<
33
%
4^ S
pRO^^
O
2
ui
O
<7
J
OFFICE OF INSPECTOR GENERAL
Catalyst for Improving the Environment
Ombudsman Report
Appropriate Testing and Timely
Reporting Are Needed at the
Hercules 009 Landfill Superfund
Site, Brunswick, Georgia
Report 2005-P-00022
September 26, 2005
-------
Report Contributors:
Christine Baughman
Tapati Bhattacharyya
Stephen Schanamann
Frances E. Tafer
Michael H. Wilson
Abbreviations
ATSDR Agency for Toxic Substances and Disease Registry
CLE Cod Liver Extract
Corps U. S. Army Corps of Engineers
DDE Di chl orodipheny 1 di chl oroethy 1 ene
DDT Di chl orodipheny ltri chl oroethane
DNA Deoxyribonucleic Acid
EPA U. S. Environmental Protection Agency
EUSES European Union System for the Evaluation of Substances
GA/DNR Georgia Department of Natural Resources
GC/ECD Gas Chromatograph/Electron Capture Detector
GC/NIMS Gas Chromatograph/Negative Ion Mass Spectroscopy
GEPD Georgia Environmental Protection Division
HCH Hexachlorocyclohexane
IRIS Integrated Risk Information System
MATT Monitoring, Analysis, and Toxicity of Toxaphene in Marine Foodstuff
MNA Monitored Natural Attenuation
NIMS Negative Ion Mass Spectroscopy
OIG Office of Inspector General
OSWER Office of Solid Waste and Emergency Response
PCB Polychlorinated Biphenyls
ppb Parts Per Billion
ppt Parts Per Trillion
RPM Remedial Project Manager
SESD Science and Ecosystem Support Division
TCDD Tetrachlorodibenzo-p-dioxin
TDI Tolerable Daily Intake
Cover photo: Aerial view of the Hercules 009 Landfill, which is in the center and left-center
of this photograph, provided by Hercules Incorporated.
-------
$
<
73
\
(J
T
^ c>s°
pRQl^
U.S. Environmental Protection Agency
Office of Inspector General
At a Gl
2005-P-00022
September 26, 2005
Why We Did This Review
The Glynn Environmental
Coalition, a nonprofit community
organization whose goal is to
have a safe and healthy place to
live and raise a family, brought
several concerns to the
Ombudsman's attention
regarding the Hercules 009
Landfill site near Brunswick,
Georgia, in EPA Region 4.
Background
Between 1975 and 1980,
Hercules Incorporated operated
the Hercules 009 Landfill to
dispose of waste material from
producing toxaphene, an
agricultural pesticide. The site
became part of EPA's Superfund
program in 1984. Construction
on the cleanup was completed in
1999, but some contaminants
remained at the site, so every 5
years the site must be reviewed to
ensure that the cleanup is
functioning as intended and it
does not adversely affect human
health and the environment.
For further information, contact
our Office of Congressional and
Public Liaison at (202) 566-2391.
To view the full report and
addendum, click on the following
links:
www.epa.qov/oiq/reports/2005/
20050926-2005-P-00022.pdf
www.epa.qov/oiq/reports/2005/
20050926-2005-P-00022A. pdf
Catalyst for Improving the Environment
Appropriate Testing and Timely Reporting Are
Needed at the Hercules 009 Landfill
Superfund Site, Brunswick, Georgia
What We Found
The testing method that EPA uses to monitor for the presence of toxaphene in
groundwater is inadequate. Toxaphene degrades overtime, changing into
other products (toxaphene breakdown products). EPA's testing method does
not identify and measure the amount of toxaphene that has degraded, only
toxaphene that has not degraded.
The available toxicity data on the toxaphene breakdown products are not
specific as to what constitutes a safe, acceptable level of exposure. Because
toxaphene breakdown products may pose a risk to human health, they should
be monitored. A different analytical method, one that monitors toxaphene
breakdown products, should be used rather than the conventional EPA
method.
EPA's report on the 5-year review of the Hercules 009 Landfill is over a year
late, because
• Region 4 was gathering additional data to reduce uncertainty about
how much toxaphene was present in the groundwater.
• The U.S. Army Corps of Engineers took longer to prepare the first
draft report than expected.
• The EPA manager for the site changed twice during the review period.
• EPA and others made extensive comments on the draft reports.
What We Recommend
We recommend that EPA Region 4:
• use an analytical method that monitors both toxaphene and its
breakdown products in the groundwater at the Hercules 009 Landfill
and take appropriate action if toxaphene breakdown products are
found.
• issue the report on the Hercules 009 Landfill 5-year review.
-------
^£0SX
i ^JL^-7 - UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
| X\\/y ? WASHINGTON, D.C. 20460
V
^ PRQl^
OFFICE OF
INSPECTOR GENERAL
September 26, 2005
MEMORANDUM
SUBJECT:
Appropriate Testing and Timely Reporting Are Needed at the
Hercules 009 Landfill Superfund Site, Brunswick, Georgia
Report No. 2005-P-00022
FROM:
Paul D. McKechnie /s/
Acting Ombudsman
Office of Congressional and Public Liaison
TO:
J. I. Palmer, Jr.
Regional Administrator, Region 4
This is our final report on our review of complaints regarding the Hercules 009 Landfill
Superfund site conducted by the Office of Inspector General (OIG) of the U.S. Environmental
Protection Agency (EPA). This report contains findings that describe the problems the OIG
identified and corrective action the OIG recommends. It represents the opinion of the OIG and
findings contained in this report do not necessarily represent the final EPA position. Final
determinations on matters in this report will be made by EPA managers in accordance with
established resolution procedures.
Action Required
In accordance with EPA Manual 2750, you are required to provide a written response to this
report within 90 calendar days of the date of this report. You should include a corrective actions
plan for agreed upon actions, including milestone dates. We have no objections to the further
release of this report to the public. For your convenience, this report will be available at
http://www.epa.gov/oig/publications.htm.
If you or your staff have any questions regarding this report, please contact me at 617-918-1471
or Christine Baughman, the Project Manager, at 202-566-2902.
-------
Table of C
At a Glance
Chapters
1 Introduction 1
Objectives 1
Site Information 1
Background Information on Toxaphene 2
Scope and Methodology 2
2 The Groundwater Should Be Tested for Toxaphene Breakdown Products .... 4
Testing for Toxaphene Breakdown Products Is Needed 4
EPA's Method Fails to Identify Toxaphene Breakdown
Products in Groundwater 5
Using a Different Method Would Identify Breakdown Products 5
Region 4 Is Concerned about Using the New Method 6
Recommendations 7
Agency Comments and OIG Evaluation 7
3 Reporting on the Cleanup's Effectiveness Was Not Timely 10
Reviews Must Occur Every 5 Years 10
Second Report's Delay Was Chiefly to Ensure Cleanup Effectiveness 11
EPA Policy Requires the Reviewer to Make a Decision 12
Further Definite Information Is Needed 12
Implement Recommendations Proposed by the Corps 13
Recommendation 13
Agency Comments and OIG Evaluation 13
Appendices
A Technical Discussion on Toxaphene 15
B Region 4 Response to the Draft Report 33
C OIG Technical Comments on the Region 4 Response 39
D Report Distribution 44
-------
Chapter 1
Introduction
Objectives
The Glynn Environmental Coalition, a nonprofit community organization whose
goal is to have a safe and healthy place to live and raise a family, brought some
issues regarding the Hercules 009 Landfill, Brunswick, Georgia, to the attention
of the Ombudsman at the U.S. Environmental Protection Agency's (EPA) Office
of Inspector General (OIG). (The OIG Ombudsman reviews and reports on
public concerns regarding EPA activities, including Superfund.)
As a result, the OIG conducted a review of the issues to determine:
If human health at the Hercules 009 Landfill Superfund site and/or
surrounding area were threatened because the analytical method
used to test for the pollutant toxaphene underestimated the amount
of toxaphene breakdown products.
If the February 2004 draft report on the 5-year review, which
addressed the above and various other issues, was issued in a
timely manner by EPA.
Site Information
Hercules Incorporated began producing toxaphene, an agricultural pesticide, in
1948 and continued production through 1980. Toxaphene was one of the most
heavily used insecticides in the United States until 1982, when EPA cancelled the
registrations for most uses; all uses were banned in 1990. A registration is a
license allowing a pesticide product to be sold and distributed for specific uses in
accordance with specific use instructions, precautions, and other terms and
conditions. Between 1975 and 1980, Hercules Incorporated operated the Hercules
009 Landfill under a permit issued by the Georgia Environmental Protection
Division (GEPD). The permit allowed the Brunswick, Georgia, Hercules plant to
dispose of waste water sludge left over from having produced toxaphene. Part of
the Hercules 009 Landfill was also used for disposing empty toxaphene drums
and toxaphene-contaminated glassware, rubble, and trash. The GEPD monitored
the Hercules 009 Landfill while it operated.
1
-------
The Hercules 009 Landfill is located in the eastern portion of Glynn County,
Georgia, near the City of Brunswick. The site is a 16.5 acre property, although
the landfill occupies only about 7 acres on the north end of the property. It is
bordered by Georgia State Highway 25 (Spur 25) on the west; an automobile
dealership on the north; a pine forest on the east; and several homes, a church, a
school, and a strip shopping center on the south/southeast. A shopping mall built
in 1985, a bank, and a restaurant are located approximately 1000 feet north of the
landfill. In the forested area in the southeastern part of the site, outside of the
protectively capped area, is a storage shed near a small pond. The property is
surrounded by a fence with locked gates.
EPA added the Hercules 009 Landfill to the Superfund National Priority List in
1984. In December 1992, EPA completed an interim cleanup action at the site.
As part of the interim action, private wells immediately down gradient of the
landfill were replaced by the municipal water system. Further site cleanup action,
known as remedial action, began at the site in June 1998. The remedial action
involved stabilizing the site and constructing a landfill cover, which was then
covered with clean fill material that was graded to promote adequate drainage,
and seeded with grass. Although active cleanup of the groundwater was not
included, groundwater monitoring was required to assess (1) any movement of
contamination through groundwater and (2) reduction of contaminants through
natural means, known as natural attenuation.
Background Information on Toxaphene
Unlike most organic environmental pollutants, toxaphene is not a single organic
compound. As manufactured, the original toxaphene pesticide was a mixture of
more than 200 closely related chlorinated organic compounds. This original
toxaphene pesticide mixture is commonly known as "technical" toxaphene.
Technical toxaphene consists mainly of polychlorinated bornanes with between
six to nine chlorines attached. The term "congener" is used to refer to a single,
structurally unique constituent of the mixture. In other words, at least 200
individual toxaphene congeners make up the original toxaphene pesticide mixture.
Individual congeners are often given their own name, such as Hx-Sed and Hp-
Sed. When the original toxaphene is released to the environment, it naturally
breaks down or degrades. These breakdown products are a different mixture than
the original toxaphene mixture, so it appears different to the testing instruments.
Scope and Methodology
The OIG began preliminary research on the site in late 2002 and completed work
in April 2004. It started the field work on this review in June 2004 and completed
it in January 2005.
We interviewed the present and two former remedial project managers (RPMs)
from EPA Region 4 and other Region 4 officials, and reviewed site files. We also
interviewed representatives of the U.S. Army Corps of Engineers (Corps) and the
2
-------
Glynn Environmental Coalition. A representative of Hercules Incorporated
provided the OIG team and Corps officials a tour of the site.
We interviewed representatives of EPA's Office of Research and Development
and Office of Solid Waste and Emergency Response. The OIG team reviewed
toxaphene testing protocols and about 50 journal articles on toxaphene toxicity
and exposure issues. The team also obtained additional information from various
State and Federal Internet sites.
On May 16, 2005, the OIG issued a draft report to EPA's Region 4 for review and
comment. The Regional Administrator responded on June 29, 2005. This
response, which included six attachments with detailed technical and legal
comments, suggested alternative actions to address the matters discussed in the
draft report. We provide a summary and general evaluation of the EPA comments
and our response at the end of Chapter 2 and Chapter 3 of this report. We
included the Regional Administrator's memorandum in Appendix B. The
memorandum, together with the six attachments and other documents later
provided by Region 4, are available as an addendum to this report. Appendix C is
the OIG evaluation of the EPA response, including the review that Region 4
requested of their proposed interim strategy for risk assessment.
We performed our review in accordance with Government Auditing Standards
issued by the Comptroller General of the United States. However, our review of
management controls and compliance was limited to those related to this
particular Superfund site.
The findings in this report are not binding in any enforcement proceedings
brought by EPA or the Department of Justice under the Comprehensive
Environmental Response, Compensation, and Liability Act to recover costs
incurred not inconsistent with the National Contingency Plan.
3
-------
Chapter 2
The Groundwater Should be Tested for Toxaphene
Breakdown Products
EPA policy states that monitoring programs should identify any potentially toxic
transformation products. Thus, the groundwater at the Hercules 009 Landfill site
should be periodically tested for toxaphene and its toxic breakdown products. To
determine if the groundwater is contaminated by toxaphene breakdown products,
the EPA needs to use a new analytical method that specifically tests for these
products. Until EPA knows whether the groundwater is contaminated by
toxaphene breakdown products, it will be unable to definitively determine if the
cleanup for the Hercules 009 Landfill protects human health. Appendix A
contains a technical discussion of the matters addressed in this chapter.
Testing for Toxaphene Breakdown Products Is Needed
Part of the cleanup at the Hercules 009 Landfill site is monitored natural
attenuation (MNA or natural attenuation). A natural attenuation remedy (or
cleanup) is one in which environmental contamination remains on-site and is
allowed to naturally decompose. For a natural attenuation cleanup to protect
human health, the environmental contaminants cannot migrate off the site
through, for example, the movement of contaminated groundwater. Therefore,
contaminants known to be at the site, as well as any toxic chemicals resulting
from the breakdown of those contaminants, should be periodically tested (i.e.,
monitored) in the groundwater. In the case of Hercules 009 Landfill, the
surrounding groundwater is being periodically tested for the original toxaphene
mixture put in the site between 1975 and 1980. However, this routine testing does
not specifically look for, or definitively identify, individual breakdown products
of toxaphene, i.e., the toxaphene breakdown products.
According to Superfund's guidance for monitored natural attenuation (OSWER—
Office of Solid Waste and Emergency Response—Directive 9200.4-17P), EPA
should look for the potential presence of toxic transformation products.
Toxaphene breakdown products are a type of transformation product.
Specifically, the guidance states:
The potential for creation of toxic transformation products
is more likely to occur at non-petroleum release sites ...
and should be evaluated to determine if implementation of
a MNA [natural attenuation] remedy is appropriate and
protective in the long term.
4
-------
Furthermore, the guidance states:
... all [MNA or natural attenuation] monitoring programs
should be designed to accomplish the following:... Identify
any potentially toxic and/or mobile transformation
products.
In other words, EPA's guidance requires groundwater monitoring for breakdown
products when the following two conditions are met: (1) the original contaminant
is known to decompose and form breakdown product(s), and (2) these breakdown
products are thought to be toxic and/or mobile in the groundwater. Since
toxaphene is known to degrade in the environment and these breakdown products
are thought to be toxic, EPA must evaluate the groundwater at the Hercules 009
Landfill site for toxaphene breakdown products.
EPA's Method Fails to Identify Toxaphene Breakdown Products in
Groundwater
The EPA conventional analytical method (known as Method 8081) is currently
used to monitor toxaphene in the groundwater at the Hercules 009 landfill site.
EPA's method is a test procedure designed to look for the original, unaltered
toxaphene mixture. However, microbes in the soil are known to decompose the
original toxaphene mixture into just two major breakdown products (i.e., Hx-Sed
and Hp-Sed) and several minor breakdown products. The original toxaphene
mixture and the two principal toxaphene breakdown products look completely
different to the analytical instruments. As a result, the groundwater monitoring
data collected at the site, using EPA's method, only identified the original
toxaphene mixture in the groundwater. The laboratory results from the EPA
method clearly indicated that groundwater surrounding the site does not contain
the original toxaphene mixture.
The groundwater surrounding the site could be contaminated with toxaphene
breakdown products. Unfortunately, EPA's method does not look for, or
specifically identify, toxaphene breakdown products. Therefore, the groundwater
monitoring by EPA's method does not meet the Agency's requirement to monitor
for toxaphene breakdown products, and the current groundwater monitoring data
from the Hercules 009 Landfill site cannot be used to determine whether the
groundwater could be contaminated by toxaphene breakdown products.
Using a Different Method Would Identify Breakdown Products
When the OIG looked at the groundwater monitoring data for evidence of
toxaphene breakdown products, the OIG found some evidence suggesting that
toxaphene breakdown products may be in the groundwater surrounding the
Hercules 009 Landfill site. However, these groundwater monitoring data are
insufficient to definitively determine the presence or absence of toxaphene
5
-------
breakdown products in the groundwater. Without such definitive data, any
determination on the effectiveness of the cleanup remedy is inconclusive.
A new analytical method using Negative Ion Mass Spectroscopy (NIMS, or called
the new method hereafter) should be used to test for toxaphene breakdown
products in the groundwater. Academia and the European Union have
successfully used the new method for at least 5 years to test for toxaphene
breakdown products in the environment.
The new method provides definitive test results because the technique generates a
mass spectrum for each compound in an environmental sample. A mass spectrum
is analogous to a chemical "fingerprint." By comparing the "fingerprint" of an
unknown compound in the Hercules 009 groundwater sample with the known
"fingerprint" of the toxaphene breakdown products, a resulting match of the
"fingerprints" would definitively identify the presence of toxaphene breakdown
products.
On the other hand, if the "fingerprints" do not match, then the new technique
would definitively determine that toxaphene breakdown products are not present
in the groundwater. Therefore, the use of the new technique for groundwater
monitoring at the Hercules 009 Landfill site provides the certainty needed to
determine whether the groundwater is contaminated by toxaphene breakdown
products. This information could then be used in the 5-year review to accurately
determine the effectiveness of the Hercules 009 Landfill cleanup remedy.
Region 4 Is Concerned about Using the New Method
Region 4 officials are concerned about using the new (or NIMS) method to test
groundwater at the Hercules 009 Landfill because the method is not approved by
EPA. When using environmental data for public health decisions, Region 4
prefers to use approved methods that have been validated by several laboratories.
The EPA conventional method (Method 8081) was validated. Consequently, EPA
knows the results will be accurate when Method 8081 is used to test for the
original toxaphene mixture.
The OIG agrees that using an EPA-approved method is better than a method not
approved by EPA. However, EPA has no approved method to identify toxaphene
breakdown products. Thus, to decide if the cleanup is effective, Region 4 must
use an unapproved method to obtain the necessary information on the presence or
absence of toxaphene breakdown products.
Under EPA's hazardous waste program, unapproved methods may be used. The
conventional method (Method 8081) is in the EPA publication SW-846, entitled
Test Methods for Evaluating Solid Waste, Physical/Chemical Methods. It is the
official collection of analytical and sampling methods that EPA has evaluated and
approved for use in complying with the hazardous waste regulations. SW-846
functions primarily as a guidance document setting forth acceptable, although not
6
-------
required, methods for the regulated and regulatory communities to use. The
related EPA program office continually reviews advances in analytical
instruments and techniques, and periodically updates SW-846 to improve method
performance and cost effectiveness. However, the methods are guidance and not
mandatory:
Except where explicitly specified in a regulation, the use of SW-
846 methods is not mandatory in response to Federal testing
requirements. ... The Agency does not intend to restrict the use of
new analytical techniques. Advances in technologies applicable to
the sampling and analysis of environmental media and hazardous
wastes outpace the ability of the Agency to promulgate revisions to
this manual. ...In summary, the methods included here provide
guidance to the analyst and the regulated community in making
judgements [sic] necessary to generate data that meet the data
quality objectives for the intended use of the results.
The EPA Region 4 laboratory has the capability to run the new (NIMS) method.
However, the Region 4 laboratory personnel will need to learn the procedures,
show they work, and practice using them before actually testing groundwater
samples from the Hercules 009 Landfill. Thus, implementing the new method
will take more laboratory resources, but is needed to obtain the necessary
information to decide if the cleanup is effective.
Recommendations
We recommend that the Regional Administrator, Region 4:
2.1 Use negative ion mass spectroscopy to definitively determine if toxaphene
breakdown products are present in the surrounding groundwater at the
Hercules 009 Landfill site, and (if so) in what amounts.
2.2 If toxaphene breakdown products are found in the groundwater, assess the
resulting risk to human health and take appropriate action.
Agency Comments and OIG Evaluation
The Agency generally agreed with our recommendations in Chapter 2. Under a
voluntary interim action, Hercules Incorporated had samples collected in March
2005 from the monitoring wells at the Hercules 009 Landfill analyzed. These
samples were analyzed using the negative ion mass spectrometry as well as EPA
method 8081. The results of these analyses, which were provided as Attachment
1 to the Region 4 response and are in the Addendum to this report, were less than
the maximum contaminant level for toxaphene in drinking water (i.e., 3
micrograms per liter). Region 4 later provided results of analytical testing they
had performed using EPA method 8081. Their results showed significantly more
7
-------
toxaphene in one well (N-l 1) at the site. These results are in the Addendum to
this report.
The OIG evaluated these analyses of the groundwater. As discussed in Appendix
C, the quality control does not adequately show that the substances to be
measured were not destroyed when the samples were prepared. We believe this
situation confirms the need for EPA to develop and validate a standard NIMS
method for analyzing samples for toxaphene breakdown products.
Region 4 confirmed they have the instrumentation necessary to perform the NIMS
procedure, but additional resources would be needed to implement the method in
the Region 4 laboratory. They would be willing to help validate the method.
Region 4's laboratory is willing to participate in a multi-
laboratory method validation study for toxaphene congeners in
environmental samples. However, since the Agency as a whole
would obviously benefit from a validated NIMS methodfor
toxaphene congeners, we believe that a multi-laboratory method
validation study should be initiated at the program level by the
Office of Solid Waste and Emergency Response. A validated
method will serve both the regulated community and the Agency by
assuring that analytical data produced by the method are
defensible, of known quality, and suitable for risk assessment
decision making.
Region 4 proposed two interim strategies to assess the risk to human health posed
by the toxaphene breakdown products, which they called weathered toxaphene.
The proposed strategy preferred by Region 4 uses toxicity criteria developed by
Region 4 staff. It is based, in part, on the sole toxicological study of weathered
toxaphene, which is the Monitoring, Analysis, and Toxicity of Toxaphene in
Marine Foodstuff (a.k.a. MATT) from the European Union, but they identified
several concerns about the study. Region 4 agreed that additional toxicity studies
would be helpful in filling data gaps, and (in Attachment 2 to the Region 4
response) suggested several studies. Region 4 requested that the OIG assess the
soundness and applicability of these criteria. The other strategy uses toxicity
criterion for technical, or original, toxaphene.
We evaluated the proposed strategies, and believe the strategy using toxicity
criteria for toxaphene is inappropriate because the toxicity of weathered
toxaphene could be significantly different. The toxicity criterion developed by
the Region 4 staff for weathered toxaphene is innovative in that Region 4
developed a reference dose for weathered toxaphene. A reference dose is the
concentration of a chemical believed to cause health problems. Thus, to ensure
no ill effects, people should not be exposed to more than the reference dose.
According to a Region 4 toxicologist (in the updated Attachment 2 to the Region
4 response, which is in the Addendum to this report):
8
-------
This use of the reference dose represents one of the first
instances in which EPA has applied a threshold-type
toxicity to a potentially carcinogenic chemical. This
procedure is consistent with EPA's new cancer guidelines
and with the goal of harmonization of cancer and non-
cancer risk assessment.
As discussed in more detail in Appendix C, because developing this reference
dose is so innovative, it should be subjected to internal and external peer review
before being applied. Given the problems in analyzing the groundwater from the
Hercules 009 Landfill, the need for a risk assessment for this site has not yet been
established.
9
-------
Chapter 3
Reporting on the Cleanup's Effectiveness
Was Not Timely
Every 5 years EPA must review the effectiveness of the Hercules 009 Landfill
cleanup, determining whether the cleanup is operating as intended and still
protects human health and the environment. EPA did not issue the report on the
second 5-year review timely; though due in September 2003, it remains in draft
form. The chief reason for delay was because Region 4 officials wanted to ensure
the cleanup was effective.
The latest report draft, dated February 2004, concluded that the cleanup was
operating as intended. We initially believed the report should be issued,
concluding that the effectiveness of the cleanup cannot be determined because the
EPA conventional toxaphene test method (Method 8081) cannot determine
whether breakdown products are present in the groundwater. However, we now
agree Region 4 can conclude the cleanup was operating as intended.
Reviews Must Occur Every 5 Years
According to Section 121(c) of the Comprehensive Environmental Response,
Compensation, and Liability Act and related regulations, cleanup actions must be
reviewed every 5 years when hazardous substances, pollutants, or contaminants
remain at the site, as was the case with the Hercules 009 Landfill. According to
EPA's June 2001 Comprehensive Five-Year Review Guidance, the purpose of the
5-year review was to evaluate, through data gathering and observation, how well
the site's remedy is working. The technical evaluation should address the
following three questions. We believe the answer to Question B would be
affected if toxaphene breakdown products were found in the groundwater
surrounding the Hercules 009 Landfill.
• Question A - Is the remedy operating as intended by the decision
documents?
• Question B - Are the exposure assumptions, toxicity data, cleanup levels,
and remedial action objectives used when the remedy was selected still
valid?
• Question C - Has any other information come to light that could call into
question if the remedy protects human health and the environment?
EPA issued the report on the first 5-year review in September 1998. Thus,
according to Agency guidance, the next 5-year review was due in September
2003.
10
-------
Second Report's Delay Was Chiefly to Ensure Cleanup Effectiveness
For the second review, the U.S. Army Corps of Engineers, or the Corps (while
working for EPA) gathered information from July 2002 through December 2002.
The Corps sent EPA the first draft 5-year review report in January 2003 and the
most recent, complete draft in February 2004. As of January 2005, the report on
the second 5-year review still had not been issued. It was delayed primarily
because of Region 4's efforts to decide if the cleanup protects human health and
the environment. Reaching such a decision was complicated by the issues
addressed in Chapter 2. Delay resulted when EPA tried to resolve some of the
technical concerns by re-evaluating some existing information and obtaining
additional data.
In September 2003, the RPM (the Remedial Project Manager) asked the Region 4
groundwater expert to review the hydrological data for the Hercules 009 Landfill
site and determine if the monitoring wells were properly located. In December
2003, this expert sent the RPM a groundwater flow map and a memorandum
noting that the monitoring wells were properly located to evaluate the
groundwater.
The 5-year review draft reports used groundwater monitoring data collected
through November 2002. In June 2004, Hercules Incorporated performed another
round of groundwater testing. Region 4's laboratory also analyzed samples from
this round, but used a different option to measure the toxaphene. Another round
of groundwater testing was performed in September 2004, with the Region 4
laboratory again analyzing the samples. The results of this round became
available in November 2004.
Other factors delaying the report were the following:
• Corps officials acknowledged that due to staffing, data gathering, and
other problems, they took 8 months to complete their fieldwork and
submit the first draft report to EPA.
• The EPA official in charge of the site, the RPM, changed in April 2002
and again in March 2003. The Corps and current RPM believe that
additional review time by new RPMs delayed the report.
• The Glynn Environmental Coalition submitted extensive comments on the
February 2004 report draft. Several sources outside the Region were
asked to submit responses, so the Region required additional time to
collate and address responses.
• In January 2004, the Glynn Environmental Coalition requested a
substantial amount of information, and the Region delayed the report to
respond to the request.
11
-------
• Once the Ombudsman notified Region 4 that we intended to review
actions at the site, Region 4 staff said they intended to delay issuing the
report on the 5-year review. They want to know what the Ombudsman's
report will contain, so the reports do not conflict.
EPA Policy Requires the Reviewer to Make a Decision
According to EPA's June 2001 Comprehensive Five-Year Review Guidance, at
the end of the technical analysis and evaluation phase of the 5-year review, the
reviewer must make a decision about the cleanup's effectiveness and document
this decision (or determination) in a report on the review. Based on the answers
to the three technical questions, the reviewer may decide that the cleanup protects
human health, will protect it once the cleanup is completed, will protect it in the
short term, or will not protect it. The reviewer can also conclude that protection
cannot be determined until further information is obtained. If a protection
statement cannot be made, a time frame should be provided when such a
statement will be made, through an addendum. If this is the case, the next 5-year
review is due 5 years from the date that the report is signed, not from the
signature date of the addendum.
Further Definite Information Is Needed
Between the first draft report of the 5-year review and the most recent one, the
conclusion on the cleanup's effectiveness changed; we agree with the Corps'
original conclusion that additional information is needed to responsibly make a
determination. The protection determination proposed by the Corps in the
January 2003 draft report stated that the cleanup action should be protective when
the groundwater cleanup goals were reached. However, it also stated that the data
being collected was not appropriate to determine if the cleanup is functioning.
The Corps believed the data were not appropriate based on three main factors:
toxicity of toxaphene breakdown products, appropriateness of the selected
groundwater sampling methodology, and appropriateness of the analytical method
used to test the groundwater.
In the February 2004 draft, the first sentence of the determination remained the
same, i.e., the cleanup actions at the site are expected to be protective upon
attainment of groundwater cleanup goals. However, the subsequent sentences in
the first paragraph of the determination were deleted, including the caveat about
the data:
The review of documents, [applicable and relevant and
appropriate requirements], risk assumptions and analytical data
and site inspections indicate that the data being collected is not
appropriate to determine if the remedy [cleanup] is functioning as
intended by the ROD [Record of Decision on site cleanup].
Ground-water [sic] data from the monitoring well network at the
12
-------
site does not appropriately represent the concentrations of total
toxaphene present in the ground water.
The Corps believed the additional information provided by Region 4 supported
this change in the protection determination. We initially did not agree. As
discussed in this report, Region 4 should identify and measure both toxaphene and
its breakdown products. However, since additional actions were taken to ensure
that the community around the site is not exposed to the groundwater, we agree
the remedy is effective.
Implement Recommendations Proposed by the Corps
Based on their review, the Corps recommended six corrective actions at the
Hercules 009 Landfill. Although we agree with all of them, we were particularly
interested in the three related to the groundwater. One of these three concerned
testing for toxaphene breakdown products, which we addressed in Chapter 2 of
this report. Another concerned the deed restrictions to control future land use.
The third concerned verifying that no one in the vicinity is using the groundwater.
The latter two recommendations should be implemented immediately because
doing so would further protect human health while the testing issue is resolved.
As part of the cleanup, restrictions were to be placed in the property deed to limit
using the land, and to prevent excavating at the site and using the groundwater.
The Corps did not find such deed restrictions in the property records. Therefore,
the Corps recommended that, with EPA oversight, Hercules Incorporated place
deed restrictions to control future land use. According to the RPM, this
recommendation is being addressed.
To address the threat of contaminated groundwater that could affect residential
drinking wells down gradient of the Hercules 009 Landfill, in December 1992, the
municipal water lines were extended to reach the residents adjacent to the site.
Later, the private wells of the residents who received municipal water were to be
properly abandoned. The document summarizing the cleanup actions taken at the
site did not confirm that all private wells were properly abandoned. The Corps
recommended that EPA make sure no one is still on private wells and that no one
has put in new wells in that area.
Recommendation
3.1 We recommend that the Regional Administrator, Region 4, issue the
report on the Hercules 009 Landfill 5-year review.
Agency Comments and OIG Evaluation
The draft report contained three recommendations in Chapter 3, all of which the
Region 4 officials disagreed with. We removed the second and third of the three
recommendations because Region 4 provided documentation that (1) the consent
13
-------
decree was recorded in the records of Glynn County, and (2) Hercules
Incorporated investigated the status of wells in the immediate area of the Hercules
009 Landfill and concluded that no new private drinking water wells were
installed. Thus, no one in the vicinity of the site was using the groundwater. A
Region 4 hydrologist evaluated the perimeter monitoring well system and
concluded it is sufficient to evaluate migration of contaminants. The
hydrologist's report was Attachment 4 to the Region 4 response.
The remaining recommendation in the draft report was that the Regional
Administrator, Region 4, issue the report on the Hercules 009 Landfill 5-year
review with the conclusion that whether the groundwater cleanup protects human
health cannot be determined at this time, and further evaluation is needed. A
timeframe should be estimated for such an evaluation.
Because of the recent evidence (see Chapter 2) provided by the NIMS data and
the toxicity criteria developed by Region 4 based on the MATT laboratory study,
Region 4 proposed that the data and toxicity review be included in the release of
the 5-year review and a determination of protectiveness be issued. They believed
this position was further supported by their above conclusions that no one was
using the groundwater and the contaminants (if any) were not leaving the site.
The OIG agrees with Region 4 that there appears to be no pathway for toxaphene
breakdown products in the groundwater to reach people near the site. Therefore,
we revised the recommendation to require the report on the 5-year review be
issued. The report can state that the remedy is effective and protects human
health, although our reasons, as discussed in more detail in Appendix C, differ
slightly from those of Region 4.
14
-------
Appendix A
Technical Discussion on Toxaphene
In order to evaluate the continued effectiveness of the Hercules 009 Landfill remedy, the Record
of Decision (EPA/ROD/R04-93/144) requires periodic testing of the groundwater for toxaphene
contamination. Both Hercules Incorporated and the U.S. Environmental Protection Agency
(EPA) use EPA's analytical Method 8081 to test for the original toxaphene pesticide mixture in
the groundwater. However, the original toxaphene pesticide mixture is known to degrade in the
environment and its degradation products are probably at least as toxic as the original toxaphene
pesticide. Since EPA's monitored natural attenuation guidance requires EPA to anticipate and to
test for the presence of potentially toxic degradation products in the groundwater, EPA is
required to evaluate the groundwater around the Hercules 009 Landfill Superfund site for the
potential presence of toxaphene degradation products. However, analytical Method 8081 was
not designed for and is inadequate to detect and measure toxaphene degradation products.
Therefore, EPA needs to use a different analytical method, such as negative ion mass
spectroscopy, to definitively assess the presence or absence of toxaphene degradation products in
the groundwater at the Hercules 009 Landfill site.
Basics of Toxaphene Chemistry
A basic understanding of the chemical structure of toxaphene is needed to address the issue.
Unlike most organic environmental pollutants, toxaphene is not a single organic compound. As
manufactured, the original toxaphene pesticide is a mixture of more than 200 closely related
chlorinated organic compounds. This original toxaphene pesticide mixture is commonly known
as "technical" toxaphene. Technical toxaphene consists mainly of polychlorinated bornanes with
between six to nine chlorines attached. The term, congener, is used to refer to a single,
structurely-unique constituent of the mixture. In other words, at least 200 individual toxaphene
congeners make up the original toxaphene pesticide mixture. Individual congeners are often
given their own names, such as Hx-Sed, Hp-Sed, p26, or p50.
Technical Toxaphene Degrades in the Environment
In the Office of Inspector General's (OIG's) review of the available scientific literature on the
environmental degradation of the original toxaphene mixture (a.k.a. technical toxaphene), we
found numerous references to biotic and abiotic degradation, and to aerobic and anaerobic
degradation. The aerobic degradation of technical toxaphene occurs at the slowest rate and has
an aerobic half-life report of about 10-14 years (Fingerling 1996). On the other hand, anaerobic
degradation of technical toxaphene occurs at a much faster rate and has an anaerobic half-life of
about 6 weeks. Therefore, since the use of toxaphene was severely restricted in 1982 (i.e., about
23 years ago), any technical toxaphene left in the environment from 1982 or before has
theoretically undergone two or more half-lives. Thus, at most, only 25 percent of the original
starting material should theoretically still be present. By contrast, the only reported condition
under which toxaphene does not degrade is autoclaved soil (i.e., all microbes in the soil have
15
-------
been killed off) (Fingerling 1996). Therefore, technical toxaphene is expected to degrade in the
environment and its degradation is mediated primarily by microbes living in the soil.
Anticipated Toxaphene Degradation Products
Upon instrumental analysis by a Gas Chromatograph/Electron Capture Detector (GC/ECD), the
original technical mixture (i.e., a mixture of 200 or more congeners) produces a complex, multi-
peaked chromatogram (see Figure IB below). However, technical toxaphene is known to
undergo microbial degradation in soil. Since the soil at the Hercules 009 Landfill site has been
stabilized with cement, the free exchange of oxygen into the soil from the air is unlikely.
Therefore, anaerobic microbial degradation is the most likely degradation process for the buried
toxaphene waste at the Hercules 009 Landfill site.
(A) Toxaphene Degradation Products in Soil
Hp-Sed
Hx-Sed
——
23:21
2
-------
a much simplified chromatogram (see Figure 1A above). Therefore, upon analysis at an
environmental laboratory, the degraded toxaphene chromatogram appears completely different
then the technical toxaphene chromatogram.
Although Hx-Sed and Hp-Sed are the major anaerobic degradation products in soil, degraded
toxaphene chromatogram (see Figure 1 A) also shows a significant number of other, less
abundant anaerobic microbial degradation products. These less abundant toxaphene congeners
in soil have been identified and are known to include the p26, p50, p40, p41, and p44 (Maruya
2001). As discussed in more detail later, these less abundant toxaphene degradation products
constitute the majority of risk to human health because they are not effectively metabolized by
the body, which causes them to bioaccumulate in the body. Hx-Sed and Hp-Sed are readily
metabolized by the body and excreted, so they should not constitute a major risk to human
health. However, since Hx-Sed and Hp-Sed are the major anaerobic degradation products, they
are easier to detect than the other less abundant toxaphene congeners and could be used to
indicate that toxaphene degradation products are present in the sample.
The implication of toxaphene's degradation is that humans are exposed to toxaphene's
degradation products and not to the original technical toxaphene mixture (de Geus 1999),
(McHugh 2003). Consequently, EPA's approach of using Method 8081 to test for the original
technical toxaphene in the environment to identify toxaphene contamination is incorrect. EPA
needs to test for individual toxaphene degradation products (i.e., specific congeners) in order to
identify the presence or absence of toxaphene contamination in the environment.
Evaluating the Potential Risk to Humans from Toxaphene Exposure
Conducting a detailed and comprehensive risk assessment for the potential exposure to
toxaphene from the Hercules 009 Landfill site is a complex task that is beyond the scope of this
OIG review. Detailed information is lacking on the potential human exposure to toxaphene
degradation products and their toxicity, which limits the ability to conduct a thorough risk
assessment. However, the potential risk to human health from toxaphene exposure can still be
conceptually understood.
In general, a major factor needed to evaluate the level of risk to human health is to determine the
major human exposure pathways to toxaphene's degradation products and to determine all
potential sources. The Hercules 009 Landfill site is just one of the potential exposure sources.
A toxaphene exposure study from the Netherlands used an EUSES (European Union System for
the Evaluation of Substances) model to estimate the exposure of the Dutch population to
toxaphene (Fiolet and van Veen 2001). This model identified that the main route of exposure to
relatively soluble toxaphene congeners is approximately 80 percent from fish and 11 percent
from drinking water. Another toxaphene exposure study by Buranatrevedh also concluded that
the main route of exposure is about 93 percent from fish and about 7 percent from surface waters
(Buranatrevedh 2004). Based on these national toxaphene exposure studies, the main exposure
risk to toxaphene is clearly from fish (i.e., 80-93 percent) and from potential sources of drinking
water (i.e., 7-11 percent). Although specific site conditions and other site specific variables at
Hercules 009 Landfill will shift the relative levels for these various exposure routes, these
national toxaphene exposure studies identify that the principle exposure routes of concern to the
17
-------
surrounding community are the fish in the diet and the potential for contaminated drinking water.
The remaining exposure routes (i.e., air and soil) are practically negligible.
Another major factor needed to evaluate the level of risk to human health is what specific
toxaphene congeners pose chronic risk to humans. The major toxaphene congeners found in fish
are p26, p50, and p62, but p40, p41, and p44 are also present to a lesser extent (Fiolet and van
Veen 2001). The major anaerobic microbial degradation products in soils that may contaminate
the groundwater are Hx-Sed and Hp-Sed, but p26, p50, p40, p41, and p44 are also found in soil
to a lesser extent (Maruya 2001).
Although humans are exposed to a variety of toxaphene degradation congeners, most of these
congeners can be rapidly metabolized via dechlorination, dehydrodechlorination, and oxidation,
primarily through the action of the mixed function oxidase system and other hepatic microsomal
enzymes (EPA 1999). For example, the primary toxaphene degradation products in soil (i.e.,
Hx-Sed and Hp-Sed) are expected to be easily metabolized and excreted with reported half-lives
in fish of 5 and 13 days respectively (Smalling 2004). However, a limited number of toxaphene
congeners (i.e., p26, p50, p40, p41) are poorly metabolized and cannot be readily excreted,
causing these congeners to accumulate in the body. These poorly metabolized congeners share a
common structural pattern of alternating single chlorine substitutions (i.e., endo, exo) on the #2,
#3, #5, and #6 carbons of the six-member ring (Maruya 2000). Specifically, the poorly
metabolized p26 and p50 congeners have half-lives of about 1 year in wild fish (Smalling 2004).
Therefore, although humans are exposed to a variety of toxaphene degradation products, the
human body can metabolize and eliminate the majority of these congeners. However, five
toxaphene congeners (i.e., p26, p50, p 40, p41, and p44) are not readily metabolized and excreted
and, thus, can accumulate in the human body.
To evaluate the level of risk to human health, EPA needs to know the concentration of these five
congeners and their metabolite precursors in the environment. Since these five toxaphene
congeners represent the long-term chronic toxaphene exposure problem to humans, the toxicity
of these five individual congeners and/or a mixture of these five congeners needs to be
determined in more detail than is currently available in the scientific literature.
Human Exposure to the p26 and p50 Toxaphene Degradation Congeners
The following two academic studies have independently identified and documented human
exposure to the individual toxaphene degradation congeners:
- In 1996, Dr. Gill used negative ion mass spectroscopy (NIMS) to detect and measure
toxaphene congeners in human blood serum (Gill, et al. 1997). Specifically, Dr. Gill's
study documented the presence of p26, p50, p44, p40, and p41 congeners at a
concentration of 2 to 200 parts per trillion (ppt) or up to 0.2 parts per billion (ppb) in
human blood serum from Native Canadian Communities. These five toxaphene
congeners represented 95 percent of the toxaphene congeners found in human serum.
- In 2003, Dr. Barr used a sophisticated analytical technique to detect and measure
toxaphene congeners in pooled human blood serum collected by the Red Cross in Atlanta
18
-------
in 1987, in Chicago in 1992, and in Cincinnati in 1994 (Barr 2004). Specifically,
Dr. Barr's study documented the presence of p26, p50, and possibly p40+41, and p44
congeners at a concentration of 3 to 30 ppt (i.e., 0.03 ppb) in human blood serum from an
undefined number of American blood donors.
These studies are critically important in identifying and simplifying the assessment of the risk to
humans resulting from environmental exposure to toxaphene. These studies dramatically
indicate that human risk is not to "technical" toxaphene's 200-plus congeners, but that the long-
term chronic toxaphene exposure in humans is limited and simplified to just five toxaphene
congeners (i.e., p26, p50, p40, p41, and p44) that the human body has difficulty metabolizing
and eliminating, causing them to accumulate in the body.
Carcinogenicity of p26 and p50 Congeners
The EPA's Integrated Risk Information System (IRIS) database identifies technical toxaphene as
a B2 probable human carcinogen with a cancer potency factor of 1.1 mg/kg/day. However, there
is limited scientific data on the carcinogenicity of persistent toxaphene degradation congeners,
such as p26 and p50. But other chemical mixtures of congeners show that individual congeners
can be significantly more carcinogenic then the original technical mixture. The classic example
is dioxin where 2,3,7,8-TCDD (tetrachlorodibenzo-p-dioxin) is up to 10,000 times more
carcinogenic than other dioxin congeners. Another example is that bioaccumulated PCB
congeners appear to be more carcinogenic than the original "commercial" PCB mixture (EPA
7C-R293-NTSX). This clearly indicates that the carcinogenicity of the original technical
toxaphene mixture cannot be applied to the carcinogenicity of the individual congeners,
specifically, p26 and p50.
The European Union has conducted an Investigation into the Monitoring, Analysis, and Toxicity
of Toxaphene in Marine Foodstuffs (i.e., MATT project). The MATT project predicted the
tumor-promoting potency of technical toxaphene and a Cod Liver Extract (CLE) containing p26,
p50, p62 in a bioassay measuring the inhibition of intracellular communication between
Hepalclc7 cells (Fair CT PL.96.3131). The CLE toxaphene congener mixture mimics the
environmental exposure to the toxaphene congeners that are found in humans (e.g., p26 and
p50). The results from this bioassay indicate that the CLE toxaphene mixture is a more potent
tumor promoter than the original technical toxaphene mixture. The MATT project estimated a
tolerable daily intake (TDI) to "weathered" toxaphene residues of 0.69 mg/kg/day. In general
terms, MATT's TDI estimate makes the toxaphene degradation products found in humans to be
about twice as carcinogenic as the original technical toxaphene mixture. However, more
definitive toxicology studies are needed to verify the carcinogenicity of the individual p26 and
p50 toxaphene congeners.
Embryotoxicitv of p26 and p50
In a 1997 study using a rat embryo culture model, the p26 and p50 toxaphene congeners caused
abnormalities in the central nervous system (Calciu 1997). The total morphological scores at 100
ng/ml for p26 and p50 were slightly worse than the total morphological score for the technical
toxaphene. The significant finding from this study is that both the target site and type of toxicity
19
-------
are highly congener-specific. Therefore, the toxicity of technical toxaphene cannot and should
not be used to predict the embryotoxic effects of the p26 and p50 congeners in humans. Thus,
more scientific research is needed to evaluate the specific embryotoxic effects of p26 and p50 to
humans.
Dr. Gill's study found concentrations of p26 and p50 at concentrations as high as 0.2 ppb in
human blood serum (described above). The lowest dose in Dr. Calciu's rat embryo culture study
was 100 ppb. The difference is a factor of 500, but the rat embryo culture study results represent
only a 48-hour exposure to the rat embryos. This short exposure time does not directly
correspond to human exposures to p26 and p50 over the full term of a pregnancy (i.e., human
fetuses are exposed to a lower dose but for a longer period of time). Furthermore, the results
from the rat embryo culture study represent dramatic development changes in which even subtle
changes in human fetal development would be considered unacceptable. Therefore, additional
research is needed to evaluate the potential for more subtle effects on embryo development when
exposed to lower doses of p26 and p50 that correspond to actual human exposure levels. For
example, in 1980, Dr. Olson observed behavior changes in rat offspring when pregnant rats were
given low doses of technical toxaphene, p42a congener (i.e., toxicant A), or p32 congener (i.e.,
toxicant B) as measured by swimming test and maze retention test (Olson 1980).
The embryotoxicity of toxaphene's persistent degradation products needs to be evaluated in the
context of co-exposure with other persistent organochlorines (i.e.,
dichlorodiphenyltrichloroethane (DDT)/dichlorodiphenuldichloroethylene (DDE),
hexachlorocyclohexanes (HCHs), and polychlorinated biphenyls (PCBs)). The amount of p26
and p50 in human milk has been found to range from a low of 6 ug/kg lipid weight (i.e., ppb) in
southern Canada to a high of 294 ug/kg lipid weight in Northern Quebec (Skopp, S., et al. 2002).
This shows that babies are exposed through the mother's exposure to toxaphene degradation
products before and after birth. Unfortunately, this observation about mother's milk is
potentially problematic because an epidemiological study by Jacobson (Jacobson 1996) indicates
that developing embryos are the most susceptible target of organochlorines (i.e., PCBs).
Jacobson's study linked organochlorine exposure during fetal development to impaired cognitive
development (e.g., low IQ scores). Therefore, the embryotoxicity from toxaphene exposure
needs to be evaluated in combination with the concurrent exposure to the other organochlorines
(i.e., DDT/DDE, HCHs, and PCBs) in order to accurately assess the risk to cognitive
impairment.
20
-------
Superfund's Remedy Requires the Evaluation of Toxic Degradation Products
Monitored natural attenuation (MNA) is part of the remedy at Hercules 009 Landfill site.
Superfund's monitored natural attenuation guidance (OSWER Directive 9200.4-17P) requires
EPA to evaluate for the potential presence of toxic transformation products. Toxaphene
degradation products are a sub-category of transformation products. Specifically, the MNA
guidance states:
"77ze potential for creation of toxic transformation products is more likely to occur at
non-petroleum release sites ... and should be evaluated to determine if implementation of
a MNA remedy is appropriate and protective in the long term."
Furthermore, the MNA guidance states:
"... all [MNA] monitoring programs should be designed to accomplish the following:...
Identify any potentially toxic and/or mobile transformation products."
Therefore, the Superfund's MNA guidance requires EPA to anticipate and to test for the
presence of potentially toxic degradation products at hazardous waste sites. Since toxaphene is
known to degrade in the environment and these degradation products are thought to be toxic,
EPA must evalaute the groundwater at the Hercules 009 Landfill site for toxaphene's
degradation products, specifically, the Hx-Sed and Hp-Sed congeners, but also the p26, p50, p40,
p41, and p44 congeners.
EPA Method 8081 Tests for Technical Toxaphene
EPA Method 8081 is an analytical testing technique that uses a laboratory instrument called a
gas chromatograph with an electron capture detector, which is referred to as a GC/ECD. When
an environmental sample is tested by the GC/ECD, the instrument products a chromatogram as a
record of what was contained in the sample (see Figure 2A).
Each peak in the chromatogram of a known technical toxaphene standard (see Figure 2B)
represents 1 of the 200 unique congeners in the technical toxaphene mixture. There are actually
so many peaks that they clump together in some areas of the chromatogram. Method 8081
detects toxaphene by identifying five peaks to the right of the red line in the environmental
sample (see Figure 2A) and comparing their shape and position to five peaks to the right of the
red line in a known technical toxaphene standard (see Figure 2B). In this example, since both
chromatograms match on the right hand side, the laboratory would report that toxaphene is
present in this hypothetical enviromental sample. EPA Method 8081 was designed and quite
dependable for detecting the original technical toxaphene mixture in an environmental sample.
21
-------
Since the late eluding peaks
match the technical toxaphene
standard, toxaphene is reported
in the enviromental sample
(A) Environmental Sample
26;48
30:15
33:42
23:21
5 Major Peaks Used to ID
"Technical" Toxaphene
(B) Technical Toxaphene Standard
26:48
30:15
33:42
23:21
Retention Time (minutes)
Figure 2: GC/ECD Chromatograms of Technical Toxaphene
EPA Method 8081 Does Not Identify Toxaphene Degradation Products
EPA Method 8081, which is listed in EPA's Test Methods for Evaluating Solid Waste,
Physical/Chemical Methods (i.e, SW-846), uses a GC/ECD to analyze for the presence of
toxaphene contamination. In order to identify the presence of a pollutant in an environmental
sample, the retention time of a peak (i.e., representing a compound) in the sample's
chromatogram is compared against the retention time of a known chemical standard. However,
technical toxaphene is not a single compound, but a mixture of more than 200 chlorinated
bornanes which produces a complex chromatogram (see Figure 3 B below). For the purposes of
detecting toxaphene in a sample, EPA Method 8081 calls for a peak profile match against at least
five peaks in the latter section of the toxaphene window (i.e., see the peaks after about 29
minutes in the chromatogram 3B). In other words, the observed relative abundance of late
eluting toxaphene congeners (i.e., octa- and nonachlorobornanes) has to closely match the
relative abundance of the octa- and nonachlorobornane congeners found in the technical
toxaphene standard. EPA Method 8081 is appropriate and highly accurate for detecting technical
toxaphene. However, EPA Method 8081 is not effective for detecting degraded toxaphene (i.e.,
"weathered" toxaphene) in environmental samples (e.g., soil, water, fish).
For demonstration purposes, chromatogram 3A below is a known chromatogram of toxaphene
degradation products in soils. When chromatogram 3A is compared by EPA Method 808l's
identification criteria for technical toxaphene, chromatogram 3 A obviously does not have the
22
-------
same late eluding peak profile (i.e., the peaks after 29 minutes) as the technical toxaphene
standard. Therefore, a match is not made and the presence of toxaphene is not reported by the
laboratory, even though specific toxaphene congeners (e.g., Hx-Sed and Hp-Sed) are known to
be present. This example demonstrates the manner in which EPA Method 8081 fails to detect
toxaphene degradation products (i.e./'weathered" toxaphene or individual toxaphene congeners)
in environmental samples.
(A) Degraded Toxaphene in Soil
Hx-Sed
Hp-Sed
i_A- vj —-/"A^ L
Same peak profile is not present
in degr aded toxaphene sample
/
23:21
26M
30:15
33:42
5 Major Peaks Used to ID
"Technical" Toxaphene
(B) Technical Toxaphene Standard
30:15
33:42
23:21
Retention Time (minutes)
Figure 3: EPA Method 8081 Analyzes for Only Technical Toxaphene
An example of EPA Method 808 l's failure to detect toxaphene's degradation products occurred
in 1997 during the Georgia Department of Natural Resources' (GA/DNR) study to measure the
toxaphene levels in several species of fish and shellfish in and around Terry Creek. Terry Creek
is another Superfund site in the Brunswick, Georgia, area that is contaminated with toxaphene
due to previous manufacturing operations by Hercules Incorporated. The results of GA/DNR's
study indicated no detectable quantities of toxaphene in every single fish sample analyzed.
However, in 2001, Dr. Maruya re-analyzed the same fish and shellfish samples that were
collected and analyzed by GA/DNR, but this time used a congener-specific technique called Gas
Chromatography/Negative Ion Mass Spectroscopy (GC/N1MS) (Maruya 2001). The GC/NIMS
analytical technique was able to identify and quantify individual toxaphene congeners that were
present in the fish samples at concentrations up to 1,420 parts per billion (ppb). The GC/NIMS's
identification of toxaphene contamination at Terry Creek is in stark contrast to the results
23
-------
obtained by EPA Method 8081 that indicated no toxaphene contamination. Therefore, this
example clearly shows that EPA's Method 8081 does not detect degraded toxaphene in the
environment.
Gas Chromatogaph/Negative Ion Mass Spectrometry Can Be Used to Identify Toxaphene
Degradation Products
Unlike the GC/ECD technique used in EPA Method 8081, GC/NIMS can definitively identify
and measure individual toxaphene degradation products in the environment. The weakness in the
GC/ECD technique used in EPA Method 8081 is that peak identification is based on only one
factor: retention time. Therefore, even if the retention times match between the sample peak and
the standard, there is still uncertainty about the identity of the peak. By contrast, the GC/NIMS
technique uses two factors to identify peaks: retention time and a mass spectrum. A mass
spectrum is analogous to a "fingerprint" of the compound. If the mass spectrum from the sample
matches the mass spectrum of the standard, this definitively identifies the compound.
The GC/NIMS methodology has been routinely used in academia since about 1993. The
European MATT project has been using the GC/NIMS method to monitor and document the
levels of toxaphene degradation products in fish from the North Atlantic. Since the GC/NIMS
method has been developed and successfully implemented by others, EPA's formal validation
and standardization of the GC/NIMS method should not present any major technical difficulties.
Also, including the GC/NIMS method in SW-846 would significantly facilitate (1) evaluating
toxaphene degradation products in the environment by the regulated community and (2)
gathering congener-specific data needed for accurate risk assessments of exposure to
toxaphene's degradation products.
Estimated Retention Time of Toxaphene Degradration Products
As described, EPA Method 8081 fails to identify toxaphene degradation products mainly
because the identification criteria are based on seeing the late eluting peaks in technical
toxaphene. However, an experienced chemist can still look for potential toxaphene degradation
products in the GC/ECD data from the Hercules 009 Landfill. Although the Hercules 009
GC/ECD data does not include standards for the Hx-Sed and Hp-Sed toxaphene degradates, the
expected retention time for Hx-Sed and Hp-Sed can be estimated from data published in the
scientific literature (Figure 4 below). Since each technical toxaphene varies slightly by
manufacturer, the technical toxaphene standard below is specifically from Hercules Incorporated
in order to allow a subsequent comparison with the Hercules 009 Superfund site data. The
estimated retention time window for the Hx-Sed and Hp-Sed toxaphene congeners (i.e., the main
toxaphene congeners expected in anaerobic soil) occurs at the front edge of the technical
toxaphene window. Notice that the Hx-Sed peak is to the left and taller than the Hp-Sed peak.
24
-------
Hx-Se.d
Estimated Retention Time Window
for Toxaphene Degradation Products
from Soil
. Toxaphene Degradation
Products Observed in
Fish from Terry Creek
0.20
Hp-Sed
0.15
o.io
0.05
0.00
20
40
50
70
Retention Time Window
for Hx-Sed and Hp-Sed
Occurs at the Front Edge
of the Toxaphene Envelope
Technical Toxaphene
Standard from Hercules
0.40
0.30
0.20
o.io
o.oo
20
30
T
40
T
50
T
60
T
70
Retention Time (minutes)
Figure 4: Estimated Retention Time Window for Hx-sed and Hp-sed
Hercules 009 GC7ECD Data Suggest Toxaphene Degradation Products May Be in the
Groundwater
On January 8, 2003, the contractor for Hercules Incorporated, RMT Incorporated, provided EPA
with the November 2002 groundwater sampling results, which were used in EPA's Hercules 009
Landfill 5-year review. KMT's subcontract laboratory (i.e., EnChem, Inc.) used EPA Method
25
-------
8081A to analyze the toxaphene groundwater samples and reported non-detect (i.e., <5.2 ug/L)
for all toxaphene results. Fortunately, the sampling results included a chromatogram for each of
the groundwater monitoring wells. This allowed for comparing the Hercules 009 Landfill
groundwater samples against the Hercules technical toxaphene standard to check for the
possibility of toxaphene degradation products (Figure 5 below).
Hercules 009 Landfill GC/ECD Data
Samples Collected November 2002
Analyzed by EnChem Inc. by Method 8081
Suspect Suspect
Peak A Peak B
Monitoring Well N-06SR (Shallow)
Decachlorobiphenyl
10 11 12
Monitoring Well N-ll (Shallow)
15
17
Technical Toxaphene Standard
(Tox-302)
12
13
Minutes
Note: Chromatograms are normalized to the surrogate compound (Decachlorobiphenyl)
Figure 5: Potential Toxaphene Degradation Products in the Groundwater at Hercules 009 Landfill
As described above in the previous section, the estimated retention time window for the Hx-Sed
and Hp-Sed toxaphene congeners occurs at the front edge of the technical toxaphene window.
When the estimated retention time windows for Hx-Sed and Hp-Sed are superimposed on the
chromatograms from monitoring wells N-06SR and N-ll, two prominent peaks are present that
resemble the Hx-Sed and Hp-Sed peak profile (i.e., the left peak is taller than the right peak).
These chromatograms provide suggestive evidence that Hx-Sed and Hp-Sed might be present in
the Hercules 009 Landfill groundwater. However, these peaks cannot be positively identified as
toxaphene degradation products due to significant limitations in the data set. First, there are no
ITx-Sed or Hp-Sed standards to establish their retention time, which is the key criterion for
26
-------
identifying compounds in a GC/ECD analysis (i.e., no standards, no identifications). Second, the
critical weakness with all GC/ECD data is the lack of a mass spectrum that could be used to
determine the structure of the compound making each of these peaks. The limitations of this
GC/ECD data set clearly show the value of GC/NIMS analysis.
If the samples had been run by GC/NIMS instead of GC/ECD, a quick review of the mass
spectra for each of the peaks could easily determine if these peaks were toxaphene congeners or
not (e.g., examples of negative ion mass spectra are provided in Figure 6). For example, the
negative ion mass spectrum for Suspect Peak A could be compared against the negative ion mass
spectrum of a hexachlorinated bornanes (Figure 6(a)). Likewise, the negative ion mass
spectrum for Suspect Peak B could be compared against the negative ion mass spectrum of a
heptachlorinated bornanes (Figure 6(b)). If the spectra matched, EPA could conclude that there
were toxaphene degradation products in the groundwater. However, with only the GC/ECD
data, a definitive determination on the identity of these suspect peaks cannot be made.
GC/NIMS Can Definitively Determine the Identity of Suspected Hx-Sed and Hp-Sed Peaks
GC/NIMS could be used to definitively determine the identity of the Suspect A and B peaks
observed in the Hercules 009 Landfill GC/ECD data (see Figure 5). The retention time and mass
spectrum for Suspect Peak A would be compared against the retention time and mass spectrum
for Hx-Sed. The mass spectrum for Hx-Sed looks like the diagram in Figure 6(a). The retention
time and mass spectrum for Suspect Peak B would be compared against the retention time and
mass spectrum for Hp-Sed. The mass spectrum for Hp-Sed looks like the diagram in Figure
6(b). The additional feature of the GC/NIMS technique of comparing and matching a peak's
mass spectrum allows for the definitive identification of the peaks.
27
-------
I00f4n
&>
¦
o
s
"
X
75 %-
u
9m
?-¦
¦
<
5ort -
c
-
£
-
_5
"
c
2:
— —
J2
lOD^-i
753c -i
50Tr
?5%
0% ¦
b)
341
343
Negative Ion Mass Spectrum
of Heptachlorobornanes
M5
r
250 275 300 325 350 375 40C ^25
Mass C.m/z.1
Figure 6: Negative Ion Mass Spectrums
for Hexachlorobornane and Heptachlorobornane
28
-------
Technical Summary of Toxaphene Issue
EPA should recognize that toxaphene degrades in the environment and that all of EPA's
toxaphene data collected using EPA Method 8081 are inadequate to screen for toxaphene's
degradation products in the environment. To address this problem, EPA should test toxaphene
contamination using a congener-specific analytical method such as GC/NIMS. EPA's validation
and standardization of the GC/NIMS method would facilitate evaluating toxaphene degradation
products in the environment. Finally, EPA should recognize that the chronic health risk to
humans is from the five persistent toxaphene congeners (i.e., p26, p50, p40, p41, and p44) that
accumulate in the human body. Therefore, EPA needs additional studies on the carcinogenicity
and embryotoxicity of these five persistent toxaphene congeners to be able to accurately evaluate
this risk associated with their exposure to humans. Without congener-specific laboratory results
and without knowing the toxicity of specific congeners, EPA is unable to definitively quantify
the risk to human health posed by the toxaphene degradation products left in the environment
and the food chain.
No Immediate Human Health Risk at Hercules 009 Landfill
Although the OIG's review indicates that the current monitoring and analytical testing at the
Hercules 009 Landfill site is inadequate to determine if any toxaphene degradation products have
leaked into the groundwater and that these toxaphene degradation products may be toxic, the
OIG believes that there is no immediate risk to the surrounding population because the remedial
action for Operable Unit Two has connected the nearby residents to the municipal drinking water
supply. However, to prevent the potential long-term contamination of the groundwater, EPA
needs to definitively determine if Hx-Sed, Hp-Sed, p26, and/or p50 toxaphene congeners occur
in the groundwater down gradient of the Hercules 009 site.
OIG Technical Conclusions
- The original "technical" toxaphene mixture degrades in the environment.
- The chronic health risk to humans is from exposure to toxaphene's persistent degradation
products (e.g., p26, p50, p40, p41, and p44) and not the original technical toxaphene
mixture.
- EPA needs to use a congener-specific analytical method (e.g., GC/NIMS) to positively
identify and quantify toxaphene degradation products in the environment. The OIG
highly recommends standardizing and validating the GC/NIMS method and inserting an
EPA GC/NIMS method into SW-846.
- EPA needs to conduct specific research into both the carcinogenicity and embryotoxicity
of the five persistent human toxaphene congeners (i.e., p26, p50, p40, p41, and p44) to
develop acceptable human exposure limits to the individual congeners and/or to the
mixture of these five congeners.
29
-------
Considering both the uncertainty as to whether the Hercules 009 Landfill sites
groundwater is contaminated with toxaphene degradation products and the uncertainty
about the human toxicity of these toxaphene degradation products, Superfund's 5-year
review for the Hercules 009 Landfill site should conclude that whether the MNA remedy
protects human health cannot be determined at this time. As a result of this uncertainty in
the data, the Region needs to definitively determine if the groundwater is contaminated
with the Hx-Sed and Hp-Sed toxaphene congeners generated by the microbial
degradation of "technical" toxaphene.
30
-------
References
Barr, J., et al., Measurement of Toxaphene Congeners in Pooled Human Serum Collected in
Three U.S. Cities Using High-Resolution Mass Spectrometry. Arch. Environ. Contam. Toxicol.,
2004, vol.46, pp. 551-556.
Braekevelt, Eric, et al., Comparison of an Individual Congener Standard and a Technical Mixture
for the Quantification of Toxaphene in Environmental Matrices by HRGC/ECNI-HRMS.
Environ. Sci. Technol. 2001, vol. 35, pp. 3513-3518.
Buranatrevedh, S., Cancer Risk Assessment of Toxaphene, Industrial Health, 2004, vol. 42, pp.
321-327.
Calciu, Cristina, et al., Toxaphene Congeners Differ from Toxaphene Mixtures in Their
Dysmorphogenic Effects on Cultured Rat Embryos, Toxicology 124 (1997), pp. 153-162.
De Geus, H., et al., Environmental Occurrence, Analysis, and Toxicology of Toxaphene
Compounds, Environ. Health Perspect, 1999, 107 (Suppl 1), pp. 115-144.
EPA 1999, Fact Sheet - Toxaphene Update: Impact on Fish Advisories. EPA-823-F-99-018,
September 1999.
FAIR CT PL.96.3131. Investigation into the Monitoring, Analysis, and Toxicity of Toxaphene in
Marine Foodstuffs. Final Report.
Fiolet and van Veen, Toxaphene Exposure in the Netherlands. National Institute of Public Health
and the Environment. RIVM Report #604502-003, January 2001.
Fingerling, G., et al., Formation and Spectroscopic Investigation of Two Hexachlorobornanes
from Six Environmentally Relevent Toxaphene Components by Reductive Dechlorination in Soil
under Anaerobic Conditions, Environ. Sci. Technol. 1996, 30, pp. 2984-2992.
Gill, U., et al., Congener-Specific Analysis of Toxaphene in Serum Using ECNI-MS,
Chemosphere, Vol. 33, #6, pp. 1021-1025.
Jacobson, J.L., et al., Intellectual Impairment in Children Exposed to Polychlorinated Biphenyls
inUtero. New England Journal of Medicine. Sept. 12, 1996; 335(11); pp. 783-789.
Maruya, K., et al., Prominent Chlorobornane Residues in Estuarine Sediments Contaminated
with Toxaphene. Environmental Toxicology and Chemistry. 2000, vol.19, #9, pp. 2198-2203.
Maruya, K., et al., Selective Persistence and Bioaccumulation of Toxaphene in a Coastal
Wetland. American Chemical Society, Chapter 12, 2001, pp. 164-174.
Maruya, K., et al., Residues of Toxaphene in Finfish and Shellfish from Terry and Dupree
Creeks, Georgia, U.S.A. Estuaries, August 2001, vol. 24, #4, pp. 585-596.
31
-------
McHugh, B., et al., The Occurence and Risk Assessment of Pesticide Toxaphene in Fish from
Irish Waters. 2003, Marine Institute, Abbotstown, Dublin 15, ISSN #1649-0053.
Olson, K.L., et al., Behavioral Effects on Juvenile Rats from Perinatal Exposure to Low Levels
of Toxaphene, and its Toxic Components, Toxicant A, and Toxicant B. Arch. Environ. Contam.
Toxicology, 1980, vol. 9, #2, pp. 247-257.
OSWER Directive 9200.4-17P, Use of Monitored Natural Attenuation at Superfund, RCRA
Corrective Action, and Underground Storage Tank Sites, April 21, 1999.
Skopp, S., et al., Enantiomer Ratios, Patterns, and Levels of Toxaphene Congeners in Human
Milk in Germany. J. Environ. Monit. 2002, vol. 4, pp. 389-394.
Smalling, K., et al., Elimination of Toxaphene Residues by the Mummichog (fundulus sp.),
2004.
32
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
REGION 4
ATLANTA FEDERAL CENTER
61 FORSYTH STREET
ATLANTA, GEORGIA 30303-8960
Jung 23, 2005
MEMORANDUM
SUBJECT: Comments on the Draft Ombudsman Report
Appropriate Testing and Timely Reporting Are Needed at the Hercules 009
Landfill Superfund Site, Brunswick, Georgia; Assignment 2004-124
FROM: J. I. Palmer, Jr.
Regional Admini
TO: Paul D. McKechnie
Acting Ombudsman
Office of Congressional and Public Liaison
Thank you for the opportunity to provide comments on the draft report. The following
comments relate to the subject document, and are provided primarily to address the action
required in EPA Manual 2750. In accordance with the memorandum transmitting the draft
report, we have followed the instructions to "address the factual accuracy of the draft report and
indicate concurrence or noncurrence with each finding and proposed recommendation. If you do
not concur with the proposed recommendation, please provide any alternative actions you wish
to be considered for the final report." The comments are a consolidation of input from Region
4's Analytical Support Branch of the Science and Ecosystem Support Division, and the
Superfund Remedial and Technical Support Branch of the Waste Management Division. In
addition, the Region has included comments from the Region 4 Office of Environmental
Accountability (OEA) that address noted excerpts in the draft report.
Excerpt from the Draft Ombudsman Report Page 8, Recommendation 2.1:
"Use negative ion mass spectroscopy to definitely determine if toxaphene breakdown products
are present in the surrounding groundwater at the Hercules 009 Landfill site, and (if so) in what
amounts."
Response to Recommendation 2.1
The Region has used negative ion mass spectrometry (NIMS) to determine the presence of
weathered toxaphene (WT). After the issuance of the Preliminary Technical Draft from the
OIG, samples were collected from the 009 Landfill monitoring wells by Hercules, Inc., under a
voluntary interim action on March 5th and analyzed by Keith Maruya at the Skidaway Institute of
\ PRO^0
trator
Internet Address (URL) • http://www.epa.gov
Recycled/Recyclable • Printed with Vegetable Oil Based Inks on Recycled Paper (Minimum 30% Poslconsumer)
-------
2
Oceanography using the NIMS method from a peer reviewed scientific publication. The
Hercules, hie., March 2005 Data Report is attached as an addendum.
Region 4's laboratory is willing to participate in a multi-laboratory method validation study for
toxaphene congeners in environmental samples. However, since the Agency as a whole would
obviously benefit from a validated NIMS method for toxaphene congeners, we believe that a
multi-laboratory method validation study should be initiated at the program level by the Office of
Solid Waste and Emergency Response. A validated method will serve both the regulated
community and the Agency by assuring that analytical data produced by the method are
defensible, of known quality, and suitable for risk assessment decision making.
Excerpt from the Draft Ombudsman Report, Page 8 Recommendation 2.2:
"If toxaphene breakdown products are found in the groundwater, assess the resulting risk to
human health and take appropriate action. "
Response to Recommendation 2.2
In response to the Preliminary Technical Draft from the OIG, Region 4 conducted a thorough
literature review for information on the toxicity of WT and reached the following conclusions:
Application of the MCL to Groundwater at 009 Landfill
The NIMS analytical results from groundwater at the 009 Landfill are all significantly
less than the MCL for technical toxpahene (TT) of 3 ug/L. Based on the Region 4
preliminary toxicity assessment, it is reasonable to assume that this MCL is protective for
WT as well as TT. The Region 4 Draft Report on WT toxicity is attached as an
addendum and a short summary is provided below.
Toxicity Criterion for Weathered Toxaphene
To develop a human toxicity criterion for WT, three choices must be made: (1) the
critical toxic endpoint; (2) the threshold dose value based on the critical endpoint; and
(3) the uncertainty/safety factors applied to determine a reference dose.
Region 4 toxicologists believe that the most appropriate endpoint for WT is tumor
promotion. This endpoint was chosen because it appears most relevant to humans, and
focusing on this endpoint is also protective of other toxic endpoints, such as immunologic
and developmental effects.
The Monitoring, Analysis and Toxicity of Toxaphene in Marine Foodstuffs (MATT)
study from the European Union is the sole toxicological study based on WT and thus
chemically is most relevant to human exposure. The critical study was performed in rats
with changes in liver cells that represent precancerous changes as the endpoint. The no
observable adverse effect level (NOAEL) from this study was 0.69 mg/kg-day as
-------
3
calculated by MATT or 0.6 mg/kg-day in the Region 4 analysis. Note these values are
very similar.
The following uncertainty factors were considered: 10 for animal-to-human extrapolation
and 10 for human variability. In spite of the paucity of studies on WT, the literature is
replete with studies on TT that cover a range of endpoints. Many of these studies
resulted in higher and less protective NOAELs. The studies that had similar values for
NOAELs were ingestion or oral gavage studies based on TT. The bulk of toxaphene
administered orally is excreted or metabolized quickly; hence, the internal doses of WT
and TT and the subsequent toxaphene body burdens end up being of similar magnitude.
Because of the large database of toxicity studies of TT, we believe no additional database
insufficiency uncertainty factor was needed. Hence, the combined uncertainty factor is
100.
This derivation of a reference dose for WT based on tumor promotion is consistent with
the recently finalized EPA cancer guidelines which state that the consideration of mode
of action vis-a-vis toxicity is paramount in the development of a toxicity criterion.
The Need for Peer Review
Unfortunately, the laboratory studies on WT toxicity that the MATT report relied on have
not yet undergone peer review because of logistical issues (one of the authors moving to
a new university). Region 4 agrees with the OIG report that additional toxicity studies
may be helpful. Peer review would also be helpful in elucidating some of the apparent
errors in interpretation of EPA cancer potency factor derivation identified in the MATT
report.
On page 19 of the OIG report, the MATT was quoted as indicating that WT is
approximately twice as carcinogenic as TT. This statement of the MATT report is
incorrect and unfortunately repeated in the OIG report. It is not entirely clear on what
basis the MATT report makes this statement, but there are two possibilities.
First, the MATT report presented a misunderstanding of the EPA TT slope factor in
which the MATT confused the units. The upshot of this misunderstanding is that the
MATT toxicity criterion appears for WT about twice as stringent at the EPA toxicity
criterion for the original TT mixture. In truth, the EPA toxicity criterion is 300 times
more stringent than the MATT criterion.
Second, the MATT interpreted some empirical data obtained from a cell culture system
and possibly made a large and unfounded conceptual leap. In this cell culture system, the
toxic endpoint was disruption of intercellular communication. Calculations indicated that
the WT was twice as effective in blocking intercellular communication as TT. However,
to make the leap of claiming that effects on intercellular communication in a cell culture
-------
4
system is tantamount to a carcinogenic response in a whole animal is a very large leap
indeed. This conceptual leap is not endorsed by Region 4 toxicologists.
Interim Strategy for Risk Assessment
In keeping with the OIG intention of using the best available science, Region 4 has two
proposals for an interim strategy. The preferred approach is to use the toxicity criterion
for WT developed by Region 4 Technical Services and based on the laboratory study in
the MATT. Presently, the Region 4 report is still in draft, but should be finalized
relatively soon. Hence, Region 4 requests that the OIG review both the MATT
laboratory study and the Region 4 derivation of toxicity criteria to determine their
soundness and applicability. An alternative approach is to use a toxicity criterion based
on TT. The EPA toxicity criterion for TT now on IRIS was last revised in 1991. The
toxicity value is based on rodent bioassays conducted in 1978 and 1979. In 2000,
Goodman1 reanalyzed these data using newer methods based on EPA guidance and
concluded that TT was actually tenfold less toxic than the IRIS value. Region 4 believes
that our preferred approach represents the best available science and would provide a
reasonable interim approach.
Excerpt from the Draft Ombudsman Report Page 13,3.1:
"Issue the report on the Hercules 009 Landfill 5-year review with the conclusion that the
protectiveness of the groundwater cleanup cannot be determined at this time, and further
evaluation is needed. A timeframe should be estimated for such an evaluation. "
Response to Recommendation 3.1:
Based on the recent evidence provided by the NIMS data and the toxicity criterion developed by
Region 4 based on the MATT laboratory study, Region 4 proposes that the data and toxicity
review be included in the release of the 5-year review and a determination of protectiveness be
issued.
Excerpt from the Draft Ombudsman Report Page 13,3.2 :
"Ensure that restrictions are placed in the property deed to control future use of the land and
groundwater."
Response to Recommendation 3.2:
When Hercules, Inc., entered into the Consent Decree for RD/RA with the U.S., it agreed to
perform all operation and maintenance activities required to maintain the effectiveness of the
1 Goodman JI, Brasick DJ, Busey WM, "Cohen SM, Lamb JC, Starr TB. (2000) Reevaluation of the cancer potency
factor of toxaphene: recommendations from a peer review panel. Toxicol Sci. 2000 May;55(l):3-16.
-------
5
Remedial Action. As set out in the Record of Decision, operation and maintenance of the multi-
media cover was to continue for a minimum of 30 years. (See Section 9.A. 1). The Consent
Decree for RD/RA requires Hercules, Inc., to not only record a certified copy of the Consent
Decree with the Glynn County registry of deeds, but to also include within any instrument
conveying any interest in the property a notice describing the restrictions applicable to the
property, the provisions of the Consent Decree with respect to institutional controls and EPA's
right of access, and the obligations of successors-in-title. In addition, Hercules, Inc., and any
successor-in-title must provide written notice to EPA of any proposed conveyance of any interest
in the title. As part of EPA's statutory mandate for a 5-year review of a remedy's effectiveness
if waste is left in place, the Consent Decree for RD/RA further obligates Hercules, Inc., to
conduct any studies and investigations that EPA might request in support of its 5-year review.
Moreover, no conveyance by Hercules, Inc., of any interest in property, however minor, will
release or otherwise affect the liability of Hercules, Inc., to comply with the terms of the Consent
Decree for RD/RA.
In light of the fact that the remedy has achieved the performance standards established in the
Record of Decision and is believed to be currently performing as designed, EPA's statutory
authorities and the enforcement-based tools arising from the Consent Decree for RD/RA are
presently believed to be adequate institutional controls. Of course, if environmental or other
conditions change, existing State-based legal authorities may in the future also be utilized to
facilitate proprietary controls, such as an environmental easement designed to protect
groundwater, if determined by EPA to be necessary.
Excerpt from the Draft Ombudsman Report Page 13,3.3:
"Confirm that no one in the vicinity is using the groundwater. "
Response to Recommendation 3.3 :
Since the perimeter monitoring wells are properly functioning as an intended early warning
system, it becomes unnecessary to document private well water use outside the perimeter.
Nonetheless, Hercules, Inc. has investigated the status of private registration of new wells in the
immediate area. In checking with the Glynn County Environmental Department, Hercules was
informed that the County has no record for the past several years of anyone advising them of the
intent to drill a private drinking water well in the immediate area of the 009 Landfill. Hercules
has also been informed by the previously affected residents that they continue to use city water
for drinking purposes. Finally, Hercules has conducted a visual canvassing of the area to locate ¦
well house structures. This effort indicates that no new private drinking water wells have been
installed in the immediate area surrounding the site.
The perimeter well monitoring system has been evaluated by a hydrogeologist in Region 4's
Technical Services Section. The ability to capture leachate was the focus of the evaluation and
the leachate capture system was found to be effective. The Final Technical Memo is attached as
-------
6
an addendum. These wells are closest to the source and allow the earliest warning signal that
downgradient wells may be affected. Since the recent NIMS and MATT comparison
demonstrate that the perimeter wells do not pose a risk and the monitoring wells act as an
effective early warning signal, it is unnecessary to obtain additional groundwater samples from
locations beyond the perimeter well system.
Attachments (6)
1. Hercules, Inc., March 2005 Data Package for 009 Landfill using NIMS
2. A Re-analysis of the European MATT (2000) Toxicity Data and Development of a Reference
Dose for Weathered Toxaphene (DRAFT)
3. Differences between Cancer and Non-Cancer Risk Assessment using Toxaphene as an
example
4. October/December 2003 Technical Memos from SRTSB/TSS describing sufficiency of the
monitoring wells to detect migration
5. Cover Memo and Complete Comments from SESD
6. Cover Memo and Complete Comments from OEA
cc w/o attachments:
Winston A. Smith, Region 4 WD
Franklin Hill, Region 4 WD
Scott Sudweeks, Region 4 WD
Ted Simon, Region 4 WD
Kay Wischkaemper, Region 4 WD
Derek Matory, Region 4 WD
Randall Chaffins, Region 4 WD
Leo Francendese, Region 4 WD
Gregory Luetscher, Region 4 WD
Gary Bennett, Region 4 SESD
Lavon Revels, Region 4 SESD
Charlie Hooper, Region 4 SESD
David Lopez, OSWER/OERR
Silvina Fonseca, OSWER/OERR
-------
Appendix C
OIG Technical Comments on the Region 4 Response
OIG Draft Recommendation 2.1:
"Use negative ion mass spectroscopy to definitely determine if toxaphene breakdown products
are present in the surrounding groundwater at the Hercules 009 Landfill site, and (if so) in what
amounts."
OIG Technical Comments on Region 4's Response:
The groundwater at the Hercules 009 Landfill was recently analyzed using negative ion mass
spectroscopy (NIMS). In March 2005, Hercules Incorporated voluntarily collected an additional
set of groundwater samples from the Hercules 009 Landfill's monitoring wells. These samples
were subsequently analyzed by EnChem using Method 8081A and also by Keith Maruya from
the Skidaway Institute of Oceanography using a NIMS methodology. EnChem found no
measurable concentration of chlorinated camphene in monitoring well N-l 1. Skidaway found an
estimated concentration of 0.74 |ig/L (a.k.a. parts per billion, or ppb) of chlorinated camphene in
monitoring well N-l 1. Thus, all reported chlorinated camphene values from EnChem and
Skidaway were below the level of concern of 3.0 |ig/L for technical toxaphene.
The OIG's review of the scientific literature regarding toxaphene found several sources that
identified potential problems with analyzing for toxaphene degradation products. The Agency
for Toxic Substances and Disease Registry's (ATSDR's) Toxicological Profile for Toxaphene
reported that sulfuric acid cleanup of toxaphene sample extracts modify the toxaphene peak
profile (ATSDR, 1996, p. 147). This suggests that the sulfuric acid cleanup procedure has the
potential to destroy or alter the ratio of toxaphene degradation congeners. In 1997, Alder
reported that gas chromatograph injector temperatures greater than 240° C have been observed to
decompose or to decrease the observed response of the toxaphene congeners p50 and p62 prior
to their measurement by the detector (Alder 1997). In 1995, Andrews found during a round-
robin study on the analysis of toxaphene that only 15 to 30 percent of the toxaphene components
were eluted from the silica or florisil columns with a nonpolar solvent (Andrews 1995). These
observations indicate that the analysis for toxaphene degradation products is difficult and the
analytical results can vary significantly depending on how the analysis was conducted.
The case narrative for the March 2005 Hercules 009 Landfill data from both EnChem and
Skidway indicated that sulfuric acid was used to cleanup the extracts and that the gas
chromatograph injector temperature may have been as high as 250° C. The OIG is concerned
that the sulfuric acid cleanup and that the gas chromatograph injector temperature, if it exceeded
240° C, may have lowered the amount of chlorinated camphene detected.
Following OIG discussions with Region 4 concerning these procedural issues with the method,
Region 4 provided additional test results from Region 4's Science and Ecosystem Support
Division (SESD). For their analyses, the sample extracts were not subjected to sulfuric acid
39
-------
cleanup and the gas chromatograph injector temperature was known to be below 240° C. In data
dated November 28, 2004, and January 13, 2005, SESD reported chlorinated camphene sample
results for monitoring well N-l 1 of 4.0 JN |ig/L and 6.4 JN |ig/L, respectively. These results are
significantly higher than Enchem's and Skidaway's chlorinated camphene results. SESD's
chlorinated camphene results of 6.4 |ig/L for monitoring well N-l 1 are 8.6 times greater than
Skidaway's chlorinated camphene results of 0.74 |ig/L. SESD's reported chlorinated camphene
results are above the level of concern of 3.0 |ig/L for technical toxaphene.
These results for chlorinated camphene show that the analysis of toxaphene is a complicated
process in which the type and skill used in the extraction, cleanup, and testing procedures can
directly impact the reported analytical results. To verify and document that all analytical
variables are under control, a laboratory normally analyzes a spiked laboratory quality control
sample (i.e., positive control sample) to confirm that the laboratory's processing of the sample
extract did not destory or lose the compound(s) of interest (i.e., Hx-Sed, Hp-Sed, p26, and p50).
Instead, all three laboratories (i.e., EnChem, Skidaway, and SESD) spiked the laboratory quality
control sample with technical toxaphene. This quality control procedure demonstrates the
performance of the method used adequately detects and measures technical toxpahene, but does
not demonstrate the method's performance on individual toxaphene congeners. To document
how well the method performs with the individual congeners of interest, such as Hx-Sed, Hp-
Sed, p26, and p50, the laboratories needed to spike and measure the recoveries for the individual
toxaphene degradation products (i.e., Hx-Sed, Hp-Sed, p26, and p50) and not technical
toxaphene.
An example of this congener-specific quality control spiked sample can be seen in Foreid's 2000
paper referenced by Keith Maruya's letter dated August 4, 2005 (Foreid 2000). Foreid
specifically spikes p26, p50, and p62 and demonstrates spiked sample recoveries for each
congener of greater than 94 percent. These quality control results clearly demonstrated that the
specific silica fractionation and sulphuric acid cleanup procedure used by Foreid did not destroy
or degrade the toxaphene congeners of interest. This type of quality control results on individual
congeners would have greatly assisted in interpreting the EnChem, Skidaway, and SESD results
from the Hercules 009 Landfill. Unfortunately, this type of quality control results were
unavailable for the Hercules 009 Landfill data because there is no standardized, validated, or
EPA approved method to analyze for toxaphene degradation products. The OIG believes that
spiked sample recoveries of the individual toxaphene congeners of interest, such as Hx-Sed, Hp-
Sed, p26, p50, and p62, should be incorporated into any NIMS method considered by the EPA
for developing and validating.
The OIG agrees with Region 4 that developing and validating a NIMS method for toxaphene
congeners needs to be initiated and supported at the national program office level. Due to the
degraded nature of the remaining toxaphene contamination left in the environment (i.e., a
substantial amount due to the bulk use of toxaphene over many years), the NIMS method is
needed by the EPA to accurately determine the extent of the remaining contamination in the
environment and to provide toxicologically-relevant analytical data for assessing the risk to
human health from these remaining toxaphene breakdown congeners. Therefore, future
groundwater monitoring at the Hercules 009 Landfill should use, if developed and validated, the
EPA-approved NIMS method.
40
-------
The OIG disagrees with the statement in the Hercules Incorporated cover letter dated April 11,
2005, that"... the ECD-NCIMS method is essentially the same as Method 2 (Total Area Under
the Curve)..The NIMS method has the ability to positively identify the presence of
individual toxaphene congeners and to measure the concentration of individual toxaphene
congeners, while Method 2 can not. Method 2 does not provide congener specific information
on toxaphene that is needed to evaluate the risk to humans from the potential exposure to the
persistent toxaphene degradation congeners. Thus, these two analytical methods are not
"essentially" the same.
OIG Draft Recommendation 2.2:
"If toxaphene breakdown products are found in the groundwater, assess the resulting risk to
human health and take appropriate action"
OIG Technical Comments on Region 4's Response:
Regarding assessing the toxicity of toxaphene, since humans are exposed to toxaphene
degradation products (e.g., mainly p26 and p50) and since individual congeners can have
dramatically different toxicities, the toxicity criteria for weathered toxaphene must be developed
and used. The OIG believes that the toxicity criteria for technical toxaphene should not be used,
because the public is no longer exposed to the original mixture of toxaphene congeners.
Regarding a critical toxic endpoint, the OIG agrees with Region 4 that weathered toxaphene acts
as a tumor promoter and not as a tumor initiator (i.e., causing direct damage to deoxyribonucleic
acid, also known as DNA). The OIG also agrees with Region 4 that the use of the reference dose
represents one of the first instances in which EPA has applied a threshold-type toxicity to a
potentially carcinogenic compound. Due to the novelty of the application under the EPA's new
cancer guideline, developing a reference dose for weathered toxaphene needs to be peer-
reviewed and a consensus reached both internal and external to the EPA before a threshold limit
for expose to toxaphene degradation products can be set and used for risk assessment.
Region 4's response stated that the MATT study conducted by the European Union is the sole
information on the toxicity of weathered toxaphene. However, Region 4's response indicated
that the MATT study may have some technical and conceptual issues that a pending peer review
may elucidate. The MATT study does not clarify the mode of action as to whether the ultimate
carcinogen is from the persistent congeners or from non-persistent congeners. Therefore, the
OIG believes that the MATT study results alone do not provide sufficiently robust toxicity data
to support setting a threshold limit for weathered toxaphene. Consequently, the OIG believes
additional research is needed to characterize and assess the mode of action of the toxaphene
degradation products to generate a sufficient quantity and of high-quality toxicity data on
weathered toxaphene to support a threshold limit. The OIG agrees with Region 4 that additional
research should focus on the toxicity of the persistent toxaphene degradation products, namely,
p26 and p50, but also the less abundant products p62, p 40, p41, and p44, because they are most
likely to produce the toxic effects.
41
-------
OIG Draft Recommendation 3.1:
"Issue the report on the Hercules 009 Landfill 5-year review with the conclusion that the
protectiveness of the groundwater cleanup cannot be determined at this time, and further
evaluation is needed. A timeframe should be estimated for such an evaluation."
OIG Technical Comments on Region 4's Response:
The Ombudsman contends that the 5-year review can be issued stating that the cleanup of the
Hercules 009 Landfill protects human health for the following reasons:
The OIG observed a reoccurring pattern in the Hercules 009 Landfill groundwater
monitoring data from 2002, 2004, and 2005. The analytical results from monitoring wells N-
06SR and N-l 1 suggest that toxaphene degradation products (i.e., measured as chlorinated
camphenes) are present and may exceed the 3.0 |ig/L level of concern for technical toxaphene at
these locations. Monitoring wells N-06SR and N-l 1 always had the highest observed
concentrations of chlorinated camphenes during each individual sampling event. By contrast, the
analytical results from the remaining monitoring wells for the site, many of which are down
gradient of wells N-06SR and N-l 1, suggest that the toxaphene degradation products in the
groundwater have not left the Hercules 009 Landfill property. Thus, the surrounding community
is not being exposed to toxaphene degradation products through the groundwater exposure
pathway.
No groundwater wells are being used in the vicinity of the Hercules 009 Landfill.
Although a deed restriction was not placed on the Hercules 009 Landfill property to
prevent drilling and using wells on the property in the future, the consent decree was recorded in
the records of Glynn County.
42
-------
References
Agency for Toxic Substances and Disease Registry (ATSDR), August 1996. Toxicology Profile
for Toxaphene. U.S. Department of Health and Human Services, Public Health Service.
Alder L, Bache K, Beck H, and Parlar H, 1997. Collaborative Study on Toxaphene Indicator
Compounds (Chlorobornanes) in Fish Oil. Chemosphere 31:1391-1398.
Andrews P, Headrick K, Pilon J, Lau B, Weber D, 1995. An Interlaboratory Round Robin Study
on the Analysis of Toxaphene in a Cod Liver Oil Standard Reference Material. Chemosphere
31:4393-4402.
Foreid S, Rundberget T, Severinsen T, Wiig O, Skaare JU, 2000. Determination of Toxaphenes
in Fish and Marine Mammals. Chemosphere 41:521-528.
43
-------
Distribution
Office of the Administrator
Regional Administrator, Region 4
Assistant Administrator for Solid Waste and Emergency Response
Assistant Administrator for Water
Assistant Administrator for Research and Development
Director, Waste Management Division, Region 4
Director, National Center for Environmental Assessment, Cincinnati
RCRA National Organic Methods Program Coordinator
Director, Office of Superfund Remediation and Technology Innovation
Agency Followup Official (the CFO)
Deputy Chief Financial Officer
Agency Followup Coordinator
Audit Coordinator, Region 4
Audit Coordinator, Office of the Administrator
Audit Coordinator, Office of Solid Waste and Emergency Response
Audit Coordinator, Office of Water
Audit Coordinator, Office of Research and Development
Associate Administrator for Congressional and Intergovernmental Relations
Associate Administrator for Public Affairs
General Counsel
Inspector General
Appendix D
44
------- |