This is the accessible text file for GAO report number GAO-10-277 
entitled 'Superfund: Information on Cost and Other Issues Related to 
the Cleanup of the Federal Creosote Site' which was released on March 
29, 2010. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

Report to Congressional Requesters: 

United States Government Accountability Office: 
GAO: 

February 2010: 

Superfund: 

Information on Cost and Other Issues Related to the Cleanup of the 
Federal Creosote Site: 

GAO-10-277: 

GAO Highlights: 

Highlights of GAO-10-277, a report to congressional requesters. 

Why GAO Did This Study: 

In the 1990s, creosote was discovered under a residential neighborhood 
in Manville, New Jersey. Creosote, a mixture of chemicals, is used to 
preserve wood products, such as railroad ties. Some of the chemicals 
in creosote may cause cancer, according to the Environmental 
Protection Agency (EPA). EPA found that creosote from a former wood-
treatment facility (known as the Federal Creosote site) had 
contaminated soil and groundwater at the site. Under the Superfund 
program—the federal government’s principal program to clean up 
hazardous waste—EPA assessed site risks, selected remedies, and worked 
with the U.S. Army Corps of Engineers to clean up the site. As of May 
2009, construction of EPA’s remedies for the site had been completed; 
however, total site costs were almost $340 million and remedial 
construction costs had exceeded original estimates. 

In this context, GAO was asked to examine (1) how EPA assessed risks 
and selected remedies for the site, and what priority EPA gave to site 
cleanup; (2) what factors contributed to the difference between the 
estimated and actual costs; and (3) how EPA and the Corps divided 
responsibilities for site work. GAO analyzed EPA and Corps documents 
and data on the cleanup effort and its costs, and interviewed 
officials from these agencies. This report contains no 
recommendations. EPA generally agreed with GAO’s findings on the agency’
s cleanup costs and actions, while the U.S. Army Corps of Engineers 
had no comments. 

What GAO Found: 

The extent of the contamination in a residential area at the Federal 
Creosote site was the primary factor influencing EPA’s risk assessment 
conclusions, remedy selection decisions, and how EPA prioritized site 
work, according to site documents and agency officials. EPA assessed 
site contamination through multiple rounds of evaluation and concluded 
that soil and groundwater contamination levels were high enough that 
EPA needed to take action. Then, EPA evaluated remedies to achieve 
cleanup goals that it had established for the site and that were 
consistent with its residential use. EPA selected off-site treatment 
and disposal of the contaminated soil and long-term monitoring of the 
groundwater contamination as the remedies for the site. In selecting 
these remedies, EPA considered a range of alternatives but ultimately 
determined that certain options would be potentially infeasible or 
ineffective due to the residential setting. For example, EPA chose not 
to implement certain alternatives on-site because the agency found 
that there was insufficient space and they would be too disruptive to 
nearby residents. In addition, EPA chose not to implement certain 
alternatives because the agency found that they would be unlikely to 
achieve the cleanup goals for the site, especially considering the 
high level of treatment required to allow for unrestricted residential 
use of the area and the high levels of contamination found at the 
site. EPA made cleanup of the site a high priority because the 
contamination was in a residential area. For example, EPA took steps 
to shorten the cleanup period and prioritized the use of regional 
Superfund resources on the Federal Creosote site over other sites in 
the region. 

The $338 million in total site costs exceeded EPA’s estimated remedial 
construction costs of $105 million by about $233 million, primarily 
because EPA’s estimates focused only on construction costs, and EPA 
discovered additional contamination during the cleanup effort. EPA 
prepared preliminary cost estimates during the remedy selection 
process; however, EPA requires that these estimates include only the 
costs associated with implementing different remedies it was 
considering, not all site costs. Also, as a result of the movement of 
contamination in the ground and sampling limitations during EPA’s site 
investigation, a greater-than-expected amount of contamination was 
discovered during the cleanup effort, which increased costs. Other 
factors, such as contractor fraud, affected total site costs to a 
lesser extent. 

EPA was responsible for managing the overall site cleanup and 
community relations, while the Corps was responsible for implementing 
the cleanup. EPA dedicated a full-time staff member to manage the site 
cleanup who, according to EPA, maintained a significant on-site 
presence to ensure that the project remained on schedule and was 
adequately funded and to work with residents. EPA also oversaw the 
work of the Corps and its costs. To conduct the actual cleanup work, 
the Corps hired contractors to design or implement cleanup activities 
who, in turn, hired subcontractors for some tasks. The Corps oversaw 
the activities and costs of its primary contractors but, according to 
Corps officials, was less involved in selecting and overseeing 
subcontractors. 

View [hyperlink, http://www.gao.gov/products/GAO-10-277] or key 
components. For more information, contact John B. Stephenson at (202) 
512-3841 or stephensonj@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

EPA's Risk Assessment, Remedy Selection, and Prioritization Decisions 
for the Federal Creosote Site Were Primarily Influenced by the Extent 
of the Contamination in a Residential Area: 

Total Site Costs Exceeded Early Construction Cost Estimates Largely 
because of the Nature of these Estimates and the Discovery of 
Additional Contamination: 

EPA Provided Overall Project Management and Communicated with 
Residents, while the Corps Oversaw Contractor Implementation of the 
Remedy: 

Agency Comments and Our Evaluation: 

Appendix I: Summary of Criminal and Civil Litigation Related to the 
Federal Creosote Site: 

Appendix II: Objectives, Scope, and Methodology: 

Appendix III: Timeline of EPA Actions at the Federal Creosote Site: 

Appendix IV: Comments from the Environmental Protection Agency65: 

Appendix V: GAO Contact and Staff Acknowledgments: 

Tables: 

Table 1: EPA's Remedial Alternative Evaluation Criteria: 

Table 2: Other Response Costs Not Included in Construction Estimates 
for the Federal Creosote Site: 

Table 3: Difference between Estimated Soil Quantities in the RODs and 
Final Soil Quantities at the Federal Creosote Site, by OU: 

Table 4: Estimated and Actual Remedial Construction Costs at the 
Federal Creosote Site: 

Figures: 

Figure 1: Map of the Federal Creosote Site in the Borough of Manville, 
New Jersey: 

Figure 2: EPA's Site Cleanup Process: 

Figure 3: Difference between Estimated Construction and Total Site 
Costs at the Federal Creosote Site: 

Figure 4: Rustic Mall Areas Remediated at the Federal Creosote Site: 

Figure 5: Residential Area Properties Remediated at the Federal 
Creosote Site: 

Figure 6: Actual Remedial Construction Costs as a Portion of the 
Difference between Estimated Construction and Total Site Costs at the 
Federal Creosote Site: 

Figure 7: Contrasting Methodologies and Contractor Fraud as a Portion 
of the Difference between Estimated Construction and Total Site Costs 
at the Federal Creosote Site: 

Figure 8: Key Events Related to Risk Assessment and Remedy Selection 
at the Federal Creosote Site: 

Figure 9: Key Events Related to Remedial Design and Action at the 
Federal Creosote Site: 

Abbreviations: 

ATSDR: Agency for Toxic Substances and Disease Registry: 

BEI: Bennett Environmental, Inc. 

CEFMS: Corps of Engineers Financial Management System: 

CERCLA: Comprehensive Environmental Response, Compensation, and 
Liability Act of 1980: 

EPA: Environmental Protection Agency: 

IDIQ: indefinite-delivery/indefinite-quantity: 

LTTD: low temperature thermal desorption: 

NJDEP: New Jersey Department of Environmental Protection: 

NPL: National Priorities List: 

OU: operable unit: 

PRAC: pre-placed remedial action contract: 

RI/FS: remedial investigation/feasibility study: 

ROD: record of decision: 

RPM: Remedial Project Manager: 

SCORPIOS: Superfund Cost Recovery Package Imaging and On-Line System: 

[End of section] 

United States Government Accountability Office: 
Washington, DC 20548: 

February 25, 2010: 

The Honorable Harry M. Reid: 
Majority Leader: 
United States Senate: 

The Honorable James M. Inhofe: 
Ranking Member: 
Committee on Environment and Public Works: United States Senate: 

In the 1990s, creosote contamination was discovered at the Federal 
Creosote site, a former wood-treatment facility in the Borough of 
Manville, New Jersey, that had been developed into a residential 
community of single-family homes and a retail mall. Creosote--a 
mixture of approximately 300 chemicals--is used to preserve wood, and 
the Environmental Protection Agency (EPA) has identified some of these 
chemicals as probable human carcinogens. Under the Superfund program-- 
the federal government's principal program to clean up hazardous waste 
sites--EPA worked with the U.S. Army Corps of Engineers (the Corps) to 
construct remedies to address site contamination. This work was 
completed in 2008, although some maintenance and groundwater 
monitoring efforts continue. Early construction cost estimates for 
these remedies totaled $105 million.[Footnote 1] However, as of May 
2009, EPA had spent almost $340 million on the cleanup, including $246 
million to construct site remedies and $92 million in other response 
costs that were not part of EPA's early estimates.[Footnote 2] The 
increase in actual remedial construction costs over the agency's 
original estimates raised questions about whether EPA had selected the 
most cost-effective remedy, given site risks and the extent of 
contamination. Furthermore, the Department of Justice (Justice) and 
the state of New Jersey are pursuing civil claims against the site's 
alleged responsible party related to cost recovery and--as part of the 
claims brought by the state of New Jersey--damages;[Footnote 3] 
Justice has also filed criminal charges against several parties 
involved in the cleanup, asserting claims related to alleged fraud in 
awarding subcontracts by contractor personnel.[Footnote 4] 

The Comprehensive Environmental Response, Compensation, and Liability 
Act (CERCLA) of 1980 established the Superfund program to protect 
human health and the environment from the effects of hazardous 
substances. Under this program, EPA has the authority to (1) clean up 
hazardous waste sites and then seek reimbursement from the parties 
responsible for contaminating them or (2) compel the responsible 
parties to clean up these sites. The Superfund cleanup process begins 
with site discovery or notification to EPA of the possible release of 
hazardous substances posing a threat to human health or the 
environment. On the basis of an initial evaluation, EPA may select a 
site for inclusion on the National Priorities List (NPL), which 
catalogs many of the nation's most seriously contaminated sites. For 
sites that EPA lists on the NPL, the agency initiates a more extensive 
investigation process to identify the nature and extent of 
contamination at the site, quantify potential risks to human health 
and the environment, and evaluate potential remedies to address site 
contamination. Selected remedies are then planned in the remedial 
design phase and implemented in the remedial action phase.[Footnote 5] 

EPA's initial investigation at the Federal Creosote site identified 
contamination from the lagoons and canals of the former wood-treatment 
facility beneath the residential portion of the site. After listing 
the site on the NPL, EPA conducted additional investigations to 
quantify site risks; evaluated potential remedies to address 
contamination, including preparing cost estimates; and selected 
remedies for the site. EPA tasked the Corps with designing and 
implementing these remedies. The Corps hired contractors to perform 
the design and construction work, and, in turn, the prime contractor 
for the construction work hired subcontractors to perform certain 
tasks, such as soil transportation, treatment, and disposal; 
landscaping; and wastewater treatment. 

In this context, you asked us to review issues concerning the Federal 
Creosote site. Our objectives were to examine (1) how EPA assessed the 
risks and selected the remedies for the Federal Creosote site, and 
what priority EPA assigned to site cleanup; (2) what factors 
contributed to the difference between the estimated and actual costs 
of cleaning up the site; and (3) how responsibilities for implementing 
and overseeing site work were divided between EPA and the Corps. We 
also summarized information on criminal and civil litigation related 
to the Federal Creosote site (see appendix I). 

To examine how EPA assessed risks and selected remedies for the site 
as well as what priority EPA assigned to the cleanup, we analyzed site 
documents detailing the results of the agency's activities and 
decisions at the site. We also interviewed relevant EPA and other 
federal, state, and local officials. To determine what factors 
contributed to the difference between the estimated and actual costs 
of site cleanup, we obtained and analyzed data on estimated 
construction and total site costs from site documents that detailed 
EPA's planned activities, EPA and Corps cost-tracking databases, and 
contractor cost summary reports. To compare estimated construction and 
total site costs, in accordance with our policy, we adjusted the 
estimated construction costs using a present value analysis and, to 
adjust for inflation, converted all dollar figures into fiscal year 
2009 constant dollars. We also analyzed site documents describing the 
cleanup effort, reviewed EPA cost-estimating guidance, and conducted 
interviews with EPA and Corps officials. To describe how 
responsibilities for implementing and overseeing the site work were 
divided between EPA and the Corps as well as EPA's oversight actions, 
we reviewed EPA interagency agreements with the Corps, site documents, 
and EPA guidance and interviewed EPA and Corps officials. To describe 
actions that the Corps took to implement its site responsibilities, we 
reviewed Corps guidance, Corps correspondence to the contractor, and 
contractor requests for approval of certain subcontracts and also 
interviewed Corps and EPA officials. To identify civil and criminal 
litigation related to the Federal Creosote site, we collected and 
reviewed information from publicly available court documents to 
summarize the cases and, where applicable, their outcome. We evaluated 
the reliability of the site actual cost data used in our analyses by, 
for example, comparing the data across different sources and reviewing 
agency data reliability controls. We determined that these data were 
sufficiently reliable for our purposes. See appendix II for a more 
detailed description of our scope and methodology. 

We conducted our work from May 2008 through February 2010 in 
accordance with all sections of GAO's Quality Assurance Framework that 
are relevant to our objectives. The framework requires that we plan 
and perform the engagement to obtain sufficient and appropriate 
evidence to meet our stated objectives and to discuss any limitations 
in our work. We believe that the information and data obtained, and 
the analyses conducted, provide a reasonable basis for any findings 
and conclusions in this product. 

Background: 

Creosote is derived by distilling tar; the type of creosote most 
commonly used for wood-treating is manufactured from coal tar. 
Polycyclic aromatic hydrocarbons--chemicals formed during the 
incomplete burning of coal, oil, gas, or other organic substances-- 
generally make up 85 percent of the chemical composition of creosote. 
EPA classifies some of the polycyclic aromatic hydrocarbons in 
creosote, such as benzo(a)pyrene, as probable human carcinogens. Some 
polycyclic aromatic hydrocarbons also may have noncarcinogenic health 
effects, such as decreased liver or kidney weight. 

From approximately the early 1910s to the mid-1950s, the Federal 
Creosote site was a wood-treatment facility. Untreated railroad ties 
were delivered to the site and, to preserve them, coal tar creosote 
was applied to the railroad ties at a treatment plant located on the 
western portion of the property (see figure 1 for an illustration of 
the site). Residual creosote from the treatment process was discharged 
into two canals that led to two lagoons on the northern and southern 
parts of the site, respectively. After treatment, the railroad ties 
were moved to the central portion of the property, where excess 
creosote from the treated wood dripped onto the ground. The treatment 
plant ceased operations in the mid-1950s. During the late 1950s and 
early 1960s, the area where the treatment plant was formerly located 
was developed into a 15-acre commercial and retail property known as 
the Rustic Mall. Through the mid-1960s, other areas of the property, 
including the former canal, lagoon, and drip areas, were developed 
into a 35-acre residential neighborhood known as the Claremont 
Development, which was made up of 137 single-family homes that housed 
several hundred residents. 

Figure 1: Map of the Federal Creosote Site in the Borough of Manville, 
New Jersey: 

[Refer to PDF for image: illustration] 

The following are indicated on a map of the Rustic Mall and Claremont 
Development residential area: 

Drip area: 
Former wood-treating facility: 
Canal A: 
Lagoon A: 
Canal B: 
Lagoon B: 

Source: GAO analysis of EPA site documents. 

[End of figure] 

Issues with creosote contamination at the site became apparent in 
April 1996, when the New Jersey Department of Environmental Protection 
(NJDEP) responded to an incident involving the discharge of an unknown 
thick, tarry substance from a sump located at one of the residences in 
the Claremont Development. Later, in January 1997, the Borough of 
Manville responded to complaints that a sinkhole had developed around 
a sewer pipe in the Claremont Development. Excavation of the soil 
around the sewer pipe identified a black, tar-like material in the 
soil. After an initial site investigation, EPA found contamination in 
both the surface and subsurface soils as well as in the groundwater 
beneath the site. In 1999, EPA placed the site on the NPL and divided 
it into three smaller units, called operable units (OU).[Footnote 6] 
OU1 consisted of the source contamination (free-product creosote) in 
the lagoon and canal areas of the Claremont Development. OU2 included 
other soil contamination in the Claremont Development, such as 
residually contaminated soil at properties over and near the lagoon 
and canal areas and the drip area of the former wood-treatment 
facility. OU2 also included contamination at a nearby day-care 
facility. OU3 included the Rustic Mall soil contamination as well as 
groundwater contamination throughout the site.[Footnote 7] 

EPA completed all major site cleanup work in November 2007, and the 
site was declared "construction complete" in March 2008.[Footnote 8] 
Ultimately, EPA performed cleanup activities on 93 of the 137 
properties in the residential area as well as on the commercial 
portion of the site. EPA's ongoing activities at the site include 
monitoring groundwater contamination, conducting 5-year reviews of 
contamination levels to ensure that the remedy remains protective of 
human health and the environment, and selling properties that EPA 
acquired during the remedial action. According to EPA officials, the 
agency could remove the site from the NPL as early as 2011; however, 
this decision will depend on the results of contamination monitoring 
at the site. 

Most Superfund sites progress through the cleanup process in roughly 
the same way, although EPA may take different approaches on the basis 
of site-specific conditions. After listing a site on the NPL, EPA 
initiates a process to assess the extent of the contamination, decides 
on the actions that will be taken to address that contamination, and 
implements those actions. Figure 2 outlines the process EPA typically 
follows, from listing a site on the NPL through deletion from the NPL. 

Figure 2: EPA's Site Cleanup Process: 

[Refer to PDF for image: illustration] 

Milestone: NPL Listing. 

Phase: Site Study: 
* Remedial Investigation; 
* Feasibility Study. 

Milestone: Record of Decision. 

Phase: Remedial Action: 
* Remedial design; 
* Remedial action. 

Milestone: Construction Complete. 

Phase: Operation and Maintenance. 

Milestone: Deletion from the NPL. 

Source: GAO analysis of EPA data. 

[End of figure] 

In the site study phase of the cleanup, EPA or a responsible party 
conducts a two-part remedial investigation/feasibility study (RI/FS) 
process.[Footnote 9] The first part of this process--the remedial 
investigation--consists of data collection efforts to characterize 
site conditions, determine the nature of the waste, assess risks to 
human health and the environment, and conduct treatability testing as 
necessary to evaluate the potential performance and cost of the 
treatment technologies that are being considered.[Footnote 10] During 
the second part of the RI/FS process--the feasibility study--EPA 
identifies and evaluates various options to address the problems 
identified through the remedial investigation. EPA also develops 
cleanup goals, which include qualitative remedial action objectives 
that provide a general description of what the action will accomplish 
(e.g., preventing contamination from reaching groundwater) as well as 
preliminary quantitative remediation goals that describe the level of 
cleanup to be achieved.[Footnote 11] According to EPA guidance, it may 
be necessary to screen out certain options to reduce the number of 
technologies that will be analyzed in detail to minimize the resources 
dedicated to evaluating less promising options. EPA screens 
technologies on the basis of the following three criteria: 

* effectiveness: the potential effectiveness of technologies in 
meeting the cleanup goals, the potential impacts on human health and 
the environment during implementation, and how proven and reliable the 
technology is with respect to the contaminants and conditions at the 
site; 

* implementability: the technical and administrative feasibility of 
the technology, including the evaluation of treatment requirements and 
the relative ease or difficulty in achieving operation and maintenance 
requirements; and: 

* cost: the capital and operation and maintenance costs of a 
technology (i.e., each technology is evaluated to determine whether 
its costs are high, moderate, or low relative to other options within 
the same category). 

After screening the technologies that it has identified, EPA combines 
selected technologies into remedial alternatives. EPA may develop 
alternatives to address a contaminated medium (e.g., groundwater), a 
specific area of the site (e.g., a waste lagoon or contaminated hot 
spot), or the entire site. EPA guidance states that a range of 
alternatives should be developed, varying primarily in the extent to 
which they rely on the long-term management of contamination and 
untreated wastes. In addition, containment options involving little or 
no treatment, as well as a no-action alternative, should be developed. 
EPA then evaluates alternatives using the nine evaluation criteria 
shown in table 1 and documents its selected alternative in a record of 
decision (ROD).[Footnote 12] 

Table 1: EPA's Remedial Alternative Evaluation Criteria: 

Criterion: Overall protection of human health and the environment; 
Description: An alternative's ability to protect human health and the 
environment through engineered systems and institutional controls that 
eliminate, reduce, or control risks.[A] 

Criterion: Compliance with applicable or relevant and appropriate 
requirements; 
Description: An alternative's ability to comply with federal, state, 
and local environmental and health regulations, and other advisory or 
guidance criteria. 

Criterion: Long-term effectiveness and permanence; 
Description: The degree of certainty that an alternative will be 
successful, considering the risks remaining after the remedial action 
and the adequacy and reliability of engineering and institutional 
controls. 

Criterion: Reduction of toxicity, mobility, or volume through 
treatment; 
Description: An alternative's ability to permanently and significantly 
reduce the toxicity, mobility, or volume of hazardous substances 
through treatment. 

Criterion: Short-term effectiveness; 
Description: An alternative's effects on and risks to workers, 
surrounding communities, and the environment during the remedial 
action as well as the time required to achieve the remedial action 
objectives. 

Criterion: Implementability; 
Description: The relative ease or difficulty associated with 
implementing an alternative, considering the technical and 
administrative feasibility of remedial technologies, and the 
availability of required labor, equipment, and materials. 

Criterion: Cost; 
Description: An alternative's capital and operation and maintenance 
costs. 

Criterion: State/Agency acceptance; 
Description: The extent to which the state and other regulatory 
agencies support an alternative. 

Criterion: Community acceptance; 
Description: The extent to which the community supports an alternative. 

Source: GAO analysis of EPA site documents. 

Note: These criteria are not weighted equally in EPA's evaluation. The 
first two are threshold criteria that must be met for an alternative 
to be eligible for selection. The third through the seventh (long-term 
effectiveness through cost) are balancing criteria used to compare 
trade-offs between the alternatives. The last two are modifying 
criteria that are considered after a public comment period on EPA's 
proposed plan. 

[A] Institutional controls, such as deed restrictions on future land 
use, are limited measures that are intended to minimize potential 
human exposure to contaminants. These measures are typically 
implemented along with other technologies because they do not reduce 
contaminant levels or prevent migration. 

[End of table] 

Next, either EPA or a responsible party may initiate the remedial 
action that was documented in the ROD. Like the RI/FS, implementation 
of the remedial action is divided into two phases. The first phase is 
the remedial design, which involves a series of engineering reports, 
documents, and specifications that detail the steps to be taken during 
the remedial action to meet the cleanup goals established for the 
site. For EPA-led remedial actions, EPA may either select a private 
contractor to perform the remedial design or, under a 1984 interagency 
agreement with the Corps, assign responsibility for designing the 
remedial action to the Corps, which may select and oversee a private 
contractor to perform the design work.[Footnote 13] The second phase 
is the remedial action phase, where the selected remedy, as defined by 
the remedial design, is implemented. Similar to the design phase, for 
EPA-led remedial actions, EPA may either select a private contractor 
to perform the remedial action or assign the remedial action to the 
Corps, which would be responsible for contractor selection and 
oversight during the remedial construction. 

When physical construction of all remedial actions is complete and 
other criteria are met, EPA deems the site to be "construction 
complete." Most sites then enter into an operation and maintenance 
phase, when the responsible party or the state maintains the remedy 
while EPA conducts periodic reviews to ensure that the remedy 
continues to protect human health and the environment. For example, at 
a site with soil contamination, the remedial action could be to build 
a cap over the contamination, while the operation and maintenance 
phase would consist of monitoring and maintaining the cap. Eventually, 
when EPA determines, with state concurrence, that no further remedial 
activities at the site are appropriate, EPA may delete the site from 
the NPL. 

EPA's Risk Assessment, Remedy Selection, and Prioritization Decisions 
for the Federal Creosote Site Were Primarily Influenced by the Extent 
of the Contamination in a Residential Area: 

The extent of the contamination in a residential area at the Federal 
Creosote site was the primary factor that influenced EPA's risk 
assessment conclusions, remedy selection decisions, and site work 
priorities. EPA determined that risk levels were unacceptable given 
the site's residential use. EPA then selected remedies for the site, 
taking into account space constraints and other challenges associated 
with a residential cleanup. Finally, EPA placed a high priority on 
scheduling and funding site work because the contaminated area was 
residential, thereby reaching key cleanup milestones relatively 
quickly. 

EPA Assessed Risks and Selected Cleanup Goals on the Basis of the 
Site's Residential Use: 

From the spring of 1997 to the summer of 2001, EPA conducted multiple 
rounds of sampling and risk assessment at the Federal Creosote site 
and concluded that human health risks exceeded acceptable 
levels.[Footnote 14] Specifically, EPA assessed the air, groundwater, 
surface soil, and subsurface soil as part of an initial site 
investigation and an RI/FS process. See appendix III for a timeline of 
EPA's risk assessment activities.[Footnote 15] 

EPA's initial investigation of site contamination, which began in 
1997, included such efforts as assessing whether contamination was 
affecting public drinking water supplies; investigating the nature of 
the bedrock and the aquifer underlying the site; collecting soil 
samples from 30 properties selected on the basis of their proximity to 
the lagoons, canals, and drip area of the former wood-treatment 
facility; and collecting approximately 1,350 surface soil samples (up 
to 3 inches below the ground surface) from 133 properties in and near 
the residential development. From this initial investigation, EPA 
concluded that site contamination posed unacceptable human health 
risks. For example, while EPA found that contamination did not pose 
short-term health risks that could require an evacuation of residents, 
EPA found that the contamination was extensive and uncontrolled; had 
impacted soil, sediment, and groundwater in the area; and likely posed 
long-term health risks. For soil contamination in particular, EPA 
determined that, in some areas, the contamination was within 2 to 3 
feet of the ground surface; in other areas, EPA found that the 
contamination was covered by little or no fill material.[Footnote 16] 
According to a site document, one resident had discovered a large 
amount of buried tar when installing a fence on his property. As a 
result of its concerns that surface soil contamination could pose a 
risk to residents, EPA developed a surface soil risk assessment in 
January 1999. EPA concluded that soil contamination levels at 27 
properties in the residential area posed long-term human health risks, 
including carcinogenic or noncarcinogenic risks (or both), that 
exceeded acceptable levels. In addition to soil contamination, EPA's 
initial investigation determined that creosote had contaminated 
groundwater in the soil as well as in fractures in the bedrock 
underlying the site, which was a potential source of drinking water. 
Furthermore, EPA's aquifer investigation showed that groundwater from 
the site had the potential to influence the Borough of Manville's 
municipal water supply wells, although Region 2 officials said the 
nature of the fractures made it difficult for EPA to determine whether 
site contamination would actually affect the wells. 

According to Region 2 officials, the purpose of a remedial 
investigation is to collect enough data to determine whether there is 
a need to take a remedial action. These officials said that an RI/FS 
for OU1 was not necessary because EPA had obtained much more 
information from its initial investigation on the extent of 
contamination at properties over the lagoon and canal source areas 
than is typically available to support taking an action. Also, 
according to EPA, the data that were collected during this initial 
investigation were equivalent in scope to that of a remedial 
investigation. Therefore, because EPA was trying to address the source 
contamination in the residential area on an expedited basis, the 
agency chose to incorporate these data into an Engineering 
Evaluation/Cost Analysis because it allowed EPA to evaluate remedial 
alternatives in a more streamlined way, as compared with an RI/FS 
report.[Footnote 17] However, for OU2 and OU3, EPA initiated an RI/FS 
process in 1998 to more fully characterize the extent of soil and 
groundwater contamination throughout the site.[Footnote 18] EPA's OU2 
soil evaluation determined that elevated levels of creosote 
contamination close to the surface in the residential area were 
generally found near the lagoons and canals, while the drip area 
generally had residual levels of contamination close to the surface. 
Underlying the site, EPA found that free-product creosote rested on a 
clay layer approximately 6 to 10 feet below the surface, although in 
some areas the layer was not continuous, and the creosote had migrated 
as deep as the bedrock, roughly 25 to 35 feet underground. On the 
basis of these findings, in April 2000, EPA developed a human health 
risk assessment for soil contamination in the residential area using a 
sample of six representative properties: two properties each 
represented the lagoon and canal areas, the drip area, and the 
remaining residential area, respectively.[Footnote 19] EPA found that 
soil contamination exceeded acceptable risk levels at the lagoon and 
canal and drip areas, but not at properties representing other areas 
of the Claremont Development. 

Furthermore, EPA's OU3 soil analysis revealed that contamination was 
generally in three main areas of the mall, with several other "hot 
spots" of contaminated material. EPA also determined that most of the 
soil contamination was within the first 2 feet below the ground 
surface; however, in certain areas, contamination was as deep as 35 
feet below the surface. EPA noted that it did not collect soil samples 
from under the mall buildings, although, according to a site document, 
EPA thought it likely that contamination remained under at least a 
portion of one of the buildings. EPA assessed the human health risks 
from exposure to soil contamination in June 2001. At the time of EPA's 
assessment, OU3 was a commercial area. However, the Borough of 
Manville and the mall owner had indicated that the area could be 
redeveloped for a mixed residential/commercial use. Therefore, EPA 
evaluated risks for OU3 under both residential and commercial use 
scenarios, and found that risks exceeded acceptable levels for 
residential use at some areas of the mall and for commercial use at 
one area.[Footnote 20] 

Finally, EPA's OU3 RI/FS investigation determined that contaminated 
groundwater in the soil above the bedrock had not migrated far from 
the original source areas of the lagoons and canals. However, free-
product creosote had penetrated as deep as 120 feet into the fractured 
bedrock, and groundwater contamination in the bedrock had moved 
through the fractures toward two nearby rivers.[Footnote 21] On the 
basis of these results, in July 2001, EPA evaluated the potential 
human health risks from groundwater contamination to on-site and off-
site residents (i.e., residents who lived on or near the site) and 
commercial workers, and found that risks for on-site residents and 
workers exceeded acceptable levels for carcinogenic and 
noncarcinogenic contaminants. 

The Department of Health and Human Services' Agency for Toxic 
Substances and Disease Registry (ATSDR) also evaluated the risks from 
site contamination and published a series of studies that expressed 
concern about site contamination levels.[Footnote 22] Between May 1997 
and February 1999, ATSDR published five health consultations that 
responded to EPA requests to answer specific questions, such as 
whether consuming vegetables grown in site soils posed a health 
threat. For example, ATSDR's first consultation concluded that 
subsurface soil contamination levels posed a threat to residents if 
the contamination was dug up, or if similar levels of contamination 
were discovered in surface soils. Then, in September 2000, ATSDR 
published a public health assessment that evaluated site contamination 
and concluded that past and present exposures to surface soil (at that 
time) did not represent an apparent health hazard.[Footnote 23] 
However, the assessment also stated that this conclusion did not rule 
out the need for remedial action because subsurface contamination 
posed a long-term hazard if soil 2 feet below the ground in certain 
areas was disturbed. 

ATSDR and EPA officials told us that ATSDR's conclusion that surface 
soil contamination did not pose a public health hazard did not mean 
that EPA's action to remediate the site was unwarranted. In 
particular, officials from both agencies cited differences in the 
agencies' risk assessment views and processes as a reason why they 
could reach alternative conclusions about site risks. For example, 
ATSDR officials indicated that ATSDR's assessment focused on 
conditions in the first 6 inches of soil to evaluate what 
contamination exposures residents may have been subject to in the past 
and at the time of the assessment. However, the officials said that 
EPA's risk assessment would have been more focused on the hypothetical 
situation where subsurface soil contamination is brought to the 
surface in the future.[Footnote 24] Therefore, the officials said 
that, in fact, ATSDR would have had very serious concerns if the site 
had not been remediated because of the potential for high levels of 
contamination in the subsurface soil to be brought to the surface 
through activities such as tree planting or house remodeling. ATSDR 
also had concerns about potential exposures to groundwater 
contamination.[Footnote 25] As a result, the officials stated that 
ATSDR's assessment recommended that EPA continue its plans to 
implement a remedial action to remove source material from the site. 

On the basis of its conclusions about site risks, EPA set cleanup 
goals for different areas of the site that, when achieved, would 
reduce risks to acceptable levels for residential use. For example, 
EPA established site-specific qualitative objectives for its remedial 
actions, such as preventing human exposure to contamination, cleaning 
up areas of source contamination to allow for unrestricted land use 
and prevent future impacts to groundwater quality, and minimizing 
disturbance to residents and occupants of the Rustic Mall during a 
remedial action. EPA also developed quantitative remediation goals to 
identify the level at which remedial actions would need to be 
implemented to protect human health. According to site documents, 
there were no federal or state cleanup standards for soil 
contamination at the time of the cleanup effort. Therefore, EPA 
established risk-based remediation goals that would reduce excess 
carcinogenic risks to a level of 1 in 1 million, and that were 
consistent with New Jersey guidance for residential direct contact 
with soil.[Footnote 26] For the groundwater contamination, EPA used 
both federal and state chemical-specific standards to set risk-based 
remediation goals. 

According to site documents and Region 2 officials, risk levels 
required a remedial action regardless of the site's future use. The 
officials said that EPA considered what level of waste could be left 
on-site while still allowing for unrestricted residential use of 
properties; however, they noted that, with unrestricted residential 
use, there is a very low threshold for the level of waste that can be 
left on-site. They said that even the residually contaminated soil was 
sufficiently contaminated that EPA dug between 10 and 14 feet deep to 
allow for unrestricted use of residents' properties. Similarly, EPA 
determined that source material in the Rustic Mall needed to be 
remediated because of the potential future residential use of the 
site.[Footnote 27] According to a site document, EPA determined that, 
under a current use scenario (at the time of its risk assessment in 
2001), there were likely no unacceptable human health risks from 
contamination under the mall because contaminants were covered by 
buildings and pavement. However, the contamination could be exposed if 
these covers were removed during site redevelopment. Therefore, EPA 
identified the level of site cleanup required on the basis of the most 
conservative future use scenario.[Footnote 28] 

EPA's Remedy Selection Decisions Were Influenced by the Residential 
Nature of the Site: 

To select remedies to address the soil and groundwater contamination 
at the Federal Creosote site, EPA identified potential remedial 
technologies from agency guidance[Footnote 29] as well as from other 
publications and databases that listed potentially available 
technologies.[Footnote 30] After identifying potential technologies, 
EPA screened out less viable technologies, combined selected 
technologies into remedial alternatives, evaluated the alternatives, 
and selected a preferred remedy for each OU. See appendix III for a 
timeline of EPA's remedy selection efforts. 

Region 2 officials told us that, to identify technologies for site 
remediation, EPA identifies a range of technologies on a site-specific 
basis. According to agency guidance, EPA prefers three technologies 
for treating the type of soil contamination found at the Federal 
Creosote site: bioremediation--using microbes to degrade contaminants 
and convert them to carbon dioxide, water, microbial cell matter, and 
other products; low temperature thermal desorption (LTTD)--heating 
contaminated material to temperatures less than 1,000 degrees 
Fahrenheit to physically separate contaminants from soils; and 
incineration--heating contaminated material to temperatures greater 
than 1,000 degrees Fahrenheit to destroy contaminants.[Footnote 31] 
EPA also identified other technologies to cap, contain, excavate, 
extract, treat, or dispose of site soil or groundwater contamination, 
including a number of emerging or innovative technologies. 

For the soil contamination, the range of technologies EPA considered 
varied among the OUs at the site. During its remedy selection process 
for OU1, EPA primarily evaluated the three technologies preferred by 
agency guidance for soil contamination at wood-treatment sites. 
According to Region 2 officials, EPA considered a limited range of 
technologies for OU1 because, originally, the agency was evaluating 
whether it would need to evacuate residents to protect them from site 
contamination. Consequently, EPA conducted a more streamlined remedy 
selection process for OU1 to speed decision making. Alternatively, for 
OU2 (and later for OU3), EPA evaluated a wider range of technologies, 
including several emerging technologies. In addition, Region 2 
officials stated that differences in the contamination between the OUs 
impacted the range of technologies considered. Specifically, the 
officials said that the OU1 material was the more sludge-like, free- 
product creosote, whereas the OU2 contamination might not have been 
visible. The officials noted that, with less contaminated soils, more 
treatment options might become viable, since some options that might 
have difficulty treating more highly contaminated material might 
successfully treat less contaminated material. However, while EPA 
considered a wider range of technologies for OU2 and OU3, in general, 
EPA screened out the emerging technologies in favor of those that were 
identified as preferred in its guidelines.[Footnote 32] Ultimately, 
EPA determined that off-site thermal treatment and disposal of the 
soil contamination would best achieve its cleanup goals and were 
consistent with residential use of the site. In implementing this 
remedy, EPA determined that it would need to purchase some houses--
where contamination was inaccessible without demolishing the houses--
and permanently relocate these residents, while residents in other 
houses would only need to be relocated temporarily.[Footnote 33] 

For the groundwater contamination, Region 2 officials said that EPA 
tried to determine how to clean up the contaminated groundwater in the 
fractured bedrock but ultimately concluded that none of the options 
would be effective; moreover, many of the options would be expensive 
and take a long time to implement.[Footnote 34] As a result, EPA 
determined that attenuation of the groundwater contamination over 
time, long-term monitoring, and institutional controls to prevent the 
installation of wells at the site would be the best alternative to 
address contamination in the fractured bedrock. To select this remedy, 
EPA invoked a waiver for technical impracticability, which allowed it 
to select an alternative that would not comply with requirements to 
clean up the groundwater to levels that would meet site cleanup 
goals.[Footnote 35] Region 2 officials stated that one of the 
presumptions EPA makes in using a waiver for technical 
impracticability is that it has put forth its best effort to remove 
source contamination. Therefore, according to the officials, on the 
basis of agency guidance, EPA needed to clean up the source material 
that was contaminating the groundwater to justify a waiver for 
technical impracticability.[Footnote 36] Moreover, the officials said 
that by removing the source material, EPA may have helped prevent the 
contaminated groundwater area from getting larger. Also, the officials 
said that, in their judgment, EPA's action would help the 
contamination in the bedrock attenuate more quickly, although they 
were unable to quantify this impact. 

In selecting these remedies, EPA's decisions were influenced by 
several challenges associated with a residential cleanup, including 
(1) space constraints that limited on-site implementation of actions, 
(2) a determination that some options would not achieve the site 
cleanup goals, and (3) concerns about some options' community impacts. 

Space constraints. According to Region 2 officials, space constraints 
posed by the residential nature of the site limited EPA's ability to 
remediate contamination on-site. For example, the officials said that 
soil contamination in the lagoons and canals was interspersed 
throughout the residential area. As a result of the lack of available 
open land and the residential nature of the site, a site document 
indicated that options for on-site treatment and disposal of excavated 
material were not considered for OU1.[Footnote 37] Also, while EPA 
considered on-site treatment technologies and alternatives for OU2 and 
OU3, Region 2 officials said EPA did not consider buying additional 
houses to create more open space.[Footnote 38] They said that once EPA 
determined that the majority of houses in the residential area could 
be saved, it tried to avoid demolishing as many homes as 
possible.[Footnote 39] The officials also noted that EPA could have 
placed a treatment facility in a corner of the Rustic Mall, but that 
the mall was still a functioning commercial area at the time EPA was 
selecting remedies. The mall was in the middle of the town, and, 
according to the officials, feedback from local citizens indicated 
that the community relied heavily on the mall. As a result, EPA did 
not formally consider taking over additional areas of the mall to 
create more open space as part of a remedial alternative. Region 2 
officials acknowledged that, after EPA began the cleanup, the owner 
decided to demolish the mall. However, they stated that, when EPA made 
its remedy selection decisions, it did not have sufficient 
justification to purchase or demolish the mall. 

In particular, EPA Region 2 officials told us that the challenge of 
space constraints was a key factor in why EPA chose not to implement 
bioremediation or LTTD--two of EPA's preferred remedies for treating 
creosote contamination--on-site. For example, the officials noted that 
bioremediation of excavated material on-site would have required a lot 
of space to store the material while it was being treated with 
microbes that would help degrade the contamination. Similarly, the 
officials said that there was not sufficient space to stockpile 
material for treatment using LTTD. That is, to operate an LTTD unit 
efficiently, the officials said that EPA would have needed to feed 
material into the unit constantly. However, they said doing so was not 
possible at the site because, while EPA might excavate 100 tons of 
soil on some days, on other days, EPA was unable to excavate as much 
since it needed to work by hand around residents' houses. Given EPA's 
inconsistent rate of excavation, the agency would have needed to 
stockpile material to ensure a constant flow into an LTTD unit. 
However, according to Region 2 officials, there was not enough space 
to stockpile contaminated material awaiting treatment, and, as a 
result, the officials estimated that EPA could have operated an on-
site LTTD unit only 25 percent of the time, which they said would not 
have been cost-effective. Specifically, the officials said that it 
would take around 60,000 square feet for all of the operations 
associated with an LTTD unit. They noted that a space roughly this 
size was available in the northeast corner of the Rustic Mall. 
However, because of constraints, such as fire code access requirements 
for a bowling alley that bordered this area, the officials estimated 
that the total available space was actually only about 43,000 square 
feet. Also, EPA would have needed additional space for other 
facilities related to the cleanup. In addition, while EPA determined 
that bioremediation and LTTD could be used to treat contamination off-
site, EPA found that they would be difficult to implement because of a 
lack of permitted commercial facilities.[Footnote 40] As a result, EPA 
relied on incineration because incineration facilities were the most 
readily available for off-site treatment of material from the site. 

Level of cleanup required. EPA had concerns about whether certain 
technologies would effectively treat contamination to required levels, 
given the residential nature of the site. For example, EPA determined 
it was unlikely that such technologies as bioremediation of 
contaminated material in place would achieve the agency's soil 
remediation goals, because EPA was uncertain whether the 
bioremediation microbes could be distributed evenly in contaminated 
areas since some of the contamination was under residents' 
homes.[Footnote 41] Region 2 officials also said it was unlikely that 
EPA could have achieved its cleanup goals using bioremediation because 
of the high levels of soil contamination at the site. They said that 
if contamination levels are high, the microbes introduced into the 
soil could be killed before they have a chance to degrade the 
contaminants. Moreover, because of the high contamination levels and 
treatment requirements at the site, the officials said they had 
concerns about the effectiveness of using LTTD. They stated that LTTD 
treats material using lower temperatures than incineration, and that 
it removes about 80 percent of the contamination each time material is 
passed through the unit. As a result, sometimes material must be 
treated multiple times before it meets residential standards. The 
officials indicated that this would have probably been the case with 
the Federal Creosote material because it was so highly contaminated. 
They said, given the nature of the contamination at the site, 
incineration was a more efficient method of treatment to achieve the 
agency's remediation goals. 

While the high treatment levels required because of the residential 
nature of the site impacted EPA's choices about individual soil 
remediation technologies, they also influenced decisions about whether 
to dispose of treated and untreated material on-site, or at an off-
site location. According to Region 2 officials, if EPA disposed of 
excavated material on-site, the agency would have had to ensure, 
through treatment and testing, that the soil met residential 
standards. Consequently, the officials concluded that if EPA disposed 
of excavated material on-site, it would have had to treat and test the 
material more extensively than it did for off-site disposal. The 
officials said that only about 35 percent of the material excavated 
from the site needed to be thermally treated before it could be 
disposed of off-site.[Footnote 42] The rest of the excavated material 
could be disposed of without treatment at a hazardous or nonhazardous 
waste landfill. However, they said, if EPA had disposed of material on-
site, it would have had to test and possibly treat 100 percent of the 
material to ensure that it met residential standards. Due to the 
potential expense of additional treatment and sampling, EPA determined 
that off-site disposal would be more cost-effective. 

For the groundwater contamination, according to site documents, EPA 
found that none of its remedial alternatives, including those based on 
extracting or treating the contamination in place, would be able to 
achieve its cleanup goals effectively and reliably within a reasonable 
time frame. For example, EPA found that some of the groundwater 
contaminants could take decades to move through the groundwater, and, 
as a result, it would take an extremely long time to remediate these 
contaminants using an extraction technology. Moreover, EPA estimated 
that the technology that was most likely to be able to achieve its 
remediation goals--extracting contaminants using steam--would cause 
significant disruption to the residential neighborhood and would be 
much more expensive than EPA's other alternatives. On the basis of its 
experience at other sites, EPA determined that complete removal of the 
groundwater contamination in the bedrock at the site was not 
practicable. In addition, EPA found that several of the treatment 
technologies it considered would not be effective at treating the 
highly contaminated free-product creosote found in portions of the 
site. 

Community impacts. The residential nature of the site and the 
importance of the Rustic Mall to the community also influenced EPA's 
remedy selection, given the effects that different technologies and 
alternatives might have on the community. For example, according to 
EPA, some of the substances that could be used to immobilize soil 
contamination in the ground were potentially more toxic than the 
creosote contamination. Also, certain options that treated 
contamination in place or extracted it from the soil or groundwater 
would have emitted heat or gas that could have posed risks to 
residents and the community. Moreover, EPA determined that some 
options would have significantly disrupted the community because of 
the need to install equipment, wells, and piping throughout the 
residential and commercial areas. 

Also, because EPA was implementing a remedial action in a residential 
neighborhood at the site, it was concerned about the length of the 
cleanup and other timing impacts on the community. Region 2 officials 
said that EPA generally does not use certain alternatives unless the 
agency has the flexibility to accomplish remediation over a long time 
frame on the basis of the current land use (e.g., the site is 
abandoned). Under these circumstances, EPA could use a remedy like 
bioremediation of contaminated material in place, which would cause 
long-term disruption if implemented in a residential neighborhood. 
Also, Region 2 officials said that, if EPA had used on-site LTTD to 
treat contaminated material, it could not have operated the unit in 
the most efficient way--24 hours a day--because the residents in 
houses within 200 feet of where the unit would have been located would 
have been negatively affected by its lights and noise during the 
night. However, the officials said, if EPA had only run the LTTD unit 
8 hours a day, the cleanup effort would have taken much 
longer.[Footnote 43] The length of time involved was a particular 
concern in EPA's evaluation of groundwater remediation alternatives. 
According to the Region 2 officials, the best alternative to extract 
contaminated groundwater from the bedrock would have taken 18 to 20 
years to implement and would have covered the site with machinery. 

Finally, EPA factored future land use impacts into its remedy 
selection decisions.[Footnote 44] For example, EPA found that options 
that relied on containment or deed restrictions, but that left 
contamination under and around the residential community, were not 
viable alternatives. Region 2 officials said capping the contamination 
would not have supported use of the land as a residential area because 
residents would have had to sign agreements not to disturb the cap, 
which would have restricted their use of the properties. Also, because 
of these restrictions, the officials said it is likely that some 
owners would have refused to sign the necessary agreements, and EPA 
would have had to take an enforcement action. Similarly, EPA avoided 
certain remedies for the Rustic Mall because of the impacts that they 
could have on the community's ability to redevelop the mall as well as 
on the operation of the mall. A Borough of Manville official told us 
that the Rustic Mall was the "hub of the town" and was located 
directly behind buildings on the town's Main Street. As a result, he 
said the community was very opposed to alternatives that would have 
left or treated contamination on-site.[Footnote 45] He said that, in 
the town's view, the contamination under the mall needed to be cleaned 
up. Otherwise, it would have been difficult to get tenants into the 
mall in the future, and the town might have ended up with a blighted 
area in the center of the community. He also said the community was 
concerned that no one would want to come and shop at the mall if there 
was a treatment facility in the parking lot. 

EPA Placed a High Priority on Cleaning Up the Federal Creosote Site: 

EPA placed a high priority on scheduling and funding the Federal 
Creosote site work because the contamination was in a residential 
area. According to Region 2 officials, it is rare to find source 
contamination, such as the free-product creosote, under a residential 
area, and most sites with the level and extent of contamination found 
at the Federal Creosote site are abandoned. The officials said EPA 
places the highest priority on addressing the principal threats at 
residential sites first. As evidence of this prioritization, EPA 
initiated efforts to study, select a remedy for, and begin cleanup of 
the residential part of the site before undertaking similar efforts 
for the Rustic Mall. For example, Region 2 officials said that EPA 
decided relatively early in the cleanup process to break the site into 
three OUs to allow work to proceed as quickly as possible. EPA 
determined that it needed to get to work immediately on OU1, and that 
the groundwater contamination and commercial area could wait until 
after EPA had decided what to do with the residential area. The Region 
2 officials said that breaking the site into different OUs was 
important because EPA knew that it needed to relocate some OU1 
residents, and this process can be time-consuming--one official noted 
that residents who must permanently relocate have 1 year to do so. 
While this process took less time at the Federal Creosote site, EPA 
did not know that would be the case initially. Moreover, the Region 2 
officials said that the first couple of years EPA spent studying the 
site caused a great deal of anxiety for residents, because they did 
not understand the risks of remaining in their homes and could not 
sell their homes if the homes would need to be demolished. The 
officials said the OU1 ROD informed residents that most of the homes 
in the neighborhood would not need to be demolished, and this helped 
reduce residents' anxiety.[Footnote 46] 

EPA also took steps to shorten the time needed to select, design, and 
implement the remedial actions. For example, Region 2 officials said 
that, because of the residential nature of the site, the site 
investigation process was both unusually extensive and expedited in 
comparison to other sites. Region 2 officials said that EPA began 
sampling early because, when the site was discovered, the agency was 
concerned that contamination risks could be so significant that 
residents might need to be evacuated. As a result, they said that the 
agency gathered a large amount of information about site contamination 
before listing the site on the NPL. The officials said this data 
collection effort helped EPA move forward with site work quickly 
because, with a large amount of data to use to gauge its overall 
approach to the site, EPA was able to compress the removal evaluation, 
listing process, and RI/FS into a relatively short amount of time. In 
addition, EPA tried to streamline work by configuring its sampling 
efforts to satisfy postexcavation requirements to confirm that 
contaminated material no longer remained on-site. Specifically, site 
documents show that to meet New Jersey requirements, EPA took samples 
on 30-by-30 foot grids to confirm that contamination was no longer 
present along the sides and bottom of an excavated area. Rather than 
wait until the excavation was completed to take additional samples to 
confirm that contamination was not present, EPA incorporated these 
requirements into earlier sampling efforts. As a result, if samples 
were clean, EPA could immediately backfill an area, which reduced the 
overall length of the cleanup effort.[Footnote 47] Finally, in an 
effort to expedite the cleanup effort, EPA Region 2 officials said 
that more of the region's resources were devoted to the site relative 
to other sites that the region needed to address at that time. 

As a result of these efforts to prioritize and expedite site cleanup 
work, the Federal Creosote site reached key cleanup milestones in less 
time than some other site cleanups. Region 2 officials said that they 
completed the three RODs for the site in about 3 years, which they 
said is a very quick time frame to complete such analyses. They noted 
that issuing a ROD is an intensive process that at another site, for 
example, took over a decade. Also, the Federal Creosote site reached 
EPA's construction complete stage more quickly than other megasites-- 
that is, sites at which actual or expected total cleanup costs, 
including removal and remedial action costs, are expected to amount to 
$50 million or more. In July 2009, we reported that, based on EPA data 
through fiscal year 2007, the median length of time it took for 
megasites to reach construction complete after NPL listing was 14.8 
years.[Footnote 48] However, according to EPA data, the Federal 
Creosote site reached construction complete in just over 9 years. 
[Footnote 49] 

Total Site Costs Exceeded Early Construction Cost Estimates Largely 
because of the Nature of these Estimates and the Discovery of 
Additional Contamination: 

Total site costs exceeded construction estimates at the Federal 
Creosote site by roughly $233 million, primarily because (1) EPA's 
early construction estimates were not designed to include all site- 
related expenses and (2) additional quantities of contaminated 
material were discovered during the cleanup effort. Other factors, 
such as methodological variation for estimating site costs and 
contractor fraud, accounted for a smaller portion of the cost 
difference. 

Total Site Costs Exceeded Construction Estimates by Approximately $233 
Million: 

According to our analysis, total site-related costs, including 
remedial construction and other response costs at the Federal Creosote 
site through the spring of 2009, were approximately $338 million, 
[Footnote 50] a roughly $233 million difference from the estimated 
remedial construction costs of $105 million.[Footnote 51] Total site 
costs were higher than construction estimates for several reasons. 
[Footnote 52] As shown in figure 3, of the $233 million difference, 
39.6 percent (or about $92 million) is due to other response costs 
that were not included in EPA's construction estimates;[Footnote 53] 
47.5 percent (or about $111 million) is from an increase in remedial 
construction costs--mostly directly related to the discovery of 
additional contaminated material; and 12.9 percent (or about $30 
million) is due to other factors--primarily differences in cost 
estimation methodology and, to a smaller extent, contractor fraud. 

Figure 3: Difference between Estimated Construction and Total Site 
Costs at the Federal Creosote Site: 

[Refer to PDF for image: pie-chart] 

Remedial construction costs potentially related to greater 
contaminated soil quantities (about $111 million): 47.5%; 

Other response costs not included in construction estimates (about $92 
million): 39.6%; 

Other factors (about $30 million): 12.9%. 

Total = $233 million. 

Source: GAO analysis of data obtained from EPA, the Corps, and court 
documents. 

Note: Site costs were drawn from several sources, including Corps data 
on construction contractor costs updated through February 15, 2009, as 
well as Corps and EPA sources updated through various dates from April 
to early May, 2009. Due to several reasons, including our methodology 
for adjusting costs to fiscal year 2009 dollars, the percentages in 
this figure are considered to be approximate. 

[End of figure] 

EPA's Construction Cost Estimates Intentionally Did Not Include All 
Site Costs: 

EPA intentionally included only costs related to the construction and 
maintenance of the selected remedies rather than total sitewide costs 
in its early cost estimates, which follows its guidance, according to 
the agency.[Footnote 54] EPA prepares these preliminary estimates 
during the remedy selection process to compare projected construction 
costs across different remedial action alternatives. Specifically, the 
National Contingency Plan directs EPA to consider the capital costs of 
construction and any long-term operation and maintenance costs as part 
of the remedial alternative screening process.[Footnote 55] According 
to EPA guidance, these estimates are not intended to include all site- 
related expenses, and certain expenses, such as early site 
investigation and EPA enforcement costs, are beyond the scope of these 
early estimates because these costs are not linked to a specific 
remedial alternative and, therefore, would not affect the relative 
comparison of alternatives. For example, while site investigation 
studies were conducted for each operable unit, these studies were 
completed prior to remedy selection to inform the selection process 
and, therefore, were not linked to any particular remedy. Similarly, 
the removal cleanup of surface soils in the residential area occurred 
prior to remedy selection and, therefore, was not related to the 
construction costs of any particular remedial alternative. Table 2 
summarizes costs for activities that were not included in EPA's 
remedial construction cost estimates--other response costs--at the 
Federal Creosote site. 

Table 2: Other Response Costs Not Included in Construction Estimates 
for the Federal Creosote Site: 

Cost description: Indirect support costs related to site construction 
work[B]; Total costs[A[(FY 2009 DOLLARS)] rs): $72,679,000. 

Cost description: Site investigation and risk assessment study; 
Total costs[A](FY 2009 Dollars): $11,038,700. 

Cost description: Indirect support costs related to other response 
efforts[B]; 
Total costs[A](FY 2009 Dollars): $4,209,100. 

Cost description: EPA payroll and travel; 
Total costs[A](FY 2009 Dollars): $2,085,100. 

Cost description: Enforcement; 
Total costs[A](FY 2009 Dollars): $1,579,800. 

Cost description: Analytical and technical support; 
Total costs[A](FY 2009 Dollars): $319,300. 

Cost description: State cooperative agreement; 
Total costs[A](FY 2009 Dollars): $278,500. 

Cost description: Emergency removal cleanup; 
Total costs[A](FY 2009 Dollars): $32,200. 

Cost description: Miscellaneous; 
Total costs[A](FY 2009 Dollars): $8,400. 

Cost description: Total; 
Total costs[A](FY 2009 Dollars): $92,230,100. 

Source: GAO analysis of EPA data. 

Note: Other response costs were drawn from EPA data through April 30, 
2009. As a result of the delay between when a cost is incurred and 
when it may be submitted and entered into EPA's cost-tracking systems, 
the total costs listed in this table are approximate. 

[A] Certain categories in this table include contractor annual 
allocation costs. These costs include money spent by government 
contractors doing site-related work not traceable to a particular 
site, such as hazardous materials training. Government contractors 
allocate these costs across the sites on which they have worked during 
the course of each year, and EPA treats them as direct costs. 

[B] Indirect support costs are overhead costs for activities related 
to managing the Superfund program, such as guidance development, that 
EPA allocates across all Superfund sites. 

[End of table] 

The Need to Remediate Greater-than-Expected Quantities of Contaminated 
Material Contributed Most to the Gap between Estimated and Actual 
Construction Costs: 

During excavation, contractors discovered greater-than-expected 
amounts of contaminated material requiring remediation across all OUs, 
which contributed most to the difference between estimated and actual 
construction costs. Based on our analysis of EPA documents, the 
initial ROD estimates for the site indicated that approximately 
154,100 to 164,400 tons of material would need to be excavated for 
treatment or disposal; however, EPA ultimately found that roughly 
456,600 tons of material needed to be excavated--an increase of at 
least 178 percent. As shown in table 3, according to our analysis, 
increased amounts excavated from the OU1 and OU3 areas contributed the 
most to the difference between the estimated and actual excavated 
amounts across the site as a whole. 

Table 3: Difference between Estimated Soil Quantities in the RODs and 
Final Soil Quantities at the Federal Creosote Site, by OU: 

Hundreds of tons: 

OU1: 
Estimated soil quantity from RODs: 66,200 to 70,700[A]; 
Final soil quantity: 210,800; 
Soil quantity increase from RODs: 140,100 to 144,600; 
Percentage increase: 198% to 218%. 

OU2: 
Estimated soil quantity from RODs: 31,700 to 33,800; 
Final soil quantity: 67,900; 
Soil quantity increase from RODs: 34,200 to 36,300; 
Percentage increase: 101 to 115. 

OU3: 
Estimated soil quantity from RODs: 56,300 to 60,000; 
Final soil quantity: 177,800; 
Soil quantity increase from RODs: 117,800 to 121,600; 
Percentage increase: 196 to 216. 

Total[B]: 
Estimated soil quantity from RODs: 154,100 to 164,400; 
Final soil quantity: 456,600; 
Soil quantity increase from RODs: 292,200 to 302,500; 
Percentage increase: 178% to 196%. 

Source: GAO analysis of EPA documents. 

Note: Soil quantities are presented in a range because EPA updated its 
conversion factor for calculating soil weights from volumes at the 
site as additional information regarding the density of the soil 
became available. The lower bound in this table generally represents 
soil quantities using a conversion factor of 1.5 cubic yards per ton, 
whereas the upper bound reflects EPA's later use of 1.6 cubic yards 
per ton. We based our analysis on EPA documents and did not 
independently verify the soil quantities reported by EPA. 

[A] EPA revised its original quantity estimate for OU1 soils as a 
result of its investigation of OU2 contamination. Based on our 
analysis of site documents, the revised quantity is approximately 
152,600 to 162,700 tons. 

[B] Totals may not add due to rounding. 

[End of table] 

According to EPA officials, it is common for EPA to remove more soil 
than originally estimated at Superfund sites because of the 
uncertainty inherent in using soil samples to estimate the extent of 
underground contamination. For example, EPA guidance indicates that 
the scope of a remedial action is expected to be continuously refined 
as the project progresses into the design stage and as additional site 
characterization data and information become available.[Footnote 56] 
However, both Corps and EPA officials stated that the Federal Creosote 
site posed a particular challenge for estimating soil quantities prior 
to excavation because of the way in which the waste moved at the site 
and, in some cases, because of access restrictions during sampling. 
According to EPA's Remedial Project Manager (RPM) for the site, soil 
contaminants generally either stay in place or migrate straight down; 
however, while some of the creosote waste at the site stayed in place, 
some of the waste migrated both horizontally and vertically. The RPM 
said that this migration made it difficult to predict the waste's 
location through sampling. For example, during excavation, contractors 
found seams of contaminated material, some of which led to additional 
pockets of creosote waste, while others did not. Given the diameter of 
the sampling boreholes (which were generally 2 to 4 inches wide) and 
the width of the seams of creosote waste (which in some cases were 
only 6 inches wide), the sampling process could not detect all of the 
creosote seams at the site, despite what EPA officials considered to 
be the extensive sampling during the early site investigations that 
formed the basis for the initial cost estimates. Additionally, 
sampling during the site investigations for the residential area as 
well as the Rustic Mall was limited by the location of buildings and 
access restrictions, according to EPA's RPM. For example, site 
documents indicate that no samples could be taken from under the mall 
during the OU3 soil investigation because the buildings were being 
used. It was not until the mall owners decided to demolish the 
existing structures as part of a town revitalization plan that mall 
tenants left and EPA was able to take samples in the areas covered by 
the buildings. These areas were found to contain additional areas of 
creosote waste, as shown in figure 4.[Footnote 57] 

Figure 4: Rustic Mall Areas Remediated at the Federal Creosote Site: 

[Refer to PDF for image: illustrated map] 

The following are indicated on a map of the Rustic Mall and Claremont 
Development residential area: 

Identified contamination areas in Rustic Mall ROD: 
Canal Area; 
Southwest Area; 
South Area; 
North Area 2. 

Additional remediated areas: 
Central Area; 
North Area 1. 

Also indicated on the map: 
Canal A; 
Lagoon A; 
Canal B; 
Lagoon B. 

Source: GAO analysis of EPA site documents. 

[End of figure] 

Although the mobility of the waste in the subsurface soil and sampling 
limitations hindered EPA's ability to determine the total quantity of 
material requiring excavation during the pre-ROD site investigation 
when the initial cost estimates were prepared, soil sampling during 
this stage was generally successful at identifying which residential 
properties contained contamination, according to our analysis of site 
documents. For example, pre-ROD soil sampling allowed EPA to correctly 
identify 83 of the 93 residential properties that would eventually 
require remediation, as shown in figure 5.[Footnote 58] 

Figure 5: Residential Area Properties Remediated at the Federal 
Creosote Site: 

[Refer to PDF for image: illustrated map] 

The following are indicated on a map of the Claremont Development 
residential area: 

OU1 properties identified in EPA’s RODs: 28 properties; 
Additional OU1 properties remediated: 2 properties; 
OU2 properties identified in EPA’s RODs: 52 properties; 
Additional OU2 properties remediated: 8 properties. 

Also indicated on the map: 
Lagoon A; 
Lagoon B. 

Source: GAO analysis of EPA site documents. 

Note: In addition to the properties highlighted in this figure, other 
areas, such as sections of streets and railroad rights of way, were 
also excavated. 

[End of figure] 

According to EPA guidance, because of the inherent uncertainty in 
estimating the extent of site contamination from early investigation 
data, cost estimates prepared during the RI/FS stage are based on a 
conceptual rather than a detailed idea of the remedial action under 
consideration. The guidance states that these estimates, therefore, 
are expected to provide sufficient information for EPA to compare 
alternatives on an "order of magnitude" basis, rather than to provide 
an exact estimate of a particular remedy's costs. For example, the 
guidance also states that preliminary cost estimates prepared to 
compare remedial alternatives during the detailed analysis phase of 
the RI/FS process are expected to range from 30 percent below to 50 
percent above actual costs. 

However, at the Federal Creosote site, actual construction costs were 
more than twice what EPA estimated. Specifically, we found that 
sitewide remedial construction costs increased by $141 million over 
EPA's estimated amounts. According to site documents, increases in the 
quantity of material requiring excavation, transportation, treatment, 
or disposal resulted in higher construction costs across all OUs. Our 
analysis of site cost data indicated that construction costs 
potentially associated with the additional quantity of contaminated 
material accounted for most of this increase ($111 million, or about 
78.7 percent).[Footnote 59] In particular, soil excavation, 
transportation, treatment, and disposal costs constituted 
approximately 56.1 percent ($62 million) of the increased construction 
costs potentially related to additional quantities of material, and 
26.7 percent of the overall $233 million difference between estimated 
construction and total site costs, as shown in figure 6. 

Figure 6: Actual Remedial Construction Costs as a Portion of the 
Difference between Estimated Construction and Total Site Costs at the 
Federal Creosote Site: 

[Refer to PDF for image: pie-chart] 

* Remedial construction costs potentially related to greater 
contaminated soil quantities: 47.5%: 
- Other sitewide construction costs (20.9%); 
- Excavation, transportation, treatment, and disposal costs (26.7%); 
* Other response costs not included in construction estimates: 39.6%; 
* Other factors: 12.9%. 

Source: GAO analysis of data obtained from EPA, the Corps, and court 
documents. 

Note: Site costs were drawn from several sources, including Corps data 
on construction contractor costs updated through February 15, 2009, as 
well as Corps and EPA sources updated through various dates from April 
to early May, 2009. Due to several reasons, including our methodology 
for adjusting costs to fiscal year 2009 dollars, the percentages in 
this figure are considered to be approximate. 

[End of figure] 

According to EPA's RPM, both the need to excavate greater amounts of 
material and the reclassification of excavated material from 
nonhazardous waste to hazardous waste affected excavation, 
transportation, treatment, and disposal costs. For example, the 
discovery of additional pockets of creosote waste increased the 
overall amount of material requiring excavation and treatment or 
disposal because, in addition to removing the waste itself, any soil 
overlying the contamination needed to be removed and disposed of to 
access the creosote waste. Additionally, if a pocket of creosote waste 
was unexpectedly discovered in an area of soil that had already been 
designated for excavation and disposal in a landfill without treatment 
because prior sampling indicated it was less contaminated, the overall 
amount of soil to be excavated would not be affected, but costs would 
increase because treatment is more expensive than landfill 
disposal.[Footnote 60] 

In addition, EPA and Corps officials said that the need to remediate 
greater quantities of material contributed to increases in other 
sitewide construction costs, such as general construction requirements 
and site restoration costs. Our analysis showed that such costs 
accounted for another 20.9 percent of the difference between estimated 
construction costs and total site costs--although the exact extent to 
which additional amounts of material contributed to the difference in 
costs is not clear. EPA's RPM stated that the effect of increased 
quantities varied, depending on the OU. However, EPA and Corps 
officials said that in general, more extensive excavation would 
increase design engineering, inspection, and other costs as well as 
costs for general construction requirements and for site restoration, 
as shown in table 4. For example, the decision to remediate additional 
contaminated material under the Rustic Mall buildings led to increased 
design engineering costs because the original excavation plans were 
created under the assumption that the mall would remain standing, and 
further rounds of design sampling were needed to identify the extent 
and location of contamination once the buildings were demolished. 
Additionally, our analysis of site documents indicated that the 
increased time required to excavate additional material could have led 
to greater project costs for general construction requirements, such 
as temporary facility rental, site security, and health and safety 
costs. Similarly, site restoration costs, such as costs for backfill 
soil, could have increased because more backfill would be required to 
restore the site after excavation. 

Table 4: Estimated and Actual Remedial Construction Costs at the 
Federal Creosote Site: 

Cost description[A]: Excavation, transportation, treatment, and 
disposal; 
ROD estimates (FY 2009 dollars)[B]: $76,110,500; 
Actual cost (FY 2009 dollars)[B,C]: $159,643,400. 

Cost description[A]: Design-engineering, inspection, and other 
costs[D]; 
ROD estimates (FY 2009 dollars)[B]: $12,027,400; 
Actual cost (FY 2009 dollars)[B,C]: $34,133,500. 

Cost description[A]: Resident relocation and home demolition; 
ROD estimates (FY 2009 dollars)[B]: $5,874,000; 
Actual cost (FY 2009 dollars)[B,C]: $9,405,500. 

Cost description[A]: General construction requirements; 
ROD estimates (FY 2009 dollars)[B]: $5,495,500; 
Actual cost (FY 2009 dollars)[B,C]: $26,613,900. 

Cost description[A]: Site restoration; 
ROD estimates (FY 2009 dollars)[B]: $4,298,300; 
Actual cost (FY 2009 dollars)[B,C]: $15,101,000. 

Cost description[A]: Site preparation; 
ROD estimates (FY 2009 dollars)[B]: $1,183,700; 
Actual cost (FY 2009 dollars)[B,C]: $926,500. 

Cost description[A]: Total; 
ROD estimates (FY 2009 dollars)[B]: $104,989,400; 
Actual cost (FY 2009 dollars)[B,C]: $245,823,800. 

Source: GAO analysis of data obtained from EPA and the Corps. 

Note: Site costs were drawn from several sources, including Corps data 
on construction contractor costs updated through February 15, 2009, as 
well as Corps and EPA sources updated through various dates from April 
to early May, 2009. Due to several reasons, including our methodology 
for adjusting costs to fiscal year 2009 dollars, the amounts in this 
table are considered to be approximate. 

[A] Some actual costs were not listed as line items in the ROD 
estimates. In these cases, we included related actual costs in the 
same category. For example, water removal costs for soil excavation 
areas below the water table and odor control were not line items in 
the estimates, but we included them in the "excavation" category 
because they are part of the process. 

[B] These amounts do not match EPA-published data as a result of our 
methodology for adjusting the data to fiscal year 2009 dollars, among 
other reasons. 

[C] This amount includes approximately $2.1 million in inflated prices 
due to contractor fraud, as we discuss later in this section of the 
report. 

[D] Some groundwater remediation costs incurred as part of the Corps' 
contract for design activities were included under the engineering, 
inspection, and other costs category. This total also includes the 
Corps' management and support fee, which according to EPA's RPM, was 
not charged to individual sites at the time the ROD estimates were 
prepared. 

[End of table] 

According to the RPM, EPA and the Corps instituted certain controls at 
the site to minimize costs. In particular, the RPM stated that the 
Corps took steps to ensure that material was not unnecessarily 
excavated and sent for treatment and disposal. For example, if 
contractors found an unexpected pocket of creosote waste during 
excavation, they were required to notify the Corps official on-site, 
who would decide whether additional excavation was required depending 
upon visual inspection and additional testing, as needed. The 
contractor was not allowed to excavate beyond the original excavation 
limits without Corps approval. According to the RPM, the Corps' 
approach of reevaluating the original excavation depth on the basis of 
additional sampling results and a visual inspection of the soil led to 
cost savings because in some areas less material needed to be 
excavated than originally planned. Furthermore, EPA and Corp officials 
stated that this process minimized unnecessary treatment and disposal 
costs that might be incurred if "clean" soil was sent for treatment or 
hazardous waste disposal. Additionally, EPA's decision in November 
2002 to allow treated soil to be disposed of in a nonhazardous waste 
facility if it met the facility's criteria for contamination levels 
helped reduce unit costs for treatment and disposal because disposing 
of soil at a hazardous waste facility is more expensive. For example, 
in a bid for a contract to treat and dispose of soil following EPA's 
decision, the selected subcontractor submitted a unit price for 
treatment and disposal at a nonhazardous waste facility that was $80 
(or 16 percent) less than its unit price for treatment and disposal at 
a hazardous waste facility--which for that particular contract saved 
$800,000.[Footnote 61] 

Furthermore, on the basis of information gathered from site documents 
and from statements made by EPA and Corps officials, EPA and the Corps 
took other steps intended to minimize costs. For example, a Corps 
official said that reducing the duration of the project could help 
minimize certain site costs. Specifically, according to our analysis 
of site documents, to reduce the amount of time spent waiting for 
sampling results prior to backfilling an excavated area, EPA and the 
Corps incorporated state postexcavation sampling requirements into 
their design sampling plans for earlier investigations. Accordingly, 
unless additional excavation was required to meet the cleanup goals, 
these samples could be used to confirm that the boundaries of the 
excavation areas had been tested for contamination. Additionally, our 
analysis of site documents showed that the Corps tested various odor 
control measures before beginning excavation at certain areas of the 
site, which allowed it to use less expensive odor control alternatives 
than originally planned and saved approximately $1.1 million in 
implementation costs. These measures also helped to speed up the 
construction work. Finally, according to the RPM, the Corps was able 
to minimize costs by managing the work to avoid costly contractor 
demobilization and remobilization expenses. For example, the Corps 
dissuaded the contractors from removing idle equipment and worked with 
the RPM to resolve administrative or funding issues or questions about 
the work as they arose to prevent an expensive work stoppage.[Footnote 
62] 

Contrasting Cost-Estimating Methodologies and Contractor Fraud Explain 
a Smaller Portion of the Difference between Estimated Construction and 
Total Site Costs: 

Other factors, including different cost-estimating methodologies and 
contractor fraud, explain a smaller portion of the difference between 
estimated construction and total site costs at the Federal Creosote 
site. In developing its estimates, EPA followed agency guidance, which 
states that as a simplifying assumption, most early cost estimates 
assume that all construction costs will be incurred in a single year. 
[Footnote 63] According to EPA, since the estimated implementation 
periods for EPA's remedial actions were relatively short periods of 
time, EPA did not discount future construction costs in its estimates, 
and, therefore, these estimates were higher than they would have been 
otherwise.[Footnote 64] In accordance with our best practices 
regarding the use of discounting, we adjusted the initial cost 
estimates to reflect that costs were projected to accrue over several 
years and that, therefore, future costs should be discounted.[Footnote 
65] However, by discounting future construction costs prior to 
adjusting for inflation, our discounted values were lower than EPA's 
original estimates in site documents. According to our analysis, 
discounting estimated costs accounted for approximately 12 percent of 
the $233 million difference between estimated construction and total 
site costs (see figure 7). 

Figure 7: Contrasting Methodologies and Contractor Fraud as a Portion 
of the Difference between Estimated Construction and Total Site Costs 
at the Federal Creosote Site: 

[Refer to PDF for image: pie-chart] 

* Remedial construction costs potentially related to greater 
contaminated soil quantities: 47.5%; 
* Other response costs not included in construction estimates: 39.6%; 
* Other factors: 12.9%: 
- Cost estimate discount adjustment (12.0%); 
- Confirmed or proven contractor fraud (0.9%). 

Source: GAO analysis of data obtained from EPA, the Corps, and court 
documents. 

Note: Site costs were drawn from several sources, including Corps data 
on construction contractor costs updated through February 15, 2009, as 
well as Corps and EPA sources updated through various dates from April 
to early May, 2009. Due to several reasons, including our methodology 
for adjusting costs to fiscal year 2009 dollars, the percentages in 
this figure are considered to be approximate. 

[End of figure] 

Contractor fraud also contributed to the difference between estimated 
construction and total site costs, but to a small degree. However, 
while some parties have pled guilty to fraud, the full extent of the 
effect of fraud on site costs will not be known until all 
investigations are complete. Court documents alleged that employees of 
the prime contractor at the site, as well as some subcontractors, were 
engaged in various kickback and fraud schemes, which resulted in 
inflated prices for certain subcontractor services. For example, a 
subcontractor for soil treatment and disposal agreed to pay 
approximately $1.7 million in restitution to EPA for fraud in 
inflating its bid prices. In addition, court documents alleged that 
fraudulent price inflation also affected other site costs, including 
certain subcontracts for items such as wastewater treatment, backfill, 
landscaping services, and utilities. To date, our analysis of 
available court documents indicated that at least approximately $2.1 
million in inflated payments may be directly attributable to fraud at 
the Federal Creosote site. On the basis of currently available 
information, this figure represents less than 1 percent of the 
difference between estimated construction and total site costs. 
However, since the fraud investigations are ongoing and additional 
charges may be filed, the full extent of contractor fraud is not 
currently known. See appendix I for more information about site-
related fraud investigations. 

EPA Provided Overall Project Management and Communicated with 
Residents, while the Corps Oversaw Contractor Implementation of the 
Remedy: 

EPA managed the overall cleanup and communicated with residents 
through a dedicated on-site staff presence, among other actions. The 
Corps implemented the cleanup work by hiring and overseeing 
contractors; the Corps was less involved in selecting and overseeing 
subcontractors at the site. 

EPA Managed the Cleanup and Communicated with Residents: 

According to a 1984 interagency agreement between EPA and the Corps 
for the cleanup of Superfund sites, EPA maintains statutory 
responsibility for implementing the Superfund program. In addition to 
selecting the remedy at a site, EPA provides overall management of the 
cleanup, ensures that adequate funding is available, and manages 
relationships with other interested parties, such as residents. If EPA 
decides that Corps assistance is needed to conduct cleanup work, EPA 
establishes site-specific interagency agreements. These agreements 
outline the specific tasks and responsibilities of the Corps at the 
site and provide a proposed budget for the activities listed. Once the 
site-specific agreements are established, EPA's primary 
responsibilities are to make sure that the work continues without 
interruption and that adequate funding is available, according to EPA 
officials. EPA officials also noted that the agency does not have the 
authority to direct Corps contractors at the site; rather, all 
instruction and direction to contractors goes through the Corps. 

To fulfill its project management and community outreach 
responsibilities, EPA dedicated a full-time RPM to the Federal 
Creosote site, according to Region 2 officials. Although RPMs 
generally have two or more sites for which they are responsible at any 
given time, Region 2 officials stated that the size and complexity of 
the site required a higher level of EPA involvement. For example, the 
officials said that the relatively large size of the site and 
stringent cleanup goals meant that a large area was excavated, and the 
complexity of the cleanup process led to a greater number of questions 
from the Corps and its contractors that required EPA's attention. 
According to the officials, the RPM was on-site at least two to three 
times per week; however, during some segments of the work, he was on-
site almost every day. They noted that the design phase in particular 
required close coordination with the Corps because design activities 
for different areas of the site occurred simultaneously and were often 
concurrent with construction.[Footnote 66] Consequently, the RPM said 
he was on-site working with the Corps and its design contractor to 
design new phases of the work; revise existing designs; and answer any 
questions regarding ongoing construction activity, such as whether to 
excavate additional pockets of waste found during the construction 
phase. According to the RPM, although the Corps was required to ask 
EPA for approval only to expand excavation to properties that were not 
included in the RODs, in practice, Corps officials kept him informed 
whenever additional excavation was required, and, in many cases, he 
made the decision regarding whether to broaden or deepen the excavated 
area. 

To monitor project progress and funding, the RPM had weekly on-site 
meetings with the Corps and received weekly and monthly reports on 
progress and site expenditures, according to EPA officials.[Footnote 
67] At the weekly meetings, the RPM would answer Corps questions 
regarding the work and be informed of any contracting or 
subcontracting issues that might delay or stop work at the site. 
Moreover, as part of EPA's oversight of site progress, the RPM said he 
reviewed Corps documents regarding any changes in the scope of the 
work. Because EPA provided funding to the Corps on an incremental 
basis, the RPM also closely monitored the rate of Corps expenditures 
to ensure sufficient funding to continue the work, according to EPA 
officials. The RPM explained that he also reviewed Corps cost 
information for unusual charges and, with the exception of a few 
instances of labor charge discrepancies, most of the time the Corps 
reports did not contain anything surprising. In the few instances 
where the RPM found a discrepancy, he contacted Corps officials, and 
they were able to explain the reason for the discrepancy--for example, 
a problem with the Corps' billing software. The RPM stated that, under 
the interagency agreement with the Corps, he did not review contractor 
invoices or expenditures because the Corps had both the responsibility 
and the expertise necessary to determine whether the contractor 
charges were appropriate, given the assigned work. 

Additionally, EPA officials stated that the residential nature of the 
site necessitated a substantial investment in community relations to 
manage residents' concerns about the contaminated material under their 
homes and the Rustic Mall. As part of these efforts, EPA used such 
tools as flyers, newsletters, resident meetings, and media interviews 
to communicate with concerned citizens.[Footnote 68] According to the 
RPM, managing community relations required the second largest 
commitment of his time, after designing the work. He said that he 
spent a great deal of time working with residents to help them 
understand the situation during the early site investigation stage, 
when it was not clear who was going to need to move out of their homes 
and residents were concerned about their health and property. The RPM 
said that he also worked personally with residents during the design 
and implementation of the remedy to minimize the impact to the 
community and to inform it of any additional actions needed, such as 
excavating contamination across a property line or closing roads. 

According to site documents and a local official, EPA's community 
relations efforts were successful at reducing residents' anxieties. 
For example, in a summary of lessons learned from the cleanup effort, 
site documents indicate that EPA's policy of promptly responding to 
community inquiries and the regular presence of EPA personnel at the 
site helped to establish and preserve a high level of public 
acceptance and trust with the community. Also, a Borough of Manville 
official noted that the continuity provided by having one RPM 
dedicated to the site for the duration of the project was particularly 
helpful in maintaining good communication because it allowed EPA 
officials to know almost all of the residents on a first-name basis 
and encouraged their participation in the cleanup process. For 
example, the RPM stated that he worked closely with residents to 
address their concerns and minimize impacts to the community during 
the excavation of contaminated material and the restoration of 
affected areas of the neighborhood. Similarly, according to the 
Borough of Manville official, EPA and the contractors effectively 
coordinated with town officials to ensure that the cleanup effort went 
smoothly. For example, to minimize disruption, EPA consulted with town 
officials about which roads would be best to use, considering the 
routes and weight limitations of trucks leaving the site. In the 
official's view, EPA's outreach efforts ensured that residents and the 
community as a whole had sufficient information to feel comfortable 
about the cleanup. Consequently, despite the size and scope of the 
cleanup effort, the official could recall very few complaints from 
residents. 

The Corps Selected and Oversaw Contractors' Design and Implementation 
of the Remedy but Had a Limited Role in the Subcontracting Process: 

At the Federal Creosote site, the Corps selected and oversaw private 
contractors' design and implementation of the remedial action; 
however, the Corps was less involved in the subcontracting process. 
Under the 1984 interagency agreement with EPA, the Corps selects and 
oversees private contractors for all design, construction, and other 
related tasks at Superfund sites, in accordance with Corps procedures 
and procurement regulations.[Footnote 69] According to Corps 
officials, the Corps selected a contractor to perform the design for 
the three OUs at the Federal Creosote site from a list of qualified 
vendors and then negotiated a price for the contracts. For 
construction, the Corps selected a prime contractor from a pool of 
eligible contractors under a cost-reimbursement, indefinite-
delivery/indefinite-quantity (IDIQ) contract.[Footnote 70] According 
to EPA and Corps guidance, this system provides more flexible and 
responsive contracting capabilities for Superfund sites, which may 
require a quick response and often lack a sufficiently defined scope 
of work for price negotiation. 

The Corps' prime contractor performed some of the work and 
subcontracted some tasks to other companies. For example, the prime 
contractor excavated contaminated material but awarded subcontracts 
for transportation, treatment, and disposal of the excavated material. 
Other subcontracted services included providing backfill soil and 
landscaping for site restoration, and treating wastewater. To 
subcontract, the prime contractor solicited bids from potential 
vendors and, for smaller subcontracts, provided the Corps with advance 
notification of the award. To award larger subcontracts, the prime 
contractor requested Corps approval. 

To carry out its oversight responsibilities, the Corps monitored 
changes in the scope of the work, contractor progress and costs, and 
work quality. For example, Corps officials stated the following: 

* The Corps had to approve any changes in project scope, such as 
excavating greater quantities of material, or any increases in other 
construction services or materials beyond the amounts originally 
negotiated between the Corps and the prime contractor. According to 
EPA officials, this chain of command helped prevent any unauthorized 
expansion of work at the site.[Footnote 71] 

* To monitor project progress and contractor costs during 
construction, the Corps reviewed prime contractor cost summary reports 
for each phase of the work. These reports contained detailed 
information on contractor costs and work progress, and, according to 
Corps officials, they were updated, reviewed, and corrected if 
necessary on a daily, weekly, and monthly basis.[Footnote 72] For 
example, Corps officials explained that they reviewed the daily 
reports primarily for accuracy and unallowable costs. For weekly and 
monthly reports, the Corps also examined whether the contractor was 
incurring costs more quickly than expected, which could indicate that 
a cost was incorrectly attributed or that a change in project scope 
was necessary (i.e., because particular aspects of the work were more 
costly than anticipated, and, therefore, a scope revision was needed 
to complete planned activities). However, Corps officials commented 
that the contractor data were generally accurate, and that errors were 
infrequent. The officials also said that, during the most active 
periods of the work, they discussed the cost reports and project 
progress, including any potential changes in unit costs, during the 
weekly meetings with the contractor. 

* The Corps also monitored work quality at the site. According to site 
documents, the Corps was required to implement a quality assurance 
plan as part of its oversight responsibilities and had a quality 
assurance representative at the site during construction. For example, 
in a July 2002 notice to the prime contractor, the Corps identified 
several workmanship deficiencies that the contractor had to address to 
retain its contract for that portion of the work.[Footnote 73] 

According to Corps guidance and officials, the Corps had a limited 
role in the subcontracting process at the Federal Creosote site. For 
example, the prime contractor was responsible for selecting and 
overseeing subcontractors. In particular, Corps guidance states that 
since subcontracts are agreements solely between the prime contractor 
and the subcontractor, the Corps does not have the authority to 
enforce the subcontract provisions.[Footnote 74] Rather, the guidance 
indicates that the Corps oversees the prime contractor's management 
systems for awarding and administering subcontracts through periodic 
reviews of the contractor's subcontracting processes and ongoing 
reviews of subcontract awards. According to Corps officials, the 
Corps' main responsibility in the subcontracting process at the 
Federal Creosote site was to review subcontract decisions and approve 
subcontracts above a certain dollar threshold. As Corps officials 
explained, subcontracts between $25,000 and $100,000 did not need to 
be approved by the Corps; rather, the prime contractor sent the Corps 
an "advance notification" package, which documented that the 
contractor had competitively solicited the work and why the contractor 
selected a particular subcontractor over others. However, for 
subcontracts greater than $100,000, the prime contractor had to submit 
a "request for consent" package to the Corps, which contained similar 
documentation as an advance notification but required Corps approval 
prior to awarding a subcontract. 

According to federal acquisition regulations and policies, when 
evaluating request for consent packages, Corps contracting officers 
should consider whether there was sufficient price competition, 
adequate cost or price comparison, and a sound basis for selecting a 
particular subcontractor over others, among other factors. Early in 
the project, the Corps identified several issues with the prime 
contractor's performance at the site, including the award of 
subcontracts. According to a letter the Corps sent to the prime 
contractor, the Corps noted that after repeated unsuccessful attempts 
to address these issues, the Corps would initiate proceedings to 
terminate the contract for site work unless the contractor took 
corrective action. However, Corps officials said the contractor 
demonstrated sufficient improvement in its documentation practices. 
Then, in 2003, the Corps raised the request for consent threshold from 
$100,000 to $500,000 because of the high volume of these packages that 
the Corps was receiving.[Footnote 75] A Corps official noted that 
while the Corps reviews and consents to the subcontracting decisions 
of its contractors as appropriate, it avoids becoming too involved in 
the subcontracting process because of bid protest rules regarding 
agency involvement in that process. According to the official, under 
these rules, a subcontract bidder cannot protest a subcontract award 
unless it can show that the overseeing agency was overly involved in 
the subcontracting process. 

Concerning contractors at the Federal Creosote site, the Department of 
Justice and EPA's Office of Inspector General have ongoing 
investigations, some of which have resulted in allegations of fraud 
committed by employees of the prime contractor and several 
subcontracting firms. For example, court documents alleged bid-
rigging, kickbacks, and other fraudulent activity related to the award 
of several subcontracts for a variety of services and materials. 
According to Corps officials, the Corps did not suspect issues of 
fraud in the subcontracting process until 2004 when, in one instance, 
a subcontract bidder objected to the award of a soil transportation, 
treatment, and disposal subcontract to another firm whose bid was 
substantially higher. Upon further review of the documents, Corps 
officials found that the prime contractor had not conducted a proper 
evaluation of the bid proposals, and the Corps withdrew its consent to 
the subcontract--ultimately requesting that the prime contractor 
solicit bids under a different process.[Footnote 76] In the revised 
bidding process, the firm that had won the earlier subcontract reduced 
its price from $482.50 to $401.00 per ton of contaminated material--
only 70 cents below the competing bid submitted by the firm that had 
protested the original subcontract. On this basis, the prime 
contractor again requested consent to subcontract with the firm to 
which it had awarded the earlier subcontract. According to a Corps 
official, the Corps was suspicious of illegal activity given how close 
the two bids were, and Corps officials discussed whether to take 
formal action against the prime contractor. However, Corps officials 
decided they did not have sufficient evidence of wrongdoing to support 
a serious action but did cooperate with others' investigations of 
fraud at the site. For more information on site-related fraud, see 
appendix I. 

Agency Comments and Our Evaluation: 

We provided a draft of this report to the Secretary of the Army and 
the Administrator of the Environmental Protection Agency for review 
and comment. The Secretary, on behalf of the Corps of Engineers, had 
no comments on the draft report. EPA generally agreed with our 
findings regarding the agency's actions and costs to clean up the 
Federal Creosote site, and provided a number of technical comments, 
which we incorporated as appropriate. EPA's written comments are 
presented in appendix IV. 

In its comments, EPA noted that the draft report accurately described 
the cleanup of the site and correctly compared the site's estimated 
and final remedial construction costs. However, EPA stated that 
comparing estimated remedial construction costs to total site costs is 
not an "apples to apples" comparison because some costs, such as 
amounts spent on removal actions or EPA personnel salaries (referred 
to as "other response costs" in this report), are purposely excluded 
from EPA's early estimates of remedial construction costs. We agree 
that to identify the extent to which site costs increased over agency 
estimates, one should only compare estimated and actual remedial 
construction costs, as we do in table 4 of this report. However, our 
objective was, more broadly, to identify what factors contributed to 
the difference between the estimated remedial construction costs ($105 
million) and the actual total site costs ($338 million). We found that 
the difference between these two amounts was $141 million in remedial 
construction cost increases--which were largely due to increases in 
the amount of contaminated material requiring remediation--and $92 
million in other response costs that were not included in EPA's 
original estimates. We believe it was necessary to provide information 
on these other response costs to more fully answer our objective and 
to provide a more informative accounting of the total costs that EPA 
incurred in cleaning up the Federal Creosote site. 

As agreed with your offices, unless you publicly announce the contents 
of this report earlier, we plan no further distribution of it until 30 
days from the date of this report. At that time, we will send copies 
of this report to the appropriate congressional committees, the 
Secretary of the Army, the Administrator of the Environmental 
Protection Agency, and other interested parties. In addition, the 
report will be available at no charge on GAO's Web site at [hyperlink, 
http://www.gao.gov]. 

If you or your staffs have any questions about this report, please 
contact me at (202) 512-3841 or stephensonj@gao.gov. Contact points 
for our Offices of Congressional Relations and Public Affairs may be 
found on the last page of this report. Key contributors to this report 
are listed in appendix V. 

Signed by: 

John B. Stephenson: 
Director, Natural Resources and Environment: 

[End of section] 

Appendix I: Summary of Criminal and Civil Litigation Related to the 
Federal Creosote Site: 

Court records show that several cases have been brought concerning the 
Federal Creosote site cleanup. First, the Department of Justice 
(Justice) and the state of New Jersey have filed claims to recover 
cleanup costs. Second, Justice has brought criminal charges in a 
series of cases against one employee of the prime contractor, three 
subcontractor companies, and eight associated individuals involved in 
the cleanup, alleging fraud, among other things. Third, the prime 
contractor has brought a civil suit against a former employee alleged 
to have committed fraud and other offenses during his employment as 
well as against associated subcontractors. The information in this 
appendix provides a brief summary of known actions related to the 
Federal Creosote site cleanup. 

Cost Recovery and Natural Resources Damages Claims: 

United States v. Tronox, LLC: The Comprehensive Environmental 
Response, Compensation, and Liability Act of 1980 (CERCLA) provides 
that parties incurring costs to respond to a release or threatened 
release of a hazardous substance may recover such costs from legally 
responsible parties, including persons who owned or operated a site, 
among others. In this regard, the Environmental Protection Agency 
(EPA) identified Tronox, LLC, the successor to the companies that 
owned and operated the Federal Creosote site, and, for 2 years, EPA 
and Tronox participated in alternative dispute resolution concerning 
EPA's cost recovery claims. In August 2008, Justice, on behalf of EPA, 
filed a civil action in the United States District Court for the 
District of New Jersey against Tronox, seeking recovery of costs that 
the government incurred for the Federal Creosote site cleanup. The 
complaint asserted that the government had incurred at least $280 
million in response costs and would incur additional costs. In October 
2008, the New Jersey Department of Environmental Protection and the 
Administrator of the New Jersey Spill Compensation Fund filed suit in 
the same court against Tronox, seeking recovery of costs incurred for 
the site, as well as damages for injury to natural resources--under 
both CERCLA and the New Jersey Spill Compensation Act--and public 
nuisance and trespass claims.[Footnote 77] In December 2008, the 
federal and state cases were consolidated. Tronox has stated its 
intent to vigorously defend against these claims. In early 2009, 
Tronox filed for voluntary Chapter 11 bankruptcy in federal bankruptcy 
court and initiated an adversary proceeding in that court, seeking a 
declaratory judgment on the status of the EPA and New Jersey claims 
with respect to the bankruptcy. Subsequently, both courts entered a 
stipulation filed by both the government plaintiffs and Tronox to stay 
the cost recovery case as well as the adversary proceeding to allow 
the parties to resolve the claims. As of the date of this report, the 
stays remain in effect. 

Criminal Cases Involving Alleged Contractor Fraud at the Site: 

United States v. Stoerr: Norman Stoerr, a former employee of the prime 
contractor at the Federal Creosote site, pled guilty to three counts 
related to his activities as a contracts administrator at the site. 
Court documents alleged that over a 1-year period, the employee 
conspired with others to rig bids for one subcontractor at the site, 
resulting in EPA being charged inflated prices. In addition, the 
documents alleged that over several years, the employee solicited and 
accepted kickbacks from certain subcontractors at the Federal Creosote 
site and another site, and allowed the kickbacks to be fraudulently 
included in subcontract prices that were charged to EPA. To date, 
Stoerr has not been sentenced. 

United States v. McDonald et al: In August 2009, the United States 
indicted Gordon McDonald--a former employee of the prime contractor at 
the Federal Creosote site--as well as representatives of two 
subcontractors who worked at the site, for various counts, including 
kickbacks and fraud. The indictment charged that the prime 
contractor's employee, a project manager, solicited and accepted 
kickbacks from certain subcontractors in exchange for the award of 
site work, and that these kickbacks resulted in EPA being charged an 
inflated price for the subcontractors' work. The indictment also 
charged that the project manager disclosed the bid prices of other 
vendors during the subcontracting process, which resulted in the 
government paying a higher price for services than it would have 
otherwise paid.[Footnote 78] One of the indicted employees (James 
Haas)--representing a subcontractor who provided backfill material to 
the site--has pled guilty of providing kickbacks and submitting a bid 
that was fraudulently inflated by at least $0.50 per ton of material. 
Haas agreed to pay more than $53,000 in restitution to EPA as part of 
his guilty plea, and has been sentenced to serve 33 months in jail and 
to pay a $30,000 criminal fine. McDonald's case is proceeding, and 
charges against a third defendant are still pending. 

United States v. Bennett Environmental, Inc.: Bennett Environmental, 
Inc. (BEI), a subcontractor providing soil treatment and disposal 
services to the Federal Creosote site cleanup, entered a plea 
agreement admitting to one count of fraud conspiracy. Court documents 
alleged that over 2 years, the company paid kickbacks to an employee 
or employees of the prime contractor, in return for receiving 
favorable treatment in the award of subcontracts, and inflated its 
prices charged to EPA. BEI was sentenced to 5 years' probation and 
ordered to pay $1.662 million in restitution to EPA, plus a $1 million 
fine.[Footnote 79] 

United States v. Tejpar: Zul Tejpar, a former employee of BEI, entered 
a plea of guilty to one count of fraud conspiracy. Court documents 
alleged that Tejpar, along with coconspirators, provided kickbacks to 
employees of the prime contractor to influence the award of 
subcontracts at the site and fraudulently inflated the company's bid 
price after an employee of the prime contractor revealed the other bid 
prices. To date, Tejpar is awaiting sentencing. 

United States v. Griffiths: Robert P. Griffiths entered a plea of 
guilty to three counts related to fraudulent activity at the Federal 
Creosote site when he was an officer of BEI. Griffiths, along with 
coconspirators, provided kickbacks to employees of the prime 
contractor to influence the award of subcontracts at the site, 
fraudulently inflated the company's invoices that the prime contractor 
charged to EPA, and fraudulently received the bid prices of other 
bidders prior to award of a subcontract. To date, Griffiths is 
awaiting sentencing. 

United States v. JMJ Environmental, Inc.: JMJ Environmental, Inc., a 
subcontractor providing wastewater treatment supplies and services, 
and John Drimak, Jr., its president, entered guilty pleas related to 
fraudulent activity at the Federal Creosote site and another site. At 
the Federal Creosote site, JMJ Environmental and Drimak, along with 
coconspirators, provided kickbacks to employees of the prime 
contractor to influence the award of subcontracts at the site, 
fraudulently inflated the company's prices that the prime contractor 
charged to EPA, and arranged for intentionally high, noncompetitive 
bids from other vendors. To date, JMJ Environmental and Drimak are 
awaiting sentencing. 

United States v. Tranchina: Christopher Tranchina, an employee of 
subcontractor Ray Angelini, Inc., which provided electrical services 
and supplies, entered a plea of guilty to fraud conspiracy for 
activities at the Federal Creosote site. Tranchina, along with 
coconspirators, provided kickbacks to employees of the prime 
contractor to influence the award of subcontracts at the site and 
fraudulently inflated the company's prices that the prime contractor 
charged to EPA. Tranchina was sentenced to imprisonment of 20 months 
and ordered to pay $154,597 in restitution to EPA. 

United States v. Landgraber: Frederick Landgraber, president of 
subcontractor Elite Landscaping, Inc., entered a plea of guilty to 
fraud conspiracy for activities at the Federal Creosote site. 
Landgraber, along with coconspirators, provided kickbacks to employees 
of the prime contractor to influence the award of subcontracts at the 
site and submitted fraudulent bids from fictitious vendors to give the 
appearance of a competitive process, resulting in EPA paying higher 
prices than if procurement regulations were followed. Landgraber was 
sentenced to imprisonment of 5 months and ordered to pay $35,000 in 
restitution to EPA and a $5,000 fine. 

United States v. Boski: National Industrial Supply, LLC, a pipe supply 
company, and co-owner Victor Boski entered guilty pleas for fraud 
conspiracy at the Federal Creosote site and another site. At the 
Federal Creosote site, National Industrial Supply and Boski, along 
with coconspirators, provided kickbacks to employees of the prime 
contractor to influence the award of subcontracts at the site and 
fraudulently inflated the company's prices that the prime contractor 
charged to EPA. The terms of the plea agreement require National 
Industrial Supply and Boski to have available $60,000 to satisfy any 
restitution or fine imposed by the court, among other items. To date, 
they are awaiting sentencing. 

Prime Contractor Civil Suit: 

Sevenson Environmental Services, Inc. v. McDonald: In 2008, Sevenson 
Environmental Services, Inc., the prime contractor at the Federal 
Creosote site, sued former employee Gordon McDonald for fraud, breach 
of fiduciary duty, breach of contract, and other claims. The suit also 
asserted various claims against other defendants, including McDonald's 
relatives and companies allegedly owned by him,[Footnote 80] and six 
subcontractors and associated individuals.[Footnote 81] The 
allegations related to the Federal Creosote site and another site. 
With respect to the Federal Creosote site, the suit alleged that the 
subcontractors provided unauthorized kickbacks to McDonald, his 
companies, or affiliates. Sevenson also alleged, among other things, 
that the company suffered substantial damage from the fraud 
perpetrated by McDonald and employees of BEI, for example, in the 
company's reliance on fraudulent change orders prepared and 
recommended by McDonald and fraudulent invoices submitted by BEI. 
Sevenson asserted similar allegations of kickbacks and fraudulent 
invoices with respect to other subcontractors. Sevenson sought 
judgment against the defendants, including punitive damages. To date, 
the parties have stipulated to a stay of the case while criminal 
charges against McDonald and others are pending (see the previously 
mentioned cases). 

[End of section] 

Appendix II: Objectives, Scope, and Methodology: 

This appendix provides information on the scope of work and 
methodology used to examine (1) how EPA assessed the risks and 
selected remedies for the Federal Creosote site, and what priority EPA 
assigned to site cleanup; (2) what factors contributed to the 
difference between the estimated and actual remediation costs of the 
site; and (3) how responsibilities for implementing and overseeing the 
site work were divided between EPA and the U.S. Army Corps of 
Engineers (the Corps). It also discusses our methodology for 
summarizing criminal and civil litigation related to the Federal 
Creosote site. 

To examine how EPA assessed the risks and selected remedies for the 
Federal Creosote site, as well as what priority it assigned to the 
cleanup, we reviewed EPA's Superfund site investigation and cleanup 
processes, including applicable statutes, regulations, and agency 
guidance. We also reviewed documentation from the site's 
administrative record, which detailed the agency's activities and 
decisions at the site.[Footnote 82] As part of this review, we 
analyzed public comments that were documented in site records of 
decision to identify key issues with the cleanup effort. To obtain 
additional information on these and other site cleanup issues, we 
interviewed EPA Region 2 officials involved with the site, including 
officials from the Emergency and Remedial Response Division, the 
Public Affairs Division, and the Office of Regional Counsel. 
Furthermore, we interviewed and reviewed documentation obtained from 
officials with the Agency for Toxic Substances and Disease Registry 
regarding its determination of site risks. We also consulted with New 
Jersey and Borough of Manville officials to obtain their views on the 
cleanup effort. Finally, we interviewed representatives of the 
potentially responsible party for the site to obtain the party's views 
on EPA's risk assessment, remedy selection, and site prioritization. 

To determine what factors contributed to the differences between the 
estimated and actual costs of site cleanup, we obtained and analyzed 
data on estimated and actual site costs from several sources.[Footnote 
83] For estimated site costs, we combined EPA's estimates for selected 
remedies from site records of decision and remedial alternative 
evaluations. In developing these estimates, EPA applied a simplifying 
assumption that all construction costs would be incurred in a single 
year, and, therefore, EPA did not discount future construction costs, 
even though work was projected to occur several years into the future 
as a result of design activities and resident relocations as well as 
EPA's estimated construction time frames.[Footnote 84] However, our 
discount rate policy guidance recommends that we apply a discount 
factor to future costs.[Footnote 85] Consequently, to convert EPA's 
estimated costs into fiscal year 2009 dollars, we (1) conducted a 
present value analysis to discount future site costs to the dollar 
year of the original estimate (base year) for each remedy, using EPA's 
recommended discount rate of 7 percent, and (2) converted the present 
value of each estimate into fiscal year 2009 dollars. To calculate the 
present value of estimated costs, we identified the projected 
construction time frames for each remedy from site documents. Because 
the documents did not provide information on how construction costs 
would be distributed over the projected time frame, we calculated the 
midpoint of a range of values, assuming that all costs for particular 
activities comprising EPA's selected remedies would either be incurred 
at the beginning of the projected time frame (the maximum value of 
these costs) or at the end of the projected time frame (the minimum 
value). To adjust the present values from the base year to fiscal year 
2009 constant dollars, we divided the present values by the inflation 
index for the base year and weighted the calculation to convert the 
base year from calendar years to fiscal years. 

To identify actual sitewide costs, we compiled data from multiple 
sources, including EPA's Superfund Cost Recovery Package Imaging and 
On-Line System (SCORPIOS) for data on site costs through April 30, 
2009;[Footnote 86] the Corps of Engineers Financial Management System 
(CEFMS) for data on Corps and contractor costs through various dates 
in April and early May, 2009; and contractor-generated project cost 
summary reports for data on contractor costs for each phase of the 
cleanup through February 15, 2009.[Footnote 87] We relied on multiple 
data sources for our analysis because none of the sources provided a 
sufficient level of specificity for us to comprehensively determine 
when and for what purpose costs were incurred. In particular, the 
SCORPIOS data provided specific dates of when EPA incurred costs but 
for some costs, especially those related to site construction work, 
the data did not generally provide detailed information on why the 
costs were incurred. Therefore, to obtain more detailed information on 
the reason for incurring certain costs, we used the data from CEFMS 
and the contractor's project cost summary reports. However, the CEFMS 
and contractor project cost summary report data did not generally 
provide specific information on when costs were incurred. 
Consequently, to determine actual site costs in fiscal year 2009 
dollars, we used two approaches. For costs taken from the SCORPIOS 
data or when detailed information on the date of a particular cost was 
available, we applied the inflation index for the particular fiscal 
year in which EPA incurred the cost. For costs taken from the other 
data sources, we used the midpoint of the range of inflation-adjusted 
values for the construction start and end dates for individual work 
phases, as recorded in site documents. 

We worked with EPA Region 2 officials to categorize site costs, 
including those that were part of EPA's original construction 
estimates as well as those that were not part of EPA's estimates. 
After identifying the costs that were not included in EPA's original 
estimates, we took the difference between estimated and actual 
construction costs, according to categories that we discussed with 
EPA, to identify where actual costs changed the most from EPA's 
estimates. Then, to identify the factors that contributed the most to 
the difference in these cost categories, we analyzed the types of 
costs in each category and interviewed EPA Region 2 and Corps 
officials responsible for the cleanup. In addition, we analyzed data 
from site documents on the estimated and actual amounts of 
contaminated material at various stages of the cleanup process to 
obtain further information on the extent to which increased amounts of 
contaminated material affected site costs.[Footnote 88] To examine the 
impact of alternative methodologies on the disparity between estimated 
and actual costs, we reviewed EPA cost-estimating guidance and 
calculated the effect of discounting future estimated costs within our 
analysis. To determine how fraud impacted site costs, we reviewed 
civil and criminal litigation documents describing the monetary values 
exchanged in various schemes.[Footnote 89] 

To ensure the reliability of the actual cost data we used for this 
report, we reviewed the data obtained from the SCORPIOS and CEFMS 
databases as well as the contractor-generated cost summary reports 
that the Corps provided. For each of these data sources, we reviewed 
agency documents and interviewed EPA and Corps officials to obtain 
information on their data reliability controls. We also electronically 
reviewed the data and compared them across all sources as well as with 
other information on site costs as available. For example, we compared 
contractor cost data provided by the Corps with similar data from the 
contractor-generated cost summary reports. Similarly, we compared 
Corps cost data from CEFMS with analogous data from EPA's SCORPIOS 
database. Generally, we found that discrepancies among comparable data 
from different sources were most likely attributable to the potential 
delay between when a cost is incurred by a contractor and when it is 
invoiced and processed, first by the Corps and later by EPA. On the 
basis of our evaluation of these sources, we concluded that the data 
we collected and analyzed were sufficiently reliable for our purposes. 
However, because some costs incurred prior to early May 2009 may not 
have been processed through the Corps and EPA's cost-tracking systems 
at the time of data collection, site cost data in this report are 
considered to be approximate. Moreover, because our methodology relied 
on calculating the midpoint of a range of costs for both the present 
value calculations and adjusting data for inflation, we consider the 
data we present in this report on estimated and actual costs and the 
difference between these costs also to be approximate. 

To describe how responsibilities for implementing and overseeing the 
site work were divided between EPA and the Corps, we reviewed the 
interagency agreement between EPA and the Corps for Superfund sites in 
general as well as site-specific agreements for the design and 
implementation of the cleanup at the Federal Creosote site. We also 
reviewed agency guidance regarding EPA's responsibilities at Superfund 
sites. To obtain information on EPA's oversight actions, we 
interviewed EPA and Corps officials responsible for site cleanup and 
contracting work. We also reviewed site meeting minutes, monthly 
progress reports, EPA correspondence to the Corps, and relevant EPA 
Office of Inspector General reports. To further describe the Corps' 
responsibilities at the Federal Creosote site, we reviewed Corps 
guidance for the cleanup of hazardous waste projects, Corps contract 
management best practices, and relevant procurement regulations. To 
obtain information on actions that the Corps took to implement its 
site responsibilities, we reviewed Corps correspondence to the 
contractor and contractor requests for approval of soil treatment and 
disposal subcontracts. We also interviewed Corps officials responsible 
for site cleanup and contracting work as well as EPA Region 2 
officials. However, we did not assess the adequacy of the Corps' 
efforts or its compliance with Corps guidance and federal procurement 
regulations. 

To examine issues regarding civil and criminal litigation related to 
the Federal Creosote site, we collected case data from the Public 
Access to Court Electronic Records system.[Footnote 90] We then 
qualitatively analyzed documents obtained from this system to identify 
the issues involved and the status of each case as well as the 
outcomes, if any, of the cases. However, because criminal 
investigations are ongoing and confidential, we could not determine 
whether any additional criminal charges were under consideration, but 
relied solely on the publicly available information for charges that 
had been filed as of November 2009.: 

We conducted our work from May 2008 through February 2010 in 
accordance with all sections of GAO's Quality Assurance Framework that 
are relevant to our objectives. The framework requires that we plan 
and perform the engagement to obtain sufficient and appropriate 
evidence to meet our stated objectives and to discuss any limitations 
in our work. We believe that the information and data obtained, and 
the analyses conducted, provide a reasonable basis for any findings 
and conclusions in this product. 

[End of section] 

Appendix III: Timeline of EPA Actions at the Federal Creosote Site: 

Figures 8 and 9 provide additional information on the timing of EPA's 
actions at the Federal Creosote site. Specifically, figure 8 shows the 
timing of key EPA actions related to its risk assessment and remedy 
selection at the site, while figure 9 provides additional information 
on the timing of key remedial design and action activities. Both 
figures demonstrate the extent to which EPA's efforts to assess risks 
and select, design, and construct remedies took place concurrently. 

Figure 8: Key Events Related to Risk Assessment and Remedy Selection 
at the Federal Creosote Site: 

[Refer to PDF for image: illustrated graph] 

Key Events: New Jersey Department of Environmental Protection (NJDEP)
identifies a black tarry substance in a leaking sump on-site; 
Start date: [Empty]; 
End date: 4/96. 

Key Events: Borough of Manville responds to report of sinkhole in 
residential area and determines that local drinking water supplies are 
not contaminated from the site; 
Start date: 1/97; 
End date: 2/97. 

Key Events: NJDEP and EPA sample for indoor air contamination in 
residential homes; 
Start date: 4/97; 
End date: 5/97. 

Key Events: Agency for Toxic Substances and Disease Registry (ATSDR) 
develops several health consultations and issues a public health 
assessment for the site; 
Start date: 5/97; 
End date: 9/00. 

Key Events: EPA conducts an initial site investigation; 
Start date: 10/97; 
End date: 11/98. 

Key Events: Responsibility for Federal Creosote site transferred from 
NJDEP to EPA; 
Start date: [Empty]; 
End date: 1/98. 

Key Events: EPA completes a Hazard Ranking System evaluation; 
Start date: 1/98; 
End date: 5/98. 

Key Events: Federal Creosote site proposed for the National Priorities 
List (NPL); 
Start date: [Empty]; 
End date: 7/98. 

Key Events: EPA takes soil samples in the Claremont Development 
Operable Unit 2 (OU2) soils as part of the remedial 
investigation/feasibility study (RI/FS); 
Start date: 12/98; 
End date: 3/99. 

Key Events: Federal Creosote site formally listed on the NPL; 
Start date: [Empty]; 
End date: 1/99. 

Key Events: EPA completes a human health risk assessment for surface 
soil; 
Start date: [Empty]; 
End date: 1/99. 

Key Events: EPA completes an evaluation of remedial technologies for 
OU1 on the basis of the results of the initial investigation; 
Start date: [Empty]; 
End date: 4/99. 

Key Events: EPA takes soil samples in the Rustic Mall (OU3) as part of 
the RI/FS; 
Start date: 8/99; 
End date: 11/99. 

Key Events: EPA takes additional groundwater samples as part of the 
RI/FS; 
Start date: 8/99; 
End date: 11/99. 

Key Events: OU1 record of decision (ROD) issued; 
Start date: [Empty]; 
End date: 9/99. 

Key Events: EPA completes a human health risk assessment and remedial 
technology evaluation for OU2 soils; 
Start date: [Empty]; 
End date: 4/00. 

Key Events: OU2 ROD issued; 
Start date: [Empty]; 
End date: 9/00. 

Key Events: EPA completes a human health risk assessment and remedial 
technology evaluation for OU3 soils; 
Start date: [Empty]; 
End date: 6/01. 

Key Events: EPA completes a human health risk assessment and remedial 
technology evaluation for groundwater[A]; 
Start date: [Empty]; 
End date: 7/01. 

Key Events: EPA completes an addendum to the groundwater evaluation 
determining that treating groundwater at the site would be technically 
impracticable; 
Start date: [Empty]; 
End date: 6/02. 

Key Events: OU3 ROD issued; 
Start date: [Empty]; 
End date: 9/02. 

Source: GAO analysis of EPA documents. 

[A] The groundwater human health risk assessment and remedial 
technology evaluation was completed in two parts. First, in July 2001, 
EPA issued a remedial investigation report for groundwater that 
contained the human health risk assessment. Second, in August 2001, 
EPA issued a feasibility study report that included the remedial 
technology evaluation for groundwater. 

[End of figure] 

Figure 9: Key Events Related to Remedial Design and Action at the 
Federal Creosote Site: 

[Refer to PDF for image: illustrated graph] 

Activities for the Claremont Development: 

Key Events: EPA conducts a removal action at certain residential 
properties with unacceptable risks from surface soil contamination; 
Start date: 7/98; 
End date: 9/998. 

Key Events: OU1 Remedial design, Phase 1; 
Start date: 9/99; 
End date: 10/00. 

Key Events: OU1 Remedial design, Phase 2; 
Start date: 9/99; 
End date: 3/02. 

Key Events: OU1 Remedial design, Phase 3; 
Start date: 9/99; 
End date: 1/03. 

Key Events: Relocation efforts begin for residents in residential
properties identified for demolition; 
Start date: 10/99; 
End date: [Empty]. 

Key Events: OU1 Remedial action, Phase 1; 
Start date: 10/00; 
End date: 6/02. 

Key Events: OU1 Remedial action, Phase 2[A]; 
Start date: 6/02; 
End date: 1/08. 

Key Events: OU1 Remedial action, Phase 3; 
Start date: 4/03; 
End date: 4/06. 

Key Events: OU2 Remedial design, Phase 1; 
Start date: 9/00; 
End date: 11/01. 

Key Events: OU2 Remedial design, Phase 2; 
Start date: 9/00; 
End date: 2/03. 

Key Events: OU2 Remedial action, Phase 1; 
Start date: 2/02; 
End date: 6/02. 

Key Events: OU2 Remedial action, Phase 2[B]; 
Start date: 8/01; 
End date: 5/06. 

Activities for the Rustic Mall: 

Key Events: OU3 Remedial design; 
Start date: 9/02; 
End date: 7/07. 

Key Events: OU3 Remedial action; 
Start date: 6/05; 
End date: 2/08. 

Key Events: EPA approval for treated waste to be disposed of
at a nonhazardous waste facility; 
Start date: [Empty]; 
End date: 11/02. 

Source: GAO analysis of EPA documents. 

[A] According to EPA, the OU1 Phase 2 remedial action was completed in 
April 2004. However, this area was used as a staging area for waste 
from OU1 Phase 3, OU2 Phase 2, and OU3 properties. As a result, 
restoration of the OU1 Phase 2 area could not be completed until all 
excavation and shipping of OU3 waste was complete (since OU3 was the 
last completed area for which the OU1 Phase 2 area was used to stage 
waste). 

[B] According to EPA, the OU2 Phase 2 remedial action occurred on an 
episodic basis. The initial remedial action that was part of OU2 Phase 
2 was a remedial action at the day-care center playground that began 
and ended in August 2001. Remedial action on OU2 Phase 2 residential 
properties was not begun until almost 2 years later and was completed 
in June 2005. Also, because the day-care center parking lot was 
connected to the Rustic Mall, the remedial action at the parking lot 
was completed, together with the OU3 remedial action, between November 
2005 and May 2006. 

[End of figure] 

[End of section] 

Appendix IV: Comments from the Environmental Protection Agency: 

United States Environmental Protection Agency: 
Region 2: 
220 Broadway: 
New York, NY 10007-1868: 

January 29, 2010: 

John B. Stephenson: 
Director: 
Natural Resource & Environment: 
United States Government Accountability Office: 
441 G Street, NW, Room 2075: 
Washington, D.C. 20548: 

Dear Mr. Stephenson: 

Thank you for the opportunity to review and comment on the draft 
report of the Government Accountability Office (GAO) entitled 
Information on Cost and Other Issues Related to the Cleanup of the 
Federal Creosote Site. Consistent with our comments on the Statement 
of Facts, the Environmental Protection Agency (EPA) agrees that the 
GAO report accurately reflects the events that describe the cleanup of 
the Federal Creosote site, located in Manville, New Jersey, from early 
removal activities and placement on the National Priorities List 
(NPL), through the selection and implementation of a series of 
remedial actions to address the nature and extent of contamination at 
the site. In the Federal Creosote cleanup, EPA addressed the remnants 
of a long-abandoned industrial complex found beneath a residential 
community of 137 single-family homes. Meeting a residential threshold 
for cleanup, and doing so while limiting the disruption to the fabric 
of the neighborhood, were two of the principal challenges for EPA 
while implementing this action, which is now completed and the site is 
nearing deletion from the NPL. 

EPA is constantly striving to improve the Superfund program, and 
welcomed GAO's review as an opportunity for an independent check of 
our methods. EPA is pleased to find that, after GAO's thorough 
evaluation of this project, it has no major recommendations on how we 
might have improved this cleanup effort. 

The GAO was asked to look into the difference between the total 
construction cost estimates EPA made prior to 2002, $105 million, and 
the final total cost of 2009, $338 million, a sum EPA is seeking to 
recover from the parties that are responsible for the contamination at 
the Federal Creosote site. The vast bulk of the difference between 
these figures results from two factors. 

First, the amount of highly contaminated soil which was actually found 
as EPA worked at the site was nearly three times the amount used when 
EPA made its original construction cost estimates. In fact, we found 
456,600 tons of contaminated soil that we had to excavate, remove, 
treat and/or dispose of, and replace, rather than the 164,400 tons on 
which the early estimates were based. This increase in quantity of 
contaminated material itself accounts for $112 million in added costs. 
When that amount is added to the original construction cost estimate, 
one gets a final total construction cost of $217 million. Another $29 
million in increased construction costs came from the added time that 
was needed to complete the cleanup, increases for commodities like 
fuel oil, and additional costs for residential relocation. This total, 
$246 million, is the figure which should be used to make an apples-to-
apples comparison of estimated and final costs. The difference is 
largely explained by the sheer increase in volume of material we dealt 
with and the costs of inflation. We are pleased to note that your 
draft report makes exactly this comparison, the correct one, in Table 
4 on page 33. 

Second, a very large portion of the $338 million final total cost 
figure is composed of categories of costs which were not — and never 
are — included in early cost estimates. Specifically, white early cost 
estimates, which are made solely for the purpose of comparing the 
costs of different potential remedies, include only projected 
construction costs, the final total includes the money EPA spent on a 
removal action at the site, on investigative work, on personnel 
salaries, on indirect (or overhead) costs, and interest, as provided 
by law, on the money the government advanced for the cleanup. These 
other nonconstruction costs total $92 million. GAO was asked to add 
them to the final total construction costs and to compare the total 
with initial construction cost estimates. Unfortunately, that 
comparison is not particularly useful as it is a comparison of apples 
to oranges, not, as explained above, an apples-to-apples comparison. 

Thank you again for the opportunity to comment on this draft report. 
EPA reiterates its commitment to improving the implementation of the 
Superfund program. If you have any questions relating to this 
response, please call Region 2's GAO liaison, John Svec, at (212) 637-
3699. 

Sincerely, 

Signed by: 

Judith A. Enck: 
Regional Administrator: 

[End of section] 

Appendix V: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

John B. Stephenson, (202) 512-3841 or stephensonj@gao.gov: 

Staff Acknowledgments: 

In addition to the individual named above, Vincent P. Price, Assistant 
Director; Carmen Donohue; Maura Hardy; Christopher Murray; Ira Nichols-
Barrer; and Lisa Van Arsdale made key contributions to this report. 
Elizabeth Beardsley, Nancy Crothers, Alexandra Dew, Richard Johnson, 
and Anne Stevens also made important contributions. 

[End of section] 

Footnotes: 

[1] This figure represents the total of EPA's estimated costs for 
remedial construction activities conducted through the spring of 2009 
to address contamination at different areas of the site, which we 
discounted using a present value analysis and adjusted to fiscal year 
2009 constant dollars. As a result of the methodology we used to 
adjust these costs, this figure does not match EPA's published 
estimates for completed and remaining work at the site, which totaled 
about $118 million in nominal dollars. As per GAO internal guidance, 
we are reporting all costs in fiscal year 2009 constant dollars, 
unless otherwise noted. 

[2] We developed these data on actual site costs using data provided 
by EPA and the Corps. As a result of the methodology we used to adjust 
the data to fiscal year 2009 constant dollars, these figures do not 
match other EPA-published information on site costs. 

[3] Justice files claims on behalf of the federal government, 
including agencies such as EPA. 

[4] In addition, a site contractor that was tasked with implementing 
the cleanup also filed a civil litigation action against several of 
these parties asserting claims related to alleged fraud. 

[5] EPA also can take removal actions at any time; these are generally 
short-term or emergency actions to mitigate immediate threats. 

[6] Sites are often divided into OUs by geography, pathways of 
contamination (e.g., groundwater), or type of remedy. 

[7] As part of the OU3 groundwater investigation, EPA also looked at 
off-site groundwater, surface water, and sediment contamination that 
may have been related to the site. 

[8] EPA Region 2, which covers New Jersey, New York, Puerto Rico, the 
U.S. Virgin Islands, and seven federally recognized Indian tribes, was 
responsible for the Federal Creosote site. 

[9] According to EPA guidance, the RI/FS process is a dynamic, 
flexible process that can be tailored to the circumstances of 
individual sites. The objective of the RI/FS is not to remove all 
uncertainties associated with the site cleanup but, rather, to gather 
sufficient information to support informed decision making regarding 
which remedy appears to be the most appropriate for a site. 

[10] To assess site risks during the remedial investigation, EPA 
generally follows a four-step process: (1) contaminant identification-
-narrowing down the list of contaminants for risk analysis on the 
basis of which contaminants present the biggest threats to human 
health; (2) exposure assessment--identifying current and future 
pathways for how individuals could be exposed to contamination on the 
basis of how the site is or could be used; (3) toxicity assessment--
examining the potential health effects of contaminants on the basis of 
their toxicity; and (4) risk characterization--estimating the 
potential risks from exposure to contaminants. EPA's acceptable levels 
for carcinogenic human health risks are based on a range of values--
between 1 in 10,000 individuals and 1 in 1 million individuals--that 
represent the excess probability a person will develop cancer over a 
lifetime as a result of exposure to contamination. If human health 
risks are greater than 1 in 10,000, they are considered to be above 
EPA's acceptable range. For noncarcinogenic risks, EPA calculates a 
hazard index value that sums the values for individual noncarcinogenic 
effects from site contaminants. If the hazard index exceeds a value of 
1, EPA assumes there is a potential for noncarcinogenic health effects 
to occur from exposure to site contamination. 

[11] Generally, final remediation goals are determined when a remedy 
is selected. 

[12] These criteria were developed to incorporate the remedy selection 
requirements and preferences of section 121 of CERCLA, which mandates, 
among other requirements, that remedial actions be protective of human 
health and the environment and be cost-effective. CERCLA also 
indicates a preference for actions that reduce the toxicity, mobility, 
or volume of contamination through treatment. 

[13] EPA referred to the 1984 interagency agreement with the Corps as 
an "umbrella" agreement. If EPA chooses to assign work to the Corps, 
EPA establishes site-specific interagency agreements that document the 
type of work to be performed and an initial budget. 

[14] Although these efforts were largely conducted by EPA contractors, 
the agency provided direction and oversight to the contractors, and, 
therefore, we refer to them as EPA efforts. 

[15] We excluded EPA's Hazard Ranking System evaluation results from 
our discussion of EPA's risk assessment because EPA officials said 
that the risk assessment process is different from the Hazard Ranking 
System evaluation that is conducted as part of the NPL listing 
process. The officials said that the goal of the listing process is to 
determine whether a site requires additional evaluation, and that a 
site's Hazard Ranking System evaluation score does not relate to EPA's 
human health risk assessment. 

[16] EPA found that 19 properties had high levels of surface soil 
contamination. To limit potential exposures at certain properties, in 
July 1998, EPA took removal actions, such as applying topsoil or other 
materials. 

[17] At the Federal Creosote site, EPA initially began evaluating 
whether to address site contamination under its removal program and, 
therefore, started preparing an Engineering Evaluation/Cost Analysis, 
which is required under the National Contingency Plan to support 
certain types of removal actions. The National Oil and Hazardous 
Substances Pollution Contingency Plan, referred to as the National 
Contingency Plan, is published in the Federal Register and can be 
found in the Code of Federal Regulations at 40 CFR Part 300. This 
plan, which was revised pursuant to CERCLA, establishes the procedures 
and standards for responding to releases of hazardous substances. 
According to EPA guidance (see EPA, Guidance on Conducting Non-Time-
Critical Removal Actions Under CERCLA, EPA 540-R-93-057 (Washington, 
D.C.: August 1993)), the goals of an Engineering Evaluation/Cost 
Analysis are to identify the objectives of a removal action and to 
evaluate various alternatives that may be used to satisfy these 
objectives on the basis of their cost, effectiveness, and 
implementability. EPA's guidance also states that an Engineering 
Evaluation/Cost Analysis, although less comprehensive, is similar to 
the RI/FS conducted for a remedial action. 

[18] To evaluate OU2 soil contamination, EPA took approximately 2,000 
soil samples, among other activities. EPA's investigation of OU3 soil 
contamination included taking almost 250 soil samples, while its 
groundwater investigation included taking samples from monitoring 
wells and collecting data on the movement of groundwater contamination 
in the soil and bedrock. 

[19] EPA officials said it would have been labor-intensive to assess 
the risks for each residential property because of the large number of 
properties and the amount of data and calculations involved. Also, 
according to EPA, an equally important consideration was the time 
involved in assessing the risks for each individual property, since 
this would have potentially delayed remediation of the site and 
prolonged residents' exposure to contamination. 

[20] To calculate risk levels, EPA broke OU3 into six areas, primarily 
on the basis of where higher levels of contamination were present 
within 10 feet of the ground surface. 

[21] According to EPA, these two rivers are a source of drinking water 
and the intake for the public water supply was approximately 2,700 
feet away from the site. 

[22] Pursuant to CERCLA requirements, ATSDR performs health 
assessments at all sites proposed for the NPL. ATSDR looks at sites 
from a public health and medical needs perspective and provides 
individuals with medical assistance to respond to conditions 
identified as stemming from exposure to site contamination. 

[23] ATSDR officials said the assessment pulled together the various 
health consultations that it had developed. 

[24] Both ATSDR and EPA officials also noted other differences between 
the two agencies' risk assessment processes, such as different 
underlying data and assumptions, which could yield different results. 

[25] ATSDR found that groundwater contamination posed a future hazard 
if public water supplies became affected; however, EPA's groundwater 
investigation was still ongoing at the time of ATSDR's assessment. 

[26] EPA's remediation goal for OU1 was based on visible levels of 
contamination. However, according to site documents and Region 2 
officials, the agency did not plan to begin excavation in OU1 areas 
until it had developed quantitative remediation goals following its 
assessment of the OU2 soil contamination. 

[27] However, EPA found that residually contaminated soil could remain 
at depths greater than 10 to 14 feet. 

[28] A contractor's study of site contamination found that applying 
nonresidential rather than residential cleanup standards in the Rustic 
Mall area would have reduced the amount of material requiring 
remediation by only about 1.4 percent of the estimated total. 

[29] The primary guidance that EPA relied on to identify potential 
remedial technologies for the soil contamination was the agency's 
guidance entitled Presumptive Remedies for Soils, Sediments, and 
Sludges at Wood Treater Sites (December 1995). EPA developed this 
guidance as part of an initiative to speed up remedy selection and 
reduce the cost and time required to clean up similar sites, among 
other purposes. Region 2 officials said that a goal of the initiative 
was to avoid conducting treatability studies to assess innovative or 
emerging technologies--technologies that have been implemented on a 
small scale and show promise for being able to remediate similar types 
of contamination--if EPA already had proven technologies. 

[30] As with efforts to assess the risks of contamination at the site, 
EPA contractors did much of the work to identify and evaluate remedial 
technologies and alternatives. However, EPA provided direction and 
oversight to the contractors, and decided which remedial alternatives 
to select for the site. Therefore, we refer to all of these activities 
as EPA efforts. 

[31] Bioremediation may be performed either in situ (in place in the 
ground) or ex situ (after material is excavated). 

[32] Region 2 officials said that innovative or emerging technologies 
were screened out because EPA found they would not be a good match for 
the site, not simply because they required a treatability study to 
assess their viability. 

[33] EPA noted that all remedial alternatives it considered, with the 
exception of a no-action alternative, would have required that the 
agency relocate residents. 

[34] EPA treated contaminated groundwater in the soil that it 
encountered while excavating the soil contamination. 

[35] Under the National Contingency Plan, EPA is allowed to select an 
alternative that does not comply with requirements applicable to a 
site, such as meeting groundwater guidelines, if the agency determines 
that certain circumstances exist. One such circumstance is when the 
agency determines that compliance with the requirement would be 
technically impracticable from an engineering perspective. 

[36] EPA, Guidance for Evaluating the Technical Impracticability of 
Ground-Water Restoration, Office of Solid Waste and Emergency Response 
Directive 9234.2-25 (Washington, D.C.: September 1993). 

[37] One area of the mall was used as a staging area to support site 
work, including EPA and contractor offices; a soil staging area to 
store excavated material; and an area for backfill soil, equipment, 
and other required materials. The need to have these facilities at the 
site contributed to the lack of open space. 

[38] Region 2 officials also stated that EPA considered placing a 
treatment facility on a nearby Superfund site. However, they said it 
would have been very difficult to get the required state and local 
approvals. 

[39] According to EPA, agency guidance indicates a preference for 
actions that address contamination risks using methods of cleanup that 
allow people to remain safely in their homes and businesses. See EPA, 
Interim Policy on the Use of Permanent Relocations as Part of 
Superfund Remedial Actions, Office of Solid Waste and Emergency 
Response Directive 9355.0-71P (Washington, D.C.: June 1999). 

[40] Region 2 officials said that EPA attempted to reach out to 
identify permitted vendors who could perform off-site LTTD, but none 
of the LTTD vendors responded to bid requests for site work. 
Similarly, EPA learned of one commercial facility that may have been 
able to treat site soils using bioremediation; however, the company 
did not respond to bid requests for site work. 

[41] EPA had similar concerns about using chemical grouting to 
immobilize contaminants. 

[42] EPA found that contamination from wood-preserving operations is a 
"listed waste" under the Resource Conservation and Recovery Act, which 
gives EPA the authority to control hazardous waste from the "cradle-to-
grave," including the generation, transportation, treatment, storage, 
and disposal of hazardous waste. As a listed waste under the act, EPA 
could not dispose of contaminated material without ensuring that it 
had been treated to specified standards. 

[43] Region 2 officials said that EPA placed an LTTD unit in a 
residential area at another site. However, at this other site, the 
officials noted that none of the residents' houses were within 200 
feet of the unit, and there was enough space to build a large earthen 
berm to lessen the unit's impacts on the community. 

[44] According to EPA guidance, remedial alternatives should be 
consistent with reasonably anticipated future land use. See EPA, Land 
Use in the CERCLA Remedy Selection Process, Office of Solid Waste and 
Emergency Response Directive 9355.7-04 (Washington, D.C.: May 1995). 

[45] Region 2 officials said the state of New Jersey was also opposed 
to on-site treatment and disposal options. The Region 2 officials said 
that while EPA occasionally signs a ROD without state concurrence, the 
agency generally tries to avoid doing so because, under CERCLA, EPA 
cannot carry out a remedial action without the relevant state's 
agreement to provide a 10 percent cost share. Also, the officials 
noted that community acceptance of EPA's remedial action is one of the 
selection criteria, and the agency prefers not to store large amounts 
of contamination for treatment near a residential area. 

[46] The officials emphasized that the OU1 remedial action was an 
early interim action and that EPA did not begin the excavation until 
the OU2 ROD was issued, which set quantitative remediation goals for 
the cleanup of the Claremont Development. The officials stated that 
when EPA's National Remedy Review Board reviewed the OU1 decision and 
recommended that Region 2 complete its sitewide RI/FS before beginning 
source material excavation, the board meant the investigation of the 
residential area and not that Region 2 should complete its RI/FS of 
OU3 before beginning excavation of OU1. Region 2 agreed with this 
recommendation. 

[47] Region 2 officials said that EPA did not have as much data for 
the Rustic Mall but maximized resources by coordinating RI/FS efforts 
for this area with design activities for other parts of the site. 
Additionally, while EPA had signed the ROD and designed the excavation 
for OU3 by the time that the owner of the mall decided to demolish it, 
EPA revised the design concurrently with the excavation rather than 
delay work. 

[48] GAO, Superfund: Litigation Has Decreased and EPA Needs Better 
Information on Site Cleanup and Cost Issues to Estimate Future Program 
Funding Requirements, [hyperlink, 
http://www.gao.gov/products/GAO-09-656] (Washington, D.C.: July 15, 
2009). 

[49] In our July 2009 report [hyperlink, 
http://www.gao.gov/products/GAO-09-656], we raised concerns about the 
usefulness of EPA's "construction complete" measure as an indicator of 
the status of site cleanups because, for example, sites with 
groundwater contamination (as at the Federal Creosote site) may take 
decades to reach selected standards, even though EPA declares the site 
construction complete. By presenting statistics in this report that 
compare the median length of time from NPL listing to construction 
complete for the Federal Creosote site with other megasites, we mean 
only to contrast the length of time it took to reach key agency 
milestones and not to compare the status of the site cleanup with 
EPA's remediation goals. 

[50] We developed these data on actual site costs using data provided 
by EPA and the Corps. As a result of the methodology we used to adjust 
the data to fiscal year 2009 constant dollars, these figures do not 
match other EPA-published information on site costs. 

[51] This figure represents the total of EPA's estimated costs for 
remedial construction activities conducted through the spring of 2009, 
to address contamination at different areas of the site, which we 
discounted using a present value analysis and adjusted to fiscal year 
2009 constant dollars. To identify the estimated costs of activities 
that had been completed as of the spring of 2009, we added EPA's 
estimates for the construction costs of these activities for each OU 
and, following GAO guidance, applied a discount factor to reflect that 
EPA's estimates pertained to future costs. In this report, when we 
discuss "estimated costs," we refer to GAO's revised estimated cost 
figure. As a result of the methodology we used to adjust these 
estimates, our calculated total does not match EPA's published 
estimates for completed and remaining work at the site, which totaled 
$118 million in nominal dollars. Unless otherwise noted, we adjusted 
all dollar values presented in this report to fiscal year 2009 dollars 
as per GAO guidance. See appendix II for more information on our 
methodology for estimating site costs. 

[52] We recognize that total site costs, which include other response 
costs, will always be different from remedial construction costs. 
However, for the purposes of this report, we determined that it was 
necessary to report total site costs as a point of departure to 
discuss how construction costs, as well as other response costs that 
were not included in EPA's original construction cost estimates, 
contributed to the total amount EPA spent on the site. 

[53] "Other response costs" refer to costs associated with the site 
that were not directly related to construction of a particular remedy 
(discussed in greater detail in the next section of this report). 

[54] EPA, A Guide to Developing and Documenting Cost Estimates During 
the Feasibility Study, Office of Solid Waste and Emergency Response 
Directive 9355.0-75 (Washington, D.C.: July 2000). 

[55] According to EPA guidance, capital costs include all labor, 
equipment, and material costs associated with construction activities 
as well as costs for professional/technical services, such as 
engineering support for the construction of the remedial action. 
Operation and maintenance costs are those postconstruction costs 
necessary to ensure or verify the continued effectiveness of a 
remedial action. 

[56] Our analysis of site documents found that additional sampling 
during the design stage was generally successful in refining the 
estimated quantity of material requiring excavation across all OUs. 
Specifically, final excavation amounts increased by only about 28 
percent over estimates at the design stage as compared with at least 
178 percent over ROD estimates. 

[57] According to site documents, EPA suspected there was 
contamination underneath the Rustic Mall buildings on the basis of 
sampling in the areas surrounding the mall. However, because EPA did 
not have access beneath the mall buildings for sampling and EPA's 
original quantity estimates presumed the buildings would remain 
standing, contamination under the buildings was not included in the 
original estimates. 

[58] According to EPA, residents at 3 properties refused EPA access to 
conduct sampling during the pre-ROD site investigation. EPA later 
gained access to the properties and found that 2 of the 3 properties 
required soil remediation. 

[59] In 2006, EPA issued an Explanation of Significant Differences for 
the site, which documented quantity increases and their associated 
cost increases to date across all three OUs. According to EPA 
officials, they waited to issue this document until after the 
completion of the OU3 design phase so that they could understand the 
differences across the entire site. Generally, EPA does not revisit 
its remedy selection decision unless new information indicates that 
the selected remedy is technically infeasible or is not sufficiently 
protective to meet cleanup goals, or that an alternative approach 
would be equally protective and more cost-effective. According to EPA 
officials, they did not find anything to indicate the agency should 
change the overall cleanup approach at the site. 

[60] Although the quantities of soil and contaminated material 
increased above EPA's estimates, our analysis found that the average 
unit costs that EPA paid for thermal treatment and disposal of 
contaminated material--the most expensive remediation option--were 
lower than originally estimated and decreased over time. For example, 
the estimated unit cost for these services ranged from $510 for OU1 to 
$650 for OU2 and OU3; however, actual average unit costs were $448 per 
ton for OU1, $426 per ton for OU2, and $354 per ton for OU3. We did 
not adjust these costs to 2009 dollars to more accurately reflect the 
actual bid prices at the time of construction. 

[61] Additionally, according to site documents, EPA did not excavate 
contaminated soil at the site if EPA determined that it did not pose a 
direct threat of human exposure. Specifically, although EPA removed 
source material that could pose a threat of groundwater contamination, 
EPA allowed residually contaminated soil to be left in some areas of 
the site at levels deeper than approximately 14 feet below the ground 
surface. According to site documents, EPA and NJDEP used deed notices 
for properties in these areas to document the remaining contamination 
and restrict excavation. 

[62] Region 2 officials said it was important to prevent a work 
stoppage because it would lead to additional contractor demobilization 
and remobilization costs and would negatively impact community 
relations. 

[63] This assumption can be modified if construction is projected to 
last over a longer period of time. 

[64] Discounting future costs to their present value accounts for the 
time value of money. In particular, having one dollar today is worth 
more than having one dollar a year from now because if it were 
invested, it could earn a return from interest. EPA's estimated 
implementation periods for site construction at the ROD stage ranged 
from 21 months to 3.5 years. However, EPA did not expect to begin 
implementing remedial actions until at least 1 year after it selected 
its remedies because of, for example, design activities or resident 
relocations. Therefore, while EPA's estimated construction periods 
were relatively short, EPA did not plan to incur construction costs 
until further into the future than is indicated simply by its 
estimated time frames. Consequently, discounting these future 
implementation costs had a relatively greater effect on our analysis 
of EPA's estimated site costs than if the costs were projected to 
occur immediately after EPA selected its remedies. 

[65] GAO, Discount Rate Policy, [hyperlink, 
http://www.gao.gov/products/GAO/OCE-17.1.1] (Washington, D.C.: May 
1991). 

[66] Design specifications provide detailed instructions to the 
contractor about how to perform the work. See appendix III for a 
timeline of key remedial design and action events at the Federal 
Creosote site. 

[67] Although concerns were identified by a 2007 EPA Office of 
Inspector General report that evaluated EPA's management of Superfund 
interagency agreements at the Federal Creosote site and certain other 
sites, officials with EPA's Office of Inspector General indicated that 
key findings did not necessarily involve EPA's oversight at the 
Federal Creosote site. According to these officials, the only issue 
related to the Federal Creosote site was a lack of detail in the 
monthly progress reports that the Corps submitted to EPA. However, the 
officials stated that, because of the RPM's on-site presence and level 
of involvement at the Federal Creosote site, they believed that the 
RPM had a good understanding of the work that was occurring at the 
site. 

[68] In addition to the RPM, EPA also assigned a community relations 
specialist to the site. 

[69] Under the 1984 interagency agreement, the Corps is also 
responsible for providing (1) technical assistance to EPA during the 
RI/FS phase and (2) site-specific cost documentation to support EPA's 
cost recovery efforts. 

[70] A cost-reimbursement contract is one in which the contractor 
receives payments for the amount of allowable costs incurred to the 
extent prescribed in the contract rather than a prenegotiated fixed 
amount, while an IDIQ contract is one in which orders for goods or 
services are placed against an established contract. The IDIQ 
contracts used at the Federal Creosote site were "pre-placed remedial 
action contracts" (PRAC). According to Corps officials, the 
construction at the site was performed under four PRACs--three for the 
prime contractor and one for the demolition contractor. 

[71] Although the Corps directed the prime contractor, Corps officials 
also noted that the RPM was heavily involved at the beginning of each 
work phase and established a precedent for when decisions regarding 
additional excavation could be made by the Corps, and for when the RPM 
needed to be consulted first. 

[72] Corps officials said that EPA specifically requested that the 
Corps track contractor charges daily because of the large scale of the 
project and of EPA's concern about maintaining forward progress due to 
the residential nature of the site. 

[73] In addition to overseeing the prime contractor charged with 
construction, site documents showed that the Corps was also required 
to oversee the design contractor and provide written reports to EPA on 
a monthly basis on changes in the scope of work, project progress, and 
costs. Additionally, the Corps was responsible for ensuring that the 
prime contractor's work was in accordance with EPA guidance and 
policies. 

[74] Although the Corps is not a party to the contract between the 
prime contractor and the subcontractor, the Corps requires that the 
prime contractor include certain contract clauses in its subcontracts 
to ensure that the subcontract follows the intent of federal 
acquisition regulations and policies. 

[75] The advance notification threshold was also increased from 
$25,000 to $100,000 at this time, according to Corps officials. Corps 
officials said that raising the notification and consent thresholds is 
not unusual, and it had been done before at other sites. 

[76] The prime contractor had solicited bids under a "Request for 
Proposals" process, which, according to a site document, allowed it to 
select the subcontractor that provided the "best value," even if it 
involved paying a higher price. However, the Corps found that the 
prime contractor did not adequately justify why a higher price should 
be paid for the subcontractor's services and requested that the prime 
contractor solicit bids using a sealed bid process, in which only a 
price per ton and a determination of a subcontractor's ability to 
perform the work were considered in the selection process. 

[77] New Jersey's related claims in state court, originally filed in 
2007, were dismissed subsequent to the filing in federal court. 

[78] The indictment alleged that these acts took place on several 
occasions and at more than one Superfund site. 

[79] Courts may impose probation on a corporation. This probation may 
involve various activities intended to prevent future wrongdoing and 
often involves court-appointed oversight. 

[80] These defendants include GMEC, Inc.; Patricia McDonald; Flowers 
By Sweetens, Inc.; Matthew McDonald; Kevin McDonald; and Thomas 
McDonald. 

[81] These defendants include BEI; John Bennett; Robert Griffiths; DCP 
Technical Services; JMJ Environmental, Inc.; John Drimak, Jr.; Elite 
Landscaping, Inc.; Frederick A. Landgraber; Ray Angelini, Inc.; Ray 
Angelini; Christopher Tranchina; National Industrial Supply, LLC; and 
Victor Boski. 

[82] We did not independently verify the accuracy of the information 
contained in these documents. 

[83] We focused on the estimated and actual costs of activities that 
had already been performed as of the spring of 2009, and excluded the 
costs of future activities at the site, such as groundwater monitoring. 

[84] Because EPA did not discount future construction costs when 
calculating its estimates, EPA's estimated costs were actually higher 
than they would have been otherwise. 

[85] GAO, Discount Rate Policy, GAO/OCE 17.1.1 (Washington, D.C.: May 
1991). 

[86] The SCORPIOS data collected from EPA included all costs that the 
agency identified as specifically related to the Federal Creosote 
site. According to an EPA official, some early investigation work was 
performed by a contractor that was working at multiple sites, and EPA 
was unable to identify how much of the contractor's costs were 
attributable to the Federal Creosote site. As a result, costs for this 
work were not included in the SCORPIOS data that EPA provided. 

[87] According to site documents, remedial construction concluded in 
February 2008; however, EPA and Corps officials noted that costs for 
some site maintenance and administrative activities continued to 
accrue. 

[88] We did not independently verify the information on soil 
quantities within the site documents. 

[89] Because some cases involved fraud at multiple sites and 
investigations are ongoing, we could not determine the exact extent to 
which fraud affected cleanup costs at the Federal Creosote site. 

[90] This system provides access to documents related to cases filed 
in federal courts and is operated by the Administrative Office of the 
U.S. Courts, which is the central support agency for the Judicial 
Branch. 

[91] We did not evaluate the accuracy of the information obtained from 
court documents. 

[End of section] 

GAO's Mission: 

The Government Accountability Office, the audit, evaluation and 
investigative arm of Congress, exists to support Congress in meeting 
its constitutional responsibilities and to help improve the performance 
and accountability of the federal government for the American people. 
GAO examines the use of public funds; evaluates federal programs and 
policies; and provides analyses, recommendations, and other assistance 
to help Congress make informed oversight, policy, and funding 
decisions. GAO's commitment to good government is reflected in its core 
values of accountability, integrity, and reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each 
weekday, GAO posts newly released reports, testimony, and 
correspondence on its Web site. To have GAO e-mail you a list of newly 
posted products every afternoon, go to [hyperlink, http://www.gao.gov] 
and select "E-mail Updates." 

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s Web site, 
[hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional 
information. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 

Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: 
E-mail: fraudnet@gao.gov: 
Automated answering system: (800) 424-5454 or (202) 512-7470: 

Congressional Relations: 

Ralph Dawn, Managing Director, dawnr@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, D.C. 20548: 

Public Affairs: 

Chuck Young, Managing Director, youngc1@gao.gov: 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, D.C. 20548: