002
Manhattan, NYC, 2nd story rooftop (unprotected), 2003-2005
Chicago, 1993-95 (most recent available)
Trail, British Columbia, Canada, historical smelter town, 1999°
Dry Deposition
Rate
(u.g Pb/m2/day)
14A
15
300
10-32
0.3
24
7.3
50
9
22
38-71
31000
Study
Sabin and Schiff, 2008
Limetal, 2006
Sabin and Schiff, 2008
Limetal., 2006
«
Sabin etal., 2006
«
Yi etal., 2006
«
Caravanosetal., 2006a
Yi etal 2001; Paode etal., 1998
Hilts, 2003
A - Rates for rest of off-shore transact from Santa Barbara to San Diego Bay ranged from 0.52 (Oxnard) - 3.3 (San Diego Bay) |ig/m3.
B - Average Pb-TSP range= 0.0056-0.01 7 |jg/m3 for urban sites; 0.0022 at Malibu site.
C - Average Pb-TSP at downwind site over sampling period -0.02 pg/m3.
D - Average Pb-TSP = 0.3 ng/m3, this reflects new technology (previously 1 .1 ng/m3) for smelter that existed in this town for past 95 years.
 3         2.3.2.3  Soil
 4          As is the case for Pb in outdoor dust, occurring in surface soils can be derived from
 5   current or historical Pb emissions, as well as leaded paint usage in older residential areas.  The
 6   relative role of these sources may vary with the type of environment and proximity to industry
 7   and population centers.
 8          In forested areas away from old urban areas, the role of leaded gasoline in soil
 9   contamination is illustrated by the documented reductions in Pb in surface soils subsequent to the
10   leaded gasoline phasout. For example, forest surface soil (litter) concentrations in a transect
11   from Vermont, through Maine and up to Gaspe in Quebec, which in 1979 exhibited a significant
12   spatial trend, ranging from 200 milligrams per kilogram, dry weight (mg/kg dw) at its
13   southernmost point down to 60 mg/kg dw at its northernmost and most remote point in Quebec,
14   declined to 32-66 mg/kg dw (with no spatial trend) in 1996 (ISA, section 3.6.1; Evans, 2005).  In
15   forests from the mid-Atlantic to southern New England, a reduction in litter Pb concentration
16   was observed between 1978 and 2004-05 (ISA, section 3.6.1; Johnson and Richter, 2010).  In the
17   latter study, the authors observed less change in concentration in the more northern sampling
18   sites, which they attribute to reduced rates of organic matter decomposition in those colder
19   temperature areas. Similarly, an additional New England study by Kaste et al (2006) also
                January 2013
2-36
Draft - Do Not Quote or Cite

-------
 1    documented a pattern in temporal reductions in soil Pb (o-horizon) related to decomposition
 2    activity. These recent findings are generally consistent with findings reported in the last review
 3    which indicated the gradual migration of deposited Pb into mineral soils (e.g., Miller and
 4    Friedland, 1994; Kaste et al., 2003; Wang and Benoit, 1997; Johnson et al., 1995; Zhang, 2003).
 5           Few studies have investigated temporal trends of surface soil Pb concentrations in more
 6    populated areas in relation to reductions in usage of leaded gasoline and paint. Current
 7    concentrations in areas of past heavy traffic powered by leaded gasoline are generally elevated
 8    above areas more distant from these areas (ISA, section 3.6.1). Current roadway-related sources
 9    of Pb, while substantially less significant than leaded gasoline, continue to provide some
10    contribution to surface dust/soil in these areas (ISA,  section 3.2.2.6). Surface soil and dust
11    concentrations are much higher in such areas  in large, older cities than surrounding suburban
12    areas (ISA, section 3.6.1).  In older residential areas, the use of leaded house paint is another
13    contributor to surface soil concentrations (ISA, sections 3.6.1 and 3.7.7.2; Yesilonis et al., 2008;
14    Brown et al., 2008; Clark et al., 2006).
15           Areas of long-term Pb emissions from point sources appear to be the areas of highest
16    surface soil Pb concentrations (ISA, sections 3.6 and 3.6.1). For example, Pb surface soil
17    concentrations within approximately 100-250 meters of long-established lead smelters have been
18    five to ten times higher than those at 3-5 km distance (ISA, section 3.6.1; 2006 CD, Table 3-4).
19    Surface soil concentrations near U.S. mines that are no longer active have also been found to be
20    elevated above more distant areas (2006 CD,  Table 3-6).  Information described in the 2006 CD
21    for areas surrounding smelters after implementation  of pollution controls, although showing
22    declines in Pb concentrations in outdoor dustfall, street dust and indoor dustfall, has not indicated
23    a noticeable decline in soil Pb concentrations  (2006 CD, pp. 3-23 to 3-24).  The continued Pb
24    emissions in such industrial areas likely influence the dynamics of Pb concentrations in the soil,
25    affecting any response to emissions reductions. Estimates of associated steady-state surface soil
26    Pb concentrations or the expected longer-term temporal pattern for this situation have not been
27    made.
28           In summary, findings to date indicate that many of those systems less influenced by
29    current point sources have and may still be responding to reduced Pb deposition rates associated
30    with reduced atmospheric emissions of Pb, including those associated with the phase-out of
31    leaded gasoline, while potential responses of soils near point sources and those involving
32    historically deposited Pb near roadways are less well characterized.

33         2.3.2.4   Biota
34           Measurements of Pb in some biota indicate reductions in biologically available Pb over
35    the past 30 years in some remote locations. For example, measurements of Pb in lichen from


                January 2013                   2-3 7            Draft - Do Not Quote or Cite

-------
 1    Golden Lake in Mount Ranier National Park (2005) and Emerald Lake basin in Sequoia/Kings
 2    Canyon National Parks (2004) indicate significant reductions (approximately 3-5-fold) since
 3    samples were previously collected in those locations in 1984 (ISA, section 3.6.6; Landers et al.,
 4    2008).  Further, Pb in teeth of juvenile and adult moose in Isle Royale National Park in northern
 5    Michigan have also declined substantially (ISA, section 3.6.8; Vucetich et al., 2009).
 6          Biota measurements have also documented sources of Pb deposition in mining areas
 7    including the transportation of materials along haul roads. Lead deposition from these mining-
 8    related activities has been documented in remote wilderness areas in Alaska. For example, Pb
 9    concentration in moss in Cape Krusenstern National Monument declines with distance from the
10    road on which product from Red Dog mine26 in northwest Alaska is transported (ISA, section
11    3.3.1.2; Hasselbach et al., 2005). In summer 2001, the median Pb concentration measured within
12    10 meters of the haul road was more then 20 times higher than sites at greater distance from the
13    road and nearly three orders of magnitude greater than those reported for this moss in other
14    Alaska locations in 1990-1992. Concentrations in moss within 10 meters of the haul road ranged
15    from 271-912 mg/kg dw (Hasselbach et al., 2005).

16         2.3.3   Aquatic Media
17         2.3.3.1  Surface Waters
18          In addition to delivery by atmospheric deposition, Pb is also carried into surface waters
19    via wastewater effluent from municipalities and industry, stormwater runoff, erosion, and
20    accidental discharges (2006 CD, p. AX7-142; ISA section 3.3.2). As a result of the phasing out
21    of leaded gasoline, reductions in Pb concentrations have been documented in surface waters of
22    the North Atlantic Ocean, as well as the relatively  less remote areas of the Great Lakes (2006
23    CD, p. 7-23). The availability of studies investigating historical trends in surface waters is
24    limited, in part due to analytical issues that challenged many monitoring programs in the past
25    (2006 CD, AX7.2.2.2).  Thus, temporal trends reported in many aquatic systems are based on
26    sediment analyses (see section 2.3.3.2 below).
27          Most Pb occurring in aquatic systems is associated with particles, with the distribution
28    between particle-bound and dissolved form being influenced by water chemistry as well as
29    suspended sediment levels (ISA, 3.3.2;  2006 CD, pp.  AX7-117 to AX7-118, Section AX7.2.2).
30    Water columns have been described as "transient reservoirs" for pollutants (ISA, section 3.7.2;
31    2006 CD, p. 2-75). Once deposited to sediments, whether Pb is available for resuspension back
32    into the water column with potential transport further down a watershed versus being buried into
33    deeper sediments depends on the aquatic system. In open ocean waters (generally characterized
            26 Red Dog mine is a zinc and Pb mine initiated in 1987. The materials are transported from the mine to a
      port on the Chukchi Sea via a 52-mile long haul road.

                January 2013                   2-38           Draft - Do Not Quote or Cite

-------
 1    by depth and distance from continental sources), resuspension to surface waters is unlikely. In
 2    more shallow systems, and additionally those influenced by land sources (e.g., stormwater runoff
 3    as well as point sources), resuspension may play a role in water column concentrations. For
 4    example, studies in San Francisco Bay, the southern arm of which has an average depth of 2 m,
 5    have indicated that Pb particles may be remobilized from surface sediments into the water
 6    column (2006 CD, AX7-141).
 7          The distribution of Pb dissolved in many U.S. surface waters has been reported by the
 8    United States Geological Survey (USGS) National Water-Quality Assessment (NAWQA)
 9    program.  The NAWQA data set encompasses data, collected from 1991-2003 on Pb
10    concentrations in flowing surface waters for more than 50 river basins and aquifers throughout
11    the country (2006 CD, Section AX7.2.2.3).  These data indicate a mean dissolved Pb
12    concentration in U.S. surface waters of 0.66  ug/L (range 0.04 to 30 ug/L) in waters affected by a
13    combined contribution of natural and  anthropogenic sources, as compared to a mean of 0.52
14    ug/L (range 0.04 to 8.4 ug/L) for waters in "forest", "rangeland", and "reference" sites (2006
15    CD, Section AX7.2.2.3).  The highest surface water Pb concentrations were observed in sites
16    impacted by land uses such as mining (2006 CD, p. AX7-131). The role of surface runoff in
17    delivering contamination to waters near metals industries presents a challenge to the
18    disentangling of atmospheric deposition contributions from those associated with surface runoff.

19         2.3.3.2  Sediments
20          Many studies have investigated temporal trends in sediment Pb  concentrations, using
21    sediment cores or surface sediment Pb concentration, with declines documented in many systems
22    and usually attributed to the phasing out of leaded gasoline (ISA,  section 3.6.2; 2006 CD, section
23    AX7.2.2).  Several studies documenting the  increased Pb deposition of the industrial age,
24    including specifically leaded gasoline usage, and the subsequent declines associated with leaded
25    gasoline phase-out were reported in the 2006 CD. They include investigations involving
26    sediment cores from the Okefenokee  Swamp in Georgia  as well as from 35 reservoirs and lakes
27    in urban and reference locations (ISA, section  3.6.2; 2006 CD, p.  AX7-141).  In the latter, the
28    median reduction in Pb mass accumulation rate in the cores, adjusted for background
29    concentrations, was 246%, with the largest decreases in lakes located in dense urban watersheds
30    (ISA, section 3.6.2; 2006 CD, p. AX7-141; Mahler et al, 2006). A third study of sediment cores
31    in 12 lakes in the Great Lakes area also documented a peak in  Pb  concentrations consistent  with
32    peak use of leaded gasoline in the U.S. in the mid 1970's and declining concentrations in most
33    lake sediments through the mid 1990's (2006 CD, p. 2-55; Yohn et al.,  2004).  Sediment surveys
34    by the USGS NAWQA have reported the highest Pb concentrations in Idaho,  Utah, and
                January 2013                   2-39            Draft - Do Not Quote or Cite

-------
 1    Colorado, with seven of the highest concentrations at sites classified as mining land use (2006
 2    CD, p. AX7-133).
 3          Among the more recent investigation of temporal trends in aquatic sediments described in
 4    the ISA is that associated with the Western Airborne Contaminants Assessment Project which
 5    documented Pb in sediment cores from 14 lakes in eight U.S. national parks in western states,
 6    including three in Alaska (ISA, section 3.6.2; Landers et al., 2010). Among the Alaska cores, in
 7    which concentrations were generally on the order of 20 itg/kg, dw, there was little variation in Pb
 8    concentration, flux or enrichment factor. The other park cores, with few exceptions, generally
 9    exhibited an increase in concentration commencing in the mid-nineteenth century, which
10    transit!oned to declining trends in the past few decades, with lower concentrations in more
11    recent, surface material. The highest concentration was recorded at core depth corresponding to
12    the mid-1970s, in one of the Sequoia/Kings Canyon lake cores (Figure 2-16; ISA, section 3.6.2;
13    Landers etal.,201027).
14
            27 Figure created from database for Landers et al (1980);
      http://www.nature.nps.gov/air/Studies/air_toxics/wacap.cfm.

                January 2013                   2-40            Draft - Do Not Quote or Cite

-------
          125 -•
          100 -•
        •S. 75 -•
          25
       I            I
 Sequoia/Kings Canyon-Pear (mid California)
 Sequoia/Kings Canyon-Emerald (mid California)
-Mt Ranier-Golden (mid Washington)
-Mt Ranier-LP19 (mid Washington)
 Glacier-Snyder (Montana)
 Glacier-Oldman (Montana]
 Rocky Mnt-Lone Pine (Colorado)
 Rocky Mtn-Mills (Colorado)
-Olympic (Washington)
-Denali (mid Alaska)
-Gates of the Arctic (n. Alaska)
•Noatak (n. Alaska)
            1800
                        1830
                                   1860
                                               1890         1920
                                               Year Material Deposited
                                                                       1950
                                                                                   1980
                                                                                               2010
 2    Figure 2-14.  Temporal trend in sediment concentration from core samples in 12 lakes at
 3                  eight National Parks or Preserves.
 4           Analyses of cores taken in several lakes and reservoirs along the Apalachicola,
 5    Chattahoochee, and Flint River Basin from north of the Atlanta, GA metropolitan area to the
 6    Gulf of Mexico indicate changes in sediment concentration both with time and in relation to
 7    influence of the large metropolitan area of Atlanta (ISA,  section 3.6.2; Callender and Rice,
 8    2000). Highest concentrations were documented just downstream from Atlanta at the core depth
 9    corresponding to the 1975-1980 time period; the corresponding concentration for the most
10    recent time period (1990-1995) at that location was approximately 50% lower (ISA, section
11    3.6.2; Callender and Rice, 2000).  These data may reflect changes in surface water discharges of
12    Pb in the Atlanta area, during this time period, as well as changes in air deposition.  The smaller
13    reduction (on the order of 20%) observed across the 20 year time period at the upstream site is
14    likely more generally reflective of changes in air deposition.
                 January 2013
                                  2-41
Draft - Do Not Quote or Cite

-------
  1     2.4   REFERENCES

  2     Brown, RW; Gonzales, C; Hooper, MJ; Bayat, AC; Fornerette, AM; McBride, TJ; Longoria, T; Mielke, HW.
  3            (2008). Soil lead (Pb) in residential transects through Lubbock, Texas: A preliminary assessment. Environ
  4            Geochem Health 30: 541-547. http://dx.doi.org/10.1007/sl0653-008-9180-y.

  5     Callender, E; Rice, KC. (2000). The urban environmental gradient: Anthropogenic influences on the spatial and
  6            temporal distributions of lead and zinc in sediments. Environ Sci Technol 34: 232-238.
  7            http://dx.doi.org/10.1021/es990380s.

  8     Caravanos, I; Weiss, A.L.; Jaeger, RJ. (2006a) An exterior and interior leaded dust deposition survey in New York
  9            City: results of a 2-year study. Environ.  Res. 100:  159-164.

10     Caravanos, J; Weiss, AL; Blaise, MJ; Jaeger, RJ. (2006b). A survey of spatially distributed exterior dust lead
11            loadings in New York City. Environ Res 100: 165-172. http://dx.doi.0rg/10.1016/i.envres.2005.05.001.

12     Cavender, K. (2008). Update of Analysis of Proposed Source-Oriented Monitoring Emission Threshold.
13            Memorandum to Lead NAAQS Review Docket (OAQ-2006-0735)
14            http://www.epa.gov/ttnnaaqs/standards/pb/data/20081015Cavender.pdf

15     Clark, HF; Brabander, DJ; Erdil, RM. (2006). Sources, sinks, and exposure pathways of lead in urban garden soil. J
16            Environ Qua! 35: 2066-2074. http://dx.doi.org/10.2134/ieq2005.0464.

17     DRRC. (Doe Run Company). (2010). DRRC multi-media Consent Decree. Herculaneum, MO: Doe Run Resources
18            Corporation.

19     Egeghy, P. P.; Quackenboss, J. J.; Catlin,  S.; Ryan, P. B. (2005) Determinants of temporal variability in
20            NHEXASMaryland environmental concentrations, exposures, and biomarkers. J. Exposure Anal. Environ.
21            Epidemiol. 15: 388-397.

22     Evans, GC; Norton, SA; Fernandez, IJ; Kahl, JS; Hanson, D. (2005). Changes in concentrations of major elements
23            and trace metals in northeastern US-Canadian sub-alpine forest floors. Water Air Soil Pollut 163: 245-267.

24     Farfel, MR;  Orlova, AO; Lees, PSJ; Rohde,  C; Ashley, PJ;  Chisolm, JJ, Jr. (2005). A study of urban housing
25            demolition as a source of lead in ambient dust on sidewalks, streets, and alleys. Environ Res 99: 204-213.
26            http://dx.doi.0rg/10.1016/j.envres.2004.10.005.

27     Gaitens, JM; Dixon, SL;  Jacobs, DE; Nagaraja, J; Strauss, W; Wilson, JW; Ashley, PJ. (2009). Exposure of US
28            children to residential dust lead, 1999-2004:1. Housing and demographic factors. Environ Health Perspect
29            117: 461-467. http://dx.doi.org/10.1289/ehp.11917.

30     Garland, JA, Nicholson, KW 1991. A review of methods for sampling large airborne particles and associated
31            radioactivity. J.  Aerosol Sci. Vol.22 (4):479-499.

32     Hasselbach, L; Ver Hoef, JM; Ford, J; Neitlich, P; Crecelius, E; Berryman, S; Wolk, B; Bohle, T. (2005). Spatial
3 3            patterns of cadmium and lead deposition on and adjacent to National Park Service lands in the vicinity of
34            RedDogMine,  Alaska. Sci Total Environ 348: 211-230. http://dx.doi.0rg/10.1016/j.scitotenv.2004.12.084.

3 5     Hilts, S. R. (2003) Effect of smelter emission reductions on children's blood lead levels. Sci. Total Environ. 303: 51-
36            58.

37     Jackson, B.P., P.V. Winger, PJ. Lasier (2004) Atmospheric lead deposition to Okefenokee Swamp, Georgia, USA.
38            Environ Poll. 130:445-451.
                  January 2013                      2-42             Draft - Do Not Quote or Cite

-------
  1     Johnson, C. E.; Siccama, T. G.; Driscoll, C. T.; Likens, G. E.; Moeller, R. E. (1995) Changes in lead
  2            biogeochemistry in response to decreasing atmospheric inputs. Ecol. Appl. 5: 813-822.

  3     Johnson, AH; Richter, SL. (2010). Organic-horizon lead, copper, and zinc contents of Mid-Atlantic forest soils,
  4             1978-2004. Soil Sci Soc Am J 74: 1001-1009. http://dx.doi.org/10.2136/sssaj2008.0337.

  5     Kaste, JM; Bostick, BC; Friedland, AJ; Schroth, AW; Siccama, TG. (2006). Fate and speciation of gasoline-derived
  6            lead in organic horizons of the northeastern USA. Soil Sci Soc Am J 70: 1688-1698.
  7            http://dx.doi.org/10.2136/sssaj2005.0321.

  8     Kaste, J.; Friedland, A.; Stirrup, S. (2003) Using stable and radioactive isotopes to trace atmospherically deposited
  9            Pb in montane forest soils. Environ. Sci. Technol. 37: 3560-3567.

10     Kenny L; Beaumont G, Gudmundsson A, Thorpe A, Koch W. (2005) Aspiration and sampling efficiencies of the
11            TSP and lowered paniculate matter inlets.  J Environ Monit. May 2005. 7 481-487.

12     Landers, DH; Simonich, SL; Jaffe, DA; Geiser, LH; Campbell, DH; Schwindt, AR; Schreck, CB; Kent, ML; Hafner,
13            WD; Taylor, HE; Hageman, KJ; Usenko, S; Ackerman, LK; Schrlau, JE; Rose, NL; Blett, TF; Erway, MM.
14            (2008). The fate, transport, and ecological impacts of airborne contaminants in western national parks
15            (USA). (EPA/600/R-07/138). Corvallis, Oregon: U.S. Environmental Protection Agency, NHEERL,
16            Western Ecology Division. http://www.nature.nps.gov/air/studies/air_toxics/WACAPreport.cfm.

17     Landers, DH; Simonich, SM; Jaffe, D; Geiser, L; Campbell, DH; Schwindt, A; Schreck, C; Kent, M; Hafner, W;
18            Taylor, HE; Hageman, K; Usenko, S; Ackerman, L; Schrlau, J; Rose, N; Blett, T; Erway, MM. (2010). The
19            Western Airborne Contaminant Assessment Project (WACAP): An interdisciplinary evaluation of the
20            impacts of airborne contaminants in western U.S. National Parks. Environ Sci Technol 44: 855-859.
21            http://dx.doi.org/10.1021/es901866e.

22     Lim, JH; Sabin, LD; Schiff, KC; Stolzenbach, KD. (2006). Concentration, size distribution, and dry deposition rate
23            of particle-associated metals in the Los Angeles region. Atmos Environ 40: 7810-7823.
24            http://dx.doi.0rg/10.1016/j.atmosenv.2006.07.025.

25     Mahler, B. J.; Van Metre, P. C.; Callender, E. (2006) Trends in metals in urban and reference lake sediments across
26            the United States, 1970 to 2001. Environ. Toxicol. Chem. 25:  1698-1709.

27     Mannino, D.M.; Albalak, R.; Grosse, S.; Repace, J. (2003) Second-hand smoke exposure and blood lead levels in
28            U.S. children. Epidemiology 14(6): 719-727.

29     Miller, E. K.; Friedland, A. J. (1994) Lead migration in forest soils: response to changing atmospheric inputs.
30            Environ. Sci. Technol. 28: 662-669.

31     Osterberg, E; Mayewski, P; Kreutz, K; Fisher, D; Handley, M; Sneed, S; Zdanowicz,  C; Zheng, J; Demuth, M;
32            Waskiewicz, M; Bourgeois, J. (2008). Ice core record of rising lead pollution in the North Pacific
33            atmosphere. Geophys Res Lett 35: L05810. http://dx.doi.org/10.1029/2007gl032680

34     Paode, RD; Sofuoglu, SC; Sivadechathep, J; Noll, KE; Holsen, TM; Keeler, GJ. (1998). Dry deposition fluxes and
3 5            mass  size distributions of Pb, Cu, and Zn measured in southern Lake Michigan during AEOLOS. Environ
36            Sci Technol 32: 16291635. http://dx.doi.org/10.1021/es970892b.

37     Russell A. and Samet, J.M. (2010) Letter from Dr. A. Russell, Chair, Clean Air Scientific Advisory Committee,
3 8            Ambient Air Monitoring Methods Subcommittee and J.M. Samet,  Chair, Clean Air Scientific Advisory
39            Committee, to Administrator Lisa P. Jackson. Re:  CAS AC Review of EPA's White Paper Approach for
40            the Development of a new Federal Reference Method (FROM) for Lead in Total Suspended Particulates
41            (Pb-TSP). November 30, 2010.
                  January 2013                      2-43             Draft - Do Not Quote or Cite

-------
  1     Sabin, LD; Schiff, KC. (2008). Dry atmospheric deposition rates of metals along a coastal transect in southern
  2             California. Atmos Environ42: 6606-6613. http://dx.doi.org/10.1016/j.atmosenv.2008.04.042

  3     U.S. Environmental Protection Agency. (1996-2011). AP-42, Compilation of Air Pollutant Emission Factors, 5th
  4             Edition. Volume 1: Stationary Point and Area Sources, Chapter 13: Miscellaneous Sources. Available at:
  5             http://www.epa.gov/ttn/chief/ap42/chl3/index.html

  6     U.S. Environmental Protection Agency. (2012) National Emissions Inventory for 2008, version 3. Office of Air
  7             Quality Planning and Standards, Research Triangle Park, NC. November 2012.
  8             http://www.epa.gov/ttn/chief/net/2008inventory.html

  9     Vucetich, JA; Outridge, PM; Peterson, RO; Eide, R; Isrenn, R. (2009). Mercury, lead and lead isotope ratios in the
10             teeth of moose (Alces alces) from Isle Royale, U.S. Upper Midwest, from 1952 to 2002. J Environ Monit.

11     Wang, E. X.; Benoit, G. (1997) Fate and transport of contaminant lead in spodosols: a simple box model analysis.
12             Water Air Soil Pollut. 95: 381-397.

13     Watmough, SA; Dillon, PJ. (2007). Lead biogeochemistry in a central Ontario forested watershed. Biogeochemistry
14             84: 143-159. http://dx.doi.org/10.1007/sl0533-007-9110-6

15     Yesilonis, ID; Pouyat, RV; Neerchal, NK. (2008). Spatial distribution of metals in soils in Baltimore, Maryland:
16             Role of native parent material, proximity to major roads, housing age and screening guidelines. Environ
17             Pollut 156: 723-731. http://dx.doi.0rg/10.1016/j.envpol.2008.06.010.

18     Yi, SM; Totten, LA; Thota, S; Yan, S; Offenberg, JH; Eisenreich, SJ; Graney, J; Holsen, TM. (2006). Atmospheric
19             dry deposition of trace elements measured around the urban and industrially impacted NY-NJ harbor.
20             Atmos Environ 40: 6626-6637. http://dx.doi.org/10.1016/j.atmosenv.2006.05.062.

21     Yohn, S., Long, D., Fett, J., Patino, L. (2004) Regional versus local influences on lead and cadmium loading to the
22             Great Lakes region. Appl. Geochem. 19:  1157-1175.

23     Zhang, Y.-H. (2003) 100 years of Pb deposition and transport in soils in Champaign, Illinois, U.S.A. Water Air Soil
24             Pollut. 146: 197-210.
                   January 2013                       2-44             Draft - Do Not Quote or Cite

-------
 1            3   HEALTH EFFECTS AND EXPOSURE/RISK INFORMATION

 2          This chapter presents key aspects of the current evidence of lead-related health effects
 3    and presents exposure/risk information from the quantitative assessment performed in the last
 4    review in the context of the currently available evidence. Staff has drawn from EPA's synthesis
 5    of the scientific evidence presented in the draft Integrated Science Assessment for Lead (USEP A,
 6    2012; henceforth referred to as the ISA) and 2006 Air Quality Criteria Document for Lead
 1    (USEP A, 2006; henceforth referred to as the 2006 CD), and from the documentation of the 2007
 8    human exposure and health risk assessment (documented in USEP A, 2007a; henceforth referred
 9    to as the 2007 REA).  The chapter is organized into sections considering the information  on
10    blood Pb as a biomarker (section 3.1), the nature of Pb effects on health (section 3.2), public
11    health implications and at-risk populations (section 3.3), and exposure and risk information
12    (section  3.4).  Presentation within these sections is organized to address key policy-relevant
13    questions for this review concerning the evidence and exposure/risk information, building upon
14    the questions included in the IRP (IRP, section 3.1).

15    3.1   INTERNAL DISPOSITION AND BIOMARKERS OF EXPOSURE AND DOSE
16          The health effects of Pb, discussed in detail in the ISA and summarized in Section 3.2
17    below, are remote from the portals of entry to the body (i.e., the respiratory system and
18    gastrointestinal tract).  Consequently, the internal disposition and distribution of Pb is an  integral
19    aspect of the relationship between exposure and effect.  Inhaled lead-bearing particles, depending
20    on size and solubility, may be absorbed into the  systemic circulation or transported to the
21    gastrointestinal tract (ISA, section 4.2.1.1). The absorption efficiency of Pb from the
22    gastrointestinal (GI) tract varies with characteristics associated with the ingested Pb (e.g.,
23    particle size and chemical form or matrix), as  well as with an individual's physiology (e.g.,
24    maturity of the GI tract),  nutritional status (e.g.,  iron, calcium, and zinc deficiencies increase
25    absorption), and the presence of food in the GI tract (ISA, section 4.2.1.2).  Once in the blood
26    stream, where approximately 99% of the Pb is associated with red blood cells (mostly bound to
27    aminolevulinic acid dehydratase, the predominant ligand), Pb is quickly distributed throughout
28    the body (e.g., within days) and is available for exchange with the soft and skeletal tissues,
29    conceptually viewed as the fast and slow turnover pools, respectively (ISA, section 4.2.2).
30    Skeletal  tissue serves as the largest storage compartment, with much less Pb stored in soft tissues
31    (e.g., kidney, liver, brain, etc) (ISA, section 4.2.2.2).
32          The role of the bone as the main storage  compartment is related to the ability of Pb to
33    form stable complexes with phosphate and replace calcium in the salt comprising the primary
34    crystalline matrix of bone (ISA, section 4.2.2.2). In infants less than a year old, the bone is

         January 2013                          3-1             Draft - Do Not Quote or Cite

-------
 1    estimated to contain approximately 60% of the total body burden of Pb (ISA, section 4.2.2.2;
 2    Barry, 1975).  Circulating Pb is taken up into the bone regions of active calcification.
 3    Accordingly, during early childhood there is rapid uptake of Pb into mineralizing bone, with
 4    somewhat more than 70% of total body burden Pb estimated to reside in bone of children aged 2
 5    to 16, increasing up to more than 90% by adulthood (ISA, section 4.2.2.2; Barry, 1975). The net
 6    accumulation of Pb in bone over a person's lifetime results in bone lead concentrations generally
 7    increasing with age (ISA, section 4.2.2.2).
 8          The distribution of Pb in the body is dynamic. Throughout life, Pb in the body is
 9    exchanged between blood and bone, and between blood and soft tissues (ISA, sections 4.3.5 and
10    4.2.2; 2006 CD, section 4.3.2). The rates of these exchanges vary with age, exposure and
11    various physiological variables. For example, resorption of bone, which results in the
12    mobilization of Pb from bone into the blood, is a gradual process in later adulthood. Resorption
13    rate is appreciably increased in pregnant or nursing women, and in association with osteoporosis
14    in postmenopausal women or, to a lesser magnitude, in older men (ISA, sections 4.3.5.2).
15    Changes in Pb exposure circumstances can also influence these exchanges, e.g., substantial
16    reductions in exposure levels contribute to increased release of Pb from the bone into the blood
17    (ISA, section 4.3.5). The studies that address the relative contributions of bone Pb  and current
18    Pb exposure to blood Pb are limited to a few human studies during the 1980s and early 1990s,
19    when leaded gasoline usage was common, and a primate study.  These studies indicate an
20    appreciable contribution from bone Pb stores to Pb in blood, on the order of 40-70% of blood Pb
21    contributed from bone, under circumstances with higher concurrent exposure than exposures
22    common today (ISA, section 4.3.5; Smith et al., 1996; Gulson et al., 1995; Manton, 1985;
23    Franklin etal., 1997).
24          During bone resorption that occurs in pregnancy, Pb is released from bone,  increasing the
25    bone contribution to maternal blood Pb levels, and maternal Pb circulates  across the placenta,
26    posing risk to the developing fetus and providing the fetal body burden (ISA, sections 4.2.2.4,
27    4.3.5, 4.7.2 and 4.7.3; 2006 CD, 6.6.2; Chuang et al., 2001).  The relative  size of contributions of
28    maternal bone Pb and current maternal exposure to maternal blood Pb and fetal body burden, are
29    influenced by the relative magnitude of current and historical  exposures. In various study
30    populations with mean maternal blood Pb levels ranging from 1.7 to 8.6 |ig/dL, average blood Pb
31    concentration in the umbilical cord (as representative of newborn blood) has been reported to
32    range from 70% to 100% of average maternal blood Pb (ISA,  section 4.2.2.4); the range is
33    similar for four study  populations with mean maternal blood Pb level below 4 |ig/dL (Amaral et
34    al., 2010; Patel and Prabhu, 2009; Lagerkvist et al.,  1996; Jedrychowski et al., 2011). The
35    relationship for individual mother-child pairs is more variable. In a low income population with
36    mean maternal blood Pb of 1.9 |ig/dL, factors associated with occurrences of relatively higher

         January 2013                           3-2            Draft - Do Not Quote or Cite

-------
 1    cord blood Pb level to maternal blood Pb level pairs included maternal blood pressure and
 2    alcohol consumption, while factors associated with relatively lower ratios of cord blood Pb to
 3    maternal blood Pb included maternal hemoglobin and sickle cell trait (Harville et al., 2005).  As
 4    a result of the contributions of maternal bone Pb to maternal blood Pb during pregnancy, the
 5    contribution to fetal body burden from maternal bone Pb relative to maternal concurrent
 6    exposure is appreciable. This maternal bone Pb contribution to fetal body burden is likely to
 7    vary, however, in response to differences between maternal historical exposures and those during
 8    pregnancy, among other factors (2006  CD, sections 4.3.1.5 and 4.3.2.5; ISA, section 6.1; Chuang
 9    et al., 2001; Gulson et al., 1999; Gulson et al., 2003; Gulson et al., 2004a; Rothenberg et al.,
10    2000).
11          Limited data indicate that blood Pb levels in some newborns may decline during the
12    period extending through the first few months of life; the extent to which this may occur may be
13    influenced by the magnitude of blood Pb level at birth and of subsequent exposure during early
14    infancy (ISA, section 4.2.2.1; Carbone et al.,  1998; Simon et al., 2007; Gulson et al., 1999,
15    2001). As infants become more mobile and engage in hand-to-mouth behavior, blood Pb levels
16    then commonly increase to a peak around two years of age (ISA, sections 4.4.1 and 6.2.1.1).
17         •  Does the current evidence continue to support blood Pb level as a useful indicator
18             of Pb exposure and dose for purposes of characterizing Pb health effects, with
19             well recognized strengths and limitations? To what extent does the evidence
20             suggest alternatives?
21          As discussed in past CDs and in the ISA, blood Pb is the most commonly used indicator
22    of human Pb exposure. For example, given the association with exposure and the relative  ease of
23    collection, blood Pb levels are extensively used as an index or biomarker of exposure by national
24    and international health agencies, as well as in epidemiological and toxicological studies of Pb
25    health effects and dose-response relationships (ISA, sections 4.3.2, 4.4.1, 5.3,  5.4, 5.5, 5.6, 5.7,
26    and 5.8).  Since 1976, the U.S. Centers for Disease Control and Prevention (CDC) has been
27    monitoring blood Pb levels nationally through the National Health and Nutrition Examination
28    Survey (NHANES).  This survey has documented the dramatic decline in mean blood Pb levels
29    in all ages of the U.S. population that has occurred since the 1970s, as shown in Figure 3-1, and
30    that coincides with actions on leaded fuels, leaded paint, lead in food packaging, and lead-
31    containing plumbing materials that have reduced Pb exposure in the U.S. (ISA, section 4.4.1;
32    Pirkle et al., 1994; Schwemberger et al., 2005). This decline has continued over the more recent
33    past.  For example, the 2009-2010 geometric  mean blood Pb level in U.S. children aged 1-5
34    years is 1.17 |ig/dL, as compared to 1.51  |ig/dL in 2007-2008 (ISA, section 4.4.1) and 1.8  |ig/dL
35    in 2003-2004, the most recent data available at the time of the last review (73 FR 67002);
         January 2013                           3-3             Draft - Do Not Quote or Cite

-------
 1

 2
 4
 5
 6
 7
somewhat less dramatic declines have also been reported in the upper tails of the distribution and

in different groups with higher levels than the general child population (ISA, Figure 4-19).
          o
          CJ

          ID
        o ^r
        c
        m
        0) CO
        E
        o
        0)
        E
        o
        0)
        0)
          CM -
          cohort birth yrs
           1900 to 1930 •
           1930 to 1940*
           1940 to 1950
           1950 to 1960
           1960 to 1970
           1970 to 1975
           1975 to 1980
           1930 to 1985
           1935 to 1990
           1990 to 1995
           1995 to 2000
           2000 to 2005
           2005 to 2003
             1970
                          1930
   1990
exam year
2000
2010
Source: Adapted from data from the NHANES (NCHS, 2010)
Note: The means of logged blood Pb were weighted to represent national averages. Data were from the publically NHANES II, NHANES III for
1988 1991 and 1992 1994, and the continuous NHANES in 1999 2000,2003 2004,2005 2006,2007 2008. Continuous NHANES data from
2001 2002 and 2009 2010 are not included because there were only 551 blood Pb samples in each of those data sets. The year plotted for
exam year was the reported exam year for NHANES II, the middle year of each of the phases of NHANES III, and the second year of each of
the continuous NHANES.
10    Figure 3-1.  Temporal trend in mean blood Pb levels for NHANES cohorts.


11           The CDC, and its predecessor agencies, have for many years used blood Pb level as a

12    metric for identifying highly exposed children at risk of adverse health effects for whom

13    recommendations might be made for their protection (e.g., CDC, 1991; CDC, 2005).  In 1978,

14    when the current Pb NAAQS was established, the CDC recognized a blood Pb level of 30 ug/dL

15    as a level warranting individual intervention (CDC, 1991).  In 1985, the CDC recognized a level

16    of 25 ug/dL for individual child intervention, and in 1991, they recognized a level of 15 ug/dL at

17    which individual intervention actions were recommended, and a level of 10 ug/dL at which more

18    general community-wide prevention activities were to be implemented (CDC, 1991; CDC,

19    2005).  In 2005, with consideration of a review of the evidence by their advisory  committee,
         January 2073
                                            5-4
                 Draft - Do Not Quote or Cite

-------
 1    CDC revised their statement on Preventing Lead Poisoning in Young Children, specifically
 2    recognizing the evidence of adverse health effects in children with blood Pb levels below 10
 3    ug/dL and emphasizing the importance of preventive measures (CDC, 2005). * Consistent with
 4    this statement, the CDC revised their approach in 2012 to one that relies on the 97.5th percentile
 5    of blood Pb concentrations in U.S. children (currently  5 |ig/dL)2 as a reference level for
 6    identifying children for whom they recommend particular follow-up actions (CDC, 2012).
 7           Multiple studies of the relationship between Pb exposure and blood Pb in children (e.g.,
 8    Lanphear and Roghmann 1997; Lanphear et al., 1998), have shown blood Pb levels to reflect Pb
 9    exposures,  including exposures to Pb in surface dust. These and studies of child populations
10    near active sources of air emissions (e.g., metal smelters), further demonstrate the effect of
11    airborne Pb on interior dust and on blood Pb (ISA, sections 4.4.1, 4.5.1  and 4.5.3; Hilts, 2003;
12    Gulson et al., 2004b). Accordingly, blood Pb level was the index of exposure in the assessment
13    of young children's risk performed in the last Pb NAAQS review (described in section 3.4
14    below).
15           A well recognized strength of using children's  blood Pb to investigate relationships
16    between Pb exposure and  health effects is the relatively lesser  uncertainty in the causality
17    aspects of such relationships than might be associated  with such relationships based, for
18    example, on media concentration. Blood Pb is an integrated marker of aggregate exposure
19    across all pathways.  For example, the blood Pb-response relationships described in
20    epidemiological studies of lead-exposed populations do not distinguish  among different sources
21    of Pb or pathways of Pb exposure (e.g., inhalation, ingestion of indoor dust, ingestion of dust
22    containing  leaded paint).  Given our focus on ambient  air contributions  to Pb exposures,
23    discussion in response to the question below considers the available information regarding
24    relationships between Pb in ambient air and the associated Pb in blood.  Additionally, the
25    exposure assessment performed for the last review of the Pb NAAQS (and described in section
26    3.4 below), employed biokinetic modeling to estimate  blood Pb levels associated with aggregate
27    Pb exposure to inform our estimates of contributions to blood Pb arising from ambient air related
28    Pb.
29           The response of adult blood Pb levels to appreciable changes in  exposure circumstances
30    is generally slower than that of blood Pb levels in young children.  For example, simulations
31    using biokinetic models indicate that blood Pb levels in adults achieve a new quasi-steady state
             1 With the 2005 statement, CDC identified a variety of reasons, reflecting both scientific and practical
      considerations, for not lowering the 1991 level of concern, including a lack of effective clinical or public health
      interventions to reliably and consistently reduce blood Pb levels that are already below 10 ug/dL, the lack of a
      demonstrated threshold for adverse effects, and concerns for deflecting resources from children with higher blood
      Pb levels (CDC, 2005).
             2 CDC intends to update the value every 4 years using the two most recent NHANES surveys (CDC, 2012).
         January 2013                            3-5            Draft - Do Not Quote or Cite

-------
 1    within 75-100 days (approximately 3-4 times the blood elimination half-life) subsequent to
 2    abrupt increases in Pb intake (ISA, section 4.3.5.2); similar models indicate a much quicker
 3    response of blood Pb levels in children both with regard to abrupt increases and reductions in Pb
 4    exposure (ISA,  section 4.3.5.1). The response in young children may reflect their much more
 5    labile bone pool associated with the rapid turnover of bone mineral in response to their rapid
 6    growth rates (ISA, section 4.3.5).  As a result of these physiological processes in young children,
 7    their blood Pb levels tend to more quickly reflect changes in their total body burden (associated
 8    with their shorter exposure history), and can also reflect changes in recent exposures (ISA,
 9    section 4.3.5).
10           The extent to which blood Pb measurements represent current exposure circumstances
11    may be uncertain, particularly where histories include exposures largely different from those
12    occurring more recently (ISA, section 4.3.2). This uncertainty may be greater for blood Pb
13    measurements taken in older children3 and adults that those for young children as a result of their
14    longer exposure histories.  For example, the bone of adults has accumulated decades of Pb
15    exposures (with past exposures often greater than current ones for current U.S. populations) such
16    that the bone may be a significant source of Pb  in blood in later years, after exposure has ended.
17    Thus, in populations with past exposures that were appreciably elevated in comparison to more
18    recent exposures, current blood Pb levels  may predominantly reflect their exposure history rather
19    than current exposures (ISA, sections 4.3.5.2 and 4.4.1).  Accordingly, the extent to which
20    studies using cross-sectional blood Pb  measurements as the exposure biomarker inform our
21    understanding of relationships between recent Pb exposures and various health effects, can differ
22    across age groups, and is greatest for very young children, as discussed in section 3.3 below.
23           Alternative biomarkers of Pb exposure include bone Pb, teeth Pb, and hair Pb (ISA,
24    section 4.3.1, 4.3.3 and 4.3.4).  Given the  role of bone as the repository of more than 90% of an
25    adult body's Pb burden,  bone Pb levels are recognized as indicators of body burden and
26    cumulative exposure history (ISA, sections 4.2.2.2 and 4.3). The currently available evidence
27    does not, however, indicate these or any other alternatives to be superior to blood Pb in young
28    children or as commonly used for the purposes  of tracking recent Pb exposures and for assessing
29    potential health risk for this age group (ISA, section 4.7.3). In summary, the current evidence
30    continues to support our conclusions from the last review regarding the use of blood Pb levels as
31    an internal biomarker of Pb exposure and  dose informative to characterizing Pb health effects in
32    young children.
             3 There is a paucity of experimental measurements of Pb biomarkers during adolescence to inform our
      characterization of Pb biokinetics (and relative roles of recent vs historic exposure on blood Pb levels) during this
      lifestage, in which individuals undergo rapid changes in sexual development, growth, food and water intake, bone
      growth and turnover, behavior, etc. (ISA, section 4.3).
         January 2013                           3-6             Draft - Do Not Quote or Cite

-------
 1         •   To what extent has new information altered scientific conclusions regarding the
 2             relationships between Pb in ambient air and Pb in children's blood?
 3           As described above, blood Pb is an integrated marker of aggregate exposure across all
 4    pathways and a reflection of exposure history. Thus, our interpretation of the health effects
 5    evidence for purposes of this review necessitates characterization of the relationships between Pb
 6    from those sources and pathways of interest in this review (i.e., those related to Pb emitted into
 7    the air) and blood Pb.  The evidence in this regard derives from analyses of datasets for
 8    populations residing in areas with differing air Pb concentrations, including  datasets for
 9    circumstances in which blood Pb levels have changed in response to changes in air Pb. The
10    control for variables other than air Pb that can affect blood Pb varies across these analyses.
11           Lead in ambient air contributes to Pb in blood by multiple exposure pathways by both
12    inhalation and ingestion exposure routes (ISA, section 4.1.1).  The quantitative relationship
13    between ambient air Pb and blood Pb, which is often termed a slope or ratio, describes the
14    increase in blood Pb (in |ig/dL) estimated to be associated with each unit increase of air Pb (in
15    Hg/m3). Ratios are presented in the form of 1 :x, with the 1 representing air Pb (in |ig/m3) and x
16    representing blood Pb (in  |ig/dL). Description of ratios as higher or lower refers to the values for
17    x (i.e., the change in blood Pb per unit of air Pb).  Slopes are presented as simply the value of x.
18           At the conclusion of the last review in 2008, the EPA interpreted the evidence as
19    providing support for use  (in informing the Administrator's decision on standard level) of a
20    range "inclusive at upper end of estimates on the order of 1:10 and at the lower end  on the order
21    of 1:5" (73 FR 67002).  This conclusion reflected consideration of the air-to-blood ratios
22    presented in the 1986 CD4 and associated observations regarding factors contributing to variation
23    in such ratios, ratios reported subsequently  and ratios estimated based on modeling performed in
24    the REA, as well as advice from CASAC (73 FR 66973-66975, 67001-67002). The information
25    available in this review, which is assessed in the ISA and is largely, although not completely,
26    comprised of studies that were available in  the last review, does not alter the primary scientific
27    conclusions drawn in the last review regarding the relationships between Pb in ambient air and
28    Pb in children's blood.  The ratios summarized in the ISA in this review span a range generally
29    consistent with the range concluded in 2008 (ISA, section 4.5.1).
30           The air-to-blood ratios for children, or for populations inclusive of children,  discussed in
31    the ISA for this review are summarized in Table 3-1. The evidence pertaining to this
32    quantitative relationship between air Pb and children's blood Pb is now, as in the past, limited by
33    the circumstances in which the data are collected. These estimates are generally developed from
34    studies of populations in a variety of Pb exposure circumstances. Accordingly, there is
             4 The 2006 CD did not include an assessment of then-current evidence on air-to-blood ratios.
         January 2013                            3-7             Draft - Do Not Quote or Cite

-------
 1    significant variability in air-to-blood ratios among the different study populations exposed to Pb
 2    through different air-related exposure pathways and at different exposure levels. This variability
 3    in air-to-blood estimates can relate to the representation of air-related pathways and study
 4    populations, including, for example, relatively narrow age ranges for the population, to reduce
 5    age-related variability in blood Pb or including populations with narrowly specified dietary
 6    sources. It can also relate to the precision of air and blood measurements and of the study
 7    circumstances, such as with regard to spatial and temporal aspects. Additionally, in situations
 8    where exposure to nonair sources covaries with air-related exposures that are not accounted for
 9    in deriving ratio estimates, uncertainties may relate to the potential for confounding by nonair
10    exposure covariance (ISA, section 4.5).
11           As was noted in the last review, population age is an important influence on the
12    magnitude of air-to-blood ratio estimates derived. Ratios for children are generally higher than
13    those for adults, and higher for young children than older children, perhaps due to behavioral
14    differences between the age groups, as well as the more rapid response of children's blood Pb to
15    exposure and body burden changes  (as discussed above) and shorter exposure history.  Similarly,
16    given the common  pattern  of higher blood Pb levels in pre-school aged children than during the
17    rest of childhood, related to behaviors that increase environmental exposures, ratios would be
18    expected to be highest in earlier childhood.  Additionally, estimates of air-to-blood ratios that
19    include air-related ingestion pathways in addition to the inhalation pathway are "necessarily
20    higher", in terms of blood Pb response, than those estimates based on inhalation alone (USEPA
21    1986, p. 11-106). Thus, the extent to which studies account for the full set of air-related
22    inhalation and ingestion exposure pathways affects the magnitude of the resultant air-to-blood
23    estimates, such that including fewer pathways  as "air-related" yields lower ratios. Estimates of
24    air-to-blood ratios can also be influenced by population characteristics that may influence blood
25    Pb; accordingly, some analyses include adjustments.
26           Given the recognition of young children as a key at-risk population in section 3.3 below,
27    as well as the influence of  age on blood Pb levels, the studies presented in Table 3-1 are grouped
28    with regard to the extent of their inclusion of young children (e.g., children barely school age).
29    Among the first group of studies, focused exclusively on young children, is the only study that
30    dates from the end  of or after the phase-out of leaded gasoline usage (Hilts, 2003). This study
31    reports changes in children's blood  Pb levels associated with reduced Pb emissions and
32    associated air concentrations near a Pb smelter in Canada (for children through age five). Given
33    the timing of this study, after the leaded gasoline phase-out, and its setting near a smelter, the
34    ambient air Pb in this study may be somewhat more comparable to that near sources in the U.S.
         January 2013                           3-8             Draft - Do Not Quote or Cite

-------
 1    today than other studies discussed here.  The study authors report an air-to-blood ratio of 1:6.5
 2    An EPA analysis of the air and blood data reported for 1996, 1999 and 2001 results in a ratio of
 3    1:6.5, and the analysis focused only on the 1996 and 1999 data (pre- and post- the new
 4    technology) yields a ratio of 1:7 (ISA, section 4.5.1; Hilts, 2003).6 The two other studies that
 5    focused on children of age  5 or younger analyze variations in air Pb as a result of variations in
 6    leaded gasoline usage in Chicago, Illinois. The study by Hayes et al.  (1994) compared patterns
 7    of ambient air Pb reductions and blood Pb reductions for large numbers of children, aged 6
 8    months - 5 years, in Chicago between 1974 and 1988, a period when significant reductions
 9    occurred in both measures.  The study reports a better fit for the log-log model which describes a
10    pattern of higher ratios with lower ambient air Pb and blood Pb levels (Hayes et al., 1994).
11    Based on the log-log model, the ratio derived for the relationship of quarterly mean air
12    concentration with blood Pb during the period is 1:8 (Table 3-1).  Another analysis for a Chicago
13    dataset, performed by Schwartz and Pitcher (1989) focused on the association of blood Pb in
14    black children (aged < 5 years) with the use of leaded gasoline from 1978 through 1980. Given
15    that leaded gasoline exposure occurs by air-related pathways, additional analyses have related
16    the leaded gasoline usage for this period to air concentrations. Using the resulting relationship of
17    blood Pb with air Pb (adjusted for age and a number of other covariates) yields a ratio of 1:8.6
18    (ISA, table 4-12, section 4.5.2). The blood Pb concentrations in the two leaded gasoline studies
19    are appreciably higher (a factor of two or more) than those in the study near the smelter (Hilts,
20    2003).
21            The second group of studies in Table 3-1 (comprised of studies including but not limited
22    to children less than or equal to five years of age) includes a complex statistical  analysis and
23    associated dataset for a cohort of children born in Mexico City from 1987 through 1992 (Schnaas
24    et al., 2004).  This  study, which was not assessed in the last review, encompasses the period of
             5 Sources of uncertainty include the role of factors other than ambient air Pb reduction in influencing
      decreases in blood Pb (ISA, section 4.5.1).  The author cited remedial programs (e.g., community and home-based
      dust control and education) as potentially responsible for some of the blood Pb reduction seen during the study
      period (1997 to 2001), although he notes that these programs were in place in 1992, suggesting they are unlikely to
      have contributed to the sudden drop in blood Pb levels occurring after 1997 (Hilts, 2003). Other aspects with
      potential implications for ratios include the potential for children with lower blood Pb levels not to return for
      subsequent testing, and the age range of 6 to 36 months in the 2001 blood screening compared to ages up to 60
      months in previous years (Hilts, 2003).
             6 This study considered changes in ambient air Pb levels and associated blood Pb levels over a five-year
      period which included closure of an older Pb smelter and subsequent opening of a newer facility in 1997 and a
      temporary (3 month) shutdown of all smelting activity in the summer of 2001. The author observed that the air-to-
      blood ratio for children in the area over the full period was approximately 1:6. The author noted limitations in the
      dataset associated with exposures in the second time period, after the temporary shutdown of the facility in 2001,
      including sampling of a different age group at that time and a shorter time period (3 months) at these lower ambient
      air Pb levels prior to collection of blood Pb levels. Consequently, EPA calculated an alternate air-to-blood  Pb ratio
      based on ambient air Pb and blood Pb reductions in the first time period, after opening of the new facility in 1997
      (ISA, section 4.5.1).
         January 2013                             3-9               Draft  - Do Not Quote or Cite

-------
 1    leaded gasoline usage and further informs our understanding of factors influencing the
 2    quantitative relationship between air Pb and children's blood Pb.  Air-to-blood ratios developed
 3    from this study are influenced by a number of factors and appear to range from roughly 2 to 6, in
 4    addition to an estimate of 9 (ISA, section 4.5.1), although this is derived from data set restricted
 5    to the latter years of the study when little change in air Pb concentration occurred, such that the
 6    role of air Pb may be more uncertain.  Estimates associated with the developmental period of
 7    highest exposure - e.g., age 2 years - range up to approximately 6, illustrating the influence of
 8    age on the ratio (ISA, section 4.5.1). Also in the Table 3-1  second group are two much older
 9    studies of populations comprising age ranges extending well beyond 6 years.  The first is the
10    review and meta-analysis by Brunekreef (1984) using datasets available at the time for variously
11    aged children as old as 18 years with identified air monitoring methods and reliable blood Pb
12    data for 18 locations in the U.S. and internationally.7 The author discusses potential confounders
13    of the relationship between air Pb and blood Pb, recognizing the desirability of taking them into
14    account when deriving an air-to-blood relationship from a community study, but noting that was
15    not feasible in such an analysis. Two models were produced, one based on the full pooled dataset
16    and a second limited to blood Pb-air Pb data pairs with blood Pb levels below 20 |ig/dL (r2
17    values were 0.692 and 0.331, respectively).  From these two logn-logn models, air-to-blood  ratio
18    estimates were derived for air concentrations corresponding to the geometric means of the two
19    sets of data pairs. At those concentrations (1.5 and 0.54 |ig/m3, respectively), the resultant ratios
20    both round to 5. The study by Schwartz and Pitcher (1989) described above also analyzed the
21    relationship between U.S. NHANES II blood Pb levels for white subjects,  aged <74 years, and
22    national usage of leaded gasoline. A separate, less specifically described, air Pb dataset was used
23    to convert the relationship of blood Pb with gas Pb to one for blood Pb with air Pb, with a
24    resultant ratio on the order of 9, adjusted for age and other covariates (Henderson, 2007, pp. D-2
25    to D-3; ISA, Table 4-12).
26           The last two studies included in Table 3-1 are focused on older children (ages 6-11). The
27    methods for characterization of air Pb concentrations (and soil Pb for latter study) also differ
28    from other studies in Table 3-1.  The first  study regressed average blood Pb concentrations  for
29    multiple locations around Mumbai,  India on average air Pb concentrations at those locations
30    (Tripathi et al.,  2001).  The values in the linear regression were 13 pairs of location-specific
31    geometric means of all the data collected over the 13-year period from 1984 to 1996; the
32    reported slope was 3.6 (Tripathi et al., 2001). The location-specific geometric mean blood Pb
             7 In the dataset reviewed by Brunekreef (1984), air-to-blood ratios from the subset of those studies that
      used quality control protocols and presented adjusted slopes include values of 3.6 (Zielhuis et al., 1979), 5.2 (Billick
      et al., 1979, 1980); 2.9 (Billick et al., 1983), and 8.5 (Brunekreef et al, 1983). The studies cited here adjusted for
      parental education (Zielhuis et al., 1979), age and race (Billick et al., 1979, 1980) and air Pb monitor height (Billick
      et al., 1983); Brunekreef et al. (1984) used multiple regression to control for several confounders (73 FR 66974).
         January 2013                           3-10             Draft - Do Not Quote or Cite

-------
 1    levels in this study (8.6-14.4 |ig/dL) indicate blood Pb distributions in this age group much
 2    higher than those pertinent to similarly aged children in the U.S. today. The second study
 3    analyzed air, soil and children's blood Pb concentrations in Duisburg, Germany during the
 4    leaded gasoline phase-out (Ranft et al., 2008).  Average blood Pb levels declined over the nearly
 5    20-year study period from 9 |ig/dL in 1983 (345 children average age of 9 years) to 3 |ig/dL in
 6    2000 (162 children average of 6 years).8 Average air Pb concentration declined from 0.45 |ig/m3
 7    to 0.06 |ig/m3 over the same period, with the largest reduction occurring between the first study
 8    year (derived from two monitoring sites for full study area) and the second study year, 1991, for
 9    which air concentrations were derived from a combination of dispersion modeling and the two
10    monitoring sites.9 For a mean air Pb concentration of 0.1 |ig/m3, the study's multivariate log-
11    linear regression model predicted air to blood ratios of 3.2 and 6.4 for "background" blood Pb
12    concentrations of 1.5 and 3 |ig/dL, respectively. In this study, background referred to Pb in
13    blood from other sources;  the blood Pb distribution over the study period, including levels when
14    air Pb concentrations are lowest, indicates 3 |ig/dL may be the better estimate of background for
15    this study population. Inclusion of soil Pb as a variable in the model may have contributed to an
16    underestimation of the these blood Pb-air Pb ratios for this study because some of the Pb in  soil
17    likely originated in air and the blood Pb-air Pb slope does not include the portion of the soil/dust
18    Pb ingestion pathway that derives from air Pb. The ISA also presents somewhat higher air-to-
19    blood ratio estimates derived using univariate linear, log-log and log-linear models on the
20    median air and blood Pb concentrations reported for the five years included in this study (ISA,
21    section 4.5.1; Ranft et al.,  2008, Table 2).  Uncertainties related to this study's estimates include
22    those related to the bulk of air concentration reduction occurring between the first two time
23    points (1983 and 1991) and the difference among the year's air datasets (e.g., two data sources
24    [air monitors] in 1983 and multiple geographical points from a combination of the monitors and
25    modeling in subsequent years).
              Blood Pb measurements were available on a total of 843 children across five time periods, in the first of
      which the average child age was 9 years while it was approximately 6 years in each of the latter years:  1983
      (n=356), 1991 (n=147), 1994 (n=122), 1997 (n=56), and 2000 (n=162) (Ranft et al., 2008).
             9 The 1983 air Pb concentrations were based on two monitoring stations, while a combination of dispersion
      modeling and monitoring data was used in the later years. Surface soil Pb measurements were from 2000-2001, but
      geo-matched to blood Pb measurements across full study period (Ranft et al., 2008).
         January 2013                           3-11             Draft - Do Not Quote or Cite

-------
1   Table 3-1.  Empirically derived air-to-blood ratios for populations inclusive of children.
Study Information
Quantitative Analysis
Air-to-Blood
Ratio A
Focused on children < 5 years old
Children, 0.5-5 yr (n = 9,604), average age 2.5 yr
Chicago, IL, 1974-1988
Urban area with lead-emitting industries, leaded gasoline usage
Hayes etal. (1994)
Children, 0.5-5 yr (1996-1999), 0.5-3 yr (2001) (n = 200-500)
Trail, BC, 1996-2001;
Small town before/after cleaner technology on large metals
smelter (at end of/after leaded gasoline phase-out)
Hilts (2003)
Black children, <5 yr (n = 5,476)
Chicago, IL, Feb 1976- Feb 1980
Area with lead-emitting industries, leaded gasoline usage
Schwartz and Pitcher (1989), U.S. EPA (1986a)
oqn-loan regression: quarterly mean PbB and PbA
[unadjusted]
PbB: 10-28 pg/dL (quarterly mean)
PbA: 0.05-1 .2 pg/m3 (quarterly mean)
linear regression: annual GM PbB and AM PbA
[unadjusted]
PbB: 4.7-11. 5 pg/dL (annual GM)
PbA: 0.03-1 .1 |jg/m3 (annual AM)
linear regression: quarterly mean PbB with gasoline Pb
(usage) [ adjusted for demographic covariates]
combined with gasoline Pb - air Pb relationship.
PbB: 18-27 pg/dL (quarterly mean, adjusted)
PbA- gas Pb relationship based on annual U.S. means
of per-site maximum quarterly means (0.36-1 .22 |jg/m3)
8.2 (@ 0.62) B
6-7.0°
8.6°
Larger age range, inclusive of children < 5 years old
Children, 96 groups of various age ranges, (n>1 90,000)
Various countries (18 locations), 1974-1983
urban or near lead-emitting industries, leaded gasoline usage
Brunekreefetal. (1984)
Children, born 1987-92 (n = 321); Mexico City, 1987-2002
Ave age increased over study period: <3yrs (1987-1992), and
increased by a year each year after that
Urban area during/after leaded gasoline usage
Schnaas etal. (2004)
U.S. NHANES II white subjects, 0.5-74 yr, Feb 1976-1980
National survey during time of leaded gasoline usage
Schwartz and Pitcher (1989), U.S. EPA (1986a)
Meta-analyses (logn-logn regression of group means):
(1) all children, (2) children <20 pg/dL [unadjusted]
PbB: 5-76 pg/dL (study group means)
PbA: 0.1-24 pg/m3 (location means)
Linear , logn-logn regressions: annual mean PbB and
PbA [unadjusted]
PbB: 5-12 |jg/dL (annual GM), aged 0.5-10 yr
PbA: 0.07-2.8 |jg/m3 (AM); 0.1-0.4 over last 6 years
Linear regression: PbB with mass Pb in gasoline as
described above, [adjusted for demographic covariates]
PbB: 11-18 pg/dL (mean per gas Pb, adjusted)
PbA- gas Pb relationship based on annual U.S. means
of per-site maximum quarterly means (0.36-1 .22 pg/m3)
Full dataset:
4.6(@1.5)E
<20|jg/dL:
4.8(@0.54)F
Linear:
9.0(0.1-0.4)
2.5 (full range)
Log-log: 4.5
(@0.4)c
9.3°
Focused on children > 6 years old
Children, 6-10 yr(n = 544)
Mumbai, India, 1984-1996
Large urban area, leaded gasoline usage
Tripathi etal. (2001)
Children, 6-1 1 yr (n=843); ave~9.5 yr (1983), -6.5 (others)
Duisburg, Germany-5 areas: 1983, 1991, 1994, 1997, 2000
Industrial urban area during/after leaded gasoline usage
Ranft etal. (2008)
Linear regression: 13-vear location-specific GM PbB and
PbA [unadjusted]
PbB: 8.6-14.4 |jg/dL (location GM)
PbA: 0.11-1.18 |jg/m3 (location GM); 0.45 (overall GM)
Linear, logn-logn. log-linear regressions: annual mean
PbB and air Pb [unadjusted]
PbB: 3.33-9.13 |jg/dL (AMs of 5 study years)
PbA: 0.06-0.45 |jg/m3 (5 AMs), 0.10 (overall median)
3.6
Log-linear G
6.4, 3.2
A - Predicted change in blood Pb (|jg/dL per |jg/m3) over range ± 0.01 |jg/m3 from study's central air Pb, which is provided in |jg/m3 in parentheses.
For linear models, this is simply the air Pb coefficient.
B - In(PbB) = In(PbA) x 0.24 +3.17 (ISA, Table4-12; Hayes et al., 1994)
C - See discussion in text and ISA, section 4.5.1.
D - Based on data for U.S. (1986 Pb CD). See ISA, section 4.5.1. Log-lin = log-linear model.
E - In(PbB) = In(PbA) x 0.3485 +2.853
F-ln(PbB) = ln(PbA)x 0.21 59 +2.620
G -Derived from regressions with separate soil Pb variable, contributing to underestimation of contribution from air Pb (see text and ISA, section 4.5.1).
GM, geometric mean; AM, arithmetic mean; GSD, geometric standard deviation; PbB, blood Pb concentration, |jg/dL; PbA, air Pb concentration, |jg/m3
       January 2013
3-12
Draft - Do Not Quote or Cite

-------
 1           In the 2008 review, in addition to considering the evidence presented in the published
 2    literature and that reviewed in the 1986 CD, we also considered air-to-blood ratios derived from
 3    the exposure assessment (73 FR 66974; 2007 REA, section 5.2.5.2).  In the exposure assessment
 4    (summarized in section 3.4 below), current modeling tools and information on children's activity
 5    patterns, behavior and physiology were used to estimate blood Pb levels associated with
 6    multimedia and multipathway Pb exposure.  The results from the various case studies assessed,
 7    with consideration of the context in which they were derived (e.g., the extent to which the range
 8    of air-related pathways were simulated, and limitations associated with those simulations), and
 9    the multiple sources of uncertainty (see section 3.4.7 below) are also informative to our
10    understanding of air-to-blood ratios. Estimates of air-to-blood ratios for the two REA case
11    studies that represent localized population exposures exhibited an increasing trend across air
12    quality scenarios representing decreasing air concentrations. For example, across the alternative
13    standard levels assessed, which ranged from a calendar quarter average of 1.5  |ig/m3 down to a
14    monthly average of 0.02 |ig/m3, the ratios ranged from  1:2 to 1:9 for the general urban case
15    study, with a similar trend although of generally higher ratio for the primary smelter case study
16    subarea. This pattern of model-derived ratios is generally consistent with the range of ratios
17    obtained from the literature.  We continue to recognize a number of sources of uncertainty
18    associated with these model-derived ratios which may contribute to high  or low biases.10
19           The evidence on the quantitative relationship between air Pb and air-related Pb in blood
20    is now, as in the past, limited by the circumstances (such as those related to Pb exposure) in
21    which the data were collected. Previous reviews have recognized the significant variability in
22    air-to-blood ratios for different populations exposed to Pb through different air-related exposure
23    pathways and at different air and blood levels, with the 1986 CD noting that ratios derived from
24    studies involving the higher blood  and air Pb levels pertaining to workers in lead-related
25    occupations are generally smaller than ratios from studies involving lower blood and air Pb
26    levels (ISA, p.  4-129; 1986 CD, p.  11-99).  Consistent with this observation, slopes in the range
27    of 3 to 5 were estimated for child population datasets assessed in the 1986 CD (ISA, p. 4-129;
             10 For example, the limited number of air-related pathways (inhalation and indoor dust ingestion) simulated
      to change in response to changes in ambient air Pb reductions in these case studies could have implications for the
      air-to-blood ratios. Additionally, with regard to the urban case study, the relationship between dust loading and
      concentration, a key component in the hybrid dust model used in estimating indoor dust Pb levels, is based largely
      on a housing survey dataset reflecting dust Pb in housing constructed before 1980 (as described in the Risk
      Assessment Report, Volume II, Appendix G, Attachment G-l). The use of leaded paint in some housing constructed
      before 1980 contributes some uncertainty due to the potential role of indoor Pb paint (compared to  ambient air Pb)
      in the relationship. The empirically-based ambient air Pb - dust Pb relationships used in the primary Pb smelter
      (subarea) case study may contribute to a potential for the ratios from this case study to more fully capture the impact
      of changes in ambient air Pb on indoor dust Pb, and consequently on blood Pb. Some have suggested, however, that
      the regression used  may not accurately reflect the temporal relationship between reductions in ambient air Pb and
      indoor dust Pb and as a result may overestimate the dust Pb reduction per ambient air Pb reduction, thus contributing
      a potential high bias to the air-to-blood Pb ratios.
         January 2013                            3-13             Draft - Do Not Quote or Cite

-------
 1    1986 CD p.  11-100; Brunekreef,  1984; Tripathi et al., 2004). Additional studies considered in
 2    the last review and those assessed in the ISA provide evidence of ratios above this older range
 3    (ISA, p. 4-130).  For example a ratio of 6.5-7 is indicated by the study by Hilts (2003), one of the
 4    few studies that evaluate the air Pb-blood Pb relationship in conditions that are closer to the
 5    current state in the U.S. ( ISA, p. 4-130). We additionally note the variety of factors identified in
 6    the ISA that may potentially affect estimates of various ratios (including potentially coincident
 7    reductions in nonair Pb sources during the course of the studies), and for which a lack of
 8    complete information may preclude any adjustment of estimates to account for their role (ISA,
 9    section 4.5).
10           In summary, as at the time of the last review of the NAAQS for Pb, the currently
11    available evidence includes estimates of air-to-blood ratios, both empirically- and model-derived,
12    with associated limitations and related uncertainties. These limitations and uncertainties, which
13    are summarized here and also noted in the ISA, usually include uncertainty associated with
14    reductions in other Pb sources during study period.  The limited amount of new information
15    available in this review has not appreciably altered the scientific conclusions reached in the last
16    review regarding relationships between Pb in ambient air and Pb in children's blood or with
17    regard to the range of ratios.  The currently available evidence continues to indicate ratios
18    relevant to the population of young children in the U.S. today, reflecting multiple air-related
19    pathways in addition to inhalation, to be generally consistent with the approximate range of 1:5
20    to 1:10 given particular attention in the 2008 NAAQS decision, including the "generally central
21    estimate" of 1:7 (73 FR 67002, 67004; ISA, pp. 4-129 to 4-130).

22    3.2  NATURE OF EFFECTS
23           Lead has been demonstrated to exert a broad array of deleterious effects on multiple
24    organ systems as described in the assessment of the evidence available in this review and
25    consistent with conclusions of past CDs (ISA, section 2.6; 2006  CD, section 8.4.1).  A sizeable
26    number of studies on Pb health effects are newly available in this review and are critically
27    assessed in the ISA as part of the  full body of evidence.  The newly available evidence reaffirms
28    conclusions  on the broad array of effects recognized for Pb in the last review (see ISA, section
29    2.10). Consistent with those  conclusions, the ISA determines that causal relationships11 exist for
             11 Since the last Pb NAAQS review, the IS As which have replaced CDs in documenting each review of the
      scientific evidence (or air quality criteria) employ a systematic framework for weighing the evidence and describing
      associated conclusions with regard to causality, using established descriptors ("causal" relationship with relevant
      exposure, "likely" to be causal, evidence is "suggestive" of causality, "inadequate" evidence to infer causality, "not
      likely") (ISA, Preamble).
         January 2013                            3-14             Draft - Do Not Quote or Cite

-------
 1    Pb12 with effects on the nervous system (cognitive function decrements and attention-related
 2    behavioral problems in children), hematological system (altered heme synthesis and decreased
 3    red blood cell survival and function), and cardiovascular system (hypertension and coronary
 4    heart disease), and on reproduction and development (postnatal development and male
 5    reproductive function) (ISA, table 2-2). Additionally, the ISA describes relationships between
 6    Pb and effects on kidney and immune system function, and with cancer13 as likely causal14 (ISA,
 7    table 2-2, sections 2.6.3, 2.6.4 and 2.6.7).
 8           In some categories of health effects, there is newly available evidence regarding some
 9    aspects of the effects described in the last review or that strengthens our conclusions regarding
10    aspects of Pb toxicity on a particular physiological system. Among the nervous system effects of
11    Pb, the newly available evidence is consistent with conclusions in the previous review which
12    recognized that "[t]he neurotoxic effects of Pb exposure are among those most studied and most
13    extensively documented among human population groups" (2006 CD, p. 8-25).  Nervous system
14    effects that receive prominence in the current review, as in previous reviews, include those
15    affecting cognitive function and behavior in children (ISA, section 5.3), with conclusions that are
16    consistent with findings of the last review.
17           Across the broad array of Pb effects for systems and processes other than the nervous
18    system, the evidence base has been augmented with additional epidemiological investigations in
19    a number of areas, including developmental outcomes, such as puberty onset, and adult outcomes
20    related to cardiovascular function, for which several large cohorts have been analyzed (ISA,
21    Table 2-8 and sections 5.4 and 5.8). Conclusions on these other systems and processes are
22    consistent with conclusions reached in the last review, while also extending our conclusions on
23    some aspects of these effects. For example, evidence in this review for the cardiovascular
24    system includes information on the role of interactions of cumulative  Pb exposure with other
25    factors such as stress in contributing to hypertension, and on a role for Pb in contributing to
26    coronary heart disease (ISA, section 5.4 and Table 2-8).
             12 In determining that a causal relationship exists for Pb with specific health effects, EPA has concluded
      that "[e]vidence is sufficient to conclude that there is a causal relationship with relevant pollutant exposures (i.e.,
      doses or exposures generally within one to two orders of magnitude of current levels)" (ISA, p. Ixiv)
             13 Lead has been classified as a probable human carcinogen by the International Agency for Research on
      Cancer, based mainly on sufficient animal evidence, and as reasonably anticipated to be a human carcinogen by the
      U.S. National Toxicology Program (ISA, section 5.10). In this assessment, EPA concludes that there is a likely
      causal relationship between Pb exposure and cancer, based primarily on animal studies, with limited human
      evidence of suggestive associations (ISA, section 5.10.6).
             14 In determining a likely causal relationship exists for Pb with specific health effects, EPA has concluded
      that "[e]vidence is sufficient to conclude that a causal relationship is likely to exist with relevant pollutant
      exposures, but important uncertainties remain" (ISA, p. Ixiv).
         January 2013                            3-15            Draft - Do Not Quote or Cite

-------
 1          Based on the extensive assessment of the full body of evidence available in this review,
 2    the major conclusions drawn by the ISA regarding health effects of Pb in children include the
 3    following (ISA, section 1.3.1).
 4          Multiple epidemiologic studies conducted in diverse populations of children
 5          consistently demonstrate the harmful effects ofPb exposure on IQ,  academic
 6          performance, learning and memory. Epidemiologic studies also demonstrate the
 1          effect ofPb exposure on inattention, impulsivity, and hyper activity  in children.
 8          The evidence in children is supported by findings in animal studies demonstrating
 9          analogous effects and biological plausibility at relevant exposure levels. Evidence
10          suggests that some Pb-related cognitive effects may not be reversible and that
11          neurodevelopmental effects ofPb may persist into adulthood (section 2.9.4).
12          Lead exposure also causes hematologic effects (such as effects on blood cells or
13          blood producing organs) in children and is associated with an increased risk of
14          internalizing behaviors (e.g., withdrawn behavior and depressive symptoms),
15          sensory and motor function decrements, atopic and inflammatory conditions (e.g.,
16          asthma and allergy) in children, as well as misconduct in older children and
17          young adults. Uncertainties arising from the lack of information about the specific
18          Pb-exposure histories which contribute to observed blood Pb levels are greater in
19          adults and older children than in young children (section 2.9.5).
20
21          Based on the extensive assessment of the full body of evidence available in this review,
22    the major conclusions drawn by the ISA regarding health effects of Pb in adults include the
23    following (ISA, section 1.3.1).
24          A large body of evidence from both epidemiologic studies of adults and
25          experimental studies in animals demonstrates the effect of long-term Pb exposure
26          on increased blood pressure (BP) and hypertension (Section 2.6.2). In addition to
27          its effect on BP, Pb exposure leads to coronary heart disease and death from
28          cardiovascular causes and is likely to cause cognitive function decrements,
29          symptoms of depression and anxiety, reduced kidney function, and immune effects
30          in adults. The extent to which the effects ofPb on the cardiovascular system are
31          reversible is not well characterized.  It is also important to note that the
32          frequency, timing, level and duration ofPb exposure causing the effects observed
3 3          in adults has not been pinpointed, and higher past exposures may well have
34          contributed to the development of health effects measured later in life.
35
36          As in prior reviews of the Pb NAAQS, this review is focused on those effects most
37    pertinent to ambient air Pb exposures. Given the reductions in ambient air Pb levels over the
38    past decades, these effects are generally those associated with the lowest levels of Pb exposure
39    that have been evaluated.  Additionally, we recognize the limitations on our ability to draw
40    conclusions regarding the exposure conditions contributing to the findings  from epidemiological
41    analyses of blood Pb levels in populations of older children and adults, particularly in light of
42    their history of higher Pb exposures.  In the last review, while recognizing  the range of health


         January 2013                          3-16             Draft - Do Not Quote or Cite

-------
 1    effects in variously aged populations related to Pb exposure, we focused on the health effects for
 2    which the evidence was strongest with regard to relationships with the lowest exposure levels,
 3    neurocognitive effects in young children. The policy-relevant questions on health effects for this
 4    review (identified in the IRP) were framed in recognition of the conclusions of the last review.
 5    Our consideration of the health effects evidence in this review is framed by policy-relevant
 6    questions building on those identified in the IRP.
 7         •   To what extent is there new scientific evidence available to improve our
 8             understanding of the health effects associated with various time periods of Pb
 9             exposures at various stages of life?
10          As in the last review, we base our current understanding of health effects associated with
11    different Pb exposure circumstances at various stages of life on the full body of available
12    evidence which includes epidemiological studies of health effects associated with population Pb
13    biomarker levels as well as laboratory animal  studies in which the effects of different exposures
14    on different lifestages are assessed under controlled conditions. The epidemiological evidence is
15    overwhelmingly comprised of studies that rely on blood Pb for the exposure metric, with the
16    remainder largely including a focus on bone Pb. Because these metrics reflect Pb in the body
17    (e.g., as compared to Pb exposure concentrations) and, in the case of blood Pb, reflect Pb
18    available  for distribution to target sites, they strengthen the evidence base for purposes of
19    drawing causal conclusions with regard to Pb  generally. The complexity of Pb exposure
20    pathways and internal dosimetry tends to limit the extent to which these types of studies  inform
21    our more  specific understanding of the Pb exposure circumstances (e.g., timing, duration,
22    frequency and magnitude) eliciting the various effects.
23          The specific exposure circumstances, including timing during the lifetime, that contribute
24    to the blood Pb (or bone Pb) measurements with which  associations have been analyzed in
25    epidemiological studies are unknown. This is particularly the case with regard to the
26    contributing role of recent exposures for which uncertainty is greater in adults and older children
27    than in younger children (ISA, section 1.4). For example, a critical aspect of much of the
28    epidemiological evidence, particularly that focused on older adults in the U.S. today, is the
29    backdrop of generally declining environmental Pb exposure that is common across many study
30    populations. An additional complication pertaining to older children and younger adults is the
31    common behaviors of younger children (e.g., hand to mouth contact) which generally contribute
32    to relatively greater environmental exposures  earlier in life (ISA, sections 4.1.1, 6.2.1).  Such
33    exposure  histories complicate our ability to draw conclusions regarding critical time periods and
         January 2013                          3-17            Draft - Do Not Quote or Cite

-------
 1    lifestages for Pb exposures eliciting the effects with which associations with Pb biomarkers have
 2    been observed in populations of adults and older children (e.g., ISA, section 2.9.5).15
 3           As at the time of the last review, assessment of the full evidence base, including evidence
 4    newly available in this review, demonstrates that Pb exposure prenatally and also in early
 5    childhood can contribute to neurocognitive impacts in childhood, with evidence also indicating
 6    the potential for effects persisting into adulthood (ISA, sections 2.9.5, 2.9.6, and 2.10).  In
 7    addition to the observed associations of prenatal and childhood blood Pb with effects at various
 8    ages in childhood, there is also evidence of lead-related cognitive function effects in non-
 9    occupationally exposed adults (ISA, section 5.3.12).  This includes evidence of associations of
10    such effects in adulthood with childhood blood Pb levels and, in other cohorts, with concurrent
11    (adult) blood Pb levels (ISA, sections 5.3.2.1,  5.3.2.7 and 5.3.12).  As the studies finding
12    associations of adult effects with childhood blood Pb  levels did not examine adult blood Pb
13    levels, the relative influence of adult Pb exposure cannot be ascertained, and a corresponding
14    lack of early life exposure or biomarker measurements for the latter studies limit our ability to
15    draw conclusions regarding specific Pb exposure circumstances eliciting the observed effects
16    (5.3.12).  Findings of stronger associations for adult neurocognitive effects with bone Pb,
17    however,  indicate the role of historical or cumulative exposures for those effects (ISA, section
18    5.3).
19           Given the relatively short exposure histories of young children, there is relatively
20    reduced uncertainty regarding the lifestages in which exposures contribute to effects for
21    associations of early childhood effects with early childhood blood Pb levels  (ISA, sections  2.9.5
22    and 5.3.12). In considering our understanding of the relative impact on neurocognitive function
23    of additional Pb exposure of children by school age or later we recognize increasing uncertainty
24    associated with limitations of the currently available evidence, including epidemiological cohorts
25    with generally similar temporal patterns of exposure.  We take note, additionally, of evidence
26    from experimental animal  studies and a small  body of epidemiologic studies that indicates that
27    Pb exposures  during different lifestages can induce cognitive impairments.  The limited
28    epidemiological evidence is of populations with blood Pb levels that are not strongly correlated
29    over time and that can, accordingly, address the issue of the role of exposure subsequent to the
             15 The evidence from experimental animal studies can be informative with regard to key aspects of
      exposure circumstances in eliciting specific effects. Accordingly, for health effects where the animal evidence base
      is extensive with regard to investigation of different patterns of exposure, it informs our interpretation of the
      epidemiological evidence. For example, the animal evidence base is particularly extensive with regard to Pb effects
      on blood pressure, demonstrating the etiologically-relevant role of long-term exposure (ISA, section 5.4.1).  This
      finding then informs consideration of epidemiological studies of adult populations for whom historical exposures
      were likely more substantial than concurrent ones to suggest that the observed effects may be related to the past
      exposure (ISA, section 5.4.1). For other health effects, the animal evidence base may be less extensive with regard
      to the role of specific exposure circumstances and thus less informative with regard to key exposure aspects.
          January 2013                           3-18             Draft - Do Not Quote or Cite

-------
 1    earliest lifestages (ISA, section 5.3.12; Hornung et al., 2009). Some animal evidence
 2    demonstrates impaired learning with infancy only, from infancy into adulthood, and postinfancy
 3    only Pb exposure (Rice, 1990; Rice and Gilbert 1990; Rice, 1992). Further, evidence that Pb
 4    exposure presents a risk during different lifestages is also consistent with our broader
 5    understanding that nervous system development continues throughout childhood. The limited
 6    analyses of this issue that are newly available in this review do not appreciably change our
 7    understanding or conclusions on this from those of the prior review (ISA, section 5.3.12).
 8          As in the last review, there is also substantial evidence of other neurobehavioral effects in
 9    children,  such as attention-related behavioral problems, conduct problems or internalizing
10    behaviors. The evidence for many of these endpoints, as with neurocognitive effects, also
11    includes associations of effects at various ages in childhood and, for some effects, into
12    adulthood, with blood Pb  levels reflective of several different lifestages (e.g., prenatal and
13    several different ages in childhood) (ISA, sections 5.3.3 and 5.3.4). There is similar or relatively
14    less extensive evidence to inform our understanding of such effects associated with specific time
15    periods of exposure at specific lifestages than is the case for effects on cognitive function.
16          Across the range of Pb effects on physiological  systems and processes other than the
17    nervous system, the full body of evidence on etiologically relevant circumstances of Pb exposure
18    eliciting increases in blood pressure and hypertension is somewhat more informative than is the
19    case with regard to many  other effects. In the case of lead-induced increases in blood pressure,
20    the evidence indicates an importance of long-term exposure (ISA, sections 2.6.2 and 5.4.7.1).
21          In summary, as in the last review, we continue to recognize a number of uncertainties
22    regarding the circumstances of Pb exposure, including timing or lifestages, eliciting specific
23    health effects.  Consideration of the evidence newly available in this review has not appreciably
24    changed our understanding on this topic. Across the full evidence base, the effects for which our
25    understanding of relevant exposure circumstances is greatest are neurocognitive effects in
26    children,  and to a lesser degree, hypertension and increased blood pressure.  Thus, we continue
27    to recognize and give particular attention to the role of Pb exposures relatively early in childhood
28    in contributing to neurocognitive effects which may persist into adulthood.
29         •   At what levels of Pb exposure do health effects of concern occur?
30         •   Is there evidence of effects at exposure levels lower than previously observed and
31             what are important uncertainties in that evidence?
32          In considering the question posed here, we recognize, as discussed in section 3.1 above,
33    that the epidemiological evidence base for our consideration in this review, as in the past,
34    includes substantial focus on internal biomarkers of exposure, such as blood Pb, with relatively
35    less information specific to exposure levels, including those derived from air-related pathways.

         January 2013                           3-19             Draft - Do Not Quote or Cite

-------
 1    Given that blood and bone Pb are integrated markers of aggregate exposure across all sources
 2    and exposure pathways, our interpretation of studies relying on them is informed by what is
 3    known regarding the historical context and exposure circumstances of the study populations. For
 4    example, a critical aspect of much of the epidemiological evidence is the backdrop of generally
 5    declining Pb exposure over the past several decades. Thus as a generality, recent
 6    epidemiological studies of populations with similar characteristics as those studied in the past
 7    tend to involve lower overall Pb  exposures and accordingly lower blood Pb levels (e.g., ISA,
 8    sections 3.5 and 4.4.1; 2006 CD, section 3.4). This has been of particular note in the evidence of
 9    blood Pb associations with nervous system effects, particularly impacts on cognitive function in
10    children, for which we have seen associations with progressively lower childhood blood Pb
11    levels across past reviews (ISA, section 5.3.13; 1986 CD; USEPA, 1990; 2006 CD; 73 FR
12    66976).
13           The evidence currently available with regard to the magnitude of blood Pb levels
14    associated with neurocognitive effects in children is generally consistent with that available in
15    the review completed in 2008. Nervous system effects in children, specifically effects on
16    cognitive function, continue to be the effects that are best substantiated as occurring at the lowest
17    blood Pb concentrations (ISA, sect 1.5).  Associations  of blood Pb with effects on cognitive
18    function measures in children have been reported in many studies across a range of childhood
19    blood Pb levels, including study  group (mean/median) levels ranging down to 2 |ig/dL (e.g., ISA,
20    sections 1.5 and 5.3.2).16 Studies in which such findings were reported for childhood study
21    group blood Pb levels below 5 |ig/dL are summarized in Table 3-2.17 In recognition of the
22    influence of age on blood Pb levels, the analyses in Table 3-2 are listed in order of age at which
23    the blood Pb measurements were taken. Findings for studies newly available in this review are
24    indicated in bold text.
             16 The value of 2 ug/dL refers to the regression analysis of blood Pb and end-of-grade test scores, in which
      blood Pb was represented by categories for integer values of blood Pb from 1 ug/dL to 9 and >10 ug/dL from large
      statewide database.  A significant effect estimate was reported for test scores with all blood Pb categories in
      comparison to the reference category (1 ug/dL), which included results at and below the limit of detection. Mean
      levels are not provided for any of the categories (Miranda., et al., 2009).
             17 Two additional studies (both newly available) that report such associations with blood Pb levels for
      which the mean is equal to 5.0 ug/dL.  The first is a study of 506 children in Detroit, MI (born, 1982-1984) at age 7
      years which observed a significant negative association with concurrent blood Pb levels for which the mean equals
      5.0 ug/dL (8.9% of childen with concurrent blood Pb above 10 ug/dL) (ISA, sections 5.3.2.1 and 5.3.3.5; Chiodo et
      al., 2007).  The second study focuses on 174 of the Rochester cohort children at age 6 years, reporting significant
      negative associations with FSIQ for four different blood Pb metrics: concurrent (mean = 5.0 ug/dL), lifetime
      average (mean=7.2 ug/dL), infancy average (mean =7.1 ug/dL) and peak (mean =11.4 ug/dL) (Jusko et al., 2008).
         January 2013                            3-20             Draft - Do Not Quote or Cite

-------
1   Table 3-2.   Associations with neurocognitive function measures in analyses with child study group blood lead levels <5 ug/dL.
Measure*
Study Group Dataset Description
Blood Pb Levels
Age
Mean6
(Mg/dL)
Range5
(Mg/dL)

N
Additional Information on Analyses
cStudies discussed in ISA (section 5.3.2) with findings of effects on neurocognitive measures reported for childhood study group PbB < 5 jjg/dL
(ordered by age of blood Pb measurements)
FSIQ
BSID/MDI
AcadPerf
AcadPerf
FSIQ
FSIQ
FSIQ
LM, EF
AcadPerf
FSIQ
LM
AcadPerf
LM
AcadPerf
Boston, prospective, age 5 yr
Subgroup with peak PbB <10 pg/dL
Bellinger and Needleman 2003
Mexico City, age 24 mo,Tellez-Rojo et al., 2006
Subgroup with PbB<5 pg/dL
Full dataset
North Carolina, 4th graders
4th grade reading scores, Full dataset
Miranda et al., 2009
Avon, United Kingdom, age 7-8 yr
standard assessment scores, Full dataset,
Chandramouli et al., 2009
Rochester, prospective cohort, age 5 yr
Subgroup with peak PbB <10 pg/dL
Canfield etal., 2003
Pooled International, age 6-10 yr
Subgroup with peak PbB <7.5 pg/dL
Lanphear et al 2005 E
New England, 2 areas, age 6-10, ,
Full dataset,
Surkan et al., 2007
Korea, 4 areas, age 8-11 yr
Full dataset
Kim et al., 2009
NHANES 111(1988-1994)
Lanphear etal., 2000
NHANES III (1991-1994)
Krieg etal., 2010
24 mo
24 mo
9-36 mo
30 mo
Syr
5-7 yr
6-1 Oyr
8-1 1yr
6-1 6 yr
12-1 6 yr
3.8
2.9
4.3
4.8
4.22
3.32
3.24
2.3
1.73
1.9
1.95
1-9.3
0.8-4.9
0.8-9.8
1-16
21%<2
52% 2-5
21% 5-10
0.5-8.4
(LOD=1)
0.9-7.4
1-10
0.4-4.9


48
193
294
57,678
488°
71
103
389
261
4,853
766-80
Regression, PbB as continuous variable; statistically significant negative
association
Regression, PbB as continuous variable; statistically significant negative
association
Linear and quantal regressions with integer PbB as categorical variable (PbB-1
reference category, includes LOD). Linear analysis: statistically significant effect
for all comparison categories. Quantal analysis: statistically significant effect in all
reading score quantiles for PbB categories greater than integer PbB=3; largest
effect in lower quantiles. Means not reported for PbB categories.
Regressions with PbB as continuous and categorical variables: continuous
analysis (statistically significant negative association); categorical (significantly
reduced scores for 5-10 |jg/dL vs reference group [ 0-2 M9/dL])
Regression, PbB as continuous variable; statistically significant negative
association
Regression, PbB as continuous variable; statistically significant negative
association
Regression, PbB grouped into categories; statistically significant negative
association for high subgroup (5-10 ug/dL, n=32) compared to reference PbB
subgroup (1-2 ug/dL, n=286).
Regression, PbB analyzed as quartiles; statistically significant difference among
quartiles; statistically significant negative association in continuous analysis of full
dataset and high blood manganese group
Continuous (unadjusted) and categorical (adjusted) analysis; significant difference
among PbB quartiles (< 1, 1.1-1.9, 2.0-3.0, >3.0 ug/dL), and neg assoc PbB <5.0.
Regression analysis; PbB as continuous variable; statistically significant negative
association.
A - FSIQ = Full Scale Intelligence Quotient; BSD = Bayley Scales of Development; MDI= Mental Development Index; LM=Learning and Memory; EF= Executive Function; AcadPerf= Academic Performance.
B - Blood Pb level (PbB) information provided here is in some cases augmented by study authors (Bellinger, 2008; Canfield, 2008a,b; Hornung, 2008a,b; Tellez-Rojo, 2008).
C - Bolded measures and studies are newly available in this review.
D - In practice, 337-425 cases were included in analysis (Chandramouli et al., 2009).
E - Blood Pb measurements of subgroup with peak PbB <7. 5 |ig/dl_ comprised of 13.6% age 5 (Boston, Cleveland), 67.0% age 6 (Rochester), 19.4% age 7 (Yugoslavia, Mexico, Cincinnati). IQ assessed at age 5 for the
single member of this subgroup from Cleveland cohort. This analysis includes blood Pb data from Rochester and Boston cohorts although for different ages (6 and 5 years, respectively) than the ages analyzed in Canfield
et al 2003 and Bellinger and Needleman 2003. For full dataset analysis (n=1333), IQ assessed at ages 5-10 yr, median blood Pb (at ages 5-7 yr) of 9.7 pg/dL and 5th -95th percentile of 2.5-33.2 pg/dL.
                       January 2013
5-21
Draft - Do Not Quote or Cite

-------
 1           Among the analyses of lowest study group blood Pb levels at the youngest ages in Table
 2    3-2 are analyses available in the last review of Pb associations with neurocognitive decrement in
 3    study groups with mean levels on the order of 3-4 |ig/dL in children aged 24 months, or ranging
 4    from 5 to 7 years (73 FR 66978-66969; ISA, section 5.3.2.2; Bellinger and Needlemann, 2003;
 5    Canfield et al., 2003; Lanphear et al., 2005; Tellez-Rojo et al., 2006; Bellinger, 2008; Canfield,
 6    2008; Hornung, 2008; Tellez-Rojo, 2008).18 Newly available in this review are two studies
 7    reporting association of blood Pb levels prior to three years of age with academic performance
 8    on standardized tests in primary school; mean blood Pb levels in these studies were 4.2 and 4.8
 9    |ig/dL (ISA, section 5.3.2.5;  Chandramouli et al., 2009; Miranda et al., 2009). One of these two
10    studies, which represented integer blood Pb levels as categorical variables, indicated a small
11    effect on end-of-grade reading score of blood Pb levels as low as 2 |ig/dL, after adjustment for
12    age of measurement, race, sex, enrolled in free or reduced lunch program, parental education,
13    and school type (Miranda et  al., 2009).
14           In a newly available study of blood Pb levels at primary school age, a significant
15    association of blood Pb in children aged 8-11 years and concurrently measured full scale IQ
16    (FSIQ) was reported for a cross-sectional cohort in Korea with a mean blood Pb level of 1.7
17    |ig/dL and range of 0.43-4.91 |ig/dL (Kim et al., 2009).19 In considering the blood Pb levels in
18    this study, we note that blood Pb levels in children aged 8-11 are generally lower than those in
19    pre-school children, for reasons related to behavioral and other factors (ISA, sections 4.3.5, 4.4.1
20    and 6.2.1.1).20 It is likely that the blood Pb levels of this study group at earlier ages, e.g., prior to
21    school entry, were higher and the available information does not provide a basis to judge whether
22    the blood Pb levels in this study represent lower exposure levels than those experienced by the
23    younger study groups. In still older children, a large cross-sectional investigation of blood Pb
24    association with effects on memory and learning, that was available in the last review was
25    focused on children  aged 6-16 years, born during 1972-1988, with a mean blood Pb of 1.9
26    |ig/dL.  A study newly available in this review, focused on a subset of the earlier study cohort
27    (ages 12-16, born  during 1975-1982),  also reports a significant negative association of blood Pb
             18 The tests for cognitive function in these studies include age-appropriate Wechsler intelligence tests
      (Lanphear et al., 2005; Bellinger and Needleman, 2003), the Stanford-Binet intelligence test (Canfield et al., 2003),
      and the Bayley Scales of Infant Development (Tellez-Rojo et al., 2006). The Wechsler and Stanford-Binet tests are
      widely used to assess neurocognitive function in children and adults, however, these tests are not appropriate for
      children under age three.  For such children, studies generally use the age-appropriate Bayley Scales of Infant
      Development as a measure of cognitive development.
             19 Limitations of this study included a lack of consideration of potential confounding by parental caregiving
      quality or IQ (ISA, Table 5-3).
             20 This study also investigated the potential role of manganese (Mn) in the blood Pb associated effects.
      When the cohort was subdivided based on Mn blood Pb levels, using the median Mn level (14 ug/L) as the break
      point, the significant association of intelligence quotient (IQ) with Pb persisted in the higher Mn group but was no
      longer significant in the lower Mn group (ISA, section 5.3.2). Separate analysis of the full study group found a
      significant negative association of IQ with blood Mn (Kim et al., 2009).
         January 2013                           3-22             Draft - Do Not Quote or Cite

-------
 1    with learning and memory test results with mean blood Pb levels of approximately 2 |ig/dL (ISA,
 2    section 5.3.2.3; Lanphear et al., 2000; Krieg et al., 2010).  In considering these study findings
 3    with regard to the question of exposure levels eliciting effects, we recognize, however, that blood
 4    Pb levels are, in general, lower among teenagers than young children and also that, for these
 5    subjects specifically, the magnitude of blood Pb levels during the earlier childhood (e.g., pre-
 6    school ages) was much higher.  For example, the mean blood Pb levels for the  1-5 year old age
 7    group in the NHANES 1976-80 sample was 15 |ig/dL, declining to 3.6 ng/dL in the NHANES
 8    1988-1991 sample (Pirkle et al., 1994; ISA, section 4.4.1).
 9          With regard to other nervous system effects in children, the evidence base at lower blood
10    Pb levels is somewhat extended since the last review with regard to the evidence on Pb and
11    attention-related behavioral problems, such as inattention, impulsivity, hyperactivity and
12    attention deficit hyperactivity disorder (ISA, section 5.3.3 and table 5-17). Several newly
13    available studies investigating the role of blood Pb levels in older children (primary school age
14    and older) have reported significant associations of attention-related behavior or conduct
15    problem measures with concurrent blood Pb levels, with mean levels generally on the order of 5
16    ng/dL or higher (ISA, section 5.3.3). One exception is the newly available cross-sectional,
17    categorical analysis of the NHANES 2001-2004 sample of children aged 8-15 years, which
18    found higher prevalence of conduct disorder in the subgroup with concurrent blood Pb levels of
19    0.8-1.0 ng/dL as compared to the <0.8 ng/dL group  (ISA, section 5.3.4 and Table 5-12). As
20    noted above, we recognize that many of these children, born between  1986 and 1996, are likely
21    to have had much higher Pb exposures (and associated blood Pb levels) in their earlier years than
22    those commonly experienced by young children today, thus precluding a conclusion regarding
23    evidence of effects associated with lower exposure levels than provided by evidence previously
24    available.
25          As summarized earlier in this section, blood Pb has been associated with a range of health
26    effects on multiple organ systems or processes.  As is the case for studies of nervous system
27    effects in children, newly available studies of other effects in child and adult cohorts include
28    cohorts with similar or somewhat lower mean blood levels than in previously available studies.
29    Categories of effects for which a causal relationship has been concluded in the  ISA and for
30    which there are a few newly available epidemiological studies indicating blood Pb associations
31    with effects in study groups with somewhat lower blood Pb levels than previously available for
32    these effects include effects on development (delayed puberty  onset) and reproduction (male
33    reproductive function) and on the cardiovascular system (hypertension) (ISA, sections 5.4 and
34    5.8; 2006 CD, sections 6.5 and 6.6). With regard to the former category, study groups in the
35    newly available studies include groups comprised of older children ranging up  to age 18 years,
36    for which there is increased uncertainty regarding historical exposures and their role in the

         January 2013                           3-23            Draft - Do Not Quote or Cite

-------
 1    observed effects.21 An additional factor that handicaps our consideration of exposure levels
 2    associated with these findings is the appreciable uncertainty associated with our understanding of
 3    Pb biokinetics during this lifestage (ISA, sections 4.3, 5.8.6). The evidence newly available for
 4    Pb relationships with cardiovascular effects in adults include some studies with somewhat lower
 5    blood Pb levels than in the last review. The long exposure histories of these cohorts, as well as
 6    the generally higher Pb exposures of the past complicate conclusions regarding exposure levels
 7    that may be eliciting observed effects (ISA, sections 5.4.2.4  and S.4.7).22
 8           In summary, our conclusions regarding exposure levels at which Pb health effects occur,
 9    and particularly with regard to  such levels that might be common in the U.S. today, are
10    complicated now,  as in the last review, by several factors.  These  factors  include the scarcity of
11    information in epidemiological studies on cohort exposure histories, as well as by the backdrop
12    of higher past exposure levels which frame the history of some study cohorts.  Recognizing the
13    complexity, as well as the potential role of higher exposure levels in the past, we  continue to
14    focus our consideration of this  question on the evidence of effects in young children for which
15    our understanding of exposure history is less uncertain.23  Within  this evidence base, we
16    recognize the lowest study group blood Pb levels to be associated with effects on cognitive
17    function measures, indicating that to be the most sensitive endpoint.  As described above, and
18    summarized in Table 3-2, the evidence available in this review is  generally consistent with that
19    available in the last review with regard to blood Pb levels  at which such effects had been
20    reported (ISA, section 5.3.2; 2006 CD, section 8.4.2.1; 73  FR 66976-66979). As blood Pb levels
21    are a reflection of exposure history, particularly in early childhood (ISA,  section 4.3.2), we
22    conclude, by extension, that the currently available evidence does not indicate Pb effects at
23    exposure levels appreciably lower than recognized in the last review.
24           We additionally note that, as in the last review, a threshold blood  Pb level with which
25    nervous system effects, and specifically neurodevelopmental effects, occur in young children
26    cannot be discerned from the currently available studies (ISA, sections 2.9.4 and  5.3.13).
27    Epidemiological analyses have reported blood Pb associations with neurocognitive effects (FSIQ
28    or BSID MDI24) in young child population subgroups, which included individual blood Pb
             21 Several of these studies involve NHANES III cohorts for which early childhood exposures were
      generally much higher than those common in the U.S. today (ISA, section 5.8.6).
             22 Studies from the late 1960s and 1970s suggest that adult blood Pb levels during that period ranged from
      roughly 10 to 30 ug/m3 (ISA, section 5.4.1).
             23 In focusing on effects associated with blood Pb levels in early childhood, however, we additionally
      recognize the evidence across categories of effects that relate to blood Pb levels in older child study groups (for
      which early childhoood exposure may have had an influence) which provides additional support to an emphasis
      nervous system effects (ISA, sections 5.3, 5.4, 5.5, 5.6, 5.7, 5.8).
             24 Bayley Scales of Infant Development, Mental Development Index. The Bayley MDI is a well-
      standardized and widely used assessment measure of infant cognitive development. Scores earlier than 24 months
      are not necessarily correlated with later FSIQ scores in children with normal development (ISA, section 5.3.16.1).
         January 2013                            3-24             Draft - Do Not Quote or Cite

-------
 1    measurements as low as approximately 1 |ig/dL and mean concentrations as low as 2.9 to 3.8
 2    |ig/dL (ISA, section 5.3.13; Bellinger and Needleman, 2003; Bellinger, 2008; Canfield el al.,
 3    2003; Canfield, 2008; Tellez-Rojo et al., 2006; Tellez-Rojo et al., 2008).  As concluded in the
 4    ISA, however, "the current evidence does not preclude the possibility of a threshold for
 5    neurodevelopmental effects in children existing with lower blood levels than those currently
 6    examined" (ISA, section 5.3.13).
 7          Important uncertainties associated with the evidence of effects at low exposure levels are
 8    similar to those recognized in the last review, including the shape of the concentration-response
 9    relationship for effects on neurocognitive function at low blood Pb levels in today's young
10    children.  Also of note is our interpretation of associations between blood Pb levels and effects in
11    epidemiological studies, with which we recognize uncertainty with regard to the specific
12    exposure circumstances (timing, duration, magnitude and frequency) that have elicited the
13    observed effects, as well as uncertainties in relating ambient air concentrations (and associated
14    air-related exposures) to blood Pb levels in early childhood, as discussed in section 3.1 above.
15    We additionally recognize uncertainties associated with conclusions drawn with regard to the
16    nature of the epidemiological associations with blood Pb (e.g., ISA, section 5.3.14), but note that
17    based on consideration of the full body of evidence for neurocognitive effects, the EPA has
18    determined  a causal relationship to exist between relevant blood Pb levels and neurocognitive
19    impacts in children (ISA, section  5.3.16.1).
20         •  To what extent does the newly available evidence alter our understanding of the
21             concentration-response relationship for neurocognitive effects (IQ) with blood Pb
22             levels in young children?
23          Based primarily on studies of FSIQ, the assessment of the currently available studies, as
24    was the case in the last review, continues to recognize a nonlinear relationship between blood Pb
25    and effects on cognitive function, with  a greater incremental effect (greater slope) at lower
26    relative to higher blood Pb levels within the range thus far studied, extending from well above 10
27    ng/dL to below 5 |ig/dL (ISA, section 5.3.13). This was supported by the evidence available in
28    the last review, including the analysis of the large pooled international dataset comprised of
29    blood Pb measurements concurrent with the IQ tests that ranged from 2.5 |ig/dL to 33.2 |ig/dL
30    (Lanphear et al., 2005; Rothenberg and Rothenberg, 2005; ISA, section 5.3.13). The study by
31    Lanphear et al (2005) additionally presented analyses that stratified the dataset based on peak
32    blood Pb levels (e.g., with cutpoints of 7.5 |ig/dL and 10 |ig/dL peak blood Pb)  and found that
33    the coefficients from linear models of the association for IQ with concurrent blood Pb were
34    higher in the lower peak blood Pb level subsets than the higher groups (ISA, section 5.3.13;
35    Lanphear et al., 2005). Stratified analyses of several individual cohorts also observed higher
36    coefficients for blood Pb relationships with measures of neurocognitive function in lower as

         January 2013                          3-25            Draft - Do Not Quote or Cite

-------
1
2
3
4
5
6
7
      compared to higher blood Pb subgroups (ISA, section 5.3.13; Canfield et al., 2003; Bellinger and
      Needleman, 2003; Kordas et al., 2006; Tellez-Rojo et al., 2006). Of these subgroup analyses,
      those involving the lowest mean blood Pb levels and closest to the current mean for U.S.
      preschool children are listed in Table 3-3 below (drawn from Table 3 of the 2008 final
      rulemaking notice [73 FR 67003]). These analyses were important inputs for the evidence-
      based, air-related IQ loss framework which informed decisions on a revised standard in the last
      review (73 FR 67005), discussed in section 4.1.1 below.

      Table 3-3.   Summary of quantitative relationships of IQ and blood Pb for analyses with
                  blood Pb levels closest to those of young children in the U.S. today.
Blood Pb Levels
(MQ/dL)
Geometric
Mean
2.9
3.24
3.32
3.8
Range
(min-max)
0.8-4.9
0.9-7.4
0.5-8.4
1-9.3
Study/Analysis
Tellez-Rojo et al (2006)B, subgroup with concurrent blood Pb <5 (ig/dL
Lanphear et al (2005)°, subgroup with peak blood Pb <7.5 (ig/dL
Canfield et al (2003) c, subgroup with peak blood Pb <10 (ig/dL
Bellinger and Needleman (2003)°, subgroup with peak blood Pb <10 (ig/dL
Median value
Average
Linear
SlopeA
(IQB points
per |jg/dl_)
-1.71
-2.94
-1.79
-1.56
-1.75
A - Average linear slope estimates here are generally for relationship with IQ assessed concurrently with blood Pb measurement. As
exceptions, Bellinger & Needleman (2003) slope is relationship for 10 year old IQ with blood Pb levels at 24 months, and the data for Boston
cohort included in Lanphear et al 2005 slope are relationship for 10 year old IQ with blood Pb levels at 5 years.
B -The slope for Tellez-Rojo et al 2006 is for BSID (MDI), a measure of cognitive development appropriate to study population age (24-mos).
C - The Lanphear et al. (2005) pooled International study also includes blood Pb data from the Rochester and Boston cohorts, although for
different ages (6 and 5 years, respectively) than the ages analyzed in Canfield et al. (2003) and Bellinger and Needleman (2003). Thus, the
ages at the blood Pb measurements used in derivation of the linear slope for the Lanphear et al (2005) subgroup shown here are 5 to? years.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
           Several studies newly available in the current review have, in all but one instance, also
     found a nonlinear blood Pb-cognitive function relationship in nonparametric regression analyses
     of the cohort blood Pb levels analyzed (ISA, section 5.3.13).  These studies, however, used
     statistical approaches that did not produce quantitative results for each blood Pb group (ISA,
     section 5.3.13).  Thus, newly available studies have not extended the range of observation for
     quantitative estimates of this relationship to lower blood Pb levels than those of the previous
     review. The ISA further notes that the potential for nonlinearity has not been examined in detail
     within a lower more narrow range of blood Pb levels than those of the full cohorts thus far
     studied in the currently available evidence base (ISA, section 5.3.13).  Such an  observation in the
     last review supported the consideration of linear slopes with regard to blood Pb levels at and
     below those represented in Table 3-3.  In summary, the newly available evidence does not
     substantively alter our understanding of the concentration-response relationship (including
     quantitative aspects) for neurocognitive impact, such as IQ with blood Pb in young children.
         January 2013
                                               3-26
Draft - Do Not Quote or Cite

-------
 1    3.3   PUBLIC HEALTH IMPLICATIONS AND AT-RISK POPULATIONS
 2           There are several potential public health impacts associated with Pb exposure.  In
 3    recognition of effects causally related to blood Pb levels somewhat near those most recently
 4    reported for today' s population and for which the weight of the evidence is greatest, the potential
 5    public health impacts most prominently recognized in the ISA are population IQ impacts
 6    associated with childhood Pb exposure and prevalence of cardiovascular effects in adults (ISA,
 7    section 2.9.1). With regard to the latter category, as discussed above, the full body of evidence
 8    indicates a role of long-term cumulative exposure,  with uncertainty regarding the specific
 9    exposure circumstances contributing to the effects  in the epidemiological studies which focused
10    largely on adult populations for whom historical Pb exposures were likely  much higher than
11    exposures that commonly occur today (ISA, section 5.4). There is less uncertainty regarding the
12    exposure patterns contributing to the blood Pb levels reported in studies of younger populations
13    (ISA, section 2.10). Accordingly, we focus the discussion of public health implications relevant
14    to this review predominantly on nervous system effects, including IQ decrements, in children.
15           We focus this discussion on IQ in recognition of IQ being a well established, widely
16    recognized and rigorously standardized measure of neurocognitive function, as well as a global
17    measure reflecting the integration of numerous processes (ISA, section 5.3.2; 2006 CD, sections
18    6.2.2 and 8.4.2). We recognize, however, that IQ is one of several measures of cognitive
19    function negatively associated with Pb exposure. Other examples include  other tests of
20    intelligence and cognitive development, and tests of other cognitive abilities, such as learning,
21    memory, and executive functions, as well  as academic performance and achievement (ISA,
22    section 5.3.2). In considering the public health significance of neurocognitive effects of Pb in
23    children, we recognize that, although some may be transient, some effects  may persist into
24    adulthood (ISA, section 2.9.S).25  We also note that deficits in neurodevelopment early in life
25    may have lifetime consequences as "[njeurodevelopmental deficits measured in childhood may
26    set affected children on trajectories more prone toward lower educational attainment and
27    financial well-being" (ISA, section 5.3.15).  Thus,  population groups for which
28    neurodevelopment is affected by  Pb exposure in early childhood are at risk of related impacts on
29    their success later in life.
30           There are important distinctions between population and individual risk such that "small
31    shifts in population means are often significant from a public health perspective" (ISA, p.  2-63).
32    For example, if lead-related decrements are manifested uniformly across the range of IQ scores
             25 The ISA states that the "persistence of effects appears to depend on the duration and window of exposure
      as well as other factors that may affect an individual's ability to recover from an insult", with some evidence of
      greater recovery in children reared in households with more optimal caregiving characteristics and low concurrent
      blood Pb levels (ISA, section 2.9.5; Bellinger et al., 1990).
         January 2013                          3-27             Draft - Do Not Quote or Cite

-------
 1    in a population, "a small shift in the population mean IQ may result in a substantial increase in
 2    the number of individuals functioning in the low range of the IQ distribution, which is associated
 3    with increased risk of educational, vocational, and social failure" as well as a decrease in the size
 4    of the subgroup achieving very high scores (ISA, section 2.9.1).
 5           In the discussion below, we use the term at-risk populations to recognize populations that
 6    have a greater likelihood of experiencing lead-related health effects, i.e., groups with
 7    characteristics that contribute to an increased risk of Pb-related health effects. These populations
 8    are also sometimes referred to as sensitive groups, as in section 1.2.1  above.  This increased
 9    likelihood of lead-related effects can result from many factors, including lifestage or age, gender,
10    race or ethnicity, diet, pre-existing disease state, or increased exposure (ISA, chapter 6).
11    Accordingly, in identifying factors that increase risk of lead-related health effects, there has been
12    consideration of evidence regarding factors contributing to increased susceptibility  (i.e.,
13    physiological or intrinsic factors contributing to a greater response for the same exposure), and
14    those contributing to increased exposure (including that resulting from behavior leading to
15    increased contact with contaminated media). As noted in the ISA, "definitions of susceptibility
16    and vulnerability vary across studies, but in most instances 'susceptibility' refers  to biological or
17    intrinsic factors (e.g., age, sex) while 'vulnerability' refers to nonbiological or extrinsic factors
18    (e.g., socioeconomic status  [SES])"  and the terms "at-risk" and "sensitive" populations have in
19    various instances  been used to encompass these concepts more generally (ISA, p. 6-1). Although
20    we emphasize the term "at-risk", we rely on the other terms in particular instances below; in so
21    doing,  our usage is consistent with these definitions.
22           Factors that increase risk of lead-related effects include, among others, behavioral
23    physiological factors. A behavioral factor of great impact on Pb exposure is the incidence of
24    hand-to-mouth activity that is prevalent in very young children and by which they transfer Pb in
25    settled particles to their mouths (ISA,  sections 4.7.1 and 6.2.1.1).  Physiological factors include
26    both conditions contributing to a group's increased risk of effects  at a given blood Pb level, and
27    those that contribute to blood Pb levels higher than those otherwise associated with a given Pb
28    exposure (ISA, sections 6.3 and 6.1, respectively).  We also considered evidence  encompassing
29    situations of elevated exposure, such as residing in  old housing with Pb-containing  paint or near
30    sources of ambient Pb,  as well as socioeconomic factors, such as reduced access to  health care or
31    low socioeconomic status (SES) that can contribute to increased risk of adverse health effects
32    from Pb (ISA, sections 2.9.7, 6.2, and 6.4).
33
         January 2013                           3-28             Draft - Do Not Quote or Cite

-------
 1         •   Has new information altered our understanding of human populations that are
 2             particularly at risk of health effects from Pb exposures?
 3           The information newly available in this review has not substantially altered our
 4    understanding of at-risk populations.  As in the last review, the factor most prominently
 5    recognized to contribute to increased risk of Pb effects is age (ISA, section 2.9.7). As noted in
 6    section 3.2 above, although the specific ages or lifestages of greatest susceptibility have not been
 7    established (e.g., ISA, section 5.3.12), the susceptibility of young children to the
 8    neurodevelopmental effects of Pb is well recognized (e.g., ISA, sections 2.9.7, 5.3,  6.2.1, 6.3.1,
 9    and 6.4). The evidence indicates that prenatal blood Pb  levels are causally associated with
10    nervous system effects, including mental development in very young children and can also be
11    associated with cognitive decrements in older children (ISA, section 5.3).  The coincidence
12    during early childhood of behaviors that increase exposure, such as hand-to-mouth  contact, and
13    the development of the nervous system contributes increased risk during this time (ISA, sections
14    5.3.2.6, 6.2.1.1, and 6.3.1.1).  The evidence also indicates a causal relationship of postnatal blood
15    Pb levels (through early childhood to school age) with cognitive function decrement in older
16    children and adolescence  (ISA, section 5.3). In epidemiological studies, associations have been
17    observed of neurocognitive, and some other nervous system effects, at various ages from early
18    childhood to school age with prenatal, early-childhood, lifetime average, and concurrent blood
19    Pb levels as well as with childhood tooth Pb levels (ISA, section 5.3). Consideration of
20    epidemiological study results for different lifestages  of exposure, particularly later in childhood,
21    is complicated by the fact that blood Pb levels in children, although highly affected by recent
22    exposure, are also influenced by their history of Pb exposure due to rapid growth-related bone
23    turnover in children (ISA, section 4.3.5). Thus, blood Pb level in children also may reflect past
24    Pb exposures and, to some extent, maternal Pb, with relative contributions varying with child and
25    maternal exposure history (ISA, section 4.2.2.4, 4.4.1 and 5.3.16; 2006 CD, section 6.6.2).
26    Collectively, however, the evidence indicates both the susceptibility of the developing fetus and
27    early postnatal years, as well as the potential for continued susceptibility through childhood as
28    the human central nervous system continues to mature and be vulnerable to neurotoxicants (ISA,
29    sections 2.9.5 and 5.3.16; 2006 CD, section 6.2.12).
30           In the collective body of evidence of nervous system effects in children,  it is difficult to
31    distinguish exposure in later lifestages (e.g., school age), and its  associated risk, from prenatal
32    and early childhood (ISA, section 5.3.12).  While early childhood is recognized  as a time of
33    increased susceptibility, a difficulty in identifying a discrete period of susceptibility from
34    epidemiological studies has been that the period of peak exposure, reflected in peak blood Pb
35    levels, is around 18-27 months when hand-to-mouth activity is at its maximum (ISA, section
36    4.4.1 and 6.2.1.1; 2006 CD, p. 6-60).  The task is additionally complicated by the role of

         January 2013                           3-29            Draft - Do Not  Quote or Cite

-------
 1    maternal exposure history in contributing Pb to the developing fetus (ISA, section 4.2.2.4.).
 2    Epidemiological analyses evaluating risk of neurocognitive impacts (e.g., reduced IQ) associated
 3    with different blood Pb metrics in cohorts with differing exposure patterns (including those for
 4    which blood Pb levels at different ages were not highly correlated) indicate associations with
 5    blood Pb measurements concurrent with FSIQ tests at ages of approximately 6-7 years, although
 6    the analyses did not conclusively demonstrate stronger findings for early (e.g., age 2 years) or
 7    concurrent blood Pb (ISA, section 5.3.12).  The experimental animal evidence additionally
 8    indicates early life susceptibility (ISA, section 5.3.16). Thus, the full evidence base continues to
 9    indicate prenatal and early childhood lifestages as periods of increased lead-related risk. In
10    summary, while uncertainties remain with regard to the role of Pb exposures during a particular
11    age of life in eliciting nervous system effects, such as cognitive function decrements, the
12    evidence continues to indicate the at-risk status of pre- and postnatal childhood lifestages (ISA,
13    sections 5.3.12 and 5.3.16).
14           Several physiological factors increase risk of lead-related health effects by contributing to
15    increased blood Pb levels over those otherwise associated with a given Pb exposure (ISA,
16    sections 4.2, 4.3 and 6.1). These include nutritional status, which plays a role in Pb absorption
17    from the GI tract (ISA, section 4.2.1.2). For example, diets deficient in iron, calcium or zinc can
18    contribute to increased Pb absorption and associated higher blood Pb levels (ISA, sections
19    4.2.1.2, and 6.1). Evidence  is suggestive of some genetic characteristics as potential risk factors,
20    such as presence of the 5-aminolevulinic acid dehydratase-2 (ALAD-2) allele which has been
21    indicated to increase blood Pb levels and or lead-related risk (ISA, sections 4.3.2 and 6.1).
22           Risk factors based on increased exposure include spending time in proximity to sources
23    of Pb to ambient air or other environmental media (e.g.,  large active metals industries or
24    locations of historical Pb contamination) (ISA, section 2.9.7, 6.2.5). Residential factors
25    associated with other sources of Pb exposure (e.g., leaded paint or plumbing with Pb pipes or
26    solder) are another exposure-related risk factor (ISA, section 2.9.7, 6.2.6). The role of
27    socioeconomic status (SES) with regard to lead-related risk is somewhat complicated.  SES often
28    serves as a marker term for one or a combination of unspecified or unknown environmental or
29    behavioral variables. Lower SES has been associated with higher Pb exposure and higher blood
30    Pb concentration (ISA, sections 5.3.16, 6.2.4 and 6.4). Further, it is independently associated
31    with an adverse impact on neurocognitive development and a few studies have examined SES as
32    a potential modifier of the association of childhood Pb exposure with cognitive function with
33    inconsistent findings regarding low SES as a potential risk factor.  Although the differences in
34    blood Pb levels among children of lower as compared to higher income levels have lessened,
35    blood Pb levels continue to be higher among lower income children indicating higher exposure


         January 2013                            3-30            Draft - Do Not Quote or Cite

-------
 1    and/or greater influence of factors independent of exposure, such as nutritional factors (ISA,
 2    sections 2.9.7, 6.2.1.1 and 6.4).
 3           In considering risk factors associated with increased Pb exposure or increased blood Pb
 4    levels, we note that the currently available evidence continues to support a nonlinear relationship
 5    between neurocognitive effects and blood Pb that indicates incrementally greater impacts at
 6    lower as compared to higher blood Pb levels (ISA, section 5.3.13), as described in section 3.2
 7    above. An important implication of this finding is that while  children with higher blood Pb
 8    levels are at greater risk of lead-related effects than children with lower blood Pb levels, on an
 9    incremental basis (e.g., per |ig/dL), the risk is greater for children at lower blood Pb levels. This
10    was given particular attention in the last review of the Pb NAAQS, in which the standard was
11    revised with consideration of the incremental impact of air-related Pb on young children in the
12    U. S and the recognition of greater impact for those children with lower absolute blood Pb levels
13    (73 FR 67002).  Such consideration included a focus on those C-R studies involving the lowest
14    blood Pb levels, as described in section  4.1.1 below.
15           Some racial or ethnic backgrounds have been identified as factors that may increase risk
16    of lead-related health effects (ISA, sections 2.9.7, 6.2.3 and 6.3.7).  For example, although blood
17    Pb levels in the U.S. general population (e.g., geometric mean level in children aged 1-5) have
18    declined, mean levels reported in recent NHANES samples continue to differ among children of
19    different ethnic backgrounds, with higher levels in non-white persons as compared to whites
20    (ISA, sections 4.4.1, 6.2.1.1 and 6.2.3).  Additionally, a study of lead-related risk of hypertension
21    among adults reported greater risk associated with blood Pb levels above  1 |ig/dL among
22    Mexican Americans and non-Hispanic blacks than among non-Hispanic whites (ISA, section
23    6.3.7; Muntner et al., 2005). The evidence available in the current review, consistent with that in
24    the last, also suggests that health status or pre-existing disease is potentially a physiological risk
25    factor for lead-related effects (ISA, section 2.9.7).  Populations with pre-existing health
26    conditions, such as hypertension, may be more susceptible (as compared to the general
27    population) for particular Pb-associated effects (ISA, section 6.3.4). For example, increased risk
28    of lead-related renal effects and heart rate variability have been reported among hypertensive
29    individuals compared to those that are normotensive (ISA, section 2.9.7).  Additionally, African
30    Americans, as a group, have a higher frequency of hypertension than the general population or
31    other ethnic groups (NCHS, 2011), and as a result may face a greater risk of adverse health
32    impact from Pb-associated cardiovascular effects.
33           Older adulthood has been identified as a lifestage of potentially greater risk of lead-
34    related health effects based primarily on the evidence  of increases in blood Pb levels during this
35    lifestage, and of potential for effects,  such as (ISA, sections 6.2.1.2, 6.3.1.2, and 6.4).
36    Contributing to blood Pb levels in the studied populations of older adults are likely to be their

         January 2013                           3-31            Draft - Do Not Quote or Cite

-------
 1    exposure histories, which included younger years during the time of leaded gasoline usage and
 2    other sources of Pb exposures which were more prevalent in the past than today (e.g., ISA,
 3    Figure 3-1 and section 3.4.3.2). Exposure history has a contributing role to blood Pb levels
 4    throughout life, and the increased rate of bone remodeling during later adulthood increases
 5    contributions of Pb from bone stores into the systemic circulation during that lifestage (ISA,
 6    sections 4.3.5 and 6.2.1.2). Additionally, limited animal evidence has indicated specific brain
 7    pathology in older animals that had substantial Pb exposures earlier in life (ISA, sections
 8    5.3.10.1 and 6.3.1). Further, the full body of evidence includes observed associations of various
 9    health effects, such cardiovascular and neurological effects with bone and blood Pb in older
10    populations, with biological plausibility for the role of Pb provided by experimental animal
11    studies (ISA, sections 5.3.6, 5.3.8, 5.3.10 and 5.4).
12           In summary, the information newly available in this review has not appreciably altered
13    our understanding of human populations that are particularly sensitive to Pb exposures.  In the
14    current review, as at the time of the last review of the Pb NAAQS, we recognize young children
15    as an important at-risk population, with sensitivity extending to prenatal exposures and into
16    childhood development. Additional risk factors include deficiencies in dietary minerals (iron,
17    calcium and zinc), some racial or ethnic backgrounds, and spending time in proximity to
18    environmental sources of Pb or residing in older houses. The evidence for SES continues to
19    indicate increased blood Pb levels in lower income children, although its role with regard to an
20    increased risk for same blood Pb level is unclear. Additionally, the currently available evidence
21    continues to indicate a potential for increased risk associated with  several other factors, including
22    older adulthood, pre-existing disease (e.g., hypertension), variants for certain genes and
23    increased stress.
24         •   Is there new evidence on health effects beyond neurocognitive endpoints in
25              children that suggest additional at-risk populations should be given increased
26              focus in this review?
27           The evidence newly available in this review supports or strengthens our previous
28    conclusions regarding the broad array of health effects of Pb (see ISA, section 2.10 which
29    compares key conclusions drawn in the last review with conclusions drawn in the current
30    assessment).  Additionally, in some categories of health effects, the newly available studies
31    extend the evidence for some aspects of the health effects described in the last review.  For
32    example, among the nervous system effects, the newly available evidence continues to support
33    the conclusions from the last review regarding Pb and neurocognitive and neurobehavioral
34    effects (ISA, section 5.3). Across the array of neurocognitive and  behavioral effects, we
35    recognize the sensitivity of the prenatal period and several lifestages of childhood, and


         January 2013                           3-32             Draft - Do Not Quote or Cite

-------
 1    particularly recognize young children as an important at-risk population in light of current
 2    environmental exposure levels.
 3           As discussed in section 3.2 above, the blood Pb levels of populations studied in newly
 4    available epidemiological studies that report associations of blood Pb with effects for systems
 5    and processes other than the nervous system (e.g., cardiovascular, developmental and
 6    reproductive) are similar, or in a few cases, somewhat lower than those assessed in the last
 7    review (ISA, sections 5.4, 5.6, 5.7, and 5.8). The greater uncertainties regarding the time,
 8    duration and magnitude of exposure contributing to these observed health effects complicate
 9    identification of sensitive lifestages and associated exposure patterns that might be compared
10    with our understanding of the sensitivity of young children to neurocognitive impacts of Pb.
11    Thus, while augmenting the evidence base on these additional endpoints, the newly available
12    evidence does not lead us to identify a health endpoint expected to be more sensitive to Pb
13    exposure than neurocognitive endpoints in children, leading us to continue to conclude that the
14    appropriate primary focus for our review is on neurocognitive endpoints in children.
15           In summary, there are a variety of ways in which lead-exposed populations might be
16    characterized and stratified for consideration of public health impacts. Age or lifestage was used
17    to distinguish potential  groups on which to focus in the last review in recognition of its role in
18    exposure and susceptibility, and young children were selected as the priority population for the
19    risk assessment (see section 3.4 below) in consideration of the health effects evidence regarding
20    endpoints of greatest  public health concern and in recognition of effects on the developing
21    nervous system as a sentinel endpoint for public  health impacts of Pb. This identification
22    continues to be supported by the evidence available in the current review.
23         •   What does  the information about air Pb concentrations available in this review
24             indicate with regard to the size of at-risk populations and their distribution
25             across the  U.S.?
26           The magnitude of a public health impact  is dependent upon the size of populations
27    affected,  as well as type or severity of the effect. As summarized above, the population group
28    that may  be most at risk of health effects associated with exposure to Pb is young children.  The
29    2010 census indicates nearly 310 million people  residing in the U.S., 74 million of which
30    approximately are children under the age of 18, with some 20 million under the age of five years.
31    Children at greatest risk from air-related Pb are considered to be those children living in areas of
32    higher air Pb exposures. Accordingly, the discussion below considers the information available
33    to inform our understanding of areas of children  potentially at risk from air-related Pb.
34           In considering the extent of this at-risk population, we turn first to consideration of those
35    areas in the U.S. with air Pb concentrations above the current standard (e.g., section 2.2.2.2
36    above).  Using the available monitoring data and U.S. census information, Table 3-4 summarizes

         January 2013                           3-33             Draft - Do Not Quote or Cite

-------
 1    the size of populations within 0.5 km of monitors in our current Pb NAAQS surveillance
 2    network at which Pb concentrations were higher than the current standard during the recent
 3    period from 2009-2011.  The distance to which concentrations exceeding the standard might
 4    extend will vary with the magnitude of the Pb concentrations and particle  size, among other
 5    factors; a half kilometer distance was selected for purposes of illustration here.  This analysis
 6    indicates approximately 2,400 children aged 5  or under to reside within 0.5 km of monitors
 7    exceeding the current standard.  To also account for the population in areas with air
 8    concentrations just at (or very near) the current standard, we have also identified an additional
 9    nine Pb-TSP monitors with 3-month average concentrations within 10% of the current standard
10    (Appendix 2D). Based on the 2010 U.S. census, 265 children aged 5 or under reside within 0.5
11    km of these additional sites.

12    Table 3-4.   Number of children aged 5 and under in areas of elevated ambient Pb
13                concentrations relative to the NAAQS.
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Population within 0.5 km of monitors
with maximum 3-month Pb concentration greater than 0.15 (ig/m3
(2009-2011)

Allsites>0.15(ig/m3
Subset of sites >0.5 (ig/m3
Number of
Counties
29
11
Number of States
or Territories
20
9
Total
population
25,344
11,753
Children,
Sand under
2,416
1,018
Data Sources: U.S. Bureau of the Census, 2010 Census of Population and Housing; recent Pb-TSP dataset
presented in Figure 2-10 above (dataset criteria and summaries included Appendices 2C and 2D, respectively).
       As the current air quality data set (analyzed in section 2.2.2 above) may not be inclusive
of all of the newly sited monitors, as discussed in section 2.2.1 above, we recognize there may be
other areas of the country where concentrations are above or just meet the current standard but
for which such data are not yet available.  To consider the potential for there to be additional, not
yet identified, areas with elevated Pb concentrations, we have separately quantified the size of
young child populations residing in areas near large Pb sources in Table 3-5. In so doing, we
recognize uncertainties and potential limitations associated with these emissions estimates for
these purposes, uncertainties both with regard to the accuracy of such estimates and also with
regard to the role of specific  source characteristics and meteorology, not explicitly considered
here, in influencing ambient  air Pb concentrations and contributing to substantial variation in air
Pb concentrations at source locations (e.g., Figure 2-11 above). Accordingly, while the summary
in Table 3-5 is informative to considering the potential prevalence of airborne Pb emissions and
potential exposure of human populations, it is limited with regard to its  ability to identify
         January 2013
                                          3-34
Draft - Do Not Quote or Cite

-------
1    populations living in areas of elevated ambient air Pb concentrations.  We interpret this analysis

2    to indicate that fewer than about 7,800 young children (aged 5 or younger) live in areas with air

3    Pb concentrations near or above the current standard, with the current monitoring data indicating

4    the size of this population to be approximately 2,700.


5    Table 3-5.   Population size near larger sources of lead emissions.
Facilities (other than airports)
Number of
Locations
Population,
all ages
Population,
aged 5 and
younger
Airports
Number of
Locations
Population,
all ages
Population,
aged 5 and
younger
All
Population,
aged 5 and
younger
                         Population Within 0.5 km of Sources Emitting at Least 0.5 tpy in 2008
                                       Facilities/'Airports estimated to emit > 5.0 tpy
                           484
                 61
         I
       0
   0
                 61
                                     Facilities/Airports estimated to emit 1.0-4.9 tpyA
            53
12,143
 731
I
 6,261
 266
 997
                                   Facilities/Airports estimated to emit 0.50 - 0.95 tpy B
            63
12,934
1,143
       52
76,105
6,699
7,842
      A - Facilities estimated to emit at least 1.0 tpy (after rounding to 1 decimal place) and less than 5.0 (after rounding to 1 decimal place).
      B - Facilities estimate to emit at least 0.50 tpy (after rounding to 2 decimal places) and less than 1.0 (after rounding to 1 decimal place).
      Sources: Population counts from U.S. Bureau of the Census, 2010 Census of Population and Housing. Emissions estimates for facilities other
      than airports drawn from 2008 NEI, version 3 (December 2012); estimates for airports reflect EPA's best estimates of piston-engine aircraft
      emissions. Piston-engine aircraft emissions inventory is available at: http://www.epa.gov/ttn/chief/net/2008neiv2/2008_neiv2_tsd_draft.pdf.
6

7
         January 2013
                              3-35
                                 Draft - Do Not Quote or Cite

-------
 1    3.4   EXPOSURE AND RISK
 2          This risk information available for this review and described here is based primarily on
 3    the exposure and risk assessment developed in the last review of the Pb NAAQS (henceforth
 4    referred to as the 2007 REA [USEPA, 2007a]), in the context of the evidence newly available in
 5    this review, as presented in the ISA.  As described in the REA Planning Document, careful
 6    consideration of the information newly available in this review, with regard to designing and
 7    implementing a full REA for this review, led us to conclude that performance of a new REA for
 8    the review was not warranted.  We did not find the information newly available in this review to
 9    provide the means by which to develop an updated or enhanced risk model that would
10    substantially improve the utility of risk estimates in informing the current Pb NAAQS review
11    (REA Planning Document, section 2.3). Based on their consideration of the REA Planning
12    Document analysis, the CASAC Pb Review Panel generally concurred with the  conclusion that a
13    new REA was not warranted in this review (Frey, 2011).26 Accordingly, the information
14    described here is drawn primarily from the 2007 REA, augmented by a limited new case study-
15    specific analysis focused on risk associated with the current standard, as described in  section
16    3.4.3.3 below.
17          The focus for the risk assessment and associated estimates presented here is on Pb
18    derived from sources emitting Pb to ambient air. As discussed in  section 1.3 above (and
19    conceptually illustrated in Figure 1-1), the multimedia and persistent nature of Pb, the role of
20    multiple exposure pathways, and the contributions of nonair sources of Pb to human exposure
21    media all present challenges and contribute significant additional complexity to the health risk
22    assessment that goes far beyond the situation for similar assessments typically performed for
23    other NAAQS pollutants (e.g., that focus only on the inhalation pathway). Limitations in the
24    available data and models affected our characterization of the various complexities associated
25    with exposure to ambient air Pb. As a result, the assessment includes a number of simplifying
26    assumptions in a number of areas and, as described in section 3.4.4 below, our estimates of air-
27    related Pb risk are approximate and are characterized by upper and lower bounds.
28          The conceptual model developed to inform planning for the 2007 REA, including
29    identification of key exposure media, target population, health endpoint and risk metric  is
30    described in  section 3.4.1.  The 2007 REA relied on a case study approach to provide estimates
31    that inform our understanding of air-related exposure and risk in different types of air Pb
32    exposure situations; the case studies included are described in section 3.4.2. In section 3.4.3, the
            26 In our evaluation presented in the REA Planning Document and consultation with CASAC, we indicated
      our conclusion that the information newly available in this review did not provide the means by which to develop an
      updated or enhanced risk model that would substantially improve the utility of risk estimates in informing the
      current Pb NAAQS review.
         January 2013                           3-36            Draft - Do Not Quote or Cite

-------
 1    analysis approach and general aspects of exposure and risk assessment methods are summarized,
 2    and the air quality scenarios simulated are described.  In section 3.4.3, we also summarize the
 3    2007 REA risk model and the interpolation approach used in the limited new analyses performed
 4    for purposes of this review. Section 3.4.4 identifies key aspects in exposure assessment and risk
 5    estimates are presented in section 3.4.5.  Treatment of key sources of variability in exposure and
 6    risk estimates is described in section 3.4.6 and the characterization of uncertainty is summarized
 7    in 3.4.7. An updated interpretation of the risk estimates for our purposes in this Pb NAAQS
 8    review section is presented  in section 3.4.8.

 9         3.4.1   Conceptual Model for Air-Related Lead Exposure and Risk
10           The focus in consideration of public health risks associated with Pb from ambient air is
11    on Pb derived from those sources emitting Pb to ambient air. The multimedia and persistent
12    nature of Pb, illustrated in Figure 1-1 above, as well as the existence of many nonair sources of
13    Pb to the environment, contribute multiple complexities to the consideration of exposure and risk
14    for ambient air-related Pb.  The conceptual model that informed planning for the 2007 REA
15    identified sources, pathways, routes, exposed populations, and health endpoints, focusing on
16    those aspects of Pb exposure most relevant to the review,  while also  recognizing the role of Pb
17    exposure pathways unrelated to Pb in ambient air (2007 REA, section 2.1).
18           As recognized in section 1.3 above, sources of human Pb exposure include current and
19    historical air emissions sources, as well as miscellaneous nonair sources, which can contribute to
20    multiple exposure media  and associated pathways (e.g., inhalation of ambient air, ingestion of
21    indoor dust,  outdoor soil/dust and diet or drinking water).27  Figure 3-2 illustrates these human
22    exposure pathways from an analytical perspective, drawing on the conceptual model for the
23    assessment (2007 REA, Figure 2-1).  As shown in Figure  3-2, in addition to  airborne emissions
24    (recent or those in the past), sources of Pb to these pathways also include old leaded paint,
25    including Pb mobilized indoors during renovation/repair activities, and contaminated soils. Lead
26    in diet and drinking water may have air pathway related contributions as well as contributions
27    from nonair  sources (e.g., Pb solder on water distribution  pipes and Pb in materials used in food
28    processing).  Limitations  in our data and modeling tools handicapped our ability to separate the
29    nonair contributions to Pb exposure from estimates of air-related Pb  exposure and risk.28
30
             27 We did not explicitly consider Pb exposure related to consumer products (e.g., toys, cosmetics, dishes) in
      the 2007 REA.
             28 The assessment grouped the exposure and risk estimates for Pb in diet and drinking water together and
      combined them with the other pathways in estimates presented for "total Pb exposure". Characterization of the risk
      assessment results in the rulemaking recognized the contribution, albeit unquantified, from air-related pathways
      within this category.
         January 2013                           3-37             Draft - Do Not Quote or Cite

-------
1


* 	 1 —
/"~ ^\
Inhalation of
ambient air Pb
(outdoors or
indoors)
Pb in ambient air

Infiltration
	 Indoor air
Deposition
indoors
t
\Resuspension
i

n
\ 0
\
\
^•Resusp
\
\
\
\
\
«Hitr.hhikina"in "
Indoor dust ^
\ ^
Ingestion of
indoor dust Pb
V J

r-*
_/PsM\_
vpby~

eposition
jtdoors
insion

Outdoor
soil/dust

(
/HistoncS


^PbfrorrN*
distribution )
processing/
i^J^x Food (crops/
livestock)
/Tan
-------
 1    and international health agencies. This reflects the association of PbB with exposure, particularly
 2    recent exposure in young children, and the relative ease of collecting PbB measurement.
 3    Although bone Pb measurements have become easier to collect and consequently, their use has
 4    been more widespread, epidemiological and toxicological studies of Pb health effects and dose-
 5    response relationships (particularly for neurodevelopmental effects in children) tend to be
 6    dominated by PbB as the exposure metric. Therefore, we focused on modeling PbB in young
 7    children, developing estimates for two PbB metrics:  "concurrent" and "lifetime average". For
 8    the former we estimated PbB at age 7 years, while lifetime average was estimated as the average
 9    of PbB levels across the  7 year period.
10           At the time of the last review, we noted that limitations precluded prediction of changes
11    in adult PbB levels (or bone Pb levels) given changes in ambient Pb levels. This reflects the fact
12    that the presence of substantial historic Pb stores in most  adults introduces uncertainty into the
13    prediction of changes in  blood or bone Pb in these adult populations resulting from  changes in
14    ambient air Pb exposure. Additionally, in considering concentration-response relationships for
15    adult PbB and adult health outcomes, we recognized the uncertainty with regard to the role of
16    historic compared to recent exposures in eliciting the observed outcomes.
17           Based on conclusions regarding the scientific evidence available in the last review
18    (presented in the 2006 CD), the assessment focused on risk to the central nervous system in
19    childhood as the most sensitive effect that could be quantitatively assessed, with decrement in IQ
20    used as the  risk metric.

21         3.4.2   Case Studies
22           Lead exposure and associated risk was estimated for multiple case studies that generally
23    represent two types of residential population exposures to air-related Pb  (see Table  3-6):  (1)
24    location-specific urban populations of children with a broad range of air-related exposures,
25    reflecting existence of urban concentration gradients; and, (2) children residing in localized areas
26    with air-related exposures representing air concentrations specifically reflecting the standard
27    level being  evaluated. Thus, the two types of case studies differed with regard to the extent to
28    which they  represented population variability in air-related Pb exposure  (as discussed further in
29    section 3.4.7 below).  Three location-specific urban case  studies focused on residential areas
30    within Cleveland, Chicago, and Los Angeles, providing representations of urban populations
31    with a broad range of air-related exposures due to spatial  gradients in both ambient  air Pb levels
32    and population density.  For example, the highest air concentrations in these case studies (i.e.,
33    those closest to the standard being assessed) were found in very small parts of the study areas,
34    while a large majority of the case study populations resided in areas with much lower air
35    concentrations.  The case studies representing the children most highly exposed via air-related

         January 2013                          3-39             Draft - Do Not Quote or Cite

-------
 1    pathways were the generalized (local) urban case study (also referred to as general urban case
 2    study) and the primary Pb smelter case study subarea.  The generalized (local) urban case study
 3    was not based on a specific geographic location and reflected several simplifying assumptions in
 4    representing exposure including uniform ambient air Pb levels associated with the standard of
 5    interest across the hypothetical study area and a uniform study population. The primary Pb
 6    smelter case study provides risk estimates for children living in a specific area that, at the time of
 7    the 2007 REA, was not meeting the then current NAAQS.  In addition to characterizing risks
 8    within a 10 km radius area surrounding the smelter (full area), we focused particularly on a
 9    subarea within 1.5 km of the facility, where airborne Pb concentrations were closest to the
10    current standard and where children's air-related exposures are most impacted by emissions
11    associated with the Pb smelter from which air Pb concentrations were estimated. Based on the
12    nature of the population exposures represented by the two categories of case study, the
13    generalized (local) urban and primary Pb smelter case study subarea include populations  that are
14    relatively more highly exposed by way of air pathways to air Pb concentrations  near the standard
15    level evaluated, compared with  the populations in the three cities or the full area of the primary
16    smelter case study.30'31

17    Table 3-6.   Types of population exposures assessed.
Type of Population Exposure
Broad range of
air-related
exposures
Generalized, high
end of air-related
exposure
Part of metropolitan area with spatially
varying air concentrations, inclusive of
location at standard or conditions
being evaluated
As above, with dominant, historically
active metals industry as ambient air
Pb source
Localized residential area with air
concentrations generally representing
the standard or conditions evaluated
Multiple exposure
zones, larger
populations
Single exposure
zone without
enumerated
population
A few exposure
zones with small
population
Case Study1
Location-specific urban:
Cleveland, Chicago, Los Angeles
Primary Pb smelter
(full area)
Generalized (local) urban
Primary Pb smelter (subarea)
             30 An additional case study (the secondary Pb smelter case study) was also developed in the 2007 REA,
      however, significant limitations associated with the use of dispersion modeling to predict ambient air, dust and soil
      Pb levels near the facility contributed to large uncertainties in the risk estimates (2007 REA, section 4.3.1).
             31 In addition to the case studies included in the Risk Assessment Report, the pilot phase of the 2007 REA
      also included a near-roadway case study, focused on subset of urban population exposed immediately near roadways
      (ICF International, 2006). Based on the pilot results and advice from CASAC, however, this case study was not
      carried into the full-scale analysis. As an alternative, we developed the generalized (local) urban case study to
      represent urban residents in a localized area exposed to air-related Pb associated with the standard being assessed.
         January 2013
3-40
Draft - Do Not Quote or Cite

-------
 1         3.4.3   Analysis Approach
 2          The approach to assessing exposure and risk for the two categories of case studies was
 3   comprised of four main analytical steps: (a) estimation of ambient air Pb concentrations, (b)
 4   estimation of Pb concentrations in other key exposure media, including outdoor soil and indoor
 5   dust, (c) use of exposure media Pb concentrations, with other pathway Pb intake rates (e.g., diet),
 6   to estimate PbB levels in children using biokinetic modeling, and (d) use of concentration-
 7   response functions derived from epidemiology studies to estimate IQ loss associated with the
 8   PbB levels. In implementing these steps for the primary smelter case study, air concentrations
 9   were estimated using dispersion modeling; indoor dust concentrations were estimated for the
10   case study subarea using a site-specific regression model.  The approach for the generalized
11   (local) urban case study and location-specific urban case studies is somewhat simpler, since it
12   does not involve fate and transport modeling for air concentration estimates and, instead, uses
13   ambient monitor levels to characterize the gradient in air Pb levels across the study area.  All
14   steps are somewhat simpler in the generalized (local) urban case study which includes a single
15   exposure zone. Figure 3-3 identifies the key input data sets, modeling steps and intermediate
16   model output in each of the four analytical steps.  The first three steps were employed in the
17   exposure assessment (discussed in section 3.4.3.1), while the fourth is the risk assessment step
18   (discussed in section 3.4.3.3).
         January 2013                           3-41             Draft - Do Not Quote or Cite

-------
1
2
RISK CHARACTERIZATION EXPOSURE ASSESSMENT

Characterizing ambient air levels and
deposition to soil
Ambient monitoring data
(general urban and location-specific urba

1
! Ambient
| concentrat
j for study a
V. 	


Characterizing soil and indoor
dust concentrations
j
Combination of (a) statistical
(regression) and/or (b) mechanistic
(compartmental) indoor dust prediction <
models
(all case studies)
	 Zi 	
.•• '•;
i Indoor residential dust j
I concentrations j
'•. 	 ./
1
Characterizing blood Pb levels
Background Pb exposure levels
•diet
• drinking water 	
• indoor paint (actually reflected in
dust modeling)
t
Demographics (distribution of
children within study areas)
(primary Pb smelter and location- '
specific urban)


Characterizing risk (IQ loss)
r"
f "\ Risk
Distribution of IQ loss for ! cha
study populations ! / ,
(partitioned between policy- k-J ^
relevant exposures and j
policy-relevant background) j


n)


Stack and
fugitive
emissions data

Air dispersion modeling L


air : 1
~ 	 J

-r^J



Nationally-representative
residential soil Pb value selected
from literature
(general urban and location-
specific urban)


i '
f 	 \
	 | Outdoor res idential L 	
1 soil concentrations j
i '
Biokinetic blood Pb
i 	 ^ modeling 4 	
	 | (all case studies)
\~ J
J— '
Probabilistic modeling of blood Pb r
levels for children within each case 	 |
study location
(all case studies)
^ J



1 ' Concentration-r
estimation - IQ ") , 	 based on log-no
nge estimation 4—1 (Lanphear poole
I case studies) from that study
	 J



Site-specific soil
monitoring data
combined with
	 statistical
extrapolation
(primaryPb
smelter)

| Ambient air j
j 	 1 concentrations 1
| (see above) j

GSD reflecting inter-
individual variability
	 in behavior related to
Pb exposure and
biokinetics

	

;sponse functions
rmal functions
d analysis), or on data

       Drawn from 2007 REA, Figure 3-3 (USEPA, 2007).

Figure 3-3.  Overview of analysis approach.

   January 2013                          3-42
                                                               Draft - Do Not Quote or Cite

-------
 1         3.4.3.1  Estimating Exposure
 2           Concentrations of Pb were estimated in ambient media and indoor dust using a
 3    combination of empirical data and modeling projections.  The use of empirical data brings with it
 4    uncertainty related to the potential inclusion of nonair source signals in these measurements (e.g.,
 5    house paint contributions to indoor dust and outdoor soil Pb). Conversely, the use of modeling
 6    tools introduces other uncertainties (e.g., model and parameter uncertainties).  The
 7    characterization of uncertainty associated with the risk assessment is discussed in section 3.4.7.
 8    Table 3-7 summarizes the exposure modeling approaches and data used to characterize Pb
 9    concentrations or input associated with each exposure pathway for each of the case studies.
10           Characterization  of Pb in ambient air relied on (a) dispersion modeling of facility-related
11    (including fugitive) Pb emissions for the primary Pb smelter case study, (b) the use of ambient
12    monitor data for the location-specific urban case studies, and (c) an assumption of uniform
13    ambient air Pb levels (matching the standard level being considered) for the generalized (local)
14    urban case study.  The use of dispersion modeling for the primary Pb smelter case study allowed
15    us to capture more spatially refined patterns of ambient air Pb over residential areas in the
16    vicinity of the facility, where ambient Pb levels were dominated by Pb released from the facility.
17    For the location-specific urban case studies, we used Pb monitors within each of the urban study
18    areas to characterize spatial gradients. By contrast, the generalized (local) urban case study is
19    designed to assess exposure and risk for a smaller group of residents (e.g., neighborhood)
20    exposed at the level of the standard and therefore, did not rely on monitor data in characterizing
21    levels, since ambient air Pb was fixed at the standard being assessed. While ambient air Pb
22    concentrations in the primary Pb smelter case study reflected only contributions from direct and
23    fugitive emissions associated with the facility, concentrations in the location-specific urban study
24    areas, which relied on empirical (monitor-based) data to define ambient air Pb concentrations,
25    reflected contributions from all contributing sources, be they currently active stationary or
26    mobile sources, resuspension of previously deposited Pb or other.  Additional detail on
27    estimation of ambient (outdoor) and indoor air concentrations is presented in section 5.2.2 and
28    Appendices A through D of the 2007 REA.
29           Characterization  of Pb concentrations in outdoor surface soil/dust, resulting from
30    deposition of airborne Pb was based on the use of (a) existing site-specific measurements
31    (primary Pb smelter case study), and (b) nationally representative residential soil measurements
32    obtained from the literature (general and location-specific urban case studies). In the case of the
33    primary Pb smelter case  study, soil Pb concentration data were available for a zone close to the
34    facility and statistical extrapolation from these data was used to predict soil levels for portions of
35    the study area beyond this zone. Additional detail on estimation of Pb concentrations in outdoor
36    surface soil or dust is presented in sections 3.1.3 and 5.2.2.2  and Appendix F of 2007 REA.
         January 2013                          3-43             Draft - Do Not Quote or Cite

-------
 1          To predict concentrations of ambient Pb in indoor dust, we relied on a combination of (a)
 2   regression-based models that relate indoor dust to outdoor air Pb and/or outdoor soil Pb and (b)
 3   mechanistic models that predict indoor dust Pb based on key mechanisms (e.g., infiltration of
 4   outdoor air indoors, deposition rates of Pb from indoor air to indoor surfaces, house cleaning
 5   rates). For the point source case study, a combination of regression-based models obtained from
 6   the literature and developed based on site-specific data were used, and a customized hybrid
 7   empirical-mechanistic model was developed for the general and location-specific urban case
 8   studies. This reflected the fact that available regression-based models had been developed
 9   largely based on residential exposures near large point sources and were not considered
10   representative of more general urban exposures. Consequently, a mechanistic model, augmented
11   with empirical data, was developed for the generalized (local) urban case  study.  Additional
12   detail on estimation of Pb concentrations in indoor dust is presented in sections 3.1.4 and 5.2.2.2
13   and Appendix G of 2007 REA.

14   Table 3-7.  Summary of approaches used to estimate case study media concentrations.
Simulation of
air quality
impacts
Concentrations
for these
media were
varied across
air quality
scenarios
(see section
3.4.3.2).
Concentrations
were constant
across air
quality
scenarios
(data/modeling
limitations)
Media
category
Ambient air
Pb levels
Indoor dust
Pb levels
Outdoor soil
Pb levels
Dietary Pb
intake
Drinking
water Pb
intake
Generalized (local)
urban case study
Single ambient air Pb
level assumed across
entire study area
(single exposure zone)
Location-specific
urban case study
Source and non-source
monitors describe
concentration gradient (6
to 1 1 exposure zones
per case study)
Hybrid model: dynamic aspect relates ambient air
Pb concentrations to indoor dust Pb; empirical
aspect represents Pb from other sources (e.g.,
paint, historical air, Pb carried indoors with people)
National dataset (HUD, for houses constructed
between 1940 and 1998).
Primary Pb smelter case
study (1.5 km subarea
and 10 km full area)
Dispersion modeling of
smelter-related emissions (
22 census block groups
and 115 blocks)
For subarea - regression of
site-specific air, dust data.
For full area - regression of
air, dust, soil data from
other, historical locations.
Site-specific data (for
subarea)
National datasets for Pb residue data (US FDA Total Diet Study) and food
consumption data (NHANES)
US and Canada datasets for residential water Pb concentrations and ingestion
rates
15
16
17
18
       Blood Pb levels were predicted from estimates of Pb contained in various media (e.g.,
ambient air, diet, water, indoor dust), and estimates of Pb intake from dietary and drinking water
pathways, using the Integrated Exposure and Uptake Biokinetic (IEUBK) model (2007 REA,
         January 2013
                                          3-44
Draft - Do Not Quote or Cite

-------
 1    sections 3.2.1.1 and S.2.4).32 Diet and drinking water intake and concentrations, as well as other
 2    model inputs, were based on the most current information (2007 REA, Appendix H). Detail on
 3    methods used to characterize media Pb concentrations and all IEUBK inputs for each case study
 4    are in the 2007 REA, sections 3.1, 3.2, 5.2.3 and 5.2.4 and Appendices C through H.  As the
 5    shortest temporal scale accepted for inputs to the IEUBK is a year, all model inputs, developed
 6    for each exposure zone in each case study, were annual average values.  For media concentration
 7    inputs, the same values were used for each year of the seven-year simulation.  Other model
 8    inputs varied as appropriate with the age of the simulated child (2007 REA, Appendix H).
 9           To simulate population variability in Pb intake and uptake, we used the IEUBK model to
10    first generate a central-tendency estimate of the PbB levels for the group of children within a
11    given exposure zone of a study area.33 Outside the IEUBK model, we then combined this central-
12    tendency estimate with a geometric standard deviation (GSD) reflecting variability in PbB levels
13    for groups of children to generate a distribution of PbB levels for a study area. The procedure
14    for combining the lEUBK-based central tendency blood Pb estimate with a GSD to generate a
15    population distribution of PbB levels differs somewhat for the categories of case studies. The
16    approach for the general urban case study is fairly simple in that we have a single lEUBK-based
17    estimate of PbB levels and this is in turn, combined with the GSD selected for this  study area to
18    produce a population-distribution of PbB levels. For both the primary Pb smelter and the
19    location-specific urban study areas, multiple polygons within the larger study area (e.g., US
20    Census blocks for the location specific urban study areas) are used as the basis for generating
21    distributions of PbB levels for the child population in each study area. These distributions are
22    generated using a MonteCarlo-based population-weighted sampling method with U.S. Census
23    child counts for each polygon and an adjustment factor distribution based on the chosen GSD
24    (see 2007 REA,  sections 3.2.2 and 5.2.2.3).
25           The GSD reflects a number of factors which operate together  to produce interindividual
26    variability in blood Pb levels, including: (a) biokinetic variability (differences in the uptake,
27    distribution or clearance of Pb), (b) differences in behavior related to Pb exposure (e.g., varying
28    hand-to-mouth activity, tap water ingestion rates, and time spent playing indoors) and (c)
             32 In predicting PbB levels, we assumed that Pb concentrations in exposure media remained constant
      throughout the 7 year simulation period.
             33 In typical IEUBK applications, the GSD is applied within the IEUBK model as part of the modeling
      process in order to generate percentiles of PbB distribution for the population simulated. However, for the NAAQS
      REA, we used IEUBK only for generating the central-tendency PbB value for a given exposure zone and then
      probabilistically combined that estimate with the GSD outside of the IEUBK model. This allowed us, in the case of
      the primary Pb smelter and the location-specific urban case studies (as noted below) to generate individual
      population-level PbB distributions for each exposure zone which could then be population-weighted and combined
      using Monte Carlo sampling to generate a single population-distribution for each study area. This was not possible
      with the typical application of the GSD within the IEUBK model.
         January 2013                           3-45             Draft - Do Not Quote or Cite

-------
 1    differences in environmental Pb exposure concentrations (e.g., spatial gradients in ambient Pb
 2    levels of a resolution beyond that simulated in each case study, differences in
 3    cleaning/vacuuming rates and air exchanges rates).34 Note, that for all of the study areas, we
 4    assumed that pathway apportionment of PbB levels based on the modeling of the central-
 5    tendency PbB level (using IEUBK) holds for all percentiles of PbB levels derived by combining
 6    that central-tendency with the GSD.  Blood Pb modeling completed for all case studies included
 7    estimates of both concurrent and lifetime-average PbB metrics, although ultimately, we focused
 8    on the concurrent PbB metric in estimating risk.35

 9              3.4.3.2  Air Quality Scenarios Included in 2007 Assessment

10            The air quality scenarios assessed in the 2007 REA for the case studies identified in
11    Table 3-6 above included conditions just meeting the NAAQS that was current at the time of the
12    last review (1.5 ug/m3, maximum calendar quarter average) and conditions meeting several
13    alternative, lower standards. Additionally, for the three location-specific urban case studies,
14    scenarios were included for then-current  conditions (2003-2005) in each location.36 These air
15    quality scenarios are characterized by quarterly or monthly averaging times and a not-to-be-
16    exceeded form.  Once the air quality dataset representing each scenario was developed, the
17    associated annual average concentrations for each exposure zone were derived for input to the
18    IEUBK model, which does not accept air quality inputs of a temporal scale shorter than a year
19    (2007 REA, Appendix H).37
20            As a result of the differing then-current conditions of the location-specific case studies,
21    the specific alternative standards assessed for each differed (see Table 3-8). To simulate the
             34 We specified GSDs for each of the case studies that reflected differences in the study areas and
      underlying study populations, as well as the availability of blood Pb measurement data (see 2007 REA, sections
      3.2.3 and 5.2.2.3).
             35 As discussed in section 2.1.5 of the 2007 REA, the concurrent PbB measurement (i.e., PbB
      measurements at the time of IQ test) and the lifetime-average blood level (i.e., average of measurements taken over
      child's first 6-7 years) were considered "stronger predictors of lead-associated intellectual deficits than was the
      maximal measured (peak) or early childhood blood lead concentrations" with the concurrent PbB level exhibiting
      the strongest relationship (2006 CD, p. 6-29).
             36 For the location-specific urban case studies of Cleveland, Chicago and Los Angeles, the maximum
      monthly average concentration was 0.56, 0.31 and 0.17 ug/m3, respectively, and the maximum calendar quarter
      average concentration was 0.36, 0.14 and 0.09 ug/m3, respectively, (2003-2005 data; 2007 REA, Appendix O).
             37 Although many different patterns of temporally varying air concentration will just meet a given potential
      alternative standard, the shortest time step accommodated by the blood Pb model is a year. Thus, the air Pb
      concentration inputs to the blood Pb model for each air quality scenario are annual average air Pb concentrations.
      For the generalized (local) urban case study, the national Pb-TSP monitoring dataset was analyzed to characterize
      the distribution of site-specific relationships between metrics reflecting the averaging time and form for the air
      quality scenarios being assessed (Table 3-5) and the annual average. The IEUBK annual average input was then
      derived by multiplying the level for a given air quality scenario by the ratio for the averaging time and form for that
      air quality scenario. For the location-specific case studies, the full temporally varying air Pb concentration dataset
      for each exposure zone was used to derive the average annual concentration for the IEUBK input.

         January 2013                            3-46             Draft - Do Not Quote or Cite

-------
 1    previous NAAQS scenario at the primary Pb smelter location, at which then-current monitoring
 2    data indicated exceedance of that standard, a proportional roll-down was performed across the
 3    area to achieve conditions that just met that standard.  Additionally, although it was considered
 4    an extremely unlikely scenario that air concentrations in urban areas across the U.S. that were
 5    well below the previous NAAQS would increase to just meet that standard (e.g., by way of
 6    expansion of existing sources or congregation of multiple sources in adjacent locations), we
 7    simulated this scenario in all case studies. In so doing, the air Pb concentrations were rolled up
 8    proportionally across the location-specific urban study areas to conditions just meeting the
 9    standard. No other scenario simulations involved rolling concentrations up.  For the primary Pb
10    smelter case study, air Pb concentrations were proportionally rolled down to conditions just
11    meeting each of the potential alternative standards assessed. In the three location-specific urban
12    case studies, the temporally and spatially varying concentrations were rolled down to conditions
13    just meeting each of the potential alternative standards that they exceeded (2007 REA, section
14    5.2.2.1).38
15           For the generalized (local) urban case study, which has a single exposure zone in which
16    air Pb concentrations do not vary spatially, we derived a single air Pb concentration estimate to
17    meet the standard assessed (e.g., specified maximum monthly or quarterly average). To reflect
18    the variability in air Pb concentrations that occur over time scales less than a year as a result of
19    temporal changes in meteorology and source and emission characteristics, the annual average air
20    concentration (input for IEUBK and dust model) was derived for the maximum monthly and
21    quarterly average metrics assessed using relationships based on the available Pb-TSP monitoring
22    data for large U. S. urban areas (2007 REA, Appendix A).
23
             38 When concentrations in the exposure zone (within the study area) that has the highest Pb concentrations
      (in terms of the metric being assessed) achieve a maximum quarterly or monthly average of the specified level, the
      potential standard is "just met".
         January 2013                           3-47            Draft - Do Not Quote or Cite

-------
 1   Table 3-8.   Air quality scenarios assessed.
Air Quality Conditions Just Meeting ...
Maximum
Quarterly Average
(pg/m3)
1.5B
0.2C




Maximum
Monthly Average
(pg/m3)


0.5
0.2
0.05
0.02
Case Studies Where Simulated
Generalized
(local)
Urban*
V
V
V
V
V
V
Location Specific
Primary
Smelter
V
V
V
V
V
V
Cleveland
V

V
V
V
V
Chicago
V


V
V
V
Los
Angeles
V



V
V
A - Conditions were set to meet the standards assessed in the single exposure zone of this case study.
B - Concentrations were proportionally rolled down to conditions just meeting the previous standard in the primary Pb
smelter case study; concentrations in the three urban location-specific case studies were proportionally rolled up to just
meet this standard.
C - Concentrations were proportionally rolled down to just meet this and the other potential alternative standards.
 3          The approaches for estimating Pb concentrations in other media varied depending on the
 4   type of case study (see section 3.4.3.1 above). Limitations in the available data and modeling
 5   tools precluded simulation of linkages between some media and air Pb, such that the full impact
 6   of changes in air Pb conditions associated with attainment of lower standards was not simulated.
 7   For example, dietary and drinking water Pb concentrations, as well as soil Pb concentrations,
 8   were not varied across the air quality scenarios in any case study (see Table 3-7).  For all case
 9   studies, however, indoor dust Pb concentrations were  simulated to change with the different air
10   quality scenarios that also provided differing ambient air Pb concentrations (outdoors and
11   indoors).

12         3.4.3.3  Methods for Deriving Risk Estimates
13          In this section, we first summarize the full risk model employed in the 2007 REA for
14   estimating risk for a broad range of air quality scenarios (section 3.4.3.3.1). Then, in section
15   3.4.3.3.2, we summarize approaches by which we have identified risk estimates pertaining to the
16   current standard, the second of which involves an analysis newly completed in this review in
17   which risk estimates are interpolated for the current standard based on the 2007 REA risk
18   estimates.
19
         January 2013
3-48
Draft - Do Not Quote or Cite

-------
 1         3.4.3.3.1  Full Risk Model in 2007 REA
 2           The risk characterization step employed in the 2007 REA involved generating a
 3    distribution of IQ loss estimates for the set of children simulated in the exposure assessment.
 4    Specifically, estimated PbB levels (for the concurrent PbB metric)39 were combined with four
 5    PbB concentration-response (C-R) functions for IQ loss based on the analysis by Lanphear et al
 6    (2005) of a pooled international dataset of blood Pb and IQ (see 2007 REA, section 5.3.1.1).
 7    Four different C-R functions were selected to provide different characterizations of behavior at
 8    low exposures.  The decision to use four different functions is in recognition of uncertainty
 9    related to modeling this endpoint, particularly at lower PbB levels for which there is limited
10    representation in the Lanphear et al (2005) pooled dataset; the 5th percentile for the concurrent
11    PbB measurements in that dataset is 2.5 |ig/dL, and the mean is 9.7 |ig/dL (73 FR 66978).  The
12    four different functions are either based directly on the lognormal model described in Lanphear
13    et al (2005), or they are derived from data presented in that study. The four functions are
14    presented in Figure 3-4 and compared in Table 3-9 with regard to total IQ loss and incremental
15    IQ loss (IQ loss per |ig/dL blood Pb) across a range of concurrent PbB levels. A brief
16    description of each of the functions is also here:
17           • Log-linear with cutpoint: log-linear function derived from the pooled analysis applied
18              down to 1 |ig/dL (concurrent PbB metric) with no IQ loss projected below that
19              exposure level.
20           • Log-linear with low-exposure linearization: log-linear function applied down to  1
21              Hg/dL (concurrent PbB metric) with linearization of the slope at that point which is
22              used to project IQ loss down to the origin.
23           • Dual linear - stratified at 10 |ig/dL:  function developed by fitting a two-piece linear
24              function stratified at 10 |ig/dL (peak PbB metric) to the log-linear function developed
25              from the pooled analysis.
26           • Dual linear-stratified at 7.5 |ig/dL: as above, but based on stratification of the two-
27              piece function at 7.5 |ig/dL (peak PbB metric).
             39 Risk estimates were also developed for lifetime average PbB levels using concentration-response
      functions derived from the Lanphear et al (2005) analysis for lifetime average PbB levels. Estimates based on the
      concurrent PbB metric are given primary emphasis due to the slightly more significant association found for
      concurrent PbB with IQ by Lanphear et al (2005) in addition to advice from CASAC.
         January 2013                           3-49             Draft - Do Not Quote or Cite

-------
1
2
3
4
5
 6
 1
 8
 9
10
11
12
14
12

1 0
8 -
 °
8
0 6

4

2
o



/'

/
*
/ ...--'•-•••
/ .---'""
/' .,'-•"'
/ - - ' ' - - - -
/',-•' _..--•-""""

'•'"' ' ^^^
•/' ^L^^—
,^7^
01 234567891
Concurrent blood Pb (ug/dL)











D




	 log-linear w ith outpoint
10 ug/dL, peak
	 	 - dual linear - stratified at
7.5 ug/dL, peak

exposure linearization



     Figure 3-4. Comparison of four concentration-response functions used in risk assessment.
     Table 3-9.  Comparison of total and incremental IQ loss estimates for blood Pb below 10
                 ug/dL based on the four concentration-response functions.
Performance Metric
Total IQ loss
Incremental IQ loss
(average # points
per |jg/dL)
at 2 |jg/dl_
at 5 pg/dL
at 7.5 |jg/dL
at 10 |jg/dL
<2 pg/dL
<5 pg/dL
<7.5 pg/dL
<10|jg/dL
Concentration-Response Function
Log-linear
with outpoint
Log-linear with
low-exposure
linearization
Dual linear -
stratified at 10
pg/dL peak
Dual linear -
stratified at 7.5
pg/dL peak
Points, IQ loss
1.9
4.3
5.4
6.2
0.94
0.87
0.73
0.62
4.6
7.0
8.1
8.9
2.29
1.41
1.09
0.89
1.6
3.9
4.3
4.6
0.80
0.80
0.58
0.47
5.9
11.1
11.5
11.9
2.94
2.24
1.55
1.20
           Of the four C-R functions provided above, we have the greatest overall confidence in risk
    estimates generated using the log-linear with low-exposure linearization (LLL) function because
    this function is (a) nonlinear, describing greater response per unit blood Pb at lower blood Pb
    levels consistent with multiple studies, (b) is based on fitting a function to the entire pooled
    dataset (and hence uses all of the data in describing response across the range of exposures), (c)
    is supported by sensitivity analyses showing the model coefficients to be robust (Lanphear et al.,
         January 2013
                                              3-50
Draft - Do Not Quote or Cite

-------
 1    2005), and (d) provides an approach for predicting IQ loss at the lowest exposures simulated in
 2    the assessment (which for some simulated children yield blood Pb levels below those studied).
 3    Risk estimates generated using the other three C-R functions are also presented to provide
 4    perspective on the impact of uncertainty in this key modeling step. For additional detail on the
 5    rationale for placing greater emphasis on the LLL function, see section 4.2.1 of the 2007 Pb Staff
 6    Paper (USEPA, 2007b).
 7           Two categories of risk metrics were generated for each of the location-specific case
 8    studies:
 9           •  Population riskpercentiles: The IQ loss associated with policy-relevant exposure
10              pathways for specific percentiles of the child population (e.g., the 50th, 90l and 95th
11              percentile modeled child). This category of metric provides perspective on the
12              distribution of IQ loss resulting from policy-relevant exposure pathways, ranging
13              from the typical or average child (50th percentile, mean) to children experiencing
14              higher exposures (90th, 95th percentiles).  Greater emphasis has been placed on the
15              median IQ loss estimates due to increased confidence in these estimates relative to the
16              higher percentile estimates, as noted in section 3.4.7.
17           •  Child frequency counts associated with specific riskpercentiles: Number of children
18              associated with each of the population percentiles (e.g., the number of children
19              predicted to have risk levels at  or above the 95th percentile).  This risk metric provides
20              a perspective on the number of children associated with various levels of IQ loss for a
21              particular case study.
22          For the generalized (local) urban case study, only the first type of risk metric, population
23    risk percentiles, was developed because a specific location with associated demographic data
24    was not modeled. In summarizing risk  estimates from the 2007 REA in this document,  we have
25    focused on the first category of risk metric, and specifically, on the median IQ loss estimates.
26          3.4.3.3.2  Air Quality Scenarios Reflecting the Current Standard
27          As the 2007 REA did not include an air quality scenario simulated to just meet the
28    standard selected by the 2008 decision,40 we have considered two approaches for identifying risk
29    estimates pertaining to conditions just meeting the now-current Pb standard for our purposes in
30    this review.  We first reviewed all the scenarios analyzed  in the 2007 REA and recognize the
31    similarity to the current standard of the  then-current conditions scenario for the Chicago case
32    study. Accordingly, we consider the risk estimates for that scenario for our purposes in this
33    review of considering risk associated with the current standard (see section 3.4.5 below).
34    Additionally, in recognition of the variation among specific locations and urban areas with
             40 The 2008 decision on the level for the revised NAAQS was based primarily on consideration of the
      evidence-based air-related IQ loss framework.  Although the specific level, averaging time and form chosen for the
      new standard were not among the air quality scenarios that had been simulated in that review, the risk estimates
      available for the range of simulated scenarios were concluded to be roughly consistent with and generally supportive
      of the evidence-based air-related IQ loss estimates (73 FR 67006; see section 4.1.1 below).
         January 2013                           3-51            Draft - Do Not Quote or Cite

-------
 1    regard to air quality patterns and exposed population, we have also newly developed estimates
 2    for an air quality scenario just meeting the current Pb NAAQS in the context of the generalized
 3    (local) urban case study to augment the risk information available in this current review. The
 4    newly developed estimates were derived based on interpolation from the risk estimates available
 5    for scenarios previously assessed for the generalized (local) urban case study. Such interpolated
 6    estimates were only developed for the generalized urban case study due to its use of a single
 7    exposure zone which greatly simplified the method employed, thus contributing relatively lesser
 8    uncertainty from the interpolation step.41
 9           In newly developing estimates for the current standard in the generalized (local) urban
10    case study, the general approach we followed was to identify the two alternative standard
11    scenarios simulated in the 2007 REA which represented air quality conditions bracketing those
12    for the current standard and then linearly interpolate an estimate of risk for the current standard
13    based on the  slope created from the two bracketing  estimates.  In representing air quality
14    conditions for these purposes,  we focused on the annual average air Pb concentration estimates
15    used as IEUBK model inputs for the various air quality scenarios. An annual average
16    concentration estimate to represent the current standard was identified in a manner consistent
17    with that employed in the 2007 REA for this case study (see section 3.4.3.2 above) with the use
18    of currently available monitoring data for relationships between air quality metrics for
19    representation of the current standard.  By this method, the air quality scenario for the current
20    standard (0.15 |ig/m3, as a not-to-be-exceeded 3-month average) was found to be bracketed by
21    the scenarios for alternative standards of 0.20 |ig/m3 (maximum calendar quarter average) and
22    0.20 |ig/m3 (maximum monthly average). A risk estimate for the current standard was then
23    derived using the slope relating general urban case study IQ loss to annual average Pb
24    concentration used for those two air quality scenarios. We used this interpolation approach to
25    develop median risk estimates for the current standard based on each of the four C-R functions.
26    Details on the method for the interpolation approach are provided in Appendix 3 A.  The
27    interpolated median estimates  of risk for the current standard for the generalized (local) urban
28    case study are provided below in Table 3-9.
             41 We did not complete interpolation of risk estimates for the current standard for the other case studies
      (i.e., the primary Pb smelter and location-specific urban case studies) because those case studies utilized a more
      complex, spatially-differentiated and population-based approach for which precludes application of the simple linear
      interpolation approach described, without introduction of substantial added uncertainty.  The simplicity  of the
      generalized (local) urban study area, however, with its single exposure zone, is amenable to the linear interpolation
      of risk described here.
         January 2013                            3-52              Draft - Do Not Quote  or Cite

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
            3.4.4  Challenges in Characterizing Air-related Exposure and Risk
             In estimating the portion of total (all-pathway) PbB and IQ loss attributable to air-related
      pathways,  we faced a variety of challenges.  Although we parsed total estimates into those for
      diet/drinking water and two air-related categories, referred to as "recent air" and "past air",
      significant limitations in our modeling tools and data resulted in an inability to parse specific risk
      estimates into specific pathways.  Although Pb in diet and drinking water sources may include
      Pb derived from Pb in the ambient air (as well as Pb from nonair sources), limitations precluded
      explicit modeling of the contribution from air pathways to these exposure pathways, such that
      the air-related component of these exposures was not estimated.42 As a result we utilized the
      estimates from recent and past air categories to create bounds within which we consider air-
      related risk to fall, as  illustrated in Figure 3-5 and described further below.
             "Recent air"
                              Ambient air
                              Indoor dust
              1 newly emitted lead
              1 resuspension of historically
                 emitted and deposited lead
                                                "Past air"
                                                              Indoor dust (other)
                                                             Outdoor soil
                                                • historically emitted Pb (still in home
                                                and/or still available for contact in
                                                external soil/dust)
                                                •
                                                                                  • Diet
                                                                                  • Drinking water
            • newly emitted Pb
            • historically emitted Pb
             Total Pb risk = recent air pathways + past air pathways + other pathways, including nonair pathways
             • Risk assessment simulated just meeting alternative NAAQS by changing recent air exposures.
             •Although unquantified, alternative NAAQS expected to also influence, to some extent, past air exposures
             and diet.
             > Air-related risk estimated to fall within range bounded by estimates of recent air and recent air + past air.
13    Figure 3-5.  Parsing of air-related risk estimates.
14           Those Pb exposure pathways tied most directly to ambient air, which consequently have
15    the potential to respond relatively more quickly to changes in air Pb (i.e., inhalation and
16    ingestion of indoor dust Pb derived from the infiltration of ambient air Pb indoors), were placed
17    into the "recent air" category. The other air-related Pb exposure pathways, all of which are
18    associated with atmospheric deposition, were placed into the "past air" category.  These include
             42 Further, although paint is not an air-related source of Pb exposure, for this analysis, may be reflected
      somewhat in estimates developed for the "past air" category, due to modeling constraints.  Fore example, technical
      limitations of the indoor dust Pb modeling may contribute to paint-related Pb in the "past air" component of indoor
      dust Pb and limitations in the available data and modeling may contribute to paint-related Pb in estimates of soil Pb.
         January 2013
                                                    3-53
Draft - Do Not Quote or Cite

-------
 1    ingestion of Pb in outdoor dust/soil and ingestion of the portion of Pb in indoor dust that after
 2    deposition from ambient air outdoors is carried indoors with humans. Additionally, while we
 3    recognized the potential for these other air-related exposures to be affected (over some time
 4    frame) by an adjustment to the Pb NAAQS, limitations in our date and tools precluded our
 5    simulation of that relationship with air Pb levels.
 6          Among the limitations affecting our estimates for the air-related categories is the
 7    apportionment of nonair pathways. For example, while conceptually indoor Pb paint
 8    contributions to indoor dust Pb would be considered background and included in a "background"
 9    category for this assessment, due to technical limitations related to indoor dust Pb modeling,  dust
10    from Pb paint was included as part of "other" indoor dust Pb (i.e., as part of past air exposure).
11    The inclusion of indoor paint Pb as a component of "other" indoor dust Pb (and consequently as
12    a component of the "past air" category) represents a source of potential high bias in our
13    prediction of exposure and risk associated with the "past air" category because, indoor paint Pb
14    is a nonair Pb source. At the same time, Pb in ambient air does contribute to the drinking water
15    and diet exposure pathways, and is likely a substantial contribution to diet.  We could not
16    separate the air contribution from the nonair contributions in the drinking water and diet
17    pathways. As a result, our risk estimate for the drinking water/diet category of pathways
18    includes some air-related risk representing a source of potential low bias in our predictions of
19    air-related risk.
20          Further,  we note that in simulating reductions in exposure associated with reducing
21    ambient air Pb levels through alternative NAAQS (and increases in exposure if the current
22    NAAQS was reached in certain case studies) only the exposure pathways categorized as "recent
23    air" (inhalation and ingestion of that portion of indoor dust associated with outdoor ambient air)
24    were varied with changes in air concentration.  The assessment from the previous review did not
25    simulate decreases in "past air" exposure pathways (e.g., reductions in outdoor soil Pb levels
26    following reduction in ambient air Pb levels and a subsequent decrease in exposure through
27    incidental soil ingestion and the contribution of outdoor soil to indoor dust).  These exposures
28    were held constant across all air quality scenarios.  This modeling/data limitation,  accordingly,
29    precluded estimates for this category from reflecting any impact of alternative standard levels
30    simulated.
31          In summary, because of limitations in the assessment design, data and modeling tools,
32    our risk estimates for the "past air" category in the last review include both risks that are truly
33    air-related and potentially,  some nonair risk. Because we could not sharply separate Pb linked to
34    ambient air from Pb from other (nonair) sources, some of the three categories of risk are
35    underestimated and others overestimated. On balance, we believe this limitation leads to a slight
36    overestimate of the risks in the "past air" category. At the same time, as discussed above, the

         January 2013                          3-54            Draft - Do Not Quote or Cite

-------
 1    "recent air" category does not fully represent the risk associated with all air-related pathways.
 2    Thus, we considered the risk attributable to air-related exposure pathways to be bounded on the
 3    low end by the risk estimated for the "recent air" category and on the upper end by the risk
 4    estimated for the "recent air" plus "past air" categories.  With regard to the latter, we are
 5    additionally cognizant of the modeling and data limitations which reduce the extent to which the
 6    upper end of these bounds reflects impacts of alternative air quality conditions simulated.
 7         3.4.5  Risk Estimates
 8           This section summarizes air-related risk estimates generated for the previous review and
 9    also risk estimates that have been newly derived in this review by interpolation from the previous
10    review estimates for the current standard (see section 3.4.3.3.2 above).  Included in this summary
11    is consideration of the following question:
12         •   What is the nature and magnitude of air-related risks remaining upon just
13              meeting the current Pb standard?
14           The air-related risk quantified in the 2007 REA is for IQ loss associated with Pb
15    exposure.  As discussed in section 3.2 above, the evidence in the last review, as well as in the
16    current review, supports identification of neurocognitive effects in young children as a
17    particularly sensitive endpoint for the exposure circumstances relevant to this review and which
18    addresses the important at-risk population, young children. With this support in the evidence for
19    quantification, the risk assessment quantified decrements in IQ,  an established indicator of
20    neurocognitive function. With regard to the nature of the risks, in addition to recognizing the
21    role of IQ as an indicator of an array of neurocognitive function impacts, we additionally take
22    note of the evidence regarding implications of neurocognitive impacts in young children with
23    regard to potential future impacts as the children age, as recognized in section 3.2 above.
24           In presenting risk estimates here, we focus on the median estimates of air-related IQ loss
25    for each case study. Estimates of air-related risk are substantially more uncertain for extremes of
26    the risk distribution, such as the 95th percentile. Those estimates and estimates for other risk
27    metrics, including population incidence for IQ loss at those case  studies with population
28    enumeration, are available elsewhere  (2007 REA, sections 4.2 and 5.3.2). In this section, Table
29    3-10 presents air-related IQ loss estimates derived in the 2007 REA for the full set of case
30    studies.  Table 3-11 provides a subset  of these risk estimates for the generalized (local) urban case
31    study in addition to estimates for air quality conditions just meeting the current standard, derived
32    by interpolation. A number of details, listed here, should be kept in mind when reviewing the
33    estimates presented in Tables 3-9 and 3-10.
34           •  The risk estimates represent IQ loss associated with air-related lead exposure for the
35              median child (exposure is  modeled through age seven years).

         January 2013                          3-55             Draft - Do Not Quote or Cite

-------
 1           • Our estimation of risk attributable to air-related exposure pathways is approximate, as
 2              described in sections 3.4.4 and 3.4.7.  We consider the air-related risk to fall within
 3              the ranges presented, bounded on the low end by estimates for the pathways
 4              categorized as "recent air" and on the upper end by the sum of the estimates for both
 5              the "recent" and "past air" pathways.43

 6           • The bolded range of risk estimates are derived using the C-R function in which we
 7              have the highest overall confidence (the log-linear with low-exposure linearization -
 8              see section 3.4.3.3.1). The wider range of risk estimates presented within the
 9              parentheses in both tables reflects the application of all four C-R functions (see
10              section 3.4.3.3.1) to exposure estimates for the recent and recent pluspast air
11              categories. Consequently, this range of risk estimates reflects both uncertainty in
12              estimation of air-related exposures as well as uncertainty in the C-R function for IQ
13              loss in children.
14
             43 The third category of pathways for which risk was estimated in the 2007 REA was comprised of diet and
      drinking water pathways. As other (nonair) sources of Pb can be appreciable contributors of Pb to these pathways,
      this category is referred to as "background" in the 2007 REA.

         January 2013                           3-56             Draft - Do Not Quote or Cite

-------
1   Table 3-10. Estimates of air-related risk from 2007 risk assessment.
4
5
Air Quality Scenario
Just meeting specified
maximum quarterly/monthly
average (ug/m3)
Median air-related IQ loss A
Generalized
(local)
urban case
study
Primary Pb
smelter
(subarea) case
study B> c
Location-s|
Cleveland
jecific urban ca
Chicago
se studies
Los Angeles
Alternative Standards Scenarios
1 .5, max quarterly D
(previous NAAQS)
0.5, max monthly
0.2, max quarterly
0.2, max monthly
0.05, max monthly
0.02, max monthly
3.5-4.8
(1.5-7.7)
1.9-3.6
(0.7 - 4.8)
1.5-3.4
(0.5-4.3)
1.2-3.2
(0.4 - 4.0)
0.5-2.8
(0.2 - 3.3)
0.3-2.6
(0.1-3.1)
<6
<(3.2 - 9.4)
<4.5
<(2.1-7.7)
<3.8
<(1.5-5.6)
<3.7
<(1.2-5.1)
<2.8
<(0.9 - 3.4)
<2.9
<(0.9 - 3.3)
2.8 - 3.9 E
(0.6 - 4.6)
0.6-2.9
(0.2 - 3.9)
0.43-2.8
(0.1-3.3)
0.6-2.8
(0.1-3.2)
0.1-2.6
(O.1-3.1)
<0.1-2.6
(<0. 1-3.0)
3.4- 4.7 E
(1.4-7.4)
F
F
0.6-2.9
(0.3-3.6)
0.2-2.6
(0.1-3.2)
0.1-2.6
(O.1-3.1)
2.7 - 4.2 E
(1.1-6.2)
F
F
0.7-2.9°
(0.2 - 3.5)
0.3-2.7
(0.1-3.2)
0.1-2.6
(O.1-3.1)
Then-current (2003-2005) Conditions
0.36, max quarterly
0.14, max quarterly
0.09, max quarterly






0.7-2.9
(0.2-3.6)



0.6 - 2.9
(0.3-3.5)



0.7-2.9
(0.2-3.5)
A - Air-related risk is bracketed by "recent air" (lower bound of presented range) and "recent" plus "past air" (upper bound of
presented range) (see section 3.4.4). Boldface estimates are generated using the log-linear with low-exposure linearization
function. Values in parentheses reflect the range of estimates associated with all four concentration-response functions.
B - In the case of the primary Pb smelter case study, only recent plus past air estimates are available.
C - Median air-related IQ loss estimates for the primary Pb smelter (full study area) range from <1 .7 to <2.9 points, with no
consistent pattern across simulated NAAQS levels. This lack of a pattern reflects inclusion of a large fraction of the study
population with relatively low ambient air impacts such that there is lower variation (at the population median) across standard
levels (see section 4.2 of the Risk Assessment, Volume 1).
D - This corresponds to roughly 0.7 - 1 .0 ug/m3 maximum monthly mean, across the urban case studies
E - A "roll-up" was performed so that the highest monitor in the study area is increased to just meet this level.
F - A "roll-up" to this level was not performed.
G - A "roll-up" to this level was not performed; these estimates are based on current conditions in this area (0.17 max monthly).
The Information in this table is drawn from the 2007 REA, Table 5-9.
       January 2013
3-57
Draft - Do Not Quote or Cite

-------
 1    Table 3-11.  Estimates of air-related risk for the generalized (local) urban case study,
 2                including interpolated estimates for current standard.
Air Quality Scenario
Just meeting specified metric (ug/m3)
Maximum
Quarterly
Average D
1.5
(previous
NAAQS)

0.2




Maximum
Monthly
Average

0.5


0.2
0.05
0.02
Maximum
3-month
Average



0.15B
(current
NAAQS)



Median Air-related IQ Loss A
for
Generalized (local)
Urban Case Study
3.5-4.8 (1.5-7.7)
1.9-3.6 (0.7-4.8)
1.5-3.4 (0.5-4.3)
1.5-3.4 (0.5-4.3)
1.2-3.2 (0.4-4.0)
0.5-2.8 (0.2-3.3)
0.3-2.6 (0.1-3.1)
A - Air-related risk is bracketed by "recent air" (lower bound of presented range) and
"recent" plus "past air" (upper bound of presented range) (see section 3.4.4 for additional
detail on these categories). Boldface values are estimates generated using the log-linear
with low-exposure linearization function. Values in parentheses reflect the range of
estimates associated with all four concentration-response functions.
B - Risk estimates interpolated - see text.
 4          Key observations regarding these air-related risk estimates across the array of air quality
 5    scenarios:

 6                                            All Case Studies

 1          •  Relative to the previous Pb NAAQS, substantial reduction in estimates of air- related
 8              risk is demonstrated across the full set of potential alternative standards simulated
 9              (Table 3-10). This is particularly the case for the lower bound (the recent air
10              estimates) which reflects only the pathways simulated to respond to changes in air
11              concentrations associated with different air quality scenarios.
12                                 Generalized (local) Urban Case Study

13          •  As described above, the general urban case study provides risk estimates for a single
14              group of children residing in a single area where air concentrations throughout area
15              are near the level of the standard being simulated. In this case study, air-related
16              median IQ loss, based on the higher confidence C-R function (bold), ranged from
17              roughly 2 to 4 IQ points for the 0.5  u.g/m3, maximum monthly average scenario, to <1
18              to roughly 3 IQ points for the 0.02 u.g/m3, maximum monthly average scenario.
19              These ranges are expanded somewhat with consideration of the full range of C-R
20              functions considered in the analysis. .
         January 2013
3-58
Draft - Do Not Quote or Cite

-------
 1                                     Location-specific Case Studies

 2           •  Compared to the other case studies, the air-related risk estimates for the location-
 3              specific urban case studies are lower because of the broader range of air-related
 4              exposures and the distribution of the population within the study areas. For example,
 5              the majority of the populations in each of the location-specific case studies reside in
 6              area8 with ambient air Pb levels wel1 below each standard assessed, particularly for
 7              standard levels above 0.05 |ig/m3, as a maximum monthly average. Consequently,
 8              risk estimates for these case studies indicate little response to alternative standard
 9              levels above 0.05|ig/m3 maximum monthly average (as shown in Table 3-10).

10           •  For the primary Pb smelter subarea, only an upper bound on risk attributable to air-
11              related exposures is provided due to uncertainties associated with dust Pb model used
12              for this case study.44 For all air quality scenarios, the population median recent plus
13              past air risk estimate is generally similar to or slightly higher than those for the
14              general urban case study, likely due to differences in the indoor dust models used for
15              the two case studies (discussed in Risk Assessment Report, sections 3.1.4).
16
17           Key observations regarding air-related risk estimates for the current standard:

18           •  The median air-related IQ loss estimate for then current conditions in Chicago study
19              area, which just met a level of 0.14 |ig/m3  as a maximum calendar quarter average,
20              falls somewhere within the lower and upper bounds of 0.6 and 2.9 points IQ loss,
21              respectively (Table 3-10).

22           •  The median air-related IQ loss estimate for the current standard in the Generalized
23              (local) Urban Case Study, newly derived by interpolation from 2007 REA results,
24              falls somewhere within the lower and upper bounds of 1.5 and 3.4 points IQ loss,
25              respectively (Table 3-11).  This estimate is derived by interpolation between the
26              estimates for the 0.2 |ig/m3 maximum quarterly and 0.2 |ig/m3 maximum monthly
27              average scenarios that were derived in 2007 REA. The newly interpolated estimate is
28              essentially the same as the estimate for 0.2 |ig/m3 maximum quarterly average
29              scenario for this  case study.

30           •  Based on results from the last review for a location-specific urban study area and on
31              those newly derived in this review based on interpolation from 2007 REA results,
32              median air-related IQ loss for the current standard is estimated, with rounding, to
33              generally fall above a rough lower bound of 1 point IQ loss and below a rough upper
34              bound of 3 points IQ loss.
35
36
             44 The regression model used for estimating dust Pb concentrations in the primary Pb smelter case study
      does not lend itself to partitioning the recent air Pb from other contributions.  Accordingly, this partitioning was not
      done for this case study (2007 REA, sections 2.4.3 and 3.1.4.2).

         January 2013                          3-59            Draft - Do Not Quote or Cite

-------
 1         3.4.6  Treatment of Variability
 2           This section discusses the degree to which the design of the previous risk assessment
 3    reflected consideration for key sources of variability associated with the scenarios evaluated
 4    (e.g., IQ loss in children associated with ambient Pb in various residential settings). In so doing,
 5    the following question is address.

 6         •  How are key sources of variability treated in the assessment?
 7
 8           Key sources of variability associated with the risk assessment include:

 9           •  Variation in ambient air Pb levels among U.S. urban residential areas: The location-
10              specific urban study areas were chosen to provide coverage for diverse residential
11              populations in U.S. urban areas with relatively elevated ambient air Pb levels (as
12              characterized using monitoring). We believe that the three cities included in those
13              case studies capture the variability in spatial patterns of ambient air Pb concentrations
14              within and across such areas reasonably well. In addition, the generalized (local)
15              urban case study provides coverage for a residential population exposed in a localized
16              area with air concentrations somewhat near the standard being evaluated, providing
17              coverage for a higher-end population risk scenario where a subset of urban children
18              are exposed without any spatial gradient in ambient air Pb levels.
19           •  Variation in the spatial distribution of children within an urban area and assignment
20              of ambient air Pb exposure:  Exposure of populations within the three location-
21              specific urban study areas was characterized to the US Census block group-level
22              characterization. Each block group was associated ambient air Pb exposure based on
23              proximity  to monitor nearest to block group centroid (Risk Assessment Report,
24              section 5.2.2). While there is uncertainty  associated with these exposures
25              (population-weighted through use of census block groups) - for example, we did not
26              consider time spent by children away from their residential block group - we believe
27              that this approach provides reasonable coverage for the potential pattern of interaction
28              between resident children located in these urban study areas and the associated spatial
29              gradients of ambient air Pb levels.
30           •  Variation among children in factors other  than media concentrations that influence
31              PbB levels: The inputs to the blood Pb model were central tendency estimates for
32              each exposure zone within each study area (pathway-specific estimates for central-
33              tendency child).  The IEUBK model then  generated PbB levels for that central-
34              tendency child within each exposure zone. These central-tendency PbB levels were
35              then combined with a GSD reflecting variability in PbB levels for young children
36              (at/near age 7 years) to generate a distribution of total PbB for a group of children
37              within each exposure zone in a given case study under the specific air quality
38              scenario. We have reasonable confidence that this modeling approach captures
39              variability in total Pb exposure (and consequently total PbB levels) for children
40              modeled for a given study area.  As noted in section 3.4.7 below, however, there is
41              uncertainty associated with parsing individual pathway contributions to total PbB
         January 2013                          3-60             Draft - Do Not Quote or Cite

-------
 1              levels (and consequently to total Pb-attributable risk). This is particularly true for
 2              high-end population percentile estimates of IQ loss.
 3           •  Inter-individual variability in IQ loss related to Pb exposure: The use of C-R
 4              functions based on a pooled analysis (Lanphear et al., 2005) which combined study
 5              populations from a number of different epidemiological studies of IQ loss in children
 6              likely provides reasonable coverage for variation across children regarding IQ loss
 7              attributable to Pb exposure. Furthermore, our estimation of risk based on several C-R
 8              functions provides further coverage for that variability, as well as also addressing
 9              uncertainty in the specification of that C-R functional form (see section 3.4.7 below).
10         3.4.7  Characterizing Uncertainty
11           Although the risk assessment utilized a number of innovative modeling elements in order
12    to generate representative estimates of risk for the various study area populations, like all risk
13    models there was uncertainty associated with the model and its output.
14         •   What are the important uncertainties associated with any risk/exposure
15             estimates?
16           One overarching area of uncertainty concerns the precision of our estimation of the
17    neurocognitive risk (as represented by IQ loss) associated with ambient air Pb. For example,
18    because of the evidence for a nonlinear response of PbB to Pb exposure and also for nonlinearity
19    in the C-R relationship for Pb-associated IQ loss, the assessment first estimated PbB levels and
20    associated risk for total Pb exposure (i.e., including Pb from  air-related and nonair exposure
21    pathways), and then separated out estimates for pathways of interest. We separated out the
22    estimates of total (all-pathway) PbB and IQ loss into three categories that included two air-
23    related categories ("past" and "recent"), in addition to a third category for diet and drinking
24    water.  However, significant limitations in our modeling tools affected our ability to develop
25    precise estimates for air-related exposure pathways.  As recognized in section 3.4.4 above, we
26    believe these limitations led to a slight overestimation of the risks for the past air category and to
27    an under representation of air-related  pathways for the recent air category. Thus, we
28    characterized the risk attributable to air-related exposure pathways to be bounded by the
29    estimates developed for the past air category and the sum of estimates for the recent air and past
30    air categories.  For air quality scenarios other than those for the previous NAAQS, this upper
31    bound is recognized as having a potential upward bias with regard to its reflection of the
32    simulated air quality conditions because modeling and data limitations precluded simulation of
33    the influence of lower air Pb concentrations on the outdoor dust and soil exposure pathways, as
34    noted in section 3.4.4 above.
35           Additional limitations, assumptions and uncertainties, recognized in various ways in the
36    assessment and presentation of results, along with a  concise characterization of their expected
37    impact on results, are listed below. The list begins with factors related to design of the
         January 2013                           3-61             Draft - Do Not Quote or Cite

-------
 1    assessment or of the case studies, followed by those related to estimation of Pb concentrations in
 2    ambient air, indoor dust, outdoor soil/dust, and blood, and estimation of Pb-related IQ loss.

 3           •  Temporal Aspects: During the 7-year exposure period, media concentrations remain
 4              fixed and the simulated child resides at the same residence (although exposure
 5              factors, including behavioral and physiological parameters, are adjusted to match the
 6              aging of the child). While these aspects of simulation introduce uncertainty into the
 7              risk estimates, it is not clear if there is a directional bias. For example, failure to
 8              consider population mobility during the simulation period could bias the overall
 9              distribution upwards unless children move to, or from a residential location with
10              higher Pb exposure, in which case this would not be the case.

11           •  Generalized (local) Urban Case Study: The design for this case study employs
12              assumptions regarding uniformity that are reasonable in the context of a general
13              description of a small neighborhood population, but would contribute significant
14              uncertainty to extrapolation of these estimates to a specific urban location,
15              particularly a relatively large one.  Thus, the risk estimates for this case study, while
16              generally representative of an urban residential population exposed to the specified
17              ambient air Pb levels, cannot be readily related to  a specific large urban population.
18              As long as these important caveats are considered in interpreting the risk estimates
19              (i.e., these risk estimates are likely represent a relatively small portion of the resident
20              child population in any given city just meeting specific air quality conditions) then
21              the potential error of extrapolating these results to a larger child population can be
22              avoided.  An additional area of uncertainty is with regard to the representation of
23              variability in air quality.  Given the relatively greater variability common in areas of
24              high Pb concentrations, the approach used to reflect variability may bias the estimates
25              high, although there is uncertainty with regard to the representativeness of the
26              monitoring dataset used to characterize this.
27           •  Location-specific Urban Case Studies: Limitations in the spatial density of ambient
28              air monitors in the three metropolitan  areas simulated limit our characterization of
29              spatial gradients of ambient air Pb levels in these case studies.  While this factor
30              introduces uncertainty into the risk estimates for this category of case study, it is not
31              clear if there is a directional  bias.

32           •  Air Quality Simulation: The proportional roll-up and roll-down procedures used in
33              some case studies to simulate air quality conditions just meeting the previous
34              NAAQS and alternative NAAQS, respectively, assume proportional changes  in air
35              concentrations across those case study areas  to create those air quality scenarios.  The
36              EPA recognizes the uncertainty with our simulation of higher air Pb concentrations
37              that would just meet the previous NAAQS in the urban location-specific case studies,
38              as well as the uncertainty in  simulation of conditions associated with the
39              implementation of emissions reduction actions to meet a lower standard. There is the
40              potential that the use of the proportional approach for adjusting monitor values
41              introduced a degree of high-bias into estimates of risk reduction. This could result if
42              we find that urban areas can target specific Pb sources impacting individual monitors,
43              thereby avoiding the need for more generalized reduction strategies resulting  in a
44              uniform pattern of reduction across all monitors in an urban area. It is important to

         January 2013                           3-62            Draft - Do Not Quote or Cite

-------
 1              point out however, that the generalized urban case study is not effected by this
 2              potential bias, since, given the single exposure zone involved, the single ambient air
 3              Pb level is fixed to reflect meeting the standard being considered (i.e., no spatial
 4              gradient is simulated in the small, localized residential area being modeled).

 5           •  Outdoor Soil/Dust Pb Concentrations: Limitations in datasets on Pb levels in surface
 6              soil/dust Pb in urban areas and in our ability to simulate the impact of reduced air Pb
 7              levels related to lowering the NAAQS contributes uncertainty to air-related risk
 8              estimates. In this case, it is likely that we have low-biased our estimates of risk
 9              reduction associated with alternative (lower) Pb NAAQS levels, since we have not
10              simulated potential changes in soil Pb related to changes in ambient air Pb.
11           •  Indoor Dust Pb Concentrations:  Limitations and uncertainly in modeling of indoor
12              dust Pb levels, including the impact of reductions in ambient air Pb levels,  contributes
13              uncertainty to air-related risk estimates.  Although our modeling of indoor dust does
14              link changes in ambient air Pb to changes in indoor dust Pb (via air exchanges and
15              indoor deposition onto surfaces), the modeling does not include a link between
16              ambient air Pb, outdoor soil Pb and subsequent changes in the level of Pb carried (or
17              "tracked") into the house. This could introduce low-bias into our total estimates of
18              air-related Pb exposure and risk.

19           •  Interindividual Variability in PbB Levels: Uncertainty related to population
20              variability in PbB levels (i.e., interindividual variability in factors other than media
21              concentration that influence PbB) and limitations in modeling of this introduces
22              significant uncertainty into PbB and IQ  loss estimates for the 95th percentile of the
23              population. We are not aware of any systematic bias introduced into the analysis from
24              this source of uncertainty.
25           •  Pathway Apportionment for Higher Percentile PbB and risks: Limitations, primarily
26              in data, prevented us from characterizing the degree of correlation among high-end
27              Pb exposures for the various pathways (e.g., the degree to which an individual
28              experiencing high drinking water Pb exposure would also experience high Pb paint
29              exposure and high ambient air-related Pb exposure).  Our inability to characterize
30              potential correlations between exposure pathways (particularly at the higher
31              percentile exposure levels) limited our ability to (a) effectively model high-end Pb
32              risk and (b) apportion that risk between  different exposure pathways, including
33              ambient air-related pathways.
34           •  IQ Loss Concentration-response Functions:  Specification of the quantitative
35              relationship between PbB level and IQ loss is subject to greater uncertainty at lower
36              PbB levels (e.g., particularly below 2.5  |ig/dL concurrent PbB).  However, we believe
37              that by considering four different models (which each treat the response at  low PbB
38              levels in a different manner), we have completed a reasonable characterization  of this
39              source of uncertainty and its impact on risk estimates. Given comparison of risk
40              estimates generated using the four models, it would appear that this source of
41              uncertainty has a potentially significant  impact on risk.
         January 2013                          3-63            Draft - Do Not Quote or Cite

-------
 1         3.4.8  Updated Interpretation of Risk Estimates
 2           As summarized in prior sections, a range of information gaps and areas of uncertainty
 3    were associated with the information available in the last review.  In the REA Planning
 4    document, staff considered the degree to which information newly available since the last Pb
 5    NAAQS review, as summarized in the draft ISA, might address specific uncertainties associated
 6    with the 2007 REA, such that an updated risk model might be developed with the potential to
 7    provide new exposure and risk estimates substantially different45 from estimates generated in
 8    2007 (USEPA, 2011).  Staff concluded that the newly available information did not provide the
 9    means by which to develop an updated or enhanced risk model that would substantially improve
10    the utility of risk estimates in informing the Pb NAAQS review.  Specifically it was concluded
11    that none of the primary sources of uncertainty indentified to have the greatest impact on risk
12    estimates would be substantially reduced through the use of newly available information
13    (USEPA, 2011).
14           Our ongoing review of the newly available  information leads us to conclude at this time,
15    that the key observations regarding air-related Pb risk modeled for the set of standard levels
16    covered in the 2007 REA, as well as the risk estimates interpolated for the current standard (as
17    discussed in Section 3.4.5) are not significantly affected by the new information.   Our overall
18    characterization of uncertainty and variability associated with those estimates (as described
19    above in sections 3.4.6 and 3.4.7) is not appreciably affected by new information. As recognized
20    at the time of the last review, exposure and risk modeling conducted for this analysis was
21    complex and subject to significant uncertainties due to limitations, data,  and models, among
22    other aspects. Further, limitations  in the assessment design, data and modeling tools
23    handicapped us from sharply separating Pb linked to ambient air from Pb that is not air-related.
24           In summary, the estimates of risk attributable to air-related exposures, with which we
25    recognize a variety of sources of uncertainty, are considered to be approximate, falling within
26    upper and lower bounds which over- and underestimate, respectively.  In scenarios for more
27    restrictive air quality conditions than those associated with the previous Pb standard, substantial
28    reductions in air-related risk were demonstrated. Focusing on the results for the generalized
29    (local) urban case study, the interpolated estimates for the scenario representing the current
30    standard are very similar to estimates for the two 0.2 |ig/m3 scenarios (maximum monthly and
31    quarterly averages) simulated in the 2007 REA,46 and are appreciably lower than those
             45 In this context, "substantially different" has been intended to mean that the degree of uncertainty is
      substantially reduced, or bias is addressed, such that the new risk estimates could convey a different message
      regarding the magnitude of public health impacts associated with the current or potential alternative standards.
             46 There is uncertainty associated with judging differences between the current standard and these potential
      alternative standards due to the difference in air quality  datasets used to estimate air concentration variability of the
      2007 REA estimates versus the interpolated risk estimate.
         January 2013                           3-64             Draft - Do Not Quote or Cite

-------
 1    associated with the previous standard.  In characterizing the magnitude of air-related risk
 2    associated with the current standard, we focusing on median estimates, for which we have
 3    appreciably greater confidence than estimates for outer ends of risk distribution (see section
 4    3.4.7) and on risks derived using the C-R function in which we have greatest confidence (see
 5    sections 3.4.3.3.1 and 3.4.7). The risk results for the current standard from the last review for
 6    one of the location-specific urban study area populations and those newly derived in this review
 7    based on interpolation from 2007 REA results for the generalized (local) urban case study, which
 8    is recognized to reflect a generalized high end of air-related exposure for localized populations,
 9    indicate air-related risk for the  current standard of a magnitude generally within the bounds of
10    roughly 1 and 3 points IQ loss.
         January 2013                           3-65            Draft - Do Not Quote or Cite

-------
  1     3.5   REFERENCES

  2     Amaral et al., 2010; Amaral, JH; Rezende, VB; Quintana, SM; Gerlach, RF; Barbosa, F, Jr; Tanus-Santos, JE.
  3            (2010). The relationship between blood and serum lead levels in peripartum women and their respective
  4            umbilical cords. Basic Clin Pharmacol Toxicol 107: 971-975. http://dx.doi.org/10.! 111/j.1742-
  5            7843.2010.00616.x

  6     Barry, PSI. (1975). A comparison of concentrations of lead in human tissues. Occup Environ Med 32: 119-139.

  7     Barry, PSI. (1981). Concentrations of lead in the tissues of children. Occup Environ Med 38: 61-71.

  8     Bellinger, D. C. and Needleman, H. L. (2003) Intellectual impairment and blood lead levels [letter]. N. Engl. J. Med.
  9            349:500.

10     Bellinger, D. 2008. Email message to Jee-Young Kim, U.S. EPA. February 13, 2008. Docket number EPA-HQ-
11            OAR-2006-0735.

12     Billick, I.H.; Curran, A.S.; Shier, D.R. (1979) Analysis of pediatric blood lead levels in New York City for 1970-
13            1976. Environ. Health Perspect. 31: 183-190.

14     Billick, I.H.; Curran, A.S.; Shier, D.R. (1980) Relation of pediatric blood lead levels to lead in gasoline. Environ.
15            Health Perspect. 34:213-217.

16     Billick, I.H. (1983) Sources of lead in the environment. In: Rutter, M.; Russell Jones, R., eds. Lead versus health:
17            sources and effects of low level lead exposure. New York, NY: John Wiley and Sons, Ltd; pp. 59-77.

18     Brunekreef, B.; Noy, D.; Biersteker, K.; Boleij, J. (1983) Blood lead levels of Dutch city children and their
19            relationship to lead in the environment. J. Air Pollut. Control Assoc. 33: 872-876.

20     Brunekreef, B. (1984) The relationship between air lead and blood lead in children: a critical review. Science of the
21            total environment, 38: 79-123.

22     Canfield, R. L.; Henderson, C. R., Jr.; Cory-Slechta, D. A.; Cox, C.; Jusko, T. A.; Lanphear, B. P. (2003)
23            Intellectual impairment in children with blood lead concentrations below 10 ug per deciliter. N. Engl.  J.
24            Med. 348: 1517-1526.

25     Canfield, R.L. 2008a. Email message to Jee-Young Kim, U.S. EPA. February 7, 2008. Docket number EPA-HQ-
26            OAR-2006-0735.

27     Canfield, R.L. 2008b. Email messages to Jee-Young Kim, U.S. EPA. August 11 and 12, 2008.  Docket number
28            EPA-HQ-OAR-2006-0735.

29     Carbone, R; Laforgia, N; Crollo, E; Mautone, A; lolascon, A. (1998). Maternal and neonatal lead exposure in
30            southern Italy. Neonatology 73: 362-366.

31     Centers for Disease Control (1991) Preventing lead poisoning in young children: a statement by the Centres for
32            Disease Control.  Atlanta, GA: U.S. Department of health and Human Services, Public Health Service;
33            October 1. http://wonder.cdc.gov/wonder/prevguid/p0000029/p0000029.asp

34     Centers for Disease Control and Prevention (2005) Preventing lead poisoning in young children: a statement by the
3 5            Centers for Disease Control and Prevention. Atlanta, GA: U.S. Department of Health  and Human Services,
36            Public Health Service. August.
          January 2013                              3-66              Draft - Do Not Quote or Cite

-------
  1     Centers for Disease Control and Prevention (2012) CDC Response to Advisory Committee on Childhood Lead
  2            Poisoning Prevention Recommendations in "Low Level Lead Exposure Harms Children: A Renewed Call
  3            of Primary Prevention". Atlanta, GA: U.S. Department of Health and Human Services, Public Health
  4            Service. June 7.

  5     Chandramouli, K; Steer, CD; Ellis, M; Emond, AM. (2009). Effects of early childhood lead exposure on academic
  6            performance and behaviour of school age children. Arch Dis Child 94:  844-848.
  7            http://dx.doi.org/10.1136/adc.2008.149955

  8     Chuang, HY; Schwartz, J; Gonzales-Cossio, T; Lugo, MC;  Palazuelos, E; Aro, A; Hu, H; Hernandez-Avila, M.
  9            (2001). Interrelations of lead levels in bone, venous blood, and umbilical cord blood with exogenous lead
10            exposure through maternal plasma lead in peripartum women. Environ Health Perspect 109: 527-532.
11            http://dx.doi.org/10.1289/ehp.Ol 109527

12     Franklin, CA; Inskip, MJ; Baccanale, CL; Edwards, CM; Manton, WI; Edwards, E; O'Flaherty, EJ. (1997). Use of
13            sequentially administered stable lead isotopes to investigate changes in blood lead during pregnancy in a
14            nonhuman primate (Macaca fascicularis). Toxicol  Sci 39: 109-119.

15     Frey, H.C. (2011) Letter from Dr. H. Christopher Frey, Chair, Clean Air Scientific Advisory Committee Lead
16            Review Panel, to Administrator Lisa P. Jackson. Re:  Consultation on EPA's Review of the National
17            Ambient Air Quality Standards for Lead: Risk and Exposure Assessment Planning Document. October  14,
18            2011.

19     Gulson, B; Mahaffey, KR; Mizon, KJ; Korsch, MJ; Cameron, MA; Vimpani, G. (1995). Contribution of tissue lead
20            to blood lead in adult female subjects based on stable lead isotope methods. Transl Res 125: 703-712.

21     Gulson, B; Mahaffey, KR; Jameson, CW; Patison, N; Law, AJ; Mizon, KJ; Korsch, MJ; Pederson, D. (1999).
22            Impact of diet on lead in blood and urine in female adults and relevance to mobilization of lead from bone
23            stores. Environ Health Perspect 107: 257-263. http://dx.doi.org/10.1289/ehp.99107257

24     Gulson, B; Mizon, KJ; Korsch, MJ; Palmer, JM; Donnelly,  JB. (2003). Mobilization of lead from human bone tissue
25            during pregnancy and lactation: A summary of long-term research. Sci  Total Environ 303: 79-104.
26            http://dx.doi.org/10.1016/80048-9697(02)00355-8

27     Gulson, B; Mizon, KJ; Palmer, JM; Korsch, MJ; Taylor, AJ; Mahaffey,  KR. (2004a). Blood lead changes during
28            pregnancy and postpartum with calcium supplementation. Environ Health Perspect  112: 1499-1507.
29            http://dx.doi.org/10.1289/ehp.6548

30     Gulson, BL; Mizon, KJ; Davis, JD; Palmer, JM; Vimpani, G. (2004b). Identification of sources of lead in children in
31            a primary zinc-lead smelter environment. Environ Health Perspect 112: 52-60.
32            http://dx.doi.org/10.1289/ehp.6465

33     Harville, E. W.; Hertz-Picciotto, I.; Schramm, M.; Watt-Morse, M.; Chantala, K.; Osterloh, J.; Parsons, P. J.;Rogan,
34            W. (2005) Factors influencing the difference between maternal and cord blood lead. Occup.Environ. Med.
35            62: 263-290.Hayes, E.B.,; McElvaine, M.D.; Orbach, H.G.; Fernandez, A.M.; Lyne, S.; Matte, T.D. (1994)
3 6            Long-term trends in blood lead levels among children in Chicago: Relationship to air lead levels. Pediatrics
37            93:195-200.

3 8     Henderson, R. (2007) Letter from Dr. Rogene Henderson, Chair, Clean  Air Scientific Advisory Committee, to
39            Administrator Stephen L.  Johnson. Re:  Clean Air Scientific Advisory Committee's (CASAC) Review of
40            the 1st Draft Lead Staff Paper and Draft Lead Exposure and Risk Assessments. March 27, 2007.

41     Hilts, S. R. (2003) Effect of smelter emission reductions on children's blood lead levels. Sci.  Total Environ. 303:  51-
42            58.
          January 2013                              3-67              Draft - Do Not Quote or Cite

-------
  1     Hornung, R. 2008a. Email message to Jee-Young Kim, U.S. EPA. February 11, 2008. Docket number EPA-HQ-
  2             OAR-2006-0735.

  3     Hornung, R. 2008b. Email message to Jee-Young Kim, U.S. EPA. August 19, 2008.  Docket number EPA-HQ-
  4             OAR-2006-0735.

  5     Hornung, RW; Lanphear, BP; Dietrich, KN. (2009). Age of greatest susceptibility to childhood lead exposure: A
  6             new statistical approach. Environ Health Perspect 117: 1309-1312. http://dx.doi.org/10.1289/ehp.0800426

  7     Jedrychowski, W; Perera, F; Maugeri, U; Miller, RL; Rembiasz, M; Flak, E; Mroz, E; Majewska, R; Zembala, M.
  8             (2011). Intrauterine exposure to lead may enhance sensitization to common inhalant allergens in early
  9             childhood: A prospective prebirth cohort study. Environ Res 111: 119-124.
10             http://dx.doi.0rg/10.1016/j.envres.2010.ll.002

11     Kim, Y; Kim, BN; Hong, YC; Shin, MS; Yoo, HJ; Kim, JW; Bhang, SY; Cho, SC. (2009). Co-exposure to
12             environmental lead and manganese affects the intelligence of school-aged children. Neurotoxicology 30:
13             564-571. http://dx.doi.0rg/10.1016/j.neuro.2009.03.012

14     Krieg, EF, Jr; Butler, MA; M-h, C; Liu, T; Yesupriya, A; Dowling, N; Lindegren, ML. (2010). Lead and cognitive
15             function in VDR genotypes in the Third National Health and Nutrition Examination Survey. Neurotoxicol
16             Teratol 32: 262-272. http://dx.doi.0rg/10.1016/i.ntt.2009.12.004

17     Lagerkvist, BJ; Ekesrydh, S; Englyst, V; Nordberg, GF; Soderberg, HA; Wiklund, DE. (1996). Increased blood lead
18             and decreased calcium levels during pregnancy: A prospective study of Swedish women living near a
19             smelter. Am J Public Health 86: 1247-1252.

20     Lanphear, BP; Roghmann, KJ. (1997). Pathways of lead exposure in urban children. Environ Res 74: 67-73.

21     Lanphear, BP; Matte, TD; Rogers, J; Clickner, RP; Dietz, B; Bornschein, RL; Succop, P; Mahaffey, KR; Dixon, S;
22             Galke, W; Rabinowitz, M; Farfel, M; Rohde, C; Schwartz, J; Ashley, P; Jacobs, DE. (1998). The
23             contribution of lead-contaminated house dust and residential soil to children's blood lead levels: A pooled
24             analysis of 12 epidemiologic studies. Environ Res 79: 51-68. http://dx.doi.org/10.1006/enrs. 1998.3859

25     Lanphear, BP; Dietrich, K; Auinger, P; Cox,  C. (2000). Cognitive deficits associated with blood lead concentrations
26             <10 microg/dL in US children and adolescents. Public Health Rep  115: 521-529.Lanphear, B. P.; Hornung,
27             R.; Khoury, J.; Yolton, K.; Baghurst, P.; Bellinger, D. C.; Canfield, R. L.; Dietrich, K. N.; Bornschein, R.;
28             Greene, T.; Rothenberg, S. J.; Needleman, H. L.; Schnaas, L.; Wasserman, G.; Graziano, J.; Roberts, R.
29             (2005) Low-level environmental lead exposure and children's intellectual function: an international pooled
30             analysis. Environ. Health Perspect. 113: 894-899.Manton, WI. (1985). Total contribution of airborne lead
31             to blood lead. Occup EnvironMed 42: 168-172. http://dx.doi.0rg/10.1136/oem.42.3.168

32     Miranda, ML; Kim, D; Reiter, J; Overstreet Galeano, MA; Maxson, P. (2009). Environmental contributors to the
33             achievement gap. Neurotoxicology 30: 1019-1024. http://dx.doi.0rg/10.1016/i.neuro.2009.07.012

3 4     National Center for Health Statistics .(2010)  National Center for Health Statistics. (2011) Health, United States,
35             2011: With Special Feature on Socioeconomic Status and Health. Hyattsville, MD.
36             2012http://www.cdc.gov/nchs/data/hus/husl l.pdf

37     Patel, AB; Prabhu, AS. (2009). Determinants of lead level in umbilical cord blood. Indian Pediatr 46: 791-
38             793.Pirkle, JL; Brody, DJ; Gunter, EW; Kramer, RA; Paschal, DC; Flegal, KM; Matte, TD. (1994). The
3 9             decline in blood lead levels in the United States: The National Health and Nutrition Examination Surveys
40             (NHANES). JAMA 272: 284-291. http://dx.doi.org/10.1001/jama.1994.03520040046039
          January 2013                               3-68              Draft - Do Not Quote or Cite

-------
  1     Ranft, U; Delschen, T; Machtolf, M; Sugiri, D; Wilhelm, M. (2008). Lead concentration in the blood of children and
  2             its association with lead in soil and ambient air: Trends between 1983 and 2000 in Duisburg. J Toxicol
  3             EnvironHealth A 71: 710-715. http://dx.doi.org/10.1080/15287390801985117

  4     Rice, DC. (1990). Lead-induced behavioral impairment on a spatial discrimination reversal task in monkeys exposed
  5             during different periods of development. Toxicol Appl Pharmacol 106: 327-333.
  6             http://dx.doi.org/10.1016/0041-008X(90)90251-O

  7     Rice, DC. (1992). Lead exposure during different developmental periods produces different effects on FI
  8             performance in monkeys tested as juveniles and adults. Neurotoxicology 13: 757-770.

  9     Rice, DC; Gilbert, SG. (1990). Sensitive periods for lead-induced behavioral impairment (nonspatial discrimination
10             reversal) in monkeys. Toxicol Appl Pharmacol 102: 101-109. http://dx.doi.org/10.1016/0041-
11             008X(90)90087-BRothenberg, S. J.; Khan, F.; Manalo, M.; Jian, J.; Cuellar, R.; Reyes, S.; Acosta, S.;
12             Jauregui, M.; Diaz, M.; Sanchez, M.; Todd, A. C.; Johnson, C. (2000) Maternal bone lead contribution to
13             blood lead during and after pregnancy. Environ. Res. 82: 81-90.

14     Schnaas, L; Rothenberg, SJ; Flores, MF; Martinez, S; Hernandez, C; Osorio, E; Perroni, E. (2004). Blood lead
15             secular trend in a cohort of children in Mexico City (1987-2002). Environ Health Perspect 112:1110-1115.
16             http://dx.doi.org/10.1289/ehp.6636

17     Schwartz, J., and Pitcher,  H. (1989) The relationship between gasoline lead and blood lead in the United States. J
18             Official Statistics 5(4):421-431.

19     Schwemberger, MS, JE Mosby, MJ Doa, DE Jacobs, PJ Ashley, DJ Brody, MJ Brown, RL Jones, D Homa. (2005)
20             Mortality and Morbidity Weekly Report 54(20):513-516. May 27, 2005

21     Simon, DL; Maynard, EJ; Thomas, KD. (2007). Living in a sea of lead changes in blood- and hand-lead of infants
22             living near a smelter. J Expo Sci Environ Epidemiol 17: 248-259. http://dx.doi.org/10.1038/sj.jes.7500512

23     Smith, D. R.; Osterloh, J.  D.; Flegal, A. R. (1996) Use of endogenous, stable lead isotopes to determine release of
24             lead from the skeleton. Environ. Health Perspect.  104: 60-66.

25     Surkan, PJ; Zhang, A; Trachtenberg, F; Daniel, DB; McKinlay, S; Bellinger, DC. (2007). Neuropsychological
26             function in children with blood lead levels <10 ug/dL. Neurotoxicology 28: 1170-1177.
27             http://dx.doi.org/10.1016/j.neuro.2007.07.007

28     Tellez-Rojo, M. M.; Bellinger, D. C.; Arroyo-Quiroz, C.; Lamadrid-Figueroa, H.; Mercado-Garcia, A.; Schnaas-
29             Arrieta, L.; Wright, R. O.; Hernandez-Avila, M.; Hu, H.  (2006) Longitudinal associations between blood
30             lead concentrations < 10 ug/dL and neurobehavioral development in environmentally-exposed children in
31             Mexico City. Pediatrics 118: e323-e330.

32     Tellez-Rojo, M. 2008. Email message to Jee-Young Kim, U.S. EPA. February 11, 2008. Docket number EPA-HQ-
33             OAR-2006-0735.

34     Tripathi, R.M.; Raghunath, R.; A.V. Kumar; V.N. Sastry; S. Sadasivan. (2001) Atmospheric and children's blood
3 5             lead as indicators of vehicular traffic and other emission sources in Mumbai, India. Sci Total Enviro 267:
36             101-108.

37     U.S. Environmental Protection Agency. (1986) Air quality  criteria for lead. Research Triangle Park, NC: Office of
3 8             Health and Environmental Assessment, Environmental Criteria and Assessment Office; EPA report no.
39             EPA-600/8-83/028aF-dF. 4v. Available from: NTIS, Springfield, VA; PB87-142378.

40     U.S. Environmental Protection Agency. (1989) Review of the national ambient air quality standards for lead:
41             Exposure analysis methodology and validation: OAQPS staff report. Research Triangle Park, NC: Office of


          January 2013                               3-69              Draft - Do Not Quote or Cite

-------
  1            Air Quality Planning and Standards; report no. EPA-450/2-89/011. Available on the web:
  2            http://www.epa.gov/ttn/naaqs/standards/pb/data/rnaaqsl eamv.pdf

  3    U.S. Environmental Protection Agency. (1990) Air quality criteria for lead: supplement to the 1986 addendum.
  4            Research Triangle Park, NC: Office of Health and Environmental Assessment, Environmental Criteria and
  5            Assessment Office; report no. EPA/600/8-89/049F. Available from: NTIS, Springfield, VA; PB91-138420.

  6    U.S. Environmental Protection Agency. (2006) Air Quality Criteria for Lead. Washington, DC, EPA/600/R-
  7            5/144aF. Available online at: http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_cr.html

  8    U.S. Environmental Protection Agency. (2007a) Lead: Human Exposure and Health Risk Assessments for Selected
  9            Case Studies, Volume I. Human Exposure and Health Risk Assessments - Full-Scale and Volume II.
10            Appendices. Office of Air Quality Planning and Standards, Research Triangle Park, NC. EPA-452/R-07-
11            014aandEPA-452/R-07-014b.

12    U.S. Environmental Protection Agency. (2007b) Review of the National Ambient Air Quality Standards for Lead:
13            Policy Assessment of Scientific and Technical Information, OAQPS Staff Paper. Office of Air Quality
14            Planning and Standards, Research Triangle Park, NC. EPA-452/R-07-013. Available at:
15            http://www.epa.gov/ttn/naaqs/standards/pb/s_pb  cr sp.html

16    U.S. Environmental Protection Agency. (2011) Review of the National Ambient Air Quality Standards for Lead:
17            Risk and Exposure Assessment Planning Document. Office of Air Quality Planning and Standards,
18            Research Triangle Park, NC. EPA/452/P-11 -003. Available at:
19            http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_2010_pd.html

20    U.S. Environmental Protection Agency. (2012) Integrated Science Assessment for Lead (Third External Review
21            Draft). Washington, DC, EPA/600/R-10/075C. Available online at:
22            http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_2010_isa.html

23    Zielhuis, R.L.; del Castilho, P.; Herber, R.F.M.; Wibowo, A.A.E.; Salle, H.J.A. (1979) Concentrations of lead and
24            other metals in blood of two and three year-old children living near a secondary smelter. Int. Arch. Occup.
25            Environ. Health 42: 231-239.
          January 2013                              3-70              Draft - Do Not Quote or Cite

-------
 1                4   REVIEW OF THE PRIMARY STANDARD FOR LEAD

 2          This chapter presents preliminary staff conclusions regarding the primary Pb standard.
 3    These preliminary staff conclusions are guided by consideration of key policy-relevant questions
 4    and based on the assessment and integrative synthesis of information presented in the ISA and by
 5    staff analyses and evaluations presented in chapters 2 and 3 herein. Final evaluations and staff
 6    conclusions, to be developed taking into consideration CASAC advice and public comment on
 7    this draft document, will be presented in the final PA. The final evaluations and staff
 8    conclusions will inform the Administrator's decisions on whether to retain or revise the existing
 9    primary  standard for Pb.
10          Following an introductory section on the general approach for reviewing the primary
11    standard (section 4.1), including a summary of considerations in the last review, the discussion in
12    this chapter focuses on the central issue of whether the information available in this review
13    supports or calls into question the adequacy of the current primary standard. Building on the
14    responses to specific policy-relevant questions on the scientific evidence and exposure-risk
15    information in chapter 3 above, presentation in section 4.2 is also organized into consideration of
16    key policy-relevant questions framing evidence-based and exposure/risk-based considerations.
17    The policy-relevant questions in this document are based on those included in the IRP (IRP,
18    section 3.1).  In section 4.3, preliminary staff conclusions are developed. Section 4.4 presents a
19    brief overview of key uncertainties and areas for future research.

20    4.1   APPROACH
21          Staffs approach in this review of the current primary standard takes into consideration
22    the approaches used in the previous Pb NAAQS review.  The past and current approaches
23    described below are both based, most fundamentally, on using EPA's assessment of the current
24    scientific evidence and associated quantitative analyses to inform the Administrator's judgment
25    regarding a primary standard for Pb that protects public health with an adequate margin of safety.
26    We note that the final decision on the adequacy of the current standard is largely a public health
27    policy judgment to be made by the Administrator. The Administrator's final decision must draw
28    upon scientific information and analyses about health effects, population exposure and  risks, as
29    well as judgments about how to consider the range and magnitude of uncertainties that  are
30    inherent in the scientific evidence and analyses.  Our approach to informing these judgments,
31    discussed more fully below, is based on the recognition that the available health effects evidence
32    generally reflects a continuum, consisting of ambient levels at which scientists generally agree
33    that health effects are likely to occur, through lower levels at which the likelihood and magnitude
34    of the response become increasingly uncertain. This approach is consistent with the

         January 2013                           4-1             Draft - Do Not Quote or Cite

-------
 1    requirements of the NAAQS provisions of the Act and with how the EPA and the courts have
 2    historically interpreted the Act.  These provisions require the Administrator to establish primary
 3    standards that, in the Administrator's judgment, are requisite to protect public health with an
 4    adequate margin of safety.  In so doing, the Administrator seeks to establish standards that are
 5    neither more nor less stringent than necessary for this purpose. The Act does not require that
 6    primary  standards be set at a zero-risk level, but rather at a level that avoids unacceptable risks to
 7    public health.  The four basic elements of the NAAQS (indicator, averaging time, level, and
 8    form) are considered collectively in evaluating the health protection afforded by the current
 9    standard.
10          The following subsections include background information on the approach used in the
11    previous review of the standard (section 4.1.1) and a  discussion of the approach for the current
12    review (section 4.1.2).

13         4.1.1  Approach Used in the Last Review
14          The last review of the NAAQS for Pb was completed in 2008 (73 FR 66964). In
15    consideration of the much-expanded health effects evidence on neurocognitive effects of Pb in
16    children available at that time, the EPA substantially  revised the primary standard from 1.5
17    ng/m3, as a not-to-be exceeded average concentration over a calendar quarter, to a level of 0.15
18    ng/m3, as a not-to-be-exceeded rolling 3-month average concentration. The 2008 decision to
19    revise the primary standard was based on the extensive body of scientific evidence published
20    over almost three decades, from the time the standard was originally set in 1978 through 2005-
21    2006.  The 2008 decision considered the body of evidence as assessed in the 2006 CD (USEPA,
22    2006) as well as the 2007 Staff Paper assessment of the policy-relevant information contained in
23    the CD and the quantitative risk/exposure assessment (USEPA, 2007a, 2007b), the advice and
24    recommendations of CAS AC (Henderson 2007a, 2007b, 2008a, 2008b), and public comment.
25          While recognizing that Pb has been demonstrated to exert "a broad array of deleterious
26    effects on multiple organ systems", the review focused on the effects most pertinent to ambient
27    air exposures, which given ambient air Pb reductions over the past 30 years are those associated
28    with relatively lower exposures and associated blood Pb levels (73 FR 66975). In so doing, the
29    EPA recognized the general consensus that the developing nervous system in children is among,
30    if not the most sensitive health endpoint associated with Pb exposures. Thus, primary attention
31    was given to consideration of nervous system effects, including neurocognitive and
32    neurobehavioral effects, in children (73 FR 66976).  The body of evidence included associations
33    of such effects in study populations of variously-aged children with mean blood Pb levels below
34    10 |ig/dL, extending from 8 down to 2 |ig/dL (73 FR 66976).  The public health implications of
         January 2013                           4-2            Draft - Do Not Quote or Cite

-------
 1    effects of air-related Pb on cognitive function (e.g., IQ) in young children were given particular
 2    focus in the review.
 3         4.1.1.1  Conclusion Regarding the Need for Revision
 4          The conclusions reached by the Administrator in the last review were based primarily on
 5    the scientific evidence, with the risk- and exposure-based information providing support for
 6    various aspects of the decision. In reaching his conclusion on the adequacy of the prior standard,
 7    the Administrator placed primary consideration on the large body of scientific evidence available
 8    in the review including significant new evidence concerning effects at blood Pb concentrations
 9    substantially below those identified when the NAAQS for Pb was initially set in 1978 (73 FR
10    66987; 43 FR 46246).  Given particular attention was the robust evidence of neurotoxic effects
11    of Pb exposure in children, recognizing:  (1) that while blood Pb levels in U.S. children had
12    decreased notably since the late 1970s, newer epidemiologic studies had investigated and
13    reported associations of effects on the neurodevelopment of children with those more recent
14    lower blood Pb levels and (2) that the toxicological evidence included extensive experimental
15    laboratory animal evidence substantiating well the plausibility of the epidemiologic findings
16    observed in human children  and expanding our understanding of likely mechanisms underlying
17    the neurotoxic effects (73 FR 66987). Additionally, within the range of blood Pb levels
18    investigated in the available evidence base, a threshold level for neurocognitive effects was not
19    identified (73  FR 66984; 2006 CD, p. 8-67).  Further, the evidence indicated a steeper dose-
20    response relationship for effects on cognitive function at those lower blood Pb levels than at
21    higher blood Pb levels that were more common in the past, "indicating the potential for greater
22    incremental impact associated with exposure at these lower levels" (73 FR 66987). As at the
23    time when the standard was  set, the health effects evidence and exposure/risk assessment
24    supported the conclusion that air-related Pb exposure pathways contribute to blood Pb levels in
25    young children by inhalation and  ingestion (73 FR 66987).  The available information also
26    suggested that the air-to-blood ratio was likely larger, when all  inhalation and ingestion pathways
27    for air-related Pb are considered, than the air-to-blood ratio (of 2 |ig/dL blood Pb to 1 |ig/m3 air
28    Pb) estimated when the standard was initially set in 1978 (73 FR 66987).
29          In the Administrator's decision on the adequacy of the 1978 standard, the Administrator
30    considered the evidence using a very specifically defined framework, referred to as an air-related
31    IQ loss evidence-based framework. This framework integrates evidence for the relationship
32    between Pb in air and Pb in young children's blood with  evidence for the relationship between
33    Pb in young children's blood and  IQ loss (73 FR 77987), as described in more detail in section
34    4.1.1.2 below.  This  evidence-based approach considers air-related effects on neurocognitive
35    function (using the quantitative metric of IQ loss) associated with exposure in those areas with

         January 2013                          4-3            Draft - Do Not Quote or Cite

-------
 1    elevated air concentrations equal to potential alternative levels for the Pb standard. Thus, the
 2    conceptual context for the framework is that it provides estimates of air-related IQ loss for a
 3    subset of the population of U.S. children (i.e., the subset living in close proximity to air Pb
 4    sources that contribute to elevated air Pb concentrations that equal the current level of the
 5    standard).  This is the subset expected to experience air-related Pb exposures at the high end of
 6    the national distribution of such exposures, not at the average.  This is the case since when a
 7    standard of a particular level is just met at a monitor sited to record the highest source-oriented
 8    concentration in an area, the large majority of children in the larger surrounding area would
 9    likely experience exposures to concentrations well below that level.
10          The two primary inputs to the evidence-based air-related IQ loss framework are air-to-
11    blood ratios and C-R functions for the relationship between blood Pb and IQ response in young
12    children.  Additionally taken into consideration in applying and drawing conclusions from the
13    framework were the uncertainties inherent in these inputs.  Application of the framework also
14    entailed consideration of an appropriate level of protection from air-related IQ loss to be used in
15    conjunction with the framework. In simplest terms, the framework provides for estimation of a
16    mean air-related IQ decrement for young children in the high end of the national distribution of
17    air-related exposures by focusing on children exposed to air-related Pb in those areas with
18    elevated air Pb concentrations equal to specific potential standard levels. The framework
19    estimates of mean air-related IQ loss are derived through multiplication of the following factors:
20    standard level (|ig/m3),  air-to-blood ratio in terms of |ig/dL blood Pb per |ig/m3 air concentration
21    and slope for the C-R function in terms of points IQ decrement per |ig/dL blood Pb.
22          Based on the application of the air-related IQ loss framework to the evidence,  the
23    Administrator concluded that, for exposures projected for air Pb concentrations at the level of the
24    1978 standard, the quantitative estimates of IQ loss associated with air-related Pb indicated risk
25    of a magnitude that in his judgment was significant from a public health perspective, and that the
26    evidence-based framework supported a conclusion that the 1978 standard did not protect public
27    health with an adequate margin of safety (73 FR 77987). The Administrator further concluded
28    that the evidence indicated the need for a substantially  lower standard level to provide increased
29    public health protection, especially for at-risk groups (most notably children), against an array of
30    effects, most importantly including effects on the developing nervous system (73 FR 77987). In
31    addition to giving primary consideration to the much expanded evidence base since the standard
32    was set, the Administrator also took into consideration the exposure/risk assessments.  In so
33    doing, he observed that, while taking into consideration their inherent uncertainties and
34    limitations, the quantitative estimates of IQ loss associated with air-related Pb in air quality
35    scenarios just meeting the then-current standard also indicated risk of a magnitude that in his
36    judgment was significant from a public health perspective.  Thus, the Administrator concluded

         January 2013                           4-4             Draft - Do Not Quote or Cite

-------
 1    the exposure/risk estimates provided additional support to the evidence-based conclusion that the
 2    standard needed revision (73 FR 66987).

 3         4.1.1.2   Conclusions on Elements of Revised Standard
 4           In considering appropriate revisions to the prior standard in the review completed in
 5    2008, each of the four basic elements of the NAAQS (indicator, averaging time, level, and form)
 6    were evaluated. The rationale for decisions on those elements is summarized below.
 7           With regard to indicator, consideration was given to replacing Pb-TSP with Pb-PMi0.
 8    The EPA recognized, however, that Pb in all particle sizes contributes to Pb in blood and
 9    associated health effects, additionally noting that the difference in particulate Pb captured by
10    TSP and PMio monitors may be on the order of a factor of two in some areas (73 FR 66991).
11    Further, the Administrator recognized uncertainty with regard to whether a Pb-PMio-based
12    standard would also effectively control ultra-coarse1 Pb particles, which may have a greater
13    presence in areas near sources where Pb concentrations are highest (73 FR 66991). The
14    Administrator decided to retain Pb-TSP as the indicator to provide sufficient public health
15    protection from the range of particle sizes of ambient air Pb, including ultra-coarse particles (73
16    FR 66991). Additionally, a role was provided for Pb-PMi0 in the monitoring required for a Pb-
17    TSP standard (73 FR 66991) based on the conclusion that use of Pb-PMio measurements at sites
18    not influenced by sources of ultra-coarse Pb, and where Pb concentrations are well below the
19    standard, would take advantage of the increased precision of these measurements and decreased
20    spatial variation of Pb-PMi0 concentrations, without raising the same concerns over a lack of
21    protection against health risks from all particulate Pb emitted to the ambient air that support
22    retention of Pb-TSP as the indicator (versus revision to Pb-PMi0) (73 FR 66991).  Accordingly,
23    allowance was made for the use of Pb-PMi0 monitoring for Pb NAAQS attainment purposes in
24    certain limited circumstances, at non-source-oriented sites, where the Pb concentrations are
25    expected to be substantially below the standard and ultra-coarse particles are not expected to be
26    present (73 FR 66991).
27           With regard to averaging time and form, consideraton was given to monthly and 3-month
28    averaging times.  The Administrator recognized the complexity related to the multimedia nature
29    of Pb and its multiple pathways of human exposure, in addition to uncertainty remaining over the
30    period of time needed for air Pb concentrations to lead to the health effects most at issue in this
31    review (73 FR 77995).  Relevant factors pertaining  to the human physiological response to
32    changes in Pb exposures and those pertaining to the response of air-related Pb exposure
             1 The term ultra-coarse is used to refer to particles collected by a TSP sampler but not by a PM10 sampler.
      This terminology is consistent with the traditional usage of' 'fine'' to refer to particles collected by a PM2 5 sampler,
      and ' 'coarse'' to refer to particles collected by a PM10 sampler but not by a PM2 5 sampler, recognizing that there
      will be some overlap in the particle sizes in the three types of collected material.
         January 2013                           4-5              Draft - Do Not Quote or Cite

-------
 1    pathways to changes in airborne Pb were considered in an integrated manner (73 FR 77996).
 2    Key considerations included the limited evidence indicating more numerous factors influencing
 3    ingestion (versus inhalation) pathways, which were considered likely to lessen the impact of
 4    month-to-month variations in airborne Pb concentrations on levels of air-related Pb in children's
 5    blood.  Such factors were considered likely to lead to response times (e.g., for the response of
 6    blood to air Pb via these pathways) extending longer than a month (73 FR 66996).  The EPA
 7    additionally considered the control that a rolling 3-month averaging time can provide (e.g.,
 8    compared to a block calendar quarter)  on month-to-month variability in air Pb  concentrations
 9    and in associated exposures (73 FR 66996).  Based on this integrated consideration of the range
10    of relevant factors, the averaging time  was revised to a rolling three-month period with a
11    maximum (not-to-be-exceeded) form,  evaluated over a three-year period.  As compared to the
12    previous averaging time and form of calendar quarter (not-to-be exceeded), this revision was
13    considered to be more scientifically appropriate and more health protective (73 FR 77996).  The
14    rolling average gives equal weight to all three-month periods, and the new calculation method
15    gives equal weight to each month within each three-month period (73 FR 77996). Further, the
16    rolling average yields 12 three-month averages each year to be compared to the NAAQS versus
17    four averages in each year for the block calendar quarters pertaining to the previous standard (73
18    FR 77996).
19          Lastly, based on the body of scientific evidence and information available, as well as
20    CASAC recommendations and public  comment, the Administrator decided on  a standard level
21    that, in combination with the specified choice of indicator, averaging time, and form,  he judged
22    requisite to protect public health, including the health of sensitive groups, with an adequate
23    margin of safety (73  FR 67006).  In reaching the decision on level for the revised standard, the
24    Administrator considered as a useful guide the evidence-based framework developed in that
25    review. As described in section 4.1.1.1 above, that framework integrates evidence for the
26    relationship between Pb in air and Pb in children's blood and the relationship between Pb in
27    children's blood and IQ loss.  Application of the air-related IQ loss evidence-based framework
28    was recognized, however, to provide "no evidence- or risk-based bright line that indicates a
29    single appropriate level" for the standard (73 FR 67006). Rather, the framework was seen as a
30    useful guide for consideration of health risks from exposure to ambient levels of Pb in the air, in
31    the context of a specified averaging time and form, with regard to the Administrator's decision
32    on a level for a revised NAAQS that provides public health protection that is sufficient but not
33    more than necessary under the Act (73 FR 67004).
34          As noted above, use of the evidence-based air-related IQ loss framework to inform
35    selection of a standard level involved consideration of the evidence with regard to two input
36    parameters. The two input parameters are an air-to-blood ratio and a C-R function  for population

         January 2013                           4-6             Draft - Do Not Quote or Cite

-------
 1    IQ response associated with blood Pb level (73 FR 67004). The evidence at the time of the last
 2    review indicated a broad range of air-to-blood ratio estimates, each with limitations and
 3    associated uncertainties.  Based on the then-available evidence, the Administrator concluded that
 4    1:5 to 1:10 represented a reasonable range to consider and identified 1:7 as a generally central
 5    value on which to focus (73 FR 67004). With regard to C-R functions, in light of the evidence of
 6    nonlinearity and of steeper slopes at lower blood Pb levels, the Administrator concluded it was
 7    appropriate to focus on C-R analyses based on blood Pb levels that most closely reflected the
 8    then-current population of children in the U. S.,2 recognizing EPA's identification of four such
 9    analyses and giving weight to the central estimate or median of the resultant C-R  functions (73
10    FR 67003,  Table 3; 73 FR 67004).  The four study groups from which C-R functions were drawn
11    in 2008, and the associated C-R slopes, are summarized in Table 3-3 above.3 The median
12    estimate of -1.75 IQ points decrement per |ig/dL was selected for use with the framework. With
13    the framework, as summarized in section 4.1.1.1 above, potential alternative standard levels
14    (jig/m3) are multiplied by estimates of air-to-blood ratio (|ig/dL blood Pb per |ig/m3 air Pb) and
15    the median slope for the C-R function (points IQ decrement per |ig/dL blood Pb), yielding
16    estimates of a mean air-related IQ decrement for a specific subset of young children (i.e., those
17    children exposed to air-related Pb in areas with elevated air Pb concentrations equal to specified
18    alternative  levels). As such, the application of the framework yields estimates for the mean air-
19    related IQ decrements of the subset of children expected to experience air-related Pb exposures
20    at the high  end of the  distribution of such exposures. The associated mean IQ loss estimate is the
21    average for this highly exposed subset and is not the average air-related IQ loss projected for the
22    entire U.S.  population of children.  Uncertainties and limitations were recognized in the use of
23    the framework and in the resultant estimates (73 FR 67000).
24           In considering the use of the evidence-based air-related IQ loss framework to inform his
25    judgment as to the appropriate degree of public health protection that should be afforded by the
26    NAAQS to provide requisite protection against risk of neurocognitive effects in sensitive
27    populations, such as IQ loss in children, the Administrator recognized in the 2008 review that
28    there were  no  commonly accepted guidelines or criteria within the public health community that
29    would provide a clear basis for such a judgment. During the 2008 review, CAS AC commented
             2 The geometric mean blood Pb level for U.S. children aged five years and below, reported for NHANES in
      2003-04 (the most recent years for which such an estimate was available at the time of the 2008 decision) was 1.8
      ug/dL and the 5th and 95th percentiles were 0.7 ug/dL and 5.1 ug/dL, respectively (73 FR 67002).
             3 One of these four is from the analysis of the lowest blood Pb subset of the pooled international study by
      Lanphear et al., (2005). The nonlinear model developed from the full pooled dataset is the basis of the C-R
      functions used in the 2007 REA (see section 3.4.3.3 above), in which risk was estimated over a large range of blood
      Pb levels. Given the narrower focus of the evidence-based framework on IQ response at the end of studied blood Pb
      levels (closer to U.S. mean level), the C-R functions in Table 3-3 are from linear analyses (each from separate
      publications) for the study group subsets with blood Pb levels closest to mean for children in the U.S. today.
         January 2013                            4-7              Draft - Do Not Quote or Cite

-------
 1    regarding the significance from a public health perspective of a 1-2 point IQ loss in the entire
 2    population of children and, along with some commenters, emphasized that the NAAQS should
 3    prevent air-related IQ loss of a significant magnitude, such as on the order of 1-2 IQ points, in all
 4    but a small percentile of the population. Similarly, the Administrator stated that "ideally air-
 5    related (as well as other) exposures to environmental Pb would be reduced to the point that no IQ
 6    impact in children would occur" (73 FR 66998). The Administrator further recognized that, in
 7    the case of setting a NAAQS, he was required to make a judgment as to what degree of
 8    protection is requisite to protect public health with an adequate margin of safety (73 FR66998).
 9    The NAAQS must be sufficient but not more stringent than necessary to achieve that result, and
10    does not require a zero-risk standard (73 FR 66998).  The Administrator additionally recognized
11    that the evidence-based air-related IQ loss framework did not provide estimates pertaining to the
12    U.S. population of children as a whole. Rather, the framework provides estimates (with
13    associated uncertainties and limitations) for the mean of a subset of that population, the subset of
14    children assumed to be exposed to the level of the standard.  As described in the final decision
15    "[t]he framework in effect focuses on the sensitive subpopulation that is the group of children
16    living near sources and more likely to be exposed at the level of the standard" (73 FR 67000).
17    As further noted in the final decision (73 FR 67000):
18           EPA  is unable to quantify the percentile of the U.S. population of children that
19           corresponds to the mean of this sensitive subpopulation. Nor is EPA confident in
20           its ability to develop quantified estimates of air-related IQ loss for higher
21           percentiles than the mean of this subpopulation.  EPA expects that the mean of
22           this subpopulation represents a high, but not quantifiable, percentile of the U.S.
23           population of children. As a result, EPA expects that a standard based on
24           consideration of this framework would provide the same or greater protection
25          from  estimated air-related IQ loss for a high,  albeit unquantifiable, percentage of
26           the entire population of U.S. children.
27
28           In reaching a judgment as to the appropriate degree of protection, the Administrator
29    considered advice and recommendations from CASAC and public comments and recognized the
30    uncertainties in the health effects evidence and related information as well as the role of, and
31    context for, a selected air-related IQ loss in the application of the framework.  In so doing, the
32    Administrator identified an air-related IQ loss of 2 points to be used in conjunction with the
33    evidence-based framework in considering an appropriate level for the standard (73 FR 67005).4
34    Given the various uncertainties associated with the framework and the scientific evidence base,
35    and the focus of the framework on the sensitive subpopulation of more highly exposed children,
             4 Also noted that "in determining what level of estimated IQ loss should be used for evaluating the results
      obtained from this specific evidence-based framework, the Administrator is not determining that such an IQ loss is
      appropriate for use in other contexts". (73 FR 67005)
         January 2013                           4-8             Draft - Do Not Quote or Cite

-------
 1    a standard level selected using this air-related IQ loss, in combination with the selected
 2    averaging time and form, was expected to significantly reduce and limit for a high percentage of
 3    U.S. children the risk of experiencing an air-related IQ loss of that magnitude (73 FR 67005). At
 4    the standard level of 0.15 |ig/m3, with the combination of the generally central estimate of air-to-
 5    blood ratio of 1:7 and the median of the four C-R functions (-1.75 IQ point decrement per |ig/dL
 6    blood Pb), the framework estimates of air-related IQ loss were below 2 IQ points (73 FR 67005,
 7    Table 4).
 8          In reaching the decision in 2008 on level for the revised standard, the Administrator also
 9    considered the results of the quantitative risk assessment to provide a useful perspective on risk
10    from air-related Pb.  In light of important uncertainties and limitations for purposes of evaluating
11    potential standard levels, however, the Administrator placed less weight on the risk estimates
12    than on the evidence-based assessment. Nevertheless, in recognition of the general
13    comparability of quantitative risk estimates for the case studies considered most conceptually
14    similar to the scenario represented by the evidence-based framework, he judged  the quantitative
15    risk estimates to be "roughly consistent with and generally supportive" of the evidence-based
16    framework estimates (73 FR 67006).
17          Based on consideration of the entire body of evidence and information available in the
18    review, as well as the recommendations of CAS AC and public comments, the Administrator
19    decided  that a level for the primary Pb standard of 0.15  |ig/m3, in combination with the specified
20    choice of indicator, averaging time and form was requisite to protect public health, including the
21    health of sensitive groups, with an adequate margin of safety (73 FR 67006). In reaching
22    decisions on level as well as the other elements of the revised standard, the Administrator took
23    note of the complexity associated with consideration of health effects caused by  different
24    ambient air concentrations of Pb and with uncertainties with regard to the relationships between
25    air concentrations, exposures, and health effects  For example, selection of a maximum,  not to
26    be exceeded, form in conjunction with a rolling 3-month averaging time over a three year span
27    was expected to have the effect that the at-risk subpopulation of children would be exposed
28    below the standard most of the time (73 FR 67005).  The Administrator additionally noted his
29    consideration of the provision  of an adequate margin of safety in making decisions on each of the
30    elements of the standard, including, for example "selection of TSP as the indicator and the
31    rej ection of the use of PMio scaling factors; selection of a maximum, not to be exceeded form, in
32    conjunction with a 3-month averaging time that employs a rolling average, with  the requirement
33    that each month in the 3-month period be weighted  equally (rather than being averaged by
34    individual data) and that a 3-year span be used for comparison to the standard; and, the use of a
35    range of inputs for the evidence-based framework, that includes a focus on higher air-to-blood
36    ratios than the lowest ratio considered to be supportable, and steeper rather than  shallower C-R

         January 2013                          4-9             Draft - Do Not Quote or Cite

-------
 1    functions, and the consideration of these inputs in selection of 0.15 ug/m3 as the level of the
 2    standard" (73 FR 67007).
 3          He additionally noted that a standard with this level would reduce the risk of a variety of
 4    health effects associated with exposure to Pb, including effects indicated in the epidemiological
 5    studies at lower blood Pb levels, particularly including neurological effects in children, and the
 6    potential for cardiovascular and renal effects in adults (73 FR 67006).  The Administrator
 7    additionally considered higher and lower levels for the standard, concluding that a level of 0.15
 8    Ug/m3 provides for a standard that is neither more or less stringent than necessary for this
 9    purpose, recognizing that the Clean Air Act does not require that primary standards be set at a
10    zero-risk level, but rather at a level that reduces risk sufficiently so as to protect public health
11    with an adequate margin of safety (73 FR 67007). For example, the Administrator additionally
12    considered potential public health protection provided by standard levels above 0.15 |ig/m3,
13    which he concluded were insufficient to protect public health with an adequate margin of safety.
14    The Administrator also noted that in light of all of the evidence, including the evidence-based
15    framework, that the degree of public health protection likely afforded by standard levels below
16    0.15 |ig/m3 would be greater than what is necessary to protect public safety with an adequate
17    margin of safety.
18          The Administrator concluded, based on review of all of the evidence (including the
19    evidence- based framework), that when taken as a whole the selected  standard, including the
20    indicator, averaging time, form, and level, would be "sufficient but not more than necessary to
21    protect public health, including the health of sensitive subpopulations, with an adequate margin
22    of safety" (73 FR 67007).

23         4.1.2   Approach for the Current Review
24          To evaluate whether it is appropriate to consider retaining the  current primary Pb
25    standard, or whether consideration of revision is appropriate, we have adopted an approach in
26    this review that builds upon the general approach used in the last review and reflects the broader
27    body of evidence and information now available.  As summarized above, the Administrator's
28    decisions in the prior review were based on an integration of information on health effects
29    associated with exposure to Pb, relationships between ambient air Pb  and blood Pb; expert
30    judgments on the adversity and public health significance of key health effects; and policy
31    judgments as to when the standard is requisite to protect public health with an adequate margin
32    of safety. These considerations were informed by air quality and related analyses, quantitative
33    exposure and risk assessments,  and qualitative assessment  of impacts  that could not be
34    quantified.
         January 2013                          4-10             Draft - Do Not Quote or Cite

-------
 1          In conducting this assessment, we draw on the current evidence and quantitative
 2   assessments of exposure pertaining to the public health risk of Pb in ambient air. In considering
 3   the scientific and technical information, we consider both the information available at the time of
 4   the last review and information newly available since the last review, including the current ISA
 5   (USEPA, 2012), as well as the quantitative exposure/risk assessments from the last review that
 6   estimated Pb-related IQ decrements associated with different air quality conditions in simulated
 7   at-risk populations in multiple case studies (USEPA, 2007a). Figure 4-1 illustrates the basic
 8   construct of our two part approach in developing conclusions regarding options appropriate for
 9   the Administrator to consider in this review with regard to the adequacy of the current standard
10   and, as appropriate, potential alternate standards. In the boxes of Figure 4-1, the range of
11   questions considered in chapter 3  above and section 4.2 below are represented by a summary of
12   policy-relevant questions that frame our consideration of the scientific evidence and
13   exposure/risk information.
         January 2013                           4-11            Draft - Do Not Quote or Cite

-------
                                                  Adequacy of Current Standard
                              Evidence-based Considerations

                     > Does currently available evidence and related
                     uncertainties strengthen or call into question prior
                     conclusions?
                        • Evidence of health effects not previously identified
                        or at lower exposures?
                        • Newly identified at-risk populations?
                        •Changed understanding of air-to-blood
                        relationships?
                        •Effects at lower air-related exposures then
                        previously understood or in conditions that would
                        have likely met current standards?
                     >Does newly available information call into question any
                     of the basic elements of standard?
                               Risk-based Considerations

                       > Nature, magnitude and importance of
                       estimated exposures and risks associated with
                       just meeting the current standard?
                       >Uncertainties in the exposure and risk
                       estimates?
                                                                                                     Consider
                                                                                                     Retaining
                                                                                           NOV   Current Standard
Does information call into question the
   adequacy of current standard?
                                           Consideration of Potential Alternative Standards
                                                Elements of Potential Alternate Standards
                                                           indicator
                                                           > Averaging Time
                                                           >Form
                                                           > Level
                                           Potentia A ternative Standards for Consideration
                                    ion   J)
2     Figure 4-1.  Overview of approach for review of current primary standard.
           January 2013
                4-12
Draft - Do Not Quote or Cite

-------
 1    4.2   ADEQUACY OF THE CURRENT STANDARD
 2            In considering the adequacy of the current Pb standard, the overarching question we
 3    consider is:
 4         •  Does the currently available scientific evidence- and exposure/risk-based
 5             information, as reflected in the ISA and REA, support or call into question the
 6             adequacy of the protection afforded by the current Pb standard?
 7          In considering the  scientific and technical information, we give our attention to both the
 8    information available at the time of the last review and information newly available since the last
 9    review, including most particularly that which has been critically analyzed and characterized in
10    the ISA. In chapter 3 above, attention was given to addressing specific questions on key aspects
11    of this information. To assist us in interpreting the currently available scientific evidence and
12    results of quantitative exposure/risk analyses to address the overarching question here, we draw
13    on discussions in chapter 3 above in our consideration of broader or more policy related
14    questions, posed within sections 4.2.1 and 4.2.2 below.
15          For the purposes of this draft PA, staff has drawn from EPA's assessment and integrated
16    synthesis of the scientific evidence presented in the third draft ISA and on the quantitative
17    exposure and risk information, based on the 2007 REA (USEPA, 2007a), described in section 3.4
18    above.  The evidence-based discussions presented in this chapter draw upon evidence from
19    epidemiologic studies and experimental animal studies evaluating health effects related to
20    exposures to Pb, as discussed in the ISA.  The exposure/risk-based discussions have drawn from
21    the quantitative health risk analyses for Pb performed in the last Pb NAAQS review in light of
22    the currently available evidence (2007 REA, REA Planning Document). Together the evidence-
23    based and risk-based considerations inform our preliminary conclusions related to the adequacy
24    of the current Pb standard.

25         4.2.1   Evidence-based Considerations
26          In considering the  evidence with regard to the issue of adequacy of the current standard,
27    we address several questions that build on the information summarized in chapter 3 to more
28    broadly  address the extent to which the current evidence base supports the adequacy of the
29    public health protection afforded by the current primary standard.  The first question addresses
30    our integrated consideration of the health effects evidence, in light of aspects described in
31    chapter 3.  The second question focuses on  our consideration of associated areas of uncertainty.
32    The third question then integrates our consideration of the prior two questions with a focus on
33    the standard, including each of the four elements.
         January 2013                           4-13            Draft - Do Not Quote or Cite

-------
 1         •   To what extent has new information altered the scientific support for the
 2             occurrence of health effects as a result of multimedia exposure associated with
 3             levels of Pb occurring in the ambient air?
 4          The current evidence continues to support our conclusions from the previous review
 5    regarding key aspects of the health effects evidence for Pb and the health effects of multimedia
 6    exposure associated with levels of Pb occurring in ambient air in the U.S.  Our conclusions in
 7    this regard are based on consideration of the assessment of the currently available evidence in the
 8    ISA, particularly with regard to key aspects summarized in chapter 3 of this PA, in light of the
 9    assessment of the evidence in the last review as described in the 2006 CD and summarized in the
10    notice of final rulemaking (73 FR 66964). Key aspects of these conclusions are summarized
11    below.
12          As at the time of the last review, blood Pb continues to be the predominant biomarker
13    employed to assess exposure and health risk of Pb (ISA, chapters 4 and 5), as discussed in
14    section 3.1 above. This widely accepted role of blood Pb in assessing exposure and risk is
15    illustrated by its established use in programs to prevent both occupational Pb poisoning and
16    childhood Pb poisoning, with the latter program,  implemented by the CDC, recently issuing
17    updated  guidance on blood Pb measurement interpretation (CDC, 2012).  As in the  past, the
18    current evidence continues to indicate the close linkage of blood Pb levels in young children to
19    their body burden; this linkage is associated with the ongoing bone remodeling during that
20    lifestage (ISA, section 4.3.5). This tight linkage plays a role in  the somewhat rapid response of
21    children's blood Pb to changes in exposure  (particularly to exposure increases), which
22    contributes to its usefulness as an exposure  biomarker (ISA, sections 4.2.2, 4.3.5, and 4.3.5.1).
23    Additionally, the weight of evidence documenting relationships between children's blood Pb and
24    health effects, most particularly those on the nervous and hematological systems (e.g., ISA,
25    sections  5.3 and 5.7) speaks to its usefulness in assessing health risk.
26          As in the last review, the evidence on air-to-blood relationships available today continues
27    to be comprised of studies based on an array of circumstances and population groups (of
28    different age ranges), analyzed by a variety of techniques, which together contribute to
29    appreciable variability in the associated quantitative estimates and uncertainty with regard to the
30    relationships existing in the U.S. today.  Accordingly, our interpretation of this evidence base, as
31    discussed in section 3.1 above, also includes consideration of factors that may be influencing
32    various study estimates, both with regard to their usefulness for our general purpose of
33    quantitatively characterizing relationships between Pb in ambient air and air-related Pb in
34    children's blood, and with regard to their pertinence more specifically to conditions and
35    populations in the U.S. today.  In so doing, we note that the current evidence, while including
36    two additional studies not available at the time of the last review, is not much changed from that

         January 2013                           4-14            Draft - Do Not Quote or Cite

-------
 1    available in the last review.  The range of estimates that can be derived from the full dataset is
 2    broad and not changed by the inclusion of the newly available estimates. Further, we recognize
 3    significant uncertainties regarding the air Pb to air-related blood Pb relationship for the current
 4    conditions where concentrations of Pb in both ambient air and children's blood are substantially
 5    lower than they have been in the past.  In considering the strengths, limitations and uncertainties
 6    associated with the full dataset, the currently available evidence appears to continue to support a
 7    range of estimates for our purposes that is generally consistent with the range given weight in the
 8    last review, 1:5 to 1:10 (ISA, section 4.7.4 and Table 4-12; 73 FR 67001-2, 67004).  We
 9    additionally note that the generally central estimate of 1:7 identified for this range in the last
10    review is consistent with the study involving blood Pb for pre-school  children and air Pb
11    conditions near a large source of Pb to ambient air with concentrations near (and/or previously
12    above) the level of the current Pb standard (ISA, section  4.5.1; Hilts, 2003).5 In so noting, we
13    also recognize the general overlap of such circumstances with those represented by the evidence-
14    based, air-related IQ loss framework,6 for which air-to-blood ratio is a key input. In
15    characterizing the range of air-to-blood ratio estimates, we recognize uncertainty inherent in such
16    estimates as well as the variation in currently available estimates resulting from a variety of
17    factors,  including differences in the populations examined, as well as in the Pb sources or
18    exposure pathways addressed in those study analyses (ISA, section 4.7.4).
19           The scientific evidence continues to  recognize a broad array of health effects on multiple
20    organ systems or biological  processes related to blood Pb, including Pb in blood prenatally (ISA,
21    section 2.6).  The currently available evidence continues to support identification of
22    neurocognitive effects in young children as  the most sensitive endpoint associated with blood Pb
23    concentrations (ISA, section 2.6.1), which as an integrated index of exposure reflect the
24    aggregate exposure to all sources of Pb through multiple pathways (inhalation and ingestion).
25    Evidence continues to indicate that neurocognitive effects in young children may not be
26    reversible and may have effects that persist  into adulthood (ISA, section 2.9.6).  Thus, as
27    discussed in section 3.2 above, we continue  to consider the evidence of Pb effects at the low end
28    of the studied blood Pb levels (closest to those common in the U.S. today) to be strongest and of
29    greatest concern for effects on the nervous system, most  particularly those on cognitive function
30    in children.
31           As in the last review, evidence on risk factors continues to support the identification of
32    young children as an important at-risk population for Pb  health effects (ISA, section 6.4).  The
             5 The older study by Hayes et al (1994) during time of leaded gasoline indicated a generally similar ratio of
      1:8, although the blood Pb levels in that study were much higher than those in the study by Hilts (2003). Among the
      studies focused on this age group, the latter study includes blood Pb levels closest to those in U.S. today.
             6 Concentrations near air sources are higher than those at more distant sites (as described in section 2.2.2);
      it is near-source locations where there is the potential for concentrations at or near the current standard.
         January 2013                            4-15             Draft - Do Not Quote or Cite

-------
 1    current evidence also continues to indicate important roles as factors that increase risk of lead-
 2    related health effects for:  nutritional factors, such as iron and calcium intake; elevated blood Pb
 3    levels; and proximity to sources of Pb exposure, such as industrial releases or buildings with old,
 4    deteriorating, leaded paint. Further, some races or ethnic groups continue to demonstrate
 5    increased blood Pb levels relative to others, which may be related to these and other factors (ISA,
 6    sections 6.1, 6.2 and 6.4).
 7           With regard to our understanding of the relationship between exposure or blood Pb levels
 8    in young children and neurocognitive effects, the evidence in this review, as in the last, does not
 9    establish threshold blood Pb level for neurocognitive effects in young children (ISA, sections
10    2.9.4 and 5.3.13).  The lowest blood Pb levels at which associations with neurocognitive impacts
11    have been observed in pre-school and school age children continue to range down below 5
12    ng/dL, with lowest group levels that have been associated with such effects ranging down to 2
13    ng/dL (ISA, sections 2.6.1.1 and 5.3.16.1).  Additionally, as in the last review, there is evidence
14    that the relationship of young children's blood Pb with neurocognitive impacts, such as IQ, is
15    nonlinear across a wide range of blood Pb, with greater incremental impacts at lower vs. higher
16    blood Pb levels (ISA, sections 2.9.4 and 5.3.13).  Accordingly, as in the last review, we continue
17    to focus our interest on C-R relationships from study groups with blood Pb levels closest to those
18    in children in the U.S. today, which are generally lower than epidemiological study groups.  The
19    currently available evidence does not identify additional C-R slopes for study groups of young
20    children (e.g., < 7 years) with mean blood Pb levels below that of groups identified in the last
21    review, 2.9 - 3.8 |ig/dL, as discussed in  section 3.2 above (ISA, section 5.3.13). Thus, the blood
22    Pb concentration - IQ response  functions or slopes identified in this review for epidemiological
23    study groups of young children with mean blood Pb  levels closest to that of children in the U.S.
24    today include the same set recognized at the time of the last review (see Table 3-3 above), the
25    median of which is 1.75 points  decrement per |ig/dL blood Pb (73 FR, 67003).
26         •  To what extent have important uncertainties identified in the last review been
27             reduced and/or have new uncertainties  emerged?
28           In our consideration of the evidence, as summarized in discussing the previous question
29    and in chapter 3 above, we have not identified any new uncertainties as emerging since the last
30    review.  However, we continue to recognize important uncertainties identified in the last review
31    that remain today. Importantly, given our focus in this review, as in the last review, on
32    neurocognitive impacts associated with Pb exposure in early childhood, we recognize remaining
33    uncertainties in our understanding of the C-R relationship of neurocognitive impacts, such as IQ
34    decrements, with blood Pb level in young children, particularly across the range of blood Pb
35    levels common in the U.S. today. With regard to C-R relationships for IQ, the evidence
36    available in this review does not include studies that appreciably extend the range of blood Pb
         January 2013                           4-16            Draft - Do Not Quote or Cite

-------
 1    levels studied beyond those available in the last review. As in the last review, the early
 2    childhood (e.g., 2 to 6 or 7 years of age) blood Pb levels for which associations with IQ response
 3    have been reported continue to extend at the low end of the range to study group mean blood Pb
 4    levels of 2.9 to 3.8  jig/dL (e.g., 73 FR 67003, Table 3). The studies at these blood Pb levels do
 5    not provide evidence of a threshold level and, as summarized in section 3.2 above, continue to
 6    indicate higher C-R slopes in these groups with lower blood Pb levels than in study groups with
 7    higher blood Pb levels (ISA, section 5.3.13).
 8           Further, we recognize important uncertainties in our understanding of the relationship
 9    between ambient air Pb concentrations and air-related Pb in children's blood. The evidence
10    newly available in this review has not reduced such key uncertainties. As in the last review, air-
11    to-blood ratios based on the available evidence continue to vary, with our conclusions based on
12    the current evidence generally consistent with the range of 1:5 to  1:10 given emphasis in the last
13    review (73 FR 67002; ISA, section 4.7.4). There continues to be uncertainty regarding the extent
14    to which this range represents the relationship between ambient air Pb and Pb in children's blood
15    (derived from the full set of air-related exposure pathways), and with regard to its reflection of
16    exposures associated with ambient air Pb levels common in the U.S. today and to circumstances
17    reflecting just meeting the current Pb standard (ISA, section 4.7.4). We note additionally the
18    significant uncertainty remaining with regard to the temporal relationships of ambient Pb levels
19    and associated exposure with occurrence of a health effect (73 FR 67005).
20         •   To what  extent does newly available information support or call into question any
21             of the basic elements of the current Pb standard?
22           We address this question for each of the elements of the standard in light of the health
23    effects evidence and other relevant information available in this review. As an initial matter,
24    however, we recognize the weight of the scientific evidence available in this review that
25    continues to support our focus on effects on the nervous system of young children, and
26    specifically neurocognitive decrements, as the most sensitive endpoint. Consistent with the
27    evidence available  in the last review, the currently available evidence continues to indicate that a
28    standard that provides requisite public health protection against the occurrence of such effects in
29    at-risk populations  would also provide the requisite public health  protection against the full array
30    of health effects of Pb. Accordingly, the discussion of the elements below is framed by that
31    background.
32           Indicator
33           The indicator for the current Pb standard is Pb-TSP. Exposure to Pb in all sizes of
34    particles passing through ambient air can contribute to Pb in blood and associated health effects
35    by a wide array of exposure pathways (ISA,  section 4.1).  These pathways include the ingestion
36    route, as well as inhalation (ISA,  section 4.1), and a wide array of particle sizes play a role in
         January 2013                          4-17            Draft - Do Not Quote or Cite

-------
 1    these pathways (ISA, section 4.1.1.1).  As at the time of the last review, we continue to recognize
 2    the variability of the Pb-TSP FRM in its capture of airborne Pb particles (see section 2.2.1.3.1
 3    above).  As in the last review, we also note that an alternative approach for collection of a
 4    conceptually comparable range of particle sizes, including ultra-coarse particles, is not yet
 5    available, although we take note of activities underway to remedy that situation in the future, as
 6    referenced in section 2.2.1.3.1 above. Additionally, the limited available information regarding
 7    relationships between Pb-TSP and Pb in other size fractions indicates appreciable variation in
 8    this relationship, particularly near sources of Pb emissions where concentrations, and potential
 9    exposures, are  greatest. Thus, the information available in this review does not address
10    previously identified limitations and uncertainties for the current indicator. Nor does the newly
11    available information identify additional limitations or uncertainties.  The evidence available in
12    this review continues to indicate the  role of a range of air Pb particle sizes in contributing to Pb
13    exposure (e.g., ISA, section 4.1.1.1), that contributes to Pb in blood and associated health effects,
14    reinforcing the appropriateness of an indicator for the Pb standard that reflects a wide range of
15    airborne Pb particles.
16          Averaging time and form
17          The averaging time and form of the current standard is a not-to-be-exceeded rolling
18    three-month average (CFR 50.16), derived from three  monthly averages calculated in accordance
19    with the current data handling procedures (CFR, Appendix R to Part 50).  As at the time of the
20    last review, evidence continues to support the importance of periods on the order of 3 months
21    and the prominent role of deposition-related exposure  pathways, with uncertainty associated with
22    characterization of precise time periods associating ambient air Pb with air-related health effects.
23    Relevant factors continue to be those pertaining to the human physiological response to changes
24    in Pb exposures and those pertaining to the response of air-related Pb exposure pathways to
25    changes in airborne Pb. The newly available evidence in this review does not appreciably
26    improve our understanding of the period of time in which air Pb concentrations would lead to the
27    health effects most at issue in this review.  Thus, there continue  to be limitations in the evidence
28    to inform our consideration of these elements of the standard and associated uncertainty.
29    However, there is no newly available information that calls into question this element of the
30    current Pb standard.
31          Level
32          The level of the current standard is 0.15  |ig/m3 (CFR 50.16). As described in section
33    4.1.1 above, this level was selected in 2008 with consideration of, among other factors, an
34    evidence-based air-related IQ loss framework, for which there are two primary inputs: air-to-
35    blood ratios and C-R functions for blood Pb - IQ response in young children.  Additionally taken
36    into consideration were the uncertainties inherent in these inputs.  Application of the framework

         January 2013                           4-18             Draft - Do Not Quote or Cite

-------
 1    also entailed consideration of a magnitude of air-related IQ loss, which as further described in
 2    section 4.1.1.2 above, is used in conjunction with this specific framework in light of the
 3    framework context, limitations and uncertainties. Additionally, selection of a level for the
 4    standard in 2008 was made in conjunction with decisions on indicator, averaging time and form.
 5           As an initial matter, we consider the extent to which the evidence-based, air-related IQ
 6    loss framework which informed the Administrator's decision in the last review is supported by
 7    the currently available evidence and information. In  so doing, we recognize the support provided
 8    by the currently available evidence for the key conclusions drawn in the last review with regard
 9    to health effects of greatest concern, at-risk populations, the influence of Pb in ambient air on Pb
10    in children's blood and the association between children's blood Pb and decrements in
11    neurocognitive function (e.g., IQ). We additionally note the complexity associated with
12    interpreting the scientific evidence with regard to specific levels of Pb  in ambient air, given the
13    focus of the evidence on blood Pb as the key biomarker of children's aggregate exposure.  The
14    need to make such interpretations in the face of the associated complexity supported use of the
15    evidence-based framework in the last review. In considering the  currently available evidence for
16    the same purposes in this review, we conclude that the evidence-based framework continues to
17    provide a useful tool for consideration of the evidence with regard to the level of the standard.
18           We next turn to consideration of the primary inputs to the framework:  air-to-blood ratios
19    and C-R functions for blood Pb - IQ response in young children.  With regard to the former, the
20    limited newly available information assessed in the ISA, and discussed in section 3.1 above, is
21    generally consistent with the information in this area that was available at the time of the last
22    review. We additionally recognize the variability and uncertainty associated with quantitative
23    air-to-blood ratios based on this information, as also existed in the last review.  As in the last
24    review, we recognize that factors contributing to the variability and uncertainty of these
25    estimates are varied, and include aspects of the study populations (e.g., age and Pb exposure
26    pathways) and the study circumstances (e.g., length of study period and variations in sources of
27    Pb exposure during the study period). We note that the full range of estimates associated with
28    the available evidence is wide, and consider it appropriate to give emphasis to estimates
29    pertaining to circumstances closest to those in the U.S. today with regard to ambient air Pb and
30    children's blood Pb concentrations, while recognizing the limitations associated with the
31    available information with regard to this emphasis. With that in mind, we consider the currently
32    available evidence to continue to support the range of estimates concluded in the last review to
33    be most appropriate for the current population of young children in the U.S., in light of the
34    multiple air-related exposure pathways by which children are exposed, in addition to inhalation
35    of ambient air, and of the levels of air and blood Pb common today.  Identification of this range
36    also included consideration of the limitations associated with the  available information and

         January 2013                           4-19            Draft  - Do Not Quote or Cite

-------
 1    inherent uncertainties. This range of air-to-blood ratios included 1:10 at the upper end and 1:5 at
 2    the lower end.  We further recognize that the limited evidence for air Pb and children's blood Pb
 3    concentrations closest to those in U.S. today continues to provide support for the Administrator's
 4    emphasis in the 2008 decision on the relatively central estimate of 1:7.
 5          With regard to the second input to the evidence-based framework, C-R functions for the
 6    relationship of young children's blood Pb with neurocognitive impacts (e.g., IQ decrements), we
 7    consider several aspects of the evidence. First, as discussed in section 3.2 above, the currently
 8    available information continues to provide evidence that this C-R relationship is nonlinear across
 9    the range of blood Pb levels from the higher concentrations more prevalent in the past to lower
10    concentrations more common today. Thus, we continue to consider it particularly appropriate to
11    focus on the evidence from studies with blood Pb levels closest to those of today's population,
12    which in the last review included studies with study group mean blood Pb levels ranging roughly
13    from 3 to 4 |ig/dL in children aged 24 months to 7 years (Table 3-3 above). As discussed  in
14    sections 3.2 above, this is also consistent with the evidence currently available for this  age group
15    of young children, and the currently available evidence does not include additional C-R slopes
16    for incremental neurocognitive decrement with blood Pb levels at or below this range.  In
17    considering whether this set of functions continues to be well supported by the evidence, as
18    assessed in the ISA (ISA,  section 5.3.2), we note the somewhat wide range in slopes
19    encompassed by these study groups, while also noting the stability of the median.  For example,
20    omission of any of the four slopes considered in the last review does not appreciably change the
21    median (e.g., the median would change from -1.75  IQ points per |ig/dL blood Pb to -1.71  or -
22    1.79).  Thus, we conclude that while differing judgments might be made with regard to inclusion
23    of each of the four study groups, these estimates are generally supported by the current review of
24    the evidence in the ISA. Further, the stability of the median to modifications to this limited
25    dataset lead us to conclude that the currently available evidence continues to support
26    consideration of -1.75 IQ points per |ig/dL blood Pb as a well founded and stable estimate for
27    purposes of describing the neurocognitive impact quantitatively on this age group of U.S.
28    children.
29          In summary, in considering the evidence and information available in this review
30    pertaining to the level of the current Pb standard, we note that the evidence available in this
31    review, as summarized in the ISA, continues to support the air-related IQ loss evidence-based
32    framework, with the inputs that were used in the last  review. These include estimates of air-to-
33    blood ratios ranging from 1:5 to 1:10, with a generally central estimate of 1:7. Additionally, the
34    C-R functions most relevant to blood Pb levels in U.S. children today continue to be provided by
35    the set of four analyses considered in the last review for which the median estimate is -1.75 IQ
36    points per |ig/dL Pb in young children's blood. Thus, we observe that the evidence available in

         January 2013                           4-20            Draft - Do Not Quote  or Cite

-------
 1    this review has changed little if at all with regard to the aspects given weight in the conclusion on
 2    level for the new standard in the last review and would not appear to call into question any of the
 3    basic elements of the standard. In so doing, we additionally recognize that the overall decision
 4    on adequacy of the current standard is a public health policy judgment by the Administrator; we
 5    discuss considerations that inform such judgments in section 4.3 below.

 6         4.2.2  Exposure/Risk-based Considerations
 7           Our consideration of the issue of adequacy of public health protection provided by the
 8    current standard is also informed by the quantitative exposure/risk assessment completed in the
 9    last review, augmented as described in section 3.4 above. We have organized the discussion that
10    follows around two questions to assist us in interpreting the results of the assessment of case
11    studies simulated to meet several different air quality conditions, including those just meeting the
12    current standard.
13         •   What is the level of confidence associated with  estimates of air-related risk
14              generated for simulations just meeting the current Pb standard?
15           As an initial matter, we recognize the significant limitations and complexity associated
16    with the risk/exposure assessment for Pb that are far beyond those associated with similar
17    assessments typically performed for other criteria pollutants. In completing the assessment we
18    were constrained by significant limitations with regard to data and tools particular to the problem
19    at hand. Further, the multimedia and persistent nature of Pb and the role of multiple exposure
20    pathways contribute significant additional complexity to the assessment as compared to other
21    assessments that focus only on the inhalation pathway. As a result, our estimates of air-related
22    exposure and risk are approximate, presented as upper and lower bounds within which we
23    consider air-related risk likely to fall.  We base our description of overall confidence in this
24    characterization of air-related risk on our consideration of the overall design  of the analysis (as
25    described in section 3.4 above), the degree to which key sources of variability are reflected in the
26    design of the analysis, and our  characterization of key sources of uncertainty.
27           In considering the degree to which key sources of variability (discussed in  section 3.4.6
28    above)  are reflected in the design of the analysis, we note the following aspects addressed by the
29    risk assessment.
30       •   Variation in distributions of potential urban residential exposure and risk across  U.S.
31           urban residential areas. This is addressed by the inclusion of three different (location-
32           specific) urban study areas that reflect a diverse set of urban areas in the U.S.
33       •   Representation of a more highly-exposed subset of urban residents potentially exposed at
34           the level of the standard.  This is addressed by the inclusion of the generalized (local)
35           urban study area.

         January 2013                          4-21            Draft - Do Not Quote or Cite

-------
 1       •   Variation in residential exposure to ambient air Pb within an urban area. This is
 2           addressed through the partitioning of the location-specific study areas into exposure
 3           zones to provide some representation of spatial gradients in ambient air Pb and their
 4           interaction with population distribution and demographics.  This was done in a somewhat
 5           more precise manner in the primary Pb smelter case study, which relied on dispersion
 6           modeling to describe gradients, as compared with the manual assignment of gradients
 7           related to air concentration differences among monitors in an area.
 8       •   Inter-individual variability in blood Pb levels. This is addressed though the use of
 9           empirically-derived GSDs to develop blood Pb distribution for the child population in
10           each exposure zone, with GSDs selected particular to each case study populations.
11       •   Inter-individualvariability in IQ response to blood Pb. This is addressed through the use
12           of C-R functions for IQ loss based on a pooled analysis reflecting studies of diverse
13           populations.
14           We also considered key sources of uncertainty (discussed in section 3.4.7 above), in
15    particular the overarching area concerning the precision of the air-related risk estimates, as
16    mentioned above. Associated sources of uncertainty include our inability to simulate changes in
17    air-related Pb as a function of changes in ambient air Pb in exposure pathways other than those
18    involving inhalation of ambient air and ingestion of indoor dust.  This  contributes to the positive
19    bias of the upper bound for the air-related risk estimates.  We additionally recognize the
20    significant uncertainty associated with estimating upper percentiles of the distribution ofair-
21    related blood Pb concentration estimates (and associated  IQ loss estimates) due to limitations in
22    available information.  Lastly, we recognize the uncertainty associated with application of the C-
23    R function at the lower blood Pb levels in the distribution; this relates to the limited
24    representation of blood Pb levels of this magnitude in the dataset from which the C-R function is
25    derived.
26           In the quantitative risk information available in this review, we have air-related risk
27    estimates for simulations just meeting the current standard from one of the location-specific
28    urban case studies (Chicago) and from the generalized (local) urban case study.  With regard to
29    the latter, we note its simplified design that does not include multiple exposure zones, thus
30    reducing the dimensions simulated. We have a reasonable degree of confidence in aspects of the
31    generalized (local) urban case study for the specific situation we consider it to represent (i.e., a
32    temporal pattern of air Pb concentrations that just meets the level of the standard), and when the
33    associated estimates are characterized as approximate, within upper and lower bounds (as
34    described in section 3.4.4 above), while also recognizing  considerable  associated uncertainty.
         January 2013                          4-22             Draft - Do Not Quote or Cite

-------
 1         •   To what extent are the air-related risks remaining upon just meeting the current
 2             Pb standard important from a public health perspective?
 3          In considering public health importance of estimated air-related risks, we consider here
 4    the nature and magnitude of such estimated risks (and attendant uncertainties), including such
 5    impacts on the affected population, and we additionally consider the size of the affected
 6    population.  Based on the evidence available in the last review and consistent with that available
 7    today, the quantitative risk estimates developed in the 2007 REA and augmented slightly in this
 8    review are for decrements in IQ, an established indicator of neurocognitive function.  In
 9    considering these estimates, we recognize that although some neurocognitive effects may be
10    transient, some effects may persist into adulthood, affecting success later in life (ISA, sections
11    2.9.5 and 5.3.15). We additionally recognize the potential population impacts of small changes
12    in population mean values  of metrics such as IQ, presuming a uniform manifestation of lead-
13    related decrement across the range of population IQ (ISA, section 2.9.1), as noted in section 3.3
14    above.
15          Exposures and risks associated with air-related Pb under several  different air quality
16    conditions were estimated in the 2007 REA. As summarized in section 3.4 above, limitations in
17    our modeling tools and data affected our ability to develop precise estimates for air-related
18    exposure pathways and contributed uncertainties. The results are approximate estimates which
19    we describe through the use of rough upper and lower bounds within which we estimate air-
20    related risk to fall. We have recognized a number of uncertainties in the underlying risk
21    estimates from the 2007 REA and in the interpolation approach employed in the new analyses
22    for this review.  We have characterized the magnitude of air-related risk associated with the
23    current standard with a focus on median estimates, for which we have appreciably greater
24    confidence than estimates for outer ends of risk distribution (see section 3.4.7) and on risks
25    derived using the C-R function in which we have greatest confidence (see sections 3.4.3.3.1 and
26    3.4.7). These risk estimates include estimates from the last review for one of the location-
27    specific urban study area populations as well as estimates newly derived in this review based on
28    interpolation from 2007 REA results for the generalized (local) urban case study, which is
29    recognized to reflect a generalized high end of air-related exposure for localized populations.
30    Taken together,  these estimates indicate air-related risk associated with just meeting the current
31    standard of a magnitude generally within the bounds of roughly 1 and 3  points IQ loss.
32          In considering the importance of such risk from a public health perspective, we also
33    consider the size of at-risk populations represented by the REA case studies. As discussed in
34    section 3.4 above, the generalized (local) urban case study is considered to represent a localized
35    urban population exposed near the level of the standard, such as a very small, compact
36    neighborhood near a source contributing to air Pb concentrations just meeting the standard.  This

         January 2013                           4-23            Draft - Do Not Quote or Cite

-------
 1    case study provides representation in the risk assessment for such small populations at the upper
 2    end of the gradient in ambient air concentrations expected to occur near sources; thus estimates
 3    for this case study reflect exposures nearest the standard being evaluated. While we do not have
 4    precise estimates of the number of young children living in such areas of the U.S. today, we have
 5    information that informs our understanding of their magnitude. For example, as discussed in
 6    section 3.3 above (Table 3-3), we estimate  there to be approximately 2,400 children,  aged 5
 7    years and younger, residing within 0.5 km of a monitor with air Pb concentrations above the
 8    current standard. We further observe several additional monitors with maximum 3-month Pb
 9    concentrations that fall below but within 10% of the current standard level (as noted in section
10    3.3 above); an estimated 265 young children reside near these monitors.  Thus, together we
11    estimate some 2700 children, aged 5 years  and younger, living in localized areas with elevated
12    air Pb concentrations that are above or near the current standard.  Based on the 2010 census
13    estimates of approximately 24.3 million children in the U.S. aged five years or younger, this
14    represents approximately one hundredth of one percent of this age group in the U.S.7  This
15    indicates the size of the population of young children of this age living in areas in close
16    proximity to areas where air Pb concentrations may be above or near the current standard to be
17    approximately 2700, which is approximately a hundredth of one percent of the full population of
18    correspondingly aged children.8
19           In summary, we recognize substantial uncertainty inherent in the REA estimates of air-
20    related risk associated with conditions just  meeting the current standard, which we characterize
21    as approximate and falling somewhat above the rough lower bound of 1 and below the rough
22    upper bound of 3 IQ points.9  This approximate estimate of risk for children living in such areas
23    is generally overlapping with and consistent with the evidence-based air-related IQ loss estimates
24    summarized in section 4.2.1  above.  With regard to the the importance of the estimated risks
25    from a public health perspective, we note, based on the  currently  available monitoring and
26    census data, the estimated size of the affected population to be approximately 2700 young
27    children, representing slightly more than one one hundredth of a percent of the full population of
             7 While these estimates pertain to the age group of children aged 5 years and younger, we additionally note
      that a focus on an alternative age range, such as inclusive of slightly older children (e.g., through age 7), while
      increasing the number for children living in such locations, would not be expected to appreciably change the
      percentage of the full U.S. age group that the subset represents.
             8 This estimate includes children in areas where recent information indicates the standard is exceeded
      because areas not in atttainment with the standard are required to attain the standard as expeditiously as practicable,
      but no later than five years after designation.  Accordingly, these areas are for present purposes treated as areas with
      air Pb concentrations just meeting the current standard and are included for purposes of identifying the size of the at-
      risk population residing in areas likely to have air Pb concentrations near the current standaard.
             9 We note that the value of the upper bound is influenced by risk associated with exposure pathways that
      were not varied with alternative standard levels, a modeling limitation with the potential to contribute to
      overestimation of the upper bound with air quality scenarios involving air Pb levels below current conditions for the
      study area (see sections  3.4.4 and 3.4.7 above).
         January 2013                           4-24              Draft - Do Not Quote or Cite

-------
 1    similarly aged children. In considering the exposure/risk estimates during the last review, the
 2    Administrator took note of the important uncertainties and limitations associated with these
 3    assessments and, while placing less weight on the assessment estimates, he recognized the
 4    quantitative risk estimates to be "roughly consistent with and generally supportive" of those
 5    estimated by the evidence-based framework. In our consideration of the risk estimates
 6    considered in the last review, as well as the estimates for the current standard that have been
 7    newly developed in this review for the generalized (local) urban case study (see section
 8    3.4.3.3.2), we agree with the previous characterization. As would be expected by the use of
 9    interpolation, the newly derived estimates are consistent with the estimates for similar air quality
10    scenarios that were available in the last review.

11    4.3   PRELIMINARY STAFF CONCLUSIONS ON THE PRIMARY STANDARD
12          This section describes preliminary staff conclusions regarding adequacy of the current
13    primary Pb standard.  These conclusions are based on considerations described above and in the
14    discussion below regarding the currently available scientific evidence summarized in the third
15    draft ISA and prior CDs and the risk and exposure information drawn from the 2007 REA. The
16    final PA will draw upon the final ISA, developed after CAS AC review and public comment on
17    the third draft ISA.  Further, staff conclusions presented in the final PA will take into account
18    advice from CAS AC and public comment on the draft PA and the preliminary conclusions.
19          Taking into consideration the discussions responding to specific questions above in this
20    and the prior chapter, this section addresses the following overarching policy question.
21         •   Does the currently available scientific evidence- and exposure/risk-based
22             information, as reflected in the ISA and REA, support or call into question the
23             adequacy of the protection afforded by the current Pb standard?
24          Our response to this question takes into consideration the discussions responding to the
25    specific policy-relevant questions in prior sections of this document (see sections 3.1-3.4, 4.2.1,
26    and 4.2.2). In so doing, we focus first on consideration of the evidence, including that newly
27    available in this review, and the extent to which it alters key conclusions supporting the current
28    standard. We then turn to consideration of the quantitative risk estimates drawn from the 2007
29    REA, and associated limitations and uncertainties.  We additionally consider the public health
30    policy judgments and judgments about the uncertainties inherent in the scientific evidence and
31    quantitative analyses that are inherent in decisions on the adequacy of the current primary Pb
32    standard.
33          We first recognize the complexity involved in considering the adequacy of protection in
34    the case  of the primary Pb standard, which differs substantially from that involved in
35    consideration of the NAAQS for other pollutants.  Unlike the case for Pb, the other pollutants for

         January 2013                          4-25            Draft - Do Not Quote or Cite

-------
 1    which NAAQS are set involve only inhalation exposure pathways, a relatively simpler context.10
 2    In addition, generally an important component of the scientific evidence base considered in
 3    reviewing the adequacy of NAAQS for other pollutants is the availability of studies that have
 4    investigated associations between current concentrations of the pollutant in ambient air
 5    (including in circumstances where the current standard is met) and the occurrence of health
 6    effects judged plausibly related to ambient air exposure to the NAAQS pollutant.  While such
 7    studies, targeting locations near air Pb sources such as smelters, were available at the time the Pb
 8    NAAQS was initially set in 1978, and to a much more limited degree at the time of the last
 9    review, such studies of health effects under air quality conditions near those reflecting the
10    current standard are not available in this review. Rather, the evidence base that supports our
11    conclusions in this review includes most prominently epidemiological studies reporting
12    associations of blood Pb levels  in U.S. populations, including the particularly at-risk population
13    of young children with health effects judged plausibly related to Pb exposures.  Support for our
14    conclusions regarding the  plausibility for ambient air Pb to play a role in such findings derives,
15    in part, from studies linking Pb  in ambient air with the occurrence of health effects. However,
16    such studies (dating from the past or from other countries) involve ambient air Pb concentrations
17    many times greater than those that would meet the current standard. Thus, in considering the
18    adequacy of the current Pb standard, rather than considering studies that have investigated
19    current concentrations of Pb in  ambient air (including in locations where the current standard is
20    met) and the occurrence of health effects, we primarily consider the evidence for, and risk
21    estimated from, models based upon key  relationships, such as those between ambient air Pb, Pb
22    exposure, blood Pb and health effects. This evidence, with its associated limitations and
23    uncertainties, contributes to our conclusions regarding a relationship between ambient air Pb
24    conditions under the current standard and health effects.
25          In considering the  currently available evidence, staff gives great weight to the long-
26    standing body of evidence on the health effects of Pb, augmented in some aspects since the last
27    review, which continues to support identification of neurocognitive effects in young children as
28    the most sensitive endpoint associated with Pb exposure, as discussed in sections 3.2 and 3.3
29    above. The evidence continues to indicate that a standard that provides protection from
30    neurocognitive  effects in young children additionally provides protection for other health effects
31    of Pb, such as cardiovascular effects later in life. Application of the evidence-based, air-related
32    IQ loss framework, developed in the last review, continues to provide a useful approach for
             10 As described in sections 1.3 and 3.1 above, exposure to Pb from ambient air, unlike exposures for other
      NAAQS pollutants, involves both ingestion and inhalation exposure pathways. As an additional complication, other
      (nonair) sources of Pb also contribute to these pathways, particularly to ingestion exposure pathways. In the case of
      Pb, the internal biomarker, blood Pb, is an indicator of exposure across different pathways and routes (as discussed
      in section 3.1 above).
         January 2013                           4-26             Draft - Do Not Quote or Cite

-------
 1    considering and integrating the evidence on relationships between Pb in ambient air and Pb in
 2    children's blood and risks of neurocognitive effects (IQ loss).  The currently available evidence
 3    base, while somewhat expanded since the last review, is not appreciably expanded or supportive
 4    of appreciably different conclusions with regard to air-to-blood ratios (section 3.1 above) or C-R
 5    functions for neurocognitive decrements (section 3.2 above) in young children.
 6          As in the last review, uncertainties remain in our understanding of important aspects of
 7    ambient air Pb exposure and associated health effects.  For example, important uncertainties
 8    remain, both with regard to air-to-blood ratios that reflect the relationship between
 9    concentrations of Pb in ambient air and air-related Pb in children's blood, and to  estimates of the
10    slope of the C-R function for neurocognitive impacts (IQ loss) at lower blood Pb levels. With
11    regard to the former, we note particularly the limitations associated with the available studies and
12    gaps in the evidence base with regard to studies that have investigated such quantitative
13    relationships under conditions pertaining to the current standard (e.g., in localized areas near air
14    Pb sources where the standard is met in the U.S. today).  Further, in considering our reliance on
15    the evidence in performing quantitative modeling of exposure and risk we additionally note
16    important uncertainties associated with relationships between  ambient air Pb and outdoor
17    soil/dust Pb and indoor dust Pb that particularly affect our quantitative estimates  of air-related
18    risk under conditions of lower ambient air Pb levels and lower blood Pb levels (73 FR 66981).
19    These critical exposure pathways are also inherent in the evidence-based air-related IQ loss
20    framework, encompassed within the estimates of air-to-blood  ratios.  Thus, we recognize
21    uncertainties related to our understanding of these and other processes that also contribute
22    uncertainty to our application of the evidence-based framework. We consider this uncertainty to
23    be increased with application of the framework for levels below the current standard given the
24    weaker linkage with existing evidence, as noted below.
25          We additionally take note of the role of public health policy judgments in the
26    Administrator's setting of the current standard in 2008,  as summarized in section 4.1.1.2 above.
27    Such judgments are inherent in setting the standard, and particularly in using the  evidence-based
28    framework to inform the decision on level for the standard, and in considering the exposure/risk
29    information. We recognize that public health policy judgments always play an important role in
30    each NAAQS review for each pollutant. One type of public health policy judgments focuses on
31    how to consider the nature and magnitude of the array of uncertainties that are inherent in the
32    scientific evidence  and analyses.  These judgments are traditionally made with a recognition that
33    our understanding of the relationships between the presence of a pollutant in ambient air and
34    associated health effects is based on a broad body of information encompassing not only more
35    established aspects of the evidence but also aspects in which there may be substantial
36    uncertainty. In the case of the Pb NAAQS review, we recognize increased uncertainty in

         January 2013                          4-27             Draft - Do Not Quote or Cite

-------
 1    characterizing the IQ response to blood Pb levels below those represented in the evidence base.
 2    We also recognize increased uncertainty in projecting the magnitude of blood Pb response to
 3    ambient air Pb concentrations at and below the level of the current standard.  We recognize this
 4    increased uncertainty particularly in light of the multiple factors that play a role in such a
 5    projection (e.g., meteorology, atmospheric dispersion and deposition, human physiology and
 6    behavior), for each of which we recognize attendant uncertainties. Collectively, these aspects of
 7    the evidence and associated uncertainties contribute to a recognition that for Pb, as for other
 8    pollutants, the available health effects evidence generally reflects a continuum, consisting of
 9    ambient levels at which scientists generally agree that health effects are likely to occur, through
10    lower levels at which the likelihood and magnitude of the response become increasingly
11    uncertain.
12           Reviews may also require judgments as to the point at which health effects become
13    important from a public health perspective.  In the case of Pb, such a judgment is explicitly made
14    in considering the public health significance  of one to two points IQ loss in at-risk populations,
15    such as young children. This type of judgment also includes consideration of the IQ loss
16    estimates yielded by the air-related IQ loss evidence-based framework for specific combinations
17    of standard level, air-to-blood ratio and C-R function.  With regard to public health significance
18    of one to two points IQ loss in young children, staff gives weight to the comments of CAS AC
19    and some public commenters in the last review which recognized such a  magnitude of IQ loss to
20    be of public health significance and recommended that a very high percentage of the population
21    be protected from such a magnitude of IQ loss (73 FR 67000).u With this objective in mind, we
22    consider the extent to which the air-related IQ loss evidence-based framework informs
23    consideration of standards that might be concluded to provide such a level of protection.  In so
24    doing, we first recognize that the IQ loss estimates produced with the evidence-based framework
25    do not correspond to a specific quantitative public health policy goal for  air-related IQ loss  that
26    would be acceptable or unacceptable for the entire population of children in the U. S. Rather, the
27    conceptual context for the evidence-based framework is that it provides estimates for the mean
28    air-related IQ loss of a subset of the population of U.S. children (i.e., the subset living in close
29    proximity to air Pb sources that contributed to elevated air Pb concentrations that equal the
30    current level of the standard). This is the subset expected to  experience air-related Pb exposures
31    at the high end of the national distribution of such exposures. The associated mean IQ loss
32    estimate is the average  for this highly exposed subset and is not the average  air-related IQ loss
             1: Our focus on IQ, as noted in section 3.3 above, reflects recognition of IQ being a well established, widely
      recognized and rigorously standardized measure of neurocognitive function, as well as a global measure reflecting
      the integration of numerous processes (ISA, section 5.3.2; 2006 CD, sections 6.2.2 and 8.4.2). Use of IQ in this
      framework is thus considered to appropriately also reflect neurocognitive effects more generally.
         January 2013                           4-28            Draft - Do Not Quote or Cite

-------
 1    projected for the entire U.S. population of children. Further, we recognize uncertainties
 2    associated with those estimates, which increase with estimates associated with successively
 3    lower standard levels.
 4           For the current standard level of 0.15 |ig/m3, an air-to-blood ratio estimate of 7 |ig/dL per
 5    ng/m3 (which  we note as reasonably representative of the range supported by the evidence) and a
 6    C-R slope of -1.75 IQ points per |ig/dL, the IQ loss estimate using the evidence-based
 7    framework is 1.8 points.12  As noted above, this value is considered to be an estimate, with
 8    attendant uncertainties, of mean air-related IQ loss of a subset of the population of U.S. children
 9    in the high end of the exposure distribution for air-related Pb.13 As noted in section 4.2.2 above,
10    our current information on numbers of young children living near monitors above or within 10%
11    of the current Pb  standard indicates the size of this population subset to be on the order of 2700
12    children aged  5 years or younger, which would correspond to approximately one hundredth of
13    one percent of the U.S. population of this age children (estimated at approximately 24 million in
14    2010 census).  A primary objective of the monitoring network for Pb is to identify and monitor
15    sites of maximum concentration in areas anticipated to be at risk of exceeding the NAAQS.
16    While we acknowledge the possibility that the monitoring data thus far available may not
17    identify every  occurrence of elevated Pb concentrations, the size of such a population subset can
18    still be concluded to fall well below one tenth  of one percent of the full population of children
19    aged 5 years or younger in the U.S. today.  Thus, we conclude that the  current evidence, as
20    considered within the conceptual and quantitative context of the evidence-based framework, and
21    current air monitoring information indicates that the current standard would be expected to
22    achieve the public health policy goal recommended by CASAC in the last Pb NAAQS review
23    that IQ loss on the order of one to two IQ points be "prevented in all but a small percentile  of the
24    population" (73 FR 67000).
25           In drawing conclusions from application of the evidence based  framework with regard to
26    adequacy of the current standard, we further recognize the degree to which IQ loss estimates
27    drawn from the air-related  IQ loss evidence-based framework reflect mean blood Pb levels
28    below those represented in the currently available evidence for young children.  For example, in
29    the case of the current standard level of 0.15 |ig/m3, multiplication by the air-to-blood ratio of 7
             12 Using an air-to-blood ratio estimate of 8 ug/dL per ug/m3 also yields an IQ loss estimate of
      approximately 2 IQ points.
             13 In giving weight to consideration of the evidence within the context of the air-related IQ loss evidence-
      based framework, we note that the air-related IQ loss estimated by the framework is for the mean of the population
      subset described above. Given the lack of data on the distribution of the air-related portion of blood Pb and on the
      extent to which distributions of air-related blood Pb levels might differ or correlate with total blood Pb that may be
      more greatly influenced by other (nonair) Pb exposure pathways,  estimates are not available  that would correspond
      to upper percentiles of the IQ loss distribution for this population subset.  Any such estimates would have
      substantial associated uncertainty.
         January 2013                           4-29            Draft - Do Not Quote or Cite

-------
 1    yields a mean air-related blood Pb level of 1.05 |ig/dL, which is half the level of the lowest blood
 2    Pb subgroup of pre-school children in which neurocognitive effects have been observed (Table
 3    3-2 above; Miranda et al., 2009) and well below the means of subgroups for which continuous C-
 4    R functions have been estimated (Table 3-3 above).14 Such an extension below the lowest
 5    studied levels may be viewed as appropriate given the lack of identified blood Pb  level threshold
 6    in the current evidence base for neurocognitive effects and the need for the NAAQS to provide a
 7    margin of safety.15 We note, however, that the framework IQ loss estimates for still lower
 8    potential  standard levels represent still greater extrapolations from the current evidence base with
 9    corresponding increased uncertainty.
10            In recognition of the role of public health policy judgments in drawing conclusions as to
11    adequacy of the Pb NAAQS, we consider the availability of new information or new commonly
12    accepted  guidelines or criteria within the public health community with regard to the public
13    health significance of specific  IQ decrements in exposed, at-risk populations that might inform
14    public health policy judgments on the appropriate degree of public health protection that should
15    be afforded to protect against risk of such neurocognitive effects in at-risk populations,16 such as
16    children living  near air Pb sources. As an initial matter, we note that no such new information,
17    guidelines or criteria are described in the ISA. In further considering the occurrence of any new
18    actions by public health  agencies that might indicate the availability of new information,
19    guidelines or criteria for interpreting public health significance of such effects, we note that the
20    CDC has revised the blood Pb level used to prioritize young children  for whom they recommend
              14 We recognize that children also have Pb in their blood derived from other (nonair) sources. The
      evidence- based air-related IQ loss framework is used, however, to estimate IQ loss attributable to air-related Pb
      because the NAAQS is intended to protect against risks from ambient air-related Pb. While children also have Pb in
      their blood derived from other (nonair) sources, the evidence indicates that the risk per increment of blood Pb is
      greater for children with the lowest blood Pb levels, as noted in section 3.2 above. Thus, the focus on estimating IQ
      loss attributable solely to air-related lead (i.e., assuming the presence of no other blood Pb) is a conservative
      approach, the use of which contributes to the margin of safety provided by the standard.
              15 As noted in section 1.2.1 above, the requirement that primary standards include an adequate margin of
      safety was intended to address uncertainties associated with inconclusive scientific and technical information
      available at the time of standard setting. It was also intended to provide a reasonable degree of protection against
      hazards that research has not yet identified. Both kinds of uncertainties are components of the risk associated with
      pollution at levels below those at which human health effects can be said to occur with reasonable scientific
      certainty. Thus, in selecting primary standards that includes an adequate margin of safety, the Administrator is
      seeking not only to prevent pollutant levels that have been demonstrated to be harmful but also to prevent lower
      pollutant levels that may pose an unacceptable risk of harm, even if the risk is not precisely identified as to nature or
      degree. The CAA does not require that primary  standards be set at a zero-risk level or at background concentration
      levels, but rather at a level that reduces risk sufficiently so as to protect public health with an adequate margin of
      safety.
              16The term at-risk populations is used here, rather then the phrase "sensitive populations" used in the last
      review. In using the term at-risk, we intend the same meaning as has traditionally been intended for the term
      "sensitive" consistent with discussion in section 1.2.1 above.

          January 2013                              4-30             Draft - Do Not Quote or Cite

-------
 1    particular follow-up health protective actions, as summarized in section 3.1 above.17  The CDC
 2    decision, while emphasizing the critical importance of primary prevention of Pb exposure,
 3    provides no new guidelines or criteria with regard to the significance of specific IQ decrements
 4    or judgments on  appropriate public health protection from risk of neurocognitive effects, and
 5    their consideration of the evidence base at the time of their decision does not substantively differ
 6    from that presented in the current draft ISA (CDC, 2012). Thus, we are aware of no new
 7    information or new commonly accepted guidelines or criteria within the public health
 8    community for interpreting public health significance of neurocognitive effects in the context of
 9    a decision on adequacy of the current Pb standard.
10           With respect to exposure/risk-based considerations, we recognize the complexity of the
11    REA modeling analyses and the associated limitations and uncertainties.  We additionally note
12    the differences among the case studies included in the REA and the extent to which they inform
13    our understanding of different aspects of the risk associated with air-related Pb in the U.S.  For
14    example, the location-specific case studies indicate the distribution of population risk in urban
15    areas with differing types of Pb sources and gradients in air Pb concentrations as well as in
16    population density, while the generalized urban (local) case study indicates the magnitude of air-
17    related risk associated with those specific localized circumstances where  air  concentrations just
18    meet the Pb standard (regardless of source type).  We agree with conclusions drawn in the 2008
19    review that the quantitative risk estimates, with a focus on those for the generalized (local) urban
20    case study, are "roughly consistent with and generally supportive" of estimates from the
21    evidence-based air-related IQ loss framework (73 FR 67006). We further take note of the
22    increasing uncertainty recognized for air quality scenarios involving air Pb concentrations
23    increasingly below the current conditions for each case study,  recognizing that such uncertainty
24    is due in part to modeling limitations deriving from uncertainty regarding relationships between
25    ambient air Pb and outdoor soil/dust Pb and indoor dust Pb (as noted in section 3.4 above).
26           Based on the above considerations, we reach the preliminary conclusion that the current
27    body of evidence, in combination with the exposure/risk information,  supports a primary
28    standard as protective as the current standard. Further consideration of the evidence and
29    exposure/risk information available in this review and its attendant uncertainties and limitations,
30    as well as consideration of the availability of other information that might also inform public
31    health policy judgments by the Administrator, leads us to reach the additional preliminary
32    conclusion that it is appropriate to consider retaining the current standard without revision.
33    These preliminary conclusions are based on consideration of the health effects evidence,
34    including consideration of this evidence in the context of the evidence-based, air-related IQ loss
             17 Uses identified for the CDC reference level include is the identification of " high-risk childhood
      populations and geographic areas most in need of primary prevention" (CDC, 2012).
         January 2013                          4-31            Draft - Do Not Quote or Cite

-------
 1    framework, and in combination with the exposure/risk information (chapter 3 and sections 4.2.1
 2    and 4.2.2 above) and the uncertainties attendant with both, recognizing the complexities and
 3    limitations in the evidence base in reaching conclusions regarding the magnitude of risk
 4    associated with the current standard, as well as the increasing uncertainty of risk estimates for
 5    lower air Pb levels.  Based on these considerations, these preliminary conclusions also recognize
 6    what may be considered reasonable judgments on the public health implications of the blood Pb
 7    levels and risk estimated for air-related Pb under the current standard, including the public health
 8    significance of the Pb effects being considered, as well as aspects of the use of the evidence-
 9    based framework that may be considered to contribute to the margin of safety.
10           In reaching these preliminary  conclusions, we additionally note that different public
11    health policy judgments could lead to different conclusions regarding the extent to which the
12    current standard provides protection of public health with an adequate margin of safety.  Such
13    public health policy judgments include those related to the appropriate degree of public health
14    protection that should be afforded to protect against risk of neurocognitive effects in at-risk
15    populations, such as IQ loss in young children, as well as with regard to the appropriate weight
16    to be given to differing aspects of the evidence and exposure/risk information, and how to
17    consider their associated uncertainties. For example,  different judgments might give greater
18    weight to more uncertain aspects of the evidence or reflect a differing view with regard to margin
19    of safety.  As noted in section 4.1 above, in establishing primary standards under the Act that, in
20    the Administrator's judgment, are requisite to protect public health with an adequate margin of
21    safely, the Administrator seeks to establish standards that are neither more nor less stringent than
22    necessary for this purpose.  The Act does not require that primary standards be set at a zero-risk
23    level, but rather at a level that avoids  unacceptable risks to public health, even if the risk is not
24    precisely identified as to nature or degree.  The requirement that primary standards provide an
25    adequate margin of safety was intended to address uncertainties associated with inconclusive
26    scientific and technical information available at the time of standard setting.  It was also intended
27    to provide a reasonable degree of protection from hazards that research has not yet identified.
28           In this context, we recognize that the uncertainties and limitations associated with the
29    many aspects of the estimated relationships between air Pb concentrations and blood Pb levels
30    and associated health effects are amplified with consideration of increasingly lower air
31    concentrations. We believe the current evidence supports the conclusion that the current
32    standard is requisite to protect public  health with an adequate margin of safety.  In staffs view,
33    based on the current evidence there is appreciable uncertainty associated with drawing
34    conclusions regarding whether there would be reductions in risk to public health from alternative
35    lower levels as compared to the level  of the current standard. Thus, we conclude that the basis
36    for any consideration of alternative lower standard levels would reflect different public health

         January 2013                           4-32            Draft - Do Not Quote or Cite

-------
 1    policy judgments as to the appropriate approach for weighing uncertainties in the evidence and
 2    for providing requisite protection of public health with an adequate margin of safety.
 3           In summary, the newly available health effects evidence, critically assessed in the ISA as
 4    part of the full body of evidence, reaffirms conclusions on the broad array of effects recognized
 5    for Pb in the last review. Further, staff observes the general consistency of the current evidence
 6    with the evidence available in the last review with regard to key aspects on which the current
 7    standard is based, including those particular to the evidence-based framework developed in the
 8    last review.  Staff additionally observes that quantitative risk estimates associated with the
 9    current standard, based on the risk assessment performed in the last review, indicate a level of
10    risk generally consistent with conclusions drawn from the evidence using the evidence-based
11    framework.  Staff additionally recognizes the limitations and uncertainties associated with the
12    currently available information.  These considerations provide the basis for the preliminary staff
13    conclusion that consideration should be given to retaining the current standard, without revision.
14    Accordingly, and in light of staff preliminary conclusions that it is appropriate to consider the
15    current standard to be adequate, we have not identified any potential alternative standards for
16    consideration in this review.

17    4.4   KEY UNCERTAINTIES AND AREAS FOR FUTURE RESEARCH AND DATA
18         COLLECTION
19           In this section,  we highlight key uncertainties associated with reviewing and establishing
20    NAAQS for Pb.  Such key uncertainties and areas for future health-related research, model
21    development, and data gathering are outlined below.  In some cases, research in these areas can
22    go beyond aiding standard setting to aiding in the development of more efficient and effective
23    control strategies. We note, however, that a full set of research recommendations to meet
24    standards implementation and strategy development needs is beyond the scope of this discussion.
25    Rather, listed below are key uncertainties and research questions and data gaps that have been
26    thus far highlighted in  this review of the health-based primary standard.
27       •   A critical aspect of our consideration of the evidence and the quantitative risk assessment
28           in this review is our understanding of the C-R relationship between blood Pb levels in
29           young children and neurodevelopmental effects, specifically IQ decrement. Additional
30           epidemiological research involving substantially-sized populations with mean blood Pb
31           levels closer to those common in today's population of young children, particularly those
32           less affected by higher Pb exposures earlier in childhood, would help to reduce
33           uncertainty in our estimates of IQ decrement associated with these lower blood Pb levels,
34           and accordingly, in characterizing Pb health effects.
35
36       •   There remains uncertainty in the evidence base with regard to the exposure circumstances
37           (timing, duration, magnitude and frequency) eliciting effects in older children and adults.
         January 2013                          4-33             Draft - Do Not Quote or Cite

-------
 1          Effects of particular focus include effects on the nervous system later in life,
 2          cardiovascular function, and delayed onset of puberty.

 3       •  Alzheimer's-like pathology has been reported in aged laboratory animals (non-human
 4          primates and rodents) exposed to Pb early in life, however uncertainly remains with
 5          regard to relationships of such pathology with altered function/behavior.

 6       •  Epidemiologic studies indicate detrimental effects of Pb on sperm production and quality,
 7          often in occupational cohorts.  Uncertainty remains regarding these effects in otherwise
 8          healthy cohorts without occupational Pb exposure or other underlying medical
 9          conditions.

10       •  Quantitative estimation of blood Pb levels in children and other at-risk subgroups in
11          response to various exposure circumstances, including air-related exposure pathways,
12          would benefit from research in a number of areas including:
13             o   Factors affecting relationships between Pb in ambient air and Pb in blood;
14             o   Temporal aspects of changes in blood Pb levels associated with changes in
15                 ambient air Pb;
16             o   Interindividual variability in blood Pb levels and methods for characterizing
17                 interindividual variability, including consideration of both empirical and
18                 mechanistic methods;
19             o   Apportionment of blood Pb levels with regard to exposure pathway contributions,
20                 particular distinctions pertinent to policy-relevant exposures and background
21                 sources; and
22             o   Blood Pb model performance evaluations, with emphasis on applications
23                 pertaining to blood Pb response to ambient air-related pathways and responses to
24                 changes in exposures for those pathways.

25       •  An important aspect of exposure and risk posed by ambient air Pb is the contribution of
26          ambient air Pb to indoor dust Pb. Research on the relationship between Pb in these two
27          media, both generally, as well  as in specific environments, and also with regard to aspects
28          associated with mechanistic modeling (e.g., air exchange rates, home cleaning frequency
29          and efficiency) would reduce uncertainty in models and methods applied in consideration
30          of this pathway in future reviews.

31       •  An understanding of the spatial gradient of ambient air Pb concentrations, and associated
32          particle sizes, in urban residential areas, as well as near Pb sources, is an important aspect
33          to our implementation of the NAAQS for Pb and a key element in assessing exposure and
34          risk.  Additional research in this area is needed.
35             o   Current limitations in this area additionally contribute uncertainty to
36                 characterization of ambient air Pb levels in the risk assessment and associated
37                 exposure and risk estimates. Research in the area of characterizing spatial
38                 variation in air Pb concentrations in different environments and related to different
39                 air sources would help to reduce this uncertainty. Of particular interest is the
40                 potential for systematic trends in the relationship between ambient air Pb
41                 concentrations and  distribution of urban residential populations. Examples of

         January 2013                          4-34            Draft - Do Not Quote or Cite

-------
 1                 locations of interest include neighborhoods downwind of airports with substantial
 2                 leaded aviation gasoline usage or in the vicinity of older roads with substantial
 3                 historical use of leaded gasoline, as well as inner city neighborhoods, both those in
 4                 which substantial reconstruction has occurred and more static areas.

 5       •  An important aspect to this review is the relationship between ambient air Pb and outdoor
 6          dust and surface soil Pb concentrations, including the temporal  dynamics of that
 7          relationship and variation in that for different environments. Research to improve our
 8          understanding of these areas would contribute to reducing associated uncertainty with
 9          regard to characterization of the relationship between air and soil Pb and the impact of
10          changes in air Pb on outdoor soil Pb levels over time.
         January 2013                           4-3 5            Draft - Do Not Quote or Cite

-------
  1     4.5   REFERENCES

  2     Bellinger, D. C. and Needleman, H. L. (2003) Intellectual impairment and blood lead levels [letter]. N. Engl. J. Med.
  3            349:500.

  4     Canfield, R. L.; Henderson, C. R., Jr.; Cory-Slechta, D. A.; Cox, C.; Jusko, T. A.; Lanphear, B. P. (2003)
  5            Intellectual impairment in children with blood lead concentrations below 10 ug per deciliter. N. Engl. J.
  6            Med. 348: 1517-1526.

  7     Centers for Disease Control and Prevention (2012) CDC Response to Advisory Committee on Childhood Lead
  8            Poisoning Prevention Recommendations in "Low Level Lead Exposure Harms Children: A Renewed Call
  9            of Primary Prevention". Atlanta, GA: U.S. Department of Health and Human Services, Public Health
10            Service. June 7.

11     Henderson, R. (2007a) Letter from Dr. Rogene Henderson, Chair, Clean Air Scientific Advisory Committee, to
12            Administrator Stephen L. Johnson.  Re: Clean Air Scientific Advisory Committee's (CASAC) Review of
13            the 1st Draft Lead Staff Paper and Draft Lead Exposure and Risk Assessments. March 27, 2007.

14     Henderson, R. (2007b) Letter from Dr. Rogene Henderson, Chair, Clean Air Scientific Advisory Committee, to
15            Administrator Stephen L. Johnson.  Re: Clean Air Scientific Advisory Committee's (CASAC) Review of
16            the 2nd Draft Lead Human Exposure and Health Risk Assessments.  September 27, 2007.

17     Henderson, R. (2008a) Letter from Dr. Rogene Henderson, Chair, Clean Air Scientific Advisory Committee, to
18            Administrator Stephen L. Johnson.  Re: Clean Air Scientific Advisory Committee's (CASAC) Review of
19            the Advance Notice of Proposed Rulemaking (ANPR) for the NAAQS for lead. January 22, 2008.

20     Henderson, R. (2008b) Letter from Dr. Rogene Henderson, Chair, Clean Air Scientific Advisory Committee, to
21            Administrator Stephen L. Johnson.  Re: Clean Air Scientific Advisory Committee's (CASAC) Review of
22            the Notice of Proposed Rulemaking for the NAAQS for lead. July 18, 2008.

23     Hilts, S. R. (2003) Effect of smelter emission reductions on children's blood lead levels. Sci. Total Environ. 303: 51-
24            58.

25     Lanphear, B. P.; Hornung, R.; Khoury, J.; Yolton, K.; Baghurst, P.; Bellinger, D. C.; Canfield, R. L.; Dietrich, K.
26            N.; Bornschein, R.; Greene, T.; Rothenberg, S. J.; Needleman, H. L.; Schnaas, L.; Wasserman, G.;
27            Graziano, J.; Roberts, R. (2005) Low-level environmental lead exposure and children's intellectual
28            function: an international pooled analysis. Environ. Health Perspect. 113: 894-899.

29     Tellez-Rojo, M. M.; Bellinger, D. C.; Arroyo-Quiroz, C.; Lamadrid-Figueroa, H.; Mercado-Garcia, A.;  Schnaas-
30            Arrieta, L.; Wright, R. O.; Hernandez-Avila, M.; Hu, H. (2006) Longitudinal associations between blood
31            lead concentrations < 10 ug/dL and neurobehavioral development in environmentally-exposed children in
32            Mexico City. Pediatrics 118: e323-e330.

33     U.S. Environmental Protection Agency. (2006) Air  Quality Criteria for Lead. Washington, DC, EPA/600/R-
34            5/144aF. Available online at: www.epa.gov/ncea/

35     U.S. Environmental Protection Agency. (2007a) Lead: Human Exposure and Health Risk Assessments for Selected
36            Case Studies, Volume I. Human Exposure  and Health Risk Assessments - Full-Scale and Volume II.
3 7            Appendices.  Office of Air Quality Planning and Standards, Research Triangle Park, NC. EPA-452/R-07-
38            014aandEPA-452/R-07-014b.

39     U.S. Environmental Protection Agency. (2007b) Review of the National Ambient Air Quality Standards for Lead:
40            Policy  Assessment of Scientific and Technical Information, OAQPS Staff Paper.  EPA-452/R-07-013.
41            Office  of Air Quality Planning and Standards, Research Triangle Park.


          January 2013                               4-36              Draft - Do Not Quote or Cite

-------
1    U.S. Environmental Protection Agency. (2012) Integrated Science Assessment for Lead (Third External Review
2            Draft). Washington, DC, EPA/600/R-10/075C. Available online at:
3            http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_2010_isa.html
         January 2013                              4-37              Draft - Do Not Quote or Cite

-------
 1             5   WELFARE EFFECTS AND EXPOSURE/RISK INFORMATION

 2           This chapter presents key aspects of the current evidence of lead-related welfare effects,
 3    and presents exposure and risk information from the last review in the context of the currently
 4    available information.  Staff has drawn from the EPA's synthesis of the scientific evidence
 5    presented in the third draft Integrated Science Assessment for Lead (USEPA, 2012; henceforth
 6    referred to as the ISA) and the 2006 Air Quality Criteria Document for Lead (USEPA, 2006;
 7    henceforth referred to as the 2006 CD), and from the screening level risk assessment performed
 8    in the last review and described in Lead Human Exposure and Health Risk Assessments and
 9    Ecological Risk Assessment for Selected Areas. Pilot Phase (documented in ICF, 2006;
10    henceforth referred to as the 2006 REA). This chapter is organized into two sections regarding
11    the currently available welfare effects evidence (section 5.1) and the exposure and risk
12    information (section 5.2) interpreted in light of currently available evidence. Presentation within
13    these sections is organized to address key policy-relevant questions for this review concerning
14    the evidence and exposure/risk information, building upon the questions included in the IRP
15    (IRP, section 3.2).

16    5.1   WELFARE EFFECTS INFORMATION
17           Lead has been demonstrated to have harmful effects on reproduction and development,
18    growth, and survival in many species as described in the assessment of the evidence available in
19    this review and consistent with the conclusions drawn in past CDs (ISA, section 2.7; 2006 CD).
20    A number of studies on Pb ecological effects are newly available in this review and are critically
21    assessed in the ISA as part of the full body  of evidence.  The newly available  evidence reaffirms
22    conclusions on  the array of effects recognized for Pb in the last review (ISA, section 2.7).  The
23    current ISA, unlike the 2006 CD, makes causal determinations for the effects  of Pb on
24    organisms.   The ISA determines that causal or likely causal relationships1 exist in both
25    freshwater and terrestrial ecosystems for Pb2 with effects on reproduction and development in
26    vertebrates and invertebrates; growth in plants and invertebrates; and survival in vertebrates and
27    invertebrates (ISA, table 2-3).  Based on these determinations and a weight of evidence
28    approach, the ISA concludes that higher level community and ecosystem effects in freshwater
              Since the last Pb NAAQS review, the ISAs which have replaced CDs in documenting each review of the
      scientific evidence (or air quality criteria) employ a systematic framework for weighing the evidence and describing
      associated conclusions with regard to causality, using established descriptors ("causal" relationship with relevant
      exposure, "likely" to be causal, evidence is "suggestive" of causality, "inadequate" evidence to infer causality, "not
      likely") (ISA, Preamble).
              In determining that a causal relationship exists for Pb with specific ecological effects, the EPA has concluded
      that "[ejvidence is sufficient to conclude that there is a causal relationship with relevant pollutant exposures (i.e., doses or
      exposures generally within one to two orders of magnitude of current levels)" (ISA, p. Ixiv)

      January 2013                            5-1              Draft - Do Not Quote or Cite

-------
 1    and terrestrial systems are likely to result from the above described population-level endpoints.
 2    The ISA also presents evidence for saltwater ecosystems but concludes that current evidence is
 3    inadequate to make causality determinations for most organisms and communities but finds the
 4    evidence to be suggestive linking Pb and effects on reproduction and development in marine
 5    invertebrates (ISA, section 1.5, Table 2-3, section 7.3.12 and section 7.4.21).
 6          Based on the extensive assessment of the full body of evidence available in this review,
 7    the major conclusions drawn by the ISA regarding ecological effects of Pb include the following
 8    (ISA, section 1.5).
 9          With regard to the ecological effects ofPb, uptake ofPb into fauna and
10          subsequent effects on reproduction, growth and survival are established and are
11          further supported by more recent evidence. These may lead to effects at the
12          population, community, and ecosystem level of biological organization. In both
13          terrestrial and aquatic organisms, gradients in response are observed with
14          increasing concentration ofPb and some studies report effects within the range of
15          Pb detected in environmental media. Specifically, observations from controlled
16          studies on reproduction, growth, and survival in sensitive freshwater
17          invertebrates are well-characterized at concentrations at or near Pb
18          concentrations occasionally encountered in U.S. surface waters.... However, in
19          natural environments, modifying factors affect Pb bioavailability and toxicity and
20          there are considerable uncertainties associated with generalizing effects observed
21          in controlled studies to effects at higher levels of biological organization.
22          Furthermore,  available studies on community and ecosystem-level effects are
23          usually from contaminated areas where Pb concentrations are much higher than
24          typically encountered in the environment. The contribution of atmospheric Pb to
25          specific sites is not clear and the connection between air concentration ofPb and
26          ecosystem exposure continues to be poorly characterized. Furthermore, the level
27          at which Pb elicits a specific effect is difficult to establish in terrestrial and
28          aquatic systems, due to the influence of other environmental variables (e.g., pH,
29          organic matter) on both Pb bioavailability and toxicity, and also to substantial
30          species differences in Pb sensitivity.
31
32          As in prior reviews of the Pb NAAQS, this review is focused on those effects most
33    pertinent to ambient air Pb exposures.  Given the reductions in ambient air Pb levels over the
34    past decades, these effects are  generally those associated with the lowest levels of Pb exposure
35    that have been evaluated. Additionally, we recognize the limitations on our ability to draw
36    conclusions about environmental exposures from ecological studies of organism-level effects as
37    most studies were conducted in laboratory settings which may not accurately represent field
38    conditions or the multiple variables that govern exposure.
39          Our consideration of welfare effects evidence in this review is framed by key policy -
40    relevant questions drawn from those included in the IRP. In the following sections, we discuss
41    the pathways by which Pb exposure occurs in ecosystems, the mechanisms that distribute Pb in

      January 2013                           5-2            Draft - Do Not Quote or Cite

-------
 1    the environment and the bioavailability of Pb in different ecosystems.  Understanding the
 2    movement of Pb in the environment is important to understanding exposure and bioavailability
 3    and, thus, informs the subsequent discussion of the effects of Pb on terrestrial and aquatic
 4    ecosystems.  Finally, we discuss the association of ambient air Pb with effects and the important
 5    role critical loads play in assessing the overall ability of ecosystems to recover from past Pb
 6    exposures and the degree to which  newly deposited Pb may affect ecosystem function and
 7    recovery.
 8         •   To what extent has the newly available evidence altered our understanding of the
 9             movement and accumulation of air-deposited Pb through ecosystems over time?
10          There is little new information on fate and transport in ecosystems specifically related to
11    air-derived Pb. Nonetheless, some new studies are available that provide additional information,
12    briefly summarized below,  on Pb cycling, flux and retention. This new information does not
13    fundamentally change our understanding from the  last review of Pb movement or accumulation
14    through ecosystems over time but rather improves  our understanding of some of the underlying
15    processes and mechanisms  in soil, water and sediment. The newly available information also
16    adds to the evidence base with regard to the timing of ecosystem recovery from historic
17    atmospheric deposition of Pb.
18          Overall, recent studies in terrestrial ecosystems provide deposition data consistent with
19    deposition fluxes reported in the 2006 CD, and demonstrate consistently that Pb deposition to
20    soils has decreased since the phase-out of leaded on-road gasoline (ISA,  section 3.3.3.1).
21    Although follow-up studies in several locations at high elevation sites indicate little change in
22    soil Pb concentrations since the phase-out of leaded on-road gasoline in surface soils, consistent
23    with the high retention reportedly associated with reduced microbial activity at lower
24    temperatures associated with high elevation sites, amounts of Pb in the surface soils at some
25    lower altitude sites were reduced over the same time period in the same study (ISA, section
26    3.3.3.1). New studies in the ISA also enhance our understanding of Pb sequestration in forest
27    soils by providing additional information on the role of leaf litter as a Pb reservoir and the effect
28    of decomposition on Pb distribution and sequestration (ISA, section 3.3.3.2).
29          Recent research on Pb transport in aquatic systems has provided a large body of
30    observations confirming that Pb transport is dominated by colloids rich in iron and organic
31    material (ISA, section 3.3.2.1). In sediments, recent research on Pb flux provides greater detail
32    on resuspension processes than was available in the 2006 CD, including research on resuspended
33    Pb largely associated with organic material or iron and manganese particles, and research on the
34    important role played by anoxic or depleted oxygen environments in Pb cycling in aquatic
35    systems. This newer research is consistent with prior evidence in indicating that appreciable
36    resuspension and release from sediments largely occurs during discrete events related to storms.
      January 2013                           5-3             Draft - Do Not Quote or Cite

-------
 1    It has also confirmed that resuspension is an important process that strongly influences the
 2    lifetime of Pb in bodies of water. Finally, there have been advances in understanding and
 3    modeling of Pb partitioning between organic material and sediment in complex aquatic
 4    environments (ISA, section 3.7.2).
 5           In summary, the newly available evidence builds on our understanding of some specific
 6    aspects of processes involved in the movement and accumulation of air-deposited Pb through
 7    ecosystems over time.  This new information, however, does not substantially alter our overall
 8    understanding of fate and transport of Pb in ecosystems.
 9         •   Does the newly available evidence further inform our understanding  of the
10             bioavailability of Pb in different types of ecosystems  and organisms?
11           As discussed in the ISA, bioavailability of Pb is an important component of
12    understanding the effects Pb is likely to have on organisms and ecosystems (ISA, section 7.3.3).
13    It is the amount of Pb that can interact within the organism that leads to toxicity and there are
14    many factors which govern this interaction. Bioavailability is defined in the Framework for
15    Metals Risk Assessment (USEPA, 2007) as "the extent to which bioaccessible metals absorb
16    onto, or into, and across biological membranes of organisms, expressed as a fraction of the total
17    amount of metal the organism is proximately exposed to (at the sorption surface) during a given
18    time and under defined conditions." The bioavailability of metals varies widely depending on the
19    physical,  chemical, and biological conditions under which an organism is exposed (USEPA,
20    2007). Characteristics of the toxicant itself that affect bioavailability  are: (1) chemical form or
21    species, (2) particle size, (3) lability, and (4) source. The bioavailability of a metal is also
22    dependent upon the fraction of metal that is bioaccessible. The bioaccessible fraction of a  metal
23    is the portion  (fraction or percentage) of environmentally available metal that actually interacts at
24    the organism's contact surface and is potentially available for absorption or adsorption by the
25    organism (USEPA, 2007; ISA, section  7.3.3).
26           New studies provide additional  insight into factors that influence the bioavailability of Pb
27    to specific organisms (ISA, section 7.3.3). In general, this evidence,  briefly summarized below,
28    is supportive of previous conclusions and does not identify significant new variables from those
29    identified previously. Section 7.3.3 of the ISA provides a detailed discussion of bioavailability in
30    terrestrial systems. With regard to aquatic  systems, a detailed discussion of bioavailability in
31    freshwater systems is provided in sections 7.4.3 and 7.4.4 of the ISA and section 7.4.14 of the
32    ISA discusses bioavailability in saltwater systems.
33           In terrestrial systems, the bioavailability of Pb has a far greater impact on the adverse
34    effects of Pb than  does the  total Pb burden (ISA, section 7.3.11).  In such ecosystems, Pb is
35    deposited either directly onto plant surfaces or onto soil where it can bind with organic matter or

      January 2013                           5-4             Draft - Do Not Quote or Cite

-------
 1    dissolve in pore water.3 The Pb dissolved in pore water is particularly bioavailable to organisms
 2    in the soil and thereby influences the impact of soil Pb on terrestrial ecosystems to a much
 3    greater extent than the total amount of Pb present (ISA, section 7.3.11). Several soil variables
 4    control the amount of Pb that is dissolved in pore water, and the ISA presents new evidence that
 5    has advanced scientific understanding of some of these variables (ISA, section 7.4.21). Studies
 6    have shown that the two most important determinants of both Pb solubility and toxicity in soils
 7    are pH and cation exchange capacity (ISA, section 2.7.1). Also,  evidence newly available in this
 8    review has confirmed the important influence of organic matter  on Pb sequestration, leading to
 9    relatively longer retention in soils with higher organic matter content, with the potential for later
10    release of deposited Pb (ISA, section 2.7.1). In general, soils with higher organic matter content
11    have the capacity for greater retention of Pb in the soil matrix and lesser availability of Pb for
12    release into pore water. Subsequent reduction in soil organic material, such as through
13    decomposition,  can contribute to subsequent increased availability. Aging of Pb in soils, both
14    under natural conditions and simulated through leaching, has also been shown to reduce
15    bioavailability to plants and soil organisms (ISA, sections 7.3.2.3 and 7.3.11).
16           In aquatic systems as in terrestrial systems, the amount of Pb bioavailable to organisms is
17    a better predictor of effect on organisms than the overall amount of Pb in the system.  Once
18    atmospherically-derived Pb  enters surface water bodies through  deposition or runoff, its fate and
19    bioavailability are influenced by many water quality characteristics, such as pH, suspended
20    solids loading and organic content (ISA, section 7.4.2). In sediments, bioavailability of Pb to
21    sediment-dwelling organisms may be influenced by the presence of other metals,  sulfides, iron
22    oxides  and manganese oxides and also by physical disturbance (ISA, section 2.7.2). For many
23    aquatic organisms Pb dissolved in the water column can be the primary exposure  route, while for
24    others sediment ingestion is significant (ISA, section 2.7.2). As recognized in the 2006 CD and
25    further supported in the ISA, there is a body of evidence showing that uptake and elimination of
26    Pb varies widely between species.
27           Although in freshwater systems the presence of humic acid in dissolved organic material
28    (DOM) is considered to reduce the bioavailable fraction of metals in the water column, there is
29    evidence presented in the ISA that dissolved organic  carbon and DOM do not have the same
30    effect on free Pb ion concentration in saltwater systems (ISA, section 7.4.24).  Recent studies
31    have broadened previous understandings of the role of DOM in Pb bioavailability and have
32    shown  that different components of DOM have different effects on Pb bioavailability in
33    saltwater systems. For example, the ISA discusses two new studies that suggest that in saltwater
34    aquatic systems, surface water DOM increases (rather than decreases) uptake of Pb by fish gill
             3 The term "pore water" refers to the water occupying the spaces among the grains of sediment or soil.
      January 2013                            5-5             Draft - Do Not Quote or Cite

-------
 1    structures, potentially through the alteration of membrane permeability.  This recent evidence
 2    supports the conclusion from the last review that factors that modify bioavailability of Pb in
 3    saltwater environments are not identical to those in freshwater systems (ISA, section 7.4.14).
 4          There is new evidence about bioavailability in some ecosystems, briefly summarized
 5    above, which further informs our understanding of bioavailability of Pb in different ecosystems
 6    and organisms, building on our fundamental understanding in the last review. Bioavailability
 7    remains an important consideration in Pb toxicity and a significant source of uncertainty relating
 8    ambient Pb and adverse effects.
 9         •   Does the current evidence alter our conclusions from the previous review
10             regarding the ecological effects associated with exposure to Pb? Does the newly
11             available evidence indicate new exposure levels at which ecological systems or
12             receptors are expected to  experience effects?
13          There is a substantial amount of new evidence in this review regarding the ecological
14    effects of Pb on specific terrestrial and aquatic organisms. On the whole, this evidence supports
15    previous conclusions that Pb has effects on growth, reproduction and survival, and that under
16    some conditions these effects can be  adverse to organisms and ecosystems.  The ISA provides
17    evidence  of effects in additional species, and in a few cases, at lower exposures than reported in
18    the previous review, but does not substantially alter the effects endpoints from the previous
19    review. Looking beyond organism-level evidence, the evidence of adversity in  natural systems
20    remains sparse due to the difficulty in determining the effects of confounding factors such as
21    multiple metals or factors influencing bioavailability in field studies. The following is a brief
22    comparison of the newly available evidence to evidence considered in the 2006 review.
23          Terrestrial Ecosystems
24          The evidence available in the last review indicated a range of biological effects of Pb on
25    terrestrial organisms that varied with type of organism and life stage,  duration of exposure, form
26    of Pb, and soil characteristics. New research  since the 2006 CD has broadened our
27    understanding of the evidence of damage to photosynthetic ability in plants exposed to Pb, and
28    provided additional evidence of oxidative stress in response to Pb exposure (ISA, section
29    7.3.4.1). Reactive oxygen  species have been found to increase in plant tissue grown in Pb-
30    contaminated soil; with increasing Pb exposure, the plant tissue responded with increased
31    antioxidant activity (ISA, section 7.3.4.1). In addition, reduced growth has been observed in
32    some experiments, as well as evidence of genotoxicity, decreased germination, and pollen
33    sterility (ISA, section 7.3.4.1).
34          In terrestrial invertebrates, previous CDs have reported adverse effects of Pb on
35    neurological and  reproductive endpoints. Recently published studies have shown some
36    additional evidence of neurological changes in  nematodes at relatively low concentrations of

      January 2013                           5-6            Draft - Do Not Quote or Cite

-------
 1     soil Pb (2.5 uM) but given the study design the relevance to environmental exposure is unclear
 2     (ISA, section 7.3.4.2). Increased mortality was found in recent studies of earthworms at
 3     concentrations similar to those in studies reviewed previously, but additional evidence indicated
 4     that effects were strongly dependent on soil characteristics including pH, cation exchange
 5     capacity, and aging (ISA, section 7.3.4.2). There is also newly available evidence for adverse
 6     effects in snails and arthropods exposed through soil or diet (ISA, section 7.3.4.2). The effects
 7     vary with species and exposure conditions, and include diminished growth and fecundity,
 8     endocrine and reproductive anomalies, and body deformities.  Increasing concentration of Pb in
 9     the exposure medium generally resulted in increased effects within each study, but the
10     relationship between concentration and effects is highly variable between studies, even when
11     the same medium, e.g. soil, was used. Current evidence suggests that aging and pH are
12     important modifiers of Pb toxicity (ISA, section 7.3.11).
13           In terrestrial vertebrates, some new evidence  is available for effects of Pb on amphibians
14    and reptiles as well as birds.  Effects reported in reptiles and amphibians include decreased white
15    blood cell counts, decreased testis weight, and behavioral anomalies (ISA,  section 7.3.4.3).
16    However, depending on various factors, studies report large differences in  effects in different
17    species at the same concentration of Pb in soil, and effects were generally smaller when field-
18    collected soils were used. In some birds, recent studies have found maternal elevated blood Pb
19    level to be associated with decreased hatching success, smaller clutch size, high corticosteroid
20    level, and abnormal behavior. Studies on some species show little or no effect of elevated blood
21    Pb level. Effects of dietary exposure have been studied in several mammalian species, with
22    cognitive, endocrine, immunological, and growth effects observed (ISA, section 7.3.11) in some
23    studies.
24           Experimental evidence of organism-level effects presented in the ISA demonstrates that
25     increased exposure to Pb is generally associated with increases in observed effects in terrestrial
26     ecosystems (ISA, section 7.3.11). It also demonstrates that many factors, including species and
27     various soil physiochemical properties, interact strongly with  Pb concentration to modify those
28     effects (ISA, section 7.3.11). In terrestrial ecosystems, where  soil is generally the main
29     component of the exposure route, Pb aging is a particularly important factor,  and one that may
30     be difficult to reproduce experimentally. Given that in natural settings, these modifying factors
31     are highly variable, the ISA notes that "[w]ithout quantitative characterization of those
32     interactions, characterizations of exposure-response relationships would likely not be
33     transferable outside of experimental settings" (ISA,  section 2.7.1). The results of the few
34     exposure-response studies conducted since the 2006 CD have been inconsistent (ISA, section
35     2.7.1).


      January 2013                            5-7             Draft - Do Not Quote or Cite

-------
 1           As reported in both the ISA and the 2006 CD, most direct evidence of community- and
 2     ecosystem-level effects is from releases by near sources of Pb to the environment, where Pb
 3     concentrations are much higher than typically observed environmental concentrations and often
 4     derive from separate releases to multiple media. Impacts of Pb on terrestrial ecosystems near
 5     smelters, mines, and other industrial sources have been studied for several decades (ISA, section
 6     7.3.12.7). Emissions of Pb from smelting and other industrial activities are commonly
 7     accompanied by other trace metals (e.g., Zn, Cu, Cd) and SO2 that may cause toxic effects
 8     independently or in concert with Pb. Those impacts have been long shown to include decreases
 9     in species diversity and changes in floral and faunal community composition. Ecosystem-level
10     field studies are complicated by these confounding exposures and the inherent variability in
11     natural systems (ISA, section 7.3.12.7).
12           New evidence of effects of Pb at the community and ecosystem scales has been
13     reported. This evidence includes several studies of the ameliorative effects of mycorrhizal fungi
14     on plant growth with Pb exposure as well as recently published research on soil microbial
15     communities, which have been shown to be impacted by Pb in both composition and activity
16     (ISA, sections 2.7.1 and 7.3.6). Many recent studies have been conducted using mixtures of
17     metals, which have attempted to separate the effects of individual metals when possible. In
18     studies that included only Pb, or where effects of Pb could be separated,  soil  microbial activity
19     was generally diminished, and was shown in some of those cases to recover with time (ISA,
20     section 7.3.6). Regardless of the recovery of such microbial  communities, species and
21     genotypical composition were  consistently altered after Pb exposure, and findings indicate that
22     those alterations were long-lasting or permanent (ISA, section 7.3.11).
23           A recent review has examined differences in species sensitivity, using blood Pb level in
24     birds and mammals as an exposure index rather than external dose (ISA, section 7.3.9.1;
25     Buekers et al., 2009). In this analysis, variation across organisms was reduced with the blood
26     Pb index and variation of Pb absorption from the diet of the  organism largely accounted for the
27     variation seen in the blood Pb. The analysis also suggests that the association between blood Pb
28     concentration and toxicity is different between birds and mammals (ISA, section 7.3.9.1).
29           Overall, new studies in this review support previous conclusions about the effects of Pb
30     on terrestrial ecosystems, namely that increasing soil Pb concentrations in areas of Pb
31     contamination (e.g. mining sites and industrial sites) can cause decreases in microorganism
32     abundance, diversity, and function. Specifically, shifts in bacterial species and fungal diversity
33     have been recently observed near long-established sources of Pb contamination (ISA, section
34     7.3.12.7). Most evidence for Pb toxicity to terrestrial plants, invertebrates and vertebrates,
35     however, is from single-species assays in laboratory studies  which do not capture the


      January 2013                            5-8             Draft - Do Not Quote or Cite

-------
 1     complexity of bioavailability and other modifiers of effect in natural systems (ISA, section
 2     7.3.12.7).
 3          Freshwater Ecosystems
 4          In fish, while there are species differences in the rate of Pb accumulation and distribution
 5     of Pb within the organism, the gills and diet are the primary routes of uptake.  Recent studies
 6     have identified the anterior intestine as an additional site of uptake of Pb through dietary
 7     exposure (ISA, section 7.4.4.3). While there are few new studies on Pb uptake through
 8     freshwater exposure by amphibians and mammals, there are some new studies presented in the
 9     ISA which seem to provide new information on trophic transfer of Pb  (ISA, section 7.4.4.4).
10     Evidence summarized in the 2006 CD indicated that measured concentrations of Pb in the
11     tissues of aquatic organisms were generally higher in algae and benthic organisms than in
12     higher trophic-level consumers, indicating that Pb was bioconcentrated but not biomagnified
13     (ISA, section 2.7.2). Some recent studies indicate transfer of Pb in aquatic food webs; recent
14     studies that have traced Pb in freshwater aquatic food webs have found that Pb concentration
15     decreases with increasing trophic level (biodilution) (ISA, sections 2.7.2 and 7.4.4.4).
16          Evidence presented in the ISA further supports the findings of past CDs that Pb in
17     freshwater can be highly toxic to aquatic  organisms, with toxicity varying with species and life
18     stage, duration of exposure, form of Pb, and water quality characteristics. The 2006 CD
19     identified evidence of adverse growth effects on several species of freshwater algae from Pb
20     exposure (2006 CD, Section 7.2.4). The ISA describes several  new studies which expand the
21     list of algal species for which these adverse effects have been identified (ISA, section 7.4.5.1).
22     For vascular plants, the ISA describes additional evidence that oxidative damage, decreased
23     photosynthesis and reduced growth occur with Pb exposure (ISA, section 7.4.5.2).
24           Since the 2006 CD, there is some additional evidence for Pb effects on cellular processes
25     in aquatic invertebrates. Recent studies of reproductive and developmental effects of Pb provide
26     further support for findings in the 2006 CD and add additional species information on
27     reproductive endpoints for rotifers and freshwater snails as well as multigenerational effects of
28     Pb in mosquito larvae (ISA, section 7.4.5.2). In the 2006 CD, study concentrations cited at
29     which effects were observed in aquatic invertebrates reflected an expansive range from 5ug/L
30     (for acute toxicity to Pb nitrate in a test system at a hardness of 18 mg/L calcium carbonate) to
31     greater than 8000 |ig/L (for acute toxicity to Pb chloride at a hardness of 280-300 mg/L calcium
32     carbonate) (ISA, p.2-38; 2006 CD, Table AX7-2.4.1). Recent studies cited in the ISA provide
33     additional evidence of effects from chronic exposures. These findings are generally consistent
34     with those  of the previous review and reflect the variability of Pb toxicity under different
35     environmental conditions and in species with varying sensitivity.


      January 2013                           5-9             Draft - Do Not Quote or Cite

-------
 1           Evidence of effects in aquatic vertebrates presented in the ISA reiterates the findings of
 2     reproductive, behavioral, and growth effects stated in previous CDs. Some additional
 3     mechanisms of Pb toxicity in the gill and the renal system offish have been elucidated since the
 4     2006 CD as well as the identification of potential new molecular targets for Pb neurotoxicity
 5     (ISA, section 2.7.2). Recent studies reported sublethal effects on tadpole growth, deformity, and
 6     swimming ability at concentrations within two orders of magnitude of environmental levels of
 7     Pb (ISA, section 7.4.5.3) based on nominal concentrations.
 8           As in terrestrial organisms, evidence presented in the ISA and prior CDs demonstrates the
 9     toxicity of Pb in aquatic ecosystems and the role of many factors, including Pb speciation and
10     various water chemistry properties in modifying toxicity (ISA, section 2.7.2). Since the 2006
11     CD, additional evidence for community and ecosystem level effects of Pb is available, primarily
12     in microcosm studies or field studies with other metals present (ISA, section 7.4.11). Such
13     evidence described in previous CDs includes alteration of predator-prey dynamics, species
14     richness, species composition, and biodiversity. New studies available in this review provide
15     evidence in additional habitats for these community and ecological-scale effects, specifically in
16     aquatic plant communities and sediment-associated communities at both acute and chronic
17     exposures involving concentrations similar to previously reported (ISA, section 7.4.4.1). In
18     many cases it is difficult to characterize the nature and magnitude of effects and to quantify
19     relationships between ambient concentrations of Pb and ecosystem response due to existence of
20     multiple stressors, variability in field conditions, and to differences in Pb bioavailability at that
21     level of organization (ISA, section 2.7.3.7).  In these field studies, the degree to which air
22     concentrations contribute to such effects is largely unknown.
23           Saltwater Ecosystems
24           Recently available evidence in the ISA on toxicity of Pb to marine algae augments the
25     2006 CD findings of variation in sensitivity across marine species. Recent studies on Pb
26     exposure include reports of growth inhibition and oxidative stress in a few additional species of
27     marine algae (ISA, section 7.4.15).
28           Recent literature provides little new evidence of endpoints or effects in marine
29     invertebrates beyond those reported in the 2006 CD.  For example, some recent studies
30     strengthen the evidence of Pb effects on enzymes  and antioxidant activity in marine
31     invertebrates and have identified an additional behavioral endpoint (i.e., valve closing speed in
32     juveniles of a marine scallop was affected by 20-day exposure to 40-400 |ig/L Pb nitrate) from
33     those discussed  previously (ISA, section 7.4.15.2). Also, as noted in the 2006 CD and supported
34     by new studies reviewed in the ISA, Pb exposure negatively affects the growth of marine
35     invertebrates (ISA, section 7.4.15.2). Recent studies also identify several species exhibiting
36     particularly low sensitivity to high acute exposures (ISA, section 7.4.15.2).

      January 2013                           5-10             Draft - Do Not Quote or Cite

-------
 1          Little new evidence is available of Pb effects on marine fish and mammals for
 2    reproductive, growth and survival endpoints that are particularly relevant to population-level of
 3    biological organization and higher (ISA, section 7.4.15).  Evidence for effects at higher levels of
 4    biological organization in saltwater habitats is primarily supported by observations in a small
 5    number of microcosm and field studies where shifts in community structure are the most
 6    commonly observed effects of Pb (ISA, section 7.4.20).  These types of studies were presented
 7    in the 2006 CD and while there are new studies presented in the ISA, they primarily expand the
 8    types of communities for which data exist but reach similar conclusions about community
 9    structure as the 2006 CD.  However, there is no evidence of adverse effects in saltwater
10    environments from current ambient  air concentrations and linkages between any level of
11    ambient exposure and effects are unknown in all but a few organisms.
12           New studies on organism-level effects from Pb in saltwater ecosystems provide little
13    evidence to inform our understanding of linkages between atmospheric concentrations, ambient
14    exposures in saltwater systems and such effects or our conclusions regarding the likelihood of
15    adverse effects under conditions associated with the current NAAQS for Pb.  Nor does the
16    currently available evidence indicate significantly different exposure levels at which ecological
17    systems or receptors are expected to experience effects.  The new evidence, as outlined above,
18    serves to support previous conclusions regarding  effects evidence and provides information on
19    additional species, but does not significantly improve our understanding of ecological effects
20    levels of Pb.
21         •  To what extent is there new information that informs our understanding of
22            critical loads of Pb, including critical loads in sensitive ecosystems?
23          Critical loads are an especially powerful tool that can link Pb input into an ecosystem,
24    including atmospheric deposition, with ecological  impairment. Critical load calculations are
25    dependent on data relating Pb released into an ecosystem from various sources with an
26    understanding of the level of Pb in an ecosystem that is likely to result in measurable effects.
27    Importantly, with these data, critical loads can account for heterogeneity in ecosystem sensitivity
28    and exposure which often results in critical load values that vary by ecosystem (e.g., aquatic-
29    water; aquatic-sediment; terrestrial), and differ by  endpoint of concern (ISA, section 7.1.3).
30    During the last review, the 2006 CD assessed the available information on critical loads for Pb
31    which was drawn primarily from Europe. Analyses were not available for U.S. locations, and
32    the European critical load values for Pb that had been developed were highly specific to the
33    bedrock geology, soil types, vegetation, and historical deposition trends in each European
34    country (2006 CD, p. E-24). As a result, the 2006 CD concluded that "[a]t this time, the methods
35    and models commonly used for the calculation of critical loads have not been validated  for Pb"
36    and that "[m]any of the methods neglect the speciation of Pb when estimating critical limits, the
      January 2013                          5-11            Draft - Do Not Quote or Cite

-------
 1    uptake of Pb into plants, and out flux of Pb in drainage water, limiting the utility of current
 2    models" (2006 CD, p. 7-46). Accordingly, the quantitative assessment for the last Pb NAAQS
 3    review did not involve critical load analyses.
 4          There are no critical loads for Pb reported in the ISA or 2006 CD for any ecosystems in
 5    the U.S.; however, as reported in the 2006 CD and further discussed in section 7.3.7 of the ISA,
 6    work has been conducted in terrestrial systems in Europe and aquatic ecosystems in  Canada
 7    (ISA, section 7.3.7 and section 7.4.8). Given that local conditions (including historic loading, soil
 8    transport and transformation processes) are key elements to critical load calculation, the utility of
 9    critical loads that are developed from other countries for application to U.S. ecosystems is
10    unclear.  Given the heterogeneity of ecosystems affected by Pb, and the differences in
11    expectations for ecosystem services attached to different land uses, it is expected that there
12    would be a range of critical load values for Pb for soils and freshwaters within the U. S.
13          In the short term, metal emissions generally have greater effects on biota in freshwater
14    systems  than in terrestrial systems because metals are more readily immobilized in soils than in
15    sediment. However,  over the longer term, terrestrial systems may be more affected by increasing
16    concentrations, particularly by those metals with a long soil residence time, such as Pb (ISA,
17    section 7.4.8). There is no new evidence regarding critical loads for Pb in saltwater ecosystems.
18         •   To what extent is there information that improves our understanding  of the
19             portion of environmental Pb derived from ambient air and the associated effects
20             on sensitive ecosystems?
21           As stated in  the last review, the role of ambient air Pb in contributing to ecosystem Pb
22    has been declining over the past several decades. However, it remains  difficult to apportion
23    exposure between air and other sources to better inform our understanding of ecosystem effects
24    that might be associated with air emissions. Considerable uncertainties also remain  in drawing
25    conclusions from evidence of effects observed under laboratory conditions with regard to effects
26    expected at the ecosystem level in the environment. In many cases it is difficult to characterize
27    the nature and magnitude of effects and to quantify relationships between ambient concentrations
28    of Pb and ecosystem response due to the existence of multiple stressors, variability in field
29    conditions, and to differences in Pb bioavailability at that level  of organization (ISA, section
30    7.5). There is no new evidence that substantially improves our understanding of the relationship
31    between ambient air Pb and measurable ecological effects.
32
      January 2013                           5-12            Draft - Do Not Quote or Cite

-------
 1    5.2  EXPOSURE AND RISK INFORMATION
 2           This risk information available for this review and described here is based primarily on
 3    the pilot ecological risk assessment developed in the last review of the Pb NAAQS (henceforth
 4    referred to as the 2006 REA [ICF, 2006]), which is described within the context of the evidence
 5    presented in the ISA that is newly available for this review.  As described in the IRP, careful
 6    consideration of newly available information in this review led us to conclude that developing a
 7    new REA for this review was not warranted.  In light of critical limitations and uncertainties that
 8    are still unresolved in the current evidence, staff concluded that currently available information
 9    does not provide the basis for developing a new quantitative risk and exposure assessment with
10    substantially improved utility for informing the Agency's  consideration of welfare effects and
11    evaluation of the adequacy of the current secondary standard or alternatives (IRP, section 3.3).
12    Based on their consideration of the IRP, the CAS AC Pb Review Panel generally concurred with
13    the conclusion that a new REA was not warranted in this review (Frey, 2011).4 Accordingly, the
14    information described here is drawn primarily from the 2006 REA.
15           The focus for the risk assessment and associated estimates presented here is on Pb
16    derived from sources emitting Pb to ambient air. While there is some new evidence that
17    improves our understanding of some of the environmental variability affecting the disposition
18    and toxicity of Pb in the environment, the information and methods to support a quantitative
19    assessment of the role of atmospheric Pb in the U.S are limited. Specific constraints include the
20    limited availability of location-specific data describing a range of U.S. ecosystems and their
21    pertinent environmental  characteristics as well as the ability to apportion effects between air and
22    non air sources, and a more complete understanding  of bioavailability and its modifiers. These
23    data gaps and areas of uncertainty in the  current evidence restrict our ability to assess
24    quantitatively the relationship between concentrations of Pb in ambient air and terrestrial and/or
25    aquatic systems, and their effect on welfare.
26           As discussed in section 1.3 above, the multimedia and persistent nature of Pb, the role of
27    multiple exposure pathways (illustrated in Figure 1-1 above), and the contributions of nonair
28    sources of Pb to exposure media all present challenges and contribute significant additional
29    complexity to the ecological risk assessment that goes far beyond the situation for similar
30    assessments typically performed for other NAAQS pollutants (e.g., that focus only on a single
31    media pathway or for which air is the only significant source). Limitations in the  available data
32    and models affected our characterization of the various complexities associated with exposure to
             4 In our evaluation presented in the REA Planning Document and consultation with CASAC, we indicated our
      conclusion that the information newly available in this review did not provide the means by which to develop an updated
      or enhanced risk model that would substantially improve the utility of risk estimates in informing the current Pb NAAQS
      review.

      January 2013                           5-13             Draft - Do Not Quote or Cite

-------
 1    ambient air Pb. As a result, the 2006 assessment was conducted as a pilot study with a number
 2    of simplifying assumptions with regard to the representativeness of the case studies for
 3    ecological exposures and our ability to isolate case studies where air-derived Pb was the only or
 4    most significant source. Therefore, this section presents a brief summary of the screening-level
 5    ecological risk assessment conducted in 2006 for the Pb NAAQS review completed in 2008 and
 6    addresses several questions relating to the current evidence and understanding that may inform
 7    our view of the results of that assessment. The discussion here also takes into consideration
 8    CAS AC recommendations in the last review with regard to interpretation of the screening-level
 9    assessment (Henderson, 2007a, b), as well as comments received from the CASAC Pb Panel in
10    the current review, as part of the consultation on the REA Planning Document (Frey, 2011).

11         5.2.1  Screening Assessment from Last Review
12          The screening-level risk assessment performed for the last review was focused on
13    estimating the potential for ecological risks associated with ecosystem exposures to Pb emitted
14    into ambient air (2006 REA, section 7). A national-scale screen was used to evaluate surface
15    water and sediment monitoring locations across the United States for the potential for ecological
16    impacts that might be associated with atmospheric deposition of Pb (described in detail in 2006
17    REA, section 7.1.2). In addition to the national-scale screen (2006 REA, section 3.6), the
18    assessment involved a case study approach, with case studies for areas surrounding a primary Pb
19    smelter (2006 REA, section 3.1) and a secondary Pb  smelter (2006 REA, section 3.2), as well as
20    a location near a non-urban roadway (2006 REA, section 3.4).  An additional case study, focused
21    on consideration of gasoline-derived Pb effects on an ecologically vulnerable ecosystem
22    (Hubbard Brook Experimental Forest), was identified (2006 REA, section 3.5).  The Hubbard
23    Brook Experimental Forest (FffiEF), in the White Mountain National Forest, near North
24    Woodstock, New Hampshire, was selected as a fourth case study because of its location and its
25    long record of available data on concentration trends of Pb in three media (air or deposition from
26    air), soil, and surface water). While no quantitative analyses were performed, summary review of
27    the literature search was included in the assessment report (2006 REA, Appendix E). For the
28    other three case studies, exposure concentrations in soil, surface water, and/or sediment
29    concentrations were estimated  from  available monitoring data or modeling analysis, and then
30    compared to  ecological screening benchmarks (2006 REA, section 7.1).
31          All three case  studies and the national-scale assessment generally considered then-current
32    or recent environmental conditions.  In  all cases but the primary Pb smelter case study, current
33    air quality conditions  were below the then-current NAAQS.  Air Pb concentrations in the
34    primary Pb smelter case study  exceeded the then-current NAAQS. A complete discussion of air
35    quality in each of the  case studies can be found in section 4 of the 2006 REA.

      January 2013                          5-14            Draft - Do Not Quote  or Cite

-------
 1          An overview of the approach developed to implement the selected elements of the
 2   conceptual model for the ecological risk assessment is provided in Figure 5-1. This figure shows
 3   the key types of information and models involved in each part of the assessment and how they
 4   are related to each other and to the other parts of the analysis. Appendix 5 A gives the locations
 5   and spatial resolution for each of the case studies and the national scale screen while summarizes
 6   the source of the screening values used or each location and media type. As indicated in Figure
 7   5-1 and Appendix 5A, the specific approach for each case study differed based on the nature of
 8   the case study (e.g., type of source, locations of populations) and the site-specific measurements
 9   available.
10
     January 2013                          5-15            Draft - Do Not Quote or Cite

-------
1    Figure 5-1.  Analytical approach for screening-level assessment in the last review (2006
2                   REA, Exhibit 2-6).
                   Exposure Assessment: Estimating Media Concentrations
                Primary Pb
                 Smelter
                Case Study
Near Roadway
  Non-urban
  Case Study
  SecondaryPb
Smelter Case Study
               Site-specific soil,
              surface water, and
                 sediment
               monitoring data
                   Stack and fugitive
                     emissions
                     estimates
                                                      Air dispersion
                                                        model
                                                    Pb deposition rates
                                                     across study area
                                                         Soil
                                                        mixing
                                                         model
                            Outdoor soil
                            concentrations
                                              Water column
                                              concentrations
  National-Scale
 Surface Water &
S e diment S ere ening
   Assessment
                                                                  Sediment
                                                                concentrations
                  Risk
            Characterization
                                          Ecological
                                            Effects
                                          Assessment
               Media-specific Hazard Quotients
4
5
6
7
     January 2013
                  5-16
                    Draft - Do Not Quote or Cite

-------
 1           To estimate the potential for ecological risk, ecological soil screening values (Eco-SSLs)
 2    developed by the EPA's Superfund program (USEPA, 2003, 2005), the EPA's recommended
 3    ambient water quality criteria (2006 REA, section 7.1) and sediment screening values developed
 4    by MacDonald and others (2000, 2003) were used, as described in detail in section 7.1.3 of the
 5    2006 REA. A hazard quotient (HQ) was calculated for adverse effects on the survival, growth,
 6    and reproduction of exposed ecological receptors to determine the potential for risk to that
 7    receptor. Ecological receptors used in the pilot are discussed in detail in section 7.1 of the 2006
 8    REA. The HQ was calculated as the ratio of the media concentration to the ecotoxicity screening
 9    value.  In each case study, HQ values were calculated for each location where either modeled or
10    measured media concentrations were available. Separate soil HQ values were  calculated for
11    each ecological receptor group for which an ecotoxicity screening value has been developed (i.e.,
12    birds, mammals, soil invertebrates, and plants), as described in detail in section 7.1.2 of the 2006
13    REA. HQ values less than 1.0 were concluded to suggest that Pb concentrations in a specific
14    medium were unlikely to pose significant risks to ecological receptors, while HQ values greater
15    than 1.0 indicated a potential for adverse effects.

16         5.2.2  Screening Assessment Results and Interpretation
17           The results for the ecological screening assessment for the three case studies and the
18    national-scale screen for surface water and for sediment indicated a potential for adverse effect
19    from ambient Pb to multiple ecological receptor groups in terrestrial and aquatic locations.
20    Below  are descriptions of the location-specific case studies and the national screening
21    assessment, key findings of the risk assessment for each, and an interpretation of the results with
22    regard to past air conditions as well as the current standard (USEPA, 2007b).
23         •  What do the key  findings of the 2006 screening-level assessment indicate
24            regarding the likelihood that adverse welfare effects would result from levels of
25            air-related Pb that would meet the current standard?
26           In addressing this question, the findings of the 2006 REA are summarized below.
27         Primary Pb Smelter Case Study
28         •  The primary Pb smelter case study location is at one of the largest and longest-
29            operating primary Pb smelters in the world (since 1892), the only one currently
30            operating in the U. S. (ICF, 2006).5
31         •  Concentrations of total Pb in several of the soil and sediment locations within the case
32            study were measured in 2000 and exceeded screening values, indicating a potential for
33            adverse effects to terrestrial and sediment dwelling organisms.
            5 As noted in section 2.1.2 above, this smelter is planning to cease existing smelting operations at this
      facility by April, 2014.
      January 2013                           5-17            Draft - Do Not Quote or Cite

-------
 1         •   While the contribution to these Pb concentrations from air as compared to nonair
 2             sources is not quantified, air emissions from this facility are substantial (ICF, 2006). In
 3             addition, this facility that has been emitting Pb for many decades, including some
 4             seven decades prior to establishment of any Pb NAAQS, such that it is likely air
 5             concentrations associated with the facility were substantial relative to the 1978
 6             NAAQS, which it exceeded at the time of the last review. Currently, concentrations
 7             monitored near this facility exceed the level of the current standard. Accordingly, this
 8             case study is not informative  for considering the likelihood of adverse welfare effects
 9             related to Pb from air sources under air quality conditions associated with meeting the
10             current Pb standard.
11          Secondary Pb Smelter Case Study

12         •   The secondary Pb smelter location falls within the Alabama Coastal Plain in Pike
13             County, Alabama, in an area  of disturbed forests. The industrial facility in this case
14             study is much younger than the primary Pb smelter, becoming active less than ten years
15             prior to the establishment of the  1978 Pb  standard.

16         •   Estimates of total Pb concentration in soils (based on fate and transport modeling using
17             1997-2000 emissions data and data for similar locations measured in a 1995  study)
18             exceeded screening values for plants, birds and mammals, indicating the potential for
19             adverse effects to these groups.

20         •   While the contributions from air-related Pb to the total Pb concentrations modeled in
21             soils at this location is unclear, the facility continues to emit Pb and the county where
22             this facility is located does not meet the current Pb  standard (Appendix 5A).  Given the
23             exceedances of the current standard,  which likely extends back over 4 to 5 decades, this
24             case study also is not informative for considering the likelihood of adverse welfare
25             effects related to Pb from air  sources under air quality conditions associated with
26             meeting the current Pb standard.
27          Near Roadway Non-urban Case Study

28         •   This case study comprises two nonurban  sites adjacent to established highways for
29             which soil Pb data were available: (1) in Corpus Christi, Texas  (ICF, 2006) and (2) in
30             Atlee, Virginia (ICF, 2006).  Measured soil concentration data were used to develop
31             estimates of Pb in soils for each location.

32         •   Estimates of total Pb concentrations taken in 1994 and 1998 in  soils in this case study
33             exceeded screening values for plants, birds and mammals, indicating the potential for
34             adverse effect to these groups.
35             These case study locations are highly impacted by past deposition  of gasoline Pb. It is
36             unknown whether current conditions at these sites exceed the current Pb standard but
37             given evidence from the past  of Pb concentrations near highways that ranged above the
38             previous (1978)  Pb standard (1986 CD, section 7.2.1),  conditions at these locations
39             during the  time of leaded gasoline very likely exceeded the current standard. Similarly,
40             those conditions likely resulted in Pb deposition associated with leaded gasoline that
41             exceeds that being deposited  under current air quality conditions that likely meet the
42             current Pb standard. Given this legacy, consideration of the potential for


      January 2013                           5-18            Draft - Do Not Quote or Cite

-------
 1             environmental risks from levels of air-related Pb associated with meeting the current
 2             Pb standard in these locations is highly uncertain.
 3         Vulnerable Ecosystem Case Study

 4         •   This case study was focused on consideration of information available for the Hubbard
 5             Brook Experimental Forest (HBEF) in the White Mountain National Forest near North
 6             Woodstock, New Hampshire, which included a long record (from 1976 through 2000)
 7             of available data on concentration trends of Pb in three media (air or deposition from
 8             air, soil, and surface water).

 9         •   While no quantitative analyses were performed, summary review  of literature search
10             indicated: (1) atmospheric Pb inputs do not directly affect stream Pb levels at HBEF
11             because deposited Pb is almost entirely retained in the soil profile; (2) soil horizon
12             analysis results showed Pb to have become more concentrated at lower soil depths over
13             time, with the soil serving as a Pb sink, appreciably reducing Pb in pore water as it
14             moves through the soil layers to streams (dissolved Pb concentrations were reduced
15             from 5 ppb to about 5 ppt from surface soil to streams).  Further, the available studies
16             concluded insignificant contribution of dissolved Pb from soils to streams (ICF, 2006,
17             Appendix E). It is unlikely that conditions have changed from the previous
18             conclusions made based on soil data through 2000 and therefore, current ambient air
19             concentrations likely do not directly impact stream Pb levels under air quality
20             conditions associated with meeting the current standard.
21         National-scale Surface Water and Sediment Screen

22         •   The national-scale screen was performed using a national  database of surface water and
23             sediment monitoring data (see Appendix 5A; ICF, 2006).
24                -  The  screen identified 15 non-mining sites at which at least one surface water
25                   dissolved Pb concentration taken between 1994 and 2004 exceeded the  chronic
26                   screening value, indicating a potential for adverse effect if concentrations were
27                   persistent over long time periods.  The acute screening value was not exceeded
28                   at any of these locations.

29                -  Analysis of the sediment concentrations for the 15 sites taken from 1991 to
30                   2000 analyzed in the surface water screen identified a subset for which
31                   concentrations exceeded the screening value, indicating a potential for adverse
32                   effects to sediment dwelling organisms.

33         •   The extent to which past air emissions of Pb have contributed to surface water or
34             sediment Pb concentrations at the locations identified in the screen is unclear. For
35             some of the surface water locations, nonair sources likely  contributed significantly to
36             the surface water Pb concentrations. For other locations, a lack of nearby nonair
37             sources indicated a potential role for air sources to contribute to observed surface water
38             Pb concentrations. Additionally, these concentrations may have been influenced by  Pb
39             in resuspended sediments and/or reflect contribution of Pb from erosion of soils with
40             Pb derived from historic as well as current air emissions.
41
      January 2013                           5-19             Draft - Do Not Quote or Cite

-------
 1           There are multiple sources of uncertainty associated with different aspects of this
 2    assessment (discussed in detail in 2006 REA, section 7.4). For example, there are significant
 3    limitations and uncertainties associated with conclusions that can be drawn from the primary
 4    smelter case study regarding the impact of atmospheric deposition under conditions associated
 5    with meeting the Pb NAAQS (past or present).  Additionally, while case study locations were
 6    chosen with the objective of including locations for which recent Pb data  were available and for
 7    which Pb exposures might be influenced by air-related Pb and not be dominated by nonair
 8    sources, there is significant uncertainty regarding the extent to which nonair sources and
 9    conditions associated with historic emissions (e.g., prior to establishment of past or present Pb
10    NAAQS) have likely contributed to the Pb exposure estimates in some locations. The screening
11    values available and used in the assessment were also sources of uncertainty with regard to
12    conclusions as to potential for ecological risk as the selected screening values (e.g., ambient
13    water quality criteria, Eco-SSLs, sediment criteria) lacked adjustment for some critical measures
14    of bioavailability. There is also uncertainty regarding the extent to which the screening values
15    could identify potential hazard for some threatened or endangered species or unusually sensitive
16    aquatic ecosystems (ICF 2006). Thus, while the assessment results are generally consistent with
17    evidence-based observations of the potential influence of Pb on ecological systems, they are
18    limited with regard to quantitative conclusions and potential hazard or risk associated with the Pb
19    NAAQS, most particularly with regard to the current standard.
20         •   What are the important uncertainties associated with interpreting the prior
21             assessment in light of newly available evidence in this review?
22           In interpreting the results from the 2006 REA, we consider newly  available evidence that
23    may inform our interpretation of risk under the current standard. Factors  necessary to alter our
24    interpretation of risk would include new evidence of harm at lower concentrations of Pb, new
25    linkages that enable us to draw more explicit conclusions as to the air contribution of
26    environmental exposures, and/or new methods of interpreting confounding factors that were
27    largely uncontrolled in the previous risk assessment.
28           With regard to new evidence of harm at lower concentrations, it is necessary to consider
29    that the evidence of adversity due specifically to Pb in natural systems is limited, in no small part
30    because of the difficulty in determining the effects of confounding factors such as multiple
31    metals and modifying factors influencing bioavailability in field studies. Modeling of Pb related
32    exposure and risk to ecological receptors is subject to a wide array of sources of both variability
33    and uncertainty. Variability is associated with geographic location, habitat types, physical and
34    chemical characteristics of soils and water that influence Pb  bioavailability, terrestrial and
35    aquatic community composition, Pb uptake rates by invertebrates, fish, and plants by species and
36    season. For wildlife, variability also is associated with food ingestion rates by species and
      January 2013                          5-20             Draft - Do Not Quote or Cite

-------
 1    season, prey selection, and locations of home ranges for foraging relative to the Pb
 2    contamination levels (USEPA, 2006b).
 3           There are significant difficulties in quantifying the role of air emissions under the current
 4    standard, which is significantly lower than the previous standard. As recognized in section 1.3.2
 5    above, lead deposited before the standard was enacted remains in soils and sediments,
 6    complicating interpretations regarding the impact of the current standard; historic Pb emitted
 7    from leaded gasoline usage continues to move slowly through systems along with more recently
 8    deposited Pb and Pb derived from non air sources.  The results from the location-specific case
 9    studies and the surface and sediment screen performed in the last review are difficult to interpret
10    in light of the current  standard and are not largely useful  in informing our judgments of the
11    potential for adverse effects at levels of deposition meeting the current standard. Under such
12    constraints it is difficult to assess the merit of the risk findings from the previous review.
     January 2013                           5-21             Draft - Do Not Quote or Cite

-------
 1    5.3   REFERENCES

 2    Buekers, J; Redeker, ES; Smolders, E. (2009). Lead toxicity to wildlife: Derivation of a critical blood
 3           concentration for wildlife monitoring based on literature data [Review]. Sci Total Environ
 4           407: 3431-3438. http://dx.doi.Org/10.1016/j.scitotenv.2009.01.044

 5    Frey, C; Samet, JM. (2011). CAS AC review of the EPA's integrated science assessment for lead
 6           (first external review draft- May 2011). (EPA-CASAC-12-002). Washington, DC: U.S.
 7           Environmental Protection Agency, Clean Air Scientific Advisory Committee.

 8    Henderson, R. (2007a) Letter from Dr. Rogene Henderson, Chair, Clean Air Scientific Advisory
 9           Committee, to Administrator Stephen L.  Johnson. Re: Clean Air Scientific Advisory
10           Committee's (CAS AC) Review of the 1 st Draft Lead Staff Paper and Draft Lead Exposure
11           and Risk Assessments. March 27, 2007.

12    Henderson, R. (2007b) Letter from Dr. Rogene Henderson, Chair, Clean Air Scientific Advisory
13           Committee, to Administrator Stephen L.  Johnson. Re: Clean Air Scientific Advisory
14           Committee's (CASAC) Review of the 2nd Draft Lead Human Exposure and Health Risk
15           Assessments. September 27, 2007.

16    ICF International. (2006) Lead Human Exposure and Health Risk Assessments and Ecological Risk
17           Assessment for Selected Areas. Pilot Phase.  Draft Technical Report.  Prepared for the U.S.
18           EPA's Office of Air Quality Planning and Standards, Research Triangle Park, NC.
19           December.

20    MacDonald, D.D., Ingersoll, C.G., and Berger, T.A. (2000) Development and evaluation of
21           consensus-based sediment quality guidelines for freshwater ecosystems. Archives of
22           Environmental Contamination and Toxicology. 39:20-31.
23
24    MacDonald, D.D., Ingersoll, C.G., Smorong, D.E., Lindskoog, R.A., Sloane, G., and Biernacki, T.
25           (2003) Development and Evaluation of Numerical Sediment Quality Assessment Guidelines
26           for Florida Inland Waters. British Columbia: MacDonald Environmental Sciences,  Lt.
27           Columbia, MO: U.S. Geological Survey. Prepared for: Florida Department of Environmental
28           Protection, Tallahassee, FL. January.

29    U.S. Environmental Protection Agency. (2005a)  Guidance for Developing Ecological Soil  Screening
30           Levels. Washington, DC: Office of Solid Waste and Emergency Response.OSWER Directive
31           9285.7-55. November.

32    U.S. Environmental Protection Agency. (2005b)  Ecological Soil Screening Levels forLead, Interim
33           Final. Washington, DC: Office of Solid Waste and Emergency Response. OSWER Directive
34           9285.7-70. Available at http://www.epa.gov/ecotox/ecossl/pdf/eco-ssl_lead.pdf

35    U.S. Environmental Protection Agency. (2006a)  Air Quality Criteria for Lead. Washington, DC,
36           EPA/600/R-5/144aF. Available online at:
37           http://www.epa.gov/ttn/naaqs/standards/pb/s_pb cr.html

38    U.S. Environmental Protection Agency. (2006b)  Analysis Plan for Human Health and Ecological


      January 2013                           5-22            Draft - Do Not Quote or Cite

-------
 1          Risk Assessment for the Review of the Lead National Ambient Air Quality Standards.
 2          Office of Air Quality Planning and Standards, Research Triangle Park, NC. Available at:
 3          http://www.epa.gov/ttn/naaqs/standards/pb/s_pb  cr_pd.html

 4    U.S. Environmental Protection Agency. (2007a). Framework for metals risk assessment [EPA
 5          Report]. (EPA 120/R-07/001). Washington, D.C. Available online at:
 6          http://www.epa.gov/raf/metalsframework/index.htm

 7    U.S. Environmental Protection Agency. (2007b). Review of the National Ambient Air Quality
 8          Standards for Lead: Policy Assessment of Scientific and Technical Information OAQPS Staff
 9          Paper. Washington, DC, EPA-452/R-07-013. Available online at:
10          http://www. epa. gov/ttn/naaqs/standards/pb/s_pb  cr.html

11    U.S. Environmental Protection Agency. (2012) Integrated Science Assessment for Lead (Third
12          External Review Draft). Washington, DC, EPA/600/R-10/075C. Available online at:
13          http: //www. epa. gov/ttn/naaqs/standards/pb/s_pb_index. html
     January 2013                           5-23             Draft - Do Not Quote or Cite

-------
 1              6   REVIEW OF THE SECONDARY STANDARD FOR LEAD

 2          This chapter presents preliminary staff conclusions regarding the secondary Pb standard.
 3    These preliminary staff conclusions are guided by consideration of key policy-relevant questions
 4    and based on the assessment and integrative synthesis of information presented in the ISA and by
 5    staff analyses and evaluations presented in chapters 2 and 5 herein. Final evaluations and staff
 6    conclusions, to be developed taking into consideration CASAC advice and public comment on
 7    this draft document, will be presented in the final PA. The final evaluations and staff
 8    conclusions will inform the Administrator's decisions on whether to retain or revise the existing
 9    secondary standard for Pb.
10          Following an introductory section on the general approach for reviewing the secondary
11    standard (section 6.1), including a summary of considerations in the last review, the discussion in
12    this chapter focuses on the central issue of whether the information available in this review
13    supports or calls into question the adequacy of the current secondary standard.  Building on the
14    responses to specific policy-relevant questions on the scientific evidence and exposure-risk
15    information in chapter 5 above, presentation in section 6.2 is also organized into consideration of
16    key policy-relevant questions framing evidence-based and exposure/risk-based considerations.
17    The policy-relevant questions in this document are based on those included in the IRP (IRP,
18    section 3.2).  In section 6.3, preliminary staff conclusions are developed.  Section 6.4 presents a
19    brief overview of key uncertainties and areas for future research.

20    6.1   APPROACH
21          Staffs approach for reviewing the current secondary standard, which involves translating
22    scientific and technical information into the basis for addressing key policy-relevant questions,
23    takes into consideration the approaches used in the previous Pb NAAQS review. The past and
24    current approaches described below are all based most fundamentally on using EPA's
25    assessment of the current scientific evidence and  previous quantitative analyses to inform the
26    Administrator's judgment with regard to the secondary standard for Pb.
27          Staff final conclusions will be developed for the next version of this document with
28    consideration of CASAC advice and public comment. In drawing conclusions regarding the
29    secondary standard for the Administrator's consideration, we note that the final decision on the
30    adequacy of the current secondary Pb standard is largely a public welfare policy judgment to be
31    made by the Administrator. The Administrator's final decision draws upon scientific
32    information and analyses about welfare effects, exposure and risks, as well as judgments about
33    the appropriate response to the range of uncertainties that are inherent in the scientific evidence
34    and analyses.  This approach is  consistent with the requirements of the NAAQS provisions of the
      January 2013                           6-1             Draft - Do Not Quote or Cite

-------
 1    Act.  Section 109 of the Clean Air Act requires the Administrator to establish a secondary
 2    standard that, in the judgment of the Administrator, is "requisite to protect the public welfare
 3    from any known or anticipated adverse effects associated with the presence of the pollutant in
 4    the ambient air".  In so doing, the Administrator seeks to establish standards that are neither
 5    more nor less stringent than necessary for this purpose.

 6        6.1.1   Approach Used in the Last Review
 7          The current secondary standard for Pb is equal to the primary standard for human health.
 8    As summarized in section 1.2.2 and described in more detail in section 4.1.1, the primary
 9    standard was substantially revised in the last review based on the much-expanded health effects
10    evidence of neurocognitive effects of Pb in children.  The level of the revised NAAQS is 0.15
11    Hg/m3.  The averaging time was also revised to a rolling three-month period with a maximum
12    (not-to-be-exceeded) form, evaluated over a three-year period.  Compared to the previous
13    averaging time of one calendar quarter, this revision was considered to be more scientifically
14    appropriate and more protective for human health. The indicator of Pb-TSP was retained,
15    reflecting the evidence that Pb particles of all sizes pose health risks (73 FR 67007).
16          The scientific assessment for the last review is described in the 2006 Air Quality Criteria
17    for Lead (AQCD; USEPA, 2006b), multiple drafts of which received review by CAS AC and the
18    public. EPA also  conducted a pilot ecological risk assessment for the review, after consultation
19    with CAS AC and receiving public comment on a draft analysis plan (USEPA, 2006c). The pilot
20    ecological risk assessment was released in December 2006 (ICF, 2006)  and the policy
21    assessment based on these assessments, air quality analyses and key evidence from the AQCD
22    was presented in the Staff Paper (USEPA, 2007b), a draft of which  also received CASAC and
23    public review. The final Staff Paper presented OAQPS staffs evaluation of the public health and
24    welfare policy implications of the key studies and scientific information contained in the 2006
25    CD and presented and interpreted results from the pilot risk/exposure analyses conducted for this
26    review. Based on this evaluation, the Staff Paper presented OAQPS staff recommendations that
27    the Administrator give consideration to substantially revising the primary and secondary
28    standards to a range of levels at or below 0.2  ug/m3.
29          In the previous review, the Administrator considered the evidence-based information
30    presented in the CD and the risk-based information presented in the 2006 REA. It was his
31    conclusion that the available data and evidence, while primarily qualitative,  suggested that there
32    was the potential  for adverse environmental impacts under the then-current standard. Given the
33    limited data on Pb effects in ecosystems, it was at the time,  as it is now, necessary to look at
34    evidence of Pb effects on organisms and extrapolate to ecosystem effects. Therefore, taking into
35    account the available evidence and current media concentrations in  a wide range of areas,  the

      January 2013                           6-2             Draft - Do Not Quote or Cite

-------
 1    Administrator concluded that there was potential for adverse effects occurring under the then-
 2    current standard; however there was insufficient data to provide a quantitative basis for setting a
 3    secondary standard different than the primary (73 FR 67011). Therefore, citing a general lack of
 4    data that would indicate the appropriate level of Pb in environmental media that may be
 5    associated with adverse effects, as well as the comments of the CAS AC Pb panel that a
 6    significant change to current air concentrations (e.g., via a significant change to the standard)
 7    was likely to have significant beneficial effects on the magnitude of Pb exposures in the
 8    environment, the secondary standard was revised to be consistent with the revised primary
 9    standard (73 FR 67011).

10        6.1.2   Approach for the Current Review
11           In evaluating whether it is appropriate to consider retaining the current secondary Pb
12    standard, or whether consideration of revision is appropriate, we have adopted an approach in
13    this review that builds on the general approach from the last review and reflects the body of
14    evidence and information now available.  As summarized above, the Administrator's decisions in
15    the previous review were based on the  conclusion that there was the potential for adverse
16    ecological effects under the previous standard. In our approach here, we intend to focus on
17    consideration of the extent to which a broader body of scientific evidence is now available that
18    would inform decisions on either the potential for adverse effects to ecosystems under the current
19    standard or the ability to set a more ecologically relevant secondary standard than was feasible in
20    the previous review. In conducting this assessment, we draw on the ecological effects evidence
21    presented in detail in the ISA, and aspects summarized above in chapter 5, along with the
22    information associated with the screening-level  risk assessment also summarized above. Figure
23    6-1 illustrates the basic construct of our approach in developing conclusions regarding options
24    appropriate for the Administrator to consider in this review with regard to the adequacy of the
25    current secondary NAAQS standard and, as appropriate, potential alternate standards.
26           In developing preliminary conclusions in this review, we have taken into account both
27    evidence-based and risk-based considerations framed by a series of policy-relevant questions.
28    These questions are outlined in the sections below and generally discuss the extent to which we
29    are able to better characterize  effects and the likelihood of adverse effects in the environment
30    under the current standard.  Our approach to considering these questions recognizes that the
31    available welfare effects evidence generally reflects laboratory-based evidence of toxicological
32    effects on specific organisms exposed to concentrations of Pb at which scientists generally agree
33    that adverse effects are likely to occur. It is widely recognized,  however, that environmental
34    exposures from atmospherically-derived Pb are likely to be lower and/or accompanied by
      January 2013                           6-3             Draft - Do Not Quote or Cite

-------
1    significant confounding and modifying factors (e.g., other metals, acidification), which increases
2    our uncertainty about the likelihood and magnitude of organism and ecosystem responses.
                                          Adequacy of Current Standard
                      Evidence-based Considerations

            > Does currently available evidence and related
            uncertainties strengthen or call into question prior
            conclusions?
               • Evidence of welfare effects not previously identified
               or at lower exposures?
               • Newly identified at-risk ecosystems?
               •Changed understanding of lead bioavailability and
               movement of through ecosystems?
               •Effects at lower air-related exposures then
               previously understood or in conditions that would
               have likely met current standards?
        Risk-based Considerations

> Nature, magnitude and importance of
screening level risk information?
> Uncertainties in the screening level risk
assessment?
                                              Does information call into
                                              question the adequacy of
                                                 current standard?
                         Consider
                         Retaining
                      Current Standard
                                                                 YES
                                   Consideration of Potential Alternative Standards
                                                        I
                                       Elements of Potential Alternative Standards
                                                   > Indicator
                                                   > Averaging Time
                                                   >Form
                                                   > Level
                                   Potential Alternative Standards for Consideration
     Figure 6-1.  Overview of approach for review of current secondary standard.

     January 2013                               6-4              Draft - Do Not Quote or Cite

-------
 1    6.2   ADEQUACY OF THE CURRENT STANDARD
 2             In considering the adequacy of the current Pb standard, the overarching question we
 3    consider is:
 4         •   Does the currently available scientific evidence- and exposure/risk-based
 5             information, as reflected in the ISA and REA, support or call into question the
 6             adequacy of the protection afforded by the current Pb standard?
 7          To assist us in interpreting the currently available scientific evidence and screening-level
 8    risk information to address this question, we have focused on a series of more specific questions,
 9    posed within sections  6.2.1  and 6.2.2 below. In so doing, we consider both the information
10    available at the time of the last review and information newly available since the last review
11    which has been critically analyzed and characterized in the ISA.

12        6.2.1  Evidence-based Considerations
13          In considering the welfare effects evidence with respect to the adequacy of the current
14    standard, we must consider the array of evidence newly assessed in the ISA and the degree to
15    which the newly available evidence supports previous conclusions about the effects of Pb in the
16    environment and the extent to which it reduces areas of uncertainty recognized in decisions on
17    the secondary standard in the last review. Additionally, as the secondary standard set in 2008 is
18    significantly  lower than the previous standard, which was the focus of evaluation in the last
19    review, we must consider the extent to which the current evidence supports conclusions about
20    the effects of current ambient levels under the  current standard and the extent to which the
21    evidence supports or calls into question the adequacy of protection afforded by the current
22    standard.
23         •   To what extent does the available information indicate that Pb-related effects are
24             occurring as a result of multimedia pathways associated with ambient air
25             conditions  that would meet the current standard?
26          The current evidence continues to support our conclusions from the previous review
27    regarding key aspects of the ecological effects evidence for Pb and the effects of exposure
28    associated with levels of Pb occurring through ecological media in the U.S.  Our conclusions in
29    this regard are based on consideration of the assessment of the currently available evidence in the
30    ISA, particularly with regard to key aspects summarized in chapter 5 of this PA, in light of the
31    assessment of the evidence in the last review as described in the 2006 CD and summarized in the
32    notice of final rulemaking (73 FR 67008). Key aspects of these previous conclusions are
33    summarized below.
34         •   Deposition of gasoline-derived Pb into forest soils has produced a legacy of slow
35             moving Pb that remains bound to organic materials despite the removal of Pb from

      January 2013                          6-5            Draft - Do Not Quote or Cite

-------
 1             most fuels and the resulting dramatic reductions in overall deposition rates (2006 CD,
 2             Section AX7.1.4.3).
 3         •   For areas influenced by point sources of air Pb that meet the standard considered in the
 4             previous review, concentrations of Pb in soil may exceed by many orders of magnitude
 5             the concentrations which are considered harmful to laboratory organisms (2006 CD,
 6             Section 3.2 and AX7.1.2.3).
 7         •   Some Pb  deposited before the previous standard was enacted is still present in soils and
 8             sediments; historic Pb from gasoline continues to move slowly through systems as does
 9             current Pb derived from both air and nonair sources.
10         •   The evidence of adversity in natural systems is very sparse due in no small part to the
11             difficulty in determining the effects of confounding factors such as multiple metals or
12             factors influencing bioavailability in field studies.
13         •   Environmental conditions exist in which Pb-associated adverse effects to aquatic
14             organisms and thereby ecosystems may be anticipated given experimental results.
15         •   Some areas with long term historical deposition of Pb to sediment from a variety of
16             sources as well as areas influenced by point sources have the potential for adverse
17             effects to aquatic communities (2006 CD, Sections AX7.2.2.2.2  and AX7.2.4).
18           The range of effects of Pb on terrestrial and aquatic organisms based on new information
19    in the current review are summarized in the ISA (ISA, section 2.7 and section 7.3 and 7.4) and
20    largely mirror the conclusions presented above from the previous review.
21           The integrated  synthesis contained in the ISA conveys how effects of Pb can vary with
22    species and life stage, duration of exposure, form of Pb, and media characteristics such as soil
23    and water chemistry. A wide range of organism effects are recognized, including effects on
24    growth, development (particularly of the neurological system) and reproductive success (ISA,
25    section 7.3 and section 7.4). Lead is recognized to distribute from the air into multiple
26    environmental media, as summarized in section 1.3 above, contributing to multiple exposure
27    pathways for ecological receptors. As discussed in section 5.1 above, many factors affect the
28    bioavailability of Pb to receptors in terrestrial and aquatic ecosystems, contributing to differences
29    between laboratory-assessed toxicity and Pb toxicity in these ecosystems,  challenging our
30    consideration of environmental impacts of Pb emitted to ambient air.
31           In studies in a variety of ecosystems, adverse ecosystem-level effects (including
32    decreases in species diversity, loss of vegetation, changes to community composition, decreased
33    growth of vegetation, and increased number of invasive species) have been demonstrated near
34    smelters, mines and other industries that have released Pb, among other materials (ISA, section
35    7.3.12 and 7.4.12).  As noted in section 5.1 above, however, our ability to  characterize the role of
36    air emissions of Pb in contributing to these effects is complicated because of coincident releases
37    to other media and  of other pollutants.  Co-released pollutants include a variety of other heavy
38    metals, in addition to sulfur dioxide,  which may cause toxic effects in themselves and may
      January 2013                           6-6            Draft - Do Not Quote or Cite

-------
 1    interact with Pb in the environment. There is also some uncertainty as to the extent to which Pb
 2    in ambient air is contributing to the exposure. These uncertainties limit our ability to draw
 3    conclusions regarding the extent to which Pb-related effects may be associated with ambient air
 4    conditions that would meet the current standard.
 5          The role of historically emitted Pb, as discussed in section  1.3.2 above, poses additional
 6    complications in addressing this question. The vast majority of Pb in the environment today,
 7    particularly in terrestrial ecosystems, was deposited in the past during the use of Pb additives in
 8    gasoline (USEPA 2007b, section 6.2.1) although industrial coal use has also been documented as
 9    a significant  source as well (ISA, section 3.2.1). This gasoline-derived Pb was emitted in very
10    large quantities (CD, pi AX7-98 and ISA, Figure 3-8) and predominantly in small sized particles
11    which were widely dispersed and transported across large distances (ISA, section 3.2).  As
12    recognized in section 2.3.1 and 2.3.3.2 above, historical records provided by sediment cores in
13    various environments document the substantially reduced Pb deposition (associated with reduced
14    Pb emissions) in many locations (ISA, section 3.2.1). As Pb is extremely persistent in the
15    environment, past environmental releases are expected to generally dominate current media
16    concentrations. There is limited evidence to relate specific ecosystem effects with current
17    ambient air concentrations of Pb through deposition to terrestrial and aquatic ecosystems and
18    subsequent movement of deposited Pb through the  environment (e.g., soil, sediment, water,
19    organisms). The potential for ecosystem effects of Pb from atmospheric sources under conditions
20    meeting the current standard is difficult to assess due to limitations on  available information not
21    only on bioavailability (as affected by the specific characteristics of the receiving ecosystem),
22    but also  information to fully characterize the distribution of Pb from the atmosphere into
23    ecosystems in long-term exposure  scenarios. Therefore, the connection between air concentration
24    and ecosystem exposure and associated potential for welfare effects continues to be poorly
25    characterized for Pb (ISA, Section 7.5), and is even harder to characterize with respect to the
26    current standard than it  was in last review with respect to the previous  standard.
27         •   To what extent have important uncertainties identified in the last review been
28             reduced and/or have new uncertainties emerged?
29          With regard to uncertainties in effects associated with Pb in the ambient air, the last
30    review illustrated the largest of these uncertainties were the ability to determine the prevalence of
31    effects in the natural environment, the ability to apportion effects between air and nonair sources,
32    and the understanding of the effects in the context of the current standard (IRP,  section 3.2).
33    While we have some new evidence with regard to bioavailability and its role in determining
34    effects, overall these remain the major sources of uncertainty in evaluating effects and exposure
35    evidence for Pb in ecosystems.

      January 2013                           6-7             Draft - Do Not Quote or Cite

-------
 1           With regard to the prevalence of effects in the natural environment, it remains difficult to
 2    assess the concentration at which Pb elicits a specific effect in terrestrial and aquatic systems,
 3    due to the influence of other environmental variables on both Pb bioavailability and toxicity, and
 4    also due to substantial species differences in Pb susceptibility. There is little new information
 5    that would facilitate extrapolation of evidence on individual  species, generally from controlled
 6    laboratory studies, to conclusions regarding adversity to populations or ecosystems within the
 7    natural environment.  For example, in terrestrial systems, evidence reviewed in the ISA (ISA,
 8    Sections 7.2.3 and 7.2.4) demonstrates that exposure to Pb is generally associated with negative
 9    effects on growth, reproduction and  survival in terrestrial ecosystems. Many factors, including
10    species composition and various soil physiochemical properties, interact strongly with Pb
11    concentration to modify effects. Also, Pb aging is a particularly important factor in terrestrial
12    systems, where  soil is the main route of exposure, and it is difficult to reproduce experimentally.
13    Without quantitative  characterization of those interactions, characterizations of exposure-
14    response relationships would likely not be transferable outside of experimental settings (ISA,
15    Section 2.7.1).
16            As reported in the ISA, there continues to be limited evidence linking current ambient
17    air sources of Pb through deposition to environmental media (e.g., soil, sediment, water,  and
18    biota) and to ecological effects (ISA, section  7.3.11). The ability to apportion effects between air
19    deposited Pb and nonair sources is difficult to assess due to a lack of information not only on the
20    bioavailability of that Pb, which is greatly influenced by the specific characteristics of the
21    receiving ecosystem,  but also on Pb  distribution in ecosystems in long-term exposure scenarios.
22           Consideration of the environmental risks associated with the current standard is
23    complicated by  the environmental burden associated with air Pb concentrations, predominantly
24    in the past, that  exceeded the current standard (USEPA,  2007b,  section 6.4.4.2). For example, a
25    large portion of Pb deposited before the standard was enacted is likely still present in soils and
26    sediments and historic Pb from gasoline continues to move slowly through systems as does
27    current Pb derived from both air and nonair sources. As a general matter, the  currently available
28    evidence, as assessed in the ISA, does not significantly reduce any of these areas of uncertainty
29    or inform our consideration of the adequacy of the current standard.
30
      January 2013                            6-8             Draft - Do Not Quote or Cite

-------
 1        6.2.2   Exposure/Risk-based Considerations
 2

 3         •  To what extent does risk or exposure information indicate that air-related
 4             ecosystem exposures important from a public welfare perspective are likely to
 5             occur with air Pb levels that just meet the current standard?
 6          The current evidence continues to support our conclusions with regard to interpreting the
 7    risk and exposure results from the previous review. Our conclusions in this regard are based on
 8    consideration of the screening-level ecological risk assessment results from the previous review
 9    in light of the currently available evidence in the ISA and as described in the 2006 REA and
10    summarized in the notice of final rulemaking (73 FR 67009). Key aspects of these results in the
11    context of the current standard are summarized below, drawing from the discussion of the 2006
12    REA results in section 5.2.2 above, which is based on the information in the 2006 REA (ICF,
13    2006).
14         Primary Pb Smelter Case Study

15         •  While the contribution to Pb concentrations from air as compared to nonair sources is
16            not quantified, air emissions from this facility are substantial (ICF 2006). Currently,
17            the county where  this facility is located exceeds the now current Pb NAAQS.
18         Secondary Pb Smelter Case Study

19         •  While the contribution from air deposited Pb to the overall Pb concentrations modeled
20            in soils at this location is unclear, the facility continues to emit Pb and the county
21            where this facility is located exceeds the now current Pb NAAQS (Appendix 5-A).
22         Near Roadway Non-urban Case Study

23         •   These case study locations are highly impacted by past deposition of gasoline Pb. It is
24            unknown whether current conditions at these sites exceed the now current Pb standard
25            but given evidence from the past of Pb concentrations near highways that ranged above
26            the previous (1978) Pb standard (1986 CD, section 7.2.1), conditions at these locations
27            during the time of leaded gasoline very likely exceeded the 1978 standard.
28         Vulnerable Ecosystem Case Study

29         •  The previous review concluded that atmospheric Pb inputs do not directly affect stream
30            Pb levels at Hubbard Brook Experimental Forest because deposited Pb is almost
31            entirely retained in the soil profile.  It is unlikely that conditions  have changed from the
32            previous conclusions made based on soil data through 2000 and therefore, current
33            ambient air concentrations likely  do not directly impact stream Pb levels under air
34            quality conditions associated with meeting the now current standard.
35         National-scale Surface Water and Sediment Screen

36         •  The extent to which past air emissions of Pb have contributed to surface water or
37            sediment Pb concentrations at the locations identified in the screen is unclear. For
38            some of the  surface water locations, nonair sources likely contributed significantly to

      January 2013                            6-9            Draft - Do Not Quote or Cite

-------
 1            the surface water Pb concentrations.  For other locations, a lack of nearby nonair
 2            sources indicated a potential role for air sources to contribute to observed surface water
 3            Pb concentrations.  Additionally, these concentrations may have been influenced by Pb
 4            in resuspended sediments and/or reflect contribution of Pb from erosion of soils with
 5            Pb derived from historic as well as current air emissions.
 6
 7          Although the available risk and exposure information continues to be sufficient to
 8    conclude that the 1978 standard was not providing adequate protection to ecosystems, that
 9    information, when considered with regard to air-related ecosystem exposures likely to occur with
10    air Pb levels that just meet the now current standard, does not provide evidence that the current
11    secondary standard is inadequate.

12    6.3   PRELIMINARY STAFF CONCLUSIONS ON THE SECONDARY STANDARD
13          This section describes preliminary staff conclusions regarding  adequacy of the current
14    secondary Pb standard.  These conclusions are based on considerations described above and in
15    the discussion below regarding the currently available scientific evidence summarized in the
16    third draft ISA and prior CDs and  the risk and exposure information drawn from the 2006 REA.
17    The final PA will draw upon the final ISA, developed  after CAS AC review and public comment
18    on the third draft ISA.  Staff conclusions presented in the final PA will take into account advice
19    from CAS AC and public comment on this document and these draft conclusions.
20          Taking into consideration the discussions responding to specific questions above  in this
21    and the prior chapter, this  section addresses the following overarching policy question:
22         •   Does the currently available scientific evidence-  and exposure/risk-based
23             information, as reflected in the ISA and REA, support or call into question the
24             adequacy of the protection afforded by the current secondary Pb standard?
25          With respect to evidence-based considerations, staff gives weight to the body of evidence
26    on the ecological effects of Pb in the natural environment, expanded in some aspects since the
27    last review, which continues to support identification of ecological effects in organisms relating
28    to growth, reproduction, and survival  as the most relevant endpoints associated with Pb
29    exposure. The currently available evidence, while somewhat expanded since the last review,
30    does not include evidence of significant effects at lower concentrations or evidence of higher
31    level ecosystem  effects beyond those reported in the last review. There continues to be
32    significant difficulties in interpreting effects evidence  from laboratory studies to the natural
33    environment and linking those effects to ambient air Pb concentrations.  Further, we are aware of
34    no new critical loads information that would inform our interpretation of the public welfare
35    significance of the effects  of Pb in various ecosystems. In summary, while new research has
36    added to the understanding of Pb biogeochemistry and expanded the list of organisms for which

      January 2013                          6-10           Draft - Do Not Quote or Cite

-------
 1    Pb effects have been described, there remains a significant lack of knowledge about the potential
 2    for adverse effects from ambient air Pb in the environment and the exposures that occur from
 3    such air-derived Pb, particularly under the current standard.
 4          With respect to exposure/risk-based considerations, we recognize the complexity of
 5    interpreting the previous risk assessment with respect to ambient air Pb under the current
 6    standard and the associated limitations and uncertainties of such assessments. For example, the
 7    location-specific case studies as well as the national screen reflect both current air Pb deposition
 8    as well as past air and nonair source contributions.  We conclude that while the previous
 9    assessment is consistent with and generally supportive of the evidence-based conclusions about
10    Pb in the environment the limitations of apportioning Pb between past and present air
11    contributions and between air and nonair sources remain significant.
12          In making the decision on revision of the secondary standard in the last review, the
13    Administrator concurred with the conclusions by CAS AC that we lack the relevant data to
14    provide a clear quantitative basis for setting a secondary Pb NAAQS that differs from the
15    primary in indicator, averaging time, level or form.  As a result, the Administrator revised the
16    secondary standard to be identical in all respects to the primary standard (73 FR 67012).
17          In now considering the adequacy of the current standard in this review, we have
18    considered evidence assessed in the ISA and discussed in sections 5.1, 2.3 and 6.2.1 above, as
19    well as information gained from the 2006 REA discussed in sections 5.2 and 6.2.2 above. As
20    summarized in section 6.2.1 above, the scientific evidence presented in detail in the ISA,
21    inclusive of that newly available in this review, is not substantively changed, most particularly
22    with regard to our needs  in reviewing the current standard, from the information that was
23    available in the last review.  Further, our updated consideration of the screening-level risk
24    information from the 2006 REA, as summarized in section 6.2.2 above, additionally does not
25    provide alternative conclusions with regard to a basis for judging the adequacy of the current
26    standard. Thus, while the current secondary standard is not expressed in elements derived from
27    ecological effects evidence,  we preliminarily conclude that the currently available information
28    does not call into question the adequacy of the current standard to provide the requisite
29    protection for public welfare. In addition, as in the last review, there continues to be insufficient
30    data to provide a quantitative basis for setting a secondary standard different than the primary
31    standard. Accordingly, we reach the preliminary conclusion that it  is appropriate to consider
32    retaining the current standard without revision.  This preliminary conclusion is based on
33    consideration of the welfare effects evidence in combination with the screening-level
34    exposure/risk information (chapter 5 and sections 6.2.1  and 6.2.2 above) and staff judgments on
35    the associated public welfare implications.

      January 2013                           6-11           Draft - Do Not Quote or Cite

-------
 1    6.4   KEY UNCERTAINTIES AND AREAS FOR FUTURE RESEARCH AND DATA
 2         COLLECTION
 3          In this section, we highlight key uncertainties associated with reviewing the secondary
 4    NAAQS for Pb. Such key uncertainties and recommendations for ecosystem-related research,
 5    model development, and data gathering are outlined below. In some cases, research in these
 6    areas can go beyond aiding standard setting to aiding in the development of more efficient and
 7    effective control strategies. We note, however, that a full set of research recommendations to
 8    meet standards implementation and strategy development needs is beyond the scope of this
 9    discussion. Rather, listed below are key uncertainties and research questions and data gaps that
10    have been thus far highlighted in this review of the health-based primary standard.

11         •   An important source of uncertainty in characterizing the effects of air related Pb in the
12             environment is the ability to apportion the sources of Pb to specific ecosystems.

13             -   It is difficult to identify the atmospheric component of Pb deposited in ecosystems
14                from other sources of Pb. Without such information, it is not possible to determine
15                the appropriately protective air concentrations of Pb.

16             -   Development of critical loads information for various ecosystems would allow for a
17                much improved understanding of the significance of air contributions to
18                ecosystems.

19         •   There is significant uncertainty regarding the timeframes under which Pb currently in
20             the environment was deposited.
21             -   In general, the connection between air concentration of Pb and ecosystem exposure
22                continues to be poorly  characterized.

23             -   There is uncertainty in characterizing the fate and transport of Pb in the
24                environment through air, water, and soil media, especially in ecosystems. This
25                limits our ability to account for past Pb exposures in current media  concentrations
26                and to link effects to current air contributions.

27         •   Our understanding of the degree of adverse impact that is occurring in the environment
28             due to Pb is incomplete and leads to uncertainty in interpreting the significance of
29             effects at specific endpoints.

30             -   Critical loads are important to capture the variability of response across ecosystems
31                to determine which ecosystems are most sensitive  and additional research is needed
32                to provide inputs for development of critical loads, particularly for a broader range
33                of ecosystem level endpoints.

34             -   The available studies on community and ecosystem-level effects are sparse and
35                usually from contaminated areas where Pb concentrations are much higher than
36                typically encountered in the environment.

37             -   The role of modifying factors on Pb toxicity continues to be a source of uncertainty
38                when determining the degree of adverse impact from Pb.

      January 2013                          6-12            Draft - Do Not Quote or Cite

-------
1            -  Evidence regarding relationships between organism-level effects of Pb (e.g.,
2                reproduction, growth, and survival) and effects at the population level and higher,
3                is lacking especially in natural systems.

4            -  Better indicators of Pb effects are needed in estuarine and marine systems to allow
5                us to understand the potential for adverse effects from Pb in these environments.
    January 2013                           6-13            Draft - Do Not Quote or Cite

-------
 1    6.5   REFERENCES

 2    Frey, C; Samet, JM. (2011). CAS AC review of the EPA's integrated science assessment for lead
 3          (first external review draft- May 2011). (EPA-CASAC-12-002). Washington, DC: U.S.
 4          Environmental Protection Agency, Clean Air Scientific Advisory Committee.

 5    Henderson, R. (2007a) Letter from Dr. Rogene Henderson, Chair, Clean Air Scientific Advisory
 6          Committee, to Administrator Stephen L. Johnson. Re: Clean Air Scientific Advisory
 7          Committee's (CAS AC) Review of the 1 st Draft Lead Staff Paper and Draft Lead
 8          Exposure and Risk Assessments. March 27, 2007.

 9    Henderson, R. (2007b) Letter from Dr. Rogene Henderson, Chair, Clean Air Scientific Advisory
10          Committee, to Administrator Stephen L. Johnson. Re: Clean Air Scientific Advisory
11          Committee's (C ASAC) Review of the 2nd Draft Lead Human Exposure and Health Risk
12          Assessments. September 27, 2007.

13    ICF International. (2006) Lead Human Exposure and Health Risk Assessments and Ecological
14          Risk Assessment for Selected Areas. Pilot Phase. Draft Technical Report.  Prepared for
15          the U.S. EPA's Office of Air Quality Planning and Standards, Research Triangle Park,
16          NC.  December.

17    U.S. Environmental Protection Agency. (2006) Air Quality Criteria for Lead. Washington, DC,
18          EPA/600/R-5/144aF. Available online at:
19          http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_cr.html

20    U.S. Environmental Protection Agency. (2012) Integrated Science Assessment for Lead (Third
21          External Review Draft). Washington, DC, EPA/600/R-10/075C. Available online at:
22          http://www.epa.gov/ttn/naaqs/standards/pb/s_pb_index.html
     January 2013                          6-14           Draft - Do Not Quote or Cite

-------
                                    APPENDICES

Appendix 2A. The 2008 NEI:  Data Sources, Limitations and Confidence	2A-1
Appendix 2B. Recent Regulatory Actions on Stationary Sources of Lead	2B-1
Appendix 2C. Criteria for Air Quality Data Analysis	2C-1
Appendix 2D. Air Quality Data Analysis Summary	2D-1
Appendix 3 A. Interpolated Risk Estimates for the Generalized (Local) Urban Case Study.. 3A-1
Appendix 5 A. Additional Detail on 2006 Ecological Screening Assessment	5 A-1
         January 2013                                 Draft - Do Not Quote or Cite

-------
                                              APPENDIX 2A
                      THE 2008 NEI:  DATA SOURCES, LIMITATIONS AND
                                              CONFIDENCE
 4          The process of identifying sources that emit Pb into the air has been ongoing since before
 5    the Clean Air Act of 1970. The comprehensiveness of emission inventories generally, and the
 6    NEI, specifically, depends upon knowledge of source types emit Pb, their locations and their
 7    operating characteristics, as well as the reporting of this information to the inventory. As noted
 8    above, the NEI relies on information that is available from a variety of sources for this
 9    information. There are numerous steps, each with its  own uncertainties, associated with the
10    development of this information for use in the emissions inventory.  First, the categories emitting
11    Pb must be  identified.  Second, the sources' processes and control devices must be known.
12    Third, the activity throughputs and operating schedules of these sources must be known.  Finally,
13    we must have emission factors to relate emissions to the operating throughputs, process
14    conditions and  control devices. The process, control device, throughputs and operating
15    schedules are generally available for each source.  However, the emission factors represent
16    average emissions for a source type and average emissions may differ significantly from source
17    to source. In some cases,  emissions testing provides source-specific information.  In others,
18    emissions factors must be estimated from similar sources or source categories, or from other
19    information. More information on emission factors and the estimation of emissions is found in
20    the introduction to EPA's Compilation of Air Pollutant Emissions Factors.l
21          The Pb  emissions information presented in chapter 2 is drawn largely from EPA's NEI
22    for 2008. The NEI is based on information submitted from State, Tribal and local air pollution
23    agencies and data obtained during the preparation  of technical support information for EPA's
24    hazardous air pollutant regulatory programs. Data in the 2008 NEI for Pb  emissions from the
25    use of leaded aviation gasoline were developed by EPA using the Federal Aviation
26    Administration's operations activity data, where available.2  The data were then reviewed by
27    State, Tribal, and local air pollution agencies. With some  additions, the information presented in
28    this document is primarily based on version 3 of the NEI for 2008, available on the EPA's
29    CHIEF website at (http://www.epa.gov/ttn/chief/net/2008inventory.html).  The NEI is limited
            1 U.S. Environmental Protection Agency. (1996-2011). AP-42, Compilation of Air Pollutant Emission
      Factors, 5th Edition. Volume 1: Stationary Point and Area Sources, Chapter 13: Miscellaneous Sources. Available
      at: http://www.epa.gov/ttn/chief/ap42/chl3/index.html. Further information on emission factors is available at:
      http://www.epa.gov/ttn/chief/ap42/
            2 Eastern Research Group (ERG), 201 la. Project report: Documentation for Aircraft Component of the
      National Emissions Inventory Methodology, ERG No. 0245.03.402.011, January 27, 2011.
            January 2013                      2A-1               Draft - Do Not Quote or Cite

-------
 1    with regard to Pb emissions estimates for some sources.  For example, we have not yet
 2    developed estimates for the NEI of Pb emissions for the miscellaneous categories of on-road
 3    emissions (e.g., combustion of fuel with Pb traces, lubricating oil, mechanical wear of vehicle
 4    components, etc.) and Pb that may be emitted from wildfires, or for emissions associated with
 5    resuspension of Pb residing in roadway dust and nearby surface soil in areas not otherwise
 6    associated with industrial facilities (see section 2.1.2).
 7           The 2008 NEI underwent extensive external review, including a review of the process for
 8    developing the inventory which includes extensive quality assurance and quality control steps
 9    (QA/QC). We provided feedback reports to point source data providers and we posted several
10    versions of the inventory on our website. We also conducted additional QA targeted at facilities
11    with appreciable Pb emissions in previous years to ensure that 2008 Pb emissions were reported,
12    and we augmented with TRI where facilities with annual emissions greater than 0.5 tpy in 2005
13    were not reported to NEI in 2008 due to Pb emissions reporting thresholds
14    (ftp://ftp.epa.gov/EmisInventory/20008v2/doc/2008neiv2_issues.xlsx). Further, there was
15    additional QA/QC  conducted for emission inventory information for facilities that are included
16    in Risk and Technology Review source categories.3 As a result we have strong confidence in the
17    quality of the data for these facilities. In summary, generic limitations to the 2008 NEI include
18    the following.
19         •   Consistency: The 2008 NEI for Pb is a composite of emissions estimates generated by
20             state and local regulatory agencies, industry, and EPA.  Because the estimates
21             originated from a variety of sources, as well as for differing purposes, they will in turn
22             vary in quality, whether Pb is reported for particular source types, method of reporting
23             compound classes, level of detail, and geographic coverage.
24         •   Variability in Quality and Accuracy of Emission Estimation Methods: The  accuracy of
25             emission estimation techniques varies with pollutants and source categories. In some
26             cases, an estimate may be based on a few or only one emission measurement at a
27             similar source.  The techniques used and quality  of the estimates will vary between
28             source categories and between area, major, and mobile source sectors. Generally, the
29             more review and scrutiny given to emissions data by states and other agencies, the
30             more certainty and accuracy there is in those estimates.
31
32
             3 The Risk and Technology Review is a combined effort to evaluate both risk and technology as required
      by the Clean Air Act (CAA) after the application of maximum achievable control technology (MACT) standards.
      Section 112(f)(2) of the CAA directs EPA to conduct risk assessments on each source category subject to MACT
      standards, and to determine if additional standards are needed to reduce residual risks. Section 112(d)(6) of the CAA
      requires EPA to review and revise the MACT standards, as necessary, taking into account developments in
      practices, processes and control technologies. For more information: http://www.epa.gov/ttn/atw/rrisk/rtrpg.html.
            January 2013                       2A-2                Draft - Do Not Quote or Cite

-------
                                         APPENDIX 2B
3
4
     RECENT REGULATORY ACTIONS ON STATIONARY SOURCES OF LEAD

      The table below identifies recent regulatory actions on stationary sources estimated to
result in reductions in lead emissions.
Action
NESHAP
NESHAP
NESHAP
NESHAP
NESHAP
NESHAP
NESHAP
NESHAP
NESHAP
NESHAP,
NSPS
EGES,
NSPS
EGES,
NSPS
NESHAP
NESHAP
EGES,
NSPS
NESHAP
Source Categories Affected
Area Sources: Polyvinyl Chloride and Copolymers Production, Primary Copper Smelting,
Secondary Copper Smelting, and Primary Nonferrous Metals (zinc, cadmium, beryllium)
Area Sources: Acrylic and Modacrylic Fibers Production, Carbon Black Production,
Chemical Manufacturing: Chromium Compounds, Flexible Polyurethane Foam
Production and Fabrication, Lead Acid Battery Manufacturing, Wood Preserving:
Area Sources: Clay Ceramics Manufacturing, Glass Manufacturing, and Secondary
Nonferrous Metals Processing
Area Sources: Electric Arc Furnace Steelmaking Facilities
Iron and Steel Foundries Area Sources
Area Sources: Acrylic and Modacrylic Fibers Production, Carbon Black Production,
Chemical Manufacturing: Chromium Compounds, Flexible Polyurethane Foam
Production and Fabrication, Lead Acid Battery Manufacturing, Wood Preserving
Area Source Standards for Plating and Polishing Operations
Area Source Standards for Nine Metal Fabrication and Finishing Source Categories
Area Source Standards for Aluminum, Copper, and Other Nonferrous Foundries
Portland Cement Manufacturing Industry
Commercial and Industrial Solid Waste Incineration Units
Standards of Performance for New Stationary Sources and Emission Guidelines for
Existing Sources: Sewage Sludge Incineration Units
Major Sources: Industrial, Commercial, and Institutional Boilers, and Process Heaters
Area Sources: Industrial, Commercial, and Institutional Boilers and Process Heaters
Standards of Performance for New Stationary Sources and Emissions Guidelines for
Existing Sources: Hospital/Medical/lnfectious Waste Incinerators
Ferroalloys Production
Citation
76 FR 22848
(2011)
72 FR 38864
(7/16/2007)
72 FR 731 79
(12/26/2007)
72 FR 74088
(12/28/2007)
73 FR 225
(1/2/2008)
73 FR 5923
(3/26/2008)
73 FR 37728
(7/1/2008)
73 FR 42978
7/23/2008)
74 FR 30366
(6/25/2009)
75 FR 54970
(9/9/2010)
76 FR 15704
(3/21/2011)
76 FR 15372
(3/21/2011)
76 FR 15608
(3/21/2011)
76 FR 15554
(3/21/2011)
76 FR 18407
(4/4/2011)
76 FR 72508
(11/23/2011)
EGES=Emissions Guidelines for Existing Sources
NESHAP=National Emission Standards for Hazardous Air Pollutants
NSPS= New Source Performance Standards
          January 2013
                                      2B-1
Draft - Do Not Quote or Cite

-------
 1                                          APPENDIX 2C


 2                     CRITERIA FOR AIR QUALITY DATA ANALYSIS


 3          Criteria for the 2009-2011 data analysis presented in Figures 2-10 through 2-13 are as
 4   listed below, with abbreviations defined following this list.

 5         •   Years utilized were 2009-2011. November and December 2008 data were also used for
 6            rolling 3-month average analysis.

 7         •   Data were extracted from EPA's Air Quality System in August and September of 2012.

 8         •   Data for parameters 12128 (STP) and 14129 (LC) were used for Pb-TSP analysis; data
 9            for parameters 84128 (STP) and 88128 (LC) were used for Pb-PM2.5 analysis; and data
10            for parameters 82128 (STP) and 85128 (LC) were used for Pb-PMi0 analysis. No
11            adjustment was made to the STP or LC data to make them more comparable to each
12            other; based on previous analysis,  the LC and STP versions of same-site-day samples
13            are extremely similar.  Collocated data - for same size cut- were combined to a site
14            basis by using the following hierarchy, evaluated each site-day:  LC parameter data
15            takes precedence over STP parameter data and lower POC numbers take precedence
16            over higher POC numbers.

17         •   For maximum rolling 3-month average analysis, the 2009-2011 databases also
18            encompassed the prior 2 months (i.e., November and December 2008 data):  3 or more
19            observations were needed to make a valid monthly average (for Pb-TSP, the daily data
20            had to be collected/analyzed via an FRM or FEM); 3 valid consecutive months were
21            required to make a valid rolling 3-month period; 6 valid rolling 3-month periods were
22            required to make a valid year; and 1, 2, or 3 valid years were required to make a valid
23            3-month metric. The rolling 3-month average metric (by size cut for 2009-2011) was
24            identified as the highest valid 3-month average across the valid (1, 2, or 3) years.

25         •   For maximum monthly analysis: 3or more observations were needed to make a valid
26            monthly average (for Pb-TSP, the  daily data had to be collected/analyzed via an FRM
27            or FEM); 6 valid months were required to make a valid year; and 1, 2, or 3 valid years
28            were required to make a valid monthly metric. The maximum monthly metric (by size
29            cut for 2009-2011) was identified  as the highest valid monthly average across the valid
30            (1,2, or 3) years.

31         •   For annual average analysis: monthly averages were used to construct annual averages,
32            3 or more observations were required to make a valid month (for Pb-TSP, the daily
33            data had to be collected/analyzed via an FRM or FEM);  6 valid months were required
34            to make a valid year (i.e., annual average) and 1, 2, or 3 valid years were required to
35            make a valid annual metric.  The annual metric (by size cut for 2009-2011) was
36            identified as the average of the 1, 2, or 3 (2009, 2010, and/or 2011) valid annual means.
37
            January 2013                     2C-1               Draft - Do Not Quote or Cite

-------
1        Definitions for Abbreviations
2        FEM - Federal equivalent method
3        FRM - Federal reference method
4        LC - Local conditions
5        POC - Parameter occurrence code
6        STP - Standard temperature and pressure
           January 2013                      2C-2               Draft - Do Not Quote or Cite

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-1. Pb-TSP concentrations at Pb-TSP sites, 2009-2011.
SITE
Max 3-month
Mean (ug/m3)
Annual Mean
(ug/m3)
Max Monthly
Mean (ug/m3)
previous
source
current
source
State
County
011090003 1.035 0.564 2.069 0 1 Alabama Pike County
040071002
0.267
0.133
0.306
0
1
Arizona
Gila County
040078000 0.061 0.033 0.077 0 1 Arizona Gila County
040134018
0.033
0.023
0.042
0
0
Arizona
Maricopa County
060250005 0.023 0.016 0.032 0 0 California Imperial County
060371103
0.016
0.011
0.018
0
0
California Los Angeles County
060371302 0.019 0.01 0.026 0 0 California Los Angeles County
060371402
0.044
0.022
0.064
0
0
California
Los Angeles County
060371403 0.116 0.061 0.148 0 1 California Los Angeles County
060371404
0.108
0.061
0.14
0
1
California Los Angeles County
060371405 0.667 0.343 0.801 0 1 California Los Angeles County
060371406
0.085
0.049
0.094
0
1
California
Los Angeles County
060371602 0.022 0.011 0.04 0 0 California Los Angeles County
060374002
0.01
0.006
0.014
0
0
California Los Angeles County
060374004 0.012 0.007 0.017 0 0 California Los Angeles County
060375005
0.008
0.003
0.01
0
0
California
Los Angeles County
060651003 0.011 0.006 0.014 0 0 California Riverside County
060658001
0.011
0.007
0.013
0
0
California
Riverside County
060711004 0.011 0.007 0.014 0 0 California San Bernardino County
060719004
0.013
0.008
0.014
0
0
California
San Bernardino County
080050007 0.02 0.014 0.028 0 0 Colorado Arapahoe County
080310025
0.01
0.006
0.013
0
0
Colorado Denver County
120570100 0.044 0.026 0.1 0 1 Florida Hillsborough County
120571066
0.979
0.282
1.959
0
1
Florida
Hillsborough County
120571073 0.423 0.116 0.977 0 1 Florida Hillsborough County
130150003
0.019
0.014
0.032
0
1
Georgia
Bartow County
130890003 0.004 0.003 0.006 0 0 Georgia DeKalb County
132150011
0.071
0.029
0.14
1
1
Georgia
Muscogee County
170310001 0.024 0.016 0.028 0 0 Illinois Cook County
170310022
0.051
0.033
0.07
0
0
Illinois
Cook County
170310026 0.034 0.023 0.038 0 0 Illinois Cook County
170310052
0.024
0.016
0.032
0
0
Illinois
Cook County
170310110 0.294 0.1 0.58 0 1 Illinois Cook County
170310210
0.053
0.037
0.092
0
1
Illinois
Cook County
170313103 0.013 0.011 0.016 0 0 Illinois Cook County
170313301
0.027
0.018
0.028
0
0
Illinois
Cook County
170314201 0.011 0.01 0.014 0 0 Illinois Cook County
170316003
0.036
0.024
0.04
0
0
Illinois
Cook County
171150110 0.199 0.082 0.386 0 1 Illinois Macon County
171170002
0.011
0.01
0.012
0
0
Illinois
Macoupin County
171190010 0.416 0.118 0.836 0 1 Illinois Madison County
171193007
0.036
0.018
0.053
0
0
Illinois Madison County
171430037 0.011 0.01 0.013 0 0 Illinois Peoria County
171430110
0.016
0.013
0.024
0
1
Illinois
Peoria County
171430210 0.015 0.011 0.026 0 1 Illinois Peoria County
January 2013
2D- 1
Draft - Do Not Quote or Cite

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-1. Pb-TSP concentrations at Pb-TSP sites, 2009-2011.
SITE
171630010
Max 3-month
Mean (ug/m3)
0.029
Annual Mean
(ug/m3)
0.019
Max Monthly
Mean (ug/m3)
0.038
previous
source
1
current
source
0
State
Illinois
County
St. Clair County
171950110 0.028 0.02 0.04 0 1 Illinois Whiteside County
172010110
0.063
0.031
0.118
0
1
Illinois Winnebago County
180350008 0.11 0.079 0.141 0 1 Indiana Delaware County
180350009
0.262
0.186
0.427
0
1
Indiana
Delaware County
180890023 0.077 0.026 0.087 0 1 Indiana Lake County
180890032
0.06
0.027
0.144
0
1
Indiana
Lake County
180890033 0.139 0.054 0.298 0 1 Indiana Lake County
180892008
0.037
0.016
0.049
0
0
Indiana
Lake County
180970063 0.079 0.026 0.125 0 1 Indiana Marion County
180970076
0.02
0.01
0.029
0
1
Indiana Marion County
180970078 0.011 0.005 0.013 0 0 Indiana Marion County
181270023
0.022
0.013
0.032
0
1
Indiana
Porter County
181630020 0.007 0.004 0.01 0 0 Indiana Vanderburgh County
191550011
0.263
0.122
0.282
0
1
Iowa
Pottawattamie County
191630015 0.012 0.01 0.024 0 0 Iowa Scott County
201690004
0.421
0.18
0.488
0
1
Kansas
Saline County
210190016 0.004 0.004 0.007 0 1 Kentucky Boyd County
211510003
0.069
0.034
0.121
0
1
Kentucky Madison County
212070001 0.036 0.021 0.076 0 1 Kentucky Russell County
220330014
0.005
0.004
0.007
0
1
Louisiana
East Baton Rouge Parish
220950003 0.053 0.037 0.066 0 1 Louisiana St. John the Baptist Parish
250250042
0.01
0.007
0.012


Massachusetts
Suffolk County
260670003 0.284 0.126 0.414 0 1 Michigan Ionia County
260810020
0.008
0.005
0.01
0
0
Michigan
Kent County
261630001 0.006 0.005 0.007 0 0 Michigan Wayne County
261630033
0.023
0.012
0.034
0
0
Michigan Wayne County
270036020 0.033 0.017 0.054 0 1 Minnesota Anoka County
270072303
0.001
0
0.002
0
0
Minnesota
Beltrami County
270177416 000 Minnesota Carlton County
270210001
0.001
0
0.002
0
0
Minnesota
Cass County
270370020 0.011 0.005 0.022 0 0 Minnesota Dakota County
270370423
0.002
0.001
0.005
0
0
Minnesota
Dakota County
270370442 0.013 0.003 0.007 0 0 Minnesota Dakota County
270370443
0.003
0.002
0.005
0
0
Minnesota
Dakota County
270370465 0.259 0.128 0.568 0 1 Minnesota Dakota County
270370470
0.002
0.001
0.004
0
0
Minnesota
Dakota County
270530963 0.006 0.003 0.01 0 0 Minnesota Hennepin County
270530966
0.016
0.004
0.044
0
0
Minnesota Hennepin County
270531007 0.004 0.002 0.006 0 0 Minnesota Hennepin County
270532006
0.003
0.001
0.005
0
0
Minnesota
Hennepin County
270750005 000 Minnesota Lake County
271230866
0.016
0.012
0.023
0
0
Minnesota Ramsey County
271230871 0.007 0.002 0.01 0 0 Minnesota Ramsey County
271231003
0.006
0.004
0.008
0
0
Minnesota Ramsey County
January 2013
2D-2
Draft - Do Not Quote or Cite

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-1. Pb-TSP concentrations at Pb-TSP sites, 2009-2011.
SITE
Max 3-month
Mean (ug/m3)
Annual Mean
(ug/m3)
Max Monthly
Mean (ug/m3)
previous
source
current
source
State
271377001 0.003 0.001 0.006 0 0 Minnesota
271377555
0.006
0.002
0.008
0
0
Minnesota
271453053 0.006 0.003 0.01 0 1 Minnesota
271630009
0.005
0.003
0.006
0
0
Minnesota
271630438 0.005 0.002 0.008 0 0 Minnesota
271630446
0.002
0
0.003
0
0
Minnesota
290930016 2.07 0.49 1.392 0 1 Missouri
290930021
1.937
0.622
2.438
0
1
Missouri
290930027 0.083 0.038 0.121 1 1 Missouri
290930029
0.115
0.037
0.271
1
1
Missouri
290930033 0.057 0.027 0.105 0 1 Missouri
290930034
0.402
0.283
0.509
0
1
Missouri
290939007 0.581 0.392 0.844 0 1 Missouri
290939008
0.387
0.256
0.538
0
1
Missouri
290970005 0.019 0.013 0.024 0 0 Missouri
290990004
1.16
0.889
1.576
0
1
Missouri
290990005 0.396 0.231 0.536 0 1 Missouri
290990009
0.067
0.043
0.087
0
1
Missouri
290990011 0.403 0.21 0.509 0 1 Missouri
290990013
0.198
0.116
0.344
0
1
Missouri
290990015 1.079 0.921 1.349 0 1 Missouri
290990020
0.943
0.646
1.28
0
1
Missouri
290990021 0.969 0.585 1.124 0 1 Missouri
290990022
0.63
0.434
0.861
0
1
Missouri
290990023 0.594 0.394 0.724 0 1 Missouri
290990024
0.659
0.374
0.862
0
1
Missouri
290990025 0.085 0.043 0.134 0 1 Missouri
290990026
0.072
0.035
0.114
0
1
Missouri
290999001 1.204 0.824 1.623 0 1 Missouri
290999002
0.431
0.232
0.593
0
1
Missouri
290999003 0.467 0.198 0.605 0 1 Missouri
290999004
0.261
0.119
0.328
0
1
Missouri
290999005 1.432 0.959 2.185 0 1 Missouri
290999006
0.091
0.049
0.131
0
1
Missouri
291790001 0.059 0.033 0.087 0 1 Missouri
291790002
0.068
0.033
0.172
0
1
Missouri
291790003 0.1 0.027 0.268 0 1 Missouri
291790034
0.093
0.068
0.142
0
1
Missouri
291870006 0.132 0.05 0.302 0 0 Missouri
291870007
0.041
0.043
0.364
0
0
Missouri
291892003 0.041 0.007 0.011 0 0 Missouri
310530005
0.072
0.051
0.149
0
1
Nebraska
311270002 0.115 0.042 0.206 0 1 Nebraska
360470122
0.019
0.014
0.02
0
0
New York
360713001 0.101 0.031 0.134 0 1 NewYork
County
St. Louis County
St. Louis County
Stearns County
Washington County
Washington County
Washington County
Iron County
Iron County
Iron County
Iron County
Iron County
Iron County
Iron County
Iron County
Jasper County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Jefferson County
Reynolds County
Reynolds County
Reynolds County
Reynolds County
St. Francois County
St. Francois County
St. Louis County
Dodge County
Nemaha County
Kings County
Orange County
January 2013
2D-3
Draft - Do Not Quote or Cite

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-1. Pb-TSP concentrations at Pb-TSP sites, 2009-2011.
SITE
360713002
Max 3-month
Mean (ug/m3)
1.027
Annual Mean
(ug/m3)
0.128
Max Monthly
Mean (ug/m3)
2.821
previous
source
0
current
source
1
State
New York
360713004 0.012 0.005 0.016 0 1 New York
390170015
0.009
0.006
0.013
0
0
Ohio
390290019 0.057 0.022 0.136 0 0 Ohio
390290020
0.025
0.015
0.035
0
0
Ohio
390290022 0.044 0.02 0.065 0 0 Ohio
390350038
0.021
0.013
0.026
0
0
Ohio
390350042 0.03 0.014 0.044 0 0 Ohio
390350049
0.531
0.165
0.719
0
1
Ohio
390350061 0.023 0.015 0.03 1 0 Ohio
390350072
0.035
0.014
0.054
0
1
Ohio
390490025 0.011 0.008 0.012 0 0 Ohio
390510001
0.178
0.073
0.21
0
1
Ohio
390910006 0.006 0.004 0.008 1 1 Ohio
391010003
0.056
0.035
0.097
0
1
Ohio
391510017 0.023 0.014 0.028 0 1 Ohio
391550012
0.011
0.007
0.017
0
1
Ohio
391670008 0.007 0.005 0.01 0 0 Ohio
391670010
0.008
0.005
0.01
0
0
Ohio
401159006 0.015 0.011 0.022 0 0 Oklahoma
401159007
0.034
0.017
0.056
0
0
Oklahoma
401210416 0.004 0.003 0.006 0 1 Oklahoma
410711702
0.035
0.018
0.037
0
1
Oregon
420030002 0.016 0.012 0.024 0 0 Pennsylvania
420030008
0.014
0.01
0.02
0
0
Pennsylvania
420030070 0.056 0.022 0.11 0 1 Pennsylvania
420031009
0.138
0.043
0.149
0
1
Pennsylvania
420070006 0.085 0.054 0.198 0 1 Pennsylvania
420070007
0.253
0.171
0.393
0
1
Pennsylvania
420070505 0.115 0.086 0.17 0 1 Pennsylvania
420110005
0.051
0.043
0.07
0
1
Pennsylvania
420110020 0.196 0.127 0.347 0 1 Pennsylvania
420110021
0.139
0.049
0.319
0
1
Pennsylvania
420110022 0.118 0.043 0.261 0 1 Pennsylvania
420110717
0.116
0.065
0.108
0
1
Pennsylvania
420111717 0.251 0.133 0.348 0 1 Pennsylvania
420210808
0.073
0.05
0.128
0
0
Pennsylvania
420450002 0.047 0.043 0.048 0 0 Pennsylvania
420450004
0.047
0.045
0.048
0
1
Pennsylvania
420550002 0.046 0.045 0.047 0 1 Pennsylvania
420630005
0.049
0.029
0.058
0
1
Pennsylvania
420730011 0.023 0.032 0.046 0 1 Pennsylvania
420790036
0.137
0.095
0.268
0
1
Pennsylvania
421010449 0.03 0.021 0.039 1 0 Pennsylvania
421011002
0.025
0.022
0.029
0
0 Pennsylvania
County
Orange County
Orange County
Butler County
Columbiana County
Columbiana County
Columbiana County
Cuyahoga County
Cuyahoga County
Cuyahoga County
Cuyahoga County
Cuyahoga County
Franklin County
Fulton County
Logan County
Marion County
Stark County
Trumbull County
Washington County
Washington County
Ottawa County
Ottawa County
Pittsburg County
Yamhill County
Allegheny County
Allegheny County
Allegheny County
Allegheny County
Beaver County
Beaver County
Beaver County
Berks County
Berks County
Berks County
Berks County
Berks County
Berks County
Cambria County
Delaware County
Delaware County
Franklin County
Indiana County
Lawrence County
Luzerne County
Philadelphia County
Philadelphia County
January 2013
2D-4
Draft - Do Not Quote or Cite

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-1. Pb-TSP concentrations at Pb-TSP sites, 2009-2011.
SITE
Max 3-month
Mean (ug/m3)
Annual Mean
(ug/m3)
Max Monthly
Mean (ug/m3)
previous
source
current
source
State
County
421290007 0.043 0.041 0.043 0 0 Pennsylvania Westmoreland County
421290009
0.046
0.027
0.047
0
1
Pennsylvania
Westmoreland County
450190003 0.007 0.005 0.013 0 0 South Carolina Charleston County
450450015
0.006
0.005
0.008
0
0
South Carolina
Greenville County
450790007 0.005 0.004 0.007 0 0 South Carolina Richland County
450790019
0.024
0.011
0.061
0
0
South Carolina
Richland County
470930023 0.165 0.114 0.208 0 1 Tennessee Knox County
470930027
0.042
0.02
0.078
0
1
Tennessee
Knox County
470931017 0.037 0.018 0.063 0 1 Tennessee Knox County
471633001
0.093
0.058
0.115
1
1
Tennessee
Sullivan County
471633002 0.06 0.039 0.069 1 1 Tennessee Sullivan County
471633003
0.06
0.041
0.066
1
1
Tennessee
Sullivan County
471633004 0.08 0.041 0.124 1 1 Tennessee Sullivan County
471634002
0.038
0.025
0.062
1
1
Tennessee
Sullivan County
480610006 0.005 0.004 0.007 0 0 Texas Cameron County
480850003
0.371
0.179
0.625
0
1
Texas
Collin County
480850007 0.199 0.105 0.24 0 1 Texas Collin County
480850009
0.774
0.453
1.178
0
1
Texas Collin County
480850029 0.18 0.076 0.335 0 1 Texas Collin County
481410002
0.04
0.023
0.087
0
0
Texas
El Paso County
481410033 0.039 0.019 0.057 0 0 Texas El Paso County
481410055
0.014
0.009
0.015
0
0
Texas
El Paso County
481410058 0.033 0.017 0.056 0 0 Texas El Paso County
482011034
0.008
0.005
0.009
0
0
Texas
Harris County
482570020 0.104 0.06 0.11 0 1 Texas Kaufman County
483750024
0.013
0.005
0.029
0
1
Texas Potter County
484790016 0.026 0.016 0.035 0 0 Texas Webb County
490351001
0.057
0.023
0.086
0
1
Utah
Salt Lake County
510090007 0.018 0.004 0.007 0 1 Virginia Amherst County
510270006
0.013
0.01
0.015
0
1
Virginia
Buchanan County
510870014 0.004 0.005 0.024 0 0 Virginia Henrico County
517700011
0.109
0.038
0.272
0
1
Virginia
Roanoke city
551170008 0.152 0.055 0.225 0 1 Wisconsin Sheboygan County
720130001
0.339
0.188
0.416
0
1
Puerto Rico
Arecibo Municipio
720210010 0.011 0.005 0.027 0 0 Puerto Rico Bayamon Municipio
January 2013
2D-5
Draft - Do Not Quote or Cite

-------
                  Appendix 2D - Air Quality Data Analysis Summary




Table 2D-2. Pb-PM10 concentrations at urban PM10 sites, 2009-2011.
SITE
040134009
040139997
060371103
060658001
060850005
080770017
110010043
120573002
120951004
121030018
121030026
130890002
170314201
171639010
210190002
210670014
250250042
261630001
261630033
295100085
360050080
360050110
360551007
410290133
410390060
410390062
410510246
440070022
482011035
482011039
490110004
500070007
510870014
530330080
550790010
Max
3-month
Mean Annual Mean
(ug/m3) (ug/m3)

0.007
0.013
0.013
0.005
0.004
0.004
0.003

0.003
0.003
0.002
0.006

0.019
0.005
0.004
0.006
0.023
0.018
0.008
0.006
0.003
0.003
0.003
0.002
0.008
0.003
0.007
0.003
0.006
0.002
0.007
0.005

0.005
0.004
0.006
0.006
0.002
0.001
0.003
0.001
0.003
0.002
0.002
0.001
0.004
0.015
0.012
0.004
0.003
0.004
0.011
0.011
0.006
0.005
0.002
0.002
0.002
0.002
0.004
0.003
0.005
0.002
0.003
0.001
0.003
0.003
0.006
Max
Monthly
Mean
(ug/m3)
0.008
0.007
0.026
0.03
0.006
0.004
0.005
0.003
0.008
0.004
0.003
0.003
0.01
0.031
0.028
0.009
0.004
0.006
0.035
0.026
0.012
0.008
0.004
0.005
0.006
0.003
0.013
0.004
0.009
0.004
0.009
0.002
0.016
0.006
0.01
State
Arizona
Arizona
California
California
California
Colorado
District of Columbia
Florida
Florida
Florida
Florida
Georgia
Illinois
Illinois
Kentucky
Kentucky
Massachusetts
Michigan
Michigan
Missouri
New York
New York
New York
Oregon
Oregon
Oregon
Oregon
Rhode Island
Texas
Texas
Utah
Vermont
Virginia
Washington
Wisconsin
County
Maricopa County
Maricopa County
Los Angeles County
Riverside County
Santa Clara County
Mesa County
District of Columbia
Hillsborough County
Orange County
Pinellas County
Pinellas County
DeKalb County
Cook County
St. Clair County
Boyd County
Fayette County
Suffolk County
Wayne County
Wayne County
St. Louis city
Bronx County
Bronx County
Monroe County
Jackson County
Lane County
Lane County
Multnomah County
Providence County
Harris County
Harris County
Davis County
Chittenden County
Henrico County
King County
Milwaukee County
January 2013
2D-6
Draft - Do not cite or quote

-------
                  Appendix 2D - Air Quality Data Analysis Summary




Table 2D-3. Pb-PMZ5 concentrations at urban CSN PMZ5 sites, 2009-2011.
SITE
Max
3-month
Mean
(ug/m3)
Annual
Mean
(ug/m3)
Max
Monthly
Mean
(ug/m3)
State
County
010730023 0.041 0.017 0.064 Alabama Jefferson County
010732003
0.032
0.008
0.078
Alabama
Jefferson County
010890014 0.003 0.002 0.003 Alabama Madison County
011011002
0.007
0.003
0.011
Alabama
Montgomery County
011130001 0.013 0.004 0.028 Alabama Russell County
020900010
0.006
0.003
0.008
Alaska
Fairbanks North Star Borough
040134009 0.004 0.003 0.005 Arizona Maricopa County
040139997
0.004
0.002
0.004
Arizona
Maricopa County
040191028 0.003 0.002 0.003 Arizona Pima County
051190007
0.004
0.002
0.006
Arkansas
Pulaski County
060010007 0.005 0.004 0.005 California Alameda County
060010011
0.012
0.005
0.023
California
Alameda County
060070002 0.003 0.002 0.003 California Butte County
060190008
0.003
0.002
0.004
California
Fresno County
060250005 0.021 0.012 0.029 California Imperial County
060290014
0.003
0.002
0.004
California
Kern County
060299001 0.001 0.001 0.001 California Kern County
060371103
0.02
0.005
0.021
California
Los Angeles County
060590007 0.042 0.017 0.043 California Orange County
060658001
0.005
0.003
0.006
California
Riverside County
060670006 0.004 0.002 0.009 California Sacramento County
060670010
0.003
0.002
0.005
California
Sacramento County
060712002 0.037 0.018 0.038 California San Bernardino County
060730003
0.005
0.003
0.006
California
San Diego County
060731002 0.004 0.002 0.005 California San Diego County
060850005
0.003
0.002
0.004
California
Santa Clara County
060950004 0.007 0.005 0.009 California Solano County
060990005
0.005
0.003
0.009
California
Stanislaus County
061072002 0.004 0.002 0.006 California Tulare County
061112002
0.003
0.002
0.003
California
Ventura County
080010006 0.003 0.002 0.003 Colorado Adams County
080310025
0.003
0.002
0.005
Colorado
Denver County
080770017 0.002 0.002 0.003 Colorado Mesa County
081230008
0.003
0.002
0.003
Colorado
Weld County
090090027 0.005 0.003 0.005 Connecticut New Haven County
100010003
0.004
0.002
0.007
Delaware
Kent County
100032004 0.004 0.003 0.006 Delaware New Castle County
110010043
0.003
0.002
0.004
District of Columbia
District of Columbia
120111002 0.002 0.002 0.003 Florida Broward County
120573002
0.003
0.002
0.003
Florida
Hillsborough County
120730012 0.005 0.003 0.009 Florida Leon County
121030026
0.003
0.002
0.003
Florida
Pinellas County
130210007 0.004 0.003 0.007 Georgia Bibb County
January 2013
2D-7
Draft - Do not cite or quote

-------
                  Appendix 2D - Air Quality Data Analysis Summary




Table 2D-3. Pb-PMZ5 concentrations at urban CSN PMZ5 sites, 2009-2011.
SITE
130590001
Max
3-month
Mean
(ug/m3)
0.002
Annual
Mean
(ug/m3)
0.002
Max
Monthly
Mean
(ug/m3)
0.003
State
Georgia
County
Clarke County
130890002 0.003 0.002 0.003 Georgia DeKalb County
131150003
0.006
0.003
0.009
Georgia
Floyd County
132150011 0.013 0.005 0.029 Georgia Muscogee County
132450091
0.005
0.003
0.008 Georgia
Richmond County
132950002 0.006 0.003 0.009 Georgia Walker County
150030010
0.002
0.002
0.002
Hawaii
Honolulu County
150032004 0.008 0.004 0.02 Hawaii Honolulu County
160010010
0.002
0.002
0.003 Idaho
Ada County
170310057 0.07 0.011 0.199 Illinois Cook County
170310076
0.005
0.003
0.005
Illinois
Cook County
170314201 0.004 0.003 0.006 Illinois Cook County
170434002
0.005
0.003
0.008 Illinois
DuPage County
171190024 0.02 0.011 0.032 Illinois Madison County
171630900
0.005
0.003
0.006
Illinois
St. Clair County
171639010 0.014 0.009 0.019 Illinois St. Clair County
180190006
0.004
0.003
0.005 Indiana
Clark County
180390008 0.004 0.003 0.005 Indiana Elkhart County
180890022
0.011
0.006
0.018
Indiana
Lake County
180892004 0.022 0.009 0.037 Indiana Lake County
180970078
0.005
0.003
0.007 Indiana
Marion County
181630012 0.003 0.002 0.003 Indiana Vanderburgh County
181630021
0.003
0.003
0.004
Indiana
Vanderburgh County
191130037 0.003 0.002 0.003 Iowa Linn County
191530030
0.003
0.002
0.005 Iowa
Polk County
191630015 0.011 0.006 0.017 Iowa Scott County
201730010
0.004
0.002
0.009
Kansas
Sedgwick County
202090021 0.006 0.004 0.01 Kansas Wyandotte County
210190017
0.005
0.003
0.007 Kentucky
Boyd County
210670012 0.007 0.003 0.012 Kentucky Fayette County
211110067
0.007
0.004
0.012
Kentucky
Jefferson County
211170007 0.006 0.004 0.01 Kentucky Kenton County
220150008
0.006
0.004
0.009
Louisiana
Bossier Parish
220330009 0.005 0.003 0.008 Louisiana East Baton Rouge Parish
240053001
0.006
0.003
0.007
Maryland
Baltimore County
240330030 0.003 0.002 0.005 Maryland Prince George's County
250130008
0.003
0.002
0.004 Massachusetts
Hampden County
250250042 0.003 0.002 0.004 Massachusetts Suffolk County
260810020
0.004
0.003
0.006
Michigan
Kent County
261150005 0.005 0.003 0.008 Michigan Monroe County
261470005
0.016
0.004
0.03 Michigan
St. Clair County
261630001 0.004 0.003 0.005 Michigan Wayne County
261630015 0.013 0.007 0.027 (Michigan |Wayne County
January 2013
2D-8
Draft - Do not cite or quote

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-3. Pb-PMZ5 concentrations at urban CSN PMZ5 sites, 2009-2011.
SITE
Max
3-month
Mean
(ug/m3)
Annual
Mean
(ug/m3)
Max
Monthly
Mean
(ug/m3)
261630033 0.017 0.007 0.029
270031002
0.009
0.006
0.013
270530963 0.003 0.002 0.005
271095008
0.004
0.002
0.006
280470008 0.003 0.002 0.003
280490019
0.003
0.002
0.004
290470005 0.003 0.002 0.003
290990019
0.015
0.005
0.018
295100085 0.011 0.006 0.013
300630031
0.003
0.002
0.003
310550019 0.004 0.003 0.008
320030020
0.002
0.002
0.003
320030540 0.003 0.002 0.004
320310016
0.003
0.002
0.003
340130003 0.004 0.003 0.006
340230006
0.003
0.002
0.005
340273001 0.003 0.002 0.003
340390004
0.004
0.003
0.007
350010023 0.003 0.002 0.005
360010005
0.005
0.003
0.007
360050110 0.004 0.003 0.005
360290005
0.011
0.007
0.016
360551007 0.003 0.002 0.003
360610134
0.004
0.003
0.004
360810124 0.005 0.003 0.011
370210034
0.002
0.002
0.003
370350004 0.003 0.002 0.004
370670022
0.006
0.003
0.011
371190041 0.003 0.002 0.003
371830014
0.002
0.002
0.003
380171004 0.003 0.002 0.005
390171004
0.011
0.006
0.012
390350038 0.009 0.005 0.013
390350060
0.009
0.006
0.011
390490081 0.003 0.003 0.005
390610040
0.01
0.004
0.015
390811001 0.009 0.006 0.012
390870012
0.007
0.005
0.01
390933002 0.01 0.006 0.018
390950026
0.004
0.003
0.005
390990014 0.012 0.006 0.018
391130032
0.014
0.004
0.034
391351001 0.003 0.003 0.004
State
County
Michigan Wayne County
Minnesota
Anoka County
Minnesota Hennepin County
Minnesota
Olmsted County
Mississippi Harrison County
Mississippi
Hinds County
Missouri Clay County
Missouri
Jefferson County
Missouri St. Louis city
Montana
Missoula County
Nebraska Douglas County
Nevada
Clark County
Nevada Clark County
Nevada
Washoe County
New Jersey Essex County
New Jersey
Middlesex County
New Jersey Morris County
New Jersey
Union County
New Mexico Bernalillo County
New York
Albany County
New York Bronx County
New York
Erie County
New York Monroe County
New York
New York County
New York Queens County
North Carolina
Buncombe County
North Carolina Catawba County
North Carolina
Forsyth County
North Carolina Mecklenburg County
North Carolina
Wake County
North Dakota Cass County
Ohio
Butler County
Ohio Cuyahoga County
Ohio
Cuyahoga County
Ohio Franklin County
Ohio
Hamilton County
Ohio Jefferson County
Ohio
Lawrence County
Ohio Lorain County
Ohio
Lucas County
Ohio Mahoning County
Ohio
Montgomery County
Ohio Preble County
January 2013
2D-9
Draft - Do not cite or quote

-------
                  Appendix 2D - Air Quality Data Analysis Summary




Table 2D-3. Pb-PMZ5 concentrations at urban CSN PMZ5 sites, 2009-2011.
SITE
391510017
Max
3-month
Mean
(ug/m3)
0.013
Annual
Mean
(ug/m3)
0.007
Max
Monthly
Mean
(ug/m3)
0.017
State
Ohio
County
Stark County
391530023 0.006 0.004 0.01 Ohio Summit County
401091037
0.002
0.002
0.003
Oklahoma
Oklahoma County
401431127 0.005 0.003 0.008 Oklahoma Tulsa County
410290133
0.003
0.001
0.002 Oregon
Jackson County
410390060 0.003 0.003 0.007 Oregon Lane County
410392013
0.002
0.002
0.004
Oregon
Lane County
410510080 0.004 0.003 0.005 Oregon Multnomah County
420030008
0.016
0.009
0.022 Pennsylvania
Allegheny County
420030064 0.03 0.017 0.052 Pennsylvania Allegheny County
420110011
0.012
0.006
0.014
Pennsylvania
Berks County
420210011 0.01 0.006 0.017 Pennsylvania Cambria County
420270100
0.003
0.002
0.004 Pennsylvania
Centre County
420290100 0.004 0.003 0.005 Pennsylvania Chester County
420430401
0.007
0.004
0.011
Pennsylvania
Dauphin County
420490003 0.005 0.003 0.007 Pennsylvania Erie County
420692006
0.007
0.003
0.01 Pennsylvania
Lackawanna County
420710007 0.005 0.004 0.008 Pennsylvania Lancaster County
420950025
0.008
0.004
0.018
Pennsylvania
Northampton County
421010004 0.007 0.003 0.012 Pennsylvania Philadelphia County
421010055
0.003
0.002
0.005 Pennsylvania
Philadelphia County
421011002 0.007 0.005 0.009 Pennsylvania Philadelphia County
421255001
0.007
0.004
0.015
Pennsylvania
Washington County
421290008 0.004 0.003 0.007 Pennsylvania Westmoreland County
421330008
0.006
0.003
0.01 Pennsylvania
York County
440070022 0.004 0.003 0.007 Rhode Island Providence County
440071010
0.003
0.002
0.003
Rhode Island
Providence County
450190049 0.003 0.002 0.003 South Carolina Charleston County
450450008
0.003
0.002
0.003
South Carolina
Greenville County
450450015 0.006 0.003 0.009 South Carolina Greenville County
450790007
0.003
0.002
0.004
South Carolina
Richland County
460990008 0.003 0.002 0.003 South Dakota Minnehaha County
470370023
0.003
0.002
0.004
Tennessee
Davidson County
470654002 0.004 0.003 0.006 Tennessee Hamilton County
470931020
0.006
0.003
0.007
Tennessee
Knox County
471251009 0.004 0.002 0.005 Tennessee Montgomery County
471570024
0.004
0.002
0.005
Tennessee
Shelby County
471570075 0.004 0.003 0.006 Tennessee Shelby County
480612004
0.002
0.002
0.003
Texas
Cameron County
481130050 0.005 0.003 0.007 Texas Dallas County
481130069
0.004
0.003
0.008
Texas
Dallas County
481390016 0.005 0.003 0.008 Texas Ellis County
481410044 0.006 0.003 0.008 Texas El Paso County
January 2013
2D- 10
Draft - Do not cite or quote

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-3. Pb-PMZ5 concentrations at urban CSN PMZ5 sites, 2009-2011.
SITE
Max
3-month
Mean
(ug/m3)
Annual
Mean
(ug/m3)
Max
Monthly
Mean
(ug/m3)
State
County
481410053 0.017 0.007 0.02 Texas El Paso County
482010024
0.006
0.004
0.008
Texas
Harris County
482011039 0.002 0.002 0.003 Texas Harris County
482450021
0.003
0.002
0.003
Texas
Jefferson County
483030325 0.007 0.003 0.018 Texas Lubbock County
483550034
0.014
0.003
0.036
Texas
Nueces County
484530020 0.006 0.002 0.014 Texas Travis County
490110004
0.004
0.003
0.007 Utah
Davis County
490353006 0.008 0.003 0.016 Utah Salt Lake County
490494001
0.007
0.003
0.012
Utah
Utah County
500070012 0.003 0.002 0.003 Vermont Chittenden County
510870014
0.005
0.002
0.009 Virginia
Henrico County
530110013 0.003 0.002 0.004 Washington Clark County
530330057
0.008
0.004
0.013
Washington
King County
530330080 0.003 0.003 0.004 Washington King County
530530029
0.005
0.003
0.008 Washington
Pierce County
530530031 0.016 0.004 0.033 Washington Pierce County
530611007
0.004
0.003
0.007
Washington
Snohomish County
530770009 0.003 0.003 0.004 Washington Yakima County
540390011
0.004
0.002
0.006 West Virginia
Kanawha County
540391005 0.005 0.003 0.008 West Virginia Kanawha County
540511002
0.005
0.004
0.006
West Virginia
Marshall County
540690010 0.008 0.005 0.01 West Virginia Ohio County
550790026
0.004
0.003
0.006
Wisconsin
Milwaukee County
551330027 0.011 0.006 0.013 Wisconsin Waukesha County
560210100
0.002
0.002
0.002 Wyoming
Laramie County
January 2013
2D- 11
Draft - Do not cite or quote

-------
                  Appendix 2D - Air Quality Data Analysis Summary




Table 2D-4. Pb-PMZ5 concentrations at non-urban PMZ5 IMPROVE sites, 2009-2011.
Max 3-mo
SITE Mean (ug/m3)
020130002
020680003
021220009
040018001
040038001
040039000
040070010
040078100
040170119
051019000
051130003
060150002
060270101
060430003
060519000
060930005
061059000
080039000
080579000
080679000
080839000
080979000
081119000
090050005
120179000
130499000
150019002
150099000
160230101
160370002
191370002
191770006
200139000
200170001
201950001
221279000
230031020
230090103
230291004
240239000
250070001
260839000
261539000
0.002
0.001
0.001
0.001
0.002
0.009
0.004
0.001
0.001
0.002
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.002
0.002
0.002
0.006
0
0.001
0.001
0.003
0.003
0.002
0.002
0.001
0.002
0.002
0.002
0.001
0.002
0.001
0.001
0.002
Annual Mean
(ug/m3)
0.001
0
0
0.001
0.001
0.004
0.003
0.001
0.001
0.001
0.001
0
0.001
0.001
0
0
0
0
0
0
0
0
0
0.001
0.001
0.001
0.001
0
0
0
0.002
0.002
0.002
0.001
0.001
0.001
0.001
0.001
0.001
0.002
0.001
0.001
0.001
Max Monthly
Mean (ug/m3) State
0.004
0.001
0.001
0.001
0.003
0.012
0.008
0.002
0.001
0.002
0.002
0.002
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.002
0.002
0.002
0.01
0.001
0.001
0.001
0.003
0.005
0.003
0.002
0.001
0.002
0.002
0.002
0.002
0.002
0.002
0.002
0.002
Alaska
Alaska
Alaska
Arizona
Arizona
Arizona
Arizona
Arizona
Arizona
Arkansas
Arkansas
California
California
California
California
California
California
Colorado
Colorado
Colorado
Colorado
Colorado
Colorado
Connecticut
Florida
Georgia
Hawaii
Hawaii
Idaho
Idaho
Iowa
Iowa
Kansas
Kansas
Kansas
Louisiana
Maine
Maine
Maine
Maryland
Massachusetts
Michigan
Michigan
County
Aleutians East Borough
Denali Borough
Kenai Peninsula Borough
Apache County
Cochise County
Cochise County
Gila County
Gila County
Navajo County
Newton County
Polk County
Del Norte County
Inyo County
Mariposa County
Mono County
Siskiyou County
Trinity County
Alamosa County
Jackson County
La Plata County
Montezuma County
Pitkin County
San Juan County
Litchfield County
Citrus County
Charlton County
Hawaii County
Maui County
Butte County
Custer County
Montgomery County
Van Buren County
Brown County
Chase County
Trego County
Winn Parish
Aroostook County
Hancock County
Washington County
Garrett County
Dukes County
Keweenaw County
Schoolcraft County
January 2013
2D- 12
Draft - Do not cite or quote

-------
                  Appendix 2D - Air Quality Data Analysis Summary




Table 2D-4. Pb-PMZ5 concentrations at non-urban PMZ5 IMPROVE sites, 2009-2011.
Max 3-mo
SITE Mean (ug/m3)
270759000
271339000
271699000
290390001
292070001
292130003
300279000
300299001
300479000
300499000
300779000
300819000
300859000
300870762
300899000
300919000
310699000
311719000
320079000
320339000
330074002
350039000
350059000
350279000
350281002
350399000
350539000
350559000
361019000
370110002
370959000
380070002
380130004
391219000
400450890
400719010
410330010
410358001
410610010
410630002
420019000
460330132
460710001
0.001
0.002
0.002
0.002
0.003
0.002
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.003
0.001
0.002
0.001
0.001
0.005
0.001
0.002
0.001
0.001
0.001
0.001
0.003
0.001
0.001
Annual Mean
(ug/m3)
0.001
0.002
0.002
0.002
0.002
0.002
0
0
0
0
0
0
0.001
0
0
0.001
0.001
0.001
0
0
0.001
0.001
0.001
0.001
0
0
0.001
0
0.002
0.001
0.001
0.001
0.001
0.003
0.001
0.001
0.001
0
0
0
0.002
0
0.001
Max Monthly
Mean (ug/m3) State
0.002
0.003
0.003
0.002
0.004
0.003
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.006
0.002
0.002
0.001
0.001
0.008
0.001
0.003
0.001
0.001
0.001
0.001
0.003
0.001
0.001
Minnesota
Minnesota
Minnesota
Missouri
Missouri
Missouri
Montana
Montana
Montana
Montana
Montana
Montana
Montana
Montana
Montana
Montana
Nebraska
Nebraska
Nevada
Nevada
New Hampshire
New Mexico
New Mexico
New Mexico
New Mexico
New Mexico
New Mexico
New Mexico
New York
North Carolina
North Carolina
North Dakota
North Dakota
Ohio
Oklahoma
Oklahoma
Oregon
Oregon
Oregon
Oregon
Pennsylvania
South Dakota
South Dakota
County
Lake County
Rock County
Winona County
Cedar County
Stoddard County
Taney County
Fergus County
Flathead County
Lake County
Lewis and Clark County
Powell County
Ravalli County
Roosevelt County
Rosebud County
Sanders County
Sheridan County
Garden County
Thomas County
Elko County
White Pine County
Coos County
Catron County
Chaves County
Lincoln County
Los Alamos County
Rio Arriba County
Socorro County
Taos County
Steuben County
Avery County
Hyde County
Billings County
Burke County
Noble County
Ellis County
Kay County
Josephine County
Klamath County
Union County
Wallowa County
Adams County
Custer County
Jackson County
January 2013
2D- 13
Draft - Do not cite or quote

-------
                  Appendix 2D - Air Quality Data Analysis Summary




Table 2D-4. Pb-PMZ5 concentrations at non-urban PMZ5 IMPROVE sites, 2009-2011.
Max 3-mo
SITE Mean (ug/m3)
480430101
481099000
490170101
490379000
490559000
500038001
511139000
511639000
530090020
530370004
530390010
530390011
530410007
530470012
540939000
560050123
560199000
560299002
560359000
560399000
780209000
0.002
0.001
0.001
0.001
0.001
0.001
0.002
0.004
0.001
0.001
0.002
0.001
0.001
0.001
0.002
0.001
0.001
0.001
0.001
0.001
0.001
Annual Mean
(ug/m3)
0.001
0.001
0
0
0
0.001
0.001
0.003
0.001
0.001
0.001
0.001
0
0
0.001
0
0
0
0
0
0.001
Max Monthly
Mean (ug/m3) State
0.002
0.001
0.001
0.001
0.001
0.002
0.002
0.006
0.002
0.001
0.002
0.001
0.001
0.001
0.002
0.001
0.001
0.001
0.001
0.001
0.001
Texas
Texas
Utah
Utah
Utah
Vermont
Virginia
Virginia
Washington
Washington
Washington
Washington
Washington
Washington
West Virginia
Wyoming
Wyoming
Wyoming
Wyoming
Wyoming

County
Brewster County
Culberson County
Garfield County
San Juan County
Wayne County
Bennington County
Madison County
Rockbridge County
Clallam County
Kittitas County
Klickitat County
Klickitat County
Lewis County
Okanogan County
Tucker County
Campbell County
Johnson County
Park County
Sublette County
Teton County

January 2013
2D- 14
Draft - Do not cite or quote

-------
                      Appendix 2D - Air Quality Data Analysis Summary

Table 2D-5. Distribution of maximum 3-month means, 2009-2011.
     (units are ug/m3)   n    min   Pct1   Pct5  PctIO  Pct25  mean PctSO Pct75  Pct90 Pct95  Pct99   max
Pb-TSPall            215    0    0.001  0.004  0.006  0.013  0.154  0.04  0.116  0.421  0.943  1.432   2.07
Pb-TSP source        112  0.004  0.004  0.012  0.019   0.05  0.273  0.11  0.355  0.774 1.079  1.937   2.07
Pb-TSP prev source     12  0.006  0.006  0.006  0.023  0.0295 0.057  0.06  0.082  0.093 0.115  0.115  0.115
Pb-TSP not source      91    0     0    0.002  0.003  0.007  0.019 0.012 0.025  0.041  0.047  0.132  0.132
Pb-PM10 urban         31  0.002  0.002  0.002  0.003  0.003  0.007 0.005 0.007  0.013 0.019  0.023  0.023
Pb-PM2.5 urban CSN   198  0.001  0.002  0.002  0.003  0.003  0.007 0.004 0.007  0.013  0.02   0.042   0.07
Pb-PM2.5 non-urban
IMPROVE            107    0    0.001  0.001  0.001  0.001  0.002 0.001 0.002  0.002 0.003  0.006  0.009
January 2013                                2D -15                  Draft - Do not cite or quote

-------
                      Appendix 2D - Air Quality Data Analysis Summary

Table 2D-6. Distribution of annual mean concentrations, 2009-2011.
 (units are ug/m3)      n    min   Pct1    Pct5   PctIO  Pct25   mean  PctSO  Pct75  Pct90  Pct95  Pct99  max
 Pb-TSPall           215    0     0     0.002  0.004  0.008  0.0798 0.022  0.058  0.198  0.434  0.889  0.959
 Pb-TSP source       112  0.003  0.003   0.005  0.013  0.027  0.1408 0.053  0.149  0.394  0.622  0.921   0.959
 Pb-TSP prev source   12  0.004  0.004   0.004  0.015  0.02  0.0306 0.033   0.04   0.041  0.058  0.058  0.058
 Pb-TSP not source    91    0     0      0    0.002  0.004  0.0113 0.008  0.016  0.023  0.041  0.05   0.05
 Pb-PM10 urban       35  0.001   0.001   0.001   0.001  0.002  0.0043 0.003  0.005  0.011  0.012  0.015  0.015
 Pb-PM2.5 urban CSN  198  0.001   0.001   0.002  0.002  0.002  0.0036 0.003  0.004  0.006  0.009  0.017  0.018
 Pb-PM2.5 non-urban
 IMPROVE           107    0     0      0      0      0    0.0008 0.001  0.001  0.002  0.002  0.003  0.004
January 2013                                2D - 16                  Draft - Do not cite or quote

-------
                      Appendix 2D - Air Quality Data Analysis Summary

Table 2D-7. Distribution of lead concentrations, 2009-2011.
 (units are ug/m3)	n   min   Pct1   Pct5  PctIO  Pct25  mean  PctSO  Pct75  Pct90 Pct95  Pct99   max
 Pb-TSPall           215    0    0.002   0.006  0.008  0.016  0.228  0.056   0.21   0.605 1.178  2.185  2.821
 Pb-TSP source       112  0.006  0.007   0.015  0.029  0.0765  0.402   0.16   0.509  1.124 1.623  2.438  2.821
 Pb-TSPprev source    12   0.008  0.008   0.008   0.03   0.0385   0.09  0.068  0.1225  0.14  0.271  0.271  0.271
 Pb-TSP not source     91     0      0     0.003  0.006   0.01   0.032  0.016  0.035  0.057 0.087  0.364  0.364
 Pb-PM10 urban        35   0.002  0.002   0.003  0.003  0.004   0.01  0.007  0.012  0.028 0.031  0.035  0.035
 Pb-PM2.5 urban CSN  198  0.001  0.002   0.003  0.003  0.004  0.011  0.007  0.011  0.021  0.033  0.078  0.199
 Pb-PM2.5 non-urban
 IMPROVE           107  0.001  0.001   0.001  0.001  0.001  0.002  0.001  0.002  0.003 0.006  0.01   0.012
January 2013                                 2D - 17                 Draft - Do not cite or quote

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-8.  Pb-TSP metrics (2009-2011).
Monitoring
SITE ID
011090003
020200051
021880001
040071002
040078000
060250005
060371103
060371302
060371402
060371403
060371404
060371405
060371406
060371602
060374002
060374004
060375005
060651003
060658001
060711004
060719004
080050007
080310025
120570100
120571066
120571073
130150003
130890003
132150010
132150011
170310001
170310022
170310026
170310052
170310110
170310210
170313103
170313301
170314201
170316003
171150110
CBSA_gt_1M
(2010 Census)






1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1


1
1
1
1
1
1
1
1
1
1

(3-yr) -
annual
mean
(ug/m3)
0.564


0.133
0.033
0.015
0.011
0.01
0.022
0.061
0.061
0.343
0.049
0.011
0.006
0.007
0.003
0.006
0.007
0.007
0.008
0.014
0.006
0.026
0.282
0.116
0.014
0.003

0.029
0.016
0.033
0.023
0.016
0.1
0.037
0.011
0.018
0.01
0.024
0.082
(3-yr) -
max 3-
month
avg
(ug/m3)
1.035


0.267
0.061
0.023
0.016
0.019
0.044
0.116
0.108
0.667
0.085
0.022
0.01
0.012
0.008
0.011
0.011
0.011
0.013
0.02
0.01
0.044
0.979
0.423
0.019
0.004

0.071
0.024
0.051
0.034
0.024
0.294
0.053
0.013
0.027
0.011
0.036
0.199
ratio of 3-yr max 3
month average to
3-year average
annual mean
1.8351


2.0075
1.8485
1.5333
1.4545
1.9000
2.0000
1.9016
1.7705
1.9446
1.7347
2.0000
1.6667
1.7143
2.6667
1.8333
1.5714
1.5714
1.6250
1.4286
1.6667
1.6923
3.4716
3.6466
1.3571
1.3333

2.4483
1.5000
1.5455
1.4783
1.5000
2.9400
1.4324
1.1818
1.5000
1.1000
1.5000
2.4268
Ratios for
CBSA_gt_1M
(2010 Census)






1.4545
1.9000
2.0000
1.9016
1.7705
1.9446
1.7347
2.0000
1.6667
1.7143
2.6667
1.8333
1.5714
1.5714
1.6250
1.4286
1.6667
1.6923
3.4716
3.6466
1.3571
1.3333


1.5000
1.5455
1.4783
1.5000
2.9400
1.4324
1.1818
1.5000
1.1000
1.5000

January 2013
2D-18
Draft - Do Not Quote or Cite

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-8.  Pb-TSP metrics (2009-2011).
Monitoring
SITE ID
171170002
171190010
171193007
171430037
171430110
171430210
171630010
171950110
172010110
180350008
180350009
180890023
180890032
180890033
180892008
180970063
180970076
180970078
181270023
181270027
181630006
181630020
191550011
191630015
201690004
210190016
211510003
212070001
220330014
220950003
250250002
260290011
260670002
260670003
260810020
261250013
261570001
261630001
261630033
270036020
270072303
CBSA_gt_1M
(2010 Census)
1
1
1



1




1
1
1
1
1
1
1
1










1







1
1
1

(3-yr) -
annual
mean
(ug/m3)
0.01
0.118
0.018
0.01
0.013
0.011
0.019
0.02
0.031
0.079
0.186
0.026
0.025
0.054
0.016
0.026
0.01
0.005
0.013


0.004
0.122
0.01
0.18
0.003
0.034
0.021
0.004
0.037



0.126
0.005


0.005
0.012
0.017
0
(3-yr) -
max 3-
month
avg
(ug/m3)
0.011
0.416
0.036
0.011
0.016
0.015
0.029
0.028
0.063
0.11
0.262
0.077
0.06
0.139
0.037
0.079
0.02
0.011
0.022


0.007
0.263
0.012
0.421
0.004
0.069
0.036
0.005
0.053



0.284
0.008


0.006
0.023
0.033
0.001
ratio of 3-yr max 3
month average to
3-year average
annual mean
1.1000
3.5254
2.0000
1.1000
1.2308
1.3636
1.5263
1.4000
2.0323
1.3924
1.4086
2.9615
2.4000
2.5741
2.3125
3.0385
2.0000
2.2000
1.6923


1.7500
2.1557
1.2000
2.3389
1.3333
2.0294
1.7143
1.2500
1.4324



2.2540
1.6000


1.2000
1.9167
1.9412
0.0000
Ratios for
CBSA_gt_1M
(2010 Census)
1.1000
3.5254
2.0000



1.5263




2.9615
2.4000
2.5741
2.3125
3.0385
2.0000
2.2000
1.6923










1.4324







1.2000
1.9167
1.9412

January 2013
2D-19
Draft - Do Not Quote or Cite

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-8.  Pb-TSP metrics (2009-2011).
Monitoring
SITE ID
270210001
270370020
270370423
270370442
270370443
270370465
270370470
270530963
270530966
270531007
270532006
271230866
271230871
271231003
271377001
271377555
271453053
271630009
271630438
271630446
290930016
290930021
290930027
290930029
290930032
290930033
290930034
290939007
290939008
290970005
290990004
290990005
290990009
290990011
290990013
290990015
290990020
290990021
290990022
290990023
290990024
CBSA_gt_1M
(2010 Census)

1
1
1
1
1
1
1
1
1
1
1
1
1



1
1
1










1
1
1
1
1
1
1
1
1
1
1
(3-yr) -
annual
mean
(ug/m3)
0
0.005
0.001
0.003
0.002
0.128
0.001
0.003
0.004
0.002
0.001
0.012
0.002
0.004
0.001
0.002
0.003
0.003
0.002
0
0.49
0.622
0.038
0.037

0.027
0.283
0.392
0.256
0.013
0.889
0.231
0.043
0.21
0.116
0.921
0.646
0.585
0.434
0.394
0.374
(3-yr) -
max 3-
month
avg
(ug/m3)
0.001
0.011
0.002
0.013
0.003
0.259
0.002
0.006
0.016
0.004
0.003
0.016
0.007
0.006
0.003
0.006
0.006
0.005
0.005
0.002
2.07
1.937
0.083
0.115

0.057
0.402
0.581
0.387
0.019
1.16
0.396
0.067
0.403
0.198
1.079
0.943
0.969
0.63
0.594
0.659
ratio of 3-yr max 3
month average to
3-year average
annual mean
0.0000
2.2000
2.0000
4.3333
1.5000
2.0234
2.0000
2.0000
4.0000
2.0000
3.0000
1.3333
3.5000
1.5000
3.0000
3.0000
2.0000
1.6667
2.5000
0.0000
4.2245
3.1141
2.1842
3.1081

2.1111
1.4205
1.4821
1.5117
1.4615
1.3048
1.7143
1.5581
1.9190
1.7069
1.1716
1.4598
1.6564
1.4516
1.5076
1.7620
Ratios for
CBSA_gt_1M
(2010 Census)

2.2000
2.0000
4.3333
1.5000
2.0234
2.0000
2.0000
4.0000
2.0000
3.0000
1.3333
3.5000
1.5000



1.6667
2.5000
0.0000










1.3048
1.7143
1.5581
1.9190
1.7069
1.1716
1.4598
1.6564
1.4516
1.5076
1.7620
January 2013
2D-20
Draft - Do Not Quote or Cite

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-8.  Pb-TSP metrics (2009-2011).
Monitoring
SITE ID
290990025
290990026
290990027
290999001
290999002
290999003
290999004
290999005
290999006
291790001
291790002
291790003
291790034
291870006
291870007
291892003
295100085
310530005
311270002
360470122
360713001
360713002
360713004
361030024
390170015
390290019
390290020
390290022
390350038
390350042
390350049
390350060
390350061
390350072
390490025
390510001
390910006
391010003
391137001
391510017
391550012
CBSA_gt_1M
(2010 Census)
1
1

1
1
1
1
1
1






1



1




1



1
1
1

1
1
1






(3-yr) -
annual
mean
(ug/m3)
0.043
0.035

0.824
0.232
0.198
0.119
0.959
0.049
0.033
0.033
0.027
0.068
0.05
0.027
0.007

0.051
0.042
0.014
0.031
0.128
0.005

0.006
0.022
0.015
0.02
0.013
0.014
0.194

0.015
0.014
0.008
0.073
0.004
0.035

0.014
0.007
(3-yr) -
max 3-
month
avg
(ug/m3)
0.085
0.072

1.204
0.431
0.467
0.261
1.432
0.091
0.059
0.068
0.1
0.093
0.132
0.041
0.041

0.072
0.115
0.019
0.101
1.027
0.012

0.009
0.057
0.025
0.044
0.021
0.03
0.531

0.023
0.035
0.011
0.178
0.006
0.056

0.023
0.011
ratio of 3-yr max 3
month average to
3-year average
annual mean
1.9767
2.0571

1.4612
1.8578
2.3586
2.1933
1.4932
1.8571
1.7879
2.0606
3.7037
1.3676
2.6400
1.5185
5.8571

1.4118
2.7381
1.3571
3.2581
8.0234
2.4000

1.5000
2.5909
1.6667
2.2000
1.6154
2.1429
2.7371

1.5333
2.5000
1.3750
2.4384
1.5000
1.6000

1.6429
1.5714
Ratios for
CBSA_gt_1M
(2010 Census)
1.9767
2.0571

1.4612
1.8578
2.3586
2.1933
1.4932
1.8571






5.8571



1.3571




1.5000



1.6154
2.1429
2.7371

1.5333
2.5000
1.3750






January 2013
2D-21
Draft - Do Not Quote or Cite

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-8.  Pb-TSP metrics (2009-2011).
Monitoring
SITE ID
391670008
391670010
401159006
401159007
401210416
401431127
410711702
420030002
420030008
420030070
420031009
420070006
420070007
420070505
420110005
420110020
420110021
420110022
420110717
420111717
420210808
420450002
420450004
420550002
420630005
420730011
420790036
421010449
421011002
421290007
421290009
450190003
450450015
450790007
450790019
470930023
470930027
470931017
471633001
471633002
471633003
CBSA_gt_1M
(2010 Census)






1
1
1
1
1
1
1
1







1
1




1
1
1
1










(3-yr) -
annual
mean
(ug/m3)
0.005
0.005
0.011
0.017
0.003
0.004
0.018
0.012
0.01
0.022
0.043
0.054
0.171
0.09
0.043
0.127
0.049
0.043
0.065
0.133
0.05
0.043
0.045
0.045
0.045
0.019
0.095
0.021
0.022
0.041
0.027
0.003
0.003
0.004
0.005
0.114
0.02
0.018
0.058
0.039
0.041
(3-yr) -
max 3-
month
avg
(ug/m3)
0.007
0.008
0.015
0.034
0.004
0.005
0.035
0.016
0.014
0.056
0.138
0.085
0.253
0.115
0.051
0.196
0.139
0.118
0.116
0.251
0.073
0.047
0.047
0.046
0.049
0.023
0.137
0.03
0.025
0.043
0.046
0.007
0.007
0.005
0.024
0.165
0.042
0.037
0.093
0.06
0.06
ratio of 3-yr max 3
month average to
3-year average
annual mean
1.4000
1.6000
1.3636
2.0000
1.3333
1.2500
1.9444
1.3333
1.4000
2.5455
3.2093
1.5741
1.4795
1.2778
1.1860
1.5433
2.8367
2.7442
1.7846
1.8872
1.4600
1.0930
1.0444
1.0222
1.0889
1.2105
1.4421
1.4286
1.1364
1.0488
1.7037
2.3333
2.3333
1.2500
4.8000
1.4474
2.1000
2.0556
1.6034
1.5385
1.4634
Ratios for
CBSA_gt_1M
(2010 Census)






1.9444
1.3333
1.4000
2.5455
3.2093
1.5741
1.4795
1.2778







1.0930
1.0444




1.4286
1.1364
1.0488
1.7037










January 2013
2D-22
Draft - Do Not Quote or Cite

-------
                  Appendix 2D - Air Quality Data Analysis Summary
Table 2D-8.  Pb-TSP metrics (2009-2011).
Monitoring
SITE ID
471633004
471634002
480610006
480850003
480850007
480850009
480850029
481130069
481410002
481410033
481410055
481410058
482011034
482011039
482570020
483750024
484790016
490351001
510090007
510270006
510870014
517700011
530330029
530610013
551170008
720130001
720210010
counts
CBSA_gt_1M
(2010 Census)



1
1
1
1





1

1


1


1






114
(3-yr) -
annual
mean
(ug/m3)
0.041
0.025
0.004
0.179
0.105
0.453
0.076

0.023
0.019
0.009
0.017
0.005

0.06
0.006
0.016
0.023
0.004
0.01
0.004
0.038


0.055
0.188
0.005
212
(3-yr) -
max 3-
month
avg
(ug/m3)
0.08
0.038
0.005
0.371
0.199
0.774
0.18

0.04
0.039
0.014
0.033
0.008

0.104
0.013
0.026
0.057
0.018
0.013
0.011
0.109


0.152
0.339
0.011
212
average — >
ratio of 3-yr max 3
month average to
3-year average
annual mean
1.9512
1.5200
1.2500
2.0726
1.8952
1.7086
2.3684

1.7391
2.0526
1.5556
1.9412
1.6000

1.7333
2.1667
1.6250
2.4783
4.5000
1.3000
2.7500
2.8684


2.7636
1.8032
2.2000

1.9463
Ratios for
CBSA_gt_1M
(2010 Census)



2.0726
1.8952
1.7086
2.3684





1.6000

1.7333


2.4783


2.7500







1.9203
January 2013
2D-23
Draft - Do Not Quote or Cite

-------
 1                                             APPENDIX 3A

 2               INTERPOLATED RISK ESTIMATES FOR THE GENERALIZED
 3                                  (LOCAL) URBAN CASE STUDY

 4           This Appendix describes the method used to develop risk estimates for conditions just
 5    meeting the current standard (0.15 |ig/m3, as a maximum 3-month average) for the generalized
 6    (local) urban case study.  These risk estimates were developed by interpolation from the 2007
 7    REA results for this case study. The general approach was to identify the two alternative
 8    standard scenarios simulated in the 2007 REA which represented air quality conditions
 9    bracketing those for the current standard and then linearly interpolate an estimate of risk for the
10    current standard based on the slope created  from the two bracketing estimates. In representing
11    air quality conditions for these purposes, we focused on the annual average air Pb concentration
12    estimates as that is the metric which had been the IEUBK model inputs for the various air quality
13    scenarios (IEUBK does not accept air quality inputs of a temporal scale shorter than a year).1  An
14    annual average concentration estimate to represent the current standard was identified in a
15    manner consistent with that employed in the 2007 REA for this case study (see section 3.4.3.2
16    above, use of 2003-2005  data) with the use  of currently available monitoring data (2009-2011)
17    for relationships between air quality metrics for representation of the current standard.  By this
18    method, the air quality scenario for the current standard (0.15 |ig/m3, as a not-to-be-exceeded 3-
19    month average) was found to be bracketed by the scenarios for alternative standards of 0.5 and
20    0.20 |ig/m3 (maximum monthly averages).  A risk estimate for the current standard was then
21    derived using the slope relating generalized (local) urban case study IQ loss to the annual
22    average Pb concentration used for those two air quality scenarios. We used this interpolation
23    approach to develop median risk estimates for the current standard based on each of the four C-R
24    functions. Details on the method  for the interpolation approach are provided below.
25        1.  Identify an estimate of annual average air Pb concentration to represent each air quality
26           scenario. For the alternative scenarios, this was done in the 2007 REA using the 2003-
27           2005 Pb-TSP  dataset for urban areas of population greater than one million. For analysis
             1 Although many different patterns of temporally varying air concentration will just meet a given potential
      alternative standard, the shortest time step accommodated by the blood Pb model is a year. Thus, the air Pb
      concentration inputs to the blood Pb model for each air quality scenario are annual average air Pb concentrations.
      For the generalized (local) urban case study, the national Pb-TSP monitoring dataset was analyzed to characterize
      the distribution of site-specific relationships between metrics reflecting the averaging time and form for the air
      quality scenarios being assessed (Table 3-8 of this document) and the annual average. The IEUBK annual average
      input was then derived by multiplying the level for a given air quality scenario by the ratio for the averaging time
      and form for that air quality scenario. For the location-specific case studies, however, the full temporally varying air
      Pb concentration dataset for each exposure zone was used to derive the average annual concentration for the IEUBK
      input.
      January 2013                          3 A-l            Draft - Do Not Quote or Cite

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14

15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
             of the current standard here, this uses the recent Pb-TSP dataset (2009-2011) for urban
             areas of population greater than one million.  The concentration in terms of the metric
             being assessed (e.g., maximum 3-month or calendar quarter average) was derived per
             monitoring site, as was the annual average concentration and also the ratio of the two
             (2007 REA, Appendix A; Appendix 2D of this document). From the average of the
             monitor-specific ratios we derived the annual average estimate. [In the 2007 REA, this
             was the IEUBK air quality input for each AQ scenario.] With the 2009-2011 data, the
             average, across monitors in urban areas greater than 1 million population, of the ratio of
             maximum 3-month average to annual average is 1.92 (Appendix 2D in this document).
                       This ratio was used to derive an annual average air Pb-TSP concentration
                       estimate to represent the current standard scenario (annual value = 0.15 ug/m3
                       * 1/1.92  = 0.078 ug/m3).  The annual average values for the scenarios
                       included in the 2007 REA and also for the current standard scenario
                       considered in this interpolation are shown in Table 3A-1 below.

      Table 3A-1. Annual average air Pb-TSP concentration estimates for different air quality
                 scenarios.
Air Quality Scenarios
Maximum
Quarterly Average D
(ug/m3)
1.5 (previous NAAQS)

0.2




Maximum
Monthly Average
(ug/m3)

0.5


0.2
0.05
0.02
Maximum
3-month Average
(ug/m3)



0.15



Annual Average Estimate
(ug/m3)
0.60
0.130
0.08
0.078*
0.05
0.013
0.005
* Derived as described in step 1 using 2009-201 1 air quality dataset.
         3.
Identify the two "bounding" alternative standard levels that will be used to derive a
slope for the risk interpolation for the current standard level. Based on comparison of
the annual ambient air Pb estimates for each of the 2007 REA air quality scenarios and
for the current standard we determine which scenarios "bound" the current standard (in
terms of the annual average ambient Pb estimate).  As Table 3A-1 above shows, the
bounding scenarios are the scenarios  for just meeting maximum monthly average
concentrations of 0.5 and 0.2 ug/m3.

Calculate the slopes of generalized (local) urban case study IQ loss per unit annual
average Pb estimate for each risk estimate of interest for the two bounding scenarios.
This calculation is (IQ Lossscenario x - IQ Lossscenario Y)/ (annual averagescenario x - annual
averagescenario Y). The risk estimates for the  4 different C-R functions and two exposure
pathway categories of interest (Recent Air and Recent + Past Air), with the derived
slopes, are in Table 3A-2.
     January 2013
                                           3A-2
                                               Draft - Do Not Quote or Cite

-------
 1   Table 3A-2. Risk estimates for bounding air quality scenarios and associated slopes.


0.2 jig/m3, max
quarterly average
0.2 jig/m3 max
monthly average
Slope *
Risk Estimates
for different C-R functions and exposure pathway categories
Log-linear with
low-exposure
linearization
Recent
air
1.53
1.21
10.69
Recent +
past air
3.37
3.16
7.03
Dual linear -
stratified
atlOug/dLpeak
Recent
air
0.54
0.41
4.10
Recent +
past air
1.18
1.08
3.38
Log-linear with
cutpoint
Recent
air
0.63
0.47
5.31
Recent +
past air
1.38
1.22
5.28
Dual linear -
stratified
at 7.5 ug/dL peak
Recent
air
1.97
1.53
14.77
Recent +
past air
4.34
3.99
11.66
* In terms of points IQ loss per unit annual average air Pb in generalized (local) urban case study (see texts for explanation).
 3
 4
 5
 6
 7
10
11
12
13
14
4.  Derive interpolated risk estimates for the current standard scenario. Using the
   slopes presented in Table 3 A-2 and the lower bounding air quality scenario risk
   estimates, derive corresponding risk estimates for the current standard scenario:
     Risk = lower bounding risk + (increment annual average Pb * slope)
     Where:
         •   lower bounding risk = risk for the lower bound air quality scenario (i.e.,
            for 0.2 u.g/m3 maximum monthly average)
         •   increment annual average Pb  is the difference between the annual average
            estimates for the current standard and the lower bounding scenario.
         •   slope is the value described in Step 4, which differs across the 8
            combinations of C-R  functions and exposure pathway categories.
    The resulting interpolated risk estimates  are presented in Table 3A-3.
15   Table 3A-3. Interpolated risk estimates for the current NAAQS scenario for the
16               generalized (local) urban case study.


Interpolated
Estimate*
Risk Estimates
for different C-R functions and exposure pathway categories
Log-linear with
low-exposure
linearization
Recent
air
1.51
Recent +
past air
3.36
Dual linear -
stratified
atlOug/dLpeak
Recent
air
0.53
Recent +
past air
1.17
Log-linear with
cutpoint
Recent
air
0.62
Recent +
past air
1.37
Dual linear -
stratified
at 7.5 ug/dL peak
Recent
air
1.94
Recent +
past air
4.32
* Points IQ loss in generalized (local) urban case study (see text for explanation).
17
     January 2013
                              3A-3
Draft - Do Not Quote or Cite

-------
1

2
                    APPENDIX 5A

ADDITIONAL DETAIL ON 2006 ECOLOGICAL SCREENING
                    ASSESSMENT





Primary
Smelter Case
C4-I i/Jif
otuay




Seconday
Smelter Case
Study



Near Roadway
Non-l Irhan
IIWI 1 Ul UQI 1
Case Study






Vulnerable
Ecosystem
Case Study





National
Surface Water
f\
Screen

Setting and spatial
extent of dataset or
modeling analysis
Herculaneum,
Missouri: soil and
waterbody samples
from study area 6 km
diameter, centered on
point source

Troy, Alabama: soil
mnrpntratioir; in
\J\JI IWOI 1 LI QLIV-M IO II 1
census blocks near
facility predicted from
dispersion and soil
mixing model
\3

Two datasets: Corpus
Christi, Texas (within 4
m from road), and
Atlee, Virginia (within 2
to 30 m distance from
road)



Hiihh^rrl Rrnnk
1 IUUUOIU LJIVJVJiX
Experimental Forest,
New Hampshire: forest
in oblong basin about 8
km long by 5 km wide



Surface water bodies in
the 47 basin study
3
units from all regions of
the United States,
covering approx. 50%
of U.S. land base
Media Screened and Screening Levels Used

Soil


Soil screening
values
developed
hocpH on 1 1 R
L/QOOVJ \JI 1 \Ji\Ji
EPA
i — i r\
Superfund
methodology
for developing
ecological soil
screening
levels
(USEPA,
2005a,b)



Freshwater
water column
U.S. EPA
freshwater AWQC
for aquatic life
adjusted for site-
specific water
hardness




NA





Freshwater
Sediment
Sediment screening
values based on
MacDonald et al.
(2000) sediment quality
assessment guidelines





NA





While no quantitative analyses were performed, summary
review of literature search indicated: (1) atmospheric Pb inputs
do not directly affect stream Pb levels at HBEF because
deposited Pb is almost entirely retained in the soil profile; (2)
soil horizon analysis results show Pb has become more
concentrated at lower depths over time and that the soil profile
serves as a Pb sink, appreciably reducing Pb in porewater as it
moves through the soil layers to streams, (3) dissolved Pb
concentrations were reduced (5 ppb to about 5 ppt) as Pb
moves from the Oa horizon to streams. Studies concluded
insignificant contribution of dissolved Pb from soils to streams
(less than 0.2 g 'ha~1 'yf 1). (IGF, 2006, Appendix E)


NA


U.S. EPA
freshwater AWQC
for aquatic life
adjusted for
hardness at site or
nearby water body
Sediment screening
values based on
MacDonald et al.
(2000) sediment quality
assessment
guidelines
AWQC= Ambient water quality criteria. NA = Not applicable; medium not part of case study.
NOTE: Information here is drawn from IGF, 2006 and EPA, 2007b.
      January 2013
                   5A- 1
Draft - Do Not Quote or Cite

-------
O ^H
o o
Si |
§ i
 11~>
fin
 O
 O
3
£
II

      ao
      o

      (L)
 (L>

   O
   •-P
   O
   (L)
 M jj
 (L) C
•g S

^ I >,
' (J Q fj
^ .b c
.ti > (u
 e c oo
^ W 
-------