EXTENT OF THE HAZARDOUS RELEASE PROBLEM AND
FUTURE FUNDING NEEDS
CERCLA SECTION 301(a)(l)(C) STUDY
Final Report
Office of Solid Waste and Emergency Response
U.S. Environmental Protection Agency
December 1984
-------
Substantial portions of this report were prepared by ICF Incorporated for
the (J.S. Environmental Protection Agency under Contract Number 68-01-6872.
-------
TABLE OF CONTENTS
Page
PREFACE
EXECUTIVE SUMMARY i
1 . KEY ELEMENTS OF THE SUPERFUND PROGRAM 1-1
2. THREATS POSED BY HAZARDOUS SUBSTANCE RELEASES 2-1
2.1 Overview 2-1
2.2 Detailed Characterization of the Threats at MPL Sites 2-4
3. SCOPE OF SUPERFUND RESPONSE 3-1
3.1 Introduction 3-1
3.2 Scope of Superfund Jurisdiction 3-1
3-3 Current Focus of the Remedial Program 3-4
4. CENTRAL ESTIMATES OF FUTURE FUNDING NEEDS :. . 4-1
4.1 Introduction 4-1
4.2 Factors and Assumptions Used in Funding Projections 4-1
4.3 Baseline Estimate of Future Funding Needs and
Sensitivity Analyses 4-8
5. POTENTIAL EXPANSION IN THE SCOPE OF THE SUPERFUND PROGRAM 5-1
5.1 Introduction 5-1
5.2 Sources of Expansion: New Problem Areas 5-2
5.3 Sources of Expansion: Policy Changes 5-13
-------
LIST OF EXHIBITS
Exhibit Page
1-1 EPA's Hazard Assessment Process 1-3
2-1 Most Frequently Reported Substances at 546 NPL Sites 2-6
2-2 Hazardous Properties of Most Frequently Reported
Chemicals 2-8
2-3 Observed Releases to Environmental Media for 546
NPL Sites 2-11
2-4 NPL Sites with Potential Releases to Groundwater 2-13
2-5 NPL Sites with Potential Releases to Surface Water 2-13
2-6 Population Estimates by Environmental
Medium at 546 NPL Sites 2-14
-
2-7 Distances to Sensitive Environments from 546 NPL Sites ... 2-15
2-8 Aquifers Potentially at Risk 2-15
3-1 Types of Activities at National Priorities List Sites 3-6
3-2 Summary of NPL Site Activities 3-8
3-3 National Priorities List Hazard Ranking System
Score Distribution 3-9
4-1 Summary of Factors and Assumptions Used in
Baseline Funding Estimate 4-2
4-2 Projected Size of the NPL 4-4
4-3 Average Remedial Cost Per Site 4-5
4-4 Baseline and Central Estimates of Future Funding Needs ... 4-10
5-1 Potential Sources of Superfund Expansion 5-3
-------
PREFACE
Section 301(a)(1)(C) of the Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA or Superfund) requires that the
President submit to Congress
a projection of any future funding needs remaining after the
expiration of authority to collect taxes, and of the threat to public
health, welfare, and the environment posed by the projected releases
which create any such needs.
This report projects the size and focus of the Superfund program and
future funding needs to address hazardous release problems. It presents
information on the numbers and types of problem sites that the Agency
anticipates and the threats posed to human health, welfare, and the
environment by hazardous releases. Baseline future funding needs are
estimated based on EPA's experience with site discovery, investigation, and
response activities. Sensitivity ranges around this estimate are developed by
altering key assumptions. All of these central funding estimates assume that
the focus of the Superfund program will stay essentially the same in the
future as it has been to date.
However, the size and focus of Superfund response could expand well beyond
the projections provided in the central estimates. This report also
identifies and discusses the types of problem sites that could contribute to a
dramatic increase in the size of Superfund.
-------
EXECUTIVE SUMMARY
Over the past four years, EPA has engaged in intensive efforts to discover
potential hazardous substance releases (about 19,000 are now listed on the
Superfund's "ERRIS" data base) and to investigate releases and list the most
serious ones on the National Priorities List (NPL) so that remedial cleanup
can begin. As of October 1984, 538 sites are listed on the NPL, with an
additional 248 sites proposed for listing.
Hazardous releases listed on the NPL are typically characteri2ed by three
factors: substances present at the site are inherently hazardous to health;
routes of exposure to the substances exist through groundwater, surface water,
or air; and target populations and environments are present that can receive
exposure to hazardous substances. The 25 substances most frequently found at
NPL sites have widely differing toxicities. However, among the properties of
these substances, nearly half are known or suspected carcinogens; seven are
very toxic to aquatic life; nine are known to be mutagens; seven are known
teratogens; and seven will ignite at room temperatures. In addition, many
hazardous waste sites contain a number of hazardous substances that may work
synergistically to cause or enhance a variety of toxic effects.
EPA expects that many of the sites that will be targeted for Superfund
cleanup in the future will pose threats resembling those that ace currently
listed on "the NPL.- The current inventory of sites and anticipated new
additions will produce an NPL of 1,500 to 2,500 sites over the next several
years. EPA's baseline estimate, using current program experience, is that the
NPL will increase to some 1,800 sites.
To address an NPL of 1,800 sites, with an average remedial cost of $8.1
million and expected responsible party contributions of 50 percent of the
costs of cleanup, future funding requirements would total $11.7 billion (FY83
dollars). This estimate includes costs of remedial as well as removal
response and associated support and enforcement costs. Potential costs
recovered from responsible parties are also factored into this estimate.
Because of uncertainties in the projections, a central range of future
funding needs is estimated as well. Depending on assumptions about the size
of the NPL, the average cost of a remedial action, and the level of
responsible party contributions to cleanup actions, future funding needs could
range from $7.6 billion to $22.7 billion, in FY83 dollars.
While EPA's response authority under CERCLA is extremely broad, these
central estimates of funding needs are based on the assumption that the
Superfund program will remain similar to EPA's experience with the program to
-------
date. The emphasis over the past four years has been on more traditional
waste sites and more obvious hazardous release problems. The focus and
character of Superfund, however, appear to be expanding to cover sites that
require more intensive discovery efforts, as well as problems that are new to
Superfund.
If EPA were to undertake a targeted, systematic discovery and
investigation effort into these problem areas, the size of the program could
increase substantially.
Among the emerging problem areas are the following:
RCRA Subtitle C Facilities. About 130 of 960 privately
owned land disposal facilities are expected to close for
financial reasons, and some 175 out of 3,520 privately
owned storage and treatment facilities are similarly
expected to close.
Municipal Landfills. There are about 12,000-18,000
currently operating municipal landfills, with up to
twice as many inactive municipal landfills; some of them
received hazardous wastes prior to RCRA notification
requirements and some still receive hazardous wastes of
small quantity generators and households.
. Industrial Landfills. Approximately 75,000 industrial
landfills are currently operating, some of which contain
and/or" continue to receive hazardous wastes.
v
Mining Waste Sites. Some 10,000 to 64,000 mines, both
active and inactive, may be potential sites of concern
to Superfund. Hazardous substances associated with
mining sites include heavy metals, radioactive
materials, and asbestos.
Leaking Underground Storage Tanks. Current estimates
range from 11,250 to 187,500 facilities with leaking
underground tanks that contain hazardous substances.
Contamination from Agricultural Uses of Pesticides. To
date, at least 15 different pesticides have been found
in the groundwater in over 20 states as a result of
agricultural uses.
Radioactive Sites. The Nuclear Regulatory Commission
administers 8,900 materials licenses; agreement states
administer an additional 13,000. EPA currently
considers former NRC licensees and state licensees as
eligible for NPL listing.
- ii -
-------
Until systematic identification and investigation of these many different
types of sites is undertaken, it is impossible to estimate the total number of
sites that could become potential Superfund problems. However, if even a
small fraction of these sites require Superfund response, the funding needed
to address them would overwhelm the central estimates currently projected for
the Superfund program.
In addition, it is possible that a large number of sites that have been
investigated and that scored below 28.5 on the Hazard Ranking System, thereby
becoming ineligible for listing on the NPL and remedial funding, may
nevertheless pose hazards that will need to be addressed over time.
- iii -
-------
1. KEY ELEMENTS OF THE SUPERFUKD PROGRAM
The Comprehensive Environmental Response, Compensation, and Liability Act
of 1980 (CERCLA) establishes a $1.6 billion reimbursable trust fund (referred
to as the "Fund") financed primarily by a tax on the manufacture, production,
or import of certain chemicals, and supplemented by general revenues. The
Fund may be used to pay for costs incurred during responses to releases of
hazardous substances, pollutants, or contaminants, that are taken in
accordance with the response authorities provided by CERCLA and the revised
National Oil and Hazardous Substances Pollution Contingency Plan (NCP) (40 CFR
Part 300). The NCP contains the basic policies and procedures that direct the
federal response to potential or actual releases of hazardous substances.
CERCLA authorizes two basic approaches to the problems posed.by hazardous
substance releases the Agency may use Fund money to respond to releases
(applying response authorization under section 104), or the Agency may use its
enforcement authorities (applying section 106 for administrative orders and
civil actions) to compel responsible parties to clean up releases. Section
107 establishes liability for hazardous substance releases for any person who
generates, transports, stores, treats, or disposes of hazardous substances.
Using negotiations, administrative orders, or litigation, the government may
encourage or compel private parties to finance response actions at sites for
which they are liable.
SPA's policy is to seek action by responsible parties so far as possible
before committing Fund monies for a government-managed response. Enforcement
authorities may also be used to recover Fund monies spent for response at
releases where there is at least one financially viable responsible party.
However, if the situation at the site requires prompt response or if
negotiations are not successful within a specified timeframe, the Agency will
respond with a Fund-financed cleanup and seek to recover the costs of the
action.
Two categories of response action are authorized by Section 104 of
CERCLA: removal and remedial actions. Removal actions are generally intended
for shorter-term responses to releases of hazardous substances and are limited
normally to 6 months in duration and $1 million in costs. CERCLA does not
require cost-sharing by the states for removal actions at privately owned
sites.
Remedial actions are longer-term, more deliberative actions that are
consistent with permanent remedy of a release. CERCLA requires cost-sharing
by the state in which the release occurs for remedial actions: a 10 percent
state share at privately-owned sites and a 50 percent state cost share at
sites owned by the state or a political subdivision. The NCP requires that
remedial actions be taken only at sites listed on the National Priorities List
-------
1-2
(MPL), which contains a list of the top priority facilities in the country
(CERCLA Section 105(8)). Sites are identified by the state or EPA regional
office and submitted to EPA. CERCLA requires, to the extent practicable, that
each state have at l^ast one site ranked among the 100 highest priority
releases.
Because the Superfund remedial action program' incorporates a systematic
process for identifying and screening releases and represents the bulk of the
Superfund cleanup capability, most of the discussion in this report concerns
the remedial action effort, which is described in more detail below.
HAZARD ASSESSMENT AND REMEDIAL ACTION
Information on the magnitude of the hazardous substance problem has
increased substantially since Congress enacted CERCLA in 1980. At that time,
EPA estimated that 30,000-50,000 hazardous substance sites existed.- Since
then, EPA has developed an inventory of hazardous substance sites called the
Emergency and Remedial Response Information System (ERRIS) and has begun
investigating these sites to determine which of these sites will require
response under CERCLA. The ERRIS data base currently contains about 19,000
sites {18,900 as of September 30, 1984) with a range of potential hazardous
waste problems.
EPA's Office of Emergency and Remedial Response (OERR) has developed a
process to evaluate hazardous waste sites to determine which sites need
response, potentially with Superfund financing, and to identify priority sites
with the most^ serious hazardous substance problems. The site evaluation
process includes several steps, as shown in'Exhibit 1-1; each step is
described below. EPA's experience with the site evaluation process is used in
this report to project the number of sites requiring response in the future.
Although the end point of this investigation process is listing eligible sites
on the NPL, removal actions may also be initiated when immediate hazards are
found at sites during the investigation.
Step 1: Preliminary Assessment. EPA (or states under an EPA
cooperative agreement) conducts a preliminary assessment for each of the ERRIS
sites to determine whether further action is needed, and whether responsible
parties exist that are ready, willing, and able to undertake a proper response.
During a preliminary assessment, EPA or the state assembles and reviews
readily available data to decide whether a reported site should be the object
of more intensive investigation. A preliminary assessment includes primarily
an office review of available data. In some cases, however, EPA or the state
may tour the perimeter of the site. EPA has conducted over 10,700 preliminary
assessments.
The primary purpose of a preliminary assessment to screen out those sites
that clearly pose no problem. If the site poses a potential threat that may
-'Fred C. Hart, Inc., "Preliminary Assessment of Cleanup Costs for
National Hazardous Waste Problem," prepared for EPA, February 19, 1979.
-------
1-3
EXHIBIT 1-1
EPA'S HAZARD ASSESSMENT PROCESS
Sices Discovered
Preliminary Assessment
So Further
Action Required
Site
Inspection
Pending
Scoring
Sites with an HRS
Ranking Below
28.5
Sites Listed
on the MPL
-------
1-4
require EPA response or if there is not enough information to make this
determination, then a site inspection will be conducted. In some cases, EPA
may determine that the .site falls under the jurisdiction of another regulatory
program such as RCRA and will refer the site to the appropriate EPA office for
follow-up. If there is no release or threat of release of hazardous
substances likely at a site, or if the site is under the jurisdiction of
another regulatory program, the site is placed in 3. "no further action"
category. If EPA determines that the likelihood of a release is low, then the
site is placed in a "pending" category. EPA or the state will monitor these
sites to identify problems if they occur. "Pending" also applies to
situations where a state, local government, or private party is concerned
about a site and may even be conducting a cleanup but which EPA is not likely
to address under CERCLA. EPA plans to complete preliminary assessments of all
currently listed ERRIS sites within the next two years.
Step 2: Site Inspections. EPA conducts a detailed site inspection
when the preliminary assessment indicates that the site may require EPA
action. About a third of all sites that have had a preliminary assessment to
date require a site inspection. A team of investigators takes field samples
and collects detailed information on the site to determine the seriousness of
the hazardous substance problem. Depending upon the time required to take
samples and receive laboratory results, this step may take from one to six
months to complete. After this detailed investigation, EPA may determine that
(1) the site requires no further action; (2) -the site ppses a threat that EPA
is unable to respond to with current funds and authority; or (3) the site
should be ranked.using the Hazard Ranking System (HRS). EPA has initiated
about 3,600 site inspections and plans to- complete 8,000 within the next two
years.
Step 3: Scoring. Sites that appear to pose a serious problem on site
investigation are scored using the Hazard Ranking System (HRS) to determine
whether the site poses a problem that the Fund should address. The HRS is
designed to measure the relative severity of the site and the probability and
magnitude of human and sensitive environment exposure to hazardous
substances. The HRS scores sites on the basis of "toxicity of hazardous
substances present at the site, observed or potential releases of hazardous
substances into environmental media, existence of potential exposure pathways,
and the size of the populations that could receive exposure. Three scores are
assigned to a site that reflect the potential harm from migration of a
hazardous substance from a site via ground water, surface water, or air.
These three scores are then weighted and combined to yield an estimated hazard
ranking. The scores can range from 0 (least hazardous) to 100 (most
hazardous).
Step 4: Listing on the NPL. At present, sites are listed on the NPL
only if the site receives a score of 28.5 or more under the Hazard Ranking
System (HRS) or is designated by a state as its top priority. The 28.5 cutoff
was originally selected to yield an initial NPL of at least 400 sites as
suggested by CERCLA; it does not necessarily represent a threshold in the
significance of the risks presented at sites. EPA will provide Fund-financed
remedial action only at sites listed on the NPL. (EPA may, however, initiate
-------
1-5
an enforcement action to secure privately funded cleanup or emergency action
at any site.)
As of September 1984, the National Priorities List contains 538 hazardous
substance releases. An additional 248 sites have been proposed for listing,
for a total of 786 proposed and final NPL sites.
In summary, £PA''s experience to date is that approximately one out of
every 12 sites receiving preliminary assessments is listed on the NPL. This
proportion has generally held over time, although at any given point in the
program, current activity levels may alter the ratio slightly. EPA's
experience with the hazard assessment process is used in Chapter 4 as a basis
for predicting the eventual size of the NPL.
-------
2. THREATS POSED BY HAZARDOUS SUBSTANCE RELEASES
2.1 OVERVIEW
Hazardous waste sites and hazardous substance releases may pose serious
threats of harm to both humans and the environment. These potential hazards
are many and varied; some are well known, some we know very little about.
Some hazards are acute and cause adverse effects within a very short time.
Other hazards, such as cancer and reproductive effects, may occur or become
apparent only over a long period of time. Environmental effects may also
become apparent only over a long period and may involve loss of vegetation,
destruction of habitat, and eventual destruction of species in a given
geographic area. Generally, it is very difficult to establish the
cause-and-effect relationships that lead to those impacts, both in human
populations and in the environment.
An accurate picture of the specific threats and the degree of risk from a
hazardous waste site requires a detailed site-by-site analysis. Such an
analysis would.be based on data collection and analytic work to precisely
quantify the chemicals arid concentrations present at the' site, the routes and
levels of exposure, and the target populations. Resources are not available
for such an intensive analysis here. Instead, this chapter provides an
overview of the types of threats posed by hazardous releases and a more
detailed characterization of the threats posed by National Priorities List
(NPL) sites.
Three major factors must be considered in assessing the threats posed by
hazardous waste sites and releases: (1) the substances that are present,
their hazardous properties and the concentrations .and volumes in which they
occur; (2) the existing or potential routes of exposure through which people
and the environment may come into contact with the substances; and (3) the
target populations and environments and particularly sensitive subgroups that
may be the receptors of these substances. Each factor is briefly described
below.
Hazardous properties of substances. Substances found at
hazardous waste sites may pose physical hazards and
biological hazards. Substances with physical hazards
may explode or ignite easily, or may be corrosive to the
skin and eyes if contacted. Biological hazards in
humans include acute toxicity resulting in death or
serious illness; ability to cause cancer, reproductive
effects, or neurotoxic effects; and other chronic health
effects. In the environment, such substances may affect
vegetation or animal populations. The concentrations in
which substances are present are also critical, since
-------
2-2
many potentially toxic substances may be present in such
small quantitites or in so dilute a form that they do
not pose a significant risk of harm.
Routes of exposure. The route of exposure is the
pathway through which humans or the environment come
into contact with hazardous substances. Critical routes
of exposure are surface water, groundwater, air, and
direct contact. People may come into contact through
ingestion, inhalation, or absorption of the substances.
In some situations, the substances may be contained so
that the possibility for exposure is very low. For
example, the substances may be present in drums which
are not leaking, or the substances may be in a
geographic location which is well contained.
Target populations or environments. "Targets" are the
people or environments most likely to be exposed to the
substances. For instance, target populations may be
those that are served by a drinking water source, or the
people who use surface water for recreation. Within any
given target population, some people or specified plants
and animals may have varying degrees of contact or
sensitivity to hazardous substances.
Non-NPL Sites
Many of the sites that score below the 28.5 HRS cutoff still pose some
threat to human health. Generally, these are sites that potentially affect
fewer people, or where there is less opportunity for exposure. A number of
these sites have actual releases into surface water, groundwater, or air, but
the surrounding population is small. Sites with a release into only one
environmental medium that is, sites with only one route of exposure ~ are
likely to score lower, even though there are surrounding populations that may
be affected by the site.
Information on non-NPL sites is much less accessible than the detailed
data available for NPL sites. In the course of this study, EPA conducted a
Site Characterization Survey in all 10 EPA regional offices. The Site
Characterization Survey was developed to call on the expertise of EPA Regional
staff and contractors who are directly familiar with the wide range of
hazardous waste sites and releases which have been identified and inspected
under the Superfund program. In general, those who were interviewed believe
that EPA has identified and is addressing the worst sites in the country, but
that many more sites pose threats to health and the environment and should
receive response, whether by EPA, the states, or the responsible parties.
Examples of sites posing potential health and environmental threats that
are not listed on the NPL are the following:
A number of sites with a direct contact threat to human
health are not listed on the NPL because direct contact
is not factored into the HRS. These sites may involve
-------
2-3
substances such as lead, or 2,3,7,8 TCDD (dioxin) in the
soil or in airborne particles which could be inhaled,
ingested or absorbed. When there is an immediate threat
through direct contact, EPA can take a removal action to
control access to the site.
Some sites located in urban areas with large surrounding
populations may fall into the lower priority group of
sites. Hazardous waste sites here may involve some
groundwater or surface water contamination, but the
population is likely to be served by municipal drinking
water supplies rather than the affected water supplies,
so there is little opportunity for contact through these
routes of exposure.
Some urban sites, as well as some sites that are
isolated, may involve drums which are in good condition
and do not pose an immediate threat. However, over time
such containers may deteriorate and cause problems.
At some sites with actual releases to surface or ground
water, the water is not used for drinking. It may have
industrial or agricultural uses or it may not provide
for human uses at all.
Some sites may be isolated from populations but could
pos.e significant environmental damage. Related to this
are sites with hazardous substance releases which may
pose threats through contamination of food chains.
Water used for irrigation or stock watering may, over
the long term, affect plants and animals which are used
for human consumption. Currently, these sites are not
addressed by Superfund, if human populations are not
involved or if there is not an immediate hazard.
In some areas, a number of small sites with minor
individual impacts may all affect the same resource.
For example, a number of sites located above the same
aquifer could have serious cumulative impacts on
groundwater.
National Priorities List Sites
As discussed in Chapter 1, EPA lists on the National Priorities List (NPL)
those sites which appear to present a significant risk to human health or the
environment. The basis for listing is the score received using the Hazard
Ranking System, a numerical scoring system which uses easily obtained data to
score the hazards associated with each site. The cutoff HRS score for listing
sites on the NPL is 28.5, which yields a list of sites with a relatively
greater potential public health threat. The HRS is weighted heavily in favor
of public health risk rather than environmental threat. While it does not
-------
2-4
look at the full range of potential environmental threats, it does score risk
to particularly sensitive environments.
The major source of information used in this chapter on the threats posed
by NPL sites is the NFL data base maintained by EPA. This computerized data
file contains numerically-coded information on approximately 50 attributes of
each of the NPL sites, compiled in che process of scoring sites under the
Hazard Ranking System. HRS determines a total "hazard score" for each site by
combining qualitative "subscores" for the attributes in a detailed arithmetic
procedure. The information used to assign these subscores comes from site
inspections and from records already available (i.e., types and quantities of
substances disposed, hydrogeologic and climatic data) about each site.
The HRS assesses the likelihood that a hazardous release will occur and
the potential hazard associated with a release separately, and then combines
these parameters along each of five environmental pathways: groundwater,
surface water, air, direct contact, and fire/explosion.!' The last two
scores are now used primarily to determine when timely, "emergency" response
is necessary at a particular site. For the purposes of ranking sites for
listing on the NPL, a composite score (on a 0-100 scale) for each site is
arrived at by a mathematical procedure that gives greatest weight to the
pathway with the highest score, but allows for a high composite score if a
site presents a moderate but balanced threat along all of the first three
pathways.
The HRS assesses the severity of each potential threat by combining
information about the .quantity of wastes present at the site and the toxicity
and environmental persistence of the most dangerous of these wastes with
information about the proximity of target populations or ecosystems and the
degree to which biological uptake of chemicals could occur (through, for
example, drinking contaminated water).
2.2 DETAILED CHARACTERIZATION OF THE THREATS AT NPL SITES
This section provides a more detailed characterization of the hazards
posed by 546 NPL sites, using the framework outlined above (inherent hazards
V The HRS treats the likelihood of release to groundwater and surface
water in one of two ways: (1) if a release of one or more hazardous
substances has already been observed at a site, the likelihood of a threat is
assumed to be 100 percent because conditions are such that hazardous
substances can defintely escape the site; or (2) even in the absence of a
documented release, HRS can assign a putative likelihood score approaching 100
percent, if waste composition, hydrogeologic, climatic, and site management
conditions suggest that a future release is probable. For the air pathway,
the likelihood of a release is assumed to be zero unless a release has already
been observed (in which case the likelihood value assigned is 1).
The components and scoring methodology of the Hazardous Waste Site Ranking
System are explained in Appendix A of the National Oil and Hazardous
Substances Pollution Contingency Plan, July 16, 1982 (40 CFR Part 300).
-------
2-5
of substances, exposure pathways, and target populations and environments).
The 546 sites included in this analysis are those that were listed on the
proposed and final NPL as of July 1, 1984.
2.2.1 Hazardous Substances Found at Sites
Exhibit 2-1 provides data on the 25 substances most frequently reported as
present at the 546 MPL sites. The first column of numbers gives the number of
sites at which each substance was reported; the remaining columns give the
number of sites for which each substance was detected in the three most
important environmental media.
The 25 specific substances listed here have widely different toxicities
and physical properties such as solubility, volatility, and environmental
persistence in the various media. However, the substances fall into four
general chemical classes:
(1) Chlorinated solvents (I 1, 6, 7, 12, 16, 17, 21, 22, 23,
24, and 25) these substances are generally very toxic
to aquatic life. More troublesome, however, is their
potential for interacting with mammalian DMA and thus to
cause cancer or genetic malformations.
(2) Aromatic solvents (#3, 4, 8, 14, and 15) many of
these substances are highly volatile and thus present
potential hazards via the air pathway. Benzene, the
building block of all aromatic compounds, is a proven
human carcinogen, according to the International Agency
for Research on Cancer (IARC). Most of these substances
are very flammable and can also solubilize otherwise
immobile substances; thus, in the event of a fire or
explosion, the presence of these aromatic solvents can
promote the dispersal of other toxic materials.
(3) Heavy metals (#2, 9, 10, 11, 13, 18, and 195 these
substances create a wide variety of toxic effects, often
highly dependent on the form in which the metal was
disposed of. If the metal is present and remains in its
elemental form, it can sometimes present little danger.
If, however, the metal reacts with other chemicals to
yield metallic compounds (e.g., lead phosphate, copper
chloride), it can become highly toxic. In particular,
if bacteria in soil convert elemental mercury or arsenic
into organic compounds such as methylmercury, an
extremely hazardous situation can result. Exposure to
lead can produce a number of systemic effects, including
neurotoxicity, anemia and other blood disorders, and
kidney toxicity. Lead exposure is especially dangerous
to children. In addition, three of the metals listed
here (arsenic, cadmium, and hexavalent chromium) are
presumed to be carcinogens.
-------
EXHIBIT 2-1
MOST FREQUENTLY REPORTED SUBSTANCES AT 546 NPL SITES
Substances Identified
at Hazardous Waste
Waste Disposal Sites
Sites a/
Detection in Environmental Media
Groundwater
Sites (Rank) b/
Surface
Water
Air
Sites (Rank) b/ Sites (Rank) b/. c/
Most Frequently Occurring
1. TrIchloroethy1ene
2. Lead
3. Toluene
4. Benzene
5. Polychtorinated Biphenyts (PCBs)
6. Chloroform
7. Tetrachloroethy1ene
8. Phenol
9. Arsenic
10. Cadmium
11. Chromium
12. 1 .1,1-Trichloroethane
13. Zinc and Compounds
14. EthyIbenzene
15. Xylene
16. Methylene Chloride
17. Trans-I,2-OlchloroethyIene
18. Mercury
19. Copper and Compounds
20. Cyanides (Soluble Salts)
21. Vinyl Chloride
22. 1,2-Dichloroethane
23. Chlorobenzene
24. 1.l-Dicnloroethane
25. Carbon Tetrachloride
179
162
153
143
121
1 11
90
84
84
82
80
79
74
73
71
63
59
54
47
46
44
44
42
42
40
127 (1)
77 (4)
61 (3)
84 (2)
39 (11)
70 (5)
57 (7)
43 (9)
45 (8)
28 (16)
34 (14)
58 (6)
28 (17)
36 (12)
32 (15)
36 (13)
42 (10)
27 (20)
17 (24)
16 (25)
28 (18)
25 (21)
23 (23)
26 (19)
25 (22)
49
64
40
36
54
24
17
28
35
28
33
20
27
14
8
17
17
20
16
16
10
17
9
9
12
(3)
(1)
(4)
(5)
(2)
(11)
(14)
(8)
(6)
(9)
(7)
(12)
(10)
(20)
(25)
(15)
(16)
(13)
(18)
(19)
(23)
(17)
(23)
(24)
(21)
8
7
16
18
6
1
3
3
2
31
1
3
2
7
9
2
1
4
0
2
4
2
0
0
2
(S)
(6)
(3)
(2)
(8)
(15)
(16)
(17)
(17)
(18)
(7)
(4)
(10)
(11)
I
ON
a/ Number of sites at which substance is present. Substances may be present in one, two. or all three
environmental media at any given site. Also, each substance has not been detected in environmental
media at all sites at which it is known to be present. Therefore the number of sites at which each
substance is detected in environmental media may not equal the number in this column.
b/ Not all ranks will be represented in i
those found most frequently at sites.
c/
II media because not all chemicals found In media are among
Volatile organics not otherwise specified were reported as being detected most often (Rank l) in the
air medium.
Source: NPL data base.
-------
2-7
Inorganic ions the cyanide ion (#20), because of its
stability, is the agent of greatest concern in this
chemical class. All soluble cyanide salts are highly
toxic to humans and to aquatic life.
In addition to these four generic classes, polychlorinated biphenyls
(PCBs) are frequently found in waste disposal sites because of the ubiquitous
use of these compounds in electrical and mechanical equipment during the
period 1930-1980. PCBs are among the most persistent organic compounds known;
only incineration at temperatures above 1500F can readily break down the
carbon-chlorine bonds. At low to moderate exposure levels, ?C3s can cause
disorders of the nervous and digestive systems, and skin lesions known as
chloracne, and they may pose risks of cancer and mutations.
Exhibit 2-2 provides information about the individual properties of these
25 chemicals. As can be seen in the exhibit, most of these substances present
serious hazards to humans and animal life, depending on the concentrations
encountered and the duration of exposure. In summary:
(1) Seven of the 25 substances are very toxic to aquatic
life (some, like mercury, are so toxic that less than 40
milligrams dissolved in 100 gallons of water will kill,
on average, half of the fish exposed for 96 hours);
(2) Nearly half of the substances are known or probable
human carcinogens, and thus theoretically pose finite
risks even in the smallest concentrations;
(3) Nine of 25 are known to be mutagens;
(4) Seven of 25 are known teratogens;
(5) Seven of the 25 will ignite at room temperatures or
lower;
(6) Seven of the 25 can be bioaccumulated in the
environment, and thus present additional risks to humans
through consumption of contaminated fish, and animals,
or plants; and
(7) Nineteen of the 25 substances resist biodegradation, and
are thus highly persistent in the environment,
especially in groundwater.
The 25 chemicals listed here are present in varying concentrations at
different sites. At each site, field personnel are encouraged to report as
present as many chemicals as they deem significant. This assessment may come
from actual sampling or from records that indicate certain chemicals have been
disposed of at the site; it is a general indication of those chemicals the
inspectors are most concerned about at a given site, not a priority ranking on
the basis of volume, toxicity, mobility, or elevated concentration. The NPL
-------
EXHIBIT 2-2
HAZARDOUS PROPERTIES OF MOST FREQUENTLY REPORTED CHEMICALS
Acute
Aquatic Toxicity
(LCso, 96-hour a)
(See Note 1)
1.
2.
3.
4.
S.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
Trichloroethylene
Lead
Tojuena
Benzene
PCBa
Chloroform
Tetrachloroethylene
Phenol
Arsenic
Cadmium
Chromium
1,1, 1-Tr ichloroethane
Zinc (and compounds)
Kthylbenzene
Xylene
Methylene Chloride
Trans -1 , 2-Dichloroethy lene
Mercury
Copper (and Compounds)
Cyanides (Soluble salts)
Vinyl Chloride
1, 2-Dichloroe thane
Chlorobenzene
1 , 1-Dlchloroethane
Carbon Tetrachloride
medium
low
aeditm
high
very high
medium
medium
medium
high
medium
edlua
aedlum
medium
medium
low
extremely
high
very high
very high
low
low
high
edlua
Carcinogenic Potential
Mammalian Acute (rated by SPA Special
Toxlcity (LDso, Carcinogen Assess- Physical
in mg/kg body ment group) (See Properties
weight) (See Note 2| . Note 3) Hutagenlclty Teratogenicity (See Note 4) Other
4920 low
160 (pigeon Lo oral)
5000
3800 low
approximately medium
4000, depending on
PCD isomer
800 low
6100 (mouse) low
414 (rat) i 669 (rat- see note 10
dermal)
8 (rat) high
225 aedlum
medium
10,300 see .note 11
350 (zinc chloride)
3500
4300
167
75,000 ppm over 2
hours (mouse - lo-
Inhalatlon)
29 mg/a3 over 30
hours ( rabbit
Inhalation)
0.120 (huaan -
lo oral)
500; 20 ppm over 30 low
minutes (guinea pig.
lo inhalation)
670 low
2910
725
2800 medium
+ flash point-90*F See
+ See
+ flauh polnt-40*F See
+ + flash point-12'P See
+ See
See
See
+ + See
+
+
See
See
* tlauh point-81*F
+
See
5,
See
See
+ + gaseous
+ flash point- S6*F See
flash polnt-84*F
flash polnt-22*F
+ See
Note a
Note S
Note 7
Note 7
Not* 5
Note 7
Note 7
Notes 5,
Note 5
Note 7
Notes
6
Note 5
Note 9
Note 7
Note 8
CO
Sources: Technical Background Document to Support Ruleraaklng Pursuant to CERCLA Section 102(b), Rockwell International Corporation) IARC Monographs on
the Evaluation of Carcinogenic Risk of Chemicals to Humans. Volumes 1 to 30 and Supplement 4| Health Effects Assessment Documents, U.S. EPA,
Environmental Criteria and Assessment Office, Cincinnati, Ohio.
-------
MOTES FOR EXHIBIT 2-2
Hot* It Aquatic toxlclty is determined by the "96-hour LCgg," the concentration, in parts per million, that when dissolved in water will kill 50 percent
of a population of test organisms (usually fathead minnows or bluegilla) within 96 hours after exposure. "Low" aquatic toxicity here designates
an LCso greater than 100 ppmi "medium)" between 10 and 100 ppmi "high,* between 1 and 10 ppmj "very high," between 0.1 and 1 ppmi and "extremely
high," less than 0.1 ppm.
Mote 2» The values given are in milligrams per kilogram of body weight; and represent the doses that will kill SO percent of a population of test
animals* Unless otherwise indicated, the test animal used was the rat, and the substance was administered orally. Other mode* of administration
include inhalation and dermal application. The designation "lo" fallowing the data entry Indicates the lowest dose or concentration known to have
a lethal effect on members of the teat species mentioned.
Note 3: The designations "high," "medium," and "low* come from the EPA Environmental Criteria And Assessment Office (ECAO) in Cincinnati, and is based
partly on research conducted by the International Agency for Research on Cancer in Lyon, France. The designations are determined by analysing a
4x4 matrix, with "sufficiency of evidence" as one axis and "potency"" as another. The IARC rates carcinogens in four groups according to the
weight of evidence definitively establishing a causal relationship between the given substance and human cancers (1) group 1 sufficient
evidence that the chemical is carcinogenic to humansi (2) group 2A probably carcinogenic to humans, based on limited information fees) human
data but sufficient or nearly sufficient data from animal studies; (3) group 2B probably carcinogenic to humans, but sufficient or nearly
sufficient evidence only exists for animal studies (the direct human data is inadequate)i and (4) group 3 not definitively classifiable as
carcinogenic, due to Inadequate information in both types of studies.
The "potency factor" is a relative assessment of the dose of the carcinogen associated with a 1 in 10 llftlme risk of cancer in the test species.
For 180 chemicals studied by IARC, the potency factors ranged from a low of 0.01 (for trypan blue and saccharin) to 200,000 (for
7,12-dimethybenx(a)anthracene). These are arbitrary reference points, but to give an Indication of where Exhibit 3-2 chemicals fit into this
scale, TCDD has a potency factor of 120,000i vinyl chloride one of 0.16. ECAO divides the ISO chemicals into four potency groups, group 1 having
the highest potency, group 4 the lowest.
Thus, the designation "high" refers to any chemical In potency group 1, or a potency group 2 chemical which also fails into evidence group 1
(definitive). The designation "medium* refers to chemicals in potency group 2 that are in evidence group 2A or 28, or to chemicals in potency
group 3 that are in evidence groups 1 or 2A. Finally, low" refers to a chemical in potency group 3 that is in evidence group 3, or to any
chemical to potency group 4.
Note 4s Flash point is the lowest temperature at which the material will ignite if exposed to a spark or flame.
Note b: Evidence indicates that the material (or a constituent of it) is bioaccumulated to toxic levels in the tissue of marine organisms, and thus has
the potential to concentrate in the food chain.
Note 6: These substances can, in some cases, be metabolized by bacteria to yield highly toxic organometalllc compounds.
Note 7i These compounds can be slowly blodegraded in the environment (all other compounds listed are persistent, especially in the groundwater
environment).
Note 81 Limited evidence of bioaccumulation exists.
Not* 9t Cyanides are highly inhibitory to microbes.
Note 10s Evidence exists that phenol may promote the development of tumors initiated by other agents.
Note Us A closely related compound, 1,1,2-trlchloroethane, was rated as a "low* carcinogenic hazard by ECAO.
tv)
I
-------
2-10
data base only codes a maximum of 15 chemicals per site regardless of how many
were reported.
Analysis of a list of more than 380 different chemicals detected at a
subset of NPL and non-NPL sites, compiled through EPA's Contract Laboratory
Program, reveals that many other classes of chemicals are found with some
regularity, including: (1) toxic pesticides such as DDT, lindane, chlordane,
hexachlorocyclopentadiene ("C-56"), and rairex; (2) strong acids and alkalis,
such as hydrochloric acid and ammonia; (3) oils and tars and their
derivatives; (4) asbestos; (5) radioactive substances, particularly radium
compounds, uranium compounds, and thorium compounds; and (6) dioxins. The
most frequently found dioxin compound is 2,3,7,8-tetrachlorodibenzo-p-dioxin
(TCDD); it is extremely toxic to aquatic and mammalian species and a potent
animal carcinogen.
Finally, it should be remembered that the specific chemicals reported do
not fully describe the range of substances found at NPL sites, because the
typical waste site contains an unclassifiable combination of mixtures. Some
materials will have entered the site in an amorphous state by virtue of the
complex manufacturing process from which the wastes emerged; other unusual
materials may form in situ when industrial wastes cross-react or are
transformed by microorganisms into new compounds. Scientific knowledge about
the health effects of these more unusual chemicals is fragmentary. Even more
important with regard to waste sites, however, is the paucity of scientific
information about the hazards of mixtures of individual substances, even when
the health threats of each compound singly are known. Very little is known
about the cumulative effects of exposure to "subthreshold" concentrations of
-each of-a variety of toxins; in particular, we are relatively ignorant of the
possible."synergistic" effects of combinations of substances, wherein chemical
or biological interactions make the hazard of the mixture greater than the sum
of the individual hazards.
2.2.2 Routes of Exposure
Observed Releases to Groundwater, Surface Water, and Air
In scoring and ranking NPL sites, the Hazard Ranking System records
whether there has been an "observed release" of one or more hazardous
substances to groundwater, surface water, or air in the vicinity of the site.
An "observed release" to one of these three media is recorded if quantitative
chemical analyses indicate that concentrations of one or more contaminants
exceed background levels in that medium. Data on observed releases to
groundwater, surface water and air for 5^6 NPL sites are reported in Exhibit
2-3. Groundwater releases were recorded at 75 percent of these sites;
releases to surface water were recorded for 56 percent of the sites; and
releases to air at 20 percent of the sites. Note, however that because the
NPL data system does not contain information on the actual concentrations
found, the statistics give no indication of the severity of releases to
environmental media.
-------
2-11
EXHIBIT 2-3
OBSERVED RELEASES TO ENVIRONMENTAL MEDIA
FOR 546 NPL SITES
Groundwater
Number
of Sites
Percent
of Total
Surface Water
Number
of Sites
Percent
of Total
Air
Number
of Sites
Percent
of Total
409 75 305 56 109 20
Source: NPL data base.
Potential Releases
In addition to the sites where releases have already been observed,
analysis of the NPL data base shows that there are a substantial number of
sites where an eventual hazardous substance release .is probable. The MRS
assigns the maximum "release value" of 45 "points in the groundwater, surface
water, or air pathway score if direct evidence exists of a hazardous substance
release from the facility in question. Even if no release has been observed,
however, MRS can assign a release value of up to 45 points in either the
groundwater or the surface water pathway,-' depending on the following
factors: (1) depth to the aquifer or distance to the nearest surface water;
(2) net annual precipitation;'(3) physical state of the waste (liquid wastes
are more likely to migrate faster than semi-solid or solid wastes); (4)
permeability of the underlying rock or soil (in the case of possible
groundwater contamination) or slope of the terrain between the site and nearby
surface water; and (5) degree of engineered containment at the site (liners,
leachate collection systems, sealed drums, etc.).
It must be noted, though, that because current scientific knowledge about
the transport and fate of chemicals in groundwater and surface water is
incomplete, and because our ability to characterize the existing groundwater
flow patterns at some sites is incomplete, it is difficult to predict with any
precision the likelihood and-extent of future releases.
p /
- HRS provides no category for "potential release" to the air medium, as
noted earlier; if no release has been observed, the entire pathway score is
scored zero.
-------
2-12
Exhibits 2-4 and 2-5 show the distribution of potential release scores for
those sites where groundwater or surface releases have not occurred. For the
groundwater pathway, 95 percent of sites have potential release scores above
15, and 70 percent have scores above 30 out of 45. For surface water, 38
percent have scores above 15 and 38 percent have scores above 30.
As a hypothetical example to provide a reference point, a site would
receive 33 points for groundwater release likelihood if it had the following
characteristics: an aquifer 21 feet below the site; 15 inches net
precipitation per year; sandy soil with a permeability of up to 300 meters per
year; liquid wastes; and the most risk-prone class of "containment" system
(unlined landfill, leaking containers, unstabilized waste piles, etc.).
2.2.3 Population Estimates
The NPL data base provides estimates on populations within a specified
linear distance of a waste site or within a specified distance of where wastes
have been found to be migrating off-site. For groundwater scores, MRS counts
persons who actually use groundwater from wells, but only those wells within
three miles of where hazardous substances are deposited or are known to have
migrated. Approximately 6.4 million people are calculated as using these
sources of drinking water. For surface water scores, HRS counts people who
use water drawn from an intake within three miles downstream of the facility
or waste site (or within one mile in static water bodies such as lakes and
ponds). .Approximately 6.0 million people use this type of water supply.
(Because of the complexity in the calculation of air scores in HRS, no
composite population estimates are reported here.)
Exhibit 2-6 summarizes these estimates of populations. The population
data for individual NPL sites fall in wide ranges. Because the values are not
normally distributed, outlying (i.e., very high population) values for a few
sites significantly elevate the average (mean) values. Therefore, median
values, most representative of populations at a typical site, are presented as
well. The corresponding median populations per site are about 5,000 for
groundwater and 3,600 for surface water.
Since linear distance is the only consideration that determines the size
of these populations, the estimates may in some cases be artificially high,
because hydrogeologic factors often .constrain the motion of pollutants in
groundwater in one direction only.
On the other hand, since so little is known about the long-range transport
of persistent compounds (or those that can be transformed into other toxic
compounds in the environment), some estimates of potentially exposed
populations may be too low. In particular, where sampling has not yet
determined the full extent of substance migration from a facility, counting
only those people within a certain distance of the known area of contamination
may fail to capture all of the possible existing risks and potential future
risks. In general, the population figures are aggregate estimates that do not
discriminate as to the severity of the risks each individual faces.
-------
2-13
EXHIBIT 2-4
NPL SITES WITH POTENTIAL RELEASES TO GROUNDWATER
Release Likelihood Score
Groundwater Pathway Number
(max. =45) of Sites Percentage
0 1 1
1-15 6 4
16-30 36 25
31-44 90 64
45 __9 __6
142 100
Source: NPL data base.
EXHIBIT 2-5
NPL SITES WITH POTENTIAL RELEASES TO SURFACE WATER
Release Likelihood Score
Surface Water Pathway Number
(max. =45) of Sites Percentage
0 116
1-15 12 6
16-30 98 50
31-44 74 38
195 100
Source: NPL data base.
-------
2-14
EXHIBIT 2-6
POPULATION ESTIMATES BY
ENVIRONMENTAL MEDIUM AT 546 NPL SITES
Environmental Number Median Mean Total
Medium of Sites b/ Population Population Population
Range
of Affected
Population
Size
Groundwater 333
Surface Water 70
5,000
3,602
19,303 6,428,000 4-640,000
85,674 5,997,000 2-999,999
a/ See text for interpretation of population estimates.
b/ Sites where population is greater than zero.
Source: NPL data base.
2.2.4 Sensitive Environments at Risk
Hazardous waste sites present real and potential hazards not only to human
populations but also to sensitive environments. The HRS defines "distances to
sensitive environments" as the distance from the location of the hazardous
substance (not the facility boundary) to an area containing an important
biological resource or to a fragile natural setting that could suffer an
especially severe impact from pollution. The NPL data base includes data on
the distance from NPL sites to three environmental settings: (a) critical
habitats of endangered species; (b) coastal wetlands; and (c) fresh water
wetlands.
Exhibit 2-7 summarizes the number of sites for which distances to
sensitive environments were evaluated, along with summary measures of
distances to the three environments. As summarized in the exhibit, freshwater
wetlands are potentially at risk from 186 NPL sites, located within a median
distance of 800 feet (0.15 miles). There are 40 NPL sites at a distance of
1,260 feet (0.24 miles) from coastal wetlands and 30 sites at a distance of
350 feet (0.07 miles) from critical habitats.
-------
2-15
EXHIBIT 2-7
DISTANCES TO SENSITIVE ENVIRONMENTS FROM 546 NPL SITES
Sensitive
Environment
Freshwater Wetland
Coastal Wetland
Critical Habitat
Source: NPL data base.
Number
of Sites
186
40
30
Median
Distance
(Feet)
800
1,260
350
Distance
Ranges
(Feet)
0-11,880
0-17,000
0-17, 000
EXHIBIT 2-8
AQUIFERS POTENTIALLY AT RISK
Aquifer Name
State(s) Served by Aquifer
Number of Sites
Cohansey
Brunswick
Raritan/Magothy
Foridan
Englishtown
Columbia/Potomac
Mount Laurel
Biscayne
Calumet
Prairie Duchien
Gulf Coast
Upper Troutdale
New Jersey
New Jersey
New Jersey, New York, Pennsylvania
Florida
New Jersey
Delaware
New Jersey
Florida
Indiana
Minnesota
Texas
Oregon, Washington
21
13
12
7
5
5
4
4
4
3
3
81
Source: MRS data.
-------
2-16
Other kinds of "environments" may also be potentially affected by MPL
sites. Many NPL sites are located on or near rivers and lakes'which are
important recreational or commercial areas, or above regional aquifers that
are important sources of water for irrigation and drinking. Others are
situated in scenic areas, forests, c.- agricultural land.
The cumulative effects of hazardous releases and waste sites on the same
resource are also a serious long-term threat to the environment and to human
populations. For example, Exhibit 2-8 lists 12 aquifers that are each
potentially at risk from 3 or more sites, based on an analysis of 877 sites
for which HRS data were available.
-------
3. SCOPE OF SUPERFDND RESPONSE
3.1 INTRODUCTION
CERCLA authorizes an extremely broad scope of Superfund response authority
to protect public health and welfare and the environment from the dangers of
hazardous substance releases. However, Congress deliberately excluded several
types of potential releases from the scope of the Superfund program. Some
types of releases were excluded entirely; other releases were excluded only
from certain portions of the program. In addition, EPA has, by policy,
limited response to certain other releases.
In the first years of the Superfund program, EPA directed primary
attention to the more conventional Superfund sites such as abandoned drums or
other waste disposal sites. This focus of the program is the basis for the
central estimates projecting Superfund needs discussed in the next chapter.
However, recently the focus of Superfund has been expanding to releases
other than conventional waste disposal sites. Most of these releases are
eligible for Superfund cleanup action under CERCLA, but fall outside EPA's
experience over the last four years. This chapter defines the scope of
Superfund response authority, the focus of the Superfund program as it has
developed over the past four years, and factors that aff.ect the' pace of
response efforts.
3.2 SCOPE OF SUPERFUND JURISDICTION
3.2.1 Releases Excluded Entirely from CERCLA
CERCLA authorities and requirements are triggered by a release or
threatened release into the environment of a hazardous substance or a
pollutant or contaminant which presents an imminent and substantial danger to
public health or welfare. Congress limited the definition of "release,"
however, by excluding three types of releases from the scope of the Act:-
Emissions from the engine exhaust of a motor vehicle,
rolling stock, aircraft, vessel, or pipeline pumping
station engine;
-Although the definition of release also excludes "any release which
results in exposure to persons solely within a workplace, with respect to a
claim which such persons may assert against the employer of such persons," the
legislative history makes clear that Congress intended that workplace releases
creating a risk of damage to human health or the environment would be subject
to all CERCLA response authorities. S. Rep. 848, 96th Cong., 2d Sess. 94
(1980).
-------
3-2
Releases of source, byproduct, or special nuclear
material from a nuclear incident, as those terms are
defined in the Atomic Energy Act of 1954 (e.g., nuclear
reactor incidents), and for the purposes of response
under CERCLA Section 104 or any other response action,
releases of source, byproduct, or special nuclear
material from any processing site designated under
section 102(a)(1) or 302(a) of the Uranium Mill Tailings
Radiation Control Act of 1978 (UMTRCA); and
The normal application of fertili2er.
Thus, even if these various releases involve ha2ardous substances (as defined
under CERCLA section 101(14)), they must be addressed through other statutory
and regulatory authorities.
In addition to limiting the definition of release, Congress limited the
definitions of "hazardous substance" and "pollutant or contaminant" by
excluding from .these terms:
(P)etroleum, including crude oil or any fraction thereof
which is not otherwise specifically listed or designated
as a hazardous substance under [section 101(14)1 ...,
and ... natural gas, natural gas liquids, liquified
natural gas, or synthetic gas usable for fuel (or
mixtures of natural gas and such synthetic gas).
(CERCLA sections 101(14) and 104(a)(2))
Therefore, releases of petroleum or related products are also not within the
scope of CERCLA. Such releases must be addressed through other authorities,
such as section 311 of the Clean Water Act.
3.2.2 Releases Ineligible for Fund-Financed Remedial Action
Section 111 of CERCLA specifies the purposes for which Fund monies may be
used. Although most government response costs may be reimbursed from the
Fund, CERCLA section 111(e)(3) provides that no money from the Fund shall be
made available for remedial action with respect to federally owned facilities
(except for certain actions listed in CERCLA section 111(c), such as natural
resource damage assessment or restoration or health effects studies).
Therefore, under CERCLA, releases at federal facilities that require more than
a short term removal action must be financed from a source other than the Fund,
Even though Fund-financed remedial actions may not be taken at federal
facilities, EPA has recently decided to list releases discovered at federal
facilities on the NPL, in order to focus public attention on the most
hazardous federal sites and to encourage appropriate cleanup actions.
-------
3-3
3.2.3 Releases for which Response Costs May Not Be Recovered from
Responsible Parties
Section 107 of CERCLA sets forth the circumstances under which a
responsible party may be held liable for response costs and natural resource
damages resulting from a release or threatened release of a hazardous
substance. Several types of releases do not subject the responsible party to
liability for response costs. Most Importantly, only costs incurred in
responding to a release or threatened release of a designated "hazardous
substance" may be recovered from responsible parties under CERCLA.
Responsible parties are not liable for costs incurred in an authorized
response to a release or threatened release of a "pollutant or contaminant."
Other releases for which response costs are not recoverable from
responsible parties under CERCLA include:
Releases resulting from actions taken or omitted in a
manner that is not grossly negligent in the course of
rendering care, assistance, or advice in accordance with
the National Contingency Plan or at the direction of a
federal on-scene coordinator in response to a release
(CERCLA section 107(d));
Releases resulting from application of a pesticide
product registered under the Federal Insecticide,
Fungicide, and Rodenticide Act (CERCLA section 107(i));
and
Releases which are "federally permitted releases" as
defined by CERCLA section 101(10) {CERCLA section
107(J)). Current EPA policy defines federally permitted
releases as releases of substances that are explicitly
included in the permit or that have been considered in
the permitting process.
3.2.4 Releases Under Other Regulatory Authorities
EPA has established a policy that it will not use Superfund authorities to
respond to certain releases when there are adequate response authorities under
other laws. Certain radioactive releases and releases from currently
operating RCRA facilities fall into this category.
Radioactive materials: Radionuclides are hazardous substances as defined
by CERCLA. As noted above, CERCLA section 101(22) specifically excludes from
Superfund certain radionuclide releases regulated under the Atomic Energy Act
and the Uranium Mill Tailings Radiation Control Act. EPA has chosen not to
use Superfund authorities to respond to releases from facilities with
effective materials licenses from the Nuclear Regulatory Commission, because
the Agency believes that the Commission has the authority to control releases
from such facilities. However, EPA will respond to releases from facilities
formerly but not currently licensed by the Commission, and from facilities
-------
3-4
licensed by states under agreement with the Nuclear Regulatory Commission, as
well as releases of mixed radionuclide and other hazardous wastes. EPA will
also respond to releases from mining and processing facilities not designated
by UMTRCA.
RCRA facilities: Where a hazardous substance is specifically identified
and made subject to a standard of practice, control procedure, or bioassay
limitation or condition in a legally enforceable final permit issued under
RCRA, a release of that hazardous substance in compliance with the permit is a
federally permitted release under CERCLA section 101(10)(E). Although other
releases from permitted RCRA facilities and all releases from interim status
facilities are within the scope of Superfund, EPA has determined that RCRA
authorities (including groundwater protection and enforcement authorities) are
more appropriate to deal with releases from regulated units of these
facilities. In fact, RCRA has recently been strengthened to require
corrective action and financial assurance for corrective action at RCRA
facilities. The Superfund program continues to focus on releases from
abandoned hazardous waste facilities and sites rather than on releases from
operating facilities.
3.3 CURRENT FOCUS OF THE REMEDIAL PROGRAM
EPA has developed an orderly process for moving from site discovery to
investigation to NPL listing, as described in Chapter 1. .While the ERRIS .data
base continues to grow with new potential hazardous release sites (from 13,000
sites in 1982 to about 18,900 as of September 30, 1984), the primary emphasis
in the program has shifted somewhat from discovery in the first two years
to investigation of potential releases and listing of the more serious sites
so that remedial cleanup can begin.
Discovery efforts were initially directed to completing the notification
process required under CERCLA section 103(c) and to identifying and
consolidating other data bases that contained potential hazardous release
sites. ERRIS grew significantly as states began to enter sites of concern.
Since then, both EPA and the states have continued to identify additional
potential sites, as well as receiving reports from the public and
notifications of hazardous substances releases.
These discovery efforts appear to be identifying many of the more serious,
the more obvious, and the more suspicious sites. However, since EPA has
placed a higher priority on completing investigations so that cleanup can
begin, the Agency has not undertaken targeted, systematic discovery efforts to
identify sites which will require more intensive detective work.
3.3.1 Types of Faculties on the NPL
EPA is currently in the third cycle of adding sites to the NPL. As of
October 1984, 538 sites are listed on the final NPL. EPA has proposed 248
additional sites for listing and is now in the process of reviewing and
responding to comments before determining which of these 248 sites will be
-------
3-5
added to the final NPL. These additional sites are identified as "Update 2"
sites. Proposed NPL sites are eligible for remedial planning activities but
not for remedial cleanup under CERCLA.
Exhibit 3-1 shows the types of facilities or activities represented by the
current NPL sites and Update 2 sites. As can be seen, about 80 percent of :he
sites involve wastes held in surface impoundments, drums, and landfills.
These types of waste facilities continue to represent the bulk of the sites,
both on the current NPL and on Update 2.
There are, however, several minor indications that the types of sites to
be addressed in the future may have a different make-up than the current NPL.
Three types of changes are discussed below.
(1) Shifts in Broad Categories of Activities. Exhibit 3-2 details the
percentage of final NPL and proposed Update 2 sites characterized by different
activities, breaking the site activities down into six categories: storage,
disposal, transportation and treatment, recovery and recycling, manufacturing
and chemical processing, and other. The exhibit shows a slight decrease in
sites characterized as disposal facilities and recovery and recycling
facilities. While the differences between the final and proposed NPL sites
are minpr and may even be a function solely of classification methods, they
may indicate a slight shift away from traditional types of Superfund sites.
(2) Shifts in HRS Scores. The proposed Update 2 sites generally score
lower on the HRS than the final NPL sites. Exhibit 3-3 shows the percentage
distribution of the HRS scores of final and proposed NPL sites. HRS scores
for Update 2 sites are concentrated between 28.5 to 45. Final NPL sites tend
to have higher and more evenly distributed scores than Update 2 sites.
(3) Clusters of Nontraditional Sites. Several new types of facilities
are present in the Update 2 sites. Federal facilities are included on the
proposed list; 36 of the 248 Update 2 sites (15 percent) are maintained by the
federal government. Sites with contamination due to the agricultural use of
pesticides are also a new entrant. Six such sites in Hawaii are included on
the Update 2 list and additional pesticide sites may become more common in the
future, especially on cotton and tobacco fields in the southeastern United
States. In addition, leaking underground storage tanks are responsible for a
growing percentage of proposed NPL sites. Of the 34 non-federal Update 2
sites in California, 19 (56 percent) represent leaking underground storage
tanks. These sites are concentrated in the South Bay area surrounding San
Jose, California, which includes "Silicon Valley," and have been identified as
causing groundwater contamination.
3.3.2 Factors Affecting the Pace of Response Efforts
Starting in FY83, the pace of the Superfund program has accelerated
significantly. In FY83, there were 115 remedial investigation/feasibility
study (RI/FS) starts; in FY84, the Agency projected approximately 120 starts.
Budget projections for FY85 are 115 starts. There are a number of constraints
-------
3-6
EXHIBIT 3-1
TYPES OF ACTIVITIES AT NATIONAL PRIORITIES LIST SITES
Final MPL Update 2» Total NPL
Activity
Surface Impoundment
Drums
Landfills, NOS
Commmercial/Industrial
Landfill
Open Dump
Municipal Landfill
Tank, Above Ground
Piles
Transporter
Chemical Manufacturing
Recycling/Recovery
Tank, Below Ground
Well Field
Chemical/Physical
Other. Manufacturing
Incineration
Wood Preserving
Waste Oil Processing
Midnight Dumping
Solvent Recovery
Spill Site
Sand/Gravel Pit
Mining Site
Military Ordinance
(Uncodable Data)
Ore Processing/Smelter
Electroplating
Storage, NOS
Biological
Open Burning
Battery Recycling
Road Oiling Site
Landfarm
Wetland
Foundry
Underground Injection
Laundry/Dry Cleaning
Drum Recycling
Number
183
134
118
107
113
82
67
54
50
45
46
34
39
33
16
32
28
32
32
25
9
15
15
3
18
16
10
7
4
3
7
7
7
3
6
4
3
4
Percent
34.01
24.91
21.93
19.89
21.00
15.24
12.45
10.04
9.29
8.36
8.55
6.32
7.25
6.13
2.97
5.95
5.20
5.95
5.95
4,65
1.67
2.79
2.79
0.56
3.35
2.97
1.86
1.30
0.74
0.56
1.30
1.30
1.30
0.56 .
1.12
0.74
0.56
0.74
Number
36
57
50
27
21
18
26
31
16
18
12
23
15
14
30
6
9
3
2
6
20
13
3
15
0
1
7
a
5
6
2
1
0
4
0
1
1
0
Percent
35.25
23.36
20.49
11.07
8.61
7.38
10.66
12.70
6.56
7.38
4.92
9.43
6.15
5.74
12.30
2.46
3.69
1.23
0.82
2.46
8.20
5.33
1.23
6.15
0.00
0.41
2.87
3.28
2.05
2.46
0.82
0.41
0.00
1.64
0.00
0.41
0.41
0.00
Number
269
191
168
134
134
100
93
85
66
63
58
57
54
47
46
38
37
35
34
31
29
28
18
18
18
17
17
15
9
9
9
8
7
7
6
5
4
4
Percent
34.22
24.30
21.37
17.05
17.05
12.72
11.83
10.81
8.40
8.02
7.38
7.25
6.87-
5.98
5.85
4.83
4.71
4.45
4.33
3.94
3.69
3.56
2.29
2.29
2.29
2.16
2.16
7.91
1.15
1.15
1.15
1.02
0.89
0.89
0.76
0.64
0.51
0.51
-------
3-7
EXHIBIT 3-1 (continued)
TYPES OF ACTIVITIES AT NATIONAL PRIORITIES LIST SITES
Final NPL Uodate 2* Total MPL
Activity Number Percent Number Percent Number Percent
Laboratory
SW Outfall
Sink Hole
Underground Fire
Explosive Disposal
2
3
3
2
0
0.37
0.56
0.56
0.37
0.00
1
0
0
0
1
0.41
0.00
0.00
0.00
0.41
3
3
3
2
1
0.38
0.38
0.38
0.25
0.13
Total Sites: 538 248 786
*Includes 4 site's continued to be proposed from Update 1.
Source: NPL data base.
-------
3-8
EXHIBIT 3-2
SUMMARY OF NPL SITE ACTIVITIES
Percent of Percent of
Final NPL Sites Proposed Update 2
Sites Characterized Sites Characterized
Activity a/ by Activity b/ by Activity b/
Storage
Disposal
Transportation and Treatment
Recovery and Recycling
Manufacturing and Chemical Processing
Other
41.9
24.6
6.1
8.0
13.9
5.4
45.1
18.2
6.3
4.1
17.5
8.8
a/ Storage includes: surface impoundment, piles, drums, above ground
tanks, below ground tanks, storage NOS, open dumps, and sink holes.
Disposal includes: Landfills NOS, commercial/industrial landfills,
municipal landfills, landfarms, underground injection, and incineration'.
Transportation/Treatment includes: Biological treatment, chemical/
physical treatment, and transporters.
Recovery/Recycling includes: Drum recovery, battery recycling, waste oil
processing, solvent recovery, and recycling/recovery.
Manufacturing/Chemical Processing includes: Chemical manufacturing, other
manufacturing, wood preserving, electroplating, laboratories, well fields,
laundry/dry cleaning, sand/gravel pits, mining sites, ore processing/smelters,
foundries, and surface water outfall.
Other includes: Midnight dumping, spill sites, military ordnance, open
burning, road oiling sites, wetlands, underground fires, explosive disposal,
and uncodable data.
b/ Since a single facility may have more than one activity, percentages
were calculated by summing the occurrences in each activity grouping and then
dividing by the total number of activities present at the 538 Final NPL sites
and the 248 Proposed Update 2 sites.
Source: NPL data base.
-------
3-9
EXHIBIT 3-3
NATIONAL PRIORITIES LIST
HAZARD RANKING SYSTEM SCORE DISTRIBUTION
28.5-33 33-39 39-:45 4S-51 51-57 57-83 63-69 69-75 75-81
MRS SCORE QROUP
Final NPL Sites <% Of 538 Sites)
Proposed Update 2 Sites (% Of 248 Sites)
Source: N0L data base.
-------
3-10
which affect the pace of the program, and which must be considered carefully
in any effort to speed up the level of activity.
Capacity of RCRA Facilities: There is a concern about the extent to
which fully permitted treatment, storage and disposal facilities will be
availaole to dispose of Superfund wastes from remedial or removal actions.
EPA's experience over the past several years has shown chat many of the
facilities that have been interim status under the Resource Conservation and
Recovery Act are in violation of requirements for such facilities, and
therefore may be unacceptable for use for storage, treatment, or disposal of
Superfund wastes.
In order to prevent future problems resulting from disposal of Superfund
wastes, EPA requires recent inspection of disposal facilities before Superfund
wastes can be disposed. In addition, EPA is considering a requirement for use
of lined land disposal units for Superfund waste and establishing a preference
for use of treatment over land disposal. This may limit the number of
facilities which are immediately available to accept Superfund waste. In
addition, while there are many new and existing facilities scheduled for
review to receive a final permit, many of these facilities are likely to be
unable to receive final RCRA permits until the facilities take additional
action to meet regulatory requirements. Because of these problems, it is
possible that there may be inadequate treatment and disposal capacity under
RCRA to meet Superfund needs. The Agency is assessing this problem further.
There are also serious concerns with the long-term impacts of continuing land
disposal of hazardous substances. The Agency's ability to shift to treatment
alternatives depends on the availability of proven treatment technologies.
Alternative Technologies: The rate at which the federal and state
governments and private industry can develop and test more advanced treatment,
recycling and destruction technologies also affects available waste disposal
capacity. The range of technologies that are used most frequently at
Superfund sites generally do not focus on advanced treatment, recycling, or
destruction. Host of the technologies that are used fit into the following
general categories:
Treatment technologies, such as activated carbon units.
Many of these still require disposal of highly
concentrated/low volume wastes that result from the
treatment process. For example, expended activated
carbon requires regeneration or the removal and further
handling or disposal of residual substances.
Containment technologies, such as slurry walls, caps and
barrier wells. The program is gaining experience with
these technologies. However, there are two important
factors that affect their long-term effectiveness. Most
of these require long-term maintenance to ensure
adequate containment of hazardous wastes. Also the
long-term effects of hazardous substances on the
materials used in containment structures are unclear and
-------
3-11
must be carefully evaluated during the remedial
investigation/feasibi1ity study.
Storage technologies, such as covered waste piles, are
used generally as interim remedies until a more
permanent remedy can be implemented.
Destruction technologies, such as incineration, are very
effective. However, their application has been limited
principally to liquids and sludges.
In situ stabilization, such as solidification or
neutralization, are effective, but generally, these
technologies are applicable only to certain waste types.
In summary, there is still a great need for technologies that will result
in handling wastes only once, without long-term operation and maintenance
requirements, and without the need to revisit a potential threat of release in
the future.
Laboratory Support: The pace of remedial actions .is affected by the rate
at which specialized support industries, most notably analytical laboratories,
will be able to expand to meet the demand from public and private sector
clients to provide rapid, accurate analysis of hazardous waste samples. EPA
has roughly doubled its demand on private sector analytical laboratories each
yeap over the past four years. With'the increased-number of site inspections
and remedial investigations that EPA has' initiated, coupled with those that
may be initiated by the states and the private sector, EPA anticipates that
there will be a strain on these industries to secure the necessary equipment
and personnel to expand rapidly and still provide high quality results.
State Capabilities: The rate at which states expand their programs in
order to provide state cost shares and assume the lead at a greater number of
sites is a key factor in the pace of the program. The record of state
participation in the program is discussed in the section 301(a)(1)(E) report
to Congress. States have made significant advances in overcoming the hurdles
of providing staffing, funding and other resources, and building a supporting
administrative structure for contracting, recordkeeping and financing.
However, there is still a gap between'state capabilities and the number of
sites that remain to be addressed. The high proportion of federal-lead to
state-lead sites does relieve states of some program costs and administrative
burden. However, the large number of sites that will be moving into the
construction phase in the next one to two years will place a heavy strain on
the states in meeting their share of remedial action costs.
Effective Program Management: Another key factor is the rate of remedial
activity which EPA can effectively oversee and the rate of cleanup
decision-making. Some degree of EPA involvement in and oversight of remedial
planning is essential to ensure that the activities remain consistent with
CERCLA goals. In addition, decision-making on the appropriate remedy involves
careful analysis and balancing of complex issues relating to the extent of
-------
3-12
cleanup, the impacts of alternative remedies, and the effectiveness of various
technologies. Accelerating the number of cleanup decisions required each year
beyond a certain leve: could have an impact on the quality of decision-making.
Private Industry Cleanup: The extent to which private industry is
willing to accept a larger share of the burden of the site cleanup could
affect program progress. During the past year, EPA and the states have
experienced a trend of increasing participation by private industry in the
cleanup program, as evidenced by an increasing number of settlements, better
organization on the part of responsible parties involved in cleanups, and the
formation of a private nonprofit consortium to manage-site cleanup for private
parties. The extent to which private industry can continue to organize its
resources will affect the pace of the overall program.
Special Expertise: A final factor is the extent to which specialized
professionals, such as chemists, toxicologists and hydrogeologists become
available to address technical problems associated with site cleanup. EPA has
encountered a shortage of experienced personnel with specialized skills. For
example, it takes years for a chemist to become fully experienced in the
techniques of operating and interpreting the data output from highly
specialized laboratory equipment. It also takes years.for toxicologists to
develop and interpret much needed information on the effects of chemical
substances on human health and the environment. While the federal and state
governments and private industry may develop the administrative capability to
initiate action at many sites, the quality of the information used to make
decisions and the rate at which decisions can be made will be seribusly
hampered if these critical skills remain in short supply.
In summary, it is difficult to predict which of these constraints are more
significant than others. However, it is apparent that any one of these can
have a serious effect on the rate of program activity. It is also difficult
to predict if additional constraints will develop over the next several years,
since some of the constraints mentioned here only became apparent recently.
It is clear, however, that maintaining even the current pace of the program
will take close cooperation between EPA, other federal agencies, the states,
and private industry to deal with these constraints.
-------
4. CENTRAL ESTIMATES OF FUTURE FUNDING NEEDS
4.1 INTRODUCTION
This chapter presents a baseline estimate of the future funding needs of
the Superfund program. The estimate is based on certain key assumptions and
estimates developed by EPA over the last four years of experience with the
program. In addition to the baseline estimate, we also present, in a
sensitivity analysis, a range of central estimates that could result if key
assumptions are changed.
The estimates of Superfund's future funding needs developed in this
chapter are based on a projected scope of the Superfund program that largely
resembles the current scope of the program, as discussed in Chapter 3- The
Agency expects some shifts in the types of problems that are represented on
the NPL, as well as an increase in the number of sites that are identified by
Superfund. However, the projections in this chapter do not take account of
massive increases in the focus and size of the Superfund program. The
potential for a greatly expanded Superfund program is addressed separately in
Chapter 5 of this report.
4.2 FACTORS AND ASSUMPTIONS USED IN FUNDING PROJECTIONS
The major factors and assumptions that are used in this study to estimate
future funding needs are discussed below under the following program headings:
Remedial action program costs;
Other program costs; and
Superfund revenues.
For easy reference, Exhibit 4-1 summarizes the factors and assumptions
discussed in the text which are used to develop the baseline estimate.
4.2.1 Remedial Action Program Costs
Remedial actions are the single largest category of Superfund expenditures
and are expected to remain so in the future. In order to project funding
needs for the remedial program, several key factors must be estimated. They
are:
(1) the projected size of the NPL;
(2) the average cost of a remedial action; and
-------
4-2
EXHIBIT 4-1
SUMMARY OF FACTORS AND ASSUMPTIONS USED
IN BASELINE FUNDING ESTIMATE
Projected Cost Elements
Value
Remedial Actions
Size of NPL
Average cost of remedial actions
Responsible party share of direct cleanups
Removal Actions
Annual number of removals
Average cost of removal action
Program Support and Claims
Program support
Mo cost projections available for natural
resource damage claims
1,800 sites
$8.1 million
50 percent
190
$350,000
Variable; based on
formula
Projected Revenues
Value
CERCLA Tax
No taxes projected beyond FY85 when
taxing authority expires
Appropriations
No appropriations projected after FY85
Interest Income
No interest projected beyond FY85
Cost Recoveries
Removal action costs
Remedial action costs
47 percent recovered
30 percent recovered
-------
4-3
(3) direct cleanup actions by responsible parties.
Each factor is discussed in turn below.
(1) Projected Size of the NPL
The size of the National Priorities List defines the scope of the remedial
action program, since only NPL sites are eligible for remedial action. As
discussed in previous chapters, the size of the NPL under current Superfund
policies has been primarily determined by two factors: the number of sites
identified in EPA's site inventory (ERRIS); and the process of screening and
inspecting sites and then ranking them for inclusion on the NPL.
Each of these factors could change in the future, thereby affecting the
future size of the NPL. First, the number of sites identified on the ERRIS
data base is expected to grow, particularly if EPA initiates a systematic and
aggressive site discovery effort. Sources of this growth will include new
discoveries of sites, newly abandoned or created sites, and previously
underrepresented sites such as radioactive wastes, mining waste sites, sites
on Native American lands, and federal facilities. While ERRIS currently
contains about 19,000 sites, over the next several years the inventory could
grow by 15-30 percent, to some 22,000 to 25,000 sites.
A second source of change in the make-up of the future NPL could come from
the process of investigating and scoring sites for inclusion on the list.
EPA's experience to date is that approximately one out of every 12 sites
receiving preliminary assessments is -eventually listed on the NPL. Whether
-this historical trend will continue is uncertain. A lower ratio (say,'15 to
1) could occur if EPA has already identified the worst sites and is now
assessing less threatening sites. On the other hand, a higher ratio could
occur (say, 10 to 1) if sites added to ERRIS in the future have already been
more thoroughly screened and are entered because there is a significant
potential for a hazardous release. It is likely that a higher proportion of
these sites would be added to the NPL.
In addition, it may be that over the next several years, a substantial
number of sites will be added to the NPL without having first been entered
into ERRIS and subjected to the assessment and investigation process detailed
earlier. For example, 19 sites proposed for the second NPL update involve
leaking underground storage tanks in California. These sites were not
initially included on ERRIS. They were identified by the state and submitted
directly for scoring. It is possible that in the future a substantial number
of "nontraditional" sites will be added in this way.
The resulting size of the NPL under each of these assumptions is
summarized in Exhibit 4-2. The results range from 1,500 sites on the NPL to
2,500 sites, based on different growth rates of ERRIS and different ratios of
ERRIS sites to NPL sites. The baseline funding estimate developed in this
analysis assumes 22,000 sites on ERRIS and a ratio of 12;1 for ERRIS sites to
NPL sites, yielding an NPL of about 1,800 sites.
-------
4-4
EXHIBIT 4-2
PROJECTED SIZE OF THE NPL
Ratio of Sites Size of ERRIS
on ERRIS 19,000 22,000 25,000
to NPL Size Sites on ERRIS Sites on ERRIS Sites on ERRIS
(Current)
12:1 (Current) 1,500 1,800 2,100
15:1 1,500 1,700
10:1 2,200 2,500
(2) Average Cost of a Remedial Action
To determine the average federal cost per site for future remedial
actions, three factors must be considered:
Average remedial planning costs (including remedial in-
vestigation/feasibility study and remedial design costs);
Average remedial action costs (including capital and
operation and maintenance costs); and
Distribution of federal vs. state cost shares.
As shown in Exhibit 4-3, the average remedial investigation/feasibility
study (RI/FS) cost is $800,000, and the average remedial design cost is
$440,000. These costs are borne solely by the federal government. The
average remedial cost per site for NPL sites is estimated-at $7.2 million for
capital costs and $0.4 million for operation and maintenance costs during the
first year. In general, the federal share of these costs is 90 percent of the
total. Thus, the average federal cost per site is estimated to be $8.1
million.
It is important to note that there is some uncertainty connected with the
estimates in Exhibit 4-3. In particular, the model used to derive the
estimate of average capital construction cost and O&M cost per site is being
updated based upon decision criteria that will be proposed in forthcoming
-------
4-5
EXHIBIT 4-3
AVERAGE REMEDIAL COST PER SITE
Cost Element
Current Estimate
Federal
Share
State
Share
Total Cost
Average cost of remedial
investigation/feasibility study a/
Average cost of remedial design a/
Average capital construction cost
per site b/ c/
Average operation and maintenance
cost per site b/ d/
Average total cost per site
$ 800,000
440,000
6,480,000 720,000
360,000
40,000
$ 800,000
440,000
7,200,000
400,000
$8,080,000 $760,0'00 $8,840,000
a/ Source: U.S. EPA.
b/ Source: Charles River Associates, Costs of Remedial Actions at
Hazardous Waste Sites, Report to EPA, December 1983t P- 31.
c/ Estimates subject to revision.
d/ First year cost only. Subsequent O&M costs assumed to be borne by the
state.
-------
4-6
amendments to the National Contingency Plan.1 Preliminary results from this
model update suggest that per site costs will be different and may be higher
than reported here. The revised model results currently are undergoing review
within EPA. Although uncertain the estimated average federal cost per site
of $S.l million per remedial act.'on appears reasonable in light of current
experience and has been used in preparing EPA's FY85 budget.
The total remedial costs for any particular site actually will be incurred
over several years -- three years on average. About 90 percent of the cost
occurs during construction in the last year. In modeling funding
requirements, a three year period is used, and the cost for each phase of the
remedial action is assumed to be incurred in the year that the phase begins.
(3) Direct Cleanup Actions by Responsible Parties
Responsible parties (RPs) contribute to the hazardous waste cleanup effort
by financing or undertaking voluntary or enforced response actions, thereby
reducing the number of remedial actions requiring government funding. EPA is
now developing and implementing policies designed to encourage RP cleanups.
If these policies are successful, EPA hopes to achieve approximately a 50
percent rate for the number of sites that will be cleaned up by RPs without
use of the Fund. These activities do not include costs recovered from
private parties for Fund-financed remedial actions. It is assumed that the
average remedial construction and O&M costs for RP actions will be the same as
the average remedial construction and O&M costs for Fund-financed activities
presented in Exhibit 4-3 (i.-e., that RP and Fund-financed sites will not
differ systematically in cost).
4.2.2 Other Program Activity Levels and Costs
Future cost estimates for removal actions are based on an expected
constant activity level of 190 removals per year. The cost for this level of
activity is estimated to be $75 million per year in FY83 dollars.
Program support costs are estimated by EPA's Superfund Forecasting Models
based upon projected removal and remedial activity levels. They include
expenditures on research and development, enforcement, management and support,
and interagency support. Program support costs vary annually as program
activities vary.
*For a description of the model and development of the estimates used in
this report, see Charles River Associates, Costs of Remedial Actions at
Hazardous Waste Sites, report to EPA, December 1983. The capital and O&M
costs were estimated based on a sample of 82 sites from EPA Regions I, II, and
III. The average cost figures were derived based on the costs associated with
cleanup of different problem types and the proportion of these problems on the
NPL. The average per site cost does reflect the cost of on-site and off-site
groundwater cleanup, where appropriate. Recent projections for the extent of
remedial action consider more comprehensive levels of cleanup, suggesting that
future costs per site may be higher than those based upon past program
experience.
-------
4-7
No future expenditure estimate is made for natural resource damage
assessments and restoration. To date, this cost element has been
insignificant. For example, EPA's FY84 budget contained $989,700 for natural
resource claims, most of which has not been spent and will be carried forward
to FY85. No natural resource damage claims have been approved for the FY85
budget.
As a cost element, natural resource damage claims could be significant
(i.e., up to the maximum statutory limit of 15 percent of the Fund), but an
estimate of the actual level of activity would be speculative at this time.
Some increase in activity is expected. However, no costs for natural resource
damage assessments and restoration are included in the central estimates
developed here. Any such costs actually occurring would increase the
estimates of future funding needs developed in this report by an amount equal
to those costs.
4.2.3 Superfund Revenues
Superfund revenues currently come from several sources: the CERCLA tax,
Congressional appropriations from general revenues, interest earned on the
invested balance of Superfund, fines, and costs recovered from responsible
parties under authority of CERCLA. The purpose of projecting funding needs in
this study is to identify total Superfund program needs after the current
taxing authority expires. Therefore, under the baseline assumptions, future
funding sources will consist entirely of cost recovery contributions.
However, it is anticipated that a reauthorized Superfund program would receive
funding from essentially the same sources as the current program.
Based on past experience, the costs of Fund-financed actions are assumed
to be recovered from responsible parties at a rate of 47 percent for removals
and 30 percent for remedial actions. A three year lag time between
expenditures and reimbursements is assumed. The estimated cost recovery rate
for remedial expenditures is subject to a range of possible estimates given
varying assumptions about: (1) the probability of successful litigation; (2)
the ability to recover interest costs from litigation; and (3) how many NPL
sites will be cleaned up by private parties without use of the Fund.
The projected 30 percent cost recovery rate for remedial actions is based
on the following rationale. To date, Fund-financed action has been taken at
sites where responsible parties are potentially capable of taking on cleanup
responsibility. The cost recovery rate for these sites is likely to be higher
than 30 percent. In future years, as EPA enforcement actions are more
successful, more RPs will undertake actual cleanups. As a result, more
Fund-financed actions will be taken at sites where there is less likelihood of
cost recovery because there are fewer RPs connected with the sites. The cost
recovery rate for these sites may be considerably lower. An overall cost
recovery rate of 30 percent balances these two phases of enforcement action.
In addition, it is important to note that cost recovery rates do not represent
total responsible party contributions to cleanup efforts. These rates do not
include projections of direct cleanup action by responsible parties described
in Section 4.2.1.
-------
4-8
4.3 BASELINE ESTIMATE OF FUTURE FUNDING NEEDS AND
SENSITIVITY ANALYSES
This section presents central estimates of future funding needs for the
Superfund program, calculated by applying the assumptions and data elements
described in the previous sections to EPA's Superfund Budget Forecasting Model
and Financial Flows Model.- We provide both a "baseline estimate" using
best available information on future funding needs, as well as a range of
central estimates using sensitivity analyses.
4.3.1 Baseline Estimate
The baseline estimate of future funding needs incorporates the most likely
assumptions and best available estimates. The baseline estimate assumes the
following values, as indicated in Exhibit 1U1:
Size of the NPL 1,800 sites;
Average federal remedial action cost $8.1 million; and
Fund vs. RP contribution 50 percent of sites are
expected to represent RP actions.
Using these assumptions, to respond to all sites on an 1,800-site NPL,
future funding needs are estimated at $11.7 billion in constant FY83- dollars.
This estimate is net of costs recovered from responsible parties over the life
of the program.
4.3.2 Sensitivity Analyses: Range of Central Estimates
Sensitivity analyses were performed on the baseline estimate of funding
needs by varying three factors: (1) size of the NPL, (2) average federal
remedial cost, and (3) responsible party contributions. The alternate values
for each factor are described below.
Size of the NPL; In addition to the central estimate of
1,800 sites, we examine the effect on future funding
needs of an NPL with 1,500 sites, 2,200 sites, and 2,500
sites, based on projected ERRIS sizes of 22,000 sites
and 25,000 sites and different ratios of ERRIS sites to
NPL sites.
2/ The Superfund Budget Forecasting Model estimates future program costs
based upon pricing assumptions and forecasts of program activity. Model
outputs include costs for administration and enforcement, research and
development, and Superfund response. These outputs and revenue estimates are
principal inputs to the Financial Flows Model. Other inputs to this
accounting model include assumptions on cost recovery lags, interest rates,
and inflation rates. Outputs of the Financial Flows Model include ending fund
balances in FY83 dollars.
-------
4-9
Average federal remedial cost: In addition to the
baseline estimate of $8.1 million, estimates of $6
million, $10 million, and $12 million are examined in
the sensitivity analyses, reflecting the costs of
cleaning up less severe sites (on the low end) and the
costs of complying with additional remedial
requirements, such as RCRA regulations and other
environmental and health standards for levels of cleanup
(on the high end).
Responsible party contribution; The baseline estimate
of 50 percent Fund-financed/50 percent RP cleanups is
varied from a low of 40 percent Fund/60 percent RP to a
high of 60 percent Fund/40 percent RP cleanups.
Ten sensitivity analyses were performed, and are briefly discussed below.
They are identified by their assumptions concerning NPL size, average federal
remedial cost, and the ratio of Fund-financed sites to RP-financed sites.
These assumptions and the resulting range of central estimates of funding
needs are summarized in Exhibit 4-4. All results are net of cost recoveries.
None of the estimates includes funds for natural resource damage claims; any
natural resource damage awards paid would directly increase the total Fund
requirements. To ease the comparison across cases, future funding needs are
described in billions of constant FY83 dollars.
Sensitivity Analyses #1 and 12 represent a range of funding requirements
from the least expensive to the most expensive likely to occur under current
policies. Assumptions in Sensitivity Analysis #1 are: $6 Million federal
remedial action cost; a 40/60 Fund/RP ratio; and 1,500 NPL sites. One
hypothesis to support these assumptions is that EPA has already identified the
nation's most hazardous waste sites. Future additions to the NPL will
therefore represent less severe sites and less costly remedies than those
currently being dealt with. Under these assumptions, Fund requirements would
total $7.6 billion (FY83 dollars). Sensitivity Analysis #2 reflects
assumptions at the other end of the spectrum: a $12 million federal remedial
action cost; a 60/40 Fund/RP split; and a 2,500 site NPL, Under these
assumptions, Fund requirements would total $22.7 billion (FY83 dollars).
SensitivityAnalyses #3» $4, and #5 vary the baseline estimate by varying
only the average federal remedial action cost per site and leaving the other
assumptions intact. At the low end, a cost estimate of $6 million results in
a decrease in program funding requirements to $10.0 billion. Increases in the
cost estimate to $10 million and $12 million result in increased funding
requirements, to $13.3 billion and $14.9 billion (FY83 dollars), respectively.
Sensitivity Analyses #6 and #7 differ from the baseline estimate by
varying the estimate of responsible party contributions 10 percent either
way. With a 40 percent Fund/60 percent RP split, Fund requirements would
total $10.5 billion (FY83 dollars). In the case of a 60 percent Fund/40
percent RP split, Fund requirements would total $12.6 billion (FY83 dollars).
-------
4-10
EXHIBIT 4-4
BASELINE AND CENTRAL ESTIMATES OF FUTURE FUNDING NEEDS
Sensitivity
Analysis
Low, Baseline, and
#1
Baseline
#2
Alternative Federal
#3
#4
*5
Alternative Fund/RP
//6
#7
Average Federal
Remedial Cost
(Millions
of FY83 $)
High Estimates
6.0
8.1
12.0
Remedial Costs
6.0
10.0
12.0
Contributions
8.1
8.1
Fund/RP
Contributions
(%)
40/60
50/50
60/40
Per Site
50/50
50/50
50/50
40/60
60/40
Size of
the NPL
{# of Sites)
1,500
1,800
2,500
1,800
1,800
1,800
1,800
1',800
Future
Funding Meeds
(Billions
of FY83 $)
7.6
11.7
22.7
10.0
13.3
14.9
10.5
12.6
Alternative Sizes of the NPL
#8
#9
#10
8.1
8.1
8.1
50/50
50/50
50/50
1,500
2,200
2,500
9.7
14.4
16.4
-------
4-11
Sensitivity Analyses #8, #9, and #10 differ from the baseline estimate by
varying only the size of the NPL. With a 1,500 site NPL, Fund requirements
would total $9-7 billion (FY83 dollars). With a 2,200 site NPL, Fund
requirements would total $14.4 billion {FY83 dollars). To address a
2,500-site NPL, Fund requirements would total $16.4 billion (FY83 dollars).
-------
5. POTENTIAL EXPANSION IN THE SCOPE OF THE SUPERFUND PROGRAM
5.1 INTRODUCTION
Since the Superfund program was enacted, the focus of the program has been
primarily on identifying, investigating, and responding to those sites which
fall under the general description of more traditional hazardous waste sites.
EPA, the States, and the public set out to identify those sites that obviously
had a hazardous waste problem or sites where there was a particular suspicion
that there might be a problem. Following this approach, EPA has identified
approximately 19,000 potential hazardous waste sites in the ERRIS data base.
Primary emphasis then shifted to investigating these sites, listing the more
serious problems on the NPL, and initiating response at these sites. Although
ERRIS continues to grow, the rate of growth has moderated.
Nevertheless, there are still many traditional hazardous waste sites that
have not been identified. Sites such as municipal and industrial landfills
require intensive record searches to identify their location, the type of
materials they were receiving, and their ownership. Many of these sites can
only be identified through facility-by-facility search and sampling. These
sites will require targeted, systematic discovery and investigation efforts
that are likely to be far more resource intensive than such efforts in the
past.
In addition, the focus and character of Superfund are expanding as the
more obvious problem sites appear to have been identified. Now Superfund is
called on to cover problem sites that received less emphasis earlier in the
program, such as mining related sites. Problems that are new to Superfund,
such as groundwater contamination from the agricultural use of pesticides, are
receiving increasing attention as well.
Systematic investigation efforts to identify these sites would expand the
size of Superfund well beyond the focus discussed in Chapters 3 and 4. In
addition to expanding the remedial action program, investigation of the
problem categories listed here would also expand Superfund's removal
activities well beyond the 190 removal actions and $75 million per year that
the Agency currently projects.
The potential for dramatic increase in the size and focus of Superfund
raises serious questions about program priorities over the next several
years. What resources should be devoted to discovery and investigation
efforts? What problems should discovery and investigation be targeted to?
How should EPA apportion resources for remedial and removal actions among the
many new sites which may be identified?
-------
5-2
This report does not attempt to answer these questions. However, this
chapter discusses those problem categories which EPA sees as the major source
of Superfund expansion. The discussion lays out the potential hazards
associated with the problem, and estimates the total numbers jf sites that
could fall under each problem category, where reasonable estimates could be
developed. It is especially important to note that the total numbers of sites
estimated do not reflect the numbers of sites that could potentially become
Superfund sites, only the total numbers of sites in each problem category.
Out of that total, some subset will require more intensive investigation, and
a subset of those could require removal or remedial response by Superfund.
5.2 SOURCES OF EXPANSION: NEW PROBLEM AREAS
The major problem categories are listed in Exhibit 5-1, along with
estimates of the number of sites of concern, where available.
5.2.1 RCRA-Related Subtitle C Facilities
There are approximately 4,800 hazardous waste storage, treatment, and
disposal facilities regulated under Subtitle C of the Resource Conservation
and Recovery Act (RCRA). Currently, 45 RCRA facilities are on the final NPL
and the second update. Land disposal facilities pose the greatest concern
because it is likely that most land disposal facilities will leak eventually,
causing hazardous'releases into the environment. RCRA facilities fall into
three categories: (1) interim status facilities that were in existence on
November-19) 1980, that have complied with notification requirements; but have
not received a final permit; (2) permitted facilities, which are subject to
more stringent monitoring and operating standards; and (3) inactive portions
of RCRA facilities that have not received wastes since January 26, 1983.
RCRA regulations are primarily design and performance standards aimed at
minimizing threats to human health and the environment by reducing the
probability of a release. Interim status facilities must comply with interim
status standards until their final permit is issued. However these standards
are less stringent than final standards. Final permits for existing
facilities require them to meet operating standards, conduct groundwater
monitoring, provide corrective action when contamination of groundwater
exceeds the groundwater protection standard, and ensure financial
responsibility for the costs of routine closure and 30 years of post-closure
care. The RCRA program prior to the 1984 amendments has not required a
permittee to provide financial assurances in advance for corrective actions
which may be needed in the future. It has not required off-site cleanup, nor
was it applicable to inactive portions of regulated facilities.
The Hazardous and Solid Waste Amendments of 1984, enacted November 11,
1984, greatly expand EPA's authority over hazardous wastes beyond the RCRA
program currently in place. These amendments increase EPA's authority to
impose requirements on interim status facilities, and to require cleanup of
off-site contamination from RCRA facilities as well as cleanup for
pre-existing releases.
-------
5-3
EXHIBIT 5-1
POTENTIAL SOURCES OF SUPERFUND EXPANSION
Number of Sites
Problem of Potential Concern
RCRA Subtitle C 605-7
Municipal Landfills 34,000-52,000
Industrial Landfills 75,000
Mining Waste Sites 9,770-63,770
LUSTs, Non-Petroleum 11,250-187,500
TOTAL 130,625-378,875
a/ Projected failures of treatment, storage, and disposal facilities
under financial assurance requirements.
-------
5-4
The following discussion first describes the regulatory structure that has
been in place and identifies those types of facilities which are likely to
"fall out" and potentially become Superfund sites. Following that is a brief
description of the expanded provisions of the 1984 amendments, and the
possible impacts they may have on the Superfund program, both in the near term
and over a longer time frame.
As long as land disposal facilities continue in operation, RCRA provides
significant authority to require owners and operators to control and clean up
hazardous releases from the facilities. However, some facilities have in the
past and will in the future go out of business. These business failures or
abandonments may result from several different causes: in some cases, the
owner may abandon a facility for a variety of reasons; some may be business
failures due to routine causes; some may be business failures due to the costs
of compliance and cleanup under RCRA.
These closed facilities have the highest potential to become Superfund
sites. Any facility that closes is subject to a post-closure permit which
specifies actions that must be taken to properly close the site, including
financial responsibility for the costs of routine closure and 30 years of
post-closure care. Post-closure permits for land disposal sites may also
require corrective action for releases from the site. However, if the owner
does not have the financial capability to correct the problem and take cleanup
action, the owner may go out of business and the site may become a potential
Superfund site.
Prior to the passage of the 1984 RCRA amendments, EPA was considering
imposing financial assurances for corrective action as part of the
requirements which an interim status land disposal facility must meet in order
to receive a final permit. Based on an analysis of the impacts of this
requirement, some projections can be made concerning the potential number of
business failures of firms owning land disposal facilities.
Of the approximately 4,800 Subtitle C facilities, about 320 are owned by
municipalities, states, or the federal government. These facilities are not
considered in projecting numbers of failures, since it is assumed that
governmental entities will remain solvent to comply with cleanup
requirements. Approximately 960 of the 4,800 facilities are active land
disposal facilities owned by non-governmental entities.
The analysis (which does not reflect the 1984 amendments) projects that
the number of business failures of land disposal firms over the next 10 years
will range from 50, with no requirement for financial assurances, to 130, if
financial assurances are required. These figures include routine business
failures, failures due to actual requirements for corrective action, and
failures due to the costs of financial assurances to provide the costs of
corrective action which might be needed in the future. Within 25 years, an
additional 20 firms or facilities might be abandoned due to routine business
failure.
-------
5-5
The remaining Subtitle C facilities consist of some 3,520 privately owned
storage and treatment facilities in the United States. If the same failure
rate is projected for these facilities as for land disposal facilities, the
number of failures within 10 years could range from approximately 180 to 475.
Since storage and treatment facilities are also subject to the financial
assurance requirements of the 1984 amendments, the number of failed storage
and treatment facilities is projected at the higher end of the range, at 475-
Therefore, the total number of Subtitle C facilities of concern is projected
to be 605 (i.e., 130 disposal facilities and 475 storage and treatment
facilities). However, there are not likely to be as many problems with
leaking storage and treatment facilities, and the problems are generally not
as serious. It is important to note that these numbers are conservative,
since EPA has not completed an analysis of the impact of the more stringent
1984 amendments.
Certain provisions of the Hazardous and Solid Waste Amendments of 1984
impose more stringent requirements on Subtitle C facilities. These amendments
expand the geographic boundaries for requiring cleanup, impose new
environmental conditions, and require more stringent treatment, design, and
post-closure standards that lead to higher costs for compliance.
The requirements that are likely to affect the number of failures include
the following: Interim status facilities and permitted facilities must meet
more stringent design standards. Interim status facilities with a continuous
release {from any solid waste management unit, regardless of when the waste
was placed in the unit) must have a schedule of compliance for corrective
action in order to receive a final permit. If they require corrective action,
they must also demonstrate financial assurance for the costs of corrective
action. EPA may require corrective action to clean up releases beyond the
boundaries of the facility, and all RCRA authorities are extended to inactive
portions of currently operating facilities.
These requirements may reduce the long-term need for Superfund. However,
the more stringent requirements also make it likely that the number of
failures in the near term will increase beyond the above projections as
owner/operators find they cannot comply with the new requirements.
Hazardous Wastes Exempted from RCRA
Until the 1984 amendments, small quantity generators of hazardous waste
(generating under 1000 kilograms per month) and recycled wastes reclaimed for
legitimate, beneficial uses were exempted from RCRA. This means that there
has been no control over disposal of hazardous wastes from these sources.
Many of these wastes may have found their way to municipal and industrial
landfills, discussed below. The 1984 amendments lowered the small quantity
generator exemption to 100 kilograms per month, and small quantity recycled
wastes will also be subject to new standards.
-------
5-6
5.2.2 Municipal Landfills
Between 12,000 and 18,000 municipal landfills are currently operating in
the United States, "he total number is likely to be toward the lower end of
this range, since over the last several years there has been movement to close
down and consolidate ma.iy smaller facilities. There may, however, be up to
twice as many (24,000 to 36,000) inoperative municipal landfills in the
country. Approximately 2,000 of the cotal number of landfills, both operating
and inoperative, are currently represented in ERRIS. This would leave an
additional 34,000 to 52,000 municipal landfills that may not be represented in
existing Superfund program data.
Operating municipal waste landfills are addressed under Subtitle D of
RCRA. While EPA has developed standards for sanitary landfills (40 CFR Part
257), enforcement of the standards (or equivalent or more stringent standards)
is the responsibility of the states. In fact, almost all municipal waste
landfills are operating under state regulatory authority, many under state-
issued permits.
Since November 1980, any waste disposal facility that did not notify EPA
of its intention to continue to accept hazardous waste is prohibited from
accepting hazardous waste from large quantity generators. However, landfills
in operation prior to November 1980 may have received hazardous wastes. In
addition, since small quantity generators of hazardous waste have not been
regulated, they may have continued to dispose of their wastes in municipal
landfills. Landfills also continue to receive household wastes which may
c'ontain some solvents, cleaning materials, and pesticides.
There are currently 163 sites on the final NPL and the second update that
are classified as containing municipal refuse or as municipally owned
landfills. Identifying landfills, especially inoperative landfills, and
investigating potential hazardous substance releases will be considerably more
difficult at Subtitle D than at Subtitle C facilities. Each state's solid
waste management program will have to be used to identify existing facilities,
and intensive record searches will be needed to identify wastes that have been
disposed of in the facilities and to identify inoperative facilities. Site
inspections for municipal landfills may be time consuming and expensive. Only
about 1,600 facilities have monitoring systems for groundwater, leachate,
and/or gas.
5.2.3 Industrial Landfills
There are approximately 75,000 operative, on-site industrial landfills in
the United States based on data from the late 1970s. Many of these sites may
have received large quantities of hazardous waste prior to November 1980. In
addition, many of these facilities may still be receiving hazardous wastes
from small quantity generators. Active industrial landfills are also
regulated under Subtitle D of RCRA (40 CFR Part 257). However, a number of
states do not regulate on-site disposal of industrial wastes.
The industrial landfills most likely to be of concern are those that have
been operative since World War II, during the enormous expansion in the use of
-------
5-7
chemicals. However, there are some older industries with a high potential for
causing problems, such as tanneries, metal processing plants, and inoperative
coal gasification plants.
As discusssed in Chapter 2, 11 of the 25 most commonly found hazardous
substances at Superfund sites are chlorinated solvents. These substances are
widely used in a variety of industrial processes. Therefore, it is likely
that a number of industrial facilities that used these and other hazardous
substances carelessly, disposed of them in landfills, or abandoned them on the
land, may pose hazardous substance problems. There are currently 107 sites
classified as commercial/industrial landfills on the NPL, and 27 more have
been proposed for the second update. In addition, as demonstrated by Exhibit
3-1, many of the activities and facilities describing NPL sites relate to
various types of industrial activity.
5.2.4 Mining Wastes
The mining industry consists of approximately 580 active metal mines and
over 12,000 non-metal mines. In addition, the Bureau of Mines has estimated
that there are between 20,000 and 200,000 inactive and/or abandoned mines.
While the exact number is not certain, approximately 70 percent of the active
non-metal mines are for sand, gravel, and stone mining. If this same
percentage is applied to all mines, the number of mines of concern to
Superfund would be reduced to approximately 9,770 to 63,770 mines. It is
likely that the number is at the lower end of the range.
Although mining wastes are not 'specifically designated as. hazardous
substances under CERCLA,- they may contain constituents which are
designated as hazardous substances under Section 101(14) of CERCLA. Mining
wastes may include heavy metals such as lead, arsenic, and cadmium, asbestos,
and radioactive materials. Acid drainage from mines, overburden, and tailings
piles may mobilize the metals so that they move through the soil and into
groundwater and surface waters. The primary concern with asbestos is
inhalation exposure through air transport of fine particles.
Releases from mines and tailings piles may cause serious environmental
damage to plants and animal populations and have significant impacts on
surface and groundwater quality. Many mining sites, especially in the West,
are relatively isolated from human populations. However, some tailings piles
have been used for construction purposes and household uses, potentially
spreading contamination to buildings and development sites.
J/ Section 101(14) of CERCLA defines hazardous substances to include "any
hazardous waste having the characteristics identified under or listed pursuant
to Section 3001 of the Solid Waste Disposal Act (RCRA) (but not including any
waste the regulation of which under the Solid Waste Disposal Act has been
suspeneded by Congress)." In the RCRA Amendments of 1980, Congress suspended
the regulation of "solid waste from extraction, benefication, and processing
of ores and materials." (Section 3001(b)(3)(A)(ii)) This suspension will
continue until at least 6 months after EPA submits studies on these wastes to
Congress. While mining waste landfills and surface impoundments are currently
exempted from Subtitle C of RCRA, they are not exempted from Subtitle D.
-------
5-8
EPA has taken action at some mining sites in response to the hazardous
constituents in the mine wastes. Recent court decisions have supported this
interpretation of EPA's authority to respond.-' However, because there was
initial uncertainty over EPA's ability to respond to mining sites, it is
likely that there has been relatively less effort to identify and investigate
potential sites.
In spite of this early lack of clarity about how to deal with mining
sites, 15 sites on the NPL are mining sites, as are 3 sites proposed on the
second update in October 1984. These sites include mining activities related
to lead/zinc, copper, silver/gold, asbestos, and uranium. Ore processing and
smelting operations are represented by an additional 16 sites on the final NPL
and one site on the second update.
5.2.5 Leaking Underground Storage Tanks
There are probably several million underground storage tanks in the United
States that are used to store petroleum products and a variety of other
substances, many of them hazardous substances used for industrial processes.
There has recently been a growing awareness of the potential problems
associated with these tanks. In general, such tanks have been installed with
little or no regulation of their design or location; with little or no
monitoring equipment to detect leaks; and without special design features to
contain leaks that might occur.
Recent evidence indicates that leakage -of underground tanks occurs
throughout the country and relatively often. For example, a study initiated
by the state of New York in 1979 indicated that up to 20 percent of active
underground petroleum tanks in the State were leaking. The results of a San
Francisco Water Quality Board study showed that 88 percent of underground
petroleum storage tanks over 7 years old had evidence of contamination. The
potential health and environmental impacts of leakage may be significant. A
1978 Michigan study estimated that 21 percent of the 268 known incidents of
groundwater contamination that year were attributable to underground petroleum
tanks.
Non-Petroleum Underground Storage Tanks
EPA estimates that there may be 500,000 to 1,000,000 underground storage
tanks used for non-petroleum substances. No studies are available documenting
the likely numbers of non-petroleum tanks, although EPA plans -to undertake
such a study in the near future. It is likely that a significant proportion
of these tanks contain hazardous substances as defined by CERCLA.
Under RCRA, EPA regulates underground tanks used for the storage of
hazardous wastes, if the tanks can be entered for inspection. Approximately
2,000 tanks fall into this category. An estimated 30-50 percent of these
tanks are leaking. Approximately 50 percent of these tanks are more than 8
years old, and 33 percent contain corrosive wastes which can reduce effective
tank life.
2/ See, for example, U.S. v. Metate Asbestos Corp. (D. Az. Ap. 10, 1984).
-------
5-9
The final NPL and second update list 79 sites potentially containing
leaking underground storage tanks. Nineteen of the second update sites are in
the South Bay area of San Francisco, which includes Silicon Valley. These
sites are of particular interest because when they were identified by the
state for addition to the NPL, they had not been entered into ERRIS. This
supports the idea that the NPL could grow significantly from sites that are
identified outside the Superfund discovery and investigation process.
Estimates of the number of facilities with leaking underground storage
tanks can be derived in the following way:
There may be 500,000 to 1,000,000 non-petroleum
underground storage tanks in the U.S.
Somewhere between 3 and 25 percent of these tanks may be
leaking, which would result in 15,000 to 250,000 leaking
tanks.
Assuming that half of the leaking tanks are found at
sites with one leaking tank, and the other half at sites
with two leaking tanks, there could be 11,250 to 187,500
sites with leaking tanks.
Underground Storage Tanks for Petroleum Products
Releases involving petroleum products are not currently covered by CERCLA.
Hpwever, some attention has been given.to expanding the scope of Superfund
coverage to include petroleum releases from underground storage tanks. There
are approximately 2,500,000 such tanks in the United States. However, further
study is needed to more precisely identify the total number of tanks.
Currently available information leads to the following estimates of the
number of leaking underground storage tanks containing petroleum:
Approximately 2,500,000 underground storage tanks in the
U.S. contain petroleum products.
Between 3 and 25 percent of petroleum underground
storage tanks are leaking, yielding an estimate of
75,000 to 625,000 leaking petroleum tanks.
Assuming that half the leaking tanks are found at sites
with one leaking tank and the other half at sites with
two leaking tanks, between 56,250 and 468,750 sites
might contain leaking petroleum tanks.
It is important to note, however, that only if CERCLA were amended to
cover petroleum releases could any of these facilities become eligible for
Superfund action.
-------
5-10
Expanding Regulatory Authorityfor Underground Tanks
The Solid and Hazardous Waste Amendments of 198M expand regulatory control
to underground storage tanks. The amendments prescribe a state-run control
program that incorporates certain standards which will be specified by EPA.
The regulatory program will include notification requirements and requirements
for leak detection systems and corrective action.
The amendments also require an inspection program, which could have a
short term impact on Superfund. Systematic inspection of underground tanks
could identify many additional hazardous release problems, as projected
above. This could accelerate the addition of non-petroleum underground
storage tanks to the NPL.
5.2.6 Pesticide Contamination from Agricultural Uses
A problem area which has only recently been raised as a possible area for
Superfund response is groundwater contamination from application of pesticides
in agriculture. A number of instances of contamination have been reported in
the past several years. These involve contamination of the soil with
persistent chemicals which eventually move into groundwater. Such
contamination may result from uses of pesticides that are approved by EPA. As
noted in Chapter 3, Superfund may respond to such problems, but EPA cannot
recover the costs of response actions.
Contamination from pesticide use may result from application of pesticides
over a broad agricultural area and from repeated applications over time. This
means that the contaminants moving through the soil and into groundwater may
affect very large areas of groundwater resources. The seriousness of the
problems will depend on a number of factors, including the hazard
characteristics of the pesticides, their persistence, and leaching rates
through various types of soil.
The proposed second update to the NPL includes six drinking water well
sites in Hawaii that are contaminated with at least two different pesticides,
DBCP and EDB. While a number of NPL facilities contain pesticide
contamination connected with manufacturing activities, these six proposed NPL
sites represent the first time that releases resulting from the agricultural
application of pesticides have been listed. The implications of extending the
focus of Superfund to these sites could be far reaching.
It is extremely difficult to estimate the size of the problem at this
time. Testing conducted by EPA, states, and private parties has identified at
least 15 different pesticides in groundwater in over 20 states as a result of
agricultural uses. A number of communities have already had to switch to
bottled drinking water supplies or have had to install water systems from
alternative sources due to pesticide contamination of groundwater drinking
water supplies.
EPA will soon undertake an extensive monitoring survey to evaluate the
frequency and severity of groundwater drinking water supplies contaminated by
-------
5-11
pesticides. This survey will monitor for the presence of approximately 45
pesticides. The criteria used to select these chemicals include environmental
transport and persistence and demonstrated occurrence in groundwater.
5.2.7 Radioactive Wastes
Radioactive wastes are encountered at all stages of the nuclear fuel cycle
in mining, milling and conversion, in fuel fabrication, in spent fuel
assemblies, and in the reprocessing of spent fuel. As noted in Chapter 3,
certain facilities licensed by the Nuclear Regulatory Commission (NRC) are
exempted from Superfund coverage. In addition, EPA has made a policy decision
not to respond to radioactive releases from facilities holding a current NRC
materials license, because the NRC has adequate authority to control such
releases. However, EPA does consider response to formerly licensed facilities
with expired licenses as well as to facilities with licenses issued by
"agreement" states (states that have the authority to issue materials licenses
under agreement with the NRC).
There have been and continue to be a number of problems associated with
radioactive releases from both currently and formerly licensed facilities.
There are 30 final NPL and proposed sites that contain radioactive materials.
As of September 1983, the NRC was administering 3,900 materials licenses and
an additional 13,000 were administered by agreement States.
It is currently not possible to estimate the number of such facilities
that might become Superfund sites, since this will depend on several factors:
first, .the degree of care exercised by .licensees in their handling of
radioactive materials; second, the stringency of requirements licensees must
meet before they close; and third, the financial capability of licensees to
fulfill any cleanup requirements which might be imposed.
5.2.8 Other Problems
Because Superfund authority is broad enough to respond to almost any
hazardous releases into the environment, there are many other types of
problems, in addition to the large problem areas discussed above, that
Superfund could be called upon to clean up. Some of these releases could be
addressed by removal actions; others would require remedial action to
correct. Even for many of the problems that are individually small, the
number of removals that EPA could undertake is potentially very large.
Federally Permitted Releases
Federally permitted releases are exempted from liability for the costs of
Fund-financed response by CERCLA section 107(J). However, EPA still has
authority to respond to such releases. Current EPA policy is that in order to
qualify for this exemption, a hazardous substance must be expressly included
in the permit or have been considered in the permitting process and in the
public record.
-------
5-12
Practically, this means that the releaser of a hazardous substance
associated with, but not covered by a federal permit, is liable for the costs
of Fund-financed response. While EPA has not actively identified federally
permitted releases which may pose a serious problem, it is likely that there
are a number of such sites resulting from hazardous substances associated with
these releases. These sites could add to the size of the NPL. If there were
a policy change to interpret a federally permitted release more broadly (i.e.,
to include hazardous substances associated with the permit but not covered by
it), the Agency's ability to recover the costs of response to such releases
would be limited.
Non-WorkplaceAsbestos Sites
During the post-World War II period, materials containing asbestos were
extensively used in the construction of schools, commercial and government
buildings, and residences. These materials were often sprayed or trowelled
onto ceilings, walls, and beams as a fire-retardant or as insulation.
Spray-applied materials containing asbestos are commonly friable (i.e., easily
crumbled or pulverized) and can result in the release of asbestos fibers into
the environment. Studies of exposure to asbestos in workplace settings have
concluded that inhalation of asbestos fibers can lead to pulmonary cancer,
mesothelioma, and fibrosis of the lung.
Spray applications of asbestos have been prohibited for all uses since
1978. However, there are many buildings still in use which contain asbestos.
EPA recently completed two surveys to provide more precise and statistically
valid estimates of the incidence of asbestos in public buildings and schools:
Asbestos in Buildings: A National Survey of Asbestos-Containing Friable
Materials, and Evaluation of the Asbestos in Schools Identification and
Motification Rule. These studies concluded that there are approximately
31,000 schools and 733,000 buildings in the U.S. with materials containing
asbestos.
In August 1984, Congress established an EPA-administered loan and grant
program for schools to enable them to remove friable asbestos from school
buildings. However, this program does not extend to other large buildings
which also may contain friable asbestos. EPA's current policy is that the
Agency will not respond to air releases within buildings. However, if EPA
were to make a policy change to address such indoor releases, this would
significantly affect the number of sites to which Superfund responds.
Single Party Sites
A significant number of hazardous releases affect only a single family
dwelling. The releases may result from such things as the presence of wood
preservatives in log homes, the contamination of homes from chlordane
applications for termite control, use of radioactive materials, and
contamination of private wells from a variety of activities.
If Superfund responds to such releases, the response would be a removal
action. While the costs per site may be relatively small, the pressure on EPA
to take action at single party sites could grow enormously, overwhelming the
-------
5-13
resources that EPA currently commits to the removal program. Expanding the
focus of Superfund's removal response activities to consistently address such
single party sites could have a serious impact on the Agency's ability to
respond to other immediate threats that affect more than a single dwelling.
Contamination of Rivers and Harbors
A problem which has received little attention from the Superfund program
to date is contamination of river bottoms and harbors. Such contamination can
result from multiple sources, both point source discharges and nonpoint source
runoff. Because the sources are often varied, it is difficult to identify
them precisely and to score rivers and harbors using the HRS. While it is
unclear how many sites with serious problems might fall into this category,
the costs of cleanup for each site could be very high. In addition, the
environmental impacts of attempting to clean up contaminated bottom sediments
could also be serious, since disturbing the sediments may mix some hazardous
substances back into the water.
Naturally Occurring Hazardous Substances
Large numbers of people currently live in areas of the United States where
naturally occurring levels of hazardous substances or pollutants are elevated
above the thresholds of safety defined for drinking water. These areas are
unlikely to be designated for a Superfund response due to EPA policy reflected
in the scoring procedures of the Hazard Ranking System. When the presence of
a substance is not above background levels established by the naturally
occurring levels of the substance, the value assigned for the observed release.
category is zero. This zero score will in turn result in a low overall HRS
rating.
The low scoring of naturally occurring contamination reflects a conscious
policy on the part of EPA to give higher priority to sites where human
activity has resulted in an actual or potential risk to health and the
environment.
5.3 SOURCES OF EXPANSION: POLICY CHANGES
The size of Superfund could also expand significantly through policy
changes that would lower the HRS cutoff score for listing sites on the NPL.
As noted in Chapter 1, the HRS cutoff score of 28.5 was originally chosen to
yield an NPL of approximately 400 sites and not necessarily to indicate a
critical threshold of risk. Sites that score above the cutoff do pose a
relatively more serious threat. However, sites that score below the cutoff
may still pose a threat to smaller populations, or to the environment.
Currently, response to sites that fall below the cutoff score is the
responsibility of states or of responsible parties who are encouraged to
undertake cleanup efforts. However, Superfund response could be extended to
such sites either by lowering the cutoff score for listing on the NPL or by
modifying the HRS to weight certain factors more heavily.
< ;-- "': orciem.nl JVrtp-t i-,«
-b:--y. u.om 2404 -V*"?
401 XStrm, sX * 'A
Washington, r,j vC^r>
-------
5-14
Referring back to the central estimates presented in Chapter 4, the size
of the NPL could grow significantly through such a policy change. If the
ratio of ERRIS sites to NPL sites were to change from the current 12 to 1 to a
ratio of 6 to 1, the NPL could range from 3,670 s tes to 4,170 sites, based on
an ERRIS size of 22,000-25,000 sites.
Until systematic identification and investigation of these many different
types of problems are undertaken, it is impossible to estimate the total
number of sites that could become potential Superfund problems. However, if
even a small fraction of these sites require Superfund response, the funding
needed to address them would overwhelm the central estimates currently
projected for the Superfund program.
------- |