Environmental Protection: More Complete Data and Continued	 
Emphasis on Leak Prevention Could Improve EPA's Underground	 
Storage Tank Program (30-NOV-05, GAO-06-45).			 
                                                                 
Leaking underground storage tanks that contain hazardous	 
products, primarily gasoline, can contaminate soil and		 
groundwater. To address this problem, the Environmental 	 
Protection Agency (EPA), under its Underground Storage Tank (UST)
Program, required tank owners to install leak detection equipment
and take measures to prevent leaks. In 1986, the Congress created
a federal trust fund to assist states with cleanups. Cleanup	 
progress has been made, but, as of early 2005, cleanup efforts	 
had not yet begun for over 32,000 tanks, many of which may	 
require state and/or federal resources to address. GAO identified
(1) data on the number and cleanup status of leaking tanks, (2)  
funding sources for tank cleanups, and (3) processes used by five
states with large numbers of leaking tanks--California, Maryland,
Michigan, North Carolina, and Pennsylvania--to identify, assess, 
and clean up sites.						 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-06-45						        
    ACCNO:   A42257						        
  TITLE:     Environmental Protection: More Complete Data and	      
Continued Emphasis on Leak Prevention Could Improve EPA's	 
Underground Storage Tank Program				 
     DATE:   11/30/2005 
  SUBJECT:   Contamination					 
	     Environmental monitoring				 
	     Federal/state relations				 
	     Hazardous substances				 
	     Pollution control					 
	     Program evaluation 				 
	     Tanks (containers) 				 
	     Environmental cleanups				 
	     EPA Underground Storage Tank Program		 
	     Leaking Underground Storage Tank Fund		 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-06-45

     

     * Report to Congressional Requesters
          * November 2005
     * ENVIRONMENTAL PROTECTION
          * More Complete Data and Continued Emphasis on Leak Prevention
            Could Improve EPA's Underground Storage Tank Program
     * Contents
          * Results in Brief
          * Background
          * States Provide EPA with Some Tank and Cleanup Data, but Do Not
            Provide Separate Information on All Known Abandoned Tanks
          * Several Funding Sources Exist For Cleaning Up Tank Releases, but
            Some States' Resources for Remediating Abandoned Tank Sites Are
            Limited
               * Tank Owners or Operators and State Indemnification Funds
                 Generally Pay for Cleanups
               * The Leaking Underground Storage Tank Trust Fund Provides
                 Support for State Programs and Abandoned Tank Site
                 Remediation
               * Some States Believe That Funding From Available Sources Is
                 Inadequate to Address All Sites With Leaking Abandoned Tanks
          * States Identify, Assess, and Clean Up Leaking Tank Sites Using
            Similar Means
               * States Identify Leaking Tank Sites Through a Variety of
                 Methods
               * Some States' Inspection Rates May Have Limited the Timely
                 Detection or Prevention of Leaks
               * States Use Risk-Based Assessments to Determine Cleanup
                 Priorities
               * States Generally Direct and Oversee Remediation, but May Ask
                 EPA to Lead or Support Certain Cleanups
          * Conclusions
          * Recommendations for Executive Action
          * Agency Comments and Our Evaluation
     * Information on the History and Status of Cleanup Activities at Five
       Underground Storage Tank Sites
          * Coca-Cola Enterprises, Yuba County, California
               * Chronology of Key Site Events from the Case Files
               * Summary of Key Information from the Case Files
          * Henry Fruhling Food Store, Harford County, Maryland
               * Chronology of Key Site Events from the Case Files
               * Summary of Key Information from the Case Files
          * Bob's Marathon, Grand Ledge, Michigan
               * Chronology of Key Site Events from the Case Files
               * Summary of Key Information from the Case Files
          * R.C. Anderson Trust, Nash County, North Carolina
               * Chronology of Key Site Events from the Case Files
               * Summary of Key Information from the Case Files
          * Tranguch Tire Service, Inc., Luzerne County, Pennsylvania
               * Chronology of Key Site Events from the Case Files
               * Summary of Key Information From the Case Files
     * Objectives, Scope, and Methodology
     * Comments from the Environmental Protection Agency
     * GAO Contacts and Staff Acknowledgments

Report to Congressional Requesters

November 2005

ENVIRONMENTAL PROTECTION

More Complete Data and Continued Emphasis on Leak Prevention Could Improve
EPA's Underground Storage Tank Program

Contents

Tables

Figure

November 30, 2005Letter

Congressional Requesters:

Leaking underground tanks that store potentially hazardous products,
primarily gasoline at service stations, can contaminate soil as well as
groundwater, the source of drinking water for nearly half of all
Americans. Some components of gasoline can pose serious health risks to
the individuals exposed to them. For example, one gasoline additive-methyl
tertiary butyl ether (MTBE)-is a potential carcinogen that can migrate
quickly through the soil into groundwater. Even in small amounts, MTBE can
render groundwater undrinkable and is difficult and costly to clean up.
According to Environmental Protection Agency (EPA) data, as of March 31,
2005, about 449,000 fuel releases (leaks) had occurred from the more than
2.2 million active (currently in use) and closed (no longer in use)
federally regulated underground storage tanks nationwide. While progress
has been made in cleaning up releases, cleanup efforts had not yet begun
to address over 32,000 of them.

To address this problem, in 1984, the Congress created the Underground
Storage Tank (UST) Program within EPA, which subsequently established the
Office of Underground Storage Tanks to manage the program. Under the
Resource Conservation and Recovery Act (RCRA), tank owners and operators
must register with a designated state or local agency underground tanks
that store petroleum or hazardous substances. EPA and the states then
track and regulate these tanks. Furthermore, EPA required tank owners to
install new leak detection equipment by the end of 1993 and new spill-,
overfill-, and corrosion-prevention equipment by the end of 1998. If these
conditions were not met, owners had to close or remove their tanks.

Tank owners and operators are ultimately responsible for cleaning up
contamination from their leaking tanks. However, in 1986, the Congress
established the Leaking Underground Storage Tank (LUST) Trust Fund to
provide money for (1) overseeing and enforcing cleanup actions taken by a
tank owner or operator, and (2) cleaning up leaks at tank sites where the
owner or operator is unknown, or unwilling or unable to take action- which
we refer to in this report as "abandoned"-or which require emergency
action. EPA or a state can proceed with cleanup using the LUST Trust Fund,
and can subsequently seek reimbursement from the owners or operators. The
fund is capitalized through a $0.001/gallon tax on gasoline and other
motor fuels and the interest that accrues to the fund balance annually. As
of September 2004, the fund balance was about $2.2 billion. The Congress
annually appropriates amounts from the LUST Trust Fund to the federal UST
Program that EPA uses to negotiate and oversee cooperative agreements with
states, implement programs on Indian lands, and support regional offices
and state programs. Appropriations from the fund have been about $70
million to $76 million annually. The majority of these funds go to states
to implement their underground storage tank programs. As of September
2004, EPA had approved 34 states, the District of Columbia, and Puerto
Rico, to operate and enforce their own underground storage tank programs
with EPA oversight and monitoring.1 Fourteen other states operate and
enforce their own tank programs under state laws with limited EPA
oversight. States receiving LUST Trust Fund money can spend it on cleanup
and related activities. In addition, the Energy Policy Act, enacted in
August 2005, authorizes states to use a portion of their federal trust
funds for inspections and other leak prevention purposes and generally
requires that states inspect their underground storage tanks at least once
every 3 years. EPA also awards states annual grants to help them fund a
portion of their inspection and enforcement costs.

In this context, we identified (1) information available on the number and
cleanup status of leaking underground storage tanks, (2) existing sources
of funding for cleanups at contaminated tank sites, and (3) processes used
to identify, assess, and clean up sites in 5 states with large numbers of
leaking tanks-California, Maryland, Michigan, North Carolina, and
Pennsylvania. In addition, to provide some perspective on how leaking
underground storage tank sites are identified and cleaned up, we are
providing information on the history and cleanup status of one leaking
tank site in each of these 5 states. This information is included in
appendix I.

To address these issues, we reviewed and evaluated program data from EPA
and interviewed program officials in EPA's Office of Underground Storage
Tanks and EPA Regions 3 (Philadelphia), 4 (Atlanta), 5 (Chicago), and 9
(San Francisco). We also reviewed and evaluated data from and conducted
interviews with state program officials in California, Maryland, Michigan,
North Carolina, and Pennsylvania. In addition, we selected one leaking
tank site in each of these 5 states for comparison of cleanup status,
costs, and other factors. We included the Tranguch Tire Service,
Incorporated, facility in Luzerne County, Pennsylvania, among the sites in
our review due to congressional interest. We selected the remaining 4
sites primarily on the basis of their high cleanup costs, cleanup status,
and other similarities with the Tranguch site for comparison. We
interviewed site project or incident managers and reviewed case files for
each site. Additional information regarding our objectives, scope, and
methodology is included in appendix II. We conducted our review between
August 2004 and November 2005 in accordance with generally accepted
government auditing standards, including assessing the reliability of the
data we obtained.

Results in Brief

EPA requires states to submit data on their tanks to the agency, including
the number of active and closed tanks, the number of confirmed tank leaks
(referred to as releases), cleanups initiated and completed, and emergency
responses. State data show that, as of March 31, 2005, more than 660,000
tanks were active and about 1.6 million had been closed since the
inception of the UST Program in 1986. In addition, the states identified
about 449,000 confirmed releases and about 416,000 initiated cleanups,
with almost 324,000 of these completed. While states also compile some
data on abandoned tanks-tanks whose owner or operator is unknown, or
unwilling or unable to clean up leaks-EPA does not require them to
separately report to the agency information on the number and cleanup
status of all of the states' known abandoned tanks. EPA officials believe
that the data the agency currently obtains from states are sufficient for
general program oversight, identifying program trends, and determining the
progress of individual states' programs. However, without separate data on
all known abandoned sites, EPA cannot assess whether these sites are being
cleaned up or deferred because of a lack of funding. In addition, because
one of the purposes of the federal LUST Trust Fund is to provide money for
cleaning up abandoned tank sites, the lack of separate and more complete
data on abandoned tanks limits EPA's ability to determine how to most
efficiently and effectively allocate trust fund dollars to the states. We
are recommending that EPA require states to separately report to the
agency information on the number and cleanup status of all of their known
abandoned underground storage tanks.

Tank owners and operators are primarily responsible for paying to clean up
their own sites, but abandoned tanks are cleaned up using available state
resources-which in some cases may be limited-and the LUST Trust Fund. EPA
estimates that the average remediation costs per site have been about
$125,000, but costs sometimes have exceeded $1 million. RCRA requires tank
owners and operators to provide assurance that, if a release occurs, they
can pay the costs of cleaning up a site and compensating third parties for
injury and property damage. For example, an owner might purchase
commercial insurance or participate in the state's financial assurance or
indemnification fund. While each state's fund differs, in most cases,
states capitalize their funds with tank registration and petroleum fees.
However, abandoned tanks generally do not qualify for funding from
commercial insurers and may not qualify for a state's assurance fund. As a
result, addressing contamination from abandoned tanks depends largely on
the availability of funds from other sources, such as state appropriations
and the federal LUST Trust Fund. Environmental officials of 2 of the 5
states we contacted told us that their states' funds are inadequate to
address contamination at some abandoned tank sites and, therefore, must
rely on LUST funding. For example, Michigan officials told us that the
state has insufficient resources to address the backlog of about 4,200
confirmed releases from abandoned leaking underground storage tanks, which
will require about $1.7 billion in public funds to remediate. Furthermore,
both Michigan and North Carolina officials said that their states defer
cleanup or let contamination at abandoned tank sites attenuate (diminish)
naturally over a number of years once immediate threats are addressed
because resource constraints preclude complete remediation. In addition,
when emergency cleanup work was required at an abandoned site in a third
state-Pennsylvania-in 1996, state officials asked EPA to take over the
cleanup because the owner was bankrupt and the state could not pay the
expected cleanup costs. Future workload increases due to as yet
undiscovered abandoned sites could place additional demands on state and
federal trust fund resources and exacerbate this situation. While EPA and
states are undertaking initiatives to make cleanup efforts less expensive
and more effective, including using contracts that tie payment to the
accomplishment of site cleanup goals, officials from all 5 states agreed
that additional federal funds would enhance their states' ability to
respond to releases from abandoned tanks.

The 5 states that we contacted identify, assess, and clean up leaking tank
sites using similar processes. For the most part, owners and operators are
responsible for these activities under state oversight. Generally, the 5
states identify leaking tanks when (1) tank owners or operators report to
the state that they have confirmed a release after leak detection
equipment was activated, after they discovered a loss of product, or after
they found leaks when upgrading, replacing, or closing their tanks; (2)
land redevelopment activities uncover unknown tanks; or (3) state
environmental or health agencies investigate contamination complaints.
Furthermore, while regular tank inspections provide the opportunity to
detect new leaks and potentially prevent future ones, as of mid-2005, only
2 of the 5 states-California and Maryland-consistently inspected all of
the state's tanks at least once every 3 years, the minimum inspection
frequency that EPA considers necessary for effective tank monitoring.
However, the Energy Policy Act, enacted in August 2005, among other
things, generally requires inspections at least once every 3 years and
authorizes federal trust funds to be available for this and other leak
prevention purposes. These provisions should allow states to place greater
emphasis on their leak detection and prevention efforts. In the 5 states
we contacted, once the state environmental agency becomes aware of leaking
tanks, it then identifies responsible parties and requires them to hire
qualified environmental consultants to perform site assessments or
characterizations and develop and implement remedial action plans to
effectively clean up the site. All 5 state agencies prioritize most sites
for cleanup generally according to the immediate threat they pose to human
health, safety, and/or the environment. While states normally oversee
cleanups, they may ask EPA to lead or support the cleanup at sites that
present an imminent threat, that have no viable responsible party, that do
not qualify for funding under a state plan, or for which the magnitude of
the cost and cleanup work is beyond state resources.

Background

Data collected from the states and reported by EPA indicate that EPA and
states have made progress in cleaning up releases from underground storage
tanks over the past decade and a half. According to EPA, of the more than
447,000 releases confirmed as of the end of 2004, cleanups had been
initiated for about 92 percent, and about 71 percent of these cleanups had
been completed. Figure 1 shows confirmed releases from underground storage
tanks, cleanups initiated, and cleanups completed annually from fiscal
years 1997 through 2004.

Figure 1: Annual Confirmed Releases from Underground Storage Tanks,
Cleanups Initiated, and Cleanups Completed, Fiscal Years 1997 Through 2004

As this figure indicates, the number of new releases confirmed annually
declined, from about 12,000 in 2003 to less than 8,000 in 2004-about 35
percent. However, while figure 1 shows a decline in the number of releases
confirmed annually over the period, it also shows a decrease in the number
of cleanups initiated and completed. According to EPA, the number of
cleanups completed each year has generally decreased over recent years and
fell by 23 percent-from more than 18,000 in fiscal year 2003 to just over
14,000-in fiscal year 2004. Furthermore, there still remains a national
backlog of almost 130,000 cleanups yet to be completed.

EPA's UST Program is primarily implemented by the states. EPA has become
directly involved in program implementation only in Indian country and
when states have been unwilling or unable to establish effective
underground storage tank programs or to address contamination at specific
sites. Instead, EPA's primary role has been to establish standards and
regulations to assist the states in implementing their programs. While all
EPA-approved underground storage tank programs must be no less stringent
than the federal program, individual aspects of each state program differ.
For example, state time frames for conducting inspections vary widely.
Also, while some states use only state environmental personnel to conduct
inspections, others use state-certified private inspectors, or both.
Furthermore, state program requirements and standards are sometimes more
stringent and inclusive than those under the federal program. For example,
states often regulate home heating fuel tanks, tanks on farms, and
above-ground tanks that RCRA generally excludes from the federal program.2

EPA's UST Program receives approximately $70 million each year from the
LUST Trust Fund, about 80 percent of which is used for administering,
overseeing, and cleaning up sites. The remaining money has been used by
EPA for negotiating and overseeing cooperative agreements, implementing
programs on Indian lands, and supporting regional and state offices. EPA
spends about $6 million annually from the LUST Trust Fund on the agency's
program implementation, management, and oversight activities. Amounts
distributed to the states from the fund each year vary depending primarily
on whether they have an EPA-approved program, the total number of each
state's tanks, and the number of releases from those tanks. Until
recently, states could use these funds only for cleanup and related
administrative and enforcement activities, and EPA awarded each state
about $187,000 annually from the agency's State and Tribal Assistance
Grant account to help administer their programs and cover inspection and
enforcement costs. Historically, states have used about one-third of their
LUST Trust Fund money for administration, one-third for oversight and
state-lead enforcement activities, and one-third for cleanups, according
to EPA.

The Energy Policy Act of 2005, enacted in August 2005, includes a number
of provisions addressing issues relating to training, tank inspections,
prohibitions on fuel deliveries to problem tanks, and funding tank
inspections and enforcement, among others. With regard to training, the
act requires EPA to publish guidelines specifying training requirements
for tank operation and maintenance personnel and authorized EPA to award
up to $200,000 to states that develop and implement training programs
consistent with these guidelines.

In addition, the act requires EPA and any state receiving federal UST
funding to inspect all regulated tanks not inspected since December 22,
1998, within 2 years of the date of enactment. After these inspections are
completed, EPA or the state must generally inspect regulated tanks once
every 3 years. The act allows EPA to extend the first 3-year period for up
to 1 additional year if an authorized state demonstrates that it has
insufficient resources to complete all inspections within the first 3-year
period. Furthermore, beginning in 2007, the act prohibits deliveries to
underground storage tanks that are not in compliance with applicable
regulations and requires EPA and states to publish guidelines for
implementing the delivery prohibition that would, among other things,
identify the criteria for determining which tanks are ineligible for
delivery. Finally, the act authorizes substantial appropriations from the
trust fund during fiscal years 2005 through 2009 for a variety of
activities, including release prevention, compliance, training,
inspections, and enforcement.

States Provide EPA with Some Tank and Cleanup Data, but Do Not Provide
Separate Information on All Known Abandoned Tanks

EPA collects data on the total number of underground storage tanks and the
status of cleanup activities relating to these tanks from all states, and
reports this information semiannually.3 Table 1 shows key tank-related
data reported by EPA as of March 31, 2005.

Table 1: Key Data on Underground Storage Tanks, as of March 31, 2005

                                        

             Tank-related data element                                 Number 
Tanks                                         
Active tanks                                                       660,274 
Closed tanks                                                     1,605,711 
Total Tanks                                                      2,265,985 
Leaks (releases)                              
Confirmed releases                                                 448,807 
Cleanups                                      
Cleanups initiated                                                 416,246 
Cleanups completed                                                 323,586 
Cleanups ongoing                                                    92,660 
Cleanup backlog                                                    125,221 

Source: GAO analysis of EPA data.

EPA's semiannual reports also include, among other data, the number of
emergency response actions taken by an implementing agency, such as the
state, to mitigate imminent threats to human health and the environment
from an underground storage tank system.

EPA, however, does not require states to provide specific data on all
known abandoned underground storage tanks. While abandoned tanks are
included in the data reported to EPA, they are generally aggregated with
the other data and cannot be separately identified. However, all 5 states
we contacted compile some limited data on abandoned tanks and report this
information separately to the EPA regional office that manages each
state's LUST Trust Fund cooperative agreement. In this regard, all 5
states separately report the number of initiated and completed cleanups of
abandoned tanks using trust fund money. However, these data do not include
separate information on cleanups of known abandoned tanks using state
funds or any known abandoned tanks where cleanup has not yet been
initiated. EPA officials believe that the data the agency currently
obtains from states are sufficient for general program oversight,
identifying program trends, and determining the progress of individual
states' programs. However, because states generally do not provide
separate data on all abandoned tanks, EPA has limited ability to assess
and track states' progress in cleaning up contamination from these tanks.
In addition, although one of the primary purposes of the LUST Trust Fund
is to provide money for cleaning up abandoned tank sites, EPA lacks
information-such as the number of releases from known abandoned tanks in
each state and how many of these releases have been or are being cleaned
up-to help it determine how to most efficiently and effectively allocate
funds to the states for this purpose. EPA allocates amounts from the fund
to each state based, in part, on the data each currently provides, but
these allocation decisions do not now take into account the specific
number of the state's abandoned sites that may require cleanup funds.4

Several Funding Sources Exist For Cleaning Up Tank Releases, but Some
States' Resources for Remediating Abandoned Tank Sites Are Limited

While tank owners and operators are primarily responsible for cleaning up
contamination from leaks in their underground storage tanks, some states
assist them through financial assurance or indemnification funds. These
funds also sometimes pay for cleanups of abandoned tank sites. However,
not all states have indemnification funds and, in 10 of the 40 states that
have such funds, claims for cleanup cost reimbursements exceeded fund
balances in fiscal year 2004. Consequently, EPA is monitoring the states'
funds to determine their viability as financial assurance mechanisms.

EPA, through the LUST Trust Fund, provides some limited support to states
for cleaning up abandoned sites as well as for administering, overseeing,
and enforcing their cleanup programs. The 5 states we
contacted-California, Maryland, Michigan, North Carolina, and
Pennsylvania-use differing approaches to ensure funding to clean up
contamination from tank leaks. Three of these states-Maryland, Michigan,
and North Carolina-are experiencing difficulties in funding cleanups of
abandoned tank sites and officials of 2 of these states told us that
available resources will be insufficient to clean up all of the abandoned
tanks in their state.

Tank Owners or Operators and State Indemnification Funds Generally Pay for
Cleanups

Owners and operators are primarily responsible for cleaning up
contamination from leaks in their underground storage tanks. However,
according to the director of EPA's UST Program, many of these
owners/operators, most of which are small, independent businesses, do not
have the financial capacity to pay for expensive cleanups. EPA estimates
that the average remediation cost per site has been about $125,000, but
costs sometimes have exceeded $1 million. Under RCRA, tank owners and
operators must maintain evidence of financial responsibility for carrying
out cleanup actions, using one or more of a variety of mechanisms,
including commercial insurance, corporate guarantee, letter of credit,
qualification as a self-insurer, or an EPA-approved state financial
assurance fund. For commercial insurance, the owner/operator usually pays
premiums as well as a deductible amount and/or co-payments before the
policy begins to cover remediation costs up to some limit of coverage per
leak incident. To assist owners/operators in funding cleanups, as of
November 2004, 40 states had established state assurance or
indemnification funds. State indemnification funds typically have
deductible and co-payment requirements similar to those for commercial
insurance, but these funds are managed by the state. Indemnification funds
are usually capitalized through gasoline and diesel fuel taxes or fees
paid by owners/operators registering or obtaining permits for underground
storage tanks, as required. Any cleanup costs above the maximum coverage
provided by insurance or the indemnification fund are borne by the tank
owner/operator. A state fund qualifies as a financial assurance mechanism
if EPA has approved it for that purpose. In deciding whether to approve a
fund, EPA considers the certainty of the availability of funds for
cleanup, the amount of funds that will be made available, the types of
costs covered, and other relevant factors.

The 5 states we contacted vary in their approaches to ensuring that
contaminated tank sites are cleaned up and that tank owners/operators, to
the extent possible, pay the remediation costs. Three of the 5
states-California, North Carolina, and Pennsylvania-currently have
financial assurance funds that reimburse owners/operators for cleanup
costs under varying conditions. Maryland and Michigan have no such funds,
and, instead, tank owners/operators rely primarily on commercial insurance
to pay cleanup costs.

California: California's Underground Storage Tank program includes a state
financial assurance program-the Underground Storage Tank Cleanup Fund-to
assist tank owners/operators in funding site cleanups. The fund,
established in 1989, is the state's primary mechanism for reimbursing
owners/operators for their costs of cleaning up leaking underground
storage tanks incurred after January 1, 1988. The fund is available to
most owners/operators of tanks subject to EPA's Underground Storage Tank
Program, as well as owners of certain small home heating oil tanks. The
California State Water Resources Control Board administers the fund, which
is primarily capitalized through a storage fee-paid by owners of regulated
and permitted underground storage tanks-for each gallon of petroleum
placed in the tanks.  According to board officials, the fund collects
about $240 million annually and, except for $200,000 per year that is used
for enforcement, monies from the fund are all used for tank cleanups,
including such activities as direct cleanup by responsible parties, agency
oversight, and replacement of drinking water wells. In fiscal year 2004,
California spent approximately $208 million to reimburse responsible
parties for direct expenses incurred in cleaning up leaking underground
storage tanks.

The fund reimburses tank owners/operators for cleanup costs up to $1.5
million per incident for "reasonable and necessary" remediation costs.5
Claimants are divided into four classes: class "A" claimants do not have
to pay a deductible before costs are reimbursed by the fund; class "B" and
"C" claimants must pay the first $5,000 in eligible corrective action
costs; and class "D" claimants are responsible for the first $10,000. An
Underground Storage Tank Petroleum Contamination Orphan Site Cleanup
subaccount was established as part of the fund in September 2004,
capitalized with $30 million ($10 million per year for 2005 through 2007)
transferred from the fund to reimburse cleanup costs incurred in cleaning
up abandoned contaminated urban brownfield sites. In addition to
reimbursing owners/operators, state officials said that $5 million a year
is transferred from the fund to a subaccount to address emergency,
abandoned, and recalcitrant tank site cleanups.

Board officials we interviewed told us that the fund is adequately
capitalized and that they do not always spend all available funds each
year. Nevertheless, these officials also said that the state is interested
in ways to minimize program costs and is experimenting with
pay-for-performance remediation contracts, which are now being used at 20
cleanup sites in the state.

Maryland: Although Maryland has 2 trust funds that have financed certain
cleanup activities, it does not have a fund that EPA has approved for use
as a financial assurance mechanism. According to state officials, owners
and operators primarily use commercial insurance to demonstrate financial
responsibility. The state's Oil Contaminated Site Environmental Cleanup
Fund has provided limited cleanup assistance to owners/operators of
federally-regulated underground storage tanks, among others. The fund
provides funding of up to $125,000 per leak occurrence from underground
storage tanks-subject to deductibles from $7,500 to $20,000-and is
primarily capitalized by a fee of 1.75 cents per barrel of oil imposed at
the first point of transfer into the state. However, the program stopped
accepting applications for reimbursement from owners and operators of
federally-regulated underground storage tanks on June 30, 2005. In
addition, the Maryland Oil Disaster Containment, Cleanup, and Contingency
Fund finances, among other things, state cleanup costs for abandoned
sites.6 Revenues for this fund, according to the fund's fiscal year 2004
annual report, are generated by a fee of 2 cents per barrel of oil
transferred into the state. From July 1, 2003 to June 30, 2004, this fund
paid out about $3.5 million.

Michigan: Michigan's state financial indemnification program for
underground storage tanks was terminated in June 1995, because it had
insufficient funds to pay existing and future claims. Since that time,
tank owners/operators have been required to annually show proof of
financial assurance to cover cleanup costs in order to operate in
Michigan. Small owners/operators usually provide this proof by obtaining
commercial insurance. The state has used a number of sources to fund
limited cleanup work at underground storage tank sites, including the
Cleanup and Redevelopment Fund, the Clean Michigan Initiative Bond Fund,
the Environmental Protection Fund; State General Funds, and the
Environmental Protection Bond Fund. Appropriations from these funds
address soil, groundwater, and sediment contamination from all sources,
including leaking tanks. Most of these funds are no longer available for
new projects. According to state officials, in the fall of 2004, state
legislators voted to establish a Refined Petroleum Fund that will be
capitalized by a 7/8 cent-per-gallon fee on refined petroleum products to
be collected through 2010. This fund is expected to accrue approximately
$60 million each year, a portion of which is expected to be used to clean
up underground storage tank sites. A Refined Petroleum Cleanup Advisory
Council was also established to provide the governor and legislature with
recommendations on how to spend the fund's revenues. State officials told
us that the council is expected to recommend an increase in the 7/8 cent
fee to implement its other recommendations.

North Carolina: North Carolina has a state fund that acts as a financial
assurance mechanism and that reimburses owners/operators for most of the
costs for site assessments, cleanups, and damages related to leaking
underground storage tanks.7 This fund applies to leaks discovered after
June 30, 1988, from commercial underground tanks containing petroleum. The
fund is primarily capitalized by a 0.297 cent-per-gallon excise tax on
motor fuel sales; a small part of the state inspection tax on motor fuel
and kerosene; and annual tank operating fees. Under provisions of the
fund, owners/operators of tanks that have upgraded corrosion, leak, and
overfill protection pay the first $20,000 of assessment and cleanup costs
and the first $100,000 in third party liability costs. The fund then pays
all other cleanup costs deemed reasonable and necessary, up to $1 million,
and an additional $500,000, with a 20 percent co-payment by the
owner/operator, after which any remaining amount is paid by the
owner/operator. The state paid approximately $21 million in reimbursements
for tank assessment and cleanup costs from this fund in fiscal year 2004.

Because the balance of the fund was not sufficient to cover all
obligations, in June 2002, the fund began operating from month to month,
paying out funds on a first-come, first-paid basis. This action resulted
in a significant backlog of claims with pending payments, according to the
fund's annual report. Consequently, EPA is currently monitoring North
Carolina's fund to determine its viability as a financial assurance
mechanism. To address concerns about the viability of the trust fund,
North Carolina officials are considering requiring tank owners/operators
to use other forms of financial assurance, such as commercial insurance.

Pennsylvania: Pennsylvania's Underground Storage Tank Indemnification Fund
was created by the state Storage Tank and Spill Prevention Act of 1989, as
amended, and is administered by the State Insurance Department, according
to the fund's 2004 annual report. The fund reimburses tank
owners/operators for reasonable and necessary cleanup costs for leaks that
occur in regulated tanks on or after February 1, 1994, the date it began
operation. The maximum amount of coverage under the fund is currently $1.5
million; however, for claims reported prior to January 1, 2002, the limit
was $1 million, according to state officials. The aggregate limit is $1.5
million for owners of 100 or less tanks and $3 million for owners of 101
or more tanks. The fund also covers bodily injury and property damage
claims that arise from a leak, indemnifies certified tank installers, and
provides loans to owners/operators for upgrading their facilities.
According to the fund's 2004 annual report, a claimant for reimbursement
from the fund must be an owner or operator of a tank registered with the
Pennsylvania Department of Environmental Protection, and must report the
claim to the fund within 60 days of the discovery of the release.
Claimants must also pay the first $5,000 per tank of allowable cleanup
costs and $5,000 per tank of third-party liability claims. State program
officials told us that Pennsylvania law requires that the fund be managed
on an actuarial basis and that the fee structure be reviewed yearly to
maintain solvency. They also said that the fund's objective is to have
positive cash flow and invested assets for a projected period of at least
5 years. The fund is primarily capitalized by (1) a 1.1 cents-per-gallon
fee (for 2004) on substances such as gasoline, new motor oil, and aviation
fuel, (2) investment income generated from fund balances, and (3) a
capacity fee of 8.25 cents-per-gallon for substances such as diesel,
kerosene, and used motor oil.

While the fund does not directly cover costs for remediating abandoned
tank sites, it is authorized to provide allocations to the Pennsylvania
Department of Environmental Protection-which manages the cleanup of
contamination from these tanks-up to a maximum of $12 million annually:
$5.5 million for general environmental cleanup, $5.5 million for
catastrophic release, and $1 million for pollution prevention. Department
officials told us that each year it must request funding from the fund's
board. The board then allocates funds to the department based on the
fund's ability to pay tank owners' claims. State fund officials told us
that the department has not requested the maximum allocation amount for
the past several years, and in some years they have not spent the full
amount of the money they requested. According to state officials, the fund
collected $68 million in 2004, and paid out $64 million. These state
officials told us that the fund is fully capitalized and is working
effectively with a balance of $215 million, as of May 2005.

The Leaking Underground Storage Tank Trust Fund Provides Support for State
Programs and Abandoned Tank Site Remediation

In addition to the states' funding sources and mechanisms, the LUST Trust
Fund assists states in (1) overseeing and enforcing corrective actions
taken by tank owners/operators and (2) cleaning up leaking abandoned tanks
or tanks that require an emergency action. EPA allocates amounts from the
trust fund to each state based on a number of criteria, such as the total
number of tanks in the state, the number of confirmed releases, and
whether EPA has approved the state's program, among other factors.8
However, these criteria do not include the number and cleanup status of a
state's abandoned tanks. According to EPA program officials, states
historically have used about two-thirds of the federal trust fund money
allocated to them each year to oversee and support the cleanups paid for
by state funds, tank owners/operators, and other financial assurance
mechanisms, while the states have used the remaining one-third to directly
pay for cleanups of abandoned tanks that are not covered by the other
funding sources.

As Table 2 shows, for the 5 states we contacted, the amount of funds that
EPA awards from the fund and the portions of these funds the states
allocate for cleaning up tank sites varies, as do the amounts of their own
funds that they spend on leak cleanups.

Table 2: Five States' Funding from EPA's LUST Trust Fund, the Amounts and
Percentages of These Funds Used To Clean Up Tank Sites, and Estimated
Amounts of Each State's Own Funds Spent on Tank Cleanups, Fiscal Year
2004a

                                        

                                    Portion of               
                                     LUST fund               
                                    allocation               
                                   used to clean             
                                      up tank                
                                      sitesb                 
State          Amounts awarded         Amount Percentage  Estimated amount 
                    to state from                              of state's own 
                  LUST Trust Fund                              funds spent to 
                                                                clean up tank 
                                                                       sitesc 
California          $3,538,351     $2,257,026        64%      $240,000,000 
Maryland             1,376,825      1,336,825        97%         1,800,000 
Michigan             1,750,000        528,596        30%        12,900,000 
North Carolina       3,208,081        990,046        31%        21,300,000 
Pennsylvania         1,487,152              0         0%         2,200,000 
Total              $11,360,409     $5,112,493        45%      $278,200,000 

Source: GAO analysis of data obtained from EPA and individual states.

aCosts are for the states' fiscal year 2004. Because states' fiscal years
differ, for Michigan, costs are for the period October 1, 2003 through
September 30, 2004; for the other 4 states, costs are for the period July
1, 2003 through June 30, 2004.

bStates use the remaining portion of their LUST funds for administrative,
enforcement, and related purposes. States do not always spend the entire
amount of the funding awarded to them during a given year and may carry
some portion of it over into subsequent years.

cThese estimates may not include, among other things, personnel and
regulatory oversight costs.

The LUST Trust Fund's contribution to state cleanup efforts is generally
small compared to amounts paid by tank owners/operators, state
indemnification programs, and other state mechanisms for cleaning up sites
each year.9 For example, in fiscal year 2004, EPA awarded $61.7 million in
trust funds to assist states' leaking tank cleanup efforts.10 However,
according to EPA, states, on average, spend a total of about $1 billion to
$1.5 billion each year on tank site cleanups. To illustrate, EPA program
officials told us that for every federal dollar spent to clean up tank
sites, states spend as much as $18 of their own funds.

If cleanup costs paid by owners/operators were included, the actual ratio
of dollars spent by other sources to federal dollars could be
significantly higher than the 18 to 1 calculation provided by EPA.
However, tank owners'/operators' costs to remediate a site are difficult
to determine since they are not always captured in state and federal
records. While state records may include deductible and co-payment amounts
paid by owners/operators under state programs, they do not typically
include any costs these parties pay that are disallowed by the state.
Furthermore, amounts that owners/operators pay in excess of program limits
are not captured in state and federal data. For example, California
program officials told us that a leaking tank site in Santa Monica
contaminated the public water supply with MTBE, which is typically very
expensive to clean up. While the owner/operator estimated that it may
require $50 million to clean up the site, the state indemnification fund
limits reimbursements for cleanup costs to a maximum of $1.5 million per
tank per leak incident. As a result, the approximately $48.5 million in
additional non-reimbursable costs paid by the owner would not be reflected
in program records.

Some States Believe That Funding From Available Sources Is Inadequate to
Address All Sites With Leaking Abandoned Tanks

Some states' indemnification funds and other resources may be insufficient
to clean up all of the leaking abandoned tanks in their state. For
example, according to a survey of states conducted for the Association of
State and Tribal Solid Waste Management Officials in early 2005, claims
for the reimbursement of cleanup costs exceeded the fund balances in 10
states.11 Of the 5 states we contacted, officials of 3-Maryland, Michigan,
and North Carolina-told us that the state is experiencing difficulties in
funding cleanups at abandoned tank sites. Furthermore, officials of 2 of
these states-Maryland and Michigan-said that available resources will be
insufficient to clean up all of them and that additional resource
allocations

from the LUST Trust Fund would help address these funding shortfalls and
enhance the states ability to clean up leaking tank sites.12

Because of funding constraints, Maryland is now prioritizing and deferring
cleanups of its abandoned tank sites. The state requires tank
owners/operators to demonstrate financial responsibility to pay for
cleanup costs, which they generally do by obtaining commercial insurance
to fund cleanups of the state's nonabandoned tank sites. However, state
officials are concerned that commercial insurance may not provide a
dependable source of funding for tank site cleanups, because insurers have
sometimes been reluctant to pay cleanup costs when leaks occur. For
example, files for the Henry Fruhling Food Store site in Harford County,
Maryland (see app. I), indicated that the site's owner/operator
experienced problems in getting the insurance company to pay for cleanup
costs because he could not prove that the leak occurred during the period
of coverage. The absence of insurance funds to pay cleanup costs may lead
to more abandoned sites-sites where the owners/operators are unable to pay
the cleanup costs themselves-which will require the state to fund cleanup
with its own funds or seek federal resources. State officials told us
that, in the absence of increased allocations of federal trust funds, they
asked the state legislature to approve an increase in Maryland's special
oil transfer fee to fund the state's tank cleanup needs. The state
legislature subsequently approved the fee increase.

Since Michigan's indemnification fund was terminated in June 1995, because
of insufficient funds to pay existing and future claims, tank
owners/operators have been required to show proof of financial assurance
to cover cleanup costs in order to operate in Michigan. Small
owners/operators usually provide this proof by obtaining commercial
insurance. However, state LUST program officials cited anecdotal evidence
showing that insurance claims for remediation costs are frequently denied
because it is often difficult to prove that the release occurred under the
period of coverage. If the owner/operator cannot or is unwilling to pay
the costs and these costs are not covered by insurance or some other form
of financial assurance, the burden for cleaning up a site will fall on the
state.

In addition, Michigan program officials told us that the state's causation
standard further exacerbates the funding problem for abandoned tanks
because it requires that the state prove that the present owner/operator
is responsible for a site's contamination before it can be held
responsible for cleanup. Proving responsibility becomes difficult in cases
where releases have occurred at some point in the past and ownership of
the property has changed. If responsibility cannot be established, the
state must then fund any cleanup of the site. In addition, state officials
said that underground storage tank owners/operators acquiring properties
after March 6, 1996, can limit their liability for pre-existing
contamination by performing a baseline environmental assessment of the
property-any contamination found at that point becomes the responsibility
of the owner/operator who caused the contamination or the state if a
responsible party cannot be identified. According to program officials,
Michigan now has a backlog of 9,000 confirmed releases from leaking
underground storage tanks, an estimated 4,200 of which are at abandoned
sites. State program officials estimate that it will require about $1.7
billion in public funds to remediate these 4,200 releases alone. However,
according to these officials, resources available from all state sources
are not adequate to remediate these releases.

North Carolina's commercial trust fund can be used to assist
owners/operators with cleanup costs and to assist landowners in cleaning
up abandoned sites where the tank owner/operator cannot be located or is
unwilling to perform the cleanup. In recent months, according to program
officials, claims against the fund have exceeded revenues, causing
timeframes for paying reimbursements to stretch out over a year. As a
result, the state is now prioritizing sites based on relative risk and
directing work only to emergency releases and those leaks that pose the
highest risks that can be funded with available resources.

While neither California nor Pennsylvania are experiencing significant
problems funding cleanups of leaking tank sites, officials in both states
said that they could use more federal funding for leak prevention
initiatives and welcome the flexibility to use federal trust funds for
that purpose, as provided by the Energy Policy Act of 2005.

In an ongoing review, we are examining the scope and magnitude of states'
workload and funding needs for cleaning up contamination from leaking
underground tanks. Specifically, for each of the 50 states, the District
of Columbia, and 5 U.S. Territories, we are examining (1) how much funding
is currently available for cleaning up contamination from leaking tanks,
(2) the extent to which tank cleanup funds have been used for purposes
other than cleanups, if at all, and (3) what future revenues will be
available to clean up contamination from leaking tanks.

States Identify, Assess, and Clean Up Leaking Tank Sites Using Similar
Means

States become aware of leaking underground storage tanks through a variety
of methods, including owner/operator reports, complaints by local
residents, incidental discovery during land redevelopment or removal of
tanks for upgrading or replacement, and compliance inspections.13 Regular
and frequent tank inspections also can detect new leaks-and potentially
prevent future ones-before they can lead to serious environmental or
health damage, and lessen or avoid the need for costly cleanups. Once
contamination from leaking tanks is detected and confirmed, the 5 states
we contacted generally use risk-based systems to prioritize sites for
cleanup according to the immediate threat they pose.14 Whether funded by
the tank owners/operators, state indemnification or other funds, or other
means, states generally direct and oversee site remediation.15 However, in
circumstances where a site presents an imminent threat, has no viable
responsible party, does not qualify for funding under a state plan, or for
which the magnitude of the cost and cleanup work is beyond state
resources, the state may ask EPA to assume oversight responsibility.

States Identify Leaking Tank Sites Through a Variety of Methods

Tank owners/operators are primarily responsible for identifying,
confirming, and reporting any leaks that occur in their underground
storage tanks and dispensing systems. EPA and the states have established
a number of requirements that tank owners/operators must follow to ensure
and facilitate the early detection of possible leaks. In this regard, in
1988, EPA issued regulations governing leak detection, among other things.
Under these requirements, tank owners/operators must notify the designated
state or local authority when they discover a release or when leak
detection equipment indicates that a leak may have occurred. This
notification must generally occur within 24 hours. Tanks must generally be
monitored for leaks at least once every 30 days.

Despite these requirements, leaks can remain undetected and/or unreported.
According to state officials, owners/operators sometimes do not conduct
proper inventory checks or leak detection procedures and may intentionally
disconnect leak detection equipment. Also, tank tightness tests are
imprecise and tanks can lose small amounts of pressure or vacuum during
the test and still pass. Such small pressure leaks can result in large
releases of the tanks contents over time. In some cases, tightness tests
have failed to detect significant leaks altogether. For example, during
investigation of the Tranguch Tire Service site in Pennsylvania, the state
Department of Environmental Protection requested tank tightness test
results for that facility as well as 3 nearby tank operating facilities.
Even though test results showed that 3 of these facilities had passed
their tests, 2 of them were ultimately found to have leaking tanks,
including all 6 tanks at the Tranguch facility (see app. I). In addition
to problems in detecting leaks, some owners/operators fail to report
suspected or actual leaks once they are discovered. For example, the tank
owner/operator of the fourth facility in the Tranguch investigation did
not provide tightness test results as requested but admitted that a leak
had occurred at the site several months earlier that he had not reported.

While owners/operators identify many leaks through established testing and
monitoring procedures, EPA and officials of the 5 states we contacted told
us that many leaks are discovered only when tanks are removed for
replacement or closure. When tanks are replaced or facilities closed, in
some states-such as California, Maryland, and Pennsylvania-a state
certified or licensed environmental consultant or contractor removes the
tanks, sometimes with state or local agency oversight. Other states, such
as Michigan and North Carolina, do not require the contractor to be
certified. In Michigan, however, any person who removes or installs a tank
must have a million dollars in pollution liability insurance, according to
state officials. As part of this process, soil samples generally are taken
from the excavation and tested to determine whether contamination is
present. However, leaks are often readily apparent because of the presence
of liquid product (gasoline or diesel fuel) and/or strong fumes; state or
local environmental or health agencies may discover leaking tanks when
investigating homeowner complaints about such odors in their residences or
gasoline contamination in their well-water.

Frequently, unknown and abandoned tanks are discovered when land is being
excavated during property redevelopment. In these cases, states generally
follow the same process of sampling and testing described above to assess
contamination at the site. However, if contamination is found, the
responsibility for cleaning up these sites differs from state to state.
For example, in Pennsylvania, the new owner of the contaminated property
would be responsible for cleaning it up, according to state officials.
However, Michigan state officials told us that Michigan law limits the
cleanup responsibility to those who actually caused the contamination.
Therefore, the state would have to pay for the cleanup unless it could
identify the party or parties who caused the contamination, which can be
difficult.16 In general, cleanup costs for abandoned tanks where no owners
or operators can be found usually become the state's responsibility.

Some States' Inspection Rates May Have Limited the Timely Detection or
Prevention of Leaks

In addition to other methods for discovering leaking tanks, state or local
environmental agencies may detect leaking tanks or indications of possible
leaks while inspecting facilities for compliance with regulatory
requirements. EPA recommended that states conduct tank inspections at
least once every 3 years. However, of the 5 states we contacted, as of
mid-2005, only 2 regularly inspected their tanks as frequently as EPA
recommended, according to state officials. State officials told us that
California requires annual inspections of all tanks; Maryland inspected
its state's tanks every 3 years; Michigan generally inspected every 3
years, depending upon the location of the tanks and state inspection
staffing levels; North Carolina inspected once every 4 or 5 years, due to
funding limits; and Pennsylvania inspected at least once every 5 years.17

EPA reported that, as of September 2004, about 35 percent of the nation's
underground storage tanks were not in "significant operational compliance"
with the applicable release detection and prevention requirements,
indicating a need for greater emphasis on inspections. EPA and state
officials agreed that regular inspections of underground storage tanks
provide the opportunity to detect new leaks before serious environmental
or health damage can occur and potentially prevent future leaks. Even if
performed on a regular basis, infrequent inspections may allow violations
of leak prevention and other tank requirements to go undetected long
enough for leaks to occur and contamination to spread, potentially
resulting in environmental and health consequences and the need for costly
cleanups.

While more frequent inspections potentially could enhance preventive
efforts, state officials in 4 of the states we contacted told us that
increasing the frequency of inspections would require additional
resources. Although EPA recommended inspections at least once every 3
years, EPA program officials recognized both the value of increased
inspections and some states' need for additional resources to conduct more
frequent inspections, and supported providing more flexibility in the use
of LUST trust funds for these purposes.

In 2001, after reviewing EPA's and states' efforts to enforce UST Program
regulations, we recommended that EPA negotiate with each state to reach a
minimum frequency for physical inspections of all its tanks and present to
the Congress an estimate of the total additional resources the agency and
states would need to conduct the inspection, training, and enforcement
actions necessary to ensure tank compliance with federal requirements.18
In addition, to strengthen EPA's and the states' ability to inspect tanks
and enforce federal requirements, we suggested that the Congress consider
(1) authorizing EPA to establish a federal requirement for the physical
inspections of all tanks on a periodic basis and (2) increasing the
resources available to the UST Program, based on a consideration of EPA's
estimate of resource needs. We noted that one way to do this would be to
increase the amount of funds the Congress provides from the trust fund and
to authorize states to spend a limited portion of these amounts on
inspection, training, and enforcement activities to detect and prevent
leaks, as long as this did not interfere with tank cleanup progress.
Generally consistent with our recommendations, the Energy Policy Act of
2005, among other things, generally requires inspections once every 3
years, increases amounts authorized to be appropriated from the fund, and
authorizes these funds to be used for inspections, training, and other
enforcement and prevention activities.

States Use Risk-Based Assessments to Determine Cleanup Priorities

The 5 states we contacted all use risk-based systems to prioritize leaking
underground storage tank sites for cleanup according to the immediate
threat they pose to human health, safety, and/or the environment.

o California prioritizes cleanup sites based on risk, with the highest
risk sites remediated first. California uses many of the same procedures
employed under the American Society for Testing and Materials' (ASTM)
risk-based corrective action process. This process has 3 tiers and tables
to determine priority rankings. The highest priority is assigned to sites
that pose a threat to human health and the next highest to those posing an
environmental threat. Under this system, immediate threats are abated
first and then sites with the likelihood of future impact are addressed.

o Maryland uses a risk-based determination to prioritize both abandoned
and nonabandoned leaking tank sites. For example, if contaminated well
water is the primary threat involved, well samples are drawn and tested
and the levels of the various compounds found are compared to EPA safe
drinking water standards. Abandoned sites whose cleanup will have to be
paid for by the state are remediated if they pose an immediate threat to
public health. The cleanup of nonabandoned sites is paid for by the tank
owners and operators and begins immediately regardless of threat level.

o Michigan uses a modified ASTM four-tier classification system to
prioritize sites according to their threat. The classification system
ranges from class 1-an immediate threat to the public or environment-to
class 4-no demonstrable long-term threat. Michigan also uses a risk-based
assessment and corrective action process, based on the ASTM process, which
allows contamination to remain on-site as long as it is possible to
demonstrate that human health and the environment are adequately
protected.

o North Carolina prioritizes leaking tank sites according to three levels
of risk: high, intermediate, and low. High risk sites are those that pose
an immediate threat to human health and the environment because, for
example, a leak presents an explosion hazard from petroleum vapors or a
release is within 1,000 feet of a drinking water well. Intermediate risk
sites include those that contaminate or potentially could contaminate
surface water, a wellhead protection area, or an area that recharges
drinking water aquifers, or have groundwater contamination levels high
enough that natural attenuation may be impeded. Low risk sites involve
releases that do not fall into the other 2 categories or that pose no
significant risk to human health or the environment. The state is now
addressing only the highest risk sites and emergency releases, with the
goal of moving them to the intermediate risk level. North Carolina also
uses a risk-based assessment and corrective action process wherein more
contamination can remain on site as long as adequate protection of human
health and the environment can be demonstrated. For instance, the state
groundwater standard for benzene is 1 part per billion, but cleanup to
5,000 parts per billion may be allowed if it can be shown that the
remaining pollution poses no threat to human health and the environment.

o Pennsylvania Department of Environmental Protection uses a modified ASTM
system that classifies abandoned tank sites based on 4 priority levels.
Priority 1 sites are those that pose an immediate threat to human health,
safety, or sensitive environmental receptors; Priority 2 sites pose
short-term (up to 2 years) threats; Priority 3 sites pose long-term
(greater than 2 years) threats; and Priority 4 sites present no such
demonstrable long-term threats. Pennsylvania does not generally prioritize
responsible party lead cleanup sites addressed under the state's
indemnification fund-all eligible sites receive funding for cleanup and
are required to follow the corrective action regulations, according to
state officials.

States Generally Direct and Oversee Remediation, but May Ask EPA to Lead
or Support Certain Cleanups

Under RCRA regulations, tank owners/operators must notify the designated
state or local authority when they discover a release or when leak
detection equipment indicates that a release may have occurred. Owners and
operators must then undertake appropriate cleanup action in accordance
with the regulations. Environmental consultants, in collaboration with the
state or local environmental agency, usually perform the site assessment,
determine the technology and approach needed to contain and remediate the
contamination, and implement and complete site cleanup. The method of
cleanup selected is tailored to the specific characteristics of the site,
including the probable pathways the contamination will follow to threaten
the soil, groundwater, and/or the health of surrounding residents.
Depending on whether the contamination has reached the local groundwater,
treatment methods can range from the removal and on-site treatment of
contaminated soil to expensive on-site pump-and-treat and vapor extraction
systems, activated carbon filtration systems for municipal water systems,
and vapor extraction and water treatment units for nearby impacted or
threatened homes and businesses, among others.

EPA seldom becomes directly involved in this process unless the site is
located at federal facilities or on Indian reservations. However, states
may ask EPA to lead or support the cleanup at sites that present an
imminent threat, have no viable responsible party, do not qualify for
funding under a state plan, or for which the magnitude of the cost and
cleanup work is beyond state resources. For example, state officials asked
EPA to assume the lead on cleaning up the Tranguch Tire Service site in
Pennsylvania after they determined that site remediation costs would far
exceed the state's resources (see app. I). This leak involved the release
of an estimated 25,000 to 50,000 gallons of gasoline. The leaking fuel
reached the aquifer and the contamination plume migrated off-site into the
sewer system of the surrounding residential neighborhood. According to an
EPA Region 3 official, gasoline and gasoline fumes seeped into the
basements of 20 to 30 homes through the sewer system as well as into a
nearby creek. After conducting an environmental investigation of the area,
the Pennsylvania Department of Environmental Resources required the
owner/operator of Tranguch to begin site characterization and cleanup
work. However, in 1995, the owner of the Tranguch facility declared
bankruptcy and the state assumed responsibility for characterizing the
site and mitigating vapors in area homes. By March 1996, the state had
spent $2 million on the site and did not have the funds that were going to
be necessary to clean up the site due to its potential magnitude.
According to state officials because of the emergency nature of the
situation and funding problems, the state asked EPA to take over as the
lead agency for remediating the site, which EPA did in late August 1996.
To date, in addition to the amounts paid by the owner/operator and the
state, the Tranguch site remediation has required

over $25 million in federal funding, primarily from the Oil Spill
Liability Trust Fund.19

Conclusions

While the data that states report to EPA on underground storage tanks
provides the agency with information it can use to determine the overall
trends and status of the UST Program, the lack of specific and complete
data on known abandoned tanks limits EPA's program oversight and its
ability to efficiently and effectively allocate LUST Trust Fund resources.
Without such information, neither EPA nor the Congress can readily
determine the number of abandoned tanks requiring cleanup nationwide,
whether this number is growing, whether states are initiating and
completing or deferring work, and what the potential impacts on state
resources and, ultimately, the LUST Trust Fund may be. Furthermore,
although one of the primary purposes of the fund is to help states clean
up releases from abandoned tanks, EPA currently allocates resources to the
states without taking into account how many abandoned tanks each state
has, how many are leaking, or how many are being cleaned up. All 5 of the
states we contacted provide data to EPA on their abandoned tanks
aggregated with other tank data and separately identify and report some
limited information on abandoned tanks to EPA regional offices. Asking the
states to separately identify information on all known abandoned tanks in
the reports they currently provide to EPA should not pose an additional
burden. In any case, we believe that requiring states to specifically
report information on all known abandoned tanks would provide EPA useful
data for overseeing the UST Program and more efficiently and effectively
allocating LUST Trust Fund resources.

While the extent to which this situation exists nationwide is unknown,
officials in 2 of the 5 states told us that the state's present resources
are inadequate to cover cleanup efforts. At the same time, the LUST Trust
Fund has continued to grow through a continuing inflow of fuel tax revenue
and accrued interest-reaching a balance of about $2.2 billion at the end
of 2004-with only about $70 million to $76 million (less than 4 percent of
the total fund balance) allocated annually to support state programs.
Furthermore, the EPA and state officials we contacted believed that
greater emphasis on leak prevention activities, such as tank inspections,
is necessary to detect compliance problems that can lead to future leaks
and uncover physical evidence of leaking tanks so that states can respond
more quickly, if warranted, to prevent or limit the potential health and
environmental impacts on nearby communities. Moreover, the cost of taking
measures to prevent a release is generally much less than the cost of
cleaning up a release after it occurs. The underground storage tank
provisions of the Energy Policy Act of 2005 may lead to increased
resources for cleanups of leaking tanks and stronger enforcement efforts
that could prevent leaks and lead to the early detection of existing
leaks, thereby reducing the need for costly cleanups.

Recommendations for Executive Action

To improve EPA's oversight of the leaking underground storage tank program
and its ability to determine how to most efficiently and effectively
allocate, LUST Trust Fund dollars to the states, we recommend that the
Administrator of EPA require that states separately identify, in their
reports to the agency, information on the number and cleanup status of all
known abandoned underground storage tanks within their boundaries.

Agency Comments and Our Evaluation

We provided a draft of this report to EPA and the states of California,
Maryland, Michigan, North Carolina, and Pennsylvania for their review and
comment. In commenting on the draft report, EPA stated that, in general,
the agency thinks that the report's findings and conclusions have merit,
and that it will assess the feasibility of implementing our
recommendation. EPA agrees that the UST Program could benefit from more
specific information about abandoned tank sites. However, EPA notes that
the process that states must conduct to establish that a tank is
abandoned-that its owner is unknown or unwilling or unable to pay for leak
cleanups-may involve ownership searches to identify the potentially
responsible party and assessment of their financial ability and
willingness to pay for cleanup. With this in mind, EPA is concerned about
placing an undue burden on states by requiring them to provide specific
data on abandoned tanks. Therefore, EPA stated that, in consultation with
the states, the agency will consider how best to incorporate our
recommendation.

We share EPA's concern about placing an additional burden on states by
asking them to determine whether a given tank is abandoned by undertaking
a potentially labor-intensive and costly effort to establish who owns the
tank and whether this owner is financially able and/or willing to pay for
cleaning up a leak. However, we are not suggesting that states should make
an effort to identify unknown abandoned tanks; rather we are recommending
that they report to EPA separately the information they currently have on
tanks that they know are abandoned, and, as new abandoned tanks are
identified in the normal course of program operations, report this
information to EPA as well. This should place no additional burden on the
states. States currently provide EPA data on known abandoned tanks
aggregated with all other tanks in the state. We are simply recommending
that the states break out the data on their abandoned tanks from total
tank data. Because a limited portion of these data-information on
abandoned tanks being cleaned up using LUST Trust Fund resources-is
currently broken out and provided to EPA's regions, the UST Program could
easily utilize these existing data and EPA would only have to require
states to break out the remaining data on the number and cleanup status of
their known abandoned sites. Having more complete data on abandoned tanks
would allow EPA to better determine the potential scope of the problem and
the progress that states are making towards addressing it. It would also
permit EPA to take this information specifically into account in
allocating LUST Trust Fund resources. Given EPA's concerns, we have
clarified our recommendation by explicitly stating that EPA should require
states to provide separate data on the number and cleanup status of all
their known abandoned tanks. EPA also provided technical comments, which
we have incorporated into this report as appropriate. Appendix III
contains the full text of the agency's comments in a letter dated November
2, 2005.

Officials from the state of Maryland said that they had no comments on the
draft report. California, Michigan, North Carolina, and Pennsylvania
officials provided a number of technical comments, which have been
incorporated into the report where appropriate. In addition, Pennsylvania
officials expressed concerns similar to those raised by EPA relating to
the additional burden on states of identifying unknown abandoned tanks. As
noted, we have clarified our recommendation to address these concerns.

We will send copies of this report to the appropriate congressional
committees and to the Administrator of EPA. We will also make copies
available to others upon request. In addition, the report will be
available at no charge on the GAO Web site at h  ttp://www.gao.gov.

(202) 512-3841 or at s [email protected] . Contact points for our Offices
of Congressional Relations and Public Affairs may be found on the last
page of this report. GAO staff who contributed to this report are listed
in appendix IV.

John B. Stephenson Director, Natural Resources and Environment

List of Congressional Requesters

The Honorable Susan M. Collins Chairman, Committee on Homeland Security
and Governmental Affairs United States Senate

The Honorable Rick Santorum United States Senate

The Honorable Arlen Specter United States Senate

Information on the History and Status of Cleanup Activities at Five
Underground Storage Tank Sites Appendix I

The following are summaries of the major events surrounding the discovery
and cleanup of contamination from leaking underground storage tanks at 5
sites: (1) Coca-Cola Enterprises in Yuba County, California; (2) Henry
Fruhling Food Store in Harford County, Maryland; (3) Bob's Marathon in
Grand Ledge, Michigan; (4) R.C. Anderson Trust in Nash County, North
Carolina; and (5) Tranguch Tire Service, Incorporated, in Luzerne County,
Pennsylvania. The summaries include a chronology of significant site
occurrences as well as additional information on the amount of leaked
fuel, the contaminants involved, the impacts of the leak on the
surrounding environment, the costs of remediating the site, the extent, if
any, of EPA involvement in the cleanup, communication between state
agencies and the affected public, and litigation relating to the site.

Coca-Cola Enterprises, Yuba County, California

The Coca-Cola distribution warehouse on this site was built around 1970,
and was originally used by another business to build mobile home units.
The 5,000 gallon underground storage tank system from which the leak at
the site occurred was reportedly installed on the east side of the
building when the facility was first constructed. The property and
surrounding area are zoned for commercial use. Yuba County Municipal
Airport is located south of the site and a public drinking water supply
well operated by the City of Olivehurst is located about 850 feet east of
where the tank system was formerly located.

Chronology of Key Site Events from the Case Files

o June 1989 - August 1989: Coca-Cola Enterprises had the 5,000 gallon
unleaded gasoline underground storage tank, piping, and dispensing system
excavated and removed. The tank and piping appeared to be in good
condition, but the soil exhibited a slight petroleum odor. According to
the case file, soil samples were collected from beneath the system during
removal, as required by California law. Field observations and analysis of
soil samples determined that gasoline was present in the soil in
concentrations that required remediation. The Yuba County Office of
Emergency Services filed a report of an unauthorized leak.

o December 1989 - May 1990: The site was assessed for contamination, which
included a soil gas survey and the drilling and installation of borings,
groundwater monitoring wells, and vapor extraction test wells. Groundwater
samples were collected as monitoring wells were installed and an analysis
indicated that the groundwater at the site was contaminated. A municipal
drinking water supply well for the City of Olivehurst was discovered 850
feet east of the tank system's former location. However, analyses of water
samples from that well did not indicate any contamination. While the
contaminant plume extended 120 feet east of the release site, it was
confined within the south and east boundaries of the property.

o May 1990 - December 1990: The Yuba County Air Pollution Control Office
authorized the on-site aeration of contaminated soils that had accumulated
during drilling activities. The soil aeration began in June 1990 and was
successfully completed in September 1990. A quarterly groundwater
monitoring and sampling program also began in June 1990. In August, a
vapor extraction pilot test was conducted at the site. Petroleum was found
in a monitoring well at the site in September and about 7.5 gallons of
gasoline were removed from the well from September through November. By
December, approximately 0.08 feet of gasoline remained in the well.

o February 1991 - August 1991: The environmental consultant for Coca-Cola
Enterprises submitted its contamination assessment report, which assessed
the extent of vertical and lateral petroleum contamination at the site
during the underground storage tank removal operations. The consultant
also completed a remedial action plan that presented a conceptual system
design for remediating hydrocarbon contamination in soil and groundwater
at the site. The plan proposed an integrated remediation system
incorporating (1) a groundwater pump and treat system with an air stripper
system to remove volatile compounds from groundwater, (2) a vapor
extraction system to remove vapors from contaminated soils in the
unsaturated zone, and (3) a thermal oxidizer to burn off the vapors. The
consultant estimated that remediating the site would take 3 years, with
one additional year for monitoring and closure. Cleanup oversight was
transferred from the Yuba County Office of Emergency Services to the
Regional Board in February 1991, and in March 1991, the Board approved the
contamination assessment report.

o April 1993 - September 1998: The recommended remediation system began
operation in April 1993. In August 1996, an oxygen release compound was
installed in three perimeter vapor extraction wells to increase the
dissolved oxygen concentrations in groundwater to enhance the natural
bio-attenuation of petroleum hydrocarbons. The pump and treat system was
operated intermittently until mid-November 1997, when the system was shut
down to prepare for site closure following approval from the Central
Valley Regional Water Quality Control Board. Groundwater monitoring and
sampling continued on a quarterly basis. In September 1998, the Board
approved initiation of closure activities.

o October 1998 - Early 2000: Monitoring indicated that the contamination
plume might have migrated southward in late 1999 and early 2000 and that
further remediation and changes in the sampling and analysis program were
warranted. The remediation system was restarted in December 1999, with
upgrades to increase flow rates and optimize efficiency and was completed
in January 2000. The revised sampling and analysis program was initiated
shortly thereafter. A survey to identify water wells within 2,000 feet of
the leak site conducted by the site's environmental contractor identified
2 municipal wells, with the nearest approximately 850 feet east of the
former tank location. However, sampling and analysis conducted by the
Olivehurst Public Utility District revealed that the closer of the 2 wells
had not been affected by the contamination plume.

o April 2002 - March 2004: Because monitoring and sampling showed no
hydrocarbon concentrations in the groundwater monitoring wells, the system
was again shut down in April 2002. Quarterly groundwater monitoring after
the system was shut down showed that the contamination plume had
stabilized and, in July 2003, the site's owner requested the Central
Valley Regional Water Quality Control Board's approval to close the
monitoring wells. The wells were abandoned by late December 2003, and the
Board closed the remediation case in March 2004.

o Status as of August 2005: Cleanup was completed.

Summary of Key Information from the Case Files

o Contaminants and compounds of concern: Gasoline (total petroleum
hydrocarbons as gasoline-TPHG), benzene, toluene, ethyl-benzene, xylenes
(BTEX)

o Size of leak: Unknown.

o Impacts of contamination: Soil and groundwater contamination occurred at
the site but was contained within the property.

o Remediation cost: The California underground storage tank fund spent
$1,202,745 to reimburse site owners/operators for site remediation costs.
Additional amounts that may have been paid by Coca-Cola Enterprises that
were not reimbursed are unknown.

o U.S. Environmental Protection Agency involvement: None.

o Communication between responsible agencies and the public: No evidence
of public meetings appears in the case files.

o Litigation: Case files show no evidence that any lawsuits were filed
relating to this site.

Henry Fruhling Food Store, Harford County, Maryland

The Henry Fruhling Food Store was a single family dwelling with an
attached small grocery store. The store had two 1,000 gallon underground
storage tanks and distribution systems for gasoline, which were installed
around 1966, and a 500 gallon underground tank and distribution system for
kerosene. Sometime prior to 1966, 2 similar underground gasoline tanks and
distribution systems were located on the site but were removed by the
previous owner.

Chronology of Key Site Events from the Case Files

o 1970: A nearby resident complained to the owners of the store about an
odor or taste of gasoline in his well water. The tank maintenance company
performed a pressure test on the tanks and distribution systems but found
no leaks.

o June 1980 - September 1980: Water samples taken by the Harford County
Health Department in response to another nearby resident's complaints
about gasoline in his well water indicated petroleum contamination. That
resident was warned not to consume water from his well and the matter was
referred to the Maryland Department of Natural Resources (the predecessor
of the Maryland Department of Environment) for action. That resident also
filed a complaint with the department concerning the presence of oil in
his well, which was referred to the Environmental Health Administration
and the Harford County Health Department for follow-up actions.

o October 1980: One of the store owners was badly burned-and later
died-when gasoline fumes ignited in his basement.

o November 1980: Water samples obtained by the Harford County Health
Department identified gasoline in the store owner's well and the case was
assigned to the Department of Natural Resources for enforcement action.

o January 1981 - February 1981: In mid-January, the Department of Natural
Resources notified the company that had maintained the tanks and pumps
since 1976 of their determination that a pollution violation had occurred.
The Department ordered the company to (1) stop discharging petroleum
products into state waters, (2) test the tightness of the tanks and supply
systems, and (3) initiate actions to recover petroleum from groundwater at
the site. In late January, tightness tests were performed on both tanks
and supply systems and no leaks were found. In early February, the
Department took auger probes at various locations throughout the Fruhling
property and found explosive vapors in the soil in the vicinity of the
tanks and pumps. In a letter dated February 27, the Department told the
owners and the tank maintenance company that (1) they had achieved
substantial compliance with their order and that no gasoline was then
leaking into the groundwater, (2) based on information they had provided
the Department, there was a "strong probability" that, at some time during
the past several years, repairs had been made to the gasoline pump that
may have eliminated a leak in the system, and (3) the results of the
recent boring tests and a survey of the area indicated "a very low
probability" that the gasoline in the groundwater was coming from any
source other than the system at Fruhling's store. Furthermore, the letter
stated that the Department's only concern was the removal of any
recoverable gasoline from the groundwater at the site.

o July 1981 - August 1981: A gasoline recovery and separator system was
installed at the site and well-pumping operations began. The accumulated
effluent from the separation process was sampled by the Harford County
Health Department on a periodic basis and was spread back on the ground
with the knowledge and approval of the County Health Department. This
process continued through January 1988, at which time a Department of
Environment official ordered the owner to stop the discharge.

o November 1981 - December 1981: The Department of Natural Resources
notified both the store owner and the tank maintenance company that they
had satisfactorily removed the gasoline from the well and complied with
their January 1981 order. The Department also advised them that the
remaining unrecoverable gasoline in the groundwater was a pollution
problem that would be referred to the Harford County and state health
departments for appropriate action. Because laboratory results of samples
taken continued to show unacceptable levels of "aromatic hydrocarbons",
the County required the owner to continue well pumping until the residuals
were reduced to an acceptable level.

o May 1983: A nearby resident raised concerns regarding contamination of
his well. Water samples taken over approximately the preceding three year
period showed minimal contamination from petroleum products.

o October 1987 - December 1987: Testing indicated that gasoline
contamination was migrating into a new well at the site.

o January 1988: In late January, an official from the Maryland Department
of the Environment (formerly the Department of Natural Resources)
investigated a complaint that effluent runoff from the site was flowing
onto a neighbor's property. The investigating official informed the store
owner that spreading effluent on the ground was no longer allowed because
this process allowed pollution to migrate back into the soil and
groundwater. The owner was issued a site complaint and was directed to
shut down well pumping operations until further notice. At this point,
pumping and effluent discharge on the ground had gone on for over 6 years
and the gasoline separation unit had been removed and water had been
pumped onto the owner's yard for the last 2 years.

o June 1988: The state took over cleanup operations at the site after the
LUST Trust Fund was established in 1988. Well-pumping from the old well at
the store site continued intermittently until mid-1988.

o October 1988: The Maryland Department of the Environment oversaw the
removal of the underground gasoline and kerosene tanks and distribution
systems, which was funded from the LUST Division account. Inspection
revealed no gasoline storage tank perforations, but soil beneath one
gasoline storage tank showed explosive readings. According to a state
official who visited the site in early December 1987, the store owner
stated that the tank maintenance company had pumped the tanks dry prior to
going bankrupt, but she did not recall the exact date.

o January 1989 - April 1989: The Maryland Department of the Environment
contracted with an environmental consulting company to perform 2 soil gas
surveys at the site to delineate the extent of subsurface gasoline vapor
contamination. Analysis of the samples revealed the presence of elevated
gasoline vapor levels at the site, with the highest concentrations
detected near the former pump island.

o Late 1989 to early 1990: The state installed charcoal filtration units
on the store owner's water system and that of a nearby neighbor.

o March 1993 - November 1993: The Maryland Department of the Environment
retained a consultant to review ongoing remediation activities at the site
and determine the adequacy of activities to control and abate
contamination. The consultant concluded that the contamination plume
appeared to be getting larger and that vapor recovery efforts were
inadequate. In November 1993, the Department had the consultant test a
combined air sparge and soil vapor extraction system.1 Test results
indicated that this system could significantly enhance the existing
recovery system.

o March 1997: The Maryland Department of the Environment continued to
operate the pump-and-treat system at the Fruhling residence. The system
had treated over 2 million gallons of water. The former Fruhling domestic
well, shallow monitoring wells, and deep monitoring wells all continued to
show elevated levels of dissolved gasoline constituents. As a result, the
Department periodically operated a soil venting system to keep the wells
within discharge guidelines. Periodic sampling of the surrounding
residences identified no additional contaminated domestic wells.

o 1999: The groundwater recovery system at the site was shut down.

o September 2001: The Maryland Department of the Environment maintained a
granular activated carbon treatment system at two residences and performed
quarterly sampling at six residences. Petroleum contamination was still
present at the Fruhling property and MTBE levels were detected at the
residence across the street from the Fruhling property. Dissolved
petroleum contamination levels had decreased at all sampling locations.

o Status as of August 2005: The site recovery system remained on site, but
was turned off. The site was being monitored in operation and maintenance
status, with sampling performed every three months. No significant
contamination had been detected in the residential wells from which
samples had been taken since July 2001. However, well-monitoring still
showed some signs of low-level contamination.

Summary of Key Information from the Case Files

o Contaminants and compounds of concern: Benzene, toluene, ethyl benzene,
total xylenes, and MTBE.

o Size of leak: Unknown; Very little liquid product (gasoline) was
recovered.

o Impacts of contamination: One death occurred from a leak-related
explosion and fire. Groundwater and residential drinking water wells were
contaminated with petroleum products. Real estate development and sales
were impeded. Residents were granted a reduction in property taxes.

o Remediation cost: Maryland spent about $708,595 in state funds to
remediate the site. Prior amounts that might have been spent by the
company that installed the tanks and the store owner are not included.

o U.S. Environmental Protection Agency involvement: None.

o Communication between responsible agencies and the public: In December
1989, officials of the Maryland Department of the Environment met with
affected parties to discuss contamination at the Fruhling site. In May
1990, the Department held a follow-up public meeting to present the
results of its initial investigation of the groundwater contamination.

o Litigation: In late July 1981, one of the affected residents filed a
lawsuit against the company that installed the tanks on the site, the tank
maintenance company, and the store owner. Consequently, the tank
maintenance company's insurer retained a consulting company to investigate
the alleged contamination of groundwater by the producer's petroleum
products. In May 1982, the consultant for the insurance company concluded
that the products of the oil company represented by both the tank
installer and the tank maintenance company were not responsible for the
groundwater contamination in the affected resident's well. In March 1988,
the lawsuit was settled out of court for $25,000, with the defendants
expressly denying liability. Prior to the settlement, the insurance
company for the store owner settled with the affected resident for $7,500.

In 1990, the owner's insurance company denied responsibility for any claim
under the owner's policy. Nevertheless, in early 1991, the state of
Maryland sued the owner to recover cleanup costs.

In October 1990, a neighbor in the area of the site sued various parties
involved in the purchase of his property, including the real estate
company, the real estate agent, and the former owners of his house for not
disclosing the groundwater contamination at the time of purchase.

Bob's Marathon, Grand Ledge, Michigan

Bob's Marathon is a gasoline service station and automobile repair shop
bordered by mixed-use commercial and residential properties in the city of
Grand Ledge, Michigan. Two reported releases occurred at the facility and
the released gasoline migrated toward a municipal water supply well
located directly down-gradient and very close (approximate 800 feet) to
the site. MTBE, benzene, and other gasoline components from this spill
potentially impacted the city's water supply for about 8,300 people.

Chronology of Key Site Events from the Case Files

o April 1986: Bob's Marathon registered all three of its underground
storage tanks with the state.

o December 1991 - February 1992: Two of the three tanks failed their
tightness tests. One of the owners reported to the Michigan State Police
Fire Marshall Division that she had discovered a leak during a routine
tank gauging inventory check. The leak involved the loss of approximately
4,500 gallons of gasoline from a 6,000 gallon tank. An environmental
consultant retained by the owners sent the Michigan Department of Natural
Resources (MDNR) the required 20-day report of initial abatement measures
and installed eleven monitoring wells at the site in December. The
monitoring wells were used to determine the directional flow of
groundwater at the site and intercept the gasoline plume. In January, the
consultant installed a product skimming system in six monitoring wells and
a passive recovery system in three wells that reportedly contained product
sheen on the water table and additional monitoring wells. Also in January,
the consultant submitted a site investigation work plan, site
characterization report, free product removal report, and interim
corrective action plan to MDNR. MDNR concluded that the interim corrective
action plan was unacceptable and provided the consultant with a list of
concerns in a deficiency letter. The environmental consultant estimated
that, by late February, 280,300 gallons of contaminated groundwater and
1,200 gallons of gasoline were removed by the skimming system.

o March 1992: Because the consultant's response to the MDNR deficiency
letter was not adequate, MDNR did not approve the work plan. The
consultant then recommended that a second consulting firm with a greater
capacity to more cost-effectively manage long-term projects take over the
work at the site. The new consultant submitted an interim corrective
action plan and a site investigation work plan to MDNR for approval. MDNR
approved the second consultant's interim corrective action plan. In an
interoffice communication, a MDNR official recommended that the first
consultant be denied payment for work conducted at the site and that MDNR
consider the consultant a potentially responsible party because of its
failure to take timely action to abate the situation at the site. Three
new underground storage tanks were installed on the west side of the
service station building-two 6,000 gallon tanks and one 15,000 gallon
tank. Approximately 400 cubic yards of soil were removed and disposed of
during excavation for these tanks.

o April 1992: MDNR tentatively approved the site investigation work plan.
The groundwater remediation system began operating.

o June 1992: An MDNR official stated that the department approved an
interim groundwater treatment system at the site because of the close
proximity of municipal wells and that this action was necessary because of
the first consultant's failure to take timely action to abate the spread
of the contamination.

o January 1993: The company operating the groundwater treatment system
decided to no longer operate and maintain it because of uncertainty
regarding reimbursement from the Michigan Underground Storage Tank
Financial Assurance program for future work.

o February 1993: The site owner replaced the second consultant with a
third after a dispute over the need to purchase the remediation equipment
and other issues. This third environmental consultant made modifications
to the existing groundwater treatment system.

o May 1993: MDNR informed the owners of Bob's Marathon that they had
failed to define the full nature and extent of the groundwater
contamination. While the groundwater plume was advancing toward the Grand
Ledge municipal well field, the leading edge of the plume had not yet been
defined. Furthermore, MDNR said that the groundwater system was
ineffective and the contamination plume continued to migrate, impacting
additional groundwater. As a result, MDNR requested that the owners
provide all information on the releases and investigations of the releases
including all soil and groundwater response actions and investigations.
MDNR conditionally approved the amended third consultant's site
investigation work plan. Approximately $850,000 of state financial
assurance program funding had been spent at the site.

o June 1993: An oil/water separator was added to the groundwater treatment
system.

o July 1993 - August 1993: The site's third environmental consultant
informed MDNR that, due to delays in payment from the state financial
assurance program, it was unable to continue site investigation activities
at the site. In light of this development, MDNR reminded the site owners
of their obligation to conduct all appropriate corrective actions to
remedy the environmental problems caused by the release of contamination
at the site including eliminating any impacts to the Grand Ledge municipal
well field. The owners' attorney informed MDNR that the owners were unable
to proceed with site investigation and remediation activities without the
assurance of funding. In response, MDNR said that if all current claims
were approved and paid by the state financial assurance program, the one
million dollar limit for reimbursement under the program would have been
reached at the site and the owners would be responsible for financing the
remaining corrective actions including a final remedy. The Michigan
Department of Public Health notified the city of Grand Ledge of MTBE
contamination of a municipal water supply well.

o October 1993: MDNR notified the owners that the site would be listed in
the "Proposed List of Michigan Sites of Contamination" for fiscal year
1995.

o November 1993: A second leak of 400 to 800 gallons of gasoline was
discovered and reported to the Michigan State Police.

o December 1993: According to the third consultant's initial abatement
report, the second leak was discovered by the owner/operator when he
noticed a strong petroleum odor in the site treatment building. When he
opened the cover of equipment used as part of the cleanup system, he
observed approximately one foot of gasoline. The owner then inspected the
underground storage tank system and found a mixture of water and gasoline
in the area of one tank, due to a pin-hole leak in a gasoline supply line.
The leak detection equipment installed on the system-that should have
detected the leak, sounded an alarm, and automatically shut off the
system-was not functioning. The tank system was taken out of service until
the perforated line could be replaced.

o January 1994: The groundwater treatment building was damaged by fire.
According to the fourth consultant's investigation report, the owner had
discovered gasoline in the equipment and a fire started, damaging the
equipment, before he could remove it. (As of June 2005, the equipment had
not been restored to service).

o March 1994: MDNR advised the owners of their responsibility to repair
the fire damaged system and conduct hydrogeological studies related to
both leaks at the site. The owners responded that they could not continue
the remediation work required to clean up the site and they terminated the
services of the consultant at the site. MDNR assumed control of the
investigation and cleanup of the leak at the site.

o April 1994: MDNR obtained emergency funds to complete the groundwater
investigation and develop a corrective action plan. MDNR hired a
consultant to update and collect additional information for the site with
the overall goal of protecting the Grand Ledge municipal water supply from
contamination originating from the site.

o December 1994: The city of Grand Ledge expressed concern that levels of
benzene in a municipal well continued to increase, indicating continuing
migration of contamination. The city asked MDNR for monitoring well test
results and a report on the current status of the remediation by the end
of the month.

o February 1995: The mayor of Grand Ledge asked a state representative to
intercede with state agencies to facilitate the issuance of all permits
and release of state funds needed to allow the design and construction of
soil vapor extraction and groundwater extraction and treatment systems to
proceed immediately.

o July 1995: Grand Ledge allowed access to the city well field to
construct and maintain a water treatment system for the contaminated
municipal well and a groundwater blocking well.

o December 1995: The system to treat well water contaminated with volatile
organic compounds began operating and a barrier well was installed to
prevent the plume from continuing to reach the municipal well.

o Status as of August 2005: Treatment facilities were operating and
cleanup was ongoing. Michigan Department of Environmental Quality
officials told us that the air sparge and soil vapor extraction systems
were recently turned off to conduct performance monitoring but carbon
treatment of the impacted municipal well was ongoing. According to these
officials, they expected to complete site cleanup between 2007 and 2010.

Summary of Key Information from the Case Files

o Contaminants and compounds of concern: Benzene, toluene, ethyl benzene,
xylenes, and MTBE.

o Size of leak: The first release was approximately 4,500 gallons of
gasoline; a second release was 400 to 800 gallons of gasoline.

o Impacts of contamination: The leak impacted the water supply for the
city of Grand Ledge and the city had to provide potable water to about
8,300 residents.

o Remediation cost: As of about March 2005, approximately $2,150,000 had
been spent to clean up contamination from the site. Approximately $950,000
of this amount came from the Michigan Underground Storage Financial
Assistance Fund to reimburse costs incurred by the owner prior to the
state taking over site remediation. According to Michigan Department of
Environmental Quality officials, continuing efforts to clean up the site
through 2007 to 2010 will involve additional costs of up to approximately
$500,000.

o U.S. Environmental Protection Agency involvement: None.

o Communication between responsible agencies and the public: None
identified.

o Litigation: A number of lawsuits have been filed relating to the leak at
Bob's Marathon, according to Michigan environmental officials. An
apartment complex east of Bob's Marathon filed a lawsuit against the
facility's owners. In addition, the owners are involved in litigation with
the first two consultants and have filed a lawsuit against the facility's
gasoline supplier, which is, in turn, suing the manufacturer of the hose
that caused the site's second leak.

R.C. Anderson Trust, Nash County, North Carolina

The R.C. Anderson Trust site was owned by R.C. Anderson from 1949 to his
death in 1984, when the property was passed on to his heirs and was
managed as a trust by a bank. Three businesses were located on the site-a
gasoline station (abandoned), a tractor dealership (subsequently a
furniture store), and an automobile repair garage. Contamination was first
reported to the North Carolina Department of Environment, Health, and
Natural Resources (DEHNR) in July 1992, during removal and closure of the
underground and above ground storage tanks and excavation of the soil
around the tanks and pump island. Land uses in the vicinity are
commercial, agricultural, and single family residential.

Chronology of Key Site Events from the Case Files

o July 1992 - September 1992: In July 1992, three underground gasoline
storage tanks and one above ground diesel tank were removed. Evidence of a
gasoline release from a 3,000 gallon tank was discovered and reported to
DEHNR. An August 1992 closure report also concluded that there was "high
potential" of petroleum contamination on-site. Accordingly, in September
1992, DEHNR notified the bank that was acting as trustee for the R.C.
Anderson property, that it was in violation of pollution control rules and
regulations and must take action to comply with corrective action rules.

o December 1992: The environmental consulting company completed a
Comprehensive Site Assessment/Corrective Action Plan to assess the
surrounding conditions and risks to area populations from the remaining
contamination at the site. This report was filed to satisfy the
requirements of the North Carolina law pertaining to investigations for
soil and water cleanup.

o May 1993: DEHNR reviewed the Comprehensive Site Assessment report,
determined it to be inadequate and required the bank to submit (1) a more
complete report which adequately identified the full vertical and
horizontal extent of the contamination plume(s) and (2) a corrective
action plan. Both reports were to be submitted by July 15, 1993.

o October 1993: The environmental consultant resubmitted the Comprehensive
Site Assessment report to DEHNR for review. The report identified several
sources of pollution affecting soils and groundwater at the site. One was
a 3,000 gallon underground storage tank that had two large rust holes.
Another source was the fuel pump island where stained soil was found when
it was removed. In addition, the soil below the 10,000 gallon above ground
storage tank (which had held diesel oil) was stained at the fill area from
apparent spills during filling operations. The above ground tank itself,
however, showed no evidence of leaks. Last, used motor oil was drained
onto the ground near the gasoline station building when it was being used
as a truck repair center. The area surrounding the site was surveyed for
water wells, public water supply intakes, and off-site monitoring wells
for potential receptors and migration pathways. Sixteen private water
supply wells and another 32 "suspected" water wells were observed within a
1,500 feet radius of the site. There were no public water supply intakes
identified within one-half mile of the site and no off-site monitoring
wells were found within 1,000 feet of the site.

o November - December 1993: Excavation and treatment began in 1993. A
smaller (530 gallon) underground gasoline storage tank was discovered
during excavation of contaminated soil and removed. While the tank was
located in an already contaminated area of the site and was rusted, it
showed no sign of leaks, according to the consultant. A closure report on
this tank dated the end of December was filed with the Raleigh Office of
DEHNR.

o January - December 1994: A second round of soil excavation with on-site
bioremediation procedures was performed in 1994. In early March, the
environmental consultant completed a Corrective Action Plan that included
the excavation and treatment of contaminated soils and an air sparging and
pump and treat facility for groundwater remediation. Under the plan (1)
soil treatment was to be completed by August 1994; (2) a groundwater
treatment system was to be installed by January 1995, and operated for 5
to 15 years; and (3) system shut-down and project completion dates would
be based upon monitoring test results and state approvals. In late March,
DEHNR approved the Comprehensive Site Assessment. In July, the Raleigh
Office of DEHNR issued a soil contaminant and treatment permit. During
1994, soil excavation and on-site bioremediation procedures were performed
on approximately 6,400 tons of contaminated soil at the site. An estimated
total of 11,792 tons of soil were excavated and treated on-site

o January - December 1995: Following completion of the soil treatment, the
environmental consultant monitored groundwater quality at the site and
reported results on an approximate quarterly basis. The samples taken in
October continue to show groundwater contamination. The contamination
plume was estimated at approximately 230 feet by 210 feet and expected to
migrate slowly to the northeast. The reports continued to recommend the
design and installation of a groundwater remediation system to remove the
contamination present.

o January 1996 - January 2003: Groundwater monitoring continued on a
semi-annual basis. In 1996, North Carolina enacted a law temporarily
suspending remediation work for low-priority underground storage tank
release sites. The R.C. Anderson Trust site was initially given a
low-priority ranking and, therefore, remediation work at the site stopped.
However, primarily because of the site's threat to uncontaminated private
domestic water supply wells, its ranking was changed to high priority in
July 1997 and remediation activities resumed. The Containment and
Treatment of Contaminated Soil permit originally issued In July 1994 by
the Raleigh Office of DEHNR was renewed in 1998. A groundwater remediation
system using pump and treat with air sparge technologies was installed and
began operation in late May 2002. Groundwater at the site was sampled 10
times through January 2003. According to the consultant's Groundwater
Monitoring Report dated September 12, 2003, as of January 2003, benzene,
lead, MTBE, and 1,2-dichloroethane were still present at on- and off-site
monitoring wells in concentrations above North Carolina groundwater
quality standards. However, according to DEHNR officials, benzene was not
detected in any off-site monitoring well during that monitoring event and
lead was detected in an off-site monitoring well but not above the NC
groundwater standards.

o Status as of August 2005: According to state officials, North Carolina
State Session Law 2004-124 suspended further work on most high risk sites
due to constrained state funds. As a result, the state is now prioritizing
sites based on relative risk and directing work only to emergency releases
and those releases that pose the highest risks that can be funded with
available resources. Initial treatment work was completed at the Anderson
Trust site and the recovery system remains on-site but is currently shut
down.

Summary of Key Information from the Case Files

o Contaminants and compounds of concern: Benzene, toluene, ethyl benzene,
xylenes, MTBE, Naphthalene, and lead.

o Size of leak: Unknown.

o Impact of contamination: One residential drinking water well adjacent to
the site was contaminated and abandoned. A potential threat of
contamination exists for 17 additional residential wells within 1,500 feet
of the site.

o Remediation cost: According to state officials, the North Carolina
Commercial Fund has reimbursed the owner (including consultants) for
$943,407.93 of reasonable and necessary expenses performed to remediate
the site. In addition, total reimbursable expenses to complete cleanup and
close the site are estimated by these officials at $1.1 million. However,
this estimate does not include deductible amounts and other expenses not
approved by the fund or otherwise deemed ineligible for reimbursement,
such as contamination from the above ground tank or used motor oil paid by
the R.C. Anderson Trust.

o U.S. Environmental Protection Agency involvement: None.

o Communication between responsible agencies and the public: No evidence
of formal public meetings was identified.

o Litigation: No information on lawsuits was found in the case files.

Tranguch Tire Service, Inc., Luzerne County, Pennsylvania

The Tranguch site was a gasoline and tire retreading service station in a
mixed commercial and residential area of northeastern Pennsylvania that
was abandoned in 1995. Several operating gasoline service stations as well
as numerous abandoned or removed underground storage tank systems lie
within the vicinity of the site. A residential neighborhood, part of which
is built over an abandoned coal mine, surrounds the site. While it is
unknown exactly when the underground storage tanks at the site began to
leak, residents' complaints of gasoline odors in their homes suggest that
leaks may have begun sometime in the late 1980s to early 1990s. By 1993,
the Pennsylvania Department of Environmental Resources (PADER) determined
that gasoline vapors from the sewer system were affecting homes. Although
PADER found contamination at four other facilities in the vicinity of the
Tranguch site, the department determined that the Tranguch facility was
primarily responsible for the leak impacting the residential area. EPA
estimated that the facility had released an estimated 25,000 to 50,000
gallons of gasoline. The resulting gasoline plume contaminated soil and
groundwater, and spread generally northeastward through the adjoining
community to encompass about a 70-acre area, including 11 businesses, two
doctor's offices, two churches, two parks, 26 vacant lots, and 359
residential properties and impacted the lives of up to a reported 1,500
neighborhood residents.

Chronology of Key Site Events from the Case Files

o Prior to 1990: Neighborhood residents near the Tranguch facility had
complained of an odor from the facility that smelled like automobile or
truck emissions as early as August 1976. However, an investigation
conducted at the Tranguch facility at that time did not reveal any
problems. Case files do not contain any additional complaints about this
site until February 1990.

o February 1990 - April 1993: Over the three-year period, PADER
investigated complaints of gasoline or other odors in residences in the
area of the Tranguch facility, including one home on three separate
occasions. In March 1993, a local Department of Public Safety
environmental protection specialist performed an investigation at this
residence and verified the presence of strong gasoline odors. Because of
the saturated condition of the soil, he was able to trace gasoline residue
to a nearby abandoned underground storage tank facility. The environmental
specialist referred the matter to PADER for follow-up. In April 1993,
PADER took a water sample from the basement sump of this residence that
tested positive for the presence of gasoline, and began efforts to
determine whether nearby underground gasoline storage tanks were the
source of the contamination. Out of eleven commercial locations in the
vicinity, PADER identified four operating facilities-including the
Tranguch facility-and one abandoned underground storage tank facility as
potential contamination sources.

o May 1993 - June 1993: PADER directed the owner of the abandoned facility
to register and either properly close (remove) or upgrade the tanks at
that site. At this time, PADER also asked the owners of the Tranguch
facility and the other two operating facilities to either provide the
department with proof that their tanks passed tank tightness tests or
conduct the tests.2 The abandoned site owner notified PADER that he
intended to close the site and PADER directed him to submit and implement
a site characterization plan. Because none of the three operating
facilities responded to the PADER request, it asked for this information a
second time.

o July 1993: The owner of one of the three operating facilities submitted
a report to PADER showing that, in June 1993, tanks at the site had passed
a tightness test. However, PADER discovered that the owner had installed
this facility's current tanks in 1991 to replace older tanks and, at that
time, 1,042 tons of fuel-contaminated soil had been removed from the site.
The older unregistered tanks remained in operation until they were
replaced in 1991. The owner filed a closure report for these tanks in
October 1993. PADER documentation indicates that the report lacked some of
the required sampling and soil analysis information.

Also in July 1993, the owner of another of the three operating facilities
admitted to having had an unreported release in April 1993. Tanks at this
site were subsequently removed. Three of these tanks had never been
registered. Soil contamination was evident during excavation, and during
the removal process an abandoned heating oil underground storage tank was
also discovered at the site. Also, when the owner of the facility removed
his tanks, PADER observed contamination at the site and the owner arranged
for a site investigation/characterization.

o August 1993: PADER requested a third time that the owner of the Tranguch
facility provide tank tightness test information and submit and implement
site assessment and remedial action plans. PADER surveyed leak detection
methods used at the Tranguch site and the owner stated that the facility
was performing these methods. These methods were required to be performed
under the state's applicable rules and regulations in force at that time.
PADER also requested that the owners of the other two operating facilities
and the abandoned facility submit and implement site assessment and
remedial action plans.

o September 1993: The owner of the Tranguch facility submitted information
to PADER indicating that tanks at the site had passed a tightness test.
PADER again requested that the Tranguch owner perform a site
characterization and report the results to the department within 14 days.
Also, the local fire department received two additional complaints of
gasoline-like fumes in area residences. PADER Emergency Response Program
personnel also began regularly monitoring vapor levels in area homes.

o October - November 1993: Beginning in October, the PADER Emergency
Response Program started to install interim remedial systems designed to
prevent gasoline vapors from entering affected homes and, by mid-May 1994,
had installed thirteen.3 Also in October, PADER identified a fourth
operating facility as a potential contributor to the area's contamination
and asked the owner of that facility to perform a site characterization.
Through late November 1993, the owner of this fourth facility took no
action in this regard, but requested to review any PADER documentation
indicating that this facility had contributed to pollution in the area, as
well as files regarding the other facilities in the area. Although PADER
arranged for access to these files, the owner never reviewed them.

PADER confirmed the presence of gasoline-like fumes in six area residences
and arranged to have a preliminary subsurface investigation performed on
the area impacted by the contamination. The investigation revealed that
gasoline had contaminated the groundwater at several locations down-slope
from the Tranguch site and that gasoline contamination had spread to the
neighborhood sewer system. The gasoline contamination was also found to
stop up-slope of the location of the Tranguch tanks, within the property
boundary. PADER officials, representatives of the city of Hazleton, and
the city fire chief met with the owner of the Tranguch site and requested
that he immediately remove all gasoline from his tanks. The following day,
the owner informed PADER that he had discovered that 375 gallons of
gasoline had been lost from his tanks early in November. The owner
admitted to PADER officials that-contrary to what he had told PADER in
August 1993-he had just started to comply with the required leak detection
methods the previous day. Accordingly, PADER issued a second compliance
order to the Tranguch site owner, requiring him to, within 24 hours,
remove all gasoline from the underground storage tanks on-site, begin
cleaning up the leaked gasoline, and take steps to monitor and mitigate
vapors in area homes. Although the owner appealed this order, he had his
tanks and lines drained and began monitoring wells to recover gasoline.
Furthermore, although the Tranguch site owner made inquiries regarding
homes impacted by vapors, he took no action to monitor or mitigate vapors
in area homes.

City personnel began venting the neighborhood sewer system.

o December 1993 - February 1994: PADER held a public meeting with area
residents, city and township representatives, and state legislators
regarding the spill. School district officials, city and township
representatives, and a state legislator were also contacted during
subsequent site activities.

In December, PADER issued a compliance order to the owner of the fourth
operating facility suspected of contributing to the contamination. The
order required a complete site investigation, including tank system
tightness testing, a review of leak detection and inventory records, and
sub-surface sampling and analysis of soil and groundwater at the site. The
owner's attorney responded with a letter stating his client's intent to
appeal the order, and included attachments with information in defense of
his decision. From the information presented in the attachments, PADER
determined that the owner had not employed any type of automatic leak
detection devices on the two underground gasoline storage tanks installed
at his site in 1962, and that he might not have

complied with applicable federal and state leak detection regulations.4 In
January 1994, the legal counsel for PADER informed the owner that the
information he had submitted was incomplete and did not satisfy the
compliance order's requirements. Later in January, the owner formally
appealed the December compliance order.

PADER documents indicate that problems with gasoline-like vapors entering
neighborhood homes and commercial establishments grew progressively worse
through the end of 1993 and, by early 1994, 28 residences and 1 commercial
building had been impacted. The Hazleton City Health Officer determined
that a neighborhood home was unfit for human habitation and the owner
temporarily relocated because of the presence of potentially harmful
gasoline vapors and the explosion potential from the collected gasoline
vapors in the basement of his home. (This homeowner had previously
reported gasoline-like odors in his residence in February 1990, March
1992, August 1992, and March 1993).

In January 1994, all product recovery efforts at the Tranguch site ceased
because of the owner's failure to pay his environmental consultant, but
resumed after he was able to secure a loan.

o March - April 1994: Local residents formed an organization called the
Group Against Gas (GAG) at about this time and announced plans for a class
action lawsuit against parties responsible for the contamination. PADER
received from the owner's consultant a proposed Tranguch site
characterization plan that included monitoring well locations. Upon
review, PADER requested that the consultant perform additional site
characterization work, including the installation of additional monitoring
wells. Also, PADER notified the owner of the fourth operating facility of
a June 1994 hearing date and the owner wrote to his state senator in an
unsuccessful attempt to stay the PADER compliance order.

o May - August 1994: In May, PADER held a second public meeting with area
residents, city and township representatives, and state legislators
regarding the spill. Also in May, to avoid compliance proceedings, the
owner of the fourth operating facility conducted a site investigation and
found gasoline in the groundwater near his underground storage tank
systems. He withdrew his appeal of the compliance order, voluntarily
removed all fuel from his underground gasoline storage tank systems, and
sent a summary report of the investigation to PADER. In August, this owner
notified PADER of his intent to conduct and complete a site
characterization by the end of September 1994.

An August court order required the Tranguch facility owner to take interim
remedial actions to recover leaked fuel, monitor and mitigate vapors in
area homes, and complete and report on site characterization by
mid-October 1994.

PADER obtained funds from the state Leaking Underground Storage Tank fund
to conduct an extensive characterization study of the impacted area.

o February 1995 - September 1995: The owner had all six tanks at the
Tranguch facility removed. Tanks were found to be perforated with
"fist-sized holes," and gasoline contamination was evident at the site.
The owner of the Tranguch facility declared bankruptcy. In addition, in
May, the owner of the fourth operating facility had two tanks removed.
Both tanks were found to be deeply pitted and the fill end of one tank had
pinholes and corrosion that extended through the steel. Evidence of
gasoline contamination was observed during the excavation.

o November 1995 - December 1995: The Pennsylvania Department of
Environmental Protection (PADEP, formerly PADER) issued another compliance
order to the owner of the Tranguch facility, who again appealed it.

o March - July 1996: PADEP's environmental consultant issued a report on,
among other things, the sources of contamination in the area. While the
consultant found contamination at all 5 of the sites, it found that the
Tranguch facility was primarily responsible for the leak impacting the
residential area. According to the consultant's report, the contamination
at the other four facilities did not significantly contribute to the
contamination plume that was affecting the area residences.

Although PADEP continued to monitor conditions in the area, it no longer
had the funds necessary to mitigate the contamination threat. Therefore,
PADEP asked EPA to lead the remediation of the Tranguch site and the
impacted area. EPA took over as lead agency for the site, while PADEP
continued to work with EPA by providing technical support.

EPA entered into an Interagency Agreement with the U.S. Army Corps of
Engineers (USACE) for contracting services for the site. The first phase
of a two-phase remedial action plan developed for the site, included soil
vapor extraction of the source area and passive oil skimmers for
collection of petroleum products.

o August - October 1996: EPA confirmed PADEP's findings regarding the
contamination plume. The EPA On-Scene Coordinator (OSC) determined that
(1) gasoline contamination at the Tranguch site impacted surface waterways
as well as groundwater at the site, (2) site conditions met or exceeded
removal criteria described in the National Oil and Hazardous Substances
Pollution Contingency Plan (NCP), and (3) the site posed an imminent and
substantial threat to the public health of residents in the area of the
plume because of the threat of fire, explosion, and direct inhalation of
benzene. The OSC estimated that over 900,000 gallons of gasoline leaked
from the tanks. EPA requested remediation funds under the Oil Pollution
Act of 1990 and received an initial $180,000 to begin removal actions,
including installation and maintenance of two underflow dams on Black
Creek to reduce the gasoline contamination which was entering the creek.5

EPA sampled and tested air quality in 53 of 362 homes in the area. The
gasoline vapors detected in 52 of the 53 homes were below EPA's benzene
action level of 21.5 ug/m3 (micrograms per cubic meter). (The state action
level at that time was 32 ug/m3). The 53rd home while above 21.5 ug/m3 was
below 32 ug/m3 . In addition, not all of the 52 homes had elevated benzene
levels. Furthermore, in the residence where benzene level was detected
above EPA's action level, EPA's contractor had observed open cans of
paint, stains, and glue to which they attributed these elevated levels .
On this basis, EPA determined that remediation of these homes was not
needed.

In addition, EPA sent the owners of several facilities identified as
"potentially responsible parties"-including the Tranguch facility-a "Legal
Notice to Suspected Discharger" requesting remedial action.

o January - February 1997: A passive basement air filtration system in a
home was changed to an active system because of a health threat.

o October 1997: EPA received approval to discharge treated groundwater
from the Tranguch site remediation plant into the Hazelton sewer system.

o November - December 1999: USACE constructed a soil vapor extraction
system on the Tranguch site.

EPA identified the Tranguch facility and three other businesses as parties
responsible for the contamination based on a USACE groundwater flow model
developed to predict the flow of spilled gasoline. According to the model,
while much of the spilled material would have come from the Tranguch
facility, some material from three other operating facilities would have
mingled with the plume from the Tranguch leak and subsequently be
transported to Black Creek. Furthermore, for part of the year, spilled
material from one of the operating facilities would flow directly to the
creek. However, EPA's Region 3 General Counsel recommended that the agency
not issue removal enforcement orders to these parties because it
considered them to be "de minimis" (small volume) contributors to the
contamination. In addition, since the Tranguch facility was in bankruptcy,
EPA believed that Tranguch would not be able to comply with the order and,
therefore, did not issue one. Consequently, Oil Pollution Act funds
continued to be used to clean up the site.

o February 2000 - September 2000: The Pennsylvania Department of Health,
concerned over air sample results, asked EPA to install ventilation and
continue testing air quality in 9 homes. EPA began installing sewer vents
at homes.

EPA, USACE, state agencies and a state Representative held public meetings
with area residents in July and August to discuss what corrective actions
EPA would take at the Tranguch facility. An additional 72 residents (32 in
July and 40 in August) requested sampling of the air in their homes.

o October 2000: A public meeting was held to discuss site sampling, health
issues, and site history for the Tranguch site and affected area.
Officials from the (1) Agency for Toxic Substances and Disease Registry
(ATSDR), (2) PADEP, (3) Pennsylvania Department of Health, (4) U.S. Coast
Guard, (5) USACE, and (6) EPA were present, among others. More than 150
Tranguch area residents were also present. The USACE official stated that
the site was complex because of the existence of an underground coal mine.
During the meeting, residents voiced concerns about health and property
values, ATSDR stated that long-term exposure to benzene (a gasoline
component) had been connected to cases of leukemia, and officials agreed
to look into the impacts of the contamination on property values in the
area and the EPA representative noted that potential buyers of homes in
the immediate area had to be informed about the contamination. Also, based
on leak data from Tranguch's tank tightness tests, he stated that 50,000
gallons or less of gasoline had leaked at the Tranguch site rather than
the 900,000 gallons originally estimated.

o December 2000: The EPA Office of Inspector General received a hotline
complaint alleging EPA mismanagement of the Tranguch site cleanup.

o January 2001: Throughout Pennsylvania, the action level for benzene is
32 ug/m3. However, according to EPA Region 3 officials, citing concerns
over the limited information on the extent of the contamination, PADOH set
a more conservative level for the Tranguch site of "non-detect". In
response, EPA identified 8.3 ug/m3 as a reliable detection limit for
benzene. After consultation with ATSDR, PADOH accepted this site-specific
benzene level as being protective of human health.

In an interview with a local newspaper, a GAG representative said that she
hoped homeowners living within the affected area would be "bought out". A
city councilman proposed several resolutions for area residents, including
a resolution to reduce temporarily the real estate assessments of affected
homes to zero, relieving the affected homeowners from property tax
payments.

o February 2001: State and local elected officials held a public meeting
to update residents affected by the Tranguch leak. According to a
newspaper article, a local councilman told those in attendance that they
had no alternative but to leave the area and that they should force
federal officials to buy their properties and move them out of the area.

The Hazle Township Supervisors commissioned a University of Pittsburgh
health study using Township funds.

EPA continued to sample air in area homes and, as of February 2001, had
sampled 308 of approximately 350 homes within the contaminated area.

Luzerne County Commissioners adopted a resolution urging the Board of
Assessment Appeals to eliminate property taxes for two years for
properties determined by the federal government and appropriate agencies
to be eligible for relief.

According to a local newspaper article, one of the state's U.S. senators
wrote to the EPA Administrator and asked the agency to buy the homes of
the area's residents and the other U.S. Senator agreed to meet with the
Administrator about the spill.

o March 2001: EPA implemented weekly "Unified Command" meetings to keep
all interested parties up to date and to provide EPA an opportunity to
address issues and/or questions any of the attending representatives might
have. While the primary members of these meetings were federal, state, and
local officials, representatives from the Group Against Gas participated
in the meetings as ex officio members.

EPA requested and received an additional $11,500,000 in funding from the
U.S. Coast Guard to install groundwater collection and soil gas extraction
systems, as well as groundwater and soil gas treatment systems. This
funding brought the Tranguch site cleanup ceiling to $25,698,188.

The mayor of Hazelton, Pennsylvania declared the area affected by the
Tranguch leak a local disaster emergency and Luzerne County Commissioners
declared the area in a state of emergency. In addition, the Pennsylvania
Emergency Management Agency was asked to determine if the area met
criteria for being declared a disaster area.

o April 2001: On behalf of EPA, the USACE completed a remediation plan for
the area sewer system and began work on a groundwater collection system, a
soil vapor extraction/biovent system, and new storm and sanitary sewer
lines.

A public meeting was held with affected Tranguch site residents. Officials
from EPA, ATSDR, Pennsylvania Department of Health, and PADEP attended the
meeting.

A state elected official introduced a resolution in the state House of
Representatives to declare the affected area a national disaster area and
to purchase the homes of citizens within the affected area. The resolution
was unanimously approved.

In a letter to the Governor of Pennsylvania, the Pennsylvania Emergency
Management Agency (PEMA) stated that the impacted area did not meet the
eligibility criteria to obtain disaster assistance from the Federal
Emergency Management Agency (FEMA). PEMA recommended that the Governor not
certify that a major disaster or emergency existed in order to request
assistance from FEMA. Further, PEMA determined that the hazard-mitigation
funding needed to "buy-out" (purchase the homes of) affected residents
would be inadequate, based on eligible costs, to address this situation.
Based on PEMA's review, the governor denied a state buy-out for residents.

o May 2001: The local board of supervisors again declared a state of
emergency for the area affected by the leak and granted property tax
relief for the affected property owners.

PADEP approved EPA's permit to discharge treated contaminated groundwater
into Black Creek. However, some residents questioned EPA's remediation
strategy. EPA and city, township, and school district attorneys met
concerning Tranguch site cleanup.

The Pennsylvania Department of Health began conducting a health study of
area residents impacted by the leak.

o June 2001: An environmental consulting company retained by EPA completed
a report on subsurface airflow modeling for the soil vapor
extraction/biovent system.

EPA saw to the installation of 288 sewer vents in area homes and allowed
residents to use a suite and hotel amenities, such as the pool, at a local
lodge to get away from the site construction noise.

o August 2001: EPA's Inspector General reported that (1) EPA managed the
cleanup of contamination from the Tranguch leak adequately, but the agency
could have better communicated with the local community and the
Pennsylvania Department of Health; (2) a federal buyout was not warranted
and that residents' desire for a buy-out was based on an inaccurate
perception of the threat posed by the leak; and (3) about $2.8 million in
remediation costs might not have been warranted.

The University of Pittsburgh Graduate School of Public Health completed a
"Preliminary Findings" report that examined whether Hazle Township
residents were at increased risk for cancer compared to that of Luzerne
County residents and residents of Pennsylvania as a whole.6 The authors of
the study stated that "these findings suggest that the incidence of
leukemia and prostate cancer in the Hazle Township is increased compared
to Luzerne County and the state of Pennsylvania". While prostate cancer
has been linked to such factors as age, race, family history and high
intake of dietary fat, research literature has linked leukemia-in
particular, acute myelogenous leukemia-to benzene exposure. However, the
study authors could not definitively identify the gasoline leak as the
source of the excess leukemia.

o December 2001: EPA completed replacement/repair of the sewer lines in
and around the plume of contamination. PADOH issued its health study
report, which provided its recommendations for determining when indoor air
monitoring would no longer be necessary. In effect, it reset the
site-specific action level for benzene of 8.3 ug/m3 back to the statewide
action level of 32ug/m3.

o January 2002: After EPA repaired the sewer lines, installed the vapor
recovery system, and sampled indoor air in area homes, the agency
determined that the air filters in homes were no longer needed. EPA
transferred ownership of the carbon air filtration units to PADEP. PADEP
provided these filters, plus additional ones, to residents requesting
them, as well as electricity to run the filters, at no cost for one year.

o June 2002: EPA held a public meeting regarding the Tranguch site
cleanup. Also attending the meeting were representatives from a U. S.
Congressman`s office and PADEP. The focus of the meeting was to answer
questions about the property reports EPA had mailed or hand-delivered to
affected residents. However, EPA also answered questioned on plans to
restore streets, ongoing health studies, and other investigations of the
spill and its potential impacts. In response to a question on how long
residents with carbon filters in their homes to purify their indoor air
should run them, the EPA representative said "...based on the sampling
that we've done throughout the community, there's no reason to run those
filters". Following the meeting, according to a newspaper account, the EPA
representative said that as long as environmental officials are capable of
eliminating potential chemical exposure for residents, federal officials
would not consider options to relocate affected area residents.

The University of Pittsburgh completed data collection efforts for its
health study, which was funded by a $100,000 grant from the Pennsylvania
Department of Community and Economic Development.

o August 2002: The Luzerne County Board of Commissioners unanimously
approved a resolution requesting the Luzerne County Board of Assessment
Appeals to approve requests to reduce to zero value, for January 1, 2003
to December 31, 2003, the real estate assessments of those properties that
were adversely affected by the Tranguch gasoline leak, as determined by
the federal government.

o September 2002: EPA requested and received an additional $600,000 in
funding from the U.S. Coast Guard. This funding brought the Tranguch site
cleanup cost ceiling to $26,298,188.

o November 2002: A PADEP report on an evaluation of the abandoned mine
under the area concluded that it had no significant environmental impact
on the community.

o January 2003: Under contract with EPA, USACE found small amounts of
petroleum contamination in one tunnel of the abandoned coal mine.

PADEP discontinued support for the air filters in residences.

The local school district sent a letter to EPA asking for $44,000 in
compensation for the economic loss resulting from not having use of the
athletic field.

The University of Pittsburgh Graduate School of Public Health staff
presented their preliminary findings of the Hazleton Health Effects study
to the mayor of the city and the community at a public meeting. According
to a local newspaper article, this study (1) included more individuals
(451 compared to 207) and more households (190 compared to 84) than in the
earlier "Preliminary Findings" study; and (2) found no statistically
significant increase in overall cancer or leukemia incidence for the
Laurel Gardens community of Hazleton residents in the area of the Tranguch
site compared to the county and state populations. The study team,
however, did stress that further investigation was warranted for both
thyroid and brain cancer, according to the newspaper account.

o February 2003 - March 2003: The local Board of Supervisors extended the
state of emergency for the area impacted by the leak through March 10,
2003, and supported a "buyout" of affected homes by the federal
government. The Luzerne County Board of Commissioners also declared that a
state of emergency continued to exist.

GAG wrote to the governor asking that PEMA reevaluate the designation of
the area as a disaster area, which had been denied earlier.

o April 2003: Local and national elected officials representing the area
sent letters to the new governor, requesting a reevaluation of the
previous governor's determination on the area's eligibility for being
declared a disaster area.

The Hazelton city council gave a property tax break to area residents
impacted by the leak for the third consecutive year.

According to EPA officials, the local school district verbally requested
that an athletic field in the affected area be restored. In addition, the
local school district sent a letter to EPA requesting a meeting in May
with the EPA on-site coordinator and USACE to discuss (1) compensation for
loss of use of the athletic field, and (2) a lack of communication between
EPA and the school district over the issue.

o May 2003: The federal government agreed to pay the local school district
$120,000 for the restoration of the athletic field.

EPA held a public meeting with attendees expressing concerns about the
cleanup.

o July 2003 - September 2003: Citing health and property concerns, more
than 250 Hazleton and Hazle Township residents petitioned for a
congressional hearing into EPA's response to and management of the leak
impacting their community.

o October 2003: PADOH completed a study showing that, of the twenty-two
types of cancers and total cancers considered, only the incidence of
leukemia and all cancers was significantly higher in the affected
community than would be expected. However, according to the PADOH study,
the relationship of leukemia incidence to the environment was unclear,
only in rare circumstances can an occurrence be causally linked to a
specific agent with certainty, and the mechanisms for the induction of
cancer from benzene exposure are not clear.

The University of Pittsburgh Graduate School of Public Health completed a
health study providing its final "Summary of Primary Findings" on area
residents' increased risk of developing cancer. This report summarized the
university's two previous studies that separately examined residents in
the affected area of Hazle Township and the City of Hazelton. In addition,
the study examined the total population of affected area residents by
combining the Township and the City. The study also examined affected
residents living in both locations classified into three potential
exposure categories (high, medium, and low) based on the proximity of
their residence to the underground gasoline plume. Most notably was the
high exposure category of those living directly over or adjacent to the
projected contamination plume. The study investigators concluded that
while the combined population did not experience an excess of all cancers,
a statistically significant excess of leukemia was observed. For the high
exposure category, the study investigators concluded that the observed
versus expected cases of leukemia was statistically significant, but that,
because of the small size in this subgroup, these results should be
interpreted with some caution. The study investigators made a number of
recommendations including long-term systematic surveillance and screening
for members of the potentially higher risk population.

The University of Pittsburgh Graduate School of Public Health staff
completed their final report of the Hazleton Health Effects Study
1990-2000. The study's findings suggest that from 1990 to 2000 no
statistically significant increase in overall cancer or leukemia incidence
in the affected area of Hazleton with the exception of brain cancer in
white males compared to the county and state populations for the period
1990 through 2000.

o December 2003: According to a local newspaper article, Laurel Garden
residents impacted by the gasoline leak asked the Luzerne County District
Attorney to investigate the case. In the request letter, the residents
said that they believe they were "needlessly and recklessly endangered" by
the owners of fuel stations in the vicinity of the impacted area, EPA, and
PADEP.

o March 2004: In a letter to local officials, EPA stated that no further
action would be taken to address contamination from the abandoned mine
based on three factors: (1) the contamination did not appear to be from
the Tranguch property, (2) the contamination did not have a pathway to
surface waters, and (3) due to the small amount of contamination present,
the vapors were not migrating from the mine location and therefore did not
threaten nearby residents.

o July 2004: EPA remained the lead agency responsible for the Tranguch
site, while PADEP agreed to provide operation and maintenance services for
the groundwater and soil vapor extraction treatment systems.

o Status as of September 2005: All leaking underground storage tanks had
been removed from the Tranguch site and only residual contamination
required remediation. In addition, cleanup efforts at all affected
residential homes had been completed and well over 95 percent of total
costs to clean up and monitor the site had been expended. Remediation
activities to remove residual contamination are expected to continue for
another 4 to 5 years, costing about $100,000 per year. The remediation
system will be shut down periodically to monitor its effectiveness and
determine whether mitigation goals for groundwater and soil contamination
have been reached. This monitoring is expected to cost about $30,000 to
$40,000 per year. The remediation system might need to be shut down a few
times before the contamination threat can be considered mitigated and the
removal project completed. However, once this determination is made,
groundwater and soil gas (vapor) monitoring will permanently stop, all
remaining monitoring wells (currently approximately 80 but likely will be
less because some are expected to close every year depending on sampling
results) will be closed (costing about $1,000 per well), the treatment
system removed, and the underground piping abandoned in place.

Summary of Key Information From the Case Files

o Contaminants and compounds of concern: Benzene, toluene, ethyl benzene,
and xylenes (BTEX). According to EPA officials, some methyl-tertiary
butyl-ether (MTBE) was identified but was never considered a contaminant
of concern.

o Size of leak: An estimated 25,000 to 50,000 gallons of gasoline was
released into the soil.

o Impacts of contamination: The leaking gasoline contaminated soil and
groundwater, entered into the sewer system through cracked pipes, and
further spread generally northeastward through the adjoining community to
encompass an area of about 70-acres, including 11 businesses, two doctor's
offices, two churches, two parks, 26 vacant lots, and 359 residential
properties.

o Remediation cost: According to EPA officials, about $25.2 million of Oil
Spill Liability Trust Fund monies have been spent to date to clean up the
contamination resulting from the Tranguch leak. In addition, according to
state officials, Pennsylvania spent about $2 to $3 million in cleanup
funds. The site owner spent an unknown additional-but relatively small
amount-on cleanup.

o U.S. Environmental Protection Agency involvement: EPA assumed
responsibility for cleaning up the site at the request of PADER in 1996.

o Communication between responsible agencies and the public: Pennsylvania
state agencies and EPA either held or participated in at least nine public
meetings and other forums regarding the Tranguch leak from 1993-when the
leak was first confirmed-through 2003. In December 1993 and May 1994,
PADER held public meetings with area residents, city and township
representatives, and state legislators regarding the contamination.
According to EPA officials, beginning in July 1996, EPA held meetings with
local officials and public meetings with area residents and others to
discuss plans for remediating the contamination at and around the Tranguch
site and to update the status of the cleanup. Through May 2003, EPA held
at least five such meetings and participated in at least one meeting
sponsored by GAG. The meetings often included representatives from the
state and other federal agencies-such as USACE-involved in cleanup
operations, among others.

o Litigation: Nearby residents affected by the contamination sued numerous
parties, including the owners of the gas stations in the vicinity of the
leak as well as certain oil companies, asserting that the contamination
had caused personal injury and property damage, among other things. These
lawsuits are still pending.

Objectives, Scope, and Methodology Appendix II

The objectives of this review were to identify (1) information available
on the number and cleanup status of leaking underground storage tanks, (2)
existing sources of funding for cleanups at contaminated tank sites, and
(3) processes used to identify, assess, and clean up sites in 5 states
with large numbers of leaking tanks-California, Maryland, Michigan, North
Carolina, and Pennsylvania. In addition, to provide some perspective on
how leaking underground storage tank sites are identified and cleaned up,
we are providing information on the history and cleanup status of one
leaking tank site in each of these 5 states.

To identify information available on the number and cleanup status of
leaking underground storage tanks, we reviewed and evaluated data from
EPA's underground storage tank program semi-annual activity reports for
the period from March 31, 2001 through March 31, 2004. Each activity
report contains data on the number of active and closed tanks, confirmed
releases, cleanups initiated and completed, and cleanup backlog for the 50
states, 5 territories, and the District of Columbia, arranged by EPA
region.1 To assess the reliability of the EPA data, we interviewed EPA
program officials at headquarters and in Regions 3, 4, 5, and 9; conducted
electronic (logic and other) reliability testing of the data itself; and
obtained and reviewed EPA's responses to questions designed to determine
the reliability of the data. In addition, we compared the semiannual data
reported by EPA with data provided by the states we visited. In general,
we found only minor discrepancies during our reliability testing of the
data. For example, the Maine and Massachusetts semi-annual data for the
period ending March 31, 2002, were inadvertently switched. Once we brought
this to EPA's attention, it was immediately corrected. We also found that
EPA reported 1,000 less closed tanks and 1,000 more active tanks than
reported by the state of Pennsylvania. The state had initially provided
EPA data on the number of closed tanks as of September 30, 2004 but the
following day provided EPA a correction to this amount. However, EPA
inadvertently did not update its records to reflect this correction. While
acknowledging these problems, we have determined that the reliability of
the semi-annual data is adequate for the purposes used in this report.

To identify funding options available for cleaning up contaminated tank
sites and the processes used for identifying, assessing, and cleaning up
leaking tanks by 5 states with large numbers of leaking tanks, we
discussed possible funding options with EPA officials, conducted
structured interviews with state program officials, and reviewed documents
provided by these officials or located on their internet sites. To select
which states to include in our review, we used data from EPA's semi-annual
activity report and applied it to our selection criteria and picked the
five states having the highest combined score. Our selection criteria
consisted of the following five quantitative indicators:

o states with the largest average number of active tanks for the last 3
years;

o states with the largest average number of confirmed releases for the
last 3 years;

o states with the largest average number of backlogs for the last 3 years;

o states with the largest increase in backlogs during the last year
divided by the number of active tanks for that state; and

o states with the largest increase in new tank releases for the last year
divided by the number of active tanks for that state.

Three of the five indicators used 3-year average data to minimize the
impact of single-year fluctuations and to reduce the effect of state
officials' periodic revisions to the data.  Two of the five indicators
included adjustments for the number of active tanks in the state so as to
reduce the possible selection bias in favor of states that have large
numbers of active tanks.

For each indicator, we assigned a numerical score corresponding to its
ranking compared to the other states. For example, because EPA data
indicated that California averaged the most releases over the last three
years, we assigned it a score of 56 out of a possible 56 points;
alternatively, because American Samoa averaged the least number of
releases over the last three years, we assigned it a score of one.2 To
determine a state's total ranking for all five indicators, we added the
scores for each state across all indicators and ranked each from highest
to lowest.

As a final step in our state selection process, we reviewed the 5 states
with the highest quantitative score to ensure that, taken as whole, these
states (1) were geographically diverse, (2) had different EPA regional
offices overseeing their LUST programs, (3) included states with and
without EPA-approved LUST programs, and (4) included states with and
without LUST assurance funds. This process resulted in our including
California rather than Ohio. While both states had the same quantitative
score, including California, in our opinion, increased geographic
diversity and added a different regional office to our selected states.

To provide information on the history and cleanup status of one leaking
underground storage tank site in each of the 5 states, we identified the 5
sites-Coca-Cola Enterprises, Yuba County, California; Henry Fruhling Food
Store, Harford County, Maryland; Bob's Marathon, Grand Ledge, Michigan;
R.C. Anderson Trust, Nash County, North Carolina; and Tranguch Tire
Service, Incorporated, Luzerne County, Pennsylvania-interviewed state
program officials responsible for overseeing or managing the cleanup at
these sites, reviewed case files, and visited each site. In identifying
these sites, we first selected the Tranguch site because of congressional
interest. For comparison with Tranguch, we selected the remaining 4 sites
primarily on the basis of the following similarities to that site:3

o Cleanup was either completed in 2004 or is relatively close to
completion based on total estimated costs;

o Remediation costs significantly exceeded the average cost of cleanup of
about $125,000; and

o The risk priority ranking for remediation was above normal.

Specifically, we selected each of the 4 remaining sites as follows:

o Coca-Cola Enterprises n Yuba County, California: From California's
GeoTracker database, we obtained a list of about 3,400 closed sites-that
is, sites where cleanup has been completed-and identified the 10 most
expensive cleanup sites for further investigation. The Coca-Cola site had
the highest risk rating among those sites closed since January 2004, and
while all of the remaining 9 sites were more costly than the Coca-Cola
site, the cost differential was not significant and all had the same or
lower risk ratings.

o Henry Fruhling Food Store in Harford County, Maryland: For site
selection, we obtained a list of 39 sites from Maryland's Oil Control
Program, administrating the state's underground storage tank program.
Because Maryland does not track cleanup costs for sites being remediated
by responsible parties, these were all sites with state-lead cleanups. In
addition, they were sites still undergoing remediation, because the state
has not closed a site within the last 2 years. From this list we selected
the Fruhling Food Store site primarily based on its high cost and high
risk. This site was the most costly to clean up and had a high risk
ranking, having impacted residential water supplies.

o Bob's Marathon in Grand Ledge, Michigan: We obtained a list of 68 sites
with remediation costs of at least $250,000 from Michigan's Department of
Environmental Quality, administrating the state's underground storage tank
program. Because Michigan does not track cleanup costs for sites being
remediated by responsible parties, these were all sites with state-lead
cleanups. We selected the Bob's Marathon site primarily based on its high
cost and high risk. Bob's Marathon was the most costly of sites that were
either completed or nearly completed and was high risk because it impacted
municipal water supplies and the community.

o R.C. Anderson Trust in Nash County, North Carolina: We obtained a list
of 108 sites with remediation costs of at least $250,000 from North
Carolina's Department of Environment and Natural Resources, administrating
the state's underground storage tank program. We selected the 4 sites with
the highest remediation costs for further analysis. From these 4, we
selected the R.C. Anderson Trust site primarily based on high risk and
high cost. This site was the only site ranked high risk because of its
potential impact on nearby water wells and on the community. While this
site was the least costly of the four high-cost sites, the cost
differential between this site and the highest cost one was only about
$24,000.

We conducted our review from August 2004 through November 2005 in
accordance with generally accepted government auditing standards.

Comments from the Environmental Protection Agency Appendix III

GAO Contacts and Staff Acknowledgments Appendix IV

John B. Stephenson (202) 512-3841

In addition, Vincent P. Price, Michael J. Rahl, and Michael S. Sagalow
made key contributions to this report. Important contributions were also
made by John W. Delicath, Wilfred B. Holloway, and Richard P. Johnson.

(360483)

www.gao.gov/cgi-bin/getrpt?GAO-06-45.

To view the full product, including the scope

and methodology, click on the link above.

For more information, contact John Stephenson at (202) 512-3841 or
[email protected].

Highlights of GAO-06-45, a report to congressional requesters

November 2005

ENVIRONMENTAL PROTECTION

More Complete Data and Continued Emphasis on Leak Prevention Could Improve
EPA's Underground Storage Tank Program

Leaking underground storage tanks that contain hazardous products,
primarily gasoline, can contaminate soil and groundwater. To address this
problem, the Environmental Protection Agency (EPA), under its Underground
Storage Tank (UST) Program, required tank owners to install leak detection
equipment and take measures to prevent leaks. In 1986, the Congress
created a federal trust fund to assist states with cleanups. Cleanup
progress has been made, but, as of early 2005, cleanup efforts had not yet
begun for over 32,000 tanks, many of which may require state and/or
federal resources to address.

GAO identified (1) data on the number and cleanup status of leaking tanks,
(2) funding sources for tank cleanups, and (3) processes used by five
states with large numbers of leaking tanks-California, Maryland, Michigan,
North Carolina, and Pennsylvania -to identify, assess, and clean up sites.

What GAO Recommends

GAO recommends that EPA require states to report to the agency information
on all known abandoned tanks. EPA agreed that the UST program could
benefit from more specific data on abandoned tanks, but had concerns about
the potential burden on states. GAO clarified its recommendation to
indicate that EPA should obtain data that states currently compile.

Data submitted to EPA by the states show that, as of March 31, 2005, more
than 660,000 tanks were in use and about 1.6 million were no longer in
use. In addition, states identified about 449,000 tank releases (leaks)
and about 416,000 initiated cleanups, with almost 324,000 of those
cleanups completed. States also compile limited data on abandoned
tanks-tanks whose owners are unknown, or unwilling or unable to pay for
their cleanup-but EPA does not require states to provide separate data on
all of their known abandoned tanks. Without this separate data, EPA cannot
effectively determine the number and cleanup status of these tanks, or how
to most efficiently and effectively allocate federal cleanup funds to the
states.

Tank owners and operators are primarily responsible for paying to clean up
their own sites, but abandoned tanks are cleaned up using state resources,
that may be limited, and federal trust funds. EPA estimates that the
average remediation costs per site have been about $125,000, but costs
sometimes have exceeded $1 million. Officials from two of the five states
we contacted reported that their state funds may be inadequate to address
contamination at abandoned tank sites. In this regard, Michigan and North
Carolina officials told GAO that, because of resource constraints, they
let contamination at abandoned tank sites attenuate (diminish) naturally
once immediate threats are addressed. Furthermore, due to limited
resources, states must sometimes find other options for cleaning up sites.
For example, Pennsylvania officials asked EPA to take over the cleanup
work at the abandoned Tranguch site in 1996 because the owner was bankrupt
and the state could not pay the expected cleanup costs.

The five states that GAO contacted identify, assess, and clean up leaking
tank sites using similar processes. Generally, owners and operators are
responsible for conducting these activities under state oversight. Leaking
tanks are identified when tank owners report leaks; land redevelopment
activities uncover unknown tanks; or state agencies investigate
contamination complaints or inspect tanks for regulatory compliance. While
regular tank inspections can detect new leaks and potentially prevent
future ones, as of early 2005, only two of the five states GAO
contacted-California and Maryland-consistently inspected all the state's
tanks at least once every 3 years, the minimum rate of inspection that EPA
considers adequate. The Energy Policy Act, enacted in August 2005, among
other things, requires inspections at least once every 3 years and
provides federal trust funds for this and other leak prevention purposes.
EPA and some state officials told GAO that increasing inspection frequency
could require additional resources. Being able to use trust fund
allocations for this purpose will help in this regard. The five states GAO
contacted, once they become aware of leaking tanks, identify responsible
parties and require them to hire consultants to conduct site assessments
and plan and implement cleanup work. The states generally prioritize sites
for cleanup according to the immediate threat they pose to human health,
safety, and/or the environment.
*** End of document. ***