&EPA
United State
EirviroiiwiU Protection
Agnncy
Welfare Risk and Exposure Assessment
for Ozone
First External Review Draft
-------
DISCLAIMER
This preliminary draft document has been prepared by staff from the Ambient Standards Group,
Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency. Any
opinions, findings, conclusions, or recommendations are those of the authors and do not
necessarily reflect the views of the EPA. This document is being circulated for informational
purposes and to facilitate discussion with the Clean Air Scientific Advisory Committee
(CASAC) on the overall structure, areas of focus, and level of detail to be included in an external
review draft Policy Assessment, which EPA plans to release for CASAC review and public
comment later this year. Questions related to this preliminary draft document should be
addressed to Travis Smith, U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, C539-07, Research Triangle Park, North Carolina 27711 (email:
smith.j travi s@epa. gov).
-------
EP A 452/P-12-004
July 2012
Welfare Risk and Exposure Assessment for Ozone
First External Review Draft
U.S. Environmental Protection Agency
Office of Air and Radiation
Office of Air Quality Planning and Standards
Health and Environmental Impacts Division
Risk and Benefits Group
Research Triangle Park, North Carolina 27711
-------
This page left intentionally blank.
-------
TABLE OF CONTENTS
Table of Contents i
List of Acronyms/Abbreviations iv
1 Introduction 1-1
1.1 History 1-3
1.2 Current Risk and Exposure Assessment: Goals and Planned Approach 1-6
1.3 Organization of Document 1-6
2 Conceptual Framework 2-1
2.1 O3 Chemistry 2-2
2.2 Sources of 63 and Os Precursors 2-4
2.3 Ecological Effects 2-5
2.4 Ecosystem Services 2-8
2.5 Conclusions 2-15
3 Scope 3-1
3.1 Overview of Exposure and risk Assessment from Last Review 3-2
3.1.1 Exposure Characterization 3-2
3.1.2 Assessment of Risks to Vegetation 3-4
3.2 Overview of Current Assessment Plan 3-6
3.2.1 Air Quality Considerations 3-8
3.2.2 National Os Exposure Surface 3-8
3.3 Ecological Effects of Exposure 3-10
3.3.1 National Scale Assessment 3-10
3.3.2 Case Study Areas 3-11
3.4 Ecosystem Services Evaluation 3-12
3.4.1 National Scale Assessment 3-13
3.4.2 Case Study Analysis 3-14
3.5 Uncertainty and Variability 3-15
4 Air Quality Considerations 4-1
4.1 Introduction 4-1
4.2 Overview of Os Monitoring and Air Quality 4-1
-------
4.3 Overview of Air Quality Inputs to Risk and Exposure Assessments 4-3
4.3.1 Recent Air Quality 4-3
4.3.2 Air Quality After simulating "Just Meeting" Current O3 Standard 4-6
5 Ecological Effects 5-1
5.1 Introduction 5-1
5.2 Relative Biomass Loss 5-2
5.2.1 Species Level Analysis 5-5
5.2.2 Abundance Weighted Relative Biomass Loss 5-14
5.2.3 Relative Biomass Loss in Federally Designated Areas 5-21
5.2.4 National Park Case Study Areas 5-29
5.3 Visible Foliar Injury 5-36
5.3.1 National-Scale Analysis of Foliar Injury 5-37
5.3.2 Updated Assessment of Risk of Visible Foliar Injury in National Parks 5-38
5.3.3 National Park Case Study Areas 5-46
5.4 Discussion 5-47
6 Ozone Risk to Ecosystem Services 6-1
6.1 Introduction 6-1
6.2 National Scale Ecosystem Serices Assessment 6-1
6.2.1 Supporting Services 6-5
6.2.2 Regulating Services 6-6
6.2.3 Provisioning Services 6-10
6.2.4 Cultural Services 6-18
6.3 Case Study Analysis 6-27
6.3.1 Southeast Region - Great Smoky Mountains National Park 6-28
6.3.2 Intermountain Region -Rocky Mountain National Park 6-33
6.3.3 Pacific West Region - Sequioa/Kings Canyon National Parks 6-36
6.3.4 Urban Case Study 6-38
6.4 Discussion 6-39
7 Synthesis 7-1
7.1 Summary of Key Results of Biomass Loss Risk Assessment 7-1
7.2 Summary of Key Results of Foliar Injury Risk Assessment 7-2
-------
7.3 Summary of Key Results for Ecosystem Services Risk Assessment 7-3
7.4 Observations 7-5
8 References 8-1
in
-------
LIST OF ACRONYMS/ABBREVIATIONS
AGSIM
AQCD
AQS
CAA
CALFIRE
CASAC
CASTNET
CH4
CMAQ
C02
C-R
CSTR
EGU
EPA
ESRI
FACA
FACE
FASOM
FASOMGHG
FHWAR
FIA
GIS
GSMNP
HNO3
IPCC
IRP
ISA
IV
Agriculture Simulation Model
Air Quality Criteria Document
Air Quality System
Clean Air Act
California Department of Forestry and Fire Protection
Clean Air Science Advisory Committee
Clean Air Status and Trends Network
methane
Community Multi-scale Air Quality
carbon dioxide
Concentration Response Function
continuous stirred tank reactors
electric generating unit
U.S. Environmental Protection Agency
Environmental Systems Research Institute, Inc.
Federal Advisory Committee Act
Free Air CC>2 enrichment
Forest and Agricultural Sector Optimization Model
Forest and Agriculture Sectors Optimization Model - Greenhouse
Gas version
Fishing, Hunting, and Wildlife-Associated Recreation
U.S. Forest Service Forest Inventory and Analysis
geographic information system
Great Smoky Mountains National Park
nitric acid
Intergovernmental Panel on Climate Change
Integrated Review Plan
Integrated Science Assessment
Importance Value
July 2012
IV
DRAFT - Do Not Quote or Cite
-------
MATS
MEA
MT
NAAQS
NCDC
NCLAN
NCore
NEI
NHEERL-WED
NO2
NOAA
NOX
NPP
NFS
NSRE
NTFP
03
OAQPS
OH
OIF
OTC
PA
ppm
RBL
REA
RMNP
SAB
SKCNP
STE
USDA
Modeled Attainment Test Software
Millennium Ecosystem Assessment
metric ton
National Ambient Air Quality Standards
National Climatic Data Center
National Crop Loss Assessment Network
National Core
National Emissions Inventory
National Health and Environmental Effects Research Laboratory,
Western Ecology Division
nitrite
National Oceanic and Atmospheric Administration
nitrogen oxides
net primary productivity
National Park Service
National Survey on Recreation and the Environment
Non-timber forest products
Ozone
Office of Air Quality Planning and Standards
hydroxide
Outdoor Industry Foundation
open top chamber
Policy Assessment
parts per million
Relative Biomass Loss
Risk and Exposure Assessment
Rocky Mountain National Park
Science Advisory Board
Sequoia/Kings Canyon National Park
stratosphere-troposphere exchange
U.S. Department of Agriculture
July 2012
v
DRAFT - Do Not Quote or Cite
-------
USFS U.S. Forest Service
USGS U.S. Geological Survey
VNA Voronoi Neighbor Averaging
VOC volatile organic carbon
WTA willingness to accept
WTP willingness to pay
July 2012 vi DRAFT - Do Not Quote or Cite
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-452/P-12-004
Environmental Protection Health and Environmental Impacts Division July 2012
Agency Research Triangle Park, NC
-------
1 1 INTRODUCTION
2 The U.S. Environmental Protection Agency (EPA) is presently conducting a review of
3 the national ambient air quality standards (NAAQS) for ozone (63) and related photochemical
4 oxidants. An overview of the approach to reviewing the Oi NAAQS is presented in the
5 Integrated Review Plan for the Ozone National Ambient Air Quality Standards (IRP, US EPA,
6 201 la). The IRP discusses the schedule for the review; the approaches to be taken in developing
7 key scientific, technical, and policy documents; and the key policy-relevant issues that will frame
8 our consideration of whether the current NAAQS for Os should be retained or revised.
9 Sections 108 and 109 of the Clean Air Act (CAA) govern the establishment and periodic
10 review of the NAAQS. These standards are established for pollutants that may reasonably be
11 anticipated to endanger public health and welfare, and whose presence in the ambient air results
12 from numerous or diverse mobile or stationary sources. The NAAQS are to be based on air
13 quality criteria, which are to accurately reflect the latest scientific knowledge useful in indicating
14 the kind and extent of identifiable effects on public health or welfare that may be expected from
15 the presence of the pollutant in ambient air. The EPA Administrator is to promulgate and
16 periodically review, at five-year intervals, "primary" (health-based) and "secondary" (welfare-
17 based) NAAQS for such pollutants. Based on periodic reviews of the air quality criteria and
18 standards, the Administrator is to make revisions in the criteria and standards, and promulgate
19 any new standards, as may be appropriate. The Act also requires that an independent scientific
20 review committee advise the Administrator as part of this NAAQS review process, a function
21 performed by the Clean Air Scientific Advisory Committee (CASAC).l
22 The current primary NAAQS for Os is set at a level of 0.075 ppm, based on the annual
23 fourth-highest daily maximum 8-hr average concentration, averaged over three years, and the
24 secondary standard is identical to the primary standard (73 FR 16436). The EPA initiated the
1 The Clean Air Scientific Advisory Committee (CAS AC) was established under section 109(d)(2) of the
Clean Air Act (CAA) (42 U.S.C. 7409) as an independent scientific advisory committee. CAS AC provides advice,
information and recommendations on the scientific and technical aspects of air quality criteria and NAAQS under
sections 108 and 109 of the CAA. The CAS AC is a Federal advisory committee chartered under the Federal
Advisory Committee Act (FACA). See
http://vosemite.epa.gov/sab/sabpeople.nsf/WebCommitteesSubcommittees/CASAC%20Particulate%20Mattei%20R
eview%20Panel for a list of the CAS AC PM Panel members and current advisory activities.
1-1
-------
1 current review of the ozone NAAQS on September 29, 2008 with an announcement of the
2 development of an ozone Integrated Science Assessment and a public workshop to discuss
3 policy-relevant science to inform EPA's integrated plan for the review of the ozone NAAQS (73
4 FR 56581). The NAAQS review process includes four key phases: planning, science
5 assessment, risk/exposure assessment, and policy assessment/rulemaking.2 A workshop was
6 held on October 29-30, 2008 to discuss policy-relevant scientific and technical information to
7 inform EPA's planning for the ozone NAAQS review. Following the workshop, EPA developed
8 a planning document, the Integrated Review Plan for the Ozone National Ambient Air Quality
9 Standards (IRP; US EPA, 201 la), which outlined the key policy-relevant issues that frame this
10 review, the process and schedule for the review, and descriptions of the purpose, contents, and
11 approach for developing the other key documents for this review.3 In June 2012, EPA
12 completed the third draft of the ozone ISA, assessing the latest available policy-relevant
13 scientific information to inform the review of the Os standards. The Integrated Science
14 Assessment for Ozone and Related Photochemical Oxidants - Third External Review Draft (ISA;
15 US EPA, 2012), includes an evaluation of the scientific evidence on the welfare effects of Os,
16 including information on exposure, physiological mechanisms by which Os might adversely
17 impact vegetation, and an evaluation of the ecological evidence including information on
18 reported concentration-response (C-R) relationships for Os-related changes in plant biomass.
19 The EPA's Office of Air Quality Planning and Standards (OAQPS) has developed this
20 quantitative welfare risk and exposure assessment (REA) describing the quantitative assessments
21 of exposure to Os and Os-related risks to public welfare to support the review of the secondary
22 Os standards. This document is a concise presentation of the conceptual model, scope, methods,
23 key results, observations, and related uncertainties associated with the quantitative analyses
24 performed. The REA builds upon the welfare effects evidence presented and assessed in the
25 ISA, as well as CAS AC advice (Samet, 2011) and public comments on a scope and methods
26 planning document for the REA (here after, "Scope and Methods Plan", US EPA, 201 Ib).
For more information on the NAAQS review process see http://www.epa.gov/ttn/naaqs/review.html.
3 On March 30, 2009, EPA held a public consultation with the CASAC Ozone Panel on the draft IRP. The
final IRP took into consideration comments received from CASAC and the public on the draft plan as well as input
from senior Agency managers.
1-2
-------
1 Revisions to this draft RA will draw upon the final ISA and will reflect consideration of CASAC
2 and public comments on this draft.
3 The ISA and REA will inform the policy assessment and rulemaking steps that will lead
4 to final decisions on the primary Os NAAQS, as described in the Integrated Review Plan for the
5 Ozone National Ambient Air Quality Standards. The policy assessment will include staff
6 analysis of the scientific basis for alternative policy options for consideration by senior EPA
7 management prior to rulemaking. The PA integrates and interprets information from the ISA
8 and the REA to frame policy options for consideration by the Administrator. The PA is intended
9 to link the Agency's scientific and technical assessments, presented in the ISA and REA, to
10 judgments required of the Administrator in determining whether it is appropriate to retain or
11 revise the current 63 standards. Development of the PA is also intended to facilitate elicitation
12 of CASAC's advice to the Agency and recommendations on any new standards or revisions to
13 existing standards as may be appropriate, as provided for in the Clean Air Act (CAA). The first
14 draft PA is planned for release around the middle of August 2012 for review by the CASAC 63
15 Panel and the public concurrently with their review of this first draft REA September 11-13,
16 2012.
17 1.1 HISTORY
18 As part of the last O3 NAAQS review completed in 2008, EPA's OAQPS conducted
19 quantitative risk and exposure assessments to estimate risks to human welfare based on
20 ecological effects associated with exposure to ambient O3 (U.S. EPA 2007a, U.S. EPA 2007b).
21 The assessment scope and methodology were developed with considerable input from CASAC
22 and the public, with CASAC generally concluding that the exposure assessment reflected
23 generally accepted modeling approaches, and that the risk assessments were well done, balanced
24 and reasonably communicated (Henderson, 2006a). The final quantitative risk and exposure
25 assessments took into consideration CASAC advice (Henderson, 2006a; Henderson, 2006b) and
26 public comments on two drafts of the risk and exposure assessments.
27 The assessments conducted as part of the last review focused on national-level Cb-related
28 impacts to sensitive vegetation and their associated ecosystems. The vegetation exposure
29 assessment was performed using an interpolation approach that included information from
30 ambient monitoring networks and results from air quality modeling. The vegetation risk
1-3
-------
1 assessment included both tree and crop analyses. The tree risk analysis included three distinct
2 lines of evidence: (1) observations of visible foliar injury in the field linked to monitored O^ air
3 quality for the years 2001 - 2004; (2) estimates of seedling growth loss under then current and
4 alternative 63 exposure conditions; and (3) simulated mature tree growth reductions using the
5 TREGRO model to simulate the effect of meeting alternative air quality standards on the
6 predicted annual growth of mature trees from three different species. The crop risk analysis
7 included estimates of crop yields under current and alternative Os exposure conditions. The
8 associated changes in economic value upon meeting the levels of various alternative standards
9 were analyzed using an agricultural sector economic model. Key observations and insights from
10 the ozone risk assessment, in addition to important caveats and limitations, were addressed in
11 Section II.B of the Final Rule notice (73 FR 16440 to 16443, March 27, 2008).
12 Prior to the issuance of a proposed rulemaking in the last review, CASAC presented
13 recommendations to the Administrator supporting revisions of the 63 secondary standard. These
14 recommendations cited the results of the quantitative risk assessment in recommending a range
15 of ozone levels below the existing standard at the time (0.084 ppm) (Henderson, 2006a). In the
16 2008 final rule, the EPA Administrator considered the results of the exposure and risk
17 assessments and the potential magnitude of the risk to human welfare given recent air quality
18 data and air quality simulated to meet the current standard and alternative standards. The EPA
19 proposed to revise the level of the primary standard to a level within the range of 0.075 to 0.070
20 ppm. Two options were proposed for the secondary standard: (1) replacing the current standard
21 with a cumulative, seasonal standard, expressed as an index of the annual sum of weighted
22 hourly concentrations cumulated over 12 daylight hours during the consecutive 3-month period
23 within the Os season with the maximum index value (W126), set at a level within the range of 7
24 to 21 ppm-hrs, and (2) setting the secondary standard identical to the revised primary standard.
25 The EPA completed the review with publication of a final decision on March 27, 2008 (73 FR
26 16436), revising the level of the 8-hour primary O^ standard from 0.08 ppm to 0.075 ppm and
27 revising the secondary standard to be identical to the revised primary standard.
28 In May 2008, state, public health, environmental, and industry petitioners filed suit
29 against EPA regarding the 2008 final decision on the O3 NAAQS, and on December 23, 2008,
30 the Court set a briefing schedule in the consolidated cases. On March 10, 2009, EPA requested
31 that the Court vacate the briefing schedule and hold the consolidated cases in abeyance. This
1-4
-------
1 request for extension was made to allow time for appropriate EPA officials appointed by the new
2 Administration to review the Os NAAQS to determine whether the standards established in the
3 March 2008 Os NAAQS decision should be maintained, modified or otherwise reconsidered. In
4 granting EPA's request, the Court directed EPA to notify the Court by September 16, 2009 of the
5 action it will be taking with respect to the 2008 Os NAAQS rule and the Agency's schedule for
6 undertaking such action. The EPA notified the Court on September 16, 2009 of its decision to
7 reconsider the primary and secondary Os NAAQS set in March 2008 to ensure they are
8 scientifically sound and protective of public health and the environment.
9 In 2010 the Administrator proposed to reconsider and revise parts of that 2008 final rule.
10 Specifically, she proposed to revise the level of the primary standard to within the range of 0.060
11 to 0.070 ppm and she proposed to revise the secondary standard by setting a new cumulative,
12 seasonal standard in terms of the W126 metric, set within the range of 7-15 ppm-hours (FR 75
13 2938). This proposal was based on the scientific and technical record from the 2008 rulemaking,
14 including public comments and CASAC advice and recommendations. The information that was
15 assessed during the 2008 rulemaking included information in the 2006 Criteria Document (EPA,
16 2006a), the 2007 Policy Assessment of Scientific and Technical Information, referred to as the
17 2007 Staff Paper (EPA, 2007a), and related technical support documents including the 2007
18 REAs (U.S. EPA, 2007b; Abt Associates, 2007a,b).4 Scientific and technical information
19 developed since the 2006 Criteria Document was not considered in the 2010 proposal.
20 On September 2, 2011, the President requested that EPA withdraw the proposal to revisit
21 and revise the 2008 Ozone National Ambient Air Quality Standards, noting that work was
22 already underway on the next review (memo from President Obama,
23 http://www.whitehouse.gov/the-press-office/2011/09/02/statement-president-ozone-national-
24 ambient-air-quality-standards).5 The proposed changes to the 2008 O3 NAAQS were not
25 finalized.
4The EPA's Office of Research and Development/National Center for Environmental Assessment
(ORD/NCEA) also conducted a provisional assessment of pertinent studies investigating the health and ecological
effects of O3 that were published after the cutoff for inclusion in the 2006 O3 Criteria Document. The provisional
assessment was conducted for the purpose of determining if any recent studies would materially change the
conclusions of the 2006 O3 Criteria Document. The provisional assessment concluded that, taken in context, results
of more recent studies did not materially change any of the broad scientific conclusions regarding the health and
ecological effects of O3 exposure made in the 2006 O3 Criteria Document. Thus, as stated above, the 2010 proposal
was based solely on the record from the 2008 rulemaking and did not consider scientific and technical information
developed since the 2006 Criteria Document.
5 Also see letter from Cass Sunstein, Administrator of the Office of Information and Regulatory Affairs, to
EPA Administrator Lisa Jackson
(http://www.whitehouse.gov/sites/default/files/ozone national ambient air quality standards letter.pdf).
1-5
-------
1 1.2 CURRENT RISK ASSESSMENT: GOALS AND PLANNED APPROACH
2 The goals of the current quantitative welfare risk assessments are (1) to provide estimates
3 of the ecological effects of O3 exposure across a range of environments; (2) to provide
4 estimates of ecological effects within selected case study areas; (3) to provide estimates of the
5 effects of O3 exposure on specific urban and non-urban ecosystem services based on the causal
6 ecological effects; and (4) to develop a better understanding of the response of ecological
7 systems and ecosystem services to changing levels of O3 exposure to inform the PA regarding
8 alternative standards that might be considered. This current quantitative risk and exposure
9 assessment builds on the approach used and lessons learned in the last 63 risk assessment and
10 focuses on improving the characterization of the overall confidence in the risk estimates,
11 including related uncertainties, by incorporating a number of enhancements, in terms of both the
12 methods and data used in the analyses. This assessment considers a variety of welfare endpoints
13 for which, in staff s judgment, there is adequate information to develop quantitative risk
14 estimates that can meaningfully inform the review of the secondary O?, NAAQS.
15 This first draft REA provides an assessment of exposure and risk associated with recent
16 ambient levels of ozone and ozone air quality simulated to just meet the current primary ozone
17 standards. Subsequent drafts of the REA will evaluate potential alternative ozone standards
18 based on recommendations provided in the first draft of the Policy Assessment.
19 1.3 ORGANIZATION OF DOCUMENT
20 The remainder of this document is organized as follows. Chapter 2 provides a conceptual
21 framework for the risk and exposure assessment, including discussions of ozone chemistry,
22 sources of ozone precursors, ecological exposure pathways and uptake into plants, ecological
23 effects, and ecosystem services endpoints associated with ozone. This conceptual framework
24 sets the stage for the scope of the risk and exposure assessments. Chapter 3 provides an
25 overview of the scope of the quantitative risk and exposure assessments, including a summary of
26 the previous risk and exposure assessments, and an overview of the current risk and exposure
27 assessments. Chapter 4 discusses air quality considerations relevant to the exposure and risk
28 assessments, including available ozone monitoring data, and important inputs to the risk and
29 exposure assessments. Chapter 5 describes the ecological effects of O3 exposure and includes
30 quantitative analyses of vegetation biomass loss and foliar injury. Chapter 6 describes the
1-6
-------
1 ecosystem services affected by the ecological effects analyzed in Chapter 5. Chapter 6 includes
2 both quantitative assessments of the effects on ecosystem services as well as qualitative
3 discussion of services for which effects are known to occur, but quantitative analyses were not
4 possible. Chapter 7 provides an integrative discussion of the risk estimates generated in the
5 analyses drawing on the results of the analyses based on quantitative analysis and incorporating
6 considerations from the qualitative discussion of ecosystem services.
1-7
-------
1 2 CONCEPTUAL FRAMEWORK
2 In this chapter, we summarize the conceptual framework for assessing exposures of
3 ecosystems to 63 and the associated risks to public welfare. This conceptual framework includes
4 elements related to characterization of ambient 63 and its relation to ecosystem, exposures
5 (Section 2.1), important sources of Os precursors including oxides of nitrogen (NOX) and volatile
6 organic compounds (VOC) (Section 2.2), ecological effects occurring in Os sensitive ecosystems
7 (Section 2.3), and ecosystem services that are likely to be negatively impacted by changes in
8 ecological functions resulting from O^ exposures (Section 2.4). The chapter concludes with key
9 observations relevant for developing the scope of the quantitative risk and exposure assessments.
10 In the previous review of the secondary standards, the focus of the ecological risk
11 assessment was on estimation of changes in biomass loss and resulting impacts on forest and
12 agricultural yields as well as qualitative consideration of effects on ecosystem services. In this
13 review, EPA is expanding the analysis to consider the broader array of impacts on ecosystem
14 services resulting from known effects of ozone on ecosystem functions. This is to address the
15 objective of this risk assessment to quantify the risks not just to ecosystems but to the aspects of
16 public welfare dependent on those ecosystems. EPA has begun using an ecosystem services
17 framework to help inform determinations of the adversity to public welfare associated with
18 changes in ecosystem functions (Rea et al, 2012). The Risk and Exposure Assessment
19 conducted as part of the Review of the Secondary National Ambient Air Quality Standards for
20 Oxides of Nitrogen and Oxides of Sulfur (U.S. EPA, 2009) presents detailed discussions of how
21 ecosystem services and public welfare are related and how an ecosystem services framework
22 may be employed to evaluate effects on welfare. In this risk assessment we will identify the
23 ecosystem services associated with the ecological effects caused by Os exposure for the national
24 scale assessment and the more refined case study areas. These services may be characterized as:
25 supporting services that are necessary for all other services (e.g., primary production); cultural
26 services including existence and bequest values, aesthetic values, and recreation values, among
27 others; provisioning services (e.g., food and timber); and regulating services such as climate
28 regulation or hydrologic cycle (Millenium Ecosystem Assessment, 2005). Figure 2- 1 illustrates
29 the relationships between the ecological effects of ozone and the anticipated ecosystem services
30 impacts. Specific services to be evaluated are discussed in the following sections.
2-1
-------
Ecological
Effects
Reduced
productivity
Foliar injury
Stomatal
functioning
Supporting Services
• Soil Formation
• Primary Productivity
• Community Structure
Regulating Services
Carbon Sequestration
Nutrient Cycling
Water Regulation
Fire Ecology
Pollination
Cultural Services *
Endangered Species
Recreational Use
Non-use Values
Provisioning Services
• Agricultural harvest
• Timber production
• Non-timber uses
4
5
Figure 2-1 Relationship Between Ecological Effects of Ozone Exposure and
Ecosystem Services
6 2.1 O3 CHEMISTRY
7 Os occurs naturally in the stratosphere where it provides protection against harmful solar
8 ultraviolet radiation, and it is formed closer to the surface in the troposphere by both natural and
9 anthropogenic sources. Oj is not emitted directly into the air, but is created when its two primary
10 precursors, volatile organic compounds (VOC) and oxides of nitrogen (NOX), combine in the
11 presence of sunlight. VOC and NOX are, for the most part, emitted directly into the atmosphere.
12 Carbon monoxide (CO) and methane (CH/i) are also important for O^ formation (US EPA, 2012,
13 section 3.2.2).
14 Rather than varying directly with emissions of its precursors, Os changes in a nonlinear
15 fashion with the concentrations of its precursors. NOX emissions lead to both the formation and
16 destruction of Os, depending on the local quantities of NOX, VOC, and radicals such as the
17 hydroxyl (OH) and hydro-peroxy (HO2) radicals. In areas dominated by fresh emissions of NOX,
2-2
-------
1 these radicals are removed via the production of nitric acid (HNO3), which lowers the O?,
2 formation rate. In addition, the scavenging of O^ by reaction with NO is called "titration," and is
3 often found in downtown metropolitan areas, especially near busy streets and roads, and in
4 power plant plumes. This titration results in local valleys in which ozone concentrations are low
5 compared to surrounding areas. Titration is usually short-lived confined to areas close to strong
6 NOX sources, and the NC>2 formed this way leads to 63 formation later and further downwind. .
7 Consequently, ozone response to reductions in NOX emissions is complex and may include ozone
8 decreases at some times and locations and increases of ozone to fill in the local valleys of low
9 ozone. In areas with low NOX concentrations, such as those found in remote continental areas to
10 rural and suburban areas downwind of urban centers, the net production of O^ typically varies
11 directly with NOX concentrations, and increases with increasing NOX emissions.
12 In general, the rate of 63 production is limited by either the concentration of VOCs or
13 NOX, and 63 formation using these two precursors relies on the relative sources of OH and NOX.
14 When OH radicals are abundant and are not depleted by reaction with NOX and/or other species,
15 O3 production is referred to as being "NOx-limited" (US EPA, 2012, section 3.2.4). In this
16 situation, Os concentrations are most effectively reduced by lowering NOX emissions, rather than
17 lowering emissions of VOCs. When the abundance of OH and other radicals is limited either
18 through low production or reactions with NOX and other species, O?, production is sometimes
19 called "VOC-limited" or "radical limted" or "NOx-saturated" (Jaegle et al., 2001), and O3 is most
20 effectively reduced by lowering VOCs. However, even in NOx-saturated conditions, very large
21 decreases in NOX emissions can cause the ozone formation regime to become NOX limited.
22 Consequently, reductions in NOX emissions (when large) can make further emissions reductions
23 more effective at reducing ozone. Between the NOx-limited and NOx-saturated extremes there is
24 a transitional region where Os is relatively insensitive to marginal changes in both NOX and
25 VOCs.
26 In rural areas and downwind of urban areas, O?, production is generally NOx-limited. This
27 is particularly true in rural areas such as national parks, national forests, and state parks where
28 VOC emissions from vegetation are high and anthropogenic NOX emissions are relatively low.
29 Due to lower chemical scavenging in non-urban areas, Oj tends to persist longer in rural than in
30 urban areas and tends to lead to higher cumulative exposures in rural areas than in urban areas.
31 (US EPA, 2012a, Section 3.6.2.2).
-------
1 2.2 SOURCES OF O3 AND O3 PRECURSORS
2 63 precursor emissions can be divided into anthropogenic and natural source categories,
3 with natural sources further divided into biogenic emissions (from vegetation, microbes, and
4 animals) and abiotic emissions (from biomass burning, lightning, and geogenic sources). The
5 anthropogenic precursors of O^ originate from a wide variety of stationary and mobile sources.
6 In urban areas, both biogenic and anthropogenic VOCs are important for O^ formation.
7 Hundreds of VOCs are emitted by evaporation and combustion processes from a large number of
8 anthropogenic sources. Based on the 2005 national emissions inventory (NEI), solvent use and
9 highway vehicles are the two main sources of VOCs, with roughly equal contributions to total
10 emissions (US EPA, 2012a, Figure 3-3). The emissions inventory categories of "miscellaneous"
11 (which includes agriculture and forestry, wildfires, prescribed burns, and structural fires) and off-
12 highway mobile sources are the next two largest contributing emissions categories with a
13 combined total of over 5.5 million metric tons a year (MT/year).
14 On the U.S. and global scales, emissions of VOCs from vegetation are much larger than
15 those from anthropogenic sources. Emissions of VOCs from anthropogenic sources in the 2005
16 NEI were -17 MT/year (wildfires constitute -1/6 of that total), but were 29 MT/year from
17 biogenic sources. Vegetation emits substantial quantities of VOCs, such as isoprene and other
18 terpenoid and sesqui-terpenoid compounds. Most biogenic emissions occur during the summer
19 because of their dependence on temperature and incident sunlight. Biogenic emissions are also
20 higher in southern and eastern states than in northern and western states for these reasons and
21 because of species variations.
22 Anthropogenic NOX emissions are associated with combustion processes. Based on the
23 2005 NEI, the three largest sources of NOX are on-road and off-road mobile sources (e.g.,
24 construction and agricultural equipment) and electric power generation plants (EGUs) (US EPA,
25 2012, Figure 3-3). Emissions of NOX therefore are highest in areas having a high density of
26 power plants and in urban regions having high traffic density. However, it is not possible to
27 make an overall statement about their relative impacts on Os in all local areas because EGUs are
28 sparser than mobile sources, particularly in the west and south and because of the nonlinear
29 chemistry discussed in Section 2.1.
30 Major natural sources of NOX in the U.S. include lightning, soils, and wildfires. Biogenic
31 NOX emissions are generally highest during the summer and occur across the entire country,
2-4
-------
1 including areas where anthropogenic emissions are low. It should be noted that uncertainties in
2 estimating natural NOX emissions are much larger than for anthropogenic NOX emissions.
3 Ozone concentrations in a region are affected both by local formation and by transport
4 from surrounding areas. Ozone transport occurs on many spatial scales including local transport
5 between cities, regional transport over large regions of the U.S. and international/long-range
6 transport. In addition, Os is also transfered into the troposphere from the stratosphere, which is
7 rich in Os, through stratosphere-troposphere exchange (STE). These inversions or "foldings" usually
8 occur behind cold fronts, bringing stratospheric air with them (U.S. EPA, 2012, section 3.4.1.1).
9 Contribution to Os concentrations in an area from STE are defined as being part of background Os
10 (U.S. EPA, 2012, section 3.4).
11 Rural areas, such as national parks, national forests, and state parks, tend to be less
12 directly affected by anthropogenic pollution sources than urban sites. However, they can be
13 regularly affected by transport of Os or Os precursors from upwind urban areas. In addition,
14 biogenic VOC emissions tend to be higher in rural areas and major sources of Os precursor
15 emissions such as highways, power plants, biomass combustion, and oil and gas operations are
16 commonly found in rural areas, adding to the Os produced in these areas. Areas at higher
17 elevations, such as many of the national parks in the western U.S., can also be affected more
18 significantly by international transport of Os or stratospheric intrusions that transport Os into the
19 area (US EPA, 2012a, section 3.7.3).
20 2.3 ECOLOGICAL EFFECTS
21 Recent studies reviewed in the ISA support and strengthen the findings reported in the
22 2006 O3 AQCD (U.S. EPA, 2006a). The most significant new body of evidence since the 2006
23 Os AQCD comes from research on molecular mechanisms of the biochemical and physiological
24 changes observed in many plant species in response to Os exposure. These newer molecular
25 studies not only provide very important information regarding the many mechanisms of plant
26 responses to Os, they also allow for the analysis of interactions between various biochemical
27 pathways which are induced in response to Os. However, many of these studies have been
28 conducted in artificial conditions with model plants, which are typically exposed to very high,
29 short doses of Os and are not quantifiable as part of this risk assessment, which is focused on
30 ambient conditions.
2-5
-------
1
2
3
4
5
6
7
10
11
12
13
14
15
16
Chapter 9 of the O3 ISA (U.S. EPA, 2012a) provides a detailed review of the effects of
Os on vegetation including the major pathways of exposure and known ecological and ecosystem
effects. Figure 9-1 of the ISA is reproduced below (Figure 2- 2) as a summary of exposure and
effects. In general, 63 is taken up through the stomata into the leaves. Once inside the leaves, 63
affects a number of biological and physiological processes, including photosynthesis. This leads,
in some cases, to visible foliar injury as well as reduced plant growth, which are the main
ecological effects assessed in this review. Visible foliar injury and reduced growth can lead to a
reduction in ecosystem services, including crop and timber yield loss, decreased C sequestration,
alteration in community composition and loss of recreational or cultural value.
03 exposure
03 uptake & physiology (Fig 9-2)
•Antioxidant metabolism up-regulated
•Decreased photosynthesis
•Decreased stomatal conductance
or sluggish stomatal response
Effects on leaves
•Visible leaf injury
•Altered leaf production
•Altered leaf chemical composition
Plant growth (Fig 9.8)
•Decreased biomass accumulation
•Altered reproduction
•Altered carbon allocation
•Altered crop quality
Affected ecosystem services
•Decreased productivity
•Decreased C sequestration
B k -Altered water cycling (Fig 9-7)
•Altered community composition
(i.e., plant, insects microbe)
Belowground processes (Fig 9.8)
•Altered litter production and decomposition
•Altered soil carbon and nutrient cycling
•Altered soil fauna and microbial communities
Figure 2- 2 Conceptual diagram of the major pathway through which Os enters
plants and the major endpoints that 03 may affect in plants and
ecosystems. Figure numbers in this figure refer to Chapter 9 of the
ISA.
Overall causal determinations are made based on the full range of evidence including
controlled exposure studies and ecological studies. Figure 2- 3 shows the 63 welfare effects
2-6
-------
1
2
3
4
which have been categorized by strength of evidence for causality in the O^ ISA (US EPA,
2012a, chapter 2). These determinations support causal or likely causal relationships between
exposure to 63 and ecological and ecosystem level effects.
Not likely
Inadequate
to infer
Reduced carbon
sequestration in
terrestrial ecosystems
Alteration of
terrestrial ecosystem
water cycling
Alteration of
terrestrial community
composition
Suggestive
Likely
Visible foliar injury
effects
Reduced vegetation growth
Reduced productivity in
terrestrial ecosystems
Reduced yield and
quality of agricultural
crops
Alteration of below
ground biogeochemical
cycles
Causal
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Figure 2- 3 Causal Determinations for Os Welfare Effects
The adequate characterization of the effects of 63 on plants for the purpose of setting air
quality standards is contingent not only on the choice of the index used (i.e. W126) to summarize
Os concentrations (Section 9.5), but also on quantifying the response of the plant variables of
interest at specific values of the selected index. The factors that determine the response of plants
to Os exposure include species, genotype and other genetic characteristics, biochemical and
physiological status, previous and current exposure to other stressors, and characteristics of the
exposure itself. Establishing a secondary air quality standard requires the capability to generalize
those observations, in order to obtain predictions that are reliable enough under a broad variety
of conditions, taking into account these factors.
Quantitative characterization of exposure-response in the 2006 Os AQCD was based on
experimental data generated for that purpose by the National Crop Loss Assessment Network
2-7
-------
1 (NCLAN) and EPA National Health and Environmental Effects Research Laboratory, Western
2 Ecology Division (NHEERL-WED) projects, using OTCs to expose crops and trees seedling to
3 63. In recent years, yield and growth results for two of the species that had provided extensive
4 exposure-response information in those projects have become available from studies that used
5 FACE technology, which is intended to provide conditions much closer to natural environments
6 (Pregitzer et al., 2008; Morgan et al., 2006; Morgan et al., 2004; Dickson et al., 2000).
7 The quantitative exposure-response relationships described in the 2006 Os AQCD have
8 not changed in the current draft ISA, with the exception of the addition of one new species, e
9 assessment of quantitative exposure-response relationships that was presented in that document.
10 The exposure-response models are summarized in the 3rd draft ISA summarizes computed using
11 the W126 metric, cumulated over 90 days. These response functions provide an adequate basis
12 for quantifying biomass loss damages.
13 Visible foliar injury resulting from exposure to 63 has also been well characterized and
14 documented over several decades of research on many tree, shrub, herbaceous, and crop species
15 (U.S. EPA, 2006, 1996a, 1984, 1978). Ozone-induced visible foliar injury symptoms on certain
16 bioindicator plant species are considered diagnostic as they have been verified experimentally in
17 exposure-response studies, using exposure methodologies such as continuous stirred tank
18 reactors (CSTRs), OTCs, and free-air fumigation. Experimental evidence has clearly established
19 a consistent association of visible injury with Oj, exposure, with greater exposure often resulting
20 in greater and more prevalent injury. This general relationship provides an adequate basis for
21 qualitative assessment of the risk of visible foliar injury, but a detailed quantitative assessment is
22 not possible because there are no concentration-response functions for foliar injury that can be
23 applied across a range of ecosystems.
24 2.4 ECOSYSTEM SERVICES
25 The Risk and Exposure Assessment evaluates the benefits received from the resources and
26 processes that are supplied by ecosystems. Collectively, these benefits are known as ecosystem
27 services and include products or provisions, such as food and fiber; processes that regulate
28 ecosystems, such as carbon sequestration; cultural enrichment; and supportive processes for
29 services, such as nutrient cycling. Ecosystem services are distinct from other ecosystem products
2-8
-------
1 and functions because there is human demand for these services. In the Millennium Ecosystem
2 Assessment (MEA), ecosystem services are classified into four main categories:
3 • Provisioning. Includes products obtained from ecosystems, such as the production of
4 food and water.
5 • Regulating. Includes benefits obtained from the regulation of ecosystem processes, such
6 as the control of climate and disease.
7 • Cultural. Includes the nonmaterial benefits that people obtain from ecosystems through
8 spiritual enrichment, cognitive development, reflection, recreation, and aesthetic
9 experiences.
10 • Supporting. Includes those services necessary for the production of all other ecosystem
11 services, such as nutrient cycles and crop pollination (MEA, 2005).
12 The concept of ecosystem services can be used to help define adverse effects as they pertain
13 to NAAQS reviews. The most recent secondary NAAQS reviews have characterized known or
14 anticipated adverse effects to public welfare by assessing changes in ecosystem structure or
15 processes using a weight-of-evidence approach that uses both quantitative and qualitative data.
16 For example, the previous ozone review evaluated changes in foliar injury, growth loss, and
17 biomass reduction on trees beyond the seedling stage using the TREGRO model. The presence
18 or absence of foliar damage in counties meeting the current standard has been used as a way to
19 evaluate the adequacy of the secondary NAAQS. Characterizing a known or anticipated adverse
20 effect to public welfare is an important component of developing any secondary NAAQS.
21 According to the Clean Air Act (CAA), welfare effects include the following:
22
23 "Effects on soils, water, crops, vegetation, manmade materials,
24 animals, wildlife, weather, visibility, and climate, damage to and
25 deterioration of property, and hazards to transportation, as well as
26 effect on economic values and on personal comfort and well-being,
27 whether caused by transformation, conversion, or combination
28 with other air pollutants." (Section 302(h))
29
2-9
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
In other words, welfare effects are those effects that are important to individuals and/or
society in general. Ecosystem services can be generally defined as the benefits that individuals
and organizations obtain from ecosystems. EPA has defined ecological goods and services as the
"outputs of ecological functions or processes that directly or indirectly contribute to social
welfare or have the potential to do so in the future. Some outputs may be bought and sold, but
most are not marketed" (U.S. EPA, 2006). Conceptually, changes in ecosystem services may be
used to aid in characterizing a known or anticipated adverse effect to public welfare. In the
context of this review, ecosystem services may also aid in assessing the magnitude and
significance of a resource and in assessing how O^ concentrations may impact that resource.
Figure 2- 4 provides the World Resources Institute's schematic demonstrating the
connections between the categories of ecosystem services and human well-being. The
interrelatedness of these categories means that any one ecosystem may provide multiple services.
Changes in these services can impact human well-being by affecting security, health, social
relationships, and access to basic material goods (MEA, 2005).
CONSTITUENTS OF WELL-BEING
ECOSYSTEM SERVICES
Provisioning
FOOD
FRESH WATER
WOOD AND FIBER
FUEL
Supporting
NUTRIENT CYCLING
SOIL FORMATION
PRIMARY PRODUCTION
Regulating
CLIMATE REGULATION
FLOOD REGULATION
DISEASE REGULATION
WATER PURIFICATION
Cultural
AESTHETIC
SPIRITUAL
EDUCATIONAL
RECREATIONAL
LIFE ON EARTH - BIODIVERSITY
Security
PERSONAL SAFETY
SECURE RESOURCE ACCESS
SECURITY FROM DISASTERS
Basic material
for good life
ADEQUATE LIVELIHOODS
SUFFICIENT NUTRITIOUS FOOD
SHELTER
ACCESS TO GOODS
Health
STRENGTH
FEELING WELL
ACCESS TO CLEAN AIR
AND WATER
Good social relations
SOCIAL COHESION
MUTUAL RESPECT
ABILITY TO HELP OTHERS
Freedom
of choice
and action
OPPORTUNITY TO BE
ABLE TO ACHIEVE
WHAT AN INDIVIDUAL
VALUES DOING
AND BEING
Source: Millennium Ecosystem Assessment
Figure 2- 4 Linkages between categories of ecosystem services and components of
human well-being that are commonly indications of the extent to
which it is possible for socioeconomic factors to mediate the linkage.
The strength of the linkages, as indicated by arrow width, and the
2-10
-------
1 potential for mediation, as indicated by arrow color, differ in different
2 ecosystems and regions (MEA, 2005).
3
4 Historically, ecosystem services have been undervalued and overlooked; however, more
5 recently, the degradation and destruction of ecosystems has piqued interest in assessing the value
6 of these services. In addition, valuation may be an important step from a policy perspective
7 because it can be used to compare the costs and benefits of altering versus maintaining an
8 ecosystem (i.e., it may be easier to protect than repair ecosystem effects). In this Risk and
9 Exposure Assessment, valuation is used, where possible, based on available data in the national
10 scale analyses and case study areas.
11 The economic approach to the valuation of ecosystem services is laid out as follows in
12 EPA's Ecological Benefits Assessment Strategic Plan: "Economi sts generally attempt to estimate
13 the value of ecological goods and services based on what people are willing to pay (WTP) to
14 increase ecological services or by what people are willing to accept (WTA) in compensation for
15 reductions in them" (U.S. EPA, 2006). There are three primary approaches for estimating the
16 value of ecosystem services: market-based approaches, revealed preference methods, and stated
17 preference methods (U.S. EPA, 2006). Because economic valuation of ecosystem services can be
18 difficult, nonmonetary valuation using biophysical measurements and concepts also can be used.
19 Examples of nonmonetary valuation methods include the use of relative-value indicators (e.g., a
20 flow chart indicating uses of a waterbody, such as boatable, fishable, swimmable); another
21 assigns values to ecosystem goods and services through the use of the common currency of
22 energy. Energetic valuation attempts to assess ecosystem contributions to the economy by using
23 one kind of energy (e.g., solar energy) to express the value of that type of energy required to
24 produce designated services (Odum, 1996). This energy value is then converted to monetary
25 units. This method of valuation, however, does not account for the premise that values arise from
26 individual or societal preferences.
27 Valuing ecological benefits, or the contributions to social welfare derived from
28 ecosystems, can be challenging, as noted in EPA's Ecological Benefits Assessment Strategic
29 Plan (U.S. EPA, 2006). It is necessary to recognize that in the analysis of the environmental
30 responses associated with any particular policy or environmental management action, some of
31 the ecosystem services likely to be affected are readily identified, whereas others will remain
2-11
-------
1
2
3
4
5
6
9
10
11
12
unidentified. Of those ecosystem services that are identified, some changes can be quantified,
whereas others cannot. Within those services whose changes can be quantified, only a few will
likely be monetized, and many will remain unmonetized. Similar to health effects, only a portion
of the ecosystem services affected by a policy can be monetized. The stepwise concept leading
up to the valuation of ecosystems services is graphically depicted in Figure 2- 5.
/
EPA Action
i i
Ecosystems
4r ^ 4>
Ecological goods and services
affected by the policy
,.••'
Planning and problem
formulation
Goods and services
identified
/
Ecological analysis
Goods and services
quantified
/
Economic analysis
---•^
Goods and
services
-A
n
S NL Goods and
\ services not
~[/^ identified
,/1 l\ Identified goods
1 y and services not
[/' quantified
j* r^ Quantified
X good and
— ijf services not
f monetized
monetized ^^. -"""*
Figure 2- 5 Representation of the benefits assessment process indicating where
some ecological benefits may remain unrecognized, unquantified, or
unmonetized. (Modified based on the Ecological Benefits Assessment
Strategic Plan report [U.S. EPA, 2006]).
2-12
-------
1 Under Section 108 of the CAA, the secondary standard is to specify an acceptable level
2 of the criteria pollutant(s) in the ambient air that is protective of public welfare. For this review,
3 the relevant air quality indicator is interpreted as ambient 63 concentrations that can be linked to
4 adverse ecological effects. The air quality analyses described in Chapter 4 explore the sources
5 and emissions, and their current contributions to ambient conditions. The national scale and case
6 study analyses (described in Chapters 5 and 6) link 63 effects in sensitive ecosystems (e.g., the
7 exposure pathway) to changes in a given ecological indicator (e.g., biomass loss to changes in
8 ecosystems and the services they provide (e.g., commercial timber production). To the extent
9 possible for effect, ambient concentrations of Os (i.e., ambient air quality indicators) were linked
10 to effects in sensitive ecosystems (i.e., exposure pathways), and then 03 concentrations were
11 linked to system response as measured by a given ecological indicator (e.g., biomass loss). The
12 ecological effect (e.g., changes in tree growth) was then, where possible, associated with changes
13 in ecosystem services and their ecological benefits or welfare effects (e.g., timber production).
14 Knowledge about the relationships linking ambient concentrations and ecosystem
15 services can be used to inform a policy judgment on a known or anticipated adverse public
16 welfare effect. For example, changes in biodiversity would be classified as an ecological effect,
17 and the associated changes in ecosystem services—productivity, recreational viewing, and
18 aesthetics—would be classified as ecological benefits/welfare effects. This information can then
19 be used to characterize known or anticipated adverse effects to public welfare and inform a
20 policy based on welfare effects.
21 The ecosystems of interest in this Risk and Exposure Assessment are impacted by the
22 effects of anthropogenic air pollution, which may alter the services provided by the ecosystems
23 in question. For example, changes in forest health as a result of Os exposure may affect
24 supporting services such as net primary productivity; provisioning services such as timber
25 production; and regulating services such as climate regulation. In addition, such changes may
26 provide provisioning services such as food; and cultural services such as recreation and
27 ecotourism.
28 Where possible, linkages to ecosystem services from indicators of each effect identified
29 in the ISA (U.S. EPA, 2012a) were developed. These linkages were based on existing literature
30 and models, focus on the services identified in the peer-reviewed literature, and are essential to
31 any attempt to evaluate air pollution-induced changes in the quantity and/or quality of ecosystem
2-13
-------
1 services provided. According to EPA's Science Advisory Board Committee on Valuing the
2 Protection of Ecological Systems and Services, these linkages are critical elements for
3 determining the valuation of benefits of EPA-regulated air pollutants (SAB CVPESS, 2009).
4 We have identified the primary ecosystem service(s) potentially impacted by 63 for
5 major ecosystem types and components (i.e., terrestrial ecosystems, productivity) under
6 consideration in this risk and exposure assessment. The impacts associated with various
7 ecosystem services for each targeted effect are assessed in Chapter 6 at a national scale and in
8 case studies.
9
10 2.5 CONCLUSIONS
11 The conceptual basis for estimating exposures to O^ and resulting welfare effects is strong. The
12 ISA provides clear scientific evidence linking ambient concentrations of Os to a number of
13 ecological effects, and science-based air quality models along with Oj, monitoring data, show
14 that important ecosystems throughout the U.S. are exposed to Os concentrations that may result
15 in adverse ecological impacts. There are field and laboratory studies that provide adequate
16 information to construct concentration-response functions that can be used to estimate risk given
17 estimates of tree or ecosystem level Os exposure.
18
19 Presented below are key observations for this conceptual overview of the assessment of ambient
20 Os exposure and welfare risk.
21
22 • Os in ambient air is formed primarily by emissions of NOX and VOC and
23 photochemical reactions in the atmosphere. Both natural and anthropogenic sources
24 contribute to Os formation. Solvents, on-road and off-road mobile sources and electric
25 power generation plants represent significant anthropogenic sources of precursors to Os
26 in ambient air. Vegetation, lightning, soils, and wildfires are significant natural sources
27 of Os precursor emissions.
28 • The ISA has determined that the evidence supports a causal relationship between
29 exposure to Os and visible foliar injury, reduced vegetation growth, reduced agricultural
30 yield, and alteration of below ground biogeochemical cycles, and a likely causal
2-14
-------
1 relationship exposure to Os and reduced carbon sequestration, alteration of terrestrial
2 water cycling, and alteration of terrestrial community composition.
3 • The causal and likely causal ecological effects identified in the ISA have an effect
4 on regulating, supporting, cultural and provisioning ecosystem services.
5
6
2-15
-------
1 3 SCOPE
2 This chapter provides an overview of the scope and key design elements of this
3 quantitative exposure and welfare risk assessment. The design of this assessment began with a
4 review of the exposure and risk assessments completed during the last 63 NAAQS review (US
5 EPA, 2007a,b), with an emphasis on considering key limitations and sources of uncertainty
6 recognized in that analysis.
7 As an initial step in the current Os NAAQS review, in October 2009, EPA invited outside
8 experts, representing a broad range of expertise to participate in a workshop with EPA staff to
9 help inform EPA's plan for the review. The participants discussed key policy-relevant issues
10 that would frame the review and the most relevant new science that would be available to inform
11 our understanding of these issues. One workshop session focused on planning for quantitative
12 risk and exposure assessments, taking into consideration what new research and/or improved
13 methodologies would be available to inform the design of quantitative exposure and welfare risk
14 assessment. Based in part on the workshop discussions, EPA developed a draft IRP (US EPA,
15 2009) outlining the schedule, process, and key policy-relevant questions that would frame this
16 review. On November 13, 2009, EPA held a consultation with CAS AC on the draft IRP (74 FR
17 54562, October 22, 2009), which included opportunity for public comment. The final IRP
18 incorporated comments from CASAC (Samet, 2009) and the public on the draft plan as well as
19 input from senior Agency managers. The final IRP included initial plans for the quantitative risk
20 and exposure assessments for both human health and welfare (US EPA, 201 la, chapters 5 and 6).
21 As a next step in the design of these quantitative assessments, OAQPS staff developed
22 more detailed planning documents, Oj National Ambient Air Quality Standards: Scope and
23 Methods Plan for Health Risk and Exposure Assessment (Health Scope and Methods Plan; US
24 EPA, 201 Ib) and Oj National Ambient Air Quality Standards: Scope and Methods Plan for
25 Welfare Risk and Exposure Assessment (Welfare Scope and Methods Plan, US EPA, 201 Ic).
26 These Scope and Methods Plans were the subject of a consultation with CASAC on May 19-20,
27 2011 (76 FR 23809, April 28, 2011). Based on consideration of CASAC (Samet, 2011) and
28 public comments on the Scope and Methods Plan and information in the second draft ISA, we
29 modified the scope and design of the quantitative risk assessment and provided a memo with
30 updates to information presented in the Scope and Methods Plans (Wegman, 2012). The Scope
5-1
-------
1 and Methods Plans together with the update memo provide the basis for the discussion of the
2 scope of this exposure and risk assessment provided in this chapter.
3 In presenting the scope and key design elements of the current risk assessment, this chapter first
4 provides a brief overview of the quantitative exposure and risk assessment completed for the
5 previous 63 NAAQS review in section 3.1, including key limitations and uncertainties associated
6 with that analysis. Section 3.2 provides a summary of the design of the exposure assessment.
7 Section 3.3 provides a summary of the design of the risk assessment based on application of
8 results of human clinical studies. Section 3.4 provides a summary of the design of the risk
9 assessment based on application of results of epidemiology studies.
10 3.1 OVERVIEW OF EXPOSURE AND RISK ASSESSMENTS FROM LAST REVIEW
11 The assessments conducted as part of the last review focused on national-level Os-related
12 impacts to sensitive vegetation and their associated ecosystems. The vegetation exposure
13 assessment was performed using an interpolation approach that included information from
14 ambient monitoring networks and results from air quality modeling. The vegetation risk
15 assessment included both tree and crop analyses. The tree risk analysis included three distinct
16 lines of evidence: (1) observations of visible foliar injury in the field linked to monitored Oi air
17 quality for the years 2001 - 2004; (2) estimates of seedling growth loss under then current and
18 alternative Os exposure conditions; and (3) simulated mature tree growth reductions using the
19 TREGRO model to simulate the effect of meeting alternative air quality standards on the
20 predicted annual growth of mature trees from three different species. The crop risk analysis
21 included estimates of crop yields under current and alternative Os exposure conditions. The
22 associated changes in economic value upon meeting the levels of various alternative standards
23 were analyzed using an agricultural sector economic model. Key elements and observations
24 from these exposure and risk assessments are outlined in the following sections.
25 3.1.1 Exposure Characterization
26 In many rural and remote areas where sensitive species of vegetation can occur,
27 monitoring coverage remained limited. Thus, the 2007 Staff Paper concluded that it was
28 necessary to use an interpolation method in order to better characterize Os air quality over broad
29 geographic areas and at the national scale. Based on the significant difference in monitor
5-2
-------
1 network density between the eastern and western U.S., the Staff Paper further concluded that it
2 was appropriate to use separate interpolation techniques in these two regions: The Air Quality
3 System (AQS; http://www.epa.gov/ttn/airs/airsaqs) and Clean Air Status and Trends Network
4 (CASTNET; http://www.epa.gov/castnet/) monitoring data were solely used for the eastern
5 interpolation, and in the western U.S., where rural monitoring is more sparse, 63 outputs from
6 the EPA/NOAA Community Multi-scale Air Quality (CMAQ) model system
7 (http://www.epa.gov/asmdnerl/CMAQ, Byun and Ching, 1999; Byun and Schere, 2006) were
8 used to develop scaling factors to augment the monitor interpolation. In order to characterize
9 uncertainty associated with the exposure estimates generated using the interpolation method,
10 monitored O?, concentrations were systematically compared to interpolated O^ concentrations in
11 areas where monitors were located. In general, the interpolation method performed well in many
12 areas in the U.S. This approach was used to develop a national vegetation 63 exposure surface.
13 To evaluate changing vegetation exposures under selected air quality scenarios, a number
14 of analyses were conducted. One analysis adjusted 2001 base year 63 air quality distributions
15 using a rollback method (Rizzo, 2005, 2006) to reflect meeting the current and alternative
16 secondary standard options. For "just meet" and alternative 8-hr average standard scenarios, the
17 associated maps of estimated 12-hr, W126 exposures were generated. Based on these
18 comparisons, the following observations were drawn: (1) current O^ air quality levels could
19 result in significant cumulative, seasonal 63 exposures to vegetation in some areas; (2) overall 3-
20 month 12-hr W126 Os levels were somewhat but not substantially improved under the "just
21 meet" current (0.08 ppm) scenario; (3) exposures generated for just meeting a 0.070 ppm, 4th-
22 highest maximum 8-hr average alternative standard (the lower end of the then proposed range for
23 the primary Os standard) showed substantially improved 3-month cumulative, seasonal Os air
24 quality when compared to just meeting the current 0.08 ppm, 8-hr average standard.
25 A second analysis described in the Staff Paper was performed to evaluate the extent to
26 which county-level O^ air quality measured in terms of various levels of the current 8-hr average
27 form overlapped with that measured in terms of various levels of the 12-hr W126 cumulative,
28 seasonal form. While these results also suggested that meeting a proposed 0.070 ppm, 8-hr
29 secondary standard would provide substantially improved vegetation protection in some areas,
30 the Staff Paper recognized that this analysis had several important limitations. In particular, the
31 lack of monitoring in rural areas where sensitive vegetation and ecosystems are located,
3-3
-------
1 especially at higher elevation sites, could have resulted in an inaccurate characterization of the
2 degree of potential overlap at sites that have air quality patterns that can result in relatively low
3 8-hr averages while still experiencing relatively high cumulative exposures (72 FR 37892).
4 Thus, the Staff Paper concluded that it is reasonable to anticipate that additional unmonitored
5 rural high elevation areas with sensitive vegetation may not be adequately protected even with a
6 lower level of the 8-hr form. The Staff Paper further indicated that it remained uncertain as to
7 the extent to which air quality improvements designed to reduce 8-hr Os average concentrations
8 would reduce Os exposures measured by a seasonal, cumulative W126 index. The Staff Paper
9 indicated this to be an important consideration because: (1) the biological database stresses the
10 importance of cumulative, seasonal exposures in determining plant response; (2) plants have not
11 been specifically tested for the importance of daily maximum 8-hr 63 concentrations in relation
12 to plant response; and (3) the effects of attainment of a 8-hr standard in upwind urban areas on
13 rural air quality distributions cannot be characterized with confidence due to the lack of
14 monitoring data in rural and remote areas.
15 The Staff Paper also presented estimates of economic valuation for crops associated with
16 the then current and alternative standards. The Agriculture Simulation Model (AGSEVI) (Taylor,
17 1994; Taylor, 1993) was used to calculate annual average changes in total undiscounted
18 economic surplus for commodity crops and fruits and vegetables when then current and
19 alternative standard levels were met. Meeting the various alternative standards did show some
20 significant benefits beyond the 0.08 ppm, 8-hr standard. However, the Staff Paper recognized
21 that the modeled economic impacts from AGSIM had many associated uncertainties, which
22 limited the usefulness of these estimates.
23 3.1.2 Assessment of Risks to Vegetation
24 The risk assessments in the last review reflected the availability of several additional
25 lines of evidence that provided a basis for a more complete and coherent picture of the scope of
26 Os-related vegetation risks, especially those faced by seedling, sapling and mature tree species
27 growing in field settings, and indirectly, forested ecosystems. Specifically, new research
28 available at the time reflected an increased emphasis on field-based exposure methods (e.g., free
29 air exposure and ambient gradient), improved field survey biomonitoring techniques, and
5-4
-------
1 mechanistic tree process models. Highlights from the analyses that addressed visible foliar
2 injury, seedling and mature tree biomass loss, and effects on crops are summarized below.
3 With regard to visible foliar injury, the Staff Paper presented an assessment that
4 combined recent U.S. Forest Service Forest Inventory and Analysis (FIA) biomonitoring site
5 data with the county level air quality data for those counties containing the FIA biomonitoring
6 sites. This assessment showed that incidence of visible foliar injury ranged from 21 to 39
7 percent of the counties during the four-year period (2001-2004) across all counties with air
8 quality levels at or below that of the then current 0.08 ppm 8-hr average standard. Of the
9 counties that met an 8-hr average level of 0.07 ppm in those years, 11 to 30 percent of the
10 counties still had incidence of visible foliar injury.
11 With respect to tree seedling biomass loss, concentration-response (C-R) functions
12 developed from Open Top Chamber (OTC) studies for biomass loss for available seedling tree
13 species and information on tree growing regions derived from the U.S. Department of
14 Agriculture's Atlas of United States Trees were combined with projections of air quality based
15 on 2001 interpolated exposures, to produce estimated biomass loss for each individual seedling
16 tree species. These analyses predicted that biomass loss could still occur in many tree species
17 when Os air quality was adjusted to meet the then current 8-hr average standard. Though this
18 type of analysis was not new to this review, the context for understanding these results had
19 changed due to recent field work at the AspenFACE site in Wisconsin on quaking aspen
20 (Karnosky et al., 2005) and a gradient study performed in the New York City area (Gregg et al.,
21 2003), which confirmed the detrimental effects of Oj exposure on tree growth in field studies
22 without chambers and beyond the seedling stage (King et al., 2005).
23 With respect to risk of mature tree growth reductions, a tree growth model (TREGRO)
24 was used to evaluate the effect of changing Os air quality scenarios from just meeting alternative
25 Os standards on the growth of mature trees.l The model was run for a single western species
26 (ponderosa pine) and two eastern species (red maple and tulip poplar). Staff Paper analyses
27 found that just meeting the then current standard would likely continue to allow Os-related
1 TREGRO is a process-based, individual tree growth simulation model (Weinstein et al, 1991) that is linked with
concurrent climate data to account for O3 and climate/meteorology interactions on tree growth. TREGRO has been
used to evaluate the effects of a variety of O3 scenarios on several species of trees in different regions of the U.S.
(Tingey et al., 2001; Weinstein et al., 1991; Retzlaff et al., 2000; Laurence et al., 1993; Laurence et al., 2001;
Weinstein et al., 2005).
3-5
-------
1 reductions in annual net biomass gain in these species. Though there was uncertainty associated
2 with the above analyses, it was important to note that recent evidence from experimental studies
3 that go beyond the seedling growth stage continued to show decreased growth under elevated 63
4 (King et al., 2005); some mature trees such as red oak have shown an even greater sensitivity of
5 photosynthesis to 63 than seedlings of the same species (Hanson et al., 1994); and the potential
6 for cumulative "carry over" effects as well as compounding should be considered (Andersen, et
7 al, 1997).
8 With respect to risks of yield loss in agricultural crops and fruit and vegetable species,
9 little new information was available beyond that of the previous review. However, limited
10 information from a free air field based soybean study (SoyFACE) and information on then
11 current cultivar sensitivities led to the conclusion that C-R functions developed in OTCs under
12 the National Crop Loss Assessment Network (NCLAN) program could still be usefully applied.
13 The crop risk assessment, like the tree seedling assessment, combined NCLAN C-R information
14 on commodity crops, fruits and vegetables, crop growing regions, and interpolated exposures
15 during each crop growing season. The risk assessment estimated that just meeting the 0.08 ppm,
16 8-hr standard would still allow Cb-related yield loss to occur in some sensitive commodity crops
17 and fruit and vegetable species growing at that time in the U.S.
18 3.2 OVERVIEW OF CURRENT ASSESSMENT PLAN
19 Since the 2008 review, new scientific information on the direct and indirect effects of O^
20 on vegetation and ecosystems, respectively, has become available. With respect to mature trees
21 and forests, the information regarding Oj, impacts to forest ecosystems has continued to expand,
22 including limited new evidence that implicates Os as an indirect contributor to decreases in
23 stream flow through direct impacts on whole tree level water use. Newly published results from
24 the Long-term FACE (Free Air CO2 enrichment) studies provide additional evidence regarding
25 chronic Os exposures in closed forest canopy scenarios including interspecies interactions such
26 as decreased growth of branches and root mass in sensitive species. Also, lichen and moss
27 communities on trees monitored in FACE sites have been shown to undergo species shifts when
28 exposed to 03. In addition, recent available data from annual field surveys conducted by the
29 USFS to assess foliar damage to selected tree species is available. In light of this new scientific
30 information, we are including additional analyses, such as combining the USFS data with recent
-------
1 air quality data to determine the incidence of visible O?, damage occurring across the U.S. at air
2 quality levels that meet or are below the current standard. Some of these analyses are not
3 included in this first draft REA, but will be included in the second draft REA. To the extent
4 warranted, based on new information regarding 63 effects on forest trees, both qualitative and
5 quantitative assessments are included in an effort to place both the estimates of risk from more
6 recent long-term studies and historic shorter-term studies in the context of ecosystem services.
7 Additional information relevant to vegetation risk assessments available includes that
8 regarding the interactions between elevated Os and CO2 with respect to plant growth and how
9 these interactions might be expected to be modified under different climatic conditions, and
10 potential reactions of O?, with chemicals released by plants to attract pollinators that could
11 decrease the distance the floral "scent trail" travels and potentially change the distance
12 pollinators have to travel to find flowers. The REA also provides an assessment of impacts
13 occurring in designated habitat for threatened or endangered species.
14 To the extent warranted, qualitative and/or quantitative assessments of ecosystem
15 services impacted by Os are considered to inform the current review. For example, the
16 ecosystem services evaluation in this review includes tree biomass and crop analyses, and where
17 possible includes impacts on ecosystem services such as impacts on biodiversity, biological
18 community composition, health of forest ecosystems, aesthetic values of trees and plants and the
19 nutritive quality of forage crops. Carbon sequestration is another important ecosystem service
20 (regulating) that may be affected by 63 damage to vegetation. New preliminary evidence of 63
21 effects on the ability of pollinators to find their target is also of special interest with respect to the
22 possible implication for ecosystem services. Impairment of the ability of pollinators to locate
23 flowers could have broad implications for agriculture, horticulture and forestry.
24 We are using the Forest and Agricultural Sector Optimization Model Greenhouse Gas
25 version (FASOM) to assess the economic impacts of 03 damage to forests, taking into account
26 the tradeoffs between land use for forestry and agricultural. FASOM is a dynamic, non-linear
27 programming model designed for use by the EPA to evaluate welfare benefits and market effects
28 of carbon sequestration in trees, understory, forest floor, wood products and landfills that would
29 occur under different agricultural and forestry scenarios. We use FASOM to model damage by
30 03 to the agriculture and forestry sectors and quantify how Os-exposed vegetation affects the
31 ecosystem service of carbon sequestration. See Appendix X for details of the model and
3-7
-------
1 methodology. [An appendix covering details of the model and methodology will be provided in
2 supplemental materials.]
3 3.2.1 Air Quality Considerations
4 Air quality analyses are necessary to inform and support welfare-related assessments. The
5 air quality analyses for this review build upon those of the ISA and include consideration of: (1)
6 summaries of recent ambient air quality data, (2) estimation approaches to extrapolate air quality
7 values for rural areas without monitors as well as federally designated Class I natural areas
8 important to welfare effects assessment, (3) air quality simulation procedures that modify recent
9 air quality data to reflect changes in the distribution of air quality estimated to occur after just
10 meeting current or alternative Os standards. . In addition to updating air quality summaries
11 since the last review, these air quality analyses include summaries of the most currently available
12 ambient measurements for the current and potential alternate secondary standard forms, and
13 comparisons among them . These air quality analyses use monitor data from the AQS database
14 (which includes National Park Service monitors) and the CASTNET network. In the last review,
15 the vegetation exposure analysis used a spatial interpolation technique to create an interpolated
16 air quality surface to fill in the gaps in ambient monitoring data, especially those left by a sparse
17 rural monitoring network in the western United States. In this review, additional approaches that
18 potentially could be used to fill in the gaps in the rural monitoring network, as well as
19 opportunities for enhancing the fusion of monitoring and modeled Os data, are explored.
20 As part of the air quality analyses supporting the assessments, it is necessary to adjust recent
21 63 air quality data to simulate just meeting the current standard and any alternative 63 standards.
22 In this first draft REA, consistent with the previous review, we are using a quadratic air quality
23 rollback approach (U.S. EPA, 2007b), but we are evaluating alternative air quality simulation
24 procedures for use in simulating just meeting the current and alternative standards for the second
25 draft REA.
26 3.2.2 National Os Exposure Surface
27 Since the last review, little has changed in terms of the extent of monitoring coverage in
28 non-urban areas. We consider both past and alternative approaches for generating estimates of
29 national Os exposures in an effort to continue enhancing our ability to characterize exposures in
-------
1 these non-monitored areas. The vegetation exposure assessments conducted include assessments
2 of recent air quality, air quality associated with just meeting the current standard and, for the
3 second draft REA, any alternative standards that might be considered.
4 In addition, given the importance of providing protection for sensitive vegetation in areas
5 afforded special protections, such as in federally designated Class I natural areas, we may also
6 consider alternative sources of 63 exposure information for those types of sites. For example,
7 portable Os monitors are being deployed in some national parks and a current exploratory study
r\
8 is underway to measure Os concentration variations with gradients in elevation. Information
9 from these monitors could potentially inform our understanding of uncertainties associated with
10 assessing O?, distribution patterns in complex terrain and high elevations. New exposure data
11 that would inform this assessment will be considered where appropriate.
12 To generate a national 63 exposure surface, staff is considering several interpolation
13 methods. We have used a previously modeled 63 surface generated by the CMAQ model based
14 on 2005 emissions at a 12 km grid resolution in conjunction with monitor data (2004-2006) to
15 create a fused surface with the Modeled Attainment Test Software (MATS).3 We have also used
16 the Voronoi Neighbor Averaging (VNA) interpolation method in the BenMAP model (Abt
17 Associates, Inc., 2010) to create a national 03 surface from more recent monitor data (e.g., 2008-
18 2010).4 Staff will also evaluate alternate interpolation methods and sources of air quality data to
19 assess which option is most appropriate given the analysis requirements, desire for consistency
20 with the health risk assessment, and available resources.
21 In order to generate the national 63 surface in terms of a particular index, the monitored
22 data and CMAQ model outputs that form the basis for the interpolation need to be characterized
23 in terms of that index. At a minimum, staff plans to generate the national surface in terms of the
24 current secondary standard. Staff recognizes that additional indices may be selected for further
25 evaluation upon review of the information contained in the ISA and may perform additional air
26 quality analyses based on those indices. Any expanded evaluation of additional indices would be
27 contained and discussed in the Policy Assessment.
For more information on portable ozone monitors in National Parks, please see
http://www.nature.nps.gov/air/studies/portO3.cfm
3 More information on CMAQ is available at http://www.epa.gov/amad/CMAQ/index.html. More information on
MATS is available at http://www.epa.gov/scramOOl/modelingapps_mats.htm.
4 More information on the VNA method in BenMAP is available at
http://www.epa.gov/air/benmap/models/BenM APManualAugust2010.pdf
3-9
-------
1 In conjunction with the health risk assessors, staff is currently considering various
2 approaches to simulate just meeting the current and alternative standards, including the quadratic
3 air quality "rollback" adjustment that was used in the last review (Johnson, 1997) and variations
4 of the proportional adjustment method. However for this first draft we have used the eVNA
5 approach for the rollback adjustment. In addition, we are currently investigating methods for
6 generating adjusted air quality in non-monitored areas.
7 The national Os surface, depicted as a GIS layer, provides the exposures needed as input to
8 the crop and tree seedling risk and ecosystem service assessments described in subsequent
9 sections.
10 3.3 ECOLOGICAL EFFECTS OF EXPOSURE
11 3.3.1 National Scale Assessment
12 3.3.1.1 Tree Seedling Concentration-Response Functions
13 We are analyzing the 11 OTC tree seedling C-R functions identified and assessed in the
14 2007 Os Staff Paper in terms of the current exposure metrics. This analysis enabled direct
15 evaluation of estimated seedling biomass loss values expected to occur under air quality
16 exposure scenarios expressed in terms of recent air quality and after simulation of just meeting
17 current the standard.
18 3.3.1.2 Estimation of Biomass Loss for Tree Seedlings
19 In the 2007 O^ Staff Paper, information on tree species growing regions was derived from
20 the USD A Atlas of United States Trees (Little, 1971). We are using more recent information
21 from the USDA Forest Service FIA database in order to update growing ranges for the 11 tree
22 species studied by NHEERL-WED. The national Os surface is combined with the C-R function
23 for each of the tree seedling species and information on each tree species growing region to
24 produce estimates of biomass loss for each of the 11 tree seedling species. We are also including
25 an additional analysis incorporating the Importance Values derived using FIA data. From this
26 information, GIS maps are generated depicting biomass loss for each species for each air quality
27 scenario.
3-10
-------
1 3.3.2 Case Study Areas
2 In order to assess the ecological effects of Os staff will analyze ecosystem level effects in
3 several case study areas. These areas have been selected to allow a more refined assessment of
4 the extent of foliar injury, biomass loss and welfare related services. Criteria that were used to
5 select case study areas include:
6 • Occur in areas expected to have elevated levels of 63 where ecological effects might be
7 expected to occur.
8 • Availability of vegetation mapping including estimates of species cover.
9 • Geographic coverage representing a cross section of the nation, including urban and
10 natural settings.
11 • Occurrence of 63 sensitive species and/or species for which 63 concentration-response
12 curves have been generated.
13 3.3.2.1 Estimation of Vegetation Effects in National Parks
14 The National Parks provide several potential case study areas. The United States
15 Geological Survey (USGS) in conjunction with the National Park Service (NPS) is actively
16 creating maps of the vegetation communities within the National Parks
17 (http://biology.usgs.gov/npsveg/index.html). This provides a consistent vegetation map to
18 compare across park units, which includes species coverage data. The NPS has also generated a
19 comprehensive list of plant species that are known to exhibit foliar injury at ambient 63 levels
20 (Porter, 2003).
21 We have selected Great Smoky Mountains National Park, Rocky Mountain National
22 Park, and Sequoia/Kings National Park. All three of these park units occur in areas with elevated
23 ambient 63 levels, have vegetation maps, and have species that are considered 63 sensitive. We
24 considered including Acadia National Park however it was determined not to fit our selection
25 criteria for OT, exposure.
26 The NPS vegetation maps are compared, using GIS, to the national 03 surface to provide
27 an overall estimate of foliar damage and total biomass loss. Potential ecological metrics that are
28 being calculated include:
29 • Percent of vegetation cover affected by foliar injury.
3-11
-------
1 • Percent of trails affected by foliar injury.
2 • Estimate of species specific biomass loss within the case study area.
3 3.3.2.2 Estimation of Effects in Urban Areas
4 Several urban areas nationally have extensive habitat management plans that include
5 resource and vegetation mapping. These data are not as consistent or as readily available as the
6 NFS units but in some cases can provide adequate vegetation maps in regions where O^ sensitive
7 species occur. We are using the iTree model developed by the U.S. Forest Service to estimate
8 impacts on vegetation in Atlanta, Baltimore, Syracuse, the Chicago region, and the urban areas
9 of Tennessee. We are presenting preliminary results for model runs representing current ambient
10 conditions and runs simulating just meeting the current standard in this draft of the REA. Model
11 runs simulating any alternative standards that may be considered will be presented in the second
12 draft REA. [The first draft results and an appendix with details regarding the model and
13 methodology will be included in supplemental materials.]
14 3.4 ECO SYSTEM SERVICES EVALUATION
15 One of the objectives of the risk assessment for a secondary NAAQS is to quantify the
16 risks to public welfare. The Risk and Exposure Assessment for Review of the Secondary
17 National Ambient Air Quality Standards for Oxides of Nitrogen and Oxides of Sulfur (U.S. EPA,
18 2009) has detailed discussions of how ecosystem services and public welfare are related and how
19 a services framework may be employed to evaluate effects on welfare. We have identified the
20 ecosystem services associated with the ecological effects described in Chapter 5 of this
21 document for the national scale assessment and the more refined case study areas. These
22 services may be characterized as: supporting services that are necessary for all other services
23 (e.g., primary production); cultural services including existence and bequest values, aesthetic
24 values, and recreation values, among others; provisioning services (e.g., food and timber); and
25 regulating services such as climate regulation or flood control. Specific services to be evaluated
26 are discussed in the following sections.
3-12
-------
1 3.4.1 National Scale Assessment
2 Depending on data and resource availability, we are attempting to develop an estimate of
3 ecosystem service impacts broadly across the United States for selected cultural, regulating, and
4 provisioning services.
5 3.4.1.1 Cultural Services
6 We are using GIS mapping developed for the ecological effects analysis to illustrate
7 where effects may be occurring and relate those areas to national scale statistics for recreational
8 use available through the National Survey of Fishing, Hunting, and Wildlife-Associated
9 Recreation (U.S. DOT, 2007) and the National Survey on Recreation and the Environment
10 (USDA,2012) . The resulting estimates of service provision are then scaled to the current
11 population and values assigned using existing meta-data on willingness to pay from the
12 Recreation Values Database available at: http://recvaluation.forestry.oregonstate.edu/
13 We are aware that these estimates are limited to current levels of service provision and
14 provide a snapshot of the overall magnitude of services potentially affected by 63 exposure. At
15 this time estimates of service loss due to Os exposure is beyond the available data and resources;
16 however, estimates of the current level of services would have embedded within them the current
17 losses in service due to OsOs exposure.
18 3.4.1.2 Regulating Services
19 The regulating services associated with O?, exposure include fire regimes and fire
20 recovery due to 63 effects on community composition and diversity, and fuel loading due to
21 early senescence and insect attack. There is data available through the CAL-FIRE on fire
22 incidence, risk, and expenditures related to fires in California.
23 We are considering using the PnET model to estimate impacts on the hydrologic cycle for
24 the second draft of this document. We considered the DLEM model however the resources
25 required proved prohibitive.
26 3.4.1.3 Provisioning Services
27 Below we outline potential methods for assessing the provisioning services associated with
28 crop yield loss and tree biomass loss, which are consistent with the methods from the previous
29 review.
30 Estimation of Yield Loss and Economic Valuation for Timber and Crops - The FASOM model
31 has been utilized recently in many evaluations of effects on the timber and agriculture market
3-13
-------
1 sectors. We are using FASOM to assess the economic impacts of O?, damage to forests and
2 agricultural crops jointly. FASOM is a dynamic, non-linear programming model designed for
3 use by the EPA to evaluate welfare benefits and market effects of 63 induced biomass loss in
4 trees that would occur under different agricultural and forestry scenarios. It is possible to use
5 FASOM to model damage by 63 to the agriculture and forestry sectors and quantify how 63-
6 exposed vegetation affects the provision of timber and crops. [An appendix with details of the
7 model and methodology will be provided in supplemental materials.]
8 FASOM has been used to calculate the economic impacts of yield changes between the
9 current ambient conditions and simulated 'just meet' scenarios for a base year. This approach
10 will also be used to calculate the economic valuation of any alternative standards under
11 consideration in the second draft.
12 3.4.1.4 Supporting Services
13 The supporting services associated with the vegetation effects of 63 exposure include
14 potential impacts on net primary productivity, and community composition. We considered
15 using the DLEM model to estimate impacts on net primary productivity however this proved
16 prohibitive in terms of resource availability. For the second draft we are exploring the possibility
17 of using the PnET model to estimate these service impacts.
18 3.4.2 Case Study Analysis
19 3.4.2.1 National Park Areas
20 We are using GIS mapping produced for the ecological effects analysis to illustrate where
21 effects may be occurring as a starting point to illustrate and, if possible, quantify the ecosystem
22 services at potential risk. These are primarily, in national parks, cultural values that include
23 existence, bequest and recreational values. We also overlay the ecological effects maps with data
24 on where hiking trails, campgrounds, or other park amenities are found to intersect potentially
25 affected areas. We then relate those areas to case study specific statistics for recreational use
26 available through the National Park Service. In addition, we have described the other nonuse
27 values associated with national parks including existence and bequest values. For the resulting
28 estimates of service provision values are then assigned using existing meta-data on willingness to
29 pay from Kaval and Loomis (2003). We are aware that these estimates will be limited to current
3-14
-------
1 levels of service provision. At this time estimates of service loss due to O?, exposure may be
2 beyond the available data and/or resources for many if not all ecosystem services listed above.
3 3.4.2.2 Urban Areas
4 We are using the i-Tree model to assess effects on ecosystem services provided by urban
5 forests, pollution removal, and carbon storage and sequestration. The i-TREE model is a publicly
6 available peer-reviewed software suite developed by the U.S. Forest Service and its partners to
7 assess the ecosystem service impacts of urban forestry (available here:
8 http://www.itreetools.org/). We are collaborating with the U.S. Forest Service to vary the tree
9 growth metric in the model, which allows us to assess the effects of O?, exposure on the ability of
10 the forests in the selected case study area to provide the services enumerated by the model. See
11 Appendix 6A for a description of the model and methodology. [Preliminary results will be
12 provided in supplemental materials.]
13 3.5 UNCERTAINTY AND VARIABILITY
14 An important issue associated with any ecological risk assessment is the characterization
15 of uncertainty and variability. Variability refers to the heterogeneity in a variable of interest that
16 is inherent and cannot be reduced through further research. For example, there may be
17 variability among C-R functions describing the relation between Os and vegetation injury across
18 selected study areas. This variability may be due to differences in ecosystems (e.g., diversity,
19 habitat heterogeneity, and rainfall), levels and distributions of Os and/or co-pollutants, and/or
20 other factors that vary either within or across ecosystems.
21 Uncertainty refers to the lack of knowledge regarding both the actual values of model input
22 variables (parameter uncertainty) and the physical systems or relationships (model uncertainty -
23 e.g., the shapes of concentration-response functions). In any risk assessment, uncertainty is,
24 ideally, reduced to the maximum extent possible, through improved measurement of key
25 parameters and ongoing model refinement. However, significant uncertainty often remains and
26 emphasis is then placed on characterizing the nature of that uncertainty and its impact on risk
27 estimates. The characterization of uncertainty can include both qualitative and quantitative
28 analyses, the latter requiring more detailed information and often, the application of sophisticated
29 analytical techniques.
3-15
-------
1 While the goal in designing a quantitative risk assessment is to reduce uncertainty to the
2 extent possible, with variability the goal is to incorporate the sources of variability into the
3 analysis approach to insure that the risk estimates are representative of the actual response of an
4 ecosystem (including the distribution of that adverse response across the ecosystem). An
5 additional aspect of variability that is pertinent to this risk assessment is the degree to which the
6 set of selected case study areas provide coverage for the range of (Vrelated ecological risk
7 across the U.S.
8 For this first draft we have not included detailed analyses of uncertainty or variability. For
9 the second draft of this document we plan to more fully differentiate variability and uncertainty
10 in the design of the risk assessment to more clearly address (a) the extent to which the risk
11 estimates represent the distribution of ecological impacts across ecosystems, including impacts
12 on more sensitive species, and (b) the extent to which risk estimates are impacted by key sources
13 of uncertainty which could prevent a clear differentiation between regulatory alternatives based
14 on risk estimates.
3-16
-------
1 4 AIR QUALITY CONSIDERATIONS
2 4.1 INTRODUCTION
3 Air quality information is used in the welfare risk and exposure analyses, described in
4 Chapters 5 and 6, to assess risk and exposure resulting from recent Os concentrations, as well as
5 to estimate the relative change in risk and exposure resulting from adjusted Os concentrations
6 after simulating just meeting the current Os standard of 0.075 ppm. To complete these analyses,
7 ambient monitoring data is provided for all AQS monitors in the U.S. for several relevant metrics
8 for 2006-2010. In addition, a national-scale spatial surface is generated that estimates W126
9 concentrations throughout the U.S. for 2006-2008 and for simulating just meeting the current Os
10 standard of 0.075 ppm. This chapter describes the air quality information used in these analyses,
11 providing an overview of monitoring data and air quality (section 4.2) as well as an overview of
12 air quality inputs to the welfare risk and exposure assessments (section 4.3).
13 4.2 OVERVIEW OF O3 MONITORING AND AIR QUALITY
14 To monitor compliance with the NAAQS, state and local monitoring agencies operate Os
15 monitoring sites at various locations, depending on the size of the area and typical peak 03
16 concentrations (US EPA, 2012, sections 3.5.6.1, 3.7.4). In 2010, there were 1,250 State and
17 Local Os monitors reporting concentrations to EPA (US EPA, 2012, Figures 3-21 and 3-22).
18 The minimum number of Os monitors required in a Metropolitan Statistical Area (MSA) ranges
19 from zero, for areas with a population under 350,000 and with no recent history of an Os design
20 value greater than 85% of the NAAQS, to four, for areas with a population greater than 10
21 million and an Os design value greater than 85% of the NAAQS.l For areas with required Os
22 monitors, at least one site must be designed to record the maximum concentration for that
23 particular metropolitan area. Since Os concentrations decrease significantly in the colder parts of
24 the year in many areas, Os is required to be monitored only during the "Os season," which varies
25 by state (US EPA, 2012, section 3.5.6 and Figure 3-20).2 Figure 4-1 shows the location and 8-h
26 Os design values (4th highest 8-h daily max Os concentration occurring within a three-year
27 period) for all available monitors in the US for the 2008-2010 period.
28
29
30
lrThe current monitor and probe siting requirements have an urban focus and do not address siting in non-urban, rural
areas. States may operate O3 monitors in non-urban or rural areas to meet other objectives (e.g., support for research
studies of atmospheric chemistry or ecosystem impacts).
2Some States and Territories operate O3 monitors year-round, including Arizona, California, Hawaii, Louisiana,
Nevada, New Mexico, Puerto Rico, Texas, American Samoa, Guam and the Virgin Islands.
4-1
-------
1
2
3
4
5
6
7
8
9
10
»*0
HAWAII
O
8-hour Ozone Design Values, 2008-2010
• 40-65 ppb (249 Sites)
O 66-70 ppb (309 Sites)
O 71-75 ppb (303 Sites)
• 76-90 ppb (168 Sites)
• 91-112 ppb (36 Sites)
1
PUERTO RICO
'•
Figure 4- 1 Individual monitor 8-h daily max Os design values displayed for the 2008-
2010 period (U.S. EPA, 2012, Figure 3-52A)
In 2010, there were approximately 112 monitoring sites being operated in rural areas.
These sites included 15 National Core (NCore) monitors, 80 Clean Air Status and Trends
Network (CASTNET) monitors, and 17 Portable Oi Monitoring Systems (POMS) network
monitors operated by the National Park Service (NPS). The location of these monitors is shown
in Figure 4-2.
4-2
-------
1
2
3
Alaska
D 231 3JO
D S61ID -\T. III!-;
.• Rural HCore
o MPS POMS
• CASTNET
i ma lines
Puerto Rico &
Virgin Islands
D 2550 ItB lilts
Figure 4- 2 U.S. Rural NCore, CASTNET and NFS POMS O3 sites in 2010 (U.S. EPA,
2012, Figure 3-22)
5 4.3 OVERVIEW OF AIR QUALITY INPUTS TO RISK AND EXPOSURE
6 ASSESSMENTS
7 The air quality information input into the welfare risk and exposure assessments includes
8 recent air quality measurement data from the years 2006-2010, as well as a national-scale
9 "fused" spatial surface of air quality data for recent air quality, 2006-2008, and adjusted to
10 reflect just meeting the current Os standard of 0.075 ppm. In this section, we summarize these air
11 quality inputs and discuss the methodology used to simulate air quality to meet the current
12 standard. More details on these data and methodologies can be found in Wells et al. (2012).
13
14 4.3.1 Recent Air Quality
15 The air quality monitoring data used to inform the first draft Oj Risk and Exposure
16 Assessments were hourly Os concentrations collected between 1/1/2006 and 12/31/2010 from all
17 US monitors meeting EPA's siting, method, and quality assurance criteria in 40 CFR Part 58.
4-3
-------
1 These data were extracted from EPA's Air Quality System (AQS) database3 on June 27, 2011.
2 Regionally concurred exceptional event data (i.e. data certified by the monitoring agency to have
3 been affected by natural phenomena such as wildfires or stratospheric intrusions, and concurred
4 upon by the EPA regional office) were not included in the assessments. However, concurred
5 exception events were rare, accounting for less than 0.01% of the total observations. All
6 concurred exceptional events in 2006-2010 were related to wildfires in California in 2008. There
7 were no concurrences of exceptional event data for stratospheric intrusions in 2006-2010.
8 4.3.1.1 Ambient Measurements and Air Quality Metrics
9 EPA focused the analysis in the welfare exposure and risk assessment on the W126 Os
10 exposure metric. The W126 metric is a seasonal aggregate of hourly Os concentrations, designed
11 to measure the cumulative effects of O^ exposure on vulnerable plant and tree species. The
12 metric uses a logistic weighting function to place less emphasis on exposure to low
13 concentrations and more emphasis on exposure to high concentrations (Lefohn et al, 1988).
14 The first step in calculating W126 values was to sum the hourly Os concentrations within
15 each month, resulting in monthly index values. Since most plant and tree species are not
16 photochemically active during nighttime hours, only O^ concentrations observed during daytime
17 hours (defined as 8:00 AM to 8:00 PM local time) were included in the summations. The
18 monthly W126 index values were calculated as follows:
AT 19 /">
19 Monthly W\ 26 = Y Y &
t^l + 4403*exp(-126*CdJ
20 where TV is the number of days in the month,
21 d is the day of the month (d = 1, 2, ..., N),
22 h is the hour of the day (h = 0, 1, ..., 23),
23 Cdh is the Os concentration observed on day d, hour h, in parts per million.
24 Next, the monthly W126 index values were adjusted for missing data. If Nm is defined as
25 the number of daytime O?, concentrations observed during month m (i.e. the number of terms in
26 the monthly index summation), then the monthly data completeness rate is Vm = Nm /12 * N.
27 The monthly index values were adjusted by dividing them by their respective Vm. Monthly index
28 values were not computed if the monthly data completeness rate was less than 75% (Vm < 0.75).
3 EPA's Air Quality System (AQS) database is a state-of-the-art repository for many types of air quality and related
monitoring data. AQS contains monitoring data for the six criteria pollutants dating back to the 1970's, as well as more recent
additions such as air toxics, meteorology, and quality assurance data. At present, AQS receives Os monitoring data collected
hourly from over 1,300 monitors, and quality assured by one of over 100 state, local, or tribal air quality monitoring agencies.
4-4
-------
1 Finally, annual W126 index values were computed as the maximum sum of their
2 respective adjusted monthly index values occurring in three consecutive months (January -
3 March, February - April, etc.). Three-month periods spanning two years (November - January,
4 December - February) were not considered because the seasonal nature of O?, dictates that it is
5 very unlikely for the maximum values to occur at that time of year. The W126 metric was
6 analyzed for each individual year of 2006 to 2008 and for the three year period of 2006-2008.
7 For the specific application of the Kohut analysis, N100 and SUM06 metric were also
8 computed. The procedures used to calculate N100 and SUM06 values are similar to the
9 calculation of the W126 metric that is described above. Hourly O^ concentrations are summed
10 within each month, resulting in monthly index values, and only Os concentrations observed
11 during daytime hours (defined as 8:00 AM to 8:00 PM local time) were included in the
12 summations. The monthly N100 and SUM06 values were calculated as follows:
O, // Cdh < 0. 1 OOpprn]
'•>*>> ^
13 Monthly
ifCdh>0.\Wppm
N 19
14 Monthly SUM06 = X X max(°> (CA ~ 0.060))
d=\ h=S
15 The monthly N100 and SUM06 values were adjusted for missing data as described above for the
16 W126 metric. Annual N100 and SUM06 values were computed as the maximum sum of their
17 respective adjusted monthly index values occurring in three consecutive months (January -
18 March, February - April, etc.). Three-month periods spanning two years (November - January,
19 December - February) were not considered because the seasonal nature of 63 dictates that it is
20 very unlikely for the maximum values to occur at that time of year.
21 The N100 and SUM06 metrics were calculated for each individual year for all 5 years
22 (2006 to 2010) and used in the Kohut analysis, which is discussed in more detail in Chapter 5. In
23 addition, the W126 and N100 value was calculated for 3-month and 7-month values for the
24 Kohut analysis and analyzed for each individual year of 2006 to 2010.
25 4.3.1.2 National-scale Air Quality Inputs
26 In addition to ambient monitoring data, the welfare risk and exposure assessment also
27 analyzed a national scale spatial surface of W 126 for the three-year period of 2006-2008 and for
28 each individual year: 2006, 2007 and 2008. This analysis employed a data fusion approach to
29 take advantage of the accuracy of monitor observations and the comprehensive spatial
30 information of the CMAQ modeling system to create a national -scale "fused" spatial surface of
3 1 seasonal average 03. The spatial surface is created by fusing 2006-2008 measured 03
32 concentrations with the 2007 CMAQ model simulation, which was run for a 12 km gridded
33 domain, using the EPA's Model Attainment Test Software (MATS; Abt Associates, 2010),
4-5
-------
1 which employs the enhanced Voronoi Neighbor Averaging (eVNA) technique (Timin et al.,
2 2010) enhanced with information on the spatial gradient of Os provided by CMAQ results. The
3 2006-2008 W126 national-scale "fused" spatial surface is shown in Figure 4-3. More details on
4 the ambient measurements and the 2007 CMAQ model simulation, as well as the spatial fusion
5 technique, can be found in Wells et al. (2012).
6
7 4.3.2 Air Quality After Simulating "Just Meeting" Current O3 Standard
8 In addition to 2006-2008 air quality concentrations for the W126 metric, the risk and
9 exposure assessments also consider the relative change in risk and exposure when considering
10 the distribution of W126 after simulating "just meeting" the current Os standard of 0.075 ppm.
11 The sections below summarize the methodology applied for this first draft REA to simulate just
12 meeting the current NAAQS by "rolling back" the baseline distribution of recent Os
13 concentrations. More details on these inputs are provided in Wells et al. (2012).
14
15 4.3.2.1 Methods
16 The "quadratic rollback" method was used in the previous Os NAAQS review to adjust
17 ambient Os concentrations to simulate minimally meeting current and alternative standards (U.S.
18 EPA, 2007). As the name implies, quadratic rollback uses a quadratic equation to reduce high
19 concentrations at a greater rate than low concentrations. The intent is to simulate reductions in
20 Os resulting from unspecified reductions in precursor emissions, without greatly affecting
21 concentrations near ambient background levels (Duff et al., 1998).
22 Two independent analyses (Johnson, 2002; Rizzo, 2005; 2006) were conducted to
23 compare quadratic rollback with other methods such as linear (proportional) rollback and
24 distributional (Weibull) rollback. Both analyses used different rollback methods to reduce
25 concentrations from a high Os year to simulate levels achieved during a low Os year, then
26 compared the results to the ambient concentrations observed during the low Os year. Both
27 analyses concluded that the quadratic rollback method resulted in an 8-hour Os distribution most
28 similar to that of the ambient concentrations.
29 In this review, quadratic rollback was used to reduce Os concentrations in all areas of the
30 U. S. with violating monitors to just meet the current NAAQS of 0.075 ppm (75 ppb). To do this,
31 a hierarchical method was used to group all monitors in the U.S. into hypothetical "non-
32 attainment" areas (Wells et al., 2012). For each of these areas, quadratic rollback was then
33 employed to simulate just meeting the current standard. Hourly Os concentrations were reduced
34 so that the highest design value in each area was exactly 75 ppb, the highest value meeting the
35 NAAQS. Finally, the 2006-2008 W126 metric was calculated from the hourly rollback
4-6
-------
1 concentrations. It should be noted that O?, concentrations were only adjusted relative to the other
2 monitors included in the hypothetical "non-attainment" area. In this way, areas with all monitors
3 below 75 ppb would not have been affected by this rollback methodology and the Os
4 concentrations in those areas would not have changed. This was true even when these monitors
5 were very close to, but outside of, other hypothetical "non-attainment" areas that were adjusted
6 to simulate just meeting the current standard.
7 To generate a national-scale spatial surface that represents 2006-2008 W126
8 concentrations when attaining the current NAAQS, the spatial surface for 2006-2008 recent air
9 quality was adjusted to reflect the rolled back W126 monitor concentrations. To do this, the
10 rolled back W126 monitor values were inserted into the spatial surface at the monitor locations
11 and the W126 surface was smoothed using the Voronoi Neighbor Averaging (VNA) spatial
12 averaging technique to minimize any sharp gradients between the national-scale spatial surface
13 that represents 2006-2008 W126 concentrations and the rollback W126 monitor concentrations.
14 This is described in more detail in Wells et al. (2012).
15 4.3.2.2 Results
16 Figure 4-3 shows the national-scale 2006-2008 W126 spatial "fused" surface created as
17 described in Section 4.3.1.1, and Figure 4-4 shows the national-scale 2006-2008 W126 surface
18 that reflects simulation of just meeting the current standard of 0.075 ppm. Figure 4-5 shows the
19 difference between the two spatial surfaces, and shows how W126 changed when simulating just
20 meeting the current standard. The state of California was most affected by the rollback, with
21 average changes in W126 of around 20. Other areas with notable changes include the areas
22 around: Atlanta, Charlotte, Denver, Phoenix, Salt Lake City and the area between Washington,
23 D.C. and Boston (all areas that had relatively high 8-hour 63 concentrations above the current
24 standard).
25
4-7
-------
1
2
Figure 4- 3 "Fused" national-scale surface of W126 metric, 2006-2008
4-8
-------
1
2
3
4
Figure 4- 4 "Fused" national-scale surface of W126 metric for 2006-2008, adjusted for
simulating just meeting the current standard of 0.075 ppm.
4-9
-------
1.69-2.96
2.97-4.61
4.62-6.84
6.85 -10.07
10.08-14.68
| 14.69-20.24
I 20.25 - 26.68
26 69 - 38.57
1
2
3
4
5
Figure 4- 5 Difference between the "fused" national-scale surfaces of W126 for 2006-
2008 and for 2006-2008 adjusted for simulating just meeting the current
standard of 0.075 ppm.
4-10
-------
5 ECOLOGICAL EFFECTS
2 5.1 INTRODUCTION
3 This chapter presents the results of ecological risk analyses based on the causal and likely
4 causal effects of O3 on vegetation and ecosystems described in the ISA. Recent studies reviewed
5 in the O3 ISA (U.S. EPA, 2012a) support and strengthen the findings reported in the 2006 O3
6 AQCD (U.S. EPA, 2006). The most significant new body of evidence since the 2006 O3 AQCD
7 comes from research on molecular mechanisms of the biochemical and physiological changes
8 observed in many plant species in response to O3 exposure. These newer molecular studies not
9 only provide very important information regarding the many mechanisms of plant responses to
10 O3, they also allow for the analysis of interactions between various biochemical pathways which
11 are induced in response to O3. However, many of these studies have been conducted in artificial
12 conditions with model plants, which are typically exposed to very high, short doses of O3 and are
13 not quantifiable as part of this risk assessment, which is focused on recent ambient levels of O3
14 exposure and O3 levels simulated to meet current and alternative O3 standards.
15 The causal findings reported in the ISA based on the current science are summarized in
16 Table 5- 1. This table includes both causal and likely causal effects. Two of the effects,
17 alteration of below-ground biogeochemical cycles and alteration of terrestrial communities are
18 not analyzed directly in this review. However both can be inferred as components of the i-Tree
19 and FASOM models discussed in Chapter 6 and the scaled-biomass loss analyses presented in
20 this chapter.
21 Table 5-1 Summary of Os causal determinations for vegetation and ecosystem effects
22 (modified from Table 9-18 in the ISA)
Vegetation and Ecosystem Effect
Visible Foliar Injury Effects on
Vegetation
Reduced Vegetation Growth
Reduced Productivity in Terrestrial
Ecosystems
Reduced Carbon (C) Sequestration in
Terrestrial Ecosystems
Conclusions from 2012 ISA
Causal Relationship
Causal Relationship
Causal Relationship
Likely Causal Relationship
2012 REA
Analyzed in this chapter at a National-
scale and within NFS Units (Section
5.3.2) and NFS case study areas
(section 5.4)
Analyzed in this chapter at a National-
scale and within NFS case study areas
(section 5.3)
Analyzed in Chapter 6 using pNET-CN
(pending)
Analyzed in Chapter 6 using pNET-CN
(pending) and i-TREE (section 6.X)
5-1
-------
Vegetation and Ecosystem Effect
Reduced Yield and Quality of
Agricultural Crops
Alteration of Terrestrial Ecosystem
Water Cycling
Alteration of Below-ground
Biogeochemical Cycles
Alteration of Terrestrial Community
Composition
Conclusions from 2012 ISA
Causal Relationship
Likely Causal Relationship
Causal Relationship
Likely Causal Relationship
2012 REA
Yield loss data are included in the
FASOM model (section 6.X), but
effects on agricultural crops are not a
focus of this review
Analyzed in Chapter 6 using pNET-CN
(pending) Relationship
Not analyzed directly in this review
Not analyzed directly in this review
2 5.2 RELATIVE BIOMASS LOSS
3 The previous O3 AQCDs (U.S. EPA, 1996, 2006) and current O3 ISA (U.S. EPA, 2012)
4 concluded that there is strong and consistent evidence that ambient concentrations of O3 decrease
5 photosynthesis and growth in numerous plant species across the U.S.
6 Meta-analyses by Wittig et al. (2007, 2009) demonstrate the coherence of O3 effects on
7 plant photosynthesis and growth across numerous studies and species using a variety of
8 experimental techniques. Furthermore, recent meta-analyses have generally indicated that O3
9 reduces C allocation to roots (Wittig et al., 2009; Grantz et al., 2006). Since the 2006 O3 AQCD,
10 several studies were published based on the Aspen FACE experiment using "free air," O3 and
11 CC>2 exposures in a planted forest in Wisconsin. Overall, the studies at the Aspen FACE
12 experimental site were consistent with many of the open-top chamber (OTC) studies that were
13 the foundation of previous O3 NAAQS reviews. These results strengthen our understanding of O3
14 effects on forests and demonstrate the relevance of the knowledge gained from trees grown in
15 OTC studies.
16 The 1996 and 2006 O3 AQCDs relied extensively on results from analyses conducted on
17 commercial crop species under the auspices of the National Crop Loss Assessment Network
18 (NCLAN) and on analyses of tree seedling species conducted by the EPA's National Health and
19 Environmental Effect Laboratory, Western Ecology Division (NHEERL/WED). Results from
20 these studies have been published in numerous publications, including Lee et al. (1994; 1989,
21 1988b, 1987), Hogsett et al. (1997), Lee and Hogsett (1999), Heck et al. (1984), Rawlings and
22 Cure (1985), Lesser et al. (1990), and Gumpertz and Rawlings (1992). Those analyses concluded
23 that a three-parameter Weibull model -
5-2
-------
2 Y = a e'
O
4 Equation 5-1
5
6 is the most appropriate model for the response of absolute yield and growth to Os exposure,
7 because of the interpretability of its parameters, its flexibility (given the small number of
8 parameters), and its tractability for estimation. In addition, if the intercept term, a, is removed,
9 the model estimates relative yield or biomass without any further reparameterization.
10 Formulating the model in terms of relative yield or biomass loss (RBL) as related to the 3-month
11 W126O3 index-
12
13 RBL = 1 - exp[-(W126/ti)p]
14 Equation 5-2
15 is essential in comparing exposure-response across species, genotypes, or experiments for which
16 absolute values of the response may vary greatly. In the 1996 and 2006 63 AQCDs, the two-
17 parameter model of relative yield was used in deriving common models for multiple species,
18 multiple genotypes within species, and multiple locations.
19 Relative biomass loss (RBL) functions for the 11 tree species used in this assessment are
20 presented in Table 5-2 (see the ISA (EPA 2012a) for a more extensive review of the calculation
21 of the C-R functions).
22
5-3
-------
1
2
Table 5- 2 Relative Biomass Loss Functions for Tree Species (modified from Table 9-18
in the ISA)
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Species
Red Maple (Acer rubrum)
Sugar Maple (Acer saccharum)
Red Alder (Alnus rubra)
Tulip Poplar (Liriodendron tulipifera)
Ponderosa Pine (Pinus ponderosa)
Eastern White Pine (Pinus strobus)
Virginia Pine (Pinus virginiand)
Eastern Cottonwood (Populus deltoides)
Quaking Aspen (Populus tremuloides)
Black Cherry (Prunus serotina)
Douglas Fir (Pseudotsuga menzeiesii)
RBL Function
1 - expKWlie/T])15]
TI (ppm)
318.12
36.35
179.06
51.38
159.63
63.23
1714.64
10.10
109.81
38.92
106.83
P
1.3756
5.7785
1.2377
2.0889
1.1900
1.6582
1.0000
1.7793
1.2198
0.9921
5.9631
Figure 5- 1 shows a comparison of W126 median RBL response functions for the tree
species used in this assessment. The figure illustrates how the two parameters affect the shape of
the resulting curves. Differences in the shape of these curves are important for understanding
differences in the analyses presented later in this chapter. The two parameters of the RBL
equation (Equation 5-2) control the shape of the resulting curve. The value of r| in the RBL
function affects the inflection point of the curve and P affects the steepness of the curve. Species
with smaller values of p (e.g. Virginia pine,) or species with r| values which are above the normal
range of ambient W126 measurements (e.g. ponderosa pine, red alder) have response functions
with more gradual and consistent slopes. This results in more constant rate of change in RBL
over a range of 63 exposure consistent with ambient exposure levels.
In contrast, the species with larger p values (e.g. sugar maple, Douglas fir) have response
functions that behave more like thresholds, with large changes in RBL over some ranges of Os
and relatively small changes at other levels. In these cases the "threshold" is determined by the t]
parameter of the model. In the example of eastern cottonwood, P is relatively low, but because r|
is also very low relative to the other species, so the resulting C-R curve has a very steep gradient
relative to other species with similar p values.
5-4
-------
O3C-R Functions
1
2
3
4
5
6
7
8
9
10
11
12
13
14
CO
0
CD
O
CD
o:
C\j
O
q
O
0
20
40
i
60
Eastern Cottonwood
Sugar Maple
Tulip Poplar
Black Cherry
White Pine
Quaking Aspen
Douglas Fir
Ponderosa Pine
Red Alder
Red Maple
me
80
100
W126
Figure 5- 1 Relative Biomass Loss Functions for 11 Tree Species
5.2.1 Species Level Analyses
5.2.1.1 Individual Species Analyses
The C-R functions listed in Table 5-2 were used to generate RBL surfaces for the 11 trees
species using GIS (ESRI®, ArcMAP™ 10). A surface was created using recent ambient 63
conditions and a scenario with 63 levels rolled back to simulate just meeting the current 8 hr
secondary standard (see Chapter 4 for a more detailed description of the O3 surfaces). The recent
ambient conditions are based on monitored data from the years 2006 to 2008 and for the
remainder of this analysis we will refer to that surface as "ambient". Two species are presented
here to illustrate the results, ponderosa pine (Figure 5-2 and Figure 5-4) and tulip poplar (Figure
5- 3 and Figure 5- 5). RBL surfaces for the remaining 8 species are presented in Appendix 5A. It
is important to note that these maps represent the RBL value for one tree species within each
5-5
-------
1 CMAQ grid cell represented, so these maps should be interpreted as indicating potential risk to
2 individual trees of that species growing in that area.
3
4
5
6
7
8
9
10
11
Three of the tree species occur entirely in the western U.S.; ponderosa pine, Douglas fir,
and red alder. Ranges for the western species were taken from the U.S. Department of
Agriculture's Atlas of United States Trees (Little, 1971) (Figure 5- 2 and Figure 5- 4). The
western tree species have more fragmented habitats than the eastern species. The areas in souther
California have the highest levels of O3, which can be seen as the very high areas of RBL in
Figure 5-2. The area of high RBL in Figure 5-2 in Idaho is a result of high O3 levels from the
2007 Idaho Forest Fires. This area is still elevated in Figure 5-4 because those areas were not
near areas considered out of attainment, so were not reduced significantly in the scenario just
meeting the current standard.
12
13
14
Ponderosa Pine
0.001 -0.018
0.019-0.035
0.036 - 0.054
0.055 - 0.074
0.075-0.102
0.103-0.150
0.151 -0.210
0.211 -0.294
Figure 5- 2 Relative Biomass Loss of Ponderosa Pine (Pinus ponderosa) seedlings
under recent ambient Os exposure levels (2006 - 2008)
5-6
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Ranges for the eight eastern species were also based on the USDA Ranges (Figure 5- 3
and Figure 5- 5, green outline). Additional work by the northern research station based on Forest
Inventory Analysis data (FIA) was used to update the range for the 8 eastern species (U.S. Forest
Service Climate Change Atlas, http://www.fs.fed.us/nrs/atlas/littlefia/index.html). These updates
can be seen in Figure 5- 3 as areas outside of the green line indicating the Little's range that are
shown to have a RBL value. For this analysis, these values were only used to expand the species
ranges and were not used to indicate absence inside of the Little's range. However, this was done
in the scaled analyses presented in section 5.2.2.
The eastern tree species had less fragmented ranges and areas of elevated RBL that were
more easily attributed to urban areas (e.g. Atlanta, GA and Charlotte, NC) or to the Tennessee
Valley Authority Region.
Potential
Biomass Loss
(Ambient)
0.006
0.043
0.078
0.114
0.149
0.185
0.221
0.256
-0.042
-0.077
-0.113
-0.148
-0.184
-0.220
-0.255
-0.291
Figure 5- 3 Relative Biomass Loss of tulip poplar (Liriodendron tulipifera) seedlings
under recent ambient Os exposure levels (2006 - 2008)
5-7
-------
Ponderosa Pine
\
\
I
1
^-v.
^^^^S^^^X
\
}
1
Potential
Biomass Loss
(Current
Standard)
^H n nm - n ns8
^| 0.039-0.074
0.075-0.111
0.112-0.147
0.148-0.184
0.185-0.221
^H 0.222-0.257
^H 0.258-0.294
1
2
3
Figure 5- 4 Relative Biomass Loss of Ponderosa Pine with O3 exposure rolled back
to meet the current (8-hr) secondary standard.
5-8
-------
Potential
Biomass Loss
(Current
Standard)
0.004
0.043
0.078
0.114
0.149
0.185
0.221
0.256
-0.042
-0.077
-0.113
-0.148
-0.184
-0.220
-0.255
-0.291
2 Figure 5- 5 Relative Biomass Loss of Tulip Poplar with Os exposure rolled back to meet
3 the current (8-hr) secondary standard.
4
5 5.2.1.2 Combined Risk Analysis of Individual Species
6 To assess the combined risk of the 11 tree species, the RBL values were compared
7 between Os exposure scenarios. The comparisons were done on using individual CMAQ 12km
8 grid cells as individual points for comparison. A linear-fit model, the equivalent of a simple
9 regression, was used to compare the RBL surfaces. The y-intercept forced through the origin so
10 that the slopes of the resulting lines would be comparable. The results for ponderosa pine and
11 tulip poplar are shown in Figure 5- 6 and the summary values for all of the species are listed in
12 Table 5- 3. Plots for the remaining species are presented in Appendix 5 A. The RBL surface for
13 recent conditions was used as the baseline for comparison between rollback scenarios. This first
14 draft includes only one Os scenario, with Os levels simulating just meeting the current standard.
5-9
-------
1 The second draft will include additional scenarios with distinct secondary standards, expressed
2 using the W126, a cumulative, seasonal index.
3 Using this approach provides two advantages. First, it will in part correct for variability in
4 Os exposures in different regions. For example, one source of variability is the difference
5 between Os concentrations measured at the height of ambient monitors and those occurring at the
6 height of the actual tree canopy. In the 2007 Staff Paper (U.S. EPA, 2007a) this difference was
7 addressed by applying a 10% reduction in hourly Os values in each grid cell. That methodology
8 introduced uncertainty, but was a useful in comparing the effects of uncertainty in the 63
9 exposure values.
10 The method used to generate the exposure surface in this assessment is not readily
11 adjusted in a similar manner so the cell-by-cell comparison allows each grid cell to be compared
12 based on the proportional change between exposure scenarios. Bias in the exposure value based
13 on elevation should be similar between Os exposure scenarios, so will be factored into the
14 proportional change. The second advantage is this provides a uniform methodology to compare
15 between endpoints. In this analysis, individual tree species are used as the endpoint of the
16 analysis. The analysis presented in section 5.2.2 uses designated critical habitat and Class I areas
17 as the endpoint, and the individual case study areas analyzed in section 5.3 can each be used as a
18 distinct endpoint, but comparable analyses can be done with all 4 different endpoints. One
19 negative of this analysis is that by forcing the model through the origin, the r-squared values are
20 difficult to interpret.
5-10
-------
Ponderosa Pine
Tulip Poplar
.s
ro
m
o_
-S
ro
m
o_
0.00
0.05
0.10
0.15
0.20
0.25
0.30
PBL - Ambient Conditions
i i i i i r
0.00 0.05 0.10 0.15 0.20 0.25 0.30
PBL - Ambient Conditions
2 Figure 5- 6 Linear fit model of RBL under recent ambient O$ exposure levels (2006 -
3 2008) conditions compared to estimated values for meeting the current (8-hr)
4 standard for ponderosa pine and tulip poplar. The dashed blue line
5 represents the one-to-one line. The red line is the fitted line.
6
7 The values presented in Table 5- 3 summarize the individual species analysis. The
8 median and maximum RBL values are listed for comparison under ambient conditions. The slope
9 of linear fit model (Figure 5- 6, red lines, Table 5- 3), can be interpreted as the average
10 proportion of ambient RBL that is expected under the rollback scenario. A similar value is
11 obtained by dividing the mean RBL under the rollback scenario by the RBL value under ambient
12 conditions. Conversely, the proportion decrease could be calculated using a paired t-test and
13 dividing the estimated difference by the mean Ambient RBL. Because some of the RBL
14 distributions are not normally distributed, the linear fit model was determined to be more robust.
15 In this analysis, the ambient RBL is used as the baseline, so the proportion at ambient conditions
16 is by definition 1, and the slope for all subsequent comparisons is always the average proportion
17 of the ambient RBL. For this 1st daft REA, we evaluate only the scenario for just meeting the
18 current secondary O?, standard. Scenarios for meeting alternative O?, standards will be evaluated
19 in the second draft REA. We have put in placeholder columns in Table 5-3 for several
20 alternative standards to provide a sense of the structure of the comparisons. The EPA has not
21 determined at this point the number of alternative standards that will be evaluated in the second
22 draft REA.
5-11
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Several values in Table 5- 3 are notable. Douglas fir is a relatively non-sensitive species
at ambient levels of 63, however the proportional value is very low (0.357). Referring to Figure
5- 1, this is because this species is only sensitive at very high O3 levels. After simulating just
meeting the current secondary Os standard, there are no areas in the country where Os levels are
high enough to cause substantial RBL for this species, so the proportional change appears very
high despite a relatively low maximum RBL value when compared to other species (Table 5- 3).
However, additional reductions in Os resulting from lower levels of the standards will not result
in similarly large proportional changes for this species because they will now be in a portion of
the RBL function where this species shows very low levels of RBL, and therefore is not
responsive to 63 changes.
Sugar maple is similar, but because the maximum RBL at ambient conditions is much higher
than for Douglas fir (see Figure 5- 1), reducing Os concentrations below the "threshold", in part
controlled by the r| parameter (see Table 5-2), for Sugar maple creates a much larger
proportional difference.
Table 5- 3 Summary of Proportional Change in RBL for 11 Tree Species
Species
Red Maple (Acer rubrum)
Sugar Maple (Acer saccharum)
Red Alder (Alnus rubra)
Tulip Poplar (Liriodendron tulipifera)
Ponderosa Pine (Pinus ponderosa)
Eastern White Pine (Pinus strobus)
Virginia Pine (Pinus virginiand)
Eastern Cottonwood (Populus deltoides)
Quaking Aspen (Populus tremuloides)
Black Cherry (Prunus serotina)
Douglas Fir (Pseudotsuga menzeiesii)
Median
RBL
(Ambient)
0.009
0.000
0.005
0.045
0.038
0.034
0.008
0.564
0.039
0.225
0.000
Maximum
RBL
(Ambient)
0.039
0.206
0.118
0.291
0.294
0.226
0.018
0.999
0.377
0.547
0.001
Proportion
at Current
Standard
0.707
0.080
0.894
0.533
0.653
0.642
0.717
0.844
0.795
0.834
0.357
Proportion
at Alt A
Proportion
at Alt B
17
18
19
The results of the individual species analyses can be combined into a single plot across
s exposure scenarios (Figure 5- 7). In this analysis, all of the values under ambient conditions
5-12
-------
1 are, by definition, 1 as this is the baseline so the box for that category is a line. After simulating
2 just meeting the current secondary 63 standard, the RBL is approximately 70% of the RBL under
3 ambient conditions. Alternatively, this could be interpreted to say that RBL with O3 exposure
4 levels simulating just meeting the current secondary Os standard is 30% lower than under
5 ambient conditions. We have put in placeholders in Figure 5-7 for several alternative standards
6 to provide a sense of the structure of the comparisons. The EPA has not determined at this point
7 the number of alternative standards that will be evaluated in the second draft REA.
Relative change in Biomass Loss
op
ci
CD
O
CN1
O
q
ci
AltB
Alt A
Current
Ambient
9
10
11
12
13
Figure 5- 7 Change in RBL across exposure scenarios for 11 tree species. Biomass
loss estimates under recent ambient Os (2006 - 2008) conditions were
used as the baseline. [Alternate levels will be included in the second
draft based on simulating just attaining alternative standards]
5-13
-------
1
2 5.2.2 Relative Biomass Loss in Federally Designated Areas
3 5.2.2.1 Importance Value Scaled Analyses
4 In order to assess the risk to ecosystems in geographic areas from biomass loss as
5 opposed to the potential risk to individual tree species, it is necessary to scale the RBL to reflect
6 the abundance of each species in specific forest ecosystems. As part of the U.S. Forest Service
7 (USFS) Climate Change Atlas (http://www.fs.fed.us/nrs/atlas/littlefia/index.html) researchers at
8 the USFS Northeastern Research Station have calculated Importance Values for eastern Tree
9 species (Prasad and Iverson, 2003). Prasad and Iverson's (2003) calculation of Importance
10 Values (IV) was based equally on relative basal area and the number of stems of each tree
11 species within each FIA plot included in their analysis area with a range for each species ranging
12 from 0 to a maximum of 100. Plot level IV's were over a 20km2 scale grid for the entire study
13 area. These values were merged with the CMAQ 12 km2 grid used for the Oj, exposure and RBL
14 surfaces, with each CMAQ grid cell assigned a weighted mean IV for each species.
15 The resulting values were used in the preceding analysis (section 5.2.1) to update the
16 Little's Ranges for the eastern species. To assess biomass loss in federally designated areas, the
17 IV's were used to scale the RBL value for each tree species. The IV surface for tulip poplar is
18 shown in Figure 5- 8. Similar to the preceding analysis, the Little's Range is included for
19 reference to illustrate where the IV indicates occurrences outside of that range; however in this
20 analysis some areas within the species range are assigned an IV of 0 and are treated as areas of
21 non-occurrence. Figure 5- 8 shows an expected abundance pattern for tulip poplar, with the
22 highest abundance (as estimated by IV) near the center of its reported range, and areas near the
23 edge of its range where the species is either very low in abundance or absent all together.
5-14
-------
0.00 - 0.77
0.78-1.
1.90-3.56
3.57 - 5.77
5.78 - 8.57
8.58-12.19
12.20-18.48
18.49-41.65
2
3
4
5
6
1
Figure 5- 8 Importance Values for Tulip Poplar. (Data from U.S. Forest Service,
http://www.fs.fed.us/nrs/atlas/littlefia/index.html)
To scale RBL, the IV was divided by 100, giving a proportional value between 0 and 1 in
each grid cell and the proportional IV was multiplied by the RBL for each tree species for each
Os exposure scenario. The resulting scaled-RBL surfaces for Tulip Poplar are shown in Figure 5-
9 (Recent Conditions) and Figure 5-10 (Current Standard).
5-15
-------
IV-Scaled
Biomass Loss
(Ambient)
0.000 -
0.002 -
0.004 -
0.006 -
0.010-
0.016-
0.025 -
0.044 -
0.001
0.003
0.005
0.009
0.015
0.024
0.043
0.089
2 Figure 5- 9 Scaled Relative Biomass Loss for Tulip Poplar under recent ambient
3 exposure levels (2006 - 2008)
5 It is important to note that the scaled-RBL values highlight different areas as being the
6 highest area relative to the un-scaled RBL. In Figure 5- 3 the areas of highest RBL for tulip
7 poplar, with values above 0.25 are predominantly in the south. In Figure 5- 9 the southern areas
8 are still high, but the areas around Washington D.C and Baltimore appear much higher, as does
9 western Pennsylvania and West Virginia, relative to the un-scaled RBL values.
5-16
-------
IV-Scaled
Biomass Loss
(Current Standard)
0.000 •
0.002 •
0.004 •
0.006 •
0.010-
0.016-
0.025 •
I 0.044 •
0.001
0.003
0.005
0.009
0.015
0.024
0.043
0.089
2 Figure 5- 10 Scaled Relative Biomass Loss for Tulip Poplar after simulating just meeting
3 the current (8-hr) secondary Os standard.
4
5 To assess the overall risk to ecosystems federally designated areas, the scaled-RBL
6 values were summed across the 8 eastern species generating a summed-RBL value, with each
7 species weighted by its scaled-RBL. Figure 5-11 illustrates these values across the eastern U.S.
8 The very high values in Figure 5-11 are directly related to the presence of Eastern Cottonwood.
9 Cottonwood is a very sensitive species and in many areas where it occurs it is a dominant tree
10 species. Figure 5- 12 shows the same summed value with Eastern Cottonwood removed. The
11 highest summed-RBL value decreases from 0.854 to 0.204, demonstrating the effect of
12 cottonwood. Figure 5- 13 and Figure 5- 14 show the summed-RBL surfaces under the current
13 standard rollback scenario for all eastern species and excluding eastern cottonwood respectively.
14 There are two important things to note with respect to the IV scaled analysis. First is that
15 the TV's do not account for total cover, only the relative cover of the tree species present. This is
5-17
-------
1
2
3
4
5
6
7
most noticeable with cottonwood, which has TV's near 100 in some areas (see Appendix 5A), but
particularly in the western portions of its range, the absolute cover is probably much lower than
100%. Although this affects the direct interpretation of the values presented here, by focusing on
the proportional changes in summed-RBL between 63 exposure scenarios, the overall effect of
the variability in absolute cover values in reduced.
The second important point is that this analysis only accounts for the 8 eastern species
with C-R functions. Other species may also be sensitive to O3 exposure and it is possible that
other species that are not sensitive may be indirectly affected through changes in community
composition and competitive interactions.
10
11
12
Eastern Tree Species (Summed)
IV-Scaled
Biomass Loss
(Ambient)
0.000-
0.009-
0.022 -
0.042 -
0.072 -
0.122-
0.213-
I 0.389-
0.008
0.021
0.041
0.071
0.121
0.212
0.388
0.854
Figure 5-11 Summed Relative Biomass Loss (scaled) for 8 Eastern tree species recent
ambient O3 exposure levels (2006 - 2008)
5-18
-------
Eastern Tree Species (Summed)
1
2
IV-Scaled
Biomass Loss
(Ambient)
^H °-°00 - °-°°4
^| 0.005 - 0.009
| 0.010-0.015
| 0.016-0.025
0.026-0.039
| 0.040 - 0.060
^B 0.061 -0.094
• 0.095-0.204
Figure 5-12 Summed Relative Biomass Loss (scaled) for 7 species, excluding
eastern cottonwood, under ambient Os conditions
5-19
-------
Eastern Tree Species (Summed)
IV-Scaled
Biomass Loss
(Current Standard)
H 0.000 - 0.008
H 0.009-0.021
| 0.022-0.041
0.042-0.071
0.072-0.121
| 0.122-0.212
H 0.213-0.388
• 0.389 - 0.854
1
2
3
4
Figure 5-13 Summed Relative Biomass Loss (scaled) for 8 Eastern tree species
after simulating just meeting the current (8-hr) secondary Os
standard.
5-20
-------
Eastern Tree Species (Summed)
IV-Scaled
Biomass Loss
(Current Standard)
^B 0.000 - 0.004
^| 0.005 - 0.009
| 0.010-0.015
| 0.016-0.025
0.026-0.039
| 0.040 - 0.060
^B 0.061 -0.094
• 0.095-0.204
2 Figure 5-14 Summed Relative Biomass Loss (scaled) for 7 species, excluding eastern
3 cottonwood, after simulating just meeting the current (8-hr) secondary Os
4 standard.
5
6 5.2.3 Potential Biomass Loss in Federally Designated Areas
7 5.2.3.1 Class I Areas
8 Federally designated Class I areas were analyzed in relation to the W126 surface and the
9 scaled RBL surfaces. Figure 5-15 shows the Class I areas and W126 values. Many of the Class I
10 areas are in the western U.S., where TV's were not available to scale the RBL values. This
11 analysis uses only the Class I areas in the eastern U.S., many of which are small, and are difficult
12 to see at the scale of Figure 5- 15, or even when expanded to show only the eastern U.S. Maps of
13 each area as in Appendix 5B.
5-21
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Class I Areas
Figure 5- 15 Recent Os conditions in Class I Areas
The analyses of Class I areas were completed in the same manner as for individual
species (see Figure 5- 6), with each designated area treated as a geographic endpoint. The areas
were analyzed using the same linear model approach and the results are summarized in Table 5-
4. We have put in placeholders in Figure 5-7 for several alternative standards to provide a sense
of the structure of the comparisons. EPA has not determined at this point the number of
alternative standards that will be evaluated in the second draft REA.
Plots of the analyses are presented in Appendix 5B. Many Class I areas occur where the
ambient O3 levels are very low and simulation of just attaining the current secondary O3 standard
resulted in very little, or no change in O3 exposure in these areas so the cumulative analysis was
done twice, first with all eastern Class I areas included (Figure 5-16A) and a second analysis
excluding areas where the ambient W126 was below 10 (Figure 5-16B).
5-22
-------
1
2
3
4
5
6
Areas in Table 5- 4 with the proportion listed as NA were not included in the analysis.
These areas were excluded either due to small sample size (e.g. Rainbow Lake Wilderness), or
because the summed RBL values in all, or all but 1, grid cells were 0.
Table 5- 4 Proportion of Ambient summed-RBL in Eastern U.S. Class I areas
Class I Area
Acadia National Park
Badlands/Sage Creek Wilderness
Boundary Waters Canoe Area Wilderness
Bradwell Bay Wilderness
Breton Wilderness
Brigantine Wilderness
Caney Creek Wilderness
Cape Roman Wilderness
Chassahowitzka Wilderness
Cohutta Wilderness
Dolly Sods Wilderness
Everglades National Park
Great Gulf Wilderness
Great Smoky Mountains National Park
Hercules-Glades Wilderness
Isle Royale National Park
James River Face Wilderness
Joyce Kilmer-Slickrock Wilderness
Linville Gorge Wilderness
Lye Brook Wilderness
Mammoth Cave National Park
Mingo Wilderness
Moosehorn Wilderness
Okefenokee Wilderness
Otter Creek Wilderness
Presidential Range-Dry River Wilderness
Mean
W126
(PPM)
6.74
7.53
5.24
6.90
16.28
13.7
9.15
12.63
11.66
13.12
7.8
7.25
7.55
16.64
6.00
7.11
9.1
14.07
10.83
6.83
13.53
13.6
1.93
8.65
7.87
7.52
Number of
Grids
9
11
67
4
4
2
2
13
5
5
2
62
2
26
4
16
2
3
o
J
4
6
4
4
21
o
J
5
Proportion
of Current
Standard
0.724
NA
1.000
0.990
NA
0.386
0.995
1.000
0.803
0.716
0.996
1.000
0.892
0.445
0.966
1.00
0.992
0.496
0.910
0.889
0.981
0.845
1.000
0.993
0.946
0.914
Proportion
at Alt A
Proportion
at Alt B
5-23
-------
Class I Area
Rainbow Lake Wilderness
Saint Marks Wilderness
Seney Wilderness
Shenandoah National Park
Shining Rock Wilderness
Sipsey Wilderness
Swanquarter Wilderness
Theodore Roosevelt National Park
Upper Buffalo Wilderness
Voyageurs National Park
Wichita Mountains
Wind Cave National park
Wolf Island Wilderness
Mean
W126
(PPM)
5
8.93
7.18
10.85
12.65
14.53
14.55
6.78
7.17
5.08
9.87
10.96
8.93
Number of
Grids
1
9
4
22
4
4
4
9
3
13
6
5
o
J
Proportion
of Current
Standard
NA
0.999
0.990
0.922
0.679
0.765
0.949
1.000
0.997
1.000
NA
NA
NA
Proportion
at Alt A
Proportion
at Alt B
1
2
3
4
5
6
The combined analyses indicate that simulating just meeting the current secondary 63
standards, the proportion of ambient summed RBL is approximately 95% relative to ambient
conditions when all eastern Class I areas are included (Figure 5-16A). When only areas with
ambient Os levels above 10 ppm are included, the proportion decreases to approximately 80%
(Figure 5-16B).
5-24
-------
Relative change in Biomass Loss
Relative change in Biomass Loss
00
0
CO
CD
CN
o
o
o
AltB
I I I
Alt A Current Ambient
CO
CD
CD
CD
O
CD
AltB
I I I
Alt A Current Ambient
2
3
4
5
6
7
8
9
10
11
12
13
A. B.
Figure 5-16 Proportion of ambient scaled biomass loss in (A) all analyzed eastern
Class I Areas and (B) eastern Class I areas with average ambient Os
W126 metric exceeding 10 ppm
5.2.3.2 Critical Habitats
Federally designated critical habitat areas for endangered species were analyzed in
relation to the W126 surface and the scaled RBL surfaces. Figure 5- 17 shows the critical habitat
areas with W126 values. Like the Class I areas, many of these are in the western U.S. where TV's
were not available, so were not used in this analysis. Also like the Class I areas, many of the
critical habitat areas are difficult to see at the scale of Figure 5-17 and are included as smaller
maps in Appendix 5C.
5-25
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
Designated Critical habitat
Figure 5-17 Recent Os conditions in designated critical habitat areas.
Analyses of designated critical habitat areas were completed in the same manner as for
Class I areas, with the linear model results summarized in Table 5- 5 and the complete analyses
including figures presented in Appendix 5C. We have put in placeholder columns in Table 5-5
for several alternative standards to provide a sense of the structure of the comparisons. EPA has
not determined at this point the number of alternative standards that will be evaluated in the
second draft REA.
Areas in Table 5- 5 with the proportion listed as NA were not included in the analysis.
These areas were excluded either due to small sample size (e.g. San Marcos gambusia), or
because the summed RBL values in all, or all but 1, grid cells were 0 (e.g. Cape Sable seaside
sparrow).
5-26
-------
1 This analysis is not intended to indicate risk to the specific endangered species within the
2 designated area; rather the intent is to use the designated critical habitat to define an endpoint for
3 evaluating risk to locations that might be more sensitive to adverse effects from O3 exposure. For
4 example, analysis of the critical habitat area for Gulf sturgeon is focused on the terrestrial
5 ecosystems within the designated habitat area, not on the aquatic system, or the Gulf sturgeon.
6 The implication in the aquatic and marine areas in particular is that effects on neighboring
7 terrestrial ecosystems will affect the aquatic or marine system, but quantifying that linkage is not
8 possible at this time.
9 Table 5- 5 Proportion of ambient summed-RBL in Eastern U.S. Critical Habitat Areas
Designated Critical Habitat Area
Gulf sturgeon
Appalachian elktoe
Reticulated flatwoods salamander
Frosted flatwoods salamander
Cape Sable seaside sparrow
Braun's rock-cress
Helotes mold beetle
Houston toad
Gray wolf
Piping plover
Salt Creek tiger beetle
Robber Baron Cave meshweaver
Madia's Cave meshweaver
Braken Bat Cave meshweaver
Virginia big-eared bat
American crocodile
Haha
Fountain darter
Niangua darter
San Marcos salamander
San Marcos gambusia
Whooping crane
Mississippi sandhill crane
Mean
W126
(PPM)
14.69
11.70
9.16
9.16
7.06
15.77
11.63
7.89
5.08
8.51
2.93
7.30
11.27
10.40
6.63
6.65
5.47
7.90
7.88
7.90
7.90
7.77
12.28
Number of
Grids
116
22
35
35
21
3
6
7
283
472
4
4
12
31
7
53
17
5
17
4
1
43
6
Proportion
at Current
Standard
0.695
0.685
0.983
0.983
NA
NA
NA
NA
1.000
1.000
NA
NA
NA
NA
0.970
NA
NA
NA
0.910
NA
NA
NA
0.759
Proportion
at Alt A
Proportion
at Alt B
5-27
-------
Designated Critical Habitat Area
Johnson's seagrass
Comal Springs riffle beetle
Mountain golden heather
Zapata bladderpod
Canada lynx
Waccamaw silverside
Spruce-fir moss spider
Government Canyon Bat Cave spider
Concho water snake
Arkansas River shiner
Cape Fear shiner
Topeka shiner
Rice rat
Amber darter
Conasauga logperch
Leopard darter
Choctawhatchee beach mouse
Alabama beach mouse
St. Andrew beach mouse
Perdido Key beach mouse
Everglade snail kite
Atlantic salmon
Mine's emerald dragonfly
Peck's cave amphipod
Comal Springs dryopid beetle
Cokendolpher Cave harvestman
West Indian manatee
Louisiana black bear
Texas wild-rice
Rhadine exilis (No common name)
Rhadine infernalis (No common name)
Mean
W126
(PPM)
9.27
7.98
11.05
4.08
4.46
8.30
14.78
10.40
5.47
12.40
14.00
5.57
5.82
18.84
19.00
7.04
12.42
17.87
10.70
18.40
9.90
3.17
9.50
8.43
8.30
7.30
9.45
9.95
7.90
11.35
11
Number of
Grids
13
5
2
4
523
2
6
31
17
78
7
235
6
7
4
21
5
o
J
11
2
58
312
30
o
J
o
J
4
211
90
6
22
30
Proportion
at Current
Standard
NA
NA
0.892
NA
1.000
0.926
0.906
NA
NA
NA
0.667
0.988
NA
0.708
0.686
0.961
NA
0.798
0.889
0.730
1.000
0.925
0.669
NA
NA
NA
0.991
0.771
NA
NA
NA
Proportion
at Alt A
Proportion
at Alt B
5-28
-------
1
2
3
4
5
The cumulative analyses indicate that across all eastern critical habitat areas, the
proportion of the ambient summed-RBL was between 90% and 95% under the current standard
rollback scenario (Figure 5-18A). When areas with ambient O?, levels below 10 ppm are
excluded, the proportion decreases to approximately 75% (Figure 5-18B).
Relative change in Biomass Loss
00
o
CD
ID
CN
O
Q
O
AltB
I I I
Alt A Current Ambient
Relative change in Biomass Loss
00
o
CD
O
CN
o
p
o
AltB
I I I
Alt A Current Ambient
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
A. B.
Figure 5-18 Proportion of ambient scaled-biomass loss in (A) all analyzed eastern
critical habitat areas and (B) eastern critical habitat areas with
average ambient Os W126 metric exceeding 10 ppm
5.2.4 National Park Case Study Areas
The National Parks provide excellent case study areas for more refined analyses of Os
exposure risks. The National Park Service (NFS) conducts ongoing Os monitoring in many
parks, and these monitors were used when possible in the creation of the Os exposure surfaces
for the parks. In addition, recreational use data are available for the parks for analyses of
recreational value presented in Chapter 6. Three parks were chosen as case study areas: Great
Smoky Mountains National Park (GSMNP), Rocky Mountain National Park (RMNP), and
Sequoia/Kings Canyon National Park (SKCNP).
Vegetation mapping has been completed in all three parks by the NFS in conjunction
with the United States Geological Survey (USGS). These maps were used to estimate the percent
5-29
-------
1 cover of the tree species included in the risk assessment. These values were then used in a similar
2 way to the TV's in the preceding section, but on a much finer scale. The vegetation maps for the
3 parks are available through the USGS Vegetation Characterization Program
4 (http://biology.usgs.gov/npsveg/apps/). The vegetation map for GSMNP was completed in 2004
5 (Madden et al. 2004).
6 The National Vegetation Community codes assigned to each vegetation community were
7 used to obtain cover estimate data through plots stored in VegBank
8 (http://vegbank.org/vegbank/index). Whenever possible, only plots from within the park were
9 used. In some cases, no plots were available from within the park and in those cases plots from
10 the same vegetation community in nearby areas were used. In a few cases there were no plots
11 available, and those communities were excluded.
12 The W126 surface for each park was intersected with the vegetation polygons and the
13 RBL values for the tree species present were scaled using the percent cover of each tree species
14 the same as in the preceding section when IV was used. These values were then summed within
15 each polygon in the GIS shapefile to create a detailed surface for each park. To assess the
16 proportional change in scaled RBL in each park a linear model was used as in the preceding
17 sections. In this analysis each polygon was treated as an individual point as opposed to CMAQ
18 grid cells as in the preceding analyses.
19
20 [GSMNP is the only park completed at present, the linear model results will be combined into a
21 combined analysis when more parks are included]
22
23 5.2.4.1 Great Smoky Mountain National Park
24 Recent (2006 - 2008) ambient O3 levels (3-month 12-hr W126) in GSMNP range from
25 9.3 PPM along the southeastern boundary to 23.3 PPM along the northwestern boundary (Figure
26 5- 19). After simulating just attaining the current secondary Os standard, (Figure 5- 20) the
27 W126 values decrease to 7.7 PPM to 13 PPM.
28
5-30
-------
1
2
3
Great Smoky Mountains National Park
W126
I 9.3
9.4-10.0
10.1 -15.0
15.1 -20.0
20.1 -25.0
25.1 -30.0
30.1 -35.0
35.1 -65.0
Figure 5- 19 Recent (2006 - 2008,12-hr 3-month W126) O3 Exposure in GSMNP
5-31
-------
1
2
3
4
5
6
7
8
9
10
Great Smoky Mountains National Park
W126
(Current Standard)
| | 7.8-10.0
Q • 10.1 - 15.0
| | 15.1 -20.0
| | 20.1 - 25.0
Q | 25.1 - 30.0
f^^| 30.1 - 35.0
• 35.1 - 65.0
Figure 5- 20 Os Exposure in GSMNP after simulating just meeting the current (8-
hr) secondary standard.
The vegetation map for GSMNP included 34 vegetation communities. Six of the eastern
tree species occurred within the park. The resulting scaled RBL values for the ambient and
current standard surfaces are shown in Figure 5-21 and Figure 5- 22. The linear model results
for GSMNP indicate a proportionally large decrease (slope = 0.493) in summed-RBL when
comparing the current standard to ambient conditions (Figure 5- 23).
5-32
-------
1
2
3
4
5
Great Smoky Mountains National Park
Potential Biomass Loss
(Scaled)
0.006
0.012
0.019
0.028
0.044
0.075
0.115-
0.011
0.018
0.027
0.043
0.074
0.114
0.165
Figure 5- 21 Summed-RBL in GSMNP, scaled using percent cover of species, under
recent Os conditions. White areas within the park represent areas where no
data were available or were developed, with minimal vegetation.
5-33
-------
1
2
3
4
Great Smoky Mountains National Park
Potential Biomass Loss
(Scaled)
0.000 •
0.006 •
0.012-
0.028 •
0.044 •
0.075 •
0.080 •
0.115-
0.005
0.011
0.027
0.043
0.074
0.079
0.114
0.165
Figure 5- 22 Summed-RBL in GSMNP, scaled using percent cover of species after
simulating just meeting the current secondary Os standard.
5-34
-------
GSMNP
1
2
4
5
6
7
8
9
10
11
12
o
(N -
O
CD O
T3
CO
c °
I o
O
DO 10
o: p _
o
O
O -
O
0.00 0.05 0.10
RBL - Ambient Conditions
i r
0.15 0.20
Figure 5- 23 Linear Fit Model comparing RBL under ambient conditions and a
scenario just meeting the current standard.
5.2.4.2 Rocky Mountain National Park
[To be added in the second draft]
5.2.4.3 Sequoia/Kings National Park
[To be added in the second draft]
5.2.4.4 National Park Case Study Area Summary
Table 5- 6 Proportion of summed-RBL in National Park Case Study Areas
Designated Critical Habitat Area
Great Smoky Mountains National Park
Rocky Mountain National Park
Sequoia/Kings National Park
Mean W126
(PPM)
16.45
Max W126
(PPM)
23.30
Proportion
at Current
Standard
0.493
Proportion
at Alt A
Proportion
at Alt B
13
5-35
-------
1
2
3 [This will include a summary of the linear model results, presented in similar to the
4 boxplots presented in preceding sections]
5 5.3 VISIBLE FOLIAR INJURY
6 Visible foliar injury resulting from exposure to Os has been well characterized and
7 documented over several decades on many tree, shrub, herbaceous, and crop species (U.S. EPA,
8 2012a, 2006, 1996, 1984, 1978). Visible foliar injury symptoms are considered diagnostic as
9 they have been verified experimentally in exposure-response studies, using exposure
10 methodologies such as CSTRs, OTCs, and free-air fumigation (see Section 9.2 of the ISA for
11 more detail on exposure methodologies). Although the majority of Os-induced visible foliar
12 injury occurrence has been observed on seedlings and small plants, many studies have reported
13 visible injury of mature coniferous trees, primarily in the western U.S. (Arbaugh et al., 1998) and
14 to mature deciduous trees in eastern North America (Schaub et al., 2005; Vollenweider et al.,
15 2003; Chappelka et al., 1999a; Chappelka et al., 1999b; Somers et al., 1998; Hildebrand et al.,
16 1996).
17 Although visible injury is a valuable indicator of the presence of phytotoxic
18 concentrations of 63 in ambient air, it is not always a reliable indicator of other negative effects
19 on vegetation. The significance of O3 injury at the leaf and whole plant levels depends on how
20 much of the total leaf area of the plant has been affected, as well as the plant's age, size,
21 developmental stage, and degree of functional redundancy among the existing leaf area. Previous
22 Os AQCDs have noted the difficulty in relating visible foliar injury symptoms to other vegetation
23 effects such as individual plant growth, stand growth, or ecosystem characteristics (U.S. EPA,
24 2012a, 2006, 1996). As a result, it is not presently possible to determine, with consistency across
25 species and environments, what degree of injury at the leaf level has significance to the vigor of
26 the whole plant. However, in some cases, visible foliar symptoms have been correlated with
27 decreased vegetative growth (Somers et al., 1998; Karnosky et al., 1996; Peterson et al., 1987;
28 Benoit et al., 1982) and with impaired reproductive function (Chappelka, 2002; Black et al.,
29 2000). Conversely, the lack of visible injury does not always indicate a lack of phytotoxic
30 concentrations of Os or a lack of non-visible Os effects (Gregg et al., 2006).
31
5-36
-------
1 5.3.1 National-Scale Analysis of Foliar Injury
2 5.3.1.1 National Summed Importance Values
3 The NFS has published a list of known and suspected Os sensitive species (NFS, 2003),
4 which was updated in 2006 (NFS, 2006). This list of species was used together with the TV's
5 from the USFS (Prasad and Iverson, 2003). A map of the eastern U.S. was generated showing the
6 summed TV's of species sensitive to foliar injury from Os (Figure 5- 24). This essentially shows
7 the abundance of trees likely to be impacted by elevated 63 levels.
9 [Analysis is not complete, waiting on data from John Coulston with the USFS to complete this
10 analysis]
11
12
13
14
Summed Importance Values of Sensitive Tree Species
Figure 5- 24 Summed Importance Values for Sensitive Species in the Eastern U.S.
5.3.1.2 Forest Health Monitoring Network
5-37
-------
1 5.3.2 Updated Assessment of Risk of Visible Foliar Injury in National Parks
2 A study by Kohut (2007) assessed the risk of O3-induced visible foliar injury on O3
3 bioindicators (i.e., Os-sensitive vegetation (NFS, 2006)) in 244 national parks as part of the NFS'
4 Vital Signs program. Kohut (2007) estimated Os exposure using hourly Os monitoring data
5 conducted at 35 parks from 1995 to 1999 and estimated Os exposure at 209 additional parks
6 using kriging, a spatial interpolation technique. Kohut (2007) qualitatively assessed risk based on
7 evaluation of three criteria: the frequency of exceedance of foliar injury thresholds1 using several
8 63 exposure metrics, the extent that low soil moisture constrains 63 uptake during periods of high
9 exposure, and the presence of 63 sensitive species within each park. Kohut (2007) concluded
10 that the risk of visible foliar injury was high in 65 parks (27%), moderate in 46 parks (19%), and
11 low in 131 parks (54%). We have updated this assessment using more recent Os exposure and
12 soil moisture data for a subset of parks with Os monitors.
13 5.3.2.1 Foliar Injury Risk Methods
14 We applied the approach used in Kohut (2007) using more recent O3 monitoring and soil
15 moisture data from 2006 to 2010. For this 1st draft REA, because we did not replicate the spatial
16 interpolation of monitor data in Kohut (2007) due to uncertainties introduced using this
17 technique, we conducted this updated risk assessment only in parks with Oj monitor data.2 As
18 noted by Kohut (2007), monitoring provides the most accurate assessment of Os exposure, but it
19 may not reflect differences in exposure throughout the park.
20 Oi Exposure: We used more recent monitoring data from 2006 through 2010 and the
21 same metrics in this analysis (i.e., SUM06 (3-month), W126 (12-hr, 7-month), N100 (7-month))
22 as Kohut (2007). In addition, we added W126 (12-hr) and N100 metrics calculated over 3
23 months to be consistent with other analyses in this REA and to determine how sensitive the risk
24 ratings were to the different W126 metrics. Each of these metrics are described in more detail in
25 Section 4.3.1. These data reflected 59 Os monitors located within park boundaries covering 43
1 Kohut (2007) uses the term "foliar injury thresholds". It is unclear whether these are true biological thresholds
below which no vegetation effects occur or whether these are simply concentration benchmarks. We use the term
"thresholds" to be consistent with the terminology in Kohut (2007).
2 For the 2nd draft REA, we anticipate expanding this updated assessment to include additional parks. One method
would assign an ozone monitor if it fell within a certain distance of a park's boundaries (e.g., 10km, 50,km, etc).
A second option would use the ozone surfaces for 2006, 2007, and 2008 described in Chapter 4. While either
method would provide ozone exposure data at parks that has additional uncertainty relative to the data at parks
with ozone monitors within their boundaries, neither would add as much uncertainty as the kriging interpolation
of monitor data.
5-38
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
separate parks, which is more than the 35 parks with 63 monitors in Kohut (2007). If a park
contained more than one 63 monitor, we used the highest monitor in the park as an indication of
the potential risk. For two parks, Badlands National Park and Glacier National Park, we used
data from an additional park monitor to fill in missing data years at the highest monitor.
Based on the foliar injury thresholds for Os exposure used by Kohut (2007), we assigned
exposure risk ratings associated with Os exposure alone to each park with an monitor. Consistent
with Kohut (2007), Os exposure must meet the criteria for both the W126 index as well as the
N100 metric in order to receive a higher risk rating. We provide the specific criteria applied in
this updated risk assessment, which are derived from Table 5-7 in Kohut (2007). Overall,
considerably more parks exceed the W126 criteria alone than in conjunction with the N100
criteria. Specifically, 35 of 37 parks exceed 5.9 ppm-hrs using the 7-month W126 metric for at
least 3 years, whereas only 5 parks exceed 6 hours using the 7-month N100 metric in any year.3
Only 3 parks exceeded 8 ppm-hrs using the SUM06 metric in any year, which corresponds to
Kohut's lowest injury threshold for natural ecosystems.
Table 5- 7
Risk Criteria for
Moisture.
Exposure Metrics, Sensitive Vegetation, and Soil
Risk Criterion and Metric
O3 Exposure
Sensitive
Vegetation
Soil Moisture
SUM06
W126/N100
(3 -mo nth)
W126/N100
(7-month)
Indicator species
Palmer Z
Higher Risk
Exceeds 8 ppm-hrs
Exceeds 4.1 ppm-hrs
AND Exceeds 6 hrs over 100 ppm
Exceeds 5.9 ppm-hrs
AND Exceeds 6 hrs over 100 ppm
Present
No relation
Lower Risk
Less than 8 ppm-hrs
Less than 4.1 ppm-hrs AND Less
than 6
Less than 5.9 ppm-hrs AND Less
than 6
Not present
Inverse
(not used to lower risk rating)
18
19
20
21
The primary difference between a high risk rating and a moderate risk rating is the
number of years that exceed the Os exposure metrics. If a park exceeded the risk criteria for 1 or
2 years, we assigned a risk rating of moderate. If a park exceeded the risk criteria for at least 3
3 In order to assess risk using the 3-month W126 metric, we calculated an adjustment to the foliar injury threshold
for highly sensitive species. Based on a regression analysis described in Appendix 5D, we determined that a foliar
injury threshold of 5.9 ppm-hrs for a 7-month W126 metric is approximately equivalent to a foliar injury threshold
at 4.1 ppm-hrs for a 3-month W126 metric.
5-39
-------
1 years, we assigned a risk rating of high. If a park did not exceed the risk criteria in any year, we
2 assigned a risk rating of low.
3 Soil Moisture: To evaluate soil moisture, we followed Kohut's approach by using Palmer
4 Z data for 2006 to 2010 (NCDC, 2012b). The Palmer Z Index represents the difference between
5 monthly soil moisture and long-term average soil moisture (Palmer, 1965). These data typically
6 range from -4 to +4, with positive values representing more wetness than normal and negative
7 values representing more dryness than normal. Values between -0.9 and +0.9 could be
8 interpreted as normal soil moisture, whereas values beyond the range from -3 to +3 could be
9 interpreted as extremely unusually soil moisture (either extreme drought or extreme wetness). As
10 described in the ISA (U.S. EPA, 2012a), plants generally uptake less 63 when soil moisture is
11 reduced, thus the risk of foliar injury is generally lower during periods of drought.
12 The soil moisture index is calculated for each of the 344 climate regions within the
13 continental U.S. defined by the National Climatic Data Center (NCDC) (NOAA, 2012a). We
14 assigned each monitored park to the climate region in which the park was located. For the
15 monitored parks that were located in more than one NCDC region, we selected the region
16 corresponding to the monitor location. We decided not to average the Palmer Z values across
17 regions because the NCDC regions are much larger geographic areas (e.g., sometimes hundreds
18 of miles in diameter) than the parks themselves. Because we did not have soil moisture data
19 outside of the continental U.S., we did not evaluate parks in Alaska, Hawaii, Puerto Rico, or
20 Guam. In addition, due to the size of these regions, soil moisture will vary within each region
21 and potentially even within a park. For example, some species along riverbanks may still
22 experience sufficient soil moisture during periods of drought to exhibit foliar injury. For this
23 reason, we provide the soil moisture data and assess the relationship with Os exposure, but we
24 have not lowered any risk ratings in the updated assessment for insufficient soil moisture. We
25 identify the regions in Figure 5- 25, and we provide the Palmer Z data for each park in Appendix
26 5D.
5-40
-------
2 Figure 5- 25 344 climate regions with Palmer Z soil moisture data (source: NCDC,
3 2012a).
4
5 Because monthly estimates of soil moisture are highly variable over time, we focused on
6 the monthly values from May to October for each year in order to be consistent with the potential
7 time period of the W126 calculation. Evaluating soil moisture is more subjective than for Os
8 exposure because Kohut (2007) did not outline specific numerical criteria for this determination.
9 We compared the soil moisture during the years of highest 63 exposure and during the years of
10 lowest exposure to determine whether there was a consistent trend. Based on our review of the
11 soil moisture data in the updated assessment, several parks showed a potentially inverse
12 relationship between high O3 exposure years and soil moisture.
13 Sensitive Vegetation Species: Consistent with Kohut (2007), we identified the parks
14 containing 63 sensitive vegetation species (NFS, (2003, 2006). Based on the NFS list, all of the
15 parks in this updated assessment contain at least one sensitive species.
16 GIS Analysis: Using GIS (ESRI® ArcMAP™ 9.3), we spatially overlaid the O3 exposure
17 monitor data, NFS boundaries (USGS, 2003), and soil moisture Palmer Z data to link these data
18 to each park. In total, 43 parks had 63 monitoring data, including 9 parks that contained more
5-41
-------
1
2
3
4
5
6
than one Os monitor. We excluded 5 parks with fewer than 3 years of monitoring data4 and one
park (i.e., Denali NP in Alaska) with an absence of soil moisture data. After these exclusions, 37
parks were included in this updated risk assessment, which are identified in Figure 5- 26. All of
the monitored parks excluded from this updated assessment received risk ratings of "low" in
Kohut (2007), except for City of Rocks, National Reservation, which had a risk rating of
"moderate".
7
8
9
10
11
12
13
14
15
Mount Ramief Wii
Figure 5- 26 37 National Parks with Os monitors included in the updated risk assessment.
5.3.2.1 Foliar Injury Risk Results and Discussion
As explained in Kohut (2007), determining the overall risk level is not quantitative, but
instead depends on a subjective evaluation of how much and how often Os exposure metrics
exceeded certain criteria, the soil moisture conditions during high exposure periods, and the
presence of sensitive vegetation species. Similar to Kohut's subjective evaluation, we also
categorized each park as at high, moderate, or low risk for foliar injury based on these criteria.
4 These 5 excluded parks for less than 3 years of ozone monitoring data are Agate Fossil Beds National Monument,
City of Rocks National Reservation, Olympic National Park, Padre Island National Seashore, and Scotts Bluff
National Monument.
5-42
-------
1 For the 37 parks assessed in the updated risk assessment, we found generally similar risk
2 levels as Kohut (2007). Based on his analysis of all 244 parks, Kohut (2007) found that the risk
3 of foliar injury was high in 65 parks (27%), moderate in 46 parks (19%), and low in 131 parks
4 (54%). Limiting the assessment to the same 37 parks in the updated risk assessment, Kohut
5 found the risk of foliar injury was high in 10 parks (27%), moderate in 4 parks (11%), and low in
6 23 parks (62%). The updated risk assessment of 37 parks found the risk of foliar injury was high
7 in 2 parks (5%), moderate in 4 parks (11%), and low in 31 parks (84%). We provide the risk
8 results for each park included in the assessment in Table 5-8, and we provide all of the 63 and
9 soil moisture data in Appendix 5D.
10 Based on our updated assessment, most parks (70%) received the same risk rating as
11 Kohut (2007), while 30% received lower risk ratings. The decrease in risk rating corresponds to
12 lower Os concentrations in more recent years, particularly for the N100 metric. In general, results
13 were insensitive to whether we used the 3-month or 7-month W126 metric. Only 1 park, Acadia
14 National Park, would have a different risk rating if we used the 3-month W126 metric rather than
15 the 7-month W126 metric.
16 In the original assessment, Kohut (2007) provided an appendix explaining the risk
17 analysis for Cape Cod National Seashore. Based on 63 exposure ranged from 17 to 25 ppm-hrs
18 using the SUM06 metric, 33.6 to 40.4 ppm-hrs using the 7-month W126 metric, and 6 to 52
19 using the 7-month N100 metric, Kohut concluded that the risk level is high because these
20 exposure levels are significantly greater than the injury thresholds using all metrics. In the
21 updated assessment, we assigned a risk level of moderate to Cape Cod National Seashore based
22 on Os exposure that ranged from <1 to 3 ppm-hrs using the SUM06 metric, 14.5 to 33.1 ppm-hrs
23 using the 7-month W126 metric, and 0 to 11 using the 7-month N100 metric because exposures
24 exceed the injury thresholds using both criteria for the W126 index (W126 and N100) in only
25 one year.
26 As another example, we assigned a risk level of low to the Great Smoky Mountains
27 National Park because 63 exposure levels exceeded the W126 injury thresholds (7-month and 3-
28 month) but not the N100 thresholds. When assessing the 3 other Oj monitors in the park, only 2
29 monitors exceeded 100 ppm using the 7-month N100 metric once apiece between 2006 and
30 2010. This is a substantial decline from the 1995 to 1999 Os data, which showed up to 107 hours
31 above 100 ppm in a single year at the highest monitor (NFS, 2004). While Os levels are still
5-43
-------
1 consistently high enough to elevate the W126 levels in the more recent monitoring data, there are
2 many fewer hours where 63 concentrations spike above 100 ppm. In addition, there appeared to
3 be a slight inverse relationship between O3 exposure and soil moisture in the Great Smoky
4 Mountains National Park using more recent soil moisture data.
5-44
-------
1 Table 5- 8 Levels of Risk of Foliar Injury in 37 Parks with an Os Monitor.
Park Name
Acadia National Park
Badlands National Park
Big Bend National Park
Blue Ridge Parkway
Canyonlands National Park
Cape Cod National Seashore
Carlsbad Caverns National Park
Colorado National Monument
Congaree Swamp National
Monument
Cowpens National Battlefield
Craters of the Moon National
Historic Park
Cumberland Gap National Historic
Park
Death Valley National Park
Devils Tower National Monument
Dinosaur National Monument
Glacier National Park
Great Basin National Park
Great Smoky Mountains National
Park
Grand Canyon National Park
Indiana Dunes National Landmark
Joshua Tree National Park
Lassen Volcanic National Park
Park Monitor
State
ME
SD
TX
NC
UT
MA
NM
CO
sc
sc
ID
KY
CA
WY
CO
MT
NV
NC
AZ
IN
CA
CA
Kohut (2007) Risk
Level
Moderate
Low
Low
Low
Low
High
Low
Low
Low
High
Low
High
Low
Low
Low
Low
Low
High
Low
High
High
Low
Updated Risk
Level
Moderate
Low
Low
Low
Low
Moderate
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
High
Low
Change
No
change
No
change
No
change
No
change
No
change
Decrease
No
change
No
change
No
change
Decrease
No
change
Decrease
No
change
No
change
No
change
No
change
No
change
Decrease
No
change
Decrease
No
change
No
change
5-45
-------
Park Name
Mesa Verde National Park
Mojave National Preserve
Mount Rainier National Park
Petrified Forest National Park
Pinnacles National Monument
Saguaro National Park
Saratoga National Historic Park
Sequoia & Kings Canyon National
Park
Shenandoah National Park
Theodore Roosevelt National Park
Tonto National Monument
Voyageurs National Park
Wind Cave National Park
Yellowstone National Park
Yosemite National Park
Park Monitor
State
CO
CA
WA
AZ
CA
AZ
NY
CA
VA
ND
AZ
MN
SD
WY
CA
Kohut (2007) Risk
Level
Low
High
Low
Moderate
High
Low
Low
High
Moderate
Low
Moderate
Low
Low
Low
High
Updated Risk
Level
Low
Moderate
Low
Low
Low
Low
Low
High
Low
Low
Low
Low
Low
Low
Moderate
Change
No
change
Decrease
No
change
Decrease
Decrease
No
change
No
change
No
change
Decrease
No
change
Decrease
No
change
No
change
No
change
Decrease
1
2
3
4
5
6
7
8
9
10
11
12
5.3.3 National Park Case Study Areas
For the National Park case study areas, staff used the 63 sensitive species list from the
preceding section and cover data from VegBank plots (see section 5.3). The resulting maps give
cover estimates for sensitive 63 sensitive species at the finer scale of the NFS vegetation map
(Figure 5- 27). It is important to note that the cover estimates are separated into vegetation strata
(herb, shrub, tree). In the preceding analyses we only used tree species, so the cover never
exceeded 100%. For this analysis we did not distinguish between strata, so the cover metric can
exceed 100. [This analysis will be completed in the 2nd draft with the addition of the 2 additional
NFS case study areas]
5.3.3.1 Great Smoky Mountain National Park
5-46
-------
Great Smoky Mountains National Park
sSti&r^
^ V '
ii«f~, - • f
v-li^^S?!^^ /'
;.l^:v >.^v^.>tr.?-v"-- •
^ll^^ rl^A,,^'
Sensitive Species Cover
Index
^B °-°° - 3-5°
^B 3-51 -9-88
9.89 - 20.85
20.86 - 39.01
39.02-52.17
^H 52.18-60.92
^B 30.93 - 77.88
I 77.89-160.08
4
5
6
Figure 5- 27 Cover Index of Sensitive Species in GSMNP
5.3.3.2 Rocky Mountain National Park
[To be added in the second draft]
5.3.3.3 Sequoia/Kings National Park
[To be added in the second draft]
1 5.4 DISCUSSION
9
10
11
12
13
14
15
For individual tree species the RBL was, on average, 30% less under the current standard
scenario. In Class I areas with higher 63 exposure this reduction was approximately 20%
and in Critical Habitat areas it was 30%.
Individual tree species show different patterns of change with respect to changes in O3.
Douglas fir has a very large proportional change when O3 is meeting the current
standard, however further reductions in O3 will likely have very little effect on that
species. Sugar maple also had a large proportional change when meting the current
5-47
-------
1 standard. Further reductions in O3 will have some effect to a point beyond which we
2 expect very little change. Other species are expected to exhibit continued gradual change
3 in RBL relative to ambient as O3 levels are reduced.
4 • Many Class I and Critical Habitat areas occur in areas of low ambient O3 and these areas
5 generally show very little change in summed RBL relative to ambient. In areas with
6 higher ambient O3 levels, the proportion of ambient summed RBL decreases by as much
7 as 20%.
8 • Within the GSMNP this value was higher, around 45%, but this analysis needs to be
9 expanded with additional parks.
10 • There are significant areas with high abundance of Os sensitive tree species. Not all of
11 these areas co-occur with areas of high 63. This is an analysis that is not complete.
12 • There are areas within GSMNP where the sensitive species cover is very high. The
13 relationship of these to areas of recreational use is presented in Chapter 6.
14 • Overall, these analyses indicate that decreasing 63 from ambient conditions to a rollback
15 scenario just meeting the Current Standard had a significant impact, but additional
16 rollback scenarios are needed to fully interpret this observation.
5-48
-------
i 6 OZONE RISK TO ECOSYSTEM SERVICES
2 6.1 INTRODUCTION
3 EPA has begun using an ecosystem services framework to help define how the damage to
4 ecosystems informs determinations of the adversity to public welfare associated with changes in
5 ecosystem functions.
6 The following sections address the risks to ecosystem services resulting from O?,
1 exposure. While most of the impacts of 63 on these services cannot be specifically quantified, it
8 is important to provide an understanding of the magnitude and significance of the services that
9 may be negatively impacted by 63 exposures. For many services, we can estimate the current
10 total magnitude and, for some, the current value of the services in question. The estimates of
11 current service provision will have embedded within them the loss of services occurring due to
12 historical and present Os exposure, and provide context for the importance of any potential
13 impacts of Os on those services. In addition, in some cases we can provide information on
14 locations where high O?, exposures occur in conjunction with significant ecosystem service
15 impairment.
16 6.2 NATIONAL SCALE ECOSYSTEM SERVICES ASSESSMENT
17 The national scale assessment will address 63 impacts on ecosystem services following
18 the framework of the Millennium Ecosystem Assessment (MEA, 2009). Following that
19 framework the subsequent sections are divided into supporting, regulating, provisioning, and
20 cultural services.
21 Two major effects of Os exposure on ecosystems considered in this assessment are biomass loss
22 (or decrease in growth rate) and visible foliar injury. Each of these ecological effects can have
23 negative effects on vegetation related to ecosystem services. To illustrate
6-1
-------
1 Table 6- 1 lists the trees identified as sensitive to 63 in studies cited in the ISA and their
2 uses.
3
6-2
-------
1 Table 6-1 O3 Sensitive Trees and Their Uses
Tree Species
O3 Effect
Uses
Black Cherry
Prunus serotina
Biomass loss,
Visible foliar injury
Cabinets, furniture, paneling, veneers,
crafts, toys
Cough remedy, tonic , sedative
Flavor for rum and brandy
Wine making and jellies
Food for song birds, game birds, and
mammals
Douglas Fir
Pseudotsuga menziesii
Biomass loss
Commercial timber
Medicinal uses, spiritual and cultural uses
for several Native American tribes
Spotted owl habitat
Food for mammals including antelope and
mountain sheep
Eastern Cottonwood
Populus deltoides
Biomass loss
Containers, pulp, and plywood
Erosion control and windbreaks
Quick shade for recreation areas
Beaver dams and food
Eastern White Pine
Pinus strobus
Biomass loss
Commercial timber, furniture,
woodworking, and Christmas trees
Medicinal uses as expectorant and
antiseptic
Food for song birds and mammals
Used to stabilize strip mine soils
Hemlock
Tsuga canadensis
Biomass loss
Commercial logging for pulp
Habitat for deer, ruffled grouse, and
turkeys
Important ornamental species
Hickory
Biomass loss
Used in furniture and cabinets, fuelwood
and charcoal
Edible nuts
Food for ducks, quail, wild turkeys and
many mammals
Ponderosa Pine
Pinus ponderosa
Biomass loss,
Visible foliar injury
Lumber for cabinets and construction
Ornamental and erosion control use
Recreation areas
Food for many bird species including the
6-3
-------
Tree Species
O3 Effect
Uses
red-winged blackbird, chickadee, finches,
and nuthatches
Quaking Aspen
Populus tremuloides
Biomass loss,
Visible foliar injury
Commercial logging for pulp, flake-board,
pallets, boxes, and plywood
Products including matchsticks, tongue
depressors, and ice cream sticks
Valued for its white bark and brilliant fall
color
Important as a fire break
Habitat for variety of wildlife
Traditional native American use as a food
source
Red Alder
Alnus rubra
Biomass loss,
Visible foliar injury
Commercial use in products such as
furniture, cabinets, and millwork
Preferred for smoked salmon
Dyes for baskets, hides, moccasins
Medicinal use for rheumatic pain,
diarrhea, stomach cramps - the bark
contains salicin, a chemical similar to
aspirin
Roots used for baskets
Food for mammals and birds - dam and
lodge construction for beavers
Conservation and erosion control
Red Maple
Acer rubrum
Biomass loss
Revegetation and landscaping esp.
riparian buffer
Red Oak
Quercus rubrum
Biomass loss
Important for hardwood lumber for
furniture, flooring, cabinets
Food, cover, and nesting sites for birds
and mammals
Bark used by Native Americans for
medicine for heart problems, bronchial
infections or as an astringent, disinfectant,
and cleanser
Short Leaf Pine
Pinus echinata
Biomass loss
Second only to loblolly pine in standing
timber volume.
Used for lumber, plywood, pulpwood,
boxes, crates, and ornamental vegetation
6-4
-------
Tree Species
O3 Effect
Uses
Habitat and food for bobwhite quail,
mourning dove, other song birds and
mammals
Older trees with red heart rot provide red-
cockaded woodpecker cavity trees
Sugar Maple
Acer saccharum
Biomass loss
Commercial syrup production
Native Americans used sap as a candy,
beverage - fresh or fermented into beer,
soured into vinegar and used to cook meat
Valued for its fall foliage and as an
ornamental
Commercial logging for furniture,
flooring, paneling, and veneer
Woodenware, musical instruments
Food and habitat for many birds and
mammals
Virginia Pine
Pinus virginiana
Biomass loss,
Visible foliar injury
Pulpwood, strip mine spoil banks and
severely eroded soils
Nesting for woodpeckers, food for
songbirds and small mammals
Yellow (Tulip) Poplar
Liriodendron tulipifera
Biomass loss,
Visible foliar injury
Furniture stock, veneer, and pulpwood
Street, shade, or ornamental tree - unusual
flowers
Food for wildlife
Rapid growth for reforestation projects
1 Sources: USDA , http://www.plants.usda.gov.plantguide: U.S. Forest Service Silvics of North
2 America, http://www.na.fs.fed.us/spfo/pubs/silvics_manual; North Carolina State University,
3 http://www.ncsu.edu/proj ect/dendrology/
4
5 The National Park Service has published a list of trees and plants considered sensitive
6 because they exhibit foliar injury at or near ambient concentrations in fumigation chambers or
7 have been observed to exhibit symptoms in the field by more than one observer. This list
8 includes many species not included in Table 6-1, such as various milkweed species, asters,
9 coneflowers, huckleberry, evening primrose, Tree-of-heaven, redbud, blackberry, willow, and
10 many others. The full list is included in Appendix X and the Os ISA (EPA, 2012). Many of
6-5
-------
1 these species are important for non-timber forest products, recreation, and aesthetic value among
2 other services.
3 6.2.1 Supporting Services
4 Supporting services are the services necessary for all other services. For example nutrient
5 cycling is required for any ecosystem service including provision of food and timber. While
6 other categories of services have relatively direct or short-term impacts on people the impacts on
7 public welfare from supporting services are generally either indirect or occur over a long time.
8 The next sections describe potential impacts of 63 on some of these services.
9 6.2.1.1 Net Primary Productivity
10 The ISA determined that biomass loss due to exposure to may have adverse effects on net
11 primary productivity (NPP). According to Pan et al. (2009) net primary productivity in U.S.
12 Mid-Atlantic temperate forests decreased 7-8% per year from 1991-2000 due to Os exposure
13 when compared to preindustrial conditions in 1860 even with growth stimulation provided by
14 elevated carbon dioxide and nitrogen deposition. In another study Felzer et al. (2004) estimated
15 Os impact on NPP for the conterminous U.S from 1950-1995 compared to a presumed pristine
16 condition in 1860. They found the largest decreases in NPP occurred in the agricultural region
17 of the Midwest during the mid-summer. This decrease was as high as 13% per year in some
18 areas. Primary productivity underlies the provision of many subsequent services that are highly
19 valued by the public including provision of food and timber. Due to data and methodology
20 limitations the loss of value to the public due to the negative effects of O3 exposure on this
21 supporting service is unquantifiable.
22 6.2.1.2 Community Composition
23 Community composition or structure is also affected by Os exposure. Since species vary
24 in their response to Os those species that are more resistant to the negative effects of Os are able
25 to out-compete the more susceptible species. For example in the San Bernardino area Arbaugh
26 et al. (2003) have shown that community composition in high Os sites has shifted toward Os
27 tolerant species such as white fir, sugar pine, and incense cedar at the expense of ponderosa and
28 Jeffrey pine. Changes in community composition underlie possible changes in associated
29 services such as herbivore grazing, production of preferred species of timber, and preservation of
30 unique or endangered communities or species among others. See Figure 5-17 for a map showing
31 current W126 Os levels in critical habitat areas.
6-6
-------
1 6.2.2 Regulating Services
2 Regulating services as defined by the MEA (2005) are those that regulate ecosystem
3 processes. Services such as air quality, water, climate, erosion, and pollination regulation fit
4 within this category. The next sections describe potential impacts of Os on some of these
5 services.
6 6.2.2.1 Climate Regulation
7 Biomass loss due to Os exposure affects climate regulation by ecosystems by affecting
8 carbon sequestration by plants and trees. Reduction of carbon uptake by forests results in more
9 carbon in the atmosphere and negative effects on climate. The studies cited in the ISA show a
10 consistent pattern of decrease in carbon uptake because of 63 damage with some of the largest
11 reductions projected over North America. In one simulation (Sitch et al., 2007) the indirect
12 radiative forcing due to Os effects on carbon uptake by plants could be even greater than the
13 direct effect of Os on climate change.
14 The Forest and Agriculture Sectors Optimization Model - Greenhouse Gas version
15 (FASOMGHG) can calculate the difference in carbon sequestration by forests and agriculture
16 due to biomass loss caused by 63 exposure. Details of the model itself and the analyses done for
17 this risk and exposure assessment are available in Appendix X. [We will be providing results in
18 terms tons carbon sequestered in supplemental materials. We will model current ambient
19 conditions and the results of simulations just meeting the current standard.]
20 In addition to its direct impacts on vegetation, O3 is a well-known greenhouse gas that
21 contributes to climate warming (U.S. EPA, 2012a). A change in the abundance of tropospheric
22 Os perturbs the radiative balance of the atmosphere, an effect quantified by the radiative forcing
23 metric. The IPCC (2007) reported a radiative forcing of 0.35 W/m2 for the change in
24 tropospheric O?, since the preindustrial era, ranking it third in importance after the greenhouse
25 gases CC>2 (1.66 W/m2) and CH4 (0.48 W/m2). The earth-atmosphere-ocean system responds to
26 the radiative forcing with a climate response, typically expressed as a change in surface
27 temperature. Finally, the climate response causes downstream climate-related ecosystem effects,
28 such as redistribution of ecosystem characteristics due to temperature changes. While the global
29 radiative forcing impact of Os is generally well understood, the downstream effects of the Os-
30 induced climate response on ecosystems remain highly uncertain.
6-7
-------
1 Since 63 is not emitted directly but is photochemically formed in the atmosphere, it is
2 necessary to consider the climate effects of different 63 precursor emissions. Controlling
3 methane, CO, and non-methane VOCs may be a promising means of simultaneously mitigating
4 climate change and reducing global Os concentrations (West et al. 2007). Reducing these
5 precursors reduces global concentrations of the hydroxyl radical (OH), their main sink in the
6 atmosphere, feeding back on their lifetime and further reducing Os production. In contrast, NOx
7 reductions decrease OH, leading to increased methane lifetime and increased Os production
8 globally in the long-term. The resulting positive radiative forcing from increased methane may
9 cancel or even slightly exceed the negative forcing from decreased Os globally (West et al.
10 2007). Of the Os precursors, methane abatement reduces climate forcing most per unit emission
11 reduction, as methane produces O3 on decadal and global scales and is itself a strong climate
12 forcer. Since they may have different effects on concentrations of different species in the
13 atmosphere, all Os precursors must be considered in evaluating the net climate impact of
14 emission sources or mitigation strategies.
15 6.2.2.2 Hydrologic Cycle
16 Regulation of the water cycle is yet another ecosystem service that can be adversely
17 affected by the effects of Os on plants. McLaughlin et al. (2007) reported that increased water
18 use by Os impacted forests decreased modeled late-season stream flow in watersheds in eastern
19 Tennessee in or near the Great Smoky Mountains. Ecosystem services potentially affected by
20 such a loss in stream flow could include habitat for species such as trout that are dependent on an
21 optimum stream flow or temperature. Downstream effects could potentially include a reduction
22 in the quantity and/or quality of water available for irrigation or drinking water, and recreational
23 use. The United States Forest Service (U.S. FS) and the National Oceanographic and
24 Atmospheric Administration (NOAA) jointly surveyed Americans age 16 and over for the report
25 on Uses and Values of Wildlife and Wilderness in the United States as part of the National
26 Survey on Recreation and the Environment (NSRE) (U.S.D.A., 2002). The NSRE (U.S.D.A.,
27 2002) specifically asked for respondents to rank the importance of water quality as a benefit of
28 wilderness. 91% of respondents ranked water quality protection as either extremely or very
29 important. Less than 1% of respondent s ranked this service as not important at all.
6-8
-------
1 6.2.2.3 Fire Regulation
2 Fire regime regulation is also negatively affected by 63 exposure. Grulke et al. (2008)
3 reported various lines of evidence indicating that O3 pollution may contribute to forest
4 susceptibility to wildfires by increasing leaf turnover rates, and litter thereby creating increased
5 fuel loads on the forest floor, Os increased drought stress, and, because both foliar and root
6 biomass are negatively affected, trees store carbohydrates in the bole over winter increasing
7 susceptibility to bark beetle attack. Taken together these factors increase susceptibility to
8 wildfire. In the United States in 2010 over 3 million acres burned in wildland fires and an
9 additional 2 million acres were burned in prescribed fires according to the National Interagency
10 Fire Center (http://www.nifc.gov/firelnfo/firelnfo statisties.html). Over the 5-year period from
11 2004 to 2008 Southern California alone experienced, on average, over 4,000 fires a year burning,
12 on average, over 400,000 acres (National Association of State Foresters [NASF], 2009).
13 The short-term benefits of reducing the Os related fire risks include the value of avoided
14 residential property damages, avoided damages to timber, rangeland, and wildlife resources;
15 avoided losses from fire-related air quality impairments; avoided deaths and injury due to fire;
16 improved outdoor recreation opportunities; and savings in costs associated with fighting the fires
17 and protecting lives and property. For example, the California Department of Forestry and Fire
18 Protection (CAL FIRE) estimated that average annual losses to homes due to wildfire from 1984
19 to 1994 were $163 million per year (CAL FIRE, 1996) and were over $250 million in 2007
20 (CAL FIRE, 2008). In fiscal year 2008, CAL FIRE's costs for fire suppression activities were
21 nearly $300 million (CAL FIRE, 2008). Figure 6- 1 shows current ambient Os levels over the
22 fire risk in California. The highest fire risk and highest Os levels overlap with each other and
23 significant portions of the California range of species sensitive to Os damage specifically
24 ponderosa pine.
6-9
-------
Fire Risk and Current Ozone (W126) Levels
PonderosaPine Range -
California
Fire Risk and W126
- 5.0
5.1-10.0
10 1 - 150
15.1 -20.0
20 1 -250
25 1 -30.0
30 1 -35.0
• 35 1 -650
Fire Risk
I '
I
2
3
2
3 Figure 6- 1 Overlap of fire risk, current Os levels and California range of
4 ponderosa pine
5 In the long term, decreased frequency of fires could result in an increase in property
6 values in fire-prone areas. Mueller et al. (2007) conducted a hedonic pricing study to determine
7 whether increasing numbers of wildfires affect house prices in southern California. They
8 estimated that house prices would decrease 9.71% after one fire and 22.7% after a second
9 wildfire within 1.75 miles of a house in their study area. After the second fire, the housing prices
10 took between 5 and 7 years to recover.
11 Additionally, long term decreases in wildfire would be expected to yield outdoor
12 recreation benefits consistent with the discussion of scenic beauty in subsequent sections.
13 6.2.2.4 Pollination
14 The ISA 03(18A) (2011 ref) identifies Os as a possible agent affecting the travel distance
15 and loss of specificity of volatile organic compounds emitted by plants, some of which act as
6-10
-------
1 scent cues for pollinators. While it isn't possible to calculate the loss of pollination services due
2 to this negative effect on scent the loss is embedded in the current estimated value of all
3 pollination services, managed and wild, in North America (U.S., Canada, and Bermuda) which is
4 $18.3 billion dollars in 2010 (Gallai et al., 2009).
5 6.2.3 Provisioning Services
6 Provisioning services include market goods such as forest and agricultural products. The
7 direct impact of Os exposure induced biomass loss can be predicted for the commercial timber
8 and agriculture markets using the Forest and Agriculture Optimization Model (FASOM). This
9 model provides a national scale estimate of the effects of 63 on these two market sectors
10 including producer and consumer surplus estimates. Non-timber forest products (NTFP) such as
11 foliage and branches used for arts and crafts or edible fruits, nuts, and berries can be affected by
12 the impact of Os through biomass loss and foliar injury. USDA has assessed the harvest and
13 market value of these products in commercial markets. There is as well a significant portion of
14 NTFP that are valuable to subsistence gatherers. Subsistence practices are much more difficult
15 to assess as these forest users are not required to obtain a permit for use of federal public lands
16 and are therefore more difficult to enumerate.
17 6.2.3.1 Commercial Timber and Agriculture
18 We used FASOMGHG (Forest and Agricultural Sector Optimization Model—Green
19 House Gas version) to calculate the resulting market-based welfare effects of 63 exposure in the
20 forest and agricultural sectors of the United States. Even though agricultural impacts are not a
21 focus of this risk assessment, a proper understanding of impacts on commercial forests requires
22 us to model the effects of Os on agriculture because of the interactions between competing
23 demands for land for forestry versus agricultural crops. We used data obtained from the Forest
24 Inventory and Analysis National Program (FIA) and the O?, related biomass loss concentration-
25 response functions from the ISA as inputs into FASOMGHG, which enabled us to adapt the
26 growth rates for tree and crop species in the model to account for the impact of 63 on vegetation.
27 See Appendix X for a full discussion of the model and methodology.
28 [We will provide results of modeling runs for current ambient conditions and meeting current
29 standards in supplemental materials. The results will be in terms of timber and crop yield loss.]
30 In addition to the direct effects of Os on tree growth Os causes increased susceptibility to
31 infestation by some chewing insects (USEPA, 2006). Chewing insects include the southern pine
6-11
-------
1 beetle and western bark beetle, species that are of particular interest to commercial timber
2 producers and consumers. These infestations can cause economically significant damage to tree
3 stands and the associated timber production. Figure 6- 2 and Figure 6- 3 illustrate the damage
4 caused by southern pine beetles in parts of the south.
5
6
7 Figure 6- 2 Southern pine beetle damage. Courtesy: Ronald F. Billings, Texas
8 Forest Service. Bugwood.org
9 According to the USDA Forest Service Report on the Southern Pine Beetle (Coulson and
10 Klepzig, 2011), "Economic impacts to timber producers and wood-products firms are essential to
11 consider because the SPB causes extensive mortality in forests that have high commercial value
12 in a region with the most active timber market in the world." The economic impacts of beetle
13 outbreaks are multidimensional. In the short term the surge in timber supply caused by owners
14 harvesting damaged timber depresses prices for timber and benefits consumers. In the long term
15 beetle outbreaks reduce the stock of timber available for harvest, raising timber prices to the
16 benefit of producers and the detriment of consumers. However, USDA estimates that these long
17 term impacts are much smaller than the short term impacts.
18 The Forest Service further reports that over the 28 years covered in their analysis (1977-
19 2004) timber producers have incurred about $1.4 billion or about $49 million per year and
20 wood-using firms have gained about $966 million or about $35 million per year due to beetle
21 outbreaks. This results in a net $15 million per year economic impact. All dollar values are
22 reported in constant $2010. These annual figures mask the fact that most of the economic
23 impacts are the result of a few catastrophic outbreaks causing the impacts to pulse through the
24 system in large chunks rather than being evenly distributed over the years. It is not possible to
6-12
-------
1
2
3
4
attribute a portion of these impacts due to the effect of 63 on trees' susceptibility to insect attack
however, such losses are already embedded within the losses quoted and any welfare gains from
decreased O3 would positively impact these numbers.
* ft
5 '' .' \ X;/VV
6 Figure 6- 3 Southern pine beetle damage. Courtesy: Ronald F. Billings, Texas
7 Forest Service. Bugwood.org
8 In the western United States Os sensitive ponderosa and Jeffrey pines are subject to attack
9 by bark beetles. Figure 6- 4 shows western bark beetle mortality from 2003- 2007. The map
10 includes Douglas fir and other western species vulnerable to bark beetles as well as ponderosa
11 and Jeffrey pine. According to the Western Forestry Leadership Coalition (2009) approximately
12 22 million acres of forest lands are at risk for bark beetle damage.
6-13
-------
7
L
i^fc-i
: *^ ,<**.
'«. ... S|«
^ %.
>^V - -il i» • ' > 1
•:»& *+ 1
A~ » ** »_ * 4
I
2 Figure 6- 4 Western bark beetle mortality obtained from State and Private
3 Forestry aerial-detection surveys (2003-2007). Source: Western
4 Forestry Leadership Coalition (2009) [This figure will be updated
5 with Os concentrations in supplemental materials.]
6
7 In 2006 the California was the largest producer of ponderosa and Jeffrey pine timber
8 from public lands. California accounted for 99 million board feet of saw logs - almost 40% of
9 the total production (U.S. Forest Service, 2009 available at:
10 http://srsfia2.fs.fed.us/php/tpo_2009/tpo_rpa_int2.php). California also experiences high 63
11 levels that may contribute to susceptibility to bark beetle attack. While it isn't possible to
12 attribute a quantified impact of 63 to economic loss due to bark beetle damage that impact is
13 already accounted for within the loss attributed to bark beetle infestation. Reducing O3 impacts
14 would likely reduce economic loss to California timber production.
15 The photographs and map above illustrate the impact insect outbreaks can have major
16 effects on aesthetic values such as scenic beauty in addition to the impacts on timber production.
6-14
-------
1 The value of the impact of 63 and insect attack susceptibility on aesthetic values, as shown in the
2 Nox/SOx Policy Assessment (EPA, 2011), may be even greater than the market value of the
3 timber. We will address those impacts in Section 6.2.4.
4 6.2.3.2 Commercial Non-Timber Forest Products
5 In addition to timber forests provide many other products that are harvested for
6 commercial or subsistence activities. These products include:
7 • edible fruits, nuts, berries, and sap
8 • foliage, needles, boughs, and bark
9 • transplants
10 • grass, hay, alfalfa, and forage
11 • herbs and medicinals
12 • fuelwood, posts and poles
13 • Christmas trees
14 For the 2010 National Report on Sustainable Forests (USD A, 2011) these products were divided
15 into several categories including nursery and landscaping uses; arts, crafts, and floral uses;
16 regeneration and silviculture uses. Table 6- 2 details selected categories of non-timber forest
17 products (NTFP) harvested by permit in 2007. These harvests are reported in measures relevant
18 to the specific articles i.e., bushels of cones, tons of foliage or boughs, individual transplants.
19
20 Table 6- 2 Quantity of non-timber forest products harvested on U.S. Forest Service and
21 Bureau of Land Management land
Product Category
Arts, crafts, and florals
Christmas trees
Edible Fruits, nuts, berries,
and sap
Unit
Bushels
Pounds
Tons
Each
Lineal foot
Bushels
Pounds
Harvest All U.S.
70,222
3,442,125
620,773
151,274
94.758
250
1,614,565
6-15
-------
Product Category
Fuelwood
Grass, hay, and alfalfa
Forage
Herbs and medicinals
Nursery and landscape
Regeneration and
silviculture
Posts and poles
Unit
Syrup Taps
ccf
Cords
Pounds
Tons
Pounds
Each
Pounds
Tons
Bushels
ccf
Each
Pounds
Tons
ccf
Each
Lineal foot
Harvest All U.S.
10,686
35,800
417,692
4,265,952
480
101,365
766,645
25,689
316
7,627
8
21,265
247,543
110,873
5,281
1,684,618
326,312
1 Note: ccf = 100 cubic feet Source: USD A 2011
2
3 According to the ISA 63 exposure causes biomass loss in sensitive woody and
4 herbaceous species which in turn could affect forest products used for arts, crafts, and florals.
5 For example, Douglas fir and red alder among others are used on the Pacific Coast for arts and
6 crafts, particularly holiday crafts and decorations. The effects of Os on plant reproduction (see
7 ISA Table 9-1, 2012) could affect the supply of seeds, berries, and cones. Foliar injury impacts
8 on Os sensitive plants would potentially affect the harvest of leaves, needles, and flowers from
9 these plants for decorative uses. Likewise the same Os effects would impact harvest of edible
10 fruits, nuts, berries, and sap. Note that this category includes blueberries, pine nuts, and sap for
11 maple syrup to name just a few. The use of native grasses as forage is a significant aspect of
6-16
-------
1 forest-land management in the western U.S. (Alexander et al. 2002). Os effects on community
2 composition particularly changes in the ratio of grasses to forbs (broad-leaved herbs other than a
3 grass) and nutritive quality of grasses can have effects on rangeland quality for some herbivores
4 (Krupa et al., 2004, Sanz et al., 2005), and therefore effects on grazing efficiency. The negative
5 impacts of Os on plants would similarly affect the harvest in the rest of the categories as well.
6 According to the Census Bureau's County Business Patterns data in 2006 this activity is
7 captured in the industry code 1132, forest nurseries and gathering of forest products, and
8 employed 2,098 people accounting for an annual payroll ($ 2006) of $71,657,000 with an
9 average annual income of $34,155 (U.S. Census Bureau, County Business Patterns, at
10 http://www.census.gov/econ/cbp/).
11 The USDA estimates the proportion of the national supply of NTFP represented by U.S.
12 FS and BLM lands is approximately 10%. Retail values for NTFPs harvested on Forest Service
13 and Bureau of Land Management lands are approximately $1.4 billion. These are very rough
14 estimates based only on permit or contract sales. These estimates could be low due to harvests
15 taken without permit or contract and sold through complex commodity chains that can combine
16 wild-harvested and agriculturally grown commodities.
17 It is important to realize that while we cannot estimate the loss of production and
18 therefore values for the loss of benefit to this sector that is due strictly to the effects of 63 those
19 losses are already embedded within the harvest and values reported here.
20 The preceding paragraphs detailed the harvest and value of permit or contract sales of
21 NTFPs on Forest Service and BLM managed lands. Since permits or contracts are not required
22 for gathering activities for personal use the analyses done by USDA are not able to account for
23 the subsistence use of non-timber forest products.
24 6.2.3.3 Informal Economy or Subsistence Use of Non-Timber Forest Products
25 Most people gathering NTFPs are doing so for personal use (Baumflek et al., 2010 and
26 USDA, 2011). In fact by one estimate (Baumflek et al., 2010) up to 80% of the people collecting
27 NTFPs in Oregon and Washington are collecting for personal reasons. Such personal use may be
28 characterized as either part of the informal economy or as subsistence activity. Participants in
29 the informal economy may earn a wage or salary and participate in gathering NTFPs for other
30 reasons than recreation (Brown et al., 1998). The term subsistence has usually been applied to
31 special groups such as Native Americans or the Hmong people. The term "subsistence" has
6-17
-------
1 generally been understood to imply an extremity of poverty such that these activities are essential
2 to a minimum of the necessities of life (Freeman, 1993). However, Freeman points out
3 researchers stress that economic goals are only a part of the impetus for these activities.
4 Brown (1998) proposed a composite definition that captures both the informal economy
5 as practiced by those who are not necessarily a part of a special population and subsistence as
6 generally referenced to those special populations. "Subsistence refers to activities in addition to,
7 not in place of, wage labor engaged in on a more or less regular basis by group members known
8 to each other in order to maintain a desired and/or normative level of social and economic
9 existence." This definition allows consideration of the cultural and social aspects of subsistence
10 lifestyles. These non-economic benefits range from maintenance of social ties and relationships
11 through shared activity to family cohesiveness to retreatism and a sense of self-reliance for the
12 individual practitioner (Brown et al., 1998).
13 While there is general acknowledgement of subsistence activities by Native Americans
14 and specific treaty rights for tribes guaranteeing access to lands for hunting, fishing, and
15 gathering there has been a lack of research focused on other populations (Emery and Pierce,
16 2005). However there are some studies that make it clear that subsistence activities provide
17 valued resources for a variety of people in the coterminous United States. Baumflek et al. and
18 Alexander et al. (2010 and 2011) have documented the collection and use of culturally and
19 economically important NTFPs in Maine and the eastern United States respectively. Brown et
20 al. (1998) reports on subsistence activities among residents of the Mississippi Delta. Emery
21 (2003) and Hufford (2000) examine activities in the Appalachians and Pena (1999) reports
22 activities by Latinos in the Southwest.
23 As with the commercial harvest of NTFPs subsistence gathering of these forest products
24 can potentially be affected by the adverse effects of Os on growth, reproduction, and foliar injury
25 to the sensitive plants in use for nutrition, medicine, cultural, and decorative purposes. It is
26 important to note that some plants may have more than one use or significance. For example, the
27 Mi'kmaq and Maliseet Indian tribes in Maine do not differentiate between blueberries'
28 nutritional, medicinal, and spiritual uses. Blueberries are a food, and a medicine that is often
29 incorporated into ceremonies (Baumflek et al., 2010). And while we cannot quantify the size of
30 the harvest of subsistence gathered items or monetize the loss of benefit due to Os effects a
6-18
-------
1 comparison to the commercial harvest may provide perspective on the significance of these
2 activities to the people who engage in them.
3 6.2.4 Cultural Services
4 Cultural services include recreation, habitat for endangered species, and non-use values
5 (i.e., existence and bequest values) that can be directly or indirectly impacted by O3 exposure.
6 The foliar injury induced by Os exposure may have a negative impact on people's satisfaction
7 with outdoor activities especially those associated with natural environments. Slowed growth or
8 changes in community composition may impact habitat for endangered species both flora and
9 fauna. Non-use values are impacted as well. According to responses to the National Survey on
10 Recreation and the Environment large majorities of Americans wish to preserve natural or
11 pristine areas even if they do not intend to visit themselves.
12 According to the National Report on Sustainable Forests (USDA, 2011) there are
13 approximately 751 m (Figure 6- 5); one-third is federally owned. All of these lands are assumed
14 to be protected to some degree but specific protections apply to wilderness areas which comprise
15 about 20% of public land, 7% is protected as national parks, 13% is designated as wildlife
16 refuges while 60% is protected managed forests including national forests, BLM lands and other
17 state and local government lands. The protections afford preservation of cultural, social, and
18 spiritual values.
„ , Corporate forest
Other corporate,.. . ,
,„,-„, \ industry, 6.8% _ Local, 1.5%
u'5% \ /
State, 9.2%
Other
noncorporate, 2.9%.
^g Forest land ownership (percent)
20 Figure 6- 5 Percent of forest land in the United States by ownership category, 2007
21 (percentages sum to 100) (Almost all forest lands are open for some form of
22 recreation, although who may have access may be restricted). Source: USDA
23 Forest Service
24
6-19
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
6.2.4.1 Non-Use Services
The National Survey on Recreation and the Environment (NSRE) (USDA, 2002) is an
ongoing survey of a random sample of adults over the age of 16 on their interactions with the
environment. NSRE surveys track American's attitudes toward various benefits derived from
the environment including non-use values. When people value a resource even though they may
never visit the resource or derive any tangible benefit from it they perceive an existence service.
When the resource is valued as a legacy to future generations a bequest service exists.
Additionally there exists an option value to knowing that you may visit a resource at some point
in the future. Data provided by the NSRE indicates that Americans have very strong preferences
for existence, option, and bequest services related to forests. Significantly, according to the
survey, only 5% of Americans rate wood products as the most important value of public forests
and wilderness areas and even for private forests only 20% of respondents rated wood products
as most important. Table 6- 3 details the survey responses to these questions.
Table 6- 3 NSRE Reponses to Non-Use Value Questions
Service
Existence
Option
Bequest
Extremely
Important
36
36
81
Very Important
38
37
12
Moderately
Important
18
17
4
Total
92
90
97
16
17
18
19
20
21
22
23
24
25
26
Studies (Haefele et al., 1991, Holmes and Kramer, 1996) indicate that the public places a
high value on protecting forests and wilderness areas from the damaging effects of air pollution.
Studies conducted to assess willingness-to-pay (WTP) for forest protection for spruce-fir forests
in the southeast from air pollution and insect damage (Haefele et al., 1991, Holmes and Kramer,
1996) confirm that the non-use values held by the respondents to the survey were in fact greater
than the use or recreation values. The survey presented respondents with a sheet of color
photographs representing three stages of forest decline and explained that, without forest
protection programs, high-elevation spruce forests would all decline to worst conditions. Two
potential forest protection programs were proposed. The first program would protect the forests
along road, and trail corridors spanning approximately 1/3 of the ecosystem at risk. This level of
6-20
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
protection may be most appealing to recreational users. The second level of protection was for
the entire ecosystem and may be most appealing to those who value the continued existence of
the entire ecosystem. Median household WTP was estimated to be roughly $29 (in 2007 dollars)
for the minimal program and $44 for the more extensive program. Respondents were then asked
to decompose their value for the extensive program into use, bequest, and existence values. This
resulted in values that represented components of 13% use value, 30% bequest, 57% existence
value (Table 6- 4).
While these studies are specific to damage due to excess nitrogen deposition and the
wooly balsam adelgid (a pest in frasier fir) the results are relevant to 63 exposure in forests. In
the southeast loblolly pine is a prevalent species and 63 foliar injury can cause visible damage.
O3 exposure may result trees to be more susceptible to insect attack which in the southeast would
include damage caused by the southern pine beetle.
Table 6- 4 Value Components for WTP for Extensive Protection Program for Southern
Appalachian Spruce-Fir Forests
Type of Value
Use
Bequest
Existence
Total
Proportion of WTP
0.13
0.30
0.57
1.0
Component Value in $2007
5.72
13.20
25.08
44.00
16
17
18
19
20
21
22
6.2.4.2 Habitat Provision
In addition to non-use values the NSRE provides data on the values survey respondents
place on the provision of habitat for wild plants and animals. Table 6- 5 summarizes the
responses to survey questions regarding the value of wildlife habitat and preservation of unique
or endangered species.
6-21
-------
1
2
Table 6- 5 NSRE Reponses to Wildlife Value Questions
Service
Wildlife Habitat
Preserving
Unique Wild
Plants and
Animals
Protecting Rare
or Endangered
Species
Extremely
Important
51
44
50
Very Important
36
36
33
Moderately
Important
9
13
11
Total
96
93
94
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
There exist meta-analyses on the monetary values Americans place on threatened and
endangered species. One such study (Richardson and Loomis, 2009) estimates the average
annual willingness to pay for a number of species. The authors report a wide range of values
dependent on the change in the size of the species population, type of species, and whether
visitors or households are valuing the species. The average annual WTP for surveyed species
ranged from $9/year for striped shiner to $26I/year for Washington state anadromous fish,
hatched in fresh water, spends most of its life in the sea and returns to fresh water to spawn, populations
in constant 2010$.
6.2.4.3 Aesthetic Value
Aesthetic services not related to recreation include the view of the landscape from
houses, as individuals commute, and as individuals go about their daily routine in a nearby
community. Studies find that scenic landscapes are capitalized into the price of housing. Studies
document the existence of housing price premia associated with proximity to forest and open
space (Acharya and Bennett, 2001; Geoghegan, Wainger, and Bockstael, 1997; Irwin, 2002;
Mansfield et al., 2005; Smith et al., 2002; Tyrvainen and Miettinen, 2000). In fact according to
Butler (2008) approximately 65% of private forest owners rate providing scenic beauty as either
a very important or important reason for their ownership of forest land.
6-22
-------
1 These services are at risk of impairment due to (Vinduced damage: directly due to foliar
2 injury, and indirectly due to increased susceptibility to insect attack. Data is not available to
3 quantify these negative effects however the damage would be included in the price premia
4 already mentioned. In other words, without such damage the associated price premia for scenic
5 beauty incorporated into housing prices would likely be higher.
6 6.2.4.4 Recreation
7 With few exceptions, publicly owned forests at all levels are open for some form of
8 recreation. Based on the analysis done for the USDA Report on Sustainable Forests referenced
9 in Section 5.1.4 almost all of the 751 million acres of forest land are at least partially managed
10 for recreation. Of the 751 million acres 44% are publicly owned (federal, state, or local).
11 Americans enjoy a wide variety of outdoor pursuits many of which are subject to
12 negative impacts due to Os exposure especially its effect on foliage, insect susceptibility, habitat,
13 and community composition. The effects related to scenic beauty (foliar injury and insect
14 damage) affect not only the scenery viewing but satisfaction with other scenery dependent
15 activities. 97% of NSRE survey respondents rated scenic beauty as an important to extremely
16 important aspect of their wilderness experience.
17 Scenic quality has been found to be strongly correlated to recreation potential and the
18 likelihood of visiting recreation settings and the correlations apply to both active and passive
19 recreational pursuits (Ribe, 1994). According to Ribe (1994), differences in scenic beauty
20 account for 90% of the variation in participant satisfaction across all recreation types.
21 Perceptions of scenic beauty are dependent on a number of forest attributes including the
22 appearance of health and the effects of air pollution and insect damage, visual variety, species
23 variety, and lush ground cover (Ribe 1989). The ISA concludes that there is a causal relationship
24 between Os exposure and visible foliar injury. Chapter 5 of this document also discusses the
25 effects of 63 on foliar injury. Figure 6- 6 shows the effects of foliar injury on ponderosa pine,
26 milkweed, and tulip poplar. The presence of downed wood, whether caused by 63 mortality,
27 insect attack, or slash from harvest activities has a negative impact on scenic beauty assessments
28 (Ribe, 1989; Buyhoff, et al, 1982). Species composition of forests may also influence
29 preferences. According to Ribe (1982) these preferences may be affected by cultural, regional,
30 or contextual expectations which would include the expectation of the presence of certain species
31 in specific areas such as the presence of ponderosa pine in California. Additionally there is a
6-23
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
positive effect for ground cover rather than bare or disturbed soil (Brown and Daniel, 1984,
1986). Thus the damage to scenic beauty 63 inflicts on sensitive plants by way of foliar injury
extends beyond large trees to the grasses, forbs, ferns, and shrubs that comprise the understory of
a forest setting.
Figure 6- 6 Examples of foliar injury due to Os exposure. Courtesy: National
Park Service
The NSRE provides estimates of participation in many recreation activities. According
to the survey some of the most popular outdoor activities are walking including day hiking and
backpacking, camping, bird watching, wildlife watching, and nature viewing. Participant
satisfaction with these activities is wholly or partially dependent on the quality of the natural
scenery. Table 6- 6 summarizes the survey results for these and other popular activities
including the percent participation and the number of participants nationally, the number of days
participants engage in recreation activities annually, and their WTP for their participation.
6-24
-------
1 Table 6- 6 National Outdoor Activity Participation
Activity
Day Hiking
Backpacking
Picnicking
Camping (developed and
primitive sites)
Visit a wilderness area
Birdwatching/Photography
Wildlife
watching/Photography
Natural vegetation
viewing/Photography
Natural scenery
viewing/Photography
Sightseeing
Gathering (mushrooms,
berries, firewood)
%
Participation
32.4
10.4
54.9
42.3
32.0
31.8
44.2
43.9
59.6
50.8
28.6
#
Participants3
69.1
22.2
116.9
90.1
68.2
67.7
94.2
93.6
126.9
108.2
60.9
#
Activity
Days"
2,508
224.0
935.2
757.5
975.4
5,828.1
3,616.5
5,720.8
7,119.7
2,055.0
852.7
Mean
WTP/Dayb
60.63
13.33
20.70
19.98
N/A
49.74
48.72
N/A
N/A
45.94
N/A
Mean Total
Participation
Value3'"
152,060
2,986
19,359
15,135
N/A
289,773
176,196
N/A
N/A
94,407
N/A
2
3
4
5
6
7
8
9
10
11
12
Source: NSRE 2000-2001 and 2003 National Report on Sustainable Forest Management 2003
National Report: Documentation for Indicators 35, 36, 37, 42, and 43 available at:
http://warnell.forestry.uga.edu/nrrt/NSRE/MontreallndDoc.PDF and Recreation Values
Database available at: http://recvaluation.forestry.oregonstate.edu/
a in millions, b$ 2010, N/A not available
The relationship between scenic beauty and recreation satisfaction for camping has been
quantified by Daniel, et al (1989) in a contingent valuation study. The authors surveyed campers
regarding their perceptions of scenic beauty, as indicated by a photo array of scenes along a
spectrum of scenic beauty, and their willingness to pay (WTP) to camp in certain areas. All else
being equal scenic beauty and WTP demonstrated a nearly perfect linear relationship (correlation
6-25
-------
1 coefficient of 0.96). This suggests that campers would likely have a greater willingness to pay
2 for recreation experiences in areas where scenic beauty is less damaged by 63. As mentioned
3 previously Ribe (1994) found that scenic beauty plays a strong role in recreation satisfaction and,
4 in fact, explains 90% of the difference in recreation satisfaction among all types of outdoor
5 recreation there is reason to believe that this linear relationship between scenic beauty and WTP
6 would hold across all recreation types. It would follow that decreases in Os damage would
7 generate benefits to all recreators. We cannot estimate the incremental impact of reducing Os
8 damage to scenic beauty and subsequent recreation demand however given the large number of
9 outdoor recreation participants and their substantial WTP for recreation even very small
10 increments of change in WTP or activity days will generate significant benefit to these
11 recreators.
12 Another resource for estimating consumer's economic value for their recreation
13 experiences is the data available on their actual expenditures for recreation and the total
14 economic impact of recreation activities. Economic impacts across the national economy can be
15 estimated using the IMPLAN® model, a commercially available input-output model that has
16 been used by the Department of Interior, the National Park Service, and other government
17 agencies in their analyses of economic impacts. For this document we will refer to analyses
18 done for the 2006 National Survey of Fishing, Hunting, and Wildlife-Associated Recreation
19 (FHWAR) (U.S. Department of the Interior and U.S. Department of Commerce, 2006) and an
20 analysis performed by Southwick and Associates for the Outdoor Industry Foundation , The
21 Economic Contribution of Active Outdoor Recreation - Technical Report on Methods and
22 Findings (OIF, 2006). See Appendix X for further detail.
23 The FHWAR and the OIF report provide estimates of trip and equipment related annual
24 expenditures for wildlife watching activities in the United States. The Outdoor Industry
25 Foundation (OIF) study provides estimates of recreationist's annual expenditures on trail-related
26 activities, camping, bicycling, snow-related and paddle sports. For this review we include the
27 data on trail-related activities and camping as the most relevant for analysis of Os related
28 damages.
29
6-26
-------
1 Table 6- 7 Expenditures for Wildlife-Watching, Trail, and Camp Related Recreation"
Expenditure
Type
Trip-Related
Equipment &
Services
Other
Expenditures
Grand Total for
all Expenditures
Wildlife-
Watching11
13.9
25.0
10.4
Trailc
31.8
3.6
Campc
108.6
9.3
Total0
153.3
37.9
10.4
200.1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
a in $ 2010 billion, data from 2006 FHWARC, data from 2006 OIF report, N/R not reported
According to these analyses the total expenditures across wildlife watching activities,
trail based activities, and camp based activities are approximately $200.1 billion dollars
annually. See Table 6- 7 for details. While we cannot estimate the magnitude of the impacts of
Os damage to the scenic beauty upon which satisfaction with these activities depend the losses
are embedded within the values reported.
The impact of these expenditures has a multiplier effect through the economy as a whole
which was estimated by OIF using the IMPLAN® model. The model estimates the flow of goods
and money through the economy at scales from local to national. According to the OIF report
(2006) trail activities generated over $83.7 billion dollars in total economic activity including
$33.4 billion in retail sales and $42.7 billion in salaries, wages, and business earnings. The same
report estimates the total economic activity generated by camping related recreation at $273
billion including $109.3 billion in retail sales and $139.2 billion in salaries, wages, and business
earnings. The total economic activity estimates also include state and federal tax revenues.
Assumptions and Caveats to the IMPLAN Results: Statistics regarding the precision of
the final economic impacts were not produced by OIF due to feasibility issues, Harris Interactive
survey results combine several parameters from the data, and outside data from the Census
population estimates and IMPLAN multipliers were used.
6-27
-------
1 6.3 CASE STUDY ANALYSIS
2 The next sections highlight four national parks and several urban areas selected as case
3 study areas to provide a more detailed analysis of the ecosystem services at risk due to O3
4 exposure in the protected areas of our country and in the urban areas where the majority of the
5 U.S. population lives.
6 National Parks are especially significant to the public welfare in that the public as a
7 whole, through their elected representatives, have designated these areas to be of special value by
8 creating the parks. While national parks supply supporting and regulating services this analysis
9 focuses on the cultural services these areas provide. The supporting and regulating services at
10 risk are described in the national scale analysis. Provisioning services generally do not apply
11 since timber harvest and agriculture are prohibited in the parks.
12 The criteria for selection of the specific parks included here are discussed in Chapter 5.
13 The methodology for the ecosystem services analysis for each park is consistent between the
14 case studies. For each park the maps generated in Chapter 5 were overlayed with the locations of
15 park amenities in order to illustrate the extent of O3 impacts on vegetation and that impact on the
16 activities important to park visitors. Park use surveys1 and public use statistics (National Park
17 Service Public Use Statistics Office, http://www.nature.nps.gov/stats/index.cfm) provide data on
18 numbers of visitors who engage in activities in the parks and recreation value surveys (Kaval and
19 Loomis, 2003) provide estimates of average willingness to pay for these activities within the
20 park region.
21 The National Park Service (National Park Service, 2011) has produced estimates of
22 visitor spending for each park and the impact of visitor spending on local economies surrounding
23 the parks. These analyses provide a total value related to the specific case study parks and do not
24 model changes in value due to O3 impacts. However the loss to the local economies due to O3
25 damage in the parks is captured in the current values. These values would likely be higher absent
26 O3 impacts.
27 The urban case study analysis utilizes the iTree model developed by the Forest Service to
28 quantify the benefits of urban forests. These urban forests are vulnerable to the adverse effects
29 of O3. The iTree model is designed to provide estimates of the effects of forests on carbon
1 These studies are conducted by the Visitor Services Project at the University of Idaho. Reports for individual
parks are available at: http://www.psu.idaho.edu/vsp.reports.htm
6-28
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
sequestration, volatile organic chemical production, and pollution removal and can be modified
to allow estimation of the biomass loss due to 63 exposure and that effect on services.
6.3.1 Southeast Region - Great Smokey Mountains National Park
Figure 6- 7 Mount Le Conte, Summer Great Smoky National Park. Courtesy:
National Park Service
Great Smokey Mountains National Park (GRSM) welcomed approximately 9.5 million
visitors in 2010 (NFS Public Use Statistics Office, http://www.nature.nps.gov/stats/index.cfm)
making it the most visited national park in America. Overlapping the border between North
Carolina and Tennessee the park is valued for the diversity of its vegetation and wildlife, the
scenic beauty of its mountains including the famous fogs that give the Smoky Mountains their
name, and the preservation of the remnants of Southern Appalachian culture. It is also subject to
high ambient Os levels.
As shown in Chapter 5 the extent of sensitive species coverage in GRSM is quite
substantial. The "whole park" services affected by such potential Os impacts include the
existence, option, and bequest values discussed in section 6.2.4.1 and habitat provision discussed
6-29
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
in section 6.2.4.2. Recreation value specific to the park is discussed later in this section. Focusing
the analysis showing the percent cover of foliar injury sensitive species in the park in Chapter 5
on the areas where recreation services are provided can give some perspective on the level of
potential harm to scenic beauty and therefore recreation satisfaction within the park.
The National Park Service 2002 Comprehensive Survey of the American Public Southeast
Region Technical Report includes responses from recent visitors to southeast parks about the
activities they pursued during their visit. By using the annual visitation rate from 2010 and the
regional results from the Kaval and Loomis (2003) report on recreational use values compiled for
the NFS estimates for visitors' willingness to pay for various activities was generated and
presented in Table 6-8. In addition to the activities listed in Table 6- 819% or 1.8 million park
visitors availed themselves of educational services offered at the park by participating in a
ranger-led nature tour suggesting that visitors wish to understand the ecosystems preserved in the
park.
Table 6- 8 Value of Most Frequent Visitor Activities at Great Smoky Mountains
National Park
Activity
Sightseeing
Day
Hiking
Camping
Picnicking
Total
%
Participation
82
40
19
50
# Participants
(thousands)
7,790
3,800
1,805
4,750
Mean
WTP
(in $2010)
53.34
69.93
29.87
42.42
Total Value of
Participation
(millions in $2010)
416
266
54
201
937
17
18
19
20
21
22
The report Economic Benefits to Local Communities from National Park Visitation and
Payroll (NFS, 2011) provides estimates of visitor spending and economic impacts for each park
in the system. Visitor spending and its economic impact to the surrounding area are given in
Table 6- 9 for the Great Smoky Mountain National Park. The median value of the components
of that spending is presented in Table 6- 10.
6-30
-------
1 Table 6- 9 Visitor Spending and Local Area Economic Impact of GRSM
Public Use Data
2010
recreation
Visits
9,463,538
2010
Overnight
Stays
393,812
Visitor Spending 2010a
All
Visitors
818,195
Non-Local
Visitors
792,547
Impacts on Non-Local Visitor
Spending
Jobs
11,367
Labor
Income3
303,510
Economic
Impact3
504,948
2
3
4
5
($000's) Source: Economic Benefits to Local Communities from National Park Visitation and Payroll (NPS,
2011) available at: http://www.nature.nps.gov/socialscience/docs/NPSSvstemEstimates2009.pdf
Table 6-10 Median Travel Cost for GRSM Visitors
Expense
Gas and Transportation
Lodging
Food and Drinks
Clothes, gifts, and
souvenirs
Total per visitor party
Median $ Amounts Spent (in $2010)
73
182
73
61
389
6
7
8
9
10
11
12
13
14
15
16
Source: The National Park Service 2002 Comprehensive Survey of the American Public
Southeast Region Technical Report (available at:
http: //www. nature. np s. gov/soci al sci ence/archi ve. cfm)
Each of the activities discussed above are among those shown in the national scale
analysis to be strongly affected by visitor perceptions of scenic beauty. As in the national
analysis it is not possible to assess the extent of loss of services due to impairment of scenic
beauty due to 63 damage however those losses are captured in the estimated values for spending,
economic impact, and WTP for the park.
On the other hand, we can quantify the extent of the hiking trails present in areas where
sensitive species are at risk for foliar injury. Of the approximately 1287 kilometers of trials in
GRSM, including a more than 114 km of the Appalachian Trail, over 1040 km or about 81% of
6-31
-------
1
2
3
4
5
6
trail kilometers are in areas where species sensitive to foliar injury occur. Figure 6- 8 maps the
hiking trails in GRSM including the relevant portion of the Appalachian Trail overlayed with the
species cover index. The accompanying pie chart, Figure 6- 9, shows the number of trail miles
in each cover category. The categories with species cover index from 60-160, the middle to
highest values, account for 635 km of trails or about 50% of trail kilometers.
7
8
Great Smoky Mountains National Park
Sensitive Species Cover
Index
0.0-3.5
3.6-9.9
10.0-20.9
21.0-33.3
33.4 - 45.6
45.7 - 60.9
61.0-77.9
78.0- 160.1
Appalachian Trail
Figure 6- 8 Hiking trails within GRSM and sensitive species cover
6-32
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Trail Length (KM) by Sensitive Species Cover
• No Data
• 0
Ito20
• 20 to 40
• 40 to 60
• 60 to 80
• 80 to 100
100 to 120
120 to 160
Figure 6- 9 Trail kilometers by species cover category
Although we cannot quantify the incremental loss of hiker satisfaction with their
recreation experience due to the effect of Os on scenic beauty along the trails this analysis
illustrates that very substantial numbers of trail kilometers are potentially at risk. With 3.8
million hikers using the trails every year willing to pay over $266 million for that activity the
even a small benefit of reducing 63 damage in the park could be significant for these park
visitors.
[We will produce other maps of amenities (camp sites) and overlays of sensitive species for 2'
draft.]
nd
6-33
-------
1 6.3.2 Intermountain Region - Rocky Mountain National Park
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Figure 6- 10 Sheep Lakes, Rocky Mountain National Park. Courtesy: National
Park Service
Rocky Mountain National Park welcomed 3.0 million visitors in 2010 (NFS Public Use
Statistics Office, http://www.nature.nps.gov/stats/index.cfm) to its 415 square miles of mountain
ecosystems. Rocky Mountain National Park allows visitors to enjoy vegetation and wildlife
unique to these ecosystems along over 300 miles of hiking trails.
[We will produce maps of amenities (hiking trails, camp sites) and overlays of sensitive species
for 2nd draft]
The National Park Service 2002 Comprehensive Survey of the American Public
Intermountain Region Technical Report includes responses from recent visitors to southeast
parks about the activities they pursued during their visit. By using the annual visitation rate from
2010 and the regional results from the Kaval and Loomis (2003) report on recreational use
6-34
-------
1
2
3
values compiled for the NFS estimates for visitors' willingness to pay for various activities was
generated and presented in Table 6-11.
Table 6-11 Value of Most Frequent Visitor Activities at Rocky Mountain National Park
Activity
Sightseeing
Day
Hiking
Camping
Picnicking
Total
%
Participation
85
51
27
38
# Participants
(thousands)
2,550
1,520
810
1,140
Mean
WTP
(in $2010)
28.17
46.03
41.47
33.77
Total Value of
Participation
(millions in $2010)
72
70
34
38
214
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
In addition to the activities listed in Table 6-11 11% or 330,000 park visitors availed themselves
of educational services offered at the park by participating in a ranger-led nature tour suggesting
that visitors wish to understand the ecosystems preserved in the park.
Each of the activities discussed above are among those shown in the national scale
analysis to be strongly affected by visitor perceptions of scenic beauty. As in the national
analysis it is not possible to assess the extent of loss of services due to impairment of scenic
beauty due to 63 damage; however those losses are captured in the estimated values for
spending, economic impact, and WTP for the park. Were 63 impacts decreased these estimates
would likely be higher.
The report Economic Benefits to Local Communities from National Park Visitation and
Payroll (NFS, 2011) provides estimates of visitor spending and economic impacts for each park
in the system. Visitor spending and its economic impact to the surrounding area are given in
Table 6- 12 for the Rocky Mountain National Park. The median value of the components of that
spending is presented in Table 6-13.
6-35
-------
1
2
Table 6- 12 Visitor Spending and Local Area Economic Impact of Rocky Mountain
National Park
Public Use Data
2010
Recreation
Visits
2,955,821
2010
Overnight
Stays
174,202
Visitor Spending 2010a
All
Visitors
170,804
Non-Local
Visitors
170,804
Impacts on Non-Local Visitor
Spending
Jobs
2,641
Labor
Income3
77,625
Economic
Impact3
129,666
3
4
5
6
7
a($000's) Source: Economic Benefits to Local Communities from National Park Visitation and
Payroll (NFS, 2011) available at:
http://www.nature.nps.gov/socialscience/docs/NPSSystemEstimates2009.pdf
Table 6- 13 Median Travel Cost for Rocky Mountain National Park Visitors
Expense
Gas and Transportation
Lodging
Food and Drinks
Clothes, gifts, and
souvenirs
Total per visitor party
Median $ Amounts Spent (in $2010)
63
100
63
45
271
8
9
10
Source: The National Park Service 2002 Comprehensive Survey of the American Public
Intermountain Region Technical Report (available at:
http: //www. nature .nps.gov/socialsci ence/archi ve. cfm)
6-36
-------
1 6.3.3 Pacific West Region - Sequoia/Kings Canyon National Parks
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Figure 6-11 Kings Canyon. Courtesy: National Park Service
Sequoia/Kings Canyon National Parks are located in the southern Sierra Nevada
Mountains east of the San Joaquin Valley in California. The two parks welcomed 1.6 million
visitors in 2010 (NFS Public Use Statistics Office, http://www.nature.nps.gov/stats/index.cfm) to
experience the beauty and diversity of some of California's iconic ecosystems.
[We will produce maps of amenities (hiking trails, camp sites) and overlays of sensitive species
for 2nd draft]
The National Park Service 2002 Comprehensive Survey of the American Public Pacific
West Region Technical Report includes responses from recent visitors to southeast parks about
the activities they pursued during their visit. By using the annual visitation rate from 2010 and
the regional results from the Kaval and Loomis (2003) report on recreational use values
compiled for the NFS estimates for visitors' willingness to pay for various activities was
generated and presented in Table 6- 14.
6-37
-------
1 Table 6- 14 Value of Most Frequent Visitor Activities at Sequoia/Kings Canyon National
2 Parks
Activity
Sightseeing
Day
Hiking
Camping
Picnicking
Total
%
Participation
81
58
33
45
# Participants
(thousands)
1,300
928
528
720
Mean
WTP
(in $2010)
24.21
27.77
124.65
76.72
Total Value of
Participation
(millions in $2010)
31
26
66
55
178
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
In addition to the activities listed in Table 6- 14 14% or 224,000 park visitors availed
themselves of educational services offered at the park by participating in a ranger-led nature tour
suggesting that visitors wish to understand the ecosystems preserved in the park.
Each of the activities discussed above are among those shown in the national scale
analysis to be strongly affected by visitor perceptions of scenic beauty. As in the national
analysis it is not possible to assess the extent of loss of services due to impairment of scenic
beauty due to 63 damage however those losses are captured in the estimated values for spending,
economic impact, and WTP for the park. Were O3 impacts decreased these estimates would
likely be higher.
The report Economic Benefits to Local Communities from National Park Visitation and
Payroll (NFS, 2011) provides estimates of visitor spending and economic impacts for each park
in the system. Visitor spending and its economic impact to the surrounding area are given in
Table 6-15 for the Sequoia and Kings Canyon National Parks. The median value of the
components of that spending is presented in Table 6- 16.
6-38
-------
1 Table 6- 15 Visitor Spending and Local Area Economic Impact of Rocky Mountain
2 National Park
Public Use Data
2010
recreation
Visits
1,320,156
2010
Overnight
Stays
438,677
Visitor Spending 2010a
All
Visitors
97,012
Non-Local
Visitors
89,408
Impacts on Non-Local Visitor
Spending
Jobs
1,283
Labor
Income3
37,299
Economic
Impact3
60,504
3
4
5
6
($000's) Source: Economic Benefits to Local Communities from National Park Visitation and Payroll (NPS,
2011) available at: http://www.nature.nps.gov/socialscience/docs/NPSSvstemEstimates2009.pdf
Table 6- 16 Median Travel Cost for Sequoia/Kings Canyon National Parks Visitors
Expense
Gas and Transportation
Lodging
Food and Drinks
Clothes, gifts, and
souvenirs
Total per visitor party
Median $ Amounts Spent (in $2010)
75
150
98
63
386
7
8
9
10
11
12
13
14
15
6.3.3 Urban Case Study
Urban forests are subject to the adverse effects of Os exposure in the same ways as
forests in rural areas. These urban forests provide a range of ecosystem services such as carbon
sequestration, pollution removal, building energy savings, and reduced stormwater runoff. The
analyses described in this section focus on carbon sequestration and air pollution removal using
the iTree model. The iTree model is a peer-reviewed suite of software tools provided by USDA
Forest Service. See Appendix X for details of the model and the methodology employed for
these case studies.
6-39
-------
1 [We will provide results in the form changes in tons of carbon sequestered and tons of
2 pollution removed due to 63 effects on tree growth in supplemental materials.]
3 6.4 DISCUSSION
4 O3 damage to vegetation and ecosystems causes widespread impacts on an array of
5 ecosystem services. Biomass loss impacts numerous services including supporting and
6 regulating services such as net primary productivity, community composition, habitat, and
7 climate regulation. Provisioning services are also affected by biomass loss including timber
8 production, agriculture, and non-timber forest products. Cultural services such as non-use
9 values, aesthetic services, and recreation are all affected by the damage to scenic beauty caused
10 by foliar injury due to 63 exposure. It is possible for several aspects of 63 effects to interact to
11 contribute to an impact on ecosystem services. For example biomass loss directly impacts timber
12 provision but other contributing effects include increased susceptibility to drought and insect
13 attack.
14 Many of these services are very difficult to quantify and even more difficult to assign a
15 quantified impact of Os exposure. For instance we were not able to quantify changes to
16 community composition due to Os or even identify the current level of service provided. Some
17 services, such as recreation, lend themselves to evaluation of total participation and measures of
18 total value but assessing the impact of 63 effects on these services is not possible at this time. A
19 very few services, such as timber provision, are amenable to quantification and monetization of
20 the actual incremental effects of 63 exposure.
21 For the supporting services identified as potentially affected by O3 exposure we were not
22 able to quantify the impacts for community composition. [However, for net primary productivity
23 we may have quantified results from PnET model runs for the second draft of this document.]
24 The regulating services indentified as potentially affected by Os exposure include
25 climate, water, pollination, and fire regulation. We will have quantified impacts of Os on carbon
26 sequestration in the form of results of model runs from FASOMGHG, national scale, and iTree
27 for the urban case studies for the 2nd draft. For the 2nd draft we are considering using the PnET
28 model to assess water cycle regulation effects. Pollination and fire effects remain unqualified
29 however we do have measures of total values of these services.
6-40
-------
1 The effects of 63 on the provision of timber and agriculture will be quantified and
2 monetized using the FASOMGHG model for the 2nd draft with related impacts of increased
3 susceptibility to insect attack described but not quantified. Non-timber forest products markets
4 are described and total values are included however subsistence use can only be described.
5 Cultural services are described in terms of total value since there are not data and
6 methods available to quantify Os impacts on these services. For example, outdoor recreation
7 activity participation rates range from 10% of the population for backpacking to 60% for natural
8 scenery viewing. The millions of participants have WTP values as high as $152 billion per year
9 for these activities and just three of these (wildlife watching, camping, and hiking) generate over
10 $200 billion per year in expenditures and over $385 billion in total economic activity. For the
11 case study national parks we are able to quantify the amenities potentially affected by O3 impacts
12 on sensitive vegetation in the parks. In Great Smoky Nation Park, for example, about 50% of the
13 trail kilometers are in the middle and highest categories for sensitive vegetation cover. [We will
14 have expanded case study analysis for the 2nd draft.]
15 The urban case studies will provide quantified and monetized iTree model results for
16 carbon sequestration and pollution removal for urban forests in supplemental materials.
17 Although we are unable to quantify the 63 impacts on these services we do know that
18 these impacts exist and that the loss of service due to those impacts is captured in the current
19 values of the services. Those values would be higher by some unknown amount were Os
20 impacts eliminated. Given the very high values for many of the services even very small
21 incremental changes in Os effects could potentially lead to large gains in benefits to the public
22 and society.
23
24
6-41
-------
7 SYNTHESIS
This assessment has estimated exposures to Os and resulting risks to ecosystems for both
recent Os levels and Os levels after simulating just meeting the current secondary 03 standard of
0.075 ppm for the 4th highest 8-hour daily maximum, averaged over 3 years, which was set to be
identical to the current primary Os standard. The results from these assessments will form part
of the basis for considering the adequacy of the current secondary Os standard in the first draft
Policy Assessment.
The remaining sections of this chapter provide key observations regarding the biomass
loss risk assessment (Section 7.1), foliar injury risk assessment (Section 7.2), ecosystem services
risk assessment (Section 7.3), and a set of integrated findings providing insights drawn from
evaluation of the full assessment (Section 7.4).
7.1 SUMMARY OF KEY RESULTS OF BIOMASS LOSS RISK ASSESSMENT
The first draft biomass loss risk assessment included two spatial scales of analysis
including a national scale analysis and several case studies focused on national parks containing
Os sensitive vegetation. The biomass loss risk assessment focused on relative biomass loss for
11 tree species for which concentration-response (C-R) functions are available. Relative
biomass loss is measured as the proportion of biomass lost relative to biomass if ozone
concentrations were zero. The assessment of individual tree species gives an estimate of the
potential relative biomass loss, calculated across the established species ranges. A second
analysis incorporated the abundance of those tree species in different ecosystems to assess the
overall ecosystem level effects of the relative biomass loss. In addition, the biomass loss risk
assessment evaluated risks occurring in several important subareas, including federally
designated Class I areas, and federally designated critical habit areas for threatened and
endangered species. The analysis provides estimates of the percent biomass loss associated with
recent (2006-2008) Os concentrations, and the proportion of the Os-related biomass loss that
would remain after just meeting the current secondary Os standard.
Key results include:
• Relative biomass loss associated with recent Os concentrations varies
substantially between species and across the ranges for individual species,
reflecting differences in sensitivity to Os and differences in Os concentrations
across the ranges of the tree species.
• Across species, the estimated potential Os-related biomass loss associated with
recent Os concentrations ranged from 0.1 percent for Douglas fir to almost 100
percent for Eastern Cottonwood. The estimated median potential Os-related
7-1
-------
biomass loss for individual species ranged from 0 percent for Douglas fir to 56
percent for Eastern Cottonwood.
• The C-R function for some species (e.g. sugar maple) demonstrates a very rapid
change in biomass loss over a small range of Os concentrations, 30 to 35 ppm for
sugar maple, that behaves similar to a threshold.
• After simulating just meeting the current secondary 63 standard, the estimated
potential Cb-related biomass loss for individual tree species was on average 70
percent of the estimated potential biomass loss at recent 63 levels, with a range
between 8 and 89 percent.
• In eastern U.S. federal Class I areas, simulating just meeting the current Os
standard resulted, on average, in a 5 percent reduction of the estimated potential
Os-related abundance-weighted biomass loss relative to estimates at recent
ambient Os exposure levels. When areas with recent ambient Os levels lower than
a W126 of 10 ppm are excluded, this reduction was on average approximately 20
percent.
• In eastern U.S. federally designated critical habitat areas, simulating just meeting
the current 63 standard resulted on average in approximately a 10 percent
reduction of the estimated potential (Vrelated abundance-weighted biomass loss
relative to estimates at recent ambient Os exposure levels. When areas with recent
ambient 63 levels lower than a W126 of 10 ppm are excluded, this reduction was
approximately 25 percent.
• In the Great Smoky Mountains National Park case study area, simulating just
meeting the current 63 standard resulted in a 51 percent reduction of the estimated
potential (Vrelated abundance-weighted biomass loss relative to estimates at
recent ambient 03 exposure levels, with weighted biomass loss estimates reduced
from as high as 16.5 percent to a maximum of 7.9 percent.
7.2 SUMMARY OF KEY RESULTS OF FOLIAR INJURY RISK ASSESSMENT
The first draft foliar injury risk assessment included two spatial scales of analysis
including a national scale analysis and several case studies focused on national parks containing
Os sensitive vegetation. The foliar injury risk assessment focused on recent ambient Oj
exposure. Two general assessments of foliar damage are included in this first draft: 1) maps of
the abundance of tree species sensitive to foliar damage from Oj, exposure, and 2) foliar injury
risk index values for 37 national parks based on the frequency of exceedance of Os exposure
benchmarks using different O?, exposure metrics (i.e., SUM06, W126, and N100), soil moisture,
and the existence of Os-sensitive species within each park.
7-2
-------
Key results include:
• In the eastern U.S., where tree cover data were available, tree species that are
considered sensitive to (Vrelated visible foliar damage account for over 80% of
the tree cover in some areas as measured by the summed importance values
(measures of relative abundance of species).
• Of the 37 parks assessed, based on the screening level risk assessment method
used in Kohut (2007), the estimated risk of foliar injury was high in 2 parks (5%),
moderate in 4 parks (11%), and low in 31 parks (84%).
• In the Great Smoky Mountains National Park case study area, there are large areas
with high cover of Cb-sensitive species based on assessment using the National
Park Service sensitive species list and vegetation mapping from the United States
Geological Survey.
7.3 SUMMARY OF KEY RESULTS FOR ECOSYSTEM SERVICES RISK
ASSESSMENT
There are a wide range of ecosystem services associated with the ecosystem effects
(biomass loss and visible foliar injury) that are causally related to Os exposure. These include
supporting, regulating, provisioning, and cultural services. The first draft risk assessment
includes both qualitative and quantitative assessments of ecosystem services. The majority of
ecosystem services impacted by Os exposures are not quantifiable using existing tools and data.
As a result, the risk assessment focuses on providing contextual information about these services
in terms of overall magnitude of the service relative to public use and where possible, economic
value of the service. We emphasize that for these ecosystem services this contextual information
does not provide estimates of the incremental ecosystem damages associated with recent Os
exposures, nor can it provide estimates of the reduction in Os-related damages that would occur
from just meeting the current Oj standards. The magnitude of ecosystem services is provided
solely to provide context for discussions of the adversity to public welfare posed by damages to
these ecosystem services from ozone exposures.
For a few ecosystem services, including commercial forestry yields, carbon sequestration,
agriculture yields, reduced productivity in terrestrial ecosystems, and alteration of terrestrial
ecosystem water cycling, models exist that can be used to estimate risks from Oj exposure. This
first draft REA includes estimates of risks associated with 1) exposure of commercial forests to
Os, including estimates of changes in yields and resulting changes in welfare for producers and
consumers of forest products and changes in carbon sequestration, using the FASOM model, 2)
changes in carbon sequestration in urban forests and changes in urban forest pollution removal,
using the iTree model. [The iTree and FASOM results will be provided in supplemental
-------
materials to be submitted for public review when the first draft Policy Assessment is released in
August 2012].
Key results include:
• While the economic costs of the Os-related impacts on ecosystem services is
largely unquantifiable, the overall economic value of the set of ecosystem services
is estimated to be large, and therefore damages from 63 have the potential to be
significant.
• Ozone-related impacts on ecosystem services associated with commercial timber
production include lost economic value due to yield losses and reductions in
carbon sequestration. The estimated value of the yield reductions associated with
recent Os levels for the 11 species for which we have concentration-response
functions are $XXX. The estimated reduction in carbon sequestration is XXX
tons of carbon per year in 2006 to 2008. [These will be provided in supplemental
materials to be released in August 2012]
• Ozone-related impacts on ecosystem services associated with urban forests
include reductions in carbon sequestration and reductions in removals of air
pollution by urban trees. For the 11 species for which we were able to model Os
damages, the estimated reduction in carbon sequestration is XXX to YYY tons of
carbon per year across the urban case study areas in 2006 to 2008. The estimated
reductions in tons of pollutants removed is XXX to YYY tons across the urban
case study areas in 2007. [These will be provided in supplemental materials to be
released in August 2012]
• Simulating just meeting the current Os standard is estimated to increase the value
of commercial forest yields by $XXX in 2006 to 2008, and to increase carbon
sequestered in commercial forests by XXX tons of carbon in 2006 to 2008. [These
will be provided in supplemental materials to be released in August 2012]
• Simulating just meeting the current Os standard is estimated to increase carbon
sequestration by urban forests in the case study areas by XXX to YYY tons of
carbon. Removal of air pollution is estimated to increase by XXX to YYY tons
across the urban case study areas. [These will be provided in supplemental
materials to be released in August 2012]
7-4
-------
7.4 OBSERVATIONS
[These observations have been prepared based on the risk estimates available for the
July public release of the first draft REA. We anticipate providing an updated draft of Chapter 9
which integrates the results of the FASOM and iTREE ecosystem service risk analyses when we
provide supplemental REA materials along with the submissions of the first draft Policy
Assessment for public review in August]
Looking across the biomass loss, foliar injury, and ecosystem service risk analyses, there
are a number of observations that can provide insight into the nature and patterns of risk. The
results suggest that due to the importance of Os sensitive species of trees in Eastern forest
ecosystems, the potential relative biomass loss associated with recent Os concentrations is high,
with median values for the most sensitive species, eastern cottonwood, as high as 56%. The
damages to forest ecosystems due to reductions in biomass loss for sensitive species include
commercial losses, but may also include losses to recreational users and to subsistence
populations. Because many of these trees are abundant near urban areas with elevated Os levels,
simulating just meeting the current Os standard results in reductions in potential biomass loss of
30% on average.
National parks and wilderness areas that have been designated as Federal Class I areas
represent important geographic endpoints (e.g. Class I and critical habitat areas) where Os
damages may be important to consider. For the Great Smokey Mountain National Park case
study area, there are areas within the park where the sensitive species cover is very high. This
park has a large number of hiking trails with heavy public use. Of the approximately 1287
kilometers of trials in the park, including more than 114 km of the Appalachian Trail, over 1040
km or about 81% of trail kilometers are in areas where species sensitive to foliar injury occur.
50 percent of the trail kilometers are in the highest class of sensitive species cover.
There are several important factors to consider when evaluating risks to ecosystems
associated with recent exposures to Os. First, there is significant variability in the sensitivity of
tree species to Os exposures. Some species, such as Douglas fir, show little response at lower
concentrations, but can have substantial response at higher Os exposure levels (W126 > 50 to 60
ppm for Douglas fir). Other species, such as sugar maple, show a distinct threshold at lower
concentrations of Os, 30 to 35 ppm, but once the threshold is exceeded show rapid response over
a very narrow range of Os concentrations. These differences in response functions have a direct
impact on the change in biomass loss that is estimated to occur after simulating just meeting the
current primary Os standard.
Second, as a result of the differences in concentration-response relationships, individual
tree species show different patterns of change with respect to changes in 03. Douglas fir has a
very large proportional change when Os is meeting the current standard, however further
7-5
-------
reductions in O^ will likely have very little effect on that species. Sugar maple also had a large
proportional change when meting the current standard. Further reductions in 63 will have some
effect to a point beyond which we expect very little change. Other species are expected to exhibit
continued gradual change in RBL relative to ambient as O^ levels are reduced.
Third, many Class I and Critical Habitat areas occur in areas where the ambient 63 is
below the level of the current standard and these areas generally show very little change in
summed relative biomass loss when exposure is simulated to just meeting the current standard
compared to recent 63 levels. In areas with higher ambient 63 levels, the proportion of ambient
summed relative biomass loss decreases by as much as 20 percent.
Fourth, the biomas loss assessments of Class I, critical habitat and national park areas are
based on C-R functions for relatively few tree species. This makes it difficult to assess the
absolute values of biomass loss because the response to 03 levels of the remaining species in
those areas is not quantifiable at this time, so the absolute values would not represent the
biomass losses for the entire community. As a resultthe assessment necessarily focuses on
proportional changes in the summed-biomass loss estimates.
This first draft REA provides preliminary estimates of exposures and risks which provide
information that can be used to begin discussions in the Policy Assessment regarding the
adequacy of the current standard. The second draft REA will further refine the estimates of
exposure and risk by incorporating additional modeling of impacts on commercial forests and
urban forests using FASOM and iTREE. In addition, U.S. Forest Service Forest Health
Monitoring data on visible foliar injury will be included, allowing for additional insights into the
impacts of recent ozone levels on this potential measure of recreational ecosystem services
(associated with enjoyment during hiking activities). We are also evaluating the pNET model for
use in estimating risks due to changes in productivity in terrestrial ecosystems, reduced carbon
sequestration in terrestrial ecosystems, alteration of terrestrial ecosystem water cycling. The
second draft REA will also evaluate any alternative 63 standards identified in the first draft
Policy Assessment following evaluation of any advice and comments on those potential
alternative standards provided during the review by the CAS AC 03 Panel. Finally, we anticipate
that the second draft REA will incorporate an improved approach to adjusting 63 concentrations
based on simulations of just meeting the current and alternative Os standards.
7-6
-------
1 8 REFERENCES
2 Abt Associates, Inc. (2010). Environmental Benefits and Mapping Program (Version 4.0).
3 Bethesda, MD. Prepared for U.S. Environmental Protection Agency Office of Air Quality
4 Planning and Standards. Research Triangle Park, NC. Available on the Internet at
5 .
6 Abt Associates, Inc. (2010). Model Attainment Test Software (Version 2). Bethesda, MD.
7 Prepared for the U.S. Environmental Protection Agency Office of Air Quality Planning
8 and Standards. Research Triangle Park, NC. Available on the Internet at
9 http://www.epa.gov/scram001/modelingapps.mats.htm.
10 Abt Associates. (2007). Technical Report on Ozone Exposure, Risk, and Impacts Assessments
11 for Vegetation. Prepared for the U.S. Environmental Protection Agency Office of Air
12 Quality Planning and Standards. Research Triangle Park, NC.
13 Acharya, G., and L.L. Bennett. (2001). Valuing Open Space and Land-Use Patterns in Urban
14 Watersheds. Journal of Real Estate Finance and Economics 22(2/3):221-23 7.
15 Adams, D.; Alig, R.; McCarl, B.A.; Murray, B.C. (2005). FASOMGHG Conceptual Structure,
16 and Specification: Documentation. Available at
17 http://agecon2.tamu.edu/people/faculty/mccarl-bruce/FASOM.html. Accessed on
18 October 22, 2008.
19 Alexander, S.j.; Oswalt, S.N.; Emery, M.R. (2011). Nontimber forest products in the United
20 States: Montreal Process indicators as measures of current conditions and sustainability.
21 Gen. Tech. Rep. PNW-GTR-852. Portland, OR: U.S. Department of Agriculture, Forest
22 Service, pacific Northwest Research Station. 36 p. Available at
23 http://www.fs.fed.us/pnw/pubs/pnw_gtr851 .pdf
24 Alexander, S.J.; Weigand, J.; Blatner, K.A. (2002). Nontimber forest product commerce in
25 Nontimber forest products of the United States, Jones, E.T.; McLain, R.J.; Weigand, J.,
26 eds. University of Kansas Press.
27 Andersen, CP; Wilson, R; Plocher, M; Hogsett, WE. (1997). Carry-over effects of ozone on root
28 growth and carbohydrate concentrations of ponderosa pine seedlings. Tree Physiol 17:
29 805-811.
30 Arbaugh, M; Bytnerowicz, A; Grulke, N; Fenn, M; Poth, M; Temple, P; Miller, P. (2003).
31 Photochemical smog effects in mixed conifer forests along a natural gradient of ozone
32 and nitrogen deposition in the San Bernardino Mountains. Environ Int 29: 401-406.
33 Arbaugh, MJ; Miller, PR; Carroll, JJ; Takemoto, BL; Proctor, T. (1998). Relationships of ozone
34 exposure to pine injury in the Sierra Nevada and San Bernardino Mountains of
35 California, USA. Environ Pollut 101: 291-301.
3-1
-------
1 Baumflek, M.J.; Emery, M.R.; Ginger, C. (2010). Culturally and economically important
2 nontimber forest products of northern Maine. Gen.Tech. Rep. NRS-68. Newton Square,
3 PA: Y.s. Department of Agriculture, Forest Service, Northern Research Station. 74p.
4 Available at
5 http://nrs.fs.fed.us/sustaining_forests/conserve_enhance/special_products/maine_ntfp/
6 Benoit, LF; Skelly, JM; Moore, LD; Dochinger, LS. (1982). Radial growth reductions of Pinus
7 strobus L correlated with foliar ozone sensitivity as an indicator of ozone-induced losses
8 in eastern forests. Can J For Res 12: 673-678.
9 Black, VJ; Black, CR; Roberts, JA; Stewart, CA. (2000). Impact of ozone on the reproductive
10 development of plants. New Phytol 147: 421-447.
11 Brown, R.B.; Xu, X.; Toth, J.F. Jr. (1998). Lifestyle options and economic strategies:
12 susbsistence activities in the Mississippi Delta. Rural Sociology 63-4: p599-623.
13 Butler, BJ. (2008). Family forest owners of the United States. (2006). Gen. Tech. Rep. NRS-
14 GTR-27. Newton Square, PA: U.S. Department of Agriculture, Forest Service, Northern
15 Research Station. 72 p.
16 Buyhoff, G.J.; Wellman, J.D.; Daniel, T.C. (1982). Predicting scenic quality for mountain pine
17 beetle and western spruce budworm damaged forest vistas. Forest Science 28:827-838.
18 Byun, D.W. and J.K.S. Ching. (1999). Science algorithms of the EPA Models-3 Community
19 Multiscale Air Qualty (CMAQ) Modeling System, (EPA/600/R-99/030 0). U.S. EPA.
20 Byun, D.W. and K.L. Schere. (2006). Review of the governing equations, computational
21 algorithms, and other components of the Models-3 Community Multiscale Air Quality
22 (CMAQ) modeling system. Applied Mechanics Reviews, 59. 51-11.
23 CAL FIRE (California Department of Forestry and Fire Protection). (1996). California Fire Plan.
24 Available at http://cdfdata.fire.ca.gov/fire_er/fpp_planning_cafireplan.
25 CAL FIRE (California Department of Forestry and Fire Protection). (2008). CAL FIRE 2007
26 Wildland Fire Summary. Available at
27 www.fire.ca.gov/communications/downloads/fact_sheets/2007Summary.pdf
28 Chappelka, A; Skelly, J; Somers, G; Renfro, J; Hildebrand, E. (1999a). Mature black cherry used
29 as a bioindicator of ozone injury. Water Air Soil Pollut 116: 261-266.
30 Chappelka, A; Somers, G; Renfro, J. (1999b). Visible ozone injury on forest trees in Great
31 Smoky Mountains National Park, USA. Water Air Soil Pollut 116: 255-260.
32 Chappelka, AH. (2002). Reproductive development of blackberry (Rubus cuneifolius) as
33 influenced by ozone. New Phytol 155: 249-255.
8-2
-------
1 Coulson, R.N.; Klepzig, K.D. (2011). Southern Pine Beetle II. Gen. Tech. Rep. SRS-140.
2 Asheville, NC: U.S. Department of Agriculture Forest Service, Southern Research
3 Station, 512p. Available at http://www.srs.fs.usda.gov/pubs/gtr/gtr_srsl40/gtr_srs!40.pdf
4 Daniel, T.C.; Brown, T.C.; King, D.A.; Richards, M.T.; Stewart, W.P. (1989). Perceived scenic
5 beauty and contingent valuation of forest campgrounds. Forest Science 35:76-90.
6 Defenders of Wildlife. (2010). 2010 Annual Report, (accessed 2011). Available at
7 http ://www. defenders. org/resources/publications/annual_reports/2010_annual_report.pdf
8 Dickson, RE; Lewin, KF; Isebrands, JG; Coleman, MD; Heilman, WE; Riemenschneider, DE;
9 Sober, J; Host, GE; Zak, DR; Hendrey, GR; Pregitzer, KS; Karnosky, DF. (2000). Forest
10 Atmosphere Carbon Transfer and Storage (FACTS-II) the Aspen Free-Air CO2 and O3
11 Enrichment (FACE) project: An overview. (General Technical Report NC-214). St. Paul,
12 MN: U.S. Dept. of Agriculture, Forest Service.
13 Duff, M., Horst, R. L., Johnson, T. R. (1998). Quadratic Rollback: A Technique to Model
14 Ambient Concentrations Due to Undefined Emission Controls. Presented at the Air and
15 Waste Management Annual Meeting, San Diego, CA. June 14-18, 1998.
16 Emery, M.R.; Ginger, C.; Newman, S.; Giammusso, M.R.B. (2003). Special forest products in
17 context: gatherers and gathering in the eastern United States., USD A Forest Service,
18 Northeastern Research Station, Newton Square, PA. Available at
19 http://www.fs.fed.us/ne/newtown_square/publications/technical_reports/pdfs/2003/gtrne3
20 06.pdf
21 Emery, M.R.; Pierce, A.R. (2005). Interrupting the telos: locating subsistence in contemporary
22 U.S. forests. Environment and Planning, 37: p. 981-993.
23 Felzer, B; Kicklighter, D; Melillo, J; Wang, C; Xhuang, Q; Prinn, R. (2004). Effects of ozone on
24 net primary production and carbon sequestration in the conterminous United States using
25 a biogeochemistry model. Tellus B Chem Phys Meteorol 56: 230-248.
26 http://dx.doi.0rg/10.llll/j.1600-0889.2004.00097.x.
27 Gallai, N; Salles, JM; Settele, J.;Vaissiere, B.E. (2009). Economic Valuation of the vulnerability
28 of world agriculture confronted with pollinator decline. Environmental Economics
29 68:810-821.
30 Geoghegan, J., L.A. Wainger, and N.E. Bockstael. (1997). "Spatial Landscape Indices in a
31 Hedonic Framework: An Ecological Economics Analysis Using GIS." Ecological
32 Economics 23:251-264.
33 Grantz, DA; Gunn, S; Vu, HB. (2006). O3 impacts on plant development: A meta-analysis of
34 root/shoot allocation and growth. Plant Cell Environ 29: 1193-1209.
35 Gregg, JW; Jones, CG; Dawson, TE. (2003). Urbanization effects on tree growth in the vicinity
36 of New York City [Letter]. Nature 424: 183-187.
-------
1 Gregg, JW; Jones, CG; Dawson, TE. (2006). Physiological and developmental effects of O3 on
2 cottonwood growth in urban and rural sites. Ecol Appl 16: 2368-2381.
3 Grulke, N. E.; Minnich, R.A.; Paine, T.D.; Seybold, S.J.; Chavez, D.J.; Fenn, M.E.; Riggan, P.J.;
4 Dunn, A. (2009). Air pollution increases forest susceptibility to wildfires: a case study in
5 the San Bernardino Mountains in southern California in Developments in Environmental
6 Science, Volume 8. A. Bytnerowicz, M. Arbaugh, A. Riebau, and C. Anderson (eds.)
7 Gumpertz, ML; Rawlings, JO. (1992). Nonlinear regression with variance components:
8 Modeling effects of ozone on crop yield. Crop Sci 32: 219-224.
9 Hanson, P. J., Samuelson, L. J., Wullschleger, S. D., Tabberer, T. A., and Edwards, G. S. (1994).
10 Seasonal patterns of light-saturated photosynthesis and leaf conductance for mature and
11 seedling Quercus rubra L. foliage: differential sensitivity to ozone exposure. Tree
12 Physiology 14, 1351 - 1366
13 Heck, WW; Cure, WW; Rawlings, JO; Zaragoza, LJ; Heagle, AS; Heggestad, HE; Kohut, RJ;
14 Kress, LW; Temple, PJ. (1984). Assessing impacts of ozone on agricultural crops: II.
15 Crop yield functions and alternative exposure statistics. J Air Pollut Control Assoc 34:
16 810-817.
17 Henderson, R. (2006a) Letter from CAS AC Chairman Rogene Henderson to EPA Administrator
18 Stephen Johnson, February 16, 2006, EPA-CASAC-06-003. Available at
19 http://yosemite.epa.gov/sab/sabproduct.nsf/69FBB!E21FBlE4428525712D004BA05D/$
20 File/casac_con_06_003.pdf
21 Henderson, R. (2006b) Letter from CAS AC Chairman Rogene Henderson to EPA Administrator
22 Stephen Johnson, June 5, 2006, EPA-CASAC-06-007. Available at
23 http://yosemite.epa.gov/sab/sabproduct.nsf/0202D7053AC6E2AC852571870075ClD2/$
24 File/casac-06-007.pdf
25 Hildebrand, E; Skelly, JM; Fredericksen, TS. (1996). Foliar response of ozone-sensitive
26 hardwood tree species from 1991 to 1993 in the Shenandoah National Park, Virginia. Can
27 J For Res 26: 658-669.
28 Hogsett, WE; Weber, JE; Tingey, D; Herstrom, A; Lee, EH; Laurence, JA. (1997).
29 Environmental auditing: An approach for characterizing tropospheric ozone risk to
30 forests. J Environ Manage 21: 105-120.
31 Hufford, M. (1995). Holding up the mountains: talk as a historical discourse. Smithsonian
32 Folklife Center, Library of Congress, Washington D.C.
33 IPCC (Intergovernmental Panel on Climate Change). (2007). Summary for policymakers. In: The
34 Physical Science Basis. Contribution of Working Group I to the Fourth Assessment
35 Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom
36 and New York, NY, USA: Cambridge University Press.
37 http://www.ipcc.ch/publications_and_data/ar4/wg2/en/spm.html.
8-4
-------
1 Irwin, E.G. 2002. "The Effects of Open Space on Residential Property Values." Land Economics
2 78(4):465-480.
3 Jaegle, L., DJ. Jacob, W.H. Brune, and P.O. Wenberg, 2001: Chemistry of HOX radicals in the
4 upper troposphere. Atmos. Environ.,35, 469-489.
5 Johnson, T. (1997). Sensitivity of Exposure Estimates to Air Quality Adjustment Procedure.
6 Letter to Harvey Richmond, Office of Air Quality Planning and Standards, U.S.
7 Environmental Protection Agency, Research Triangle Park, North Carolina.
8 Johnson, T. (2002). A Guide to Selected Algorithms, Distributions, and Databases Used in
9 Exposure Models Developed by the Office of Air Quality Planning and Standards.
10 Prepared by TRJ Environmental, Inc. for U.S. Environmental Protection Agency, Office
11 of Research and Development, Research Triangle Park, NC.
12 Karnosky, DF; Gagnon, ZE; Dickson, RE; Coleman, MD; Lee, EH; Isebrands, JG. (1996).
13 Changes in growth, leaf abscission, biomass associated with seasonal tropospheric ozone
14 exposures of Populus tremuloides clones and seedlings. Can J For Res 26: 23-37.
15 Karnosky, DF; Pregitzer, KS; Zak, DR; Kubiske, ME; Hendrey, GR; Weinstein, D; Nosal, M;
16 Percy, KE. (2005). Scaling ozone responses of forest trees to the ecosystem level in a
17 changing climate. Plant Cell Environ 28: 965-981.
18 Kaval, P. and J. Loomis. (2003). Updated outdoor recreation use values with emphasis on
19 National Park recreation. Final Report, Cooperative Agreement 1200-99-009, Project
20 number JJVIDE-02-0070. Fort Collins, CO: Colorado State University, Department of
21 Agricultural and Resource Economics.
22 King, JS; Kubiske, ME; Pregitzer, KS; Hendrey, GR; McDonald, EP; Giardina, CP; Quinn, VS;
23 Karnosky, DF. (2005). Tropospheric O3 compromises net primary production in young
24 stands of trembling aspen, paper birch and sugar maple in response to elevated
25 atmospheric CO2. New Phytol 168: 623-635.
26 Kohut, R. (2007). Assessing the risk of foliar injury from ozone on vegetation in parks in the US
27 National Park Service's Vital Signs Network. Environ Pollut 149: 348-357.
28 Krupa, S.; Muntifering, R.B.; Chappelka, A.H. (2004). Effects of ozone on plant nutritive
29 quality characteristics for ruminant animals. Botanica 54, 129-140.
30 Lee, EH; Hogsett, WE. (1999). Role of concentrations and time of day in developing ozone
31 exposure indices for a secondary standard. J Air Waste Manag Assoc 49: 669-681.
32 Lee, EH; Hogsett, WE; Tingey, DT. (1994). Attainment and effects issues regarding alternative
33 secondary ozone air quality standards. J Environ Qual 23: 1129-1140.
34 Lee, EH; Tingey, DT; Hogsett, WE. (1987). Selection of the best exposure-response model using
35 various 7-hour ozone exposure statistics. Research Triangle Park, NC: U.S.
36 Environmental Protection Agency.
3-5
-------
1 Lee, EH; Tingey, DT; Hogsett, WE. (1988a). Evaluation of ozone-exposure indices for relating
2 exposure to plant production and for estimating agricultural losses. (EPA/600/3-88/039).
3 Washington, DC: U.S. Environmental Protection Agency.
4 Lee, EH; Tingey, DT; Hogsett, WE. (1988b). Evaluation of ozone exposure indices in exposure-
5 response modeling. Environ Pollut 53: 43-62.
6 Lee, EH; Tingey, DT; Hogsett, WE. (1989). Interrelation of experimental exposure and ambient
7 air quality data for comparison of ozone exposure indices and estimating agricultural
8 losses. (EPA/600/3-89/047). Corvallis, OR: U.S. Environmental Protection Agency.
9 Lesser, VM; Rawlings, JO; Spruill, SE; Somerville, MC. (1990). Ozone effects on agricultural
10 crops: Statistical methodologies and estimated dose-response relationships. Crop Sci 30:
11 148-155.
12 Little, E.L., Jr. (1971). Atlas of United States trees, volume 1, conifers and important hardwoods:
13 Misc. Pub. 1146. Washington, D.C.: U.S. Department of Agriculture. 9 p., 200 maps.
14 Madden, M., R. Welch, T. Jordan, P. Jackson, R. Seavey, and J. Seavey. (2004). Digital
15 Vegetation Maps for the Great Smoky Mountains National Park. Prepared for the United
16 States Geological Survey and National Park Service Inventory and Monitoring -
17 Vegetation Mapping Program.
18 Mansfield, C.A., S.K. Pattanayak, W. McDow, R. MacDonald, and P. Halpin. 2005. Shades of
19 Green: Measuring the Value of Urban Forests in the Housing Market. Journal of Forest
20 Economics 11(3): 177-199.
21 MEA (Millennium Ecosystem Assessment Board). (2005). Ecosystems and Human Well-being:
22 Current State and Trends, Volume 1. Edited by R. Hassan, R. Scholes, and N. Ash.
23 Washington: Island Press. Available at http://www.millenniumassessment.Org/8
24 documents/document. 766. aspx .pdf.
25 Morgan, PB; Bernacchi, CJ; Ort, DR; Long, SP. (2004). An in vivo analysis of the effect of
26 season-long open-air elevation of ozone to anticipated 2050 levels on photosynthesis in
27 soybean. J Plant Physiol 135: 2348-2357.
28 Morgan, PB; Mies, TA; Bollero, GA; Nelson, RL; Long, SP. (2006). Season-long elevation of
29 ozone concentration to projected 2050 levels under fully open-air conditions substantially
30 decreases the growth and production of soybean. New Phytol 170: 333-343.
31 Mueller, J., J. Loomis, and A. Gonzalez-Caban. (2007). Do Repeated Wildfires Change
32 Homebuyers' Demand for Homes in High-Risk Areas? A Hedonic Analysis of the Short
33 and Long-Term Effects of Repeated Wildfires on House Prices in Southern California.
34 Journal of Real Estate Finance and Economics, 1-18.
35 National Association of State Foresters (NASF). (2009). Quadrennial Fire Review 2009.
36 Washington, DC: NASF. Quadrennial Fire and Fuel Review Final Report 2009.National
37 Wildfire Coordinating Group Executive Board January 2009.
8-6
-------
1 National Climatic Data Center (NCDC). (2012a).U.S. Climatological Divisions. National
2 Oceanic and Atmospheric Administration. Available at http://www.ncdc.noaa.gov/temp-
3 and-precip/us-climate-divisions.php
4 National Climatic Data Center (NCDC). (2012b). Historical Palmer Drought Indices. National
5 Oceanic and Atmospheric Administration. Available at
6 http://wwwl .ncdc.noaa.gov/pub/data/cirs/
7 National Interagency Fire Center (MFC). (2012). Statistics. Available at
8 http://www.nifc. gov/firelnfo/firelnfo_statistics. html
9 National Park Service (NFS). (2003). Ozone Sensitive Plant Species on National Park Service
10 and U.S. Shish and Wildlife Service Lands: Results of a June 24-25, 2003 Workshop.
11 Baltimore, MD.
12 National Park Service (NPS). (2004). Assessing the Risk of Foliar Injury from Ozone on
13 Vegetation in Parks in the Appalachian Highlands Network. October. Available on the
14 Internet at http://www2.nature.nps.gov/air/Pubs/pdf/03Risk/aphnO3RiskOct04.pdf
15 National Park Service (NPS). (2006b). Ozone Sensitive Plant Species, by Park, November 2006.
16 Available on the Internet at
17 http://www.nature.nps.gov/air/Permits/ARIS/docs/Ozone_Sensitive_ByPark_3600.pdf
18 National Park Service (NPS). (2011). Economic benefits to Local Communities from National
19 Park Visitation and Payroll, 2010. Natural Resources Report NPS/NRSS/EGD/NRR-
20 2011/481. Available at
21 http://www.nature.nps.gov/socialscience/docs/NPSSystemEstimates2009.pdf
22 National Park Service (NPS). (2006a). Ozone bioindicators. Washington, DC.
23 http://www.nature.nps.gov/air/Pubs/bioindicators/index.cfm.
24 National Wildlife Federation. (2011). National Wildlife Federation Annual Report and Financial
25 Statements, (accessed 2011). Available at http://www.nwf.org/About/Annual-
26 Report/Annual-Report-Archive, aspx
27 Odum, H.T. (1996). Ecological Accounting. New York: Wiley and Sons. Outdoor Industry
28 Foundation. (2006). The Economic Contribution of Active Outdoor Recreation -
29 Technical Report on Methods and Findings. Available at
30 http://www.outdoorfoundation.org/pdf/ResearchRecreationEconomyTechnicalReport.pdf
31 Palmer, W.C. (1965). Meteorological drought. Research Paper No. 45. U.S. Weather Bureau.
32 NOAA Library and Information Services Division. Washington, D.C. Available at http://
33 ncdc.noaa.gov/temp-and-precip/drought/docs/palmer.pdf.
34 Pan, YD; Birdsey, R; Horn, J; McCullough, K. (2009). Separating effects of changes in
35 atmospheric composition, climate and land-use on carbon sequestration of US Mid-
36 Atlantic temperate forests. For Ecol Manage 259: 151-164.
37 http://dx.doi.0rg/10.1016/j.foreco.2009.09.049.
3-7
-------
1 Pena, D.G. (1999). Cultural landscapes and biodiversity: the ethnoecology of an Upper Rio
2 Grandewatershed commons in Ethnoecology: Situated Knowledge/Located Lives Ed.
3 V.D. Nazarea, University of Arizona Press, Tucson, AZ, pp 107-132.
4 Peterson, DL; Arbaugh, MJ; Wakefield, VA; Miller, PR. (1987). Evidence of growth reduction
5 in ozone-injured Jeffrey pine (Pinus Jeffrey! Grev and Balf) in Sequoia and Kings
6 Canyon National Parks. J Air Waste Manag Assoc 37: 906-912.
7 Prasad, A. M. and L. R. Iverson. (2003). Little's range and FIA importance value database for
8 135 eastern US tree species.
9 http://www.fs.fed.us/ne/delaware/4153/global/littlefia/index.html, Northeastern Research
10 Station, USD A Forest Service, Delaware, Ohio.
11 Pregitzer, KS; Burton, AJ; King, JS; Zak, DR. (2008). Soil respiration, root biomass, and root
12 turnover following long-term exposure of northern forests to elevated atmospheric Co-2
13 and tropospheric O-3. NewPhytol 180: 153-161.
14 Rawlings, JO; Cure, WW. (1985). The Weibull function as a dose-response model to describe
15 ozone effects on crop yields. Crop Sci 25: 807-814.
16 Rea et al, (2012). Using Ecosystem Services To Inform Decisions on U.S. Air Quality Standards.
17 Environ. Sci. Technol.46 (12), 6481-6488.
18 Ribe, R. G. (1989). The Aesthetics of Forestry: What has empirical preference research taught
19 us? Environmental Management 13: 55-74.
20 Ribe, R.G. (1994). Scenic Beauty along the ROS. Journal of Environmental Research 42:199-
21 221.
22 Richardson, L.; Loomis, J. (2009). The total economic value of threatened, endangered and rare
23 specis: and updated meta-analysis. Ecological Economics, 68:1535-1548.
24 Rizzo, M. (2005). A Comparison of Different Rollback Methodologies Applied to Ozone
25 Concentrations. November 7, 2005. Available at
26 http ://www. epa.gov/ttn/naaq s/standards/ozone/s_o3_cr_td.html
27 Rizzo, M. (2006). A Distributional Comparison between Different Rollback Methodologies
28 Applied to Ambient Ozone Concentrations. May 31, 2006. Available at
29 http ://www. epa.gov/ttn/naaq s/standards/ozone/s_o3_cr_td.html
30 SAB CVPESS (Science Advisory Board Committee on Valuing the Protection of Ecological
31 Systems and Services). (2009). Valuing the Protection of Ecological Systems and
32 Services. EPA-SAB-09-012. May. Available at
33 http://vosemite.epa.gov/sab/sabproduct.nsf/WebBOARD/SAB-09-
34 012/$File/S AB%20 Advi sorv%20Report%20full%20web .pdf.
35 Samet. JM. (2009). Consultation on Ambient Air Monitoring Issues Related to the Ozone
36 NAAQS. EPA-CASAC-09-005. March. Available at
-------
1 yosemite.epa.gov/sab/sabproduct.nsf/64B88B99C37A68CF852575710072D8CO/$File/E
2 PA-CASAC-09-005-unsigned.pdf.
3 Samet. JM. (2011). Consultation on EPA's Ozone National Ambient Air Quality Standards:
4 Scope and Methods Plan for Health Risk and Exposure Assessment (April 2011) and
5 Ozone National Ambient Air Quality Standards: Scope and Methods Plan for Welfare
6 Risk and Exposure Assessment (April 2011). EPA-CASAC-11-008. June. Available at
7 yosemite.epa.gov/sab/sabproduct.nsf/0594FCC1374FCC5D852578B60069CDB2/$File/E
8 PA-CASAC-ll-008-unsigned.pdf
9 Sanz, J.; Muntifering, R.B.; Bermejo, V.; Gimeno, B.S.; Elvira, S. (2005). Ozone and increased
10 nitrogen supply effects on the yield and nutritive quality of Trifolium subterraneum.
11 Atmospheric Environment 39: 5899-5907.
12 Schaub, M; Skelly, JM; Zhang, JW; Ferdinand, JA; Savage, JE; Stevenson, RE; Davis, DD;
13 Steiner, KC. (2005). Physiological and foliar symptom response in the crowns of Prunus
14 serotina, Fraxinus americana and Acer rubrum canopy trees to ambient ozone under
15 forest conditions. Environ Pollut 133: 553-567.
16 Sierra Club Foundation (2011). Annual Reports, (accessed 2011). Available at
17 http ://www. sierraclub. org/foundation/downloads/2010-annual-report.pdf
18 Sitch, S; Cox, PM; Collins, WJ; Huntingford, C. (2007). Indirect radiative forcing of climate
19 change through ozone effects on the land-carbon sink. Nature 448: 791-794.
20 http://dx.doi.org/10.1038/nature06059.
21 Smith, V.K., C. Poulos, and H. Kim. (2002). Treating Open Space as an Urban Amenity.
22 Resource and Energy Economics 24:107-129.
23 Somers, GL; Chappelka, AH; Rosseau, P; Renfro, JR. (1998). Empirical evidence of growth
24 decline related to visible ozone injury. For Ecol Manage 104: 129-137.
25 Southwick Associates. (2006). The Economic Contribution of Active Outdoor Recreation-
26 Technical Report For: Outdoor Industry Foundation. Available at
27 http://www.outdoorfoundation.org/pdf/ResearchRecreationEconomyTechnicalReport.pdf
28 Timin B, Wesson K, Thurman J. Application of Model and Ambient Data Fusion Techniques to
29 Predict Current and Future Year PM2.5 Concentrations in Unmonitored Areas. (2010). Pp.
30 175-179 in Steyn DG, Rao St (eds). Air Pollution Modeling and Its Application XX.
31 Netherlands: Springer.
32 Tyrvainen, L., and A. Miettinen. (2000). "Property Prices and Urban Forest Amenities." Journal
33 of Economics and Environmental Management 39:205-223.
34 U.S. Department of Agriculture. (2004). National Report on Sustainable Forests-2003
35 Documentation for the 2003 National Report: Indicators 35, 36, 37, 42, & 43 Available at
3 6 http ://warnell .forestry .uga. edu/nrrt/NSRE/MontreallndDoc.PDF
8-9
-------
1 U.S. Department of Agriculture. (2011). National Report on Sustainable Forests-2010 Available
2 at http://www.fs.fed.us/research/sustain/
3 U.S. Department of Agriculture. (2002). National Survey on Recreation and the Environment
4 (NSRE): 2000-2002. The Interagency National Survey Consortium, Coordinated by the
5 USDA Forest Service, Recreation, Wilderness, and Demographics Trends Research
6 Group, Athens, GA and the Human Dimensions Research Laboratory, University of
7 Tennessee, Knoxville, TN.
8 U.S. Department of the Interior, Fish and Wildlife Service and U.S. Department of Commerce,
9 U.S. Census Bureau. (2006). 2006 National Survey of Fishing, Hunting, and Wildlife-
10 Associated recreation. Available at http://www.census.gov/prod/www/abs/fishing.html
11 U.S. EPA (U.S. Environmental Protection Agency). (1978). Air quality criteria for ozone and
12 other photochemical oxidants (pp. 373). (EPA/600/8-78/004). Washington, DC.
13 U.S. EPA (U.S. Environmental Protection Agency). (1984). Air quality criteria for ozone and
14 other photochemical oxidants, Vol. 3 (pp. 405). (EPA/600/8-84/020A). Research Triangle
15 Park,NC.
16 U.S. EPA (U.S. Environmental Protection Agency). (1996). Review of national ambient air
17 quality standards for ozone: Assessment of scientific and technical information: OAQPS
18 staff paper. (EPA/452/R-96/007). Research Triangle Park, NC.
19 U.S. EPA (U.S. Environmental Protection Agency). (2006). Air quality criteria for ozone and
20 related photochemical oxidants (pp. 2118). (EPA/600/R-05/004AF). Research Triangle
21 Park,NC.
22 U.S. EPA (U.S. Environmental Protection Agency). (2007). Review of the national ambient air
23 quality standards for ozone: Policy assessment of scientific and technical information:
24 OAQPS staff paper. (EPA/452/R-07/007). July. Research Triangle Park, NC.
25 U.S. EPA (U.S. Environmental Protection Agency). (2009). The Risk and Exposure Assessment
26 for Review of the Secondary National Ambient Air Quality Standards for Oxides of
27 Nitrogen and Oxides of Sulfur. (EPA-452/R-09-008a).. September. Available at
28 http://www.epa.gov/ttn/naaqs/standards/no2so2sec/data/NOxSOxREASep2009MainCont
29 ent.pdf
30 U.S. EPA (U.S. Environmental Protection Agency). (201 la). Integrated Review Plan for the
31 Ozone National Ambient Air Quality Standards - Final. (EPA 452/R-l 1-006). April.
32 Available at http://www.epa.gov/ttn/naaqs/standards/ozone/data/201 l_04_OzoneIRP.pdf.
33 U.S. EPA (U.S. Environmental Protection Agency). (201 Ib). Ozone National Ambient Air
34 Quality Standards: Scope and Methods Plan for Health Risk and Exposure. (EPA-452/P-
35 11-001). Research Triangle Park, NC. April. Available at
36 http://www.epa.gov/ttn/naaqs/standards/ozone/data/20 ll_04_HealthREA.pdf.
8-10
-------
1 U.S. EPA (U.S. Environmental Protection Agency). (201 Ic). Ozone National Ambient Air
2 Quality Standards: Scope and Methods Plan for Welfare Risk and Exposure. (EPA 452/P-
3 11-002). April. Research Triangle Park, NC. Available at
4 http://www.epa.gov/ttn/naaqs/standards/ozone/data/20 ll_04_WelfareREA.pdf
5 U.S. EPA (U.S. Environmental Protection Agency). (201 Id). Integrated Science Assessment of
6 Ozone and Related Photochemical Oxidants (Second External Review Draft).
7 (EPA/600/R-10/076B). September. Research Triangle Park, NC. Available at
8 http://cfpub.epa.gov/ncea/isa/recordisplay.cfm?deid=226363.
9 U.S. EPA (U.S. Environmental Protection Agency). (2012). Integrated Science Assessment for
10 Ozone and Related Photochemical Oxidants (Third External Review Draft). (EPA/600/R-
11 10/076C). July. Research Triangle Park, NC. Available at
12 http://cfpub.epa.gov/ncea/isa/recordisplay.cfm?deid=242490.
13 U.S. Geological Survey (USGS). (2003). FEDLANP020 - Federal Lands and Indian
14 Reservations of the United States. Reston, VA. Available at
15 http://coastalmap.marine.usgs.gov/GISdata/basemaps/boundaries/fedlands/fedlanp020.zip
16
17 Vollenweider, P; Woodcock, H; Kelty, MJ; Hofer, R, -M. (2003). Reduction of stem growth and
18 site dependency of leaf injury in Massachusetts black cherries exhibiting ozone
19 symptoms. Environ Pollut 125: 467-480.
20 Wegman, L. 2012. Updates to information presented in the Scope and Methods Plans for the O3
21 NAAQS Health and Welfare Risk and Exposure Assessments. Memorandum from Lydia
22 Wegman, Division Director, Health and Environmental Impacts Division, Office of Air
23 Quality Planning and Standards, Office of Air and Radiation, US EPA to Holly
24 Stallworth, Designated Federal Officer, Clean Air Scientific Advisory Committee, US
25 EPA Science Advisory Board Staff Office. May 2, 2012.Wells, B., Wesson, K., Jenkins,
26 S. (2012). Analysis of Recent U.S. Ozone Air Quality Data to Support the O3 NAAQS
27 Review and Quadratic Rollback Simulations to Support the First Draft of the Risk and
28 Exposure Assessment. Available on the Internet at:
29 http://www.epa.gOv/ttn/naaqs/standards/ozone/s o3 2008 rea.html
30 Wells, B., Wesson, K., Jenkins, S. (2012). Analysis of Recent U.S. Ozone Air Quality Data to
31 Support the O3 NAAQS Review and Quadratic Rollback Simulations to Support the First
32 Draft of the Risk and Exposure Assessment. Available on the Internet at:
33 http://www.epa.gOv/ttn/naaqs/standards/ozone/s o3 2008 rea.html
34
8-11
-------
1 West, J. I, A. M. Fiore, V. Naik, L. W. Horowitz, M. D. Schwarzkopf, D. L. Mauzerall (2007).
2 Ozone air quality and radiative forcing consequences of changes in ozone precursor
3 emissions. Geophys Res Lett, 34, L06806, doi:10.1029/2006GL029173.
4 Western Forestry Leadership Coalition. (2009). Western bark beetle assessment: A framework
5 for Cooperative Forest Stewardship. 2009 Update. Available at
6 http://www.wflccenter.org/news_pdf/325_pdf
7 Wildlife Conservation Society. (2011). Financial Documents, (accessed 2011). Available at
8 http://www.wcs.org/about-us/fmancial-documents.aspx
9 Wittig, VE; Ainsworth, EA; Long, SP. (2007). To what extent do current and projected increases
10 in surface ozone affect photosynthesis and stomatal conductance of trees? A meta-
11 analytic review of the last 3 decades of experiments [Review]. Plant Cell Environ 30:
12 1150-1162.
13 Wittig, VE; Ainsworth, EA; Naidu, SL; Karnosky, DF; Long, SP. (2009). Quantifying the impact
14 of current and future tropospheric ozone on tree biomass, growth, physiology and
15 biochemistry: A quantitative meta-analysis. Global Change Biol 15: 396-424.
16 World Wildlife Fund (2011). 2010 Annual Report, (accessed 2011). Available at
17 http://www.worldwildlife.org/who/financialinfo/2010fundingandfmancialoverview.html
8-12
-------
United States Office of Air Quality Planning and Standards Publication No. EPA 452/P-12-004
Environmental Protection Air Quality Strategies and Standards Division July 2012
Agency Research Triangle Park, NC
------- |