United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-454/R-97-010 Y
September 1997
EPA
National Specialty Workshop
on Technical Tools for
Air Toxics Assessment
-------
f
'•&
EPA-454/R-97-010
National Specialty Workshop on Technical Tools
for Air Toxics Assessment
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Floor
Chicago. It. 60604-3590
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Emissions, Monitoring, and Analysis Division
Research Triangle Park, NC 27711
September 1997
-------
This report has been reviewed in accordance with the U.S. Environmental Protection Agency's peer and
administrative review policies and approved for presentation and publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
2?
-------
Executive Summary
The National Specialty Workshop on Technical Tools for Air Toxics Assessment was held on
June 17-20, 1997 in Research Triangle Park, North Carolina. The purpose of the workshop was to
provide a forum for Regional, State and local agency personnel to learn about air toxics technical
tools that are available to them, and make comments and suggestions on these tools. Over 80
participants attended the workshop, representing 33 State and local agencies, five Regional EPA
Offices, and a number of other staff from various EPA groups. Workshop sessions were focused
in six main areas: Elements of the Office of Air Quality Planning and Standards (OAQPS) Air
Toxics Program, Ambient Air Monitoring and Data Analysis, Emission Inventories, Dispersion
Modeling, Exposure Assessment, and Education and Outreach. Participants volunteered for panel
discussions held for each of the subject areas. Several weekly conference calls were held prior to
the workshop and panel members worked to identify issues for discussion at the workshop. The
lists of issues were compiled into an issues paper for each area.
At the workshop, half-day sessions were devoted to each of the six main areas. Each session
consisted of presentations, followed by a panel discussion. The presentations were intended to
provide information on the current status of many activities relating to air toxics. During panel
discussions, panel members presented their issue paper to the workshop attendees for discussion.
Issue papers were finalized at the end of the workshop.
In general, four major themes emerged from this workshop: the strategy of the Air Toxics
Program, personnel training, guidance on proper use of the tools and technical and budgetary
resources. A summary of each of these main areas is as follows:
1. An overall strategy for dealing with air toxics issues needs to be developed and technical tools
should be designed to fit this strategy.
• The overall air toxics strategy would include how the various activities within Section 112
relate to each other, such as MACTs, NESHAPs, Urban Air Toxics Program, etc. If these
activities could share the same data (e.g., emissions inventory and monitoring data), rather
than having to recreate these data for each application, duplication of efforts could be
reduced and the technical tools could be used more effectively.
• Air toxics programs should be designed to promote closer working relationships and
technical tools should promote the sharing of air toxics data (i.e., monitoring and inventory
data should be provided in a format that can be used in dispersion modeling, which, in turn,
should be able to be input easily in exposure models). There is a need for a coordinated
working relationship between exposure modelers, the risk assessors and the decision makers
at the Region and State levels in order to address the specific questions that need to be
answered.
in
-------
2. Training should continue to be recognized as a critical element of the Air Toxics Program.
• State and local agency personnel need to be advised of the availability of training materials
and courses. It would be hepful to have a continually updated clearinghouse of training
materials, accessible to agency personnel.
• Alternative forms of training, such as Internet courses, are useful for situations where travel
time or funding is limited.
• Training materials need to be user friendly. Training is a separate function from writing
guidance documents and policy memos. The training program need to identify audiences,
develop learning objectives and measurement methods, and determine suitable delivery
systems.
• Training should be built into the Air Toxics rule development process in the early stages
since training is essential to rule implementation.
• Air toxics is recognized as a priority program and additional training, i.e., MACT-specific
training, is planned in FY98.
3. Guidance on the availability, purpose, and use of air toxics technical tools is needed.
• On a new or existing website, provide an EPA-approved list of exposure models, how and
when to use them, EPA expert contacts, and suggested default inputs.
• Specific guidance on assessing HAPs is needed, such as guidance on emission inventories
for point, area, and mobile sources. The role of the Emission Inventory Improvement
Program (EEP) should be expanded to specifically address HAPs and guidance and tools
from State and local agencies should be obtained.
• A guidance document is needed to instruct agencies on how to effectively collect, analyze,
and present data.
4. As Federal, State, and local governments continue to downsize and are asked to do more with
less, innovative air toxics monitoring tools and procedures need to be developed.
• Opportunities, such as this workshop, could be provided for agency personnel to meet each
other and develop contacts for various technical issues.
• A contacts list for technical area experts within air toxics would be helpful as a quick
reference for technical questions and would promote interaction between the various
technical disciplines. An initial list of area experts based on workshop participants was
developed but a more comprehensive list is needed.
IV
-------
In addition to the general themes that were discussed at the workshop, specific issues within the
panel subject areas were identified. The major issues for the panel areas include the following:
• Monitoring and Data Analysis: More monitoring data are needed to define the community
health concerns caused by air toxic pollutants. In many instances, monitoring has been
performed, but the data from the States performing the monitoring are not centrally located
or available for use by others. In those cases, the data need to be compiled for future use.
For many pollutants, monitoring has not been previously or adequately performed and
additional monitoring is needed.
Update sampling and analytical methods need to be completed for monitoring pollutants
where methods have not been previously available. For example, the update of EPA's
Compendia of Methods for monitoring organic and inorganic pollutants in air is needed.
These compendia are necessary to promote uniformity in monitoring procedures from state
to state. These revised documents should be widely distributed for ease in locating
information.
As monitoring methods become available for pollutants not previously monitored, analysis
methods and tools must be developed to organize and analyze the data. Projects such as the
Ambient Air Quality Characterization Project help to address needs for new analysis tools.
• Emission Inventories: There is an overall need for more guidance on how to provide a
complete emissions inventory for mobile, area, and point sources. Upgraded tools/software
are needed, as well as recommended procedures for applying the tools for area, mobile, and
point sources. Inventories should be consolidated where possible to reduce duplicate data
collection and reporting. Overall guidance is needed on:
• Estimating toxic emissions and uncertainty levels from highway vehicles, nonroad
equipment, aircraft, commercial marine engines, and locomotives.
• How to do a point, area, on-road and nonroad mobile sources, and biogenic toxics
emissions inventory.
• An emissions inventory suitable for modeling HAPs is needed.
• Dispersion Modeling: For the purposes of assessing impacts of specific sources, the
existing technical tools are generally adequate, however, more guidance is needed on how to
use the tools. Tools are needed to address new issues that result from modeling air toxics.
For example, better algorithms are needed to characterize wet deposition, particle/gas
partitioning, gas scavenging coefficient and particle size distribution, etc. More technical
guidance is needed on using the models to help promote consistency in applications. For the
purposes of nationwide modeling assessments, existing tools need to be assessed or new
ones developed.
-------
• Exposure Assessment: EPA needs to develop better tools to address exposure assessment
needs. Guidance is needed for all personnel on the steps involved in performing exposure
assessments. Specific areas that the tools need to consider include:
• Assessing the exposure through the inhalation pathway versus all routes of exposure;
• Assessing exposure to humans versus including exposure to animals and plants; and,
• Assessing exposure in urban settings.
At the completion of the workshop, participants were asked to evaluate the workshop. The
evaluation showed that the workshop succeeded in bringing together many technical disciplines to
voice opinions on issues relating to areas involving air toxics assessment. Sixteen percent
reported that their primary job function is monitoring and data analysis, 32 percent dispersion
modeling, 20 percent emission inventory and permitting, and 32 percent classified themselves as
environmental engineers. The participants gained information on how their work affects others
and on the available technical tools that they may not have known about prior to the workshop.
Participants provided EPA with information regarding their current and future needs. They
indicated that one of the benefit of the workshop was learning who the experts are in various
technical areas to help promote interaction between technical disciplines. Participants wanted to
learn basic information about all of the disciplines so that they could see how their work fits into
the overall air toxics approach. Those speakers presenting new information (e.g., the dispersion
model AERMOD) were more helpful because they gave an idea of the current activities and what
is coming, relating to air toxics, in the near future. Participants indicated that they were eager to
learn more about performing exposure assessments and how EPA plans to approach exposure
assessment in the future. They felt that the workshop was helpful in informing personnel of the
availability of technical tools. For example, prior to the workshop, panel members developed lists
of tools that would be helpful in performing their work. During the workshop, they learned that
tools were already available to do many of the things they suggested.
Participants stated that they felt that the workshop was helpful in opening lines of communication
between States and EPA to help get agency personnel more involved and interested. They felt
that the workshop was organized and productive, and was successful in meeting its goals. The
participants liked the idea of workgroup discussion prior to the workshop. They suggested that
the workshop become an ongoing activity in order to promote interaction between technical
personnel.
V]
-------
CONTENTS
Executive Summary iii
1.0 Introduction 1
2.0 Workshop Proceedings 3
2.1 Introductions 3
William F. Hunt, Director, Emissions, Monitoring and Analysis Division
5
Bruce Jordan, Director, Emissions Standards Division (ESD) 7
2.2 Elements of the OAQPS Air Toxics Program 8
2.2.1 Presentations 8
MACT - Overview of Statute, Al Wehe, US EPA 8
Summary of the OAQPS Air Toxics Strategy (OATS), Al Wehe, US EPA . 9
The Air Toxics Program in OAQPS, Melissa McCullough, US EPA 9
2.2.2 Panel Discussion 10
Should Offsite Consequence Analysis of Accidental Releases Be
Incorporated into a Permitting Program, Tom Rogers, Florida DEP .... 10
Examples of Situations Where Risk is Communicated to the Public,
Denis Lohman, Region HI 11
Emission Inventory Tools for Air Toxics, Lee Page, Region IV 12
Combining Modeling and Inventory Tools, Mike Pokorny, Maryland DEP 12
Using Tools for Targeting Sources, Linda Lay, OECA 12
New Jersey's Risk Assessment Procedures, Joann Held, New Jersey DEP . 13
2.2.3 Group Discussion 14
2.3 Ambient Air Toxics Monitoring and Data Analysis 14
2.3.1 Presentations 14
Update of EPA's Compendia of Methods for Monitoring Organic and
Inorganic Pollutants in Air, Bill McClenny, US EPA 14
State and Local Participatory Programs, Dave-Paul Dayton, ERG 15
State and Local Participatory Program: Paniculate Monitoring
Participatory Program - Saturation Monitoring Repository,
Stan Sleva, TRC 15
Hurricane Fran Video from the FEMA Burn Site, Neil Berg, US EPA 16
Chemical Mass Balance for Toxic Species (VOC and Particulates),
CMB8 Software, Charles Lewis, US EPA 16
New Data Analysis Tools for Ambient Air Quality Data,
James Hemby, US EPA 17
2.3.2 Panel Discussion 17
2.4 Air Toxics Emission Inventories 19
Vll
-------
.4.1 Presentations 19
Introduction - National Air Toxics Emissions Inventory,
Anne Pope, US EPA 19
Emission Inventory Improvement Program (EIEP),
Steve Bromberg, US EPA 20
Emission Estimation: Techniques and Tools. Guidance and Assistance
from the Emission Factor and Inventory Group, Anne Pope, US EPA,
and Mary Ann Barckhoff, Lockheed Martin 22
2.4.2 Panel Discussion 25
2.5 Dispersion Modeling 27
2.5.1 Presentations 27
Air Toxics Modeling, Joe Tikvart, US EPA 27
Dispersion Modeling of Pollutant Impacts, John Irwin, US EPA 28
Meteorological Data and Electronic Data Transfer,
Dennis Atkinson, US EPA 29
AERMOD - Source-Specific, Near-Field Dispersion Model,
Russ Lee, US EPA 30
Source Attribution Modeling for Dioxin, Andy Roth, RAPCA 31
Practical Problems in Combustion Air Modeling, Dr. Steven Ehlers,
US EPA 32
New Jersey DEP Risk Assessment Procedures for Minor Sources,
Alan Dresser, New Jersey DEP 34
2.5.2 Panel Discussion 34
2.6 Exposure Assessment Techniques 35
2.6.1 Presentations 35
Exposure Modeling: Components, Status, and Uses,
Michael Zelenka, NERL 35
Microenvironmental Modeling Issues, Dr. Alan Huber, NERL 37
Comparing Cumulative Annual Dose Estimates from pNEM to
"Literature" Exposure Factors, Thomas McCurdy, NERL 37
Real-time Estimation of Polycyclic Aromatic Hydrocarbons,
Dr. Nancy Wilson, NERL 38
Minnesota Air Toxics Indexing System,
Dr. Greg Pratt, Minnesota PCA 39
2.6.2 Panel Discussion 40
2.7 Case Histories of Multi-pathway Modeling Efforts 40
The Mercury Study: An Integrated Risk Assessment, Martha Keating,
US EPA 40
The Screening Risk Assessment Evaluation for the Proposed Incineration
Project at the Anniston Army Depot Chemical Demilitarization Facility,
Leigh Bacon, Alabama DEM 42
2.8 Education and Outreach 42
2.8.1 Presentations 42
vm
-------
Training Opportunities in Air Toxics for You Through the Education and
Outreach Group, Howard Wright, US EPA 42
Unified Air Toxics Website, Dr. Nancy Pate, US EPA, ITPID 43
2.8.2 Panel Discussion 44
3.0 3.0 Issue Papers 45
3.1 Elements of the OAQPS Air Toxics Program, Panel One 45
3.2 Ambient Air Toxics Monitoring and Analysis Data, Panel Two 46
3.3 Air Toxics Emission Inventories, Panel Three 51
3.4 Dispersion Modeling, Panel Four 55
3.5 Exposure Assessment Techniques, Panel Five 57
3.6 Education and Outreach, Panel Six 6O
4.0 4.0 Conclusions 66
Appendix A Letter of Invitation A-l
Appendix B Workshop Agenda B-l
Appendix C Attendance List C-l
Appendix D Workshop Presentation Handouts D-l
Appendix E Contacts List for the Workshop Participants E-l
Appendix F Workshop Evaluation Summary F-l
Appendix G Contacts List for EPA Air Regulations and Enforcement and
Compliance Assurance G-l
IX
-------
1.0 Introduction
The National Specialty Workshop on Technical Tools for Air Toxics Assessment was held on
June 17-20, 1997 at EPA's Administration Auditorium, 79 Alexander Drive, Research Triangle
Park, North Carolina. Over 80 participants attended this workshop. Participants included
representatives from 33 State and local agencies, five EPA Regional Offices, and a number of
other EPA staff from various groups. The purpose of the workshop was to provide participants
with an opportunity to learn about a variety of technical tools that are needed for air toxics
assessment and provide an opportunity for Regional, State and local agency personnel to voice
their opinions and define their needs as they relate to these tools. This report describes the
preparation for the workshop, provides a summary of the presentations, and outlines the
recommendations received from the participants.
Plans for the workshop had been discussed for many months On March 25, 1997 a letter of
invitation was sent to Regional, State and local agency personnel defining the purpose of the
workshop. This letter is included in the report as Appendix A. The workshop mission statement
was defined as:
Regional and State contacts have expressed an interest in learning about the technical tools
required to support many sections of the Clean Air Act regarding the air toxics program so that:
1. They can better support requirements under Title III such as the newly published Risk
Management Plans under Section 112(r), residual risk and de-listing requests under MACT
rules, the Urban Area Source Study, the Great Waters Studies, etc.
2. They know who the air toxics topic area experts (office level identification) are across EPA
and the State and local agencies in order to obtain rapid assistance.
3. They can better communicate the magnitude of air toxics issues facing the technical analyst
and the success in solving air toxic problems.
Based on this invitation, very positive response was received from over 75 persons. Based on this
response, the workshop date was set to begin on June 17, 1997. The workshop was divided into
seven sessions. The first day devoted half-day sessions to Elements of the OAQPS Air Toxics
Program and Ambient Air Toxics Monitoring and Data Analysis. Half-day sessions on June 18,
1997 were devoted to Air Toxics Emission Inventories and Dispersion Modeling. On June 19,
1997 sessions were devoted to Exposure Assessment, Case Histories of Multi-Pathway Modeling,
and Education and Outreach. On the final day of the workshop, June 20, 1997 two hands-on
computer training sessions were provided on the use of dispersion models and emission estimation
tools. The final workshop agenda is included in this document as Appendix B. A list of all
workshop attendees is included as Appendix C.
-------
The first part of each session included speakers who provided information regarding the current
status of many activities relating to air toxics. Many speakers provided handouts of their
presentation slides either prior to or during the workshop. Copies of these presentations are
included in this report as Appendix D. The second part of each session was used for panel member
discussions. Panel members had been in contact prior to and during the workshop to define issues
in each subject area as they relate to air toxics assessment. The panel discussions were framed
around three basic questions:
1. When are the present technical tools adequate in helping you do your job?
2. When are they deficient?
3. List possible solutions that may alleviate some of the items on the deficiency list.
A series of conference calls were held for each panel where panel members answered the questions
and developed issue papers. The final issue papers were presented at the workshop during panel
discussion. Workshop participants commented on the issues and added suggestions. Panels met at
the end of the day following the workshop discussion and revised the issue papers according to the
discussion. The final versions of the papers are included in Section 3.0 of this report.
Over the course of the workshop, each participant was asked to indicate his or her area of
expertise in air toxics issues. The list was designed to meet the second workshop mission of
helping provide a quick reference so that personnel working in air toxics can stay in contact with
each other after the workshop. This list is included in this report as Appendix E. At the end of the
workshop, participants completed a Workshop Evaluation Form that gave them an opportunity to
voice their opinions on the workshop and make suggestions for improvement for future
workshops. A brief summary of the responses can be found in Section 4.0 of this report. The
evaluation form and a more complete listing of the responses is included in this report as
Appendix F.
-------
Overall, this workshop will provide an opportunity for Regional, State and local agency personnel
to inform EPA of their needs over the next few years. Major issues will be summarized into a
workshop report, and major recommendations will be presented to senior management. A copy of
all presentations will also be included in this workshop summary. Mr. Touma closed by discussing
workshop logistics with participants and introduced the first speaker, William F. Hunt.
William F. Hunt, Jr.,Director, Emissions, Monitoring and Analysis Division (EMAD)
Mr. Hunt, welcomed participants to the workshop and reiterated the importance of the role of
technical tools in the air toxics program. The purpose of Mr. Hunt's presentation was to provide
an overview of the air toxics infrastructure, discuss challenges ahead for the air toxics program,
and give a few examples of how these tools have been used in the past. Mr. Hunt began by
describing the current sense of direction of the Air Toxics Program and the current activities
involving infrastructure development. A significant amount of technical tools are available to the
Regional, State and local agency personnel for monitoring and data analysis, dispersion modeling,
and emission inventories. Although these tools were designed for criteria pollutants, they may be
applicable to air toxic pollutants also. Although there is no explicit mandate in the Clean Air Act
(CAA) for infrastructure development, technical tools are needed to carry on several mandated
programs including air toxics, which pose unique challenges. Since funding is limited, efforts to
develop or improve existing tools must be prioritized and efforts must build upon the programs
already in use.
Mr. Hunt provided information on some of the current activities within the EMAD in support of
technical tool development. These are:
• Emission Inventories and Factors Programs - National, State and local air toxics
inventories and emission factors are being developed to meet various Section 112
requirements. There are gaps in available air toxics data. Data from State and local
agencies are needed to fill these gaps. Nationwide, approximately 39 percent of emissions
are from mobile sources, 26 percent are from point sources, and 35 percent are from area
sources.
• Dispersion Modeling - Some of the tools available for dispersion modeling are still
applicable. There are, however, some new issues that need to addressed within this area.
These issues include dense gases, particle and gas deposition, and atmospheric chemical
transformations of toxic pollutants.
• Air Monitoring - There is a lack of technology related to monitoring all toxic pollutants.
Current efforts have been unable to develop quality assurance (QA) procedures. There
would be great expense in monitoring all pollutants in all parts of the country. A project is
underway that is considering requiring States to submit air toxics data.
-------
Mr. Hunt reviewed an example of how interpretation of monitoring data can make a difference in
the results. The example is from the Urban Air Toxics Program (UATP) and shows how
determining trends using monitoring data is difficult due to changes in monitoring site location. In
1995, sampling for 1,3-butadiene was performed at 16 sites across the nation. The Port Neches,
Texas, site results conflicted with emission inventory results. The emission inventory reported a
decrease in emission from 1990 to 1995, while the monitoring data showed an increase. A team of
specialists is needed to understand the complex issues involved. These complex issues included the
fact that different sampling sites were used for monitoring in 1990 and 1995 and the emissions
inventory had undergone changes.
Mr. Hunt continued with a discussion of air toxics indicators and trends. The goals of air toxics
indicators include synthesizing data collected from different programs and showing national trends
and patterns. Two challenges in meeting these goals include limited data availability on air toxics
and a lack of consensus on calculating relative risk. Some ongoing work in this area includes the
Government Performance and Results Act, developing a strategy to measure progress towards
reduced health risks resulting from programs (e.g., MACTs), and exploration of indicators. One
example of how indicator information is used includes analysis of some sampling data from PAMS.
In this example, between 1994 and 1995, 19 sites were sampled for various air toxics pollutants.
Between these times, the Reformulated Gasoline regulation came into effect. For these sampling
sites, all pollutants tested decreased except for ozone and formaldehyde. In a comparison of the
PAMS data and monitoring data from a California location, the data showed similar trends. A
comparison was performed in 1993 of chemicals in the New England states and the Midwest
states. This comparison also showed that during episodes of increased ozone, the formaldehyde
concentrations also increased. These examples show how the data can be used as indicators and
for trending.
Within the area of education and outreach, EMAD is currently working to do the following:
• Expand opportunities for education and outreach;
• Develop additional training on technical tools for monitoring, emission inventories,
modeling, and exposure assessment,
• Prepare training materials;
• Utilize multi-media training vehicles such as the Air Pollution Distance Learning Network
telecourses, internet, CD-ROM, and classroom lectures;
• Provide workshops that bring together multi-disciplines to foster communication and
information exchange; and,
• Establish a list of area experts to allow for quicker communications and solutions.
-------
Dialogue is needed with the State and local agencies as to what type of Education and Outreach
needs exist. Some training materials are available (through the internet as well as written
materials), but guidance is needed on which other types of training are needed.
Mr. Hunt listed current activities within ORD relating to exposure assessment as:
• Continuing to develop guidance and tools to evaluate multi-pathway human exposures for
multi-pollutants. Scientific capabilities must be expanded to measure multi-media
environmental and human exposures and biological markers of exposures; and,
• Characterizing residential multi-media exposures. Quantifying contributions from indoor
and outdoor sources including source-receptor relationships are needed, as well as
understanding relationships between exposure, dose and effect.
Some of the accomplishments include:
• Expanding the National Air Toxics Emission Inventory database;
• User friendly air dispersion models capable of addressing new concerns;
• Several years of PAMS monitoring data and easier ways to collect data;
• Developing useful air quality indicators;
• Developing a strategy to measure effectiveness of Air Toxics Program;
• Continuing research on exposure assessment methodologies; and,
• Expanding education and outreach opportunities.
The principle challenge ahead for the Air Toxics Program is the declining budget and the
increasing demand for infrastructure development. Effective partnerships with States, local
agencies and other stakeholders is needed to overcome these problems Improved methods are
also needed for technology transfer through the evolving medium of the internet making strides in
improving these methods Mr. Hunt then introduced Bruce Jordan.
Bruce Jordan, Director, Emissions Standards Division (ESD)
Mr. Jordan gave a brief overview of currents efforts in ESD. These include:
• Characterizing the air toxics problem;
• Defining the goal of the air toxics program;
• Tieing all pieces of the Air Toxics Program to a common goal (e.g., MACTs - how do they
fit in with other requirements); and,
• Measuring progress in reaching the common goal (i.e., what are the best indicators).
-------
Mr. Jordan stated that data is needed to direct the Air Toxics Program. ESD is currently looking
at how to use available funding to meet the four items listed above. One issue related to using the
available funding includes how different technical tools relate to each other. For example, in order
to prove a model is providing valid results, the results must be compared to actual monitoring data.
If monitoring programs can be brought in to validate the models, the results could be used to
prioritize activities and funding. He urged the workshop to try to find solutions that are cost
effective. Following those remarks, Mr. Hunt provided closing remarks and again welcomed the
participants to the workshop.
2.2 Elements of the OAQPS Air Toxics Program
2.2.1 Presentations
MACT- Oven'iew of Statute, Al Wehe, US EPA
Mr. Wehe provided an overview of the Clean Air Act of 1990 air toxics provisions including the
history of the 1990 air toxics provisions, the statutory provisions, the basic requirements, and the
maximum achievable control technology (MACT) program for reducing emissions of air toxics.
This overview was oriented toward reviewing Section 112 of the 1990 Clean Air Act (CAA) in
order to provide attendees background for further discussion at the workshop.
In 1990 as Congress reviewed the history of the air toxics program addressed in the Clean Air Act
under Section 112, it found that for two decades, efforts to lower these toxic emissions had been
stymied by argument, conflict and litigation concerning risk-based analysis and decisions leading to
few successful efforts to control exposures to these hazardous pollutants. Between 1970 and 1990
EPA set standards for only seven hazardous air pollutants.
In 1990 Congress replaced this risk-based approach with an approach predicated on the best
demonstrated technologies of all sources in a category. It was not because risk considerations
were not important that these changes were made (they are still integral elements of the 1990
amendments), but because a reliance on only risk had made rule promulgation nearly impossible.
Congress included a risk-based component which calls for assessment of risk remaining after
application of performance-based standards and remediation through a risk-based residual risk
program.
These standards are known as Maximum Achievable Control Technology (MACT) standards, and
Generally Available Control Technology (GACT) standards. In essence, EPA is required to
establish control requirements to assure that all major sources of air toxics achieve the level of
control already being achieved by the best performing sources in each category. Further, by
establishing performance levels rather than mandating particular control methods, Congress
ensured that results would prevail over red tape. Congress provided a 10-year schedule in which
to promulgate these MACT standards, with certain standards being promulgated in the first two
-------
years, 25 per centum in the first four years, an additional 25 per centum promulgated not later than
the seventh year and the remaining 50 per centum not later than the tenth year.
The workshop slides used in Mr. Wehe's presentation provide an outline of the provisions of
Section 112, a view of the MACT standards promulgated to date, and an outlook for the future of
the MACT program. Further the slides summarize an important tool in MACT standard
development, MACT partnerships, which emphasize stakeholder involvement, leveraging the
resources, experience, skills, and knowledge of all stakeholders in the program. A description of
the MACT Partnerships program is available in the Federal Register (60 FR 16088, dated
March 29, 1995).
The just-released Second Report to Congress on the Status of the Hazardous Air Pollutant
Program under the Clean Air Ad, EPA-453/R-96-015, August 1997 provides an up-to-date
summary of the Section 112 program. This report will be available through the National Technical
Information Service (NTIS); the executive summary and a fact sheet will be available on the OAR
policy and guidance website (http://www.epa.gov/ttn/oarpg/rules.html)
Summary of the OAQPS Air Toxics Strategy (OATS), Al Wehe, US EPA
Mr. Wehe continued the presentations with a summary of the OAQPS Air Toxics Strategy
(OATS). This strategy is needed because:
• We need a comprehensive vision of how to integrate the various air toxic requirements of
the CAA (i.e., fitting them with the criteria programs as best as possible); and,
• We need to improve regulatory and economic efficiency by matching solutions with
particular air toxics problems.
The Air Toxics Strategy should
• Use an air management model to organize the program and its activities;
• Reflect the fact that HAPs have many different kinds and degrees of effects, exposures, and
sources; and,
• Build on proactive stakeholder participation.
The impact of the OATS has not yet been defined since OATS is still in draft stage. Specific
program elements are being defined (i.e., defining goals of the strategy and how to measure
progress).
The Air Toxics Program in OAQPS, Melissa McCullough, US EPA
-------
Ms. McCullough began her presentation with an organizational chart of the OAQPS. This chart is
included in Ms. McCullough's presentation handouts in Appendix D of this report.
The four groups within the OAQPS and the purposes of each group are as follows:
• The Emissions Standards Division (ESD), whose purpose is to establish air toxics
standards;
• The Air Qualities Strategies and Standards Division (AQSSD), which assesses air toxics;
• The Emissions Monitoring and Analysis Division (EMAD), which measures ambient air, as
well as estimates source emissions; and,
• The Information Transfer and Program Integration Division (ITPID), which manages and
transfers air pollution and pollution control information and develops and delivers training
courses and educational materials.
Ms. McCullough reviewed the purposes of each group and provided examples of the kinds of
projects conducted within each group.
Question: Food and drug industries must get approval from the FDA before they release a
product. Why not have similar air pollution regulations that require industry to
obtain approval beforehand?
Answer: This issue is quite complex and controversial, and focuses on where the burden of
proof falls. Whose responsibility is it to prove that a process is safe or unsafe?
With the FDA, the burden of proof falls on the industry marketing a drug to have
specific exposures for specific effects, and for which they bear responsibility In air
toxics, the issue is much more complex scientifically and already the subject of
discussion, especially for environmental groups
2.2.2 Panel Discussion
The panel discussion began with introductions of panel members by Joann Held. Panel members
are identified in the Workshop Agenda in Appendix B of this report and on the Attendees List in
Appendix C. Panel members agreed prior to the workshop to each provide a short talk on
different aspects of the issues defined in the panel issue paper. Tom Rogers was the first presenter.
Summaries of the panel presentations follow.
Should Off site Consequence Analysis of Accidental Releases Be Incorporated into a Permitting
Program, Tom Rogers, Florida DEP
Mr. Rogers began his presentation by describing how an offsite consequence analysis (OCA) could
be incorporated into a permitting program. There are a number of ways in which such an analysis
10
-------
could be used. These range from using toxics' concentration levels as an absolute determining
factor (i.e., greater than or less than a health-based criteria level) in deciding the permitability of a
source (like is done for the criteria pollutants) to using it solely as public information with no effect
on the issuance of the permit. Alternatively, the analysis could be considered as a weighted
element, in some manner, of a decision to issue a permit. For example, the results of the analysis
could be used to force control or work practice requirements or used to develop emergency
management planning without being used as a reason to deny permit issuance.
To help determine how an offsite consequence analysis should be used, if at all, Mr. Rogers
explained that the more fundamental question of "what are you trying to accomplish" should be
addressed. Are you trying to provide complete protection of the general public from any kind of
accident; are you trying to reduce risk through better facility design and work practices; are you
trying to reduce risk through facility location and waste minimization; or do you simply want better
contingency and emergency planning and public information?
Mr. Rogers also addressed the technical tools available for air toxics assessment. He stated that
the tools should be consistent with chemical and physical laws and principles. Parameterizations of
physical and chemical effects should be based on validated data sets and be peer reviewed, and the
air dispersion algorithms should be consistent with those used in the Guideline on Air Quality
Models. Data input to these models should be reasonably obtainable and output should
correspond to the form of the endpoint criteria being evaluated. The tools need improvement in
the areas of emissions characterization and quantification, chemical and physical transformations,
multiple compounds analysis, time dependent changes, and less-than-lifetime exposures.
Mr. Rogers continued by listing what he expected to gain from this workshop as follows:
• The knowledge of a suite of models that cover the range of sources and scenarios that
toxics modeling may be performed for;
• A better understanding of the tools (models) that are available and what they can and
cannot do;
• The knowledge of whether some models are considered "better" than others, based on
evaluation studies or physics, that essentially address the same sorts of sources, and
• Better knowledge of what is the best technique at this time for completing a risk
assessment analysis.
Question: Should OCA be used in permitting?
Answer: Mr. Rogers stated that, from his perspective:
11
-------
• Do not use an OCA as an independent means of deciding permit issuance or denial of a
permit;
• Use the OCA to force risk reduction measures; and
• Enhance existing contingency or emergency planning processes to address the risk of
accidental releases.
Examples of Situations Where Risk is Communicated to the Public, Denis Lohman, Region 111
The next panelist, Denis Lohman, provided examples of situations where risk is communicated to
the public. Examples included when models are used to determine where to put monitors for
landfill monitoring and Superfund Sites.
Emission Inventory Tools for Air Toxics, Lee Page, Region IV
The next panelist, Lee Page, continued by defining the steps involved in the MACT development
process. One step in this process is identifying sources emitting HAPs.
Tools that are available for identifying air toxics sources include the following.
• "Compiling air Toxic Emission Inventories," an EPA document that is available on the
EPA's Technical Transfer Network (TIN). The TTN is avilable on the web at
http: //www. epa.gov/ttn;
• Chemical spcific EPA documents (e.g., locating and estimating documents), available on
theTNN;
• "Source Identification Procedures for Source Subject to Regulation Under 112d of the
Clean Air Act as Amended in 1990," an EPA document that is not yet available on the
TTN. It is referred to as the "Source ID Cookbook." Copies should be available from
Regional Offices, however, if they are not, contact Lee Page at (404) 562-9131 to request
a copy.
Combining Modeling and Inventory Tools, Mike Pokorny, Maryland DEP
The next panelist, Mike Pokorny, provided a brief example of how modeling and inventory tools
can be combined. Maryland has developed a tool, named the Air Quality Integrated Management
System (AIMS), that integrates modeling, inventories, monitors, etc., to identify the top 30
pollutants with the highest cancer risk in an urban area from all sources (e.g., mobile, area, point,
etc.). The AIMS is coded by source category and gives the component of risk for source
12
-------
categories. The assessment is performed from the individual source risk components. The benefit
of this system is that it combines all of the groups together. AIMS also converts to VOC
assessment and attainment strategy.
Using Tools for Targeting Sources, Linda Lay, OECA
Linda Lay outlined the organization of the Office of Enforcement and Compliance Assurance
(OECA). There are two major Divisions within OECA, the Office of Compliance and the Office of
Regulatory Enforcement. The OECA activities on MACT standards include preparing guidance
documents, Compliance Assistance Centers and tools, support of rule development, applicability
determination support, ADI, enforcement tools, and ATIS support.
Ms. Lay provided a list of compliance assistance tools developed and under development through
OECA's Office of Compliance. This list is provided in Ms. Lay's presentation handouts in
Appendix D of this report. A brochure is also included which lists contacts for small business
compliance assistance centers. These centers help small and medium-sized businesses understand
and comply with federal environmental requirements. The Compliance Assistance Centers listed in
the brochure are:
• Printing: http://www.pneac.org
• Automotive: 1-888-GRN-LINK, or http://www.ccar-greenlink.org
• Agriculture: http:///es.inel.gov/oeca/ag/aghmpg.html
• Metal Finishing: 1-800-AT-NMFRC, http://www.nmfrc.org
• Access to all Centers: http://es.inel.gov/oeca/mfcac.html
Ms Lay also provided a table that list names and phone numbers of OECA personnel and their
various areas of expertise This listing is included in Appendix G of this report.
New Jersey's Risk Assessment Procedures, Joann Held, New Jersey DEP
Ms. Held gave a brief overview of the risk assessment procedures that are completed in New
Jersey as a "quick and dirty" way of assessing safety. Alan Dresser provided a detailed summary
of these procedures later in the workshop. Ms. Held explained that New Jersey's risk assessment
procedure is really a first cut at deciding if the emissions are safe. They perform a pre-model
assessment for 50 carcinogens. This smaller list makes the first cut assessment easier. The
assessment results in an order of magnitude number to determine risk. New Jersey gave this
procedure to industry and it helped to reduce permitting efforts since source emissions requested
on the application were reduced.
13
-------
2.2.3 Group Discussion
The group discussed the issues brought forth in the panel's issue paper. Items included in the
paper were refined, and the final version of this paper is included in Section 3.0 of this report. One
major discussion point was exposure assessments versus dispersion modeling. The distinction
between these two areas includes who is affected in the analysis. Exposure assessment takes
people into account in the analysis. Dispersion modeling transports pollutants across land but does
not really deal with the people it affects. Also, dispersion models are fairly good at predicting peak
concentrations, but not as good at predicting where those concentrations occur. This would be an
important issue for applying dispersion model results to exposure assessment.
2.3 Ambient Air Toxics Monitoring and Data Analysis
2.3.1 Presentations
Update of EPA 's Compendia of Methods for Monitoring Organic and Inorganic Pollutants in
Air, BillMcClenny, US EPA
Dr. McClenny reviewed the EPA's organizational structure and the guidance documents that have
been originated by different groups. The revision of the EPA Compendium of Methods for the
Determination of Toxic Organic Compounds in the Ambient Air to add three new methods and to
update several of the existing methods was then discussed in some detail. All three new methods
involve the monitoring of volatile organic compounds in the ambient air: TO-15 addresses the
monitoring of the 97 volatile organic compounds listed in the Title III of the Clean Air Act of 1990
by sample collection with canisters and analysis by GC/MS; TO-16 addresses the monitoring of
trace gases by FTIR-based open path monitoring; and, TO-17 addresses the use of solid adsorbents
and associated analytical equipment for monitoring of toxic VOCs. Both TO-15 and TO-17
present a set of performance criteria that establish the level of performance expected in terms of
criteria such as method detection limits, replicate precision, audit accuracy, and data completeness
The use of performance-based methods is consistent with the definition provided by the
Environmental Monitoring Management Council's (EMMC) of "a monitoring approach that
permits the use of appropriate analytical methods that meet preestablished demonstrated method
performance standards based on Agency data quality needs." The EPA is moving towards
performance-based methods because:
• Procedures for approving modifications to methods create high barriers to new, better,
and/or less costly technologies for analysis;
• Specific methods may not give the sensitivity (i.e., 10"6 risk levels) needed to meet project
Data Quality Objectives (DQOs); and,
• Laboratories are forced into more expensive technology than might otherwise be used.
14
-------
Dr. McClenny reviewed the EPA Compendium methods that have been or are being revised and
indicated that completion of the Compendium in a consolidated volume is in doubt because of
retirements and reorganization at the Center for Environmental Research Information (CERI),
which was the primary sponsoring agency. Project summaries for the compendia are available in
the presentation handouts in Appendix D of this report.
State and Local Participatory Programs, Dave-Paul Dayton, ERG
Mr. Dayton presented information regarding the State and local Participatory Programs on the
National Volatile Organic Monitoring Contract. The program supports monitoring programs
concerning three distinct components: urban air toxics measurements, speciated non-methane
organic compounds (NMOC) (and nonspeciated NMOC), and photochemical assessment
monitoring for volatile organics. Each program component provides the following:
• Sampling and analysis support,
• Sampling equipment (UATMP/SNMOC/Carbonyls);
• Site installation and training;
• Site support;
• Central laboratory analysis;
• Quarterly and annual reports; and,
• AIRS-AQMS data entry.
Mr. Dayton reviewed the methods included in the central laboratory analysis and the compounds
evaluated. Contacts for this program are:
• Neil Berg, EPA Delivery Order Manager
• Kathy Weant, EPA Project Officer
• Dave-Paul Dayton, ERG Program Manager
• Tim Hanley, ERG Site Coordinator
Details of costing for the programs are provided in Mr. Dayton's presentation handouts in
Appendix D of this report.
State and Local Participatory Program: Particulate Monitoring Participatory Program -
Saturation Monitoring Repository, Stan Sleva, TRC
Mr. Sleva continued the session by providing a brief overview of the State and local Participatory
Program for the Saturation Monitoring Repository. Mr. Sleva's presentation handouts (found in
Appendix D of this report) provide a more detailed explanation of the program.
15
-------
The EPA is encouraging State and local air pollution control agencies to conduct short-term,
multi-site ambient air pollutant monitoring studies using saturation monitoring. Saturation
monitors are noninterference or nonequivalent sampling methods for measuring paniculate matter,
carbon monoxide, and NOX. The samplers are small, portable, and relatively easy to set up and
operate. It is helpful to "saturate" a geographical area with the samplers to assess the air quality in
areas where high concentrations of pollutants are possible. These monitors help air pollution
control agencies in evaluating their ambient air monitoring networks for consistency with the 40
CFR Part 58 air quality surveillance regulations. The Saturation Monitor Repository (SMR)
includes a pool of saturation samplers and related equipment which agencies use when it becomes
necessary for them to upgrade or borrow saturation sampling equipment.
Mr. Sleva reviewed some past and ongoing studies and project costs. Cost tables are included in
Mr. Sleva's presentation handouts. To inquire about a request for services, contact:
Neil Berg (SMR) Project Manager
U.S. Environmental Protection Agency (MD-14)
Research Triangle Park, North Carolina 27711
(919)541-5520
Hurricane Fran Video from the FEMA Burn Site, Neil Berg, US EPA
Mr. Neil Berg provided a videotape describing the sampling procedures and equipment used
during the Hurricane Fran debris burn. A map is provided in the presentation handouts to show
placement of the sampling equipment. The burn site was used to dispose of wood and debris from
Hurricane Fran. Burning and chipping operations were being performed at the site. Testing was
being performed for PM10. Additional copies of this video can be obtained from Bill Stancil at
(919)541-2862.
Chemical Mass Balance for Toxic Species (I'OC and Participates), CMB8 Software,
Charles Lewis, US EPA
Charles Lewis presented and reviewed the modeling software CMB8. Chemical Mass Balance
(CMB) uses source profiles to separate the measured ambient mass concentrations of paniculate or
VOC into their quantitative source contributions. The U.S. EPA's CMB8 is an updated version of
the CMB software for this purpose that has been supported and distributed by the EPA for several
years.
The U.S. EPA has tacitly approved CMB as a regulatory planning tool through the Agency's
approval of numerous State Implementation Plans which have had a CMB component.
Revisions to CMB8 should significantly increase the convenience, efficiency and accuracy of
performing receptor modeling by the CMB method. The CMB8 Features:
16
-------
Windows Environment - 16-bit and 32-bit versions are available for Windows 3.x and
Windows 95, respectively.
Species and Source Selection Defaults - the user can specify any one of up to ten sets of
fitting species that can be chosen at random by mouse-clicking on a software button.
Input file options - the direct input of ambient and source data files may be done in a variety
of formats: blank-delimited ASCII (*.txt), blank-delimited ASCII with columns and rows
interchanged (*.car), comma-separated value (*.CSV), xBase (*.dbf), and Lotusl23
spreadsheets (*.wks and *.wkl).
Output file options - the output result files can be renamed (both prefix and suffix) and
placed in different directories during a session, allowing for a better grouping and
identification of results generated during a lengthy session.
Collinearity Treatment - users can select collinearity parameter values directly in CMB8 to
use in detecting and handling collinearity problems.
Britt-Leuke Algorithm - the Britt-Leuke algorithm without approximations is used in
CMB8 to take account of uncertainty in all variables (e.g., the source compositions as well as
ambient concentrations)
CMB8 can be obtained via anonymous ftp for the 16-bit Windows 3.x version at
eaf.sage.dri.edu:/cmb80/rnodel/16bit/win3mmdd exe, or for the 32-bit Windows 95 version at
eaf.sage.dri edu:/cmb80/model/32bit/wn32mmdd exe, where mmdd is the month (mm) and day
(dd) of the version label. The file win3mmdd.exe is a PKZIP self-extracting file that includes the
program executable, sample input files, and a tutorial/user's manual.
New Data Analysis Tools for Ambient Air Quality Data, James Hemby, US EPA
Mr. James Hemby reviewed and demonstrated two data analysis tools for ambient air quality data,
PAMSDAS and VOCDAT. Examples and screen captures are available in Mr. Hemby's
presentation handouts in Appendix D of this report.
VOCDAT is a data analysis tool that performs the following:
• Provides a general platform to display VOC data and to perform QC tasks;
• Import/export AIRS format files,
• Prepares summary of species groups; and,
• Offers options to apply reactivity factors or create weight percent NMHC.
17
-------
VOCDAT can be obtained via anonymous ftp at ftp.crl.com/users/so/csonomati/outgoing/vocdat.
The software files are vocdl 16a.zip, vocdl 16b.zip, and vocdl 17.zip. Demonstration data files
are under testdata.zip. A password is required to unzip these program files. Requests for a
software password and technical questions should be directed to Sonoma Technology, Inc. at
(707) 527-9372.
The data analysis tool PAMSDAS facilitates exploratory PAMS analysis. The capabilities of
PAMSDAS include creation of boxplots of VOC species and criteria pollutants, examination of
diurnal profiles, development of tables of summary statistics and correlation analyses. The
software is currently undergoing significant revision and upgrades. Once PAMSDAS is complete
(projected in October 1997), it will be posted on the web and distributed to Regions and States.
For additional information on PAMSDAT, contact Mark Schmidt, EPA, at (919) 541-2416 or
schmidt. mark@epamail. epa. gov.
Mr. Hemby reviewed the goals and products of the Ambient Air Quality Characterization Project.
The goals of this project are to identify, catalogue, and collect ambient air quality data on toxics
which are not stored in AIRS and to eventually use these data to assess the air toxics issue. The
two major aspects involved in meeting this goal include collecting and analyzing ambient data, and
developing and applying indicators. The products of this project include an annual data catalogue
and an annual trend/indicator assessment. All State and local agencies which have previously
collected ambient air quality toxics data or are currently involved in such measurements are invited
to participate in this project. For additional information or to participate in the project,contact
James Hemby, EPA at (919) 541-5459 or hemby.james@epamail.epa.gov.
2.3.2 Panel Discussion
Thomas Shoens, Panel Chair, and Peter Kahn reviewed the panel's issue paper with the workshop
group. The paper was revised according to group discussion and the final version of the paper is
included in this report in Section 3.0. Recommendations made by the panel include:
• Tools - use FTIR monitoring to look at "Hotspots," saturation monitoring, mobile tools for
continuous type monitoring, satellite imagery, "Cookbook" approach for VOC data, list of
acceptable ambient air level concentrations,
• FTIR may be more beneficial than other methods for doing sampling. Benefits in using
FTIR versus canisters or saturation sampling are that they are transportable and provide
real-time data.
18
-------
2.4 Air Toxics Emission Inventories
2.4.1 Presentations
Introduction - National Air Toxics Emissions Inventory, Anne Pope, US EPA
Ms. Pope gave an introduction of the requirements for air toxics inventories to support Section
112, the current status of air toxic inventories, and the needed improvements to air toxic
inventories for various Section 112 programs.
Ms. Pope informed the participants that there is no authority in the CAA that requires State
agencies to require collection of air toxics emission inventories.
She discussed some consistency problems with the emission inventories. She indicated that a
fundamental problem with the inventories is the definition of toxics. The definition of air toxics
varies from agency to agency, as well as the type of pollutants submitted. She also explained that
all State agencies do not submit the same data. Some States only submit data for MACT or Major
Sources, while other States submit data for MACT and Major Sources along with mobile and area
sources.
Ms. Pope then presented information on the following. National Toxic Inventory (NTI),
Compilation of NTI, Limitations of NTI, Planned Improvements and Data Outputs from NTI.
The information is summarized on Ms. Pope's presentation handouts in Appendix D of this report.
Ms. Pope also discussed the list of 40 potential HAPs developed by a team called the Risk
Identification Process (RIP). She then gave example data outputs for the 188 HAPs from the 1993
NTI, Version 2.1 All sources combined totaled 4 4 million tons per year (pulled from 1993 NTI,
Version 2.1 August 1997) for the 188 HAPs.
Currently, the inventory data allows you to perform analysis for all HAPs in the NTI and look at
source categories. For FY 1998, the EPA plans to add facility-specific data such as
latitude/longitude and stack heights.
Ms. Pope also indicated that there was a survey mailed out to State and local toxic coordinators
from STAPPA/ALAPCO that was due by June 4, 1997. At the time of the workshop, EPA had
not received surveys from 50 percent of the States. Ms. Pope stressed that the information from
the States is very important for data analysis. The survey was aimed at Residual Risk and
Emission Reduction, and without data from the States, the EPA will not have accurate data for
analysis.
19
-------
Emission Inventory Improvement Program (EHP), Steve Bromberg, US EPA
Mr. Bromberg discussed the emission inventory (El) procedures and tools. He indicated that the
inventories have came along way since 1990 when there was no unanimity. He stated that looking
through the issue papers brought back memories of where they were after the first inventory.
Since that time, several factors have helped to improve the emission inventory. One of these
factors was the development of the Emission Inventory Improvement Plan (EIIP). EIIP is a
cooperative effort between State/local agencies, industry and EPA to improve estimates of
emissions from various sources. The program was introduced in 1990. In 1993, it was approved
and funded by State/local agencies with 105 grant funds. Another factor affecting the emission
inventory is the agreement of common terminology (i.e., facility, point or area source). After the
first inventory, it was realized exactly what it takes to perform a criteria pollutant inventory.
Mr. Bromberg continued with his presentation of the EIIP process. The EIIP process results in:
• Higher quality inventory data (consistency);
• Empowerment of State/local agencies;
• Cost savings for State/local agencies and industry; and,
• Improved communications and understanding among all inventory groups.
EIIP avoids policy making and research. EIIP teams are concerned with improving data quality
because:
• Emission inventories are the foundation of many decisions;
• If mistakes are made early in the process, it will interject errors in downstream calculations;
• Redoing work to address problems is costly and embarrassing; and,
• Unrealistic regulations may result from these errors (also, cannot achieve an emission
reduction if emissions are underestimated)
The EIIP membership is a nationwide effort between large and small agencies for standardization
and data transfer. The EIIP products include a seven volume series which contains: Planning,
Point Sources, Area Sources, Mobile Sources, Biogenic Sources, Quality Assurance and Data
Handling and Reporting. Documents can be obtained from the EIIP web page at
http://www.epa.gov/oar/oaqps/eiip/.
A preferred method should result in the best estimate and can be accomplished by a typical agency
(usually most costly and timely with less uncertainty). An alternative method may be less
demanding to accomplish, but may provide a less dependable data set The preferred method is the
correct choice when it meets the data users needs. The choice of preferred or alternative method is
not a unilateral decision, those who know what the data are going to be used for should be
consulted. Mr. Bromberg stated that OAQPS will use EIIP guidance to compile inventories
20
-------
Question. Will the States be able to use EIIP in place of AP-42 or along with it?
Answer: EIIP is a cookbook of how to estimate emissions. Usually, there are no emission
factors in the EIIP. To estimate emissions, both the EIIP guidance and the AP-42
should be used. First, you should consult the EIIP documents for an answer on
the procedures for estimating an emission, and the AP-42 document should be
used to select the emission factor.
EIIP has a data transfer prototype pilot test designed to move data from the prototype pilot to an
EPA system. The goal is to have agencies transfer data between themselves and EPA.
The prototype takes State/local agencies data systems through the following steps:
State/local Data System=t> Application Interface ^EDI Translator (creates a standard data
setH>X12 Data Set (agency is done)=£>EPA picks up^EDI Translator=£>Creates Electronic
File^EPA Application Interface^EPA Data System.
If the prototype works well for EPA and State/local agencies, it will be expanded to industry to
transfer data and allow States an easy uniform way of sending data.
Question: Will EPA create an interface with AFS so that States will not have to re-invent an
in-house interface like most States previously had to do?
Answer: Yes, but AFS contains only point source data, which is about one third of the
HAP data. The EPA is currently looking at the inventory system as a whole from
a Federal standpoint. AFS is available, and is not going away anytime in the
immediate future.
Mr. Bromberg made the point that, a few years back, the EPA tried to consolidate emission
inventory reporting and data requirements. Reporting requirements are scattered throughout the
CAA. There were several issues that the EPA could not overcome a few years ago. Since then,
the EPA has worked through some of these issues and prepared a consolidated emission inventory
draft reporting rule.
Several State/local agencies are collecting toxics data, but State/local agency legislatures have
commented that EPA has no authority to require this data collection. Because of their legislatures,
several State/local agencies are having difficulty continuing to collect toxics data.
Mr. Bromberg passed out TABLE 1 Proposed Emission Inventory Rulemaking Provisions. This
table is included in Appendix D of this report. He stated that the rule will require that every three
years State/local agencies generate a toxic inventory report. The rule split the information
requirements into thirds. Every year, State/local agencies have to report Major or Significant
21
-------
sources. Every third year, State/local agencies have to add the remaining sources to the toxic
inventory report.
Question: Thresholds for actual or potential emissions?
Answer: Data to be reported under the proposed rule are actual emissions. Reporting
thresholds are for actual emissions.
Question: A lot of companies reporting toxic data feel the information is confidential. How
do we handle confidential information?
Answer: We do not want any confidential information. This means for toxic inventory
reporting you may only get facility level data, sources may request that process
level information not be released. Modelers need process level information to do
their job, but from an inventory standpoint, the facility level is enough.
Question: What about data uncertainty9
Answer: One of the comments in the issue paper alluded that it was not important
Mr. Bromberg feels it is an important issue that has not been addressed.
Knowledge of the amount of data uncertainty and having some indication of the
quality of data that you obtain is very important. Unfortunately, we have not
been given any money to determine quality of data. As more people become
aware of the data limitations, we will begin receiving better data.
Question: With the schedule for inventory every three years, are there any plans for a link
between meteorological and monitoring data with emissions data?
Answer: Mr Bromberg stated that he was not aware of any plans for this type of link.
Emission Estimation: Techniques and Tools. Guidance and Assistance from the Emission Factor
and Inventory Group, Anne Pope, US EPA and Mary Ann Barckhoff, Lockheed Martin.
Ms. Pope began by indicating that, from looking through the issue papers, there was obviously a
lack of knowledge of the current tools available through the EPA. All EPA documents, including
EIIP documents, guidance documents or Toxic Chemical Release Inventory System (TRIS) data,
are available through the Info CHIEF Help Desk. CHIEF stands for the Clearing House of
Information for Emission Factors. This number is available for obtaining documents, as well as
technical support. The hotline also has resources available for training by the support staff. If
training is needed, the help desk staff will go out to State and local agencies and provide onsite
training.
22
-------
There are several tools available for emission estimates. These tools include:
• Continuous Emissions Monitoring (CEM);
• Source Tests;
• Material Balances (for solvent applications); and,
• Emission Factors (AP-42).
Ms. Pope noted that speciation (using the database SPECIATE) is not a recommended EPA
method for compiling air toxics or HAP emission inventories. The SPECIATE database was never
intended to be used for emission estimation The database should not be used for emission
estimation because it was developed as a modeling tool for a very broad class of VOCs (it does not
contain a complete list). The second reason is the database only contains approximately 700
profiles, so all of the needed data is not available for emission estimation. Profiles are assigned to
other SCCs. For example, an original profile for a smelter may be assigned to all types of smelters.
There was no distinction between a smelter used for processing chromium, brass or bronze ore.
All smelters were assigned only one profile, which means the correct emissions estimates cannot be
obtained when using the SPECIATE database.
Ms. Pope introduced Ms. Mary Ann Barckhoff, who gave a brief overview of EPA's various tools.
Ms. Barckhoff described the various EPA Emission Estimation Tools that are currently available as
follows.
• FaxCHIEF (AP-42 is loaded on the fax machine)
Fax: (919) 541-5626 or 0548
• EFIG Web Site (gateway to other sources)
http://www.epa.gov/oar/oaqps/efig/
• EFIG List Server (receive CHIEF Newsletter via email) subscribe via email to:
List server@unixmail.rtpnc.cpa.gov, type "Subscribe CHIEF Firstname Lastname"
• AirCHIEF CD-ROM (contains EPA reports and databases)
Call InfoCHIEFfor more information (919) 541-5285
• National Emission Trends Viewer CD-ROM (includes El data from 1985-1995)
• TANKS Version 3.0 (based on the petroleum industry - calculational software to estimate
emissions from organic liquid storage tanks)
• FIRE Version 5.1b (Factor Information Retrieval System - EF from AP-42, old AIRS,
locating and estimating documents, updated through AP-42 Supplement A&B)
23
-------
• Info CHIEF Help Desk Telephone (919) 541-5285, e-mail address
infochief@epamaiL epa.gov
Ms. Pope explained that the AP-42 document is composed of two volumes. Volume I, fifth
edition, was completed in 1995. Volume I has been updated since that time with supplements A &
B. The update is available online. Volume II, fifth edition, will be available online in August. The
AP-42 Background Documents are also available online, through the CHIEF web page. The
AP-42 Background document explains the methodology behind the development of the emission
factors. Currently, the EPA is creating a link between AP-42 and EIIP guidance, which will make
it much easier to search for information. The link will be available in the Fall of 1997 along with
the new version of the AirCHIEF CD-ROM.
The predominant new feature of the AirCHIEF CD-ROM (Version 5.0) will be the Adobe Acrobat
Format. One of the advantages of the Adobe format is that reader software used to view programs
are available through www.adobe.com. The readers are free, but there is a charge for the
development software. Also, the new version contains a search engine that will allow you to
search the text of the document. The search engine will also allow you to search between chapters
in the AP-42 document that the flat format and other databases do not allow.
Due to the overwhelming demand for the CHIEF web page, the dial in version of CHIEF TTN will
no longer be available after September 1, 1997. If you have any concerns regarding this, please
contact Mr. Ron Myers at (919) 541-5407.
The EPA is currently working on developing a new version of FIRE. It will be a Windows™-
based system. The menu items will contain a description of the old codes used. The EPA is
looking for user input, so if you have any suggestions, please call the help desk. The help desk will
be able to provide the name and number of the person you need to send comments to.
Question: What is the hierarchy of emission factor tools? If I understand correctly, the EIIP
guidance was to be used before the AP-42.
Answer: The two are supplemental to each other. There are no point source emission
factors in the EIIP guidance, therefore you should use AP-42 for point sources.
For area sources, AP-42 contains inadequate emission factors, so you would
need to refer to the EIIP guidance. EIIP guidance should be used over the AP-42
for area sources. If the emission factors were adequate, they would not be in the
EIIP. AP-42 should be used over the FIRE database, depending on when your
version was updated. If you have the latest version of the AP-42, then use that.
For the approaching 1996 emission inventory, the most frequently asked
questions were what is the hierarchy and where do I go to find emission factors?
Which one do I use? We are going to establish a sequence to walk you through
this.
24
-------
Question: Will there be a link between FIRE and AP-42?
Answer: No. Due to all the different sources and records, it is to difficult to create a link
between the two. Hopefully, the EPA will be able to develop a link in the future.
Question: How will we know which source is more recent, in order to determine which one
precedes the other?
Answer: Usually, there is a revision date on the web page.
Question: Is it standard practice that you put updates in the Alert section of the web page?
Answer: No, The web page currently does not have an alert section, just a revision date
for the document.
Question: Where can the hierarchy sequence for the emission factors be located?
Answer: It is not available yet, but it will be by the end of the year with the 1996 emission
inventory. If you would like to keep checking to see when the sequence becomes
available, the e-mail address is wwv.epa.gov/oar/oaqps/efig/ei.
Question: What about draft documents on the web page? Could we use the draft AP-42 for
emission factor information?
Answer: The EPA does not recommend using drafts for emission factor information. The
drafts are online for comment purposes only, they have not been finalized. If you
use the draft documents, do so at your own risk.
Ms. Barckhoff ended with a slide for the Info CHIEF Help Desk. Questions or comments should
be directed to (919) 541-5285 or email at infochief@epamaiLepa.gov
2 4.2 Panel Discussion
Mike Fishburn, Pane) Chair, presented slides with each of the issue paper topics and explained the
panel groups reason for including each comment Mr. Fishburn's presentation handouts are
included in this report in Appendix D.
Mr. Fishburn added that there were a lot of comments regarding AIRS. Several panel members
felt that it was very important to input data into AIRS for the people who need it. The AIRS
area/mobile source module will be replaced with the NET database and arrangements will be made
to take care of the acceptance of the AIRS data into the NET database for facilities that have spent
a lot of time creating an interface with AIRS.
25
-------
Comment: There is a need for emission inventory consistency and methodology on a national
level, as well as for State and local agencies.
Question: Are you looking for a specific guidance document on how to compile an
inventory or a tool?
Answer: Yes and No. EIEP is currently adding toxic emission factors, so there is no need
for emission factor guidance, but there is a need for how to and where to go for
activity data. Most point sources do not have a specific person responsible for
emission inventories, therefore the need exists for guidance and/or a simplified
document for someone that does not deal with toxic inventories on a daily basis.
There is a need for a reading list on outside source information for State and local
agencies, as well as industry.
Comment: There is a need for continuous work on emission factors because they are
inadequate when the factors for a particular process or item is not available.
Response: A lot of work on emission factors is inadequate. For example, the Novas study
focused primarily on recreational marine vehicles. Outboard usage seems to be
declining, while jet ski usage appears to be increasing. Emissions may be
offsetting each other or jet ski emissions may be overtaking marine vehicles.
There are no emission factors available at this time for jet skis, so there is no way
of estimating.
Comment: Hopefully, there will be some guidance and prioritization of pollutants for
emission inventories. Some agencies felt it would be nice to have guidance from
EPA on which categories to go after and how much time should be spent on
emission inventories. If the guidance is only a short list of chemicals and the
categories we need to be concerned with, the prioritization is very important.
Some agencies are getting everything, including very small numbers, which is very
time consuming
Response: Until the EPA gets more data, we can not put the problem in perspective. There
needs to be a national effort from both EPA and State and local agencies for
standardization.
The final version of the issue paper is included in this report in Section 3.0.
26
-------
2.5 Dispersion Modeling
2.5.1 Presentations
Air Toxics Modeling, Joe Tikvart, US EPA
Mr. Tikvart began his presentation with an overview of dispersion modeling and air toxics. In the
past, the emphasis has been on criteria pollutants. There has been somewhat of an "identity crisis"
for air toxics. As this issue is being defined, there is a greater focus in modeling towards air toxics.
This promotes a hopeful future as there is emphasis in CAA Title III on risk and the "lessons
learned" from criteria pollutants can be used.
Mr. Tikvart listed ideas for discussion for the modeling session of the workshop. These ideas are
as follows:
• Areas that need to be addressed in the models for air toxics assessment include similarities
to criteria pollutants, dense gas problems (a family of models is needed to address dense
gases), chemistry (transformation of pollutants; Gaussian models do not deal well with
this), deposition (multi-pathways) problem, and general evaluations (need to know how
good a model is).
• Input data - emissions/air quality/meteorological. Obtaining quality data is always a
problem.
• Screening Models (Techniques)
• Models - EPA vs. Others - There is no specific program to dictate which model to use.
• Averaging Times
• Sources of Meteorological Data
• Chemical Transformation
• Variable Emission Rates (spikes in emissions)
Mr. Tikvart continued by discussing the current directions of Air Quality Modeling Group
(AQMG) Program. The directions listed include:
• National assessment to demonstrate risk reduction to include urban applications and
evaluation, and comparisons with national/regional assessments;
• Assessments for individual source categories;
27
-------
• Outreach (to include workshops, a model clearinghouse, and the SCRAM website), and,
• Assistance to States on Risk Assessments.
Dispersion Modeling of Pollutant Impacts, John Irwin, US EPA
Mr. Irwin began his presentation by listing the goal of his presentation as to provide a basic
understanding of the kinds of air dispersion models available and the purposes for which these
dispersion models are best used. Many of the EPA dispersion models are available on the SCRAM
BBS on the TTN or via internet at http://www.epa.gov/scram001/. Reviews of Eulerian and
Lagrangian air dispersion models can be found in Mr. Irwin's summary paper in the presentation
handouts in Appendix D.
Mr. Irwin continued by listing considerations for model selection. Questions to be considered in
selection include.
• Are we following the dispersing material, or do we watch it go by? (Lagrangian vs.
Eulerian, and puff/plume vs. grid);
• Do we allow more than one source? (single vs. multiple sources); and,
• Do we allow more than one source type? (point, area, line, volume).
Basically, the modeling tools are very simple. The usefulness of modeling results depends on how
clever you are in the use of these tools. Mr. Irwin provided examples of puff vs. plume models and
then reviewed some data to show comparisons of modeling results with observations. Observed
data and modeled data do not match exactly In the "real world" (observed) data, there are
irregularities that are not simulated by the models.
Assumptions and limitations in Gaussian plume modeling include:
• Plumes go in straight lines;
• Bell-shaped concentration curves exist in the vertical and horizontal planes;
• Wind speed dilutes plumes; and,
• There is enough wind blowing to move the plume downwind (hence, plume models are
inappropriate for calm wind conditions).
A factor of 2 between observed vs. estimated concentrations for Gaussian models is considered
reasonable, given all the stochastic effects not simulated by standard dispersion models The
general trend in modeling seems to slightly overestimate 1-hour concentration maximum
concentration values and to slightly underestimate annual concentration values.
28
-------
Mr. Irwin listed some side topics related to dispersion modeling and provided discussion on each
topic. The topics listed are as follows:
Why do things go wrong on estimated vs. actual concentrations?
• Coherent structures (eddies from surface to mixing levels in wind or turbulence field) cause
differences in concentrations. As the eddy rolls downwind, it captures the plume, and
disperses it.
• Wind directions change.
• It is possible for a plume to split - which may lead to very low concentrations in the center
of the plume.
Models simulate the ensemble of many representations of like conditions.
• Uncertainty can only be reduced so far. There is no way of knowing exactly what the wind
is going to do.
• Transport directions vs. narrow plumes.
Model evaluations define caveats.
• Comparison data is sparse.
• To validate a model is to confirm that the physics as modeled is indeed happening This is
impossible in the atmosphere. Hence, all we can do is evaluate a model. To evaluate a
model is to compare modeling results with observations (with no proof that the physics
modeled is indeed happening).
• Models produce estimates of what is expected on average. We compare these estimates of
average dispersion with observations which contain all the stochastic effects not modeled.
We should expect and do find large differences in such comparisons.
Air quality modeling is a skill
• Modeling is the artful use of science. The models themselves are simplistic; it is wise
application that is problematic.
Meteorological Data and Electronic Data Transfer, Dennis Atkinson, US EPA
29
-------
Mr. Atkinson began his presentation with a brief discussion of meteorological data and data
processing. Meteorological data is needed for refined modeling. The two types of data needed are
surface data and mixing height data. A preprocessor is needed - either PCRAMMET for NWS
data or MPRM for on-site data. Mr. Atkinson reviewed formats for a preprocessor.
Current Meteorological Projects Include:
• Revisions to the mixing height algorithm which includes creating two front-end programs
to accept surface and upper air CD-ROM data;
• SAMSON CD-Update to include the period 1990 to 1995 that will include Automated
Surface Observing System (ASOS) data starting in 1992; and,
• ASOS vs. NWS observer data analysis that will analyze the effects of limited cloud
information on dispersion modeling and will compare concentration estimates using NWS
observer and ASOS data in ISCST3.
Mr. Atkinson noted that the SCRAM has the latest modeling information available - including
models (user's guides), programs (user's guides), guidance documentation, meteorological data,
and State/local modeling contacts. This information can be found electronically at
www.epa.gov/scram001 or www.epa.gov/ttn.
AERMOD: Source-Specific, Near-Field Dispersion Model, Russ Lee, US EPA
The purpose of Mr. Lee's presentation was to describe the dispersion model AERMOD.
AERMOD is the first product of the AMS/EPA Regulatory Model Improvement Committee
(AERMIC). AERMIC was formed with the goal of improving science in applied models.
The model design criteria used for developing AERMOD includes:
• It must be based on up-to-date science,
• It must be simple, yet capture the essence of physical processes;
• It must provide robust concentrations (estimates over a wide range of sources and
meteorological conditions);
• It must be easy to update; and,
• It must have reasonable implementation. Inputs to the program should be reasonably
available.
30
-------
Mr. Lee compared AERMOD to the current regulatory model, ISCST3, as follows:
• Both models compute steady-state, near-field impacts;
• ISCST3 assumes Gaussian distribution of the plume in both horizontal and vertical planes,
while AERMOD uses a non-Gaussian form when appropriate;
• ISCST3 associates turbulence and dispersion with Pasquill-Gifford stability classes, while
AERMOD uses turbulence from modern planetary boundary layer theory and/or from
measurements;
• ISCST3 makes no distinction between dispersion from surface and elevated releases, while
AERMOD accounts for the differences realistically; and,
• ISCST3 has a discontinuous approach to the treatment of complex terrain while AERMOD
uses a more up-to-date continuous approach.
Expectations and future work for AERMIC include:
• Full-featured, updated, applied AERMOD in late 1997;
• Performance evaluation of three or four independent databases;
• Comparison of AERMOD to other applied models; and,
• Future work including urban dispersion, building downwash, wet and dry deposition, non-
steady-state, etc,
The evaluation of AERMOD is being completed in two phases. The first is called the
developmental evaluation, and is repeated during the development of the model. The second is a
final independent evaluation, which will be conducted this fall after AERMOD is finalized. The
developmental evaluation included data sets from both surface and elevated releases, rural and
urban environments, and flat and complex terrain Although AERMOD is still undergoing
revisions, the current version shows less bias than ISCST3.
Question: Does AERMOD handle stagnant conditions?
Answer: Maybe. If conditions hold true from examples studied, it will give closer to
observed results than ISCST3 (i.e., the results will be lower than ISCST3).
Source Attribution Modeling for Dioxin, Andy Roth, RAPCA
The purpose of Mr. Roth's presentation was to review a study that was performed for the City of
Dayton, Ohio, for dioxins. Dioxin is the term used for any or all of 48 polychlorinated
31
-------
dibenzodioxins and 87 polychlorinated dibenzofbrans. Measurement of dioxin was performed in
September 1995 using SW-846 Method 8290.
Six samples were taken at four different sampling sites around Dayton. ISC was used and the
predicted concentrations were compared to actual sample measurements. ISC results conflicted
with the actual measurements. Possible sources of error include:
• Dioxin emission rate estimates were based on log-normal correlation of dioxin emission
rates and ESP temperatures for testing in 1995, 1994, and 1988. If 1988 is not included,
the new estimate is -20% of the model input.
• Perhaps good resolution was not obtained by collecting 24 meteorological data points per
model run/sampling period.
• NWS calms protocol does not include winds below 1.54 m/s. These are set equal to 0 m/s.
This ignores dispersion during periods of light winds.
An alternative source attribution technique, the Chemical Mass Balance (CMB) Model, was
performed to compare results. The result of this comparison was that everything compared -
except Sample 5, which was found to include a source afterburner upset that led to the increased
emission rates.
Mr. Roth listed two discussion items for the group.
• Receptor modeling techniques may help in dioxin source attribution studies. Source profile
library? Mr. Roth stated that, during this study, he was impressed with CMB. This could
stand alone from NWS data for short term uses due to the availability of portable met
towers.
• ISC should be modified to employ five minute average met data. The availability of ISC
source code would be an issue in this modification.
The group discussed a way to "fake out" ISC by giving it five minute data. The user would give it
five minute data, ask for a period average, and then, at the bottom, the average would be a 24-hour
average.
Practical Problems in Combustion Air Modeling, Dr. Steven Ehlers, US EPA
Dr. Ehlers began his presentation by explaining that the ISCST3 modeling tool requires many
parameters and he was unable to find any information available on the sensitivity of modeled
results over the range of values presented as defaults for the different values of the parameters.
Dr. Ehlers performed a sensitivity analysis of different parameters on the ISCST3 model. The
results of Dr. Ehlers' analysis are summarized below. The sensitivity of the model to changes in
32
-------
the parameter were listed in four categories, severe (>50% difference from baseline Region 6
Protocol results), moderate (10-50%difference), slight (<10% difference), and none.
The analysis results that met Dr. Ehlers' "Severe" criteria are:
• Elevated vs. flat terrain: include terrain 10 km away.
• Surface roughness at measurement site: EPA-required value for NWS site
• Scavenging coefficients, isolated events 300%, but rare occurrence in Region VI.
The following parameters met the "Moderate" criteria:
• Anemometer height: underestimates <1 km.
• Particle size distribution and density: require stack test data for particle size and density.
This parameter was borderline and was considered close enough to "Severe" to be given
special attention at all facilities.
The following parameters met Dr. Ehlers' "Slight" criteria:
• Polar vs. cartesian grid nodes: applicant selects grid.
• Noon-time albedo: specify default values.
• Bowen ratio: specify default values.
• Minimum Monin-Obukhov length: specify default values
The following parameters did not have any impact on the model results:
• Terrain grid file. This is based on the use of the "ELEV" switch in the Control (CO)
section and the elevation component in the Receptor Grid files.
• Anthropogenic heat flux
33
-------
• Fraction of net radiation absorbed.
New Jersey DEP Risk Assessment Procedures for Minor Sources, Alan Dresser, New Jersey DEP
The purpose of Mr. Dresser's presentation was to explain the risk assessment procedures used in
the New Jersey Department of Environmental Protection. Applicability to minor source risk
assessment procedures include air permit applications for new or modified sources which emit one
or more of 56 specific carcinogens. A list of these carcinogens can be found in Mr. Dresser's
presentation handouts in Appendix D of this report.
All sources require a NJDEP Senior Permit Review Engineer to perform a Level I risk assessment.
Level I includes calculating ton per year emission rates for each carcinogen, using NJDEP
nomographs to obtain long-term ambient concentration of each carcinogen, using NJDEP unit risk
factors to obtain cancer risk factors for each carcinogen, and adding incremental risk factors
together to obtain the total cancer risk. If the total cancer risk is less than or equal to 1 in a
million, the facility passes the assessment. If the risk is greater, then a Level 2 assessment must be
performed.
During a Level 2 analysis, the applicants provide information on the sources, NJDEP performs
Good Engineering Practice (GEP) stack height analysis with Building Profile Input Program
(BPIP), and NJDEP does ISCST3 modeling with 2-3 years of hourly meteorological data. The
review is expanded to include 121 carcinogens and 119 noncarcinogens. If the facility fails the
parameters provided for a Level 2 assessment (i.e., maximum incremental cancer risk greater than
1 in a million), then the facility is not issued a permit. If the maximum incremental risk is less than
or equal to 1 in a million, then the facility passes. If the maximum incremental cancer risk of each
carcinogen is 1 in 10,000 to 1 in a million, or the maximum cancer risk is greater than 1 in
100,000, or the cancer risk at any sensitive receptors is greater than 1 in a million, the application
goes to a Risk Management Committee
The Risk Management Committee examines the feasibility of better pollution controls or
refinement in emission estimates, reducing the proposed hours of operation, modifying stack
characteristics for better dispersion (i.e., higher stack height, or change from horizontal to vertical
release), or conducting a more refined risk assessment to better define impacts or include other
sources at the facility.
2.5.2 Panel Discussion
Mr. Daniel Wise, Panel Chair, led the discussion of the issues presented in the panel issues paper.
He stated that the basic need in dispersion modeling is to have more technical guidance and
consistency in applying models. Other issues that were discussed during the panel session include
34
-------
• AERMOD is not competition for CALPUFF.
• More technical assistance is needed for complex terrain and sparse weather data.
CALPUFF can help in these situations. Also, NWS is producing data on winds for every
40 km every 3 hours. The data may be available every hour this year.
2.6 Exposure Assessment Techniques
2.6.1 Presentations
Exposure Modeling: Components, Status, and Uses, Michael Zelenka, NERL
The purpose of Mr. Zelenka's presentation was to provide background on exposure assessment
techniques. Mr, Zelenka reviewed the structure of the Office of Research and Development
(ORD) and the purpose of each of the four divisions within ORD. The four divisions are the
National Exposure Research Laboratory (NERL), the National Health and Environmental Effects
Research Lab (NHEERL), the National Risk Management Research Laboratory (NRMRL), and
the National Center for Environmental Assessment (NCEA). Mr. Zelenka outlined the role of risk
assessment to these groups and defined the focus of each. Presentation handouts can be found in
Appendix D of this report.
Mr. Zelenka outlined the steps involved in a risk assessment as follows:
• Hazard Identification - describe the adverse health effects that might occur due to exposure
to an environmental contaminant;
• Dose-Response Assessment - determine the toxicity or potency of a contaminant. The
dose-response assessment describes the quantitative relationship between the amount of
exposure to a contaminant and the extent of injury or disease,
• Exposure Assessment - describe the nature and size of the population exposed and the
magnitude and duration of exposure; and,
• Risk Characterization - use the information from the above listed steps to predict the
effects of exposure to the contaminant. Estimates are made of the likelihood that a
population will experience any of the adverse effects associated with the contaminant,
under known or expected conditions of exposure.
Human risk assessment research has evolved to a new approach driven by multimedia legislation,
growing recognition of the importance of multi-pathway risk assessments, increased capability to
measure multimedia environmental and human exposure metrics, and increased understanding of
biokinetics and the use of biological markers of exposure (biomarkers). Mr. Zelenka reviewed the
differences between the historic approach to exposure assessment and the evolved approach. The
35
-------
differences are that, rather than the historic approach of using single factors, the evolved approach
considers factors from multiple sources, pathways, routes of exposure, and health endpoints. Risk
assessment decisions also now have a multi-media and multi-pollutant focus. Mr. Zelenka
reviewed the evolving approaches to several factors involved in exposure assessment, such as
exposure pathways, temporal variability at specific locations, etc. Details of these approaches are
provided in Mr. Zelenka's presentation handouts in Appendix D of this report.
Human exposure simulation models used at NERL are:
• pNEMTX" (probabilistic NAAQS Exposure Model for pollutant "X") - simulates the
movements of people through zones of varying air quality to approximate the actual
exposure patterns of people living in a defined study area;
• HAPEM-MS (Hazardous Air Pollutant Exposure Model for Mobile Sources) - provides
long-term (1 year or more) average exposure estimates for pollutants generally attributed
to mobile sources; and,
• HAPEM-PS (Hazardous Air Pollutant Exposure Model for Point Sources) - provides
annual average estimates of exposures to hazardous air pollutants and resulting cancer
incidence for populations residing near point sources.
Mr. Zelenka reviewed comparisons between pNEM and HAPEM in areas such as for indoor
sources, residential patterns, and ventilation estimates. These comparisons are listed in the
presentation slides in Appendix D.
Mr. Zelenka listed the key elements for exposure model development as:
• Basic research: needed to improve estimates, including those that are modeled, of exposure
(e.g., microenvironmental concentrations, activity patterns, dose metrics)
• Improvements to the models;
• More, and better use of GIS,
• Faster and more user friendly models;
• Additional data (measured and derived) and computing platforms that allow the ease and
flexibility of performing human exposure research; and,
• Develop a modeling framework so that models that characterize exposure and dose at
target sites can be linked, and future exposure models will include the ability to be used as
tools for prioritizing the major routes and media of concern.
36
-------
Microenvironmental Modeling Issues, Dr. Alan Huber, NERL
Dr. Huber began his presentation by stating that the purpose of his presentation was to identify
issues with microenvironmental modeling; not to provide any solutions. Microenvironments are
the intersection of human presence/activity with a characterizable environment. Our daily activity
takes place in a series of microenvironments. Our total exposure is the sum of the exposure from
our local microenviroment and the regional environment. The goals of a microenvironment
assessment include population-distribution of exposure, special group and/or special
microenvironment assessment, effective reduction of exposure, and residual risk and/or relative
risk assessment. Model development/application and measurement studies must intersect.
Dr. Huber listed perspectives of models as:
• Regional scale models and ambient air;
• Urban scale models and ambient air; and,
• Local scale (microenvironmental) models - where the exposure begins and ends.
Characteristics used in model selection include pollutant emission sources, chemistry of interest,
human activity patterns for exposure, spatial and temporal scales of interest, available
meteorology, and numerical methods (i.e., Gaussian plume formulation vs. numerical simulation
techniques). Dr. Huber provided examples of exposure modeling activities. Dr. Huber's
presentation handouts can be found in Appendix D of this report.
Question: Any research on multiple buildings in urban microenvironments?
Answer: No. Program on regional models. Some research at university level.
Comparing Cumulative Annual Dose Estimates from pNEM to "Literature " Exposure Factors,
Thomas McCurdy, NERL
Mr McCurdy began his presentation by reviewing places in the CAA that require exposure
assessment There are requirements in the CAA to perform ecosystem health risk and impact
assessments. Mr. McCurdy stated that risk assessment procedures are a decision-making process
based on science - not science itself.
Mr. McCurdy explained that exposure is the time series of a joint set of air pollution and people-
activities at various ventilation rates. Direct data from personal exposure monitors (PEMs) or
exposure/intake dose generally are not available, especially for hypothetical scenarios (such as
reducing emissions or attaining standards). Therefore, exposure modeling must be undertaken.
37
-------
Mr. McCurdy listed two general concepts of intake dose and endpoints in exposure assessment as:
• Health impacts, spirometric effects, and symptomatic responses vary greatly with a
person's "dose profile." The severity of the effects is a risk assessment issue.
• The major endpoint is the biologically effective dose for organs.
Mr. McCurdy reviewed what constitutes a bad approach to a risk assessment. The bad approach
involves making assumptions (i.e., concerning the concentrations of pollutants) throughout the
assessment process that leads to false end estimations.
Question: What are the best tools to use at the State and local level?
Answer: pNEM
Real-time Estimation ofPolycyclic Aromatic Hydrocarbons, Dr. Nancy Wilson, NERL
The purpose of Dr. Wilson's presentation was to provide overviews of two projects involving
monitoring of polycyclic aromatic hydrocarbons (PAH). The monitoring can be performed using
either of two instruments manufactured by EcoChem - the PAS 1002i and PAS2000 The
instruments perform real-time monitoring of PAH on the surface of fine airborne particles
(aerodynamic diameter less than about 1 fim) by ionizing the surface of PAH using ultraviolet light,
removing the emitted photoelectrons, and measuring the positive current from the remaining
photoionized positive particles. The instruments are not highly precise quantitative instruments.
They can best be described as semi-quantitative. The agreement is good between integrated
sampling results and the results obtained from the PAS 1002i (the PAS 2000 model was not
compared). Dr. Wilson listed some differences in the two instruments. These differences are
outlined in her presentation handouts that are in Appendix D of this report.
Dr. Wilson reviewed examples of results obtained using the instruments in collecting real-time
values for ambient air. The examples included measurements in a non-smoking household, a
smoking household, and the exterior of a veterinarian's office The results showed peaks with a
response time of seconds, corresponding to the occupants' PAH-producing activities (for example,
smoking in a home or a diesel car driving by the veterinarian's office).
Current activities involving the PAH monitors include:
• Comparing the two instruments to see if the conversion factors are consistent.
• Monitoring total exposure to PAH of preschool children. Studies are including home and
day care centers so exposures can be combined to give the total exposure.
38
-------
The results shown by Dr. Wilson of children's total PAH exposures suggest that studies need to
include a multi-media approach, instead of just air, because people's daily activities make a big
difference in their exposures.
Minnesota Air Toxics Indexing System, Dr. Greg Pratt, Minnesota PC A
Dr. Pratt began his presentation by explaining a project currently being performed in his office.
Minnesota received a grant from EPA to do a measurement and source apportionment of human
exposures to toxic air pollutants. This study compares modeling and canister results to results
obtained from personal badge monitoring. The study includes 15 VOCs from two inner city
neighborhoods and one suburban neighborhood with 20 participants each.
The project involves measuring the personal exposures of individual study participants to the
VOCs using personal monitoring badges. Simultaneously, indoor and outdoor concentrations of
the same substances will be measured at each participant's home (using badges), in each study
neighborhood using canister methodology, and at central sites Additionally, the exposures will be
modeled using standard dispersion modeling methods. The hypotheses tested involve comparisons
of each of the quantifications of exposure to the "gold standard" of exposure measured on the
badge worn by the participant
Dr. Pratt continued with a presentation of an indexing system that was developed in his office.
Minnesota implemented a Toxic Air Pollutant Indexing System to rack chemicals by their
environmental hazard potential. The system assigns a substance an index number, where the index
is the hazard potential found by dividing the potential exposure by the toxicity of the pollutant. To
date, 195 pollutants have been indexed and the indexes have ranged over 21 orders of magnitude.
The potential exposure is estimated with Level HI Fugacity Model using chemical and physical
properties of the substance. The system uses concentrations of the substance estimated in six
compartments: air, aquatic biota (fish), terrestrial flora, water, soil, and soil water. Standard
toxicity indicators for human health (IRIS2 and HEAST), ecological effects, and other data are
used The index values are being used to do the following:
• Set thresholds for reporting requirements (registration, inventory, environmental
monitoring, and enhanced emission estimates);
• Set air emission fees using hazard-based fee rates (rather than flat rate);
• Identify persistent and bioaccumulating substances that require further study,
• Refine environmental monitoring goals; and,
• Set priorities for facility review.
39
-------
2.6.2 Panel Discussion
Ms. Aimee Kennedy, Panel Chair, provided an introduction of panel members and a brief overview
of the issue paper for this panel. Panel members provided presentations of certain aspects of the
issue paper and the issues were discussed with the workshop group. Ms. Kennedy revised the
issue paper, and the final version is included in Section 3.0 of this report. Some general comments
made during the panel discussion include:
• Some type of model for use in urban settings would be helpful;
• There is concern that a standardized approach to exposure assessment will stifle State
innovation.
2.7 Case Histories of Multi-pathway Modeling Efforts
This section provides summaries of the presentations for this workshop session. There was no
panel group for this topic, and, therefore, no issue paper.
The Mercury Study: An Integrated Risk Assessment, Martha Keating, US EPA
The purpose of Ms. Keating's presentation was to describe risk assessment procedures used in a
nationwide study of mercury emissions The Clean Air Act as amended in 1990, Section
112(n)(l)(B), required the U.S. Environmental Protection Agency (EPA) to prepare a report to
Congress on mercury emissions. A risk assessment was performed to evaluate mercury emissions
and a report was prepared for Congress. The assessment was started in 1992, and the report has
not yet been released. The purpose of this presentation was to describe the approach which was
used to model and characterize the problem, how limitations in the assessment were described, and
the lessons learned from applying these tools to a problem that has far-reaching policy implications.
The Mercury Study was designed to integrate a number of different types of information including
data on type, sources, and trends in emissions, identification of major pathways of exposure to
humans and wildlife, and estimates of control technology efficiencies and costs. The Mercury
Study exposure assessment relied on a number of models to predict mercury fate and transport in
the atmosphere, watershed, water body and aquatic and terrestrial food chains. The study included
electric utilities, municipal waste combustors, and other sources (including area sources).
The study was aimed at answering many questions posed in the beginning of the assessment, which
include:
• In considering the sources, pattern, and magnitude of deposition, is there a link between
deposition and elevated environmental concentrations?
• What are the adverse effects?
40
-------
• What are the control technologies available? How much do the controls cost?
• What are the research needs?
The components of the study included.
• Emissions inventory;
• Environmental fate and transport;
• Exposure (numerous exposure pathways were assessed, including wildlife);
• Health effects;
• Risk characterization, and,
• Control technologies and costs.
Each facet of the Mercury Study has it's own set of uncertainties - either due to uncertainties
inherent in the approach (e.g., using emission factors for the emissions inventory) or uncertainty
related to the models themselves or the parameters used in the models.
Ms. Keating summarized the "lessons learned" from this study as:
• "Science guides, but policy decides" The risk assessment is only one piece of the
information needed by the risk manager. Other risk management factors such as economic
considerations, technical feasibility, and other non-risk related factors are always part of
policy considerations.
• Extensive justification and documentation of model parameters is essential for a
"transparent" analyses. The downside is that there is much information for critics to
challenge.
• A discussion of uncertainty, even if it is qualitative, is required. Describe the limitations of
the assessment. Know your models.
• Sensitivity analyses are useful to test a range of possibilities when there is no one correct
answer. Using a range is more defensible than point estimates.
• Use measured data as much as possible to test and corroborate modeled predictions.
• Qualify conclusions, if necessary, with statements about the level of confidence in the
results.
The presentation handout in Appendix D provides a more detailed summary of the study.
41
-------
The Screening Risk Assessment Evaluation for the Proposed Incineration Project at the
Anniston Army Depot Chemical Demilitarization Facility, Leigh Bacon, Alabama DEM
The purpose of Ms. Bacon's presentation was to describe a screening risk assessment procedure
used in the evaluation of Anniston Army Depot Chemical Demilitarization Facility. Screening risk
assessments are comprised of a multi-pathway human health risk assessment (HHRA) and an
ecological risk assessment (ERA). The ADEM Air Division did not evaluate impacts for the ERA.
HHRAs include estimated risks and hazards from direct exposure through inhalation of emissions
and estimated risks and hazards from indirect exposures through soil, water, and food products
exposed to emissions depositions.
Ms. Bacon explained that the U.S. Department of Defense currently stores lethal unitary chemical
agents (which includes the blister agents HD and HT and the organophosphate nerve agents GB
and VX) at the Anniston Army Depot. These agents are stored as bombs, cartridges, mines,
projectiles, rockets, and ton containers. The proposed demilitarization of these agents would be
achieved through incineration. The HHRA included definition of the study area, definition of the
exposed individuals, determination of the concentration of each substance in the environmental
media, estimation of the amount of substance each individual is exposed to, and assessment of the
toxicity of media concentrations. Methods of exposure evaluated from the incineration include
paniculate concentrations via air, vapor concentration via air, dry deposition via food and soil, and
wet deposition via water and milk products. Emission rates for AAD sources were established
using data at the Johnston Atoll Chemical Agent Disposal System (JACADS). Maximum lifetime
incinerator/source specific emissions were determined for the chronic assessment. Maximum
lifetime term emissions were used for the acute assessment. Air dispersion modeling tools were
used to predict the paniculate and vapor concentrations for the acute portion of the HHRA. An
electronic spreadsheet developed by USACHPPM was used to verify media concentrations for a
specific pollutant for a specific exposure scenario. Through dispersion modeling and the use of the
electronic spreadsheet, the ADEM Air Division was able to verify media concentrations used in the
health effects portion of the SRA. The U.S. Army submitted a SRA for the proposed
demilitarization activities at the Anniston Army Depot Ms. Bacon's presentation handouts are
included in Appendix D of this report.
2.8 Education and Outreach
2.8.1 Presentations
Training Opportunities in Air Toxics for You Through the Education and Outreach Group,
Howard Wright, US EPA
Mr. Wright began his presentation by defining the Education and Outreach Group's (EOG)
missions as follows:
42
-------
• Provide technical air pollution training to primary clients (State and local air agency
personnel);
• Promote environmental education to teachers and students, with a focus on K-12; and,
• Enhance outreach activities by forging new partnerships and strengthening existing ones
(e.g., small business support, Air Quality Learning Centers, Public Information Officers).
Mr. Wright explained that training needs are assessed through joint needs assessment surveys, a
STAPPA/ALAPCO/EPA training committee, course evaluations, and an annual training
conference. Ron Townsend explained the need to conduct needs assessment surveys. Needs
assessments feed to training priorities, which feed to the Strategic Planning Process. Training
ideas can be voiced during the assessment surveys for consideration during the planning process.
Mr. Wright listed ways that training can be provided as:
• Self-instructional materials;
• Classroom courses; and,
• Satellite broadcasts (on the Air Pollution Distance Learning Network (APDLN)).
He then listed several APDLN Broadcasts that are scheduled for the next few months. This type
of information, as well as other EOG training information, can be obtained through the following:
• The OAQPS Home Page on the internet,
• The APTI BBS on TTN (dial 919-541-5742);
• The WESTAR Home Page;
• EPA's LAN services menu;
• The APTI 1997 classroom schedule; and,
• State and local site coordinators and training contacts.
Mr. Wright added that an EOG web page will soon be available.
Unified Air Toxics Website, Dr. Nancy Pate, US EPA, ITPID
Dr. Pate began her presentation by defining the purpose of the Unified Air Toxics Website as to
provide the general public, federal, State, and local governments and emitting facilities,
comprehensive air toxics information (from basic to very technical) in a centralized location on the
internet. The website also encourages sharing of information in order to reduce duplication of
effort. The website can be accessed at http://www.epa.gov/oar/oaqps/airtox.
43
-------
The six technical areas either already on the website or soon to be included are:
• Basic Facts;
• EPA Rules and Implementation;
• Pollutants and Sources;
• Technical Resources;
• EPA Programs; and,
• State and Local Agency Programs.
Explanations of these six areas can be found in Dr. Pate's presentation handouts in Appendix D of
this report.
2.8.2 Panel Discussion
For the Panel Six discussion, Brian Haugstad reviewed the points made in the issue paper with the
group. The panel listed seven basic deficiencies/recommendations to be included in the paper.
Examples and discussion regarding each of these issues is provided in Mr. Haugstad's presentation
handout in Appendix D of this report. The seven main issues include:
• Training should focus upon the needs of the regulatory agencies implementing the
regulations, while clearly understanding who is the target audience.
• Many of the video materials available to States and local agencies are out-of-date.
• Make EPA training and outreach materials more accessible.
• Publications such as quarterly newsletters, fiber-optic television links throughout the state
(for state specific education and outreach programs), and informational brochures mailed to
regulated facilities to aid in MACT compliance, are helpful in educating regulated facilities.
• Small Business Assistance Groups are often helpful in providing confidential, on-site
assistance with regulatory compliance and educational materials to regulated facilities.
• A plain English Guidance Document is needed to instruct agencies on how to effectively
collect, analyze, and present data.
• Make training materials and specifically satellite/video training more interactive.
Comment: A plain English guide to the Clean Air Act is available from the Public Assistance
Office.
44
-------
Mr. Haugstad incorporated comments from the discussion. The final version of this paper is
included in Section 3.0 of this report.
3.0 Issue Papers
Panel group members worked collectively to develop lists of issues to be addressed during the
workshop in each of the workshop areas. The issue papers are structured to be in response to
three basic questions. These questions are:
• When are the present technical tools adequate in helping you do your job?
• When are they deficient?
• List possible solutions that might alleviate some of the items on the deficiency list.
The following sections include the final issue papers for each panel. These final papers incorporate
discussion and comments made by workshop participants during the workshop.
3.1 Elements of the OAQPS Air Toxics Program, Panel One
What are some suggestions for desirable qualities of toxic tools, in general?
• A whole spectrum of tools is needed, from simple screening tools to sophisticated, complex
tools.
• Packages that integrate several tools can make tools easier to use and can address up-front
any incompatibility between tools. For example, a combination of tools, such as the
Maryland Air Quality Implementation Management System (AIMS), would be helpful in
developing an urban air toxics program
• Caution should be used if an attempt is made to force reconciliation between monitored
and modeled data.
• Cooperation among various groups with different expertise is often necessary, e.g., EPA,
State agencies, city governments, and research facilities Multimedia approaches may also
be necessary.
• Ways to communicate risk predicted with these tools should also be available to specialists
working with all types of tools.
When are the present technical tools adequate in helping you do your job?
• They are adequate when conducting a screening analysis for a single air toxic contaminant.
45
-------
• They are adequate when evaluating alternative scenarios for comparative purposes (e.g.,
cleaning up Superfund sites using a stripper versus other methodologies). These
evaluations are good in a relative sense rather than an absolute sense.
When are they deficient?
• They are deficient when less-than-lifetime risk evaluation is required.
• They are deficient when multiple contaminants are involved (cumulative risk analysis for an
area).
• They are deficient when source characteristics or emissions are unknown or intermittent.
• They are deficient when they do not address chemical and physical transformations.
• They are deficient when they cannot address time-dependent changes
List possible solutions that might alleviate some of the items on the deficiency list.
• Regional, State, and local agencies are in need of training on which technical tools are
available to them, e.g., what is the proper equipment, and the proper use of the tools.
• More validation studies are needed systematically structured to specify characteristics for a
range of sources to boost confidence in results of tools especially for more complicated
small sources, such as for short stacks, fugitive emissions, etc.
• Prepare materials that can be distributed to small businesses to provide assistance in
complying with air, surface and groundwater, wastewater, and waste management
regulatory requirements. This is helpful in educating people on air toxics programs.
• It would be helpful to leverage training funding by providing training materials that can be
used again in training of other personnel
• An electronic bulletin board system providing written descriptions of real-life solutions to
issues would be useful.
• Development of a complete chemical database that includes unit conversions is needed.
• Since there are already so many activities (e.g., Section 112, NESHAPs, MACTs, etc.)
going on, a way to mesh together the activities so that work can be performed more
efficiently would be helpful.
3.2 Ambient Air Toxics Monitoring and Analysis Data, Panel Two
When are the present technical tools adequate in helping you do your job?
46
-------
• The guidance documents that are available for performing air pathway analyses at
Superfund sites, which could also be used to some extent for non-Superfund activities, are
adequate.
• The monitoring data for some toxics are adequate to assess the general trend in toxics
levels. They are also adequate for cross-checking estimates of exposure using models for
reasonableness.
• Ambient air toxics monitoring tools are available to measure actual ambient levels of
specific toxics in those communities and estimate the impacts. The data will allow us to
determine which, if any, toxics need to be looked at closer.
• Guidance documents, the Compendium of Methods, and consultation with HQ expert staff
have been useful at regional level to the further understanding of the air toxics problem.
• Examples of technical efforts that have worked well are as follows:
• An air toxic sampling effort was initiated with the cooperation of several State and county
agencies. Each agency conducts specialized analyses. For example, one agency performs
analysis for VOCs via canister sampling/analysis, another performs benzo-a-pyrene (BAP)
analyses, and another conducts aldehydes' analyses
• In a recent example, a school was closed due to odor complaints and children and staff
getting sick. Through utilizing expert staff, the State's air toxics mobile lab, and regional
in-house knowledge, the problem was alleviated through the installation of a permanent
sampling location on site and cooperation with the local industries in announcing to the
school when upsets occurred.
• There have been successful efforts in initiating air toxics monitoring projects to assist in
the multimedia effort to study the problems in communities. Support from the State
agency to do the field work helped to get three stations working.
When are they deficient?
• At times, there are State air quality standards or risk-based action limits that are established
for particular projects that are much lower than the available method detection limits.
Guidance is needed on what to do in these situations.
• Guidance on how to combine available air toxics monitoring and analysis tools into a
comprehensive air toxics review is lacking. This guidance should include which monitoring
and modeling methods are appropriate in certain situations and how to interpret the
monitoring data.
47
-------
Emissions data that are temporally or spatially grided to an adequate resolution are also
lacking, which makes it difficult to compare ambient versus reported emissions.
The biggest problem we run into is with the data that is used with the computer software
technical tools. An example of a general problem that we have is that the Air Quality
Division as a whole has a difficult time collecting and entering accurate data into a single
database that everyone can use. We are currently addressing the problems internally
through a data management workgroup. An example of a specific problem is that we can
use our global positioning system to precisely locate emission sources but the electronic
maps that we currently have are not accurate enough to properly represent these source
locations.
Toxics monitoring data are not adequate for developing estimates of exposure in most
parts of the country, because there is so little of it and many monitors are located in places
where criteria pollutant readings are expected to be high.
1 True background monitoring data is needed to evaluate high toxic levels.
1 Without consistent health/risk data on air toxics, adequate analysis of the data is difficult.
• The cost of monitoring is high. In addition, most monitoring staff and resources in our
state are geared toward criteria pollutants. We typically can only do air toxics monitoring
through grants from EPA or by asking the source to pick up the cost for community
relation purposes. Typically, we have to rely on past monitoring studies or studies from
EPA or other State and local agencies.
Recommendations
Air toxic pollutants and their impact on communities within urban areas has become an
issue that has been getting more attention. Developing a process to address the concerns
of the community is needed. This process should include recommendations on appropriate
monitoring, modeling, and analysis methods. The following issues could be discussed:
How do you bring all interested parties/stake holders together to provide information and
data that are necessary to address the health concerns of a community?
How do you develop a plan that fulfills all aspects of the stated objective? The plan could
include the following:
- What are the health or environmental concerns of the community?
- Is there a high incidence of cancer in the community?
48
-------
- Can the identified cancer be traced to specific air toxic pollutants?
- What air polluting sources within the community or surrounding communities could be
contributing to the specific health problems and is there emissions inventory data
available for these sources?
- Should or can inspections be performed at facilities within the community or
surrounding communities to obtain information that could be used to link the health
problems?
- Could emissions tests be performed at facilities suspected to be emitting contaminants
that are significantly impacting ambient air quality and how would this be done?
- With the aid of emissions and air dispersion modeling data what ambient air sampling
or monitoring program could be implemented'7
- How could the information and data collected be compiled, assessed for data quality,
and compared to health standards?
• As local, State, and Federal governments continue to downsize and are asked to do more
with less, innovative air toxic monitoring tools and procedures need to be developed.
• Mobile type air toxic monitoring equipment capable of continuously monitoring a wide
range of pollutants is needed to identify those sources impacting air quality and
characterize what the air quality is within a defined area as it relates community-based
health concerns.
• The following are needed to enhance the ability of field personnel to obtain data and
information quickly in the field:
• A computer database that can provide aerial photographs. These photographs could be
used for evaluating the history of land use at a specific site or industry facility.
Response. There is a database of all aerial photos in the federal government's possession. It
is the "Aerial Photography Summary Record System" operated by the USGS in
Reston, Va. The phone number is (703) 648-5903.
• A user friendly computer database or cellular phone type system to access information such
as: business phone and address directory, health data, meteorological data, emissions
inventory data, ambient air data for both criteria and noncriteria pollutants, mapping system
capable of overlaying the data described above.
49
-------
Response: EPA's Envirofacts System, which will soon be completely accessible via the
world wide web, may help in resolving this issue.
• Continue to develop FTIR, UVDOAS/open path and laser-based air monitoring
technology.
• Guidance is needed on what kinds of requirements may be made in the future so that
necessary changes or purchases can be made. Examples of these future requirements are
speciate particulate analyses, metals analyses, aldehyde and polar compound analyses.
Some of these would require using PM2.5 filters, which are difficult to use due to the size
and are also expensive.
Response: There should be some guidance available on fine particulate speciation in the
IMPROVE Visibility Transport Monitoring Program.
• A comprehensive federal list of acceptable ambient concentrations for air toxics is needed.
At the Workshop a recommendation was made to change the term, "acceptable
ambient concentrations" to a more friendly term. Someone in the group indicated
that Region III has an appropriate term.
• There needs to be an AIRS spatial scale for site-specific monitoring (i.e., neighborhood or
city). The scale that is used may not be representative of what the air mass actually is.
• It would be helpful to have training classes of technical tools for FTIR and UVDOAS
monitoring. These classes would address identifiable tools and uses, program uses, and
where users can go to obtain information.
• Develop and improve remote sensing capabilities, such as satellite imagery, to produce
"real-time" flux measurements.
• A comprehensive listing of method detection limits and concentration units are needed to
adequately analyze air toxic data.
• Integration of air toxic data into multimedia programs needs to be developed to help
evaluate impacts to ecosystems.
• There needs to be an Air Toxics Monitoring Strategy developed.
50
-------
3.3 Air Toxics Emission Inventories, Panel Three
When are the present technical tools adequate in helping you do your job?
• They are adequate for targeting source categories for toxics analyses.
Response: NO. Source categories have been chosen from experience and background
knowledge. Inventories may help in ranking categories.
• The toxic emission inventory data for all source categories are adequate to provide a rough
estimate of the relative contribution of various source categories and to assess trends over
time.
• They are adequate and helpful in determining rule applicability based on emission level.
When are they deficient?
• We need upgraded tools/software as well as recommended procedures for applying the
tools for area, mobile and point sources. The ideal would be to have a reporting system for
air emissions that would provide the same details as we now have for criteria pollutants
(i.e., stack locations and parameters, etc.).
• The inventories are not as helpful when no uncertainty estimates are included in the
inventory. Reliability ratings are somewhat related but are not necessarily enough
information. The inventory should provide a quantitative estimate of the uncertainty
distribution associated with each chemical and each industry. Such a data set would
facilitate Monte Carlo simulations of uncertainty.
• The inventories are not adequate when the base year of the inventory is not coordinated
with the available meteorological data. A limited number of meteorological data sets can
be produced due to resource constraints.
• The available tools are too poor to develop estimates for mobile source categories that are
precise enough to use in comparing the relative impacts of various control strategies, such
as fuel reformulation, and on emissions from on highway and nonroad vehicles.
• Speciation profiles in the SPECIATE database are in need of updating for highway vehicles
and aircraft. The highway vehicle profiles are in need of improvement, especially for
reformulated fuels. These profiles are merely baseline profiles where relative contributions
of certain compounds have been adjusted. VOC profiles for diesels are by carbon number
rather than specific compounds. Such profiles where contributions of specific compounds
are quantified for diesel engines exist and should be incorporated. Newer speciation data
51
-------
for aircraft engines should be reevaluated and possibly incorporated. There is nothing for
nonroad engines in SPECIATE.
• A new toxic emission model for mobile sources, MOBTOX, needs to be developed and
guidance provided on how to use the model for State and local government agencies.
• Overall guidance is needed on estimating toxic emissions and uncertainty levels from
highway vehicles, nonroad equipment, aircraft, commercial marine engines, and
locomotives.
• Once an accounting of HAP emissions has occurred, shouldn't there be some decision-
making guidelines concerning the pursuit of evaluating ground level impact? For example,
EPA could put the HAPs into "bins" that determine what emission rates of specific HAPs
are significant enough to pursue a risk assessment strategy. These bins could also be used
to determine what emission levels are significant enough to even include in the account-
ability portion of an inventory. Do we need to report emissions less than 1 Ib (for example)
for many of the HAPs? There needs to be some consideration of credibility in regards to
the extent that regulatory agencies pursue emissions data. There also needs to be some
guidelines for regulatory agencies in this matter in order to eliminate man-hours wasted in
data entry and the hard copy forms needed to provide this information to the data entry
personnel.
• There may be a need to develop a training program that would include the development of
a reading list of references that would provide the requisite background to understand the
issues, what inventory work, modeling, and research has been done, what the research
needs are, and so on.
• There may be a need to develop specific training programs on how to do a point, area, on-
road and nonroad mobile source, and biogenic toxic emissions inventory.
• Since EPA staff has been insistent on the point that the SPECIATE database is not reliable
for point source toxic emissions inventories, and should not be used for that purpose, it
would be very helpful if EPA could provide speciation profiles that are more appropriate,
i.e., either update SPECIATE or provide a separate toxics speciation database.
• There may be a need to develop a consensus-type process, similar to the Emissions
Inventory Improvement Program's (EIIP), to compile existing inventory methodologies
and/or develop new methodologies. This would help assure that different States and locals
used the same steps or procedures in developing toxic emissions estimates.
Response: Few agencies have the resources to become involved in emission factors research.
It is anticipated that there will be continued reliance on EPA for research
52
-------
List possible solutions that might alleviate some of the items on the deficiency list.
• Continue making toxics data (including government and industry data) more accessible and
easy to find (internet).
• Encourage good documentation including QA and error estimates.
• Provide some means of dealing with secondarily formed pollutants without running a
regional oxidant model. Current tools are not very amenable to doing a good analysis for
things like formaldehyde.
Response: There needs to be a policy on handling of "condensibles," air toxics that are solid
paniculate, air toxics adsorbed on particles, and air toxics that are gases. There
should be a distinction between soluble fractions readily absorbed into the body
and insoluble fractions. There should be distinctions in inventories between:
TSP - Total Suspended Paniculate
PM10 - Inhalable paniculate
Condensible. Tars, liquid heavy hydrocarbons, naphthalene, etc., that are emitted as liquid
drops or vapor and condense in ambient air.
Secondary Particles. Sulfates, nitrates, etc., that are a product of chemical reactions in air.
This is mostly a monitoring issue where this fraction should be
identified.
Solid Fraction. Insoluble-silica, carbon, silicates.
Soluble-benzene soluble organics, soluble metal compounds, e.g., Pb, Cr, As.
Harmless- salt, sulfates, etc.
Liquid-acid mist.
We need to differentiate in inventories between particles according to origin, composition,
and health impacts as well as by size.
• Keep emissions estimation and inventory tools (i.e., emission factor databases) updated and
make the updated versions available on a regular basis.
Response: We have previously proposed a "look-up" table in AFS where emission factors
that have been used nationwide could be accessed by SCC Code to see what all of
the agencies across the country have used to estimate emissions. This would
make the most data available to the widest range of users and help achieve
nationwide consensus. We should collect information from other nations and
incorporate it in the system. They also have some good data and good ideas.
53
-------
• Do whatever is regulatorily necessary to provide support to existing State air toxics
programs. Do this in a hurry before these programs are watered down (or unfunded) due
to lobbyists' claims that EPA will be coming forth with their own risk assessment values in
the near future and that State programs are duplicative.
• Ensure inventories are compatible with the models likely to utilize the emissions
inventories. Meteorological data, deposition data, and inventory data set production
should all be coordinated to produce the highest quality coordinated data sets. All three
data sets are expensive to generate and can not be performed on a model application by
application basis.
OTHER ISSUES:
• There is a need for investigation and resolution of the impact(s) of area source emissions
area-wide vs. "hot spots." Is it appropriate to spread emissions over an area (e.g.,
county-wide) as opposed to looking at the health impacts of a specific location? Is it
accurate to apply typical area source type emission factors/ methods based on population
or employment rather than doing the detailed type of analysis that might be required for a
permit application or a point source inventory?
• There is a need for some way to meaningfully relate health risk assessment and impacts to
emissions levels so that emissions data become more meaningful as a consequence (a
weighting scheme, screening model?).
• There is a need to tie in with the EPA Consolidated Reporting initiative to clarify what
Emissions Inventory responsibilities will be regarding HAPs as opposed to TRI reporting
requirements. TRI currently does not collect enough information to perform modeling for
health risk assessment.
• There is a need for a model/spreadsheet to apply toxic fractions or speciate on-road and
nonroad mobile source emissions and aggregate the emissions of specific toxic chemicals.
• Guidance is needed on inclusion/exclusion of categorically insignificant activities at Title V
facilities.
• Activity Data is needed for nonroad mobile sources.
54
-------
3.4 Dispersion Modeling, Panel Four
When are the present technical tools adequate in helping you do your job?
• EPA guideline dispersion models are powerful, yet easy to use, tools for predicting the
ambient impacts of air toxic chemicals for most situations. The two models that are
commonly used for predicting air toxics impacts are SCREENS and ISCST3.
• ISC3 tends to do a competent job in handling HAPs analyses. Risk assessments comparing
modeled concentrations to health-based standards are adequate.
• Receptor modeling is a complimentary tool to ISC3.
• Dispersion models in general are adequate.
When are they deficient?
• The existing guidance on using models specifically designed for hazardous/toxic air releases
is not adequate. The use of these models is much more complicated than the use of ISC3
and SCREEN3, especially when quantifying source data. Also, none of the public domain
hazardous/toxic refined dispersion models, e.g., ADAM, ALOHA, DEGADIS,
HGSYSTEM, SLAB, and AFTOX, are capable of treating multiple sources or more than
one set of meteorological conditions in one run. These will become important issues as
Section 112(r) programs are implemented.
• There is some concern within the modeling community regarding the accuracy of the
available algorithms for modeling downwash caused by nearby structures. The downwash
algorithm in EPA's ISC3 and SCREEN3/TSCREEN models is too simplistic and contains
a number of unrealistic discontinuities (treatment of plumes just above and below the GEP
stack height, the Huber-Snyder/ Schulman-Scire algorithm division point, and the
cavity/wake region discontinuity) There is no guideline method for estimating cavity
impacts for averaging periods greater than one hour. Regulatory agencies are therefore
required to develop their own methods.
• ISCST3 is deficient when determining less than one hour average concentrations from
continuous, "steady-state" toxic releases.
• When modeling HAPs, no chemical transformation is assumed. Some of the volatile
organic HAPs may be reactive enough to warrant the use of a decay factor in the modeling.
• ISC3 is deficient in simulating valley or channeling effects.
• Regarding deposition modeling
55
-------
• Calm hours are ignored (which could underestimate the total flux).
• Mercury valence states create difficulty. There is not any good data on this and how it gets
deposited.
• For wet deposition, rain out is not considered - only wash out.
• Gas scavenging coefficient and particle size distribution data is limited.
• There is some discontinuity between guidance regarding Superfund sites such as:
• Which receptor height to use to represent the human breathing zone (zero feet versus 3-6
feet);
• What constitutes a secured property line; and,
• Using maximum ground-level concentration versus some statistical analysis instead.
• Guidance for modeling non-point air toxic sources is limited.
List possible solutions that might alleviate some of the items on the deficiency list.
• Develop a guideline method for predicting the impacts of air toxics within building cavities
that has the support of the modeling community and that can be used to predict impacts for
different averaging periods. A new building downwash algorithm known as PRIME has
been developed that addresses most of the technical problems with the ISC3/SCREEN3
downwash algorithms. The PRIME algorithm is now undergoing an independent
evaluation. If it performs better than the EPA's current recommended downwash
algorithm in this evaluation, it will be recommended as the guideline technique. There are
plans to incorporate PRIME in both the ISC3 and CALPUFF models, which will allow for
the prediction of long-term impacts using sequential meteorological data. Rhode Island
and Texas have developed a computer program that predicts long-term cavity impacts
using sequential meteorological data.
• EPA should evaluate each of the HAPs and, where chemical transformation rate is
significant, provide guidance on a decay factor for that specific compound.
• Guidance is needed for performing more accurate HAP modeling; for example,
characterization of non-point sources.
• Modeling guidance between different EPA offices should be consistent.
56
-------
• A section on the SCRAM web page for air toxics Frequently Asked Questions (FAQs) is
needed.
3.5 Exposure Assessment Techniques, Panel Five
When are the present technical tools adequate in helping you do your job?
• ISC3 modeling coupled with basic inhalation pathway exposure calculations:
r
• They are adequate when assessing the incremental risk/exposure caused by only one
chemical, through the inhalation pathway.
• They are adequate for prioritizing the risk from specific sources by comparing the
risk/exposure caused by a chemical at a specific source to another source with the same
chemical.
• They are adequate for comparing the risk/exposure caused by one chemical at a specific
source before and after compliance.
• HAPEM-MS3 for toxic emissions from mobile sources:
- Appropriate for estimating annual average human inhalation exposure by hour of the
day and season for a population.
- Appropriate for assessing exposure to toxics that are variable in time or space
providing that measured, or modeled, ambient data for that toxic pollutant are
available.
- Adequate when CO can be used as a surrogate for toxics.
• Fugacity Model:
- Appropriate for assessing concentration of pollutants in different media in order to
prioritize HAP list (used by Minnesota to develop their HAP index)
• HEM:
- Panel members were not as experienced in dealing with this model. Therefore,
comments on its adequacy could not be assessed.
- Interest in learning more about how to apply this model was expressed.
57
-------
When are they deficient?
• When the exposure to more than one chemical needs to be assessed. The interaction
between chemicals is hard to predict.
• When the exposure through multiple pathways need to be assessed. There is no clear
guidance on what tools or protocols should be used (expected to be addressed by TRIM).
• When the exposure from multiple sources needs to be assessed. Tools and data are not
readily available to determine HAP emissions from all sources in a given area (expected to
be addressed by TRIM).
• When the risk assessment results need to be interpreted for use in setting limits or to be
communicated to the public.
• How to understand uncertainties.
• How to account for lack of multi-pathway or background data.
• How to enforce limits calculated from risk or exposure assessment results.
Li si possible solutions that might alleviate some of the items on the deficiency list.
• Provide clear guidance on when Air Toxics Assessments should be conducted.
• Case-by-case MACTs.
• To find the residual risk for existing MACTs to ensure safety.
• Regional air toxics studies.
• Provide clear guidance on who (State, EPA, company) should conduct the exposure
assessments (i.e., who has the burden of proof to demonstrate that the limits are
protective?).
• Develop a standardized guidance on what should be included in these exposure
assessments.
• Multi-pathway vs. inhalation exposure.
• Multiple chemical exposure vs. single chemical exposure.
• Uncertainty analysis (qualitative vs. quantitative).
• Specific modeling techniques or defaults.
• Point source modeling vs. area and mobile sources.
58
-------
• Provide training for the standardized approach to conducting the Air Toxics Exposure
Assessment (instead of standardizing the entire program, a participant recommended
developing a web site or bulletin board similar to SCRAM that would have an EPA
approved list of exposure models, how and when to use them, EPA expert contacts, and
suggested default inputs).
• Provide guidance on how to interpret risk assessment results.
• How to understand uncertainties.
• How to account for lack of multi-pathway or background data.
• How to compare the results to either an absolute incremental risk limit (i.e., 10"6) or a
relative risk measurement (i.e., cancer incidents, etc.).
• How to use the results effectively.
• Develop better data to be used in the exposure assessments.
• Toxicity data
• Particle size distributions (will need to require the facilities to do it).
• HAP emissions data from sources to evaluate regional exposure (State will need to help
out with this part).
• Software that incorporates exposure routes, toxicity data and GIS and/or population data
to evaluate regional exposures (expected to be resolved with TRIM model).
• Biological pharmokinetic data.
• Provide guidance and/or training on how to communicate the risk results.
• Address the following policy questions
- What role will States play in residual risk determination?
- What steps (gathering information, training, etc.) do the States need to be doing in
order to prepare for their involvement with residual risk?
- What type of data will be required for the new exposure models, so that the States
can start collecting the data if not already available?
- How will the residual risk program affect States that already have a risk-based air
toxics program?
59
-------
3.6 Education and Outreach, Panel Six
PART ONE:
When are present technical tools adequate in helping you do your job?
1. In providing staff training specific air toxics standards or procedures, and in providing
information on general air toxics issues.
2. In providing education, outreach and training to the regulated community. Printed
training and outreach materials, and live or taped educational videos are helpful, when
provided in a timely manner.
PART TWO:
When are technical tools inadequate?
1. Training often focuses upon general facility operations, and upon pollution control
methods. Instead, there should be greater emphasis placed upon regulatory requirements,
and the means necessary to regulate these requirements.
2. When training is not provided in a timely basis. This happens frequently.
3. When telecourse speakers provide vague information, and are unable to answer specific
questions or details.
4. When communicating risk assessment to the public.
5. When there are no viewing sites equipped to receive satellite courses. When local
network sites are not available, otherwise interested participants must either travel to
more distant sites, or otherwise not participate
PART THREE:
What recommendations may solve these identified problems?
1. Training should focus upon the needs of the regulatory agencies implementing the
regulations, while clearly understanding who is the target audience. Different target
audiences have different training needs, (i.e., permit writers vs. compliance staff vs. the
public).
60
-------
Recommendations:
A. EPA should coordinate with STAPPA/ALAPCO and establish an early development stage,
that makes greater use of annual and special interest surveys, to more clearly identify target
audiences and their training needs.
B. These surveys should include annual updates on State/local toxics expenditures, air toxics
monitoring and modeling efforts, emission inventories, exposure assessment efforts, and
special studies. A listing of State and local program contacts should also be updated
annually.
C. STAPPA/ALAPCO has a listing of State air toxics contacts. State and local agency staff
should review these contact lists, to ensure correct staff are listed.
D. Survey the targeted audience well prior to developing the training materials
E. Different State and local regulatory agencies have different goals and training needs.
Surveys should identify these differences
F. Survey STAPPA/ALAPCO, EPA Regions, and Trade Associations to determine what type
of training is needed, and to specifically identify when the training is most needed.
G. Surveys should include long-term strategic goals and these long-term ideas forwarded to
EPA. This will force State and local agencies to actively think and prioritize their training
needs long-term.
H. State and local agencies should actively solicit and prioritize their training needs from their
identified targeted groups, and this information should be forwarded to EPA.
I. Training goals should be identified for specific regulatory standards.
J. Develop written surveys that may be distributed to State and local agencies, that identifies
and prioritizes requisite and required training
K. State and local agencies should conduct their own surveys of the agency staff, to determine
their agencies' training needs.
L. Identified training needs could be shared through the use of regional air toxics conference
calls.
M. Newly released educational materials should be screened early after receiving it, to insure it
is useful for targeted groups
61
-------
N. StatesAocal programs should be offered the opportunity to participate in the development
of educational material.
2. Many of the video materials available to State and local agencies are out-of-date.
Recommendations:
A. A quarterly update on all current available EPA training materials should be provided to
primary State and local agency air toxics contacts.
B. Primary air toxics coordinators should be contacted directly by APTI, and informed when
this information is available on EPA's Unified Web Site.
C. All video and written materials provided by EPA to State and local programs should
include dates of publication, and a brief synopsis.
D. The EPA should develop new training materials as soon after State/local surveys are
evaluated and significant training needs are identified
E. APTI videos should not be repetitive, but should reflect current policies and procedures.
F. Remember... it is difficult for State and local agency staff to provide industry with training,
until State and local program staff first are provided a clear understanding themselves.
3. Make EPA training and outreach materials more accessible.
Recommendations:
A. Provide an on-line clearinghouse of all available course materials, including video tapes,
training courses, and other training resource materials, by accessing the Unified Air Toxics
Web Site.
B. Continually (or quarterly) update course materials on-line, as these materials become
available, so that most current training materials are always up-to-date.
C. Separate training materials by discipline or MACT standards, to more easily locate
resource materials.
D. The EPA and STAPPA/ALAPCO should contact State and local air toxics staff contacts
directly, as opposed to strictly through State/local educational coordinators.
62
-------
E. Regulatory agencies should solicit input from trade associations and small business
assistance groups. Training workshops can then be co-sponsored by all three participating
groups.
F. Provide a means whereby training materials already developed by State and local programs
may be shared electronically on the Unified Web Site.
G. Provide an easily identified on-line heading (i.e., "air toxics education"), so that on-line
browsers may more easily locate educational training materials.
H. Greater publicize the Unified Air Toxics Web Sites. Promote its use among State and local
air toxics contacts, and among State and local air directors.
I. For Regional EPA updates, State and local agencies should greater utilize EPA's Regional
web sites. Provide regional EPA contacts with recommendations, suggesting how their
region may improve their EPA web sites.
J. Make better use of fiber-optic satellite feeds, to involve government agencies, regulated
facilities and the general public.
K. To improve training attendance, EPA and State sponsored informational meetings should
require minimal long distance travel
L. Class workshops may be improved by supplementing class materials with important
instructional video tape materials.
M. In developing effective workshops, descriptive video materials should be included in
workshops.
N Use more conference calls. It is cheap and convenient.
O. Workshops are not commonly well attended by the public. Suggestions are needed to
improve public participation. Use State/federal advertising and marketing resources?
P. Send important notices to all air toxics contacts, using a list server (i.e., similar to Federal
Registrar notices, and Mobile Source e-mailings).
Q. Link other good web sites to the Unified Air Toxics Web Site, for easy access.
R. The Unified Air Toxics Web Site should have a comments section, for agency staff to
comment upon education materials, and to make further suggestions.
63
-------
4. Publications such as quarterly newsletters, fiber-optic television links throughout the state
(for state specific education and outreach programs), and informational brochures mailed to
regulated facilities to aid in MACT compliance, are helpful in educating regulated facilities.
Recommendations:
A. Newsletters may be used to educate and provide all types of compliance standard
information to regulated facilities.
B. In order to more effectively solicit comment from the public and from regulated facilities,
additional marketing and advertising promotions may be necessary.
C. EPA written materials can be modified and simplified by State/local agencies for State
agency instructional purposes (i.e., brochures and newsletters).
D. Publications should be clearly written, in shortened form.
E. Do not include needless language and too great of detail.
F. If materials are too long or detailed, often regulated facilities do not bother reading it, and
it is all wasted effort.
G. Use "catchy and colorful graphics" that bring attention to the publication or brochure, and
prioritize the ideas that you wish to relay to the reader.
H. In addition to regulator news, newsletters should also include health risk data based upon
health studies, that draws the reader further into the article.
I. To reduce duplication of efforts, EPA and State agencies should contact each other when
significant communications are forwarded to regulated facilities.
J. Video training should not include extraneous details
64
-------
5. Small Business Assistance Groups are often helpful in providing confidential, on-site
assistance with regulatory compliance and educational materials to regulated facilities.
Recommendations:
A. Examples of the nation's most highly effective programs should be summarized and shared
nationally.
B. STAPPA/ALAPCO should survey how many small business assistance groups are currently
active nationwide. Investigate their effectiveness.
C. Small Business Assistance Programs should be expanded, by increased funding of these
sources, and by further sharing of technical ideas between regulated facilities and
government agencies.
D Alternatives to directly interacting with government regulatory agencies Regulated
facilities often call upon Small Business Assistance Groups, rather than interacting with
government agencies. Improved compliance of MACT standards.
6. A Plain English Guidance Document is needed to instruct agencies on how to effectively
collect, analyze, and present data.
Recommendations:
A. Statistical analysis training is needed
B. More training opportunities need to be provided on the Education and Outreach home
page.
C. Risk Assessment More training is needed for Training the Trainers target groups
D. Training materials should use non-technical instruction, whenever possible, to keep training
as simple as possible and easy to understand
7. Make training materials and specifically satellite/video training more interactive.
Recommendations:
A. Question and answer format.
B. Provide additional MACT standards satellite/ video training. One MACT standard per
video.
65
-------
C. When possible, film on-site at regulated facilities, to demonstrate how to perform required
MACT compliance activities.
D. Make video presentations interesting, by using state-of-the-art, color graphics.
4.0 Conclusions
In general, four major themes appeared for discussion during the workshop. These include the
strategy of the Air Toxics Program, personnel training, guidance on proper use of the tools, and
technical and budgetary resources. A summary of the major themes identified through pre-
workshop panel discussion and discussion at the workshop is as follows:
1. An overall strategy for dealing with air toxics issues needs to be developed and technical
tools should be designed to fit this strategy.
• The overall air toxics strategy should include how the various activities within Section
112 relate to each other, such as MACTs, NESHAPs, Urban Air Toxics Program, etc. If
these activities can share some of the same data (e.g., inventory and monitoring data),
rather than having to recreate these data for each application, duplication of efforts could
be reduced and the technical tools can be used more effectively.
• 'The different technical disciplines need to be made more aware of each other's needs.
Air toxics programs should be designed to promote closer working relationships and
technical tools should be structured to promote the sharing of air toxics data (i.e.,
monitoring and inventory data should be provided in a format that can be used in
dispersion modeling, which, in turn, should be able to be input easily in exposure
models). There is a need for a closer working relationship between exposure modelers
and the risk assessors at the Regional and State levels. The exposure modelers are
developing algorithms that are scientifically sound, but which may not address the
specific questions that need to be answered by the decision-makers at the State and local
levels.
2. Training should continue to be recognized as a critical element of the Air Toxics Program.
• State and local agency personnel need to be advised of the availability of training
materials and courses. A clearinghouse of training materials should be accessible to
agency personnel and those materials should be kept up-to-date
• Alternative forms of training, such as internet courses, should be provided for situations
where funding is limited and travel cannot be performed.
• Training materials should be kept as easy to understand as possible. Moreover, training
does not mean issuing guidance documents and policy memos, but is a thoughtful process
66
-------
that involves identifying audiences, developing learning objectives and measurement
methods, and determining suitable delivery systems.
• Training must be included as a vital part of the Air Toxics Program and should be built
into the rule development process in the early stages. Rule developers must recognize
that rule implementation is at least as important, if not more important, than rule
promulgation, and training is essential to the implementation.
• Air toxics is recognized as a priority program and the need for additional training
materials for this program is recognized. In FY98, additional training, i.e., MACT-
specific training, is planned.
3. Guidance on the availability, purpose, and use of air toxics technical tools is needed.
• On a new or existing website, provide an EPA-approved list of exposure models, how
and when to use them, EPA expert contacts, and suggested default inputs.
• Specific guidance on assessing HAPs is needed, such as guidance on emission inventories
for point, area, and mobile sources. The role of the Emission Inventory Improvement
Program (EIIP) should be expanded to specifically address HAPs and guidance and tools
from State and local agencies should be obtained. For example, a plain English guidance
document is needed to instruct agencies on how to effectively collect, analyze, and
present data.
• All publications should be clearly written, in shortened form, and should not include
needless language and detail.
4. As Federal, State, and local governments continue to downsize and are asked to do more
with less, innovative air toxics monitoring tools and procedures need to be developed.
• Opportunities, such as this workshop, should be provided for agency personnel to meet
each other and develop contacts for various technical issues
• A contacts list for technical areas within air toxics would be helpful to promote
interaction between the various technical disciplines. A list of workshop participants'
areas of expertise was developed at the workshop, but it only includes workshop
participants. A much more comprehensive list is needed to provide agency personnel
with a quick reference for technical questions.
In addition to the general themes that were discussed at the workshop, specific issues within the
panel subject areas were identified. The major issues for the panel areas include the following:
67
-------
• Monitoring and Data Analysis: More monitoring data are needed to define the community
health concerns caused by air toxic pollutants. In many instances, monitoring has been
performed, but the data from the States performing the monitoring is not centrally located
and available for analysis. In those cases, the data needs to be compiled for future use. For
many pollutants, monitoring has not been previously performed or has not been adequately
performed, and additional monitoring is needed.
Sampling and analytical methods should be completed for monitoring pollutants where
methods have not been previously available, and outdated methods should be updated. For
example, the update of EPA's Compendia of Methods for monitoring organic and inorganic
pollutants in air should be completed. These compendia are needed to promote uniformity in
monitoring procedures from state to state. The entire revised compendiums should be widely
distributed for ease in locating information.
As monitoring methods become available for pollutants not previously monitored, analysis
methods and tools must be developed to organize and analyze the data. Projects such as the
Ambient Air Quality Characterization Project help to address needs for new analysis tools.
Emission Inventories: The overall need in the emission inventory subject area is for more
guidance on including mobile, area, and point sources to provide a complete inventory.
Upgraded tools/software are needed, as well as recommended procedures for applying the
tools for area, mobile, and point sources. There is a need to review the EPA Consolidated
Emission Inventory Reporting initiative to clarify what emission inventory responsibilities will
be regarding FIAPs as opposed to Toxic Release Inventory (TRI) reporting requirements. The
process of meeting inventory reporting requirements needs to be streamlined to reduce the
burden of the reporting requirements. Inventories should be consolidated where possible to
reduce duplicate data collection and reporting. Overall guidance is needed on:
• Estimating toxic emissions and uncertainty levels from highway vehicles, nonroad
equipment, aircraft, commercial marine engines, and locomotives
• How to do a point, area, on-road and nonroad mobile sources, and biogenic toxics
emissions inventory.
• Dispersion Modeling: In general, existing technical tools are adequate, however, more
guidance is needed on how to use the tools. Tools are needed to address new issues that
result from modeling air toxics. For example, better algorithms are needed to characterize
wet deposition, particle/gas partitioning, gas scavenging coefficient and particle size
distribution, etc. More technical guidance is needed on using the models to help promote
consistency in applying the models.
68
-------
• Exposure Assessment: EPA needs to develop better tools to address exposure assessment
needs. Guidance is needed for all personnel on the steps involved in performing exposure
assessments. Specific areas that the tools need to consider include:
• Assessing the exposure through the inhalation pathway versus all routes of exposure;
• Assessing exposure to humans versus including exposure to animals and plants; and,
• Assessing exposure in urban settings.
The workshop succeeded in bringing together many technical disciplines to voice opinions on
issues relating to areas involving air toxics assessment. Sixteen percent of workshop participants
reported that their primary job function is monitoring and data analysis, 32 percent were dispersion
modelers, 20 percent perform emission inventory and permitting work, and 32 percent were
classified as environmental engineers The technical disciplines gained information regarding how
their work affects work done in other areas. Participants gained knowledge of the many technical
tools available to them that may not have known about prior to the workshop. Participants also
provided EPA with information by defining their needs for the current tools and for future efforts.
On the final day of the workshop, participants completed a workshop evaluation form expressing
opinions about the workshop. Participants indicated that the greatest benefit of the workshop was
learning who the experts are in various technical areas to help promote interaction between
technical disciplines. Participants wanted to learn basic information about all of the disciplines so
that they could see how their work fits into the overall air toxics approach. Those speakers
presenting new information (e.g., the dispersion model AERMOD) were more helpful because they
gave an idea of the current activities and what is coming, relating to air toxics, in the near future.
Many comments were received stating that participants were eager to learn more about performing
exposure assessments and how EPA plans to approach exposure assessment in the future. They
felt that the workshop was helpful in informing personnel of the availability of technical tools. For
example, prior to the workshop, panel members developed lists of tools that would be helpful in
performing their work. During the workshop, they learned that tools were already available to do
many of the things they suggested.
Participants stated that they felt that the workshop was helpful to open lines of communication
between States and EPA to help get agency personnel more involved and interested. They felt that
the workshop was organized and productive, and was successful in meeting the goals of the
workshop. The participants stated that it was helpful to have groups working on developing issues
for discussion prior to the workshop. It was suggested that the workshop become an ongoing
workshop/training session to help promote interaction between disciplines and to help keep
personnel informed of the technical tools that are available to them.
The Workshop Evaluation Forms also allowed participants to make suggestions for future
workshops. A summary of the major points identified for consideration in future workshops
includes:
69
-------
The duration of this workshop (three days) was adequate to address the issues.
Screen workshop speakers prior to the workshop to ensure that the presentation material is
relevant to the purpose of the workshop.
Provide a complete, good quality copy of the overheads used by presenters. All slides should
be provided to participants and the copies of the slides should be readable.
• Hold conference calls with panel members early in the planning process.
• Hold an ice breaker meeting to facilitate participants meeting each other. Some type of
activity, other than during workshop hours, would provide for a more informal atmosphere
that would promote interaction with other attendees. Ice breakers activities could include a
social hour at the hotel the night before the meeting or a ball game.
• Include fewer, but more detailed, presentations in the workshop. Do not use technical
jargon in the presentations that most participants will not understand.
• State and local participants want more structure in the panel discussion. The panel
discussion during the conference calls prior to the workshop were intended to be a free
exchange of ideas relating to the various technical areas. It was suggested that more of a
structured approach to these discussions may have been more productive. A list of specific
questions for each technical area may have helped to spark discussion.
70
-------
Appendix A
Letter of Invitation
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK. NC 27711
MAR " J -5W" A« QUALITY PLANNING
""^ " ** *••<•' AND STANDARDS
MEMORANDUM
SUBJECT: Invitation to a National Specialty Workshop on
Technical Tools for Air Toxics Assessment
FROM: Joe Touma, Meteorologist J"^ '
Air Quality Modeling Group, EMAD (MD-14)
Dennis Doll, Meteor o log ist
Air Quality Modeling Group, EMAD (MD-14)
TO: Regional Air Toxics Coordinators (Region I-X)
Regional Modeling Contacts (Region I-X)
Regional Emissions Inventory Contacts (Region I-X)
Regional Toxics Monitoring Contacts (Region I-X)
We have previously explored with you the feasibility of
conducting a specialty workshop on technical tools for air toxics
assessment. Your response was generally favorable. We are
planning on having the Workshop at Research Triangle Park, NC, on
June 17-19, 1997. Attendance is open to Regional, State and
local agency staff. The workshop is designed for senior staff
environmental scientists, engineers, chemists, meteorologists and
toxicologists with experience in the topics covered in the
workshop.
Attached is a draft agenda for this workshop. In order to
plan for a successful workshop, we need full participation from
all attendees. The following are some of the parameters:
1. The workshop will be limited to 50 participants. Names will
be selected in the order in which they are received; however
we will strive to achieve a balance among the Regions,
States and local agencies and among those with experience in
the topic areas of the workshop.
2. A goal of the workshop is to include reports from the
Regional Offices, States and local agencies on internal air
toxics initiatives. Attendees are expected to participate
in at least one of the following activities: a) as speakers
providing Regional, state and local perspectives on either
monitoring, emission inventories, modeling or exposure
analysis tools; b) as panel members on sessions dealing with
these tools; c) as speakers providing case histories of
multi pathway analyses efforts, or d) as participants for a
focus group, described later. Presenters are responsible
A-l
-------
for preparing hard copies of their presentation materials.
Panel members are expected to meet at the end of the day to
prepare a summary report. Attendees should indicate their
preference (s) within 30 days of receipt of this letter, and
will be assigned based on their preferences.
J. Once panel members have been identified, several
teleconference calls will be scheduled to develop draft
issue papers for discussion during the conference. Panel
members are expected to coordinate with other experts in
their groups on the development of these issue papers.
4. The presentation materials, as well as panel summary reports
will be published in workshop proceedings. Certain segments
may be video-taped for use in a future APDLN telecast.
5. The workshop will be held in the EPA's Administration
Auditorium on 79 Alexander Drive, RTP (see map attached).
Attendees are responsible for all travel arrangements and
accommodations. A packet containing a list of hotels in the
area can be provided. If you wish to car pool locally,
please indicate on the registration form and your name will
be distributed.
6. The workshop spans three full days (Tuesday, Wednesday and
Thursday). In order to have a successful workshop, we ask
that all participants commit to staying for the duration of
the workshop and plan for traveling on Monday and Friday.
7. An optional one-half day can be arranged for those
requesting hands-on computer training on technical tools in
using emissions factors to develop emissions inventories
with exercises in Air CHIEF and TANKS, or modeling (TSCREEN
and Exlnter models). In order to assure availability of
facilities and course instructors, at least 5 people are
needed to enroll in each session.
8. In addition to discussing the issues addressed in this
workshop, OAQPS also invites approximately 10 non-federal
attendees to participate in a focus group to aid in the
evaluation of the national air toxics problem. The purpose
of this focus group is to clarify the perspectives of
stakeholders on the issue of defining an air toxics problem
by addressing such questions as:
- What characterizes air toxics emissions, ambient
concentrations and deposition as "a perceived problem"? Is
there a consensus on whether there is one? At what scale?
- What is an acceptable level of uncertainty in the emission
estimates? Ambient measurements? Deposition estimates?
Exposure estimates? Health effects data?
- Given probable action at some level of uncertainty, and
the need for reducing uncertainties where possible, for what
measures (emissions, health benchmarks, etc) are
uncertainties acceptable, and where should efforts/resources
be concentrated to reduce uncertainties?
A-2
-------
The results of the focus group discussions will, we believe,
provide a valuable perspective, given the important issues of
data availability and uncertainty. Group discussions will
require participation through teleconferencing before the
workshop itself. Non-federal attendees interested in
participating in this focus group should indicate their interest
on the registration form and whether they are interested in
meeting in the evening or an optional one-half day session on
Friday morning.
Please circulate this invitation to your respective State
and local agency contacts. A final agenda including the topics
and assigned names will be mailed by April 30. To ensure
seating, participants should send their registration forms within
30 days to: Joe Touma or Dennis Doll< USEPA, OAQPS (MD-14),
Research Triangle Park, NC 27711. A copy should be sent also to
the Regional contact.
If you have any further questions or comments concerning
this speciality workshop please contact either of us at:
Joe Touma (919) 541-5381; E-roail: touma.joe@epamail.epa.gov
Dennis Doll (919)541-5693; E-mail:doll.dennis@epamail.epa.gov
Attachment
cc: F. Dimmick (MD-13)
D. Guinnup (MD-14)
B. Hunt (MD-14)
J. Irwin (MD-14)
B. Jordan (MD-13)
D. Misenheimer (MD-14)
D. Mobley (MD-14)
D. Pagano (MD-15)
S. Shaver (MD-15)
J. Tikvart (MD-14)
Workgroup Members
A-3
-------
Appendix B
Workshop Agenda
-------
Workshop Agenda
National Specialty Workshop on
Technical Tools for Air Toxics Assessment
June 17-19,1997
79 Alexander Drive, Administration Bldg. Auditorium,
Research Triangle Park, North Carolina
A workshop on the technical tools developed by OAQPS, Regional Offices, State and local
agencies for meeting the goal of the air toxics program: a discussion of issues, initiatives,
adequacy, limitations, tool needs and data gaps, and future direction.
Target Audience: Regional Offices, State and Local Agency Technical Personnel
Workshop Size: 50 people
Workshop Mission Statement: Regional and State contacts have expressed an interest in
learning about the technical tools required to support many sections of the Clean Air Act
regarding the air toxics program so that:
1. They can better support requirements under Title m such as the newly published Risk
Management Plans under Section 112(r), residual risk and de-listing requests under
MACT rules, the Urban Area Source Study, the Great Waters Studies, etc.
2. They know who are the air toxics topic area experts (office level identification) across
EPA, and the States and local agencies in order to obtain rapid assistance.
3. They can better communicate the magnitude of air toxics issues facing the technical
analyst and the success in solving air toxic problems.
This interactive workshop includes half-day sessions devoted to Emission Inventories,
Dispersion Modeling, Ambient Monitoring and Data Analysis, Exposure Assessment Case
Histories, and Education & Outreach as related to supporting the air toxics program. The
sessions will be interactive in order to obtain feedback from all attendees, rather than one way
OAQPS presentations.
Input from the participants is important to provide direction to air toxics technical needs and to
obtain assistance from the State and local agency partners in an era of reduced resources.
B-l
-------
Tuesday June 17
7-8 am Workshop Registration
8-9 am Welcome and Announcements, 30 minutes, Joe Touma, Dennis Doll
Introductory Remarks- Bill Hunt, 30 minutes
9 am -12 noon: Elements of the OAQPS Air Toxics Program
Framing the discussion issues in this segment:
7. An Overview of Section 112(r), MACT rule, the Urban Air Toxics Study, and the Great
Waters Study, etc.
2. What technical tools are needed in support of these rules?
3. Who are the agency experts in air toxics, how can I get a hold of them?
4. Overview of strengths and weaknesses in the technical tools available for doing all
adequate technical analysis now and in the future.
Speakers:
9:00-9:30 The Clean Air Act Framework for the Air Toxics Regulations. Al Wehe, 30
minutes
9:30-10:00 Learn About How EPA Plans to Bring Together the Federal Air Toxics
Program (The OATS Report). Al Wehe, 30 minutes
10:00-10:30 How Is the Air Toxics Program Structured in OAQPS and Who Has What
Functions? Melissa McCullough, 30 minutes
Panel Discussion: Regional/State/local Perspectives (90 minutes)
Chair: Joann Held (New Jersey)
1. Tom Rogers (FL): Should Offsite Consequence Analysis of Accidental Releases be
Incorporated into a Permitting Program?
2. Denis Lohman (Reg 3): Examples of Situations In Which Risk Is Communicated to the
Public (e.g. Superfund Remedial Actions; Place-Based Studies; Proposed Permits; and
Incident Analysis).
3. Lee Page (Reg 4): Emission Inventory Tools: Present and Future
4. Mike Pokorny (MD): Combining Modeling & Inventory Tools
5. Linda Lay (EPA OECA): Using Tools for Targeting Sources
6. Joann Held (NJ): Risk Screening Tools
12-1: Lunch Break
B-2
-------
1 pm- 5 pm: Ambient Air Toxics Monitoring and Data Analysis
Framing the discussion issues in this segment:
1. Why perform ambient air monitoring for air toxics ?
2. What are some of the ambient air monitoring methods and costs for air toxics?
3. How can I use monitoring data from a subset of air toxic pollutants (such as the PAMS
network) in my analyses?
4. What ambient air toxics monitoring data are available ? Where can I find them ? What
current summaries and analyses are available?
Speakers:
1:00-1:30 Compendium Methods for Monitoring Organic and Inorganic Compounds. Bill
McClenny, 30 minutes
1:30-1:45 State and Local Participatory Programs: Volatile Organic Toxic Compounds,
Including Speciated NMOC, UATMP, Carbonyls, and PAMS. Dave-Paul
Dayton, 15 minutes
1:45-2:00 Paniculate Compounds Including PM10, PM2.5, and Filter Analysis for Toxic
Species. Stan Sleva, 15 minutes.
2:00-2:30 Video Demonstration of Ambient Sampling and Analytical Techniques for
Hurricane Fran Debris Burn Site. Neil Berg, 30 minutes
2:30-3:00 Chemical Mass Balance for Toxic Species (VOC, and F'articulates). Chuck
Lewis, 30 minutes.
3:15-3:50 New Data Analysis Tools. James Hemby, 35 min
Panel Discussion: (45 minutes)
Chair: Thomas Shoens (Wayne County)
Panel Members: Peter Kahn (Region 1)
Todd Jackson (W. Virginia)
Victor Guide (Region 3)
5 pm: Adjourn
5-6 pm: Panel members only: Ambient Air Toxics Monitoring and Data Analysis, final
summary discussion
B-3
-------
Wednesday June 18
8am -12 noon: Air Toxics Emission Inventories
Framing the discussion issues in this segment:
1. What are potential air toxic pollutants for urban, rural, and mobile sources ? What are
typical sources, and source categories?
2. Are emission factors available for all air toxics pollutants ? What do I do for sources
that do not have emission factors?
3. Where is the information that allows me to compute the total emissions from sources in
rural areas, urban areas, mobile sources, fugitive sources, etc.
4. Where are the emission tools that I can use? What pollutants do they cover? What is
the difference between EPA tools and those from other organizations.
What are the new procedures to collect and report emissions data ?
5.
Speakers:
8:00-9:00
9:00-10:00
10:10-11:00
Introduction- National Air Toxics Emission Inventory. Anne Pope, 60 minutes
Emission Inventory Improvement Program (EIIP). Steve Bromberg, 60 minutes
Emission Estimation: Techniques and Tools. Guidance and Assistance Tools
from the Emission Factors and Inventory Group. Anne Pope, Mary Ann
Barckhoff, 50 minutes
Panel Discussion: (70 minutes) 11:00-12:10
Chair:
Panel Members:
Mike Fishburn (Texas)
Jennifer DeMay (Washington)
Bob Ragland (Forsyth County, NC)
Diane Scher (Tennessee)
Paul Cocca (EPA OW)
Richard Cook (EPA QMS)
Craig Bresson (Lane Regional)
Brian Haugstad (Iowa)
Jim Lefik (Allegheny County)
Mohamed Basher (Allegheny County)
12-1: Lunch Break
B-4
-------
1-5 pm: Dispersion Modeling
Framing the discussion issues in this segment:
1. What dispersion models are appropriate for use in air toxic assessment? What are the
advantages and disadvantages of each model?
2. What data do I need to run these models?
3. What is a screening study and when can I use a screening vs a refined model?
4. When do I use EPA models vs other types ?
5. What is the averaging time to be considered for air toxics?
6. Where can I get the meteorological data to run these models?
7. Do I need to consider chemical transformation in the dispersion models ? If so, are the
same models appropriate?
8. How do I model sources with variable emission rates?
Speakers:
1:10-1:25 Overall Perspective of Air Toxics Modeling Issues-Program Management
Issues. Joe Tikvart, 15 minutes
1:25-2:10 Introduction to Issues Typically Encountered in Modeling. John Irwin, 45
minutes
2:10-2:25 Meteorological Data Needs and Electronic Bulletin Board Systems. Dennis
Atkinson, 15 minutes
2:30-3:00 AERMOD, the Next Generation Air Dispersion Model. Russ Lee, 45 minutes
3:00-3:25 Dioxin CMB Modeling. Andrew Roth (Dayton), 25 minutes
3:35-3:55 Practical Problems in Combustion Air Modeling. Steven Ehlers, Region 6, 20
minutes
4:00-4:15 New Jersey DEP Risk Assessment Procedures for Minor Sources. Alan
Dresser, 15 minutes
Panel Discussion: (45 minutes) 4:15-5:00
Chair: Daniel Wise (Rhode Island)
Panel Members: Patrick Pakunpanya (Louisiana)
Jeffrey Sprague (Illinois)
Alan Dresser (New Jersey)
Henry Feingersh (Region 2)
Samuel Bell (Jefferson County)
Dom Ruggeri (Texas)
Steve Kish (Michigan)
5:00 pm: Adjourn
5-6 pm: Panel members only: Air Toxics Emissions Inventories, Dispersion Modeling, final
summary discussions
B-5
-------
Thursday June 19
8 am -11:30 am: Exposure Assessment Techniques
Framing the discussion issues in this segment:
1. What are the components of exposure models?
2. What types of exposure assessment models are available ?
3. Can additive exposures be accounted for?
4. What is the role of CIS is undertaking exposure and risk assessments ?
5. Inhalation versus multi-media exposure.
6. Addressing uncertainty and variability.
Speakers:
8:05-8:55 Exposure modeling: Components, Status, and Uses. Michael Zelenka (NERL),
50 minutes
8:55-9:30 Micro environmental Modeling Issues. Alan Huber (NERL), 35 minutes
9:30-9:55 Comparing Cumulative Annual Dose Estimates from pNEM to "Literature "
Exposure Factors. Thomas McCurdy (NERL), 25 minutes
9:55-10:20 Real-Time Estimation ofPolycyclic Aromatic Hydrocarbons. Nancy Wilson
(NERL), 25 minutes
10:20-10:40 Minnesota Air Toxics Indexing System. Greg Pratt, Minnesota, 20 minutes
Panel Discussion: (45 minutes) 10:45-11:30
Chair: Aimee Kennedy (W. Virginia)
Panel Members: Jerry Ebersole (Oregon)
Jeffrey Hayward (North Carolina)
Jim Snyder (EPA Region 8)
Greg Pratt (Minnesota)
Daniel Gray (Kentucky)
ll:30am -1:30 pm: Case Histories of Multi-pathway Modeling Efforts
12-1: Lunch Break
Framing the discussion issues in this segment:
7. What recent studies were performed by EPA Headquarters, Regional Offices, and
State and local agencies. What are the lessons learned?
2. What role did indirect exposure assessments have in the evaluation of hazardous waste
incineration and other site-specific assessments in recent years? What are the lessons
learned from the WTI incinerator assessment in Region V?
3. Examples of visualization techniques to best convey results?
B-6
-------
Speakers:
11:30-12:00 The Mercury study; an Integrated Risk assessment. Martha Keating, 30
minutes
1:00-1:30 The Screening Risk Assessment Evaluation for the Proposed Incineration
Project at the Anniston Army Depot Chemical Demilitarization Facility. Leigh
Bacon, 30 minutes
1:30 pm-3:30 pm: Education and Outreach
Framing the discussion issues in this segment:
1. Availability of EPA training and resource materials, where do I go for more
information ?
2. What are the reference links for information and training?
3. Other sources of training ?
4. Communication/Outreach Tools (e.g. INTERNET, etc.); etc.
5. Training Needs in Air Toxics
Speakers:
1:35-2:25 Training Opportunities in Air Toxics for You through the Education and
Outreach Group. Howard Wright, 50 minutes
2:25-2:35 The Unified Air Toxics Web Site. Nancy Pate, 10 minutes
Panel Discussion: (45 minutes) 2:35-3:20
Chair: Brian Haugstad (Iowa)
Panel Members: Ellen Morris (Connecticut)
Samuel Bell (Jefferson County)
Leigh Bacon (Alabama)
Jimmy Johnston (Georgia)
3:20 pm: Summary, next steps
3:30 pm: Workshop Adjournment
B-7
-------
Appendix C
Attendance List
-------
NATIONAL SPECIALTY WORKSHOP ON TECHNICAL TOOLS FOR
AIR TOXICS ASSESSMENT
June 17-20,1997
Name
Angela Andrews
Dennis Atkinson
Leigh Bacon
Mary Ann Barckoff
Mohamed N. Basher
Samuel L. Bell
Neil Berg
Craig Bressan
Steve Bromberg
Charles Buckler
Paul Cocca
Richard Cook
Dave-Paul Dayton
Jennifer DeMay
Mark Derf
Jim Dicke
Dennis Doll
Agency
US EPA
US EPA
ALDEM
Lockheed Martin
Allegheny County
Jefferson County
US EPA
LRAPA
US EPA
NCDAQ
EPA/OW/OST/SASD
EPA QMS
ERG
Washington
IN DEM
OAQPS
US EPA
Phone
919-541-4565
919-541-0518
334-271-7861
412-578-8112
205-930-1366
919-541-5520
541-726-2514
919-541-1000
202-260-8614
313-741-7827
919-461-1214
360-407-6825
317-232-8449
919-541-0083
919-541-5693
Fax
334-279-3044
412-578-8144
205-939-3019
541-726-1205
202-260-9830
313-741-7439
919-461-1579
360407-6802
317-233-5967
Email
Ibb @adem.state.al.us
dns@city-net.com
jcdh @bham.mindspring.com
lrapa@worldnet,att.net
cocca.pauJ@epamaiJ.epa.gov
cook.rich @epamail.epa.gov
ddayton@erg.com
jdem461 @ecy. wa.gov
aynos@opn.dem.state.in.us
Panel Number
Six
Three
Four, Six
Three
Three
Three
Three
C-l
-------
Name
Alan Dresser
Laurel Driver
Jerry Ebersole
Steven Ehlers
Spence Erickson
Henry Feingersh
Mike Fishburn
William G. Gillespie
Daniel J. Gray
Victor Guide
Brian Haugstad
Jeffrey J. Hayward
Joann Held
James Hemby
Alan Huber
Ann Ingram
John Irwin
Todd Jackson
Jimmy Johnston
Peter R. Kahn
Agency
NJDEP
US EPA
ORDEQ
US EPA
ORDEQ
Region 2
TNRCC
DC Air Resources
KYDAQ
Region 3
IADNR
NCDEHNRAQ
NJDEP
US EPA
NERL
US EPA
US EPA
WVDEP
GAEPD
Region 1
Phone
609-633-2675
919-541-2859
503-229-6974
214-665-8312
503-229-6458
212-637-3382
512-239-1934
202-645-6091
502-573-3382
215-566-2733
515-281-4927
919-733-1475
609-633-1113
919-541-5459
919-541-1338
919-541-1584
919-541-5682
304-558-4022
404-363-7127
617-860-4392
Fax
609-292-7793
503-229-5675
214-665-6762
503-229-5675
212-637-3901
512-239-1515
202-645-6102
502-573-3787
515-242-5094
919-733-1812
609-292-7793
304-558-3287
404-363-7100
617-860-4397
Email
adresser@dep.state.nj
gerald.ebersole@state.or.us
ehlers.steven@epamail.epa.gov
spence.l.erickson@state.or.us
feingersh.henry@epamail.epa.gov
mfishbur @tnrcc .state, tx.us
gray@mail.nr.state.ky.us
guide.victor@epamail.epa.gov
bhaugst@max.state.ia.us
jeffjiayward@aq.ehnr.state.nc.us
jheld@dep.state.nj.us
jimmyjohnston@mail.dnr.state.ga.us
kahn.peter@epamail.epa.gov
Panel Number
Four
Five
Four
Three
Five
Two
Three, Six
Five
One
Two
Six
Two
C-2
-------
Name
Martha Keating
Aimee C. Kennedy
Sieve Kish
Linda Lay
Russ Lee
James A. Lefik
Chuck Lewis
Denis M. Lohman
Ken McBee
Bill McClenny
Melissa McCullough
Thomas McCurdy
Ray Merrill
Evelina C. Morales
Ellen P. Morris, PhD.
Donnette Olds
Lee Page
Sirisak P. Pakunpanya
Nancy Pate
Michael Pokorny
Agency
US EPA
WVDEP
MIDEQ
USEPA/OECA
US EPA
Allegheny County
US EPA
Region 3
VADEQ
US EPA
US EPA
NERL
ERG
OKDEQ
CTDEP
OAQPS
Region 4
LADEQ
ITPID
MDDEP
Phone
919-541-5340
304-558-1213
517-335-4794
202-564-8577
919-541-5638
412-578-8132
919-541-3154
215-566-2192
804-698-4024
919-541-3158
919-541-5646
919-541-5774
919^61-1236
405-962-2206
860^24-3412
404-562-9131
504-764-0202
919-541-5347
410-631-3232
Fax
304-558-1222
517-335-3122
202-564-0068
412-578-8144
215-566-2124
804-698-4510
919-541-0942
919-461-1579
405-962-2200
860^24-4063
404-562-9095
504-765-0222
919-541-4028
410-631-3202
Email
kish @deq.state.mi.us
lay.linda@eparnail.epa.gov
achd_aq_eq @compuserve.com
lohman.denny@epamail.epa.gov
klmcbee@deq.state.va.us
mccullough.melissa@epamail.epa.gov
rmerrill@erg.com
evelina.morales @ oklaosf.state.ok.us
ellen.morris @po.state.ct.us
page.lee@epamail.epa.gov
patrickp@deq.state.la.us
pate.nancy@epamail.epa.gov
Panel Number
Five
Four
One
Three
One
Six
One
Four
One
C-3
-------
Name
Anne Pope
Greg Pratt
Bob Ragland
Joanne Rice
Tom Rogers
Andrew J. Roth
Dom Ruggeri
Diane Scher
Thomas Shoens
Stan Sleva
Jim Snyder
Jeffrey Sprague
William K. Steinmetz
Andrew Stewart
Joe Tikvart
Joe Touma
Stanley C. Tracey
AlWehe
Nancy Wilson
Daniel Wise
Agency
US EPA
MNPCA
Forsyth County
US EPA
FLDEP
RAPCA
TNRCC
TNDEC
Wayne County
TRC
US EPA
ILEPA
NCDEHNR AQ
WIDNR
US EPA
US EPA
DC Air Resources
US EPA
NERL
RIDEM
Phone
919-541-5373
612-296-7664
910-727-8060
904-921-9554
937-225-4118
512-239-1508
615-532-9193
313-833-3596
313-312-6689
217-524-4692
919-715-7713
608-266-5499
919-541-5562
919-541-5381
202-645-6093
919-541-5623
919-541-4723
401-277-2808
Fax
612-297-8701
910-727-2777
904-922-6979
937-225-3486
512-239-1123
615-532-0614
313-833-3561
303-312-6065
217-524-4710
919-733-1812
608-267-0560
919-541-0044
202-645-6102
919-541-0942
401-277-2017
Email
gregory.pratt @pca.state.mn .us
raglanre@co.forsyth.nc.us
rogers_t@dep.state.fl.us
rothaj@rapca.org
drugged @ smtpgate.tnrcc.state.tx.us
tshoens @co.wayne.mi.us
snyder.james@epamail.epa.gov
epa2 1 02 @epa.state.il.us
stewaa@dnr .state, wi.us
touma.joe@epamail.epa.gov
wehe.al@epamail.epa.gov
daniel.wise@navc.ci.net
Panel Number
Five
Three
One
Four
Three
Two
Five
Four
Four
C-4
-------
Name
John White
Howard Wright
Michael ZeJenka
Agency
NCDAQ
US EPA
NERL
Phone
919-541-5584
919-541-1326
Fax
Email
Panel Number
C-5
-------
Appendix D
Workshop Presentation Handouts
-------
APPENDIX D
TABLE OF CONTENTS
Bill Hunt, The Role of Technical Tools in the Air Toxics Program D-l
Al Wehe, MACT- Overview of Statute D-13
Al Wehe, Summary ofOAQPSAir Toxics Strategy (OATS) D-21
Melissa McCullough, The Air Toxics Program in OAQPS D-23
Tom Rogers, Should Offsite Consequence Analysis of Accidental Releases be Incorporated
into a Permitting Program? D-27
Linda Lay, Using Tools for Targeting Sources D-35
Bill McClenny, Compendium of Methods for Monitoring Organic and Inorganic
Compounds D-47
Dave-Paul Dayton, State and Local Participatory Programs: Volatile Organic Toxic
Compounds, Including SpedatedNMOC, UATMP, Carbonyls, andPAMS D-99
Stan Sleva, State and Local Participatory Program: Particulate Compounds Including
PM10> PM25, and Filter Analysis for Toxic Species D-ll 1
Neil Berg, Video Demonstration of Ambient Sampling and Analytical Techniques for
Hurricane Fran Debris Bum Site D-125
Dr. Charles Lewis, Chemical Mass Balance for Toxic Species (VOC and Particulate) . D-127
James Hemby, New Data Analysis Tools for Ambient Air Quality Data D-I31
Anne Pope, Introduction- National Air Toxics Emission Inventory D-143
Steve Bromberg, Emission Inventory Improvement Program (EIIP) D-l59
Anne Pope and Mary Ann Barckhoff, Emission Estimation: Techniques and Tools.
Guidance and Assistance Tools from the Emission Factors and Inventory Group D-I61
Mike Fishburn, Presentation Notes for Air Toxics Emission Inventories Panel D-175
Joe Tikvart, Overall Perspective of Air Toxics Modeling Issues - Program Management
Issues D-183
John Irwin, Dispersion Modeling of Pollutant Impacts D-187
-------
APPENDIX D
TABLE OF CONTENTS
Dennis Atkinson, Meteorological Data and Electronic Data Transfer D-201
Andrew Roth, Source Attribution Modeling for Dioxin D-207
Steven Ehlers, Practical Problems in Combustion Air Modeling D-219
Alan Dresser, New Jersey DEP Risk Assessment Procedures for Minor Sources D-221
Michael Zelenka, Exposure Modeling: Components, Status, and Uses D-221
Alan Huber, Micro environmental Modeling Issues D-241
Nancy Wilson, Real-Time Monitoring of Polycyclic Aromatic Hydrocarbons D-247
Greg Pratt, Minnesota Air Toxics Indexing System D-251
Martha Keating, The Mercury Study: An Integrated Risk Assessment D-263
Leigh Bacon, The Screening Risk Assessment Evaluation for the Proposed Incineration
Project at the Anniston Army Depot Chemical Demilitarization Facility D-269
Howard Wright, Training Opportunities for You Through the Education and Outreach
Group D-275
Nancy Pate, The Unified Air Toxics Website D-279
Brian Haugstad, Panel Six Issues and Recommendations D-283
-------
The Role Of Technical Tools In
The Air Toxics Program
National Specialty Workshop on
Technical Tools for Air Toxics
Assessment
William F. Hunt
June 17,1997
PURPOSE
Technical Tools are Designed to
Provide the Infrastructure of the Air
Toxics Program
Today, I Will Discuss Our Infrastructure
Development Activities, Where We Are
Strong, Weak, & Challenges Ahead
I Will Give You Examples Of Some
Success Stories
D-l
-------
Issues With Air Toxics Infrastructure
Development
• No Clear Mandate In CAA For
Infrastructure Development
• However, As You Know, This Is
Needed To Carry-On Several
Mandated Programs: Title V, Urban Air
Toxics, Residual Risk, Etc.
• Funding is Limited: We need
Consensus and Priorities On Needs
• This Workshop, I Hope Begins Building
A Dialogue On All Three Aspects
Emission Inventories & Factors
National, State & Local Air Toxics
Inventories & Emissions Factors Are
Being Developed to Meet Various
Section 112 Requirements
Data Gaps Can Not Be Filled Without
State & Local Agency Assistance
D-2
-------
Total National 188 HAP Emissions (tpy)
by Source Type
,-POINT (25.86%)
MOBILE (39.36%)
AREA (34.78%)
Dispersion Modeling
• Some Of The Available Dispersion
Modeling Tools Are Still Applicable
• However, New Issues Related To
Dense Gases, Particle & Gas
Deposition, Atmospheric Chemical
Transformations Of Toxic Pollutants
Pose New Challenges
D-3
-------
Air Monitoring
Lack Of Technology For Monitoring All
Toxic Pollutants
Inability To Develop QA Procedures
Expense For Monitoring All Pollutants
In All Parts Of The Country
Consider Requiring State Air Toxics
Monitoring Data being Reported to
AIRS
PAMS:
- Site #2 Is Ideal Platform For Extended
Toxic Monitoring
Interpreting Ambient Air
Monitoring Data
An Example from the Urban Air Toxics Program
m Examples of Data Resources
- Emission Inventories
- Fate and transport
- Historical monitoring data
- New/current air monitoring results
D-4
-------
Example of Spatial Variations
1,3-butadiene
• What can the National UATMP program
results say about this compound in
urban environments
-16 monitoring stations in 1995 UATMP
program
ef On miPAHOMnhiihi fflilln
D-5
-------
(MtrtoTiktoMhri
fcr
hta IMS MINT
lUttn ».<•)
i I i i i i I I 1 I 1 I I i I i
Port Neches, Texas -Example
. ATSDR data indicate Texas and Louisiana
as the lead manufacturing states for 1-3
butadiene
• TRI data show nine (9) industrial sources of
1-3 butadiene within 10 miles of the Port
Neches, Texas UATMP Site
• TRI data show decreasing trend in emissions
data from 1990 to 1995
D-6
-------
AlrEafaihu»ri>B«l»dkMwfcU
Ton (nm) Material
PortNKb«f,TX \
*• (W5DATMP) \
MomitoriBC Stedott*
(1990 ud 1991DATMP)
CTW.19M
D-7
-------
Port Neches, Texas -Example
• Site Location is critical
-1990 and 1991 PNTX data showed
• much less 1-3 butadiene than 1995 data
-Why?
• difference in locations of the PNTX
monitoring stations - a major factor
TOdeM
Geometric Mean l>BnUdkne Coictntntioni
Meuutd it Pert Necfce*. TOM
Proflrvin Ynr
1990
1991
1995
Geometric Men l^Butu&ne
^rwir^fi ttmtifui /njJiu\
OJ9
1.11
1.82
NunlMrofVaEd
Stmnln Cnllfctni
29
30
Note AtihowBinFisutc 5-2. the 1995 UATMP mootoriag KHion to PNTX wtj u »(fif&rem
locMioa thu tbc 1990 ud 1991 UATMP moatoang mtion.
D-8
-------
Air Toxics Indicators & Trends
Goals:
- Synthesize Data Collected from Disparate
Programs
- Show National Trends & Patterns
' Challenges:
- Limited Data Bases
- No Consensus on Calculating Relative Risk
• Ongoing Work:
- Government Performance & Results Act-Pilot
- Developing a Strategy to Measure Progress
Towards Reduced Health Risks Resulting from
Programs (e.g., MACT)
- Received Funds to Explore Indicators
Education & Outreach
Expand Opportunities for Education & Outreach
Develop Additional Training on Technical Tools
for Monitoring, Emissions Inventory, Modeling
and Exposure Assessment
Prepare Training Materials
Utilize Multi-Media Training Vehicles Such as
APDLN Telecourses, Internet, CD-ROM,
Classroom Lectures
Workshops that Bring Multi Disciplines Together
to Foster Communication & Information
Exchange
Establish a List of Area Experts to Foster
Quicker Communications and Solutions.
D-9
-------
Exposure Assessment
Continue Developing Guidance & Tools to
Evaluate Multi-Pathway Human Exposures for
Multi-Pollutants
- Must Expand Scientific Capabilities to Measure
Multi-Media Environmental & Human Exposures
& Biological Markers of Exposure
' Need to Characterize Residential Multi-Media
Exposures:
- Quantify Contributions From Indoor & Outdoor
Sources Including Source-Receptor
Relationships
- Understand Relationship Between Exposure,
Dose & Effect
Accomplishments:
Expanding National Air Toxics Emissions
Inventory Data Base
User Friendly Air Dispersion Models
Capable Of Addressing New Concerns
Several Years Of PAMS Monitoring Data;
Easier Ways To Collect Data
Developing Useful Air Quality Indicators
Developing Strategy to Measure
Effectiveness of Air Toxics Program
• Continuing Research on Exposure
Assessment Methodologies
• Expanding Education and Outreach
Opportunities
D-10
-------
Challenges Ahead
Declining Budget and Increasing
Demand for Infrastructure
Development
Effective Partnership With States, Local
Agencies & Other Stakeholders To
Improve All Aspects Of Our Work
Improved Methods For Technology
Transfer Through The Evolving Medium
Of The Internet
D-ll
-------
CLEAN AIR ACT
Air Toxics Provisions
Fred Dimmldt
VS. EPA, OAQPS
dimmick.rred@epamaQ.epa.gov
&EPA:
OUTLINE
History
Statutory Provisions
Basic Requirements
MACT
History
1970 SECTION 112 PROVISIONS
Directed EPA to Identify air pollutants with
hazardous effects and establish standards to prevent
any adverse effects "with >n ample margin of
safely."
In the first two decades under KESHAPS, EPA tcl
standards for only eight pollutants (most recently
benzene in 1989):
Arsenic Beryllium Radon
Asbestos Mercury Vinyl chloride
Benzene Ridionuelides
History
EPA's SOLUTION
• For HAPs that are carcinogenic, EPA deemed that
there v.is no safe level of exposure, to to establish
standards with "ample margin of safety" was
impossible.
• Risk assessment was not formalized in the 1970s, and
when standard procedures became available in the
1980s the data contained significant uncertainty.
EPA began risk assessment studies which take years
to complete.
D-13
-------
History
THE VINYL CHLORIDE CASE
• EPA and EOF mutually agreed u pon a
schedule to develop a zero-risk goal in 1977
• EPA took no action for 8 years
• EPA withdrew rule in 1985 dtlng:
- High costs for add itional VC control
- No technology could reach zero emissions
History
THE VINYL CHLORIDE CASE <«mt-d)
• NRDC sued EPA In 19S5
• Court rejected NRDC's position in 1984
• Remanded back lo EPA for:
. Giving loo much attention lo control technology
• Ko( providing fen ample margin of safely
. EPA must:
1. Determine safe level! without technological or
costs concerns
2. Determine an ample margin of safely (more
stringent, but could consider costs)
Statutory Provisions
TITLE HI HAZARDOUS AIR
POLLUTANTS
[CAA Amendments of 1990]
Sec. 301 Hazardous Air Pollutants (Section 112)
Sec. 302 Conforming Amendment (Bookkeeping)
Sec. 303 Risk Assessment and Management Commission
Sec. 304 Chemical Process Safety Management
Sec 305 Solid Waste Combustion (Section 129)
Ser. 306 Ash Management and Disposal
SECTION 112
Statutory Provisions
Sec. Il»a) Definitions
Stc. 11Kb) List of Pollutants
Sec. IIZ(c) List of Source Categories
Sec. UKd) Emission Standards
Sec. llJ(e) Schedule for Standards and Review
Sec. 11VO Standard to Protect Health and UK Environment
Sec. ll«i) Modifications
Sec. 112(h) Work Practice Standards and Other Requirements
Sec. 112(0 Schedule for Compliance
Sec. U2IJ) Equivalent Emission Limitation by Permit
D-14
-------
Statutory Provisions
SECTION 112 (eonYd)
Sec. 112(k) Arc* Some Program
Sec. 112(1) Suit Frofrim*
Sec. I12(m) AtiMfpherfc Deposition lo Crul Lakes •ml
Coastal Water*
Sec. ll«n) Other Provisions
Sec 112 Eslabl isK M ACT Tor each category
• Expected 75-90% reduction below existing emission
levels
• All standards must be promulgated by No>. 15, 2000
Basic Requirements
189 HAPs
• List found in Section 112(b)(l)
• Has been corrected since initial publication
• List can be revised through Administrator
rc\icw, per Section 112(b)(2) (e.g., caprolactum)
• Any person may petition the Administrator to
modify the list, per Section 112(b)(3)
Basic Requirements
SECTION 112 SOURCE
CATEGORY LIST
• Section U2(c) required list of categories emitting
one or more HAP
EPA published initial list July 16, 1992
- 166 categories of major sources
• i categories of area sources
List can be revised per Section 112
-------
Basic Requirement!
MACT
• National Standards
• Case-bj-C«s« Determinations
• Implementation/Delegation
Basic Requirements
RESIDUAL RISK
• Within 8 to 9 years of a MACT standard,
residual risks must be assessed and EPA must
determine whether more stringent standards
are required to provide an "ample margin of
safety."
• EPA must consider standards for any carcinogen
with >10'4 cancer risk
Basic Requirements
ACCIDENTAL RELEASES
For 100 extremely hazardous air pollutants,
facilities must develop a risk management
plan to address how to deal with any
possible hazards.
Basic Requirements
STUDIES
Great Waters Study
Urban Air Toxics Study & Strategy
NAS Study
Utility Air Toxics Study
Mercury Study
Risk Assessment Study
D-16
-------
ALL 4-YEAR MACTs ARE DONE
Cleafl Air Act -- Section U2 4-Year MACT Standards
ALL 4-YEAR MACTs ARE DONE
Clcaji AirAct - Section 112 4-Year MACTSundards
ALL 4-YEAR MACTs ARE DONE
Qcan Air Act -Section 112 4-Year MACT Standards
ALL 4-YEAR MACTs ARE DONE
Clean Air Act •• Section 112 4-Year MACT Standards
D-17
-------
ALL 4-YEAR MACTs ARE DONE
Clean Air Act -• Section 112 4-Year MACT Standards
MACt •»•*>«
ALL 4-YEAR MACTs ARE DONE
dean Air Act •• Section 112 4-Year MACT Standards
MACT
MACTPROGRAM -
What Does The Future Hold?
• Budget reductions in 1993,1994,1995 and 1996
• Work slowed on most NESHAP due in 1997
Need to address potential "Permit Hammers"
• Potential budget stability in 1997 and 1998
MACT
MACT PARTNERSHIP -
The Basic Approach
• UPFROVT planning and working with all
parties
• LEVERAGE resources /experiences /skills/
knowledge
- Slate and Local Regulators/Permit Specialist
• Industry and Environmentalists
• CONSENSL'S-bascd standards dCN
D-18
-------
MACT
MAXIMUM ACHIEVABLE
CONTROL TECHNOLOGY
• Section 112(d)
• Section 112(g)
• Section 112(j)
MACT
NATIONAL STANDARDS
Section 112(d)
- For new sources, best controlled similar
source
- For existing sources, the average
emission limitation achieved by the best
performing 12 percent (at least)
- Additional reductions based on costs and
noo air quality impacts
MACT
CASE-BY-CASE MACT
Section U2(j) • the "permit hammer" provisions
- Occurs when MACT standard is late by 18
months after due date
- Permit contains case-by-casc MACT
Section 112(g) • sources with increases in HAP
emissions
- Changed operations
- Offset and dc minimts allowances (not
currently operational)
COMPLIANCE HAS BEGUN
FOR 2-YEAR MACTs
Clean Air Act -Section 112 2-Year MACT Standards
D-19
-------
SUMMARY
• Air toxics program is driven by CAA
• Many MACft are being developed
• Traditional rcg development Is too expensive
and permit hammers may fall
• MACT partnerships is one answer
D-20
-------
SUMMARY OF OAQPS
AIR TOXICS
STRATEGY (OATS)
A MID COURSE ADJUSTM ENT THAT
Organizes the program around a management model
Remembers -11 APs differ
Works based on building consensus
WHY IS EPA DEVELOPING
THIS STRATEGY?
, •— j -• . *•
" to protect the public health and environment
• we need a comprehensive vision of how lo
integrate the various air toxics requirements of
the CAA (fitting them with the criteria programs
as bcsi as possible)
• we need lo improve regulatory and economic
efficiency by matching solutions with particular
air IOMCS problems
of Draft OATS
OVERVIEW OF STRATEGY
. — ,' f -r I -
The Air Topics Program Should ... ~ .. -
• Use an air management model to organic the
program and us activities
• Reflect the fact that HAPs have many different
kinds and degrees of effects, exposures, and sources
• Uuild on proactive slaveholder participation
Sxnnun of Drad OATS
A Model for the Air Toxics
Program
Define Which Reduction! Are
N«*d«d for HHE PfOfckou
- An r
D-21
-------
Saw** of Drat OATS
[The Air Toxic Problem Varies
| -By Scales,.
1 -!'—. - | r~' ^
pewit sources; h
••dividual niks ""
URBAN • multiple sources •
stationary and mobile. elevitcd
wdntdvil risks across Urje
•umbers of pcopk
REGIONAL • sources h«M«t
effects whtch trots
jurndicuonal boundaries
NATIONAL • multiple sources.
population risks ind
Sianman- of Draft OATS
i WHY PARTNERSHIPS?
i Uncertainty in science and information
i Need for consensus based decision making
i Availability of skills, knowledge and abilities —
plenty of leverage opportunities
i Upfront plans to address stakeholder interests
r~
r
HOW DOES OATS BUILD
ON CAA?
• A crcdibfelirjtoxlcs program roust stf eoa'» »nd"h*a\ e a~$frong
information base to measure progress m (he overall success of the
program.
• The e%o!ving air tones program should prioritize its efforts such
(hat (he most hazardous pollutants and problems are dealt H ith
first and best.
• The air toiics program should continue 10 use the MA.CT
approach as its cornerstone with residual risk standards used in a
very fucuscd manner. We need to reduce the need fur residual risk
standards
OATS' IMPACT
I OATS report is still in :\ draft stage
I Elements of OATS plan arc in progress
— Can we describe the problem?
- What arc the goals of the program?
— How can progress be measured?
D-22
-------
The Air Toxics Program
in OAQPS
o
K)
OJ
Who Does What From Where
on Air Toxics?
-------
The Office of Air Quality Planning
and Standards
Office of Air Quality Planning and Standards |
Director: John Seitz, Deputy Lydia Wegman J
Emissions Standards Division
Director Bruce Jot dan. Deputy: Jack Edwarrlson
to
Coalings and Consumer Products Group
Lead Linda Herring
Combustion Group
Lead Doug Bell
Metals Group
Lead Al Vervaerl
Minerals and Inoiganic Coatings Group
Lead. Jim Ciowder
Organic Chemicals Group
Lead Susan Wyart
Waste and Chemical Processes Group
Lead K.C Hustvedt
Planning. Policy and Standards Group
Lead Fred Dimmick
| Air Quality Strategies and Standards Division J I Emissions Monitoring and Analysis Division ] [Information Transfer and Program Integration Div]
I Director Sally Shaver, Deputy: Bill Harnett J I Director Bill Hunt, Deputy Henry Thomas J I Director: Tom Curran, Deputy: Bob Kellam
Health Effects and Standards Group
Lead Karen Mailin
Innovative Strategies and Economics Grp
Lead Ron Evans
Ozone Policy and Strategies Group
Lead Tom Helms
Integrated Policy and Strategies Group
Lead Joe Paisie
Risk and Exposure Assessment Group
Lead Dianne Byrne
Visibility and Ecosystem Protection Group
Lead. Eric Ginsburg
Air Quality Modeling Group
Lead: Joe Tikvart
Monitoring and Quality Assurance Group
Lead. Rich Scheffe
Air Quality Trends Analysis Group
Lead. Dave Guinnup
Emission Inventory and Factors Group
Lead Dave Mobley
Source Characterization Group A
Lead. Bill Lamason
Source Characterization Group B
Lead Conniesue Oldham
Education and Outreach Group
Lead. Howard Wright
Information Management Group
Lead. Ed Lillis
Information Transfer Group
Acting Lead: Bob Blaszczak
Integrated Implementation Group
Lead: Karen Blanchard
Operating Permits Group
Lead. Steve Mine
Program Review Group
Lead: Raqueline Shelton
-------
ESD Establishes Air Toxics
Standards
J-Establishes national emissions standards and
manages Federal Programs for controls of air toxics
from stationary sources
juDevelops new source performance standards; control
techniques guidelines; standards for hazardous
wastes under RCRA; and guidance for implementing
standards at the State and local level
j< Develops strategies and policies for regulatory
programs, including co-controls for HAPs and criterfa
pollutants
Examples of
ESD Air Toxics Projects
+ Development of MACT standards
*• Air Toxics long term goals and strategy
*• Characterization of the Air Toxics Problem
> Innovative approaches, e.g. Generic MACT
and the Consolidated Air Rule
AQSSD Assesses Air Toxics
Develops national and geographically focused
strategies and programs based on assessment of
effects, exposure, risk, and economics
Provides expertise for health and environmental
effects, exposure and risk assessement
Responsible for special studies
Provides expertise in benefits assessment,
economics and regulatory impact
Examples of
AQSSD Air Toxics Projects
* Mercury, Great Waters and Utility studies
> Urban Area Source Program and strategy
development
" Risk Identification Project and development of
Residual Risk program
*• Toxics Contingent Valuation Project __
• _J • j • .A- > •
D-25
-------
EMAD Measures and Estimates
J> Develops and.'assists in using air models, as well
as emission factors and inventories
-° Develops methods and assists in measurement
of ambient air, as well as source emissions
^Analyzes air quality data and documents
progress
/"Source of technical expertise on measurement
and estimation methods
Examples of
EMAD Air Toxics Projects
*• National Toxics Inventory
+• Ambient Air Quality Data Catalogue and
Analysis
*• Urban Air Toxics Monitoring Program
*• Workshop on Technical Tools for Air Toxics
• J- €>•&• >•
ITPID's Stock in Trade is
Information
& Manages and transfers air pollution, and pollution
control, information
S Manages information transfer systems
3 Develops and delivers training courses and
educational materials on technical and
management aspects of air pollution control
S Manages the integration of air toxics programs
under section 112 with operating permits
programs
Examples of
ITPID Air Toxics Projects
• MACT implementation strategy development
•AirToxics Health Effects Summary
•Annual Air Toxics Implementation Working
Meeting and Air Toxics Workshop
•AirToxics informational video
• J • J • &• > •
D-26
-------
Should Offsite Consequence
Analysis of Accidental Releases
Be Incorporated Into A
Permitting Program?
Florida Department of Environmental Protection
nd, If So ... How?
SSRSEBWKSSSSSSSEI
* To meet specific criteria (in the sense of a
standard) and used to decide permit
issuance
TI As a weighted element of the decision to
issue a permit, but not have to meet specific
criteria
Florida Department of Environmental Protection
D-27
-------
d,IfSo...How?(cont.)
TI Used to force control or work practice
requirements, the degree to which is based
on the severity of the OCA results
TI Used for emergency planning or public
information only
TI Not required as part of a permit application
Florida Department of Environmental Protection
\|/ What Are You Trying To
Accomplish?
TI Complete protection of the general public
from any kind of accident
TI Complete protection of the general public
from most accidents (risk levels)
7i Better protection of the general public from
any kind of accident
Florida Department of Environmental Protection
D-28
-------
\|/ What Are Trying To
Accomplish? (cont.)
TI Reduction of risk through facility design
and work practices
TI Reduction of risk through facility location
and waste minimization
TI Better contingency and emergency planning
TI Public right to know (information)
Florida Department of Environmental Protection
nalysis Tools Are Important
TI What are the capabilities of the tools
available to support what you are trying to
accomplish?
TI How good do the tools have to be to be
usuable in a permitting program?
Florida Department of Environmental Protection
D-29
-------
Public Health Criteria Data Is
Important
TI Is the quality of the health and risk
information sufficient to support what you
are trying to accomplish?
TI How do we reconcile the different
uncertainties among the tools and the public
health data used in an OCA?
Florida Department of Environmental Protection
egulatory (Political) Questions?
TI Should we be "permitting" accidents?
TI What is not being done through 112(r),
MACT, NESHAPS, or other programs that
we believe need to be accomplished in a
permit program OCA requirement?
TI Who is not covered by an existing program,
but should be?
Florida Department of Environmental Protection
D-30
-------
W What Should We Expect From
7l\lhe Tools?
TI Consistent with chemical and physical laws
and principles.
TI Parameterizations of physical and chemical
effects should be based on validated data
sets and be peer-reviewed.
7i The air dispersion algorithms should be
consistent with those used in the Guideline
for Air Quality Models.
Florida Department of Environmental Protection
What Should We Expect From
Tools? (cont.)
TI Data input used in the models should be
reasonably available.
TI Output data should correspond to the form
of the endpoint criteria being evaluated.
Florida Department of Environmental Protection
D-31
-------
eaknesses of Current Tools
7i Emissions characterization and
quantification.
TI Chemical and physical transformations.
7i Multiple compounds analysis.
TI Time dependent changes.
TI Less than lifetime exposure
Florida Department of Environmental Protection
orkshop Expectations
TI The knowledge of a suite of models that
cover the range of sources and scenarios
that toxics modeling may be performed for.
TI A better understanding of the tools (models)
that are available and what they can and
cannot do. What situations and substances
are each applicable?
Florida Department of Environmental Protection
D-32
-------
orkshop Expectations
7i The knowledge of whether some models.
are considered "better" than others, based
on evaluation studies or physics, that
essentially address the same sorts of
sources.
TI Better knowledge of what is the best
technique at this time for completing risk
assessment analysis?
Florida Department of Environmental Protection
orkshop Expectations
The development of a complete chemical
database that contains all of the chemical
and physical characteristics that are used in
any of the tools. This includes a units
conversion program since every model uses
different units.
Florida Department of Environmental Protection
D-33
-------
-------
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF ENFORCEMENT AND COMPLIANCE ASSURANCE
OFFICE OF COMPLIANCE
LIST OF COMPLIANCE ASSISTANCE TOOLS DEVELOPED AND UNDER
DEVELOPMENT
SECTOR FACILITY- INDEXING PROJECT
• The Manufacturing Branch initiated the Sector Facility Indexing Project. This Project is a
pilot data integration effort that synthesizes environmental records from several data
sources into a system that allows facility-level and sector analysis. The Project focuses on
five major industries: petroleum refining, iron and steel, primary non-ferrous metals, pulp
mills, and automobile assembly. Using national databases, the Project provides the
following information for each of the 699 facilities within the five industry categories:
name, location, permits held under major statutory programs, compliance history,
self-reported pollutant releases under TRI, the relative toxicity of chemicals released,
toxicity-weighted pollutant releases, a description of the population surrounding the
facility, and the actual production or production capacity. Contact: Mike Barrett 202-
564-7019 or Maria Eisemann at 202-564-7016.
SECTOR NOTEBOOKS
• In October 1995, the Office of Compliance published 18 Sector Notebooks. Each
notebook contains: a national industry summary, description of the industrial process, a
waste release profile, pollution prevention opportunities, a summary of statutes and
regulations, a compliance and enforcement profile, a list of compliance activities, and a
directory of contacts. During the first year of availability, approximately 45,000 copies
were distributed. More than 25,000 copies were requested and distributed by OC to other
Federal, state and local government offices; more than 14,000 copies were downloaded
from EPA's Internet site, Enviro$en$e; more than 5,000 copies were sold by the
Government Printing Office to requestors in the private sector; and over 100 countries
requested and received full sets of 18 notebooks. Contact: Seth Heminway 564-7017.
• Based upon the success of this project, OC is developing eight new notebooks for release
in FY 1997. They address the following industry sectors: power generation, federal
facilities, transportation, plastics and synthetic resins, pharmaceuticals, textiles,
shipbuilding and foundries.
D-35
-------
DRY CLEANING
Tools Currently Available:
1. Plain English Guide for Perc Dry Cleaners: A Step by Step Approach to Understanding
Federal Environmental Regulations-This document assists owners and operators of dry
cleaning shops in understanding and complying with federal air, hazardous waste, and wastewater
regulations. The Korean and the Spanish versions of the Plain English Guide will be available in
February 1997. Contact Joyce Chandler 564-7073.
2. Multimedia Inspection Guidance - This document assists field personnel in State, local, and
EPA Regional offices to conduct multimedia inspections and compliance assessments of dry
cleaners. A multi-media inspection checklist is included in the Guidance for on-site visits. Contact
Joyce Chandler 564-7073.
3. Compliance Strategy - The Strategy assists State, local, and EPA Regional officials by
providing multimedia information on the requirements applicable to the dry cleaning industry. It
contains a comprehensive list of EPA products relating to dry cleaners. In addition, it discusses
techniques and agency policies designed to promote compliance assistance. Contact Joyce
Chandler 564-7073.
4. Compendium of Education Materials - This is a searchable database of existing educational
environmental materials and activities for dry cleaners. It may be used by Regions, States and
other entities providing compliance assistance, as well as by members of the dry cleaning
community. Contact Joyce Chandler 564-7073.
5. Generic Dry Cleaning Equipment Chvner's Manual - Proper maintenance and operation of
equipment can be one of the first and easiest steps to help achieve compliance and undertake
pollution prevention. The Office of Air and Radiation developed this manual for dry cleaners with
older machines who could not obtain the manufacturer's instructions for their specific equipment.
Contact George Smith at 919-541-1549
6. Profile for the Dry Cleaning Industry - This sector notebook provides information of general
interest regarding environmental issues associated with the dry cleaning industrial sector. To
order call US Government Information at 202-512-1800. Order number 055-000-00512-5.
7. NESHAP Regulation Translations - Since a significant percentage of the dry cleaning
community does not speak English as its first language, the NESHAP for Perc Dry Cleaners was
translated into Korean, French, Chinese, Vietnamese and Spanish. Contact Joyce Chandler 564-
7073.
D-36
-------
Tools/Activities Under Development:
1. Virginia Partnering Project'. This project is a partnership between the State and the
drycleaning industry in Virginia. The project will prepare and provide documentation to improve
compliance in the state and undertake pilot projects to determine most effective methods to
improve compliance for drycleaners. The project will also provide support during implementation
of new regulatory requirements. Contact: Joyce Chandler 564-7073.
PRINTING
Tools Currently Available:
1. Printers National Environmental Assistance Center. This Center is a partnership between
industry and state environmental experts to create an environmental assistance center for the
printing industry, which comprehensively addresses the environmental compliance needs of smalt
printers, plus provides information on how to achieve compliance by reducing waste and
emissions. The intent of this telecommunications-based Center is to work in collaboration with
printing trade associations and other printing industry experts, regulators and technical assistance
providers to develop and deliver environmental resources for printers. The Center can be found
at Internet: http://www.hazard.uiuc.edu/pneac/pneac.html.
2. Multimedia Compliance/Pollution Prevention Assessment Guidance for Lithographic
Printing Facilities. This guidance helps Regions and States determine the compliance status of
printing facilities and identify ways to bring them into compliance and go beyond compliance.
This document highlights opportunities to implement pollution prevention and innovation
technology. Furthermore, the document can be provided to the printing community for
conducting self-assessments/self-audits and can be valuable in assisting printers to develop
methods to incorporate pollution prevention into their everyday practices. The guidance was pilot
tested at several printing facilities in the State of Washington. This product supports the
Division's work m the printing sector, and it also accomplishes several items that are important to
OC. First, it provides a sector based assessment guidance in a multimedia format. Second, it
assists both regulatory agencies and individual printers audit printing operations; and third, it
provides pollution prevention information to printers. Contact: Doug Jamieson 202-564-7041
Tools/Activities Under Development:
1. Partnering Project with Maryland. To partner with State and printing industry in Maryland.
Prepare and provide documentation and pilot projects to improve compliance with printers. To
pilot the use of customer demands as an incentive for printers to come into compliance and adopt
pollution prevention methods.
D-37
-------
INDUSTRIAL ORGANIC CHEMICAL MANUFACTURING
Tools Currently Available:
1. Joint EPA/CMA/SOCMA/API Subpart CC Rule Compliance Assistance Tool: The
compliance assistance tools being developed in partnership with industry are a compliance
assistance manual, training module and self-audit checklist for use by the regulated community
and inspectors. The Subpart CC Rule requires control of organic emissions from tanks,
containers, impoundments and miscellaneous units. Contact Walt Derieux 564-7067.
2. Process-Based Self-Assessment Tool for the Organic Chemical Industry: This tool has been
developed primarily for small to medium-sized facilities in the organic chemical manufacturing
industry. It will be an effective tool for environmental control officers, managers, and other
facilities personnel in developing and implementing a strategy to identify and track waste streams
associated with a particular, or group of, units, production operations and the associated waste
stream treatment or disposal operation that may impact facilities' environmental compliance.
Contact Jeff KenKnight 564-7044
3. Chemical Industry Baseline Report (1990-1994): The purpose of the baseline report is to
provide a general profile of the chemical industry and to present a 5-year base period reference
point of its compliance history. The report presents an overview of compliance trends and status
by analyzing enforcement-related data for industries that are classified under the 2800 series of the
SIC system. Contact Walt Derieux 564-7067
4. EPA/CMA Root Cause Analysis. A joint EPA/CMA project, to understand underlying factors
that contributed to noncompliance and to find assistance tools, regulatory reform and
management recommendations to promote and improve compliance. Contact Sally Sasnett 564-
7074.
5. Chemical Compliance Assistance Center. The center, in part, will support the states and local
agencies by providing one-stop location for accessing federal regulatory, compliance and technical
assistance information. Contact: Emily Chow 202-563-7071.
6. CCSMD EnviroSenSe Homepage A free public integrated environmental information system
to provide information on pollution prevention, innovative technology and compliance assistance.
Internet address http://es.inel.gov/oeca/ccsmd.html
7. Sector Notebook "Profile of the Organic Chemical Industry". EPA published in September
1995 a sector notebook (EPA/310-R-95-012) on the organic chemical industry that included
general industry information (economic and geographic), a description of industrial processes,
pollution outputs, pollution prevention opportunities. Federal statutory and regulatory framework,
compliance history and a description of partnerships that have been formed between regulatory
D-38
-------
agencies, the regulated community and the public. Order from US Government Information 202-
512-1800 Order #055-000-00515-0.
8. Hazardous Organic NESHAP Compliance Assistance Tool: This tool will address the
hazardous organic NESHAP (HON) rule. This rule under 40 CFR part 63 sets the Maximum
Achievable Control Technology standards applicable to the portion of the organic chemical
industry that is the synthetic organic chemical manufacturing industry. Contact Marcia Mia 564-
7042.
Tools Under Development:
1. Corporate Executive Guide to Facility Environmental Management. Designed for small
business owners, this document will help in planning for a comprehensive management program.
Contact Rich Satterfield 564-2456.
2. Catalogue of Industry Environmental Compliance/Management Documents. A joint
compendium of compliance, technical and other resources. Contact Rich Satterfield 564-2456.
3. Protocol for Conducting Environmental Audits'. A guide for small and medium sized
business on how to conduct compliance and management audits. Contact Rich Satterfield 564-
2456.
4. Chemical Industry State Root Cause Analysis. To understand underlying factors that
contributed to noncompliance in state enforcement actions and to find assistance tools, regulatory
reform and management recommendations to promote and improve compliance. Contact Reggie
Cheatham 564-2425.
FOOD AND KINDRED PRODUCTS
Tools Under Development:
1. Compliance Guidance. This project will develop a multi-media compliance guidance for
small businesses in the food and kindred products fields. Contact: Becky Barclay
at (202) 564-7063 or barclay.rebecca@epamail.epa.gov.
AGRICULTURAL PRACTICES
Tools Currently Available:
/. The Agriculture Compliance Assistance Center is operational and providing outreach, fact
sheets, compliance assistance and liaison activities to the agricultural community on sector
requirements An Internet-available home page is currently on line. The Agriculture Compliance
D-39
-------
Assistance Center relies heavily on existing providers of agricultural information. Internet address:
es.inel.gov/oeca/ag/aghmpg. html.
2. The Agricultural Issues Forum Conference Call is held monthly by the Ag Center. Each call
provides a link between Regions to discuss agriculture sector compliance/enforcement issues and
to communicate what is going on in the Regions.
3. "Major Existing Laws and Programs that Could Affect Producers of Agricultural
Commodities" was issued in September of 1996 by the Ag Branch. It is being made available on
the Internet. This document provides a succinct, general description of EPA's requirements that
may impact farms and is intended as a guide to EPA's requirements. The Internet version will be
cross linked to specific information on requirements. Internet address:
es.inel.gov/oeca/ag/aghmpg. html
4. A CAFO compliance/enforcement strategy is currently under development and a draft is
expected to be available in the Spring of 1997. This strategy will provide guidance on
performance expectations (see the CAFO MO A guidance.) OECA will sponsor a national
meeting for CAFO compliance. Contact: Al Havinga 202-564-4147.
5. The Agriculture and Ecosystems Division has entered into an fnteragency Agreement
(LAG) with USDA-Cooperative State Research, Education, and Extension Service. The
activities being funded in FY 97 under the Interagency Agreement will further the development
and dissemination of compliance assistance information to the agricultural community, which can
serve as national models. The following grants have been selected:
Delaware—This project will develop a plain-language guide book that describes EPA
regulations which impact agriculture.
Cornell—This project will identify agricultural water quality impacts in a watershed;
assess local perceptions of regulatory criteria; and develop outreach materials on
compliance. The project will be utilized as a national model for community based
ag/urban interface issues. The watershed selected will include operations subject to
NPDES permit requirements for Concentrated Animal Feeding Operations (CAFOs).
Kentucky—This project will assess the educational needs of small and minority farmers
regarding EPA issues and regulations. A small and minority farmer's handbook and
one-on-one Extension educational programs will be developed.
Missouri—This project will address air quality requirements affecting farmers and
agribusiness. The project will develop a broad-based educational program on air quality
issues such as the storage and handling of grain, fugitive emissions, volatilization, burning,
and air-stream opacity. Information relating to uncontrolled air quality at animal
D-40
-------
operations and safety issues associated with air quality at agricultural operations will also
be covered. Contact: Ginah Mortensen: 913-551-7864.
Tools Under Development:
/. CAFO Inspector Guidance/Self-Auditing Handbook. Guidance on identifying the need for
permit, compliance with permit, compliance with "no-discharge program," and implementation of
best management practices. Usable by farmers. Contact: AlHavinga 202-564-4147.
AUTO SERVICE AND REPAIR
Tools Currently Available:
/. The Automotive Compliance Assistance Center- GreenLink. Contact Everett Bishop (202)
564-7032. Managed by the Coordinating Committee for Automotive Repair (CCAR) a
consortium of 40+ automotive affiliates representing a wide range of interests within the auto
service industry. Greenlink provides a toll-free phone number with fax-back capability, a private
voice box for the user to leave an anonymous question which in turn will provide a private answer
to the caller within four days, or human interaction with a CCAR representative. For those users
with Internet capability, GreenLink has a web site. Internet capability, Greenlink has a web site,
Internet address, http://www.ccar-greenlink.org. Their phone number is 1-888-GRN-LINK .
Both of these systems provide a repository of compliance requirements, information and
assistance providers available to automotive shop owners and technicians.
Tools Available This Year:
/. Consolidated Environmental Screening Checklist w/Support Documentation. Contact:
Julie Tankersley (202) 564-7002. A one page screening checklist with the major environmental
regulatory requirements associated with this industry. A support booklet will be available with
the checklist to assist users in "how to use" the checklist and including more information for each
requirement, and where to seek further assistance.
2. Educational Modules for Shop Chvners and Technicians. Contact: Everett Bishop (202)
564-7032. A series of environmental educational modules which identifies the major activities
associated with auto repair and service. The modules will discuss the historical nature of the
regulation, the regulation and all that is involved with it, health and safety issues associated with
the activity, pollution prevention opportunities and how to get more information if the user has
more detailed questions.
D-41
-------
ELECTRONICS AND COMPUTERS
Tools Currently Available:
L Printed Wiring Board Compliance Assistance Center. OC in with partners to be determined
will be opening a compliance assistance center for Printed Wiring Board Manufacturers that will
serve as a "one-stop" source of information for federal regulations and pollution prevention
practices. The Center will include a "walk through a plant" feature in which users can click on
parts of the manufacturing process and find out the applicable regulatory requirements and
potential pollution prevention and control means for compliance. Contact: Keith Brown 202-564-
7124.
2. Electronic Database, Developed an electronic database of almost 1,000 Federal and State of
Texas environmental recordkeeping and reporting requirements that may apply to manufacturers
in the computer and electronics industry. These basic requirements were further refined to
include information at the data element level, and the database includes over 8,000 reported data
elements. This project will be expanded to enable businesses to query the database and determine
their regulatory responsibilities. Contact: Keith Brown 202-564-7124.
WOOD PRESERVING
Tools Currently Available:
7. Wood Preserving RCRA Compliance Guide. This guide was designed to provide both
industry and inspectors with a "plain English" summary of the applicable statutory and regulatory
requirements, general information on the industry and the processes used, and the geographic
distribution of the industry. Contact: Seth Hemingway 202-564-7017
METAL FINISHING
Tools Currently Available:
/. The National Metal Finishing Resource Center. The Office of Compliance in conjunction
with the Department of Commerce and the metal finishing industry, established this virtual
compliance assistance center which opened in October of 1996. The purpose of the Center is to
provide industry with "one-stop" shopping for federal environmental regulatory, compliance and
pollution prevention information.
D-42
-------
CLEAN AIR ACT APPLICABILITY DETERMINATIONS
Tools Currently Available:
I, The Office of Compliance updated the Clean Air Act Applicability Determination Index
(ADI) to include 325 new determinations and expanded it to include the MACT program. The
ADI is a computerized, menu-driven compilation of all policy and technical determinations issued
pursuant to the Clean Air Act in the following categories, asbestos, CFC's, NESHAP, NSPS,
woodstoves and Part 63 MACT standards. Within each category, users can search by date,
subpart, references, and string word searches. The compilation is currently available on the
Agency's Technology Transfer Network, which receives an average of 1,200 calls per month on
the Index. Contact Belinda Breidenbach 202-564-7019.
CFCS
Tools Currently Available:
7. Section 608 Compliance Assistance Pilot Project Through this effort, an overview of the .
amendments to the 608 requirements, training module, and self-audit checklist were developed.
These compliance assistance tools were made available to the public through CMA and the
Stratospheric Ozone Information Hotline. Contact. Dawn Banks at 202-564-7034.
SHIP BUILDING
To be available this year:
7. Compliance Guide: In conjunction with the Office of Air and Radiation, OC will produce a
guidebook to complying with the NESHAP for Shipbuilding and Ship Repair standard. Contact:
Suzanne Childress 202-564-7018.
2. Ship Building Sector Notebook: In-depth profile of demographics, compliance profile,
economic trends, p2 opportunities. Contact: Suzanne Childress 202-564-7018.
TEXTILES
To be available this year:
7. Textiles Sector Notebook: In-depth profile of demographics, compliance profile, economic
trends, p2 opportunities. Contact: Brenda Briedenbach 202-564-7019
D-43
-------
TRANSPORTATION
Tools Currently Available:
7. "Fuel for Thought..H
-------
NATIONAL AGRICULTURE
COMPLIANCE ASSISTANCE CENTER
1 he At; (Vnlrr pmvules mini illation to help pio-
diiicfsol .ii;iu iilliir.il (Oininndities and their sup-
poilini; businesses meet thrir e.nvironnienl.il re-
quirements, preve.nl pollution belore it oci.urs, .uul
reduce; i osls by identtlying flexible, common-sense
ways Id .K.hie.ve, ( onipli.mce.
r.onimmm.ilions e.llorls.in- dire.c.le.d prim.irily at
agricultural inlormation providers — im hiding (ed-
er.il .ind state agencies, land gr.int universities, trade
asso( lalions, industry repre.se.nUilives, product and
service providers, l.irm worker associations, environ-
mental advocacy groups, and the .iprir.iilliir.il press
and trade journals — who ultimately convey infor-
mation lo agricultural (armors and farmworkers
The Ag Center is coordinated from El'A's Kansas
City Regional Office, giving it direct access to a laige
segment of the, agricultural community and first-
hand information on factors (hat a!1e.( t the compli-
ance of producers and agribusinesses.
T
>
ccess
^^
o
&>
i^
n
n
3
n
M
in
•3
n
5
rt
TO
O
O
n
o
3
FT
O
tr
1
T
^
s
1 ' 1
M
Ul
3"
3
era
1-800-AT-
z
^
S
TJ
^
*•
5
3
3
3"
0
ira
T
>
o
pj*
C
n
•3
•//esine!
no
o
o
(»
o
tu
TO
IU
3"
3
•o
TO
3-
3
3"
•3
f*
^
5
<
0
C
-greenlm
*"
O
T
>
c
rt
O
3
o
£J
n
•
is
9°
O
r;
2
7;
CTl
O'
cn
Ul
T
•a
3
3
TO
31
.-•
^
•b
3
B>
O
O
PRINTERS INATIONAL
ENVIRONMENTAL
ASSISTANCE CENTER
I'N("A(, addresses the environmental compliance
needs of sni.ill printers ,md provides information
on hoiv to .11 lueve ( omphance by reducing waste
and emissions Information is provided throiiRh a
World Wide Wrb site, satellite training, -besl-in-
c.lass" pollution prevention videoconferences, and
two listservsorVh.it rooms " PRINTECI I, designed
(or technical printing issues, and PKINTKEC".,
which loruse.son environmental laws and regula-
tions aflectmg the printing industry.
PNEAC is operated by the University of Wis-
consin-Extension, Solid f-f I Inzardnus Waste Edu-
cation Center, and the Waste Management and
Resource Center of the Illinois Department of
Natural Resources, with the collaboration of 12
industry associations, nonprofit*, and govern-
ment agencies.
"THROUGH PNCAC'
UNDERLAYMENT fDR
FID PRINTED MATERIAL
RDQERTS Cn.
NEW CENTERS
IN 1 99Y/9B:
Four new (.enters will be established
in IW/'W. El'A is seeking partners in
the private, sector to jointly plan and
operate the following centers:
>• IVansportation- Contact Virginia
Lathrop at 202-564-7057.
>• Printed Wiring Board
Manufacturers: Contact Keith
Urownat 202-SR4-712-I.
>• Small Chemical Manufacturers:
Contact Emily Chow at 2(l2-.S(vl-707l.
» Local Governmental Contact .lohn
202-.SO-l-7n;»fi,
SMALL
COMPLIANCE
-------
3" -
'< -
5. 5 i S" 5
I i. i 5 |
"5^-33
?' 3 n'
n -3
r =
c. ?.
M » ^ S «r •• r* — -f
•* 5 = * s
f!S fit MS nHii* idliil li-rH
IH n| 'I! Hlifff Hflli Hill
~ 3 s. ^2 B. = 2. I =• =' I j •<• S. 3 s ^ =• r. i y? §. i' >: 5
= =• p 2 ?. 3 r. =r i-5^3f=^ Is £ V =• E. = :• ? 2 -5
2S ?t *t:£;2^^" 3 — 7-1 ~^-
= z ~ = 3 §• §. i =• -. s
vi B 2 1' rs
c. r
r- r -= ? ". s- " 5 2
y
~
'1
^
X.
1
a-
s
0_
3
^
n
3
*
X.
c
•3
1
r.
3
«
•§.
^_
s
G
3
•a
L.
n
i RESO
C
n
PI
f-
METAL
o1
a
h auto service activity.
1
JJ
ions, and health and environi
3
o
3
P
— «
S
C
8
v^
•3
o
wners and technicians thro
c
TO.
(A
ff
£
_«
T
n
3
^
nmcnfcil curriculum modul
£
Cfl
3"
£j
~
*
5?
vt
n
3
"y"
eirown multi-media checkl
—•
w
o
3-
VI
late inspectors as a buildin;
wx
O"
o"
n
7!"
O
C.
n
c
VI
n
c.
3"
w_
O
•3
O
3
i
u
»
u
2.
7»
u
s
Sfl
3
n
3
3
T
O
VI
O
c.
1J
c?
c.
3
3
Q.
n
3"
n
n
^
>
c
o
|
o
:o
s
i
C
3
Q
^^
a
_^
r
VJ
•i1 i-
q ri
-• C*
:partments.
::d hy CCAK, the Coordinat
~*
tra
n
O
3
3
•F
rs
c.
c
^;
-^
V)
C.
vt
C
c.
n
3
,»
t
3
C.
n
^
r.
^
c.
*•
-i
5T
u
2
ti
utomotivc trade associatic
3
0;
g
S"
3
2
rrsand I
n
clinicians of auto body and
3
•3
-
"^
3-
C
^3
_y>
O
£J
O
q
lotivi; service industry, pri
2
^>
(^^
o
S
3
jii
3
•3
ii
1
3
n
y.
!•*
CJ
^
h
^
r.
3
ik' ASSISTANCE CENT
PI
a
r
u
9B
nl
| COMPLIANCE INFTJR
f
>
*1
Q
s
1^
X THE AUTOMOTIVE
1
0
a
PI
PI
r
\
1
T T
COMPLIANCE:
ASSISTANCE; CENTERS:
One-stop shopping for your environmental
compliance information needs.
^ General Into: Lynn Vendinello, 202-56
-------
United States Center for Environmental
Environmental Protection Retearch Information
Agency
Research and Development EPA-625/R-96/010» February 1997
& EPA Project Summary
Compendium of Methods for the
Determination of Inorganic
Compounds in Ambient Air
This Project Summary is the announcement of the availability of a collection of methods for
measurement of inorganic pollutants of interest in ambient air. These methods have been prepared to
provide regional, state and local environmental regulatory agencies and other users with step-by-step
sampling and analysis procedures for the determination of selected inorganic pollutants in ambient air.
The methods comprise the Compendium of Methods for the Determination of Inorganic Compounds in
Ambient Air, which is intended to assist those persons responsible for sampling and analysis of ambient
inorganic pollutants.
Determination of pollutants in ambient air is a complex task, primarily because of the wide variety
of compounds of interest coupled with the lack of standardized sampling and analysis procedures. Many
inorganic compounds can be sampled and analyzed by several techniques, often with different interferences
and detection limitations. This Compendium contains a set of 17 methods (in 5 categories) presented in a
standardized format, with a variety of applicable sampling methods and various analytical techniques for
specific classes of inorganic pollutants, as appropriate to their ambient levels and potential interferences.
Consequently, this treatment allows flexibility in selecting alternatives to complement the user's
background and laboratory capability. These methods may be modified from time to time as
advancements are made.
This Project Summary was developed by EPA's Center for Environmental Research Information
(CERJ), National Rjsk Management Research Laboratory (NRMRL), Office of Research and
Development (ORD), U.S. EPA, Cincinnati, OH, with assistance from the ORD's National Exposure
Research Laboratory (NERL) at Research Triangle Park, NC. Its purpose is to announce key findings of
the research project, which is fully documented in a separate report of the same title (see Project Report
ordering information on the last page).
Introduction
The Clean Air Act Amendments of 1990 (CAAA of 1990) have significantly increased the
responsibilities of both federal and state agency programs for evaluating and maintaining air pollutant
emissions compliance. In turn, this increased responsibility has generated a need for more personnel
trained to interpret, enforce, and respond to regulatory initiatives. Consequently, the Agency has
restructured its technology transfer program to more effectively provide technical assistance in the form of
publication of technical documents, presentations and workshops, and development of tools to assist
Agency personnel in keeping their skills up-to-date so that they may efficiently cope with the many changes
evolving in new programs, equipment, sampling and analytical methodology, and available enforcement
tools.
D-47
-------
Limited guidance has been available to state and local agencies or to other organizations concerned
with the determination of inorganic pollutant concentrations in ambient air. As a result, state and local
agencies and others responding to air pollution problems have had to develop their own monitoring
strategies, including selection of monitoring methods, sampling plan design, and specific procedures for
sampling, analysis, logistics, calibration and quality control. For the most pan, these procedures were
based on professional judgments rather than adherence to any documented uniform guidelines. Many
governmental agencies and professional or research organizations have developed ambient air monitoring
methods and procedures, mostly to respond to specialized needs. But these methods and procedures have,
in general, been neither standardized nor readily available to other agencies involved with ambient air
monitoring for various pollutants.
To meet these needs, EPA's ORD, through CERI and NERL has supported technology transfer
programs involving standardized, peer reviewed monitoring methods for regulatory and industrial
personnel via publications of a series of methods Compendia. Other Compendia in this series, reflecting
EPA's commitment to technology transfer, are:
• Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air.
EPA 600/4-89-017, June 1988.
• Compendium of Methods for the Determination of Air Pollutants in Indoor Air,
EPA 600/4-90-010, April 1990.
• Compendium of Methods for the Determination of Toxic Organic Compounds in Ambient Air.
Second Edition, EPA 625/R-96/010b, February 1997.
These Compendia have historically assisted Federal, State, and local regulatory personnel in
developing and maintaining necessary expertise and up-to-date technology involving sampling and analysis
of both indoor and organic hazardous air pollutants (HAPs). The objective of this project was to develop
and standardize methods for measuring inorganic pollutants of interest in ambient air and publish them in
this Compendium of Methods for the Determination of Inorganic Compounds in Ambient Air. This fourth
Compendium adds much needed methods for measurement of inorganic pollutants in ambient air to the
series.
Consistent with past practices, the Compendium methods are provided as guidance only in
appropriate monitoring situations. In particular, these methods are not intended to be used as specific
regulatory guidance for measurement or monitoring purposes and are offered with no endorsement for
suitability or recommendation for any specific application; rather, this is merely to document the methods
and to make them more widely available.
Structure and Content of the Inorganic Compendium
This Compendium has been prepared to provide regional, state and local environmental regulatory
agencies, as well as other interested parties, with specific guidance on the determination of selected
inorganic compounds and pollutants in ambient air. A visual guide to the organization of the Inorganic
Compendium is illustrated in Figure 1, while Table 1 lists the 17 methods which comprise the 5 chapters of
the Compendium. The 17 methods have been compiled from the best elements of methods developed or
used by various research or monitoring organizations. They are presented in a standardized format, and
each one has been extensively reviewed by several technical experts having expertise in the methodology
presented.
D-48
-------
Determination
Sampling Methodology
Analytical Methodology
Average TSP
Conoantration
ConoantfMion
Average PM*
Concentration
Intantaneous
PM,
Conoentratnn
FlBnw A1OWC
Abtofption (FAA)
(Method 10-3.2]
HiVol Swnplar
ferPaftiriM
R«T«™ne«M*ho<
TSPofPM.)
Graph!* Fumaoi Alemie
fxMhod K« J]
Low Vobma Sampler
for Pwlctas
•Ochotomous
(CrtaptarlO-2)
AiyonPhcma
SpaetroeopY OCP)
(Method 1O-3 4)
IndudM*^ CoupW
Pta ii i n Sp«etro»eopY
OCPiMS)
f-(*r«>d IO-35]
Contnuous PM«
Sampler
•6«U Afl»nuaion
-TtOM®
(Chapler 10-1]
Proton kiduo»d
X-ray EmcBion
Specoocopy (PlXE)
•Method IOO 6]
N«utn>n Aacaton
Ana^asfKAA)
(Mxriod IOO 7]
Almospheric Acdrty,
Reactive Aadc and
Base Gases
Annubr Oenuder
Technology
(Chapler IO-4]
Ion Chromaiograpriy (1C)
fcr Atmcphcnc Acdty
ton Chromaiography (1C)
fcr Readwc ActJ e
end Bate turn
Alnosph«nc Mercury.
Vapor and Partxao
Quaro^ber Filer
and GoU-coaled
Bead Trap
(CnapierlO-5]
Duakamalgamaion Cold
Vapor Atomc
Fluorescence
Spectroecopy (CVAFS)
(Method ICVS]
Figure 1. A visual guide to the organization of the Inorganic Compendium,
D-49
-------
Table 1. Methods in the Inorganic Compendium
Chapter IO-1 Continuous Measurement of FM10 Suspended Paniculate Matter (SPM) in Ambient Air
Method 10-1.1 Determination of PM10 in Ambient Air Using the Graseby Continuous Beta Attenuation
Monitor
Method 10-1.2 Determination of PM10 in Ambient Air Using the Thermo Environmental Inc. (formerly
Wedding and Associates) Continuous Beta Attenuation Monitor.
Method 10-1.3 Determination of PM10 in Ambient Air Using a Continuous Rupprecht and Patashnick
(R&P) TEOM* Particle Monitor
Chapter 1O-2 Integrated Sampling of Suspended Particuiate Matter (SPM) In Ambient Air
Method IO-2.1 Sampling of Ambient Air for Total Suspended ParticuJate Matter (SPM) and PM10
Using High Volume (HV) Sampler
Method IO-2.2 Sampling of Ambient Air for PM10 Using a Graseby Dichotomous Sampler
Method 10-2.3 Sampling of Ambient Air for PM10 Concentration Using the Rupprecht and Paiashnick
(R&P) Low Volume Partisol* Sampler
Method IO-2.4 Calculations for Standard Volume
Chapter IO-3 Chemical Species Analysis of Filter-Collected Suspended Particuiate Matter
Method 10-3.1 Selection, Preparation and Extraction of Filter Material
Method 10-3.2 Determination of Metals in Ambient Particuiate Matter Using Atomic Absorption (AA)
Spectrometry
Method 10-3.3 Determination of Metals in Ambient Particuiate Matter Using X-Ray Fluorescence
(XRF) Spectrometry
Method 10-3.4 Determination of Metals in Ambient Particuiate Matter Using Inductively Coupled
Plasma (ICP) Spectrometry
Method 10-3.5 Determination of Metals in Ambient Paniculate Matter Using Inductively Coupled
Plasma/Mass Spectrometry (TCP/MS)
Method 10-3.6 Determination of Metals in Ambient Particuiate Matter Using Proton Induced X-Ray
Emission (PKE) Spectroscopy
Method IO-3.7 Determination of Metals in Ambient Particuiate Matter Using Neutron Activation
Analysis (NAA) Gamma Spectrometry
Chapter IO-4 Determination of Reactive Acidic and Basic Gases and Strong Acidity of Atmospheric Fine
Particles in Ambient Air Using the Annual Denuder Technology
Method IO-4.1 Determination of the Strong Acidity of Atmospheric Fine Particles (< 2.5^m)
Method IO-4.2 Determination of Reactive Acidic and Basic Gases and Strong Acidity of Atmospheric
Fine Particles
Chapter 10-5 Sampling and Analysts for Atmospheric Mercury
Method 10-5.1 Sampling and Analysis for Vapor and Panicle Phase Mercury in Ambient Air Utilizing
Cold Vapor Atomic Fluorescence Spectrometry (CVAFS)
D-50
-------
N-NITROSODIMETHYLAMINE
CRESOLS/PHENOLS
SPECIFIC
PESTICIDES
ALDEHYDES/KETONES
SEMI-
VOLATILES
PESTICIDES/PCB's
FORMALDEHYDE
VOLATILES
SEMl-VOLATILES
VOLATILES (-158' TO 170'C)
VOLATILE (-10-C to 20
-------
D-52
-------
EPA/625/R-96/010&
Compendium of Methods
for the Determination of Toxic
Organic Compounds
in Ambient Air
Second Edition
Compendium Method TO-16
Long-Path Open-Path Fourier
Transform Infrared Monitoring
Of Atmospheric Gases
Center for Environmental Research Information
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
January 1997
D-53
-------
Method TO-16
Acknowledgements
This Method was prepared for publication in the Compendium of Methods for the Determination of'Toxic
Organic Compounds in Ambient Air, Second Edition, (EPA/625/R-96/010b), which was prepared under
Contract No. 68-C3-0315, WA No. 3-10, by Midwest Research Institute (MRI), as a subcontractor to
Eastern Research Group, Inc. (ERG), and under the sponsorship of the U.S. Environmental Protection
Agency (EPA). Justice A. Manning and John Burckle, Center for Environmental Research Information
(CERI), and Frank F. McElroy, National Exposure Research Laboratory (NERL), both in the EPA Office
of Research and Development, were the project officers responsible for overseeing the preparation of this
method. Additional support was provided by other members of the Compendia Workgroup, which
include:
• John Burckle, U.S. EPA, ORD, Cincinnati, OH
• James L. Cheney, Corps of Engineers, Omaha, NB
• Michael Davis, U.S. EPA, Region 7, KC, KS
• Joseph B. Elkins Jr., U.S. EPA, OAQPS, RTP, NC
• Robert G. Lewis, U.S. EPA, NERL, RTP, NC
• Justice A. Manning, U.S. EPA, ORD, Cincinnati, OH
• William A. McClenny, U.S. EPA, NERL, RTP, NC
• Frank F. McElroy, U.S. EPA, NERL, RTP, NC
• Heidi Schultz, ERG, Lexington, MA
• William T. "Jerry" Winberry, Jr., MRI, Gary, NC
This Method is the result of the efforts of many individuals. Gratitude goes to each person involved In
the preparation and review of this methodology.
Author(s)
• George M. Russwurm, ManTech Environmental Technology, Inc., Research Triangle Park, NC
Peer Reviewers
• Robert L. Spellicy, Radian International, Austin, TX
• William F. Herget, Radian International, Austin, TX
• Judith 0. Zwicker, Remote Sensing Air Inc., St. Louis, MO
• William W. Vaughn, Remote Sensing Air Inc., St. Louis, MO
• Robert J. Kricks, RJK Consultant, Cranford, NJ
• Robert H. Kagaan, AIL Systems Inc., Deer Park, NY
• J.D. Tate, Dow Chemical, Freeport, TX
Finally, recognition is given to Frances Beyer, Lynn Kaufman, Debbie Bond, Cathy Whitaker, and Ksthy
Johnson of Midwest Research Institute's Administrative Services staff whose dedication and persistence
during the development of this manuscript has enabled it's production.
DISCLAIMER
This Compendium has been subjected to the Agency's peer and administrative review, and it has been
approved for publication as an EPA document. Mention of trade names or commercial products, does
not constitute endorsement or recommendation for use.
D-54
-------
METHOD TO-16
Long-Path Open-Path Fourier Transform Infrared
Monitoring Of Atmospheric Gases
TABLE OF CONTENTS
Page
1. Scope 16-1
2. Summary of Method 16-2
3. Significance 16-3
4. Applicable Documents 16-4
4.1 ASTM Standards 16-4
4.2 EPA Documents 16-5
5. Definitions 16-5
6. Apparatus and System Requirements 16-8
6.1 Summary 16-8
6.2 FT-IR Sensor Requirements 16-8
6.3 Computer Requirements 16-8
6.4 Software Requirements 16-9
7. Materials and Supplies 16-9
8. Standard Procedures for Processing of Infrared Spectra 16-9
8.1 Summary 16-9
8.2 Suggested Order of Generation of FT-IR Concentration Data 16-10
8.3 Selection of Wave Number Regions for Analysis in the Presence of
Interfering Species 16-10
8.4 Generation of a Background Spectrum 16-13
8.5 Production of a Water Vapor Reference Spectrum 16-15
8.6 Subtraction of Stray Light or Black Body Radiation 16-17
8.7 Generation of an Absorbance Spectrum 16-19
8.8 Correction for Spectral Shifts 16-19
8.9 Analysis of the Field Spectra for Concentration 16-21
8.10 Post-Analysis Review of the Data 16-22
9. Quality Assurance 16-24
9.1 Summary 16-24
9.2 The Determination of Method Noise or Method Noise Equivalent Absorption . . 16-24
9.3 The Measurement of the Return Beam Intensity 16-25
9.4 The Measurement of Stray Light 16-27
9.5 The Measurement of Black Body Radiation 16-27
9.6 The Determination of the Detection Limit 16-28
9.7 The Determination of Precision 16-30
111
D-55
-------
LIST OF TABLES (continued)
Page
9.8 The Determination of Accuracy 16-31
9.9 The Measurement of Resolution 16-33
9.10 The Determination of Nonlinear Instrument Response 16-34
9.11 The Determination of Water Vapor Concentration 16-36
10. References 16-37
IV
D-56
-------
METHOD TO-16
Long-Path Open-Path Fourier Transform Infrared
Monitoring Of Atmospheric Gases
1. Scope
1.1 Fourier transform infrared (FT-IR) spectroscopy used for open-path monitoring of atmospheric
gases is undergoing a vigorous development and growth period. Until now the developmental effort and
the most of the data acquisition have been performed by highly trained individuals experienced in the
fields of instrument development and spectroscopy. In the future, operators trained at the technician level
will be required to perform the operation routinely. This method is intended to address that need.
Specifically, the method is intended to allow trained technicians to acquire data in a standardized way and
to process that data to obtain atmospheric gas concentrations. The primary intent is that the results will
be obtained in a consistent fashion.
1.2 This method is intended for the use of an FT-IR system that acquires data using a long, open air path
and does not require the acquisition of a sample for subsequent analysis. The system produces data that
is a time sequence of the path-averaged atmospheric concentrations of various gases. Because the FT-IR
can potentially measure the concentration of a large number of atmospheric gases, this method does not
address the requirements for measuring a particular gas or a set of gases. Rather, it is intended to be a
generalized method.
1.3 The method is intended to be instrument independent in that it discusses the processing of spectra
so that gas concentrations can be obtained. The primary geometric configurations of FT-IR instruments
that are commercially available are the monostatic configuration and the bistatic configuration. These
configurations are shown schematically in Figures 1 and 2. This method can be used to process data
from either of these types. It is assumed that the FT-IR is under computer control and that the
controlling software will allow the manipulation of the spectra. This method is specifically designed to
process spectra that will be analyzed by the commonly called classical least-squares technique. If the
classical least-squares technique is to be used, the spectra must be processed in a specific way, and this
document describes the steps of that processing. Although there are other ways to analyze the spectra,
such as partial least squares, iterative least squares, spectral subtraction, principal component analysis,
and peak height and peak area calculations, the use of these techniques requires that the spectra be
processed in a different way than is described here. While some of the procedures given here are
applicable to the other analysis techniques, this method addresses only the classical least-squares
technique.
1.4 The method is not intended as a tutorial for the use of the computer software or the instruments
themselves. Inclusion of this type of explanation would make this document excessively long. When
certain features from the software packages are called for, it is assumed that the user has read or can read
the appropriate description in the specific manual. As far as the instruments are concerned, it is assumed
that the operator has participated in instrument training provided by the specific instrument manufacturer
and that this training has been sufficient to enable the operator to produce spectra and to save them on
a disk.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-1
D-57
-------
Method TO-16 VOCs
1.5 Since this method in this document is considered to be a set of operational procedures, the document
does not contain an in-depth explanation about the origin or the rationale for the inclusion of particular
steps. For a more complete and rigorous discussion of the FT-IR technique, the user of this method is
referred to EPA's FT-IR Open-Path Monitoring Guidance Document (1).
1.6 The intent of this document is to provide the operator with step wise procedures producing
concentration data from spectra taken with an FT-IR. To accomplish this, items such as background
spectra, water vapor reference spectra, and stray light are discussed. In keeping with the concept of a
procedure, these quantities must be specified directly. However, the entire area of FT-IR remote sensing
of the -atmosphere is undergoing rapid change, and parts of these procedures will without doubt need
revision in the future. Throughout mis document the user must keep in mind that for each procedure in
the TO-16 method there may be other equally valid procedures that are currently being used that are not
described here.
1.7 Finally, a statement about computer automation of these procedures is in order here. The method
does not address the problem of automation directly and implies that an operator is available to perform
the individual steps. Some operational software packages already exist that incorporate many of these
routines in an automatic way. It is felt that each procedure potentially can be automated, but the steps
listed here are those that need to be incorporated in any automated procedure.
2. Summary of Method
2.1 For the purpose of this document the operation of an FT-IR remote sensor is divided into two parts.
The first is initial data acquisition after the system has been set up by the manufacturer and the second
is what is considered to be routine data acquisition. The first of these data acquisition periods is intended
to produce data that will form the basis of a quality assurance data set. The second is devoted to the
production of time sequences of atmospheric gas concentration data.
2.2 There are several items that need to be determined before the FT-IR system can be put into routine
service. These items have been selected to determine how the system is functioning initially and include
the shortest path length that will saturate the detector, the ambient black body radiation level for the
bistatic configuration, the stray light inside the instrument for the monostatic configuration, and the return
intensity as a function of distance. Beyond these steps there is a survey set of data that should be
acquired. Data from this survey set will form the basis of the routinely monitored quality control checks
for the instrument.
2.3 In addition to the FT-IR data it is required that the ambient temperature and the relative humidity
be monitored on a continuous basis so that the water vapor concentration as a function of time can be
determined. It is to be clearly understood that relative humidity measurements alone are not relevant to
this operation but the amount of water is. These data should be acquired at the site where the FT-IR data
is taken. Use of data taken at airports miles away is not appropriate.
2.4 The initial step in the procedure for determining the concentration data for various gases is the
production of a set of interferograms, and it should be the interferograms that are saved as the primary
data. The various procedures given in Section 8 of this document use the single beam spectrum that is
created from the interferogram. A single beam spectrum taken with a monostatic system over a 414-m
Page 16-2 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-58
-------
VOCs Method TO-16
path length is illustrated in Figure 3. Various atmospheric constituents as well as a stray light component
are pointed out. However, it is the interferograms that are considered the most important data. If they
are not saved they cannot accurately be reproduced by simply performing the inverse Fourier transform.
Once a set of target gases has been selected, the wave number regions to be used in the analysis are
chosen. For the monostatic instrument geometry, the stray light component must next be subtracted from
each single beam spectrum. For the bistatic case, the black body radiation spectrum must be subtracted
from each single beam spectrum. One spectrum from this set is then chosen to be the background, or
/Q, spectrum, and this can be turned into a synthetic background spectrum. A second spectrum is then
used to create a water vapor reference spectrum, and all the remaining spectra are then converted to
absorbance spectra. All the spectra to be analyzed are then checked for wave number shifts. Finally,
the absorbance spectra are analyzed by the classical least-squares technique.
2.5 It is suggested that, if possible, twice each day a short cell filled with a known quantity of gas should
be inserted in the infrared beam, and four spectra should then be recorded. The instrument must be
operating in exactly the same manner as it is when it is routinely acquiring data but this time with the
cell. These spectra are analyzed in the same way as all other spectra, but for the particular gas in the
cell. This data is then added to the appropriate control charts. No exact procedure for using this cell
and no specifications for the cell are provided at this time within this document. Not all the instruments
that are commercially available can accommodate a cell, and many gases cannot be easily used in such
a cell.
2.6 A subset of each day's spectra is then selected and the following two items are determined: the root
mean square (RMS) noise in three wave number regions and the return beam intensity at two wave
numbers. The range of water vapor concentrations over the time period during which the subset of data
was taken is calculated. These data are also then added to the appropriate control charts.
2.7 The remainder of the data can then be checked as described in Section 8 and then against the data
quality objectives provided by the monitoring program.
3. Significance
3.1 VOCs enter the atmosphere from a variety of sources, including petroleum refineries, synthetic
organic chemical plants, natural gas processing plants, and automobile exhaust. Many of these VOCs
are acutely toxic; therefore, their determination in ambient air is necessary to assess human health
impacts.
3.2 The environmental impacts from the release of airborne VOCs is a topic of great interest among air
pollution scientists. It is important that measurement methods be developed to accurately assess the
impact of airborne chemical emissions on the environment. Until now, traditional air sampling/analytical
techniques (i.e., solid adsorbents, treated canisters, portable gas chromatographs, etc.) have been used
to characterize emission impacts of airborne toxic chemicals in the environment.
3.3 The method of trace gas monitoring using FT-IR-based, long-path, open-path systems has a number
of advantages that are significant over traditional methods. Some of these advantages are related to the
path monitoring aspect of this method which, by its very nature, distinguishes the method from all point
monitoring methods. The main advantages of these systems are the following:
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-3
D-59
-------
Method TO-16 VOCs
• Integrity of the sample is assured since no sampling actually occurs.
• Multi-gas analysis is possible with a single field spectrum.
• Path-integrated pollutant concentrations are obtained.
• Spatial survey monitoring of industrial facilities is possible if scanning optics are used.
• Coadding of spectra to improve detection capabilities is easily performed.
• Rapid temporal scanning of line-of-sight or multiple lines-of-sight is possible.
• Monitoring of otherwise inaccessible areas is possible.
3.4 Applications include the monitoring of atmospheric gases along the perimeters of industrial facilities
or, from an elevated, centrally located platform, monitoring over the industrial facility to infrared sources
or retroreflectors placed along the facility edge. Other applications include monitoring (1) at hazardous
waste sites during remediation or removal operations to provide warnings of high concentrations and to
verify that back-to-work conditions have been achieved; (2) in response to accidental chemical spills or
releases; (3) in workplace environments to develop concentration profiles at the worker level; and (4) in
the ambient air for some compounds. It is theoretically applicable to the measurement of all gaseous
compounds that exhibit absorption spectra in the mid-infrared region of the electromagnetic spectrum.
3.5 Significant advances have been made in recent years to develop the FT-IR systems into practical
remote sensing tools, particularly in the understanding of the importance of water vapor interference
associated with FT-IR methodology. As indicated in this method, the generation of a background
spectrum for a given measurement and the generation of water vapor spectra to account to water vapor
interference in mid-infrared measurements are features of the FT-IR measurement technique that deserve
more attention. The significance of Compendium Method TO-16 is that it is the first such method to
address all the features that are required to make a field measurement using FT-IR-based systems. As
such, it provides a guide to field measurement as well as a basis for improvement and further
consideration.
3.6 The ultimate significance of remote sensing with FT-IR systems is a matter of cost effectiveness and
of technological advances. Technological advances are required in at least two important areas: (1) the
improvement in the characteristics of the instrumentation itself and (2) the development of "intelligent"
software. The software is required to improve the means for short-term adjustment of background and
water vapor spectra to account for the continual variation of ambient conditions that can adversely affect
the accuracy and precision of FT-IR based systems.
4. Applicable Documents
4.1 ASTM Standards
• Method D1356 Definition of Terms Relating to Atmospheric Sampling and Analysis.
Page 16-4 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-60
-------
VQCs Method TO-16
4.2 EPA Documents
• Technical Assistance Document for Sampling and Analysis of Toxic Organic Compounds in Ambient
Air, U. S. Environmental Protection Agency, EPA-600/4-83-027, June 1983.
• Quality Assurance Handbook for Air Pollution Measurement Systems, U. S. Environmental
Protection Agency, EPA-600/R-94-038b, May 1994.
• Open-Path Monitoring Guidance Document, U. S. Environmental Protection Agency,
EPA 600/4-96-040, April 1996.
5. Definitions
[Note: This section contains a portion of the glossary of terms from the guidance document (l)for remote
sensing that is applicable to Compendium Method TO-16. When possible, definitions of terms have been
drawn from authoritative texts or manuscripts in the fields of remote sensing, air pollution monitoring,
spearoscopy, optics, and analytical chemistry. In some cases, general definitions have been augmented
or streamlined to be more specific to long-path, open-path monitoring applications and to Compendium
Method TO-] 6. These definitions were intended to remain scientifically rigorous and still be generally
applicable to the variety of FT-IR open-path remote-sensing issues that must be addressed by the
operator.]
5.1 Absorbance— the negative logarithm of the transmission. A = -\n(I/I^), where 7 is the transmitted
intensity of the light and 70 is the incident intensity. Generally, the logarithm to the base 10 is used,
although the quantity / really diminishes exponentially with A.
5.2 Apodization— a mathematical transformation carried out on data received from an interferometer to
alter the instrument's response function. There are various types of transformation; the most common
are boxcar, triangular, Happ-Genzel, and Beer-Norton functions.
5.3 Background Spectrum— 1. With all other conditions being equal, that spectrum taken in the
absence of the particular absorbing species of interest. 2. Strictly, that radiant intensity incident on the
front plane of the absorbing medium. 3. A spectrum obtained from the ambient black body radiation
entering the system. This background must be considered in FT-IR systems, in which the IR beam is not
modulated before it is transmitted along the path. For FT-IR systems that do not use a separate source
of infrared energy, the background is the source of infrared energy.
5.4 Beer's Law— Beer's law states that the intensity of a monochromatic plane wave incident on an
absorbing medium of constant thickness diminishes exponentially with the number of absorbers in the
beam. Strictly speaking, Beer's law holds only if the following conditions are met: perfectly
monochromatic radiation, no scattering, a beam that is strictly collimated, negligible pressure-broadening
effects (2,3).
5.5 Bistatic System— a system in which the receiver is some distance from the transmitter. This term
is actually taken from the field of radar technology. For remote sensing, this implies that the light source
and the detector are separated and are at the ends of the monitoring path.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-5
D-61
-------
Method TO-16
5.6 Fourier Transform—a mathematical transform that allows an aperiodic function to be expressed as
an integral sum over a continuous range of frequencies (4). The Fourier transform of the interferogram
produced by the Michelson interferometer in an FT-IR is the intensity as a function of frequency.
5.7 FT-IR—an abbreviation for "Fourier transform infrared." A spectroscopic instrument using the
infrared portion of the electromagnetic spectrum. The working component of this system is a Michelson
interferometer. To obtain the absorption spectrum as a function of frequency, a Fourier transform of the
output of the interferometer must be performed. A brief overview of the FT-IR is provided in FT-IR
Theory (5). An in-depth description of the FT-IR can be found in Fourier Transform Infrared
Spectrometry (6).
5.8 Intensity—the radiant power per unit solid angle. When the term "spectral intensity" is used, the
units are watts per steradian per nanometer. In most spectroscopic literature, the term "intensity" is used
to describe the power in a collimated beam of light in terms of power per unit area per unit wavelength.
5.9 Interference—the physical effects of superimposing two or more light waves. The principle of
superposition states that the total amplitude of the electromagnetic disturbance at a point is the vector sum
of the individual electromagnetic components incident there. For a two-component system of collinear
beams of the same amplitude, the mathematical description of the result of addition is given by I(p) =
2/0(l -I- cos/XT), where 70 is the intensity of either beam, and A is the phase difference of the two
components. The cosine term is called the "interference term" (7,8). See also "Spectral Interference."
5.10 Interferogram—the effects of interference that are detected and recorded by an interferometer; the
output of an FT-IR and the primary data that is collected and stored (6,8).
5.11 Interferometer—any of several kinds of instruments used to produce interference effects. The
Michelson interferometer used in FT-IR instruments is the most famous of a class of interferometers that
produce interference by the division of an amplitude (9).
5.12 Light—strictly, light is defined as that portion of the electromagnetic spectrum that causes the
sensation of vision. It extends from about 25,000 cm"1 to about 14,300 cm"1 (4).
5.13 Minimum Detection Limit—the minimum concentration of a compound that can be detected by
an instrument with a given statistical probability. Usually the detection limit is given as 3 times the
standard deviation of the noise in the system. In this case, the minimum concentration can be detected
with a probability of 99.7% (10,11).
5.14 Monitoring path—the actual path in space over which the pollutant concentration is measured and
averaged.
5.15 Monostatic System—a system with the source and the receiver at the same end of the path. For
FT-IR systems, the beam is generally returned by a retroreflector.
5.16 Reference Spectra—spectra of the absorbance versus wave number for a pure sample of a set of
gases. The spectra are obtained under controlled conditions of pressure and temperature and with known
concentrations. For most instruments, the pure sample is pressure-broadened with nitrogen so that the
Page 16-6 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-62
-------
VOCs Method TO-16
spectra are representative of atmospherically broadened lines. These spectra are used for obtaining the
unknown concentrations of gases in ambient air samples.
5.17 Retail > e Absorption Strength—a term used exclusively in Compendium Method TO-16 to describe
the relation of absorption due to interfering species to the absorption of the target gas.
5.18 Resolution—the minimum separation that two spectral features can have and still, in some manner,
be distinguished from one another. A commonly used requirement for two spectral features to be
considered just resolved is the Raleigh criterion. This states that two features are just resolved when the
maximum intensity of one falls at the first minimum of the other (5,6). This definition of resolution and
the Raleigh criterion are also valid for the FT-IR, although there is another definition in common use for
this technique. This definition states that the minimum separation in wave numbers of two spectral
features that can be resolved is the reciprocal of the maximum optical path difference (in centimeters) of
the t\vo interferometer mirrors employed.
5.19 Retroreflector—the CIE (Commission Internationale de 1'Eclairage) defines retroreflection as
"ladiation returned in directions close to the direction from which it came, this property being maintained
over wide variations of the direction of the incident radiation." Retroreflector devices come in a variety
of forms and have many uses. The one commonly described by workers in remote sensing uses total
internal teflection from three mutually perpendicular surfaces. This kind of retroreflector is usually called
a corner cube or prismatic retroreflector (12).
5.20 RMS N'oise— this quantity is actually the statistical quantity rms deviation. In Compendium Method
TO-16 the rms noise (deviation) is calculated by using a least squares fit to the baseline. Because of this
calculation, the rms noise in Compendium Method TO-16 uses the quantity N-2 in the denominator rather
than N-l as normally described.
5.21 Single Beam Spectrum—that spectrum which results from performing the Fourier transform on
the interferogram. It is not a transmission spectrum. The term "single beam" is a holdover from older
instruments that were double beam instruments.
5.22 Source—the device that supplies the electromagnetic energy for the various instruments used to
measure atmospheric gases. These generally are a Nernst glower or globar for the infrared region or a
xenon arc lamp for the ultraviolet region.
5,23 Spectral Intensity—see Section 5.8.
5.24 Spectral Interference—when the absorbance features from two or more gases cover the same wave
number regions, the gases are said to exhibit spectral interference. Water vapor produces the strongest
spectra] interference for infrared spectroscopic instruments that take atmospheric data.
5.25 Synthetic Background—a spectrum that is made from a field spectrum by choosing points alon°
the baseline and connecting them with a high-order polynomial or short, straight lines. The synthetic
bacKgiound is then used to find the absorbance spectrum.
5.26 Wave Number—the number of waves per centimeter. This term has units of reciprocal centimeters
(cm'1).
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-7
D-63
-------
Method TO-16 VOCs
6. Apparatus and System Requirements
6.1 Summary
6.1.1 Compendium Method TO-16 is a procedure that deals with how spectra taken with an FT-IR
are to be processed in order to obtain various atmospheric gas concentrations. General requirements for
FT-IR instrumentation is being prepared by a committee of L'Organisation Internationale de Metrologie
Legale (OIML) and will be available in the very near future.
6.1.2 The instrument requirements listed here are limited to those that will define a rudimentary but
operational system. The requirements are delineated into three categories: those of the FT-IR sensor itself
(see Section 6.2), the computer associated with it (see Section 6.3), and the software that allows for data
analysis (see Section 6.4).
6.2 FT-IR Sensor Requirements
6.2.1 The system should be capable'of making spectral absorption measurements along an open air
optical path.
6.2.2 The system can be either of the monostatic or the bistatic geometry.
6.2.3 The system must be able to produce and save an interferogram and a single beam spectrum.
6.2.4 The system must be able to operate with a resolution of at least 1 cm"* over the mid-infrared
region (700-4200 cm"1).
6.2.5 The system must be capable of acquiring data by co-adding individual interferogram scans in
one-scan increments. As a minimum, the system must be able to acquire data from a one-scan
interferogram to an interferogram made up of sufficient co-added scans so that at least 5-min
concentration averages can be obtained.
6.2.6 The system must be able to perform the mathematical procedure of Fourier transformation on
the interferogram, thereby producing a so-called single beam spectrum. The transform can be performed
as part of post-acquisition processing or in quasi-real time. If performed in quasi-real time the process
of transformation should not add significantly to the data acquisition time.
6.2.7 Although there is no agreed upon procedure for the use of a gas cell with these systems, the
system may have provisions for installing an ancillary gas cell in the optical beam. If that is the case,
the installation must allow for the entire beam to pass through the cell. The cell can be of any of several
designs: short, either single or double pass; multi-pass capable of producing a relatively long optical path;
or a multi-chambered cell with the individual chambers interconnected and in parallel with one another.
6.3 Computer Requirements
6.3.1 The computer must be capable of acquiring data in the form of interferograms with sufficient
speed so that the system is able to operate in quasi-real time.
6.3.2 The computer must have provisions for storing of the data acquired in one 24-h period. The
storage must accommodate the interferograms.
6.3.3 The computer must have sufficient RAM to operate the controlling software and the data
manipulation software.
Page 16-8 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-64
-------
VOCs Method TO-16
6.4 Software Requirements
6.4.1 The software must have provisions for manipulating the spectra so that all the individual
procedures listed here can be accomplished.
6.4.2 The software must be able to perform the analysis for concentration using classical least
squares.
7. Materials and Supplies
7.1 Only a small number of materials are required in addition to the basic instrument for this method.
However, the basic instrument operation may have specific material requirements such as liquid nitrogen
or nitrogen, etc. A listing of any specific instrument's requirement for material must be obtained from
the manufacturer.
7.2 A set of gases may have to be purchased in order to acquire spectra with a cell. This set of gases
is intended to allow the operator to determine the precision and accuracy of the data obtained from the
field spectra, but at the present time no procedure using a cell has been developed. The specific gases
required are dependent on the particular monitoring program. If necessary, the gases can be purchased
as pure gases, which are then diluted with nitrogen for use, or they can be a mixture of gases that are
properly mixed at purchase. The dilution step can be quite cumbersome and it is recommended that
appropriate mixtures of gas be acquired directly whenever possible. The required concentrations of the
gases are dependent on the anticipated concentration of the target gas in the atmosphere and the ratio of
the actual path length used to the length of the cell. Many applications will require that these gases be
purchased with certifications traceable to the National Institute of Standards and Technology.
7.3 The only other material that may be required is a set of screens of varying mesh that will be used
when determining whether the system is responding linearly. This screening can be regular aluminum
window screen or made of other opaque metallic materials. The size of the mesh is not really important,
but the screen should be large enough to cover the entire beam. The mesh itself should be chosen so as
to change the transmitted intensity by an easily measured amount (on the order of 25% or more). The
screen must not be made of any plastic materials as they transmit infrared energy. This in itself is not
a problem but the plastic materials introduce absorbance at specific wave numbers and may not provide
the desired result.
8. Standard Procedures for Processing of Infrared Spectra
8.1 Summary
The specific procedures that are required to produce atmospheric gas concentration data are included in
this section. They start with the general operations procedure that describes how the other individual!
procedures should follow one another.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-9
D-65
-------
Method TO-16 VQCs
8.2 Suggested Order of Generation of FT-IR Concentration Data
8.2.1 This section provides the FT-IR operator with a systematic approach to the generation of FT-IR
concentration data. These procedures are recommended for operators with little experience with FT-IR
operation. As the operator gains more experience with the production of FT-IR concentration data, he
may want to reorder the sequence of events to better fit his experimental schedule.
8.2.2 Assumptions
8.2.2.1 Compendium Method TO-16, in general, does not describe the general planning that is
necessary to conduct a field program. It is felt that each data acquisition program is different and all
programs cannot be covered in depth with a single procedure. For example, the time for acquiring a
single spectrum can vary from a single scan of a few seconds up to a half hour. The actual time required
for any one program is dependent on that program and therefore is not discussed further in this
procedure. Much of the planning for the acquisition of data is connected to the generation of a detailed
quality assurance/quality control (QA/QC) program plan, and this method is not considered to be such
a plan. Section 9 presents individual items that should be addressed as a minimal quality assurance effort.
The procedures in Section 8 cover only the production of concentration data.
8.2.2.2 It is also assumed that water vapor concentration data for the data acquisition period is
available. This method does not discuss how to acquire that data, however. The water vapor
concentration data is used for post-analysis review and for some of the QA/QC checks.
8.2.2.3 From this point on, it is assumed that the individual spectra have already been acquired.
It is assumed that the interferograms have been converted to single beam spectra. It is further assumed
that no other data manipulation has occurred.
8.2.3 The suggested order for the production of concentration data utilizing FT-IR is given below.
• Selection of wave number regions for analysis in the presence of interfering species (see
Section 8.3).
• Generation of a background spectrum (see Section 8.4).
• Production of a water vapor reference spectrum (see Section 8.5).
• Subtraction of stray light or black body radiation (see Section 8.6).
• Generation of an absorbance spectrum (see Section 8.7).
• Correction for spectral shifts (see Section 8.8).
• Analysis of field spectra for concentration (see Section 8.9).
• Post-analysis review of the data (see Section 8.10).
8.3 Selection of Wave Number Regions for Analysis in the Presence of Interfering Species
8.3.1 Purpose. This section instructs the operator on how to select the wave number regions that
are to be used in the analysis of field spectra. This section includes the process of working with
interfering species because the absorbance spectrum of any one particular gas frequently overlaps with
Page 16-10 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-66
-------
VOCs Method TO-16
the absorbance of another species. This section also provides the operator with a measure of the strength
of the interference.
8.3.2 Assumptions. One of the most important requirements when utilizing classical least squares
as an analysis technique is the identification of all possible compounds whose absorption spectrum can
interfere with the absorption feature being analyzed. It is therefore imperative that the operator has as
complete knowledge as he can of the compounds that are expected to be present during the measurement
period. The assumption made here is not only that this knowledge exists bur that reference spectra for
all the potentially present compounds also exist. The operator should be aware that absorption spectra
from unexpected chemical compounds may appear during the data acquisition phase and that these must
be accounted for in the analysis for the most accurate data.
8.3.3 Additional Sections Referenced. Section 9.2 is referenced in this section.
8.3.4 Methodology. While FT-IR spectra can in fact be acquired before it is known exactly what
wave number region to use for any particular gas in the analysis, this is never a good idea. If on-line
analysis is a requirement then the wave number regions must be selected first.
When starting, the operator must be aware that this is likely to be an iterative procedure and some wave
number regions may be rejected in the process. The selected wave number region can be quite narrow,
but there are some dangers in selecting a very narrow region. It is best if the operator at first selects the
entire absorbing band structure, using the end points as the 1% absorbance values relative to the peak.
If narrowing the wave number region becomes necessary, the operator should be aware that the selected
wave number region should always encompass the largest possible range in absorbance.
This procedure starts by an examination of the absorption spectrum of the target gas and selection of the
absorption feature that has the highest absorption coefficient and is outside the strong absorption regions
of water vapor and carbon dioxide. The wave number region to use is the region that is covered by the
entire peak under study. It is not necessary to include any wave number region whose relative
absorbance is less than 1 % of the peak. The absorption coefficient is calculated and the expected
absorbance is calculated by using the anticipated concentration at the site. This absorbance is compared
to the noise equivalent absorbance obtained from Section 9.2. If the expected absorbance is not 3 times
higher than the noise equivalent absorbance, then that wave number region should be rejected. It is likely
that if the anticipated absorbance does not meet this criterion then measurement of that particular gas will
have to be rejected because the remaining absorption coefficients will be too small.
If that test is passed the procedure continues. The absorption features of all the other gases known to be
present at the measurement site are then compared to the target gas for possible interferences. If the total
interference is thought to be too strong, the wave number region is rejected and the process is started
over with a different absorption feature. If all of the features in the target gas are rejected, the gas
concentration cannot be measured by FT-IR.
To calculate the absorption coefficient for any particular feature, the operator must measure the
absorbance of the feature being used at the peak of the feature. This is done by using the reference
spectrum. Then by using the expression a = A/cl the absorption coefficient a is determined. The A in
this expression is the peak absorbance measured from the reference spectrum, and the d is the
concentration-path length product also obtained from the reference spectrum.
Once a is obtained, an estimate of the peak absorbance can be made as follows. Use the expression
A = acl, where the c is the anticipated concentration at the site and / is the anticipated path length. The
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Pa°e 16-11
D-67
-------
Method TO-16 VQCs
c and the / used here must have the same units (e.g., ppm, meters) as the reference spectra. The
calculated A is then the anticipated peak absorbance at the site.
To judge whether a particular gas is a possible interfering species, a comparison of the absorption features
must be made. This is initially done by simply comparing the spectra of all the other compounds known
to be present at the site with the absorption feature under study. If any overlap between the two spectra
exists the gas must be considered an interfering species.
To judge the strength of any interfering species the absorption coefficients of the interfering species must
be calculated as above and an estimate of the anticipated absorbance at the measuring site made. In
measuring the correct absorbance to use for the interfering species, the operator should use the highest
absorbance of the interfering species spectrum within the overlapping wave number region. Note that
the actual peak absorbance of the interfering gas may very well fall outside the overlap region. The
absorption coefficient and the anticipated absorbance at the site for the interfering species is then
determined exactly as described above. Then the fractional overlap of the spectra must be determined
and the estimated impact on the actual measurement is made.
To determine the fractional overlap, measure the wave number region of the overlap in the spectra and
then divide that by the entire wave number region selected for the target gas. The measurement should
be made by using the 1 % relative absorbance wave numbers of the interfering species.
To estimate the strength of the overlapping absorbing feature, multiply the fractional overlap of the
interfering species by the anticipated absorbance at the site (for the interfering species). Then divide that
product by the anticipated absorbance at the site for the target gas. The total interfering strength is then
the sum of all the strengths for the individual interfering species.
The classical least squares technique is a very powerful tool for analysis and can determine the presence
of very small quantities of gas in the presence of a fairly large interference. While no hard rule can be
given, the operator should be concerned and at least attempt to find another wave number region if the
total strength of the interfering species is more than 5 times the anticipated absorption of the target gas
at the site.
If the operator rejects the wave number region, then the process is repeated with the next highest
absorption coefficient and so on until a suitable wave number region is found. The operator is advised
to record a table of these wave number regions in a permanent notebook for the specific gases that he is
working with. It is likely that these calculations will have to be done only once for any particular target
gas.
8.3.5 Procedure
8.3.5.1 Examine the reference spectrum of the target gas and select the absorbance feature with
the highest absorbance that is outside the strong absorbance of water and carbon dioxide.
8.3.5.2 Record the wave number region using the relative 1 % absorbance peaks as the end points.
8.3.5.3 Calculate the absorption coefficient a using the peak absorbance.
8.3.5.4 Calculate the anticipated absorbance at the field site using the a from Section 8.3.5.3, the
concentration anticipated at the field site, and the path length anticipated at the field site.
8.3.5.5 Compare the result of Section 8.3.5.5 with 3 times the RMS noise calculated from
Section 9.2.
Page 16-12 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-68
-------
VOCs Method TO-16
8.3.5.6 Compare the absorbance spectra of all the gases known to be present in the atmosphere
at the site. Record any overlaps with the selected region.
8.3.5.7 Calculate the following.
83.5.7.1 The absorption coefficient a of the interfering species using the peak absorbance in the
overlap region.
83.5.7.2 The fractional overlap.
83.5.7.3 The anticipated strength of the interfering species.
83.5.7.4 The sum of the interfering strengths.
83.5.8 Accept or reject the wave number region.
83.5.9 If necessary repeat Sections 8.3.5.1 through 8.3.5.8 with the next highest absorbance peak.
8.4 Generation of a Background Spectrum
8.4.1 Purpose
8.4.1.1 This section instructs the operator on how to generate a background spectrum that can then
be used as 70 in Beer's law. A background spectrum can be generated by several methods. These
methods are (a) the upwind background, (b) the cross-path background, (c) the zero target gas
background, and (d) the synthetic background. The first three backgrounds are generally used with no
further processing, but the synthetic background has to be made. Each is briefly discussed below.
8.4.1.2 Since the synthetic background is the only one that requires computer processing, it is the
one for which the actual steps are given in this procedure.
8.4.2 Assumptions
8.4.2.1 The wave number regions for the analysis have previously been chosen.
8.4.2.2 Field spectra have been acquired, and one of them is to be used for a synthetic
background.
8.4.2.3 Software is available that allows a synthetic background to be made.
8.4.3 Additional Sections Referenced. No other sections are referenced.
8.4.4 Methodology. In the derivation of Beer's law, one calculates how much the intensity of the
infrared source diminishes as the energy traverses an absorbing medium. To calculate the concentration
of the gas, the operator must compare the initial intensity obtained in the absence of the target gas with
the measured intensity obtained when the target gas is present. This initial intensity is called the
background, and it is the response of the instrument to the infrared source in the absence of any
absorbance due to the target gas. A variety of phenomena are responsible for the shape of the
background curve. A number of these phenomena are related to the instrument, but the predominant
atmospheric process that shapes the background is the absorbance due to water vapor.
If any absorbance due to the target gas remains in the background, the absolute values of the gas
concentration cannot be measured. In this case, only values relative to the concentration in the
background will be obtained.
The upwind background is one that is predominantly used at smaller sites, where it is fairly simple to
move the system from one side to another. Once the wind direction is known, the system is set up so
that the path is along the upwind side and a spectrum that is to be used as a background is acquired. This
procedure is normally done twice a day (morning and evening), and these spectra are generally used as
backgrounds with no further processing.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page
D-69
-------
Method TO-16 VOCs
The argument is made that an upwind background will contain only target gas concentrations from
upwind sources. The remaining downwind field spectra will then give correct values from gases at the
site alone. While this is a valid argument, it is not a very strong argument for the use of an upwind
spectrum, and any variability in the upwind sources may erroneously be interpreted as variability of the
target gas concentration at the site itself. Also to be noted when usjng such a spectrum is that it may not
be valid for the entire time period the operator intends. As the water vapor concentration changes, the
curvature of the baseline in the spectra changes also. This will give rise to high error bars (as calculated
from classical least squares) and to variability in the target gas concentration that follows that of water.
When that occurs, this background (or any background) may no longer be valid.
The cross-path background is taken with an optical path placed along one side of the site and with the
wind velocity parallel to it. This background is generally used when the geometry of the site allows it,
and it supposedly has no target gas concentration. This type of background may also pose some
unwanted problems. If the wind is very light, then the gases from the site can indeed diffuse into the
optical path. Target gas concentrations from elsewhere may be present, and interpretation may present
the same problems described above.
Some researchers have obtained a background spectrum by simply waiting long enough for the gas
concentration to go to zero. This method will not work for gases that are always present in the
atmosphere, such as methane or carbon monoxide. This procedure may be used if there is sufficient time
in the program for the waiting period and if real-time analysis is not an immediate requirement, but it
is not clear whether the water vapor concentration will be in a satisfactory range. Also, the operator
should not expect any one background to remain valid for more than a few days, and then a new
background must be obtained.
If the measurement program merely requires a "yes" or "no" response to the question of whether a
compound is present, then any spectrum that is taken may possibly be used for a background. If
possible, the operator should use a spectrum that has a minimum of the target gas, but that is not
necessary if the analysis software allows negative numbers. (Note that it is the difference of the
concentration in the two spectra that will be measured.)
If, however, the absolute values of the concentration of the target gas in the optical path are required by
the measurement program and no applicable spectrum can be found that is void of the target gas
absorption, a synthetic background must be used. A synthetic background is one that is made from a
single beam field spectrum, and it may have some of the target gas in it. Once a field spectrum has been
selected, a new baseline is made to replace all the absorbance features in the wave number region used
for analysis with a new curve that resembles the instrument baseline as closely as possible. This new
baseline is made by connecting the data points along the original baseline with straight line segments, or
by some other appropriate fining procedure, thereby removing any absorbance features. The difficulty
with this method is knowing where the baseline actually is. No points within any absorbance feature of
the original spectrum can be used. A portion of a field spectrum and the synthetic background made
from it by connecting the data points with very short straight line segments is illustrated in Figure 4. The
original field spectrum has absorption lines due to water vapor in it.
Selecting the points for the baseline for a synthetic background may be quite difficult when large wave
number regions are used or when the curvature of the baseline is high. This is a problem with the wave
number region used for the analysis of ozone, for example.
Page 16-14 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-70
-------
VOCs Method TO-16
8.4.5 Procedure
8.4.5.1 From the set of available spectra, select one spectrum by using the following criteria.
8.4.5.1.1 The target compound concentration should be near a minimum.
8.4.5.1.2 The interfering species concentration should be at a minimum.
8.4.5.1.3 The vapor pressure concentration should be in the mid range of water vapor
concentrations during the period for which the background is to be used.
8.4.5.1.4 The return intensity at 987 cm"1, 2520 cm"1, and 4400 cm"1 should be normal for this
instrument and for the particular path length used.
8.4.5.2 Once the candidate spectrum has been selected, use the available software to create a
synthetic background.
8.5 Production of a Water Vapor Reference Spectrum
8.5.1 Purpose. This section instructs the FT-IR operator on how to create a water vapor reference
spectrum from a single beam field spectrum. Absorption due to water vapor represents an interference
to the spectral region of the target gas, and these interferences must be accounted for in whatever analysis
routines that are finally used. Water vapor presents the predominant absorption features in the spectra
acquired by the FT-IR, and the operator can expect it to interfere with the target gas spectrum. It is
essentially impossible to create a water vapor reference spectrum in the laboratory by using a cell because
the concentrations required are not normally attainable, and measuring the amount of water vapor in the
cell is very difficult. Therefore, the water vapor reference spectrum has to be made from the acquired
field spectra. Fairly large changes in the atmospheric concentrations of water vapor can occur rapidly,
and that generally implies that a new water vapor reference spectrum has to be created.
8.5.2 Assumptions
8.5.2.1 The wave number regions for the analysis of the remaining spectra have previously been
chosen.
8.5.2.2 Field spectra have been acquired, and one of them is to be used for a water vapor
reference spectrum.
8.5.3 Additional Sections Referenced. Activities and evaluations performed in Section 8.4 are
referred to in this section.
8.5.4 Methodology. The first step in the process of creating a water vapor reference is the selection
of a single beam spectrum from the set available. The selection is based on a number of criteria. During
the process, any absorbance due to the target gas must be subtracted from the water vapor reference.
This can be done in a number of ways, but until the operator gains some familiarity with the FT-IR
analysis process, it is best to do this by starting with a spectrum that contains a reasonable amount of the
target gas and any other interfering species. In this way it may be possible for the operator to see the
absorbance feature and do the subtraction interactively. Otherwise, the concentration of the target gas
needs to be measured by using an analysis routine and then subtracted by using the library reference
spectrum.
There is also some argument that can be made to specifically acquire a spectrum with a large number of
scans and to use that as a water vapor reference spectrum. The large number of scans ostensibly gives.
a smaller noise value. This argument is not generally true with the FT-IR systems because the calculated
RMS noise is not usually generated by the system electrical noise. The majority of the calculated RMS
noise seems rather to be the result of slight changes in the water vapor concentration and other
atmospheric constituents from one spectrum to the other.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-15
D-71
-------
Method TO-16 VOCs
From the set of available spectra, one spectrum must be selected by using the following criteria.
• The target compound concentration should not be near a minimum. As the operator gains more
experience at creating a water vapor reference he may want to minimize the target gas absorption
if possible.
• The interfering species concentrations should not be near minima.
• The vapor pressure concentration should be in the mid range of water vapor concentrations during
the period for which this particular water vapor reference spectrum is to be used. It should be
remembered that many of the water vapor lines may be saturated as far as the instrument response
is concerned. Tha{ implies that the time period that can be covered with any one water vapor
spectrum must be carefully chosen. However, at the present time no explicit guidance concerning
the length of time that a single water vapor reference is valid can be given. Perhaps the best advice
is to compare the curvature of the baselines of the single beam spectra. If that is changing rapidly,
a new water vapor reference spectrum may have to made.
• The return intensity at 987 cm" , 2520 cm" , and 4400 cm"^ should be normal for this instrument
and for the particular path length used. Any spectrum that has been acquired in foggy or rainy
conditions should not be used.
The last criterion is included as a check to determine that the instrument is operating correctly.
Once the candidate spectrum has been chosen, it must be turned into an absorption spectrum by using the
background spectrum created in Section 8.4.
The new water vapor absorbance spectrum must now be analyzed for the presence of absorbance due to
the target gas. To accomplish this, the normal analysis procedure can be used if an older version of the
water vapor reference spectrum already exists. It is likely that using the older water vapor reference will
result in somewhat higher error bars from the analysis. At the present time this can be ignored. The
results of this analysis should be zero, but it can give a positive result if there is an absorbance due to
the target gas in the newly created water vapor reference. If a positive value exists then that amount of
the target gas must be subtracted from the water vapor reference spectrum. The exact procedure to use
for the subtraction process will depend on the software that the operator has.
If no other water vapor reference exists, the following procedure must be used. A set of 15 pairs of
spectra must be acquired with the FT-IR. They should be taken so that no time elapses between them.
They should be acquired with the same number of scans and the same resolution as the newly created
water vapor reference spectrum. The individual 15 pairs are used to create 15 absorbance spectra. These
spectra should not contain any of the target compound absorbance because they have been taken back-to-
back, and it is hoped that each will contain the same amount of the target gas absorption. These spectra
must then be analyzed for the target compound by using the newly created water vapor reference.
The average value of the results of this analysis should be zero. If it is not but some positive or negative
bias exists, some amount of the target compound absorbance is still in the water vapor reference
spectrum.
Page 16-16 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-72
-------
VOCs Method TO-16
There are two possibilities to consider if a bias exists. The first is that the baseline of the newly created
water vapor reference is not quite correct, and the second is that some of the target compound must be
subtracted from the newly created reference spectrum. (This can give rise to either a negative or a
positive bias.) At the present time no procedure exists to correct for curvature of the baseline. If the
operator decides that baseline curvature is the primary problem, then there is little that he can do to
correct the problem.
If a bias exists that is not from a baseline curvature then the operator must subtract some of the target
gas from the newly created water vapor reference. If an interactive software mode for subtraction exists,
the subtraction can be done in an interactive mode using the target gas reference spectrum as the
subtrahend. If an interactive software mode is not available, the target gas reference can be used as
follows. The target gas reference spectrum can be multiplied by an appropriate factor and the result
subtracted from the newly created water vapor reference. The path length at which the water vapor
reference spectrum was acquired is known and the target gas concentration is known in parts per million
from the analysis above. The reference spectrum absorbance is given in terms of parts per million
meters. So the operator must divide the absorbance of the spectrum by the path length in meters and by
the ratio of the concentrations (reference/calculated). The resulting spectrum can then be subtracted from
the created water vapor reference.
Repeat the analysis procedure and this process until the target gas concentration is zero.
8.5.5 Procedure
8.5.5.1 Select the single beam spectrum that is to be used for a water vapor reference using the
criteria listed above.
8.5.5.2 Create an absorbance spectrum using the appropriate background spectrum.
8.5.5.3 Analyze the newly created water vapor reference for the target gas.
8.5.5.4 If necessary, subtract the proper amount of the target gas absorption from the water vapor
reference.
8.5.5.5 Reanalyze the water vapor spectrum.
8.5.5.6 Repeat Sections 8.5.5.3 through 8.5.5.5 until the target gas concentration is zero.
8.6 Subtraction of Stray Light or Black Body Radiation
8.6.1 Purpose. This section instructs the operator how to subtract the stray light or black body
radiation measured by the instrument from the field spectra. This procedure can be used by operators
using either the monostatic or the bistatic instrument configurations. The subtraction for either
configuration is performed by using single beam spectra.
8.6.2 Assumptions. Assumptions. For both the stray light component and the black body radiation
component measurement the instrument must be operating at its equilibrium conditions. That is, the FT-
IR must have been allowed to warm up. As long as the operating conditions are not changing rapidly,
the spectra should be acquired by using a large number of scans so as to provide a good signal-to-noise
ratio. Since these spectra have to be subtracted from the field spectra, noise will be added to the analysis,
and a longer acquisition time minimizes the electrical noise. Acquiring data for up to one half hour is
satisfactory. Not much is gained in the signal-to-noise ratio by acquisition times longer than that.
8.6.3 Additional Sections Referenced. No other sections are referenced.
8.6.4 Methodology. The procedure for subtracting stray light is primarily to be used for the removal
of a spurious signal from FT-IR instruments using the monostatic configuration with a second beam
spliner. While it is possible to have scattered light that gives rise to unwanted signals in instruments
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-17
D-73
-------
Method TO-16 VOCs
using the other geometric configurations, this component is very difficult to measure and is considered
to be a difficulty that the manufacturer has to deal with. This type of stray light subtraction will not be
discussed further in this method. Instrument manufacturers strive to have the stray light as small as
possible compared to the intensity returning from the retroreflector, but to remove it all can be a
formidable task, and it should therefore be measured and subtracted. It is fairly simple to show
mathematically that, whatever percentage of the return intensity the stray light intensity represents, that
percent error will be carried through to the final result in the analysis. The presence of stray light can
sometimes be detected visually in the single beam spectrum as is shown in Figure 3. Therefore, it has
to be subtracted from the spectra if the errors in the data are to be minimized. The intent of the specific
program may indicate it is not necessary to subtract the stray tight spectrum from the field spectra; an
example is when only the identification of compounds is necessary.
Once the stray light intensity is known and measured it should not change unless some component of the
optical system is changed or reoriented. Therefore, the stray light spectral subtraction can easily become
part of the routine analysis. Since the stray light component is generated inside the instrument, its
intensity is not path-length dependent. This means that the stray light will change its intensity relative
to the return intensity as the path length changes. It can easily be measured by simply slewing the
instrument away from the retroreflector and acquiring a spectrum.
The need to subtract the black body radiation arises only in bistatic systems that have an unmodulated
source at one end of the physical path. It is convenient to think that the black body radiation comes from
the fact that the field of view of the receiving telescope is larger than the angle that the infrared source
subtends; therefore, the instrument allows the infrared energy from the surroundings into the system.
This is only partially true, and if the instrument is at the same temperature as the surroundings, the black
body radiation can be thought of as coming entirely from the instrument enclosure. That is because all
black body radiators at the same temperature radiate the same amount of energy per unit area. Therefore,
the easiest way to measure the black body spectrum is to turn the source off and then acquire a spectrum.
There is an additional problem with the black body radiation curve that occurs when the instrument is
pointed at the sky. When this situation occurs it is very likely that there will be an emission spectrum
superimposed on the black body curve. The emission spectrum arises from several atmospheric gases
and is quite variable. Even the smallest amount of cloud cover will dramatically change the intensity of
this spectrum. That fact makes it almost impossible to subtract the emission spectrum totally. It is
advisable to avoid pointing the instrument so diat it has the sky in the field of view. If that cannot be
avoided, the operator should be aware that higher than normal errors can occur in the data in the region
below about 1050 cm'*.
Small changes in the ambient temperature (10 °K) are not thought to be significant in the black body
radiation, and thus one spectrum should be usable for an extended period. These spectra should be
subtracted from the field spectra after the single beam spectra have been obtained. If the interferograms
are subtracted and the single beam is then calculated, a different result is obtained. The reason for that
is not fully understood at this time.
Page 16-18 Compendium of Methods for Toxic Organic Air Pollutants January' 1997
D-74
-------
VQCs Method TO-16
8.6.5 Procedure
8.6.5.1 Measure the stray light in the instrument by slewing the instrument off the retroreflector.
8.6.5.2 Subtract this spectrum from each single beam field spectrum before proceeding with the
analysis.
8.6.5.3 Measure the black body radiation spectrum by turning the source off.
8.6.5.4 Subtract this spectrum from each single beam field spectrum before proceeding whb the
analysis.
8.7 Generation of an Absorbance Spectrum
8.7.1 Purpose. This section instructs the operator on how to generate an absorbance spectrum from
the field spectra and an appropriately chosen background spectrum.
8.7.2 Assumptions. The following assumptions are made.
8.7.2.1 An appropriate background spectrum is available.
8.7.2.2 All the field spectra have been converted to single beam spectra.
8.7.2.3 All the field spectra have been corrected for stray light and the black body radiation if
necessary.
8.7.3 Additional Sections Referenced. No other sections are referenced.
8.7.4 Methodology. Beer's law is the underlying physical law that governs the way the least squares
analysis is performed. Mathematically, Beer's law is written as I(v) = IQ(v~)t\p(-aCL). In order to
calculate C, the concentration of the gas in the atmosphere, one must divide by 70 and take the logarithm
of the result. That gives ln(/o/7) = aCL. The spectrum described by the term InC/o/T) is called the
absorbance spectrum. The FT-IR analysis is actually done by using the logarithm to the base 10, but this
is normally transparent to the operator.
All software packages that are available for least squares analysis allow the generation of an absorbance
spectrum. The operator is generally asked to supply the background spectrum, but then the process is
mathematically performed by the computer. It is important to understand that some correction may be
necessary to the field spectra before they are converted to absorbance spectra.
8.7.5 Procedure. Use the available software to create the absorbance spectra.
8.8 Correction for Spectral Shifts
8.8.1 Purpose. This section instructs the operator on how to align two spectra so as to minimize
the errors involved with spectral shifts.
8.8.2 Assumptions. The field spectra have been acquired and are in the single beam format. A
water vapor reference that is to be used for analysis is available. A background spectrum has been
prepared and is available for use. In order to check for a shift between the field spectrum and the
reference spectrum, an absorbance spectrum must be used if the reference spectrum is an absorbance
spectrum.
8.8.3 Additional Sections Referenced. No other sections are referenced.
8.8.4 Methodology. There are three ways that a spectral shift will affect the FT-IR data analysis.
The first is when a spectral shift between the field absorbance spectrum and the water vapor reference
spectrum exists. The second is when a spectral shift between the field absorbance spectrum and the
library reference spectrum for the target gas exists. The third is when a nonsynthetic background is used
and a spectral shift exists between the background and the field spectra. A spectra! shift compared to the
instrument may also be noticed when new reference spectra are purchased or produced on an instrument
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-19
D-75
-------
Method TO-16 VQCs
other than the one used for data acquisition. The first two of these comparisons are done using
absorbance spectra, but the third must be done with single beam spectra.
When a synthetic background is used, any spectral shift between the field spectrum (single beam) and the
background spectrum (single beam) is irrelevant. That is because the synthetic background generation
process does away with all spectral features of interest.
A question arises as to what sort of a shift is really important to the analysis. Some researchers discuss
this in terms of absolute quantity of wave numbers. This is not really satisfactory because then apparently
small shifts are important for some spectral features while at other times they are not. If a Gaussian
shape is used to describe the absorption line shape, then it is possible to show mathematically that when
the absorption feature of interest shifts by about 10% of the line width (FWHH), a 5% error occurs in
the least-squares analysis. If a Lorentzian line is used to describe the actual line shape, the shift can be
about 15% of the line width (FWHH) before a 5% error occurs when least-squares analysis is used.
Experimentally, if a 5 % error is acceptable, it is only seldom that line shifts will be important. However,
if a 1 % error is all that is allowed by data quality objectives, then the same calculations show that a 0.5 %
shift (FWHH) of the line is all that can be tolerated. This really implies that wave number shifts will
probably not be important when broad absorption features (such as presented by ozone) are used but will
be crucial for narrow absorption features (such as presented by carbon monoxide). The predominant
spectral feature in the FT-IR open path field spectra is water vapor, and the pressure-broadened lines of
water have a line width (FWHH) of about 0.2 cm~ *. Since water is the predominant feature, the errors
produced by the classical least-squares technique will be primarily caused by how well water is handled
in the analysis. That means that water vapor must always be checked for shifts.
Experience has shown that when a spectral shift occurs, the magnitude of the shift is different in the C-H
(2900-3000 cm'1) stretch region than it is in the fingerprint region. This implies that all line shifts are
caused by some change in the interferometer and/or the system optics. If that is truly the case, then the
shift is linear in wave number, and a linear correction must be applied when the correction is made
throughout the field spectrum. Some computer software automatically identifies a wave number shift and
then shifts the entire spectrum by the proper amount. If that software is available it should be used.
The best place to determine whether a shift has occurred is in the low-wavelength or high-wave-number
end (in the region of the C-H stretch) of the spectrum. It may also be possible to automatically
determine during the acquisition phase whether a shift has occurred and then shift each individual
spectrum as it is being acquired. To do that, some known spectral feature present in every spectrum must
exist. Thus it may be possible to select some water vapor line that is present in all the spectra covering
a particular time period and compare all the spectra with that particular line. A shift of that kind
guarantees that ail the field spectra are aligned one to the other but does not automatically guarantee that
the field spectra and the reference spectra will be aligned. At any rate, at the present time no such line
has been agreed upon, and it may not be possible to select a single line for all occasions.
If shifting software is not available, two problems are presented to the operator. The first is how to
recognize a shift and the second is how to correct for it over the entire wave number region.
Recognizing that a shift has occurred can be facilitated by subtracting one spectrum from another. If a
small (less than the line width) shift has occurred, the difference will appear as an "S"-shaped curve.
This kind of curve is closely related to the first derivative of the line shape if the shift is small.
Page 16-20 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-76
-------
VOCs Method TO-16
Determining the absolute magnitude of the shift can be a difficult task, and no simple mathematical
relation exists between the features of the S-shaped curve and the magnitude of the shift. At the present
time, the best estimate of the magnitude of the shift is obtained from measuring the difference of the peak
positions of the two lines. However, this is best done on spectra that have been interpolated to increase
the number of data points. Or, if the operator so chooses, he may zero fill the interferogram by a factor
of 2 or 4 in order to increase the number of data points.
Since the correction for a shifted spectrum is most likely linear in wave number, the shift must be done
in steps if appropriate software is not available. A shift between the individual library reference spectra
and the field absorbance spectra can be overcome because the library spectra can be individually shifted.
It may also be possible (depending on the software available) to rename the water vapor reference
spectrum so that there are two or three of them, each with its own shift, and then do all the analysis
simultaneously. The same procedure can be used to overcome a shift between the field spectra and the
background spectrum when a synthetic background is not used. However, if the shift is small (less than
the data point spacing) but significant, then all the spectra may have to be interpolated or zero filled to
correct for the shift.
8.8.5 Procedure (applicable when shifting software is not available).
8.8.5.1 Subtract the two spectra and examine the residual for an S-shaped curve. Do the
background and the field spectra first because these have to be done with single beam spectra.
8.8.5.2 Determine the magnitude of the shift by comparing the peaks of the individual lines.
8.8.5.3 Shift one spectrum with respect to the other. This will have to be done in the target gas
analysis regions and may have to be done several times.
8.8.5.4 Create absorbance spectra from the field spectra and the background and repeat
Sections 8.8.5.1 through 8.8.5.3.
8.8.5.5 Perform a correction for shift to the water vapor reference spectrum, the reference
spectrum, and the background if necessary. The field spectra should not be shifted, as this requires the
most time.
8.9 Analysis of the Field Spectra for Concentration
8.9.1 Purpose. This section instructs the user on the procedures used for the analysis of FT-IR
absorbance spectra in order to produce gas concentration values.
8.9.2 Assumptions. The spectra have been convened to absorbance spectra and all changes and
corrections listed in the above sections have been made to them. A set of reference (library) spectra that
includes the target gas, the interfering gases, and a water vapor reference is available for use. A software
package that is capable of performing least squares analysis on the spectra is available.
8.9.3 Additional Sections Referenced. No other sections are referenced.
8.9.4 Methodology. There are a number of ways to analyze the spectra in order to obtain
concentration data. These include peak height or peak area analysis, spectral subtraction, partial least
squares, iterative least squares, principal component analysis, etc. While these methods are all usable,
this procedure uses classical least squares as described mathematically by Haaland and Easterling (13).
The use of classical least squares requires that the spectra be prepared in a specific way for the analysis
to work efficiently and effectively. Thus the majority of Compendium Method TO-16 is concerned with
preparation of the spectra.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-21
D-77
-------
Method TO-16 VOCs
It is likely that, when the other techniques cited above are used, the data will have to be prepared in a
different manner. Under those conditions the steps of this procedure that deal with spectral preparation
are not usable.
Whatever software is available to perform the classical least-squares technique, it must be able to perform
the analysis of a single target gas in the presence of interfering species. It is only seldom that the range
of wave numbers used for the analyses will be free of absorbances due to interfering species. This is
particularly true of water, and the analysis routines must be able to perform a multiple linear regression
of the field spectra.
There are a number of software packages that are in use that perform classical least-squares analysis of
the spectra. These all have somewhat different user interfaces and operating conditions, but in all cases
the mathematical algorithms are transparent to the user. Therefore, the software packages are not
described in detail here. Since the classical least-squares analysis is a multiple linear regression, it must
have certain items available for it to function. The items that are common to all available analysis
packages include the target gas reference spectrum, the background (or /Q) spectrum, the water vapor
reference spectrum, and whatever interfering gas reference spectra are necessary. Most software
packages are, however, only available with the FT-IR instrument itself. The primary concern for this
procedure is that the analysis itself follows the classical least squares described mathematically by Haaland
and Easterling (13).
8.9.5 Procedure. The individual steps in this section are dependent on the specific software
available to the operator. Since the individual packages are not described here, the specific steps required
for any one package are not either.
8.10 Post-Analysis Review of the Data
8.10.1 Purpose. The purpose of this section is to provide the operator with a way to check the data
for possible problems. This procedure primarily makes use of plotted data in the form of the
concentration of one gas plotted against the other and of time sequence plots. There is one statistical
determination that can be used to determine if correlations exist between pairs of data. The primary tool
used here is for the operator to look for trends in the data where none should exist. The specific tests
of the data are described below.
8.10.2 Assumptions. The only assumption is that all of the spectra have been analyzed by use of
the least-squares analysis software.
8.10.3 Additional Sections Referenced. No other sections are referenced.
8.10.4 Methodology. The operator should make several plots of the concentration data. The first
should be a set of plots of target gas concentration versus time. These plots should be examined for any
expected trends in time. For example, ozone in rural areas generally follows a diurnal pattern with a
minimum at about 0600 hours and a maximum at about 1500 hours. The concentration values should
not go negative to any great extent; although around zero concentration the values may go slightly
negative, the average value over time should be zero. Suppression of negative values should never be
done in the analysis because then a zero average can never be achieved. If values go negative with time
in a regular fashion, then something is amiss with the data. The most likely case is that there is a small
remaining absorbance due to the target gas in the water vapor reference spectrum. If the concentration
values are much higher than the anticipated values, there may also be a problem with the water vapor
reference spectrum. In this case there may have been too much of the target gas absorbance subtracted
Page 16-22 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-78
-------
VOCs Method TO16
from the water vapor reference. If that is so, the water vapor reference should be fixed and then the data
reanalyzed.
Plots should be made of the target gas concentration versus the water vapor. If the variability of the
target gas and the water vapor are correlated and this is not expected, the water vapor reference spectrum
must, in most cases, be corrected.
The next step is to plot the concentration values of those gases whose concentrations are expected to be
correlated. This includes any gases that are derived from the same source. If the variability of these
gases is not correlated, the data must be carefully examined for the cause. There are no good guidelines
to judge what is causing that problem, but a nonlinear response of the instrument for one of the gases is
a possibility. If that is suspected, the operator must carefully examine the QA data for possible clues.
Another check of the instrument can be made by analyzing the spectra for ^0. Nitrous oxide is present
naturally in the atmosphere with a concentration very close to 300 ppb. The variability in this
concentration should be less than ±10%. If this is not the case then all the data must be suspect.
Another gas that is always present in the atmosphere is methane. The variability of methane can be fairly
large, particularly in the proximity of landfills. That means it is somewhat more difficult to use as a
quality check of the data but it can still be used. The value of the atmospheric concentration of methane
should never fall to less than about 1.7 ppm.
If the FT-IR instrument is a bistatic one and there is any possibility that the instrument was admitting
energy from the sky when the black body radiation measurement was made, there might be a problem
with the observed detection limits. If that occurs, it is possible that the analysis is flawed because of
emission spectra in the black body radiation.
Another check for the quality of the data can be obtained by examining the errors calculated by the least-
squares analysis routine. If there is an abrupt change in the relative error and no obvious reason such
as an abrupt change in the water vapor concentration, it may be that a new interfering species, not
accounted for in the analysis, has been measured.
Once these checks have been made on the data, the operator must follow the data quality checks that have
been written for the specific program that is being studied.
8.10.5 Procedure.
8.10.5.1 Plot the data as a function of time and check for unexpected trends.
8.10.5.2 Plot the target gas data concentration as a function of water and determine if the
variability is correlated.
8.10.5.3 Determine whether N2O and CH4 have been correctly measured.
8.10.5.4 Determine whether correlation of the data exists where correlation is expected.
8.10.5.5 Review all the QA/QC data taken in compliance with the specific data quality objectives.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-23
D-79
-------
Method TO-16 VQCs
9. Quality Assurance
9.1 Summary
The section provides guidance to the operator in determining how well the FT-IR sensor is operating.
While this section is labeled "quality assurance", it is by no means adequate to serve as a quality
assurance project plan or program plan. Project and program plans are meant to address the specific data
quality objectives of a monitoring program, and the final use of the FT-IR data and cannot be adequately
covered in this document.. Some of the procedures are limited in scope because a satisfactory procedure
has not been developed at this time.
9.2 The Determination of Method Noise or Method Noise Equivalent Absorption
9.2.1 Purpose. The purpose of this section is to allow the operator to determine the method noise.
This determination should form part of the routine quality assurance checks made of the instrument. It
should be made at least once a day for extended programs and every time the instrument is moved or
otherwise changed. This procedure is used to judge whether the instrument is operating properly but not
as a gauge of the quality of the data.
9.2.2 Assumptions
9.2.2.1 This procedure assumes that spectra have been acquired with the same operating parameters
(number of co-added scans, resolution, etc.) as the field spectra. The one exception is that the spectra
used to determine the method noise should be taken so that no time elapses between them.
9.2.2.2 It is also assumed that software exists that will allow this determination to be made
automatically by computer.
9.2.3 Additional Sections Referenced. No other sections are referenced.
9.2.4 Methodology. Instrumental noise is generally considered to be the random fluctuations in the
recorded signal. That is not exactly true for the FT-IR system when the data are acquired along a long,
open path. Evidently, the time required to allow small but measurable changes in the gaseous
atmospheric constituents is short compared to the normal acquisition time of the spectra. Because of that,
when two spectra are used to create an absorbance spectrum there is a variability in the result that is not
electronic noise alone. This is defined here as the method noise. It is important because it cannot easily
be done away with and will contribute to the error of the measurement.
The determination of method noise uses the statistical quantity called the RMS deviation. The
mathematical routine normally used for this calculation performs a linear least-squares fit Qinear
regression) using the data points over a specified wave number region and calculates the RMS deviation
from that line. The RMS deviation is defined as the square root of the sum of the differences squared
divided by the quantity N-2. The number N is the total number of data points. The differences are
calculated by taking the difference between the actual data point and the line; they are then squared and
added.
The actual range of wave numbers that can be used changes with resolution, but the number of data
points does not. The number of data points used should be 80 points. Thus for a 1-cnf1 resolution, the
range of wave numbers is 40, because the instrument acquires a data point every half resolution unit.
Since this measurement is considered to be the determination of an instrument parameter, the wave
Page 16-24 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-80
-------
VQCs Method TO-16
number region or regions should be chosen to minimize the effect of water vapor. The water vapor
concentrations along the path are known to change rapidly, and that will perhaps cause most of the
variability in the signal.
The two single beam spectra that are used to measure the noise should be taken without any time lapse
between them. These two spectra are then used to create an absorbance spectrum. Which of the two that
is used as the so-called background is irrelevant. Three wave number regions are then used for this
determination. For this procedure, the regions are based on a l-cm'1 resolution and are 968-1008,
2480-2520, and 4380-4420 cm'1, respectively. Other regions may be used, but the operator should try
to cover the range of wave numbers that are being measured. The 80 data points used in the
measurement should also be adhered to. This data should then be recorded and plotted on a quality
control chart for comparison purposes.
9.2.5 Procedure
9.2.5.1 Record two spectra with the same operational parameters that will be used for the
acquisition of the field spectra. Do not allow any time to elapse between these spectra.
9.2.5.2 Create an absorbance spectrum by using either of the two spectra taken in Section 9.2.5.1
as a background.
9.2.53 Analyze this absorbance spectrum for the RMS deviation in the three wave number regions
968-1008 cm'1, 2480-2520 cm'1, and 4380-4420 cm'1.
9.2.5.4 Record this data in a notebook and plot it on a quality control chart.
9.3 The Measurement of the Return Beam Intensity
9.3.1 Purpose. This section provides guidance to the measurement of the return beam intensity in
the case of the monostatic system or the intensity of the IR source at the FT-IR in the case of the bistatic
system. This procedure needs to be done only once as long as the detector or the infrared source does
not change.
9.3.2 Assumptions. In order that these measurements be realistic, the stray light component or the
black body radiation should be subtracted from the spectra. This means those measurement results should
be available to the operator or should be made in conjunction with this measurement.
9.3.3 Additional Sections Referenced. Refer to Section 9.4, Measurement of Stray Light, and
Section 9.5, The Measurement of Black Body Radiation, if applicable.
9.3.4 Methodology. The return beam intensity determines the operational signal-to-noise ratio of
the FT-IR system. This intensity is a variable and depends on the path length chosen, the water vapor
in the atmosphere, and other atmospheric conditions. The primary atmospheric conditions that make the
return beam intensity change are fog, rain, snow, and sleet. Of these, fog has by far the largest effect.
Another cause for a change in the return beam intensity is pollen in the atmosphere. This happens in the
spring in areas where there are a large number of pine trees. Finally, for the monostatic geometry, which
uses a retroreflector, condensation on the mirror can make dramatic changes in the return beam intensity.
There are also instrumental causes of changes in the return beam intensity but they are beyond the scope
of this document.
For these reasons, it is prudent to include in a quality assurance program the measurement of the return
beam intensity. If the return energy has been degraded by an unacceptable amount, the operator must
change the length of the path. Whether the return is acceptable or not is dependent on the data quality
objectives from the quality assurance program plan.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-25
D-81
-------
Method TO-16 VQCs
This procedure is separated into two parts. The first is a procedure for measuring the return beam
intensity as a function of path length. The second is the measurement of return beam intensity as a
function of time.
There are two reasons to measure the return beam intensity as a function of path length. The first is to
determine when the energy becomes intense enough to saturate the detector. The second is to determine
when the infrared energy becomes too small to measure. These measurements then determine
experimentally the minimum and maximum usable path length. There are a number of reasons why the
return beam intensity should be monitored as a function of time. The primary one is that the return beam
intensity will change according to varying weather conditions. The operator must become familiar with
the magnitude and the rapidity of these changes.
9.3.5 Procedure
9.3.5.1 Return Beam Intensity as a Function of Path Length.
93.5.1.1 Place the light source or the retroreflector at a short distance, say 25 meters, from the
detector.
93.5.1.2 Align the system to maximize the return signal.
93.5.13 Record a spectrum and convert this spectrum to a single beam spectrum.
93.5.1.4 Record the intensity levels in the 987-cm"1 region and in the 2,500-cm"1 and the
4,400 cm"1 regions. The reason the wave numbers are not given specifically is that the operator should
select a maximum in the baseline return intensity in these regions.
93.5.1.5 Examine the detector cutoff region at about 650 cm"1. If a dip occurs in this region or
the baseline is elevated above zero, then the detector is already saturated.
93.5.1.6 If there is no indication of saturation, move the light source or the retroreflector so that
the distance separating it and the detector is smaller. Repeat Sections 9.3.5.1.1 through 9.3.5.1.5.
93.5.1.7 Continue this process by cutting the distance in half until the single beam spectrum
exhibits saturation as described above in the 650-cm"1 region. Record this distance. This distance
represents the minimum path length that can be used with this particular instrument without altering the
instrument.
93.5.1.8 Next, move the light source or retroreflector to a distance of 100 m.
93.5.1.9 Realign the instrument to maximize the signal.
93.5.1.10 Record a spectrum and convert it to a single beam spectrum.
9.3.5.1.11 Record the intensity levels at the same wave numbers as used above.
93.5.1.12 Repeat Sections 9.3.5.1.8 through 9.3.5.1.11 by increasing the path length in 50-m
increments until the intensity levels no longer change. For the monostatic geometry mode, this will occur
when all the energy being recorded comes from the stray light in the instrument. For the bistatic mode,
the return signal will diminish to zero in the 4000-cm region and then will evolve into the black body
radiation spectrum.
93.5.1.13 Plot a graph of the return intensity versus path length.
9.3.5.2 Measurement of the Return Beam Intensity as a Function of Time. At least once every
day of operation the return beam intensity should be recorded at the wave number regions given above.
More frequent measurements should be made when the atmospheric or other conditions listed in
Section 9.3.4 occur. The atmospheric conditions should also be recorded. Water vapor plays an
important role in the recorded beam intensity so that the partial pressure of water should also be
calculated and recorded. A continuous plot of these data should be made showing the intensity as a
function of time. The graph should include notations for the various atmospheric conditions listed above.
Page 16-26 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-82
-------
VOCs Method TO-16
9.4 The Measurement of Stray Light
9.4.1 Purpose. The monostatic FT-IR systems are prone to having stray light in the instrument.
To obtain the best possible accuracy, this stray light component must be subtracted from the field spectra
before the analysis is performed. This section describes how to measure the stray light component and
is applicable to only the monostatic geometries that modulate the beam with the interferometer before the
infrared energy is transmitted along the path and that use a second beam splitter to direct the beam.
There are other sources for stray light that arise from overfilled optical components. These are a problem
for the manufacturer and are not addressed in this document.
9.4.2 Assumptions. The primary assumption for this procedure is that the FT-IR system being used
is of the monostatic geometry. The system has been operating sufficiently long to be past any warm-up
periods and it is using the same resolution as when it is acquiring the normal field spectra.
9.4.3 Additional Sections Referenced. No other sections are referenced.
9.4.4 Methodology. The easiest way to measure the stray light with a monostatic system that
modulates the beam before the energy is transmitted along the path is to simply slew the system away
from the retroreflector. Once this is done, the operator should acquire a spectrum using a large number
of co-added scans. If the time required to acquire this spectrum cannot be at least 4 times the length of
time to acquire the field spectra, he should then use the longest time possible. The issue here is one of
electronic noise, and the electronic noise should diminish as the square root of the time needed to acquire
the spectrum. Changes in the atmospheric constituents play no role in this measurement.
This spectrum should be saved as a single beam spectrum with an appropriate name, and it must be
subtracted from the single beam field spectra before they are converted to absorbance spectra.
A second way to measure the stray light is to cover the receiving telescope with some opaque, non-
reflecting material. Any material that is reflecting acts as a mirror and will give erroneous readings.
Any material that is not opaque will allow some of the beam returning from the retroreflector to be
transmitted to the detector. This method is not recommended.
This measurement must be done at the beginning of operation and every time the instrument is altered
in any way. For those programs that are short-term field programs, the measurement should be made
at the beginning of each field program.
9.4.5 Procedure.
9.4.5.1 Set up the instrument in exactly the same way as it will be used to acquire field spectra.
9.4.5.2 Slew the transmitting telescope off the retroreflector so that there is no beam return signal,
9.4.5.3 Acquire a spectrum.
9.5 The Measurement of Black Body Radiation
9.5.1 Purpose. FT-IR systems generally have a field of view larger than the solid angle that the
light source or the retroreflector subtends at the far end of the path. The bistatic systems (or those that
do not transmit the beam through the interferometer before it transmitted along the path), therefore, admit
radiation to the detector from the surrounding background. These systems also respond to any radiation
coming from the instrument itself (the instrument is also a radiator of energy). This radiation is
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-27
D-83
-------
Method TO-16 VQCs
commonly referred to as the black body radiation and it must be subtracted from the single beam spectra
before the analysis is performed. This procedure describes how to measure that radiation.
9.5.2 Assumptions. The instrument must be set up in the same manner and with the same
background that will be in its field of view during the acquisition of the field spectra.
9.5.3 Additional Sections Referenced. No other sections are referenced.
9.5.4 Methodology. The FT-IR systems available today use some form of a heated element as a
source of infrared energy. These elements generally have a temperature in the vicinity of 1500"K. The
terrestrial surroundings in which the FT-IR operates generally have a temperature around 300eK. All
things above absolute zero radiate energy according to their temperature and have a very well known
energy distribution in wave number. The distribution of energy peaks at a wave number that is
temperature dependent with the cooler body having a peak at lower wave numbers. Also, the energy
distribution of a cooler body is lower in intensity at all wave numbers than the distribution of a hotter
body. The question arises of what the ratio of intensities is of these two sources. This ratio at the peak
of the 300 degree source is about 5. That is, the surroundings can represent about 20% of the energy
of the source. Therefore, it must be subtracted from the spectra before the analysis is performed.
The simplest way to acquire a black body spectrum is to set up the instrument in exactly the same way
it will be run to take the field spectra. A spectrum should then be acquired with the light source turned
off. This spectrum should be saved with an appropriate name.
This spectrum does not appear to change dramatically when the instrument is pointed at terrestrial targets
such as buildings or trees. Nor does it change dramatically with slight changes in the ambient
temperature (± 10°K). A 10% change in temperature will shift the peak by 10%, and that may become
important. Remember, however, that a 10% change in temperature is about 30°C. The black body
spectrum does, however, change dramatically if the sky is included in the instrument's field of view. In
this case an emission spectrum appears from the atmosphere, and this is very difficult to handle. The
black body spectrum can also change dramatically if hot sources other than the primary light source are
allowed into the field of view of the instrument. The operator is advised to take precautions so that these
conditions are avoided.
In order to minimize the noise introduced by the subtraction process, the number of scans used to acquire
the spectrum should be large. An acquisition time of more than 15 min is probably excessive. It is
prudent to run such a spectrum at least once every day during the study. These spectra should be
investigated for changes, particularly when there are large swings in temperature. This is possible during
the early fall, when the temperature can range from cold at night to quite warm in the daytime.
9.5.5 Procedure
9.5.5.1 Set up the FT-IR along the same path that will be used to acquire the field spectra.
9.5.5.2 Acquire a spectrum over a long acquisition time with the infrared source off.
9.5.5.3 Store the spectrum with an appropriate name.
9.6 The Determination of the Detection Limit
9.6.1 Purpose. The purpose of this routine is to provide the operator with a mechanism for
determining the detection limits for the various gases. The definition of the detection limit is given here
as the minimum concentration of the target gas that can be detected in the presence of all the usually
encountered spectral interferences.
Page 16-28 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-84
-------
VOCs Method TO-16
9.6.2 Assumptions. The instrument is operating with the same parameter settings as those used for
collecting the field spectra. That is, the path length, resolution, number of co-added scans, and the
apodization function are the same in both cases. If the instrument has an ancillary gas cell, this must be
empty.
9.6.3 Additional Sections Referenced. No other sections are referenced.
9.6.4 Methodology. The detection limit of the FT-IR systems is a dynamic quantity that will change
as the atmospheric conditions change. The variability of the target gas, water vapor, and all of the other
interfering species concentrations contributes to the variability of this measurement. Some researchers
have suggested that the 70 spectrum used to create the absorption spectrum when measuring the detection
limits be the same as that used for the field spectra. However, that cannot be done if a synthetic
background is used since the field spectra are expected to contain some quantity of the target gas. If any
other arbitrary background is used the measurement will certainly reflect the variability of the target gas,
at least. To overcome most of the effects of this problem, the operator should use spectra whose
acquisition times are no longer than about five minutes. If the field spectra are acquired at shorter times
then the shorter time should be used. If the field spectra are acquired at longer times because the
anticipated variability of the target gas is small, then it is appropriate to use the longer times.
The detection limit as determined in this procedure is the result of a calculation using a set of
15 individual absorption spectra. The 16 individual single beam spectra used for this determination are
acquired in 5-min intervals and no time is allowed to elapse between them. The absorption spectra are
then created by using the first and the second single beam spectra, the second and the third, and the third
and the fourth, and so on until the 15 absorption spectra are obtained. These absorption spectra are
analyzed in exactly the same way that all field spectra are to be analyzed and over the same wave number
region. The analysis should result in a set of numbers that are very close to zero because most of the
effects of the gas variability have been removed. The numerical results should be both positive and
negative and for a very large set of data should average to zero. Three times the standard deviation of
this calculated set of concentrations is defined to be the detection limit.
There is reason to believe that this procedure gives the most optimistic (lowest) value for the detection
limit because it removes most of the effects of the interfering species. However, the other suggested
procedures seem to introduce as much uncertainty, and this procedure may actually be used for further
diagnostics of the post-analysis review of the data (see Section 8.10).
9.6.5 Procedure.
9.6.5.1 Acquire a set of 16 single beam spectra in exactly the same manner that will be used for
the field spectra.
9.6.5.2 Use the first spectrum as a background to create an absorbance spectrum from the second
spectrum.
9.6.5.3 Use the second spectrum as the background and create an absorbance spectrum from the
third spectrum.
9.6.5.4 Continue this process until all 15 absorbance spectra have been created.
9.6.5.5 Analyze each of the spectra for the target gas concentration.
9.6.5.6 Calculate the standard deviation of the set of concentration values.
9.6.5.7 Multiply the result of Section 9.6.5.6 by 3 to obtain the detection limit.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-29
D-85
-------
Method TO-16 VOCs
9.7 The Determination of Precision
9.7.1 Purpose. Precision is a measure of the FT-IR system's ability to make repeatable
measurements when challenged with the same sample. This section provides guidance to the operator
on how to make that determination for some gases.
9.7.2 Assumptions. The FT-IR system has the capability for installing a gas cell in the beam so that
the entire beam passes through it. This is something that the manufacturer has to build into the design
of the system and is not under the control of the operator. While the measurements are being made, the
instrument is operating in the same way that it is used to collect the field spectra.
9.73 Additional Sections Referenced. No other sections are referenced.
9.7.4 Methodology. The precision with which a measurement is made with the FT-IR instruments
is, at the present time, very difficult to measure. The best method that has been suggested is one that
uses a cell of some sort that is filled with a high concentration of gas and is then placed in the beam.
However, this process is quite error-prone and it has not been shown to work well with a mixture of
gases. A second difficulty is that it cannot be used for all the gases that can be potentially measured with
the FT-IR. The primary reason is that the concentration of the gas in the cell has to be high in order to
produce a measurable absorption. Many gases have a vapor pressure that is too low to achieve these
concentrations. No procedure has been established for making these measurements of polar compounds.
Additionally, not all the commercially available instruments have at the present time been designed to
accept a cell in an appropriate position of the optical path.
However, for those instruments and for those gases that can be measured, the procedure is as follows.
A cell whose length is short compared to the path length is filled with a high concentration of gas. The
cell is placed in the infrared beam so that all of the energy passes through the cell. Then a set of spectra
is acquired and these are converted to absorption spectra. These absorption spectra are analyzed for the
target gas. The relative standard deviation of this set of measurements is given as the precision.
This procedure is also quite similar to the procedure for the measurement of accuracy. The measurement
of precision, however, does not require an exact knowledge of the concentration of the gas, but rather
the gas concentration must remain constant. Thus the gas concentrations used can be made up in the field
at a lower cost to the monitoring program.
Determining the precision of the FT-IR monitoring system is complicated by the fact that the
measurements are made over an open path in the atmosphere. It cannot be assumed that the concentration
of the various atmospheric gases will be constant in time, and this fact will impact the precision
measurements. This procedure calls for the precision measurements to be made by using the same path
length that is generally used for acquiring field spectra. Therefore, the precision will vary in time and
will be dependent on the variability of not only the target gas but also the variability of the interfering
species. The precision measurement described is therefore a method precision and includes all of the
parameters that must be considered in the field spectra analysis.
The cell can be filled in a number of ways, but the preferred way is to use a gas of the appropriate
concentration from a prepared cylinder that has been purchased for this purpose. The proper mixture can
be calculated as follows:
• The absorption coefficient of the gas can be calculated from the reference spectrum by using a =
A/cl, where A is the absorbance at the peak of the reference spectrum and d is the
Page 16-30 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-86
-------
VOCs Method TO-16
concentration-path length product, which is supplied with the reference spectrum for the reference
gas.
• Next, the desired absorbance when the cell is filled is selected. This can be set at 0.05.
• Then c is calculated from c = A/a/, where A = 0.05, a is the absorption coefficient calculated
above, and / is the length of the cell in meters if the reference gas has a concentration-path length
product in parts per million per meter.
• The concentration calculated above has units of parts per million if the concentration-path length
product for the reference gas has units of parts per million per meter. This is the concentration to
use when purchasing a cylinder of gas. The fill gas of the cylinder must not absorb in the infrared,
and the gas preferred for this is nitrogen.
Before the target gas is introduced into the cell, the cell should be flushed with nitrogen until at least five
volumes of the cell have passed through it. At the present time the preferred method for introducing the
target gas is with a flowing system. The gas should remain flowing during the measurement.
9.7.5 Procedure.
9.7.5.1 Calculate the appropriate concentration for the target gas and obtain a cylinder of that
concentration.
9.7.5.2 Set up the instrument as it will be used to acquire the field spectra.
9.7.5.3 Place the cell in the instrument and flush it with dry nitrogen so that at least five volumes
of the cell have passed through it.
9.7.5.4 Flow the target gas through the cell, and after three volumes of the cell have passed
through, acquire a set of 15 spectra.
9.7.5.5 Analyze these spectra for the target gas.
9.7.5.6 Express the relative standard deviation of this set of concentrations as the precision.
9.8 The Determination of Accuracy
9.8.1 Purpose. Accuracy is a measure of the ability of the FT-IR to measure a known concentration
of gas. This procedure may allow the operator to determine the accuracy of the FT-IR measurements
for some gases. This measurement is very difficult to make and no exact procedure has been accepted.
9.8.2 Assumptions. The FT-IR must have the capability for installing a gas cell that is short
compared to the path length in the instrument so that the entire infrared beam passes through it. This
must be included in the manufacturer's design of the instrument, and whether or not the cell can be
placed in the beam is not under the control of the operator. The measurements for accuracy should be
made with the instrument operating in the same way as it is when acquiring normal field spectra.
9.8.3 Additional Sections Referenced. No other sections are referenced.
9.8.4 Methodology. The general procedure to be used for the determination of accuracy is
essentially identical to the procedure for the determination of precision. The difference is that for the
measurement of accuracy the concentration of the gas in the cell must be known. Obtaining this
knowledge poses some special problems, and preparation of the sample gas by the individual operators
is not recommended at this time. Rather, whenever possible a cylinder of prepared gas should be
purchased; for convenience, this prepared mixture is called the reference gas for the rest of this
procedure. However, the vapor pressure of some gases is too low to allow the purchase of appropriate
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-31
D-87
-------
Method TO-16 VOCs
concentrations. Even if a cylinder is purchased, there is some difficulty with knowing what the
concentration in the cell is, particularly for the polar compounds.
If a cell is to be used for this measurement then the first step is to calculate the concentration that is
required. It is anticipated that the accuracy of the measurement is dependent on the concentration that
is being measured. Therefore, the operator must make some judgement of what that concentration is to
be. To obtain the concentration in the cell, the operator must multiply the anticipated concentration by
the ratio of the path length used for the monitoring program to the cell length. Thus if the path length
to be used in the acquisition phase is 100 m and the cell length is 20 cm, then the operator must multiply
the anticipated concentration by 500 to get the required concentration of the reference gas in the cell.
Once the proper mixture of gas has been obtained, the operator must introduce it into the cell. At the
present time it is recommended that the gas should be flowed through the cell continually during the
measurement. Before the measurement is attempted, the gas should be allowed to flow through the cell
until at least five volumes of the cell have passed through it.
At the beginning of this measurement the cell should be flushed with dry nitrogen and then a spectrum
should be acquired. The reference gas is then flowed through the cell and a second spectrum acquired
while the gas is flowing. The cell should then be flushed with dry nitrogen again and a third spectrum
recorded.
The average value of the target gas concentration found from the first and third spectra is subtracted from
the value determined for the target gas from the second spectrum. This value is then used as the recorded
value for the measurement. This procedure is repeated five times in a day, and the average value of these
five measurements is used as the accuracy measurement. The percent accuracy is then defined as the
average value found above divided by the known concentration of the cylinder gas value times 100. This
value should be recorded and plotted on a control chart made for that purpose.
If a flowing system is used, the flow rate must be small so that there is no measurable pressure change
in the cell. Flow rates of a few cubic centimeters per minute are acceptable and would require no
measurement of the pressure. When the cell is purged to remove the target gas, the volume of purge gas
used should be at least 5 times the volume of the cell.
The procedure described here has not been studied in depth, and little written material exists in the
literature. Questions such as what the material of the lead lines are to be made of, whether the pressure
must be measured in the cell, and whether the lines have to be heated have not been answered at this
time. It is also not clear whether this procedure can be used with a mixture of gas or if only a single
species must be used at a time. It seems possible that, in the future, a procedure using the water in the
atmosphere can be used for this measurement. Absorbance due to water is in every important part of the
spectrum that is used with FT-IR measurements, and it will be in every spectrum. Water can also be
measured independently with techniques other than the FT-IR so that a verification step can be performed.
However, the use of water has not been explored at all.
9.8.5 Procedure.
9.8.5.1 Calculate the required concentration of the reference gas and obtain a cylinder with that
concentration.
9.8.5.2 Set up the FT-IR with the same operating conditions used to acquire the field spectra.
9.8.5.3 Install the cell in the beam if necessary and flush it with dry nitrogen.
Page 16-32 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-88
-------
VOCs Method TO-16
9.8.5.4 Acquire a spectrum.
9.8.5.5 Flow the reference gas through the cell so that at least five volumes of the cell pass
through it.
9.8.5.6 Acquire a second spectrum with the reference gas flowing.
9.8.5.7 Flush the cell with dry nitrogen again and acquire a third spectrum.
9.8.5.8 Analyze all three spectra for the target gas by using the same background as used for die
field spectra.
9.8.5.9 Find the concentration of the reference gas from the result of analyzing the second
spectrum minus the average value of the first and third spectra.
9.8.5.10 Repeat Sections 9.8.5.3 through 9.8.5.9 five times in any one day of operation.
9.8.5.11 Determine the percent accuracy as the average value of the five measurements divided
by the known concentration of the reference cell times 100.
9.9 The Measurement of Resolution
9.9.1 Purpose. The purpose of this procedure is to provide the operator with a means for measuring
the resolution of the FT-IR instrument.
9.9.2 Assumptions. The spectra used to make this determination have been acquired with the same
instrumental parameters as those used for the field spectra. Particularly, the apodization function and the
path length must be the same.
9.9.3 Additional Sections Referenced. No other sections are referenced.
9.9.4 Methodology. The resolution of the FT-IR is an important parameter in that it determines the
specificity of the measurements. The instrument resolution does not exhibit dramatic changes from day
to day and needs to be measured infrequently. However, whenever any change is made to the instrument
optics, including the light source, the resolution must be remeasured. The resolution can also change
when the path length changes if the instrument does not have an appropriate field stop to clearly define
the field of view regardless of the optical path length. If that is the case, the resolution should be
measured at whatever path lengths are used. The FT-IR resolution is also dependent on the apodizatioa
function that is used when single beam spectra are created from interferograms, and if more than one
apodization function is used then the resolution should be measured for each. The operator needs to be
aware of the instrument resolution for a number of reasons. The spectra from two instruments cannot
be compared if the resolutions are not the same. The use of reference spectra at resolutions different
from that of the instrument creates problems with accuracy. Subtracting one spectrum from another with
different resolutions is also a problem. The manufacturers of these devices list the nominal resolution,
but a listed resolution of, for example, 1 cm"^ should not be interpreted as an exact number.
To measure the resolution, an absorption spectrum must be used. An absorption line that is narrow in
comparison to the instrument's line function must be used, and the spectral line used must be a single
line. If changes in the instrument resolution occur, they should be noticeable in the high-wave-number
region first.
Six primary atmospheric constituents are present in every spectrum. They are water vapor, methane.
carbon dioxide, nitrous oxide, ozone, and carbon monoxide. Of these, only the absorption features of
water vapor and carbon monoxide can be used to measure the instrument resolution. If the path length
is great enough and the water vapor concentration is large enough, then the atmospheric constituent
deuterated water can also be used.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-33
D-89
-------
Method TO-16 VQCs
In addition to these, absorption features from other gases in high concentrations in conjunction with a
short cell can be used. The important feature of any line that is selected for resolution measurements is
that it be a single line and be narrow compared to the instrument's nominal operating resolution. Thus
methane cannot be used because the lines are not single lines. Whatever feature is chosen, it must not
be impacted by any interfering species, as this has the same effect as having double lines. The absorption
features of ammonia or hydrogen chloride can be used. HC1 is actually a good choice because it absorbs
in the high-wave-number region. However, it is not generally present in high enough quantities in the
atmosphere to be measured in every spectrum.
There are a number of lines that can be used in the water vapor spectrum that can be used for this
measurement. They are at the wave numbers 1014.2, 1149.46, 1187.02, and 2911.88. It should be
noted that many of the water lines are already saturated as far as the instrument response is concerned
at a vapor pressure of 3 torr. So any line used must be checked to make sure it is not saturated. For
carbon monoxide there is at least one line at 2168.9 that can be used. These lines should easily be
observed in spectra that have been taken with path lengths greater than 100 m (total).
The resolution for the FT-IR is defined as the full width at half maximum (FWHM) for either of the these
lines. Thus to determine the instrument's resolution, an absorbance spectrum must be created with a
synthetic background. The operator needs to have a large number of data points across the line in order
to make this measurement, and it should be remembered that the system takes only two data points per
nominal resolution element. The best way to create this absorption spectrum is to record an
interferogram and then zero fill by at least a factor of 4 before computing the Fourier transform. If that
is not possible, then the absorbance spectrum must be interpolated to increase the number of data points.
The absorbance at the peak must be measured, and any non-zero baseline value must be subtracted from
that measurement. The result of this subtraction is the peak height. Then the entire width of the line at
one-half the peak absorbance is measured in wave numbers. This is the required measure of the
resolution of the instrument.
9.9.5 Procedure.
9.9.5.1 Obtain an interferogram with the FT-IR operating at the same path length as will be used
for the acquisition of the field spectra.
9.9.5.2 Zero fill the interferogram by at least a factor of 4.
9.9.5.3 Perform the Fourier transform on the interferogram.
9.9.5.4 Create an absorbance spectrum using a synthetic background.
9.9.5.5 Isolate one of the lines and measure the peak height.
9.9.5.6 Subtract any non-zero baseline measurement.
9.9.5.7 Measure the full width of the line at one half the absorbance measured in Section 9.9.5.6.
This is the resolution.
9.10 The Determination of Nonlinear Instrument Response
9.10.1 Purpose. The FT-IR instrument can respond nonlinearly to changes in the light intensity for
several reasons. There are two instrumental conditions that must be guarded against, and these are
discussed here. The first is that the electrical gain is set too high, and this can cause the analog-to-digital
(A/D) converter to be saturated. The second is that the light source itself is too intense, and this causes
the detector response to become nonlinear. This procedure is intended to give the operator a means for
determining when either of these conditions exist.
Page 16-34 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-90
-------
VQCs Method TO-16
9.10.2 Assumptions. The instrument is operating under the same conditions as it will be to acquire
the field spectra.
9.10.3 Additional Sections Referenced. No other sections are referenced.
9.10.4 Methodology. A nonlinear response can be caused by excessive source intensity or amplifier
gain. All of the FT-IR systems that are used for remote sensing use A/D converters to convert the analog
detector signal to a digitized form. Most use either a 16 bit or an 18 bit converter, and that defines the
range of voltages that can be monitored. If the source intensity and amplifier gain combination is too
high, then the A/D convener can be saturated. This manifests itself as a sudden drop in the signal being
recorded when the source or the retroreflector is moved closer to the detector. When this happens, the
system gain must be lowered, if that possibility exists, or the path length must be changed.
The second type of nonlinear response is somewhat more difficult to determine. This occurs if too much
light falls on the detector. The detector converts the incident light photons to an electrical signal. There
is a limit for how many photons can be convened to electrons, and when this limit is exceeded the
detector response becomes nonlinear.
There may also be a nonlinear response from the fact that HgCdTe detectors exhibit nonlinear behavior
in their response to infrared energy. This circumstance is not covered here and should be corrected by
the manufacturer.
In everyday operation, the easiest way to detect the second kind of nonlinearity is to examine the portion
of the single beam spectrum at wave numbers below the detector cutoff. This is in the 650-680-cnf *
region for most HgCdTe detectors. If a dip below zero occurs in that region or if the signal is above
zero at wave numbers below that region, the system's response may be nonlinear.
There are two ways to check the system's response. Both involve the use of screens to diminish the light
intensity while the response is being viewed. If the screens have meshes that reduce the intensity by
known amounts, the response should be diminished by that amount also. If the instrument responds
differently, the system is nonlinear.
Wire screens can be purchased in a number of mesh sizes, and the mesh size determines how much light
will get blocked. Plastic screens should not be used because they may exhibit selective absorption.
Aluminum screening that is used for window screening is satisfactory but may not reduce the intensity
enough. It is best to use screens of different mesh when using the procedure described below rather than
two layers of screening with the same mesh.
The following procedure needs to be done only if the operator suspects that the system is operating In
a nonlinear way.
9.10.5 Procedure.
9.10.5.1 Set the FT-IR system up as it will be used for acquisition of the field spectra.
9.10.5.2 Move the source or the retroreflector to twice the original distance.
9.10.5.3 Examine the signal. If a sudden increase in the signal strength occurs, then the A/D
converter is saturated.
9.10.5.4 With the source on and the retroreflector at the distance used for the field spectra, acquire
a single beam field spectrum and examine the intensity in the detector cutoff region. If a dip occurs, the
detector may be saturated.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-35
D-91
-------
Method TO-16 VOCs
9.10.5.5 If the dip that is described in Section 9.10.5.4 occurs, insert a wire screen in the beam
so that it covers the entire beam and record the signal level.
9.10.5.6 Insert a second screen in the beam and record the signal again. If the screens are .the
same, each should diminish the beam in the same ratio. If that does not happen, the system is nonlinear
in response and the infrared energy must be decreased by some means such as increasing the path length,
closing the iris in the instrument, etc.
9.11 The Determination of Water Vapor Concentration
9.11.1 Purpose. It is suggested that the water vapor content in the atmosphere be monitored
independently of the FT-IR measurements. This is not an individual procedure like the preceding
portions of this method in that it does not explain the siting criteria for making water vapor
measurements. It is rather a discussion as to why the measurement is important.
9.11.2 Assumptions. There are no assumptions about the FT-IR system associated with this process.
9.11.3 Additional Sections Referenced. No other sections are referenced.
9.11.4 Methodology. Absorption due to water vapor is the predominant feature in the spectra
acquired by the FT-IR remote sensor. It also seems to be one of the most difficult compounds to deal
with in the analysis. There are measurable changes in the observed water vapor from spectrum to
spectrum. Data from a local airport weather service is not satisfactory to understand the changes and
their effects on the analysis. Large and abrupt changes in the water vapor content can be expected.
When that occurs it is likely that the background spectrum and the water vapor reference spectrum will
have to be remade. But the only way to know that these changes have occurred is to measure the water
independently.
Some argument can be made that the water vapor concentration can be obtained by simply adding water
to the list of analyzed gases. However, it is not simple to make that measurement. Many of the water
vapor lines are very strongly absorbing when the vapor pressure is 3 or 4 torr. The atmospheric vapor
pressure in most areas is at least 5 times higher than that. That makes line selection for analysis quite
problematic.
Since the water vapor can easily be measured in a continuous fashion it seems prudent to make the
measurement independently of the FT-IR. One post-analysis check of the data is to look for a correlation
between the concentrations of the target gas and water vapor. To accomplish that, the operator must
determine what the water vapor concentration is. The following discussion describes a way of doing that.
The water vapor concentration can be obtained by measuring the relative humidity and the ambient
temperature. These values, along with the Smithsonian psychrometric tables, are then used to calculate
the water vapor concentration. The psychrometric tables can be found in the Handbook of Chemistry and
Physics (14), which is published yearly.
There are solid-state devices available today that allow reliable measurements of the relative humidity and
the ambient temperature. These devices give results to within a few percent of relative humidity and a
few tenths of a degree for the temperature. The operator will need a way to record the output from these
devices. This can be accomplished with a data logger that allows for multichannel, multiday recording.
The sensors can be placed anywhere along the path but must be shielded from the sun. A complete
description of how to configure the placement of these devices is well outside the scope of this document.
Page 16-36 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-92
-------
VQCs Method TO-16
For a complete discussion of these measurements, the operator should consult the following document:
Quality Assurance Handbook for Air Pollution Measurements, Volume IV—Meteorological Measurements
(15). Once the water vapor concentration is known, it should be plotted as a function of time and men
compared with the target gas concentration as discussed in the procedure for post-analysis data checking
(see Section 8.10). The operator should pay particular attention to the periods where abrupt changes in
the water vapor occur.
10. References
1. Russwurm, G. M., and Childers, J.W., FT-IR Open-Path Monitoring Guidance Document, U. S.
Environmental Protection Agency, Research Triangle Park, NC, EPA/600/R-96/040, April 1996.
2. Pfeiffer, H. G., and Liebhafsky, H. A., "The Origins of Beer's Law," J. Chem. Educ., Vol.
28:123-125, 1951.
3. Lothian, G. F., "Beer's Law and Its Use in Analysis," Analyst, Vol. 88:678, 1963.
4. Champeney, D. C., Fourier Transforms and Their Physical Applications, Academic Press, London,
1973.
5. Nicolet Analytical Instruments, FT-IR Theory, Nicolet Analytical Instruments, Madison, WI, 1986.
6. Griffiths, P. R., and deHaseth, J. A., Fourier Transform Infrared Spectrometry, John Wiley and
Sons, New York, 1986.
7. Halliday, D., and Resnick, R. Fundamentals of Physics, Wiley and Sons, New York, 1974.
8. Stone, J. M., Radiation and Optics, McGraw-Hill, New York, 1963.
9. Tolansky, S., An Introduction to Interferometry, John Wiley and Sons, New York, 1962.
10. Calvert, J. G., "Glossary of Atmospheric Chemistry Terms (Recommendations 1990)," PureAppL
Chem., Vol. 62(11):2167-2219, 1990.
11. Long, G. L., and Winefordner, J. D., "Limit of Detection: A Closer Look at the IUPAC
Definition," Anal. Chem., Vol. 55(7):712A-724A, 1983.
12. Rennilson, J. J., "Retroreflection Measurements: A Review," Appl. Opt., Vol. 19:1234, 1980.
13. Haaland, D. M., and Easterling, R. G., "Application of New Least-Squares Methods for the
Quantitative Infrared Analysis of Muiticomponent Samples," Appl. Spectrosc., 36(6):665-673,1982.
14. Handbook of Chemistry and Physics, CRC Press, Cleveland, OH.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants Page 16-37
D-93
-------
Method TO-16 VOCs
15. Quality Assurance Handbook for Air Pollution Measures, Volume IV—Meteorological
Measurements, U.S. Environmental Protection Agency, EPA-600/R-94-038b, Research Triangle
Park, NC, May 1994.
Page 16-38 Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-94
-------
VOCs
Method TO-16
Detector
r
•••MB* •- *
Receiving
Optics
IR Path
^_ I
t
t
^^ %
%
*
«
TrwtsfTvtftnQ
Opfes
« t
-•-•"IR
Absorbing Medium
B
"-»•«: -^'
IR Path
i
\ \
*V'
t
i
0
Trans nwong
Optics
.**
IR
Source
Detector
Figure 1. The bistatic configuration of an FT-IR system.
January 1997 Compendium of Methods for Toxic Organic Air Pollutants
Page
D-95
-------
Method TO-16
VOCs
..*:-"•— .%........
^r /'Transmiting IRjPath
. . .•*"•• ""»»
tt •' . ;
• \ i^_
\ :::
Translating
Retroreflector
Retroreflector
\
Absorbing Medium
Transmitting
Optics
Receiving •
Optics '
IR Path
Intarfaromoier
;*
•• IR
Source
Detector
Additional
Beam Splitter
Transmitting/
Receiving
Optics
^
/
Interferometer
^
tt
Source
Detector
Figure 2. The monostatic configuration of an FT-IR system.
Page 16-40
Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-96
-------
VOCs
Method TO-16
CO
o
a
9
m
O)
JC
CO
400
1000
2000 3000
Wave Number (cm-')
4000
Figure 3. Single-beam spectrum acquired by using a monostatic system and a 414-m path.
fNote: S indicates stray light.]
January 1997 Compendium of Methods for Toxic Organic Air Pollutants
D-97
Page 16-41
-------
Method TO-16
VOCs
Original Spectrum
900 927 954 981 1008 1035 1062 1089 1116
Wave Number (crrr1)
Figure 4. Synthetic I0 spectrum for an FT-IR absorbance.
[Note: The peak at 1110 cm'1 has intentionally been left in as afducial point.]
Page 16-42
Compendium of Methods for Toxic Organic Air Pollutants January 1997
D-98
-------
National Specialty Workshop
Technical Tools for Air Toxics Assessment:
* f.
t
m State and Local Participatory Programs #
- National Volatile Organic Monitoring Contract
«•
•»
. Speciated NMOC
• Urban Air Toxics '
• Carbonyls <
»Photochemical Assessment Compounds
> Dave-Paul Dayton ]
Eastern Research Group
Momsville NC ]
5
National Volatile Organic Monitoring Contract
I
What the Program Includes
Compounds
Relationship to Toxics
Options and Costs
How Can I get Access
D-99
-------
National Volatile Organic Monitoring Contract
Program Supports Three Distinct Components *
>
- Urban Air Toxics Measurments {
4
*
- Speciated NMOC (and Nonspeciated NMOC) :
f,
- Photochemical Assessment Monitoring for Volatile Organics;
•;
a
National Volatile Organic Monitoring Contract
- Each Program Component Provides
- Sampling and Analysis Support
- Sampling Equipment (UATMP/SNMOC/Carbonyls)
- Site Installation and Training
- Site Support
- Central Laboratory Analysis
- Quarterly and Annual Reports
-AIRS-AQMS data entry
D-100
-------
National Volatile Organic Monitoring Contract
\
m Central Laboratory Analysis Includes: -
; "
- Compendium Method TO-14 for Air Toxics (thru 96 program) >
- Compendium Method TO-15 for Mr Toxics (starting 97 program) f
• Polar Fuel Additives *
- Compendium Method TO-11 for Carbonyls *
- Compendium Method TO-12 for NMOC :
- Speciated Nonmethane Organic Compounds
- Photochemical Assessment Monitoring Compounds J
National Volatile Organic Monitoring Contracit
»
Urban Air Toxica
-49 compounds from TO-14
-13 carbonyl compounds
- 9 additional "polar" compounds from TO-15 :
ft
• includes 4 fuel additive compounds !
D-101
-------
National Volatile Organic Monitoring Contract
Speciated NMOC '.
":
- 80 Hydrocarbon species including ,
• Saturated Hydrocarbons '
. Unsaturated Hydrocarbons ,
• Single ring aromatics }
• Includes 12 UATMP target compounds \
i
\
National Volatile Organic Monitoring Contract
Photochemical Assessment Monitoring
- QA/QC Support
• Retention Time Standards
• Site Setup Ongoing Support for automated GC monitoring
- 57 Hydrocarbon species from PAMS Technical Assistance
Document
• Includes 11 UATMP Target Compounds ;
-13 of the Compendium Method TO-11 carbonyl compounds j
j
D-102
-------
I.
I
National Volatile Organic Monitoring Contract
Program Costs ;
- State Accessable through 105 Grant Funds
- Independent Access through Non-Federal Funds \
- Base Sites Include: {
. UATMP I
. SNMOC |
.PAMS {
- Site Options are available for each base j
• Toxics ;
• Carbonyls •
.SNMOC *
*
National Volatile Organic Monitoring Contract
- Program Access and Contacts
- Neil Berg - EPA Delivery Order Manager
- Kathy Weant - EPA Project Officer
- Dave-Paul Dayton - ERG Program Manager
- Tim Hanley - ERG Site Coordinator
D-103
-------
Table 1
NMOC, UATMP, and PAMS Networks Base Program
and Associated Costs (1997)*
o
Jk.
Line If cm
1.0
2.0
3.0
NMOC Base Site Support
• (Frequency, 4 months, 5 days per week, I sample per day
ctjiiuls l>2 samples)
• Canister Cleaning and Shipping
• Instrument Certification and Installation
• Travel
• NMOC Analysis
• QA/QC Program and Standards
• Data Validation, Reduction, Reporting, and AIRS Input
Speciated Base NMOC Site Support
• (Frequency, 4 months, 5 days per week, 1 sample per day
equals 92 samples)
• Canister Cleaning and Shipping
• Instrument Certification and Installation
• Travel
• Speciuled Analysis
• QA/QC Program and Standards
• Data Validation, Reduction, Reporting, and AIRS Input
Toxics Option to NMOC Base Site*
• (10 Samples)
• Sampler Certification
• Toxics or Polar Organic Compound Analysis
• QA/QC Program and Standards
• Data Validation, Reduction, Reporting, and AIRS Input
Cost ($)
15,869
30.058
5,614
Shipping ($)
3,840
3,840
NA
Total ($)
'Cannot Ix selected utf slund-ulone program.
19,709
33.89K
5,614
I
i
I
X
f
-------
Table 1
(Continued)
Line Item
Co$t($)
Shipping (S)
Total ($)
4.0
Carbonvl Option to NMOC Base Site*
• (10 samples)
• Carbonyl Analysis for 16 Target Compounds
• QA/QC Program and Standards
• Data Validation, Reduction, Reporting, and AIRS Input
1,663
NA
1,663
D
>—•
o
5.0
Speciatcd NMOC Option to NMOC Base Site*
• (10 samples)
• Specialed Hydrocarbon Analysis
• QA/QC Program and Standards
• Data Validation, Reduction, Reporting, and AIRS Input
2,872
NA
2,872
6.0
Toxics Base Sile Support
• (1 year, 1 sample per 12 days, includes 10% duplicates
equals 34 samples)
• Canister Cleaning and Shipping
• Instrument Certification and Installation
• Travel
• Toxics or Polar Organic Compound Analysis by GC/MSD-
FID
• QA/QC Program and Standards
• Dala Validation, Reduction, Reporting, and AIRS Input
28,254
1,564
29,8 IK
'Cannot be selected m stand-alone program.
v
4
I
-------
Table 1
(Continued)
r
t
o
O\
Line Hem
7.0
8.0
9.0
Carbonvl Option |o Toxics Base Site*
• (I year. I sample per 12 days, includes 10% duplicates
equals 34 samples)
• Instrument Certification and Installation
• C.'aibunyl Analysis for 16 Compounds
• QA/QC Program and Standards
• Data Validation, Reduction, Reporting, and AIRS Input
Technical Site Support
• Any of a combination of the following PAMS site support
activities
- On-Sile Instrument Setup,
- QA/QC Support, Standards, & Round Robin Support,
On-sitc Technical Assistance and Consultation
PAMS Site - h'reqiiencv A
• (Eight, 3-hr every 3rd day and one 24-hr sample every 6th
day during (he monitoring period of June, July, and August
equals 284 samples, includes 10% duplicates)
• Canister Cleaning and Shipping (canisters provided by
Stale or local agency)
• Instrument Certification and Installation (instrument
provided by State or local agency)
• VOC Analysis
• QA/QC Program and Standards
• Data Validation, Reduction. Reporting, and AIRS Input
Cost (5)
3.798
24,650
70.014
Shipping (S)
NA
NA
11.060
Total($)
3.79K
24.650
81,974
'Cannot be selected us stand-alone program.
-------
Table 1
(Continued)
Line Item
Cost($)
Shipping ($)
10.0
1'AMS Site - Frequency B
• (Eight, 3-hr every day during the monitoring period of
June, July, and August and one 24-hr sample every 6th day
year round equals 818 samples, includes 10% duplicates)
• Canister Cleaning and Shipping (canisters provided by
State or local agency)
• Instrument Certification and Installation (instrument
provided by Stale or local agency)
• VOC Analysis
• QA/QC 1'rogram and Standards
• Data Validalion, Reduction, Reporting, and AIRS Input
146,814
36,662
183,47h
o
-j
11.0
I'AMS Site - Frequency C
• (Bight, 3-hr samples on S peak ozone days plus each
previous day, eight 3-hr samples every 6(h day and one 24-
hr sample every 6th day during the monitoring period of
June, July, and August equals 236 samples, including 10%
duplicates)
• Canister Cleaning and Shipping (canisters provided by
Slate or local agency)
• Instrument Certification and Installation (instrument
provided by Stale or local agency)
• VOC Analysis
• QA/QC Program and Standards
• Data Validation. Reduction, Reporting, and AIRS Input
37,181
10.028
47.209
'Cannot be selected as stand-alone program.
-------
Table 1
(Continued)
Line Item
: Shipping ($)>
Total ($)
12.0
PAMS Site -Frequency D
• (Eight, 3-hr samples every 3rd day during the monitoring
period of June, July, and August equals 266 samples,
including 10% duplicates)
• DNI'I I Cartridges
• Carbonyl Analysis for 16 Compounds
• QA/QC Program
• Data Validation, Reduction, Reporting, and AIRS Input
21,109
552fc
21.661
o
oo
13.0
PAMS Site - Frequency E
• (Eight, 3-hr samples, every day during the monitoring
period of June, July, and August equals 810 samples,
includes 10% duplicates)
• DNPH Cartridges
• Carbonyl Analysis for 16 Compounds
• QA/QC Program
• Data Validation, Reduction, Reporting, and AIRS Input
61,608
1,500C
63,1 OK
14.0
PAMS Site - Frequency F
• (Eight, 3-hr samples on 5 peak ozone days plus each
previous day, eight 3-hr samples every 6th days, and one
24-hr samples every 6th day during the monitoring period
of June, July, and August equals 236 samples, includes
10% duplicates)
• DNPH Cartridges
• Carbonyl Analysis for 16 Compounds
• QA/QC Program
• Data Validation, Reduction, Reporting, and AIRS Input
18,054
250'
18.304
•Cannot be selected at stand-alone program.
-------
Table 1
(Continued)
Line Kent
Cos<($)
Shipping ($)
Total ($)
o
o
15.0
PAMS Supplemental Auto GC Canister Support Frequency
• (One 24-hr sample every 6th day equals 60 samples)
• Canister Cleaning and Shipping (canisters provided by
Stale or local agency)
• VOC Analysis
• QA/QC Program and Standards
• Data Validation, Reduction, Reporting, and AIRS Input
19,373
2,944
22.317
•Select base and optional programs desired and return the information to Neil Berg at U.S. EPA Office of Air Quality Planning and Standards,
MD-14,
Research Triangle Park, North Carolina 27709.
blf selected in combination with Line Item No. 9.0, shipping for this item is not applicable.
clf selected in combination wilh Line Item No. 10.0, shipping for this item is not applicable.
'If selected in combination with Line Item No. 11.0, shipping for this item is not applicable.
*CMiix>( to felccled »f stand-alone program.
-------
NATIONAL SPECIALTY WORKSHOP ON TECHNICAL TOOLS
FOR AIR TOXICS ASSESSMENT
State and Local Participatory Program:
Paniculate Compounds Including PM,0, PMZJ,
and Filter Analysis for Toxic Species
EPA Contract 68D30029
Work Assignment 3-118
Neil Berg
Work Assignment Manager
Monitoring and Quality Assurance Group
Emissions, Monitoring, and Analysis Division
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Stan Sleva
TRC Environmental Corporation
6340 Quadrangle Drive, Suite 200
Chapel Hill, NC27514
Tuesday Afternoon Session
June 17, 1997
D-lll
-------
1.0 INTRODUCTION
The U.S. Environmental Protection Agency (EPA) is encouraging State and local air
pollution control agencies to conduct short-term, multi-site ambient air pollutant monitoring
studies using the technique known as saturation monitoring. Saturation monitors are non-
reference or non-equivalent sampling methods for measuring paniculate matter, carbon monoxide
and NOx. The samplers are small, portable, and relatively easy to set up and operate. Because
they are relatively inexpensive, it is customary to "saturate" a geographical area with the samplers
to assess the air quality in areas where high concentrations of pollutants are possible. Such
information is helpful to air pollution control agencies in evaluating their ambient air monitoring
networks for consistency with the 40 CFR Part 58 air quality surveillance regulations, which
require an annual network review and approval by EPA. EPA initiated the Saturation Monitor
Repository (SMR) in 1993. The SMR includes a pool of saturation samplers and related
equipment which agencies use when it becomes necessary for them to upgrade or borrow
saturation sampling equipment. The SMR provides limited services to quality assure the samplers
and assist in Held studies upon request.
Occasionally, State and local agencies will find that the concentrations of paniculate
mutter measured at locations in their ambient air monitoring networks exceed the PM-10 national
ambient air quality standard (NAAQS). In order to understand the nature of the paniculate matter
responsible for the exccedance, EPA has provided filter analysis assistance to the agencies
operating their paniculate matter network.
Under work assignment 3-118, TRC Environmental Corporation (TRC) is providing
D 112
-------
\
technical assistance to EPA in operating and maintaining the SMR, conducting special field
studies using the saturation samplers, and analyzing paniculate matter filters.
2.0 MINI VOL PORTABLE AIR SAMPLER
The Airmetrics portable minivol air sampler can be used to collect ambient air samples
for TSP, PM10, and PM2.5 and CO and NOx. The sampler is compact, lightweight, battery-
operated, and constructed from durable PVC. It can be operated from either AC or DC power
supplies. An internal programmable timer can make up to six sampling runs within a 24 hour
period, while an elapsed time totalizer records total pump operation time. The sampler also
continuously monitors the battery voltage and air flow rate, automatically shutting off sampling
if limits are exceeded. Figure 2-1 shows the sampler and hanging apparatus. Figure 2-2 shows
the preseparator and filter holder. Figure 2-3 shows the battery/sampler connection points and
Figure 2-4 shows the integrated gas sampling arrangement.
D-113
-------
ao>_.p o p o o _g=a
. HANGING BRACKET
HOISTING
HOOK
. BALE HANBfcE
Y - BRACKET ASSEMBLY '
HOIST
POLE
Figure 2-1 Sampler and Hanging Apparatus
D-J14
-------
RAN HAT
Ml 10 MPACTOft (COMPLETE)
202-003 :
i
STANDOFF
MULTIPLE IMPACT ADAPTER
j
2.5 MPACTOR (COMPLETE)
.-^,
PRESEPARATOR ADAPTOR
SUCONE GASKET
SUP RING
FILTER
DRAIN DISK j
FILTER HOLDER
FEMALE QUICK DISCONNECT
Figure 2-2 Preseparator and Filter Holder Assembly
D 115
-------
f.V » '
BATTBIY PAK LATCH
BANANA PLUGS
MOUNTNG CATCH
CHARGNG JACK
BANANA JACKS
BATTBOfPAK
Figure 2-3 Battery/Sampler Connection Points.
-------
CCMCCT MOS6 RTTN3
BAG MOGULS ICCTflETB
Figure 2-4 Integrated Gas Sampling Arrangement
D-117
-------
3.0 STATUS OF THE SATURATION MONITORING REPOSITORY
Table 1 shows the status of the SMR as of May 31, 1997:
TABLE 1. STATUS OF SATURATION MONITORING REPOSITORY
Location
In-House
Cincinnati
City of Indianapolis - ERM
EPA - RT1 Special Study
Missouri - Jefferson City
New Jersey - Holland Tunnel
NC DEHNR - Asheville
NC DEHNR - Washington
Region 3 - Philadelphia
Region 7 - Kansas City
Region 8 - Tribes
USDA Agricultural Research Service
Washington State University
Wl DNR - Milwaukee
TOTAL
No. of Monitors
35
2 (PM-2.5)
8 (PM*)
6 (PM-2.5)
6 (PM-2.5)
2 (1 PM, 1 CO)
2 (PM-2.5)
1 (PM)
4 (PM-2.5)
5 (PM-2.5)
5 (PM-2.5)
20 (PM*)
15 (PM*)
4 (PM-2.5)
115
Status
Available
Due June 1997
Due January 1998
Due August 1997
Due December 1997
Due October 1997
Due May 1998
Due September 1997
Due December 1997
Due October 1997
Due July 1997
Due June 1997
Due January 1999
Due October 1997
* Includes PM-2.5 inlets
4.0 COMPLETED STUDIES
The following studies were completed during the last six months:
Study #1:
Agency:
Request:
Purpose:
PM-10/2.5 Saturation Study in Washington N.C.
NCDEHNR - Washington Regional Office
I 1 PM-10 and 2.5 samplers
To determine PM-10 and 2.5 levels at selected sites
D-118
-------
Time Period:
Pre- and Post-
Filter Weighing:
Study #2:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
Study #3:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
1996 through January 1997
No
Region VIII/ORD Reference Sampler Study
Region VIII
3 PM-2.5 samplers
To compare with proposed FRM
November 1996 through January 1997
No
Ventura County California
Ventura County Air Pollution Control District
10 PM-2.5 samplers
To determine PM 2.5 levels at selected sites
Summer 1996 through March 1997
No
5.0 ONGOING STUDIES
The following studies were ongoing as of May 31. 1997:
Study #1:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
PM-2.5 Special Study
Cincinnati
2 PM-2.5 samplers
To determine PM concentrations around burning landfill (-1 month),
followed by periodic sampling at selected sites
July 1996 through June 1997
No
D-119
-------
Study #2:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
Study #3:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filler Weighing
Study #4:
Agency:
Request:
Purpose:
Time Period:
Pre- and Posi-
Filler Weighing:
Holland Tunnel PM/CO Study, New Jersey
Municipal Environmental Commission
1 CO, I PM-2.5 sampler
To determine PM and CO outflow from the Holland Tunnel.
February 1996 through September 1997
No
PM-2.5 Saturation Study in Asheville, North Carolina
NC DEHNR
2 PM-2.5 samplers
To determine the impact of suspended PM sources near residences
located at valley exit
May 1996 through May 1998
No
Saturation Sampler Study in Washington, North Carolina
NC DEHNR
1 PM sampler
To determine sampler/battery operation
March 1997 through September 1997
No
Study #5:
Agency:
Request:
Purpose:
Time Period:
Pre- and Posi-
Filter Weighing:
PM-2.5 Study
EPA Region VII
5 PM-2.5 samplers
To determine PM levels in urbanized area
March 1996 through September 1997
No
D-120
-------
Study #6:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
Study #7:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
PM-10 Study in Big Springs, Texas
USDA Agricultural Research Center
20 PM samplers (8 with PM-10 inlets, 6 with PM-2.5 inlets, 6 with no
size selection on the inlet)
To determine the relationship'between PM-10 readings and soil erosion
by wind.
February 1994 through June 1997
No
PM-10 Study
Washington State University
15 PM samplers
To determine PM emissions from windblown dust at three different
elevations at a site near Spokane, Washington.
February 1994 through December 1998
No
Study #8:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Wei«hiii2:
PM Saturation Study in Jefferson City
Missouri Department of Natural Resources
6 PM-2.5 samplers
To determine PM levels at selected sites in Missouri
November 1996 throuah November 1997
No
10
D-121
-------
Study #9:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
Study #10:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
Study #11:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
Study #2:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
PM-2.5 Saturation Study - Northern Cheyenne, Confederated Salish and
Kootenai Tribes
EPA Region VI11
5 PM-2.5 samplers
To better characterize the PM emissions in Montana's tribal
nonattainment areas to allow EPA to better target control strategies.
November 1996 through June 1997
No
PM-2.5 Monitoring Study
Wl DNR - Milwaukee
4 PM-2.5 samplers
To determine PM levels at selected urban area sites
November 1996 through September 1997
No
PM-2.5 Monitoring Study
City of Indianapolis - ERM
8 PM-2.5 samplers
To determine PM levels at selected urban area sites
November 1996 through September 1997
No
PM Monitoring Study
EPA Monitoring and Quality Assurance Group
6 PM-2.5 samplers with different PM2.5 inlet configurations
To compare with proposed FRM
April
1997 through August 1997
Yes
6.0 PLANNED STUDIES
The following saturation studies are planned for the upcoming period:
Study #1:
PM 2.5 and PM 10 Monitoring Study
D-122
-------
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
EPA Region II
10 PM-2.5 samplers and 3 PM 10 samplers
Determine PM2.5 levels in Queens, NY
To be determined
Yes
Study #2:
Agency:
Request:
Purpose:
Time Period:
Pre- and Post-
Filter Weighing:
PM 2.5 Monitoring Study
Puerto Rico Air Quality
15 PM-2.5 samplers
Determine PM2.5 levels selected areas of Puerto Rico
To be determined
No
7.0 FILTER ANALYSIS COST ESTIMATES
Analysis
Gravimetric Analysis
(Pre and post sampling weighing)
Elemental Analysis (34 elements)
(X-ray Fluorescence)
Total Carbon. Organic Carbon,
Elemental Carbon
(Thermal/Optical Reflectance)
Total Carbon. Organic Carbon,
Elemental Carbon, Carbonate Carbon
(Thermal/Optical Reflectance)
Sull'aie, Nitrate, and Chloride
(Ion Chromatographic Analysis. 1C)
Ammonium or Nitriie Ion
(Automated color!metric analysis. AC)
Cost S/Sample
8-11
40
35
70
30
30
Single Elements, Sodium or Potassium
(Atomic Absorption Spectroscopy, AA) 30
12
D-123
-------
D-124
-------
o
Lf 6t'A/D
o*.
Jn 2,3,W,5
r» ^ -on/)
(r, i
.-cJ lec.H •
__ : NOoCit S.-I^
M U'Su.
-------
D-126
-------
In: Proceeding* of the EPA/A6WMA Symposium, "Measurement of Toxic and Related Air
Pollutants," April 29.- May 1, 1997, Research Triangle Park, HC
CMB8: New Software for Chemical Mass Balance Receptor Modeling
Charles w. Lewis
National Exposure Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park NC 27711
ABSTRACT
The Chemical Mass Balance (CMS) method for receptor modeling of
ambient air pollutants has been in use for over two decades. Over the
past year the U.S. Environmental Protection Agency's Office of
Research and Development and Office of Air Quality Planning and
Standards have jointly sponsored the development of a new generation
of CMB software, CMB8. Developmental work has been performed by the
Desert Research Institute, Reno, NV. Changes embodied in CMB8 include
(1) switch from a DOS-based to a Windows-based environment, (2)
increased attention to volatile organic compounds (VOC) applications,
(3) correction of some flaws in the previous version (CMB7), (4) more
options for input and output data formats, (5) addition of a more
accurate least squares computational algorithm, (6) a new treatment of
source collinearity, (7) multiple defaults for sources and fitting
species, and (8) choice of fitting criteria. Details of the changes
and the procedure for obtaining CMB8 are given.
INTRODUCTION
The Chemical Mass Balance (CMB) method for receptor modeling of
ambient air pollutants has been in use for over two decades. The CMB
model can be written as
p
Ci = V aw 5j, i - 1, Ji (1)
where C, is the ambient concentration of specie i, ag is the fractional
concentration of specie i in the emissions from source j , S, is the
total mass concentration contributed by source j, p is the number of
sources, and n is the number of species, with n > p. The Ct and av are
known and the S; are found by a least squares solution of the
overdetermined system of equations (1). While the system is simple in
appearance its solution is not, because of the need to take account of
uncertainties in both the C,'s and the atf's.
The U.S. Environmental Protection Agency (EPA) has tacitly . . .
approved CMB as a regulatory planning tool through the Agency's
approval of numerous State Implementation Plans (SIPs) which have had
a CMB component. Since about 1990 EPA has supported and freely
distributed the current version of the CMB software (CMB7) . Over the
past year EPA's Office of Research and Development and Office of Air
Quality Planning and Standards have jointly sponsored the development
of a new generation of CMB software, CMB8, and associated
documentation and guidance. The work is being performed by the Desert
Research Institute, Reno, NV. During the development period the
evolving versions of CMB8 have been available to an informal network
of over 20 interested individuals for testing.
In the following an overview of CMB8 is presented^ The procedure
for obtaining a copy of the software and documentation is also given.
D-127
-------
CMB8 FEATURES
Windows Environment
CMB8 is designed for use with Microsoft Windows*. Both 16-bit
and 32-bit versions are available for Windows 3.x and Windows 95,
respectively. CMB8 is not available as a DOS application. The switch
to a Windows environment offers a number of advantages. Species and
profile selections can be done more quickly under mouse control, and
the user can choose to immediately view the current selections in
display windows (or have them continuously present) during a session.
In general, use of a mouse has encouraged a richer choice of options
and gives more immediate access to them.
Species and source Selection Defaults
In CMB8 the user can specify any-one of up to ten sets of fitting
species that can be chosen at random by mouse-clicking on a software
button. A set can be created or modified at any time during a
session. Similarly, up to ten sets of source profiles can be defined
and accessed. It is anticipated that this feature will provide a vast
improvement in the efficiency of searching through many source and
fitting species combinations in order to find the best fit — a very
time-intensive aspect of using CMB7.
Input File options
Necessary inputs for a CMB calculation include an ambient data
file, which lists all the measured chemical information for a series
of ambient samples, and a source profile data file, which lists the
chemical composition of a collection of candidate source profiles.
CMB7 required these files to be blank-delimited ASCII text files
(supplementary software was available to externally convert
spreadsheet- and database-type files to the required CMB7 format). In
CMB8 the direct input of ambient and source data files may be done in
a variety of formats: blank-delimited ASCII(*.TXT), blank-delimited
ASCII with columns and rows interchanged (*.CAR), comma-separated
value (*.CSV), XBASE (*.DBF), and Lotus 123 spreadsheet (*.WKS and
*.WK1).
Output File Options
The calculated results from a CMB8 session are available in two
output files whose default names are CMBOUTRP.TXT and CMBOUTDB.TXT.
The "report" file CMBOUTRP is an electronic copy of the output results
that are displayed and printable during the session. File
CMBOUTRP.TXT is the analogue of CMBOUT.DT2 in CMB7. Similarly, the
"database" file CMBOUTDB.TXT is the analogue of the CMB7 file
CMBOUT.DT2.
A new feature of these files in CMB8 is that not only their names
(both prefix and suffix) but also their directories may be changed by
the user during a session, allowing for better grouping and
identification of results generated during a lengthy session. The
last choices in a session become the defaults for the next session.
In the case of the file CMBOUTDB, the only allowed suffixes are .TXT,
.CSV, .DBF and .WKS. Each can be selected with a software button,
which causes the file to have the corresponding format type.
Collinearity Treatment
Collinearity in a CMB context refers to a situation in which the
compositions of two or more source profiles are so similar that it is
difficult to infer the separate impact of each. Collinearity is a
form of mathematical ill conditioning, and its presence is associated
with additional serious problems h^vond the one just noted. The basis
-------
than either x2 °r *? alone.
OBTAINING CMB8
Until CMB8 is finalized, and is formally accepted and supported
by EPA, the current version of the CMB8 software and documentation is
available from the developer, DRI, via anonymous ftp. The address is:
eaf.sage.dri.edu:/cnbB0/model/I6bit/win3MMDD.exe
where MMDD stand for month and day to identify the development version
(the directory may contain a few earlier versions in addition to the
latest one). The file win3MMDD.exe is a PKZIP self-extracting file,
so the user should use a binary mode of file transfer. In addition to
the CMB8.EXE program the file also contains sample input files and a
tutorial/users manual. A Windows 95 version may be obtained by
changing the last part of the address to 32bit/vrn32MMDD.exe.
CONCLUSIONS
Features of the new version of CMB software (CMB8) have been
described. It is anticipated that CMB8 will significantly increase
the convenience, efficiency and accuracy of performing receptor
modeling by the CMB method.
ACKNOWLEDGEMENTS
The DRI CMB8 development team consists of John Watson, Norman
Robinson and Judy Chow. We are indebted to those individuals who made
substantial contributions to improving the CMB8 software through their
suggestions and testing efforts during the development period. They
include Teri Conner, Tom Coulter, William Ellenson, Eric Fujita,
Douglas Lowenthai, Elizabeth Vega, and Robert Willis. Richard Scheffe
was instrumental in coordinating the joint support of EPA's Office of
Research and Development and Office of Air Quality Planning and
Standards to serve both the research and regulatory interests of the
Agency.
DISCLAIMER
The information in this document has been funded wholly or in
part by the United States Environmental Protection Agency under
Purchase Order No. 5D1607NAEX to the Desert Research Institute. It
has been subjected to Agency review and approved for publication.
Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
REFERENCES
1. Henry, R.C. "Stability analysis of receptor models that use least
squares fitting," in Proceedings of the AWMA Specialty Conference
on Receptor Models Applied to Contemporary Pollution Problems,
SP-48, Air & Waste Management Association: Pittsburgh, 1982; pp.
141-157.
2. Henry, R.C. Atmos. Environment 1992 26A, 933-938.
3. Britt, H.I.; Luecke, R.H. Technoroetrics 1973 15f2^. 233-247.
4. Watson, J.G.; Cooper, J.A.; Huntzicker, J.J. Atmos. Environment
1984 18(7). 1347-1355.
D-129
-------
D-130
-------
HT:y^™^fgs^$
v^. '.•'*'! •'••••: ••"• t^'^f
£•2 •'.-.. .••'•/•-jv:-^--««»^>«fibagaa!
££ • - -•.-./.•>;• .•"••'^i^i^
sO • : •••"• •'-"'• //^-'ci^^?>*nii
Ambient Air
Presentation,
Technical Tools for Air Toxics Assessment"; ..'• •_• ; ^;J;:;^^
June 17, J997
Hemby
Air Quality Trends & Analysis Group
OAQPS
^OVERVIEVP»»
' ' ^—" * J—/JLV T AJL/ T T. .-.•'f-.*;.ii'^->j.'---HviI^;11£'ihJ'itti
Demonstrate two
software tools
+PAMSDAS
Other miscellaneous tools
Ambient air quality data characterization
project
Air Qualify Trends & Analysis Group
OAQPS
D-131
-------
•Desci0ij»ti|
''•'-••>:'•'!. '•''•:•:'••'„' •^^•^fSKs^gssfS
Software module;^lacilita^^
PAMS analysis;;
Design
4 Menu driven; easy to learn &
4No programming experience needed ^?
Once complete (October '97), the module will be
posted on the web and distributed to
Regions/States
Air Quality Trends Si Analysis Group
OAQPS
PAMSDAS
.
System Requirements..
4 PC based (Pentium)
4 Minimum 16 MB RAM
Operating Environment , / V.'
4 S-PLUS front-end menu-driven system
4 Windows NT/Windows 95/Windows 3.x
Capabilities (demonstrated on next several slides)
Air Quality Trends £ Analysis Group
OAQPS
D-132
-------
:S KUSlur Wuidnv
FJt £dl lools
§-¥LVS T Copyright 1988. 199*
S : Copyright ATIT. .,
Version 3.3 Release 1 for MS.;
Barking data will be in _Dat«
rPUMWrtx
ra0ru*von not
Box Hot*
OUnaiProflet
E«e E* loo** Options &*ow t**> PH Summaries
S-P1US: Copyright 1988."1995 ***L>,
-------
•S I'LUbioi Windows (Ginptm
E* loot* Qpfem ^xtow tje* Hot SunMrie* UMfes QuIMMS
1995 Baton Rougt - Captol
.. . b* Plot Sumwie* UBfet OutPAMS
S-PLUS • Copyright"l98~8"." 1995 HathSoft ' Inc
S : Copyright ATiT.
Version 3.3 Release 1 for KS Windovs 3.1 • 1995
Working data will be in Data
IceprtoLSSvoc
Acetone
3-MethyH-Butene
D-134
-------
I -
(MCI
:S PLUS lor Window* - (Gfdplucii 3 (Cunrnt>]
jpe Edl looU Qplions y^ndow tM> PM SunnMes UKe* OulPAMS
Ivatn*
S4>LUS forWindow*. \J,
D-135
-------
1995 Pride
Wt*M«y
: j
: i
9:
• • • •
!•••*
Fie &a looU Qptons ^ndow \j/0f PU
S^PLUS : Copyright" 19 e"6. 199S~Ma" Typical Ranges T«Ue
S : Copyright ATiT %
Version 3.3 Release 1 (or MS Vim]
Working data vill be in _Da.ta
>
Error in dialogbox display(choicebox): Cancelled
Dunped
W
& Ffedance Graphka^'[PA.'.' |||^S-PLUS for Window*
D-136
-------
Q Po(|r3?c.4?4 • Kutcpad
:apitol.95uoc
5
6
7
8
9
1t
11
12
13
1*
15
16
17
18
19
20
21
22
23
2*
26
27
28
UOC
ncetaldehyde
acetone
Acetylene
Alpha-Pinene
3-Hethyl-1-Butene
2-Nethyl-2~Butene
1-Butene
Benzene
Beta-Pinene
2.2-DlHethyllMitane
2.3-DiHethylButme
.2.3-TriHethylBenzene
.2, *-TriHethylBenzene
,3,5-TriHethylBenzene
Cis-2-Butene
Cis-2-Hexcne
Cis-2-Pentene
CycloPentane
CycloPentene
CycloHexane
EthylBenzene
Ethane
Ethylene
Air Oualily Trends & Analysis Group
OAQPS
D-137
-------
Wmdows-base••••'••--*
Import/Export AIRS format files' (in'cjudihg m
codes). " T.1 •"'r>.*7S'""
Prepares summary of species groups.
Options to apply reactivity factors or create wt%
NMHC.
Air Quality Trends & Analysis Group
OAQPS
1 VOCDat:|i9S
Planned
a Modify AIRS output to include 0
for PAMS target species that are missing;!;b^
• Create sum of target species arid^^ output to ~2
• Allow user customization of target species list.
• Create module(s) to run statistical checks and :
comparisons to enhance QC/QA activities.
• Finalize program in October/November - plans to
make available via EPA Internet site
Air Quality Trends &. Analysis Group
OAQPS
D-138
-------
\
WN
Time mies plot of several (oecitt |raupi tl Stafford. CT in 1994. Etampk of nuudentification of i paraffin
for u unidentified peak. (Level 0, preliminary 6»u. CT DEP)
ti<* Un un wt) *rlj
Example of Hndmi tpecies mtMdtnnr»C4tton in a data KI u*inj a time «n** plot (top)
*nd scutcr plot t bottom) Daii *tre collecied ai the Laic Cliuon. MO f*AMS me
dunni June 1995 {U\etO. AtRSl
D-139
-------
Fui{erprim ploa of 1992 umptet u Chuniul (El Paso). TX. Eumplct of typical (up)
•rid calibration (bouom) finfnpnnu. (Level I,MRS diu)
- -.-. -. • c •.. !',-~.'xvtsift-wjjzKtiiss
• . - •'.•• ".-*.' -.'.-.;:.'•,»•"• ii-^<-Sira*;>5s3
.,;,-;.•:•-.•• - -. ^-"^i-.-.--.^..:;^.::;.^^-,-^
, ' v -•- . •*-;• •_ i_ ; '.. i^-t-n .^-S«fc*Ap ^••*r»1=lr
C*" " " * *« *• 9"'1*-**" j.'1" "i / *Tvv*J ^. ?"J'^\.-'
':'? '• " ' •• > i ': |N"- .. 2._••'*! I. . *'*:'\f t-''.''.'t
OTHER ANAI?^SIS^$>0B
MVOYAGER
• Environmentalstats
• Output Format Software
AirQualiiy Trends & Analysis Group
OAQPS
D-140
-------
CHAK&CTERI
W • Products:
Goal: Assess amtoxics-iss^|s5i$gKp
quality data
Two aspects:
4 Collect and analyze ambient;
4 Develop and apply indicators
4 Annual data catalogue
4 Annual trend/indicator assessment
Air Quality Trends & Analysis Group
OAQPS
D-141
-------
-------
National Air Toxics
Emissions Inventory
June 18, 1997
Air Toxics Modeling & Exposure
Workshop
CAA Requirements for NIT
- Section 112(d),MACT
- Source category & HAP listing/delisting
- MACT Tracking
Section 112(k), Urban Area Source Program
Section 112(c)(6)
Section 112(f), Residual Risk Program
Section 112(m), Great Waters Program
Section 112(n), Special Studies, e.g., Hg
D-143
-------
Nil Database Estimation Tools and Stakeholders
MAnONALTDOCS DATABASE
AIR TOXICS MANAGEMENT
MODEL
Establish and
Consider Hsaithsnd
Envtronmtntat Goals
What Reductions ArsNs«d«d?
• Ar» reductions nssdsd?
• From which source*?
Evaluate
RssuKs
.Implsmsnt
and Enforce
1
How toAehisv*
Rsductions/Solvs
Probwfn
p-i44
-------
112(c)(6) HAPs
• Alkylated lead
• Hexachlorobenzene
• Mercury
•PCBs
•POM
• 2,3,7,8-TCDD
• 2,3,7,8-TCDF
112c6 Category Listing
Open Burning of Scrap Tires
Gasoline Distribution Stage I Aviation
Wood Treatment/Wood Preservatives
D-145
-------
Potential 112(k) HAPs
Acetaldehyde
Acrolein
Acrylamide
Acrylonitrile
Arsenic compounds
Benzene
Beryllium compounds
Bis(2-ethylhexyl)phthalate
1,3-Butadiene
Cadmium compounds
Carbon tetrachloride
Chloroform
Chromium compounds
Coke oven emissions
1,4-Dichlorobenzene
1,2-Dichloropropane
1,3-Dichloropropene
Dioxins/furans
Ethyl acrylate
Ethylene dibromide
Ethylene dichloride
Ethylene oxide
Formaldehyde
Hydrazine
Lead compounds
Manganesecompounds
Mercury compounds
Methyl chloride
Methylene chloride
Methylene diphenyl diisocyanate.
Nickel compounds
POM
Quinoline
Styrene
1,1,2,2-Tetrachloroethane
Tetrachloroethylene
1,1,2-Trichloroethane
Trichloroethylene
Vinyl chloride
Vinylidene chloride
Risk Identification Process - List
of 37 HAPs of Interest
Acetaldehyde
Acrolein
Acrylamide
Acrylonitrile
Antimony compounds
Arsenic compounds
Benzene
Beryllium compounds
bis(2-chloroethyl) ether
1,3-Butadiene
Cadmium compounds
Chloroform
Chromium compounds
Coke oven emissions
Ethylene dibromide
Ethylene dichloride
Ethylene oxide
Formaldehyde
Glycol ethers
Hydrazine
Lead compounds
Managanese compounds
Mercury compounds
Methylene chloride
Methylene diphenyl diisocyanate
Nickel compounds
Phosgene
POM
Styrene
2,3,7,8-TCDD
2,3,7,8-TCDF
Tetrachloroethylene
Toluene
2,-4-Toluene diisocyanate
Trichloroethylene
Vinyl chloride
Xylenes (include o,m,
and p)
D-146
-------
NTI Data Sources:
Inventory Approach
Source-specific data where available
State/local HAP inventories
MACT data
Industry data
TRI data
Estimates using efs and source activity data
Types of NTI Data
' Annual emission estimates in ton/yr
1 Physical location of emissions
1 Data sources
1 Miscellaneous fields related to data processing
- Data entry dates
- Allocation codes
Auxiliary information used to estimate emissions
- Activity data
- Population
-Number of facilities
- Employment data
D-147
-------
NTI System Construction
1 FIPs state and county codes
SCCs
CAS numbers/Pollutant names
SIC codes
Allocation codes to define source categories as area,
point or mobile
Primary and secondary references
US Census population data to define inventory areas
as either urban or rural
VOC or PM codes to identify VOC and PM-HAPs
Date of inventory
NTI System Design
Database uses SAS
Hardware requirements: Pentium 200 MHz or Unix
system
Data Storage Space
Auxiliary Data - 120 Megabytes
Local Inventory Data - 50 Megabytes
Dictionary Data - 15 Megabytes
Inventory Data - 200 Megabytes
Data Processing Space - 1 -2 Gigabytes
System Memory 64 Gigabytes
D-148
-------
NTI Pollutant and Source
Category Coverage
188HAPS
780 unique point, area, and mobile (on-road &
off-road) source categories
3141 counties in 50 states
Example Data Outputs
D-149
-------
Planned Improvements to NTI
MACT data
Facility specific data for major sources
Update NTI with 112(k) and 112(c)(6) data
State/Local HAP inventories
Title V data
Gap-filling in source/pollutant data
Better correlation between monitoring and emission
inventory data (Qualitative)
STAPPA/ALAPCO Survey
Source Characteristic
Toxic Chemical Characteristics
Risk Assessment
Ambient Data
Data Accessibility
D-150
-------
Dear State and Local Air Toxics Regulators:
EPA is beginning to work on the implementation of the residual risk requirements mandated by
Section 112(f) of the Clean Air Act. In order to understand the residual risks that a source or
source category may present after implementing MACT, it is necessary to know what toxic air
pollutants they emit. The best sources of the type of information EPA needs for this effort are the
state and local regulatory agencies who work most closely with the faculties.
The following is a short survey developed by STAPPA, ALAPCO and EPA. It attempts to gather
information about which state and local air agencies collect air toxics emission data and what
types of data they have. IT DOES NOT ACTUALLY REQUEST THAT YOU SUBMIT THE
DATA! It is merely an attempt to determine what information may be available. The
questionnaire also asks that you provide the name of a person to contact if we need follow-up
information — someone familiar with your agency's air toxics efforts as a whole and who can
direct us to specific technical staff people, if needed.
Please send your responses to Mary Sullivan Douglas of STAPPA/ALAPCO at m4ckiair@sso.org
by June 2,1997. Thank you for your participation!
If you have questions about this survey in general, please direct the,. 10 Carol Piening,
Washington State Department of Ecology, Air Quality Program, email: cpie461@ecy.wa.gov
phone: (360)407-6858 fax: (360)407-6802
If you have questions about the ambient air quality portion of the survey, please direct them to
James Hemby, EPA, OAQPS email: hemby.james@epamail.epa.gov (919)541-5459 (voice)
(919) 541-1903 (fax)
***AIR TOXICS DATA SURVEY - May, 1997 ***
DEFINITIONS:
For the purposes of this survey, please consider the following definitions:
TOXIC CHEMICAL - any chemical on the list of hazardous air pollutants found in Section
112(b) of the Clean Air Act.
SOURCE or SOURCE CATEGORY - a source category listed by EPA for development of
MACT standards pursuant to Section 112(c) of the Clean Air Act. Categories of AREA sources
include chrome electroplaters, dry cleaners, commercial sterilization facilities, halogenated solvent
cleaners, and secondary lead smelting. MOBILE sources includes both on-road and off-road
vehicles.
BENCHMARK - the guidelines or standards currently in use by your state or local agency.
D-151
-------
PART 1 - SOURCE CHARACTERISTICS:
In general, for what kinds of sources do you collect source characteristic information? (X all that
apply)
Point _
Area
Mobile
Some specific-source categories
•
In general, how often are data for a particular source reported?
ie. annually
Does your agency collect emission point information? Yes No
Does your agency collect emission unit information? (X all that apply)
Stack height: Yes No__
Inside diameter: Yes No
Exit temperature: Yes No
Exit velocity: Yes No
Distances) to property line: Yes No
Geographical location: Yes No (e.g. latitude and longitude for GIS applications)
Emission rates: Yes No
Maximum hourly emission rate: Yes No
Are facility data coded by (X all that apply)
scc_
AMS_
SIC_
Plant or facility ID: name , address , ID , FIPS county codes
In general, how accurate and complete are the data collected?
1 (poor) 2 _ 3_ 4 5 (good) _
PART 2 - TOXIC CHEMICAL CHARACTERISTICS
Does your agency collect information for toxic chemical emissions by CAS number?
Yes No On a case-by-case basis.
In what units are the toxic chemical emissions information? (X all that apply)
Lb/hour
Lb/year
Kg/hour
Kg/yr
Other
D-152
-------
PART 3 - RISK ASSESSMENT INFORMATION
Has your agency performed or evaluated any risk assessments on source categories?
Yes - approximate number: No
Has your agency performed or evaluated any risk assessments on individual sources -within those
source categories?
Yes - approximate number: No
What type of benchmarks are used by your agency for cancer endpoint evaluations? (X all that
apply)
Cancer: EPA unit risk values
State/local unit risk values
Both_
Cancer unit risk value not used
Derived from occupational guidelines
If your cancer benchmarks are derived from occupational guidelines, what factors do you apply to
them?
What type of benchmarks are used by your agency for non-cancer endpoint evaluations? (X all
that apply)
Non-cancer: EPA reference concentration/reference dose
State-derived (NOAEL/LOAEL approach)
Derived from occupational guidelines
If your non-cancer benchmarks are derived from occupational guidelines, what factors do you
apply to them?
What averaging period is used for your benchmarks? (X all that apply)
Annual
24 hour
8 hour
1 hour
Other
D-153
-------
PART 4: AMBIENT DATA
Has your agency collected any ambient data on air toxics?
Yes No
Are these monitoring efforts best characterized as
Special studies directed at particular sources
Characterizations of representative ambient conditions
Both
What options) below best describes the duration of the monitoring effort
One year (please indicate year )
Multi-years (please indicate years )
Ongoing (please indicate initial year )
Other
Does your agency have plans to conduct any ambient toxics monitoring in the future?
Yes No
If yes, please provide a brief description (objectives, duration, etc.). Ongoing, as part of ozone
precursor study.
PARTS: DATA ACCESSIBILITY
Is the data available in a computerized file format? (i.e. ASCII, spreadsheet, database files?)
Source Characteristic data Yes some No
Chemical Characteristic data Yes some No
Risk Assessment information Yes some No
Ambient data Yes some No
Are/were ambient data submitted to the Aerometric Information Retrieval =
System (AIRS) database?
Yes No
Are the data accessible via the Internet?
Yes No
Is the data organized in a central location, or dispersed among several people in your agency?
Centralized Dispersed
D-154
-------
PART 6: ONE CONTACT PERSON
If there are follow-up questions about your air toxics program and the data you have collected,
who would be the appropriate primary contact?
Name and title:
Agency or organization:
Mailing address:
Phone:
Fax:
E-mail:
YOU ARE FINISHED WITH THIS SURVEY! Please send your responses, electronically if
possible, to Mary Sullivan Douglas of STAPPA/ALAPCO.
email: m4cbair@sso.org
phone: (202) 624-7864
fax: (202) 624-7863
THANK YOU for your time and assistance. This information will be compiled and made
available to state and local air pollution control agencies, EPA regional offices, and EPA
headquarters. As we get a better idea of what information we need, and who has more
information available, we will likely ask the contact person for additional help.
D-155
-------
Status of STAPPA/ALAPCO Survey - June 16
Alabama
Alaska
Arizona rec'd
Arkansas
California
Bay Area rec'd
Ventura rec'd
Monterey rec'd
Colorado rec'd
Connecticut rec'd
Delaware rec'd
Florida
Georgia
Hawaii
Idaho
Illinois rec'd
Indiana
Iowa
Kansas
Wyandote rec'd
Kentucky rec'd
Louisiana rec'd
Maine rec'd
Maryland
Massachusetts
Michigan rec'd
Minnesota rec'd
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York rec'd
North Carolina rec'd
North Dakota rec'd
Ohio
Oklahoma rec'd
Oregon
Pennsylvania rec'd
Rhode Island
D-156
-------
South Carolina rec'd
South Dakota rec'd
Tennessee
Texas
Utah
Vermont rec'd
Virginia rec'd
Washington rec'd
Puget Sound rec'd
West Virginia
Wisconsin rec'd
Wyoming
DC
D-157
-------
TABLE 1. PROPOSED EMISSION INVENTORY RULEMAKING PROVISIONS.
Ir-.i-
! Provision
I^I^^^^^^^^^HIH^H
CAA citation
Frequency of reporting
Estimating period
Areas to which
provision applies
Pollutants and source
size thresholds
; SlotewWe Point
*Sfgnif icant SojUrce»!^ >
§110(a)(2)(F)
Annual
Annual
Entire U.S.
Pollutant tpy3
SO, 2,500
NO, 2,500
VOC 250
PM-10 250
CO 2,500
Source Inventory
; (.Major Sources2 ;/ ..
§110(aX2)(F),§112
Every three years
Annual
Entire U.S.
Pollutant
-!EY
SO, > 100
NO, ;• 100
VOC > 100
PM-10 .- 100
CO > 1,000
Pb > 10
10
Major source
thresholds vary in
nonattainment areas
depending on
classification.
Emission Statements
§ 182{aX3)(B)
Annual
Annual and Dairy1
Ozone NA areas and
attainment areas in
ozone transport
regions
Pollutant tpy
Ozone NA areas:
VOC , 25
NO, > 100
Attainment areas in
ozone transport
region:
VOC > 50
NO, >. 100
*'% :
' "' -$
3-Year Inventory 'I
§172(cX3),§182(a)(3)(A),and
§187(aK5)
Every three years
Annual and Dairy*
Ozone, CO, and PM-10
NA areas
Pollutant
Ozone NA areas: tpy
VOC > 10
NO, ?. 100
CO s 100
CO NA areas:
CO 2 100
PM-10 NA areas:
PM-10 * 70
Inventory includes:
• Point sources 2 specified
tpy.
• Area sources < specified
tpy.
• On-road mobile sources.
• Nonroad mobile sources.
• Biogenic sources.
-------
D-160
-------
Great Lakes Regional Air Toxics Emissions Inventory
--Grcac Lakes
Commission
Great Lakes Regional Air Toxics Emissions
Inventory
Overview
Air regulatory agencies in the eight Great Lakes states
agree that a collaborative effort is vital to successfully
implement a compatible regional database of airborne
toxic pollutant emissions. They have been working
cooperatively toward this goal since 1987. Once a quality
controlled/quality assured data inventory has been
established, the states and the U.S. Environmental
Protection Agency can begin to work separately and in
concert to define and regulate sources; evaluate control
technology; establish guidelines for siting new facilities;
and reduce airborne deposition of persistent toxic
chemicals to the Great Lakes.
Purpose
This inventory will assist in the successful implementation of key provisions of the Great Lakes Toxic
Substances Control Agreement, signed by the Great Lakes governors in 1986. In addition, this work is
consistent with the state activities for the implementation of the Urban Area Source Program required
under sections 112(c) and 112(k) under the Clean Air Act Amendments of 1990 and the assessment of
atmospheric deposition to the Great Lakes under the efforts of U.S. EPA's Great Waters Program.
• Section 112 of the Clean Air Act
* Great Lakes Governors' Toxic Substances Control Agreement of 1986
Background/History
An Overview of Inventory Development
The Great Lakes states, working together through the Great Lakes Commission, are creating this
regional database with funds from the Great Lakes Protection Fund and U.S. EPA. Specifically, the
program's goal is to establish a baseline using 1993 data on point and area source emissions of 49
toxic air pollutants that have been identified as significant contributors to the contamination of the
Great Lakes.
Steering Committee for the Great Lakes Regional Air Toxics Inventory Project
The steering committee guides development of the regional inventory and associated products,
including RAPIDS and the Air Toxics Emissions Protocol. Emission inventory specialists from the
Great Lakes states, U.S. EPA and the province of Ontario work together closely.
D-161
-------
Great Lakes Regional Air Toxics Emissions Inventory
49 Toxic Pollutants
The Great Lakes Regional Air Toxics Emissions Inventory targets 49 compounds that have been
identified as significant contributors to the contamination of the Great Lakes. These compounds,
and the programs or agencies that have targeted them, are listed.
AirToxics E-mail Archive
The Great Lakes Regional Air Toxics Emissions Inventory Project successfully demonstrates the
cost-effective and time-efficient use of the Internet as an aid to solving regional environmental
problems. Project members use an e-mail list server to discuss project developments, drastically
reducing the need to travel to meetings and hold expensive conference calls. An archive of the
e-mail messages provides a history of this interaction.
Products
Air Toxics Emissions Inventory Protocol for the Great Lakes States
The protocol provides instructions necessary for the Great Lakes states to develop a regional
inventory that is complete, accurate and consistent from state to state. The protocol was tested
during the Southwest Lake Michigan Pilot Study. It is being used to guide the efforts of all eight
Great Lakes states in 1995-96 as they prepare the first full statewide toxic air emissions point and
area source inventories and populate the regional repository, Further additions and refinements may
be expected after the first full eight-state inventory.
Regional Air Pollutant Inventory Development System (RAPIDS) client/server software
RAPIDS, the Regional Air Pollutant Inventory Development System, is the first-ever multi-state
pollutant emissions inventory software. Designed and implemented for the Great Lakes
Commission and the eight Great Lakes States, RAPIDS was tested by the lead states of Illinois,
Indiana and Wisconsin in their joint development of the Southwest Lake Michigan Urban Areas Air
Toxics Emission Inventory. At least 49 toxic pollutants were targeted, in addition to several other
important nontoxic compounds.
Southwest Lake Michigan Pilot Study
Illinois, Indiana and Wisconsin, working together through the Great Lakes Commission, have
completed the first multistate inventory of emissions of toxic air contaminants that are identified as
being potentially harmful to the Great Lakes ecosystem or human health. Specifically, these states
created an inventory of small point and area sources of toxic air emissions from the combined
12-county urban areas of Chicago, Gary and Milwaukee.
Scope Study for Expanding the Great Lakes Toxic Emission Regional Inventory to include Estimated
Emissions from Mobile Sources
This study explores expanding the Great Lakes Toxic Emission Regional Inventory to include
estimated emissions from mobile sources.
Comments or questions about the regional inventory? Contact Project Manager Carol Ratza,
cratza()elc.org.
Air Quality in the Great Lakes Region
Search Index
D-162
-------
Overview of the RAPIDS System
-Great Lakes
Commission
Overview of the RAPIDS System
RAPIDS is a client/server system
consisting of an ORACLE
back-end database (version 7.0),
designed using ORACLE CASE
tools, and a suite of front-end
applications developed using
various software tools (e.g.,
primarily PowerBuilder and
S AS). The RAPIDS system
includes the following
components:
• RAPIDS Database
• FIRE Upload
client/application
• Data Import and Data
Export client applications
• Query
• OC Checker client/application
• Report Generator
• Data Converter client/application
• Emission Estimator client/application
Emissions
Calculation
FIRE
Upload
AIRSJAFS
Converter
Exit RAPIDS
RAPID
The RAPIDS Database
The RAPIDS database includes the following components:
o An ORACLE back-end database consisting of various
ORACLE data tables of emissions data and estimates
located on a separate (i.e., separate from the front-end
client applications) file server at each participating state;
o A set of Data Entry client/applications developed in
PowerBuilder that consists of various forms/screens to
enter various types of emissions data, and emission
estimates derived external to RAPIDS;
D-163
-------
Overview of the RAPIDS System
\
o A set of ORACLE data tables of emissions data and
estimates located at the Great Lakes National Program
Office (GLNPO) containing emissions data and estimates
obtained from each of the participating states (i.e., a
regional database of emissions data and estimates).
RAPIDS includes a client/application that uploads
(GLNPO Upload) each state's set of ORACLE tables to the
regional repository located at GLNPO.
RAPIDS Data Model
The RAPIDS data model is the cornerstone of the RAPIDS
system and includes entities that are shared by most mission
critical applications. As more applications are added to the
RAPIDS system, the core data model will be extended to
support them.
FIRE Upload
A FIRE Upload client/application (this application is under
development) that will upload the emission factors contained in
FIRE into a reference table used to calculate emissions. The
Factor Information Retrieval System or FIRE is an emission
factor database repository developed by U.S. EPA. The emission
factors contained in FIRE have been incorporated into RAPIDS
and used within the system to compute emission estimates for
certain source categories.)
Data Import and Data
Export
Data Import and Data Export client applications facilitate the
import of emissions data and estimates maintained by the states
external to RAPIDS into the back-end database, and the export of
data from the back-end database into ASCII files (i.e., import file
format).
Query
The QUERY application provides the user with access to
InfoMaker (the InfoMaker software is not provided with
RAPIDS), a powerful, easy-to-use ad-hoc reporting tool that lets
users query the RAPIDS database and create custom reports of
D-164
-------
Overview of the RAPIDS System
data. This ad-hoc query and reporting tool supplements the
pre-defined reports created by the Reporting client/application.
~ QC Checker
A QC Checker client/application that perform various statistical
checks on the emissions data and estimates contained in the
ORACLE back-end database. A sample QC Checker screen
shows the various options available to the user for performing
these statistical checks. Additional statistical checks are expected
to be added to the QC Checker application over time.
Report Generator
A Report Generator consisting of various
client/applications that generate summary reports
of the emissions data and estimates contained in the
ORACLE back-end database. A sample Tier 1, 2, 3
Reporting screen shows the various reporting
options available to the user for this particular
report.
Data Converter
A Data Converter client/application converts the
emissions data and estimates into U.S. EPA's AIRS
Facility Subsystem (AFS) transaction records. A
sample AIRS Converter screen shows the various
options available to the user for creating AFS
transaction record from RAPIDS.
Emission Estimator
An Emission Estimator client/application allows the
user to compute emission estimates using a variety
of emission estimation techniques [e.g., product of
activity data and an approved emission factor,
speciation of either paniculate matter (PM) or
VOC emission estimates or user-defined
algorithms] that match pre-established
D-165
-------
Overview of the RAPIDS System
SCC/compound-specific methodologies listed in the
Air Toxics Protocol. (The protocol is a
comprehensive document that describes the
methodologies the participating states will use to
compile the regional inventory, including the
procedures to resolve differences of opinion). A
sample emission estimation screen shows the
various options available to the user for estimating
emissions.
RAPIDS software was designed with several features:
• An easy-to-use Windows format built with state-of-the-art database management tools.
• A menu-oriented design with an interface that maintains a consistent "look and feel" throughout the
various system modules.
• Data storage and management for point source and area source emissions estimates. Although a
fugitive emissions data model is not presently a RAPIDS feature, the system is designed to
accommodate this component at a later date.
• Specific emissions estimates for states, facilities, devices, processes or streams can be incorporated.
• Source-specific emission factors can be stored and used to calculate emission estimates.
• Batch and interactive emission estimation capabilities, using emission factors, activity data, mass
balance and speciation.
• Batch input and on-line updates.
• Various QA/QC checks and routines on the emissions data and estimates are performed, including
range checks, outlier checks, acceptable value checks, missing value flags and other techniques, to
identify suspicious or unacceptable data.
• Point and area source data and estimates stored in RAPIDS are converted into the U.S. EPA's
Aerometric Information Retrieval System (AIRS) transaction format.
• Emissions summary report generation.
Comments or questions about RAPIDS? Contact Project Manager Carol Ratza, cratza@elc org.
Return to the RAPIDS Home Page
D-166
-------
A747
Table 1: List of Target Compounds for the Regional Toxic Air
Emissions Inventory
|iMi&|&
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Arsenic
Atrazine
Benz(a)anthracene
(1 ,2-Benz(a)anthracene
Benzo(a)anthracene)
Benzo(a)pyrene
Cadmium
Carbon tetrachloride
Chlordane
Chromium
Chrome VI
Chrysene
(Benz(a)phenanthrene)
Cobalt
Coke oven emissions
Copper
1,2-Dichloroethane
Diethlyhexyl phthalate .
(Bis(2-ethylhexyl) Phthalate)
Di-n-butyl phthalate •'
Di-n-octyl phthalate
Dioxins
Ethyl benzene
Fluoranthene
(1 ,2-Benzacenapthene
Benzo(jk)fluorene)
Heptachlor
Hexachlorobenzene
Hexachlorobutadiene
Hexachloroethane
Lead
AiKylated lead compounds
Manganese & compounds
Mercury
Methoxychlor
Dimethoxy-DDT
Methylene Chloride
Methane Dichloride
Freon 30
Naphthalene
)«E BSsjSSm
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
' •jyffifr^svrTTox to ' LJstr ^a*
sgl^GTMiltfiilcw^S
"^P^>oran}&sim^^^
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
<$ijM3jCAA£ij&<5
^j^^ffiijji
Yes
Yes
Yes
i3£i&&ii&&£&
7440-38-2
1912-24-9
56-55-3
50-32-8
7440-43-9
56-23-5
57-74-9
7440-47-3
18540-29-9
218-01-9
7440-48-4
8007-45-2
7440-50-8
107-06-2
117-81-7
84-74-2
117-84-0
100-41-4
206-44-0
76-44-8
118-74-1
87-68-3
67-72-1
7439-92-1
7439-92-1
7439-97-6
72-43-5
75-09-2
91-20-3
10
D-167
-------
A747
gsugg
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
••*]"Mgsi ssgsa^MHW'PjM*^^i!8a'ai^Bj
'viffiySp'J^ffy^wia^iSScSiK^
Nickel and compounds
NI carbonyl
Ni cyanide
NI subsuffide
Parathion
Pentachtoronrtrobenzene
(PCNB) (Quintobenzene)
Pentachlorophenol (PCP)
Phenol (Carbofic Acid)
Total polychlorinated
biphenyls (PCBs)
Total polychlorinated
dibenzodioxins (PCDDs)
Total polychlorinated dibenzofurans
(PCDFs)
Total polycydic aromatic
hydrocarbons (PAHs)
Porycyclic organic matter (POM)
2,3,7,8-tetrachlorodibenzo-p-dioxin
(TCDD)
2,3,7,8-tetrachlorodibenzofuran
(TCDF)
Tetrachloroethene
(Tetrachloroethylene
1 , 1 ,2,2-Tetrachloroethylene
Perchloroethylene PERC)
Trichloroethene
(Trichloroethylene)
1,1,1-trichloroethane
2,4,5-trichlorophenol
2,4,6-trichlorophenol
Trifluralin
(2,6-Dinitro-n.n-dipropyl-4-(trifluoro-
rnethyl) benzenamine)
gjgsffijjassB
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
fifiSmbWtUstf^
^^Pnm^Soi^j
-
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
3|Pli^lil|j|| |j|l
wK$lg?$iJii!$$-
13463-39-3
557-19-7
12035-72-2
56-38-2
82-68-8
87-86-5
108-95-2
1336-36-3
1746-01-6
51207-31-9
127-18-4
79-01-6
71-55-6
95-95-4
88-06-2
1582-09-8
1. Compounds listed (among others) on U.S. EPA Great Waters Program's list of targeted toxic chemicals.
2. Compounds originally targeted by the Great Lakes Commission. The full GLC list now includes all 49 compounds Csted
above.
3. Compounds identified (among others) in the U.S. Clean Air Act. as amended in 1990 (Section 112 (c)(6)).
D-168
-------
niques and ools
Guidance and Assistance Tools from the
Emission Factor and Inventory Group,
OAQPS, U.S. EPA
Presented by:
Anne Pope, U.S. EPA
Mary Anne Barckhoff, Lockheed Martin
* r^^
k*s*f*
^4f$*&&%
-j--•-> ••• ' • "' •-••-
Continuous Emissions Monitoring
Source Tests
Material Balances
Emission Factors
ON
vo
-------
o
s
An Emission Factor is a ratio
that relates quantity of pollutant
emitted to the activity level of a
specific source.
(Example: kg PM/Mg coal burned)
Volume I — Point and area sources
Volume II - Mobile sources
Contains criteria and HAP emissions information
' Types of sources emitting criteria air pollutants
v General process descriptions
' Identification of potential release points
v Emission factors for controlled and uncontrolled
processes
•ndooAJrCHEFCD
IPO, EPI. CHIEF MS, WW> P*t. Hx CHIEF
T & es"
Pollutant or source category specific reports
that contain detailed emission Information on:
•/ Types of sources emitting hazardous air
pollutants
^General process descriptions
v Identification of potential release points
v Emission factors for controlled and
uncontrolled processes
v Reviewed by S/L agencies, industry,
consultants, and EPA
-------
http://www.epa.gov/oar/oaqps/efig/
Emisiion Factor and Inventory Group
Office of Air Quality Planning and Standards,
US EPA
Receive CHIEF Newsletter (no longer available in
print format) via email
Receive timely EFIG news updates via email
How to Subscribe:
- Email a message to:
lists0rverQunixmail.rtpne.epa.gov
- type "Subscribe CHIEF F/rsfname Lastname"
v What is It?
- AP-42 Is loaded on our fax machine to be retrieved
from your fax machine
v How do I use K?
• CaH (from a fax machine) (919) 8414626 or 0548. A
recording will coach you through the process.
v Suggestion!!!
- Choose document 000001 (the code list for the
Document Index) as your first request
v' For help...
• Call Info CHIEF (919) 641-5285
v Contains EPA reports and databases in
one easy-to-use CO-ROM including:
-FIRE
- L&E's
-AP-42
- SCO List
- SIC Ust
» AP-42 Background'
Documents
Install copies
-FIRE
-TANKS
NEW (version 5.0)1
• ENP Documents
-Adobe Acrobat
Reader *
-------
|»^
»4
NJ
Version 5.0 will be available
Fall 1997
'NEW! Adobe
Acrobat format
v Verity search engine
' Links between EIIP
and AP-42
documents
System
Requirements
• Requires
IBM-compatible PC
• 486 or better
- Microsoft Windows
• CD-ROM reader
Call Into CHIEF for more Information,
(919) 541-5285
Calculational software to estimate emissions from
organic liquid storage tanks
Based on revised equations in Chapter 7 of AP-42
Calculates emissions for
' Horizontal and Vertical Fixed Roof Tanks
- External and Internal Floating Roof Tanks
•• Underground Storage Tanks
Calculates total VOC and individual HAP emissions
Calculates monthly and annual emissions ^^
Microsoft Access program
Includes emission inventory data for 1985-1995
Emission reports available by:
•• Pollutant source-type grouping (chemical
manufacturing)
•• National, state, county or non-attainment area
-Year
Summary reports export to dbf, txt, and wkl file
types.
Searchable database containing EPA's
recommended emission factors
•> Criteria pollutants
* Toxic pollutants
'Complete Source Classification Code (SCC)
Usting
v Standard Industrial Codes (SIC)
v Available for download from EFIG web page
(www.epa.gov/oar/oaqp8/efig/)
-------
.:%f^3(ffl|F;fl
Staffed under contract by
LOCKHtCD MARTIH
Locate Emissions and Inventory Material
Help with using databases and CD-ROM
Help with CHIEF BBS and Internet problems
On-site training
-------
D-174
-------
Presentation Notes
Mike Fishburn
Panel #3: Air Toxic Emissions Inventory
When Are Emissions Inventory Tools Adequate?
Perceived Deficiencies, Problems and Some Solutions:
Training and Materials
Reading List
Satellite Training/Video Tapes/PC-Based Training
Accessibility of Data and Available Tools
There were a number of tools and sources mentioned in the comments (AP-42, FIRE,
L&E documents, TTN BBS (CHIEF, CAA, EMTIC, etc), SPECIATE), but there was
also a need expressed for improvement in tools in a variety of areas.
New SPECIATE Database
Guidance
List of Chemicals?
List of Categories?
Level Setting
Help with Methodology and Format(s):
What type of inventory information and format is most useful to the modelers and the
people that do exposure and health risk assessment?
Is a holistic inventory (every possible source category) necessary or only an inventory
of selected categories?
What about hot spots?
D-175
-------
Consistency
There is a need for a concise list of toxic area source categories to be studied that
would be consistent across states and locals (or a list of chemicals to be inventoried
and related source categories).
There's a perceived need for EPA to assist in establishing consistent across-the-board
"levels of significance" in regard to HAPs reporting, i.e., how quantity (1 Ib. per year,
10 Ibs. per year, 10 tons per year, etc.) of a particular HAP should be reported because
levels at or above that should be of concern.
Consistency (cont.)
There may be a need to develop a consensus-type process, similar to the Emissions
Inventory Improvement Program's (EHP), to compile existing inventory methodologies
and/or develop new methodologies. This would help assure that different states and
locals used the same steps or procedures in developing toxic emissions estimates.
Where should inventories be sent (i.e., what database)? What format(s) should be
used?
Specific or Special Guidance Issues
Improved speciation for on-road and off-road
Guidance in how to use the new MOBTQX module
Relating Emissions Levels of HAPs to Exposure and Health Risk Assessment
There is a need for some way to meaningfully relate health risk assessment and impacts
to emissions levels so that emissions data become more meaningful as a consequence (a
weighting scheme, screening model?) In this way those category/chemical combinations
that are more significant could be put into priority bins that would allow everyone
involved to better focus their efforts.
Atmospheric Transformation/Secondarily-formed Pollutants
Provide some means of dealing with secondarily formed pollutants without running a
regional oxidant model. Current tools are not very amenable to doing a good analysis
D-176
-------
for things like formaldehyde.
TRI/EI Overlap
Concerns were expressed about the seeming overlap in responsibilities between
Emissions Inventory and the Toxic Release Inventory (TRI). There needs to be a
clarification of the areas of responsibility.
Responsibility for Establishing New Toxic Emissions Inventory Requirements
Federal Role
Role of the State and Local Agencies
The Legalistic Approach-Because It's Required
Industry Accountability
Health Risk
Conclusion
If the priorities/needs of Emissions Inventory personnel in regard to toxic emissions
inventories could be reduced to no more than a handful of major points they would be:
• Update current tools and make them easily accessible (i.e.TTN-BBS and CD-ROM),
and provide assistance in developing emission inventories to state and locals by
developing a reading list and conducting training.
A particular emphasis should be placed on updating SPECIATE if it is
inappropriate for use with point source toxic emssions inventories.
Continue to provide updated HAP emission factors in FIRE and AP-42.
Conclusion (cont.)
• Create a uniform method for database development, develop an air toxic list of
pollutants that are important, develop a significance level for each pollutant to
minimize database size and maximize its use, and recommend a particular method for
developing emissions so that there is a consistency within the state as well as between
states (i.e. use AP-42 for point sources, etc.)
• Develop a requirement (regulation) that sources above certain significance levels must
report emissions and create standards or some sort of ambient levels as a goal for
specific toxic pollutants. This could also require that industry be held accountable for
D-177
-------
its toxics emissions as well as criteria pollutants. And, if everyone uses these same
parameters for inclusion in an inventory, then this would be one step towards providing
consistency in both data collection and comparison.
Conclusion (cont.)
• Do whatever is regulatorily necessary to provide support to existing state air toxics
programs.
• Ensure inventories are compatible with the models likely to utilize the emissions
inventories.
D-178
-------
Panel 3: Toxic Air Emissions Inventory
Additional Comments
Who should have the responsibility for establishing any new toxic emissions inventory
requirements? Is there a need?
Federal Role?
"Should the federal government be the primary party responsible for promulgating
programs and regulations aimed at emissions reductions of hazardous air pollutants?"
Role of the States and Locals
"How many states and locals are actually doing anything with toxic emissions
inventories on a regular basis and what level of involvement do they want? For
instance, is there the political will on the part of states to develop their own regulations
if that was necessary to "fill the holes" in any federal rules that might be promulgated
(Urban Area Source Program Team)?
"Who will bear the expense of any additional inventory requirements?"
Legalistic Approach-Because It's Required
"There is currently no federal requirement for area, non-road and on-road mobile
source inventories toxic emissions inventories. What's on the horizon legally? What
will be the outcome of the work on implementing sections of CAAA 112 that is being
done through EPA's Urban Area Source Program?" What form should the any new
guidance take?
"A recommendation is that EPA concentrate on a umbrella-type structure or framework
(e.g., scientific research shows that a certain number of chemicals from certain
categories cause health problems at X levels and a standard is to be established not to
exceed X level) while leaving states and locals the flexibility to develop and implement
the regulations and programs required to achieve reductions. The initial presumption
should be that the states and locals should be and will want to be involved in
developing the regulations and programs appropriate to their areas. That level of
flexibility, built into the program, will assure that any new toxic emissions programs
and regulations are more targeted and address local area-specific problems."
"Another point of view on the issue of EPA standard-setting is that it may not be
workable to rely on the EPA to set the standards because it may take too long and may
D-179
-------
not be workable. The history of the Maximum Achievable Control Technology
(MACT) standards was cited to support this view of EPA in the standard-setting role.
In this view states would assume more of the standard-setting role rather than waiting
for EPA to take action."
"There is a legal requirement that all point sources establish their permitting category
(Title V or non-Title V). To do this, a fairly thorough accounting of HAPs must occur
to establish whether facilities are over or under the 10/25 ton thresholds for Title V
permitting"
"Some States have air toxics programs that require an accounting and evaluation of
many of the HAPs listed in the federal program.
"The experience of developing the 1990 Base Year Inventory and the 1993 Periodic
Inventory have made it clear that while area source and nonroad mobile sources are
becoming increasing larger percentages of the overall anthropogenic emissions
inventory the quality of data available about activity levels is relatively poor. While
some states may require that area sources report if they exceed fairly low threshold
amounts there may be an equal or larger number of states that may be estimating
emissions based on surrogates such as population or employment because of the paucity
of actual information about activity levels. Since the area source and nonroad mobile
source categories may either not lend themselves to survey work or survey work would
be too resource intensive and expensive, there may be an issue of how to get reliable
data to estimate toxic emissions. It may be counterproductive to raise people's level of
concern based on emissions estimates allocated to a county-wide area when it may be
that only the gas station at 12th and Vine is the source of the problem in its proximity
to a daycare center.
There may be very little enthusiasm on the part of the federal government to mandate
that such information be provided and collected. And, given the political climate and
proximity to small business association lobbies, it seems highly unlikely that states will
be requesting additional authorities to collect such information.
Unless there is a clear federal mandate or requirement placed on states or locals to
develop toxic emissions inventories and/or the grant funding to develop inventories
some panelists are of the opinion that states and locals that already have toxic emissions
inventory programs in place will be likely to continue, some additional states and locals
will comply with a voluntary program, but still others will not develop the inventories
and may legitimately make their case that their agencies are resource constrained and
are already having difficulty in meeting federal requirements."
Industry Accountability
D-180
-------
"Industry has been held accountable for its emissions of the criteria pollutants since
the onset of the Clean Air Act. Why shouldn't there be a comprehensive accounting of
what toxics are being emitted into the air?"
Health Risk
"Is there a significant threat? The lesson of the new NAAQS: A frequently heard
comment made in response to the proposed revisions in the NAAQS is that EPA has
failed to make its case that there is a significant health risk sufficient to outweigh the
costs entailed in achieving even greater emissions reductions than are currently
mandated. If regulations are promulgated that mandate SIP-like targets for reductions
in HAPs, then it seems critical to the successful implementation of those regulations
that EPA make a strong, scientifically-based case that there is, in fact, a significant
health risk from the exposure to various HAPs. It might be, given the availability of
the evidence, that the risk from HAPs should be prioritized or that tiers, bins, or other
groupings be established that would convey the level of associated risk. Anything else
might result in inefficiencies, i.e., trying to inventory and control everything."
"Washington State has developed Ambient Source Impact Levels (ASILs) which are
protective screening concentrations for TAPs to identify situations requiring further
risk assessment. These numbers are based on EPA's IRIS database or occupational
safety standards (threshold limit values (TLVs) if IRIS data not available. Sources
undergoing New Source Review (new sources and modified sources) must undergo
screening using ASILs."
"Once an accounting of HAP emissions has occurred, shouldn't there be some
decision-making guidelines concerning the pursuit of evaluating ground level impact?
For example, EPA could put the HAPs into "bins" that determine what emission rates
of specific HAPs are significant enough to pursue a risk assessment strategy. These
bins could also be used to determine what emission levels are significant enough to
even include in the accountability portion of an inventory. Do we need to report
emissions less than 1 Ib (for example) for many of the HAPs? There needs to be some
consideration of credibility in regards to the extent that regulatory agencies pursue
emissions data. There also needs to be some guidelines for regulatory agencies in this
matter in order to eliminate man-hours wasted in data entry and the hard copy forms
needed to provide this information to the data entry personnel."
D-181
-------
D-182
-------
AIR TOXICS MODELING
- OVERVIEW -
Prior Emphasis on Criteria Pollutants
Greater Focus on Air Toxics
Past "Identity Crisis" for Air Toxics
Hopeful Future
— Emphasis in CAA Title III on Risk
— Lessons learned from criteria pollutants
D-183
-------
AIR TOXICS MODELING
- DISCUSSION ISSUES --
Models for Air Toxics Assessments
— Similarities to criteria pollutants
— Dense gas
— Chemistry
— Deposition
— Evaluation
Input Data - Emissions/AirQuality/Meteorological
Screening Models
Models - EPA vs Others
Averaging Times
Sources of Meteorological Data
Chemical Transformation
Variable Emission Rates
D-184
-------
DIRECTIONS OF AQMG PROGRAM
National Assessment to Demonstrate Risk
Reduction
— Urban applications & evaluation
— Comparisons with national/regional
assessments
Assessments for Individual Source Categories
Outreach
— Workshops
-- Model Clearinghouse
- SCRAM Website
Assistance to States on Risk Assessments
D-185
-------
D-186
-------
Lecture Notes For The National Speciality Workshop of Technical Tools for Air Toxics
Assessment, Held June 17-19, 1997 in Research Triangle Park, NC
Dispersion Modeling Of Pollutant Impacts
by
John S. Irwin1
Atmospheric Sciences Modeling Division
Air Resources Laboratory
National Oceanic and Atmospheric Administration
Research Triangle Park, NC 27711
United States of America
ABSTRACT
Since 1973 the U.S. Environmental Protection Agency (EPA) has provided a
selection of numerical air quality dispersion modeling methods. Where once these
models were distributed using computer magnetic tape, the distribution now is made via
the Support Center for Regulatory Modeling which is available either through an
electronic bulletin board service (telephone modem access, (919)541-5742), or through
an Internet web page, http://www.epa.gov/scram001/. The goal of this presentation is
to provide a basic understanding of the kinds of air dispersion models available and the
purposes for which these dispersion models are best used.
For pollutants that can reasonably be treated as inert, Lagrangian models are
made available, as such models provide direct tracking of individual impacts from each
of the sources simulated. For chemically active pollutants, where the pollutant of
interest is primarily the result of reactions during transport (as opposed to being emitted
directly into the atmosphere), Eulerian grid models are made available for best
treatment of the often complex chemistry.
The spread (or dispersion) of particles or gases within a smoke plume as it is
transported downwind is often approximated as having a Gaussian distribution in the
vertical and horizontal. This approximation regarding dispersion has been adapted into
many Lagrangian air dispersion models, each of which has been tailored to treat a
different set of situations, such as: dispersion from building complexes, near roads and
intersections, near coastal areas, etc. Following a brief overview of Eulerian and
Lagrangian modeling, a discussion is presented of the strengths and limitations of the
Gaussian plume dispersion assumption.
1 On assignment to the Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency
D-187
-------
BACKGROUND
The passage of the Clean Air Act in 1963 and its amendments (U.S. Congress,
1963), in particular the amendments of 1970 and 1977, formalized the need for routine
access to air quality simulation models. This act and its amendments established the
concept of air quality goals to protect public health and welfare. These goals were
expressed in the form of National Ambient Air Quality Standards (NAAQS), which were
specific maximum permitted air concentration levels. Limiting existing and future
emissions such that these national standards were attained became the responsibility
of the states. The states were charged with developing detailed plans and procedures
that when enforced would assure attainment of the NAAQS. Air quality simulation
models became the means by which demonstrations could be made that a source's
emissions would not endanger attainment of the NAAQS.
Figure 1 provides a schematic representation of some of the concepts
envisioned in the Clean Air Act. At the bottom of the cycle depicted, the basic
decisions are whether human health and welfare are sufficiently protected. If not, then
knowing the costs associated with alternative plans for reducing the health and welfare
impacts, provides useful information for deciding next steps. The goal of effective
costs/benefit analyses is to address at least the major consequences and relationships
of alternative air pollution prevention strategies. Removing pollutants from stack gases
before they are released to the atmosphere, creates a need and associated cost for the
disposal of the pollutants removed. And the disposal of these pollutants must be
properly addressed, otherwise the air pollution problem has only been transferred to
another medium, such as the soil, which may in fact have a more severe and costly
environmental impact.
If it is determined that adjustments are needed to attain the NAAQS, the next
logical step (moving counter-clockwise), is to develop an emission inventory.
Developing an inventory is usually an iterative process. An initial accounting for major
known sources is developed. This initial inventory is used as input to an air quality
simulation model. The modeled concentration values are compared to available
measured concentration values. Then a series of analyses are considered to address
whether the differences seen are 1) within acceptable bounds given known levels of
natural variability, 2) resulting from deficiencies in the application of the air quality
model, or 3) deficiencies in emission inventory. This iterative process may involve use
of at first rather simplistic air quality simulation models, to provide first order checks of
the emission inventory. Then progressively more refined air quality simulation models
can be used as needed to further refine both the emission inventory and the air quality
simulation to the level needed for the determination of health and welfare impacts.
Reducing ambient air impacts from specific sources can be accomplished by
various means, as: changing to different fuels, adding control equipment to trap
D-188
-------
material before it is released to the atmosphere, upgrading industries to use more
modern (less polluting) equipment or practices. Each option alters the emissions and
each option carries with it an associated cost.
ACCESS TO MODELING TOOLS
To provide access to available dispersion models, a User's Network for Applied
Modeling of Air Pollution (UNAMAP) was developed in 1972 (Turner et al., 1989). Six
air quality simulation models were made available to four mainframe computers that
were accessible to the public. However, the system proved too awkward to manage,
and in 1973 it was replaced by making available a computer tape of the FORTRAN
source codes for the six models. Copies of the computer tape were distributed to the
public by the National Technical Information Service (NTIS). Over the succeeding
years new versions of the computer tape were provided to include additional air
simulation models and updates to existing codes. In 1988, version 6 of UNAMAP
provided codes for 24 different simulation models, and 19 associated processors and
utility programs. The utility programs served various needs from reformatting
meteorological data from standard archive formats, to processors that provided listings
and analyses of generated results.
By the late 1980's, many of the simulation models made available through
UNAMAP had attained the status of being routinely accepted for various regulatory
assessment activities (Environmental Protection Agency, 1986). The updating of these
models had to be carefully managed and coordinated within the Environmental
Protection Agency and with the public. These demands and the advent of electronic
Bulletin Board Systems (BBS) prompted a renovation in the manner in which simulation
modeling products were provided. The Support Center for Regulatory Models
(SCRAM) BBS started operation in May, 1989. Initially SCRAM consisted of 18 models
and utility programs, with a total of about 50 files available for download. At that time,
access to the SCRAM BBS was through 4 phone modem lines. Download activity was
around 30 files per week during the first few months and has grown to an average of
about 2100 files per week today.
Today, 40 modem lines are available at any given moment via a Technology
Transfer Network (TTN). And the TTN now consists of over 15 separate BBS systems,
providing a vast variety of information and services related to air pollution control. Over
time the purpose of SCRAM was adapted and expanded to meet new user needs. In
addition to the software for various air simulation models, SCRAM now also provides
modeling guidance documents. Today, there are 46 models and utility programs made
available via SCRAM, with over 3500 files available for download. On April 23, 1996,
SCRAM became accessible via Internet. The web page address is:
hUp://www. epa.gov/scram001 /
D-189
-------
AIR DISPERSION MODELING
Air quality simulation models can be characterized broadly by the manner in
which the fate and transport of pollutants within the air are simulated. Models which
employ algorithms of the physical processes that are stated relative to the dispersing
material during transport downwind can be referred to as Lagrangian dispersion
models. Lagrangian models are typically not used if the pollutant of concern is
primarily formed through chemical and physical processes during transport, such as
ozone. The chemical rates of production of secondarily formed pollutants are
controlled by the combined effects of the emissions from all sources. For numerical
efficiency, the simulation of atmospheric chemistry is typically stated relative to a
volume of air fixed in space, through which the air moves. Such models can be
referred to as Eulerian grid models. As discussed below, selecting an air simulation
model for use for a particular situation involves not only an awareness of the pollutants
being modeled, but also includes among other factors, the emission control options
being considered and the chemical and physical process being simulated. Since this is
a general discussion to emphasize principles, specific references have purposely not
been provided. A useful general reference that discusses all and more than will be
introduced here, and itself provides an excellent cross reference to other references, is
Randerson (1984).
Eulerian Models
Shown in Figure 2 is a first-order implementation of a grid model. In this
implementation, the entire modeling domain is one box. The emissions within this
volume from area, point and line sources are assumed to be instantaneously well-
mixed throughout the volume. The chemical reactions resulting from the interaction of
the emitted species with each other and with incoming solar radiation are then
simulated to produce volume average concentrations as a function of time. One-
volume grid models are often used as aids in seeing if more comprehensive model
simulations are worthwhile. A discussion and illustration of a popular single-cell grid
model often used for screening analyses is provided on the Internet; the URL address
is http://www.shodor.org/ekma/. A more complete characterization of the processes is
obtained by dividing the modeling domain into horizontal grid cells, stacked in the
vertical. This allows elevated point source emissions to be injected into the simulation
into cells aloft and near-surface emissions to be injected in the grid cells next to the
surface, to provide a more realistic characterization of the processes, (Environmental
Protection Agency, 1990). Within these multi-grid models, the emissions are well-
mixed within each cell they are emitted. Dispersion between grid cells occurs over
time, and is typically a function of the time-dependent three-dimensional meteorological
conditions within the modeling domain.
The impact of secondarily formed pollutants, typically involves small amounts of
primary emissions from a multitude of widely dispersed sources, such as automobiles
D-190
-------
or home heating. The formation through chemical and physical processes of pollutants
during transport downwind takes time. The development of emission control strategies
for secondarily formed pollutants rarely involves adjustments unique to particular
individual sources. Hence, Eulerian grid models are very useful for assessing the
effects of secondarily formed pollutants.
Laorangian models
Shown in Figure 3 is a popular implementation of a Lagrangian model, namely
the Gaussian plume model (Turner, 1970). Based on field tests of tracer gas
dispersion, it has been determined that the crosswind concentration profile along the
surface, downwind from a point source release, appears to generally have a bell-
shaped profile (Gaussian). Even the vertical profile can be often approximated as
having a Gaussian profile. Empirical data provide a basis for estimating the growth in
the vertical and lateral dimensions of the dispersing plume as it moves downwind.
Adjustments can be made to account for the fact that the vertical and lateral growth of
the dispersing material will be faster over large trees and bushes versus over large
lakes or bodies of water. Algorithms have been developed to account for the fact that
most industrial source emissions are hotter than the surrounding ambient air, and tend
to rise. As shown in Figure 3, the physical stack height, h, is less than the effective
height of the plume, H, due to buoyant plume rise effects. Through the years, other
adaptations have been developed from accounting for possible capture of the
emissions upon release within the wake of nearby buildings, to accounting for alteration
in the downwind course of the plume and rate of dispersion due to a hill being in the
path of the plume. The plume model is limited to downwind distances for which the
dispersive state of the atmosphere can be assumed to be steady-state. This limits the
plume model to distances of order 30 km. However, the same Lagrangian concepts
can be adapted to a puff model, where the emission is simulated as a series of
overlapping puffs, which then allows nonsteady-state conditions (in time or space) to be
appropriately simulated.
The major impacts from primary emissions are usually near the release, typically
involving transport downwind of less than 15 km. Developing effective plans to mitigate
impacts from primary emissions often involves unique limitations being placed on
individual sources. One of the strengths of Lagrangian models is that they can be
fashioned to address very local (source-specific) factors that greatly affect the local
(near-field) dispersion and related concentration values. Since Lagrangian models
track independently the contributions from each source, they are very useful for
assessing the effects of primary emissions of pollutants.
D-191
-------
NATURAL VARIABILITY VERSUS MODEL ACCURACY
Natural Variability
The differences seen in comparisons of model predictions and observations of
atmospheric air concentrations may largely reflect an inherent uncertainty caused by
the stochastic nature of turbulence within the atmosphere. This component of variance
is inherent in that it is the component of the variance that can not be reduced
significantly by improving the physics of the air quality models. At best air dispersion
models provide an unbiased estimate of the average concentration expected over all
realizations of an ensemble. An estimate of an ensemble can be developed from a set
of experiments having fixed external conditions (Lumley and Panofsky, 1964). To
accomplish this, the available concentration values are sorted into classes
characterizing ensembles. For each of the ensembles thus formed, the difference
between the ensemble average and any one observed realization (experimental
observation) is then ascribed to natural variability, whose variance, on2, can be
expressed as (Venkatram, 1988):
o£ = <(C° - )2> (1)
where C° is the observed concentration seen within a realization; the angle brackets
refer to an average over all available realizations within a given ensemble, so that
is the estimated ensemble average. In (1) the ensemble refers to the infinite population
of all possible realizations meeting the chosen characteristics of the ensemble. In
practice, we will only have a small sample from this ensemble. Measurement
uncertainty in C° in most tracer experiments is much less than , hence its inclusion
in the above on is deemed negligible. Available estimates suggest on may be large, of
order (Hanna, 1993).
An illustration of concentration variability is presented in Figure 4. Project
Prairie Grass (Barad, 1958, and Haugen, 1959) is a classic tracer dispersion
experiment, where sulfur-dioxide (S02) was released from a small tube placed 46 cm
above the ground. Seventy 20-minute releases were conducted during July and
August, 1956, in a wheat field near O'Neil, Nebraska. The wild hay was trimmed to a
uniform height of 5 to 6 cm in height. Sampling arcs were positioned on semicircles
centered on the release, at downwind distances of 50, 100, 200,400 and 800 m. The
samplers were positioned 1.5 m above the ground, and provided 10-minute
concentration values.
For the purpose of illustrating concentration variability, small ensembles of six
experiments along the 400-m arc have been grouped together in Figure 4 using the
inverse of the Monin-Obukhov length, L, a stability parameter (as 1/L approaches zero,
surface layer of the atmosphere approaches neutral stability conditions).
D-192
-------
Concentration values from near-surface point sources are inversely proportional to the
transport wind speed, U, and directly proportional to the emission rate, Q. To group the
results of the six experiments together, the concentration values have been normalized
by multiplying the concentration values by U/Q, where U was defined as the value
observed at 8 m above the ground. The solid line shown for each group is a Gaussian
fit to the results for each six experiment group. As mentioned earlier in this discussion,
the lateral dispersion is seen to be well approximated by a Gaussian shape. The
scatter of the normalized concentration values about this Gaussian fit can be
statistically analyzed, to provide an estimate of the concentration variability not
characterized by the Gaussian fit. From analyses of this and another tracer study
(involving tracer injected into the emissions of a 186 m stack of an operating power
plant), the stochastic fluctuations (natural variability), the distribution of C°/ for
centerline concentration values was found to have a somewhat log-normal distribution
with a standard geometric deviation of order 1.5 to 2 (Irwin and Lee, 1996).
Model Accuracy
To set the context for the discussion to follow, it is important to realize that most
of the model evaluation results currently available in the literature are for applied
dispersion models that use ensemble average characterizations of the vertical and
lateral dispersion, the chemical transformations, and the physical removal processes.
Thus these applied dispersion models only provide a description of the average fate
and dispersion of pollutants to be associated with each possible ensemble of
conditions. Natural variability can and will result in large deviations to be seen in
comparisons of individual observations (which are individual realizations from an
ensemble of realizations) with modeling results (which are characterizing the ensemble
average result).
Figure 5a shows a comparison of observed and simulated SO2 concentration
values in the vicinity of the Cliffy Creek power plant which is located in a rural area
near Madison, Indiana in the United States. There were 6 air quality monitors near this
power plant, ranging in distance from 3 to 15 km from the power plant. This power
plant has six coal-fired boilers each capable of producing 217 megawatts of power. For
each pair of boilers there is one 208-m stack. The average hourly S02 emission rate
for 1975 and 1976 was 8.67 kg/s. For each year (1975 and 1976) the maximum
observed and simulated SO2 concentration value for each receptor is shown for several
averaging times, ranging from the 1-hour maximum to the annual average. Data files
from Environmental Protection Agency (1982) for the CRSTER dispersion model
(Environmental Protection Agency, 1977) were used to construct the data shown in
Figure 5a. Through the years, CRSTER has been replaced by other Gaussian plume
dispersion models, but the comparisons shown in Figure 5a are typical of those
summarized in Hanna (1993) for characterization of S02 emissions from isolated power
plants located in rural areas of the United States.
D-193
-------
The trend to increasingly underestimate the maximum concentration as
averaging time increases can be summarized using the average fractional bias (FB =
2(e-o)/(e+o), where e is the simulated value and o is the observed value). The average
FB for each averaging time is: 24% (1-hr), 5% (3-hr), -6% (12-hr), -9% (24-hr), -43%
(720-hr), and -77% (annual average). The standard deviation of FB (a measure of
relative scatter) is relatively independent of averaging time, ranging from 34 to 47%. If
we summarize these results by stability, it can be shown that there is a bias to
overestimate concentrations during unstable conditions and to underestimate
concentrations during stable conditions. The bias seen in the model performance may
in part be traceable to the use of the Pasquill-Gifford dispersion parameters (Turner,
1970), in which the lateral dispersion has an implied averaging time of 3-minutes
(hence the simulated lateral dispersion is too narrow for characterization of 1-hour
average dispersion), and the vertical dispersion is typical for ground-level releases
(which has various deficiencies for characterizing dispersion from tall stack emissions).
Studies during the late 1980's of tall stack dispersion in convectively unstable
conditions have shown that standard Gaussian plume models can not properly simulate
the effects on dispersion of the highly organized convective eddies. As a
consequence, standard Gaussian plume models tend to underestimate maximum
surface concentration values from tall stack emissions during convectively unstable
conditions.
Figure 5b shows the comparison of observed and simulated SO2 concentration
values for 1976 in the vicinity of St. Louis, Missouri in the United States. Seven of
these monitors were within 13 km of downtown St. Louis, five were within 13 to 36 km,
and the remaining site was about 50 km north of the downtown area. The SO2
emission inventory included 208 point sources and 1989 area sources. Although the
area sources only accounted for about 3.5% of the 29.76 kg/s of total S02 emissions,
they are simulated to account for as much as 14 to 67% of the annual average
concentration values for the 13 downtown monitors (Irwin and Brown, 1985). The
comparison results shown in Figure 5b are listed in Turner and Irwin (1983), and are for
RAM, a Gaussian plume dispersion model that is tailored for urban applications. The
average FB for each averaging time is: 0% (1-hr), -18% (3-hr), -35% (24-hr), and -2%
(annual average). The standard deviation of the FB varies as averaging time varies,
as: 59% (1-hr), 46% (3-hr), 60% (24-hr), and 33% (annual average). The results
shown in Figure 5b are typical of those summarized in Hanna (1993) for
characterization of SO2 emissions in urban areas of the United States. The dispersion
parameters in RAM are based on tracer studies of dispersion conducted in an urban
area. The tracer releases were conducted from ground-level releases and releases
from the roof of a three-story building. The averaging time of these tracer experiments
(and hence the urban dispersion curves in RAM) is approximately 1 -hr.
8
D-194
-------
CONCLUSIONS
Within the United States emphasis has been placed on the development,
evaluation and application of air quality simulation models that allow development of air
quality management plans to achieve defined national air quality goals. These plans
involve development of emission control strategies sometimes for individual sources
('primary* impacts associated with pollutants emitted directly into the atmosphere) and
sometimes for classes of sources ('secondary* impacts associated with pollutants
formed during transport). Part of the decision of which model to select is dictated by
insuring that the appropriate physical processes are addressed by the model. But,
another part of the decision in model selection is the recognition that every model is a
compromise, in that not all processes are included or else the computational demands
would be excessive. Hence, model selection often involves expert judgment based on
actual experience in the use and application of the various models available.
Typically Eulerian grid models can not treat individual source impacts, unless
these impacts are several grid cells from the source. This limitation arises from the fact
that current grid models uniformly mix the emissions within the grid cell, and thus do not
properly address the initial growth and dispersion of the pollutants. Since grid models
are most often addressing impacts from pollutants that are formed from other primary
emissions, the lack of treatment of initial dispersion affects is typically assumed to be
tolerable. Studies are showing that better characterization of the chemical reactions
may require more direct and complete treatment of the initial dispersion effects. Hence,
in the future we might expect the more advance chemical models to be 'plume-in-grid1
models, where initial 'sub-grid' dispersion will be treated by a plume model. Grid model
estimates of maximum ozone (a secondarily formed pollutant) are typically within 20 to
30% of those observed. Since the chemistry is strongly forced to track the presence of
sunlight and the precursor emissions are often strongly correlated with surface
temperatures, the model estimates show good correlation in time. The spatial
correlation is dependent on the sophistication brought to bear on the analysis and
characterization of the time and space varying three-dimensional wind field. Typically
one of the more problematic inputs for grid models is the spacial and temporal
characterization of the precursor emissions, which are most often deduced from
assumptions of land use, activities patterns, traffic flows, etc., rather than on direct
measurements of emissions.
Typically Lagrangian plume and puff models can at best only treat chemical
processes that can be approximated as simple linear transformations in time. But
Lagrangian plume and puff models can track individual source impacts, allowing for
development of source specific air pollution control strategies. Since the uncertainties
in the characterization of the direction of transport are of the order of the actual plume
width, large differences are seen when concentrations are paired in time and space.
But when comparisons are made of observed and simulated frequency distributions for
.D-195
-------
fixed receptors, current applied plume and puff models typically provide estimates of
maximum concentration values within a factor of two or three of those observed. These
differences are an order of magnitude larger than those seen for estimates of
secondary pollutants, but are understandable when one considers the lack of any
strong constraints in the dispersive processes that would insure correlation in time.
Regardless of whether the model is Eulerian or Lagrangian, modeling the transport and
dispersion of inert pollutants is an order of magnitude more difficult than simulation of
secondarily formed pollutants whose reaction rates are strongly correlated by the
availability of sunlight.
Recent investigations of the differences seen in comparisons of modeling results
with observations for inert tracers suggest that much of the differences seen may be a
result of natural variability. Future development of dispersion models and the methods
used to characterize performance of these models will likely increasingly devote more
attention to the characterization of the stochastic effects associated with atmospheric
natural variability.
REFERENCES
Barad, M.L (Editor) (1958): Project Prairie Grass, A Field Program In Diffusion. Geophysical Research
Paper, No. 59, Vol I and II, Report AFCRC-TR-58-235, Air Force Cambridge Research Center, 439
PP.
Environmental Protection Agency, (1977): User's manual for single-source (CRSTER) model. EPA-450/2-
77-013. U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
Research Triangle Park, NC 27711, 303 pages.
Environmental Protection Agency, (1982): Evaluation of Rural Air Quality Simulation Models. EPA-450/4-
83-03. U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
Research Triangle Park, NC 27711, 300 pages
Environmental Protection Agency, (1986): Guideline On Air Quality Models (Revised). EPA-450/2-78-027R.
U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC 27711,290 pages
Environmental Protection Agency, (1990): User's Guide for the Urban Airshed Model. Volume 1: User's
Manual for UAM (CB-IV). EPA-450/4-90-007A. U.S. Environmental Protection Agency, Office of
Air Quality Planning and Standards, Research Triangle Park, NC 27711,260 pages.
Hanna, S.R.. (1993): Uncertainties in Air Quality Model Predictions, Boundary Layer Meteorology, Vol. 62,
pp. 3-20.
Haugen, D.A. (Editor) (1959): Project Prairie Grass, A Field Program In Diffusion, Geophysical Research
Papers, No. 59, Vol III, AFCRC-TR-58-235, Air Force Cambridge Research Center, 673 pp.
Irwin. J.S. and Brown, T.M., (1985): A sensitivity analysis of the treatment of area sources by the
Climatological Dispersion Model. Journal of Air Pollution Control Association. (35):359-364.
10
D-196
-------
twin, J.S. and Lee, R.F, (1996): Comparative Evaluation of Two Air Quality Models: Within-Regime
Evaluation Statistic, 4th Workshop on Harmonisation within Atmospheric Dispersion Modelling for
Regulatory Purposes. Oostende, Belgium, May 1996, WTO Report E&M, RA9603, pages 535-542.
(In press) International Journal of Environment and Pollution
Lumley, J.L. and Panofsky, HA, (1964): The Structure of Atmospheric Turbulence. Wiley Intersdence,
New York, 239 pp.
Randerson, D., (Editor) (1984): Atmospheric science and power production. DOE/TIC-27601 (NTIS
document DE84005177). National Technical Information Service, U.S. Department of Commerce,
Springfield, VR, 850 pp.
Schere, K.L. and Demerjian, K.L., (1984): User's guide for the photochemical box model (PBM). EPA-
600/8-84-022a. U.S. Environmental Protection Agency, Office of Research and Development,
Research Triangle Park, NC 27711,119 pages
Turner, D.B., (1970): Workbook of atmospheric dispersion estimates. Publication AP-26, (NTIS PB191
482). Office of Air Programs, U.S. Environmental Protection Agency, 84 pp.
Turner, D.B. and Irwin, J.S., (1983): Comparison of sulfur dioxide estimates form the model RAM with St
Louis RAPS measurements. Air Pollution Modeling and Its Application II. Plenum Publishing
Company, N.Y. pages 695-707.
Turner, D.B., Bender, L.W., Pierce, T.E., and Petersen, W.B., (1989): Air Quality Simulation Models from
EPA. Environmental Software, 4(2):52-61.
U.S. Congress, (1963): Clean Air Act as Amended by Air Quality Act of 1967 (Public Law 90-148); The
Clean Air Act Amendments of 1970 (Public Law 91-604); The Technical Amendments of 1973
(Public Law 92-157, Public Law 93-15, and Public Law 93-319); and The Clean Air Act
Amendments of 1977 (Public Law 95-95), United States Code, Title 42, §1857 et. seq.
Venkatram, A., (1988): Topics in Applied Modeling. \nLecturesonAirPollutionModeling, A.Venkatram
and J.C. Wyngaard (Editors). American Meteorological Society. Boston, MA. pp. 267-324.
11
D-197
-------
DISPERSION
MODEL
CONCENTRATION
IMPACTS
EMISSION
INVENTORY
HEALTH GOALS
&
CONTROL COSTS
Figure 1. Illustration of the cycle of steps that often occur during an assessment of
the consequences of emissions of pollutants.
12
D-198
-------
Figure 2. Schematic illustration of a photochemical box modeling domain (Schere
and Demerjian, 1984).
13
D-199
-------
Figure 3. Schematic illustration of a Gaussian plume model (Turner, 1970).
14
D-200
-------
1.000 -3
8
Prairie Grass
- 400 Meter Arc
(6,11,34,45,48,5
.263 < L < -S2 m
O
I
000
Y(metere)
Figure 4a. Near-neutral unstable
ensemble normalized concentration
values. Experiment numbers are shown
within the figure.
10000 •
1000
I
O
—' 1 ' r
Prairie Grass
400 Meter Arc
(21,22,23,24,42,55)
164
-------
Meteorological Data
and
Electronic Data Transfer
Dennis G. Atkinson
U.S. Environmental Protection Agency
June 18, 1997
Meteorological Data
Required for refined modeling
Data needed (2 types):
- surface data
- mixing height data
Preprocessor needed (either):
*PCRAMMET- NWS data
- MPRM - on-site data
D-201
-------
Meteorologic
PCRAMMET processing
- surface data - 3 formats
-SCRAM - EPA (1984-1992)
-SAMSON - NCDC (1961-1990)
-TD-1440 (CD-144) - NCDC (all years)
mixing height data -1 format
- SCRAM - EPA (1984-1991); NCDC (all) ||
m%
xm
'.^.iig
:S!-:V'-*
••"!#.• :*
:'$;;>?
•> • ^ : ••"'•''•' • "••'•• -'•«!"
Meteorological Projects
Mixing Height Program
SAMSON CD - update
ASOS vs. NWS data analysis
D-202
-------
Based on studies performed by George
Holzworth (1972)
j-- j-j.«T.>
••.siTcilij
• Created two front-end programs to accept |||
surface and upper air CD-ROM data Ht
• • *'iVl«-*,,|
• HOLZI.ZIP file in the Non-EPA Models of |j|
the SCRAM website §i
SAMSON CD-Update
EPA/NCDC project
SAMSON update - (1990-1995)
- ASOS data during this period - starting 1992
(Automated Surface Observation System)
-cloud data limited to 12,000 feet
- NCDC will fill temperature/station pressure
for 10 or less consecutive missings
D-203
-------
ASOS data sensitivity analysis WA m
•HfclKa
•Si*F'j7*
Analyze affects of limited cloud f||
information on dispersion modeling ||!
(12,000 feet limit) ffff
- ceiling height, sky cover ipfi
Compare concentration estimates using l|j
NWS and ASOS data in ISCST3
1ft
^^•!ift--^p3
Electrohid Data
Support Center for Regulatory Air Models
(SCRAM) - latest modeling information
available
models (user's guides)
programs (user's guides)
guidance documentation
meteorological data
state/local modeling contacts
D-204
-------
Electronic Data Transfer
SCRAM - part of the Technology Transfer «s
Network (TTN); 24 technical areas
IS!
• URL for SCRAM:
www.epa.gov/scram001
• URL for TTN:
ttnwww.rtpnc.epa.gov
',. -;^r'';';?
D-205
-------
vo
o
(N
-------
Source Attribution
Modeling
for
DIOXIN
Dioxin = PCDD/PCDF
Term used for any or all of
49 polychlorinated dibenzodioxins
87 polychlorinated dibenzofurans
(4 to 8 Chlorine atoms per molecule)
D-207
-------
City of Dayton
Montgomery County, Ohio
September, 1995
25 measurements of
PCDD/PCDF concentrations in
ambient air (24-hour samples)
Montgomery County, Ohio, 1995 PCDD/PCDF Sampling Locations
* mSD>Um&i
D-208
-------
Measured Total PCDD/PCDF Concentrations, pg/m3
Sample
1
2
3
4
5
6
Site A
1.9
1.2
1.6
1.7
4.2
2.3
Site B
9.4
8.6
6.3
8.2
29.7
14.0
SiteC
3.3
4.1
2.2
2.9
5.2
5.0
SiteD
2.4
1.5
2.2
2.6
22.9
5.3
Industrial Source Complex (ISC) Dispersion Model
* Stack parameters...from source tests, permit
applications
* PCDD/PCDF emission rates...from
MWC tested emission rates correlated with
ESP temperature
* receptor grid and elevations...from USGS
topo maps
* meteorological data...from National
Weather Service (NWS) stations at two Dayton
airports
D-209
-------
ISC Predicted PCDD/PCDF Concentrations, pg/m3
(due only to the North or South MWCs)
Sample
1
2
3
4
5
6
Site A
0
0
26.3
13.0
5.2
1.3
SiteB
0
0
0.6
0
0.1
2.3
SiteC
0
4.0
0.5
4.3
5.6
0.2
SiteD
34.6
0
0
0
0
1.5
SHEESH! Why so bad?!?!
Some possible (extrinsic) sources of error:
1. PCDD/PCDF emission rate estimate was based on log-
normal correlation of PCDD/PCDF emission rate and
ESP temperature. Testing in 1995, 1994, 1988. Throw
out 1988, new estimate - 20% of model input.
2. NWS meteorological data consist of short-term readings
on the hour. Therefore only 24 met data points per
model run/sample period. Not good enough resolution?
3. NWS calms protocol. Any wind speed reading at or
below 1.54 m/s (3.0 knots) is set equal to 0 m/s. Ignores
dispersion during periods of light winds.
D-210
-------
Alternative Source Attribution Technique:
Chemical Mass Balance (CMB) Model
Receptor modeling technique developed for
particulate matter source attribution. Instead of
source dispersion characteristics, measurements of
ambient air particulate concentrations and constituent
species (40 elements, organic carbon, sulfate, and
nitrate). Library of source emission profiles of
particulate matter species are mathematically
compared to ambient measurements for source
attribution by source type (coal-fired boilers, diesel
transportation, road dust, etcetera).
&EPA Receptor Model
Technical Series,
Volume III (Revised)
CMB User's Manual
(Version 6.0)
D-211
-------
Why not PCDD/PCDF?
Grouping PCDD/PCDF congeners by
number of chlorine atoms results in a
profile with 5 polychlorinated
dibenzodioxin congener groups and 5
polychlorinated dibenzoiuran congener
groups.
Write pairs of simultaneous equations for
each PCDD/PCDF congener group pair:
and
XT = Fmwc>fMmwc + Fos>fMos
where:
xd, xf = measured ambient air concentrations of
dioxin and furan congener groups, pg/m3
FmWc,d> Fmwc>f = dioxin and furan congener group
fractions calculated from North MWC stack test
profile
K
D-212
-------
Mmwe = ambient total mass PCDD/PCDF contributed
••mwc
by the county MWCs, pg/m3 (unknown)
FOStd, FOJif = dioxin and furan congener group fractions
calculated from other source profile
M0, = ambient total mass PCDD/PCDF contributed
by other source, pg/m3 (unknown)
"Other source" emission profile???
Conceptual, aggregate, non-county MWC source of
PCDD/PCDF. No stack data (of course). Best
represented by Site B sample 1. Wind direction, distance
from county MWCs, high PCDD/PCDF.
CMB Results, pg/m3 (ambient total PCDD/PCDF from
County MWCs)
Sample
1
2
3
4
5
6
Site A
1.7
0.9
1.0
1.2
0.2
0.9
SiteB
—
1.1
1.2
0.8
-2.4
1.7
SiteC
4.4
4.0
1.7
2.4
3.4
4.8
SiteD
1.6
0.8
1.8
2.3
-9.2
2.8
What's it all mean??
D-213
-------
Site A and Site C: generally low
PCDD/PCDF levels, suburban sites -
MWCs should be responsible for
majority of PCDD/PCDF
Site B: urban industrial location, other
PCDD/PCDF sources likely, relatively
distant from MWCs, relatively minor
contribution from MWCs
Site D: commercial area close to South
MWC, majority of ambient
PCDD/PCDF from MWC, EXCEPT
sample 5
SAMPLE 5: Two highest levels
measured (out of 25 samples). ISC
bombed, CMB bombed. MWC
operations normal. Site B and Site D
north and south of South MWC.
D-214
-------
NWS reported calms for 11 (out of 24)
readings. No dispersion? Portable met
tower - light and variable winds.
Polyvinyl chloride burnoff oven:
batch unit, relatively low temperature
and oxygen, short capped stack, batch
run during sample 5 has evidence of
afterburner upset.
...no stack data...
It
Open Discussion:
Receptor modeling techniques may help
in PCDD/PCDF source attribution
studies. Source profile library?
ISC should be modified to employ 5-
minute average met data. Availability
of portable met towers. Availability of
source code.
D-215
-------
.#
SAMPLE 5
/NWS hourly
60 -I
50 •
40 -
30
20 -
10-
n i
•
>
-«
•
•
•
kV
0 60 120 160 240 300
WW Dfrufam,
360
SAMPLE 5
5 min. avg
60
120
180
300
360
D-216
-------
SAMPLE 5 fa* C*W
5 min. avg
60 -j
50 •
40
30
20
10
0)
, X
v«v\V
xifei"
A (P.U
i, Us')
0 60 120 180 240 • 300 360
D-217
-------
D-218
-------
Sensitivity Analysis Results
PARAMETER
Elevated vs. Flat Terrain
Rural vs. Urban Air Dispersion
Coefficients
Surface Roughness at Application
Site
Watershed Size and Proximity
Anemometer Height
Particle Size Distribution and
Density
Polar vs. Cartesian Grid nodes
Terrain Gird File
Minimum Monin-Obukhov
Length
Surface Roughness at
Measurement Site
Noon-Time Albedo
Bowen Ratio
Anthropogenic Heat Flux
Fraction of Net Radiation
Absorbed
Scavenging Coefficients
SENSITIVITY
Severe
Severe
Severe
Severe
Moderate
Moderate
Slight
None
Slight
Severe
Slight
Slight
None
None
Severe
RECOMMENDATION
Must include terrain < 1-2 km; Hills >
stack height only if > 5 km
Detailed land use analysis required
EPA-required method, or site-specific
justification
Use actual watershed area near source;
use representative points > 10 km away
Under estimates < 1km
Require stack test data for particle size
and density
Applicant selects grid
No impact on model results
Specify default values
EPA-required value for NWS site
Specify default values
Specify default values
No impact on model results
No impact on model results
Isolated events 300%, but rare
occurrence in Region 6
D-219
-------
o
-------
NEW JERSEY DEP RISK
ASSESSMENT PROCEDURES
FOR MINOR SOURCES
(since 1989)
MAJOR SOURCES (ex. Municipal Solid
Waste, Sewage Sludge, and Hazardous Waste
Incinerators; Coal-Fired Power Plants; etc.)
Risk Assessment Responsibility of Applicant
Protocol *F Analysis
Inhalation or Multi-pathway
MINOR SOURCES (ex. Chemical Batch
Plants, Degreasers, Air Strippers, etc.)
Risk Assessment Responsibility of NTDFp
Level-1/screening ¥Level-2/refined modeling
Inhalation Pathway Only
APPLICABILITY TO
MINOR SOURCE RISK
ASSESSMENT PROCEDURES
Air Permit Applications For New or Modified
Sources Which Emit One or More of the
Following 56 Carcinogens:
Acetaldehyde Hexachlorobenzene
Acrylamide Hexachloroeihane
Acrylonitrile Hydrozine
Allyl Chloride Lindane
Arsenic Methyl Chloride
Asbestos Meihylene Chloride
Benzene 4,4-Methylenedianiline
Bcnzidine Nickel
Bcr>2o(a)pyrene Nickel Subsulfide
Benzyl Chloride Nitroben/enc
Beryllium 2-Nitropropane
Bis(2-Chloroeihyl)*iher N-Niirosodimeibylan>in«
Bis(chloromcihyl)ether N-Nitroso-n-meihylurea
1,3-Buiadiene N-Nilrosomorpholinc
Cadmium Pemachlorophenol
Carbon Tetrachloride Polychlorinaied Biphenyls
Chlordane Propylene Oxide
Chromium VI 2.3,7'.S-TCDD (Dioxin)
1,2-Dichloropropane l.l.Z.I-Teirachloroeihane
Heptachlor
Formaldehyde
Ethylene Oxide
Elhylenc Dichloride
Eihylene Dibromidc
Ethyl Acrylaie
Epichlorohydrin
1,2-Diphenylhydrazine
1,4-Dioxane
Vinylidene Chloride
Vinyl Chloride
2,4,6-Trichlrophenol
Trichloroeihylene
1,1,2-Trichloroethane
Toxaphcne
Tetrachloroeihylene
Chloroform
Styrene
D-221
-------
NJDEP LEVEL-1 RISK
ASSESSMENTS
[Conducted by the NJDEP Air Permit Review
Engineer]
1. Calculate the Ton per Year Emission Rate for
Each Carcinogen
2. Use NJDEP Nomograph to Obtain Long-Term
Ambient Concentration of Each Carcinogen
3. Use NJDEP Unit Risk Factors to Obtain
Cancer Risk Factor for Each Carcinogen
4. Add Incremental Cancer Risk of Each
Carcinogen
Total Cancer Risk # 1 hi a million » PASS
Total Cancer Risk > 1 in a million w FAIL
go to Level-2 Risk Assessment
NJDEP NOMOGRAPHS
ISCLT2 Model -
5-Year Composite STAR Deck
Flat Terrain
Dense Receptor Grid
CONSERVATIVE ASSUMPTIONS:
* No Plume Rise
# Significant Building Downwash
Stk. Hts.# 30 ft ^Building Ht. = 2/3 Stack Ht.
Stk. Hts> 30 ft -^Building Ht. = 1/2 Stack Ht.
* Higher Impact of Either the Urban or
Rural Mode Used
# If More than One Stack, Add Maximum
Impacts Regardless of Location
D-222
-------
NJDEP LEVEL-2 RISK
ASSESSMENTS
[Conducted by the NJDEP Air Modelers]
1. Applicant Must Provide:
~ Detailed Plot Plan
~ Location of All Emission Points
~ Building Dimensions.
- Location of Nearby Sensitive Receptors
(i.e. homes, schools, etc.)
2. NJDEP - GEP Stack Height Analysis with
BPIP
3. NJDEP - ISCST3 Modeling with 2-3
Years of Hourly Meteorological Data
4. Review Expanded to Include 121
Carcinogens and 119 Noncarcinogens
NJDEP LEVEL-2 RISK
ASSESSMENTS (Continued)
If the Maximum Incremental Cancer Risk of
Each Carcinogen:
-*• less than/equal to 1 in a million «• PASS
-» greater than 1 in 10,000 **• FAIL (no permit)
-* 1 in 10,000 to lin a million «* Go to Step 2
Maximum Cancer Risk less than 1 in a
100,000 and Cancer Risk at All Sensitive
Receptors less than 1 in a Million «• PASS
Maximum Cancer Risk greater than 1 in a
100,000 or_ Cancer Risk at Any Sensitive
Receptors greater than 1 in a Million *& go
to Risk Management Committee
D-223
-------
RISK MANAGEMENT
COMMITTEE
Meeting Attended by: Chief of New Source
Review, Chief of Air Quality Evaluation, Air
Permit Review Engineer, Modeler, and Regional
Enforcement (facility inspector)
EXAMINES FEASIBILITY OF:
* Better Pollution Controls or Refinement in
Emission Estimates
* Reduce Proposed Hours of Operation
•* Modifying Stack Characteristics for Better
Dispersion:
~ Higher Stack Height
° ~ Change from Horizontal to Vertical
Release
* Conduct a More Refined Risk Assessment to
Better Define Impacts or Include Other
Sources at the facility (applicant or NJDEP)
D-224
-------
.>
(0
RISK SCREENING NOMOGRAPH A
Annual Impact - Stack Height 10-30 ft
o
Z
_ too too too ooo Km
Nearest Distance to Property Line (ft)
RISK SCREENING NOMOGRAPH B
Annual Impact - Stack Height >30ft
;:::;:::;::J;;j;;r ;;:">;
;....;..-'I---J--i--f-i--i-
D-225
-------
D-226
-------
EXPOSURE MODELING:
COMPONENTS, STATUS,
AND USES
Michael Zelenka
U.S. Environmental Protection Agency
National Exposure Research Laboratory
Research Triangle Park, North Carolina
National Speciality Workshop on Technical Tools
for Air Toxics Assessment
June 19, 1997
PRESENTATION OUTLINE:
1. Brief overview of the structure of laboratories conducting risk
assessment work at EPA's Office of Research and Development.
2. Define "exposure" and illustrate the "risk assessment paradigm.
3. Illustrate the evolution of approaches to human exposure
assessment efforts and human exposure modeling in particular.
4. Overview of the basic construct of exposure models currently in use
by the EPA's National Exposure Research Laboratory (NERL).
5. Discuss the needs and directions of future exposure modeling
efforts at the NERL.
D-227
-------
OFFICE OF RESEARCH & DEVELOPMENT
>• National Exposure Research Laboratory
1. Characterize the physical and chemical properties that govern exposures.
2. Develop, test, evaluate, and demonstrate mathematical models of exposure.
*- National Health and Environmental Effects Research Lab
1. Address health and/or ecological effects of exposures to man-made
stressors.
2. Determine the likelihood that these effects will occur under conditions of
environmental exposure.
> National Risk Management Research Laboratory
1. Characterize pollutant sources that require management.
2. Identify, develop, and evaluate tools and technologies for prevention, control
restoration, and remediation of environmental problems.
*- National Center for Environmental Assessment
1. National resource center for the overall process of risk assessment:
integrating hazard, dose-response, and exposure data and models to
produce nsk characterizations.
RELEVANCE OF HUMAN RISK ASSESSMENT
RESEARCH AT EPA
The Clean Air Act Amendments of 1990 address human exposures (for example):
Title I, Sec. 103(d):
• The Administrator must prepare environmental health assessments for the
hazardous air pollutants listed under Section 112(b).
... These assessments shall include an examination, summary, and
evaluation of available... information for the pollutant to ascertain the levels
of human exposure which pose a significant threat to human health and the
associated acute, subacute, and chronic adverse health effects.
D-228
-------
HUMAN EXPOSURE:
• The contact at one or more personal physical boundaries (skin,
lungs, and other organs) between a human and contaminant(s)
having a specific concentration for a specified period of time.
TOTAL HUMAN EXPOSURE:
' Human exposure that consists of inputs from all media (air, water,
food, soil, house dust) that contain the contaminants} and all routes
of entry into the body (inhalation, ingestion, dermal absorption).
SCIENTIFIC CONTEXT: HUMAN RISK
ASSESSMENT RESEARCH
i Focus of Near-Term Human Exposure Assessment Research:
> NERL:
• Uncertainties in exposure measurements (particularly,
microenvironmental data), measurement-derived exposure
models, and activity-pattern data.
> NHEERL:
• Uncertainties in mechanistic information (e.g.,
pharmacokinetics) for hazard identification and dose-response
assessment.
D-229
-------
RISK ASSESSMENT
Hazard
Identification
Dpse-Responsrx /^ Exposure
Assessment r* *< Assessment
Risk
Characterization
-—. —
Hazard Identification: Describe the adverse health effects that might occur due to exposure to an
environmental contaminant
Dose-Response Assessment: Determine the toxicity or potency of a contaminant The dose-response
assessment describes the quantitative relationship between the amount of exposure to a contaminant and the
extent of injury or disease.
Exposure Assessment: Describe the nature and size of the population exposed and the magnitude and
duration of exposure. Exposure assessment includes a description of the pathways (e.g., air, food, water, house
dust) by which the contaminant travels through the environment, the environmental concentrations of the
contaminant relative to time, distance, and direction from its source (including transformations), potential routes
of exposure (i.e, dermal, inhalation, or ingestion). and information on sensitive subgroups (i.e, children or
pregnant women).
Risk Characterization: Use the information from above steps to predict the effects of exposure to the
contaminant. Estimates are made of the likelihood that a population will experience any of the adverse effects
associated with the contaminant, under known or expected conditions of exposure.
EVOLUTION: HUMAN RISK ASSESSMENT
RESEARCH
i HISTORIC APPROACH:
• Reliance on media-specific legislation, single pathway
assessment, individual source emitters, individual (i.e., single)
contaminants.
• Dominated by ambient monitoring and regulations on sources and
ambient concentrations.
i EVOLUTION OF NEW APPROACH DRIVEN BY:
• Multimedia legislation.
• Growing recognition of the importance of multipathway risk
assessments.
• Increased capability to measure multimedia environmental and
human exposure metrics.
• Increased understanding of biokinetics and the use of biological
markers of exposure (biomarkers).
D-230
-------
EVOLUTION: HUMAN RISK ASSESSMENT
RESEARCH
HISTORIC APPROACH EVOLVING APPROACH
Problem Identification & Investigation
• Single source of contaminant • Multiple sources of contaminants
• Single environmental pathway • Multiple environmental pathways
• Single route of exposure • Multiple routes of exposure
• Single (presumptive) health • Multiple health endpoints
endpoint
Risk Assessment Decisions
• Single media focus • Multi-media focus
• Single pollutant focus • Multi-pollutant focus
IMPLICATIONS OF RISK ASSESSMENT
EVOLUTION ON EXPOSURE MODELING
i The SAB (EPA-SAB-IAQC-95-005) recommended the following:
1. Develop and implement a strategic plan to collect critical input
data for exposure models and validate the entire methodology.
2. Establish a framework to ensure that the entire range of potential
risks from multiple sources are addressed holistically within that
framework.
• To achieve these goals, exposgre model development should:
a. be linked within comprehensive models that characterize
exposure and dose at target sites, and,
b. include the ability of future exposure models to be used as
tools for prioritizing the major routes and media of concern.
D-231
-------
EVOLVING APPROACHES TO HUMAN
EXPOSURE MODELING AND ASSESSMENT
EFFORTS
Exposure Pathways:
. HISTORIC APPROACH: Source emissions and concentrations of
contaminants reach a receptor (human) through a single pathway.
. EVOLVING APPROACH: A contaminant can reach humans
through multiple routes of exposure.
Predictive Models based on Source Emissions and Transport:
> HISTORIC APPROACH: Exposure^ assessed by using either source
emissions or concentrations of contaminants in media and environmental
transport models, along with a limited number of exposure situations
(scenarios).
+ EVOLVING APPROACH: Time-activity patterns significantly influence
exposures and need to be included in modeling efforts in order to include all
exposure pathways and account for the large variability across the diverse
population of the United States.
• Much more direct measurement exposure data is needed to validate model
assumptions regarding the most significant routes of exposure, and about
how contaminants move into or through environmental compartments.
D-232
-------
• Temporal Variability at Specific Locations:
»HISTORIC APPROACH: The concentration of contaminants in
media remains constant over time.
. EVOLVING APPROACH: Contaminants can come from multiple
sources whose concentrations can vary greatly, causing
responses that are substantially different from those assessed on
a "per source" basis.
Independence pf Exposures to Multiple Contaminants or
Sources:
• HISTORIC APPROACH: Exposures assessed on a chemical-by-chemical
basis.
- EVOLVING APPROACH: Effects from multiple contaminants may be other
than additive.
D-233
-------
• Infiltration of Outdoor Air into Buildings:
»• HISTORIC APPROACHES: Buildings do not significantly affect exposures
to outdoor pollutants.
a. Building conditions are static, and indoor concentrations of airborne
contaminants can be modeled using steady state conditions.
*• EVOLVING APPROACHES: Indoor exposures from outside air are affected
by factors related to:
1) the building itself (e.g., type of construction, ventilation method),
2) environmental conditions (season, temperature differences
between indoors and outdoors),
3) contaminant morphology (reactive or non-reactive, gas or
particle), and
4) human habits and practices (adjustment of windows,
sweeping, vacuuming).
a. Infiltration of outdoor contaminants have substantial inter- and
intra-building variability making steady-state conditions less likely.
Human Exposure Simulation Models at NERL
1. pNEMr'X" (probabilistic NAAQS Exposure Model for pollutant "X")
v Different versions for different pollutants
• Currently the "Xs" are: CO*, Os*. NOz, SC-2, and particles.
* Most application has been done with the CO and Os versions.
v pNEM is used by OAQPS when considering NAAQS and for evaluating
alternative NAAQS proposals.
•j pNEM provides hourly estimates of the distribution of exposures,
Goal: To simulate the movements of people through zones of varying air
quality to approximate the actual exposure patterns of people living in a
defined study area.
D-234
-------
2. HAPEM-MS (Hazardous Atr Pollutant Exposure Model for Mobile
Sources)
* This model provides long-term (1 year or more) average exposure estimates
for pollutants generally attributed to mobile sources.
• HAPEM-MS uses ambient CO data for input and uses CO-to-chemical Y
ratios to estimate exposures to other mobile source pollutants.
^ HAPEM-MS is used by QMS and program offices - most recently used for
the Motor Vehicle-Related Air Toxics Study* (required under S. 202 of the
CAA for motor vehicle air toxics regulation).
*U.S. EPA (1993) Motor Vehicle-Related Air Toxics Study (Draft).
Office of Air and Radiation, Ann Arbor, Ml.
HAPEM-PS
' Outgrowth of 1991 critical evaluation of the Human Exposure
Model (HEM).
1 Provides annual average estimates of exposures to hazardous air
pollutants and resulting cancer incidence for populations residing
near point sources.
' Ambient pollutant concentrations are determined by dispersion
model using point source emission rates and local meteorology.
Census unit centroids are used as dispersion model receptors.
1 Model accounts for residential patterns, activity patterns, ME's,
and home-work commutes.
D-235
-------
Comparison of pIMEM and HAPEM
Ambient Air Quality Data:
*pNEM: fixed-site
* HAPEM-MS: fixed-site
« HAPEM-PS: dispersion model data
Microanvlronmental Concentrations:
• pNEM: mass balance model
• HAPEM: regression equations [indoor = (a) (outdoor) + b]
i Time/activity patterns:
* pNEM: strict time sequence of assignments to ME's, districts,
and breathing rates.
• HAPEM: fraction of time spent in each combination of ME and
district. Also, accounts for season and hour of day.
Comparison of pNEM and HAPEM - 2
Indoor Sources:
» pNEM: probabilistic algorithms for simulating emission
patterns within the mass balance model.
• HAPEM: treats as additive factor in regression equation.
Residential Patterns:
• pNEM and HAPEM-MS: people are assigned to large districts
surrounding fixed-site monitors.
• HAPEM-PS: people are assigned to small census units near
point source and larger census units away from source.
Commuting Patterns (based on census data):
• pNEM and HAPEM-MS: workers commute between large
districts.
• HAPEM-PS: workers commute between individual census
units.
D-236
-------
Comparison of pNEM and HAPEM - 3
1 Ventilation Estimates:
• pNEM: equivalent ventilation rate estimated for each exposure
event [EVR = (vent.rate}/(body surface area).
• HAPEM; does not estimate ventilation rate.
i Microenvironments:
• p/VEM: typically less than 10.
• HAPEM, 37 available.
i Time Activity Data:
• pNEM: data base currently contains diary data from 10
studies.
• HAPEM: Cincinnati, Denver, and Washington, D.C. only.
Comparison of pNEM and HAPEM - 4
Census Data:
• pNEM and HAPEM: use 1990 census data to extrapolate cohort
exposures to general population.
1 Health Effects Resulting from Exposure:
» pNEM: CO version estimates COHb level resulting from each individual
exposure event.
• HAPEM: estimates one-year and 70-year cancer incidence within exposed
population.
1 Adjustment of Ambient Air Quality Data:
• pNEM: statistical techniques are applied to fixed-site monitoring data to
simulate attainment of each NAAQS under evaluation.
• HAPEM-MS: pNEM techniques can be used to adjust input fixed-site
monitoring data.
• HAPEM-PS: model is typically used to simulate "as is" conditions - no
adjustment of ambient data. Analysts can change emission rates in
dispersion model to simulate point source controls.
D-237
-------
Comparison of pNEM and HAPEM - 5
Estimate fmnrlftl output): .^icuohw
p/VEAf: model provides pollutant concentfation and E VRby
AXDosure event for each population cohort (pNEM/CO also
r eac
i OOHb level). Output tables list numbers of
people exposed and exposure occurrences by concentration
range and EVR range for various averaging times.
HAPEM-PS: output tables list annual average pollutant
concentration and cancer incidence for each population cohort
and for total population.
HAPEM-MS: produces HAPEM-PS outputs plus tables of
average pollutant concentrations by season, hour or day,
demographic group, and microenvironment.
Comp"ter Reoiilrements;
pNEM and HAPEM. IBM main-frame.
REQUIRED OF THE NEXT GENERATION OF
EXPOSURE MODELS
EPA's exposure models play an important role in the risk paradigm and
therefore must:
1. characterize the full range of the distribution of human exposures
that can occur,
2. identify microenvironments that must be considered in exposure
assessments to urban air toxics,
3. integrate all pertinent information into a rational human exposure
framework,
4. and, ultimately, produce scientifically-based estimates of human
exposures with the ability to emphasize sensitive or "at risk" subgroups
of the population such as children, asthmatics, and others.
D-238
-------
FUTURE NERL HUMAN EXPOSURE MODEL
DEVELOPMENT:
KEY ELEMENTS FOR EXPOSURE MODEL DEVELOPMENT;
I. Basic research: needed to improve our estimates, including those that
are modeled, of exposure.
Examples:
microenvironmental concentrations
activity patterns
dose metrics (e.g., breathing rates)
II. Improvements to the models:
• More, and better use of CIS.
• Faster and more user friendly models.
• Additional data (measured and derived) and computing platforms that
allow the ease and flexibility of performing human exposure research.
• Develop a modeling "framework" so that:
a. models that characterize exposure and dose at target sites can
be linked, and,
b. future exposure models will include the ability to be used as tools
for prioritizing the major routes and media of concern.
D-239
-------
D-240
-------
MICROENVIRONMENTAL
MODELING ISSUES
Alan Huber
US EPA NERL, MD-56
Phone/FAX 919-541-1338
E-mail (preferred)
huber.alan@epamail.epa.gov
WHAT ISSUES?
Definition of a Microenvironmenl
Assessment Goals
Modeling "Art" versus "Real" Dat;
Perspective on Models
Examples
Discussion
D-241
-------
DEFINITION 01
MICROENYIRONMENT
Intersection of human presence/activity
with u characleri/able em ironment
The world of our dialy activity lakes place
in a series of microenvironmenls
While all exposure is "local"
ET(total) = EL(local )+ F:R(regional)
Assessment Goals
Population-distribution of exposure
Special group and/or
special microem ironmenl
Effective reduction oi'exposure
Residual Risk and/or Relative Risk
D-242
-------
pun suiooq amsodx? oiji o.ioq,\\
poi\i
sr« \puin isi)d pun ur?isop .\pnis o.id
qioq ui unn.iodun si oouoios [U.insiin]^
)snui
-------
Characteristics in Model Selection
Pollutant limission Sources
Chemistry o! Interest
Human Aclivily Patterns lor Hxposiire
Spaetial and Temporal Scales of Interest
Available Meteorology
Numerical Methods
SOURCE IMPACT
Q=1 g/s(35tons/yr)
0.9
--. 0.7
1
3 n«
•— • 0.5
O
S** 03
0.3
C
O
o
§ 01
U
-0.1
....
~~
\
•1>
S^, ....
M £
f\ 1
m e
„
n 1
— Point
4km2Area
n 16km2Area
DISTANCE (km)
D-244
-------
SOURCE IMPACT
Q»1 g/s (35 ton»/yr)
m
0.60
0.50
«" 0.40
4 0.30
10.06
0.05
0.04
0.03)
»
~\
-,\
,. i - j
\
*. •
•.
*.
—
^
.
— _
_j
s,
'•
..u..j 1.
...] L
1 !
s
S
~*7i
"
-
•>
^
;i
T
!
-T
"
X^
-i su^
""*••,
i • ""•
T
s
* • t
" V1
,
*
-h
!
i
•
-
•!^,
•> to IA M tn iqwintan *n w w in iqinvMnn
" N ft *V 10 (Dt^flDOJO Q O O O 9OOOO
»• W r> ^ 85 o h»09)O
T-
DISTANCE (km)
Point
•• 4km2Af»a
16km2Area
BOX MICROENVIRONMENT
Ambient Air Exchange
0 9
0
1 07
ft a'
O
U
OOf
U
S
01 03
0 1
t
i "0°
A 1* nc
0 D
! D
1 D
o Q
a !
1
.. 00 \
n •>
2 4 1
)fj(j<5f>Oi
pDaC
3°°
0 1
lODD'-'"
5 2
)QQ0nr>«
2 2
too
8 3
O Ambient
D 0.1 ACM
0 0.2 ACM
4 A 1 ACM
Time (minutes)
D-245
-------
BOX MICROENVIRONMENT
In-Box Source (0*1 ug/m3)
[ i
! j
! 1 _,00^
c-°
o i
0 j
0
o 1
j,J|aoOoC"jlc'r'aiJU1'
-f\r~\ —
,0oooo<
innnnni
)OOOOO<
rannoDt
>oo
,00—
O6o~6ooo 660066 60600000606000606
1 ! ! ! i
-2
10 16 22
Time ((Minutes)
28
O 0.1 ACM
Q 0 2 ACM
34 O 1 ACM
D-246
-------
Real-Time Monitoring of Polycydic
Aromatic Hydrocarbons
EcoChem PAS 10021 and 2000
> Real-time monitoring of PAH on the surface of fine
airborne particles
"Principle of operation:
-Photoionization of surface-adsorbed PAH
-Photoelectrons diffuse rapidly to the wall
-Positively charged particles continue to a collection
electrometer
-Measure the positive current in pA or fA
-Output to PC or datalogger
D-247
-------
Conversion factors, pA to ng/m3
> Depend on the source
> Range from 1000 to 3000 ng/mA3 per pA
»Published work (Burtscher, Niessner, and others)
substantiates the utility and range of conversion
factors
> Semi-quantitative in ambient air
> Greatest accuracy requires specific calibration
Response
response to vapor
•> Particles less than about 1 um
•>PAH with 3 or more rings
> Highest for 5-7 ring PAH, such as benzo[a]pyrene
* Limit of detection about 10 ng/m~3 for PAS 10021;
much lower for PAS 2000
D-248
-------
Differences between PAS 10021 and 2000
>PAS1002i
-Ultraviolet lamp, 187nm
-4 L/min
-LOD10ng/mA3
-PAH, a few metals, and
some organ ics
-Lamp intensity calibration
required
>PAS 2000
- Excimer laser, 222 nm
-2 L/min
-LOD<1 ng/mA3
-PAH, perbromo- and
periodo-alkanes
-Software compensation
Comparisons with integrated sampling
(PAS 1002i)
Hn more than 40 indoor and outdoor
microenvironments, agreement was good, with a
conversion factor of 1000 ng/mA3 per pA.
> Agreement was best (RA2 > 0.6) with the sum of the
carcinogenic (B2) PAH as determined by GC/MS.
D-249
-------
Activity patterns and ambient monitoring
> Rapid response and high sensitivity make the PAS
valuable for following activities that generate PAH,
-Example: PAH at a veterinarian's office
* Ability to collect real-time averages make it valuable
for monitoring ambient concentrations of PAH, traffic
patterns, other combustion.
-Example: Ambient monitoring in DC, summer of 1994
Current activities
^•Comparison of PAS 1002i, ZOOOi, and integrated
sampling for PAH inside and outside 8-9 RTP homes
-Denuded and non-denuded integrated sampling to
determine phase distributions
D-250
-------
Toxic Air Pollutants
iased MROjyiieiii
Sherry IK. Livingston1, Fardin Oliaei1,
George L. Bollweg1, Sally Paterson2, and
Donald Mackay2
1. Division of Air Quality, Minnesota Pollution Control Agency,
520 Lafayette Road, St Paul, MN 55155, USA
D-251
-------
^v^SiS*
Does not predict actual impacts
^Ss&S&fe
Estimat
chemical andI physical^^properties of substance:
- molecular weight r
- solubility
- vapor pressure
- octanol-water partition coefficient
- melting point
- food chain multipliers
— degradation half lives in air, water, soil and sediment
v. *'•• J 4 -. - t •••-*•- -I T» "•*•. . •
D-252
-------
- digestion Cancer Slope Factor (SF)
- Threshold Limit Value (TLV)
Ecological Effects:
, .-ifAquatjgLJb
""•" "- -- -'
-i^l
m^
Water/ingestion RfD
Water/lngestion SF
Aquatic Biota/lngestion RfD
Aquatic Biota/lngestion SF
Soil/lngestion RfD
SoMngestipriSF;
D-253
-------
Air
Wat*r
Soil
2.178*11
1.380*10
2.040*11
1.380*10
1.0*3
8.0*0
2.0*-2
2.178*14
1.104*11
2.040*10
2.760*8
0.03
0.15
1.2
1000
1650
1170
D-254
-------
Predicted concentration (ug/g) in each compartment
CAS* a}r water ao.blota soil terr.flora
Substance
1,1,1-McMorMtham 71454 Z7>E-03 JJ8E-06 6.S1E-05 1J8E-05 7J5E-05
dtoxlni (total 21S,7,8t) 1940«-7*-3 4.18E-07 2J9E-06 ZOOE-01 4.82E-01 4.11 E+02
total chnmlum 744(M7-3 2J2E-17 340£41 1JOE-01 3J6E*00
hexacMoratiwuWM 118-74-1 3J8E44 2.12E-08 2J9E41 7JOC-OS - 4X7E-01
"
D-255
-------
water aa.blota soil
Substance
terr.flora
1.1,1-trichloroethane -2.99
dlpJdns (total 2,3J,8sX ^.81
-6.94
:-7.16
-7.60
-4.07;
-10.40 -8.64
RfC/RfD Cancer Unit Cancer Slop* -
Inhalation Ingestlon Rlsk(lnhal.) Factor(lngMt) TLV
Substance (rng/m3) (tng/kg/d) 1/(ug/m3) 1/(mo/kg/d) tmo/m3)
U1.1,1-TCA 1.0E+00 9.0E-02
. dloxlns
1910.
1.3E+00
6.2E*03
D-256
-------
HjjSubstahce ^Inhataooni
lqg(toxicity in mg/kg/day
Substance alr/lnhal.RfC alr/lnhal.UR alr/TLV watar/lngesLRfD waterflngesLSF
-2.44
-5.90
1.63
D-257
-------
Substance IribestRfD IngestSF thgestRfb IngestSF IngwtRfD IngettSF
-6.55 -9.36 -7.69
2.91
6.90
Substance
Aquatic Ufe Pisc-WIIdlife \ v "
Criterion(ug/» Criterionfuq/ll water/ALC water/PWC
263.0
1,1,1-TCA
dioxlns
totaj chromium
hexachlorobenzene 0.02
8.50E-09
-4.82
4.02
5.45
D-258
-------
1,1,1-TCA
dioxins
.total
HCB
1.08E+09
1.23E+21
1.ME+00
.^-MIEfll'i'-:: _.
alr/RfC
terrflora/SF
dioxins '-.'.-
mercury
PCBs
silver
benzo(a)pyrene
cadmium
dibromochloropropane
chromium (VI)
Index (log)
21.09
19.80
16.92
16.12
16.05
16.00
16.94
15.63
15.65
Basis..
terrflora/SF
water/PWC
waterfPWC
water/ALC
terrflora/SF
water/ALC
air/UR
water/ALC
D-259
-------
benzole acid 6.53
phthallc anhydride 6.03
dlchlorobenzldine (3,3'-) 5.75
cyclohexanone 6.02
anthracene 4.05
torrflora/RfD
terrflora/RfD
torrflora/SF
watat/RfD
I-" _- . ;.-•, •" *.-;••, v1'"^, ~^ ^"T-' ^CjW^- -? *^"'."WX.l'x'Tp'iiw-t;f *\^«^3 ;!•-?--»/'-Sl:
• Set thresholds for reporting requirements
(registration, inventory, environmental
monitoring, and enhanced emission estimates)
• Set air emission fees using hazard-based fee
rates (rather than flat rate)
• Identify persistent and bioaccumulating
D-260
-------
ai^-s
-"tt-siSrt-JfK*
acetone
ocnzcuc
chlorophcnol
7.16
11.16
: 8.31:
0
0
9,240
O.OOE-HX)
O.OOE+00
3^85,000
3,522
4.75E+13
5.09E+14
• A systemrfor ranking chemicals by their .L. -:
environmental hazard potential "
• Combines human and ecological toxicity data
with environmental fate (fugacity) modeling of
potential exposure
• Hazard Potential = potential exposure/toxicity
•.Ranges over 21 orders of magnitude >v; ,
' '''l^ubstiric^iridej^fo^^
D-261
-------
D-262
-------
The Mercury Study: An Integrated Risk Assessment
Martha H. Keating
-------
as both "state of the science" and completely off base.
The study had a statutory deadline of November 1994 for submittal to Congress..
but has yet to be submitted. Currently the EPA is awaiting the final written report of the
Science Advisory Board which reviewed the study in February 1997. It is hoped a
Mercury Study will be released by the end of 1997.
This paper is intended for state and regional agencies who may wish to design a
multimedia risk assessment. The discussion does not focus on specific models or
results. The intent is to describe the approach which was used to model and
characterize the problem, how limitations in the assessment were described and the
lessons learned from applying these tools to a problem that has far-reaching policy
implications.
Design of the Mercury Study
The Mercury Study was designed to integrate a number of different (but linked)
types of information:
Data on type, sources, and trends in emissions;
* evaluation of the atmospheric transport of mercury to locations distant from
emission sources;
assessment of potential impacts of mercury emissions close to the source;
* identification of major pathways of exposure to humans and wildlife;
identification of the types of human health consequences of mercury exposure
and the amount of exposure likely to result in adverse effects;
evaluation of mercury exposure consequences for ecosystems and for wildlife
species;
* identification of populations especially at risk from mercury exposure due to
innate sensitivity or high exposure; and
estimates of control technology efficiencies and costs.
Because of the paucity of monitoring data for mercury, particularly around source
types of particular interest, the Mercury Study exposure assessment relied on a number
of models to predict mercury fate and transport in the atmosphere, watershed, water
body and aquatic and terrestrial food chains. Quantitative modeling analyses examined
the following ; (1) the long range transport of mercury from emission sources through
the atmosphere; (2) the transport of mercury emissions through the local atmosphere;
D-264
-------
(3) the aquatic and terrestrial fate and transport of mercury at hypothetical sites; and (4)
finally the resulting exposures to hypothetical humans and animals that inhabit these
sites. Exposure to mercury from seafood was estimated using a cross sectional survey
with a three day sampling period and central tendency estimates of mercury
concentrations in the tissues of seafood.
Figure 1 illustrates how the various exposure models were combined to estimate
both human and wildlife exposure.
Describing the Limitations of the Assessment
Each facet of the Mercury Study has it's own set of uncertainties - either due to
uncertainties inherent in the approach (e.g., using emission factors for the emissions
inventory) or uncertainty related to the models themselves or the parameters used in
the model. ;
The approach used to describe the limitations of the analysis was to list and
discuss what the factors were that created uncertainty, the effect of these factors on the
analysis (e.g., over- or underestimation), and the means by which the uncertainties
were addressed. It is extremely important that the researcher understand the data -
their derivation, quality and how a change might influence the overall results.
For example, the emissions inventory was the basically the input for the long
range transport analysis. The quality of the emissions data for each of the source
categories varied. It was important to know what the basis for the emissions estimate
was (e.g., stack test data or engineering judgement), how the estimate compared to
other estimates in the published literature or provided by industry, whether the estimate
was current or how it might change over time and most importantly how changes in the
emissions estimate (either lower or higher) would affect the results of the modeling
analysis.
For the exposure analysis, a table of factors important for estimating mercury
exposure was provided along with how important they were and how they were handled
in the analysis. Some were addressed by sensitivity analyses, different scenarios were
used for others and quantitative uncertainty analyses were performed for those where
the data were sufficient.
Limitations in the assessment should be acknowledged and considered when
drawing conclusions. It is also helpful to list areas of research or further study or
analyses needed to reduce the uncertainty and variability of the data.
D-265
-------
Lessons Learned
1. "Science guides, but policy decides". The risk assessment is only one piece of
the information needed by the risk manager. Other risk management factors such as ""
economic considerations, technical feasibility and other non-risk related factors are
always part of policy considerations.
2. Extensive justification and documentation of model parameters is essential for a
"transparent" analyses. The downside is that there is much information for critics to
challenge.
3. A discussion of uncertainty, even if it's qualitative is required. Describe the
limitations of the assessments. Know your models.
4. Sensitivity analyses are useful to test a range of possibilities when there is no
one correct answer (always!). Using a range is more defensible than point estimates.
5. Use measured data as much as possible to test, and hopefully corroborate,
modeled predictions.
6. Qualify conclusions, if necessary, with statements about the level of confidence
in the results.
D-266
-------
Figure I
Transport and Exposure Modeling Conducted in Jlie Combined COMPDEP and RKLMAP Local Impact Analysis'
to
ON
Local Hg Source
*«IWH-I 4 t*»
SAH Ri:VII-\V DKAI-T
-------
a
to
oo
-------
Multi Pathway Human Health Risk Assessments
Evaluation afth S&e&yiig
IA multi pathway human health risk
assessment (HHRA)
• An ecological risk assessment (ERA) j-
r Verification of die emissions rates used in die ehronic :j
: and acuteHHRA's.;:::;.-"";;! ':-.;'!;:.?!^'--;":-.:-"''
Verification of ^ modeling methodology, model
iVerification.of the! methodology usedindetennining
^. depositions. •'::.
'~1 • The ADEM- Air Division was not responsible for the"";
health effects assessment • £*
• Estimated risks and hazards from direct
exposure through inhalation of emissions.
• Estimated risks and hazards from indirect -v
exposures through soil, walerand .food ^ _.
" products exposed to emissions depositions. 2e
4^- --r^r --yyr
fj • The U.S. Department of Defense (DOD)
-J currently stores ledial unitary chemical
' I agents at the Anniston Army Depot
t... • These include the blister agents HD and HI.
and the organophosphate nerve agents GB
andVX. -....'
Leigh Barb Bacon
D-269
-------
Afttlti Pathway Human Health Risk Assessments
County, ;
: • Hfita*>7:r:-'«w£:ivw^.TO^«s.> OT.^:,= .-J::I--V .-;..'.,:• .
-. r;approximately 4 miles west of Anniston,
u • ;'AL. ' :;-"'v.V:.-"" :•"•':. -.r" V \-'-" •' :"-"- •
"
:.• r -:'-I - .-.;../'-A-...:..:..- r. V.:-:..- -t'-^-
• : ..r ...'.I. ..' .:.~n:--.-- : .: - :•• .•:."..-...:::Vr ::,";.
»iM^i ,>*ii *•»•!! mttatm ^^-iio^-gibir?^
Definition of the study area. .
H •Definition of the exposed individuals
J : . . , •_•••- • •
~J • Determination of me concentratipn of each
£] ^substocem the environmental media
ip/l • Esmnation of the amount of substance each
- individual is exposed to."''..;; ; : ::
• Assessment of the toxicity of media
concentrations. ' :
9
» Chronic ". :'.
*J •OperatJowJdm«ionof6yea«.
- • Acute :"_; " :. .£ '••.
•Opentxaal duration of Ihour.
•*3
i For the chronic portion of the HHRA, it was
limited to all the acreage within 50
kilometers.
I For the acute portion of the HHRA, it was ^r
limited to all the acreage within 10 j*
kilometers. w
U • Adult Resident
4 • Child Resident
•j » Subsistence Farmer
°. • Subsistence Fisher-
£ '"" - toginMartm/Neely Henry Lake
f - Eastaboga Fish Hatchery
j - Abandoned Catfish Farm
Leigh Barb Bacon
D-270
-------
Multi Pathway Human Health Risk Assessments
•V •••• •• •.•• •_
£ ir Adult Resident living on the AAD _._ ; ;l
••t '-• w-r • ' '•:••' •' ••;.-•:*:.• :.-
^•Onsrte Worker
i:. liS-tf:'- €'• v
*
•!. -- ' ••-•
fteAAD ;i:V
•'.' .. ' • : :••..'.• :
..;:• ,•:; ' , -:,-•
~- ' ••
::- "•"-.,:-;.: :':
"•^^\~'-i
• Paniculate Concentration
• Wet Depostion' / •
. -Water ... ''•"••
*
- i
• Emission Rates for the AAD sources were
established from data at the Johnston Atoll
Chemical Agent Disposal System
• Maximum lifetime incinerator/source
specific emissions were determined for the ,%;
chronic assessment . W'
• Maximum short term emissions were used **
for the acute assessment '-i
f ;^» - x .-!/ -V
,* .-,?* •'*-,"=!
^*r__^l
.,_• 1 • To account for increased emissions due to
^ startup/shutdown, upsets or malfunctions,
.*. iipset factors were applied to the emission
-rj estimates for each pollutant ;;;'j;:^-
• Upsets were assumed to occur 5% of the
time for metals and 20% for organics.
• Emissions were scaled by 2.8 and 1.45 for
organic and metal emissions, respectively.
''.^f^ -, -• .»' ^^j
**"•
• (10/60) (10) + (50/60) (1)-
Vi*" -I
• '.t • Example- MetoU
- (055) (1) * (0.05) (10)- 1.45
: • The only exception is the BRA, which can H
right itself, or be shut down within 10 \-
ij- -• minutes. ••: .= >..•:•..•. .,...
-
• f .
^-4
• To account for increased emissions due to
startup/shutdown, upsets or malfunctions,
upset factors were applied to die emission
estimates for each pollutant
• Assumed that only one source would be in
upset at a tune. "'"'" '-"' '
• The source in upset is scaled up by a factor
of 10, while all other processes run under
normal conditions.
Leigh Barb Bacon
D-271
-------
Multi Pathway Human Health Risk Assessments
• For the chromcportipn of tbc HHJRA
rV rconceotrationfcwetanddwdepositioas.cv!a^£ 15
i>-', • . •^•^•rrK;"^^?^&«-^:rv ••^^•rM:ft;5~*e^~~r 1,
gj m F^^ejscuteJppinpjn ofthe HHRA, modejing
E,:^^^jf^^f^^^^T^^X:^o^^r^
™ predict ambient vapor and particulate "7-n-^:\'
^d concentrations. L ; -• .-'• f L
• to Addition tq the chronic ^jd acute
r^ modeling,
•>,:'-~--':•:.1-!:• -T£
was run
' "'"•iViTin»iV*i>'tntji' fj^r'''nllf"iTJniiTrinaiin''oii' tttm fiiVnttinr • -
I HH*» i
^1 • In lieu of on-site data, 5 years of surface, . .
>^ upper air and precipitation data from the .
•^ Birmingham, Alabama municipal airport was d
-'.i used in the modeling. ;:;::. ;;.;I- '•.!:;•:••
• Three receptor grids were generated for the
•Chronic-polar receptor grid . ';.•'•-• ^
•Acute-polar receptor grid, and a fenceline grid "^
•Chronic/Acute-discrete receptors ^
** '-^ ,jF s£f •'" v
».«• •...». •—^•>'«i*
• Based on the exposure scenario, media
concentrations can be verified for a specific
. pollutant for a specific year. . .
i Using the unit emission rate maximum
predicted concentration for each source, the
pollutant specific emission rate is applied to *
the unit concentration for each individual
source.
• All sources are upset and the maximum "
predicted concentration is the greatest of
these scenarios.
• Using the unit emission rate maximum
predicted concentration for each source, the
pollutant specific emission rate is applied to
the unit concentration for each individual
source.
• Each source is upset individually, while tbe '' It
other sources are run under normal *
conditions. Tbe maximum predicted **
concentration would be the greatest of these.
Barb Bacon
D-272
-------
Multi Pathway Human Health Risk Assessments
s&*f
"v3W$
--*-' m.All ambient concentrations assumed tojbcco-^
f^ol-aM^T^T:^-;-
• All incinerators assumed to run 6000 hours
liTheAAD li
' froin 4 years'to> years.
' ' '
Chronic upset condirjohs
and 20% tot
^•i • Conservative stack parameters.
• Non detect substances are assumed to be
present at the detection level for regulated
pollutants, and at 1/2 the detection level for
.unregulated pollutants, .
• The U.S. Army submitted a SRA for the
proposed demilitarization activities at the
Anniston Army Depot. : •
• Through dispersion modeling and the use of
an electronic spreadsheet, the ADEM Air
; Division was able to verify media
concentrations used in the health effects
portion of the SRA.
Leigh Barb Bacon
D-273
-------
D-274
-------
f WHAT IS EOG?
' THE EDUCATION AND OUTREACH GROUP (EOG)
'ACKNOWLEDGES THE IMPORTANCE OF EDUCATION,
OUTREACH, AND TRAINING (APTI) TO THE AIR PROGRAM
MERGED INTO ONE GROUP IN NOVEMBER, 1994
MULTI-DISCIPLINARY GROUP OF ENVIRONMENTAL
ENGINEERS, SCIENTISTS, SPECIALISTS, AND EDUCATORS
D-275
-------
• PROVIDE TECHNICAL AIR POLLUTION TRAINING TO
PRIMARY CLIENTS (State and local air agency personnel)
• PROMOTE ENVIRONMENTAL EDUCATION TO TEACHERS
AND STUDENTS, WITH A FOCUS ON K-12
ENHANCE OUTREACH ACTIVITIES BY FORGING NEW
PARTNERSHIPS AND STRENGTHENING EXISTING ONES
• Small Business Support
• Air Quality Learning Centers
• Public Information Officers
4
NEEDS
• JOINT NEEDS ASSESSMENT SURVEY
STAPPA/ALAPCO/EPA TRAINING COMMITTEE
COURSE EVALUATIONS
ANNUAL TRAINING CONFERENCE
D-276
-------
• SELF INSTRUCTIONAL MATERIALS (obtained from EOG
and air quality learning centers)
CLASSROOM COURSES (taught at area training centers
and academies)
• SATELLITE BROADCASTS (on the Air Pollution Distance
Learning Network)
. ":::^";:':;;W^i^lii::^
WHAT'S AVAILABLJE? (more)
" SELF INSTRUCTIONAL COURSES
SI: 400 Introduction to Risk
Assessment/Risk Management
SI: 404 Urban Air Toxics
SI: 422 Air Pollution Control Orientation
SI: 458 Hazardous Waste Calculations
D-277
-------
• OAQPS HomePage on INTERNET
• APTI BBS on TTN (dial 919-541-5742)
• WESTAR Homepage
• EPA's LAN Services Menu
• APTI 1997 Classroom Schedule
• S/L Site Coordinators and Training Contacts
D-278
-------
Unified Air Toxics
Web Site
www.epa.gov/oar/oaqps/airtox/
Dr. Nancy a Pate
EPA/OAQPS/1TPID
Information Transfer Group
919-541-5347
pate.nancy@epamail.epa.gov
Basic Facts
EPA Rules and Implementation
Pollutants and Sources
Technical Resources
EPA Programs
State and Local Agency Programs
Basic Facts
Why "Unified"?
What are Air Toxics?
What are the Health Effects?
"Criteria" vs "HAP"
Citizen Guides
Links to more health info
EPA Rules and
Implementation
List of all Air Toxics Rules and related
rules
Links to all Preamble and Rule texts
Policy Guidance
Implementation Resources
Pollutants and Sources
Listofll2(b)(l)HAP's
Modifications
Current Source Category List
Technical Resources
Hotlines
Hotlinks
Document Information
Resources
Selected List of Documents
D-279
-------
EPA Programs
Alphabet Soup
Road map to EPA Offices
Functional Statements
State and Local Agengy
Programs
Describes the Partnership
Links to State and Local Web Sites
Share Library
D-280
-------
Unified Air Toxics Website
Purpose of website - To provide the general public, federal, state and local governments and
emitting facilities, comprehensive air toxics information (from basic to very technical) in a
centralized location on the internet. To encourage sharing of information in order to reduce
duplication of effort
Website address: http//www.epa.gov/oar/oaqps/airtox/
Brief description of the 6 areas of information either presently or soon to be on this website:
1. Basic Facts
- describes Unified Air Toxics Website
- briefly explains what are air toxic pollutants and their effects
- provides links to citizen guides and brochures on health risks of air toxics and other
basic sources of information
2. EPA Rules and Implementation
- lists all air toxics rules and closely related rules
- provides links to all preamble and rule texts, related policy guidance and access to
implementation materials produced by federal, state or local agencies
3. Pollutants and Sources
- lists the 188 toxic air pollutants regulated by the EPA
- defines types of air toxic pollutant sources
- lists 174 source categories of industrial and commercial sources of air toxics
4. Technical Resources
- lists telephone information resources (such as hotlines) and internet technical
information resources (such as centers, clearinghouses) with air toxics information
- gives access to select published technical documents and links to document information
resources
5. EPA Programs
- describes air toxics program and its major components
- lists all EPA offices with air toxics responsibilities and describes their roles
6. State and Local Agency Programs
- describes federal, state and local air toxics program partnership
- lists state and local air agency program websites
- provides a "sharing library" for states and local agencies without websites to display
work products in one centralized location
D-281
-------
D-282
-------
National Speciality Workshop
on Technical Tools for Air Toxic Assessment
EPA Environmental Research Center
Research Triangle Park, NC
(June 17- 20,1997)
Education and Outreach
Panel Group Summary
Panel Members:
Brian Haugstad: Chair (Iowa)
Dr. Ellen Morris (Connecticut)
Samuel Bell (Jefferson County, AL)
Leigh Bacon (Alabama)
Jimmy Johnson (Georgia)
EPA Coordinators:
Jim Dicke (EPA-EOG)
Ron Townsend (EPA-EOG)
Howard Wright (EPA-EOG)
D-283
-------
PART ONE:
When are present technical tools adequate in helping you do your job?
1. In providing staff training on specific air toxics standards or procedures, and in providing
information on general air toxics issues.
2. In providing education, outreach and training to the regulated community.
• Printed training and outreach materials, and live or taped educational videos are
helpful, when provided in a timely manner.
3. Telecourses filmed at regulated facilities, that demonstrate regulated compliance.
PART TWO:
When are technical tools inadequate?
1. When the training focuses not on the regulatory requirements, but instead upon the general
operations of the facility, or upon other methods by which the facility reduced air pollution.
2. When training is not provided in a timely basis.
• This happens frequently.
3. When telecourse speakers provide vague information, and are unable to answer specific
questions or details.
4. When communicating risk assessment to the public.
5. When there are no viewing sites equipped to receive satellite courses.
• When local network sites are not available, otherwise interested participants must
either travel to more distant sites, or otherwise not participate.
Pagel
D-284
-------
PART THREE:
What recommendations may solve these identified problems?
1. Training should focus upon the needs of the regulatory agencies implementing
the regulations, while clearly understanding who is the target audience.
• Different target audiences have different training needs, (i.e. permit writers v.
compliance staff v. the public)
Recommendations;
A. EPA should coordinate with STAPPA/ALAPCO and make greater use of annual and
special interest surveys, to more clearly identify target audiences and their training needs.
B. STAPPA/ALAPCO has a listing of state air toxics contacts.
• State and local agency staff should review these contact lists, to ensure correct staff
are listed.
C. Survey the targeted audience well prior to developing the training materials.
D. Different state and local regulatory agencies have different goals and training needs.
Surveys should identify these differences.
E. Survey STAPPA/ALAPCO, EPA Regions, and Trade Associations, to determine what type
of training is needed, and to specifically identify when the training is most needed.
F. Surveys should include long-term strategic goals and these long-term ideas should be
forwarded toEPA. This will force state and local agencies to actively think and prioritize
their training needs long-term.
G. States and local agencies should actively solicit and prioritize the training needs from their
identified targeted groups, and this information should be forwarded to EPA.
H. Training goals should be identified for specific regulatory standards.
I. Develop written surveys for distribution to state and local agencies that identify and
prioritize requisite and required training.
J. Local and state agencies should conduct their own surveys of their agency staff, to
determine their agency's training needs.
K. Identified training needs could be shared through the use of regional air toxics conference
calls.
L. Newly released educational materials should be screened early after receiving it, to insure
it is useful for targeted groups.
Page 2
D-285
-------
2. Many of the video materials available to states and local agencies are
out-of-date.
Recommendations;
A. A quarterly update on all current available training materials should be provided to
primary state and local agency ah- toxics contacts.
B. Do not rely upon the state educational coordinator to pass on training updates to air toxics
staff. Primary air toxics coordinators should be contacted directly, or informed where to
find this information on the Unified Web Site.
C. References to all written and video materials should include a date of publication, and a
brief synopsis.
D. New training materials should be developed by EPA as soon after surveys are received
and significant training needs are identified.
E. Videos should not be repetitive, but should reflect current policies and procedures.
F. Remember... it is difficult for state and local agency staff to provide industry with training,
until state and local program staff first are provided a clear understanding themselves.
PageS
D-286
-------
3. Make EPA training and outreach materials more accessible.
Recommendations;
A. Provide an on-line clearinghouse of all available course materials, including video tapes,
training courses, and other training resource materials, by accessing the Unified Air Toxics
Web Site.
B. Continually (or quarterly) update course materials on-line, as these materials become
available, so that most current training materials are always up-to-date.
C. Separate training materials by discipline or MACT standards, to more easily locate
resource materials.
D. EPA and STAPPA/ALAPCO should contact state and local air toxics staff contacts directly,
as opposed to strictly through state/local educational coordinators.
E. Solicit input from trade associations, and have trade associations co-sponsor training
programs.
F. Provide a means whereby training materials already developed by state and local programs
may be shared electronically on the Unified Web Site.
G. Provide an easily identified on-line heading (ie. "air toxics education"), so that on-line
browsers may more easily locate educational training materials.
H. Greater publicize the Unified Air Toxics Web Sites.
• Promote its use among state and local air toxics contacts, and among state and local
air directors.
1. For regional EPA updates, state and local agencies should greater utilize EPA's regional web
sites.
• Provide regional EPA contacts with suggestions on how regional EPA web sites may
be improved.
.1. Make better use of fiber-optic satellite feeds, to involve government agencies, regulated
facilities and the general public.
K.. To improve training attendance, EPA and state sponsored informational meetings should
require minimal long distance travel.
L. Use more conference calls.
• It is cheap and convenient.
M. Workshops are not commonly well attended by the public.
• Suggestions are needed to improve public participation.
• Use state/federal advertising and marketing resources?
N. Establish additional automatic e-mailing methods, sent to air toxics contacts.
• ie. Similar to Federal Register notices, and Mobile Source e-mailings.
O. Link other good web sites to the Unified Air Toxics Web Site, for easy access.
P. The Unified Air Toxics Web Site should have a comments section, for agency staff to
comment upon education materials, and to make further suggestions.
D-287
-------
4. Publications such as quarterly newsletters, fiber-optic television links
throughout the state (for state specific education and outreach programs), and
informational brochures mailed to regulated facilities to aid in MACT
compliance, are helpful in educating regulated facilities.
Recommendations:
A. Newsletters may be used to educate and provide all types of compliance standard
information to regulated facilities.
B. EPA written materials can be modified and simplified by state/local agencies for state
agency instructional purposes, (ie. Brochures and newsletters)
C. Publications should be clearly written, in shortened form.
D. Don't include needless language and too much detail.
E. If materials are too long or detailed, often regulated facilities don't bother reading them, and
the effort is wasted.
F. Use "catchy and colorful graphics" that bring attention to the publication or brochure, and
prioritize the ideas that you wish to relay to the reader.
G. In addition to regulator news, newsletters should also include health risk data based upon
health studies, that draws the reader further into the article.
H. To reduce duplication of efforts, EPA and state agencies should contact each other when
significant communications are forwarded to regulated facilities.
I. Video training should not include extraneous details.
5. Small Business Assistance Groups are often helpful in providing confidential,
on-site assistance with regulatory compliance and educational materials to
regulated facilities.
Recommendations:
A. Examples of the nation's most highly effective programs should be summarized and shared
nationally.
B. STAPPA/ALAPCO should survey how many small business assistance groups are currently
active nationwide.
• Investigate their effectiveness.
C. Small Business Assistance Programs should be expanded, by increased funding of these
sources, and by further sharing of technical ideas between regulated facilities and government
agencies.
D. An alternative to directly interacting with government regulatory agencies.
• Regulated facilities often call upon Small Business Assistance Groups, rather than
interacting with government agencies.
• Improved compliance of MACT standards.
Page 5
D-288
-------
6. A Plain English Guidance Document is needed to instruct agencies on how to
effectively collect, analysis and present data.
Recommendations;
A. Statistical analysis training is needed.
B. More training opportunities provided on the Education and Outreach home page.
C. Risk Assessment.
• More training needed for Training the Trainers target groups.
D. Training materials should use non-technical instruction, whenever possible, to keep training
as simple as possible and easy to understand.
7. Make training materials and specifically satellite/video training more
interactive.
Recommendations:
A. Question and answer format.
B. Provide additional MACT standards satellite/ video training.
• One MACT standard per video.
C. When possible, film on-site at regulated facilities, to demonstrate how to perform
required MACT compliance activities.
D. Make video presentations more interesting, by using state-of-the-art, color graphics.
Page 6
D-289
-------
D-290
-------
Appendix E
Contacts List
Participants in the National Specialty
Workshop on Technical Tools for Air
Toxics Assessment
-------
CONTACTS LIST FOR THE PARTICIPANTS IN THE
NATIONAL SPECIALTY WORKSHOP ON TECHNICAL TOOLS FOR
AIR TOXICS ASSESSMENT
June 17-20,1997
Name
Leigh Bacon
Mohamed N. Basher
Samuel L. Bell
Paul Cocca
Richard Cook
Jennifer DeMay
Mark Derf
Jerry Ebersole
Steven Ehlers
Henry Feingersh
Daniel J. Gray
Brian Haugstad
Jeffrey J. Hayward
Alan Huber
Agency
ALDEM
Allegheny County
Jefferson County
EPA/OW/OST/SASD
EPA QMS
Washington
IN DEM
ORDEQ
US EPA
Region 2
KYDAQ
IADNR
NCDEHNRAQ
EPANERL
Phone
334-271-7861
412-578-8112
205-930-1366
202-260-8614
313-741-7827
360^07-6825
317-232-8449
503-229-6974
214-665-8312
212-637-3382
502-573-3382
515-281-4927
919-733-1475
919-541-1338
Fax
334-279-3044
412-578-8144
205-939-3019
202-260-9830
313-741-7439
360-407-6802
317-233-5967
503-229-5675
214-665-6762
212-637-3901
502-573-3787
515-242-5094
919-733-1812
919-541-1338
Email
lbb@adem.state.al.us
dns@city-net.com
jcdh@bham.mindspring.com
cocca.paul @epamail.epa.gov
cook.rich@epamail.epa.gov
jdem461 @ecy. wa.gov
aynos @ opn .dem.state.in.us
gerald.ebersole@state.or.us
ehlers.steven@epamail.epa.gov
feingersh .henry @epamail .epa.gov
gray@mail.nr.state.ky .us
bhaugst@max.state.ia.us
jeff_hayward@aq.ehnr.state.nc.us
huber.alan@epamail.epa.gov
Area of Expertise*
Air Dispersion Modeling
Title V Permits
Dispersion Modeling, Area Toxics, Title V
Permits
Toxics Deposition to Watersheds
Mobile Source Air Toxics
Point Source Inventories
Air Dispersion Modeling
Air Toxics/Education and
Outreach/Exposure Assessments
Air Dispersion Modeling and Risk
Assessment
Air Dispersion Modeling
Permitting/Title V
Air Toxics Policy Development Rules
Writing, Education and Outreach
Toxicology/Risk Assessment/Risk Policy
Human Exposure Air Toxic Models
E-l
-------
Name
Jimmy Johnston
Peter R. Kahn
Aimee C. Kennedy
Steve Kish
Linda Lay
James A. Lefik
Denis M. Lohman
Ken McBee
Thomas McCurdy
Evelina C. Morales
Ellen P. Morris, PhD.
Lee Page
Sirisak P. Pakunpanya
Nancy Pate
Michael Pokorny
Greg Pratt
Bob Ragland
Agency
GAEPD
Region 1
WVDEP
MIDEQ
USEPA/OECA
Allegheny County
Region 3
VADEQ
US EPA
OKDEQ
CTDEP
Region 4
LADEQ
ITPID
MDDEP
MNPCA
Forsyth County
Phone
404-363-7127
617-860-4392
304-558-1213
517-335-4794
202-564-8577
412-578-8132
215-566-2192
804-698-4024
919-541-5774
405-962-2206
860^24-3412
404-562-9131
504-764-0202
919-541-5347
410-631-3232
612-296-7664
910-727-8060
Fax
404-363-7100
617-860^397
304-558-1222
517-335-3122
202-564-0068
412-578-8144
215-566-2124
804-698-4510
405-962-2200
860-424^063
404-562-9095
504-765-0222
91 9.54 1.4028
410-631-3202
612-297-8701
910-727-2777
Email
jimmy_johnston@mail.dnr.state.ga.us
kahn.peter@epamail.epa.gov
kish @deq.state.mi.us
lay.linda@epamail.epa.gov
achd_aq_eq@compuserve.com
lohman.denny@epamail.epa.gov
klmcbee@deq.state.va.us
evelina.morales @ oklaosf .state.ok.us
ellen.morris@po.state.ct.us
page.lee@epamail.epa.gov
patrickp @deq.state.la.us
pate.nancy@epamail.epa.gov
gregory.pratt@pca.state.mn.us
raglanre@co.forsyth.nc.us
Area of Expertise*
Air Toxics Regulatory Implementation
Air Toxics Monitoring
Air Emissions from Hazardous Waste
Combustion
Modeling
Air Toxic Enforcement/Policy
Air Toxics/Air Monitoring
Air Dispersion Modeling
Air Dispersion Modeling
Exposure/Intake Modeling
Toxicology/Epidemiology
Toxicology/Risk Management
Air Toxics Program
Modeling/Engineering
Air Toxics Health Effects and Regulation
Air Toxics Regulation/Implementation/
Modeling/Inventories
Dispersion Modeling, Exposure
Assessment, Environmental Fate of Air
Pollutants
Point Source Emission Inventories and
Point Source Modeling
E-2
-------
Name
Tom Rogers
Andrew J. Roth
Dom Ruggeri
Tom Shanley
Thomas Shoens
Jeffrey Sprague
William K. Steinmetz
Andrew Stewart
Stanley C. Tracey
John White
Daniel Wise
Agency
FLDEP
RAPCA
TNRCC
MDEQ/AQD
Wayne County
BLEPA
NCDEHNR AQ
WIDNR
DC Air Resources
NC DEHNR AQ
RIDEM
Phone
904-921-9554
937-225-4118
512-239-1508
517-335-7056
313-833-3596
217-524-4692
919-715-7713
608-266-5499
202-645-6093
919-715-6285
401-277-2808
Fax
904-922-6979
937-225-3486
512-239-1123
517-335-3122
313-833-3561
217-524-4710
919-733-1812
608-267-0560
202-645-6102
919-733-3340
401-277-2017
Email
rogers_t@dep.state.fl.us
rothaj@rapca.org
druggeri@tnrcc.state.tx.us
shanleyt@deq.state.mi.us
tshoens @co.wayne.mi.us
epa2 1 02 @epa.state.il.us
stewaa@dnr.state.wi.us
tracey@mail.environ.state.dc.us
john_white@aq.ehnr.state.nc.us
daniel.wise@navc.ci.net
Area of Expertise*
Air Dispersion Modeling
MACT/Air Toxics/Modeling
Air Pollution Meteorology, Dispersion
Modeling
Deposition Modeling
Air Toxics, Source Sampling, Air
Monitoring
Dispersion Modeling, Receptor Modeling
Ambient Monitoring
Air Toxics Policy Development, General
Point of Contact Engineering Evaluation
Title V/l 12(r) Regulations
Dispersion Modeling
Modeling
E-3
-------
Appendix F
Workshop Evaluation Summary
-------
Workshop Evaluation Form
National Specialty Workshop on Technical Tools for Air Toxic Assessment, June 17-20,1997
Please take a few minutes to complete this evaluation of the workshop. We will use your feedback
to improve workshops in the future. Thank you!
1. Overall, I found the workshop to be:
circle one
1
2
3
4
5
very beneficial somewhat helpful a waste of time
2. The workshop sessions that were most helpful are:
These sessions were better than others because:
3. The workshop sessions that I liked the least were:
These sessions were not very helpful because:
4. I found the duration of the workshop to be:
circle one
1
2
3
4
5
too long about right too short
Comments:
5. If I were in charge of this workshop, I would have done the following differently:
6. The role of the panel discussions in the workshop was:
circle one
1
2
3
4
5
very beneficial somewhat useful not very useful
Comments:
7. Job Title/Specialty: Years of experience:
8. Please add any additional comments or suggestions you wish to share:
F-l
-------
Workshop Evaluation Form
National Specialty Workshop on Technical Tools for Air Toxic Assessment, June 17-20,1997
1. Overall, I found the workshop to be:
circle one
1
2
3
4
5
very beneficial somewhat helpful a waste of time
t
Average Response: 2
2. The workshop sessions that were most helpful are:
These sessions were better than others because:
• All workshop sessions were helpful. The most helpful speakers were those that
addressed specific tools or specific projects that were done.
• The most helpful information included how different stated do their own studies and
policies and what they have done to apply/respond to air toxics issues and air toxics
regulations.
• Those speakers presenting new and unreleased information, i.e., EFIG, TOOLS,
AERMOD, were helpful because they provided fresh information to keep ours heads up
about in the future. In particular, dispersion modeling and exposure assessment sessions
were educational and gave an idea of what is coming down the pipe and who are experts
in the field.
• The emissions inventory, exposure assessments, and dispersion modeling sessions were
the most helpful sessions. The speakers were better able to communicate their
information. Also, the problems/deficiencies were brought out in the open and
discussed somewhat.
• The exposure assessment session was helpful because it provided for shorter, but concise
discussions. The discussions were more coherent than others. The technical
presentations were clear and informative. Mike Zelenka's presentation was helpful
because he gave the basic tenets of the field so those not in the field could gain a basic
understanding.
• The dispersion modeling session was helpful because the discussion was more general. It
was easy to relate their strong and weak points.
• Presentations made by John Irwin, Greg Pratt, Anne Pope, Martha Keating, and Mike
Zelenka were exceptionally helpful.
3. The workshop sessions that I liked the least were:
These sessions were not very helpful because:
F-2
-------
• The least helpful sessions were those that were general overview topics. The Elements of
OAQPS air toxic program was not helpful because it was shallow. It should either have
been shortened or removed, as it was not very helpful.
• The ambient air modeling session was not as helpful because the material presented (i.e.,
listing methods) was not clearly explained and should have been presented as a handout
instead. Some presentations were too technical. There did not seem to be any issues that
need immediate attention.
• Some sessions were not as helpful because they got very technical and were bogged down
with very specific details that did not have wide encompassing themes that applied to
States. The very technical sessions and panel sessions used terms that are not familiar
and were often less organized and informational.
• The air toxics emission inventories sessions seemed to start in the middle.
4. I found the duration of the workshop to be:
circle one
1
too long
234
about right
t
5
too short
Average response: 2.7
Comments:
• Perhaps a bit too long.
• Needed to schedule in breaks!
• We have brought out a lot of issues. Is there a plan, other than submitting comments to
upper management, to seriously address these issues?
• The length of the workshop totally (3'/z days) was good.
• I think the next workshop could be orly two days. An additional two days could be used
for information transfer.
5. If I were in charge of this workshop, I would have done the following differently:
• Had more breaks, sessions with out breaks were too long.
• Not a thing - very well organized.
• Better organization of panel discussions, provide more time for open discussion.
• The panel conference calls needed to be more focused with the product in mind and the
intent/needs clearly defined and implemented.
• Fewer but more detailed talks.
• Shorten the length of time for speakers to 20-30 minutes.
• After workshop social functions - not necessarily formal, but dinner or a ball game would
help to increase interaction.
F-3
-------
• I would have preferred more time allotted for "give and take" between the panels and the
audience. It might have been helpful if each panel had more time for internal discussion.
• Direct focus of presentations on State roles or new air toxics policies.
• Would have each StateAocal give status of current toxics program. .
• Have the meeting at a hotel to avoid transportation issues. Plan a ice breaker meeting the
evening before the formal meeting begins or the first evening of the meeting.
6. The role of the panel discussions in the workshop was:
circle one
1
2
3
4
5
very beneficial somewhat useful not very useful
T
Average response: 2.7
Comments:
• Panels could have more focused on brainstorming or long term goal
setting/improvements/action (follow up).
• Panel discussion sessions were not very helpful because we could read the summaries.
• Panels were a lot of work for State and local participants. Please improve
communication about goals.
• The panel discussions are essential to a dialogue between feds and States. State opinions
should be voiced.
• The panels need more mix so that all the experts are not on the same panel. This defeats
the purpose of having comments from the audience.
• They were helpful to open lines of communication between States. Gets you more
involved and interested.
• The panels turned out to be just additional individual presentation. A panel that actually
debates an issue among themselves may have been better.
• Seemed to bring up many issues that related to all State and local situations. Would like
to see a summary in writing of panel discussions.
7. Job Title/Specialty, Years of experience:
• Permits Manager, 19
Chemist/AirTox, 25
• Engineer, 11
• Modeler/Engineer, 15
EPS/Permits SMACT, 5
El, 20+
• Research Scientist/Dispersion Modeling, 15
• Emissions Inventory, 6
F-4
-------
• Environmental Engineer, 5
• Staff Epidemiologist, 3
• Modeling, 2
• Engineer, 3
• Modeling, 5
• Environmental Engineer, 1
• Environmental Scientist/Modeler, 5
• Hazardous Air Pollution Analyst, 6
• Hazardous Waste Combustion Engineer for Air, 3
A.Q. Modeling, 17
• Engineer/Ambient Air Monitoring, 14
• A.Q. Modeling, 5
Envt. Prot. Spec Hi/Modeling, 9
• Environmental Engineer, 9
• Toxics Unit Manager/Regulatory Implementation, 11
8. Please add any additional comments or suggestions you wish to share:
• It would have been helpful to have time set aside before the meeting began for the panel
members to meet and finalize their comments. It was too much of a push to have a
meeting in the day. Policy issues would be good for the next meeting. Some talks were
too technical (jargon). The greatest benefit of the workshop was the opportunity to
network with others with different expertise/responsibilities. Perhaps consideration
should be given to smaller breakout sessions.
• Appreciated opportunity to attend and see other's viewpoints.
• This is the most productive workshop I've attended so far.
• I hope the issues raised and communication and dialog between the various disciplines
can be kept alive until resolution and not allowed to fizzle after this conference and the
communication lines remain open.
• Overall, an excellent forum for discussing and evaluating the available tools for air toxics
assessment. I learned a great amount about this area and welcomed the opportunity to
relay my concerns and experiences to other people working in this field.
• Overall, I found the workshop to be worthwhile. With a few minor changes, it should
become an ongoing workshop/training session.
The workshop was well organized and appeared to be well attended by the State and local
agencies.
A fairly successful workshop. I came to gain information and feel like I've gained quite a
bit.
F-5
-------
Appendix G
Contacts List
EPA Air Regulations and Enforcement and
Compliance Assurance
-------
CONTACTS LIST FOR THE EPA AIR REGULATIONS
AND ENFORCEMENT AND COMPLIANCE ASSURANCE
(current as of June 1997)
Phone #
919-541-2452
919-541-2837
919-541-5264
919-541-5264
919-541-5264
919-541-2421
919-541-5265
919-541-5426
919-541-5251
919-541-5264
919-541-5264
OAQPS Contact
Jim Szykman
Steve Fruh
Walt Stevenson
Walt Stevenson
Walt Stevenson
Martha Smith
Rick Copland
Jim Eddinger
Fred Porter
Walt Stevenson
Walt Stevenson
Regulation
NSPS Subpart A: General
Provisions
NSPS Subpart CC: Glass
Manufacturing
NSPS Subpart Ca: Emission
Guidelines for Large MWC
Units
NSPS Subpart Cb: Sulfuric
Acid Plants
NSPS Subpart Cbb: MWC
Rules
NSPS Subpart Cc: Emission
Guidelines & Compliance
Times for MSW Landfills
NSPS Subpart D, Da, Db,
DC: Boilers
NSPS Subpart E:
Incinerators
NSPS Subpart Ea: Municipal
Waste Combustors
NSPS Subpart Ebb:
Municipal Waste
Combustors
OC Contact
Sally Mitoff
Belinda Breidenbach
Chris Oh
Joyce Chandler
Joyce Chandler
Phone #
(202) 564-7012
(202) 564-7022
(202) 564-7004
(202) 564-7073
(202) 564-7073
G-l
-------
Phone #
919-541-5446
919-541-5435
919-541-
919-541-5025
919-541-5416
919-541-2364
919-541-0881
919-541-5289
919-541-0881
919-541-0881
919-541-5602
919-541-2364
OAQPS Contact
Joe Wood
Bill Neuffer
Mary Johnson
Mark Morris
Kevin Cavender
Eugene Grumpier
Phil Mulrine
Eugene Grumpier
Eugene Grumpier
Al Vervaert
Kevin Cavender
Regulation
NSPS Subpart F: Portland
Cement
NSPS Subpart G: Nitric Acid
Plants
NSPS Subpart H: Sulfuric
Acid Plants
NSPS Subpart I: Hot Mix
Asphalt Facilities
NSPS Subpart J: Petroleum
Refineries
NSPS Subpart K, Ka, Kb:
Storage Tanks
NSPS Subpart L: Secondary
Lead Smelters
NSPS Subpart M: Brass &
Bronze
NSPS Subpart N, Na: BOF
NSPS Subpart O: Sewage
Treatment Plants
NSPS Subpart P: Primary
Copper Smelters
NSPS Subpart Q: Primary
Zinc Smelters
NSPS Subpart R: Primary
Lead Smelters
OC Contact
Scott Throwe
Jeff Kenknight
Dawn Banks-Waller
Scott Throwe
Tom Ripp
Dan Chadwick
Everett Bishop
Jane Engert
Jane Engert
Maria Malave
John Dombrowski
Jane Engert
Jane Engert
Jane Engert
Phone #
(202) 564-7013
(202) 564-7033
(202) 564-7034
(202) 564-7013
(202) 564-7003
(202) 564-7054
(202) 564-7032
(202) 564-5021
(202) 564-5021
(202) 564-7027
(202)564-7036
(202) 564-5021
(202) 564-5021
(202) 564-5021
-------
Phone #
919-541-2837
919-541-5515
919-541-1084
919-541-1512
919-541-5289
919-541-5289
919-541-5427
919-541-2837
919-541-2379
919-541-5263
919-541-5446
OAQPS Contact
Steve Fruh
David Painter
Juan Santiago
Conrad Chin
Phil Mulrine
Phil Mulrine
JeffTelander
Steve Fruh
Mohamed Serageldin
Sims Roy
Joe Wood
Regulation
NSPS Subpart S: Primary
Aluminum Reduction
NSPS Subpart T,U,V,W,
X: Phosphate Fertilizer
NSPS Subpart Y: Coal
Preparation
NSPS Subpart Z: Ferroalloy
Production
NSPS Subpart AA: Steel
Plants, EAF
NSPS Subpart AAa: Steel
Plants, EAF & AOD
NSPS Subpart BB: Kraft
Pulp Mills
NSPS Subpart CC: Glass
Manufacturing
NSPS Subpart DD: Grain
Elevators
NSPS Subpart EE: Surface
Coating, metal furniture
NSPS Subpart GG:
Stationary Gas Turbines
NSPS Subpart HH: Lime
Manufacturing
OC Contact
Jane Engert
Cletis Mixon
Steve Howie
Chris Oh
Jane Engert
Maria Malave
Maria Malave
Maria Eisemann
Scott Throwe
Ken Harmon
Scott Throwe
Chris Oh
Scott Throwe
Phone #
(202) 564-5021
(202) 564-4153
(202) 564-4146
(202) 564-7004
(202) 564-5021
(202) 564-7027
(202) 564-7027
(202) 564-7016
(202) 564-7013
(202)564-7049
(202) 564-7013
(202) 564-7004
(202) 564-7013
G-3
-------
Phone #
919-541-2364
919-541-5435
919-541-0859
919-541-5515
919-541-5515
919-541-0859
919-541-5305
919-541-2379
919-541-5261
919-541-1084
919-541-0837
919-541-5261
OAQPS Contact
Kevin Cavender
Bill Neuffer
Dave Salman
David Painter
David Painter
Dave Salman
Dan Brown
Mohamed Serageldin
Gail Lacy
Juan Santiago
Dave Markwordt
Gail Lacy
Regulation
NSPS Subpart KK: Lead
Acid Batteries
NSPS Subpart LL: Metallic
Mineral Processing
NSPS Subpart MM: Surface
Coating, Auto
NSPS Subpart NN:
Phosphate Rock
NSPS Subpart PP:
Ammonium Sulfate
Manufacturing
NSPS Subpart QQ: Graphic
Arts, Rotograve Printing
NSPS Subpart RR: Pressure
Sensitive Tape & Label
Coating
NSPS Subpart SS: Surface
Coating, Large Appliances
NSPS Subpart TT: Surface
Coating, Metal Coil
NSPS Subpart UU: Asphalt
Roofing
NSPS Subpart W: VOC
leaks, SOCMI
NSPS Subpart WW: Surface
Coating, Beverage Cans
OC Contact
Jane Engert
Keith Brown
Suzanne Childress
Cletis Mixon
Steve Howie
Scott Throwe
Ginger Gotliffe
Seth Heminway
Scott Throwe
Scott Throwe
Andrew Cherry
Jeff Kenknight
Scott Throwe
Phone #
(202) 564-5021
(202) 564-7124
(202) 564-7018
(202) 564-4153
(202) 564-4146
(202) 564-7013
(202) 564-7072
(202) 564-7017
(202) 564-7013
(202) 564-7013
(202)564-5011
(202) 564-7033
(202) 564-7013
G-4
-------
Phone #
919-541-5397
919-541-5427
919-541-5439
919-541-5608
919-541-5305
919-541-0837
919-541-5608
919-541-5124
919-541-5397
919-541-0837
919-541-0837
919-541-5124
OAQPS Contact
Steve Shedd
JeffTelander
Tony Wayne
Bob Rosensteel
Dan Brown
David Markwordt
Bob Rosensteel
Warren Johnson
Steve Shedd
David Markwordt
David Markwordt
Warren Johnson
Regulation
NSPS Subpart XX: Bulk
Gasoline Terminals
NSPS Subpart AAA:
Woodstoves
NSPS Subpart BBB: Rubber
Tire Manufacture
NSPS Subpart ODD:
Polymer Manufacture
NSPS Subpart FFF: Flexible
Vinyl and Urethane
NSPS Subpart GGG:
Equipment Leaks, Petroleum
Refineries
NSPS Subpart HHH:
Synthetic Fiber
NSPS Subpart ffi: SOCMI,
Air Oxidation
NSPS Subpart JJJ: Dry
Cleaning
NSPS Subpart KKK:
Equipment Leaks, Onshore
Natural Gas
NSPS Subpart LLL: Onshore
Natural Gas
OC Contact
Julie Tankersley
Peter Bahor
Robert Marshall
Maria Malave
Sally Sasnett
Ginger Gotliffe
Tom Ripp
Belinda Breidenbach
Jeff Kenknight
Joyce Chandler
Dan Chadwick
Dan Chadwick
NSPS Subpart NNN: SOCMll Jeff Kenknight
Distillation Operations |
Phone #
(202) 564-7002
(202) 564-7029
(202) 564-7021
(202) 564-7027
(202) 564-7074
(202) 564-7072
(202) 564-7003
(202) 564-7022
(202) 564-7033
(202) 564-7073
(202) 564-7054
(202) 564-7054
(202) 564-7033
G-5
-------
Phone #
919-541-5435
919-541-5435
919-541-5402
919-541-5499
919-541-5124
919-541-5261
919-541-5408
919-541-5435
919-541-5305
919-541-2421
919-541-4516
OAQPS Contact
Bill Neuffer
Bill Neuffer
Randy McDonald
Elaine Manning
Warren Johnson
Gail Lacy
Ellen Ducey
Bill Neuffer
Dan Brown
Martha Smith
Mary Tom Kissell
Regulation
NSPS Subpart OOO: Non
metallic Mineral Processing
NSPS Subpart PPP: Wool
Fiberglass Insulation
NSPS Subpart QQQ: VOC
from Petroleum Wastewater
NSPS Subpart RRR: SOCMI
Reactor Processes
NSPS Subpart SSS: Surface
Coating, Magnetic Tape
NSPS Subpart TTT: Surface
Coating, Plastic Parts for
Business Machines
NSPS Subpart UUU:
Calciners and Dryers
NSPS Subpart VVV:
Polymeric Coating of
Supporting Substrates
NSPS Subpart WWW: MSW
Landfills
NSPS Subpart YYY: VOC
Emissions from the Synthetic
Organic Chemical
Manufacturing Industry
Wastewater
OC Contact
Keith Brown
Scott Throwe
Tom Ripp
Dan Chadwick
Jeff Kenknight
Steve Hoover
Maria Malave
Keith Brown
Maria Malave
Phone #
(202)564-7124
(202) 564-7013
(202) 564-7003
(202) 564-7054
(202) 564-7033
(202) 564-7007
(202) 564-7027
(202) 564-7124
(202) 564-7027
G-6
-------
Phone #
919-541-2452
202-233-9370
919-541-5602
919-541-5602
919-541-08817
919-541-5308
919-541-5124
202-233-9370
202-233-9370
919-541-5254,
919-541-5262,
919-541-0837
202-233-9370
919-541-2910
OAQPS Contact
Jim Szykman
ORIA (Office of
Radiation & Indoor
Air)
Al Vervaert
Al Vervaert
Eugene Grumpier/
111 i am Rosario
Warren Johnson
ORP
ORP
*Jan Meyer, **Rick
Colyer, Dave
Markwordt
ORP
Lula Melton
Regulation
NESHAP Subpart A:
General Provisions
NESHAP Subpart B: Radon
from underground Uranium
Mines
NESHAP Subpart C:
Beryllium
NESHAP Subpart D:
Beryllium Rocket Motor
Firing
NESHAP Subpart E:
Mercury
NESHAP Subpart F: Vinyl
Chloride
NESHAP Subpart H:
Radionuclides from DOE
NESHAP Subpart I:
Radionuclide Emissions
NESHAP Subpart J: Benzene
Leaks
NESHAP Subpart K:
Radionuclide from Elemental
Phosphorous
NESHAP Subpart L: Coke
OC Contact
Belinda Breidenbach
Dan Chadwick
Jane Engert
Virginia Lathrop
Jane Engert
Jeff Kenknight
Joanne Callahan
Virginia Lathrop
Joanne Callahan
Virginia Lathrop
Rafael Sanchez
Joanne Callahan
Virginia Lathrop
Maria Malave
Phone #
(202) 564-7022
(202) 564-7054
(202) 564-5021
(202) 564-7057
(202) 564-5021
(202) 564-7033
(202) 564-5009
(202) 564-7057
(202) 564-5009
(202) 564-7057
(202) 564-7028
(202) 564-5009
(202) 564-7057
(202) 564-7027
G-7
-------
Phone #
919-541-5167
919-541-2837
919-541-0881
919-541-5602
202-233-9370
202-233-9370
919-541-5254,
919-541-5262,
919-541-0837
202-233-9370
919-541-5416
919-541-0837
919-541-0884
OAQPS Contact
S. Fairchild-Zapata
Steve Fruh
Eugene Grumpier
Al Vervaert
ORP
ORP
*Jan Meyer, **Rick
Colyer, Dave
Markwordt
ORP
Mark Morris
David Markwordt
Bob Lucas
Regulation
NESHAP Subpart M:
Asbestos •
NESHAP Subpart N:
Inorganic Arsenic From
Glass Manufacture
NESHAP Subpart O: Arsenic
from Primary Copper
Smelters
NESHAP Subpart P: Arsenic
NESHAP Subpart R: Radon
from Phosphogypsum Stacks
NESHAP Subpart T: Radon
from disposal of Uranium
Mine Tailings
NESHAP Subpart V:
Equipment Leaks,
NESHAP Subpart W: Radon
from Operating Mill Tailings
NESHAP Subpart Y:
Benzene Storage Vessels
NESHAP Subpart BB:
Benzene from Transfer
Operations
NESHAP Subpart FF:
Benzene Waste Operations
OC Contact
Tom Ripp
Scott Throwe
Jane Engert
Jane Engert
Joanne Callahan
Virginia Lathrop
Joanne Callahan
Virginia Lathrop
Rafael Sanchez
Joanne Callahan
Virginia Lathrop
Rafael Sanchez
Rafael Sanchez
Rafael Sanchez
Phone #
(202) 564-7003
(202) 564-7013
(202) 564-5021
(202) 564-5021
(202) 564-5009
(202) 564-7057
(202) 564-5009
(202) 564-7057
(202) 564-7028
(202) 564-5009
(202) 564-7057
(202) 564-7028
(202) 564-7028
(202) 564-7028
G-8
-------
Phone #
919-541-0164
919-541-2452
919-541-5254
919-541-5268
919-541-1549
919-541-5420/
919-541-5289
919-541-0837
919-541-5289
919-541-5397
919-541-0283
919-541-5608
OAQPS Contact
Barbara Driscoll
Jim Szykman
Jan Meyer
Amanda Agnew
George Smith
*Lalit Banker / **Phil
Mulrine
David Markwordt
Phil Mulrine
Steve Shedd
Paul Almodovar
Bob Rosensteel
Regulation
NESHAP Subpart GG:
Aerospace Man and Rework
Facilities
MACT Subpart A: General
Provisions
MACT Subpart F-H, The
HON
MACT Subpart L: Coke
Oven Batteries
MACT Subpart M: Perc Dry
Cleaners
MACT Subpart N:
Chromium Electroplating
MACT Subpart O: Ethylene
Oxide Sterilizers
MACT Subpart Q: Industrial
Process Cooling Towers
MACT Subpart R: Gasoline
Distribution
MACT Subpart T:
Halogenated Solvent
Cleaning
MACT Subpart U: Polymers
& Resins Group I
OC Contact
Belinda Breidenbach
Jeff Kenknight
Maria Malave
Joyce Chandler
Scott Throwe
Karin Leff
Mimi Guernica
Julie Tankersley
Tracy Back
Sally Sasnett
Phone #
(202) 564-7022
(202) 564-7033
(202) 564-7027
(202) 564-7073
(202) 564-7013
(202) 564-7068
(202) 564-2415
(202) 564-7002
(202) 564-7076
(202) 564-7074
G-9
-------
Phone #
919-541-5402
919-541-2364
919-541-0837
919-541-5672
919-541-2363
919-541-5261
919-541-2452
919-541-2379
919-541-0283
919-541-0859
919-541-5608
OAQPS Contact
Randy McDonald
Kevin Cavender
David Markwordt
Jim Durham
Michele Aston
Gail Lacy
Jim Szykman
Mohamed Serageldin
Paul Almodovar
Dave Salman
Bob Rosensteel
Regulation
MACT Subpart W: Epoxy
Resins & Non-nylon
Polymides
MACT Subpart X:
Secondary Lead Smelters
MACT Subpart Y: Marine
Vessel Loading
MACT Subpart CC:
Petroleum Refineries
MACT Subpart DD: Off-Site
Waste & Recovery
Operations
MACT Subpart EE:
Magnetic Tape Manufacture
MACT Subpart GG:
Aerospace
MACT Subpart H: Ship
Building and Repair
MACT Subpart JJ: Wood
Furniture Manufacturing
MACT Subpart KK: Printing
and Publishing
MACT Subpart JJJ:
Polymers & Resins Group IV
OC Contact
Sally Sasnett
Jane Engert
Virginia Lathrop
Tom Ripp
Ann Stephanos
Seth Heminway
Suzanne Childress
Suzanne Childress
Robert Marshall
Ginger Gotliffe
Sally Sasnett
Phone #
(202) 564-7074
(202) 564-5021
(202) 564-7057
(202) 564-7003
(202) 564-7043
(202) 564-7017
(202) 564-7018
(202) 564-7018
(202) 564-7021
(202) 564-7072
(202) 564-7074
"indicates the Primary Project Lead
""Indicates the Secondary Project Lead
n. m
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
l.REPORTNO. 2.
EPA-454/R-97-010
4. TITLE AND SUBTITLE
National Specialty Workshop on Technical Tools for Air Toxics Assessment
7. AUTHORS)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Pacific Environmental Services, Inc.
5001 S. Miami Blvd.
Research Triangle Park, NC 27709
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Emissions, Monitoring & Analysis Division
Research Triangle Park, NC 2771 1
3. RECIPIENTS ACCESSION NO.
5. REPORT DATE
September 1997
«. PERFORMING ORGANIZATION CODE
S. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
1 1 . CONTRACT/GRANr NO.
EPA Contract No. 68D30032
13. TYPE OP REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
EPA Work Assignment Manager: Joe Touma
16 ABSTRACT
This report is the final report documenting proceedings and recommendations from the National Specialty Workshop on Technical
Tools for Air Toxics Assessment. The workshop was held from June 17 through June 20, 1997 in EPA's Administration
Auditorium, 79 Alexander Drive, Research Triangle Park, NC. The purpose of the workshop was to provide a forum for discussion
on the technical tools developed by OAQPS, Regional Offices, State and local agencies for meeting the goal of the air toxics
program. This interactive workshop included half-day sessions devoted to Emission Inventories, Dispersion Modeling, Ambient
Monitoring and Data Analysis, Exposure Assessment, Case Histories, and Education and Outreach as related to supporting the air
toxics program.
1 7 KEY WORDS AND DOCUMEOT ANALYSIS
i. DESCRIPTORS
Air Toxics Program Emission Inventories
Dispersion Modeling Exposure Assessment
Monitoring and Data Analysis
Education and Outreach
18. DISTRIBUTION STATEMENT
Release Unlimited
k IDENTIFIERS/OPEN ENDED TERMS
19 SECURITY CLASS (Kiport>
Unclassified
20. SECURITY CLASS (Pa,,)
Unclassified
c. COSAT1 HcU/Onxw
21 NO OFPAGES
391
22. PRICE
EPA Form 12HM (Sti. 4-TT) PREVIOUS EDITION IS OBSOLETE
------- |