c/EPA
United States
Environmental Protection
Agency
Office of Emergency and
Remedial Response
Washington DC 20460
EPA/540/1-89/003
November 1989
Superfund
OSWER Comparative Risk
Project
Executive Summary and
Overview
-------
EPA/540/1-89/003
November 1989
OSWER Comparative Risk Project
Executive Summary and Overview
Superfund
Sites
Industrial
Facilities
Oil and gas
Operations
Municipal
Waste
Combustion
Underground
Injection of
Hazardous
Waste
Office of Solid Waste and Emergency Response
U.S. Environmental Protection Agency
Washington, DC 20460
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Floor
Chicago, IL 60604-3590
-------
TABLE OF CONTENTS
EXECUTIVE SUMMARY v
SCOPE AND ORGANIZATION OF THE STUDY v
FINDINGS ' vi
PROGRAM RECOMMENDATIONS vii
CHAPTER 1. INTRODUCTION 1
1.1 ORGANIZATION 1
1.2 RELATIONSHIP OF STUDY TO AGENCY COMPARATIVE RISK EFFORTS . . 2
1.3 RELATIONSHIP OF STUDY TO STRATEGIC PLANNING 3
CHAPTER 2. STUDY APPROACH 4
2.1 OVERALL APPROACH 4
2.2 OSWER PROBLEM AREAS 5
2.2.1 Problem Area Definitions 5
2.2.2 Problem Areas Addressed 6
CHAPTER 3. STUDY RESULTS 7
3.1 GENERAL RESULTS 7
3.1.1 Acute Events 7
3.1.2 Other Health Effects 7
3.1.3 Ecological Effects 8
3.1.4 Welfare Effects 9
3.1.5 Ground Water Valuation 9
3.2 KEY FINDINGS 9
3.2.1 Final Rankings 9
3.2.2 Data Quality and Confidence Levels 13
3.2.3 Overall Confidence in Rankings 14
3.3 METHODOLOGICAL AND DATA FACTORS AFFECTING WORKGROUP
RANKINGS 14
3.3.1 Acute Events 14
3.3.2 Other Health Effects 15
3.3.3 Ecological Effects 15
3.3.4 Welfare Effects 16
3.3.5 Ground Water Valuation 16
3.4 RANKINGS AND PUBLIC PERCEPTION OF RISKS 16
CHAPTER 4. INFORMATION NEEDS 24
4.1 GENERAL INFORMATION NEEDS 24
4.1.1 Program Recommendations 24
4.1.2 Information Needs 25
4.1.3 Ongoing Information Collection Efforts 25
4.2 ACUTE EVENTS 27
4.3 OTHER HEALTH EFFECTS 27
4.4 ECOLOGICAL EFFECTS 27
4.5 WELFARE EFFECTS 27
4.6 GROUND WATER VALUATION 27
4.7 PUBLIC PERCEPTION OF RISK 28
111
-------
APPENDIX A GLOSSARY
APPENDIX B DEFINITION OF OSWER PROBLEM AREAS
APPENDIX C INDIVIDUAL WORKGROUP SUMMARIES
IV
-------
EXECUTIVE SUMMARY
The Office of Solid Waste and Emergency Response (OSWER) Comparative Risk
Project was initiated to serve strategic planning needs by exploring the
comparative risks posed by various waste management practices regulated by
and/or under OSWER purview. The study was also undertaken to gain experience
in performing comparative analysis, identifying information needs, and assisting
in establishing research priorities.
In evaluating study goals, a number of objectives became apparent,
including:
• Identifying Data Needs: provide for more consistent evaluations of
relative and absolute risks across different program areas that
OSWER addresses;
• Improve on Explicit Use of Data: allow for reevaluating rankings
as new information becomes available (algorithms are made explicit
where possible);
• Public Perception and Risk: address the apparent disparity between
the public's perception of risks and the actual risks that may
require priority OSWER attention in the future; and
• Type of Decision: provide a clearer separation of risk assessment
from risk management.
The outputs/products of this study are expected to serve to rank important
waste management problems facing OSWER; identify public health and environmental
problems for future investigation; and identify data needs and uncertainties
associated with information gaps. The findings of this study are not intended
as an assessment of the effectiveness of existing regulatory controls or as
justification for more or less regulatory involvement.
SCOPE AND ORGANIZATION OF THE STUDY
This study was organized with a study chair and five Workgroups. The five
Workgroups are: (1) Acute Events, (2) Other Health Effects, (3) Ecological
Effects, (4) Welfare Effects1, and (5) Ground Water Valuation. Each workgroup
produced a separate report. The Workgroups divided the universe of public.
health and environmental problems into different problem areas. These problem
areas generally reflected organizational structure and program priorities. The
The Welfare Effects Workgroup examined impacts on welfare (which were
defined as damages to economic resources that result in a reduction in the value
of commercial activities or in the value of human well-being). This approach
was chosen because of the lack of studies associating economic impacts with OSWER
problem areas. While the supporting workgroup report is titled "Report of the
Economic Effects Workgroup", the Workgroup recommended that "Welfare Effects"
be used for purposes of the overview report.
v
-------
Workgroups believe that the list of problem areas accounts for the principal
risks associated with OSWER's programs.
FINDINGS
Table ES-1 summarizes both the relative rankings (e.g., high, medium, and
low) and the level of confidence of the rankings. The workgroup rankings were
developed through a process that involved the evaluation of available
information and judgments to estimate the relative risks associated with each
problem area.
The problem areas in Table ES-1 are ordered by decreasing severity--that
is, those judged to represent the highest risks are at the top and those judged
to pose the least risks are at the bottom. Because it provides a range of
results associated with the individual workgroup rankings, Table ES-1 can only
be used to stratify problem areas into broad groups.
For purposes of this overview report, the workgroup rankings have been
stratified into three bands:
• High: Those problem areas that have at least two rankings of high
and were assessed by at least four workgroups. This band includes
Municipal Landfills; Industrial Facilities; Hazardous Waste Storage
Treatment Tanks, Drums and Containers; and Land Disposal of
Hazardous Waste.
• Medium: Those problem areas not grouped as "high" or "low" relative
risk for purposes of this overview report.
• Low: Those problem areas that did not receive any "high" rankings.
Problem areas grouped into as "low" were ranked by no more than
four workgroups and had a minimum of at least two low rankings by
the Workgroups. (While Ocean Dumping was ranked as high (with high
confidence) within the Ecological Effects Workgroup, the definition
of Ocean Dumping included activities OSWER programs do not address.
As such, the ranking of Ocean Dumping within this matrix has been
adjusted downward.)
Had Superfund activities been grouped into one single category, it appears
that the ranking would have been: medium by the Acute Events workgroup and high
for the other four workgroups. Given this combination, Superfund activities
would be moved from the "medium" risk grouping to "high".
Because risk is a continuum, these three broad bands of problem areas
reflect professional judgment, not risk thresholds. Table ES-1 suggests that
those problem areas with better supporting information tended to be associated
with higher confidence and also were actually ranked (as opposed to not being
ranked or being combined with other problem areas).
The underlying information used to support rankings shown in Table ES-1
differs significantly both within and across Workgroups. The differences
vi
-------
include such factors as measures of risk used for ranking and the quality of
information needed for ranking. These rankings provide risk estimates
associated with current emissions that reflect in-place controls. Obviously,
rankings would differ without controls. Selected highlights that may assist in
interpreting Table ES-1 include:
• Industrial Facilities: All Workgroups except the Ground Water
Valuation Workgroup stated that they believed that risks may be
substantially higher than the data suggest and that more data should
be collected.
• Superfund Sites: The Other Health Effects Workgroup considered
current National Priorities List [NPL] sites (currently about 1,200
sites) and sites that are to be evaluated for NPL listing (currently
over 30,000 sites). The Welfare Effects and the Other Health
Effects Workgroups combined remedial and removal actions at both NPL
and non-NPL sites. The other workgroups considered only removal
actions at NPL or potential NPL sites.
• Chemical and Industrial Manufacturing Facilities: The Other Health
Effects Workgroup called this problem area "Chemical and Industrial
Manufacturing Facilities" or "Accidental Releases at Chemical
Facilities." The Welfare Effects Workgroup called this problem area
"Accidental Releases at Chemical Facilities." The Acute Events
Workgroup addressed only risks posed to workers at chemical and
industrial facilities.
• Underground Storage Tanks: This problem area was only considered
by the Ground Water Valuation Workgroup. It includes: hazardous
substance product storage tanks, petroleum underground storage
tanks, and exempt underground storage tanks.
• Underground Injection of Hazardous Waste: The Other Health Effects
Workgroup considered injection of hazardous waste into Class I
injection wells. The other workgroups considered all underground
injection of hazardous waste regulated under 40 CFR Part 267 Subpart
G and 40 CFR Part 146.
Table ES-1 suggests that the relative rankings of problem areas are
generally consistent with program priorities, and that some problem areas not
necessarily perceived as high program priorities could be ranked high even
though they are supported by relatively uncertain information.
PROGRAM RECOMMENDATIONS
The following long-term recommendations for the continued support of
strategic planning are:
• Risk Rankings: should be reviewed periodically to incorporate new
information gathered.
Vll
-------
• Centralized Repository: should be established for the maintenance
of information used to perform the rankings.
To provide a more consistent basis to support decisions on program
direction, the Workgroups recommend developing better information in the
following areas:
• Broader Range of Information Collection: Information was especially
limited and of low quality for ecological effects, welfare effects,
and ground water valuation. Consideration should be given to
enhanced data collection and methodological work in these areas.
• Cross Media Analysis: It is recommended that efforts be made to
better assess major risks from all pathways as opposed to those
pathways easiest to evaluate.
• Low Confidence Areas: The Workgroups expressed particularly low
confidence in their findings in four problem areas (i.e., Mixed
Wastes, Other Treatment, Storage, and Disposal Facilities,
Industrial Facilities, and Solid Waste Management Units). Future
analyses should benefit from more and better data in these areas.
• Ground Water Valuation: The Ground Water Valuation Workgroup
eliminated 13 problem areas from consideration, partly due to lack
of information. The Ecological Effects Workgroup did not consider
Removal Actions at Non-NPL Sites due to insufficient data. To
improve future studies, EPA should benefit from obtaining new data
sources for these problem areas.
Finally, to improve confidence of the findings of this and future studies,
sensitivity analyses of the use of alternative methodologies and the
corresponding impacts on final rankings would be useful.
Vlll
-------
ES-1
COMPARISON OF WORKGROUP RANKINGS OF OSWER PROBLEM AREAS
Workgroup Ranking
Municipal Landfills
Industrial Facilities
Hazardous Waste Storage Treatment Tanks.
Drums, and Containers
Petroleum Underground Storage Tanks
Land Disposal of Hazardous Waste
Superfund Sites
Remedial Activities
Transportation of Hazardous Materials
Chemical and Industrial
Manufacturing Facilities
Underground Storage Tanks
Oil Spill Response
Oil and Gas Operations
Solid Waste Management Units
Exempt Storage Tanks
Mining Wastes
Other Treatment. Storage.
and Disposal Facilities
Removal Activities
Combustion of Hazardous Waste
Municipal Waste Combustion
Hazardous Substance/Product
Underground Storage Tanks
PCS Wastes
Land Treatment of Hazardous Wastes
Underground Injection of Hazardous Waste
Mixed Wastes
Ocean Dumping
Ocean Incineration
Acute
Events
L
H
M
H
M
-
£
H
X
-
H
H
M
M
L
M
M
M.
£
M
M
£
M
L
L
L
Other
Health
Effects
M
H
H
M
M
H
-
£
L
-
L
M.
X
M
M
M
-
M
M.
M
L
M
L
L
L
-
Ecological
Effects
X
M
M.
M£
X
-
H
JA£
M£
-
M£
M
ML
M£
MM
M
-
M
MH
M,
MH
M
ML
M
H
M£
Welfare Ground-Water
Effects Valuation
H H
H
H
H
H M
11 TT
—
L
L
H
L
M £
M. M,
M.
M, M.
M £
-
L L
L L
L
L
L
L
L
L
L
1
1
3
5"
IX
-------
Notes on the Comparison Table
Letters reflect relative risk rankings judged for problem areas:
ft High or higher relative risk potential
Moderately high relative risk potential
Medium or moderate relative risk potential
Moderately tow relative risk potential
Low or tower relative risk potential
The workgroup dd not consider this problem area.
Uncertainty is reflected in lettering as:
H Low uncertainty (high confidence)
3{ Medium uncertainty (medium confidence)
H High uncertainty (tow confidence)
No attempt was made to 'add* risks across Workgroups or compare the importance
of one Workgroup as more significant than another. Note that differences in
problem areas addressed is due to a combination of available data and scope.
X
-------
CHAPTER 1. INTRODUCTION
The Office of Solid Waste and Emergency Response (OSWER) Comparative Risk
Project was initiated to serve strategic planning needs by exploring the
comparative risks posed by various waste management practices regulated by
and/or under OSWER purview. The study was also undertaken to gain experience
in performing comparative analysis, identifying information needs, and assisting
in establishing research priorities.
The statutory goals for the solid waste program direct the Agency to
promote protection of health and the environment. The statutory language varies
among the different programs, the emphasis is consistently on developing
standards to protect the health and the environment or by adopting standards of
other Agency programs.
In evaluating study goals, a number of objectives became apparent,
including:
• Identifying Data Needs: provide for more consistent evaluations of
relative and absolute risks across different program areas that
OSWER addresses;
• Improve on Explicit Use of Data: allow for reevaluating rankings
as new information becomes available (algorithms are made explicit
where possible);
• Public Perception and Risk: address the apparent disparity between
the public's perception of risks and the actual risks that may
require priority OSWER attention in the future; and
* Type of Decision: provide a clearer separation of risk assessment
from risk management.
1.1 ORGANIZATION
This report is organized as follows:
• Chapter 1: Introduction -- addresses the goal of this study, its
relationship to strategic planning, and its relationship to Agency
comparative risk efforts.
• Chapter 2: Study Approach -- addresses specific problem areas
covered and the overall approach used.
• Chapter 3: Study Results -- addresses the confidence in the data
sources and methodology used, differences among the Workgroups
(including problem areas addressed and methodologies used), future
data needs, and the comparison of risks.
-------
• Chapter 4: Information Needs -- addresses the links between
information needs based upon the rankings of problem areas and the
type and quality of information available to support rankings.
• Appendix A: Glossary -- defines key terms.
Appendix B: Definition of OSWER Problem Areas -- defines the areas
addressed by the Workgroups.
• Appendix C: Individual Workgroup Methodologies -- addresses the
different methodologies used by the Workgroups to determine the
relative risk of the OSWER problem areas.
These materials constitute the project overview. The original workgroup
reports that provide the foundation for this overall report should also be
reviewed for detailed evaluations of information bases, assumptions, and
references.
1.2 RELATIONSHIP OF STUDY TO AGENCY COMPARATIVE RISK EFFORTS
The OSWER Comparative Risk Project is a follow on to an earlier Agency
effort. In 1986, the Administrator commissioned a special task force of senior
EPA managers and technical experts to compare the risks associated with major
environmental problems. In February 1987, the task force issued its report,
Unfinished Business: A Comparative Assessment of Environmental Problems. The
report was an Agency-wide effort that examined all environmental problems facing
the nation and estimated the relative risk to public health and the environment
posed by each problem area, assuming existing levels of regulatory control.
In many instances, the rankings in Unfinished Business did not correspond
closely with EPA's current program priorities. For example, areas of
substantial Agency effort identified as posing relatively low risk included
hazardous waste and Superfund sites, underground storage tanks, and municipal
non-hazardous waste sites. The report, however, suggested that this seemingly
anomalous result was not necessarily inappropriate. Rather, some problem areas
appeared to pose relatively low risks precisely because of the high levels of
Agency resources were devoted to controlling them. Furthermore, Unfinished
Business noted higher levels of effort may remain necessary to hold risks at
current levels. In reviewing Unfinished Business, the Workgroups determined
that at least one reason that OSWER problem areas did not rank high was the lack
of information to support development of national chemical-specific rankings.
Solid waste programs often focus on waste streams and may lack pollutant-
specific information on risks. Moreover, information on Superfund risks may
exist but is sometimes difficult to obtain. As such, the OSWER Comparative Risk
Project was viewed as an opportunity to make fuller use of newer program-related
information including information under development when Unfinished Business was
being prepared.
Unfinished Business relied upon professional judgment (supported by
background materials) to develop rankings. Thus, it is not possible to modify
the rankings in Unfinished Business based on additional information. The OSWER
-------
Comparative Risk Project was designed to generate rankings through a process
that better documented the algorithm and rationale used to generate rankings.
If the rationale for the algorithms and rankings could be documented, both
algorithms and the rankings could be updated as new information becomes
available.
1.3 RELATIONSHIP OF STUDY TO STRATEGIC PLANNING
In June 1985, the Administrator initiated a strategic planning process.
The process was intended to assist in developing a consensus around
environmental goals across all media and help establish program priorities. The
resulting OSWER strategic planning initiative was based on the concept of
preventing waste transfers from one location to another and one medium to
another. Given the absence of a cross-media perspective on waste management
issues, it is possible that a specific regulation in one media may shift waste
to other media without reducing risks to human health or the environment. This
study supports this larger planning process.
Comparing risks from releases to different media allows OSWER to begin to
assess the consequences of regulation in different media. In the long-term,
this study will support OSWER strategic planning by: identifying public health
and environmental problems for future investigation; and identifying data gaps
and associated uncertainties.
This study should be seen only as an early step in the larger strategic
planning process. It was undertaken to gain experience in performing
comparative analysis, as much as for the results of the analysis itself. This
study is an opportunity to make fuller use of available information and to build
upon the experience of Unfinished Business. While the results are expected to
assist in the identification of information gaps and in establishing research
priorities, no other applications of the results have been determined at this
time. Thus, the findings are not intended to assess the effectiveness of
existing regulatory controls or justify more or less regulation.
-------
CHAPTER 2. STUDY APPROACH
This chapter addresses the general study approach. The overall
methodology (Section 2.1) and the OSWER problem areas (Section 2.2) are also
discussed.
2.1 OVERALL APPROACH
This study was organized with a study chair and five workgroup chairs.
The five Workgroups are: Acute Events; Other Health Effects; Ecological
Effects; Welfare Effects; and Ground Water Valuation. Each workgroup produced
separate reports (which are summarized in Appendix C). Workgroup assumptions,
references employed and other useful information is available in the individual
reports.
To promote consistency, the Workgroups began with the same set of
environmental problem areas. A centralized effort was initiated to collect
information and seek review for completeness and technical accuracy. This
information was used as a baseline for ranking the problem areas.
The evaluation relied primarily upon readily available information, where
possible. However, in discussing possible approaches they might take, the
Workgroups decided to modify the approach employed in Unfinished Business.
Recognizing that direct comparisons across problem areas are difficult, the
following steps were undertaken to improve the plausibility of such comparisons:
• promoted the consistent use of information across Workgroups,
• requested assumptions be explicitly stated and
• characterized uncertainties associated with the rankings.
Assessing health risks is generally easier than assessing risks associated
with other measures of environmental damage (e.g., ecological risks) because of
availability of information and acceptance of consistent methods to evaluate
risks. The assessment process is further complicated by the difficulties in
relating risks to the types of activities. For example, it is difficult to
relate specific health risks (typically assessed by a single chemical) to
different types of heterogenous hazardous waste.
Another difficulty in comparing health risks with other measures of
environmental damage is the difficulty in comparing endpoints that must be
associated with differing levels of confidence (or uncertainty). The
uncertainty of a specific endpoint (economic impact) is a function of the
uncertainties associated with both the inputs and algorithm employed. As
previously stated, measures of environmental risk such as economic impacts may
be the most relevant measures for some audiences, but are also the most
uncertain and the most limited in the support of comparisons. Thus, the
strength of conclusions in this report reflects consideration of the relative
uncertainties of the input parameters and the algorithm employed to rank problem
areas.
-------
2.2 OSWER PROBLEM AREAS
OSWER problem areas generally reflected organizational structure and
program priorities. The 24 problem areas represent the principal risks
associated with OSWER's programs (as defined in Appendix B) . These problem
areas are not mutually exclusive; the same release may be addressed in more than
one problem area (e.g., disposal of PCB wastes at a site on the National
Priorities List may be reflected in both the problem areas PCB Wastes and
Remedial Actions). This potential for "double counting" is noted wherever
possible.
2.2.1 Problem Area Definitions
Differences in the Workgroups' definitions for some problem areas created
difficulty in comparing relative risk rankings. These differences can be
divided into two categories:
• those related to different types of wastes or waste management
practices and
• those related to different types of events.
A number of the workgroups included different types of waste and waste
management practices in the same problem area. For the Underground Injection
of Hazardous Waste problem area, the Other Health Effects Workgroup considered
injection of hazardous waste into Class I injection wells, while the other
workgroups considered all underground injection of hazardous waste regulated
under 40 CFR Part 267 Subpart G and 40 CFR Part 146. In the Ocean Dumping
problem area, Workgroups diverged in their definitions of what constituted ocean
dumping. The Other Health Effects and Welfare Effects Workgroups considered
risks from ocean dumping of municipal waste, dredge spoil, and discharges of
industrial waste into the ocean, the Ecological Effects Workgroup considered
risks from ocean dumping of municipal waste and dredge spoil, and the Acute
Events Workgroup only considered risks from ocean dumping of municipal waste.
As to the types of events and associated risks considered, the Acute
Events Workgroup took a different approach. For example, in the Oil and Gas
Operations problem area, the Acute Events Workgroup considered only the effects
of blow-outs at oil wells, while other workgroups considered the effects of
releases from wastes resulting from oil and gas activities.
In the Chemical and Industrial Manufacturing Facilities problem area, the
Acute Events Workgroup addressed only risks posed to workers at chemical and
industrial facilities, whereas other workgroups considered the impacts of
releases both within and outside chemical and industrial manufacturing
facilities.
-------
2.2.2 Problem Areas Addressed
Every workgroup did not address all of the 24 OSWER problem areas. In
fact, only five problem areas were addressed by all the Workgroups (Combustion
of Municipal Waste, Land Disposal of Hazardous Wastes, Mining Wastes, Other
Treatment, Storage, and Disposal Facilities, and Municipal Landfills). Four
other problem areas (Remedial Activities, Hazardous Substance Product Storage
Tanks, Petroleum Underground Storage Tanks, and Exempt Storage Tanks) also were
addressed by all Workgroups, but were merged with existing/new problem areas.
For the Remedial Activities, the Other Health Effects Workgroup and Welfare
Effects Workgroup combined this problem area with Removal Activities to create
a new problem area called Superfund Removal and Remedial Activities. Similarly,
the Ground Water Valuation Workgroup combined Hazardous Substance Product
Storage Tanks, Petroleum Underground Storage Tanks, and Exempt Storage Tanks,
into a single problem area called Underground Storage Tanks.
Workgroups excluded problem areas from consideration for two reasons:
• insufficient data were available to properly assess the relative
potential risk posed by a problem area or
• the relative risks were too small to merit analysis.
Based on these criteria, the Ground Water Valuation Workgroup eliminated 13
problem areas: Land Treatment of Hazardous Waste; Hazardous Waste Storage
Treatment Tanks, Drums, and Containers; PCB Wastes; Mixed Wastes; Underground
Injection of Hazardous Waste; Transportation of Hazardous Materials; Removal
Activities; Industrial Facilities; Municipal Waste Combustion; Oil Spill
Response; Chemical and Industrial Manufacturing Facilities; Ocean Incineration;
and Ocean Dumping. The other four workgroups evaluated most of the other
problem areas.
-------
CHAPTER 3. STUDY RESULTS
This chapter addresses general results reported from each of the
Workgroups; specific rankings of the Workgroups; methodological and data factors
affecting the rankings; and data quality and confidence.
3.1 GENERAL RESULTS
This section draws upon the highlights of the results reported in the
individual reports. While Appendix C summarizes approaches, data use and
results, the individual reports contain additional information.
3.1.1 Acute Events
The Acute Events Workgroup provided recommendations in three key areas:
new data collection efforts; revisiting risk rankings/estimates of relative risk
over time; and greater data centralization.
• New Data Collection: Two criteria were employed to formulate
recommendations:
• data were especially poor and
• the relative risks may be greater than the numeric score may
reflect.
With few exceptions, data needs are greatest for problem areas in
the low relative risk ranking group. The Workgroups stressed that
the following problem areas appeared to warrant additional work:
Ocean Dumping, Combustion of Municipal Wastes, Mining Waste,
Municipal Landfills, and Mixed Wastes.
• Revisiting risk rankings/estimates of relative risks over time: The
Workgroup recommended that all the problem areas be assessed
periodically to determine if regulatory or technological changes
have affected the relative risks posed by acute events.
Greater Data Centralization: The Workgroups concluded that
considerable information is available through State agencies,
Regional EPA offices, different offices within EPA, and other
federal agencies. While many problems are involved in developing
an accurate, reliable centralized reporting system, such a system
should make a significant difference through improving confidence
in rankings, providing a basis for re-visiting rankings, and
providing for more reliable priority setting.
3.1.2 Other Health Effects
The Other Health Effects Workgroup provided both substantive and general
recommendations. The substantive recommendations are:
-------
• Risks associated with Mixed Wastes, Ocean Incineration, and
Underground Injection, appear to be very low. Additional work based
on health risks, in the absence of new information, is not
recommended.
• Improve on cross media analysis, especially the evaluation of air
risks. In particular, air risks associated with Land Disposal of
Hazardous Wastes and Municipal Landfills are substantially higher
than anticipated.
Other recommendations that addressed general risk assessment activities
included:
• Continue efforts to model risk associated with waste-related
problems. Such modeling offers valuable insights into health
problems and is critical to informed regulatory decision-making and
priority-setting.
• Develop consistent assumptions for use in assessing exposure and
characterizing the quality of the information.
• Evaluate all pathways for their potential risks to public health.
Air or direct contact risks could be larger than risks associated
with the drinking of contaminated ground water.
• Address risks associated with noncarcinogenic health effects more
critically (there appears to be a bias towards cancer risks).
• Use modeling results, not just monitoring data, to assess exposure,
especially for air and drinking water.
3.1.3 Ecological Effects
The Ecological Effects Workgroup stressed that the absolute numerical
rankings in its report should not be taken as absolute because of the
assumptions employed to develop rankings and the data limitations. In general,
data were limiting for many problem areas. The Workgroup provided several
recommendations, including:
• In presenting data on ecological risks, the associated
uncertainties, limited meaningful measures of pollutant-related
stress, and the lack of quantitative ecological risk information,
all make the assessment of ecological risks difficult. As such, it
is recommended that the uncertainties of this analysis be carried
forward in subsequent presentations of the data.
• Conduct a comprehensive review to identify situations with the
greatest potential for significantly damaging to important bird and
mammal populations. In addition to the scientific literature,
information generated by the Fish and Wildlife Service, NOAA, EPA
Regional Offices, and State environmental and natural resources
agencies will enable EPA to better focus its efforts.
-------
• Form a task force to examine inter- and intra-Agency projects that
have incorporated ecological effects data and risk assessment
methodologies. The task force should: (1) coordinate Federal
efforts both within EPA (e.g., the Office of Pesticides and Toxic
Substances) and other departments (e.g., Department of Agriculture,
Department of the Interior); (2) assess available data bases and
evaluate international efforts to obtain data specific to ecosystem
impacts; (3) coordinate with the Agency-wide groups (specifically
the eco-risk planning group on Reducing the Uncertainty in Risk
Assessments (RURA); and (4) following completion of the above
steps, organize an international conference.
3.1.4 Welfare Effects
The Welfare, Effects Workgroup recommended that welfare damages be
investigated carefully for these high risk problem areas if any new comparative
risk analysis is initiated. The Workgroup further recommended that if assessing
welfare impacts is a priority for future regulatory activities, then the
uncertainties associated with rankings and data should be explicitly considered.
3.1.5 Ground Water Valuation
The Ground Water Valuation Workgroup focused on the approach and analysis
employed to assess resource damage to drinking water supplies on the basis of
replacement costs.
3.2 KEY FINDINGS
This section addresses the key findings of each workgroup analyses. Final
rankings, data quality and confidence levels, and overall confidence in rankings
are discussed.
3.2.1 Final Rankings
Table 1 summarizes both the relative rankings (e.g., high, medium, and
low), and the level of confidence (indicated by text face) of the rankings.
The workgroup rankings were developed through a process that involved the
evaluation of available information and judgments to estimate the relative risks
associated with each problem area.
The problem areas in Table 1 are ordered by decreasing severity--that is,
those judged to represent the highest risks are at the top and those judged to
pose the least risks are at the bottom. Because it provides a range of results
associated with the individual workgroup rankings, Table 1 can only be used to
stratify problem areas into broad groups.
For purposes of this overview report, the workgroup rankings have been
stratified into three bands:
-------
• High: Those problem areas that have at least two rankings of high
and were assessed by at least four workgroups. This band includes
Municipal Landfills; Industrial Facilities; Hazardous Waste Storage
Treatment Tanks, Drums and Containers; and Land Disposal of
Hazardous Waste.
• Medium: Those problem areas not grouped as "high" or "low" relative
risk for purposes of this overview report.
• Low: Those problem areas that did not receive any "high" rankings.
Problem areas grouped as "low" ranked by no more than four
workgroups and had a minimum of at least two low rankings by the
Workgroups. (While Ocean Dumping was ranked as high (with high
confidence) within the Ecological Effects Workgroup, the definition
of Ocean Dumping included activities OSWER programs do not address.
As such, the ranking of Ocean Dumping within this matrix has been
adjusted downward.)
Had Superfund activities had been grouped into one single category, it
appears that the ranking would have been: medium by the Acute Events workgroup
and high for the other four workgroups. Given this combination, Superfund
activities would be moved from the "medium" risk grouping to "high".
Given that the ordering of problem areas shown in Table 1 has been derived
from the ranking efforts of the individual workgroups, the ordering of problem
areas shown in Table 1 reflects the assumptions associated with the rankings of
each workgroup, and then some additional assumptions. For this reason, rankings
within a workgroup (e.g., high, medium high, medium, etc.) are more certain
than the compilation of rankings across workgroups. Due to the limitations of
the data and methods used to develop the rankings, minor differences in rankings
among problem areas should not be construed to represent genuine differences in
aggregate, national risks. Moreover, these three broad bands of problem areas
reflect professional judgment, not risk thresholds.
The underlying information used to support rankings shown in Table 1
differ significantly both within and across workgroups. The differences include
such factors as measures of risk used for ranking and the quality of information
needed for ranking. These rankings provide risk estimates associated with
current emissions that reflect in place controls. Obviously, rankings would
differ without controls. Selected highlights that may assist in interpreting
Table 1 include:
• Industrial Facilities: All workgroups except the Ground Water
Valuation Workgroup stated that they believed that risks may be
substantially higher than the data suggest and that more data should
be collected.
• Superfund Sites: The Other Health Effects Workgroup considered
current National Priorities List [NPL] sites (currently about 1,200
sites) and sites that are to be evaluated for NPL listing (currently
over 30,000 sites). The Welfare Effects and the Other Health
Effects Workgroups combined remedial and removal actions at both NPL
10
-------
and non-NPL sites. The other workgroups considered only removal
actions at NPL or potential NPL sites.
• Chemical and Industrial Manufacturing Facilities: The Other Health
Effects Workgroup called this problem area "Chemical and Industrial
Manufacturing Facilities" or "Accidental Releases at Chemical
Facilities." The Welfare Effects Workgroup called this problem area
"Accidental Releases at Chemical Facilities." The Acute Events
Workgroup addressed only risks posed to workers at chemical and
industrial facilities.
• Underground Storage Tanks: This problem area was only considered
by the Ground Water Valuation Workgroup. It includes: hazardous
substance product storage tanks, petroleum underground storage
tanks, and exempt underground storage tanks.
• Underground Injection of Hazardous Waste: The Other Health Effects
Workgroup considered injection of hazardous waste into Class I
injection wells. The other workgroups considered all underground
injection of hazardous waste regulated under 40 CFR Part 267 Subpart
G and 40 CFR Part 146.
Table 1 suggests that the relative rankings of problem areas are generally
consistent with program priorities, and that some problem areas not necessarily
perceived as high program priorities could be ranked high even though they are
supported by relatively uncertain information. Table 1 also suggests that those
problem areas with better supporting information tended to be associated with
higher confidence and also were actually ranked (as opposed to not being ranked
or being combined with other problem areas).
The Acute Events Workgroup ranked Transportation of Hazardous Materials,
Oil & Gas Operations Waste, Oil Spill Response, Accidental Releases at Chemical
Facilities, Petroleum Underground Storage Tanks, and Industrial Non-Hazardous
Waste Facilities highest; and ranked Ocean Dumping, Remedial Actions at NPL
Sites, Municipal Waste Combustion, Land Treatment of Hazardous Waste, Mining
Waste, Ocean Incineration, Mixed Waste, and Municipal Landfills lowest.
The Other Health Effects Workgroup ranked Superfund Activities, Industrial
Non-Hazardous Waste Management Facilities, Solid Waste Management Units, and
Hazardous Waste Storage and Treatment Tanks, Drums, and Containers at the top
of their list. In contrast to the Acute Events Workgroup, Oil Spill Response
was among the lower risk problem areas. It is worthwhile noting that Solid
Waste Management Units received a high ranking in this workgroup, but there was
a lack of data to support this ranking at the time that the rankings were
developed. However, the Regulatory Impact Analysis (released after drafting of
this report) for the Corrective Action rule generally supports the rankings
assigned by this workgroup.
In contrast to the Acute Events Workgroup, the Ecological Effects
Workgroup ranked Ocean Dumping as the highest. This ranking assumed that Ocean
Dumping includes practices that are not and are not expected to be addressed by
OSWER because they are addressed by other programs (e.g., air or water).
11
-------
However, like the Other Health Effects Workgroup, the Ecological Effects
Workgroup ranked Remedial Actions at NPL Sites in the high risk group. Other
problem areas that were ranked as posing relatively high risk were: Land
Disposal of Hazardous Waste, Municipal Landfills, PCB Wastes, and Mining Wastes.
Problem areas that particularly low included: Underground Injection of
Hazardous Waste, Oil Spill Response, Solid Waste Management Units,
Transportation of Hazardous Materials, Exempt Underground Storage Tanks, and
Accidental Releases at Chemical Manufacturing Facilities.
The Welfare Effects Workgroup placed problem areas into three broad
relative risk categories. The Workgroup concluded that comparisons beyond this
level of detail were not justified, considering the quality of data supporting
individual scores. At this level of comparison, however, the Workgroup
expressed reasonable confidence in its conclusions. Superfund Activities, Solid
Waste Management Units, Land Disposal of Hazardous Waste, Industrial
Non-Hazardous Waste Facilities, Petroleum Underground Storage Tanks (Including
Used Oil), Municipal Landfills, Hazardous Waste Storage and Treatment Tanks
were all ranked relatively high by the Welfare Effects Workgroup. Combustion
of Hazardous Waste, Transportation, Hazardous Substance Underground Storage
Tanks, Land Treatment of Hazardous Waste, Oil Spill Response, Underground
Injection, Ocean Dumping, PCB Wastes, Accidental Releases at Chemical
Manufacturing Facilities, Mixed Waste, Ocean Incineration were all ranked
relatively low.
The Ground Water Valuation Workgroup report estimated aggregate national
costs of releases or potential releases to ground water from OSWER problem
areas. The principal findings were:
• Ground Water resource damage is extremely variable across problem
areas. It is also extremely variable across facilities in problem
areas due to site-specific differences in use patterns, pollutant
releases, hydrogeological characteristics, and availability of
substitute water sources.
• Nationwide, underground storage tanks produce the highest total
resource damage ($15 billion) in present value terms.
• NPL sites have the highest mean resource damage (present value of
$9.7 million per site).
• The size of the facilities is highly correlated with high resource
damage.
Other highly ranked problem areas were: Municipal Landfills, Solid Waste
Management Units, Land Disposal of Hazardous Waste, Mining Waste, Oil and Gas
Operations Waste, Industrial (Subtitle D) Landfills, and Land Treatment of
Hazardous Waste.
12
-------
3.2.2 Data Quality and Confidence Levels
The level of confidence that Workgroups expressed in their rankings varied
greatly. While they may have had confidence in broad groupings of problem areas
(e.g., high vs. low), they reported considerable uncertainty in relative risk
rankings for most problem areas. Thus, while it is possible to derive rankings
for problem areas that subdivide the problem areas presented, the differences
in the rankings would not be as robust.
The Other Health Effects Workgroup, the Ecological Effects Workgroup, and
the Acute Events Workgroup all developed separate scales for assigning
confidence levels to problem areas. The Other Health Effects Workgroup scale
represents the confidence the Workgroup expressed in the accuracy of its
assessment of the relative risk posed by a problem area. The rankings,
therefore, reflect staff judgments, based upon available data, of the relative
risk posed by a problem area. It should be stressed that the confidence levels
are not simply data quality indicators. In many problem areas, data were
lacking or of extremely poor quality, but the Workgroup expressed a high degree
of confidence in its assessment of the relative potential risk posed.
The Acute Events Workgroup and the Ecological Effects Workgroup developed
a confidence level scale based on the quality of the data used to develop
relative risk assessments. Both used a similar method for assessing data
quality. The Workgroups assigned a score to each parameter used to calculate
relative potential risk. Factors such as the degree of documentation supporting
reported effects, the generalization of data, and the degree to which data were
based on professional judgment were used to assess the quality of data
supporting each parameter. The number of parameters supported by good,
moderate, or poor quality data then determined the overall data quality score
a problem area received.
Confidence levels for the Welfare Effects Workgroup report were derived
from two sources: problem area summary sheets and the text of the workgroup's
report. These sources were reviewed by the Workgroup to determine the degree
to which findings were based on comprehensive historical data, modeling data,
or professional judgment. The problem area summary sheets and the text of the
report then were reviewed for any indications of the degree of confidence the
Workgroup expressed in the data used to estimate problem area rankings.
• Historical data existed and/or the Workgroup expressed a high degree
of confidence: high.
• When little historical data existed or the Workgroup used model data
and/or the Workgroup expressed only medium confidence: medium.
• When the Workgroup findings were based primarily on professional
judgment (i.e., no data existed): low.
Confidence levels for the Ground Water Valuation Workgroup report were
derived from Exhibit 13 of its report. This exhibit lists the sources of data
for each problem area and provides a relative assessment of data quality. As
13
-------
such, the characterizations of good, fair, or poor were used because they
corresponded directly with rankings of confidence.
3.2.3 Overall Confidence in Rankings
Workgroups expressed particularly low confidence in four problem areas:
• Mixed Wastes;
• Other Treatment, Storage, and Disposal Facilities;
• Industrial Facilities; and
• Solid Waste Management Units.
With the exception of Other Treatment, Storage, and Disposal Facilities, all the
Workgroups stated that they either had low confidence in their estimate of
relative risk or did not have enough data to assess the risk posed by the
problem area. For the Other Treatment, Storage, and Disposal Facilities problem
area, all stated that they had low confidence in their estimate of relative
potential risk, except the Acute Events Workgroup, which that its estimate was
based on moderate quality data only.
All Workgroups expressed a high/moderate level of confidence that they had
accurately identified the risks associated with Oil and Gas Operations Waste.
Workgroups also generally reported a high to moderate level of confidence in the
findings for the problem areas Petroleum Underground Storage Tanks and Exempt
Storage Tanks.
3.3 METHODOLOGICAL AND DATA FACTORS AFFECTING WORKGROUP RANKINGS
The type and quality of information affect the type of method that might
be appropriate for supporting the rankings of problem areas.
The methodology each workgroup ultimately used had considerable impact on
the final ranking. In some cases, differences in one or two parameters in a
workgroup's methodology had a substantial impact on the relative risk score a
problem area received. Below, these critical parameters, or "driving factors",
are identified for each workgroup. Refer to Appendix C for a more detailed
summary of each workgroup methodology.
3.3.1 Acute Events
The Acute Events Workgroup's method assumed the overall relative risk
posed by a problem area is comprised of four effects:
14
-------
• acute exposure health risks;
• chronic health risks from acute events;
• welfare effects; and
• ecological risks.
For each effect, relative risks were assessed by summing measures of the
observed frequency of occurrence of each event, the population potentially
exposed, and the inherent hazard (based on the effect type).
3.3.2 Other Health Effects
The main driving force in the Other Health Effects Workgroup's methodology
was cancer risks. Two scales (i.e., individual and population) were used to
estimate cancer risk scores while only one scale was used to assess noncancer
scores. As a result, cancer risks tended to drive the final relative risk
rankings. Thus, scores used to determine ranking were primarily formulated from
the application of an algorithm that used information from reports.
The potential biases associated with combining high and low estimates was
mitigated by the workgroup's use of professional judgment. If relative risk
scores were thought to be disproportionately influenced by a single parameter,
the Workgroup would review the input parameters to ensure that reasonable
factors had been considered. For example, the Workgroup acknowledged that for
several problem areas (i.e., Mixed Wastes, PCB Wastes, Hazardous Materials
Transportation, and Underground Injection), the relative risk scores were
initially heavily influenced by the toxicities and volumes of chemicals
released. The Workgroup thought this gave an inflated estimate of the relative
risk associated with these problem areas and revised risk estimates downward.
3.3.3 Ecological Effects
The Ecological Effects Workgroup estimated relative risk as a function of
four composite parameters:
• number of sources and releases in a problem area;
• contaminant concentration;
• receptors; and
• toxicity.
Each composite parameter was derived by taking the mean score of a series of
data elements. The first three composites were then added together, and the
result was multiplied by the toxicity composite. This approach made the
relative risk score heavily dependent on the toxicity composite score.
Furthermore, as the approach used for developing the toxicity composite scores
was based on selecting the highest ranked contaminant in each toxicity category,
15
-------
the effect of the toxicity composite on the relative risk score may have been
multiplied even further.
3.3.4 Welfare Effects
The Welfare Effects Workgroup's methodology was, in many ways, the most
flexible and least prone to being overly influenced by a single parameter.
However, the methodology does seem to weight natural resource damages higher
than other parameters, at least in the problem areas ranked low.
The Workgroup considered welfare effects for three general resource
categories:
• residential and commercial property (four types);
• natural resources (two types); and
• surface water (two types).
Within each category, potential types of damage were identified and the
Workgroup assigned a welfare damage score to each type. As a result, the total
welfare score received by a problem area could be heavily influenced by the
welfare damage score for natural resources (because of its four types of
damage).
3.3.5 Ground Water Valuation
The number, distribution, and location of facilities within a problem area
were three key factors that affected final relative risk rankings for the Ground
Water Valuation Workgroup. The value users placed on ground water located in
different areas was not considered, however. Thus, a problem area was likely
to be highly ranked if it affected a large number of aquifers, irrespective of
what value users placed on the individual aquifers. Therefore, problem areas
where facilities or sources were distributed over many aquifers were much more
likely to receive higher rankings than those that had a narrower distribution.
3.4 RANKINGS AND PUBLIC PERCEPTION OF RISKS
The importance of public perception and effective communication is a
priority for successful program management. Program history suggests that the
public wants to know that a site is "safe" before EPA or a State leaves the
site. Moreover, it is likely that the judgment as to what constitutes safe
varies with the site and the people surrounding that site. Given this premise,
there is an increasing concern for better understanding the public, improving
communication with the public, and improving methods for addressing public
concerns.
16
-------
During the preparation of Unfinished Business, EPA -reviewed puMi<* polling
data conducted over a two-years by the Roper Organization. The survey focused
on perceptions of 15 environmental problems. Because surveys did not directly
match the 31 EPA problem areas; they did, however, match several problem areas.
As a result, Unfinished Business employed professional judgment to allow
comparison of Roper's results with EPA's ranking of problem areas.
As with Unfinished Business, the OSWER Comparative Risk Study found some
difference between its rankings of problem areas and public opinion (Table 2).
Discrepancies can be accounted for by a number of factors. One of the most
obvious is that the public's perception of the risk a problem area poses seems
to be closely linked to the visibility and number and density of sites within
a problem area. The degree to which a release is noticeable or produces
dramatic effects is likely to affect public perceptions of risk. This probably
accounts for the fact that many problem areas that were ranked high by the Acute
Events Workgroup often tended to be ranked high in the Roper poll. The
comparison of the overall results of the OSWER Comparative Risk Project with the
results of the Roper study suggest that the public may not place a lesser value
on risks posed by a number of problem areas where the damage caused is less
obvious or more long-term. This would suggest that future attempts to increase
public understanding of the risks posed by various problem areas should focus
on explaining the less visible risks and the mitigation of long-term or chronic
effects. The more effective that EPA is in characterizing risks and what
actions have been taken to reduce them, the more likely it is that the public
and EPA can come into a common alignment that supports the rationale behind
Agency decisions.
17
-------
TABLE 1
COMPARISON OF WORKGROUP RANKINGS OF OSWER PROBLEM AREAS
Workgroup Ranking
Acute
Events
Other
Health
Effects
Ecological Welfare Ground-water
Effects Effects Valuation
Municipal Landfills
M
H
H
Industrial Facilities
H
H
H
oo
Hazardous Waste Storage and Treatment //
Tanks, Drums, and Containers
Petroleum Underground Storage Tanks
H
H
M
M,
H
H
All workgroups except the Ground-Water
Valuation Workgroup stated that they
believed that risks may be substantially
higher than the data suggests and that
further data collection activities should
be undertaken. The Other Health Effects
Workgroup called this problem area
"Industrial Non-Hazardous Waste
Facilities." The other workgroups called
this problem area "Industrial Facilities
(RCRA Subtitle D)."
The Economic Effects Workgroup called this
problem area "Hazardous Waste Storage and
Treatment Tanks".
The Other Health Effects Workgroup called
this problem area "Petroleum Underground
Storage Tanks." The other workgroups
called this problem area "Petroleum
Underground Storage Tanks (RCRA Subtitle
I)."
Land Disposal of Hazardous Waste
M,
H
-------
TABLE 1
COMPARISON OF WORKGROUP RANKINGS OF OSWER PROBLEM AREAS (CONTINUED)
Workgroup Ranking
Acute
Events
Other
Health
Effects
Ecological Welfare Ground-water
Effects Effects Valuation
Superfund Sites
H
H
Remedal Activities
The Other Health Effects Workgroup
considered current NPL and potential NPL
sites (over 30,000 sites). The Economic
Impacts and Other Health Effects
Workgroups combined remedial and removal
actions at both NPL and non-NPL sites. The
other workgroups considered only removal
actions at NPL or potential NPL sites.
The Ground-water Valuation Workgroup
called this problem area "NPL Sites".
Transportation of Hazardous Materials
Chemical and Industrial
Manufacturing Facilities
K
H
L
L
- The Other Health Effects Workgroup called
this problem area "Chemical and Industrial
Manufacturing Facilities" or "Accidental
Releases at Chemical Facilities." The
Economic Effects Workgroup called this
problem area "Accidental Releases at
Chemical Facilities." The Acute Events
Workgroup addressed only risks posed to
workers at chemical and industrial
facilities.
Underground Storage Tanks
TT This problem area was only considered by
the Ground-water Valuation Workgroup. It
includes: hazardous substance product
storage tanks, petroleum underground
storage tanks, and exempt, underground
storage tanks.
09 Spin Response
H
L
-------
TABLE 1
COMPARISON OF WORKGROUP RANKINGS OF OSWER PROBLEM AREAS (CONTNUED)
Workgroup Ranking
Acute
Events
Other
Health
Effects
Ecological Welfare Ground-water
Effects Effects Valuation
Oil and Gas Operations
H
M
M
The Acute Events Workgroup only addressed
the effects of blow-outs at oil wells.
Other workgroups considered the effects of
releases from wastes produced as a result
of oil and gas activities.
Solid Waste Management Units
Exempt Storage Tanks
M
M
H
M
ML
M.
K)
O
Mining Wastes
M,
Other Treatment Storage,
and Disposal Facilities
Removal Activities
Combustion of Hazardous Waste
M
M
M
M
M
The Other Health Effects Workgroup called
this problem area "Exempt Tanks." The
Economic Workgroup called this problem
area "Exempt Storage Tanks." The other
workgroups called this problem area
"Unregulated (Agricultural and
Residential) Underground Storage Tanks."
The Acute Effects Workgroup excluded
mining wastes located at NPL sites; other
workgroups did not explicitly exclude
mining wastes at NPL sites.
In addition to wastes considered by other
workgroups, the Other Health Effects
Workgroup included wastes from smelting
and refining along with other types of
mining waste considered by the other
workgroups.
-------
TABLE 1
COMPARISON OF WORKGROUP RANKINGS OF OSWER PROBLEM AREAS (CONTNUED)
Workgroup Ranking
Acute
Events
Other
Health
Effects
Ecological
Effects
Welfare Ground-Water
Effects Valuation
Municipal Waste Combustion
Hazardous Substance/Product
Underground Storage Tanks
PCB Wastes
Land Treatment of Hazardous Wastes
Underground Injection of Hazardous Waste
£
M
M
£
MH
M
L
M.
L
MH
M
ML
L
L
L
L
L
Mixed Wastes
Ocean Dumping
L
L
L
L
M
H
L
L
The Other Health Effects Workgroup called
this problem area "Hazardous Substance
Tanks." The other workgroups called this
problem area "Hazardous Product
Underground Storage Tanks (RCRA Subtitle
I)-"
The Other Health Effects Workgroup con-
sidered injection of hazardous waste into
class I injection wells. The other
workgroups considered all underground
injection of hazardous waste regulated
under 40 CFR Part 267 Subpart G and 40
CFR Part 146.
The Other Health Effects and Economic
Effects Workgroups considered risks from
ocean dumping of municipal waste, dredge
spoil, and discharges of industrial waste
into the ocean. The Ecological Effects
Workgroup considered risks from ocean
dumping of municipal waste and dredge
spoil. The Acute Events Workgroup
considered risks from ocean dumping of
municipal waste only. The Other Health
Effects Workgroup called this problem, area
"Ocean Disposal." The other workgroups
called this problem area "Ocean Dumping."
-------
TABLE 1
COMPARISON OF WORKGROUP RANKINGS OF OSWER PROBLEM AREAS (CONTNUED)
Workgroup Ranking
Acute
Events
Other
Health
Effects
Ecological Welfare Ground-Water
Effects Effects Valuation
Ocean Incineration
Notes on the Comparison Table
Letters reflect relative risk rankings judged for problem areas:
3t High or higher relative risk potential
Moderately high relative risk potential
Medkjm or moderate relative risk potential
Moderately tow relative risk potential
Low or tower relative risk potential
The workgroup did not consider this problem area .
•AL
Uncertainty is reflected in lettering as:
H Low uncertainty (high confidence)
${, Medum uncertainty (medium confidence)
H High uncertainty (tow confidence)
No attempt was made to "add" risks across Workgroups or compare the importance
of one Workgroup as more significant than another. Note that differences in
problem areas addressed is due to a combination of available data and scope.
-------
TABLE 2. ROPER SURVEY AREAS AND OSWER PROBLEM AREAS
Public
Perception of
Risk and Ranking
High
Roper
Survey Area
Chemical Waste
Disposal
Chemical Plant
Accidents
Combustion
Moderate
Lower
Oil Tanker
Spillage
Effects
Strip Mining
Corresponding Problem Area
(and Workgroup)
Land Disposal of Hazardous
Waste (all Workgroups)
Hazardous Waste Storage and Treatment
Tanks, Drums, and Containers (all but
Ground Water Valuation)
Other Treatment, Storage, and Disposal
Facilities (all)
Chemical and Industrial Manufacturing
Facilities (Acute Events and
Ecological Effects)
Accidental Releases at Chemical and
Industrial Manufacturing Facilities
(Other Health Effects and Welfare
Effects)
Air Pollution Combustion of Non-
Hazardous Waste including Non-
Hazardous Municipal Waste (all but
Ground Water Valuation)
Ocean Incineration (all but Ground
Water Valuation)
Oil Spill Prevention and Response
(Acute Events and Ecological Effects)
Oil Spill Response (Other Health
and Welfare Effects)
Mining Wastes (all but Ground Water
Valuation)
Mining Waste Sites (Ground Water
Valuation)
23
-------
CHAPTER 4. INFORMATION NEEDS
In general, the composition of data often guides analysis and affects its
results, just as data deficiency limits analysis. Also, the Workgroups reported
that they felt that limitations in the data precluded estimating the absolute
risks posed by problem areas.
4.1 GENERAL INFORMATION NEEDS
For this study, information needs have been prioritized by considering the
types of information needed to support program activities and the cost of
obtaining the information. , The rankings in Table 1 have been developed by
averaging the workgroup rankings. Thus, uncertainties associated with the
rankings are reflected in the text face but were not used to array problem
areas.
Table 1 suggests that OSWER has generally obtained higher confidence
information in the highest ranking problem areas. The rankings of problem areas
make it clear that there are significant information gaps exist.
In terms of future data needs, the Workgroups expressed low confidence in
rankings for: Mixed Wastes, Other Treatment, Storage, and Disposal Facilities,
Industrial Facilities, and Solid Waste Management Units. Thus, future work is
recommended to improve understanding of the risks these problem areas pose. In
addition, several workgroups expressed many data needs related to the individual
effects they had studied. The Ground Water Valuation Workgroup eliminated 13
problem areas from consideration since adequate .data were not available.
The relatively high risk rankings of problem areas that were ranked high
for reasons other than risks to human health suggest that protection of human
health may not be the most sensitive indicator of concern. This supports the
position that OSWER programs are taking in assuming that ecological and other
welfare impacts may be more sensitive to environmental stress than public
health. This assumption has resulted in shifts in program activities and
development of guidance materials for program assessment efforts.
As discussed later in this Chapter, OSWER has initiated efforts to collect
and refine information in a number of areas to assist in better understanding
risks to public health and the environment. Nevertheless, there are a number
of areas where additional information is important to the development of sound
strategic planning. For example, given that uncertainty is high for mixed
waste, it is possible that risk estimates associated with mixed wastes could
change significantly with additional information.
4.1.1 Program Recommendations
The following long-term recommendations for the continued support of
strategic planning are:
24
-------
• risk rankings be reviewed periodically based on information gathered
and
• a centralized repository should be established for the maintenance
of information used to perform the rankings.
4.1.2 Information Needs
In order to provide for a more consistent basis to support decisions on
program direction, it is recommended that options for improving information in
the following areas be developed:
• Information was especially limited and of low quality in the areas
supporting workgroup efforts for Ecological Effects, Welfare
Effects, and Ground Water Valuation. Consideration should be given
to enhanced data collection/methodological work in these areas.
• Accepting the finding that some pathways for exposure have not
consistently been assessed, it is recommended that efforts be made
to better assess major risks from all pathways.
• For four problem areas (i.e., Mixed Wastes, Other Treatment,
Storage, and Disposal Facilities, Industrial Facilities, and Solid
Waste Management Units), the Workgroups seem to have expressed
particularly low confidence in their findings. The extent of future
analyses might improve if the Agency acquired more data in these
areas.
• The Ground Water Valuation Workgroup eliminated 13 problem areas
from consideration, partly due to lack of information. The
Ecological Effects Workgroup did not consider Removal Actions at
Non-NPL Sites since insufficient data were available to estimate the
ecological effects of releases during removal actions. To improve
future studies, EPA may want to address the problem of obtaining new
data sources for these problem areas.
Finally, in order to improve upon the confidence of the findings of this
and future studies, sensitivity analyses of the use of alternative methodologies
and the corresponding impacts on final rankings would be useful.
4.1.3 Ongoing Information Collection Efforts
The need for much of this information was recognized even before this
effort was initiated; OSWER has embarked on efforts to fill some of these needs.
More specifically, information on chemical-specific concentrations in hazardous
waste streams is being addressed through the following information collection
efforts: the Treatment, Storage, Disposal and Recycling Survey; and the
Generator Survey. There has also been an increasing awareness of the need for
consideration of the risks associated with exposure to different media. This
issue is being addressed though improved guidance in the review of the Superfund
Public Health Evaluation Manual and the RCRA Facility Investigation Guidance.
25
-------
Moreover, OSWER is piloting an effort to determine the reasonableness of
combining information from different surveys that the Agency has collected
(including the air and water programs) in order to better characterize both
hazardous and non-hazardous waste streams through the OSWER Waste System Model.
The development of this model requires extensive coordination across different
Agency groups and should aid in presenting information in a consistent form.
Recognizing the importance of public health risk information in supporting
decisions for all OSWER programs, we are undertaking studies to better
understand metals movement in the environment, development of probabilistic risk
estimates (along with associated uncertainties), and improvement, of guidance for
the implementation of existing OSWER programs.
While the limitations of information that has been vised to support
economic estimates of environmental impacts may be most relevant to some
readers, they are also generally the product of more analyses and are subject
to great uncertainties. Decisions for the collection of such information are
likely to consider the need for such information to support regulatory
activities, the likelihood that such information can be made credible, and the
resource requirements associated with the collection of the information.
Finally, recognizing the costs associated with the cleanup of
radiologically contaminated waste (especially soils), OSWER programs are working
with other Agency programs and other Federal agencies to assess information on
technological methods for cleanup of existing sites and assessment of risks.
These efforts may lead to a unified Agency program to develop technologies for
the cleanup of radiologically contaminated soils as well as support for both
international and EPA decisions on ocean disposal of low level radioactive
wastes.
To better understand the relationship between public and Agency
assessments of risks, OSWER has initiated two studies to examine public
perception of risks associated with its programs. The first of these studies
is examining the relationship between different types of sites associated with
chemical risk (e.g., hazardous waste treatment facilities and Superfund sites)
and the influence of different socioeconomic variables on risk perception. The
second study is examining public perception of the risks associated with
alternatives for disposal of low-level radioactive wastes (including land and
ocean disposal options) is to undertake. One outcome of these; studies should
address the issue of how the Agency might improve on its approach to addressing
the public or different segments of the public when faced with a decision for
cleanup.
These studies of public perception, should provide insight, as to the types
of variables that affect the public thinking on issues related to OSWER
problems. In addition to the above studies, a variety of studies on risk
communication are designed to improve dialogue between the Agency and the
public. Taken together, these studies should provide a basis for improved
communication and support of cleanup options selected.
26
-------
4.2 ACUTE EVENTS
A key problem in applying the Acute Events comparative risk methodology
was that data for nearly every OSWER problem area were incomplete and not
focused on the event. Very little data were available on potential population
exposure and the frequency of acute events. Therefore, the accuracy of
parameter estimates varied greatly among problem areas.
Often the Acute Events Workgroup experienced difficulty in obtaining
information on the location of a facility. Also it was difficult to identify
which chemicals and what quantities were typically released during an acute
event. When possible, this information was obtained from data bases or case
study reports, but often no data were available. The Acute Events Workgroup
expressed concern over the absence of any discussion concerning illegal or
unpermitted dumping. Very little data are available in this area, and no study
methods have been developed.
4.3 OTHER HEALTH EFFECTS
For the Other Health Effects Workgroup, comprehensive risk modeling
studies rarely covered all of the potential exposure pathways through which
health risks might occur. The available exposure data were very limited and of
variable quality. There was also considerably less information on noncancer
risks than on cancer risks, and far fewer national modeling studies have been
conducted to assess noncancer risks. For most problem areas, the Workgroup
could only synthesize existing information on non-carcinogenic chemicals and
their potential effects. The availability of methods for the characterization
of risks as probabilities, fuller characterization of the severity of effects,
and characterization of the uncertainties is desired. Such information would
not only be useful in supporting more relevant decisions within the Agency but
should also improve our ability to communicate risks to the public.
4.4 ECOLOGICAL EFFECTS
Data availability for the Ecological Effects Workgroup was a major
limiting factor. For many problem areas, the scores for concentration and
receptor composites were based largely on professional judgment.
4.5 WELFARE EFFECTS
The Welfare Effects Workgroup's methodology relied heavily on its best
professional judgment. Future analyses may also want to consider aesthetic and
potential use values, where sufficient data were not available.
4.6 GROUND WATER VALUATION
Data deficiencies required the Ground Water Valuation Workgroup to make
assumptions about the environmental settings of various types of facilities or
27
-------
sites, hydrological parameters, exposed populations, types of waste handled,
facility size, and water-use patterns. Future comparative risk analyses may
want to consider the effects of ground water contamination prevention programs,
which were not considered in this project.
4.7 PUBLIC PERCEPTION OF RISK
It is clear from the differences in the rankings between public opinion
polls and the Agency's efforts to address environmental concerns that there is
considerable work to be done in improving communication with the public. As
part of the recently released A. Management Review of the Superfund Program, also
known as the "90 Day Study of the Superfund Program", the Administrator
identified a goal of "encourage full participation by communities." This and
other activities within the Agency signal an increased emphasis on efforts to
provide effective communication with the public.
28
-------
APPENDIX A
GLOSSARY
Term
Definition
acute event
aquifer
beneficiation
(mining)
bioconcentration
absolute risk An estimated level of risk intended as a representation of the
true risk (see relative risk).
The sudden, unplanned release of hazardous substance(s) that
pose a threat to public health, welfare, and/or the environment
and result in either acute or chronic exposures.
The permeable rock strata or sediment that is saturated with
ground water and may allow free movement of ground water.
The treatment of raw materials (i.e., iron ore) to improve
the material's inherent properties in preparation of smelting.
The process by which organic chemicals and certain metals
accumulate in tissues of exposed organisms. When the organisms
are consumed by predators, some of these pollutants can
increase in concentration in the predator.
Any waste, hazardous or nonhazardous, existing in large
quantities.
The potential or probability of contracting cancer due to
exposure to carcinogenic substances. Cancer risk often is
expressed as expected number of cases (population risk) or
excess lifetime probability per person (individual risk).
ecological impact The damage to natural components of an ecosystem resulting from
the contamination of environmental media, such as soil, surface
water, ground water, and air.
bulk waste
cancer risk
economic effect
The loss in value of materials, property, natural resources,
food supplies, or recreational resources due to damages caused
by environmental problem areas.
A-l
-------
GLOSSARY (Continued)
Term Definition
extraction The act or process of removing natural resources.
(mining)
ground water The contamination of subsurface waters that result in
damage unusable drinking water supplies and the need for an
alternative source of drinking water.
health effect The human injury, illness, or death that are the result of
exposure to hazardous substances.
low-level (radio- Radioactive wastes that consist of radionuclides, lab
active) waste wastes, or transuranic nuclides at concentrations of less than
100 nanocuries/gram.
municipal waste Non-hazardous waste, typically household waste and waste from
small commercial firms, regulated under RCRA Subtitle D.
microbial The transformation of substances in aquatic or soil
degradation environments by microscopic organisms.
mixed waste Wastes that contain both radioactive waste and RCRA hazardous
waste and are regulated by EPA as hazardous waste under RCRA.
National Priority EPA's list of the most serious uncontrolled or abandoned
List (NPL) hazardous waste sites identified for possible long-term
remedial response.
National Response The center responsible for collecting and maintaining reports
Center (NRG) of releases of hazardous substances into the environment.
Notification to the NRG marks the beginning of the response
process.
noncancer risk The potential or probability to incur a noncancer health effect
(e.g., lead poisoning, neurological disorders, liver disease)
due to exposure to noncarcinogenie hazardous substances. Often
indicated by comparing an estimated dose to a reference dose.
welfare effects Effects upon economic resources that affect the value of
commercial activities or in the value of human well-being.
A-2
-------
GLOSSARY (Continued)
Term
Definition
non-NPL site
photooxidation
reactivity
reference dose
(RfD)
relative risk
reportable
quantity (RQ)
RCRA Subtitle C
RCRA Subtitle D
RCRA Subtitle I
solid waste
transuranic
nuclides
An uncontrolled or abandoned hazardous waste site identified
for possible long-term remedial response, but not considered
a national priority.
The process by which materials or substances undergo oxidation
as a result of radiant energy (i.e., light).
The characteristic assigned to a material that is either
explosive, reacts violently with water, or generates toxic
gases when exposed to water or other liquids that are
moderately acidic or alkaline.
The dose of a substance above which adverse health
effects may occur.
An estimated level of risk intended only for comparison to
other relative risk estimates derived using the same
methodology (see absolute risk).
A threshold quantity set by EPA for certain hazardous
substances. RQs serve as triggers for notification when a
release of that hazardous substance occurs.
Regulates any material that may be characterized as hazardous
waste pursuant to 40 CFR Parts 260-272.
Regulates any material that may be characterized as solid, non-
hazardous waste pursuant to 40 CFR Parts 240-257.
Regulates petroleum products and CERCLA hazardous substances
that is contained in an underground storage tank.
Any material that is discarded, to include solids, semi-solids,
liquids, or gases.
Radioactive materials that contain alpha-emitting nuclides
with an atomic number greater than 92.
A-3
-------
-------
APPENDIX B
DEFINITIONS OF OSWER PROBLEM AREAS
B-l
-------
TABLE B-l
OSWER PROBLEM AREAS
Problem Areas
Definition
Combustion of Hazardous Waste
Hazardous Waste Storage and Treatment
Tanks, Drums, and Containers
(Subtitle C)
Land Disposal of Hazardous Waste
Land Treatment of Hazardous Waste
Industrial Facilities
Mining Wastes
Combustion of waste material that is
hazardous, including waste burned for
energy recovery (e.g., industrial
furnaces and industrial and utility
boilers) (40 CFR Part 266, Subpart
D), hazardous waste incinerators (40
CFR Part 264, Subpart 0 and Part 265
Subpart 0), and open detonation.
Hazardous waste storage and treatment
tanks, drums and containers, which
are found both at private hazardous
waste management companies and at
individual industrial sites are
regulated under Subtitle C of RCRA.
These land sites are hazardous waste
landfills, surface impoundments, and
waste piles under RCRA Subtitle C (40
CFR Parts 264, Subparts M and N).
The use of processes (e.g.,
photooxidation and microbial
degradation) to reduce the toxicity
or quantity of waste deposited at a
site. These land treatment sites are
subject to RCRA requirements in 40
CFR Sections 264.270-.317.
Nonhazardous waste landfills, surface
impoundments, land treatment, and
incineration (not including municipal
incineration) regulated under
Subtitle D of RCRA (surface
impoundments are regulated under 40
CFR Parts 264 and 267, Subpart K and
Part 267 Subpart D).
Wastes include solid waste from the
extraction, beneficiation, and
processing of ores and minerals.
B-2
-------
TABLE B-l (continued)
Problem Areas
Definition
Mixed Waste
Municipal Landfills
Municipal Waste Combustion
Oil and Gas Operations Waste
Other Treatment, Storage
or Disposal Facilities
Mixed waste consist of radioactive
wastes that are mixed with a RCRA
hazardous waste component and are
regulated by EPA as hazardous waste
under RCRA. For the most part, these
are low-level wastes consisting of
naturally-occurring radionuclides and
transuranic nuclides at
concentrations less than 100
nanocuries/gram.
Landfills used to dispose of
household waste and waste from small
commercial firms. These sites are
regulated under Subtitle D of RCRA.
This area includes the incineration
of municipal waste, which is not
hazardous.
Oil and gas operations generate
drilling fluids, produced waters, and
other wastes associated with the
exploration, development, or
production of crude oil, natural gas,
or geothermal energy as described in
RCRA section 3001(b)(2)(A). (Oil
spills into navigable waters of the
U.S are covered by section 311 of the
Clean Water Act.)
Such facilities are involved in waste
treatment, storage, and disposal not
included in those identified above.
These include: waste recycling
facilities, waste transfer stations,
container storage areas, or any
subject to systematic and routine
release.
B-3
-------
TABLE B-l (continued)
Problem Areas
Definition
Solid Waste Management Units
Oil Spill Response
Removal Activities
Remedial Activities
SWMUs are on-site land disposal units
associated with permitted RCRA
facilities. These classes of
facilities are subject to
requirements as established by
Section 3004U and 3008H of RCRA.
Activities linked to establishing
procedures and requirements for
preventing oil discharges from
vessels and offshore and onshore
facilities for containing such
discharges, and for responding to
discharges or threats of discharges
to waters of the U.S., adjoining
shorelines, waters of the contiguous
zone, and waters of the high seas
under U.S. jurisdiction CWA §311.
Activities related to Superfund
removal actions taken pursuant to
CERCLA §104 (40 CFR Part 300) and CWA
§311. Removal actions will include
emergency response actions to
transportation accidents involving
releases of hazardous substances.
Activities related to Superfund
remedial actions taken pursuant to
CERCLA §104 and 40 CFR Part 300.
Remedial actions will include RI/FS
activities and cleanup at NPL and
non-NPL sites.
B-4
-------
TABLE B-l (continued)
Problem Areas
Definition
Chemical Industrial Manufacturing
Facilities
Ocean Dumping
Ocean Incineration
FOB Wastes
Chemical industrial manufacturing
facilities that process/manufacture
chemicals, especially those types of
chemicals that manufacture, process,
or store types of chemicals that
could pose a threat in terms of an
accidental release that could pose a
public health threat. (Focus on
accidental releases that would be
addressed under SARA Title III.)
Ocean dumping involves the disposal
of bulk waste (usually untreated) at
sea (40 CFR 227). Ocean dumping is
only allowed for bulk nonhazardous
wastes as no additional treatment of
the waste is permitted.
Ocean incineration involves the
incineration of waste materials at
sea. Given that future risks
associated with ocean incineration
are highly uncertain but appear to be
small (ocean incineration may not be
chosen as a popular mechanism to deal
with hazardous waste), the principle
reason for including this source
category is that ocean incineration
affects the capacity question.
Polychlorinated biphenyls (PCBs) and
PCB wastes primarily found in
electrical transformers and
capacitors. They are regulated under
the Toxic Substances Control Act 40
CFR Part 761.
B-5
-------
TABLE B-l (continued)
Problem Areas
Definition
Transportation
Trucks, railroads and barges used to
transport hazardous substances.
Most hazardous substances are
transported by truck and are
regulated by the Department of
Transportation.
Underground Injection
Underground injection is defined as
disposal of liquid waste material
into isolated geologic strata,
placing the wastes in portions of the
earth's crust that are free from the
usual effects of the hydrologic cycle
(regulated under 40 CFR Part 267,
Subpart G and Part 146).
Exempt Storage Tanks
Heating oil, farm, and residential
tanks that are exempted from
regulation under Subtitle I.
Hazardous Substance Storage Tanks
(RCRA Subtitle I)
Underground storage of substances
designated as hazardous (under
CERCLA).
Petroleum Storage Tanks (RCRA
Subtitle I)
Tanks that store petroleum products
in underground storage tanks such as
gasoline stations (includes used
oil).
B-6
-------
APPENDIX C
INDIVIDUAL WORKGROUP SUMMARIES
This appendix addresses the scope of effort, methodology, data sources and
uncertainty levels, and results of each of the five Workgroups (i.e., Acute
Events, Other Health Effects, Ecological Impacts, Welfare Effects, and Ground
Water Valuation). A short discussion on caveats and other issues is also
provided for each Workgroup.
C.I ACUTE EVENTS
Scope of Effort. The Acute Events Workgroup assessed and compared risks
arising from acute events in 24 OSWER problem areas. Acute events were defined
as sudden, unplanned releases of hazardous substances that pose a threat to
public health, welfare, and the environment. These events represent a
qualitatively different kind of threat to human health and the environment as
compared to long-term or chronic environmental contamination. Acute events have
the potential to cause immediate injury, long-term health problems, significant
environmental contamination, and extensive property damage.
Methodology. The methodology used (see Figure C-l) was based on the
assumption that the overall relative risk a problem area poses is comprised of
four different effects:
(1) effects associated with acute exposures;
(2) effects associated with chronic exposures;
(3) welfare effects; and
(4) ecological effects.
For each effect type, relative risks were assessed by combining the
observed frequency of occurrence of an acute event with its inherent hazard
(based on the effect type) and the size of the population potentially exposed.
A series of indices were developed to represent exposure, frequency, and severity
(inherent hazard). The three indices were summed for each effect type to get
an effect type score. Then, the four effect type scores were summed to get an
overall acute events score for a problem area.
Frequency indices are based on historical records of the annual occurrence
of acute events. Exposure indices are derived from 1980 Census-based estimates
of the number of people within one mile of a facility at which acute events may
occur. Severity indices were estimated as a function of the quantity of a
chemical released and the primary criteria reportable quantity (RQ) for that
chemical (i.e., an indicator of inherent toxicity). The RQ provides the minimum
quantity of a chemical spilled or released at which a report must be made to the
National Response Center and is based on five primary criteria: mammalian
toxicity, chronic toxicity, carcinogenic toxicity, aquatic toxicity, and
ignitability or reactivity. The severity indices, therefore, are based on the
typical quantity released during an acute event, and the inherent toxicity
(represented by the RQ) of typical releases in each OSWER problem area.
C-l
-------
Data Sources and Uncertainty Levels. The Acute Events Workgroup used
various data sources to estimate index scores. A key problem in applying the
acute events comparative risk methodology, however, was that data for nearly
every OSWER problem area was incomplete and not focused on acute events.
Therefore, the data sources and the procedures used to calculate parameter
estimates varied from problem area to problem area. Where data proved to be
unobtainable or nonexistent, the Workgroup primarily relied on the Acute Events
Survey. This survey was conducted specifically for the OSWER Comparative Risk
Project on EPA personnel in program offices responsible for regulating or
monitoring facilities in individual OSWER problem areas.
Population exposure estimates were obtained primarily from two sources:
U.S. EPA Graphic Exposure Modeling System (GEMS) and U.S. Bureau of the Census,
1980 County Population estimates. If a specific facility or source location was
known, then exposure estimates were developed from GEMS. If only the general
distribution of facility or source locations were known, then exposure estimates
based on U.S. Bureau of the Census data were developed to estimate the potential
population exposure.
Frequency estimates were obtained, when possible, from either case studies
(such as Emergency Response Division Weekly Pollution Reports) or data base
estimates (such as the Department of Transportation's Hazardous Material
Transport Incident Reporting System). When case studies were used, it was
assumed that they accurately reflected the number of events that occurred in each
problem area. Frequency estimates derived from case study data were all
extrapolated to one year. In most cases, this was done by either dividing by
the total number of years reported in the data or by extrapolating from monthly
data to one year. Annual frequency estimates for some problem areas were derived
from existing data bases. In each case, annual frequency estimates were based
on the mean multi-year data.
Severity estimates, such as the Human Injuries Acute Effects Index, were
also obtained, when possible, from either case study reports (such as EPA's
Summary of State Reports on Releases from Underground Storage Tanks') or data
bases (such as the Acute Health Effects Data Base). However, there was less
information on human injuries and deaths associated with acute events than for
any other parameter. In cases where no information was available, no attempt
was made to estimate this parameter and the Human Injuries Acute Health Effects
Index was dropped from consideration altogether.
A key part of the methodology was the identification of chemicals and the
quantities of those chemicals that were typically released during an acute event.
When possible, this information was obtained from data bases or case study
reports. However, estimates from the Acute Events Survey were used more widely
in this part than in any other.
Results. The efforts of the acute events workgroup culminated in a set
of rankings that reflects the relative risks from acute events for the twenty-
four OSWER problem areas (Table C-l) . The problem areas are divided into three
relative risk groups on the basis of their overall rank. The high risk group
includes the OSWER problem areas that received scores of 45 to 54; the medium
C-2
-------
risk group includes OSWER problem areas that received scores of 36 to 43; and
tbe low risk group includes those OSWER problem areas that received scores of
25 to 34.
Discussion. Several assumptions were made in determining the relative
risks of acute events associated with the 24 OSWER problem areas. Some typical
assumptions made by the Workgroup include the following.
• An acute event observed in the past was used to predict
an acute event in the future.
• Annual numbers of events are assumed to be constant.
• The frequency of acute events occurring in one area was
extrapolated to determine the frequency of acute events
for the nation as a whole.
• The location of sensitive ecosystems (as reflected in
the exposure index for ecological risks) was assumed to
be more likely in areas with low human population
density.
• The potential loss in property value (as reflected in
the welfare effects severity index) was assumed to be
associated with the ignitability of the substance
released.
Data for nearly every OSWER problem area was incomplete in the area of
acute events. As such, the workgroup made a series of assumptions about missing
data. Therefore, caution should be exercised in interpreting the study
findings.
C-3
-------
FIGURE C-1
ACUTE EVENTS METHODOLOGY
n
i
Acute Risk Parameters
A Population Exposure
B Frequency
C Severity (Death and Injury
Score or Mammalian Toxicity
Score. Whichever is Higher)
Acute Health
Effect Composite
Chronic Risk Parameters
A Population Exposure
B Frequency
C Severity (Release and
Chronic Toxicity Score)
Chronic Health
Effect Composite
Overal
Risk Score
for
Problem Area
Welfare Risk Parameters
A Property Exposure
(Proportional to Population
Exposure)
B Frequency
C Severity (Release and
IgnitabSty /Reactivity Score)
Welfare Effect
Composite
Ecological Risk Parameters
A Ecosystem Exposure
(Inversely Proportional to
Population Exposure)
B Frequency
C Severity (Release and
Aquatic Toxicity Score)
Sum
Ecological Effect
Composite
-------
TABLE C-l
ACUTE EVENTS RANKINGS
PROGRAM AREA/SOURCE
Acute Chronic Total
Exposure Exposure Welfare Ecological Score
HIGH
Transportation of Hazardous Materials
Oil & Gas Operations Waste
Oil Spill Response
Accidental Releases at Chemical Facilities
Petroleum Underground Storage Tanks
Industrial Non-Haz. Waste Facilities
16
12
14
14
12
12
14
13
13
12
12
11
13
13
13
11
12
12
11
13
11
11
10
10
54
51
51
48
46
45
MEDIUM
Solid Waste Management Units 11 11 11 10 43
Hazardous waste storage and treatment tanks 10 11 10 10 41
Land disposal of hazardous waste 10 9 10 12 41
Other TSDFs 9 11 10 10 40
Removal Activities/Other Emergency Response 10 11 10 9 40
UST Hazardous Substance Tanks 11 12 9 7 39
Exempt Storage Tanks 9 11 9 10 39
PCS Wastes 9 10 9 9 37
Underground Injection of Hazardous Waste 10 9 9 9 37
Hazardous Waste Combustion 8 10 9 9 36
LOW
Ocean Dumping
Remedial Actions at NPL Sites
Municipal Waste Combustion
Land Treatment of Hazardous Waste
Mining Waste
Ocean Incineration
Mixed Waste
Municipal Landfills
8
8
8
7
8
7
7
5
7
9
9
9
8
8
7
7
8
8
7
6
6
7
7
5
11
9
8
8
7
7
5
8
34
34
32
30
29
29
26
25
C-5
-------
C.2 OTHER HEALTH EFFECTS
Scope of Effort. The Other Health Effects Workgroup addressed both noncancer
and cancer risks associated with exposures that were not addressed in acute
events. This constitutes the majority of the types of risks that are addressed
by OSWER program offices in the support of various regulations. Twenty-two
problem areas were addressed.
Methodology. The Other Health Effects Workgroup (see Figure C-2) sought to
(1) build upon readily available information and (2) draw from the experience
of Unfinished Business. The Workgroup examined both cancer and noncancer risks.
The analysis of noncancer risks was limited to those risks not addressed in the
Acute Events Workgroup. The approach to developing a ranking of problem areas
involved (1) collecting information for each of the problem areas, (2) assembling
that information to support the rankings, (3) developing an algorithm for the
assignment of points associated with risk factors for problem areas, and (4)
convening a meeting to exercise the methodology. The development of the
algorithm involved considerable professional judgment on technical risk and value
judgments on the relative importance of differing types of risks (e.g., cancer
risks and noncancer risks were both used to develop one score).
Data Sources and Uncertainty Levels. Comprehensive risk modeling studies were
not available for many of the problem areas. For areas with available studies,
all of the potential exposure pathways through which health risks might occur
were rarely covered. Therefore, after all available comprehensive risk
information was reviewed, additional data for the problem areas on the individual
parameters affecting risk were collected. These data included types of wastes
and waste management units, chemicals or constituents of concern, types and
numbers of releases, exposure pathways, concentrations of constituents at
exposure points, potentially exposed populations, and numbers and distributions
of sources.
The percent of the potential cancer risks covered by the models and the
uncertainties associated with the modeling effort were assessed. For some
problem areas, several sources of risk estimates provided conflicting results.
Potential reasons for these conflicts were identified by analyzing the modeling
approaches used. The estimates obtained with the model having the greatest
technical accuracy were adopted by the Workgroup.
Generally, there was considerably less information on noncancer effects than
on cancer risks, and far fewer of these effects have been studied through
national modeling efforts. Where data were available on individual chemicals
for which noncarcinogenic health risks have been associated, a method similar
to that used in Unfinished Business was used. Sufficient information was
available to implement this scoring approach for very few problem areas, however.
Therefore, for most problem areas, the existing information on chemicals
associated with noncarcinogenic health risks was summarized, and a qualitative
assessment of the likelihood that the problem areas would result in exposures
exceeding RfDs was based on the findings of previous investigators and
professional judgment.
C-6
-------
Results. The results of Workgroup deliberations for this effort are presented
in Table C-2. As explained in the Workgroup report, the extent and quality of
data had little influence on the relative ranking of problem areas. Moreover,
a comparison of the rankings of OSWER rankings with those of the Unfinished
Business report suggest that there are at least some similarities among the
reports.
The results of the Other Health Effects Workgroup placed Superfund Activities,
Industrial Nonhazardous Waste Management Facilities, Solid Waste Management
Units, and Hazardous Waste Storage and Treatment Tanks, Drums, and Containers
at the top of the ranked list. In contrast to the Acute Events Workgroup, Oil
Spill Response was among the lower risk problem areas. It is worthwhile noting
that Solid Waste Management Units received a high ranking in this Workgroup but
there was a lack of data to support this ranking at the time that the rankings
were developed. Recently, however, the Regulatory Impact Analysis for the
Corrective Action rule has been released, which generally supports the rankings
assigned by this workgroup.
Discussion. A number of assumptions were made or limitations existed for the
assessment of other health effects.
• Hazard (e.g., toxicity, potency) information is not available for most
substances.
• Where hazard information exists, it is based largely on animal data at
high doses, resulting in uncertainty when extrapolation to humans and
low doses occurs.
• Information on interactive effects (e.g., synergism, antagonism) is not
available for most substances.
• Data limitations and general nonacceptance of a quantitative noncancer
risk assessment methodology led the Workgroup to employ a highly
qualitative noncancer methodology.
• When modeling was conducted for a representative facility or site for
a problem area, it was often difficult to then scale the results up to
the entire problem area.
• Assumptions regarding corrective action or averting action vary widely.
• Assumptions regarding exposed populations vary widely.
• Data on concentrations of contaminants actually monitored at exposure
points did not exist or was not used for the most part.
C-7
-------
FIGURE C-2
OTHER HEALTH EFFECTS METHODOLOGY
n
i
co
Workgroup
Develops
Population Risk
Consensus Score
Each Workgroup
Member
Assessment of
Cancer Risks
A Release
B Transport
C Exposure
D Potency
Assigns
Population Risk
Score
Cancer
Risk
Score
hdrvidual Work-
group Member
Assigns Maximum
Exposed Individual
(ME!) Risk Score
Workgroup
Develops
MB Risk
Consensus
Score
Overall
Problem Area
Score
Assessment of
Noncancer Risks
(More Quaitative
than Cancer Risks)
A Release
B Transport
C Exposure
D Toxicity
Workgroup
Develops
Noncancer
Risk Consensus
Score
Each Workgroup
Member
Assigns
Noncancer Risk
Score
Noncancer
Risk Score
-------
TABLE C-2
RANKINGS FROM THE OTHER HEALTH EFFECTS WORKGROUP
Cancer Risk
PROGRAM AREA/SOURCE Population MEI
Superfund Remedial and Removal Activities
Industrial Non-Hazardous Waste Facilities
Hazardous Waste Storage and Treatment Tanks
Land Disposal of Hazardous Waste
Mining Waste
Hazardous Waste Combustion
Municipal Landfills
Municipal Waste Combustion
Land Treatment of Hazardous Waste
Oil & Gas Operations Waste
UST Hazardous Substance Tanks
Other TSDFs
Petroleum Underground Storage Tanks
Exempt Storage Tanks
Mixed Waste
Oil Spill Response
Accidental Releases at Chemical Facilities
PCB Wastes
Ocean Dumping
Ocean Incineration
Transportation of Hazardousaterials
Underground Injection of Hazardous Waste
4
4
2
2
1
3
2
3
2
2
1
2
1
1
1
1
1
1
1
1
1
1
5
5
5
4
5
3
5
4
3
3
4
2
3
3
2
2
2
1
1
1
1
1
Non- Cancer
Risk
4
3
3
3
3
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Score
13
12
10
9
9
8
8
8
6
6
6
5
5
5
4
4
4
3
3
3
3
3
C-9
-------
C.3 ECOLOGICAL IMPACTS
Scope of Effort. The Ecological Impacts Workgroup addressed the relative
ecological risk potential and ecosystem impacts associated with 23 of the OSWER
problem areas. Because the OSWER worksheets providing the requisite data were
riot available, no evaluation of removal actions at non-NPL sites was attempted.
The Workgroup used a semi-quantitative risk characterization approach that relied
both on available data and professional judgment.
Methodology. For each problem area, the Ecological Impacts Workgroup
characterized the actual impacts and potential risks associated with each problem
area (see Figure C-3). These data elements were grouped into the following four
data composites: (1) number of sources and releases in a problem area; (2)
contaminant concentration; and (3) receptors.
For each data element, direct information or estimates from the
appropriate OSWER Comparative Risk Project work sheet were recorded with the
resulting quantitative score. Where there was a range of values or the available
data were limited, a best estimate was made. Where data were lacking,
professional judgment was used, if appropriate. Otherwise the category was left
blank. When available, a list of the most common constituents associated with
each problem area was provided. For each problem area, data elements within the
toxicity composite were scored according to the specific contaminant associated
with that problem area that resulted in the maximum score for that data element
(i.e., the bioconcentration score was obtained from the constituent with the
highest bioconcentration factor, and the aquatic toxicity score was obtained from
the constituent that was most toxic to freshwater organisms).
The score for each composite was obtained by taking the average score for
the data elements within that composite for which data were available (missing
values did not affect the score). For each problem area, an indication of the
availability of data also was provided.
Data Sources and Uncertainty Levels. For most ecological problem
areas,the sources/releases composite and the toxicity composite reflected high
levels of quantitative information. The concentration composite scores and
receptor scores were generally based on more subjective information.
Information for each of the respective parameters for the sources/releases
composite was compiled from the appropriate worksheet for each problem area, and,
where necessary, from other information sources. For each of the six ecological
data categories, the direct information from the OSWER worksheet was recorded
with the appropriate quantitative rankings. Where there was a range of values
or the available data were limited, a best estimate of the midpoint for the range
was made for the category. Where data were lacking, professional judgement was
used, if appropriate.
Of the four catagories of toxicity values (i.e., toxicities to freshwater,
marine, and terrestrial organisms, and bioconcentration), toxicity to terrestrial
organisms was the most available. Toxicity values for mice were used when
available and toxicity to rabbits was used secondarily. In compiling freshwater
and marine toxicity values, a hierarchy of data was created to use the best
C-10
-------
validated data when EPA ambient chronic water quality criteria values were not
available. In general, the data for freshwater organisms was more abundant than
data for marine organisms. Bioconcentration values were less available than
other toxicity values.
Results. The results of the ranking procedure are shown in Table C-3.
In general, the final ranking did not depend on the availability of data,
although several problem areas in the moderate risk potential category (problem
areas 3, 7, 8, 13, and 20) might have received a higher relative ranking if more
data were available. The final rankings were dependent largely on the scores
for two data composites: (1) toxicity; and (2) number of sources and releases.
A ranking based on the product of these two scores (not shown) correlated well
with the ranking shown in Table C-3 (r = 0.89, p < 0.01), and the groupings of
problem areas into high, medium, and low risk categories was largely identical.
As described earlier in Section 3.2.2, the Workgroup employed a broader
definition for ocean dumping and showed a striking difference (higher risks) as
compared with most workgroups.
Discussion. The ideal approach to a ranking procedure of this nature
would be to use a quantitative risk assessment methodology to characterize risks
to certain species or ecosystems. In this project, such an approach was limited
by theoretical, data, and resource limitations. Because of these limitations,
a qualitative risk characterization approach that relied heavily on professional
judgment was used. Data availability was a major limiting factor. For many
problem areas, the score for several data elements (particularly those within
the concentration and receptor composites) were based largely on professional
j udgment.
C-ll
-------
FIGURE C-3
ECOLOGICAL EFFECTS METHODOLOGY
Sources/Releases
Categories
Number of Sources
Types of Sources
C Geographic Density/
Distribution
Frequency of Releases
Duration of Releases
Area of Sites
Sources/Releases
Composite
Concentration
Categories
Magnitude of Releases
Size of Area Affected
Compartments
Receiving Releases
Fate of Contaminants
Concentrations at or
Distance to Receptor
Concentration
Composite
Receptor Categories
Receptor
Composite
Compartments Con —
taininy Affected
Organisms
Exposed Populations
Size of Affected
Populations
Sum
Bio-concentration
Category:
Highest Value for
All Contaminants
Present at
Problem Area
Toxicity to
Freshwater
Organisms
Category:
Highest Value
Combined
Exposure
Factor
Toxicity
Composite
Toxicity to Manr
_| Organisms
Category:
Highest Value
Toxicity to
Terrestrial
Organisms
Category:
Highest Value
Lcological Risk
Characterization Score
-------
TABLE C-3
ECOLOGICAL IMPACTS
PROGRAM AREA/SOURCE SCORE
Ocean Dumping 53.4
Remedial Actions at NPL Sites 53.1
Land Disposal of Hazardous Waste 52.0
Municipal Landfills 50.0
PCB Wastes 43.4
Mining Waste 42.5
Municipal Waste Combustion 40.2
Mixed Waste 38.3
Hazardous Waste Storage and Treatment Tanks 36.2
Hazardous Waste Combustion 35.8
Oil and Gas Operations 33.2
Land Treatment of Hazardous Waste 31.1
Other TSDFs 30.7
Industrial Non-Hazardous Waste Facilities 30.6
UST Hazardous Substance Tanks 30.6
Ocean Incineration 28.7
Petroleum Storage Tanks 27.8
Underground Injection of Hazardous Waste 24.8
Oil Spill Response 24.7
Solid Waste Management Units 23.0
Transportation of Hazardous Materials 22.2
Exempt Underground Storage Tanks 21.4
Accidental Releases at Chemical Facilities 19.8
C-13
-------
C.4 WELFARE EFFECTS
Scope of Effort. The Welare Effects workgroup exam5_ned welfare effects
(which were defined as damages to economic resources that result in a reduction
in the value of commercial activities or in the value of human well-being) . This
approach was chosen because of the lack of studies associating economic impacts
with OSWER problem areas. While the supporting workgroup report is titled
"Report of the Economic Effects Workgroup", the workgroup recommended that the
welfare effects be used for purposes of the overview report.The objective of the
Welfare Effects Workgroup was to rank OSWER problem areas according to their
potential relative negative welfare impacts. Negative welfare damages were
defined as the loss in value placed upon an economic resource when environmental
contamination associated with releases from OSWER problem areas occurs.
Data on the dollar value of welfare damages for each problem area were often
unavailable. Consequently, the Workgroup did not attempt to estimate the.
absolute dollar value of welfare losses for each problem area. As such, the
Workgroup decided only to estimate relative welfare damages associated with
different problem areas. The limitations available information also required
that the Workgroup had to rely extensively on best professional judgment when
assessing the relative damages. Furthermore, the absence of rigorous,
quantitative data meant that the Workgroup was unable to estimate damages with
a high level of precision and instead used ordinal scales (i.e., high, medium,
and low) to assess welfare damages.
Methodology. The Welfare Effects Workgroup's methodology relied heavily on
the Workgroup's own best professional judgment (see Figure C-4). The Workgroup
considered welfare effects for each problem area for three general categories
of resources:
« residential and commercial property;
• natural resources; and
• surface water.
Damage to residential or commercial property was defined as material or
structural damages to either residential or commercial properties and loss of
commercial land for business activities. Damage to natural resources was defined
as damages to agricultural crops, livestock, timber resources, and fisheries.
Damage to surface water was defined as damages to surface waters used for
drinking purposes or for recreation. For each problem area, the Workgroup
assigned a Welfare Damage Score for each type of damage based on their estimate
of the degree of welfare damage likely to be associated with releases in that
problem area. Thus, for each problem area, the Workgroup assigned a Welfare
Damage Score for damages to residential property, commercial property, timber
resources, crops, livestock, fisheries, recreational use of surface water, and
surface water used as drinking water. The Workgroup summed all eight scores to
devise an index of total welfare damage associated with a problem area (overall
scoring range 8 to 80).
The Welfare Damage Scores assigned to each type of damage ranged between 1
and 10. A score of 1 was assigned when welfare damages were estimated to be
relative low, a score of 5 was assigned when welfare damages were estimated to
C-14
-------
be moderate, and a score of 10 was assigned when welfare damages were estimated
to be high.
In assigning Welfare Damages Scores, the Workgroup agreed on a series of
assumptions or guidelines to estimate the magnitude of welfare damages. The
Workgroup assumed that welfare damages would be greater when one or more of the
following was true:
• a large volume of waste is released;
• substances released have a high toxicity;
• the concentration of constituents of concern in substances
released is high;
• the degree of regulation in a problem area is relatively low;
• the number of sites or facilities in a problem area is
relatively high;
• population density surrounding typical sites or facilities
is relatively high; and
• there is a high density of natural resources close to the
problem area.
Whenever possible, quantitative data were used as the basis for estimates
of welfare damages. When no quantitative data were available, the Workgroup
used its best professional judgment to estimate values for these parameters.
Data Sources and Uncertainty Levels. The key data source relied on to
develop welfare damage estimates were problem area fact sheets prepared by the
Office of Program Management and Technology. The fact sheets contained data on
most parameters considered in the analysis. When data were not available from
the fact sheets, attempts were made to identify other data sources. If no other
data could be identified, the Workgroup relied on its own best professional
judgment to estimate welfare damages.
Results. The final ranking of the Welfare Effects Workgroup is presented
in Table C-4. The Workgroup divided the problem areas into three groups.
(1) High: problem areas with total welfare damage scores of 25
or greater.
(2) Medium: problem areas scoring between 15 and 25.
(3) Low: problem areas scoring below 15.
The Workgroup concluded that comparisons beyond this level of detail were
not justified considering the quality of data supporting individual scores. At
this level of comparison, however, the Workgroup expressed reasonable confidence
in its conclusions.
C-15
-------
Discussion. The Welfare Effects Workgroup chose not to consider four other
potential welfare effects:
• aesthetic values;
• potential use values;
• secondary or indirect effects; and
• economic effects arising from ground water contamination.
Aesthetic and potential use values were not considered because the
Workgroup judged that sufficient data were not available to realistically assess
these effects. Secondary or indirect effects were not considered because the
Economic Effects Workgroup wished to avoid overlap with other Workgroups. For
example, the Welfare Effects Workgroup did not consider costs from medical
expenses of exposed individuals because it was felt that this should be dealt
with by the Other Health Effects Workgroup. Similarly, it was felt that
potential welfare loss resulting from ecological damages should be addressed by
the Ecological Effects Workgroup. By the same rationale, welfare effects
arising from ground water contamination was not considered as this was to be
addressed by the Ground Water Valuation Workgroup.
C-16
-------
FIGURE C-4
ECONOMIC EFFECTS METHODOLOGY
o
I
Property Damage
Composite
Property Damage
Category
A Commercial
B Residential
Natural Resource
Damage Category
A Timber
B Crops
C Livestock
D Rsheries
Surface Water
Damage Category
A Drinking Water
B Recreational
Water
Sum
Natural Resource
Damage Composite
Sum
Surface Water
Damage Composite
Sum
Overal
Welfare Damage
Score
-------
TABLE C-4
WELFARE EFFECTS WORKGROUP RESULTS
o
1
;— »
co
PROGRAM AREA/SOURCE
Superfund Activities
Solid Waste Mgmt Units (SUMUs)
Land Disposal of Hazardous Waste
Industrial Non-Hazardous Waste Facilities
Petroleum USTs (Including Used Oil)
Municipal Landfills
Haz. Waste Storage & Treatment Tanks
Mining Wastes
Exempt Storage Tanks
Oil & Gas Operations Waste
Other TSOFs
Municipal Waste Combustion
Combustion of Hazardous Waste
Transportation
Hazardous Substance USTs
Land Treatment of Hazardous Waste
Oi 1 Spi 1 1 Response
Underground Injection
Ocean Dumping
PCS Wastes
Accidental Releases 3 Chem. Manufacturing Fac.
Mixed Waste
Ocean Incineration
Surface Water
Recreational Drinking
Water
10 10
10 5
5 5
5 5
5 5
5 5
5 5
5 1
5
1
1
1
1
1
1
1 1
1 1
0 1
1 0
0 0
0 0
0 0
0 0
Natural
Resources
Timber Crops Livestock Fisheries
15 1 5
15 5 5
15 5 5
11 1 1
01 1 1
11 1 1
11 1 1
55 1 1
01 0 1
15 5 1
11 1 1
11 1 1
11 1 1
11 1 1
01 1 1
10 0 1
00 0 1
00 0 0
00 0 1
00 0 1
00 0 0
00 0 0
00 0 0
Property Overall
Commercial Residential Score
55 42
55 41
55 36
10 10 34
10 10 33
10 5 29
10 5 29
1 1 20
55 18
1 1 16
55 16
1 5 12
1 1 8
1 1 8
1 1 7
1 1 6
00 2
0 0 1
00 2
0 0 1
00 0
00 0
00 0
Scale:
0 = No, or insignificant, level of economic impacts
1 = Relatively Ion level of economic impact
5 = An Average (or medium) level of economic impacts
10 = Relatively high level of economic impacts
-------
G.5 GROUND WATER VALUATION
Scope of Effort. The objectives of the Ground Water Valuation Workgroup
were to assess and compare the extent to which OSWER problem areas contribute
to ground water resources damage. The Workgroup analyzed nine problem areas.
Nine other problem areas considered by other OSWER Comparative Risk Workgroups
were not analyzed since either presented only a minimal threat to ground water
or because no reliable data were available on the potential of these problem
areas to contribute to ground water contamination.
Methodology. The methodology used by the Ground Water Valuation Workgroup
was based on a methodology developed by the Office of Solid Waste's (OSW)
Economic Analysis Staff for quantifying resource damage associated with ground
water contamination (see Figure C-5). This methodology is based on the concept
of resource damage and replacement costs. In this approach it is assumed that
ground water used as drinking water is generally a free resource. When such a
resource becomes unusable due to contamination, there is a cost associated with
its replacement. This cost is taken to represent the "value" users place on a
contaminated resource, or how much they would be willing to pay for the least
expensive substitute resource.
For each problem area, the Ground Water Valuation Workgroup developed a
series of scenarios representing the range of potential contamination likely to
occur from releases in that problem area. Each scenario was based on different
runs of a series of problem area-specific ground water contamination models.
These models simulated pollutant release from a facility or waste management unit
and predicted concentrations of pollutants at various distances downgradient from
sources of contamination. The models also predicted plume size, duration of
contamination, and assumed well density. Based on the results of these model
runs, the Ground Water Valuation Workgroup then developed engineering
specifications for replacement water supply (based on a new system's design
capacity, available water sources, and period of operation) for each scenario.
A series of engineering cost algorithms were then applied to predict capital and
operating costs for the replacement supply. These costs (expressed in present
value terms) were assumed to be an approximation of the resource value of the
ground water under that scenario. Aggregate national costs were then calculated
by extrapolating estimated costs to the universe of facilities or sites in a
problem area based on the estimated distribution of scenarios in each problem
area.
Data Sources and Uncertainty Levels. No single data source was used to base
model estimates on. Generally, cost, distribution, failure potential, and types
of contamination were taken from data in regulatory impact analyses (RIAs) or
similar reports on impacts of various waste management practices. In most cases
these data were based on samples of sites or facilities. Generally, data
deficiencies required that the Workgroup had to make assumptions about
environmental settings of various types of facility or site, hydrological
parameters, exposed populations, types of waste handled, facility size, and
water use patterns.
Current and future ground water use was estimated on the basis of U.S.
Geological Survey water supply data covering the 1975-1980 period. The
C-19
-------
probability of future use was estimated based on the assumption of an increased
rate of use of 1.4 percent per year.
Results. The Ground Water Valuation final ranking is presented in
Table C-5. Aggregate national costs (not shown) ranged from $0 for Combustion
of Hazardous Wastes to $15 billion for USTs. Other significant findings
included:
* NPL sites have the highest mean resource damage per site
(present value costs of $9.7 million per site).
• The number of facilities in each problem area plays a significant
role in determining aggregate national resource damage cost
estimates.
• The size of facilities within a problem area is highly correlated
with high resource damage cost estimates.
Discussion. A number of important caveats or limitations should be noted
about the results of the Ground Water Valuation Report:
• The Ground Water Valuation Workgroup did not consider the effects
of ground water contamination prevention programs, although such
regulations are already in place in a number of problem areas.
The effect of this assumption is to probably overestimate
resource damage in some problem areas. If regulatory
environments were considered, it is likely that the ranking of
the problem areas would change.
• The Workgroup assumed that no ground water clean-up activities
were performed for any problem area except NPL sites. The effect
of this assumption is that contamination spreads more widely
(i.e., plumes are modeled to grow larger and last longer) than
they actually do given current regulations. As a result,
replacement water supplies are assumed to have larger design
capacity and operate for longer periods than they probably
actually have or do. This has the overall effect of increasing
resource damage cost estimates.
* In many cases the final resource damage estimates were based on
poor data and numerous assumptions. Generally, the weakest
aspects of the data were related to plume size and water use
patterns.
• The Workgroup considered using only a nearby aquifer as a
substitute source for contaminated drinking water. Other
alternatives (e.g., temporary use of bottle water or point of use
treatments) were not considered as alternatives.
• The Workgroup did not consider regional differences in resource
damage. As such, regional variation in the cost of replacing
contaminated water systems was not considered. Such variations
C-20
-------
might have considerable impacts on the overall ranking of problem
areas.
Resource damage models focused entirely on human use. Other
forms of potentially significant damage from contaminated ground
water (e.g., the loss of ground water for irrigation) were not
considered. Including other economic losses might change overall
rankings.
The models used to calculate the impacts of contamination are not
sensitive to the magnitude of contamination beyond identified
thresholds. Resource damage is held to occur the moment a
threshold is passed. At that point it is assumed that users will
stop drinking contaminated water and seek alternative supply the
moment the threshold is exceeded. Including magnitude in the
analysis might alter the overall ranking of the problem areas.
C-21
-------
FIGURE C-5
GROUND-WATER VALUATION METHODOLOGY
o
I
to
to
Develop Potential
Scenarios
Representing
the Range of
Potential
Determine
Thresholds at
Which Chemical
Constituents
Cause Drinking
Water Supples
to be Unusable
Use Avafeble
Ground-water
Models to Predict
Concentrations
Downgraded
from
Release Point
Determine
Score by
Setting
Replacement
Cost Equal
to Resource
Damage
Preofct
Capital and
Operating Costs
for
Water Supply
Replacement
Compare
Concentrations
to Thresholds
-------
TABLE C-5
GROUND WATER VALUATION
Natural Mean
Resource Resource
Problem Area Damage* Damage**
Underground Storage Tanks 15000 11
NPL Sites 8600 9700
Municipal Landfills 2580 420
Solid Waste Management Units 1750 310
Land Disposal of Hazardous Waste 190 365
Mining Waste 178 360
Oil and Gas Waste 173 0.8
Other Treatment, Storage and Disposal Facilities 69 26
Hazardous Waste Combustion 0 0
The following problem areas were not evaluated but judged to rank between other
treatment, storage and disposal facilities and hazardous waste combustion:
Industrial (D) Landfills
Land Treatment of Hazardous Waste
Underground Injection
Non-NPL Sites
Transportation
Similarly, the following problem areas were not modeled but judged to rank after
hazardous waste combustion:
Municipal Trash Incineration
Hazardous Waste Cleanup
Ocean Incineration
*Natural Resource Damage is reported as water supply replacement costs by
problem area in millions of dollars.
**Mean Resource Damage is reported as water supply replacement costs by problem
area in thousands of dollars.
- 23 &U.S GOVERNMENT PRINTING OFFICE: 1989 - 748-159/00363
------- |