United States
Environmental Protection
Agency
Office of
Emergency and
Remedial Response
EPA/ROD/R02-88/054
February 1988
$EPA
Superfund
Record of Decision
York Oil Company, NY
-------
50275 101
REPORT DOCUMENTATION ' «EPO*T NO.
2.
PAGE
EPA/ROD/R02-88/054
I 3. Recipients Accession No.
4. Title and Subtitle
SUPERFUND RECORD OF DECISION
York Oil, NY
irst Remedial Action
5. Re
'. Author(s)
1 8. Performing Organization Rept. No.
9. Performing Organization Name and Address
10. Preject/Task/Work Unit No.
12. Sponsoring Organization Name and Address
U.S. Environmental Protection Agency
401 M Street, S.W.
Washington, D.C. 20460
11. Contracted or Grant(G) No.
(o
13. Type of Report & Period Covered
800/000
14.
IS. Supplementary Notes
16. Abstract (Limit: 200 words) ' : ~ "
The York Oil Company site, encompassing 17 acres, is located in the Hamlet of Moira,
Franklin County, New York. Wetlands and woodlands are the principal land use in the
vicinity of the site. Residences exist along the main roads interspersed with
active/inactive agricultural and pasture land. The now dissolved York Oil Company
operated a waste oil recycling facility from 1964 to 1977, Crankcase industrial oils,
with some containing PCBs, were collected from sources throughout New England and New
York, then stored and/or processed at the site in eight above-ground storage tanks, a
series of three earthen-damned settling lagoons, and at least one below-ground storage.
tank. The recycled PCB-contaminated oil was either sold as No. 2 fue^ oil or used in =
dust control for the unpaved roads in the vicinity of the site. During heavy rains and
spring thaws, the oil-water emulsion from the lagoons would often overflow onto the
surrounding lands. In lieu of paying damages to adjacent farm-owners, in 1964 the oil
company purchased land in the area of the spills. The York Oil site contamination was
first discovered in 1979 by a New York state Department of Transportation road crew, who
then notified the New York State Department of Environmental Conservation (NYSDEC).
(See Attached Sheet)
17. Document Analysis a. Descriptors
Record of Decision
York Oil, NY
First Remedial Action
Contaminated Media: gw, sediments, soil, sw
PCBs, phenols, VOCs
c. COSATI Field/Group
Availability Statement
19. Security Class (This Report)
None
20. Security Class (This Page)
None
21. No. of Pages
210
22. Price
(See ANSI-Z39.18)
See Instructions on Reverse
OPTIONAL FORM 272 (4-77)
(Formerly NTIS-35)
Department of Commerce
-------
DO NOT PRINT THESE INSTRUCTIONS AS A PAGE IN A REPORT
INSTRUCTIONS
Optional Form 272, Report Documentation Page is based on Guidelines for Format and Production of Scientific and Technical Reports.
AHSI Z39.18-1974 available from American National Standards Institute, 1430 Broadway, New York, New York 10018. Each separate!,
bcund reportfor example, each volume in a multivolume setshall have its unique Report Documentation Page.
1. Report Number. Each individually bound report shall carry a unique alphanumeric designation assigned by the performing orga-
nization or provided by the sponsoring organization in accordance with American National Standard ANSI Z39.23-1974, Technical
Report Number (STRN). For registration of report code, contact NTIS Report Number Clearinghouse, Springfield, VA 22161. Use
uppercase letters, Arabic numerals, slashes, and hyphens only, as in the following examples: FASEB/NS75/87 and FAA/
R0-75/09.
.?. Leave blank.
3. Recipient's Accession Number. Reserved for use by each report recipient.
4. Title and Subtitle. Title should indicate clearly and briefly the subject coverage of the report, subordinate subtitle to the main
title. When a report is prepared in more than one volume, repeat the primary title, add volume number and include subtitle for
the specific volume.
5. Report Date. Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g.,
date of issue, date of approval, date of preparation, date published).
6. Sponsoring Agency Code. Leave blank.
7. Author(s). Give name(s) in conventional order (e.g., John R. Doe, or J. Robert Doe). List author's affiliation if it differs from
the performing organization.
8. Performing Organization Report Number. Insert if performing organization wishes to assign this number.
9. Performing Organization Name and Mailing Address. Give name, street, city, state, and ZIP code. List no more than two levels of
an organizational hierarchy. Display the name of the organization exactly as it should appear in Government indexes such as
Government Reports Announcements & Index (GRA & I).
iO. Project/Task/Work Unit Number. Use the project, task and work unit numbers under which the report was prepared.
H. Contract/Grant Number. Insert contract or grant number under which report was prepared.
12. Sponsoring Agency Name and Mailing Address. Include ZIP code. Cite main sponsors.
13. Type of Report and Period Covered. State interim, final, etc., and, if applicable, inclusive dates.
14. Performing Organization Code. Leave blank.
15. Supplementary Notes. Enter information not included elsewhere but useful, such as: Prepared in cooperation with . . . Translation
of... Presented at conference of ... To be published in ... When a report is. revised, include a statement whether the new
report supersedes or supplements the older report.
16. Abstract. Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the
report contains a significant bibliography or literature survey, mention it here.
17. Document Analysis, (a). Descriptors. Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms
that identify the major concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b). Identifiers and Open-Ended Terms. Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c). COSATI Field/Group. Field and Group assignments are to be taken from the 1964 COSATI Subject Category List. Since the
majority of documents are multidisciplinary in nature, the primary Field/Group assignment(s) will be the specific discipline,
area of human endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group
assignments that will follow the primary posting(s).
i'J. Distribution Statement. Denote public releasability, for example "Release unlimited", or limitation for reasons other than
security. Cite any availability to the public, with address, order number and price, if known.
l-T>. & 20. Security Classification. Enter U.S. Security Classification in accordance with U.S. Security Regulations (i.e., UNCLASSIFIED).
21. Number of pages. Insert the total number of pages, including introductory pages, but excluding distribution list, if any.
21. Price. Enter price in paper copy (PC) 'and/or microfiche (MF) if known.
: '.Ji3 0 - 33i-5J!6 (3393) OPTIONAL FORM 272 BACK (4-77)
!l^^
-------
EPA/ROD/R02-88/054
York Oil, NY
First Remedial Action
16. ABSTRACT (continued)
Several removal actions have been undertaken at the site to contain the
PCB-contamination problem. Currently, soils, sediments, ground water, and surface water
are contaminated with PCBs, VOCs, metals, and phenols.
The selected remedial action for this site includes: excavation of 22,000 yd-* of
contaminated soil and 8,000 yd^ of contaminated sediments with onsite solidification
followed by onsite disposal of treatment residuals; installation of deep ground water
drawdown wells and shallow dewatering wells to collect the sinking contaminant plume and
the oil during excavation, with onsite treatment and subsequent discharge in accordance
with New York State NPDES requirements; offsite thermal treatment of approximately
25,000 gallons of contaminated tank oils in addition to other soils collected at the
site; cleaning and demolition of the empty storage tanks; and grading. Treatability
studies will determine the effectiveness of the solidification process and optimal
treatment system for groundwater. The estimated capital cost for this remedial action
is $6,500,000, with present worth O&M of $500,000.
-------
DECLARATION FOR THE RECORD OF DECISION
SITS NAME AND LOCATION
York Oil Comoany site, Hamlet of Moira, Franklin County, New York
STATEMENT OF PURPOSE AND BASIS
This decision document represents the selected remedial action
for the York Oil Company site, developed in accordance with
the Comprehensive Environmental Response, Compensation, and
Liability Act of 1980 (CERCLA) 42 U.S.C. S 9601, et_. sea. , as
amended by the Superfund Amendments and Reauthorization Act
of 1986, and to the extent practicable, the National Oil and
Hazardous Substances Pollution Contingency Plan (NCP) 40 C.F.R.
Part 300, November 20, '1985.
This decision is based upon the Administrative Record for the
York Oil Company site. The attached index identifies the items
which comprise the Administrative Record upon which the selection
of a remedial action is based. Among the documents relied upon
are: .
Remedial Action Master Plan, CDM, November 1982.
0 Remedial Investigation/Feasibility Study, Erdman, Anthony,
Associates, August 1985.
Addendum Feasibility Study, Erdman, Anthony, Associates,
November 1987.
* Attached Summary of Remedial Alternative Selection for the
York Oil Company Site.
0 Attached Responsiveness Summary, January 1988.
A copy of the Administrative Record is located at the following
locations:
0 Moira Town Hall
North lawrence Road
York 12957
New York State Department of
Environmental Conservation
Route 86
Ray Brook, New.. York 12977
0 U.S. Environmental Protection Agency
Emergency and Remedial Response Division
26 Federal Plaza
New York, New York 10278
-------
-2-
DESCRIPTION OF SELECTED REMEDY
The remedial alternative presented in this document is the first
operable unit of a permanent remedy for the York Oil site. The
remedy includes:
0 Excavation of approximately 30,000 cubic yards of contami-
nated soils followed by on-site solidification of this
material. This volume includes excavation of approximately
8,000 cubic yards of contaminated sediments from the adjacent
strip of land west of the site;
* Installation of deep groundwater drawdown wells along the
southern and western perimeter of the site to collect the
sinking contaminant plume, and installation of shallow dewatering
wells to collect contaminated groundwater and oil during
excavation;
*
* On-site treatment of the collected contaminated groundwater
and oils with subsequent discharge of the treated groundwater
in accordance with New York State Pollutant Discharge and
Elimination System requirements;
»
0 Off-site thermal treatment of approximately 25,000 gallons of r
contaminated tank oils in addition to other oils collected at
the site in accordance with the Resource, Conservation and
Recovery Act 40 CFR S 264 Subpart 0 and the Toxic Substances
Control Act 40 CFR S 761.70 requirements;
* T
Cleaning and demolition of the empty storage tanks;
* On-site disposal of the solidified soil and surface grading; and
0 Since the solidified soil will remain on-site, the remedy will
be reviewed every five years to assure that human health and the
environment are being protected.
*
0 Treatability studies during the remedial design to determine
the effectiveness of the solidification process and to determine
the optimal treatment system for the contaminated groundwater.
Should the treatability study determine that solidification
would not grovide the desired degree of treatment, then a
treatability study would be performed to determine the effec-
tiveness of thermally treating the soils at the site.
This Record of Decision (ROD) addresses only source control
measures for the York Oil site. An additional operable unit
remedial investigation/feasibility study for the contamination
pathways is in progress to further define the extent of the
contamination migration from the site. If additional remedial
actions are determined to be necessary, a ROD will be prepared
for approval of future remedial action.
-------
-3-
DECLARATIONS
Consistent with CERCLA as amended, and the NCP, I have determined
that the selected remedy is protective of human health and the
environment, attains federal and state requirements that are
applicable or relevant and appropriate, and is cost-effective.
This remedy satisfies the preference for treatment that reduces
toxicity, mobility or volume as a principal element. Finally,
it is determined that this remedy utilizes permanent solutions
and alternative, treatment technologies to the maximum extent
practicable.
The action will require future operation and maintenance (O&M)
activities to ensure the continued effectiveness of the remedy.
These activities will b'e considered part of "the aporoved action
and eligible for Superfund monies for a period of up to one year.
The O&M associated with the first three years of related deep
well pumping and associated groundwater treatment is, however,
part of the proposed remedial action. Therefore, the O&M costs
during these three years will be cost-shared with the State.
After the three years, any additional O&M pumping costs will »
be funded utilizing state funds. '
The State of New York has been consulted with and agrees with the
approved remedy (see attached).
I have also determined that the action being taken is appropriate
when balanced against the availability of Superfund monies for
use at other sites.
Datie ' ~ ^Christopher J./Daggett
Regional Administrator
-------
ROD DECISION SUMMARY
YORK OIL COMPANY SITE
NEW YORK
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
Region 2
New York
-------
TABLE OF CONTENTS
Page
Site Location and Description 1
Site History 8
Current Site Status 9
Risk Assessment 15
Risks Under Present Conditions 16
Response Requirements 17
Enforcement 17
Community Relations . 18
Alternatives Evaluation 19
Development of Alternatives/Initial Screening ... 21
Description of Alternatives 25
Detailed Evaluation and Comparison of Alternatives. 32
Selected Remedy 41
Statutory Determinations 47
Operation and Maintenance 48
Schedule ,...-. ..:. .-49
Future Actions . . . 50
ATTACHMENTS
A - Administrative Record Index
B - NYSDEC Letter of Concurrence
C - Responsiveness Summary
-------
FIGURES
Figure Page
1 - Site Location Map 2
2 - Addendum FS Study Area 3
3 - General Site Conditions 5
4 - Site Drainage Areas 6
5 - Descriotion of Selected Remedy 42
6 - Cross Section of Selected Remedy 43
-------
TABLES
Table Page
1 - Groundwater Monitoring Well Compounds
Exceeding ARARs and/or Other Criteria/Guidance . . 12
2 - Surface Water Compounds Exceeding ARARs and/or
Other Criteria/Guidance 13
3 - Summary of Soil/Sediment Data 14
4 - Inappropriate Remedial Technologies 22
5 - Appropriate Remedial Technologies 23
6 - Remedial Alternatives Summary 26
7 - Long-Temt Monitorirxj Program 28
8 - Cost Estimate for Selected Remedy - Alternative 5C. .. .45-46
-------
ROD DECISION SUMMARY
YORK OIL COMPANY SITE
NEW YORK
This summary addresses the site prooer ooerable unit, which
involves addressing the source of contamination to prevent
further contaminant migration from the site, and to eliminate
the direct contact threat posed by the site (1). A subsequent
operable unit remedial investigation/feasibility study (RI/FS)
is in progress to further define the extent of the contaminant
migration from the site.
SITE LOCATION AND DESCRIPTION
The York Oil site was placed on the National Priorities List
of known or threatened-releases in July 1982. The 17-acre
site proper, is located in a rural area in northeastern New
York State, approximately one mile northwest of the Hamlet of
Moira in Franklin County, New York (see Figure 1). It was
formerly used as a waste oil recycling facility, and eventually
the site became a tank and lagoon storage facility for PCB-laden
waste oil. The site is bounded on the north and east by private ,
homes, and on the west and south by wetlands and woodlands.
Wetlands and woodlands are the principal land use in the vicinity
of the York Oil site. Residences exist along the main roads
interspersed with active/inactive agricultural and pasture
land. Although the area is rural, there are an estimated 1,700
inhabitants within a three mile radius of the site. There are
thirteen residences, housing aoproximately forty persons,
located within one half-mile of the site, with the nearest being
approximately 300 feet from the eastern boundary of the site.
Private wells serve as the sole drinking water supply for area
residents.
The Town of Moira'3 Highway Department garage and sand/salt/
gravel stockpile is located on North Lawrence Road adjacent
to the site to the northwest, and emoloys approximately ten
persons. Adjacent to the site to the southeast exists an
abandoned milk plant. An abandoned railroad embankment, running
east/west, is the southern boundary of the site. This track
provides access to the woodlands and wetlands around the site
for hunters', hikers, and recreational vehicles.
(1)The site proper is defined as all York Oil Company lands,
including the fenced portion of the site, and the 1000 ft.
by 200 ft. strip of land located to the west of the fenced
area (see Figure 2).
-------
-2-
York Oil Rtmtdlal
Investigations Report
LOCUS PLAN
rioune NO. i
-------
FENCED LIMITS
OF THE
YORK OIL SITE
AOJACEN
WESTERN
DRAINAGE
AREA
LEGEND
. «>*f
APPROX. LIMITS OF FORMER YORK
OIL CO.,INC. (TAKEN FROM A MAP
PREPARED BY FOURTH COAST
POLLUTION CONTROL, INC. - I983)"
APPROX. LIMITS OF
PROJECT STUDY AREA
0 200 4QQ 6CX)
SCALE r»4CO'
YORK OIL SITE
MOIRA, NEW YORK
ADDENDUM FS
SITE BOUNDARIES
AND STUDY AREA
OCT. 1987
-------
-4-
The York Oil site is located within the Lawrence Brook watershed,
which drains portions of northwestern Franklin County and north-
eastern St. Lawrence County. Two major tributaries, Alburg Brook
and Joy 3rook flow north merging to form Lawrence Brook. Lawrence
Brook flows north, turning northwest near the site and .the closest
point is about 1250 feet from the site. Approximately six miles
downstream of the site, Lawrence Brook flows into Deer River.
Deer River flows into the St. Regis River and ultimately into the
St. Lawrence River.
The site is situated on the side of a glaciated hill that slopes
from the northwest to the southwest. Lawrence Brook meanders around
the glaciated hill to the east. The ground surface away from the
site decreases gradually to a drainage trench along the abandoned
railroad grade (see Figure 3). This trench connects to a poorly
defined drainage pathway flowing northwesterly through a series
of wetlands and ultimately to Lawrence Brook. South of the
railroad grade is a wetland area where_ su_rface waters_flow to the
south into Lawrence Brook, and to the north towards a culvert
under the~railroad grade, which finally drains into the western
wetlands. *
The site drainage is influenced by a man-made interceptor trench
located west of the consolidated lagoon areas, that has been con-
structed on the site during remedial work in 1980 and 1981. The
interceptor trench channels runoff from part of Area A toward
the southern portion of the site and into a drainage ditch
which runs parallel and adjacent to the north side of the
railroad embankment (see Figure 4). This drainage ditch
also picks up the remainder of Area A runoff not flowing
into the interceptor trench. It then carries flow toward
the wetlands to the west and northwest of the project site.
This runoff flows to a larger wetland area six miles north
of the site. The direction of flow beyond the wetlands
include Lawrence Brook, the Deer River and terminating with the
St. Lawrence River.
The runoff from Area B is toward the south-southeast. Near the
southern coen«r of the site, a divided flow condition exists at
the drainaq^trench with a slight easterly flow under high flow
conditions. ""Ea»f6f the'abandoned milk plant, there exists a
ditch approximately 250 feet long that provides some retention
capacity for this flow. When this capacity is exceeded, the
water then flows through an 18-inch diameter culvert beneath the
abandoned railroad tracks near the intersection of Mill Road
and North Lawrence Road.
Runoff from Area C which includes the Moira Town garage property,
flows onto the York Oil site. Approximately half of this runoff
enters Area A and is channeled into the drainage ditch which
-------
^
==?^^S^3 '/*Xt' If?/-' / - '-' Cotttot IP4T* p '
?3!i£?^=£^t*-l!*///> Xx^> lAfcoow* \
^^flgg^^-^- -.pii.
...-.-.^...^ss-v
-------
y ^
e^v
14 7 V\\ ^K 1^^ ^C^
^«J»?\ ^
ABANDONDEO R.R.
LEGEND
Q«n«f«Ui«d 8urUc« Flow
i* Culvert
York Oil Remedial
Investigations Report
SITE DRAINAGE
AREAS
FIGURE 4
fit!
> i
-------
-7-
flows toward the western wetlands. The oortion entering
Area B flows toward the 18-inch diameter culvert oreviously
described.
The Lawrence Brook 100 year storm "flood zone is not within the site
prooer boundaries. Areas adjacent to the site exoerience oeriodic
flooding as a result of high intensity storms and raoid snow melt,
but are not affected by the Lawrence Brook 100 year floodway.
RI activities indicate that subsurface conditions at the
site generally consist of glacial overburden deoosits that
overlie sedimentary bedrock. While the thickness of the over-
burden varies, it is about forty feet thick within the fenced
boundaries of the site.
0
The overburden and bedrock occur in layers of more permeable and
less oermeable materials that are capable of oroducing and/or
transmitting ground water, or are barriers to groundwater flow,
resoectively. Five soil deoosits have been identified with
resoect to groundwater flow.
A significant geohydrologic factor which exists throughout the ,
site overburden is the oresence of a glacial till layer ranging
in thickness from five to twenty feet. This confining till
layer overlying bedrock has an average" hydraulic conductivity
of approximately 3xlO~4 ft/day and is a barrier to groundwater
flow. This hydraulic characteristic is very important for the
York Oil Company site since it orotects residential wells
that are screened in the bedrock, from contamination emanating
from the site. Another factor which affects groundwater flow
is the cobble and boulder layer which may potentially be the
prominent water bearing zone.
The bedrock, narticularly along discontinuities and fracture
zones, is a transmisalve,zone capable of producing and carrying
groundwater. The majority of the nearby residents obtain
potable water from the bedrock at depths of approximately 100
feet. The bedrock groundwater contours reveal a divide located
immediately north of the site. Thus, groundwater flow in the
bedrock at tljt* »ite is to the south.
.>
Regional groundwater flow directions in the overburden deoosits
indicate a mound, within the.glaciated hill beneath the site.
The flow from this mound is generally radial and thus, the flow
generally enters the site from the north. Overburden groundwater
flows are altered by1 several localized on-site features, including
drainage trenches and the mounded fill covering the former lagoons.
Groundwater flow off the mound is generally southward, and a
divergence was observed in this area, with flows to both the
southeast and southwest. The groundwater divergence aooears to
be related to the more permeable cobble and nested boulder layer.
-------
-8-
On-site qroundwater and oil either mix with the regional ground-
water flow and continue southward, or discharge to surface locations.
The surface discharge locations include the wetlands south of the
site, the east-west trending drainage channel along the railroad
grade or the north-south trending drainage channel.
Overburden groundwater levels revealed seasonal variations during
the RI activities that range from approximately 2 to 5 feet. The
observed seasonal variations do not, however, alter groundwater
flow directions.
The quantity of water, including incoming groundwatac and infiltra-
tion, is difficult to estimate as the thick:.esc ana lateral extent
of the cobble and boulder layer is variable. Based on explorations,
however, it is estimated that the average water quantity may range
from 10 to 15 gal/min, averaged over an entire year.
A six foot high chain 'link fence surrounds the York Oil site which
was intended to limit access during remedial activities. This
fence does not, however, limit access to the adjacent strip of land
and the wetlands. The dominant surface feature at the site is an
approximately twenty-five foot high mound of sandy materials
covering former lagoons II and 12 (see Figure 3). This mound
represents materials (i.e., contaminated oil'sludges and soils) »
collected during previous site clean-uo activity, and then '
temporarily consolidated with kiln dust and sand. West of
the mound, sand covers former lagoon 13. This area is relatively
flat and contains PCB-contaminated soils.
North of the former lagoons exists two, 25/000 gallon above ground
horizontal steel storage tanks, designated as tank No. 2 and No. 3.
They contain a combined total capacity of approximately 25,000 gallons
of contaminated oil, and sludge, which was pumped into them during
past clean-up efforts. One empty subsurface horizontal steel storage
tank, with a volume estimated to be 5,000 to 10,000 gallons, also
remains on-site. Within the bermed areas near Tank 2 and 3, approxi-
mately ninety 55 gallon drums containing site cleanup equipment, are
stored, awaiting future disposal.
SITE HISTORY
The now dissolved York Oil Company, Inc. of Waltham, Massachusetts
operated a ifoste oil recycling facility in Moira, New York from
approximately 1964 to 1977. Crankcase industrial oils, some contain-
ing PCBs, were collected from sources throughout New England and
New York, then stored and/or processed at the site proper in eight
above-ground storage tanks, a series of three earthern-dammed
settling lagoons, and at least one below-ground storage
tank. The recycled 'PCB-contaminated oil was either sold as
No. 2 fuel oil or was used in dust control for the unpaved
roads in the vicinity of the site.
-------
-9-
Durinq heavy rains and soring thaw, the oil-water emulsion from
the lagoons would often overflow onto the surroundinq lands. In
lieu of oaying damages to adjacent farm-owners, in 1964 the York
Oil Comoanv ourchased land in the area of the soills.
The York Oil site contamination was first discovered in 1979 by
a road crew of the Mew York State Deoartment of Transoortat ion,
who then notified the New York State Deoartment of Environmental
Conservation (NYSDEC). The site was then an abandoned oil tank
farm and ownershio of the orooerty was confirmed as belonging to
Kenneth Peirce. Initial government resoonse to oil soills at
the site was oerformed under the the State of New York Oil Spill
Comoensation Fund, and involved various surveys, insoections,
and limited cleanuo and containment work. Analysis taken at
this time indicated that the site oresented a PCB-contamination
problem.-.
-:-.--i ,.. , . ^
Because the FCBcontaminated oils were being washed from the site
with the »uirfrac-e runoff, the U.S. Environmental Protection Agency
(USEPA)- wasr notified by the NYSDEC and subsequently undertook
several emergency actions at the site since 1979. As part of »
the US EPA emergency action, the lagoons were drained and PCB-* »
contaminated oil was,, transferred to the above-ground storage
'tanks (see Figure 2) » -the contaminated soils from the adjacent
western strio of. land, were temporarily consolidated in lagoons 1
and 2 wi'tlvkiln dust«and sand, and lagoon 3 was graded with soil
and sandy oil seeoaqe control ©Derations were initiated utilizing
drainaqe and interceotor trenches, weir/inverted oioe arrange-
ments, sorbent oads and oil booms; and a six foot chain link fence
was erected "around the site to reduce the direct contact threat
during remedial activities.
In 1980, the site^erooerty was transferred from Kenneth Peirce to
Steven 'Wood and Charles Lawrence, residents of Moira.
Current response -actions being funded by the USEPA to orevent PCB-
contaminated oil from entering the wetlands area involve, periodic
collection of contaminated surface oil in the drainage trench at
the site propvr»«nd the changing of oil sorbent pads
" ' '
CURRENT
Through a Coooertive Agreement with .tfce. USEPA, the NYSDEC comoleted
an RI/FS for the York Oil site through its contractor, Erdman,
Anthony, Associates (EAA), and it was released for public comment
in August 19fr5V Based uoon the finding's presented in the RI/FS,
a ROD was prepared, recommending source containment (i.e., perimeter
leachate" collection, deep groundwa^ej: drawdown wells, on -site
leachate treatment and. installation, of , a cap. in accordance with
RCRA). The ROD signing was oostponed, however, due "to a change
in EPA policy which now reguires pursuit of resoonse actions
that oermanently destroy, treat or recycle wastes, rather than just
-------
-10-
contain them, and the failure of the RI/FS to adequately
assess the extent of contamination at the site proper.
Following delays which were due to liability insurance problems
in New York State, an addendum FS to further define the
extent of contamination at the site proper and to evaluate
alternatives that utilize permanent solutions and treatment
technologies to the maximum extent practicable was initiated
in December 1986. The addendum FS was completed by EAA and
released for public comment in November 1987. The following
is a brief summary of the types and concentrations of contam-
inants detected at the site.
The site presently consists of two above-ground steel storage
tanks containing approximately 25,000 gallons of PCB-contaminated
oil, one empty subsurface steel storage tank, one graded
unlined lagoon, and two consolidated unlined lagoons forming-
a mound approximately twenty-five feet high, containing
PCB-contaminated soils and sludges, temporarily capped with
kiln dust and sand. Approximately 30,000 cubic yards of
contaminated soil are present at the site and this volume is
based on a cleanup level of 10 ppm for PCBs in soils. This
level is derived based on the current New York State goal to
cleanup uncontrolled PCB sites in residential areas to a
level less than 10 ppm and on the USEPA Toxic Substances
Control Act (TSCA) PCB Spill Clean-Up Policy which states
that PCB spills in residential areas be cleaned-up to 10 ppm.
-------
-11-
Th e RI/FS indicates that the primary contaminants at the
site are PCBs (maximum concentration of 230 opm), heavy
metals, including copper, lead (maximum concentration of
16,000 ppm) and zinc, volatile organics and total ohenolics.
These contaminants have been detected in oils in tanks,
soils/sediments, surface waters and groundwater monitoring
wells. A list of compounds detected in groundwater
monitoring wells and surface water at the site orooer
that exceed their respective applicable or relevant and
appropriate requirements (ARARs) and/or other criteria/
guidance to be considered are listed in Tables 1 and 2,
respectively. Residential wells currently in use have
not been impacted by the site. Table 3 lists the range
of concentrations of eompounds detected in soils/
sediments along with their frequency of detection.
Air quality monitoring conducted during the RI/FS was
performed to assess baseline concentrations of hazardous
materials in the air, to characterize air-borne contaminants
released during drilling activities, and to assess require-
ments for worker health and safety. The analytical
results reveal no elevated background concentrations of
hazardous materials in the air. It is, therefore, unlikely
that airborne contamination is a primary machanism for
migration of contaminants.
The contaminant pathways from the site are primarily
through surface water as overland flow to drainage paths
and low lying areas, and through the groundwater, either
as dissolved or floating contaminants, depending on their
chemical nature. The extent surface water contamination,
based upon oil and sediment analysis for PCBs, is wide-
spread. Oil was observed floating in a drainage trench
over two miles downstream of the site. Sediments from
these locations have measureable PCS concentrations
indicating the liklihood that oil from the site migrated
in these surface waters for long distances, and have
probably discharged to Lawrence Brook. The contamination
pathway RT/FS .will more accurately define the contamination
in these areas.
PCBs tend to be in higher concentrations in lower lagoon
#3, in the floating oil layer beneath the mound (lagoon
II and #2) and in oils that have migrated via overland
flow from the site. The highest level of PCBs in soil/
sediments is located in the adjacent 1000 ft x 200 ft strip
of land west of the site at 210 ppra. Even though this
area has been previously cleaned, migration of PCBs from
-------
-12-
Table 1
GROUNDWATER MONITORING WELL COMPOUNDS EXCEEDING ARARS
AND/OR OTHER CRITERIA/GUIDANCE TO BE CONSIDERED
Comoound
Arsenic 9/16
Benzene 1/20
Cadmium 1/16
Chromium 4/16
Copper 9/16
2,4 Dimethyl
phenol 1/9
Lead 9/16
Nickel 9/16
Phenol 1/9
Total Phenolics 7/9
Polychlorinated
Biphenyl (PCS) 10/27
Trans- 1-2
Dichloroethene 6/20
Xylene 1/20
Zinc 16/16
Range of Concen.
in Detected
(1) Samples (ppb)-
2-110
63
14
70-950
28-920
260
66-15,000
42-940
. 2,000
7-3,420
1-26,900(4)
90-2,000
210
68-2,500
Standard/Guideline
(ppb)
25/0
ND/0
10/0
50/50
200/1000
1/50 & 100(3)
25/50
-/15.4
1/50 & 100(3)
1/50 & 100(3)
.1/0
-/SO & 100
-/50 & 100
300/5,000
Source(2)
703.5/CWA
703.5/RMCL
MCL/CWA
MCL/CWA
170.4/703.5
703.5/*
703.5/CWA
-/CWA
703.5/*
703.5/*
703.5/CWA
-/*
-/* .
170.VCWA
(1) Frequency is the number of positive observations divided by the
number of samples analyzed. Does not include duplicates, blanks
or matrix spikes. : ' .
(2) Sources for enforceable standards are as follows:
703.5- NYS Water Quality Regulations- Standards for groundwater
170.4- NYS Sources of Water Supply- Standards of raw water quality
MCL- EPA National Primary Drinking Water Regulations- Maximum
Contaminant Levels
Sources for the guidelines are as follows:
CWA- Clean Water Act Water Quality Criteria
RMCL- EPA Recommended Maximum Contaminant Levels
143.3- EPA National Secondary Drinking Water Regulations- Secondary
MCLs
* -The NYS ambient water quality guideline is 50 ppb for any single
organic compound or 100 ppb for the total of designated organic
compound**^
(3) The New York State groundwater standard for phenols is 1 ppb,
however, this is a secondary standard for aesthetic purposes
only. For this site, the NYSDOH has advised that the NYS ambient
water quality guideline of 50/100 ppb be utilized.
(4) PCBs are concentrated in the floating oil layer in the shallow
water table, however, they are reported here as being present in
the groundwater.
-------
-13-
Table 2
SURFACE WATER COMPOUNDS EXCEEDING ARARS
AND/OR~OTHER CRITERIA/GUIDANCE TO BE CONSIDERED
Maximum Concen.
Frequency in Detected Criteria (2)
Compound (%) (1) Samples (ppb) (ppb)
Arsenic 100 4 0
Lead 100 5,900 50
Nickel 100 430 15.4
Polychlorinated
Biphenyls (PCS) 100 120,000 .013
(1) Frequency is the number of positive observations divided
by the number of samples analyzed times 100. Does not
include duplicates, blanks or matrix spikes.
(2) Water Quality Criteria (WQC)- Established under the Clean
Water Act as non-regulatory criteria to protect human
health from exposure to contaminated water, fish, and fish
products, as well as to protect aquatic life.
ND - Analyzed but:not detected. -
-------
I
c
c
c
c
0
i
-u-
OF SOIL /SEDIMENT DATA
Range of
Concentrations
in Detected
Acenaphthylene 2/1i
Acetone !/,? 1.25 - 2.8
Araenic 7/f, 1.6*
Barium i>ia 2.5 - 8.2
Benzene ?,,, 2.2*
Bis (2-ethylhexyl) ' «-7i - 9.4
Phthalate 2/li
Chloroform J/JJ »-28 - 22
4/11 -'- " 2S
1/13
18/1]
-- 9/11 -
pibromochlorcmethane 1/13 3-7.-.25*
H/U _-
Copper 9/u *« - '9
Chromium" Jifr *»*
9/11
1/13
3/11
5/13 § 85 _ j~
nnorene */** 3 /4
1,2 Dichlorobenzene 3/11 . M1-f*
Bthylbenzene c/t? .82 - 6.5
Lead 4ii?, 1.5 - 3."4
A»/.U 4.8 - 16,3 M
Mercury 7/11 i.ii4 - i.i64
Methyl ethyl ketone 1/13 8.3*
2-methyl napthalene 3/5 18 - 62
4-methyl-2-pentanone 1/13 "
f *-*«**«« 3/iJ 8.896-26
Maphthalene 5/16 7.7-64
lickel 8/11 3.5-18
Pentachlorophenol 2/18 5.4 - 7.1
Phenanthrene 4/18 3.3 - 8.8
Phenol 2/18 2.7 - 3.2
Polychlorinated
Biphenyla (PCBa) 46/186
8.1 - 210
SUver J^tJ 1.1*
tjtr.chroro.thyl.ne 7/13 ....n^
Tran«-l,2 V" 1«2 - 188
dichloro.theae 3/11 . ««
lrlrl Trlttloco.than. 2/13 I'll ' f'?
TricMoro*thyl.n. 8/13 1'fi ' hi
Xylene «/tt .13 ^58
Xinc JiJt,
. U/u 3^8.-2.520
«/ii *«» A^V
um "« - »
* - One sample with compound detected.
frequencyi
-------
-15-
the site into this area continues to occur. Of the total
estimated 30,000 cubic yards of contaminated soils/sediments,
approximately 8,000 cubic yards is attributible to this area.
The second primary contaminant pathway from the site is
through the groundwater. Soluble and insoluble contaminants
at the site are migrating through the .groundwater. Insoluble
or floating contaminants (PCB-laden oil and volatile organics)
were detected in the shallow water table monitoring wells and
well points. The contaminant plume is concentrated around
former lagoon #3 and is moving and spreading southwards
towards the drainage trench along the abandoned railroad
grade and the southern wetlands. As noted in Table 1, PCBs
are reported as being*present in the groundwater, however,
they are concentrated in the floating oil layer in the shallow
water table. PCBs were not detected in the deep monitoring
wells screened in bedrock.
The water soluble contaminants (total phenolics ) are migrating
southward. As they migrate, they tend to sink into the deeper »
groundwater sources. The phenolics plume is the only deep .r
plume exceeding an ARAR. The New York State groundwater
standard for phenols is 1 ppb, however, this is a secondary
standard for aesthetic purposes only. For this site, the New
York State Department of Health (NYSDOH) has advised that for
phenols, the NYS ambient water quality guideline of 50 ppb for
any single phenolic comoound or 100 ppb for the total of phenolic
compounds be utilized.
RISK ASSESSMENT
The health risk assessment conducted in the FS evaluates
potential risks associated with threatened or actual exposure
at the York Oil site. This baseline risk assessment evaluates
the possible impacts of si'te conditions on human health under
baseline conditions of a projected no-action remedial alternative,
Since ther« is a contamination pathway RI/FS underway at the
site to evaluate the potential contaminant pathway impacts,
the following-risk assessment is based upon site proper
environmental data only. Using maximum concentrations observed
for each environmental media (i.e., groundwater, surface water
and soils;, the risks associated with a "worst case" scenario
are presented. The findings of the contamination pathway RI
will permit the computation of a more likely scenario based
on actual contaminant concentrations at the point of exposure.
-------
-16-
The health risk assessment for the York Oil site has identi-
fied and analyzed potential health risks posed by the presence
of hazardous chemicals at the site. A brief summary of the
process used in this anaysis is presented below.
A characterization of contaminant sources was prepared based
on both the data collected in the field study portion of the
current study and on the results of past studies. A selection
process was then develooed and imolemented to choose indicator
compounds for the detailed toxicological and risk evaluations.
Indicator compounds were chosen usinq criteria which eliminated
from consideration those chemicals whose presence at the site
did not significantly contribute to the hazard posed.
Potential receptor populations were identified, where possible,
and screened based on assumed exoosure oathways. Exposure
levels for pathways which passed the screening process were
estimated for the most sensitive identified receptors.
Exposure levels were calculated using a number of assumptions
regarding the absorption of the contaminant and the frequency
of exposures. Risk levels were then calculated for each >
exposure pathway under the worst case scenario. The scenario r
chosen was conservative in nature, and it was developed using
assumptions which may overestimate risks.
Risks Under Present Conditions
Risk calculations were undertaken for each of the screened
exposure pathways considered possible under current conditions.
All evaluated site pathways were identified as potentially
posing risks under a "worst case* scenario. Based on estimated
non-carcinogenic hazard indices and cancer risk estimates,
the relative significance of impacts of pathways is as follows:
4 groundwater consumption as drinking water;
0 direct contact with surface water;
0 accidental ingestion of soils;
0 accidental ingestion of surface water; and
0 direct contact with soils.
-------
-17-
In summary, this indicates that PC3s and lead oose the greatest
hazard associated with soils and surface water ingestion and
dermal absorption. The major hazard associated with ingesting
groundwater is due to the oresence'of ?CBs, cadmium, lead,
arsenic and benzene. Therefore, contamination leaving the
site proper in its present condition would not provide adequate
protection of human health and the environment.
Response Requirements
Based on the results of the health risk assessment, in order
to provide site control, the following actions should be addressed:
1. Surface water infiltration should be reduced.
*
2. Contaminated surface-water .runoff should be eliminated.
3. Contaminated waste oils contained in the tanks, and oil
migrating to the south on top of the water table should
be collected and detoxified.
....... »
4. On-site contaminated soils and sludge should be collected *
and detoxified.
5. Deeper olumes of phenols migrating to the south should
be collected and detoxified.
6. Contaminated surface sediments and soils in the 1000 ft
x 20.0 ft strip of land should be collected and detoxified.
ENFORCEMENT
The USEPA has idetified three potentially responsible parties
(PRPs) for the York Oil site at this time. These PRPs were
sent notice letters prior to the. initiation of the 1985 RI/FS.
Following negotiations, the PRPs were not willing to conduct
the required work. On December 9, 1983, the United States
brought a civil action under Section 104 and 107 of CERCLA,
seeking recovery of past and future monies expended for
response activities at the site. The suit was filed against
Kenneth Peiroarrtformerrowner of the York Oil Company), the
Aluminum Company of America (ALCOA), and the Reynolds Metals
Cqmoany. Upon the comoletion of the 1985 RI/FS, the PRPs
were sent notice letters informing them of, among other
things, their potential liability at the site, the availability
of the 1985 RI/FS report and the close of the public comment
period.
-------
-18-
In November 1987, the USEPA orovided the. PRPs with the addendum
FS and notified them of the oreferred remedial action for the
site as well as the close of the public comment period.
The USEPA intends to send notice to the PRPs uoon aporoval of
the ROD.
COMMUNITY RELATIONS
Various oublic informational meetings have been conducted and
fact sheets have been distributed throughout the site 's NPL
history to inform the community of the remediation process.
The draft 1985 RI/FS was made available for oublic review and
comment on August 22, 1985, at the following locations:
Moira Town Hall, the NYSOEC Region 5 Office, and the Franklin
County Office of Emergency Preoaredness. The oublic was
notified of the RI/FS availability by public notice which
were mailed to thirteen homeowners in the immediate area, and
by cress releases which appeared in the Watertown Daily Times
and the Massena Observer. Following a request for extension
of the twenty-one day public comment oeriod by the PRPs, the
comment period was extended an additional 30 days to October
11, 1985. A public meeting was held on August 28, 1985 aft
the Moira Town Hall which was attended by the NYSDEC, USEPA,
Association of Concerned Citizens of Moira, elected officials,
the press, and area residents. " .
Following USEPA's decision to conduct an addendum FS to further
evaluate oermanent remedies and better define the extent
of contamination, local citizens were notified by mail of
this proposed additional investigation and the concurrent
investigations that would be conducted for the contamination
pathways. The draft addendum FS was released for public
comment on Movember 27, 1987, followed by the release of the
proposed remedial- action plan for the site. The public was
notified o the addendum FS availability by oublic notice
which were mailed to the thrirteen homeowners and by press
release which aooeared in the Malone Telegram.
The public repositories for the Administrative Record, which
includes the&-addendum FS, are the Moira Town Hall, the NYSDEC
Region 5 Of $ce in. Rant Brook, New York and the USEPA Region 2
Office in New-'York City, the NYSDEC and USEPA held a public
meeting on December 16, 1987 to discuss the addendum FS and
the proposed plans. The comment oeriod was scheduled to end
on December 18, 1987, however, in response to a request by
the PRPs, the comment period was extended until January 15, 1988,
-------
-19-
A summary of the comments raised concerning the RI/FS and
public.meeting are contained in the attached responsiveness
summary. The local community prefers a permanent remedy for
the site. The responsiveness summary also includes a copy
of the press release announcing the meeting and an attendance
list for the public meeting.
A transcript of the public meeting was kept in accordance
with Section 117(a)(2) of CERCLA and is available to the
public at the above-mentioned Administrative Record repositories,
ALTERNATIVES EVALUATION
The remedial alternatives for the York Oil site were developed
and evaluated using CERCLA as amended, the NCP 40 CFR §300.68,
"Guidance on Feasibility Studies Under CERCLA" and EPA's
Interim Guidance on Selection of Remedy (December 24, 1986
and July 24, 1987) as guidance.
The major objective of the FS is to evaluate remedial
alternatives using a cost-effective approach consistent with
the goals and objectives of CERCLA. According to Section 121
of CERCLA, the recommended remedial alternative should protect
human health and the environment, should be cost-effective,
and should utilize permanent solutions "and alternative treatment
or resource recovery technologies to the maximum extent
practicable. The proposed remedy must also attain ARARs that
have been identified for the site. Section 300.68(e) of the
NCP and Section 121(b)(1)(A-G) of CERCLA outline procedures
and criteria which were used in evaluating and selecting the
appropriate remedy for the site.
A five step process was developed and used to meet the FS
objectives. The following is a summary of that process.
The first step is to evaluate human health and environmental
effects associated with releases and threatened releases of
hazardous substances from the site. Criteria to be considered
are outlined in Section 300.68(e) of the NCP and include such
factors as actual or potential direct contact with hazardous
material, degree of contamination of drinking water, and
extent of isolation and/or migration of the contaminants.
-------
-20-
The next step is to develop a range of potential available
remedial technologies that could be used to remediate the
site. Section 121(b) of CERCLA requires that remedial
technologies in which treatment permanently and significantly
reduces the toxicity, mobility or volume of the hazardous
substances as a principal element, are to be preferred over
remedial technologies not involving such treatment. These
technologies are initially screened on a technical basis.
Based on the screening, a list of individual remedial
technologies appropriate to site conditions and consistent
with the remedial action objectives is.developed.
The site-appropriate remedial technologies are then combined
into a number of preliminary remedial alternatives.__The
basis for the various Combinations are: the technical and
logical interrelationship between separate technologies; NCP
Section 300.68(f) requirements that generel categories of
alternatives must be considered and CERCLA Section 121
provisions regarding the preference for remedial actions that
utilize permanent solutions and alternative treatment or
resource recovery technologies. USEPA is in the process of
revising the NCP to reflect these new provisions added by
CERCLA. USEPA13 "Interim Guidance on Superfund Selection of
Remedy" (December 24, 1986 and July 24, 1987) is intended to
aid the Agency in the selection of remedial actions pending
USEPA's upcoming revisions of the NCP. This summary reflects
that guidance. USEPA1s interim guidance requires analysis of
alternatives involving: 1)treatment options with different
degrees of long-term management; 2)containment of waste
option with little or no treatment, but providing protection
of human health and the environment primarily by preventing
exposure or reducing the mobility of the waste and 3)the
no-action alternative. In most circumstances, these three
categories of alternatives must be carried through the detailed
evaluation process, and should not be eliminated during
previous screening processes.
The fourth step in the process is to provide an initial
screening of these alternatives as delineated in Section
300.68(g) oC£th« NCP. The three broad criteria that should
be utilized -trr the screening are: the relative effectiveness
in minimizing threats; the engineering feasibility and
impleraentability of the alternatives and the cost of implement-
ing the remedial action. This general screening is intended
primarily to reduce the number of remedial alternatives which
will subsequently be* evaluated in detail.
-------
-21-
The final step as outlined in Section 300.68(h) of the NC? is
to conduct a detailed analysis of the limited number of
alternatives that remain after the initial screening. In
most circumstances, a range of treatment alternatives, a
containment and no-action alternative should be included in
this analysis. These alternatives are then evaluated using
evaluation criteria derived from the NCP and CERCLA. These
criteria relate directly to factors mandated by CERCLA in
Section 121 including Section 121(b)(1)(A-G) and EPA's Interim
Guidance on Selection of Remedy (December 24, 1986 and July
24, 1987). The criteria are as follows:
0 Compliance with legally applicable or relevant and
appropriate requirements (ARARs)
Reduction of toxicity, mobility or volume
Short-term effectiveness
Long-term effectiveness and permanence
Implementability
Cost
State acceptance
Community acceptance
Protection of human health and the environment
DEVELOPMENT OF ALTERNATIVES AND INITIAL SCREENING
Source control remedial responses for the York Oil site will
address contamination of the site proper. A further definition
of the extent of contaminant migration from the site is
currently being assessed in a contamination pathway RI/FS.
The objective of the proposed source control remedial action
is to prevent further contaminant migration from the site and
thus minimize the threat to human health and the environment.
Criteria established to obtain this objective are:
0 Eliminate the potential for human/animal direct contact
with site wastes;
0 Eliminate the migration of PCB-contaminated oils and other
contaminants through surface and groundwaters; and
Eliminate the potential for precipitation/infiltration
with the wastes.
Source control technologies that are not considered appropriate
for utilization at the York Oil site and a brief discussion
of the reasons for their exclusion are Isited in .Table 4.
Table 5 lists and briefly describes the technically appropriate
remedial technologies for the York Oil site. These technologies
were accepted on the basis that they are compatible with the
specific site conditions and the remedial action objectives
for this operable unit. These technologies were then combined
into source control alternatives. As a result, nine remedial
action alternatives were developed for evaluation.
-------
-22-
Table 4 - -
INAPPROPRIATE REMEDIAL TECHNOLOGIES
0 Grout injection - Best suited for sealing voids in rocks,
rather than for containing groundwater flow in unconsolidated
materials around the site. The compatibility of most grouts
with hazardous waste and leachate has not been determined.
Slurry walls are less costly and have lower permeabilities
than grouted barriers. In addition, the variability of the
permeabilites of the soils at the site are not considered
suitable for grouting^.
0 Steel or wood piling cut-off wall - Wood is an ineffective
groundwater barrier, and steel barriers are rejected as a
containment response on the basis that the pilings cannot
be driven into rocky soil with boulders. The boulders
encountered within the overburden and the boulder/cobble
layer at the York Oil site would damage the piles, rendering
the cut-off wall ineffective.
0 In-situ solidification - Varying permeabilities at the site
limit complete interaction of the fixation additives and
the contaminants in the soil. :
0 Soil flushing - Flushing the soil with solvents or surfactants
will create problems with surfactant recovery. Steam
enhanced oil recovery is ineffective since the confining
pressures of the site overburden is not great enough to
contain the steam.
Oxidation .- Inappropriate due to violent reactions between
oxidant (e.g., ozone) and metals, oxidation of non-target
compounds which may lead to changes in soil hydraulic
conductivity, and partial oxidation of target compounds
resulting in more mobile and/or more toxic by-products.
0 In-situ ch4iaical dechlorination - Due to the non-uniform
distribution of PCBs in the soils, complete mixing of the
reagents and PCBs utilizing in-situ treatment would not
,occur. It is also limited due to the higher moisture
content in the soil/sludge layer.
-------
-23-
Table 5
APPROPRIATE REMEDIAL TECHNOLOGIES
° Slurry wall - Would be keyed into the intact glacial till
layer at a depth of about 35 feet, and would physically
contain the site. The slurry wall would be installed around
the site perimeter to prevent the entry of clean groundwater
into a leachate collection drain.
8 French drain.- It is designed to collect contaminated
groundwater and oil in the overburden and transport it to a
treatment facility. The inclusion of a hydraulic barrier
on the outer portion of the drain is intended to reduce the
amount of clean groundwater entering the drain.
; »
0 Drawdown wells - Shallow wells would collect contaminants
in the overburden and oil layer and deeper wells would
collect the phenolics plume. The collected groundwater
would be piped to a treatment facility at the site.
* On-site storage tanks - During remedial response/ temporary
storage tanks could be utilized for storage of oil and
soils and permanent tanks could be utilized for a proposed
water treatment facility at the site.
0 Water treatment system - It is essential for the cleanup of
contaminants in the groundwater. Groundwater would first
pass through an oil/water separator whereby oils would be
set aside for PCB-treatment. The water could then be
treated with an activated carbon filter to remove organics
and a deionizer to remove heavy metals.
Surface sealing - If the site is contained, surface sealing
utilizing clays, bentonite, cement and membrane liners
helps to prevent direct contact and reduces infiltration
and leachate generation. If hazardous wastes are left on-
site, the surface seal must comply with Section 264.310
requirements of RCRA.
0 Excavation^of wastes - Would remove all source contaminants,
(approximately 30,000 cubic yards of soil) except the deep
phenolics plume, thus eliminating further migration from
the site. A shallow well dewatering system would be required
to control groundwater in the mound during excavation.
0 Off-site disposal of wastes - Aside from the fact that
landfills do not reduce the toxicity or volume of the
contaminants and the future fate of these wastes are not
certain, off-site disposal in a PCB-permitted landfill is a
technically feasible technology.
-------
-24-
Table 5 - continued
0 Chemical dechlorination - Dechlorination of PCB-contaminated
soils and oils in an off-site facility and in a slurry mode
following excavation, is a feasibile technology, however,
volatile organics and metals would still be present in.the
waste.
0 Biological treatment - It is a viable treatment method for
the contaminated wastes at the site and could only be
performed in a slurry mode, following excavation. The
microorganisms are sensitive to low temperatures, however,
covered and heated tanks could provide optimal conditions.
Most microbes identified are not capable of degrading highly
chlorinated PCBs.
* Solidification - Following excavation, on-site slurry mode
solidification provides permanent immobilization of the
soils at the site. The toxicity and mobility of the wastes
is reduced resulting in a non-leachable solidified soil matrix.
0 Thermal treatment - The waste would be excavated, then *
thermally treated on-site. This process would destroy both
PCBs and organics, however, the fate of the metals is
questionable. Further treatment of the ash may, therefore,
be required. Thermal treatment must comply with RCRA Section
26.4 Subpart 0 and XSCA Section 761.70 incineration requirements.
-------
-25-
The nine remedial alternatives have been subjected to an
initial screening consistent with 40 CFR Section 300.68(g)(1),
(2) and (3) of the NCP to narrow the list of potential remedial
actions for further detailed analysis. This screening
concluded that the excavation and off-site disposal option of
the wastes in a PC3-permitted landfill is not as effective in
minimizing threats to human health and the environment as
containing the wastes on-site. On-site containment does not
require excavation and transport of the wastes thereby
providing significantly fewer environmental risks than off-
site landfilling. Also the implementability difficulties and
significantly higher cost for landfilling is a major factor
in eliminating this option from further consideration.
The other remedial option that is eliminatd from further
evaluation is the excavation of the soils followed by either on-
site or off-site chemical dechlorination of these soils arrd
oils in a slurry mode. The public health and environmental
problems with the excavation and transport of the soils has
been discussed above. In addition, dechlorination (either on-
site or off-site) only addresses the PCBs and would not affect
the volatile organics and heavy metals. This would only result *
in some reduction of. the toxicity of the waste, however, there r
would be no change in either the mobility or volume of wastes.
Other treatment options (solidification, thermal and biological
treatment) provide better effectiveness in minimizing threats.
DETAILED EVALUATION OF ALTERNATIVES ?
As a result of the screening process, a total of seven remedial
action alternatives were developed for detailed comparative
evaluation at the York Oil site. Three containment options,
three treatment options and the no-action alternative were
carried through to. this step. These seven feasible remedial
alternatives, and their associated capital costs, operation
and maintenance present worth costs and total present worth
costs are provided in Table 6. This table also provides the
estimated time to implement each remedial alternative from the
completion of the ROD, and the time it takes until full protection
is achieved*,
y"
Description "of Alternatives
This section provides a brief description of the seven feasible
remedial alternatives. A more detailed description of the
alternatives can be found in the addendum PS.
-------
table 6
alternative
Number
Remedial Alternative* Summary
Total Coat* (| x 10*)
Component*
'.'-'i1
Capital O a N
Praaant
Worth
Time to
Implement
Crew ROD
Time Until
Pull Protect.
la Achieved
ante
4.2
I
I No Action with Bit* Monitoring 0.3
?TA1HMBHT OPTIONS
";1perimeter slurry Hall* Interior Perimeter Drain* 5.1
Lateral Dralna Extending into Pill* Treatment of
Collected Groundwater, O(C-8lte Thermal Treatment
of Tank Oil*, Cleaning and Demolition of Tank*,
and an Impermeable Cap in Accordance with RCRA
3 Prench Drain with Hydraulic Barrier Around the
Bit* Except Cor Morthweat Portion, Lateral Drain*
Extending into Pill, Deep Drawdown Nell*, Treat-
ment of Collected Groundwater, Off-Bit* Thennal
Treatment of Tank Oil** Cleaning and Demolition
of Tank*, and an Impermeable Cap in Accordance
with RCRA
4 Shallow and Deep Drawdown Hell Syatem, Treatment 2.3
of Collected Groundwater, Off-Bite Thermal Treat-
ment of Tank Oila, Cleaning and Demolition of Tank*
and an Impermeable Cap in Accordance with RCRA
rRBATMENT OPTIONS ' ,. ,
~TSBoll Excavation, On-Site Thermal Treatment of 15.0
Boil* and Tank Oila, Deep Drawdown Halle, Treat-
ment of Collected Groundwater, Cleaning and Demo-
lition of Tankr, On-8ite Dlapoaal of Reaidual Aah,
and Surface Grading
SB Boil Excavation, On-Bit* Biological Degradation 13.0
of Soil* and Tank Oil*, Deep Drawdown Halle,
Treatment of Collected Groundwater, Cleaning and
Demolition of Tank*, On-Bit* Dlapoaal of Treated
Boil, and Surface Grading
SC* Boll Excavation, On-Site Solidification of Soil*, (.5
Thermal Treatment of Oil*, Deep Drawdown Hell*,
Treatment of Contaminated Groundwater, Cleaning
and Demolition of Tank*, On-8it* Diapoaal of
Solldifaffftmsoil and Surface Grading
1.1 1.3 f mo. Undetermined Mill not protect
human health and
environment.
1.7 (. 3 yr*. 15 yra. Phyaically containa
the alte but doea
not capture deep plume
(Doesn't meet ARARa).
Bouldera limit excav.
1.* «.l 3 yra. IS yra. Hydraulically containa
the elte, however,
drain provides limited
oil recovery. High
long-term OtM require-
ments.
/
Hydraulically containa
15 yra. shallow i deep plumee.
Pumps collect excess
amounta of clean water
OtH Intensive.
5 yra. Reduces toxlcity and
mobility. Deatroya PCB
and organlca. Netala
may inhibit process.
Further treatment of
ash may be required.
Degradee most PCBs and
organlca but Ineffect-
ive in degrading high
Chlorinated blphenyls.
0.5 7.0 3 yra. 5 yra. Reduces toxiclty and
v ' mobility. Permanently
immobllltes the waete.
Protects humajbJiaalth
l.B
0.5
4.1
15.5
0.5
13.5
3 yra.
3 yra,
3 yra,
5 yra.
Preferred Remedial Alternative.
ft enylronme/i
I
ls>
ON
OfcH
-------
-27-
Alternative 1 - No-Action with Site Monitoring - This alternative
consists of allowingthe site to remainTnits existing condition
and maintaining the current containment system. The present
containment system consists of an -interceptor ditch that runs
along the northwest border of the site and a filter fence
located in a seep area south of the lagoon S3. The existing
containment system and existing conditions were previously
shown on Figure 2. The ditch collects surface runoff and
intercepts groundwater during periods of high groundwater. The
filter fence collects oils migrating southwards from lagoon #3.
A long-tern groundwater monitoring program would be initiated
with the no-action alternative and each of the other six options,
to allow periodic reassessment of human health and environmental
risks posed by the site-. The monitoring would incorporate the
installation of aoproximately 15 wells at five locations. Each
well c'-sr.ar would consist of at least 3 points/ screened in
the overburden, cobble layer and bedrock. The exact location
of the wells would be a function of the selected alternative.
In general, 2 to 3 well clusters would be located downgradient
along the south side of the site and 1 to 2 well clusters would
be located upgradient in the northwest and northeast corners .
of the site. The sampling would occur on a quarterly basis and
would include groundwater/ surface sediments in the wetlands,
and for the no-action alternative, oils collected in the
interceptor-ditch. Table 7 summarizes the proposed long-term
monitoring program at the site.
Alternative 2 - Slurry Wall with Interior Drain, Groundwater
Treatment, Thermal Treatment of Oils/ and an Impermeable Cap
The perimeter slurry wall would be keyed 5 feet into the intact
glacial till layer at a depth of about 35 feet/ and would
physically contain the on-site contaminants. An interior drain
would be placed around the site perimeter inside the slurry
wall at a depth of approximately six to seven feet with lateral
drains extending into the fill. This drain system would partially
protect the slurry wall from migrating oils which could possibly
erode the wall, reduce the occurrence of oil and groundwater
overtopping the slurry wall by draining the groundwater mound
into a sump,,Tamd provide the necessary inward groundwater
gradient. The drainage would then be pumped into an oil/water
separator and water treatment system along with oils collected
from the on-site storage tanks.
-------
SronndMttr Bwiitorin?
Eiistinq Ktlli:
TQ1R
T04
Y07(S)
Y0710)
YW
TOM
TQ12 (3 clustirD
TQ14 (3 clustir*)
Y027(B) -
ProoMH Ktils:
T0101 (3 clostirs)
Y0102 (3 clastiri)
Y0103 (3 clMtirf)
YOI04 (3 clastirs)
Y0109 (3 clutirs)
PiriMtirt:
voUtiit orqanics
Kta
Oil Mtf 9TIIM
ZJK
ItU
'i flnitorta^
u
3 ittfimt uiplH ii
-------
-29-
Oils collected from the on-site tanks and groundwater would be
thermally treated off-site. The contaminated water would be
treated by carbon adsorption and deionizer units and discharged
into the surrounding wetlands or injected into the ground. The
PCB-laden sediments would be collected from the adjacent strip
of land and placed on-site within the slurry wall for containment.
An impermeable RCRA cap would be constructed over the contaminated
on-site area as a component of this alternative. The monitoring
program outlined for Alternative No. 1 would also be incorporated
with this alternative, in addition to cleaning and demolition
of the on-site tanks.
Alternative 3 - French Drain with Hydraulic Barrier, Deep
Drawdown Wells, Groundwater Treatment, Thermal Treatment of
Oils, and an Impermeable Cap
With this alternative, a french drain would encircle the site,
with lateral drains extending into the fill. The proposed
drain, which would lower the water table and collect migrating
oils, would be placed at a depth of approximately 15 to 20 feet »
in the down-gradient (southern) area of the site. The north, *
or up-gradient drain would be placed approximately 3 to 5 feet
below the ground to control the level of the water table. The
hydraulic barrier, on the outside of the french drain serves to
limit the amount of clean groundwater entering the system, thus
reducing treatment quantities and costs. The barrier would
also prevent migration of any contaminants through the drain.
Another component of this alternative is the addition of
approximately 20 deep drawdown wells places along the south,
east, and west sides of the site. These wells would be pumped
for three years to collect the phenolics within the weathered
till and cobble/boulder layer located beneath the french drain.
If, after three years, the phenolic olume is cleaned-up to a
level tht satisfies the ARARs (50 ppb for any single phenolic
compound and 100 ppb for total phenolics), the deep wells would
then be used for monitoring only.
Groundwater and oil collected in the french drain and wells
would be storied, treated, and disposed of using the oil/water
separator, water treatment system, and .off-site thermal treatment
as outlined under Alternative No. 2. * The tanks would be cleaned
and demolished as discussed in the previous option.
The PCB-laden sediments would be collectd from the 1000 ft. x
200 ft. strip of land and placed on-site within the limits of
the french drain for containment. An impermeable RCRA cap
would be constructed over the contaminated on-site areas.
-------
-30-
Alternative 4 - Drawdown Well System, Groundwater Treatment,
Thermal Treatment of Oils, and an Impermeable Cap
The system of drawdown wells would consist of 29 shallow wells
and 23 deep wells set above the till layer. The shallow wells
would be installed UD to twenty feet deep and would be constructed
of 2-inch diameter stainless steel. This system of wells would
be connected to a manifold and would be pumped by a series of 3
suction pumps. The deep wells would be set on top of or extend
up to 5 feet into the till layer, which is approximately 35-40 feet
deep. Each deep well would be constructed of 6-inch diameter
stainless steel and each would have an individual submersible
pump. As with the french drain alternative, the deep wells
would be pumped for the first three years to collect the
phenolics plume, and then used for monitoring.
As with the slurry wall and french drain alternatives, this
system would be coupled with an oil/water separator and water,
treatment system. Recovered oil would be treated off-site
using thermal treatment as outlined in Alternative #2 and 3.
The contaminated sediments would be collected from the strip *
of land and placed on-site within the drawdown well zone of r
influence for containment. An impermeable cap would be con-
structed in accordance with RCRA over the on-site contaminated
area. The long-term groundwater monitoring program would also.
be incorporated with this alternative. The PCB-laden sediments
would be collected from the 1000 ft x .200 ft strip of land and
placed on-site within the limits of the french drain for con-
tainment. An impermeable RCRA cap would be constructed over
the contaminated on-site area.
Alternative 5A - Soil Excavation, On-Site Thermal Treatment of
Soils and Tank Oils, Deep Drawdown Wells, Groundwater Treatment,
On-Site Disposal of Ash and Surface Grading
For this alternative, an estimated 30,000 cubic yards of
contaminated soils and sludges would have to be excavated from
the site. This volume would include approximately 8,000 cubic
yards of material from the 1000 ft x 200 ft strip of land west
of the site
-------
-31-
For three years, deep drawdown wells would be pumped to collect
the phenolics plume, and after that time they would be used for
monitoring purposes. Temporary shallow drawdown wells would be
utilized to collect migrating pollutants in addition to dewatering
the mound to allow efficient excavation.
As is the case with all the alternatives, the collected groundwater
would be treated and the tanks would be cleaned and demolished.
Alternative SB - Soil Excavation, On-Site Biological Treatment
of Soils and Oils, Deep Drawdown Wells, Groundwater Treatment,
On-Site Disposal of Treated Soil and Surface Grading
As with Alternative 5A, under Alternative 5B, all contaminated
soils would be excavated. However, Alternative 58 would then
involve biological treatment of the soils in an on-site treatment
unit. This process could not be applied in-situ due to the
inability to assure complete mixing of the biological organisms
and the contaminated media.
The biological treatment would be done in mixing tanks or ^
lagoons constructed on-site. The microorganisms would be added »
to the excavated soils and nutrients and catalysts would be
added as required to aid in the reaction. After treatment and
testing, the detoxified soils would be"returned to the site and
graded. Oil within the storage tanks would be treated separately
or added to the excavated soil for treatment.
During the remedial design, a treatability study would be
conducted to determine the appropriate nutrients, catalysts,
and reaction conditions. In addition, laboratory testing would
be required to evaluate the leachability of metals from the
treated soils.
A series of shallow wells would also be included during
construction with this alternative to collect floating oils and
to allow efficient excavation. Deep drawdown wells would also
be necessary to collect the phenolics plume in the deeper
groundwater zone.
As discussed?previously, the groundwater would be treated in-
site and the tanks would be cleaned and demolished.
-------
-32-
Alternative 5C - Soil Excavation, On-Site Solidification of
Soils, Thermal Treatment of Oils, Deep Drawdown Wells, Groundwater
Treatment, On-Site Disposal of Solidified Soils and Surface Grading
Alternative 5C involves the solidification/fixation of contaninated
soils and sludges. The contaminated soil and sludge in the
1000 ft by 200 ft strip of land as well as the contaminated
soils within the fenced area would be excavated and dewatered
prior to treatment. In-situ treatment is not feasible because
of the highly permeable soils, therefore, resulting in incomplete
mixing.
The solidification treatment would oe performed in a mobile
unit located on-site. The mobile unit would mix the fixing
agents/additives, and would blend the waste in mixing tanks
with the fixing additives, thereby permanently immobilizing the
waste. The stabilized material w-uld be tested to verify its
non-leachability, followed by bn-site disposal and grading.
A series of shallow wells would also be included with this
alternative during construction to collect floating oils and
to allow efficient excavation. The oils collected during t
excavation as well as the tank oils would be thermally treated
off-site. Deep drawdown wells are also necessary to collect
the groundwater plume within the boulder zone. The collected
groundwater would be subsequently treated. .
A treatability study during the design phase is recommended to
determine the effectiveness of the fixing agents to permanently
immobilize the entire waste stream.
Detailed Evaluation and Comparison of Alternatives
A detailed evaluation of each of the seven alternatives remaining
after the initial screening was conducted consistent with 40
CFR Section 300.68(h) of the NCP. A comparative discussion of
the seven alternatives using the evaluation criteria listed
previously is summarized below.
Compliance with Applicable or Relevant and Appropriate
Requirements
Section 121(d) of CERCLA requires that remedial actions comply
with all applicable or relevant and appropriate Federal and
State requirements (ARARs) to the extent that hazardous sub-
stances are present 'on-site.
-------
-33-
Each of the seven alternatives would comply with ARARs, with
the exception of the no-action. If the no-action alternative.
is selected/ both surface and groundwater at the site would
continue to exceed federal and state standards. The slurry
wall option (Alternative 2) does not meet the State groundwater
ARAR for phenols (i.e., 50 ppb for any single phenolic compound
and 100 ppb for total phenolics), since the deep groundwater
extraction wells are absent from this option. However, this
alternative does comply with RCRA Part 264.310 requirements
which specify certain thickness and composition for final
covers at hazardous waste sites.
While permits are not required for on-site remedial actions at
Superfund sites, any on-site action must meet the substantive
technical requirements of the permit process. Thermal treat-
ment of the oils at the site would comply with all the appli-
cable federal requirements,of Part 264 Subpart 0 of RCRA
(Subpart 0 specifies design requirements for operation of
hazardous waste incinerators) and Part 761.70 of TSCA which
specifies requirements for incineration of PCBs. Specifically,
operation of an on-site thermal treatment unit would require
that the transportable unit undergo waste specific trial or
demonstration burns to demonstrate satisfactory destruction
of the toxic components of the waste. The trial or demonstra-
tion burn must show that the unit achieves 99.9999% destruction
and.removal efficiency (six 9s ORE), and controls air emissions
of products of incomplete combustion, acid gases and particu-
lates to specified levels.
Containment Alternatives 3 and 4 include deep Dumping to collect
the migrating phenolics plume and these options comply with the
RCRA provisions regarding final covers.
The three treatment alternatives will comply with state and
federal PCB-cleanup policies, requiring PCB cleanup in soils
in residential areas to 10 ppm. (Even though these cleanup
policies are not promulgated ARARs, they are consistently
ulilized when: developing cleanup levels for PCBs in residential
areas.) Thai* options also mitigate both the shallow and deep
groundwater qiiality problems. Thermal and biological treatment
and solidification are also consistent with the proposed NCP as
permanent treatment remedies. The treatment options would comply
with all federal and state requirements concerning potential
air emissions (particulates and volatiles) during the excavation
of contaminated soils and sludges.
-------
-34-
The applicable Mew York State air requirements for excavation
include Part 257 and Part 373, which regulate ambient air
standards, and control of particulates from waste piles,
respectively. The federal requirements to be complied with
during excavation include 40 CFR Part 50 and Part 264.25(f),
which control ambient ^ir standards and control of particu-
lates from waste piles, respectively.
The State applicable requirements include compliance with Part
212 which controls air emissions of contaminants to the outdoor
atmosphere, Part 373-2.15 regulating operating standards for
hazardous waste incineration and the State Air Guide-1 Guidance.
All alternatives utilizing thermal treatment of oils would also
comply with these requirements. However, if an off-site, out-
of-state treatment facility is used for the oils, then the
thermal treatment alternatiaves" would comply with the applicable
State requirements where the unit is located.
Each of the treatment options (Alternatives 5A, 5B, and 5C),
which include the incorporation of deep wells, will meet the
State groundwater standard for phenols.
»
Discharge of the treated groundwater into the wetlands would *
comply with the New York State Pollutant Discharge Elimination
System requirements.
Consistent with CERCLA .Section 121 requirements, the continued
effectiveness of containment options 2, 3 and 4 would have to
be evaluated at least every 5 years to assure continued protection
of human health and the environment. The treatment options would
also have to be evaluated every five years since this material
would remain on-site.
°Reduction of Toxicity, Mobility or Volume
This evaluation criteria relates to the performance of a remedial
alternative in terras of eliminating or controlling risks posed
by the toxicity, mobility or volume of hazardous substances.
Solidification would permanently immobilize the soil/waste
matrix, therfby eliminating any associated toxicity due to the
contaminants.'-" Any future leaching of contaminants from the
solidified soil would also be eliminated by this option. Thermal
and biological treatment of the soils'would destroy the PCBs
and organics, however, the toxicity associated with the heavy
metals would remain unchanged.
: ' fc
The no-action alternative would not result in a reduction of
either the toxicity, mobility or volume of the waste since the
site would be left as is. The containment options (Alternative
-------
-35-
2, 3 and 4) would result in some toxicity and mobility reduction
due to the groundwater collection and treatment system, however,
the volume of waste material would not be reduced. The volume
of waste material would not be affecated by either thermal
or biological treatment. With solidification, however, due to
the addition of the fixation agents, the volume of waste material
would likely increase, but not substantially.
"Short-Term Effectiveness
Short-term effectiveness measures how well an alternative is
expected to perform, the time to imolement the action, and the
potential adverse impacts of its implementation.
The implementation of the no-action alternative causes no
adverse impacts over the short-term, since the only short-term
construction activities associated with this alternative i.s
the installation of additional groundwater monitoring wells
for long-term site monitoring. The estimated time to implement
the no-action alternative is six months from the signing of the ROD.
Alternatives 2 and 3 require limited excavation of the soils, »
and the .installation of a slurry wall and french drain, respect- »
ively. Excavation could result in short-term air emissions
and installation difficulties due to large boulders at the
site. Both alternatives provide limited recovery of contaminated
oils due to the low porosity of the soils.
r . \.
Alternative 4 "provides a greater degree of protection over the
short-term than Alternative 2 and 3 since only deep and shallow
drawdown wells would be installed, thereby not requiring soil
excavation. However, excessive amounts of clean groundwater
would be collected and oil recovery via the pumping system
would be limited. Installation of an impermeable cap (Alterna-
tives 2, 3 and 4) would increase the short-term air emissions
due to the necessity of grading the mound at the site.
The three treatment options (Alternatives 5A, SB and 5C) require
excavation, thereby increasing the short-term risk from air
emissions. Thermal treatment may result in air emissions, how-
ever, as not*$T above, strict measures would be implemented to
ensure that sach emissions would not be harmful to human health
and the environment. Alternative 5A may also require additional
materials handling on-site, such as pretreatment (e.g., shred-
ding and crushing) of the contaminated soils prior to feeding
to the thermal treatment unit.
The time to imolement each remedial option, except for the no-
action, is approximately three years from the signing of the
ROD.
-------
-36-
"Lonq-Term Effectiveness and Permanence
Long-term effectiveness and permanence addresses the long-term
protection and reliability of an alternative.
Over the long-term, the on-site treatment ootions orovide
essentially equivalent protection to the local communtiy, since
the residuals are not expected to pose a hazard from a health
and environmental perspective. However, the long-term
effectiveness of thermal treatment to destroy the organics and
to fuse the high concentration of lead into the residual ash in
a non-hazardous form is questionable at this time. Further
treatment of the residual ash may, therefore, be required.
The residuals following thermal, biological and solidification
treatment processes would be anlayzed to determine the
effectiveness each treatment procedure has in rendering the
material into a non-hazardous form. In addition, tests would be
conducted to ensure that the residuals pose no direct contact
risk.
Each alternative, except the no-action and slurry wall options, »
is designed to clean-up the deep phenolics groundwater »
contamination within three year of pumoing.
-.
The treatment options achieve full protection of human health
and the environment, with minimal O&M, within five years from
the signing of the ROD, while the containment options require a ^
high degree of O&M and take approximately fifteen years to
achieve full protection. Alternatives 2, 3 and 4 only contain
the wastes while Alternatives 5A, SB and 5C treat the wastes,
resulting in a permanent remedy for the site. It is believed
that Alternative 5C, solidification, would provide a higher
degree of long-term.effectiveness than biological treatment
since solidification would permanently immobilize the waste.
Biological treatment is not effective in degrading the highly
chlorinded PCBs and metals. As mentioned above, further
treatment of the residual ash following thermal destruction may
be needed.
The containnipnt options, once implemented, need to be evaluated
every five years to ensure their continued effectiveness. This
evaluation would determine the effectiveness of these options
to lower the groundwater mound at the site and determine
the continued effectiveness of the cap to reduce the direct
contact threat.
-------
-37-
The no-action a-lternative orovides minimal long-term effective-
ness sj.nce the wastes would remain at the site without any con-
tainment or treatment.
°Imolementabilitv ....
Implementability addresses how easy or difficult it would be to
carry out a given alternative. This covers implementation from
design through construction and O&M.
The implementability of the alternatives is evaluated in terms
of technical and administrative feasibility, and availability of
needed goods and services.
Each alternative evaluated here is technically feasible, however,
each treatment option '(Alternative 5A, SB, and 5C ) would require
a treatability study to determine the optimal conditions to
render the residuals into a non-hazaradous form. The tr;stability
study would be conducted during the remedial design.
Pull-scale operation of transportable thermal treatment units
at hazardous waste sites has been limited. Units have also »
experienced extended periods of downtime, therefore, it is likely'
that operation of a unit at the York Oil site would also result
in some extended downtime oeriods. However, in all situations,
transportable units have been repairable and have been brought
back- up to full-scale operation.
The effectiveness of thermal treatment to fuse the metals in
the residual ash without further treatment of this material has
not been demonstrated at this time. Frequent monitoring of
residuals during full-scale operations would be needed to
ensure the systems's effectiveness and reliability.
Transportable units for thermal, biological and solidification
are currently available for use at hazardous waste sites and
could be used at York Oil. Sufficient area exist on-site to
set-up these mobile treatment units for operation and there is
ample land area available on-9ite for disposl of the treated soil.
Due to the savere and elongated winters in this area, the construc-
tion season "for each alternative would be limited. These conditions
might also result in hampered maintenance operations, especially
with the groundwater collection component (pumps, drains) of the
containment options. Due to the decreased winter temperatures,
both solidification and biological treatment may require additional
precautions (heated'tanks ) to maintain optimal reaction rates.
-------
-38-
°Cost
Costs are evaluated in terms of caoital, O&M and present worth.
In comparing treatment Alternatives 5A, SB and 5C which result
in the same degree of remediation (i.e., final end-product
which does not pose a threat to human health and the environ-
ment), solidification of the soils has been identified as a
cost-effective alternative. The present worth cost for these
three options ranges from approximately $7 million for solidifi-
cation to $15 million for thermal treatment. For thermal
treatment of soils (Alternative 5A), if it is determined during
the treatability study that further treatmeant of the residual
ash is needed, an estimated additional $2 million could be
required. The three treatmeant options include costs for
excavation and treatment of approximately 30,000 cubic yards of
contaminated soils.
The total present worth cost for the containment options
(Alternatives 2, 3 and 4) vary from approximately $4 million
to $7 million, but do not meet the preference for treatment.
They do, however/ reduce the exposure, to the contaminants. m
. r
As mentioned previously, Table 6 provides a summary of the
capital, OSM and total present worth cost for each of the
seven alternatives. A more detailed breakdown of these costs
are provided within the addendum FS.
State Acceptance
This section addresses any concerns and degree of support
the State has expressed regarding the remedial alternatives
being evaluated.
The States supports a permanent solution for the York Oil
site. Its preference is on-site thermal treatment of the
contaminated soils and oils (Alternative 5A). The State's
primary concern with this option is the ability or inability
of the thermal treatment unit to render the high concentra-
tion of metals in the residual ash in a non-leachable form.
Alternative 5C includes on-site solidification of the soils
and off-site thermal treatment of the oils. This option
would permanently destroys the contaminants in the oils
through' thermal treatment and would immobilize the contaminated
soils at the site. The State's concern regarding this
option is whether the contaminants in the solidified soil
would leach, therefore, not providing a permanent solution.
Prior to implementation of either of these options, a treatability
study would be performed to answer the concerns mentioned above.
-------
-39-
"Community Acceptance
This evaluation criteria addresses the degree to which members
of the local community suport the remedial alternatives
being evaluated.
Both the DEC'S preferred option (Alternative 5A) and EPA's
proposed remedial action plan (Alternative 5C) were made
available during the public comment period and were oresented
at the public meeting. In general, the community indicated
a preference for a permanent remedy at the site. Some residents
expressed concern that solidification only contains the
wastes and is not a permanent technology.
Concerns were also raised regarding the quality of the residential.
wells and the need for a residential water supply in the
area of the site. Since all residential wells are free of
any contamination attributable to the York Oil site, and the
potential for future impact of residential wells is minimal
due to the nature and movement of groundwater in the area, a
residential water supply is not justified at this site. ?
*
Detailed responses to the community concerns are contained
in the attached responsiveness summary..
"Protection of. Human Health and the Environment
*» ' s,
Protection of human health and the environment is the central
mandate of CERCLA. Protection is achieved primarily by re-
ducing health and environmental threats to acceptable levels
and taking appropriate action to ensure that there will be no
unacceptable risks to hemah health and the environment through
any exposure pathways.
Except for the no-action alternative, all of the alternatives
evaluated here are protective of human health and the environment
to some degree. However, options 2, 3 and 4 only physically or
hydraulically contain the contaminants at the site, thereby
resulting in-, the potential for some continued migration of PCBs and
other contaminants into the groundwater and surface water.
Thermal treatment of the soils by itself (Alternative 5A) would
not address the health risks associated with leaving heavy
metal contamination in the residual ash, however, it would
destroy the organic contaminants in the soils.
-------
-40-
Appropriate measures would need to be taken during site excava-
tion (Options 5A, SB and 5C) to protect workers and the com-
munity. In addition, prior to implementing treatment under
Alternatives 5A, SB and 5C, measures would have to be taken to
assure that implementation of these treatment processes does
not pose a threat to human health or the environment. A few
of the potential problems are outlined below.
workers would be protected through measures outlined in project
specific health and safety plans and through contractor ad-
herence to Occupational Safety and Health Act (OSHA) regulations.
An on-site transportable thermal treatment unit and/or associ-
ated air pollution control equipment, materials handling equip-
ment, or materials pretreatment equipment may generate noise
during routine operation. However, proprietors of these units
have indicated a willingness to house or insulate any noisy
pieces of equipment or take any other measures necessary to
eliminate the generation of noise.
Both solidification and biological treatment would not generate »
substantial noise during actual operations. However, noise
from excavation equipment and any potential materials pretreat-
ment would be generated. Again, measures could be taken to
minimize noise generation.
Dust and particulate matter could be generated during materials
handling pretreatment. The potential for air releases of
products of incomplete combustion during thermal treatment,
also exists. Measuares would be taken to reduce these potential
hazards prior to full-scale operation.
Of the treatment options, solidification (Alternative 5C)
permanently immobilizes the soils and eliminates any future
leaching of both organic and inorganic contaminants. All
threats associated with soils ingestion and dermal contact,
and surface water runoff, would be eliminated. The contaminated
oils at the site would be thermally treated in an off-site
facility, therefore, eliminating any future toxicity of these
wastes. T
-------
-41-
.SELECTED REMEDY"
Based upon CERCLA, the detailed evaluation of the alternatives,
and public comments, the EPA has determined that Alternative 5C,
excavation and on-site solidification is the most cost-effective,
environmentally sound remedy for this source control operable
unit. This remedy consists of the following 'components:
1. Excavation of approximately 30,000 cubic yards of contami-
nated soils followed by on-site solidification of this
material. (Figures 5 and 6 indicate the contaminated
areas to be excavated.) This volume includes excavation of
approximately 8,000 cubic yards of contaminated sediments
from the adjacent strip of land west of the site. On-site
analytical testing of the soil would be performed, to
screen the soils te help determine the required excavation
limits. The treated soils would be returned to the same
waste management unit from where they were removed, and
then graded. Since the solidified soil will remain on-site,
the remedy will be reviewed every five years to assure that
human health and the environment are being protected.
2. Installation of thirteen deep groundwater drawdown wells .
along the southern and western perimeter of the site to
collect the sinking phenolics contaminant plume (see
Figure 5), and installation of shallow dewatering wells
to collect contaminated groundwater and oil during excava-
tion. The deep wells would be screened at the top of the
underlying till layer.
3. On-site treatment of the collected contaminated groundwater
with subsequent discharge of the treated groundwater in
accordance with New York SPDCS permit requirements. The
proposed treatment system would consist of an oil skimmer and
oil/water separator that would concentrate the PCB-laden
oils floating on the groundwater. Water from the separator
would be discharged into a modular water treatment unit.
The collected oils would be thermally treated off-site.
4. Off-site thermal treatment of approximately 25,000 gallons
of contaminated tank oils in addition to other oils collected
at the site, in accordance with the RCRA 40 CFR $264 Subpart
0 and the TSCA 40 CFR §761.70 requirements. The empty
storage tanks would be cleaned and demolished.
5. Treatability studies would be conducted during the remedial
design to determine the effectiveness of the solidification
process and to determine the optimal treatment system for
the contaminated groundwater. Should the treatability study
-------
,11V"*
I «Ml»3*'
\ .t»;x..:"J
York Ol Addsndum
Feasibility Study Report
SITE EXCAVATION
ApproilMt* Llall* Of fCI
Distribution U Shallow
Ground M»t«»
ApproilMt* Halt* or Oil
SludM
___ APOIOM|B*I« Llall* or fCI
ConI«Bln«tloo la Soli/
COtJfOL/PAre
-------
«»l
400
390}
H
M0\
LEGEND:
APPROXIMATE TRANSITIONS
BETWEEN 801 TYPES
NOTE:
I. SCHEMATIC ILLUSTRATES COMCMTUAL DESIQN-
NOT INTENDED FOR FBIAL
t. SUBSURFACE PROFLC ADAPTED FROM
SUBSURFACE MFORMATION PROVDED Bt QZA
York Oil Addendum
Feasibility Study Report
SITE EXCAVATION
FIGURE
((y **-m»t>um t«NO
«Uc. Ill
VMNONIP AMP ONBIO «LACIAL fat
fMOIMNT TO-i an
IONOCK
-------
-44-
determine that solidification would not provide the desired
degree of treatment, then a treatability study would be
performed to determine the effectiveness of thermally treat-
ing the soils at the site.
6. An additional operable unit RI/FS for the contamination
pathways is in progress to further define the extent of the
contamination migration from the site. If remedial actions
are determined to be necessary to address the contamination
pathways, a subsequent ROD will be prepared.
Cost estimates for the selected remedial action are listed in
Table 8.
-------
i
i
P
I
I
k
I
I
l
i
i
i
-45-
Alisraativt 5C - Sit* excavation «;th SoiidiMCatio
-------
i
I
I
t
I
»
I
-46-
. V
j
IF
Tutt 8
(cont'd)
»TEH
«Mro9tolo
-------
-47-
5TATUTORY DETERMINATIONS
The selected remedy satisfies the nine evaluation criteria to
the greatest extent of any of the alternatives evaluated.
The solidification alternative would comply with federal and
state clean-up levels for PCBs, all groundwater ARARs, and
surface water discharge requirements (New York SPDES). The
contaminated oils would be thermally treated off-site in accor-
dance with RCRA Section 264 Subpart 0 and TSCA Section 761.70
requirements. In addition, the wastes in the solidified
soil.would be permanently immobilized in a non-leachable
form. The treated soil would undergo the extraction procedure
toxicity test.and/or the proposed toxicity characteristics leach-
ing procedure test to ensure the permanance of the treatment.
The selected remedy involves placement and treatment of soils
and debris wastes. Placement of wastes or treated residuals
is prohibited under RCRA Land Disposal Restrictions (LDR)
unless certain treatment standards are met. LDR standards »
have not been promulgated for soil and debris wastes, but '
when published/ the standards may be applicable or relevant
and appropriate. Despite the absence of specific treatment
standards, the treatment method employed as part of this
remedial action satisfies the statutory requirement to sub-
stantially diminish th« toxicity of the waste or substantially
reduce the likelihood of migration of hazardous constituents'
from the waste so that short-term and long-term threats to
human health and the environment are minimized.
EPA believes that soils solidification is an availabe and
reliable technology for the treatment of wastes types identified
at the York Oil site. The treatability study would ensure the
site-specific technical feasibility and operational reliability
of the solidification process.
The selected remedy satisfies CERCLA Section 121 requirements
for utilizing a remedy that permanently immobilizes and reduces
the toxicity^ and mobility of the waste at the site. Solidification
of the soil^provides the same degree of protection to human
health and the environment as the other treatment options, but
at a lower cost.
To summarize, EPA's selection of on-site solidification of
soils, off-site thermal treatment of oils and deep drawdown
wells (Alternative 5C) is protective of human health and the
environment, will attain ail ARARs, and is cost-effective.
Since this option utilizes solidification and thermal treatment
to eliminate the principal threat at the site, this alternative
would also satisfy CERCLA preference for remedies which employ
treatment, as their principal element to reduce toxicity,
mobility or volume of the contaminants at the site.
-------
-48-
Operation and Maintenance
O&M are those costs required to operate and maintain the reme-
dial action throughout its lifetime. These activities ensure
the lifetime effectiveness of the remedial alternative selected.
The selected alternative involves the installation of deep
wells to collect the phenolics plume with subsequent treatment
of the contaminated groundwater in a water treatment system.
It is estimated that thirteen deep drawdown wells would be
installed along the southern, eastern, and western perimeter
of the site. These wells would be screened at the top of the
underlying till layer and are anticipated to consist of six
inch diameter stainless steel wells with 25 gallon/minute pumps.
The recommended water treatment system is anticipated to consist
of an oil/water separator followed by a modular water treatment
unit. The oils would be thermally treated and the contaminated
groundwater would be treated and discharged to the wetlands in
conformance with New York SPDES permit requirements. A treata-
bility study would determine the optimum water treatment system.
The O&M associated with the first three years of deep well
pumping is part of the proposed remedial action, since it is
currently projected that it would take three years to restore
the quality of the aquifer and cleanup the deep groundwater
phenolics plume to a health-based level of 50 ppb for any
single phenolic compound and 100 ppb for total phenolics.
Therefore, the related O&M costs during those three years would
be cost-shared with the State. Subsequent to three years, any
additional O&M pumping costs would be incurred utilizing State
funds.
As part of the remedial action, a 30-year groundwater sampling
program is included to monitor changes in the nature and extent
of contamination at the site to determine the effectiveness of
the operation.
One hundred percent of the remedial design will be funded by EPA.
Cost sharinf^for construction of the remedy is 90% Federal and
10% State. -i-v
-------
-49-
SCHEDULE*
Activity
8 Public Meeting
0 Regional Administrator Signs
Record of Decision
0 PRP Negotiations
If no PRP pick-up, then:
*
0 Contractor Procurement Process
for Remedial Design
0 Treatability Study and
Remedial Design Begins
0 Potential PRP Negotiations
Completion
0 Contractor Procurement Process
for Construction
0 Implement Remedy
0 Construction Complete
Date
December 16, 1987
January 29, 1988
December 1987 to
February 1988
February 1988 to
May 1988
June 1988
December 1988
December 1988 to
March 1989
April 1989
October 1990
* This is a projected schedule for this site and it is, therefore,
subject to future modification.
-------
-50-
FUTURS ACTIONS
This ROD addresses the source control operable unit for the
York Oil site. An additional operable unit RI/FS for the
contamination pathways is in progress to further define the
extent of the contamination migration from the site. Furthe.r
evaluation of the groundwater quality/ the extent of contamination
in the wetlands and Lawrence Brook/ and the potential impact
of wildlife and aquatic life such as bioaccumulation in the
food chain is being conducted during this study. If additional
remedial actions are determined to be necessary/ a subsequent
ROD will be prepared.
Residential wells currently in use have not been impacted by
the site. It is also unlikely that these residential bedrock
wells will become contaminated once the remedy has been imple-
mented* The glacial till layer above the bedrock is also a low
permeable confining layer. Continued monitoring of the ground-
water would be conducted/ however/ to assure the quality of
the groundwater. m
. »
The area around the York Oil site has a high sensitivity for
the discovery of cultural resources. Accordingly/ under the
National Historic Preservation Act/ a cultural resources survey
would be performed during the remedial design phase of the source
control operable unit to determine if any historical landmarks
or additional cultural resources exist within the undisturbed
portions of the York Oil site area.
Because of the similarity of the cultural resource issues
associated with the entire area/ both'the contamination path-
ways and source control cultural surveys would be conducted
concurrently.
-------
New York Stai? Department of Environmental Conservation
^cad. - :'--. N«w ^c ^ -:223- 4011
35 C .;.-.-
JAN 2 8 1988
Mr. Stephen 0. Luftig, P.E.
Director
Office of Emergency and Remedial Response
U.S. Environmental Protection Agency
Region II
26 Federal Plaza
New York, New York 10278
Dear Mr. Luftig:
Re: York Oil Company - Moira, New York »
» .
The New York State Department of Environmental Conservation
(NYSOEC) has re-examined the remedial action alternatives for the York
Oil Company site. The purpose of this letter is to advise you that the
NYSDEC will conditionally concur with the U.S. Environmental Protection
Agency's (USEPA) recommended Remedial Action Plan which includes site
excavation, on-site solidification, groundwater extraction wells,
treatment of collected groundwater, thermal treatment of oils stored in
tanks, cleaning and. demolition of the tanks and surface grading. It
appears to us that the on-site solidification alternative is a feasible
source control measure for the York Oil site.
As you know, the NYSDEC preferred on-site thermal treatment for the
contaminated soils and this is stated in the Record of Decision (ROD).
The State, as does USEPA, supports a permanent solution for the York Oil
site. Recognizing the potential problems associated with implementation
of either solidification or thermal treatment, a treatability study must
be performed to answer all uncertainties. We understand that EPA will
require treatability studies shall be conducted during the remedial
de**$n to determine the effectiveness of the solidification process for
the soils and determine the optimal treatment system for the
contaminated groundwater. Should the study indicate that solidification
of the soils is ineffective, then the effectiveness of thermally
treating these soils shall be investigated. Provisions to authorize
such an expansion of the treatability study must be stipulated in the
ROD.
-------
Mr.
"he beiec:t?
c :.i i alternative .»;!]
require cng teroi ^on tor ' ^~
ana maintenance. ihou'C the rcon;tor;.ig 'naicate tnac contaminants .are
.e'rvg r-^'edseo to the eov ; --ornienc . then, pursuant to SARA, the se'ectec
riieci 3 ' ^"c^ze snou'o be re-evaiuated oy the USEPA r'or suosecuent
jrac:at '-=Tiedy. iec-fcn 1^1 (:) of 3ARA i-eauires tne -jSEPA to aeterrnvne
.ne ef rect i veness of" th;s type of" remeoiai action or a fi/
-------
ATTACHMENT C
RESPONSIVENESS SUMMARY
Responses to the public comments on the Draft Addendum
Feasibility Study for the York Oil Company Inactive
Hazardous Waste Site in Moira, New York.
-------
York Oil Company Public Meeting
Moira, New York - December 16, 1987 -,
In attendance were twenty four (24) local residents;
three (3) newspaper/radio reporters; three (3)
representatives from the New York State Department of
Environmental Conservation (NYSDEC) or the "Department"; two
(2) representatives of United States Environmental
Protection Agency (EPA) or "Agency"; two (2) representatives
from Erdman, Anthony Associates (EAA); and one (1)
representative from Goldberg, Zoino and Associates (GZA).
Mr. Daniel *L. Steenberge, P.E. of the New York State
Department of Environmental Conservation chaired the
meeting. The meeting began with an introduction of" *
the Government representatives and the Government's
consultant with an explanation that the purpose of the
meeting was to gather public comment so that it could be
used in the decision making process of choosing a remedy for
the York Oil Company Inactive Hazardous Waste Site. The »
introductions were followed by a technical presentation from
Mr. Carl W. Eller, P.E. of Erdman, Anthony Associates. Mr.
Eller's presentation included a^brief history of the York
Oil Company Site activities prior to the beginning of work
on the Addendum Feasibility Study and dstailed discussion.of
the findings of the Addendum Feasibility Study. An overhead
projector was used as a visual aid by Mr. Eller and his
presentation was followed by a question, answer and comment
period.
A verbatim transcript of the meeting was kept in
accordance with the Comprehensive Environmental Response
Compensation and Liability Act, 42USC9601, (CERCLA) Section
117, and is attached as Exhibit I. Written comments were
received from Reynolds Metals Company and Alcoa Aluminum
Company of America and are attached as Exhibit II. Attached
as Exhibit III is a general summary of the significant
questions, comments and concerns expressed at the public
meet£ixr and in written comments along with the Department of
Environmental Conservation's responses to the questions,
comments and concerns. Attached is Exhibit IV which
includes the New York State Department of Health responses
to specific comments of Alcoa and Reynolds.
-------
PUBLIC MEETING ATTENDANCE LIST
YORK OIL SITE - TOWN HALL MO IRA
12/16/87
NAME Address
Ken Mason (Suburban Propane) Malone, NY
Lee Mulverhill (Suburban Propane) Malone, NY
Doug Wilson Massena, NY
Brian Schafer Massena, NY
Mike Healy Newark, NY
Wilfred Boyea Moira, NY
Jane Boyea Moira, NY
Louie LaRock Moira, NY
Joseph J. Terry Moira, NY
Lona Wells " Moira, NY
Mary Wells Moira, NY
Pete Warner Moira, NY
Gary Bradford Moira, NY
Rita Hutchins Brushton, NY
Win "Jake" Hughes Malone, NY
Wayne H. Brockway Moira, NY
Thomas Finnegan Moira, NY
Thomas R. Grady Malone Evening Telegram
Joe Lyon Moira, NY
Monte Coughlin " WICY
Tammy Wells Moira, NY
GOVERNMENT REPRESENTATIVES
NAME Affiliation
Daniel L. Steenberge, P.E. NYSDEC
Stephen Hammond, P.E. NYSDEC
Robert Howe USEPA
Joel S1ngermarH~- USEPA
Carl W. Ell er *£*»£.. Erdman Anthony Associates
Mary Kenny ~~~ Erdman Anthony Associates
Raymond Laport Goldberg Zoino Associates
James Connolly NYSDEC
-------
TERMS/ABBREVIATIONS
PCB - Polychlorinatsd Biphenyls - Once widely used industrial chemicals
whose high stability contributed to both their commercial usefulness
and their long-term deleterious environmental and health effects.
Common uses: In electrical -capacitors and electrical transformers
and as hydraulic fluids, lubricants and cutting oils.
ppm - Parts per million - Abbreviation standing for 1 part in 1 million
parts, also written PPM.
mg/1 - Milligrams per liter - Abbreviation standing for 1 part in 1 million
parts. Applies to aqueous samples.
ug/1 - Microgram per liter - Abbreviation standing for 1 part in 1 billion
parts. Applies to aqueous samples.
#
mg/kg - Milligrams per kilogram - Abbreviation standing for 1 part in 1
million parts. Applies to non-aqueous samples such as soil.
ug/kg - Microgram per kilogram - Abbreviation standing for 1 part in 1
billion parts. Applies to non-aqueous samples such as soil.
»
RI/FS - Remedial Investigation/Feasibility Study-The remedial investigation
is the report on the scientific investigation and the feasibility
study is the pre-design planning document.
ARAR - Applicabl or Relevant and Appropriate Requirements - Promulgated
Federal and State standards that'remedial actions must comply
with. " . *
EA - Erdman, Anthony and Associates - This is the State of New York's
consulting engineer for this project.
GZA - Goldberg Zoino Associates - This is the sub-consultant to Erdman,
Anthony Associates.
Portable GC - Portable Gas Chromatograph - This is a scientific instrument used
to Identify various chemicals.
Phenolics - Class of aromatic organic compounds.- The phenolics present at the
Ybrk 011 Site are most likely due to the use of high-boiling point
pftenols as fuel-oil sludge inhibitors.
Aroclors - Another name for Polychlorinated Biphenyl - The Aroclors are charac-
terized by four digit numbers, the first two digits indicate that the
mixture contains biphenyls (12) and the last two digits give the
weight percent of chlorine in the mixture (e.g., Aroclor 1254 contains
biphenyls with approximately 542 chlorine).
10"4 or 10"6- This is called scientific notation - For example 10" is .0001 or one
ten thousanths and 10 is .000001 or one one milliona^h.
-------
-2-
CERCLA - Comprehensive Environmental Response Compensation and Liability
Act - This is the federal law commonly known as Superfund.
SARA - Superfund Amendments . and Reauthorization Act - This, is the
Federal law which ammended CERCLA.
-------
EXHIBIT I
(see attached transcript)
-------
EXHIBIT II
-------
; i -\ i ' : ' ' ^ ' i v . " -^ '.
,!«'' v 1 s» ^/ . i . J-j--.. < I "w ("~ /A . « . £L £"» W.H
1988 January 19
BY EXPRESS MAIL
Daniel L. Steenberge, P.E.
Senior Sanitary Engineer
NYS department of Environmental
Conservation
Route 86
Ray Brook, New York 12977
RE: York Oil Site
JAW 2 0 1930
Dept of Environmental Conservation
REGIONAL ENGINEER - REGION 5
RAY BROOK, NEW YORK 12972
Dear Mr. Steenberge:
In Alcoa's comments on the above referenced "Site (in my letter
dated December 17, 1987 and during the December 28, 1987
meeting in-.Albany) we referred to recent {information on the
relationship between the degree of chlorination of byphenyls
and PCB toxicity. We have been able to obtain a copy of the
attached comments of Dr. Stephen M. Roberts, et al., made to
the California Scientific Advisory Board dated October 20,
1987. As we stated during the December 28, 1987 meeting we are
forwarding a copy of these comments to you for inclusion in the
administrative record on this matter.
Since
Ralph W. Waecttter
Attorney
WPPRWW140
Attachment
cc:
James Ludlani,' NYSDEC
R. Howe, U.S. EPA
Elena T. Kissel, Esq., U.S. EPA
-------
ORAL PCB COMMDJTARY
Before the Scientific Advisory Panel
"Safe Drinkidg Water and Toxic Enforcement Act of 1986"
State of California
by
Or. Raymond D. Harbison
Dr. Robert C. Jaajcs
Or. Stephen M. Roberts
Division of Interdisciplinary Toxicology
University of Arkansas
School of Medicine ;
Little Rock, Arkansas
on behalf of
National Electrical Manufacturers Association (NEMA)
Chemical Manufacturers Association (CMA)
Utility Solid Vast* Activities Group (USVAG)
October 30, 1987
Sacraaento, California
-------
PCB Commentary:
State of California Scientific Advisory Panel -
'Safe Drinking Water and Toxic Enforcement Act of 1986".
Prior to the originally scheduled August meeting of the Scientific
Advisory Psnel to review polychlorinated biphenyls (or PCBs), we
submitted ts tie panel a detailed review of the published literature
relevant to the carcinogenicity of PCBS. This review was submitted on
behalf of three national trade associations, who are concerned about
their industries, customers, and the geaeral public who may be broadly
affected by this panel's sctions. The conclusions of our review are:
PCBs sre not mutagenie.
»
The only conclusive evidence from chronic animal studies »
that PCBs are carcinogenic is limited to PCB mixtures
containing 60% chlorine.
Epidemiological studies of highly exposed worker populations
have shown no consistent cancer increases attributable to PCB
exposures.
Therefore, the existing data are inadequate to characterize
PCBs as probable human carcinogens.
I would like to take this opportunity to briefly review the basis
for these conclusions, and I ask this scientific panel to consider the
following information before rendering their decision.
All PCB congeners or commercial PCB mixtures are not equal and that
it is scientifically incorrect and unjustified to consider all PCB
mixtures as- equivalent. It is now widely recognized that the chemical
properties and toxicities of the 209 discrete PCB congeners are dictated
primarily by the degree and position of chlorination; and in general terms,
it may be stated tha.t the toxicity increases as the average chlorine con-
tent of the PCB mixture increases. Therefore, there are significsnt
and important differences among commercial PCB mixtures, and they should
not be evaluated by a single, all inclusive, generic rule that is based
on animal data specific to only some PCB mixtures.
-------
PCBs are not genotoxic. A vide range of commercial PCB mixtures and
specific PCS congeners have been shown to be without activity in bacterial
mutation assays. PCfis have also been found to be inactive is test
systeas measuring DNA daaage, chromosomal damage, and the ability to
induce cell transformations. Also, consider the fact that the positive
animal carcinogenicity data is specific for highly chlorinated PCB
mixtures. Only 3 animal studies have unequivocally demonstrated that
PCBs induce liver cancer in rats, and all involved commercial PCB
mixtures of 60% chlorine.
Epideaiologic studies have not causally linked PCB exposure to
cancer of any type in humans. For the currently available studies to
provide evidence for human carcinogenicity would require:
That these studies show a consistent increase in *
cancer mortality. *
The increase in cancer to be positively correlated to either
the magnitude or the duration of PCB exposure.
The increase in cancer to show evidence of latency.
The evidence for carcinogenicity to demonstrate a consistent
pattern, that is, specific sites of cancer should be
consistently elevated.
A review of the available mortality studies reveals that none of the
four criteria for a positive epidemiologic association have been met.
In conclusion, I ask the panel to recognize that PCB mixtures,
either manufactured as commercial products or inadvertently generated
in processing, are not equivalent. For this reason a generic ruling that
treats all PCS mixtures as equivalent-human health hazards is scientifically
inappropriate and should be replaced by an approach that considers each
PCB mixture separately. Consideration of PCB mixtures separately would
be consistent with the approach this scientific panel has taken when
evaluating other chemical classes and would be consistent with the
-------
scientific literature describing the toxicities PCB mixtures of different
chlorine content. PCBs «re not genotoxic, and PCB mixtures of_less than
60% chlorine are not carcinogenic in rata; therefore, clear distinctions
should be made between PCB mixtures with a chlorine content of 601 or
greater from those mixtures having a chlorine content of leas than 601.
Further, none of the PCB mixtures have been found to be carcinogenic by
epidemiologic study of capacitor workers exposed to high levels of
commercial mixtures. Therefore, I ask this panel to consider esch
mixture separately or defer any final decision on PCBs until such time
that it can consider each mixture on the basis of the data available for
the chlorine content «f that mixture. Based upon our review of the
current scientific literature, we submit that the listing of onl-y cotnoercial
PCB mixtures with an average molecular chlorine content of 60% or greater
is scientifically justified.
-------
Aw8 I 9 13S7
ornea* or
MOBSlflOIT & FOZBSTSB
nujrcwco
"OCVl*;* SUM l^i Jkjagjj.. C
aus
Via Telefacaimile *i/tr ^tf*"t ..1 213-^21-9423
August 18, 1987
Dr. Steven leek »
Executive Secretary '
Scientific Advisory Panel
1600 9th Street, Room 450
Sacramento, California
Re: Scientific Advisory Panel
Meeting Slated for August 28, 1987
Dear Dr. Booki
Z am writing en behalf of three trade associations:
the National Electrical Manufacturers Association ("KEMA");
the Chemical Manufacturers Assoeation CCMA"); and the
Utility Solid Waste Activities Group CUSWAC"). NXMA,
CMA, and USWAG hereby request that they be permitted to
submit a brief joint presentation to the Scientific Advisory
Panel, as the panel considers listing polyehlorinated
biphenyls (ceamonly known as PCB's) as possible known
carcinogen* on August 28, 1987.
" WEMA, CMA and USWAG will be jointly represented
by Dr. Robert James, a toxicologist with the Department
of Interdisciplinary Toxicology, University of Arkansas
Medical School, Little Reck, Arkansas. Dr. James will
be accompanied by Dr. John Craddock, of the Monsanto Chemical
Corporation and Steven S. Rosenthal, Esq. of this firm.
Or. James intends to offer a brief (three to five minute)
oral presentation, and will be available to respond te
any questions from the panel prompted by the oral presentation
or a prior written submission being sent to the Panel
this week. por Distribution hy CMA
SPECIAL PROGRAMS '
From
yr.-f.gg>
-------
August 20, 1987
Dr. Stevea A. Book
Health aad Welfare Agency
1600 9th Street, Room 450
Sacramento, CA 95814
Dear Dr. Book:
The National Electrical Manufacturers Association (NEHA), the
Chemical Manufacturers Associstioa (CMA), and the Utility Solid Waste
Activities Group (USVAG) Jointly commissioned independent toxicology
consultants to review the published technical literature oa carcino-
geaicity of PCBs for the upcoming Safe Drinking Water aad Toxic
Enforcement Act Scientific Advisory Paael (SAP) review oa August 28.
Eaclosed is a copy of the consultants' report for consideration by the
SAP in reviewing PCBs. Copies of this document also have been seat
xader separate cover to each SAP member.
This technical review is based upon animal studies sad
epidemiological studies of exposed populations with particular emphasis
oa workplace exposures. The conclusions of the eoasultsats are aa
follows:
. The dsta do not indicate PCBs are probable human carcinogens;
existing evidence is inadequate to coafirm PCBs as carcinogenic.
. Epidemiological studies of highly exposed worker populations
have shown no consistent cancer increases attributable to PCB
exposures.
. PCBs are aoa-snitageaic.
. Bioassay reaults are inconsistent and ahow no more than
limited, inconclusive evidence of animal carcinogeaicity except
in studies using higher chlorinated PCBs, e.g., 60% chlorine.
We believe that the evidence fro* animal studies does not justify
regarding all PCBs aa carcinogenic to humaas. At. most, the dsta
indicate that only the "higher chlorinated" PCBs, e.g., 60% chlorine
and higher, may be claaaified aa auspect animal carcinogens. Reviews
of human health effects studies do not reflect any statistically
significaat human cancer effects resulting from PCB exposure.
-------
ATTACHMENT
JCiabrough, R. D., 1987. Hunt an Health Effects of
Polychlorinated Biphenyls (PCBs) and Polybrominated
Biphenyls (PBB«). Ann. Rev. Pharaaeol. Toxleol. 27:87-111.
N«w York State Department of Health, June 1987. An
Investigation of the Health Status of Workers 'with Potential
PCS Exposure from Hudson River Sediment.
The Massachusetts Department o£ Public Health. June 1987.
The Greater New Bedford PCS Health effects Study 1984-1987.
-------
BIOLOGICAL DATA RELEVANT TO THE
EVALUATION OF CARCINOGENIC RISK TO HUMANS
Prepared for
Scientific Advisory Pine1,
Safe Drinking Water and Toxic Enforcement Act
State of California
Prepared by
Dr. Raymond D. Harbison
Dr. Robert C. Janes
Dr. Stephen M. Roberts
Division of Interdisciplinary Toxicology
University of Arkansas
School of Medicine
.Little Rock, Arkansas
U.S.A.
August 1987
-------
TABLE OF CONTENTS
Executive Summary . ill
1.0 Carcinogen! city and Related Studies
In Animali 1
1.1 Studies in Mice 1
1.2 Studies in Rats 2
1.3 Studies in Other Species 11
1.4 The Effects of PCBs on Other
Liver Carcinogens 11
2,0 Epideraiological Studies of PCS Exposure
and Cancer Mortality 15*
3.0 Other Relevant Human Data .18
3.1 Health Effects Information from
Accidental Inqestion 18
I ' :
3.2 Health Effects Information from
Environmental Exposure 19
3.3 Health Effects Information from
Occupational Exposure 20
3.4 Summary of Non-Cancer Human
Health Effects of PCBs 21
4.0 Cenotoxic Effects of PCBs 22
4.1 Bacterial Mutagenicity Studies 7.2
4.2 Clastogenic/Chromosomal Studies 23
4.3 Dominant Lethal Tests 24
4.4 DMA Damage Studies ' 25
4.5 Cell Transformation Studies 25
4.6 Summary 25
- i -
-------
DECOTXVE
Evaluation of Animal Data
The Investigation of PCBs' carcinogenic potential in mica is
limited to two short studies, while aome eight to ten studies have
been reported using various strains of rat. The PCS mixtures'tested
thus far are : 1) in mice - Kanechlor 300, Kaneehlor 400, K&nachlor
500 and Aroclor 1254; and 2) in rats - Kanechlor 300, Kanechlor
400, Kanechlor 500, Clophen A30, Clophen A60, Aroclor 1254 and
Aroclor 1260.
PCBs have been found to be tumorigenic in mice with Aroclor
1254 producing hepatooas after 11 months of exposure and Kanechlor
500 (similar in composition to Aroclor 1254) inducing
hepatocellular carcinomas after 8 months of exposure. These lesions
were shown to be reversible and specific for the dose (500 ppm) and
chlorination of the PCS mixture.
Zn rats, Aroclor 1260 or its equivalent, Clophen A60, have
produced hepatocellular carcinomas in three studies at doses of
approximately 100 ppm, a dose which appears to represent the
maximally tolerated dose for rats. A review of these three studies
indicates that the tumors occur very late in the life of the
animal, with a significant incidence of tumors only beginning to
appear after about two years of exposure. Of interest is the fact
that in all three studies the PCS treatment, while increasing the
incidence of live* cancer, did not increase the total tumor
incidence. The total tumor incidence was not increased in these
studies because in each case the incidence of other tumor types had
been significantly decreased. This suggestion of antitumor activity
of PCBs has also been demonstrated in a study examining the effect
of PCS exposure on the final tumor incidence in animals following
the transplantation of the Walker 256 sarcoma. The effects of
chronic PCB-treatment was not life-shortening, and in fact in two
of the studies the morbidity and mortality of the animals was
actually decreased by PCS treatment. Furthermore, while the tumors
are described as malignant, i.e. hepatocellular carcinomas, in none
of the three studies did the liver tumors metastasize to other
organs even though metastases would be expected if the tumors were
malignant. So from these studies it is evident that PCB-treatment]
does not increase the total cancer risk in these animals, rather itI
shifts the incidence of the type of tumors observed byj
significantly decreasing some tumor types while enhancing the
incidence of liver tumors. Lastly, PCB mixtures of lesser
chlorination, i.e., Aroclor 1254 and Clophen A30 (similar in
composition to Aroclor 1242, see table 1.4, page 8, of Brinkman and
DeKok, 1980), have been examined in two separate studies and found
not to be carcinogenic. Thus, conclusions to be drawn from the rat
data, like the mouse data, are specific for the dose and degree of
chlorination of the PCS mixture being tested.
iii
-------
While Bertazzi et al. (1986), in their study of Italian
PCa-exposed workers/ found a statistically significant elevation in
the rate of cancer mortality among both male and female" worJeers,
the much larger study of Brown, found no increase in cancer
mortality. In the Brown study/ there were higher-than-expected
incidences of rectal and liver cancer. Evidence for an association
between PCS exposure and these malignancy types cannot be
considered strong/ however/ in that: 1) no cas'es of rectal cancer
were observed after the initial report/ suggesting that this
increased rate was anomalous/ 2) the number of cases of liver
cancer observed in this study is not appreciably greater than
expected when examined without total number of liver neoplasms used
by Brown which includes those liver cancers that have metastasized
from other organs/ 3) in comparison/ the study by Bertazzi et al.
found only one case of liver cancer and no cases of rectal cancer
in their cohort/ and 4) the absence of a clear association with
latency or relationship with duration of exposure.*
The primary sites of neoplasms contributing to the
higher-than-expected cancer mortality rates in the Bertazzi studies
were located in the digestive system and the hematopoietic and
lymphatic systems. These were not increased in the larger cohort
reported by Brown. Further/ in the Bertazzi cohort/ there was no
evidence of latency or relationship between cancer mortality and
exposure to PCBs. It should also be noted that in the larger of the
two subdivisions of their cohort/ the female workers/ differences
in incidences of causes of death presumably unrelated to PC3
exposure (viz./ increases in accidental death and decreases in
deaths from cardiovascular disease) were of similar magnitude as
increases in',death from malignant tumors. This suggests/ at least
for this group/ that other confounding variables may exist.
In summary/ epidemiological evidence for human earcinogenicity
of PCBs is at present weak and mostly negative. As such/ until
larger epidemiological studies can be completed/ the data must be
Considered Inadequate gJ gJiagg^g ±tm Pgg^ as human cm~~
-------
PXTX M!T.gVXNT TO
QP 'gVRgTKQGgNTg ftTSIC TO
1.0 Carciaogenicity aad Related Studies Za Animal*
1.1 Studies in Mice
Nagasaki et al. (1972) initially -examined the
hepatocarcinogenic effects of Japanese brands of PCB fluids by
feeding male mice dietary levels of 100 ppm, 250 ppm, and 500 ppm
for 32 weeks. As this particular citation represents a short
communication, detailed discussions of the experimental design and
histopathological examinations of this investigation were lacking.
An increased incidence of liver tumors was found in only one of the
treatment groups, those mice receiving the 500 ppm diet of
Kanechlor 500, where hepatomas were identified in 58% or 7/12 o£
the animals. The livers from animals in thia group also contained
nodular areas and many necrotic foci. Zn sharp contrast to iaaae
findings, no hepatomas and none of these histopathologic changes
were observed in animals receiving lower doses of Kanechlor 500 or
in any of the animals receiving Kanechlor 400 or Kanechlor 300.
The results of the above study also appear to have been
reported in two other journals under Zto et al. (1973a4b) . As in
the previous study it was reported that male mice of the ddY strain
were fed diets containing Kanechlor 300, Kanechlor 400 or Kanechlor
500 at dietary levels of 100, 250 or 500 ppm for 32 weeks. .PCBs
significantly' increased the liver weights of the animals, and at
the highest- dose of Kanechlor 500 the liver: body weight ratio had
increased 3-fold with histopathological examination of the livers
revealing a focal hypertrophy in the centrilobular hepatocytes of
non-neoplastic areas. Another change observed in all PCB
treatment groups, except those receiving 500 ppm of either
Kanechlor 500 or 400, was a marked amyloid degeneration of the
liver in the space between the sinus endothelium and the
hepatocytes. For some reason, the liver tumors in these more
recent publications (Ito et al, 1973a4b) have been reclassified,
and the tumors reported as hepatomas in the previous communication
(Nagaski et al., 1972) are now listed as carcinomas (Zto et al.,
1973 a&b) . That is, in the group receiving 500 ppm of Kanechlor 500
(originally described as having hepatomas in 7/12 animals) nodular
hyperplasia was found in 7/12 animals (58.3%) while 5/12 of the
livers (41.7%) now have well-differentiated hepatocellular
carcinomas. The carcinoma cells were reported to be comprised of
irregularly shaped cells with pyknotic nuclei and occasionally had
mitotic nuclei. The reason for changing the classification of
these neoplasms from hepatoma to carcinoma is not provided. Of
some interest to discussions of the carcinogenic potential of PCBs
is the fact that all of the other doses of all three Kanechlors
tested failed to produce even nodular hyperplasia. Thus, the
tumorigenicity reported was quite specific for the dose and the
extent of chlorination of the Kanechlor being tested.
Kimbrough and Linder (1974) examined the effects of PCBs in
-------
contained fatty degenerative changes, and while 6/10 of the livers
from female animals had adenomatous nodules, none of the livers of
the male animals contained such nodules. However,the liver nodules
observed in the female animals do not appear to be related to the
PCB treatment as 2/5 (40%) of the livers from the control female
animals also contained adenomatous nodules.
In a second study Kimura et al. (1976) fed 12 female Oonryu
rats diets containing 400 ppm Kanechlor 400 for 6 months. The
estimated dose corresponded to a total of 531 mg of PCBs during
this period. Eight of the 12 animals were then sacrificed 590
days after the feeding began. None of these animals developed
hepatocellular carcinoma, and 9/12 of the livers were normal in
appearance, suggesting that the degenerative changes observed in
the previous study are reversible.
Ito et al. (1974) fed male Wistar rats Kanechlor 300,
Kanechlor 400 or Kanechlor 500 at dietary levels of 100, 500 or
1,000 ppm for up to 52 weeks. Mo hepatocellular carcinoma was
found in the livers of any of the treated rats. The highest dose
of all three Kanechlors did produce a cholangiofibrosis of the
liver, but this effect was not observed at the lower doses with any
of the Kanechlors. Nodular hyperplasia was observed in 30-40% of
the rats exposed to the two highest doses of Kanechlor 500 (i.e.
doses of 500 ppm and 1,000 ppm) and in animals receiving a diet
containing 1,000 ppm of Kanechlor 400. Oval cell proliferation and
proliferation of the bile duct cells were observed in all treatment
.groups. Hypertrophy of the centrilobuiar cells- was also evident in
animals receiving the highest dose of the two most heavily
chlorinated PCB mixtures. Fatty changes and fibrosis were also
observed in the livers of animals of several of the treatment
groups. The fatty changes, hypertrophy and fibrosis of the liver
all tended to be present and correlate with the observation of
nodular hyperplasia, suggesting that these changes may have been
contributory factors.
Kimbrough et al. (1975) published the first major positive
study demonstrating that Aroclor 1260 can produce hepatocellular
carcinoma in the rat. In this study 200 female Sherman strain rats
were fed Aroclor 1260 at a dietary level of 100 ppm for
approximately 21 months. There was a statistically significant
6-7% decline in the weight gain of the animals exposed to PCBs in
this study suggesting that the dose used approximated the
maximally-tolerated dose. The incidence of the histopathological
findings from this study are summarized in Table 1. The most
consistent histopathologic difference in the PCB treatment group
was the finding of hyperplastic or neoplastic nodules in 144/184
(78%) of the livers. More importantly, however, was the finding of
hepatocellular carcinoma in .26/184 (14%) of the PCB-treated
animals. The tumors were well-differentiated neoplasms of the
trabecular type, except in three of the animals which had tumors
with a glandular, papillary pattern. Foci of coagulative necrosis
were occasionally observed in the cancerous areas, but there was no
fibrosis or other evidence of chronic degenerative changes. Tumors
-------
one nodular hyperplasia was observed and it was in the group fed
100 ppm of Aroclor 1260 for 12 months. Mortality in this, study was
high and approximately one-third of the 105 animals anticipated to
be exposed for two years at each dietary level died. Hepatomas
were observed in 7/25 livers from animals fed 100 ppnr Aroclor 1260,
in 4/26 fed Xroclor 1254, 3/19 fed Aroclor 1242, and only 1/168
animals receiving the 1-10 ppm diets. Nodular hyperplasia was
twice as prevalent as hepatomas in the high dose animals,
particularly in the Aroclor 1254 group.
In 1978 the National Cancer Institute examined the carcinogenic
potential of Aroclor 1254 (NCI, 1978) . Groups of 24 male and 24
female Wistar rats were fed Aroclor 1254 at dietary levels of 25,
50 or 100 ppm for 105 weeks. Clinical signs of toxicity including
hair loss, facial edema and cyanosis occurred by week 72 in the
high dose animals and the mean body weights were roughly only
2/3-3/4 that of their respective controls. This decrease, in-body
weight exceeds the no more than 10% weight loss guideline for the
estimated maximally tolerated dose that is part of the NCI
guideline for cancer bioassays (NCI, 1979). Several histopathologic
changes occurred in the livers of animals receiving PCBs that
appeared to be related to the PCB treatment, particularly the
incidence of hyperplastic nodules and adenomas. Male animals 'had
one hepatocellular carcinoma in the 50 ppm group and only 2 in the
100 ppm group. Although the incidence of these tumors was not
significant, the occurrence of proliferative lesions did appear to
be dose related. In reviewing this bioassay, the Data Evaluation/
.Risk Assessment subgroup of the Clearinghouse on Environmental
Carcinogens responsible for providing peer review of NCI studies
concluded the following :
It is concluded that, under the conditions of the
bioassay, Aroclor 1254 was ««» earginoganie in Fischer
344 rats; however, a high incidence of hepatocellular
proliferative lesions in both male and female rats were
related to treatment. la addition, the carcinomas of the
gastrointestinal tract may be associated with treatment
in both males and females. Based on the liver
proliferative lesions in the treated rats and published
reports, it is suggested that Aroclor 1254 may be a
tissue prompter."
Morgan et al. (1981) have taken the same tissue sections that
originated in the NCI bioassay (NCI, 1978), stained the stomach
sections for alkaline phosphatase, and then re-sectioned these
tissues for histological evaluation in conjunction with those
provided in the NCI study itself. The final incidence of alkaline
phosphatase rich areas was 6.4% in controls, 10.4% in animals fed
25 ppm Aroclor 1254, 16.7% in the SO ppm group, and 35.4% in the
100 ppm group. These changes were most often noted in the pyloric
region of the stomach and duodenum (88% of the lesions were found
in these areas), suggesting a toxicity specific to the cells of
these areas. Gastric adenocarcinomas comprised six of the 33 total
-------
the cellular appearance of the liver tumors of PCB-treated animals.
Schaeffer et al. (1984) used a total of 432 weanling Wistar
rats to examine the effects produced by chronically feeding rats
Clophen A60 (equivalent to Aroclor 1260) or Clophen A30 (similar in
composition to Aroclor 1242; Brinkman and DeXok, 1980). The study
consisted of three groups. Group 1, a control group of 139 animals
receiving the normal diet, Group 2 with 152 animals receiving a
diet containing 100 ppm of Clophen A30, and Group 3 which consisted
of 141 animals fed a diet containing 100 ppm of Clophen A60.
After day 801 animals were randomly selected and killed, and the
experiment was terminated on day 832. The Clophens used in this
study were reported to be free of any chlorinated dibenzofuran
contamination, but the level of detection for this analysis was not
specified. Zn those animals necropsied prior to day 800,
hepatocellular carcinomas were only identified in the PCS treatment
groups, one in Group 2 and a total of 9 were observed in Group 3.
This latter number was statistically significant for the Clophen
A60 treatment, but represented a liver cancer incidence of only 7%
for the entire group. Zn contrast, the incidence of thymoma was
significantly reduced by PCB treatment declining from 12% in the
control group to 3-4% in the treatment groups. Likewise the total
number of the remaining types of neoplasms was significantly
reduced by the PCB treatment, with Clophen A60 causing the greatest
reduction (from 52 in controls down to 18 in the Clophen A60
group). The final results of this study are shown below in Table
2. The incidence of hepatocellular carcinoma was significantly
increased only in those animals receiving the Clophen A60. Thus,
the results of this study were consistent with the previous rat
studies, i.e. Aroclor 1260, or its equivalent, was reported to have
induced hepatocellular carcinoma in rats while a lesser chlorinated
PCB mixture was not carcinogenic.
Table 2
Frequency of Xepatocellular Alteration* Induced by Chronic
feeding Studies with Clophea A30 and Clophen A60
f of fagj Meeclagfig Medulla HecaEeeellulap eareinoma
Controls 6/131 5/131 1/131
(group 1) (4.5%) (3.8%) (0.8%)
Clophen A30 63/130* 38/130* 4/130
(group 2) . (48%) (29%) (3%)
- *
Clophen A60 3/126 63/126* 61/126*
(group 3) (2.4%) (50%) (48%)
denotes a significant difference from the control value (P<0.05)
-------
and significantly decreased the total tumor load of the exposed
rats. In essence/ he questions the human relevance of tumors which
occur only very late in the life of the animal" are not
life-shortening and do not metastasize to other organs of the body.
Thus, to quote Young :
"If the purpose of long-term studies is to extrapolate to
humans, then one finds it difficult -to infer dire
consequences to humans when the treatment is beneficial
in the model system. Is the model only useful for
inferring bad events? The model should be equally valid
for detrimental and beneficial effects."
Lastly/ the analysis of Young also calls to question a suggestion
made by Schaeffer et al. (1984)/ which was that the decrease in the
incidence of thymoma might be caused by immunosuppressive effects
of PCBs. While PCBs cause thymic atrophy at certain doses/ any
proposed immunosuppressive effect cannot be considered to have a
significant clinical impact when the treated animals did not
ultimately suffer a greater incidence of morbidity or mortality
from either infectious diseases or the cancer induced by this
treatment.
The last major rat study reported is that of Norback and
WeItman (1985) . These investigators fed 70 male and 70 female
Sprague-Dawley rats a diet containing 100 ppm Aroclor 1260 for 16
months followed, by a reduction to 50 ppm for the next 8 months.
The animals were then fed a control diet for the remaining 5 months
of their lives. All results were compared to a control .group which
initially consisted of 126 animals/ 63 of either sex. At various
time points throughout this study two control animals of each sex
and three PCB-treated animals of each sex (10 animals in total)
were anesthetized with ether and the medial left lobe of the liver
of each animal was surgically removed. These tissue samples were
taken at I/ 3, 6/ 9/ 12/ IS/ and 18 months. At 24 months a similar
group was killed and at the end of 29 months all remaining animals
were sacrificed. The induction of liver hypertrophy in the
centrilobular area of the lobule was evident at the first
observation period made one month -after the PCS diet was initiated.
By the 18th month the liver:body weight ratio had increased from 4%
to 12% in the female animals. Macroscopically these investigators
noted evidence characteristic of neoplastie nodules near the
capsular surface/ hepatocellular carcinomas and adenofibrosis. In
the PCB-exposed group/ the observed lesions appeared in the
following sequence : centrilobular hypertrophy at 1 month/ foci of
cells appeared at 3 months/ foci of altered cells in the
centrilobular and midzonal regions at 6-9 months/ neoplastie
nodules appeared at 12 months/ trabecular carcinoma was observed
after 15 months and adenocarcinoma at 24 months. Simple cystic
cholangioma and adenofibrosis appeared in animals 18-23 months
after the exposure began. There was no evidence of metastases to
the lungs. All trabecular carcinomas had cell arrangements with a
glandular/ ductal or cystic pattern and all adehocarcinomas had
some elements of the trabecular pattern of growth. The lumens of
-------
it would appear that a number of early deaths occurred only in the
group of male control animals. These same observations were also
noted by the authors, who stated in the discussion of this paper :
m Although the tumors met the morphological criteria for
malignancy, their biologic behavior was relatively
unaggressive. The neoplasms did not metastasize to
distant organs nor invade blood vessels. Mortality of the
animals was not increased. The lack of greater morbidity
or mortality is likely due to slow progression of the
neoplastic process and late appearance and slow growth of
the hepatocellular carcinoma."
The authors further noted that it remains to b« established whether
PCBs have an initiating effect or whether the neoplasms observed
result from the promotion of a background incidence of initiated
cells.
1.3 Studiea In Other Species
Calandra (1976) reported, in summary form, a carcinogenicity
bioassay performed in dogs. Groups containing 4 male and 4 female
dogs were fed 1, 10 or 100 ppm of Aroclors 1242, 1254 or 1260 for
two years. While this exposure interval is for a period of time
that is considerably less than the lifetime of the species being
tested, no remarkable findings were reported.
1.4 The Effects of PCBs oa Other Liver Carcinogen*
PCBs have been tested in numerous studies to determine whether
they alter the carcinogenicity of other chemicals. The evidence
obtained from such studies is sometimes inconclusive, and in
several instances similar studies have produced conflicting
results. The results of have been summarized in the following
paragraphs and in table 5 located on page 13 :
A number of studies indicate that PCBs administered
after a carcinogenic dose of a liver carcinogen enhance
the incidence of tumors. Nishizumi (1976) found that
Kanechlor 500 promotes liver tumors initiated by
diethylnitrosamine. Kimura et al. (1976) demonstrated
that Kanechlor 400 can promote the liver cancer induced
by 3'-methyl-4-dimethylaminoazobenzene (MeOAB). Zto et
al. (1978) initiated liver tumors with
N-2-fluorenylacetamide and then increased tumor incidence .
by feeding the rats diets containing 1,000 ppm of PCBs.
Preston et al.. (1981) have reported that Aroclor 1254
with or without polychlorodibenzofuran contaminants can
promote the liver cancer initiated by diethylnitrosamine.
Osterle and Demi (1984) demonstrated that this effect can
be produced in weanling rats/ although the immature
animal is less sensitive. Demi et al. (1983) have also
found that an initial PCB pretreatment to increase
11
-------
Wistar rats and CD-I nice from diethylnitroaamine-induced
liver tumors (Nishizumi, I960, Anderson et al., 1983).
Likewise Kimura et al. (1976) found that PCBs given
before or during MeOAB exposure decreased the incidence
of MeDAB-induced liver cancer. Similarly, Makiura et al.
(1974) have shown that the co-administration of PCBs with
the liver carcinogens MeDAB, N-2-fluorenylacetamide and
diethylnitrosamine decreases tumor incidence. Trout fed
diets containing PCBs prior to their exposure to
aflatoxin had a reduced frequency of liver tumors induced
by the aflatoxin (Hendricks et al. 1977). This effect
is perhaps explained by Stott and Sinnhuber (1978) who
demonstrated that PCBs reduced the bioactivating ability
of trout microsomal fractions used to test the
autagenicity of aflatoxin B1.
In dermal initiation-promotion assays PCBs have been
found to diminish the initiating activity of certain skin
carcinogens and to be devoid of any promoting activity
(DiGiovanni «t al., 1977; Berry et al., 1978). Similarly
Hayes et al. (1985) examined the effects of PCBs on the
proliferating hepatocytes in livers of young animals or
adult animals after partial hepatectomy. They concluded
that short-term PCS exposures do not have an initiating
action in an in vivo assay that detects both hepatic and
extra-hepatic initiating carcinogens.
In contrast to its reportedly adverse effects on the
immune system, Kerklivet and Kimeldorf (1977a&b).
demonstrated that PCBs administered either in the diet or
by injection reduced the size of Walker 256
carcinosarcomas transplanted to rats and increased the
lifespan of the PCB-treated animals.
Table 5
xy of PCS Interactions With Other Carcinogens
Study Carcinogen Results
»he Cargineon
Nishizumi diethylnitrosamine promotion (liver tumors)
(1976)
Kimura et al., Me-DAB promotion (liver rumors)
(1976)
Ito et al., N-2-fluorenylacetamide promotion (liver tumors)
(1978)
13
-------
this effect has also been demonstrated for phenobarbital
and is * no doubt also true for most other liver enzyme inducing
chemicals that produce a hypertrophy or hyperplasia of tihe liver.
Additionally, recent work by Demi and Oesterle (1987) has shown
that the promotional effects of PCBs has a threshold, as one might
expect, and this threshold appears to be well above the dose
commonly encountered by most persons. In contrast to its
promotional effects on liver carcinogens known to be initiators,
PCBs also act as anticarcinogens when administered prior to the
exposure of an initiating carcinogen. This effect is most probably
the result of PCB induction of liver metabolism which then acts to
decrease -the carcinogenic effects of a number of chemicals by
increasing their degradation and elimination.
2.0 Xpideaiologieal Studies of 9CB Zzposuze and Caaeer
Mortality *
Published epidemiological studies addressing the carcinogenic
potential of PCBs have been confined to two cohorts, both
consisting of employees of plants where PCBs were used in the
manufacture of capacitors.(Kote : Bahn et al. (1976), in a letter
to the editor, reported an analysis of eight deaths among 92
research and development workers at a refinery, and suggested an
association between PCB exposure and incidence of malignant
melanoma. The extremely limited size of their cohort, the inability
to adequately assess exposure,of the cohort to other, potentially
carcinogenic substances or other known risk factors, and the
inconsistency of their observations with the results from the two
larger cohorts, precludes any utility of this report in the
assessment of the human carcinogenicity of PCBs.]
Brown and Jones (1981) have published a retrospective cohort
mortality study of employees from two plants. Minimal exposure
period to PCBs was three months, and the types of PCBs used were
Aroclors 1016, 1242, and 1254. Workers who had potential exposure
to trichloroethylene, which was also used in both plants, were
excluded from the cohort. Of the 2567 employee cohort, 163 were
known to be deceased and the vital status of 55 (2%) was unknown.
Righer-than-expected incidences of cancer in the rectum
(primarily among females in Plant #2) and liver were noted,
although the differences were not statistically significant. The
overall incidence of malignancies observed was slightly less than
expected. Hone of the causes of death showed a clear association
with latency, and no relationship was observed between the duration
of employment in jobs involving PCB exposure and risk of mortality
due to cancer. The causes of mortality and incidences of specific
malignant neoplasms are summarized in Table 6.
The study of mortality in this cohort has recently been
updated, with the number of deceased increasing from 163 to 295
15
-------
the study, and exposure was primarily to Aroclor 1254 and Pyralene
1476. -While some members of the cohort were also exposed to
trichloroethylene, alkylbenzene, and epoxy resins, the "number was
stated to be few. A total of 27 deaths were observed in the cohort
between 1954 and 1978. The overall incidence of cancer observed
was significantly higher than expected (14 versus 5.65 expected),
contributed primarily by elevated (though not statistically
significant) incidences of neoplasms of the 'digestive organs and
lymphatic and hematopoietic systems.
Bertazzi et al. (1986) have recently published an update of
their study of PCB-exposed capacitor workers with a modified
cohort. Non-production workers at the plant were added to the
cohort, and the minimum period of employment was reduced from six
months to one week. The size of the cohort was consequently
increased to 2100, and the number of deceased to 64. The results
from male and female workers were analyzed separately/ and are
summarized below in Tables 7 and 8.
Table 7
Mortality from Selected Cauaes) of
Male Workers Kxpoeed to FCB
Cause of daath Observed Expected SMR Cxp«ct«d SMR
All causes
Malignant tumors
Cancer of G.Z. tract
Lung cancer
Hemato logic neoplasms
Cardiovascular disease
Accidents
Confidence limits (95%)
30
14
6
3
3
a
c
b.
27.8
5.5
1.7
1.2
0.8
'»
. «-B
144-415
104-300
108
233*
34«c
250
3?3
101
88
c- 141-721
*- 112-572
29.8
7.fi
2.2
l.C
1.1
9.4
5.8
101
183b
27 4d
187
263
95
103
For males, mortality from cancer was significantly greater
than expected (14 versus 5.5 expected). Among specific tissues, the
observed incidence of neoplasms of the GI tract was significantly
higher than expected, and hematologic neoplasms were also greater
17
-------
symptoms reported included acneform lesions and hyperpigmentation.
The incidence of hyperpigmentation was high (9/10) among offspring
born to mothers pregnant during%ingestion of the contaminated rice
oil (Higuchi, 1976). This hyperpigmentation gradually faded over
the 2-5 months after birth. Developmental tooth and bone defects
were noted among some of the offspring (e.g. eruption of teeth at
birth, larger-than-usual frontal and occipital fontanelles, and
maintenance of a wider sagittal suture than usual). However,
postnatal physical and mental development of infants born to
contaminated rice oil-exposed mothers paralleled that of healthy
infants (Higuchi, 1976).
While it was originally assumed that the rice oil was
contaminated with only PCBs (Kanechlor 400), subsequent analysis
revealed that there was also extensive contamination with
polychlorodibenzofurans (PCDFs) and polychlorinated quaterphenyls
(PCQs). Studies in animals have clearly indicated that PCDFs are
more toxic than PC3s,- and. a number of lines of evidence support the
contention that the adverse health effects suffered by the rice oil
poisoning victims were a result of PCDF rather than PCS exposure.
First, the concentrations of PCBs measured in patients with
symptoms of Yusho were substantially lower than those observed' in a
number of studies of healthy, occupationally-exposed workers
(Xunita et al, 1984; Masuda et al, 1985). The occupational studies
have also shown workers to have undetectable or
minimally-detectable concentrations of PCDFs (Kashimoto et al,
1985), while rice oil victims in Japan and Taiwan had significant
concentrations of PCDFs, including a number of extremely toxic
isomers. These human observations are supported by a study in which
monkeys were fed PCBs with levels of PCDF and PCQ contamination
similar to that of the Yusho oils (Xunita et al, 1984). Monkeys fed
this diet developed dermatologic symptoms resembling those seen in
patients with Yusho. Monkeys fed PCBs without PCDF/PCQ
contamination, or fed PCQs alone, did not develop dermatologic
lesions. Collectively these observations have led to the conclusion
that the symptoms experienced by the rice oil poisoning victims
were the result of PCDF rather than PCS exposure (Xuratsune, 1980;
Kashimoto et al, 1981; Drill et al, 1982; Masuda et al, 1982;
Masuda and Yoshimura, 1984; Chen -et al, 1984; Xunita et al, 1985;
Miyata et al, 1985, and Bandiera et al, 1984).
3.2 Health effects Information from Xaviroaaeatal Ixposure
A number of studies have attempted to measure health effects
from environmental exposure to PCBs. The most common source of
environmental exposure in these studies was the consumption of Lake
Michigan fish (e.g. Humphrey, 1980; Fein et al, 1984; Kreiss et al,
1981; Smith, 1983.), although the results from other sources of
environmental exposure have also been studied (e.g. Baker et al,
1980). The results from these studies have generally been negative.
Interpretation of the meaning of the few positive observations has
been hampered by either the absence of a control population matched
for known risk factors for the parameter studied, a strong positive
correlation between PCB body burden and the body burden of other
19
-------
conditions. Which respect to each of these categories/ the
following observations and conclusions were noted:
o logical -mt*mc*9* Of 11 Studies Of PCB-expOSed
workers which reported dermatologic findings, dermal
symptoms were noted in all 11. Correlation of dermal
symptoms with blood PCS concentrations were generally poor
or non-existent. However, collectively the evidence
strongly suggests that chloracne may occur when PCS blood
levels exceed 150-200 jig/ml.
t.ivey function? Some abnormality in liver function
indicated by a change in one or more relevant clinical
chemistry parameters , was observed in five of seven
studies of occupational exposure to PCBs. (Though not
mentioned by Gaf fey, it should be noted that while
differences in some parameters indicative of liver
function were observed to be statistically significant in
some of these, studies, these differences were uniformly
quite small. Further, with chemical-induced hepatotoxicity
one would expect to find a consistent pattern of
abnormalities among overlapping indices of liver function.
No such consistent pattern was observed.] Zn no case was.
the abnormality associated with any measurable adverse
health effect. The remaining two studies found no
evidence of liver abnormalities. A ninth study found
evidence of induction of drug metabolism among PCB-exposed
workers (Alvares et al, 1977) .
Fat; mggaboliam! There appears to be a correlation between
serum triglycerides and PCS exposure in most studies.
Results concerning cholesterol are equivocal, with one
study showing an increase, one a decrease, and three no
change. Conflicting results have also been observed with
HDL-cholesterol. Changes in fat metabolism produced by
PCBs, if they exist, do not appear to be of clinical
significance.
Bloed and blood preaaure; None Of the five Studies which
examined blood chemistry noted abnormalities associated
with PCB exposure. One study of PCB-exposed workers
measured blood pressure, but found no association with
PCBs.
Symptoms. illnaaa. and othgy eondifciona; Five Studies
report a variety of symptoms among PCB-exposed workers.
Most of these symptoms appear to be unrelated to PCB
exposure. None of the reports have found significant
clinical effects to be associated with PCB exposure.
3.4 Summary of Non-cancer Human Health Effects of ?CBs:
PCBs appear to be of low potency in producing adverse human
health effects. Among workers with demonstrated high body burdens
21
-------
TUBLI 9
Summary of Mierobial Mutageaicity Test Results
Strain _______________
Product C3078 03052 Gtt TA93 TA1CO TA1000 TA1535 TA1S36 TA1S7 TA1S38 WP2 WPZuvi
Aredor ....
12M
Ifag. tfag. Ncg. tfag. Nig. Ifag. tag. . »fag. Ifag. (fag. Ntg.
1254
KvweMor - --NMg.tfag.-~ - - -
500
- tag.
tatrictiloro-
Diphwiyt
- tag. tag. - tag. tag. tag, tag. tag. -
Arodor - -.'-.- - - - - - tag.' - -
1221 v ' ' l
Source: L«vinsJcas (1981)
* Based on re-evaluation of Wyndhaa';et al. (1975) by Or. Safe.
Levins leas, 1981), all of which have been summarized in Table 9, it
is concluded that PCBs are not mutagenic in bacterial test systems.
4.2 Clastogenic/Chromoaomal ~ Studies
Several test systems designed 'to measure chromosomal damage
have been used to test the potential genotoxicity of PCBs. The
only human cell line tested were human lymphocytes in the study of
Hoopingarner et al. (1972). The test system used phytohemagglutin
stimulated cells which were treated with 100 ppm Aroclor 1254 for
the first 24 hours, then again during the last eight hours of the S
and 62 stages of > the cell cycle and during the three hours of
mitosis. Cytological examinations of the cells during these stages
of the cell cycle failed to reveal any effect of PCB treatment on
chromosomal integrity... -.. --
Green et al. (1975a) tested Aroclors 1242 and 1254 for their
23
-------
mutations in the rat. Aroclor 1254 administered for five days at
dosages of 75, 150 or 300 mg/kg or at dietary levels of 25 or 100
ppm for 70 days was likewise without effect. Keplinfer et al.
(1972) and Calandra et al. (1976) tested Aroelors 1242, 1254 and
1260 in mice at dosages of 500 or 1000 mg/kg and also found them to
be without dominant lethal effects.
4.4 DNA Oaaage Studies
On the basis of sedimentation rates, Stadnicki et al. (1979)
have reported that the epoxide of tetrachlorobiphenyl caused
single-stranded chromosomal breaks in the DNA of L-929 cells at
concentrations ranging from 1 ug/ml to 100 ug/ml. A mixture of two
hydroxylated metabolites, and to a much lesser extent
tetrachlorobiphenyl, caused some damage at 20 ug/ml and what was
reported as significant damage at 100 ug/ml. The significance of
this single in vitro test is questionable, given the fact that the
epoxide metabolite was the only chemical species demonstrating a
strong activity in this test system. This conclusion has been
reached in part because the authors, on the basis of this study,
similarly concluded that the epoxide metabolite is the only
chemical species of interest with regards to its potential
carcinogenicity. However, all mammalian tests as well those io
vitro tests containing some activation system were negative,
indicating that either the epoxide is not genotoxic in other test
systems or quantities of the epoxide sufficient to produce
genotoxicity are not generated in vivo.
4/5 Cell Transformation Studies
Norback et al. (1981) reported that Aroclor 1254 transformed
C3H10T11/2 cells to Type ZZI foci after six weeks of continuous
exposure to 10 ug/ml of Aroclor 1254, while a concentration of 1
ug/ml did not. The authors suggested that these results indicate
that the effects .of PCBs in cell culture may include promotion. In
contrast to this study, Pienta (1980) reported Aroclor 1254 did not
induce cell transformations after eight days of exposure when
utilizing Syrian hamster embryo cells. The highest dose tested
was 50 Jig/ml, which was five times the highest dose later used by
Norback et al. (1981).
4.6 Summary : .
The results of the preceding studies have been summarized
on the next page in Table 10. Given the fact that the only test
reported as positive is one of questionable significance, the
number of times PCBs have been tested and found to be without
significant genotoxicity lead inevitably to the conclusion that
these compounds should be considered to be without evidence of
genotoxic activity.
25
-------
incidence of liver cancer, did not increase the total tumor
incidence. The total tumor incidence was not increased in these
studies because in each case the incidence of other tumor types had
been significantly decreased. This suggestion of antitumor activity
of PCBs has also been demonstrated in a study examining the effect
of PCS exposure on the final tumor incidence in animals following
the transplantation of the Walker 256 sarcoma. The effects of
chronic PCB-treatment was not life-shorteningr and in fact in two
of the studies the morbidity and mortality of the animals was
actually decreased by PCB treatment. Furthermore, while the tumors
are described as malignant, i.e. hepatocellular carcinomas, in none
of the three studies did the liver tumors metastasize to other
organs even though metastases would be expected if the tumors were
malignant. So from these studies it is evident that PCB-treatment
does not increase the total cancer risk in these animals, rather it
shifts the incidence of the type of tumors observed by
significantly decreasing some tumor types while enhancing the
incidence of liver tumors. Lastly, PCB mixtures of lesser
chlorination, i.e. Aroclor 12S4 and Clophen X30 (similar in
composition to Aroclor 1242, see table 1.4, page 8, of Brinkman and
DeKok, 1980), have been examined in two separate studies and found
not to be carcinogenic. Thus, conclusions to be drawn from the rat
data, like the mouse data, are specific for the dose and degree of
chlorination of the PCB mixture being tested.
Possibly because PCBs produce liver hypertrophy, they enhance
the tumorigehesis of certain liver carcinogens if given after the
carcinogen in question has had an opportunity to initiate tumors.
However, if the PCB exposure precedes or is concomitant with
exposure to liver carcinogens the tumorigenic response is typically
decreased, probably as a result of an enhanced metabolic
detoxification of the carcinogen.
To summarize, the qualitative human relevance of the
carcinogenic activity of PCBs based on the animal data is limited.
The studies providing some evidence of its carcinogenic activity
are specific for the degree of chlorination of the PCB mixture, the
total tumor incidence is not increased, and the tumors produced
occur only very late in the life of the animal and have no adverse
effect on the morbidity or mortality of the animal. There are other
considerations that limit concern for the carcinogenic data in rats
as well. PCBs are not mutagenic, and the mechanism of tumorigenesis
for these compounds therefore would appear to involve an epigenetic
mechanism. There is also substantial evidence that the doses used
to induce tumors in rats are hepatotoxic, and evidence indicating
the neoplasms induced by PCBs are reversible if the exposure is
terminated before the animal has been exposed for a considerable
portion of the animal's lifespan. All of these findings seriously
undermine the human relevance of the animal carcinogenicity data.
This is particularly true as the human dosages from past and
present human exposures are far lower than those used in the animal
studies. Given these considerations, it is concluded that the
animal data provides sufficient mvidmnem of -^ifflitgrf human reJgvangg
that PCBs of 60% chlorine content (e.g. Aroclor 1260/Clophen A60)
27
-------
for this group, that other confounding variables may exist.
In' summary, epidemiological evidence for human carcinogenicity
of PCBs is at present weak and mostly negative. As such, until
larger epidemiological studies can be completed, the data must be
Considered -fnarfgonaga to charaftsriftt Pf8s ms human ear
5.3 Evaluation
There is sufficient gvirfgnee of limited human r
for the carcinogenicity of Aroclor 1260 in animals.
The human evidence for carcinogenicity of this compound is
Xgaelag 12S4 /TTmnaghlag gQQ
There is inadequate mvrdmnf* for the carcinogenicity of
Aroclor 1254 in animals.
The human evidence for .carcinogenicity is negative but
4QQ
There is no evidence/insufficient evidence for the
carcinogenicity of Aroclor 1248 in animals.
The human evidence for carcinogenicity is negative but
eioph+n X3Q/1TaaaehlQg SOQ/Xroelog 1243 ;
There is either no mvidmncm or insufficient mvidmnem for
the carcinogenicity of these mixtures in animals.
The human evidence for carcinogenicity is negative but
29
-------
Biph^nyla. OSEPA report 560/6-75-004.
Chen, P.H.S. «t al. 1984. Poly chlorinated biphenyla, ~*
dibenzofurans, and quaterphenyls in the toxic rice-bran oil and
PCBs in the blood of patients with PCS poisoning in Taiwan. Am. J.
Znd. Med. 5:133.
Demi, E. et al. 1983. Benzo(a)pyrene initiates enzyme-altered
islands in the liver of adult rats following aingle pretreataent
and promotion with polychlprinated biphenyls. Cancer Lett. 19:301.
Demi, E. 'and 0. Oesterle. 1987. Dose-reponse of promotion by
polychlorinated biphenyls and chloroform in rat liver foci
bibassay. Arch. Toxicol. 60:209.
DiGiovanni, J. et al. 1977. Tumor-initiating ability of
2,3,7,8-tetrachlorodibenzo-p-dioxin and Aroclor 1254 in the two
stage system of mouse skin carcinogenesis. Bull. Environ. Cent am.
Toxicol. 18:552.
Drill, V.A. et al. 1982. Comments and Studies on the use of
Polychlorinated Biphenyls in Response to an order of the United
States Court of Appeals for the District of Columbia. Drill,
Freiss, Hays, Loomis and Shaffer Inc., Consultants in Toxicology,
Arlington, VA.
EPA (Environmental Protection Agency). 1980. AmM»*f wa^er Qualify
Criteria for Polyehloginaged Biph»nyla. EPA- 440/5-80-068,
PB81-117798.
Fein, 6.6. et al. 1984. Prenatal exposure to polychlorinated
biphenyls: Effects on birth size and gestational age. J. Ped.
105 (2). -315.
Fischbein, A. et al. 1979. Clinical findings among PCB-exposed
capacitor workers. N.Y. Acad. Sci. 320:703.
Fischbein, A. et al. 1982. Derma tological findings in capacitor
manufacturing workers exposed to dielectric fluids containing
polychlorinated biphenyls (PCBs). Arch. Environ. Health 37:69.
Fischbein, A. et al. 1985. Oculodermatological findings in
workers with occupational exposure to polychlorinated biphenyls
(PCBs). Brit. J. Zndust. Med. 42:426.
6affey, W.R. 1983. The epidemiology of PCBs. Zn: 9ms Human
Hazards. P.M. D'ltri and M.A. KamriA [EdS.],
Butterworth Publ., 'Woburn, MA.
6ans, J.B. and S.J. Pintauro. 1986. Liver scarring induced by
polychlorinated biphenyl administration to mice previously treated
with diethylnitrosamine. Proc.Soc. Exp. Biol. Med. 183:207.
Greim, H. et al. 1985. Studies on the evaluation of tumor
31
-------
growth in rats by feeding a polychlorinated biphenyl, Aroclor 1254.
Bull. Environ. Contain. Toxicol. 18:243. _,
Kimbrough, R.O. and R.E. Linder, 1974. Induction of adenofibrosis
and hepatomas of the liver in Balb/cJ mice by polychlorinated
biphenyls (Aroclor 1254). J. Nati. Cancer Inatit. 53:547.
Kimbrough, R.O. et al. 1975. Induction of liver tumors in Sherman
strain rats by polychlorinated biphenyl Aroclor 1260. J. Natl.
Cancer Inst. 55:1453.
Kimura, R.T. et al. 1976. Polychlorinated biphenyls as a promoter
in experimental hepatocarcinogenesis. 2. Krebsforseh. Klin. Onkol.
87:257.
Kiaura, N.T. and T. Baba, 1973. Neoplastic changes in the rat
liver induced by polychlorinated biphenyl. Gann 64:105.
Kreiss, K. et al. 1981. Association of blood pressure and
polychlorinated biphenyls. J. Am. Med. Assoc. 245:2505.
Kunita, N. et al. 1984. Causal Agents of Yusho. An. J. Ind. Med.
5:45.
Kunita, N. et al. 1985. Biological effect of PCBs, PCQs and PCDFs
present in the oil causing Yusho and Yu-Cheng. Environ. Health
Persp. 59:79.
Kuratsune, M. 1980. Yusho. In: Haloygnaggd Biphenyla. Triphenylaj
Nap^halgnaa. Pibangodiexing and Related Product^,. R. Kimbrough
-------
Shelton, D.W. et al. 1984a. Effect of dose on the inhibition of
carcinogenesis/mutagenesis by Aroclor 1254 in Rainbow trout fed
aflatoxin Bl. J. Toxicol. Environ. Health 13:649. -
Shelton, O.W. et al. 1984b. The hepatocarcinogenicity of
diethylnitrosamine to Rainbow trout and its enhancement by Atoclors
1242 and 1254. Toxicol. Lett. 22:27.
Ward, J.M. 1985. Proliferative lesions of the glandular stomach
and liver in F344 rats fed diets containing Aroclor 1254. Environ.
Health Persp. 60:89.
Warshaw, R. et al. 1979. Decrease in vital capacity in
PCB-exposed workers in a capacitor manufacturing facility. N.Y.
Add. Sci. 320:277.
Wolff, M.S. 1982. Body burden of polychlorinated biphecyIs'among
persons employed in capacitor manufacturing. Int. Arch. Occup.
Environ. Health. 49:199.
Young, S.S. 1985. Letter to the Editor. Toxicol. Appl. Pharmcol.
78:321.
35
-------
LIST Of TABLES
Table 1: Results of Carcinogenesis Bioassay
Of Aroclor 1260
Table 2: Frequency of Hepatoeellular Alteration*
Induced by Chronic Feeding Studies with
Clophen A30 and Clophen A60
Table 3: The Incidence of Bepatocellular Carcinoma,
Other Neoplastic Lesions, and Mortality by
Tine Period
Table 4: Incidence of Repatocellular Neoplasms
Table 5: Summary of PCS Interactions with Other
Carcinogens
Major Causes of Death for Plant Workers
Exposed to Aroclor 1016, 1242, and 1254
Mortality from Selected Causes of Male
Workers Exposed to PCB
%
Mortality from Selected Causes of Temale
Workers Exposed to PCB
Summary of Microbial Mutagenicity Test
Results
Table 6:
Table 7:
Table 8:
Table 9:
Table 10: Type of Genotoxic Activity
8
10
13
16.
17
18
23
26
- 36 -
-------
ALUMINUM COMPANY OF AMERICA
ALCOA BUILDING
PITTSBURGH, PENNSYLVANIA 13213
'-EGA1. OEPART.VSNT
1988 January 15
BY EXPRESS MAIL
Daniel L. Steenberge, P.E.
Senior Sanitary Engineer
NYS department of Environmental
Conservation
Route 86
Ray Brook, New York 12977
RE! York Oil Site
JAN 10 <
0«pt of Environmental Conservation
REGIONAL ENGIN£ER REGION 5
RAY 8800*! NEW YORK 12977
Dear Mr. Steenberge:
Following a meeting held in Albany on December 28, 1987 among
representatives of EPA, DEC, Alcoa and Reynolds the comment
period on the proposed remedial action for the above-referenced
Site was extended until January 15, 1988. Alcoa has continued
its review of the scope of the clean-up alternatives as they
are presented in the recent draft Addendum Feasibility Study
Report ("1987 AFS"), the 1985 Remedial Investigation Report
("1985 RI"), the-. 1985 Feasibility Study Report ("1985 FS"),
NYSDEC's November 27, 1987 Public Announcement which presented
the Department's preferred remedy, and the EPA Proposed
Remedial Action Plan.
Alcoa submits the following comments which, together with
Alcoa's past comments, including those expressed verbally at
the December 28, 1987 meeting, its written comments submitted
by letter dated December 17, 1987 and prior written and oral
comments made to EPA and NYSDEC, we request be incorporated in
the administrative record related to the York Oil Site.
I. Recommended remedial alternatives far exceed, in scope and
cost, those necessary to protect human health and the
environment and are inconsistent with the National Contingency
Plan.
It is abundantly clear from information contained in the
1985 studies that surface contamination at the York Oil
-------
Daniel L. Steenberge, P.E.
January 15, 1988
Page 2
Site occurred as a consequence of a spill of oil, most
likely due to tanks being cut open and lagoon dikes being
breached. (1985 RI, App. RI-1, page 3; 1985 RI, page
1-9.) Subsequent sampling and analysis demonstrated that
the only ground water which exhibits any level of
contamination is that confined to the operational area of
the site. Indeed, no significant ground water
contamination by PCB's exists anywhere at the Site.
Further, the contractor which prepared the 1985 studies
and the 1987 AFS has noted that the movement of
contaminants in the groundwater is slow and that no
groundwater drinking wells are threatened by the Site.
Although we were not provided with, nor have we had an
opportunity to review, records of the NYS Department of
Health, we believe that such records would confirm the
absence of a threat to drinking water wells.
Any alternative adopted must be the most cost-effective
remedy which will protect human health and the
environment. The absence of a genuine threat to human
health and the environment and the negligible to
non-existent migration of contaminants off-site or into
the groundwater strongly suggests that the appropriate
remedy, not inconsistent with the NCP, is one far less
extensive and costly than those preferred by EPA and
NYSDEC. I
II. Recommended remedial alternatives employing site
excavation present the highest exposure risks to workers and
the community.
Several of the remedial alternatives involving excavation
of site soils, considered in the 1985 FS and the 1987 AFS,
present the highest risk of community and worker exposure
of any of the alternatives considered:
"Installation of this system [excavation/off-site
disposal] would result in the most severe exposure
risks of the alternatives considered. As with the
slurry walls and french drain, on-site excavation
could result in worker or- community contact with the
exposed materials. During implementation of this
alternative, excavation would be concentrated on the
lagoons and the most contaminated materials ... "
(1985 FS, page 6-60)
Similarly, such worker and community exposure would occur
under the excavation and on-site incineration options
-------
Daniel L. Steenberge, P.E.
January 15, 1988
Page 3
presented in the 1987 AFS. The impact of such exposures
were not, but should be, evaluated as part of the risk
assessment. - - --.-
It is highly likely that exposure risks to both workers
and the community would be magnified due to on-site
incineration. Such "disposal" will also produce emissions
that must be of concern. For example, if 50,000 tons of
soil containing an average of 2 ppm PCB are incinerated,
250 pounds of PCB will be destroyed. However, because the
oil in the lagoons at the site was contaminated with lead
(used motor oil), some lead and other metal emissions will
result. Lead is particularly easy to volatilize because
of its low melting point and the chemical nature of lead
compounds used as motor fuel additives. If the lead
content of the contaminated soil is 500 ppm, then 50,000
pounds of lead could be volatilized at the high
incineration temperature? tha"C will be used. Much of this
lead could escape control devices as a fine fume, thus
producing an emission situation of. concern.
While the increased, exposure risks resulting from site
excavation may have been presented, albeit buried within a
technical discussion,_in the 1965 FS, the enhanced risks
were not addressed in the 1987 AFS.. Therefore, the Moira
and surrounding community may .not now be sufficiently ..,
apprised of these potential emissions, nor will" they, in
the future, become aware or have any further opportunity
to comment upon them since the anticipated incineration
units will not require RCRA_ permj.ts.
III. Information currently available indicates that significant
error is contained.in,.the cost estimates for Option 5A,
resulting in a substantial understatement of the true .costs of
this alternative. :
The 1985 FSW note's that the excavation, transportation and
spreading of 2700 cubic yards of wetlands soils will cost
$92.87 per cubic yard. (1985 FS, p.B-3) However, the
1987 AFS states that excavation, on-site transportation
and incineration can be accomplished at a cost of $145.00
per cubic yard (excavation at' $26/cy and on-site
transportation-and incineration at $119/cy). (1987 AFS,
App. I-D, p.2) If the 1985 excavation cost figures are
correct then''the 1987 estimated excavation costs appear to
be far to low.
Further, the 1987 AFS used $215 per ton as the unit cost
for incineration of on-site soils (1987 AFS, App. I-D,
p.7) and a much lower figure of $119 per ton for
-------
Daniel L. Steenberge, P.E.
January 15, 1988
Page 4
incineration of soils in the wetlands area (1987 AFS,
App. I-D, p.2). Available data indicates that neither of
the estimates employed in the 1987 AFS is realistic.
The attached Waste Treatment Technology Newsletter from
BBC, Inc., Publications (Attachment I, hereto) contains a
report (pp.7-8) on mobile incinerator equipment which we
assume would be similar to the equipment envisioned by the
consultant for the Site. This BCC report states that the
EPA mobile incinerator operates at a unit cost of $300 to
$1000 per ton, depending upon waste characteristics. An
ENSCO mobile system is reported to operate at $300 to $400
per ton of contaminated soil. We are aware that
IT Corporation and Waste Tech also plan to have mobile
equipment in the future and estimate costs to be within
these ranges.
The incineration cost estimates employed in evaluating
total Option estimated cost are critical since over 38,000
cubic yards (50,000 tons) are included in the clean-up -
estimate. If the true cost is twice the consultant's
estimate (at best a conservative estimate of the magnitude
of the error since the soil has no fuel value and high
moisture content), the onsite incineration estimate is
understated by at least $10 million. If realistic on-site
incineration estimates had been employed, Option 5A, the
consultant's recommendation/ would cost in excess of $35
million.
Such a gross error must be considered in evaluating
appropriate, cost-effective remedies.
IV. The absence of Treatability. Studies for the Alternative 5
Options renders coat estimates for those Options incomplete and
inaccurate.
_ The contractor has not included any information on the
treatability of the soils at the Site or in the wetlands
area. Fixation and incineration are complex processes.
It is virtually impossible to estimate the true costs or
effectiveness of the Alternative 5 Options without
treatability worJc prior to design of a construction
program and prior to..selection of a remedial alternative.
In contrast stands the. alternative of closure and capping
of the site in situ which will not require additional
treatability study prior to design. Such an alternative
can immediately progress to design of a construction
-------
Daniel L. Steenberge, P.E.
January 15, 1988
Page 5
program. For any remedy requiring further study prior to
construction design, the procedures outlined in the
'recently published EPA Remedial Action Coating Procedures
Manual (see Attachment II, hereto) be employed to better
understand and estimate the costs of the various
alternatives.
V. Adoption of NYSDEC recommended PCB clean-up levels does
not comply with legally applicable or appropriate and relevant
requirements.
In its December 17, 1987 comments Alcoa stated that
promulgated EPA PCB clean-up standards, rather than
unpublished NYSDEC clean-up "policies" are the applicable
standards to which the Site should- be remedied. NYSDEC's
policy is neither applicable nor appropriate and relevant.
Since the DEC "policy" is not promulgated it may not -
qualify as a potential applicable standard. See "Interim
Guidance on Compliance With Other Applicable or Relevant
and Appropriate Requirements", 52 Fed. Reg. 32496,
32496-97, August 27, 1987. The EPA promulgated standards
are the applicable standards. Neither can the DEC policy
be considered in determining the necessary level of
clean-up, because there exists an applicable,
. chemical-specific standard for PCB, Thus, the NYSDEC
policy is also neither appropriate nor relevant.
VI. Alternative 5 Options conflict, with cleanup standards
required under CERCLA amendments..
While recent CERCLA amendments shift clean-up preferences
toward treatment, this preference does not exist in a
vacuum. Treatment remedies must be those which
permanently and significantly reduce the volume, toxicity
or mobility of site contaminants. SARA §121(b)(1).
Incineration remedies, which at best reduce small and
non-threatening levels of PCB- while leaving an ash residue
containing the true contaminants of concern metals,
simply do not provide a significant reduction of volume,
toxicity, or mobility. Encapsulation or fixation, at a
site where the available evidence suggests negligible risk
of movement, is of questionable value as a permanent or
significant solution.
These conclusions are reinforced by the fact that other
suitable remedies are available which will adequately
protect human health and the environment. Further, and as
-------
Daniel L. Steenberge, P.E.
January 15, 1988
Page 6
described above, the lack of treatability studies at the
time of remedy selection means that the significance and
permanence of the recommended alternatives is highly
speculative, as are the short-term exposures associated
with the remedy. Indeed, the alternatives requiring
excavation, transportation and incineration, in contrast
to other remedies, are those most likely to conflict with
the assessment factors required under SABA
§121(b) (1) (A)-(G) .
Suitably closed and capped, the site presents a very low
risk of human exposure and meets the SABA standard by
interposing a remedy which will greatly reduce the already
low mobility of site contaminants. The applicable
clean-up standard should be 25 ppra for PC" ur.der E£A
promulgated rules. That standard indicates that no
excavation or "treatment" of excavated material is
required. Cost-effectiveness, together with maximum
practicable protection, remains a requirement of any
Superfund clean-up. Treatment, for treatment's sake, has"
never been a goal or requirement of CERCLA. Where
available data indicates that less extensive remedies will
provide equal-and possibly greater protection, selection
of the currently recommended alternatives is inconsistent
with the National Contingency Plan and CERCLA.
In sum, the alternatives evaluated and recommended by the
consultant, by NYSDEC and by EPA the Alternative 5 Options
are the result of flawed technical analysis and improper
application of the standards applicable to Superfund Program
clean-ups. Option 5A is simply the most egregious example,
since it would incinerate the largest quantity of Site
materials (thereby presenting the highest risk of worker and
community exposures) and is the most seriously cost undervalued
incineration alternative.
In the face of far le»a expensive, yet equally protective
alternatives, selection of any of the Alternative 5 Options
contravenes the fundamental requirement that remediation be
done in the most cost-effective fashion that will protect human
health and the environment. To adopt an extraordinarily costly
remedial alternative, with no practical nor demonstrable
increase in the protection of human health and the environment,
over far less costly and eminently suitable alternatives, is an
inherently capricious exercise of agency power. That such an
exercise may, in fact, increase community and worker exposures
-------
Daniel L. Steenberge, P.E.
January 15, 1988
Page 7
(over those presented by more suitable remedies) confounds
reason. It is an exercise which itself poses a threat to human
health and the environment.
Sincerely,
^]^Vxy»xU
Ralph W. Waechter
Attorney
WPPRWW136
Attachments
cc: James Ludlam, NYSDEC
R. Howe, U.S. EPA
Elena T. Kissel, Esq., U.S. EPA
-------
related parts for complete filter arrays. Prices for arrays range from 5413 for a single
cartridge in-tank filter rated at 300 gallons-per-hour to $4,327 sixty-cartridge systems
rated at 6,000 gallons-per-hour. Depth filter media include polypropylene, cotton with
a polypropylene core, glass fiber, pleated melamine or granular carbon or resin.
[Indexers: Membranes. Company announcement, Memtek, Raytheon]
RAYTHEQN PURCHASES ADVANCED WASTEWATER TREATMENT SYSTEM FROM
MEMTEK CORPORATION. Raylheon's new monolithic microwave integrated circuit
facility is located in Andover, MA, where it is configured to remove arsenic, fluoride,
copper and nickel from wastewater. The Memtek system consists of the company's
Teflon tubular membrane. Each module contains ten membrane tubes with a pore
size of 0.1 11. According to Memtek's Chris Founier, the system works by first
chemically pretreating wastewater to form a suspended solid slurry which is settled
and then pumped through the membrane filter. The turnkey system cost approximately
$250,000.
[Indexers: Patents, Membranes]
RECENT PATENTS: MEMBRANES
Japan JP 61 2Q0.81Q fCI. BQ1P131 Sentembgr 5. 19Bfi
Kurita Water industries, Ltd.
MEMBRANE FILTRATION OF WASTEWATER
The filtrate from ultrafiltration of wastewater is treated by 2 reverse osmosis units both having 10-60%.salt
rejection. A biological treatment effluent was ultrafittered and the filtrate containing 360 COO was
subjected to reverse osmosis at 40 times concentration and five times concentration in two units.
Japan JP 61 129 Q94 fCI CQ2F3/19> June 17 1986 . .
Nino Electric Industrial Company, Ltd.
MEMBRANE IN BIOLOGICAL TREATMENT OP WASTEWATER
Wastewater is treated in an aerating tank containing an uttrafilter which is connected to a suction pump or
pressurized tank The aeration prevents contamination and concentration-polarization on the membrane.
The aeration prevents contamination and concentration-polarization on the membrane.
East Germany DP 233 823 fCI Cn2F9/QQ> March 12. ^9,86
Forschungszentrum Wassertechnik Dresden
BIOLOGICAL AND NATURAL PHYSICOCHEMICAL PRETREATMENT OF CONTAMINATED RAW
WATER
A slow flow-through water reservoir is divided into many reaction chambers by flexible membranes, raw
water is fed to the first chambers, depending on water quality the water is aerated In the first and optionally
subsequent chambers, and the water Is allowed to undergo natural self-cleaning in the last chamber, and a
water preparation plant Is added to the outlet of the reservoir. The mesh size of the first membrane may be
larger than subsequent membranes.
[Indexers: Incinerators]
CALIFORNIA DEPARTMENT OF HEALTH-SERVICES COMPARES ADVANCED.
MOBILE INCINERATOR DESIGNS. The comparison was presented in "Alternative
Technology for Recycling and Treatment of Hazardous Wastes." The discussion
contrasts four types bf thermal destruction systems applicable to site cleanup
ATTACHMENT NO. \
-------
operations. "Each ot the processes can readily destroy wast*. Dils and solve t
residues. The rotary kiln incinerator, the infrared incinerator anc the electric kiln
pyrolyzer can detoxify contaminated soils. The rotary kiln and supercritical solubilizer
units can detoxify wastewaters.
Rotary kiln units: The EPA has developed a transportable incineration
system for on-site thermal destruction of hazardous materials at remote spill
or disposal sites. The system can be broken down into three sections: the
waste shredder and rotary kiln, the afterburner with venturi gas squelch unit
and the scrubber with sound suppressor and fold-down stack. The sections
can be transported on semitrailers. Such a system can detoxify up to two
tons-per-hour of contaminated dirt (about 15,000 tons-per-year) or up to 60
gallons-per-hour of liquid waste oil. Successful trial burns of heavily
chlorinated wastes (like dioxins) were conducted at Times Beach, Missouri
in 1985 with destruction removal efficiencies of 99.999% or better.
Operating costs for the mobile incinerator range from $300 to $1,000-per-ton
depending on waste characteristics. Consequently, these wastes can be
disposed of on-site. Design specifications for the EPA mobile incinerator are
available nationwide to assist federal and state personnel for Superfund
cleanup projects. A few private companies have constructed rotary kiln
incinerators. Several have been permitted by the EPA and are operating
outside California. For example, Ensco Environmental Services (Little Rock,
AR) offers a mobile (seven trailers) unit that can simultaneously incinerate
up to four tons-per-day of contaminated solid material together with 150
gallons-per-hour of liquid hazardous wastes. Ensco rotary kiln units have
been operating in Florida for on-site cleanups. Operating costs are in the
range of $300 to $400 per-ton of contaminated dirt treated.
Infrared incinerators: This comparatively new technology offers tremendous flexibility
for the decomposition of solid hazardous wastes. In a typical infrared incinerator, a
woven metal conveyor belt carried the wastes under infrared heating elements
equally spaced along the length of the insulated furnace. At the end of the furnace,
ash residue is discharged to a hopper. Off-gases are sent to a secondary burner
where they are destroyed by a propane burner. The flexibility of this technology
comes from the operator's ability to vary the residence time (the belt speed), the
temperature (the intensity of the heating elements), and the aeration. The infrared
furnace's capability to operate with very low air flows is a particular advantage. The
process can be operated to minimize emissions of volatile organic compounds and
particulates. Infrared incinerators can be constructed as a mobile unit. Shirco
Infrared Systems of Dallas, TX, is manufacturing several mobile units expressly for
contaminated site cleanup applications. One Shirco unit is employed at a fixed
location for hazardous waste treatment. Several other units are used for activated
carbon regeneration. Shirco reports that their units can achieve greater than
99.999% ORE of PCBs and dioxin. A mobile unit with a capacity of 100 tons-per-day
costs approximately $2.5 million.
Advanced Electrical Reactor The AER unit uses infrared radiation to
pyrolytically destroy.incoming wastes. The reactor consists of a tubular core
surrounded by an annular radiant zone. A porous sleeve separates the two
8
-------
United States
Environmental Protection
Agency
Hazardous Waste Engineering
Research Laboratory
Cincinnati OH 45268
Research and Development
EPA/600/S8-87/Q49 Dec. 1987
SEPA Project Summary
Remedial Action Costing
Procedures Manual
Brian Burgher, Mike Culpepper. and Werner Zieger
The full manual provides specific
procedures for the cost estimating and
economic analysis steps required for
preparing engineering cost estimates
for selecting remedial action alterna-
tives in response to the requirements
of the Comprehensive Environmental
Response. Compensation, and Liability
Act (CERCLA) and the National Con-
tingency Pleh (40 CFR 300). The
manual la designed to be used in
conjunction with EPA's manual
entitled Guidance on Feesibility Stu-
dies Under CERCLA. The audience for
this manual Includes EPA Regional
Project Officers, remedial investiga-
tion/feasibility study contractors, and
state and local remedial action person*
nel end other Pederel agenciea.
Detailed procedures are provided for
generating estimated capital and
annual operating coat*, calculating
annual costs and present worth for
each remedial action alternative, and
performing sensitivity analyses of the
cost estimates to determine the Impact
of changes to varioua cost input
parameters. Worksheets are provided
to assist the user in developing the
feasibility cost analyses. An example
coat analysis is provided to illustrate
these procedures.
Thit Project Summary wet devel-
oped by EPA't Heierdout Witt* Engi-
neering Reteetch Leboretory. Cincin-
netl. OH. to announce key finding! of
the reteerch project thet it fully doc-
umented in e separate report of the
teme title (tee Project Report ordering
infotmetion et beck).
Introduction
As the number and scope of remedial
actions at uncontrolled hazardous waste
disposal sites increase, there is a cor-
responding need for standard procedures
for preparing engineering cost estimates
for proposed remedial solutions. The
National Contingency Plan(NCP) of 1982
outlines a general approach for conduct-
ing remediation at Superfund sites. The
purpose of the manual is 'to provide
specific procedures for the cost estimat-
ing and economic analysis steps required
for the various remedial action planning
phases. These phesos are:
Preparing an initial assessment of
remedial action alternatives to estab-
lish a general cost for the remedial
investigation/feasibility process and
initial remediel meesures,
Screening remedial action alterna-
tives during feasibility analysis to
eliminate those alternatives for which
the costs are substantially greater
than other alternatives and yet do not
provide e commensurate public health
or environmental benefit,
Preparing detailed cost estimates for
feasibility studies to sid in selecting
a remedial action alternative.
The full manual presents -procedures
and provides worksheets to accomplish
the cost analysis objectives of the
phases. The guidance presented has
been developed for generalized condi-
tions at uncontrolled hazardous waste
disposal sites. The user should modify
these procedures where necessary to
accommodate site-specific conditions.
ATTACHMENT MO. 2
-------
The anticipated audience for this
document includes those persons
responsible for Superfund remedial
actions: EPA Regional Protect Officers.
remedial investigation/feasibility study
contractors, state and local remedial
action personnel, and other Federal
agencies. This manual should also be
useful to those involved m planning
state- and private-lead and voluntary
actions, particularly to demonstrate to
the public and other interested parties
that the selection of a remedial action
alternative was conducted according to
EPA approved procedures.
Discussion
Chapter 2 of the full manual address**
initial site planning -&n estimating. Site
response assessment costing ir impor-
tant because it sets the tone for the
remedial investigation and feasibility
study, and gives a rough estimate of the
level of effort and funding necessary to
address sue problems under Superfund.
It is a fairly straightforward procedure
requiring minimal time and effort to
complete. It should be noted that these
initial costs will be superseded by costs
derived during the feasibility study
process.
The purpose of Chapter 2 is to define
the steps of the site response assess-
ment process and offer guidelines as to
how costs for each of these steps may
be derived. With this dats the reader
should be able to:
Assign 'order-of-magnitude costs to
the applicable sections of the site
response assessment report format.
Determine total order-of-magnitude
costs for site RI/FS activities and
remedial alternatives available at the
site assessment stage.
Chapter 3 describes standard proce-
dures for preparing cost estimates for
remedial action alternatives being eval-
uated during feasibility analysis. Initial
site planning cost estimating efforts.
discussed in Chapter 2. are based solely
on existing data and are performed to
provide general cost data associated with
activities. Remedial Removal
measures, and remedial alternatives.
Feasiliby study cost estimating is aimed
solely at providing remedial altnernj
costs using information collected di __
the investigation. The sections in Chapter^
3 provide an overview of the costing
process for both screening and detailed
feasibility analysis of alternatives: proce-
dures for economic analysis: guidance on
the use of sensitivity analysis: and »n
example illustrating these cost analysis
procedures. Additionally, worksheets are
provided to assist the user to estimate.
analyze, and present costs for remedial
action alternatives.
Britn Burgher and Werner Zieger art with JR8 Associates. McLean. VA 22102:
Mike Culpepper is with CH^M Hill. Atlanta. GA 30333.
Douglas Ammon is the SPA Project Officer (see below).
The complete report, entitled "Remedial Action Costing Procedures Manual."
(Order No. PB 88-113 691/AS; Cost: 914.95. subject to change! will be
available only from
National Technical Information Service
S28S Port Royal Road
Springfield. VA22161
Telephone: 703-487-4650
For information Donald Sanning can be contacted af
Hazardous Waste engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati. OH 45268
-J
Construct a site response assessment
report outline and identify areas
which require estimating.
United States
Environmental Protection
Agency
Canter fo» Environmental R«*«arcn
Information
Cincinnati OH 45268
BULK RATE
POSTAGE t FiES PA!
EPA
PERMIT No. G-3S
Official Business
Penalty lor Private U*e »300
EPA/600/S8-87/049
-------
V O w r 0 > c
REYNOLDS ALUMINUM
Metals Company P.O. SOA 27003 Ricivnona Virginia 232
January 5, 1988
Ms. Elena T. Kissel
Assistant Regional Counsel
Office of Regional Counsel
:U. S. Environmental Protection Agency
Region II
26 Federal Plaza
New York, New York 10278
RE: York Oil Superfund Site
Dear Ms. Kissel:
On December.28, 1387 at the invitation of EPA, Reynolds
Metals Company attended -a meeting in Albany, New York regarding
the York Oil Superfund Site. At this meeting, Reynolds learned
for the first time that EPA was strongly considering the selec-
tion of onsite incineration as the remedial technology at the
York Oil site. This fact contradicts the information contained
in the letter which Reynolds received from the EPA dated November
25, 1987, which states that:
"Following a review of the preliminary recommended
alternative and other alternatives, the United States
Environmental Protection Agency ("EPA") and the New York
State Department of Environmental Conservation ("NYSDEC")
have selected a preferred remedial action for the York Oil
site which at this time includes; site excavation, onsite
solidification of soils..."
Reynolds relied on this letter in it's preparation of
comments on the November, 1987 Addendum Feasibility study, and
focussed the limited time available in commenting on the solidi-
fication option. We frankly do not understand the basis for the
Agency's apparent shift in it's position.
Because of this apparent shift, we believe that we have beer.
denied an effective opportunity to comment on all the options
addressed in the study. Further, we feel that the EPA has not
allowed adequate time for people in the Moira area to publicly
review the study. We believe that if the residents of the area
knew that onsite incineration was the possible technology of
choice, they would have likely commented much more fully on an
option that could adversely impact their community.
-------
22/10
Ms. Elena T. Kissel ....
Page Two
January 7, 1988
Additionally, Reynolds believes that the costs and technical
feasibility of onsite incineration are grossly underestimated in
the Addendum Feasibility Study, and that if onsite incineration
is implemented, the final costs will be substantially higher.
In view of the above, Reynolds requests that the EPA allow
another 30 days for the submission of additional comments on this
Addendum Feasibility Study. We also ask that this letter, and
the Agency's response, be made a part of the Administrative
Record on this site.
If you have any questions, feel free to contact our Mr. John
L. Ooyle at (804) 281-3865, or our Mr. Stephen 0. Ryan at (804)
281-3952.
Lawrence C. Tropfeal/ Jr., P.E.
Director, Environmental Control
Corporate Environmental Control
Department
Ict/sdr
-------
32/19
.^ ;-VC COMPANY OF
. PENNSYLVANI
LCG.it. OCPA.sV .'£*«.
December 29, 1987
ALCOA
via Federal- Express
Ms. Elena T. Kissel
Assistant Regional Counsel
Office of Regional Counsel
U. S. Environmental Protection Agency
Region XI
26 Federal Plaza
New York, New York 10278
Re: York Oil Site, Moira, New York
Addendum feasibility Study
r-
Dear Ms. Kissel:
Aluminum Company of America (Alcoaj. submitted a number of
comments to you on th« Addendum Feasibility Study for the York
Oil Site in a letter dated December 17, 1987. These comments
also were elaborated upon at a meeting held in the New York DEC
Offices in Albany, New York, on December 28, which we
attended. One purpose of this letter is to add clarification
to several of the issues tha-t were discussed at the December 18
meeting, and to that end we have enclosed for your review
documents which support many=-of the comments Alcoa made in its
response to the Addendum Feasibility Study. The other purpose
of this letter is to reiterate to you our concern over the fact
that we were not advised prior to this meeting that the EPA &
DEC were considering thermal destruction of PCS contaminated
soils and sludges as an alternative remedial action.
The letter from Elena Kissel, dated November 25, 1987, which
accompanied the Addendum Feasibility -Study clearly states in
the last paragraph on page one, "Following a review of the
preliminary recommended alternative and other alternatives, the
United States Environmental Protection Agency ("EPA") and the
New York State Department of Environmental Conservation
("NYSDEC") have selected a preferred remedial action for the
York Oil site...." Based on this statement, Alcoa did not
spend time evaluating the various alternatives outlined in the
-------
1 .
Ms. Elena T. Kissel
December 29, 1987
Page 2
report, we felt that such an exercise would be fruitless since
the letter clearly indicated that both the EPA and NYSDEC had
agreed or. the remedial action plan. Since this is evident-ly
not the case, as was stated during the December 28 meeting,
Alcoa requests the opportunity to review and comment on the
various options noted in the Addendum Feasibility Study. We
believe a 30-day extension of the comment period would be
appropriate in this particular instance because the parties
were not made aware of the divergence of the opinion between
DEC.S EPA on the Remedial Action Plan, we are particularly
concerned about the cost estimates used in Alternative 5A since
they appear to be very crude and not suitable as a basis for
decision making. Since no treatability studies have been
conducted, it is difficult to understand how management options
can be evaluated in enough depth to determine which methods can
work, and which method will be most effective. It studies are
planned or are underway, we would appreciate receiving
information on the results so that we may review it as well.
Alcoa's comments and much of the discussion at the December 28
meeting centered acound the "softness" of the data, the lack of
extensive data, the disclaimers that the contractor attached to
the data and his interpretations of the data, and the misuse of
the data for risk assessment purposes. In light of the
magnitude of the decisions that will be made concerning site
cleanup options, Alcoa feels that the data and analyses are
inadequate to provide a basis foe. such decisions. We feel that
a defensible risk assessment should be prepared and that a
better definition of the potential for off-site contamination
should be established before multi-million dollar decisions are
made concerning actions at the gibe.
If the currently available data la to he used for decision
making, Alcoa believes that the data as presented in the RTFS
and Addendum Feasibility Study supports the following actions:
material In the tanks should be removed and disposed of (by
incineration off-site), If PCB levels are above 50 ppm)j that
the tanks should be cleaned and removed; that the drums
containing PCS contaminated solids should be disposed of
'off-site; that the existence of an underground tank system
should be evaluated, and If appropriate*, It should be removed;
that contaminated "hot spots" (above 25 ppm PCS) In the area
south of the site should be excavated and disposed of on the
fenced site; and that the fenced site should be graded and
capped to control infiltration and surface runoff. Our
recommendations are based on the sample data which show no
significant groundwater contamination off-aite, limited
migration of pollutants In groundwater and no migration of PCBs~
-------
Ms. Elena T. Kissel
December" 29, 1987
Page 3
in groundwater, no on-site soil or sludge 'samples which fail
the EP-TQX test, and very low PCB contamination with the
exception of some surface oil samples collected in the drainage
pathway.
Our reading of the reports leads us to conclude that the sludge
lagoons were used primarily for treatment and reclamation of
motor oils and that the "bulking" process used by the agencies
in 1981-1982 have been effective. The surface contamination
appears to be sporadic which supports an inference that the PCB
containing substances were spilled at the site sometime after
the oil reclamation operations ceased. This should clearly be
treated as a spill in accordance with the EPA spill cleanup
policy, and we believe that our recommendations are in
accordance with that policy.
Thank you for the opportunity to comment on this proposed
program. We are interested in seeing that the environment is
protected and that the most ctist effective approach is taken to
insure that the site is properly addressed.
Very truly yours,
Marlene w. Jackson
General Attorney
MWJ/rv
Pncs.
cc: Robert Howe (w/encs.)
James Ludlam (w/encs.)
Melissa Marshall (w/enca
-------
SUPERFUND UPDATE.
December 7, 1987
Page M65
NCP AMENDMENTS 70 LIMIT
USE OF RCRA STANDARDS
The Environmental Protection Agency will not require contaminated
ground water at Superfund sites to be restored to the exacting background
level in EPA's Resource Conservation and Recovery Act standards, says a
draft of proposed amendments to the National Contingency Plan, unless the
cleanup involves a waste regulated by Subtitle C or requires an activity
like one regulated by RCRA.
Significantly, the NCP draft would not define the capping and con-
tainment of hazardous wastes at the Superfund site itself long a
favored agency cleanup method as a disposal subject to regulation
under RCRA. Only waste actually removed from a site would potentially be
governed by RCRA's strict management standards.
The applicability of other environmental laws would be similarly
limited. For example, the NCP would prohibit the use of national
emission standards for hazardous air pollutants under the Clean Air Act
in the cleanup of inactive asbestos disposal sites unless the site is a
former asbestos mill or contains asbestos mill waste. The CAA standard,
however, could be relevant and appropriate.
The RCRA regulations became final in May 1980 and took effect in
November 19BO. Any waste generated before May 1980, therefore, is not a
RCRA hazardous waste and not subject to regulation under Subtitle C;
disposal sites containing waste generated before May 1980 (meaning most
sites on the National Priorities Liatl ary np^ d«fiB»d »* "BCftA-peguiatod
, units. exempting them from Subtitle C also. But toxic wastes deposited
in unlicensed landfills and othf* nnr»gnTat.»»nrfli under Superfund.
The distinction is critical for polluters, who already face enoraoi
cleanup coats and long-tern legal and financial liabilities under
Superfund. Agency regulations require contaminated ground water at
post-1980 waste units to be cleaned up to expensive, environmentally
stringent background levels. They allow EPA to waive the standard by
granting an alternate concentration level (ACL) on a site-by-slte basis
or requiring a maximum contaminant level (MCL) for specific substances
under the Safe Drinking Water Act.
The source or prior use of many wastes at [Superfund] sites cannot
be positively identified. Yet the [Superfund] waste may be identical to
a listed RCRA waste. If [Superfund] waste would not otherwise exhibit
the characteristics that would make it a RCRA hazardous waste ... the
RCRA regulations for hazardous waste would not be applicable to manage-
(Continued)
-------
*« u66 Hazardous Waste News
EPA TO LIMIT USE OF RCRA STANDARDS (Continued)
ment of the [Superfund] waste. Certain RCRA regulations, especially the land-
disposal restrictions and design and operating requirements, would very likely
be relevant and appropriate to management of [Superfund] waste," it-Jays.
RCRA ground-water standards are potentially relevant and appropriate only
when "two basic Jurisdictional prerequisites" are satisfied: 1.) The substance
being cleaned up must be a RCHA hazardous waste; 2.) it must b« removed from a
RCRA-regulated waste unit or destined for final treatment, storage or disposal
in a RCRA-regulated unit. '
"If the Jurisdlotlonal prerequisites are not fulfilled, then none of the
[RCRA] treatment, storage and disposal regulations will be [legally] applicable.
Note that these requirements may still be relevant and appropriate, the NCP
draft continues.
"Because many [Superfund] actions address RCRA waste that is not in a
RCRA-regulated unit, a key question in evaluating whether the basic Jurisdic-
tlonal prerequisites are fulfilled is the determination of which [Superfund
cleanups] involve movement of RCRA hazardous waste in such a manner as to
constitute treatment, storage or disposal. EPA has determined that any physical
movement, alteration or disturbance of RCRA hazardous waste due to remedial
action may constitute a new act that fulfills the XCRA definition of treatment,
storage or disposal." (Emphasis added)
NCP To fleflne Treatment. Storage and Disposal
Treatment will Involve any physical, chemical or biological change that
neutralizes the waste, recovers energy or other resource from it or renders it
nonhazardous, according to the NCP amendment.
Storage covers any Superfund cleanup activity resulting in "the holding of
hazardous waste for a temporary period, at the end of which the waste is
treated, stored or disposed of elsewhere," it says. Disposal occurs only when
RCRA hazardous waste is removed from a non-RCRA unit at a Superfund site and
disposed of in a RCRA-regulated unit. Disposal occurs when waste is removed
from the Superfund "unit/* treated and returned to its original location.
The shifting of hazardous waste within a non-RCRA unit is not disposal.
Similarly, the- covering and sealing off of hazardous waste at a non-regulated
[Superfund] unit, called *capping with waste in place,' is also not considered
disposal, and RCRA disposal requirements would not be applicable."
RCRA Tiffnd-Diapcsal Restrictions May Apply
If a cleanup involves the excavation and treatment of hazardous waste banned
by RCRA from land disposal, the disposal prohibitions would apply, says the
draft NCP. "If, however, banned waste is merely moved, graded or treated
entirely within the original [non-RCRA] unit, then ... placement has not
occurred, and the land-disposal restrictions are not applicable," it say a.
"Nevertheless, depending upon the site, [EPA or the state] may find that, while
the land-disposal restrictions are not applicable, they may still be relevant
and appropriate."
Another restriction on the use of RCRA standards (the background level, MCL
or ACL) would limit them to sites containing RCRA hazardous wastes that were
deposited after July 1982 the date federal ground-water monitoring rules took
effect and that are leaking into ground water beneath the,site. But the
corrective-action requirements in the 1964 RCHA amendmenta for releases into
off-site ground water would apply regardless of when the waste was deposited at
the aite.
-------
1792
ENVIRONMENT REPORT^
Compensation, tad Liability Act (superfund law) for costs at
the Conservation Chemical Co. hazardous waste site in
Missouri (Armco Inc. v. Maryland Casualty Co., US
SupCL No. 87-744).
A federal district conn held in favor of the insurance
company, determining that the insurer had no duty to defend
or indemnify because the government's suit was for injunc-
live relief, and not for damages that would normally be
covered under the policy (Maryland Casualty Co. v.
Armco Inc., DC Md: 24 ERC 1980).
The U.S. Court of Appeals for the Fourth Circuit rejected
the lower court's distinction between legal and equitable
forms of action, holding instead that the insurance policy's
own definition of damages excluded coverage of response
costs (Maryland Casualty Co. v. Armco Jnc_ CA 4; 26
ERC 1281).
In its petition, Armco Inc. the waste generator, argued
that the court rulings were contrary to the purposes of the
superfund law. creating an artificial distinction between
action under the law for damages to natural resources.
which would be covered under a policy, and actions for
response costs, which would not
Armco said the approach used by the courts frustrated the
congressional policy behind the superfund law's enactment
and ignored decisions by other courts that held that response
costs were damages for the purpose of comprehensive gen*
eral liability policies. Armco also argued that the approach
taken by the lower courts would create obstacles to the
efforts of parties to take remedial actions at hazardous
waste sites.
Drinking Wittr
SCIENTIFIC BASIS FOR REGULATIONS
DISCUSSED BY SCIENCE ADVISORY BOARD
The diversity of the more than 200 types of polychlori-
nated biphenyl compounds can lead to regulatory headaches
and scientific wrangles, such as those Environmental Pro-
tection Agency regulators experienced at a Science Adviso-
ry Board meeting Nov. 19.
The EPA science advisers met to suggest how the agency
can regulate PCBs found in drinking water. According to
Krishan Khanna. a pharmacologist in the criteria and stan-
dards division of EPA's Office of Drinking Water, the agen-
cy could
IB humans is another challenge, Because RIBs of ten art
contaminated with polychlorinated dibenzofurans, EPA offi-
According to Stephen Hamilton, manager of environmen-
tal science and technology * General Electric Corp., recent
research on PCs-containing sediments in waterways shows
that heavily chlorinated compounds are being converted Into
PCBs with less chlorine. _
Dechlorination is important because PCBs containing
fewer chlorine atoms are more soluble, more volatile, and
break down more easily than heavily chlorinated biphenyls,
Scientists speculate that anaerobic bacteria in sedimes
with no free oxygen remove tome of the chlorine from tht
PCBs. he said.
Hamilton told SNA Nov. 23 that GE researchers found
heavily chlorinated PCBs in the first Inch or so of sediment
from a number of waterways. They expected to find the
same contaminants lower down, but instead found PCBs
containing fewer chlorine atoms, he said.
Most of the lesser-chlorinated PCBs found in the sediment
are orthosubstituted. he said, meaning that the dechlorinat-
ing process tends to work on meta- and para- substituted
compounds. Meta and para-substituted PCBs are the types
of biphenyls that tend to bioaccumulate and are retained in
humans, be explained.
Researchers at Michigan State University are conducting
further study to determine the ecological conditions re-
quired for the establishment of enough bacteria to dechlori-
oate PCBs at an observable rate, Hamilton said.
Hazardous Wait*
INTERIM PROCEDURES FOR SEEKING ASSESSMENTS
OF HEALTH EFFECTS OF TOXIC RELEASES ISSUED
Interim procedures for individuals to follow when re-
questing a health assessment of releases of hazardous sub-
stances were issued Nov. 24 by the Agency for Toxic Sub-
stances and Disease Registry.
The health effects assessment process is required under
the Superfund Amendments and Reauthorization Act of 1986
and the Resource Conservation and Recovery Act. Com-
ments are being requested on the interim procedures to help-
the agency prepare final regulations for the initiation and
conduct of health assessments and other health effects stud-
ies (52 FR 45018).
ATSDR said the purpose of a health assessment is to
evaluate information on the release of hazardous substances
by assessing current or future impacts on public health;
developing health advisories or other public health recom-
mendations; and identifying studies or actions needed to
evaluate, mitigate, or prevent human health effects.
The request should contain:
» The name, address, and telephone number of the person
requesting the assessment;
» The organization that the person represents, if any,
»The name and location of the facility or release of
concern,
» A statement of what is the perceived potential health
pnpblem associated with toe release;
statement requesting HHS to perform a health assess-
it; and
Any other information, such as the nature and amount of
release, substances of concern, media contaminated,
pathway for human exposure, and agencies which
been notified or have investigated the facility or
release.
The requests should be sent to Associate Administrator,
Agency for Toxic Substances and Disease Registry, Chamh-
lee. Building 27.1600 Clifton Road NJL, Atlanta, Ga. 30333.
Written comments on the interim procedures should be
submitted by Feb. 22, 1981, to Director, Office of External
Affairs, Agency for Toxic Substances and Disease Registry,
Chamblee 28 South, at the address above; specify docket
number ATSDR-3.
For more information, contact Georgi Jonea, director.
Office of External Affairs. Agency for Toxic Substances and
Disease Registry. Chamblee 28 South, at the above address;
telephone (404) 454-4620.
11-37-17
Copyngm «1W7 Dy Ttw BUTMU el Nttor* AfUrt, inc. WuMngvn. O.C.
001S-«11/f7/>»*JO
-------
COMMENTS OF THE
CHEMICAL MANUFACTURERS ASSOCIATION
ON EPA'S PROPOSED NATIONAL PRIMARY
DRINKING WATER REGULATION FOR PCBS
Table of Contents
INTRODUCTION.
I. THE ONLY EVIDENCE THAT PCBS ARE
CARCINOGENIC IS FROM ANIMAL STUDIES
THAT ARE LIMITED AND INCONCLUSIVE.. ,
A. Epidemiclogic Studies of Highly
Exposed Worker Populations Have
Found No Significant Cancer
Increases .. ..
B. Contrary to EPA's Discussion,
PCBs Have All Been Found to Be
Non-Mutagenic.
C. The Inconsistent Bioassay
Results Show No More Than
Limited, Inconclusive Evidence
of Animal Carcinogenicity.
II. UNJUSTIFIED EXPENDITURE OF SOCIETAL
RESOURCES WOULD BE MANDATED BY
ESTABLISHMENT OF A ZERO RMCL FOR
PCBS 9
CONCLUSION ' 10
-i-
-------
PRIVILEGED & CONFIDENTIAL
DRAFT 2/2496
COMMENTS OF THE
CHEMICAL MANUFACTURERS ASSOCIATION
ON EPA'S PROPOSED NATIONAL PRIMARY
DRINKING WATER REGULATION FOR PCBS
INTRODUCTION
The Chemical Manufacturers Association (CMA) PCS Pro-
gram Panel welcomes this opportunity to comment on EPA's
proposed National Primary Drinking Water Regulations.
50 Fed. Reg. 46936 (Nov. 13, 1985). CMA's PCB Panel has.
been an active participant in numerous EPA rulemakings, par-
ticularly pursuant to the Toxic Substances Control Act
(TSCA), concerning PCBs. Throughout its history, the Panel-
has worked closely with EPA to assure that any PCB regula-
tions are both protective of public health and consistent
with reasonable assessment of potential risks.
As detailed in these comments, we urge that EPA revise
its proposal establishing a Recommended Maximum Contaminant
Level (RMCL) of zero for PCBs in drinking water. EPA errs
in its conclusion that PCBs are a Category I probable human
carcinogen. To the contrary, there is only limited, incon-
clusive evidence of animal carcinogenicity. Accordingly, an
RMCL of zero is unjustified. We propose that the Agency
instead, rely on significant work of the Food and Drug Ad-
ministration (FDA) to assess potential risks of PCB in the
diet and, following its lead, adopt an RMCL of at least 6.5
micrograms (t&) per liter.
-------
-2-
I. THE ONLY EVIDENCE THAT PCBS ARE
CARCINOGENIC IS FROM ANIMAL STUDIES
THAT ARE LIMITED AND INCONCLUSIVE.
Extensive data exists on PCBs. Highly exposed worker
populations have been studied, and the compound has been
administered in a wide variety of toxicology studies. Based
upon this extensive data base, it is possible to reach rea-
sonable conclusions about any potential risk that might be
posed from the presence of PCBs in drinking water. Although
it is not possible to conclude that absolutely no adverse
health effects would be expected, more than enough data ex-
ists to indicate that EPA would err by concluding that it"
should establish a PCS RMCL of zero.
A. Epidemiologic Studies of Highly
Exposed Worker Populations Have '
Found No Significant Cancer
Increases.
Several epidemiology studies of highly exposed worker
populations have found no significant cancer increases. I/
Studies by Brown and Jones and by Bertazzi, et al., of 2,567
I/ Earlier reliance on studies in Yusho is no longer
considered appropriate. Yusho Oil is not representative of
PCB heat transfer fluids, as it contained, in addition to
PCBs, high levels of polychlorinated dibenzofurans and
polychlorinated quarterphenyls. In consideration of these
facts, several scientists have concluded that PCBs were not
the cause of poisoning in Japan or Taiwan. See Kuratsune,
"Yusho," in Halogenated Biphenvls. Triohenvls, Napthalenes.
Dibenzodioxins and Related Products (1980); Chen, et al.,
(Footnote Continued)
-------
-3-
and 1,310 workers, respectively, exposed for up to more than
20 years, have not demonstrated any consistent relationship
between exposure to PCBs and the development of any particu-
lar cancer. 2/ In addition, the more recent epidemiological
review by Lawton, et al., 3/ of medical records of General
Electric workers with PCS exposure in electrical capacitor
manufacture found only minor dermatological effects, unre-
markable physical examinations, and no signs of chloracne
past or present. The study demonstrates that extensive ex-
posure to PCBs in the occupational setting, sufficient to
have established blood levels in excess of 200 ppb, has not
led to any clinical disease states.
(Footnote Continued)
"Polychlorinated Biphenyls, Dibenzofurans, and
Quarterphenyls in the Toxic Rice-Bran Oil and PCBs in the
Blood of Patients with PCS Poisoning in Japan," 5 Amer. J.
Ind. Med. 133 (1984); Kashimoto, et .al., "Role of
Polychlorianted Oibenzofuran in Yusho (PCB Poisoning),"
36 Arch. Environ. Health 321 (1981); Masuda and Yoshimura,
"Polychlorianted Biphenyls and Dibenzofurans in Patients
with Yusho and Their Toxicological Significance," 5 Amer. J.
Ind. Med. 31 (1984); and Kunita, et al., "Causal Agents of
Yusho," 5 Amer. J. Ind. Med. 45 (1984).
2/ Brown, D. P., and M. Jones, "Mortality and Industrial
Hygiene Study of Workers Exposed to Polychlorinated
Biphenyls," 36 Arch. Environ. Health 120-29 (1981);
Bertazzi, P. A., et al., "Mortality Study of Male and Female
Workers exposed to PCBs," Paper presented at the
International Symposium on Prevention of Occupational Cancer
(April 21-24, 1981, Helsinki, Finland).
3/ Lawton, R. W., jet .al., "Studies of Employees
Sccupationally Exposed to PCBs," General Electric Progress
Report (1981).
-------
-4-
The reference in EPA's Health Advisory draft on PCBs
(at page 4) to an increase in incidence of malignant
melanoma in petrochemical workers exposed to PCBs and other
identified chemicals should be afforded .no weight. This
study 4/ is not adequate to reach conclusions of an associa-
tion with carcinogenicity due to the small number of indi-
viduals exposed and the confounding variable of exposure in
the workplace to many other chemicals. Indeed, the study
was subsequently withdrawn for revision and has not been
reissued. 5_/
In sum, substantial human data indicate even heavily
exposed workers have not developed serious clinical diseases
attributable to PCBs. No significant human evidence of PCS
association with any cancer exists.
*
£
B. Contrary to EPA's Discussion,
All PCBs Have Been Found to
Be Non-Mutaqenic.
There is also no significant evidence that PCBs are
mutagenic in test systems or in human populations. It is
thus quite unlikely that PCBs would exert mutagenic activity
4/ Bahn, A. K., et al., "PCBs and Melanoma," 296 New
England J. Med. 108 (1977); Bahn, A. K., et .al., "Melanoma
After Exposure to PCBs," 295 New England J. Med. 450 (1976).
5/ See NIOSH -1977 Criteria for a Recommended Standard:
Occupational Exposure to Polychlorinated Biphenyls (PCBs),
USDHEW, NIOSH Publication No. 77-225.
-------
-5-
in humans. This conclusion is consistent with the lack of
excess cancer incidence observed in PCS-exposed human popu-
lations.
EPA's reference to a mutagenicity assay of
4-chlorobiphenyl as being positive, SO Fed. Reg. at 47002,
is inaccurate. That study is contradicted by numerous other
mutagenicity assays of PCDs. 6/ More significantly, the
co-author of that study has, since its initial release,
6/ McMahon, et al., "Assay 855 Test Chemicals in Ten
fester Strains Using a New Modification of the Ames Test for
Bacterial Mutagens," 39 Cancer R«3. 682 (1979);
Environmental Protection Agency,"Ambient Water Quality
Criteria for Polychlorinated Biphenyls," EPA 440/5-80-068 PB
81-117798 (1980); Bruce and Heddle, "The Mutagenic Activity
of 61 Agents as Determined by the-Micronucleus, Salmonella
and Sperm Abnormality Assays," 21 Can. J. Genet. Cytol. 319
(1979); Schoeny, et a_l., "Non-mutagenicity for Salmonella of
the Chlorinated Hydrocarbons Aroclor 1254,
1,2,4-trichlorobenzene, mirex and kepone," 68 Mutat. Res.
125 (1979); and, Rinkus and Legator, "Chemical
Characterization of 465 Known or Suspected Carcinogens and
Their correlation with Mutagenic Activity in the Salmonella
Tvphimuriua System." 39 Cancer Res. 3289 (1979).
Other studies that support the lack of mutagenic activity
for PCBs include an absence of effects in the micronucleus
test and sperm morphology tests. (Bruce and Heddle (1979)
previously cited; Jensen and Ramel, "The Micronucleus Test
as Part of a Short-Term Mutagenicity Test Program for the
Prediction of Carcinogenicity Evaluated by 143 Agents
Tested," 75 Mutat. Res. 191 (1980).
Others have shown that PCBs did'not produce chromosomal
damage in the testes and bone marrow of rats (Dikshith, et
al., "Effect of Polychlorinated Biphenyl (Aroclor 1254) on
Rat Testes," 22 EXP. Mol. Path. 376 (1975); and, Carthoff,
et al.. "Biochemical and Cytogenetic Effects in Rats Caused
by Short-Term Ingestion of Aroclor 1254 or Firemaster BP6,"
3 Toxicol. Environ. Health 769 (1977).
-------
-6-
indicated that he had been unable to replicate the result
and that it was his belief that PCBs should not be consid-
ered mutagenic. 7/
In sum, no prediction of PCB carcinogenicity can be
based on any mutagenicity results.
C. The Inconsistent Bioassay
Results Show No More Than
Limited, Inconclusive Evidence
of Animal Careinogenicity.
EPA's placement of FCBs in Category I based 'upon*
strength of evidence of carcinogenicity is inappropriate.
The existing data do not provide "sufficient human or animal
evidence of carcinogenicity" to warrant the regulation of
PCBs "as probably human carcinogens." 50 Fed. Reg.
at 46948. To the contrary/ the only possible evidence indi-
cating Carcinogenic risk is in the animal data and that data
should place PCBs in Category III as it is "inadequate" evi-
dence of carcinogenicity. Id. At a minimum, PCBs should be
placed in Category II given that the evidence is "limited
inconclusive evidence ... from animal data." Id.
A sizeable volume of animal experimental work has been
directed toward chronic studies of PCBs in several species:
the mouse, the rat, ahd the dog. The data shows: (a) the
?/ See Affidavit of Dr. Steven Safe submitted to EPA in
1980 by the Dow Chemical Company. A copy of that affidavit
is attached as Appendix A to these comments.
-------
-7-
evidence on carcinogenicity is negative for gastrointestinal
carcinoma and bladder carcinoma in rats and hepatocellular
carcinoma in the dog; and (b) some studies have reported an
increase in hepatocellular carcinoma in mice and rats ex-
posed to commercial PCBs, whereas other studies have afford-
ed negative results. §/
Animal studies thus do not provide convincing evidence
that PCBs induce liver cancer. Of the major studies in the
rat, one has been judged positive and two have been nega-
tive. Hepatic carcinoma was not produced in the dog.
Chronic administration of Aroclors in the rat did not induce
bladder cancer, gastrointestinal carcinomas, or cancer of
§/ See, Ito, N., et al.. "Histopathologic Studies on Liver
Tumorigenesis Induced in Mice by Technical Polychlorinated
Biphenyls and Its Promoting Effect on Liver Tumors Induced
by Benzene Hexachloride," 51 J. Natl. Cancer Inst. 1637
(1973); Kimbrough, R. 0. and R. E. Linder,"Induction of
Adenofibrosis and Hepatomas of the Liver in BALB/cJ Nice by
Polychlorinated Biphenyls (Aroclor 1254)," 53(2) J. Natl.
Cancer Inst. 547-52 (1974); Kimbrough, R. D., ±t jil.,
"Induction of Liver Tumors in Sherman Strain Female Rats by
Polychlorinated Biphenyl (Aroclor 1260)," 55 J. Natl. Cancer
Inst. 1453-59 (1975); Calandrm, J. C., "Summaryof
Toxicological Studies on Commercial PCBs," Proe. of the
National Conference on Polychlorinated Biphenyls Chicago,
111. (Nov. 19-21, 1975); Calandra, J. C., "Summary of
Toxicological Studies on Commercial PCBs," Proc. of the
National Conference on Polychlorinated Biphenyls 35-42
(EPA-560/6-75-004) (1976); National Cancer Institute,
Bioassay of Aroclor 1254 for 'Possible Carcinogenieity,
No. 38, NEW (1978); Preston, B. D., et al., "Promoting
Effects of Polychlorinated Biphenyls (Aroclor 1254) and
Polychlorinated ' Dibenzofuran-free Aroclor 1254 on
Oiethyl-Nitrosamine-Induced Tumorigenesis in the Rat," 66(3)
J. Natl. Cancer Inst. 509-25 (1981).
-------
-8-
the thyroid gland, pituitary gland, adrena: gland, uterus,
lung, hematopoietie system, or other organs.
Much of the controversy surrounding this issue involves
interpretation of the significance of hepatocellular prolif-
erative lesions noted in some but not all chronic studies.
The liver is a target organ for PCBs. Chronic administra-
tion of sufficiently high levels of PCBs can induce liver
injury in animals leading to a proliferative response in
hepatic tissue. This response may be reparative in nature
and not due to genotoxic interactions, since the majority of
short-term assays demonstrate that PCBs are not mutagenic.
Based on these considerations, exposure to moderate environ-"
mental levels of PCBs would hot pose a significant
carcinogenic risk to man.
. In aum, EPA, errs by concluding that PCBs are a probable
human carcinogen. There is "inadequate" evidence of
carcinogenicity-.to reach-such a conclusion. At best, PCBs
t. » * . j "- , ...-.-
should be placed in SDWA Category XI given the "limited in-
conclusive evidenceu^..*- from animal data." Accordingly, -
consistent, with. JRMCta proposed for other substances at this
time, a numerical number other than zero should be estab-
lished as the RMCL for PCBs.
-------
-9-
II.
UNJUSTIFIED EXPENDITURE OR SOCIETAL
RESOURCES WOULD BE MAN^ATEU\BY >,
ESTABLIS>iMENT\Or A\2£RO\RMCLNfOR PCS a.
of zero foX PCBs, it
be
Should EPA \adopt an RM
uired to fashion a Maximura:ontaminant \evel (MCL)\for
PCBs
will, in
istance.
water
riate to
zero
limit*
require\either monitoring and/oi
water in\this country for this
expected tbi be found \n most
very low levels, it would be
»gulatory ac\ion based ok an
inevitably muc\ monitorinc
Much more Xppropriate, ^iven the.
evidence of carcrnogenicity, frould be
time of\a reasonably limit for a\PCB
RttCL. -
CMA thus urges that the RMCL for PCBs not be zero. To
the extent the Agency-feels a need to establish any RMCL, an
appropriate basis for such a value would be in FDA's conclu-
sion that the tele-ranee for PCBs in the diet should be
13 micrograas per-day. The Food and Drug Administration has
set a tolerance limit of 2 -ppm PCBs in*(21 C.F.R. Part 109).
Assuming an average person's daily consumption of 8CBa >»
. Ǥ..-
fishyis 6.5 Jb/d*-y*. the average daily consumption of PCBs
t*i+lfp~
-------
-10-
would be 13 vg/day. Accordingly, given average human, con-
sumption of 2 liters of water per day, the RMCL would be
6.5 ug/liter.
CONCLUSION.
CMA urges that an RMCL for PCBs be established at at
least 6.5 ug/liter. Such an RMCL would be more than ade-
quately protective of the nation's drinking water supplies
and would more accurately reflect the substantial epidemi-
ologic and toxicologic information on this substance.
-------
FEVEN P. MILLARD'
Room ror Improvement
over environmental degradation have led to leg-
luion such as the Clean Water Act, the Clean Air Act,
.e Resource Conservation and Recovery Act, and the Com-
-ehensive Environmental Response, Compensation, and
lability Act. Federal and state laws have mandated the
-cation of numerous monitoring programs to ensure or de-
:rmine the integrity of the environment. Unfortunately,
lany of the resulting monitoring programs shed linle light
n the questions that they were meant to help answer because
icy are based on poor experimental designs and a super-
cial knowledge of statistics. This article briefly discusses
3me past environmental-monitoring designs and current
:gulations for monitoring at hazardous-waste sites. The
efficiencies of many monitoring programs are linked to the
eed to improve the status of the career category "statis-
ician" in the sphere of the environmental sciences.
CEY WORDS: Environmental Protection Agency;
jroundwater monitoring; Power. Sampling design; Statis-
ician; Type I error Type II error.
1. INTRODUCTION
«past two decades have witnessed a flurry of legis-
in the realm of environmental protection, such as the
Water Act, the Clean Air Act, the Resource Con-
servation and Recovery Act (RCRA), and the Comprehen-
sive Environmental Response. Compensation, and Liability
Act (CERCLA). Anderson (1983) gave a more comprenen-.
sive list of federal environmental legislation, and Loftis et
al. (1986) discussed legislation pertinent to groundwater
monitoring. Federal, state, and local statutes have mandated
extensive environmental monitoring, ostensibly in Uw in-
terest of environmental protection. Consequently, mere has
been an increase in the demand for professional* U> design
monitoring programs, analyze the resulting data, and im-
plement environmental policy decisions based on these re-
sults. But who designs these monitoring programs? Who
analyzes these data? And how much statistical reasoning
-goes into environmental policy decisions?- Based on my
limited experience, it appears mat many environmental
monitoring programs lack grounding in fundamental statis-
Strvta P. Milted i* DBMOT. Bwatiiocsl Coocotens. UM.
i of 800000. Uwvmcy at Wnhnfn. Snofe. WtMntva «I93.
The oritiaai dnn of ito nek WM wrioea white to Mdnr « Vuioat
Auinm Prafcuor. Sm«uc*«d Applied Probability PiofnB. Uarwmty
of CttifonuA. Sttti BotMn. It ww picpmd fren (tit mt of > tridreu
Qupw of tte Southern California Qupur of (he
AuacittOB. Tht mtar it fractal to te inuabm
poop. Dams Lcncnmaicr. N. Phillip ROM. Robot Hindi. ma
i referee* for tfwr helpful couunenu. uiuciuat. «d wg-
tical principles, precluding any meaningful data analyses
meant to address the nominal network objectives. Qualified
statisticians simply are not consulted in the design, analysis.
or policy-making phase of environmental-monitoring pro-
grams, or else they are ignored. My goal in this article is
to alert th« statistical community to the sometimes shoddy
state of statistics in environmental-monitoring, in the hope
of encouraging more participation and research in this area.
2. STATISTICAL NATURE OF
ENVIRONMENTAL MONITORING
Environmental monitoring may have many objectives.
including estimation of baseline physical or biological pa-
rameters (e.g.. what is the background concentration of
mercury in a particular stream?), detection of standards vi-
olations (e.g., does the concentration of mercury. ID the.
effluent from a certain paper mill constitute a violation of
the effluent permit?), and/or determination of the presence
or absence of a change or impact hi an area (e.g.. has Love
Canal contaminated the soil in adjacent neighborhoods?).
Environmental-monitoring data refect multiple sources of
variability, including natural fluctuations in space and time.
as well as deviations resulting from Held measurement, stor-
age and handling, and laboratory analysis techniques. If it
is important to be able to detect a change on the same order
of magnitude as the variability inherent in the variable, then
careful consideration of statistical issues is essential.
The usual procedure for analyzing environmental moni-
toring data is to set a (the probability of making a Type I
error) to some fixed value (conventionally .01. .05. or. 10).
apply some standard statistical test, and then appropriately
reject or not reject the null hypothesis. Often, little or no
attention is given to the power of the test, and rarely are
the underlying assumptions of the Kt questioned. An "ideal"
procedure for designing an cnvironmental-roonitoring pro-
grim might be somehow to quantify the cost of making a
Type I error and a Type U error and then to balance the
levels of a and ft (the probability of a Type U error) to
reflect these costs, perhaps through a decision-theory ap-
proach (Berger 198S). it should be stressed, however, that
the basic question of how to balance both kinds of errors is
not a statistical one but a political one. Examples in the next
section show how this issue is improperly handled or com-
pletely ignored in many environmental-monitoring pro-
grams.
3. EXAMPLES OF STATISTICAL
SHORTCOMINGS IN ENVIRONMENTAL-
MONITORING PROGRAMS
In this section, three examples are given that illustrate
some of the statistical problems that prevail in environ-
mental-monitoring programs and regulations.
O 1987 American Statistical Atsorfon'M
77k* American Statistician. November 1987. Vet. 41. He. 4 249
-------
Difference and Minimal Detectable Difference
In 1979, the Analytic Sciences Corporation developed
what dtey termed optimal environmental-monitoring pro-
grams for the New England Power Service Company (Duffy
and Luden 1979). The aim of these programs was to detect
any change over time in the abundance of certain organisms
in Mount Hope Bay in southern Massachusetts. The effec-
tiveness of a monitoring program was defined in terms of
"average differences in the [between times] . . . catches
which are detectable with a given confidence level" (Duffy
and Luden 1979, pp. 2-5). Suppose X, and X2 denote the
mean catch in yean 1 and 2. respectively, based on n, and
/ij observations. Let 7 » Xj - X\ denote the difference in
means between these two yean, and let r? denote the es-
timated standard deviation of 7. Duffy and Luden (1979,
p. A-2) stated that the smallest detectable population change,
A. can be estimated as
A - r.j?. (3.1)
where ta denotes the critical value of the t statistic for a
two-sided- level-o test with (n, + «2) - 2 df. Apparently.
Duffy and Luden believe that if a * .05. nt » /ij » 12,
and ip « 1, then a change of A - 2.074 could be detected
with 95% confidence. Actually, the correct value is A
(2.074 + 1.717) - 3.79 {.assuming that observations be-
tween sampling occasions are independent. Equation (3.1)
is of course incorrect because account has not been taken
of the power of the t test; it is the formula for the least
significant difference (SokaJ and Rohlf 1981. p. 244), not
the minimal detectable difference (Sokal and Rohlf 1981.
p. 262). In fact, A in (3.1) is identical to the minimal
detectable difference for a power of .5. Ironically, Duffy
and Ludcts (1979, p. A-2) stated that "account was not
taken of the Type II error. ..."
Duffy and Luden (1979) based part of their research on
dw work of Saila. Pikanowski, and Vaughan (1976), who
confused minimal detectable difference with least significant
difference in exactly the same way. The "optimal" mon-
itoring designs of Duffy and Luden were approved by Re-
gion I of the U.S. Environmental Protection Agency (EPA)
(Duffy..Luden, and Ketschke 1981). Millard and Lertea-
maier (1986) demonstrated an approach to optimizing sam-
pling designs based on the more appropriate minimal
detectable difference. ~ .*
3.2 Example 2: Enrinwinettaf Monitoring at
LovtCuri ; - - "'; ;'... .'".
In the early 1980s, in response to t state-of-emergency
order by then President Jimmy Cuter, the EPA conducted
a study to determine whether a specific residential area
surrounding Love Canal, termed thtEmergency Declaration
Area, was contaminated (EPA 1982). Chemical concentra-
tions in the toil and odier media were compared among the
Emergency Declaration Area, the Love Canal Area, and a
control area. Problems with the poor experimental design
of this study and the subsequent chemical analyses were
discussed by Anderson (1983). Brass (1985). Office of
(1986. 1987). *\ .. ^nse to thr*. . isms was given by
Seilkop and Akland (1987).
The basic flaws of the study included a lack of sufficient
control sites and samples, a lack of adequate number of
replicates, poor quality control, and pseudoreplication (i.e..
treating correlated observations as independent: see Hurlben
1984). For example, becasueof the small number of cor
observations, in most cases there was no significant dL
ence between the Love Canal area and the control area (Ol
1983). Despite the shortcomings of this study, it was the
major input to the habitability decision reached by the U.S.
Department of Health and Human Services (DHHS). which
concluded in 1982 that the Emergency Declaration Area "is*
as habitable as the control area* with which it was com-
pared" (DHHS 1982; also cited by OTA 19X3. p. 6).
3.3 Example 3: Current Legislation Requiring
Groundwater Monitoring at
Hazardous-Waste Sites
Sections 264 and 265 of Title 40 of the Code of Federal
Regulations (Office of the Federal Register I986a), estab-
lished under the authority of RCRA, mandate groundwater
monitoring in the vicinity of hazardous-waste sites. Section
265 outlines a nominal-groundwater-monitoring program for
an interim period (before the site gains an approved permit)
and requires the submission of a more comprehensive mon-
itoring program for subsequent operation. Section 264 out-
lines regulations for monitoring during the active life of the
facility.
- Section 264 addresses three kinds of mon
grams: detection monitoring, compliance monitoring
,cprrecove-acti«n monitoring. Section 265 deals only
detection monitoring. Only detection monitoring under Sec-
tion 264 will be discussed in this example. The aim of
detection monitoring is to determine whether die concen-
trations of certain chemicals have increased above their nor-
mal background levels in die groundwater near the hazardous-
waste site. The regulations provide that background level
is to be defined by one year of quarterly sampling of the
control welKs) (hydrauucauy upgradient from the waste ate).
Sampling at the compliance-point wells (hydraulically
downgradient fronj the waste site) must occur at least semi-
annually during the life of the facility. (Detection monitoring
under Sec. 265 requires at least one upgradient well and
dute downgradient wells.)
Each time diat a single well sample is taken at a com-
pliance-point well, at least four subsamples must be taken
(i.e., four laboratory replicates are measured). These sub-
samples are referred to as "portions" in Section 264, tnd
as "aliquots" in die groundwater-monitoring literature (see
Groenevdd and Duvai 1985; McBean and Rovers 1985).
When sampling is performed at die control well(s). Section
264 requires at least one sample per well and at least four
samples from all of the control wells. Hence, if only one
control well exists, four "replicates" from this
be taken each quarter during the sampling year
determine the background levels. It is ooc clear
2SO Tht American Statiaieim. Novtmbtr 1987. Vol. 41. No. 4
-------
these four "replicates" for a single control well should be
well (Held) replicates or laboratory replicates (aliquot*) The
mandate of aliquots for compliance-point wells, however.
apparently leads most or all hazardous-waste operators to
use aliquots (McBean and Rovers 1985).
On each sampling occasion, for each compliance-point
well and each chemical of interest, the hazardous-waste
operator must compare the average of the four aliquots at
(he compliance-point well with the average background level.
If the background level coefficient of variation (cv) is less
than or equal to 1 (a check against gross nonnormaliry), the
comparison is made using Cochran's approximation to the
Behrens-Ftsher Student-r test at the .05 level. If the cv is
greater than 1 or if the regional EPA administrator so directs.
an "equivalent" statistical procedure is used. It is not clear
what observations and what avenges are used. Should the
operator treat the 4 aliquots from the compliance-point well
as 4 independent observations from one population and the
16 observations from the single control well (4 aliquots for
4 quarters) as independent observations from another pop-
ulation? This is what the EPA has directed (Loftis et al.
1986, p. 87). The more appropriate analysis uses the av-
enges over aliquots as the observations, in which case there
are four observations for the control well and one for the"
compliance-point well. It is also not clear how data from
different quarters for die same compliance-point well are
combined. Apparently, past data at a compliance-point well
are ignored.
If the results of the t test yield a significant difference
between a compliance-point well and the control well(s).
the hazardous-waste operator is instructed to obtain a new
compliance-point well sample and four aliquots from this
new sample and to repeat die t test based on these new
aliquots. If the result is again significant, only then does
die operator conclude that a statistically significant change
has occurred and subsequently notify the regional EPA ad-
ministrator of the finding.
There are obviously several deficiencies in these require*
menu for groundwater monitoring. Fust is the idea of pre-
ferring aliquots (lab replicates) to well replicate* (field
replicates). McBean and Rovers (1985) recognized that both
intersample and intrasampie variability should be taken into
account when comparing compliance-point and control wells.
They present analyses of certain groundwater constituents
and show that the concentration of these constituents in a
well sample increase* with die number of bailings, at me.
well. They also incorrectly conclude that intersampk (be-
tween bailings) variability is significantly greater (p < .05)
than intrasampie (between aliquots) variability by using a
conventional F test. What diey actually show is that inter-
sample variability is significantly greater than 0. because,
that is what the F test is testing. Results given in their paper,
however, can be used to compute unbiased estimates of both
components of variability. Table 1 shows diat, as is to be
expected, intersample variability can be orders of magnitude
larger dun intrasampie variability. The requirement of treat-
ing aliquots as independent observations is in essence man-
dated pseudoreplication (Hurlben 1984). Groeneveld and
Duval (1985) suggested that it is more cost-effective to
simple more frequently and with fewer aliquots, because
-. at
Vc8««n VK) Rovtrs (1093)
in
Pwnettr
Conductivity
BetwMn
Within
CMorid* (mo/1)
BetwMn
Within
SS
9,207
500
55
24
at
3
8
3
8
MS EMS
3,069 «i * 3
-------
hd geotechnical consulting firms, who usually employ
professionals in areas of expertise other than statistics. As
a result, many reports generated by these firms attempt to
apply textbook statistics, sometimes incorrectly, and almost
always without consideration of the power of the monitoring
design network. Many introductory courses and textbooks
in statistics do not emphasize the concepts of power and
experimental design for lack of time and space (Huribert
1984). For instance, the first edition of Zar (1974) contains
only a brief mention of the concept of power, whereas the
second edition (Zar 1984) devotes numerous pages to the
topic.
. What is the explanation for the widespread misuse of
statistics in environmental monitoring? The problem clearly
is not a lack of research in development of statistical meth-
ods for environmental data (e.g.. Caselton and Zidek 1984;
Eynon and Switzer 1983; Gilbert 1987; Hawkins and Cressie
1984; Hirsch and Slack 1984; van Belie and Hughes 1984).
It is not clear that the problem can be explained by a lack
of technology-transfer mechanisms either The American
Statistical Association (ASA). has sponsored eight symposia
on statistics, law, and the environment (see The American
Statistician. November 1985. Part 2).
Part of the answer may be explained by a difference in
attitude toward statistics between the environmental-moni-
toring field and other professions that have much closer ties
with statistics. The medical and pharmaceutical professions.
for instance, have long recognized the advantages and im-
portance of employing statisticians (Meier 1978). For ex-
ample. the May 1986 issue of Amsiat News contains 10
advertisements for openings for biostatisticians in the non-
academic sector but only one opening in the area of envi-
ronmental monitoring. Salsburg (1985) even argued that in
some cases the medical profession overemphasizes the im-
portance of classical inferential statistics.
On the other hand, "statistician" does not appear to be
a well-defined career category in the sphere of environ-
mental monitoring. For example, the Federal Code of Reg-
ulations requires that any alternative groundwater-monitonng
design for the interim period of a hazardous-waste facility
(other than the one presented ia Section 265) be approved
by qualified geologist or feotechnical erapaeeK even though
this is a sampling-design problem* the tervioM of a statis-
tician are not required as well, hmhennore. rt is up to a
regional EPA administrator to dictate what statistical tests
are to be used for thc-monitoring programs at each hazard-
ous-waste site, yet only I region -out of 10 (Region V)
employs anyone with- the position tide of statistician or
mathematical statistician (EPA Headquarters Personnel Of-
fice, personal communication. Aufust 20, 1986). (In all
fairness, the EPA does employ 28 mathematical statisticians
and 51 statisticians who are based at the EPA headquarters
in Washington, D.C., and me three laboratories in Las Ve-
gas. Cincinnati, and Durham.) :'
Hiring more statisticians at the EPA regional offices may
not solve the problem of respectability, however. U.S. Of-
fice of Personnel Management requirements define a math-
ematical statistician as someone who performs "professional
work involving the development and adaptation of mathe-
matical statistical theory and methodology." yet the basic
requirement i- - \. or B.S. degree (noif necessarily in
statistics) with 12 semester hours of mathematics and 6 of
statistics (U.S. Office of Personnel Management 1981). There
are no additional degree or course requirements specified
for increasing CS levels.
Why is the practice of statistics stronger in drug and
medical research than in the area of environmental
toring? One possible explanation is that the former;
research are much older than the latter. The Food and Drug*
Administration was established in 1931. although its be-
ginnings date back to the Food and Drug Act of 1906. The
area of drug research has had time to build a strong link
with statistics. In comparison, the EPA is a young agency."
established in 1970. and environmental monitoring is a young
science. It will take time for a statistical tradition to develop
in this field.
Minton (1983) noted the general lack of visibility of sta-
tistics as a discipline. He points out that much of the problem
is due to "nonstatisticians" thinking of a statistician as
someone who merely performs computations, not someone
who can also help with the intellectual part of research (such
as the design). The widespread availability of statistical
packages for personal computers and the growing trend to-
ward expert systems may well lead to much less involvement
of qualified statisticians in any phase of environmental-
monitoring studies.
5. HOPE FOR THE FUTURE
There are a few dim rays of hope for an improved role
for statisticians and statistics in environmental monitoring:
I. In September 1983. the EPA announced a new
on the determination of the habitability of the
Declaration Area surrounding Love Canal (Periez 1983).
ostensibly because it found significant migration of chem°
icals during the summer of 1983. but almost certainly be-
cause of statistical criticisms leveled against the past study
as well.
2. Recently, the ASA was asked to review EPA-funded
acid precipitation research. The results of this review were
summarized by the ASA Coordinating Committee (1985).
3. In August 1986, the EPA announced that it was con-
sidering revising the regulations for groundwater monitoring
at hazardous-waste sites and requested "comments from the
public to assist in the regulatory development process" (Of-
fice of the Federal Register 1986b. p. 29812). The public
includes statisticians hired by industry to analyze these reg-.
ulations.
4. The ASA has to date organized three research con-
ferences supported by the ASA and the EPA on environ-
mental monitoring and statistics.
What can statisticians do as a community to nurture the
growing awareness of the importance of statistics in envi-
ronmental monitoring and in general? In the academic sec-
tor, instructors of statistics courses should emphasize to their
students that a few statistics courses will not train
be statisticians but will enable them to communicate
statisticians. (Do a few law courses qualify someone to be
an attorney?) Moreover, as Huribert (1984) suggested, basic
232 The Amfhean Starimeian. Nmtmtxr 1987. Vol. 41, No. 4
-------
c^_. cspu ol experimental design should be included in el-
ementary courses and explained in concise, nonmathemat-
icaJ terms. In the private sector, statisticians should emphasize
their ability to help contractors save money by helping them
to create cost-effective, statistically sound designs. In the
government sector, statisticians should play an integral pan
in the creation of any statutes that involve statistical tests.
They can also emphasize the useful role thai they can play
by helping to provide scientific evidence to justify certain
regulatory actions. In this age of environmental litigation,
statisticians will be appearing more often in the courtroom
to help defend the government's or industry's actions (see
Meier 1986).
It has been more than 10 years since Boen and Smith
(1975) called for the formation of an ASA committee to
study whether statisticians should be certified. We are no
closer to a certification procedure, and in many areas our
profession still lacks the respect it deserves (Minion 1983).
In light of the increasing role statistics plays in public wei-
fare. we owe it to the public to assure quality control in our
profession.
(Received February 1986. Kevited No\-ember 1996.}
REFERENCES
ASA Coordinating Commmee (1983). "ASA Review* of EPA-Funded
Acid Precipitation Research." TV American Statistician. 39. 243-229.
Aadenon. R. L. (1983). "Suiiuici and (ht Environment." AmaaiHe**.
September-October. 3-4.
Berger. J. O. (19*3). Sratmieat Deriao* Tneorr o*d fluiuiari AnaMt.
New Ywte Sprinter-Verlaf.
Boca. J.. and Smith. H. (1979). "Should Statistician* Be Certified?" TV
America* Statistician, 29. 113-114.
Bran. I. 0. (1983). "Why Proof of Safety It Much Mort Difficult Than
Proof of Haari." Biometha. 41. 783-793.
Caaclton. W. p.. md Zidek. J. V. (1984). "Optiwel Moninrinf Nctwort
Dai|ns." Suiiaiet ami Probability Lmen. 2.223-227.
Duffy. J. J.. and Laden, G. (1979), "Optimization of Bnyton Point
Environmental Monitorial Program*." Technical Report 1462*1. An-
alytical Sciences Corporation. Bttif'"|. Man.
Duffy. J. J.. Luden. G.. and Ketschke. t (1911). "Cou*aer»flr Opti-
mization for Biota Mouuoflap, Prefnau." w luwet Attormed With
Impact AuMjmrw. ed. L. B. Jemea. Sparta. MD: E. A. ComnNOt-
cations, pp. 133-147.
Eynon. B. P.. and Swiocr. R. (1913). "Tta V^riatefcy of Rara/allAc-
idity." TV Combat Jam* ofSiaiain. II. 11-24.
Gilbert. R. O. (1987). SmUmie* Mr*** ft* CMmrnmtmal foOmmmm
Monitoring. New Yorlu-Van MuiCTal
Graenevcid. L.. tad DwraL IL.Q. (19ISK "Stabaical Procedures and
Gxundenaons for Environmental Manafement (SPACEMAN)." teca-
oical report. Florida Dna»umm of En»iromaenul R*nlariort. Talla-
Hawkim. D. M.. and Cream. N. (1984). "Robot KrifrnfA Pro-
posal." matntmmieai Ctatofy. 16. 3-18. ,-.
Hindi. R. M.. and Slack. J. R. (1984). "A Nonparamemc Tread Test
for Seasonal Data With Serial Dependence,;'Water RetomrertReifarc*.
20. 727-732.
Hurfben. S. H. (1984). "rSeudorrpttcanon and Urn Design of Eca4opcal
Field ExperwMB." Ecotot**MoHtfrvfta, 54. 187-211.
-, Urfris. J. C.. MuniatMivry. R. H.. H«rrl». J.. Nenies. D., Porter. P, S..
W«tf. R. c.. and Sander*. T. G. (1986). "Monitoring Strategies for
Groundwiter Quality Management." Report 140. Colorado Wtter Re-
sources Research Institute. Fort Collins.
McBean. E. A., and Rovers. F. A. (1983). "Analysis of Variances as
Determined From Replicates vs. Successive Sampling." Ground Water
Monitoring Review. 3. No. 3. 61-64.
Meier. P. (1978). "The Biggest Public Health Experiment Ever The 1954
Field Trial of the Salk Poliomyelitis Vaccine." in Statistics: A Guide
10 the Unknown (2nd ed.). ed. J. M. Tanur. San Francisco: Holdcn-
Day. pp. 3-13.
(1986). "Damned Lian aad Expert Witnesses." Journal of the
American Statistical Aueciaaom, 81. 269-276.
Millard. S. P.. and Lenenmaier. D. P. (1986). "Optimal Design of Bi-
ological Sampling Programs Using the Analysis of Variance." Enuar-
me. Coatial aad Shelf Science. "22. 637-636.
Millard. S. P.. Yearsley. J. R.. and Lenenmaier. D. P. (1983). "Space-
Time Correlation and Its Effects on Methods for Delecting Aquatic
Ecological Change." Canadian Journal of Fisheries and Aquatic Sci-
ences. 42. 1391-1400.
Minton. Paul D. (1983). "The Visibility of Statistics as a Discipline."
Tne American Siatiaicia*. 37. 284-289.
Periex. Jane (1983). "Love Canal Chemicals Leak: Resenlement Plans Set
Back." New York Time*. September 28. pp. Al. 89.
Office of the Federal Register (National Archives and Records Adminis-
tration) (I986a). "Code of Federal Refutations. Title 40. Pans 190 to
399. Protection of Environment." (July I ed.). Washington. DC Au-
thor.
(I986b). "Federal Register. Vol. 31. No. 161.. Part IV. Environ-
mental Protection Agency." August 20. Washington. DC: Author.
Office of Technology Assessment (U.S. Congress) (1983). "Habitabiliiy
of the Love Canal Area: An Analysis of the Technical Basil for the
Decision on me Habitabiliiy of the Emergency Declaration AreaA
Technical Memorandum." Technical Report OTA-TM-M-13. Wash-
ington. DC: U.S. Govunmtm Printing Office.
Prentice. R. L.. and Marek. P. (1979). "A Qualitative Discrepancy Be-
tween Censored Data Rank Tests." Biometries. 33. 861-867.
Saila. S. B.. Pikanowtki. R. A., and Vaughan. D. S. (1976). "Optimum
Allocation Strategies for Sampling Benthos in me New York Bight."
Estuarine and Coastal Marine Sciences. 4. \ 19-128.
Salsburg. D. S. (1983). "The Religion of Statistics as Practiced in Medical
Journals." TV American Statistician. 39. 220-223.
Seilkop. S. K.. and Akland. G. (1987). "Comment oa Swline and Cook
(1986)." TV Amtrtram Suaitticia*. 41. 164.
Sokal. R. R.. aad Rohlf. F. J. (1981). Biometry (2nd ed.). San Francisco:
W. H. Freeman.
Stoline. M. R.. and Cook. R. J. (1986). "A Study of Statistical Aspects
of the Love Canal Environmental Monitoring Study." TV America*
Statistician. 40. 172-177!
(1987). Reply » "Comment OR Stoliae and Cook (1986)." by
S. K. Seilkop and G. Akland. TV America* Suturieia*. 41. 165.
U.S. Department of Health aad Human Services (1982). "DHHS Eval-
uation of Results of Environmental Chemical Testing by EPA in die
Vicinity of Love CanalImplications for Human HealthFurther Con-
lidentions Concerning Habitabiliiy." Washington. DC: Author.
U.S. Environmental Protection Agency (1982). Environmental Monitoring
a Lav* COM/ (Vol. I). Washington. DC: U.S. Government Prinnai
Office.
U.S. afffte of Personnel Management (1981). "Mathematics and Related
Positiuns: Qualificatiom Information Sheet." Form QM500. Washing-
ton. DC: U.S. Ooxrament Printing Office.
.. Vaa Belle. G.. aad Hughes. J. P. (1984). "Nonperametnc Tests for Trend
in Water Quality." Water Resources Retearen. 20. 127-136.
' Zar. J. H. (1974). ft#*»*ia>coM/wf^EMlewood Cliffs. NJ: Prentice-
Hall.
(1984). iiuuiimni Anmfvsu. (2ad ad.). Englewood Cliffs, NJ:
Prenhce-HalL- . _ .
Tne American Slaiiaicim. Novtmbrr 1987. Vol. 41. No. 4 223
-------
N. PHIU.IPROSS*
Comment
1. INTRODUCTION
Dr. Millard's article provides a timely reminder to the
statistical community that statistical methods as well as stat-
isticians have not yet been totally accepted as essential com-
ponents of environmental decision making. Most
environmental program managers, analysts, and scientists
agree that well-designed environmental monitoring systems
are essential to making informed decisions. Therefore, it is
difficult to understand why statistics has not become better
integrated into the process of environmental program man-
agement and decision mking I believe that Dr. Millard
touches on ?omr' aspects r>f this paradoxical problem. I
would liks ;o focus on two areas that I believe can provide
additional insight: the socio-political process that affects
environmental programs and the role of the statistician in
working within 'his socio-political system.
2. THE SOCIO-POLITICAL PROCESS
Societal concern about our environment and its future is
a major force in molding the direction in which our envi-
ronmental programs will go. In the United States we often
deal with these societal concerns by enacting laws that at-
tempt to correct or regulate our perception of the causes.
These laws are enacted through the political process, carried
out by executive agencies, and interpreted by the courts.
Green (1986) cited from a dissenting opinion of Justice
Rehnquist in American Textile Manufacturers, Inc., v. Don-
ovan (1981): ''Statutory law more often than not is based
on emotion, politics, and compromise rather than on co-
herent principles that objective experts would regard as
valid. ..." Most of our monitoring programs are based on
statutes enacted by federal, state, or local legislatures. As
Chief Justice Rehnquist has said, they are often based on
emotion, politics, and compromisevery subjective and
not always amenable to the imposition of sound scientific
design in the implementation process. In fact, many of the
statutory requirements are so detailed they leave the imple-
menting agencies very little room for monitoring design
considerations (National Academy of Sciences 1977). This
does not mean that the lack of flexibility in many of our
environmental statutes totally excuses managers from using
good design principles; however, it does mean that the abil-
ity of the program managers to use objective scientific ap-
proaches can be severely limited. They must operate within
these limitations; they have no choice.
Dr. Millard has chosen an excellent example in Love
Canal. It provides insights into how both political and social
forces can impose themselves on what initially appears to
be a scientific problem. The shortcomings in the Love Canal
N. Phillip Rosa if Chief. Statistical Policy Brand). Office of Policy.
Planing, ind Evaluation. Environmental Protection Agency. Washington.
D.C. »4oO.
study, some of which Dr. Millard has cited, did not result
from the lack of qualified statisticians participating
study, nor were their recommendations totally
of the shortcomings resulted from the attempt by EPA sci-
entists to deal with the realities of the socio-political con-
straints that were imposed on the study.
For a variety of reasons, detailed by the EPA (1982). a
series of events resulted in President Carter declaring 'on
May 21. 1980. that a state of emergency existed at Love
Canal. This led to the initiation of the EPA Love Canal
monitoring studies. EPA site-sampling activities began on
August 8. 1980. and concluded on October 31. 1980. The
entire data-collection activity took place in less time than it
would have taken to develop statistically valid sampling
designs and to arrange for the logistical support to undertake
the sample-collection activities. It would have been desir-
able to have had several months to just work on the study
design; however, this was not possible. The political and
social pressures demanded that visible federal action be
taken immediately. We know that front-end planning would
have eliminated many of the study's shortcomings; we know
it now. and we knew it then. At die time of the "crisis."
however, the socio-political pressures and constraints did
not permit any other actions to be taken. The EPA (1982)
described its Love Canal monitoring effort as follows:
. The monitoring program performed by EPA at Love Canal iavolj
collection and analysis of approximately 6.000 field samples.
Love Canal study the most comprehensive multimedia monitoring!
ever conducted by EPA at a hazardous waste site. The precisios ad
accuracy of the cBvuoBmemaJ neasurements "^unf
through application of aa extensive quality assurance program. Ai a result.
this study exemplifies die design and execution of a snuB-of<
-------
Scientists are trained to use logic and objective thinking:
in contrast, decision making requires taking action with a
minimum of information; often theories art modified with
each new problem. Both art legitimate approaches; bow-
*we must keep in mind that they serve different pur-
. Green (1986) quoted Philip Handler, then president
i National Academy of Sciences, regarding the EPA's
decision on DDT:
I have uudettvne dx masochistic exercise of reading a considerable part
of those hundreds of pates of mtimotiy which were taken by the heating
'rummer (who would oow be called ta "administrative law judge"]. . . .
I tay "masochistic" quite deliberately. Two-thirds of the to-called "sci-
entific evidence" that I read could not have found acceptance by (he
editorial board of » reputable journal. The tiniation was summed up. I
thought, in the remarkable words of the Administrator of EPA in the
decision statement in which DDT was effectively banned for most pur-
poses. As close u I can recall it. he said "DDT constitutes an unquan-
ofiabte hazard of uncertain nature." ... I am not disagreeing with the'
decision itself. It is the basis for decision making, the inadequacy of the
data, which I found so very troublesome. I still do. ... (p. 122)
Green concluded by pointing out that "U is clear that Dr.
Handler was unable to accept the legal culture" (p. 222).
Statisticians-may not wish to accept the "legal culture"
either, however, we need to understand that it is a way of
thinking that strongly influences what is and what can be
done in environmental monitoring programs.
3. THE STATISTICIAN IN THE
SOCIO-POLITICAL ARENA
Marquardt (1987) defined the primary role of the statis-
tician as that of a "purveyor" of the scientific method.
Statisticians must become "entrepreneurial practitioners"
and consultants for the scientific method. You cannot lit.
comfortably in your office and expect the managers and
decision makers to come knocking on your door insisting
that you get involved early enough to ensure that the sci-
entific method will be followed and. that die resulting in-
formation will be acceptable for the types of decisions that
need to be made. Statisticians need to raise their concerns
to their management and not wait for management's aware-
i of the problem to result from concern filtering up through
the chain of command. Most managetvwant to. know when
they are heading for disaster, it is the obligation of their
statistical staffs to help them become aware of impending
disaster long before they reach a point of no, or not easy.
return. The managers are operating ondex the forces of the
socio-political environment. They ate not going to seek out
the ttatifliiniM! the statisticians most, if they want to have
an impact on the decision making early enough in the pro-
cess, seek out the manager. The statistician ha* to convince
the decision makers that statistics can help them make better
and more defensible decisions and that .statistical dunking
is not so rigid as to preclude the flexibility needed in decision
making.
ust be careful not to fall victim to the "shoot
the messenger" syndrome. To avoid being the bearer of
bad news, you must be aware of what is happening and to
try to anticipate where your expertise may be needed. Too
often 1 hear the statement dutt 1 am only concerned with the
analyses, not the policy decisions. This is a very
narrow v\e/w to :ake of one's responsibility, if the statistician
is not aware of t^e policy context in which decisions are
made, then it is doubtful that he or she will be given the
responsibility of being concerned about the statistical anal-
yses. If every time a statistician is asked by the manager to
assist in the analysis or interpretation of imperfect data (usu-
ally data that you were not responsible for collecting), they
respond by stating that there is nothing~you can do to help
the data were not collected properly, or there is no time to
design and implement a survey of that sizemanagers will
stop asking statisticians for assistance. It does little good to
point out errors in the analysis or inadequacy of the data
after time and money (usually a great deal of both) have
been expended and decisions are seconds away from being
made. It may be true that what the information manager
wants from the data or study may not be obtainable. But
the problem may be amenable to a modified approach
some logical answer to the question may be possible or.
perhaps, a different question can be asked. The statistician
has the training and the ability to make these judgments and
to define approaches that may not have the rigor of a clas-
sical design but nonetheless will provide a better answer
than attempting to impose inappropriate or inadequate anal-
yses on the data or data-collection activity.
Recently there have been some innovative approaches for
dealing with imperfect sets of .data. Paul. Rao, and Zelen
-(1983) and Hennemuth. Paul, andTaille (1985) introduced
the concept of "encountered'data" as a possible approach
to dealing with observational studies in which the data do
not fall into experimental or replicattble categories. The
approach is particularly useful for environmental problems.
More often than not we are faced with using available data
rather than planning for the collection of new data. Inno-
vative thinking and approaches of this type to environmental
data collection and analyses should "be an integral pan of
the environmental statistician's arsenal of tools.
': 4. . IMPROVED TRAINING
Another area where improvement is needed is in the way
applied statisticians are trained, which contributes to the
probiemof- adjustment to the real world of applied statistics.
McCulloch. Boroto, Meeter. Polland, and Zahn (1985) pointed
out that many students have difficulty making the change
from academia to government or industry. Statisticians tend
to be trained as research scientists, most of whom are com-
fortable doing theoretical research or reviewing and com-
mentinf on the misappJication of theory. Applying the results
of theoretical research to real world problems is difficult.
however, and very little time, if any, U devoted to providing
statisticians experience and training in real-world applica-
tions. Some attempts have been made to provide graduate
students with training in. the "an" of consultation. Mc-
Culloch et al. (198S) discussed an expanded approach to
the education of statisticians being used at Florida State
University, where courses on statistical consultation are of-
fered as part of the curriculum. They found in their course-
evaluation interviews with graduate students that the con-
sulting courses had given them an opportunity to gain ex-
perience in actual consulting, to learn more about applied
Tht Amerinn StaUuicion. Navtmbtr 1987. Vol.41. No. 4 255
-------
.v«u!':j :iow to worx
non-
.iicians." Carter. Scheaffer. and Maries (1986) strongly
advocated Che role thai the consulting units in statistics de-
partments play in providing in excellent opportunity for
graduate students to receive on-the-job training in the prac-
tice of statistics. Unfortunately, the "practice" is often lim-
ited to consulting in the controlled environment of academia.
Both of these articles discussed consulting experience under
controlled situations. But there is nothing like the real ex-
perience of working in the government or an industrial set-
ting. Statisticians should have this type of exposure during
their academic training. Primary- and secondary-school
teachers are required to take at least one semester of an
internship during which they actually teach within a school
system. They must take this internship before they can be
certified to teach. If statistics departments could arrange for
statistical internships and traineeships with government or
industry, this would provide the student with pre-«mploy-
ment experience (i.e.. help them decide what they want to
do with the education they are getting) and. at the same
time, provide feedback to the statistics department on the
success of their mining courses.
5. FEDERAL-AGENCY INITIATIVES
Dr. Millard's article provides a timely reminder that sta-
tistical methods as well as statisticians have not yet been
totally accepted as essential components of environmental
decision making, but it is important to note that there have
been major improvements in statistical quality and statis-
tician involvement in environmental monitoring programs
in the last few yean (particularly at the EPA). Keep in mind
that the monitoring activities that Or. Millard reviews in his
article took place from three to five yean ago. Many lessons
have been learned from these experiences, and positive
changes have been initiated in a number of environmental
programs. The EPA has initiated a number of activities to
foster better quality assurance and analyses of their envi-
ronmental data. For example, in 1984 the EPA formally
initiated a new and more rigorous agency-wide quality as-
surance program (EPA 1984). Through a process called Data
Quality Objectives (DQO's) the EPA is working toward
institutionalizing quality-assurance requirements for ail pro-
grams. The DQO's will provide basis for developing qual-
ity-assurance protocols for their existing and plumed EPA
environmental-monitoring activities. Oversight responsibil-
ity for this program resides in the EPA Office of Research
and Development's Quality Assurance Management Staff
and the EPA Office of Policy, naming, and Evaluation's
(OPPE) Office of Standards and Regulation. Born of these
offices are staffed with highly qualified statisticians and
technical experts. In addition, the assistant administrator for
Me U«Vb has established a Statistical Policy Advisory Com-
mittee. which will provide policy guidance and advice to
the assistant administrator on statistical issues and problems.
This committee is composed of 25 senior-level statistical
managers and statisticians from EPA program offices, lab-
oratories, and regions. The committee represents a major
step forward for statistics at the EPA. and the impac^^its
activities will be reflected in continued improvemen^Bie
way EPA deals with its environmental-monitoring anoT4ta-
'collection activities.
6. CONCLUSION
In conclusion. Dr. Millard demonstrates (hat there defi-
nitely is room for improvement in the use of statistics and
statisticians in environmental monitoring; however, the rays
of hope are far from dim. Statisticians need to recognize
that environmental programs are functions of complex social
and political forces. Once we recognize the socio-political
pressures and how these constraints mold environmental
programs, then we can carve out a meaningful role for
statistics and statisticians that will provide the program man-
agers and the public with the tools to make better and more
informed decisions.
ADDITIONAL REFERENCES
Ammca* Trailf Momrfoet*rtrt. lot., v. Donovan 11981 >. 452 U.S. 490.
Carter. R. L.. Scheaffer. R. L. and Marks. R. G. (1986). "The Role of
Consulting Units in Statistics Departments." The Amtntan Statistician.
40. 260-264.
Green. H. P. (1986). "The Role of Courts in Environmental
Making." Journal of ike Watnington Academy of Science!.
223.
Henaemuta. R. C. Petti. G. P.. and Taille. C. (1983). "Can We Design
Our Enconam?" unpublished paper presented to the International Council
for the Eiptoration of the Sea. London.
House. P. W.. and Shull. R. (1983). Managing Metearrii am Demand.
Boston: Abt Books, pp. 33-47.
Marquardl. D. W. (1987). "The Importance of Statisticians." Jommai of
the America* Stanttirot Association. 82. 1-7.
McCulIoch. C. E.. Boroto. D. R.. Meeter. D.. Polland. R.. and Zahn.
D. A. (1983). "An Expanded Approach to Educating Statistical Con-
sultants." TV Amrwm Sunaieim. 39. 139-167.
National Academy of Sciences (1977). Entiro*nnmal Momioritiiy. 4. I-
III.
Paiil. G. P.. Rao. C. R.. and Zefcn. M. (1983). "Proposed Advanced
Research Conference on Weighted Distributions and Related Weighs*
Methods for Statistical Analysis and Interpretation of Encountered Data.
Observational Studies and Resulting Inferences." technical report. Penn-
sylvania State University. Center for Statistical Ecology and Environ-
mental Statistics.
U.S. EiuiiomntmaJ Protection Agency (1984). "QA Rcsponsnilibes for
Each Major Agency Organuation." EPA Order 3360.1. EPA. Wash-
ington. D.C.
256 77t» America* Statistician. November 1997. Vol. 41. No. 4
-------
BERTRAM PRICE1
1. INTRODUCTION
Dr. Millard identifies a serious problem that affects the
ability of the U.S. Environmental Protection Agency (EPA)
to carry out effectively the mandates of major environmental
legislation. The problem is the failure of some (possibly
many) environmental-monitoring programs to produce data
containing information sufficient to support the types of
analyses needed for making effective regulatory decisions.
Millard focuses on statistical issues in environmental mon-
itoring, demonstrating with three examples that statistical
principles either are not used where needed or. if used, are
often applied incorrectly. One of his major points, that sta-
tistical procedures based on data obtained from monitoring
networks are used almost always without consideration of
the power of the monitoring design network, is a particularly
important criticism. There is little question that Type II
statistical error rates are too often ignored when environ*
mental-monitoring programs are being developed. Since most
environmental-monitoring data are used ultimately in a sta-
tistical test (e.g.. either testing compliance with a regulatory
standard or determining if a change in pollution level has
occurred) designs for data collection should be focused on
the control of statistical decision error rates. Designing data-
collection plans for monitoring programs that control the
likelihood of Type II errors requires formal monitoring ob-
jectives to be stated. Environmental managers responsible
for developing regulatory programs (and possibly even some
statisticians) may be surprised to find that more attention
-devoted to controlling statistical error rates can lead to a
unified approach for planning and implementing environ-
mental-monitoring programs.
2. NATIONAL ACADEMY OF
SCIENCES REPORT
This themedeveloping formal objectives for environ-
mental-monitoring programs and designing data-collection
plans to control error rateswas addressed in some detail
in me report prepared by the Study Group on Environmental
Monitoring and published by the National Academy of Sci-
ences (NAS. 1977). Chapter 3 of the report. ~EPA's Man-
agement of Scientific Dan." and Appendix C, "Incorporating
Uncertainty Into Environmental Guidelines," prepared by
J. Stuart Hunter, deal with problems in applications of sta-
tistics to environmental monitoring similar to those identi-
fied by Millard. (Millard's article would have more relevance
if be had related his observations to those reported by the
Study Group.) As discussed later in this comment, progress
since 1977 toward solving some of the problems identified
is apparent. Millard's examples, however, are a reminder
Bema Phct is President. Price AUOCUMS. Inc.. Suite 400.2100 M
Sow. Nmih»m. Washington. O.C. 20037. The author is member of
BB ASA Commute* on frr'"^ end the Environment.
that deficiencies in statistical practice applied to environ-
mental monitoring identified in 1977 have not been elimi-
nated. To place in perspective some oflhe criticisms. 1 have
briefly summarized findings of the Study Group and noted
a few important areas of progress.
The Study Group reported that "broad purposes must be
taken into account in the design of monitoring networks.
But designing a network m'at is cost-effective requires one
to define specific operational objectives and criteria in ad-
vance" (NAS 1977. p. 23). Establishing operational objec-
tives for monitoring may be easy in some situations and
much more difficult in others, depending on the "broad
purposes."
3. AMBIENT AND SOURCE MONITORING
The Study Group's classification of monitoring systems
into two categoriesambient monitoring systems and source
monitoring systemsreflects different "broad purposes."
Ambient systems are intended to determine how well en-
vironmental standards are being met and how environmental
quality is changing over time. Source monitoring systems
are used to develop inventories of pollutants discharged to
(he environment and to test compliance of individual or-
ganizations with regulatory standards. An example helps
clarify ..this classification system. Ambient water-quality
standards limiting chemical concentrations in surface waters
have been established under section 304
-------
rent regulations or control methods. In addition, they are
intended to assess, in the future, regulations that have r.ct
been implemented or. in some cases, have not been for-
mulated.
4. PROBLEMS AFFECTING EPA
MONITORING SYSTEMS
The Study Group indicated three reasons for the problems
affecting EPA monitoring systems: (a) "the apparent lack
of connection between the data EPA acquires and the de-
cisions the Agency makes." (b) quality assurance practice
that is "often inadequate to guarantee that the data produced
wifl be accurate and relevant to decision making." and (c) "the
scarcity of statistical talent within the Agency" (NAS 1977,
p. 63). The Study Group was more specific in its criticisms
of environmental sampling and chemical analysis, which
are components of a monitoring system. Regarding sam-
pling. it identified the need for "a central point of scientific
responsibility for sampling protocols; network designs; pro-
cessing and audit of sampling activities; and the preparation
of guides,- instructions and standards" (NAS 1977. p. 66).
Regarding chemical analysis, the Study Group expressed
concern that no estimates of bias and precision existed for
many of the EPA's reference measurement methods and that
ruggedness testing, which is conducted according to a for-
mal experimental design to investigate consequences that
small changes in analysis protocols have on the resulting
measurements, was not used often enough. Furthermore,
the Study Group called for interiaboratory comparison pro-
grams. noting the necessity of these programs to control the
quality of analytical data used in major regulatory initiatives
such as the tens of thousands of NPDES permits issued to
limit discharge of pollutants into the nation's- waters. The
Study Group also called for a certification program based
on performance for laboratories that, by performing mea-
surements, determine whether industries and state and local
governments are complying with standards established in
all EPA programs. Effective resolution of most, if not all.
of these problems integrally involves statistical concepts and
the need for statistical design.
5. FEDERAL-AGENCY INITIATIVES
Have EPA monitoring programs and the role played by
statistics improved since the report was written? Notwith-
standing Millard's three examples, progress in some areas
is apparent.
In 1982 the EPA adopted a broad set of quality assurance
(QA) guidelines requiring, among other things, the prepa-
ration of a QA Program Plan
issued for environmental data collection and for each spe-
cific project that involves environmental data collection.
These QA plans are intended to be compieuiiuiv*. covering
in operational detail project organization and responsibility,
sampling design including samplr jpeeificiBKJnJ. lathpfe-
collection protocols, statistical sampling specifications, and
analytical measurement protocols including quality control,
such as die analysis and interpretation of replicates, blanks.
and referee laboratory samples. The general QA objective
is to collect data with quality sufficient for the intended
application. Statistical concepts such as bias, precision, sample
size. Type I and Type II error rates, and minimum detectable
difference must be addressed explicitly if the QA objective
is to be realized.
In a related action, in 1984 a program was
requiring data quality objectives (OQO's) to be estai
for all environmental-monitoring and data-collection^ro-
grams undertaken by the EPA. As in the QA program, the
guiding principle for data quality is to produce data sufficient
for the intended application. DQO's address characteristics
of the raw data and summary statistics based on the raw
data including bias, precision, representativeness, and sta-
tistical decision error rates when the data are used for testing
compliance with a regulatory standard.
Regarding the standardization and testing of reference
measurement methods, in 1986 the EPA drafted guidance
for analytical-method development and validation. Devel-
opment and validation activities, which are to be completed
before an analytical-measurement method is released for use
in a regulatory monitoring program, include ruggedness test-
ing, conducting an interiaboratory study, and estimating the
method's bias and precision. Prior to 1986, the EPA took
action to improve control in the NPOES permit program, a
program specifically addressed by the Study Group, by con-
ducting a series of interiaboratory studies of the analytical
methods needed for the program. The EPA developed a
laboratory qualification program to screen laboratories in-
tending to analyze samples to determine NPOES pemu't
compliance, and it established a laboratory performance
.program to monitor ongoing performance of
that pass the qualification test. In addition to these
for the NPDES permit program, the EPA has implemented
laboratory qualification and ongoing performance testing in
the Contract Laboratory Program, which is a network of
laboratories under contract to the EPA to analyze samples
taken at hazardous-waste disposal sites that are being con-
sidered for cleanup under "Superfund" legislation.
Each of these three actions indicates improvements af-
fecting environmental-monitoring programs in areas criti-
cized by the Study Group. In addition, each of the actions
requires substantial statistical involvement. Nevertheless,
Millard's examples provide strong evidence that statistics
still is not always used effectively.
6. ADDITIONAL EPA CONSTRAINTS
.
In 1977 the Study Group listed a number of reasons why
statistics had a limited effect on monitoring programs. In-
cluded were the following:
(a) the miintar afjybff^eif^l acWcurir* the Afeacy to too mil;
(b) (here is no prafcniMltanft^iMHiie£Mi^«d AuubcaJ eipeniw
is sevtendJBiilUfhovi the Agency. (c) rwpomtbUity far eaMiiRmeet snd
fMatttMiaoB. jf food stttiflical pneticts is uncoortinaied: (d) eo 00*-
ticua with EPA has tufficim nak to influence Agency ~
delay them on te besis of misticsi deficiency; (c) no
tutisucml greup within the Agency OB consider the whole nnge <
tittical concerns idsted to eavnonmenul problem* in onkr to decide
astittkal eipcRitt could be brought to bear most effectively or to
ts iUiittical advisor » the Agency'* top msattemem. (NAS 1977. p. 93 >
2SS TV Amthcai Sututieia*. Novtmbtr 1987. Vol. 41. No. 4
-------
The Study Group als-. . :: * » broad recommendation that
"EPA develop. . . talent ut sufficient quantity and character
to meet its many statistical responsibilities and that the Agency
establish at least one senior statistician to advise EPA man-
agement on the entire range of Agency's statistical con-
cerns" (NAS 1977, p. 96).
7. IMPROVING COMMUNICATION
The Study Group looked inside the EPA for reasons and
solutions. Millard's assessment and recommendations are
primarily focused outside EPA. His assertion that "Many
introductory courses and textbooks in statistics do not em-
phasize the concepts of power and experimental design"
(p. 252) and his recommendation that "basic concepts of
experimental design should be included in elementary courses
and explained in concise, nonmathematical terms" (p. 222)
need to be considered seriously by the teaching community.
This is one way to help those developing regulatory policies
to recognize the importance of statistics in environmental
programs. Of equal or perhaps more importance, the con-
sulting st;ui«i6>an muit take the responsibility for-com-
municating, in terms understandable by regulators, o^I>^s.
and environmental scientists, the role of statistical design
principles in setting objectives and providing efficient so-
lutions to problems of environmental regulation.
Ten years have passed since the Study Group completed
its report. The EPA has recently developed new concepts
and implemented new approaches to improve the quality of
environmental-monitoring programs. Another formal re-
view of environmental monitoring is justified at this rime.
The review should not be limited to evaluating progress
relative to previous recommendations, but it should provide
guidance for the effective-application of statistics in the
continuing development of meaningful and efficient mon-
itoring systems for the future.
ADDITIONAL REFERENCE
National Academy of Sciences (1977). Environmental Monitoring: Ana-
lyiifal Studies for tht U3. Environmental Protection Agency (Vol. 4).
Wellington. DC: Author.
7ft* American Styiiiiciati. Novtmbrr 1987. Vol. 41. No. 4 259
-------
REYNOLDS ALUMINUM
Reynolds Me«a» Company . PO Bo* 27003 Rcnmono Virgins 23261 /
._. I
,. X
December 17, 1987
r>. .
EMERY EXPRESS
Mr. Janes Ludlam
New York State Department of Environmental Conservation
Bureau of Remedial Action
50 Wolf Road, RM 222
Albany, New York 12233-0001
RE: York Oil Site - Moira, New YorJc
Dear Mr. Ludlam:
Reynolds Metals Company has reviewed the Addendum
Feasibility Study Report on the York Oil Company site, dated
November, 1987, and has the following comments to offer:
1. The data contained in this report, along with that of the
RI/FS done in 1985, shows that a substantial number of sediment
and soil samples were analyzed for. PCBs. However, of the 266
samples analyzed, only 9 had concentrations above the 50 PPM
level that both the State of New Yorfc and the EPA use as a
criteria for hazard determination. The States ^arbitrary
selection of the 10 PPM cleanup policy goal at this site is thus
in question, what with other superfund sites in the State using
the more normal 50 PPM PCB as a remedial action level
(specifically the Hudson River sites, where PCB hotspots above 50
PPM were remediated by removal) .
New York State's 10 PPM PCB cleanup policy dramatically
increases the cost of the York Oil cleanup, by increasing the
amounts of soils alleged to be contaminated. This is evident in
the soil volumes estimated necessary for excavation, now at
30,000 cubic yards, up froa the 2,700 cubic yards in the original
1985 RZ/FS.
A more practical approach would be to just excavate the
hotspots above 50 PPM, as has been successfully done at
other superfund site cleanups, with the same overall risk
minimization to the public. This approach would act to keep
the costs of any remedial action down to a more reasonable
level, and would adequately protect the public from any
perceived danger at this site. ...
-------
Mr. James Ludlam
December 17, 1987
2. The Agency has selected as it's preferred alternative the
option labeled "5C", an approach utilizing soil excavation,
solidification, and groundwater treatment. Nowhere in the
addendum study does Reynolds find any cost breakdown of the
solidification process.
In fact, Reynolds has serious reservations about the
cost estimates that the EPA has developed for this
alternative. Specifically, page 6-74 gives the cost for
excavation and solidification as being $2,746,800. Using
the EPA's soil quantity estimate of 30,000 cubic yards, one
arrives at an average figure of $91.56 per cubic yard for
both excavation and treatment. On page 5-32 of the 1985
study, the Agency itself says that excavation costs range
from $50 - $60 per ton. Reynolds has had some contact with
solidification companies, and has never seen a quote as low
as the $31 per ton figure that is left over after the costs
for excavation are subtracted out.
Reynolds feels that the costs for option 5C are
understated, especially for the solidification end of the
process.
3. The PCBs that the EPA has identified in all the
studies of the York Oil site are PCS 1248, 1254, and 1260. As
pointed out in the recent study commissioned by the State of
California, these aroclors are not considered to be carcinogenic
to animals, or to humans.
Specifically:
Aroclor 1260 - "The human evidence for carcinogenicity of this
compound is inadequate."
Aroclor 1254 - "The human evidence for carcinogenicity is
negative."
Aroclor 1248 - "The human evidence for carcinogenicity is
negative."
Sines these compounds are the ones discovered at the
York Oil site, their lack of carcinogenicity can be used as a
basis for the 50 PPM cleanup goal being chosen.
4. Page 2-4 of this addendum states that the EPA conducted
a cleanup of the site in 1981, where soils from the wetlands and
sediments from the lagoons were solidified with cement and cement
products and left' onsits. Reynolds questions why these materials
have to be excavated again for the solidification treatment, if
they have already undergone the process.
-------
.--
Mr. James Ludlam
December 17, 1987
If you wish to have additional information on
these comments, feel free to call our Mr. S. D. Ryan at
(804) 281-3952.
Sincerely,
Lawrence C. Tropea, Jr. P.if.
Director, Environmental Control
Corporate Environmental Department
LCT/SDR
cc: Keith Teel - Covington & Burling
Robert Howe - U. S. EPA
-------
.UMINL'M COMPANY OF AMERICA'
ALCDA
December 17, 1987
VIA EXPRESS MAIL
Mr. James Ludlam, Project Engineer
New York State Department of Environmental
Conservation
Bureau of Remedial Action
SO Wolf Road - Room 222
Albany, New York 12233-0001
Re; York Oil Site. Moira> New York - Addendum FQaaibility Study
Dear Mr. Ludlam:
Attached are the comments of Aluminum Company of America to the
Addendum Feasibility Study prepared for the York Oil Site. A
copy this letter and the attachment is also being sent to
Mr. Robert Howe, EPA's Project Manager. By separate letter we
are also transmitting a copy of these comments to Ms. Elena T.
Kissel, EPA's Assistant Regional Counsel, and, in that letter,
have requested a meeting between DEC/EPA and Alcoa to discuss
the proposed remediation. We therefore request that the
comment period be?extended until at least such time as a
meeting can be held. If you have any questions concerning
these comments, please feel free to contact me at (412)
553-4259.
truly yours,
ilph W. Waechter
Attorney
Attachment
RWW/mit
117RWW
cc:
-------
ALUMINUM COMPANY OF AMERICA
ALCOA BUILDING
PITTSBURGH, PENNSYLVANiA 15219
LEGAL S£=ARTV.£NT
ALf'QA
December 17, 1987
VTA EXPRESS MAIL
Ms. Elena T. Kissel
Assistant Regional Counsel
Office of Regional Counsel
United States Environmental Protection Agency
Region II
26 Federal Plaza
New York, New York 10278
Re? York Oil Site. Moira. New York
Dear Ms. Kissel: x
Enclosed for your convenience is a copy of the comments of
Aluminum Company of America to the draft Addendum Feasibility
Study for the referenced site.
With regard to this site and the Addendum Feasibility Study and
the Agency's decision making process,, I would like to request
that representatives of EPA and NYSDEC meet with
representatives of Alcoa to discuss the Consultant's report and
recommendations and EPA's and DEC'S recommended remediation.
We are free to meet at any time prior to issuance of a ROD. He
anticipate that such a meeting will benefit the agencies in
their decision making and request that additional comments
offered by Alcoa'at such a meeting be included in the
administrative record on this matter. In that regard, we
request that the comment period be extended until such time is
we are able to meet. Please contact me at (412) 553-4259 so
that we can discuss mutually convenient dates.
jruly yours,
Ralph W. Waechter
Attorney
Attachment
RWW/mit
116RWW
-------
ALOMINDM COMPANY OF AMERICA'S
COMMENTS ON THE DRAFT ADDENDUM FEASIBILITY
STUDY FQRTHE YORK OIL SITE. MOTRA. NEW YORK
I. CTTNFRAL COMMENTS
1. EPA' 3 1984 Draft PCS Cleanup Policy is referenced
throughout the Report. This Policy was never
"7" adopted by EPA and is now obsolete. The Agency
N* % 2u Other than stating that it is a DEC "action level",
-"= ' \ ,», '^
^ Uj p* ^i^ the Report gives no basis for choosing a 10 ppm
$ ^ ^J & *
- jo r- clean-up level. The scientific and environmental
- ^ v^ . '--
c basis for choosing this level, or any level, should
-' r -'
' V- » r?> be discussed and documented.
V)
3. The groundwater ingestion risk assessment employed a
26.9 ppm PCB concentration. Risk assessment for
surface water accidental ingestion used a 120 ppm
concentration. Each of these values are orders of
magnitude above the- PCB solubility limit, as
»
referenced in the Report. It appeared, therefore,
-------
that if these are valid samples, they are actually
samples of "free oil". Indeed, Sample YO-DSW in
Appendix C is labeled as an "oil" sample at 120 ppm
not as surface water. Only one surface water sample
» « f
is stated in the data - the water phase of Sample
YO-DSW at a concentration of 0.014 ppm. Additional
data on surface and groundwater PCB levels is
necessary to perform a meaningful and scientifically
justifiable risk assessment.
4. Because zinc will not be regulated as a drinking
water standard by EPA, using zinc as a factor in any
of the risk assessments is inappropriate.
5. The most recent literature on PCB carcinogenicity
(e.g., that of Or. Stephen Roberts, a University of
Arkansas Toxicologist) indicates that only those
isomers of PCB with 60% or greater chlorine content
are suspected carcinogens. The majority of PCB
isomers found at the site are Aroclors 1242 and
1248. In light of the recent scientific findings,
only Aroclor 1260 should be used in cancer risk
assessments. Such an approach would follow that
employed in California's Proposition 65.
6. An assumed gastrointestinal adsorption of 100% was
: ' *
applied in assessing the risk of drinking PCB
-------
II.
contaminated groundwater. The assumed dermal
adsorption factor for organics was 50%. More
realistic and appropriate factors for
gastrointestinal and dermal adsorption are 15 to 30%
and 1% respectively. For further information on the
appropriate adsorption factors, Dr. Renata
Kimbrough, formerly of the CDC and now with U.S.
EPA-HQ, an expert in this area, should be consulted.
1. Executive Summary, Page 2 - Top of Page -
"Alternative No. 5A was ranked highest based on
non-cost criteria." Feasibility Studies require cost
evaluation for each alternative. What then is the
significance of this statement?
2. Executive Summary, Page 2 - Bottom - Until the
contamination pathway RI/FS is performed the Erdman,
Anthony ("EA") recommendation that "on- and off-site
remedial action be completed at the same time" is
suspect and not supportable.
3. Page 1-1 - The site is located near Moira, not in
the Town of Moira. The last paragraph suggests that
more than one report is to be referenced, but only
one is listed.
-------
4. Page 1-1 - Last Paragraph - The preparer of the RI
Report should be identified.
5. Page 2-1 - Does the Report include information from
groundwater wells installed earlier, or only the 57
new wells? If none of the earlier data was used,
why not?
6. Page 2-2 - The results of field screening using the
portable GC should be, but are not, discussed in the
Report.
7. Page 2-2 - First Paragraph - The "Site Exploration
and Conditions . . ." is incorrectly dated. That
Report is dated October, 1987.
8. Page 2-3 - Paragraph 2 indicates that lab data from
the 1985 Study are not included. However, it
appears that only selected results from the 1985
Study are mentioned in this Report. What was the
criteria employed in selecting the data to be
included?
9. Page 2-4 - Paragraph "5'suggests that cleanup
activities have continued since 1981. The systems
installed at that time have not been maintained and
.' * *
were essentially non-functional as early as 1985.
-------
10. Page 2-5 - Graphics would help this subsurface
discussion.
11. Page 2-6 - First Paragraph - How many is "many"?
12. Page 2-6 - Observations were made about oil and oil
staining. How are these terms defined? How much
oil is needed to be "oil" rather than oil stained?
13. Page 2-7 - This table contains clearly subjective
observations rather than measurements. It' should be
so stated and the terms "oil soaked", "oil sludge",
"oil and oil staining", and "oil pockets" should be
defined. Also, the term "possible oil stains" is
confusing.
14. Page 2-7 - Boring number YO-17A> depth and elevation
do not agree for 23.0-24.0 and 382.7 and 380.7.
15. Page 2-7 - and 2-8 - Under remarks - What does EA
mean by "suspected" and "possible"?
16. Page 2-9 - Paragraph 3 - States that the flow from
the mound is generally radial but states that the
flow is generally from the North. What does this
mean? Paragraph 4 states that there is a bedrock
grbundwater divide immediately north of the site.
-------
17. Page 2-11 - The basis for the determination that the
groundwater flow from the site is only 10-15 GPM is
not clear.
Which report is discussed in the first paragraph of
section 2.03. It contains 37 individual chemicals
found at the site. Two aroclors are mentioned, but
the appendix to this report mentions that three were
found.
18. Page 2-11 - Third Paragraph - PCB aroclors need to
be identified. (They are not listed on Table 2-3
either.)
19. Page 2-13 - The MCLG for PCBs is listed as 0, but no
MCL is listed.
20. Page 2-15 - Paragraph 2 - Mentions widespread
surface water contamination. However, the data set
suggests that only one or two surface water samples
were analyzed. How can widespread contamination be
determined with so few samples?
The termv "hotspots" Is used to describe areas of PCB
j term should be defined.
r
i
,/ \*
-------
The last paragraph discusses how insoluble materials
are moving through the groundwater system. It is
generally believed that insoluble materials move
extremely slowly, if at all, in a typical
groundwater 'system. Oil and PCBs will probably not
move with the water, and perhaps will not move at
all.
21. Page 2-15 - First Paragraph - "Apparently" should be
eliminated.
22. Page 2-15 - Second Paragraph - How long is "long"?
23. Page 2-16 - The statement is made that the water
soluble components tend to sink . . . What is the
basis for this .statement?
24. Page 3-1 - The term "hotspot" is used at the bottom
of the page.
25. Page 4-1 - At the bottom of the page a statement is
made that "It has been established that PCB levels
in the 0-10 ppm range, should be managed to
eliminate direct human exposure. . . . This is not
in accord with the SPA most recent and final cleanup
policy that calls for spill cleanup and to various
levels . . . some above 10 ppm.
-------
26. Page 4-1 - This table is the same as Table 2-3.
27. page 4-2 - Paragraph 1 states that it as anticipat
that the cleanup limits will approach the lower end
of the established limits. What is the basis for
this determination . . . particularly since the
indications are that surface water does not contain
contaminants and that the groundwater in the usable
zone is not impacted?
Borings YO-54 and YO-32 are mentioned as having PCS
levels above the action level. However the data in
Appendix C show that the levels found are not, above
the 10 ppm action level.
Paragraph 3 suggests that data is available to draw
contours of PCB contamination, yet the data are not
included in the report. It discusses areas where it
is suspected that PCB contamination may have
occurred. Such speculation has no place in this
type of report. Data are needed to support major
decisions, not speculation. The writer also refers
to Appendix A and B which are boring logs, not test
*
data.
28. Page 4-3 - Top of Page - This doesn't make sense.
- -
How and why will PCB concentrations in the 0-10
-------
ppm range be ej^*»«-fed during cleanup? What
established limits are referred to in the last
sentence?
29. Page 4-5 - The data presentation in Table 4-2 is
unsatisfactory. In addition to the high and low
values, it would be helpful and appropriate to
present the average of all samples taken. It would
also be helpful to have all the results tabulated in
an easily understood table. Evidently, at least 60
soil samples taken at the site showed zero
contamination of PCBs, and many showed levels that
were only slightly above the detection limits.
30. Page 4-6 - Paragraph 3 - It is stated that the 1985
study determined that the oil sludge layer cay be
serving as a reservoir for potential contaminants.
However, the data presented in this report do not
seem to confirm that. The soil samples indicate
that the PCB levels are below the 50 ppm level and
that the sludges pass the EP Toxicity testa (top of
pp 4-6). Shouldn't this be mentioned? In
Paragraph 3, reference is made to PCB levels in
surface water of 120 ppm. This is impossible since
PCB solubility in water is at least three orders of
magnitude less. This should be reported as PCB
levels in oil.
-------
31. Page 4-8 - Since only one water sample is involved
in this table, why is a frequency number included in
Table 403.
32. Page 4-9 - A "plume" of phenolic contamination is
discussed in Paragraph 4. However, the number of
samples available seem to be inadequate to define a
"plume". Table 4-4 suggests that phenol was
detected in only one of nine groundwater samples
analyzed.
*
33. Page 4-13 - In Paragraph 1, the "worst case"
scenario approach is described. However, it seems
inappropriate to use a concentration (31.5 mg/Jcg)
r>
found 20 feet below the grtfnd surface as the "worst
case" surface soil concentration for ingestion risk
analysis. Likewise, it seems inappropriate to use
an oil sample (120 ppm) as the basis for a "worst
\
case" water ingestion or water contact risk
- t .
analysis. Realistic samples should be used in these
scenarios.
34. Page 4-15 - The last sentence of the paragraph
before Section 4.03.022 discusses PCBs as a class.
' Recent statements by EPA officials indicate that the
Agency does not fconsider PCBs as a class, but
rather as a series of compounds with widely varying
-------
degrees of toxicity. PCBs with more than 60 percent
chlorine are considered to be toxic (i.e., Aroclor
1260) while PCBs with less chlorine (i.e., Aroclors
1248, 1252) are of less concern.
35. Page 4-16 - In Paragraph 2, the indicator scores
should be based on realistic samples, not worst case
numbers such as a sample found 20 feet below the
surface. r
36. Page 4-18 - The 31.5 mg/Jcg concentration for sils is
from a sample 20 feet below the surface. It should
not be used for soil ingestion risk analysis.
37. Page 4-20 - The 26.9 mg/1 PCB concentration for
groundwater is not realistic for two liters of
consumption per day. The sample is obviously
contaminated with oil and would not be used as a
constant drinking water source.
38. Page 4-22 - PCBs are listed as a probable human
carcinogen. There is not data to suggest this. It
should be listed as a suspected human carcinogen or
as an animal carcinogen.
39. Page 4-25 - This is the same as Table 2-4.
-------
40. Page 4-28 - In Paragraph 3, it is noted that PCB3
were found in 43 percent of the soil samples
analyzed. It should also be noted that^the average
level in all samples- with PCBs detected is less thar
3 mg/Jcg.
In Paragraph 4, the concentration found in one
sample, 20 feet below the surface is used for
surface soil ingestion risk analysis. This is
inappropriate.
41. Page 4-30 - In Paragraph 5, a groundwater
concentration of 26.9 mg/1 is used as the basis for
ingestion risk analysis. This sample was obviously
contaminated with oil -since the level is at least
100 times the solubility limit for'PCBs. It is an
inappropriate sample to use for risk analysis.
42. Page 4-31 - At the bottom of the page, a PCB level
of 120 mg/1 is selected for worst case surface water
risk analysis. However, the data in Appendix C
clearly note that sample as an oil sample, with the
associated water sample containing 0.014 mg/1 PCBs.
It is inappropriate.to use the 120 mg/1
concentration for surface water risk analysis.
-------
43. Page 4-33 - In Paragraph 2, the possible exposure of
people who work in the town garage is mentioned.
However, since the town garage is a maintenance
facility for motorized equipment, oils with high
lead, other metals and phenol content will be
present in and around the facility. Exposure to
workers and visitors to the garage will probably be
much higher from the activities in that facility
than from this site.
44. Page 4-36 - In Paragraph 4, mention is made of risks
associated with ingestion and dermal absorption of
contaminants in on-site soils. Yet, in Paragraph 2
on Page 4-35, the writer concludes that the presence
of contamination in the subsurface soils prevents
the release of site contaminants as particulate
matter, effectively eliminating potential inhalation
exposure and minimizing the potential for direct
contact with or accidental ingestion of soil. In
spite of this, soils found at a depth of 20 feet
were used in the ingestion risk analysis.
45. Page 4-37 - The concept of using water on the site
as undiluted drinking water is introduced in
Paragraph 2. This -seems inappropriate since there
are no surface intakes for many miles from the site.
-------
46. Page 4-38 - Section 4.03.034 begins by stating that
the maximum contaminant concentrations detected in
samples were used. Only appropriate samples should
be used.
47. Page 4-40 - In Paragraph 1, it is stated that
observed maximum concentrations for various
compounds exceed the ARAR by at least 600 times. It
should be noted that the PCB concentrations used are
in oil, while the ARARs cited are for water. If the
water sample concentrations were used the ratios for
PCBs would be several orders of magnitude less.
48. Page 4-41 - Zinc should be removed from Table 4-11
since it has been dropped as a compound in drinking
water standards.
49. Page 4-42 - The PCB concentration in the surface
water sample was reported as 0.014 mg/1 not 120 mg/1
as shown. The zinc number is reported as 0.088 mg/1
not the 0.88 mg/1 concentration shown.
50. Page 4-43 - Same comments as Page 4-42.
' *
51. Page 4-48 - Or. Renate Kinbrough of the Center for
Disease Control suggests that gastrointestinal
-------
absorption should be 15-30 percent rather than the
100 percent rate assumed.
52. Page 4-49 - Same comments are Page 4-48.
53. Page 4-51 - The carcinogenic potency factor for PCBs
is listed as 4.34. This seems high for a suspected
carcinogen. What is the basis for that factor?
54. Page 4-59 - In Paragraph 1, the estimated risk range
of 10~4 to 10~7 is noted as the range that, may be
considered to be a public health concern. Normally,
a risk smaller.than 10~6 is considered acceptable.
What is the basis for the selection of the range
10'4 to 10"7?
s.
55. Page 4-60 - Paragraph 2Lifetime cancer risk
estimates for arsenic, PCBs and trichloroethylene
are within the target risk range only if the worst
case assumptions include the errors pointed out
earlier concerning assumed absorption factors, the
use of oil samples rather than water samples and the
use of sample data collected at a depth of 20 feet
to represent surface soil conditions.
-------
56. Page 5 - 47 - Third Paragraph - EA has assumed that
there will be clean up associated with the pathway
3tudy, duch an assumption is inappropriate at this
time.
57. Page 5-49 - Alternative 4 calls for a total of 52
groundwater extraction wells. However, the report
concludes that the total groundwater flow from the
site is 14 gpm. The number of wells specified seems
very large for this type of flow system.
58. Page 5-50 - Last Sentence - A recommendation of
thermal treatment, when treatability studies haven't
been conducted, is wholly inappropriate and
unsupportable.
59. Page 5-52 - Last Sentence - Same comment regarding
treatability studies.
60. Page 6-1 - The referenced FS document is outdated.
A June 1986 edition is available.
61. Page 6-60 - Paragraph 2 discusses the New York DEC
Law that regulates materials containing PCBs in
excess of 50 ppm. It should be noted that only some
surface sediments (probably from the oil spill which
.- *
occurred when the tanks were overturned at the site)
contain PCBs in excess of 50 ppm.
-------
62. Page 6-61 Paragraph 1 speaks of the EPA cleanup
policy for residential areas. It should be noted
that this policy refers to residential areas where
* * * .
people live nearby and have direct access, such as
homes with utility poles nearby. Sites with
controlled access, such as fenced areas, are
considered non-residential. The York Oil site would
be considered non-residential under the EPA policy.
63. ??ge 6-61 - Paragraph 2 states that PCB levels on
and adjacent to the site are above 50 ppm. The
report indicates that the maximum soil sample was
31.5 mg/kg, the maximum groundwater sample was
26.4 mg/1 and a surface water sample,.when analyzed
for oil and water separately showed. 120 ppm PCB in
the oil and 0.014 ppm in the water fraction. (The
surface water sample was taken from a location off
the site.) Therefore, it can be concluded that
soils and water on the site are not contaminated
above the action limits for PCBs, and that only the
sediments in the drainage areas resulting from an
oil spill sppear to be contaminated.
64. Page 6-62 - Reference is made in Paragraph 2 to the
EPA Draft PCB Cleanup Policy. That draft policy has
-------
been superseded by a formal cleanup policy. The
formal policy should be referenced here.
65. Page 6-72 - Cost estimates are difficult to
challenge without detailed study. However, the cost
estimate to excavate, incinerate and backfill
30fOOO cubic yards of moist soil for $9,165,000
seems very modest. If this represents 50,000 tons
of material, the estimated cost is $183 per ton.
Incineration costs alone usually range above $500
per ton for materials with low or no BTU value.
More details on the cost estimates should be
provided.
66. Page 7-8 - Second Paragraph - Under cost comparison,
»,
Alternative No. 5A is 2-1/2 times as expensive as
the least expensive alternative.
III. COMMENTS ON REPORT "SITE EXPLORATION AND CONDITIONS"
PREPARED BY GOLDBERG-ZOINO ASSOCIATES
1. Page 10 - If sediment samples were collected in
one foot intervals, were steps taken to prevent
cross contamination when the sample tube was
withdrawn, between intervals?
-------
2. Page 11 - Are the results of the field testa for PCS
recorded in the report?
3. Page 14 - The statement is made that moisture
content estimates were made in the soil and sediment
samples, and that the estimates were purposefully
made higher because high estimates would make the
PCS content results higher as well. Such procedures
are inappropriate and reduce the credibility of the
results.
4. Page 17 - Reference is made in Paragraph 2 to the
fact that during the current and previous field
activities by GZA, this system (a filter chamber
near the outlet of Lagoon 3 to collect oils exiting
%
Lagoon 3) was generally free of water with no oil
visible on the sorbent pad. .This statement suggests
that the contamination found in the drainage areas
is from an earlier oil spill and not from the
release of contaminants from the material currently
in the fixed lagoons.
5. Page 23 - The different between oil and oil-stained
soils is unclear. Thetse terms should be defined.
6. Page 31 - The method used to determine soil and
: ' *
sediment moisture should be stated in the last
-------
paragraph, since PCB levels are mentioned. If the
soil or sediment moisture contents are off by a
factor of two, the PCB levels will be off by that
same factor.
7. Page 32 - Paragraph 2 states that boring Y054 showed
PCB levels of 20 mg/kg. The table in Appendix C
shows ND and BOL as the levels measured by GZA. The
split spoon data also indicate ND levels of PCB in
this boring.
8. Page 32 - Paragraph 2 stated that boring Y032 showed
PCB levels above 10 mg/kg. the data in the first
table of Appendix C indicate that the surface sample
showed a level of 10 mg/kg and that all other levels
were less than 5 mg/kg. No samples greater than
10 mg/Jcg were reported. 'r
9. Page 32 -There is no basis for the statement at the
bottom of the page that "it is assumed that much of
the former Lagoon 3 contains PCBs greater than
10jrfmg/kg.N The evidence is to the contrary. The
referenced sample, boring Y015 (not available in
this report) evidently showed some PCB contamination
in the surface sample, but essentially none in the
sample below the cap layer.
-------
10. Page 33 - Paragraph 3 attempts to describe possible
analytical problems associated with the CLP test
procedures used for PCB determination. If
appropriate spiked were used and spike recoveries
were within acceptable limits, the test results
should be valid. If not, the results should be
invalidated, not speculated as being too low.
Paragraph 4, same comment as above.
11. Page 38 - The PCB levels reported in the monitoring
wells are obviously not possible in water.
Therefore, oil must have been present. Is there any
information on how the samples were collected (were
the wells bailed fully), how much oil was present in
the samples, and were the oil and water fractions
analyzed separately?
12. Page 39 - In Paragraph 1, the writer speculates
about the "PCB plume." However, he fails to state
the very likely possibility that since almost all
the detected PCB contamination is on or near the
surface, the PCBs at the site came from the oil
. tanks which were tipped over and emptied 'by unknown
persons after the site was abandoned by Mr. Pearce.
-------
Vlie evidence points toward a surface oil spill as
the source rather than PCB contamination in the
lagoons.
* » » .
13. Page 47 - Conclusion 5 states that contaminated
soils are predominantly in the south central portion
of the site covering former Lagoon 3. This
conclusion seems to be based on comments on Page 32,
Paragraph 1, which refer to PCB levels in boring
Y039 to a depth of six feet. However, the data in
Appendix C indicate that -Implicate tests of the deep
samples in Y039 showed 17.3 mg/kg by ERCO and
0.5 mg/kg by GZA. At best, the numbers should be
averaged, which is then below the action level.
Actually, the data should be disregarded.
* ;_
Certainly, it should not be the basis for a
conclusion.
14. Page 48 - Conclusion 7 speaks of PCBs found in Y052
from 7-12 feet. However, the data in Appendix C
indicate that the PCB levels in &052 in samples S-4
(6-8 feet), S-6 (10-12 feet) and S-9 (16-18 feet)
were all below 5 mg/kg. This should be clarified.
15. Page 50 - Conclusion 8 discusses a "plume" of PCB
contamination in the shallow groundwater. The
: *
implication is that the "plume" has a PCB
-------
concentration of 120 mg/1. However, only one oil
sample, collected in a drainage trench, was at that
level. How can that sample be used to define a
"plume?"-
16. Page 50 - Conclusion 8 discusses a phenol plume that
appears to sink as groundwater flows to the south
across the site. Insufficient information is
presented to support this conclusion. Only one
groundwater test of the nine reported contained any
phenol.
17. Page 52 - Paragraph 4 under limitations appears to
imply that the writer's confidence in the data is
limited. If this is the case, appropriate caveats
should be included in the report.
118RWW
-------
John W.Johnson
County Manager
OFFICE OF
FRANKLIN COUNTY MANAGER
Court House, Malone, New York
Phone: (518)483-6767 Ext. 26S
January 13, 1988
Mr. Daniel L. Steenherge
Senior Sanitary Engineer
NYS Dept. of Environmental Conservation
Route 86
Ray Brook, New York 12977
'Dear Mr. Steenherge:
REGIONAL ENGINE
f....
The Franklin County Legislature has had the opportunity to review both
Deparanent of Environmental Conservation and Environmental Protection Agency1 s
remedial plans for clean up of the York Oil Company Hazardous Waste Site
located in the Town of Moira.
i '
Resolution No. 5 passed by the County Legislature on 1/7/88 outlines their
recarmendation. Specifically the County prefers removal of all hazardous materials
from the site as the first option. DEC'S proposal for on-site thermal treatment is
our second preference; providing heavy metals are not present in the soil. The EPA
recommendation for solidification is the least preferred of the three options.
We are concerned that solidification does not reclaim the soil but simply
solidifies the hazardous material. The site still remains hazardous. The residents
in the area must still be neighbors to hazardous waste and the County will forever
hold title to this unattractive plot of land. What future liability will the site
pose to the County? Does EPA intend to relieve the County of all liability asso-
ciated with the hazardous material stored in a solid state on County owned property?
These 4uestions need to be answered before any option is chosen.
JOHN W. JOHNSON
FRANKLIN COUNTY MANAGER
JWJ:rtro
Enc.
-------
RESOLUTION NO. 5
Offered by Legislators Donaldson, Frenette, Jackson, McCann
Pounore. Jones A Smith
Relating to York Oil Site Decontamination Procedures
WHEREAS: The York 011 Site has been for years a major concern of
local, state and federal authorities because of the contaminated
material deposited there; and
WHEREAS: Several methods of detoxification have been proposed
including removal from the site, thermal treatment and solidification;
and "* - .
.WHEREAS: Removal from the site is no longer on the options list
and the Department of Environmental Conservation believes that thermal
treatment would be the best alternative as it would allow the site and
surrounding area to be environmentally safe whereas sol 1-d-; fixation
would render the site useless and 1t would in fact remain a hazardous
waste site exposing the County and its citizens to possible adverse
environmental and liability questions; and
WHEREAS: The liability problem of the York Oil Site has already
caused an insurance company to cancel the County policies; and
WHEREAS: The people of Franklin County and particularly those in
the Township of Moira deserve to have their community restored to an
environmentally safe place to live; NOW, THEREFORE, BE IT
RESOLVED: That the Franklin County Board of Legislators requests
that DEC and the EPA accept the alternative that involves thermal
treatment in the event removal 1s no longer a viable option; and, be
it '
-------
EXHIBIT III
-------
-1-
EXHIBIT III
Comment:
A series of questions dealt with the quality of the
area's groundwater and the water obtained from residential
wells, potential future impact on the wells and the need for
residential water supply in the area immediate to the York
Oil Site.
Response;
All sampling done to date on the residential wells
shows they are free of contamination due to the York Oil
Site. If the site were left as it is, there might be a
potential for future impact on residential wells. However,
the remedial program proposed for the site and the
monitoring of the remedial program's effects will eliminate
this potential. Based upon this, a residential water supply
system does not appear to be justified at the York Oil Site.
Comment;
The time from identification of the site until the
present is approaching ten years. " When can we anticipate
.that the final portions of this project will be completed.
~~- .''.'. i
Response;
We can now anticipate that we would begin the design
phase of this project during the first or second quarter of
1988 and be ready to commence construction in 1989.
Comment;
A significant portion of discussion at the public
meeting centered around the appropriate permanent remedy to
be implemented at the site.
Response;
There was not yet agreement between the USEPA and
NYSDEC at the time of the public meeting. The NYSDEC
through its consultant had recommended a thermal
destruction alternative whereas the USEPA had chosen a
solidification.process as the preferred alternative. The
consensus of those attending the public meeting appeared to
be that the solidification process was really not a
permanent technology.
-------
-2-
Comment;
111 " ^^^» %
There were several comments that the choosing of a ten
ppm cleanup level for PCB was not appropriate nor
established by regulation.
Response;
A national PCB cleanup policy of ten ppm for PCB in
residential areas was established on April 2, 1987 at (52
FED.REG.10688). USEPA has adopted the National PCB Cleanup
Policy.
Comment;
It was asked why the material at the site could not be
left in place since they had previously been solidified with
cement and cement products in 1981.
Response;
The findings of our study were that the .solidification
process undertaken in 1981 was a temporary remedial measure
which was not effective in permanently stabilizing or
immobilizing these materials.
Comment; -,
Several of the comments dealt with the fact that the
report used the highest found concentration of various
contaminants in its health risk assessment.
Response;
The health risk reassessment was done to assess the
worst case. It is EPA's policy to use the highest values
found.
Comment;
Several comments dealt with the fact that the report
did not include information which had previously been
included in the earlier Remedial Investigation/Feasibility
Study.
Response;
The current report is an Addendum Feasibility which is
a supplement to the original Remedial Investigation/
Feasibility Study (RI/FS).
-------
-3-
Comment;
One comment stated that the sample analysis did «ot
show contamination of the groundwater above the 10 ppm
action level. (Note: This comment is in reference to the
groundwater standard for PCS).
Response;
The regulatory action level for PCB in groundwater is
.0001 ppm (i.e. 6 NYCRR 703).
Comment;
Several of the comments dealt with the use of a limited
data base to draw conclusions.
Response;
No matter how many data points or samples are
collected, the data base will always be limited in some
aspects. The Department has determined there was a
sufficient data base to draw valid conclusions.
Comment;
T-
One comment states that the report introduced a concept
of not being able to use water on the site .as unpolluted
drinking water in the future. The cdmmenta-tor further
stated that it seems inappropriate since there are no
surface water intakes for many miles from the site.
Response;
The mere fact that there are no current uses of this
surface water for drinking at this time does not preclude
the possibility of that use in the future.
Comment;
A recommendation of thermal treatment, before
treatability studies have been conducted, is inappropriate
and unsupportable.
-------
-4-
Response: _
A recommendation for use of thermal treatment was based
on our consultant's literature search. The literature
search findings indicates that the thermal treatment option
was the most promising of the three identified permanent
technologies. A treatability study will be a necessary part
in the design phase of this project. Similarly, if either
one of the other two permanent options is selected, a
treatability study would also have to be completed during
the design phase.
Comment;
Reference was made to the statement in paragraph two oo
page 17 of the RI/FS that during current and previous field
activities by GZA, the existing collection system (filter
chamber near the outlet of lagoon 3 to collect oils exiting
the lagoon) was generally free of water with no oil visible
on the sorbent pads. This statement suggests that the
contamination found in the drainage areas is from an earlier
oil spill and not from the release of contaminants from the
material currently in the fixed lagoons.
Response;.
The contamination found in the drainage area around the
southern fence of the site is not from earlier oil spills.
The drainage ditch which runs along the southern boundary of
the fenced area at the railroad currently contains floating
oils which are seeping out of the ground and are
periodically removed by a contractor working for the
Department.
Comment;
EPA's 1984 Draft PCS Clean Up Policies are referenced
throughout the report. This policy was never adopted by EPA
and is now obsolete. The agency issued a final policy
establishing a National PCS Clean Up Policy on April 2,
1987 (52 Fed. Reg. 10688). Most recent final policy should
be referenced and applied. It should be noted that the
Agency's present clean up level for soils and low contact
restricted access areas is 25 ppm.
-------
-5-
Response;
The commenter was correct that the draft PCB clean up
policy was referenced throughout rather than an appropriate
referenced final rule. However, this specific issue is not
substantive since the USEPA, NYSDEC and our consultants were
cognizant of the final rule. Briefly, the final rule was
used in the Agency's decision making, however the report did
not incorporate the final rule. In response to the second
portion of the comment regarding, use of a 10 ppm versus a
25 ppm clean up level in the soils, the Agency's do not
agree with the commenters position that the York Oil site is
a restricted access area as described in the final rule.
The Agency's reasons for making this determination are that
the area to be remediated is not completely fenced and it
does not meet-the intent of a restricted zone (i.e., there
is no commercial or industrial activity currently at the
site which would provide continuous control over access to
the site).
Comment; .
Other than stating that it is a DEC action level, the
report gives no basis for choosing a 10 ppm clean up level.
The scientific and environmental basis for choosing this
level or any level should be discussed and documented.
Response;
,V
In response to a previous comment, it was established
that a PCB clean up policy level had been implemented by
EPA. Using a 10 miligram per kilogram concentration of PCB
to determine carcenogenic risk associated with leaving PCB's
at 10 ppm in soils, the risk analysis formulas produce the
following numbers:
The residual risk due to soil ingestion is an increase risk
of 4.3 in a 100,000 and the residual risk due to soil/skin
direct contact is an increase in risk of 8.25 in a
1,000,000.
Thereby the health risk assessment is supportive of a 10
milligram per kilogram action level.
Comment;
Because zinc is not regulated in the drinking water
standards by EP'A, using zinc as a factor in any of the risk
assessments is inappropriate.
-------
-6-
Response:
NYSDEC has established a groundwater standard for zinc
of 5 parts per million.
Comment;
An assumed gastrointestinal absorption of 100 percent
was applied in assessing the risk of drinking PCS
contaminated groundwater. The assumed dermal absorption
factor for inorganics was 50 percent. More realistic and
appropriate factors for gastrointestinal and dermal
absorption are 15 to 30 percent and 1 percent respectively.
Response;
At the time of writing the health assessment, USEPA was
contacted by our consultant. EPA's response to the
inquiries regarding absorption factors was that there were
no published absorption rates. Thereby, it was appropriate
for our consultant to use a conservative number for
assessing absorption.
Comment;
Does the.report include .information from groundwater
wells installed earlier, or only the 57 new wells? If none
of the earlier data was used, why not?
Response;
There appears to be some confusion on the part of the
commenter since this current report is an Addendum to the
original RI/FS which means that in any assessment all data
collected since the beginning of all studies was used.
Comment;
Page 2-3, Paragraph 2, indicates that lab data from the
1985 study are not included. However, it appears that only
selected results from the 1985 study are mentioned in this
report. What was the criteria employed in selecting the
data included?
Response;
All available data from the 1985 study was assessed.
-------
-7-
Comment;
Page 2-7-This table contains subjective observations
rather than measurements. The terras "oil soaked"/ "oil
sludge", and" oil and oil stained", and "oil pockets" should
be defined. Also, the term "possible oil stains" is
confusing.
Response:
The terms oil soaked or oil sludge were used to record
the visible observations as to the nature of the material
collected or sampled. Use of any one of the descriptors
using the word oil was an indicator to those persons doing
the assessment of contamination to make determinations to
what samples would be analyzed by the laboratory or by the
on-site portable GC. The presence of "oil" is not the basis
for determining the extent of the remedial action, rather
the results of laboratory analyses regarding the presence
and concentrations of PCBs.
Comment;
Page 2-7 and 2-8 under Remarks - What does Erdman
Anthony Associates mean by suspected and possible?
Response; r
The subjective qualifier terms of "suspected" and
"possible" were used by the field personnel to identify
samples that were not obviously visibly contaminated or
laden with oil. However, these samples contained some type
of staining which indicated the need to have them analyzed
to determine whether or not they were actually contaminated.
Comment; -
Two of the comments provided by Alcoa numbered 16 and
17 in their section number II labeled "Specific Comments"
dealt with the hydrogeology of the site. These questions
dealt with the proper interpretations of the hydrogeology.
For these specific comments you should refer to the
questions as shown in the exhibits.
Response.:
The Department assessed these questions and believes
the report is correct as written.
-------
-8-
Comment;
Page 2-11 - 3rd paragraph - PCS aroclors need to be
identified. (They are not listed on table 2-3 either).
Response;
From a regulatory standpoint only total PCS is
regulated thereby listing of specific aroclors is not
needed.
Comment;
Page 2-15 - paragraph 2 - mentions widespread surface
water contamination. However, the data sets suggested that
only one or two surface water samples were analyzed. How
can widespread contamination be determined with so few
samples?
The term "hot spots" is used to describe areas of PCS
contamination. This term should be defined.
The last paragraph discusses how insoluble materials
are moving through the groundwaters system. It is generally
believed that insoluble materials move extremely slowly, if
at all in a typical groundwater system. Oil and PCB's will
probably not move with the water or perhaps will not move at
all.
Response;
As discussed in response to a previous question the
data base is not limited to that as presented in this
Addendum. It is the entire data base as presented in both
the Addendum and Remedial Investigation/Feasibility Study
(RI/FS). The term "hot spot" was used- to describe areas of
PCS contamination in soils of greater than the 10 milligrams
per kilogram. In response to the last item of the question
the findings of our studies have shown that insoluable
materials are moving through the hydrogeological system.
Comment; - '
Page 2-15 - 1st paragraph - the term apparently should
be eliminated.
-------
-9-
Response; ^
.The qualifier "apparently" is appropriate in this
situation since it indicates that contamination was not
identified at the analytical method detection limits.
Comment;
Page 2-16 - the statement is made that the water
soluble components tend to sink ... what is the basis for
this statement?
Response:
A rereading of the entire first paragraph on page 2-16
indicates that the contaminant being discussed is total-
phenblics which forms a density plume of the type that
tends to sink. Further the studies and the analytical
chemistry showed this to be a valid assumption.
Comment:
Page 4-1 - At the bottom of the page a statement is
made that "it has been established that PCS levels in the
0-10 ppm range, should be managed to eliminate human
exposure ... this,is not in accord with the. EPA's most
recent and final cleanup policy that calls for spill cleanup
to various levels ... some above 10 ppm.
Response;
The use of the term "management" in this situation is
similar to the EPA cleanup policy which specifies that
material is to tie covered to prevent direct human contact.
Comment;
Page 2-2 - the results of the field screening using the
portable gas chromatagraph (GO should be, but are not,
discussed in the report.
Response;
The results of the field screening is presented in the
report (See Addendum Feasibility Study Report, Appendix Vol.
1, Appendix C):'
-------
-10-
Comment;
«^_««»l>MB^_» ^
Page 4-2 - Paragraph one states that it is anticipated
that the cleanup limits will approach the lower end of the
established limits, what is the basis for this
determination ... particularly since the indications are
that surface water does not contain contaminants and that
the groundwater in a usable zone is not impacted?
Response;
The statement that it is anticipated that the cleanup
limits will approach the lower end of the established limits
since that area of contamination is quite well defined.
Contaminated material will either be detoxified or
stabilized if it is in excess of the 10 ppm action level.
Less contaminated materials on the remainder of the site
will end up being under the final cover placed at the site.
Comment;
Boring YO-54 and YO-32 are mentioned, as having PCS
levels above the action level. However, the data in
Appendix C shows that the levels found are not above the 10
ppm action level.
Response;
A rereading of Appendix C indicates that levels of PCB
greater than 10 ppm were not found by the portable GC.
However, confirming work at the contract laboratory
identified PCB's in excess of the 10 ppm action level.
Comment;
Paragraph 3 suggested, that data is available to draw
contours of the PCB contamination, yet the data are not
included in the report.
Response;
Taking into consideration all data presented in the
Addendum and the original Remedial Investigation/Feasibility
Study there is sufficient data available to develop PCB
concentration isopleths.
-------
-11-
Comment; ^
Page 4-5 - the data presentation in Table 4-2 is
unsatisfactory. In addition to the high and low values,
it would be helpful and appropriate to present the average
of all samples taken. It would also be helpful to have all
the results tabulated in an easily understood table.
Evidentially, at least 60 soil samples taken at the site
showed zero contamination of PCB's, and many showed levels
that were only slightly above the detection limits.
Response;
To present the average of all samples taken is
inappropriate since the methodology used in the field to
assess a site entails taking a number of samples to
determine a area of 0 contamination. Averaging all the
results of these essentially "clean" samples would give an
erroneous low number.
Comment; .
Page 4-6 - paragraph 3 - it is stated that the 1985
study determined that the oil sludge layer may be serving as
a reservoir for potential contaminants. However, the data
presented in this report do not seem to confirm that.
Response;
Based on the findings of this study it appears
that the oil sludge layer serves as a reservoir for the
phenolics plume which is exiting the site. We feel this
interpretation is correct.
Comment;
Page 4-9A "Plume11 of phenolic contamination is
discussed in paragraph 4. However, the number of samples
available seem to be inadequate to define a plume. Table
4-4 suggests that phenol was detected only one of nine
groundwater samples analyzed.
Response;
The 1985 RI/FS includes the additional data on the
phenolic compounds in groundwater.
-------
-12-
Comment; _
Page 4-22 - PCB's are listed as a probable human
carcinogen. There are no data to suggest this. It should
be listed as a suspected human carcinogen or as an animal
carcinogen.
Response;
The term "probable" was taken from USEPA's guidance
document on writing health risk assessments entitled the
Superfund Public Health Evaluation Manual of October 8, 1986
at Table 4-7 on Page 4-23. NYSDEC and USEPA believe this
.term is appropriate.
Comment:
Page 4-40 - in paragraph one, it is stated that
observed maximum concentrations for various compounds exceed
the Applicable or Relevant and Appropriate Standard (ARAR),
by at least 600 times. It should be noted that the PCS
concentration used are in oil, while the ARAR's sited are
for water. If the water sample concentrations,were.used the
ratios for PCB's would be several orders of magnitude less.
- *
Response: '..-.- ....
It is possible" for a water supply to provide,water, that
would contain oil which would not be noted visibly and
thereby contaminants could be ingested at the concentration
indicated in the samples used in table 4-11.
Comment; ~~a.> .--* - -..;.;*:.;*;-: :. .,-;..>.,..
-'" ' -' - . - -; .? ;*:-...: ;.. ,.< . ,.. 1
Page 5-47-- 3rd-paragraph - EA has-assumed that there "".
will be cleanup associated with the pathway -study« such an
assumption is inappropriate at this *ime.: .-. *- ,.,.,,. , .
Response: ' -'-"* - '-- -*- - .- ........
EA suggestion that on-site cleanup be performed in
conjunction with a pathway cleanup activities is in the form
of a recommendation to the State. Such, a recommendation is
appropriate based on*1 their' findings-. Briefly the purpose of
EA's work was to-assess the situation, draw conclusions, and
make recommendations. ^- .
-------
-13-
Comment ;
Page 50 - Conclusion eight discusses the "plume" "of PCS
contamination in the shallow groundwater. The implication
is that the plume has a PCB concentration of 120 mg/1.
However, only one oil sample, collected in a drainage
trench, was at that level. How can that sample be used to
define a plume?
Response;
Oil was found in various monitoring wells, also a
visual observation of the oil seeping out of the soil into
the drainage trench along the railroad grade indicates that
a plume exists.
Comment:
The contamination present on the site is due solely to
spilled material caused by the overturning of tanks at the
site.
Response:
The findings of the RI/FS study do not support this
contention. -
Comment ; . * i
Two commentators suggested that the preferred
option should be grading the site and putting on cover
material and reseeding along with disposal of the oils
stored at the site in the tanks.
Response ;
We believe such a remedy is not complete, for example,
it would not address subsurface contamination and would not
recover PCB contaminated oils, etc. It would also not be a
preferred remedy as required required by CERCLA Section
Comment ; .
Two verbal comments were received concerning the
appropriate remedy, one at the public meeting and another
through a telephone conversation with the Department. Both
comments were in favor of using a thermal treatment option
rather than a solidification option. They felt that the
solidification option was not a permanent remedy. Further
even if solidification was technically permanent, those
commenting had reservations regarding the impact on their
property values .since the site may continue to be perceived
as being hazardous.
-------
-14-
Response;
USEPA believes that soils 30!idification is *n available
and reliable technology for the treatment of waste types identi-
fied at the site. Solidification satisfies CERCLA Section 121
requirements for utilizinq a remedy that oernanently immobilizes
and reduces the toxicity an.1 nobility of the waste *t the site.
Solidification of the soils provides th
-------
-15-
Comment;
Page 14 of GZA's Report - The statement is made that
moisture content estimates were made in the soil and
sediment samples, and the estimates were purposefully made
higher because high estimates would make the PCB content
results higher as well. Such procedures are inappropriate
and reduce the credibility of the results.
Response;
NYSDEC believes that the possible difference between
the estimated and actual moisture content would only
introduce a possible variation of approximately ten percent
between the actual and reported concentrations.
Additionally several duplicate samples were analyzed by the
contrac: laboratory. Goldberg Ziono Associates values for
the GC work performed in the field were typically lower
than the contract laboratory results reported for the
duplicate samples. NYSDEC believes that even if the results
reported by Goldberg Ziono Associates were slightly high due
to an estimated moisture content, this variation did not irr
any manner influence the decision-making process. Comment:
Comment;
^^^^^^^^^^^MM* ,_
Page 32, Paragraph 2, states that boring YO54 showed
PCB levels of 20 mg/kg. The table in Appendix C shows NO
(i.e., Non-Oetect) and BDL (i.e., Below Method Detection
Limit) as the levels measured by GZA. The split spoon data
also indicate ND levels of PCB in this boring.
Response;
The data referenced in the comment was produced by a
portable GC in the field. The recorded data of 20 mg/kg was
as reported loy the contract laboratory.
Comment:
Page 32 - There is no basis for the statement at the
bottom of the page that "it is assumed that much of the
former Lagoon 3 contains PCBs greater than 10 mg/kg."
Response;
If all data.available are examined, it will be found
that the statement regarding Lagoon No. 3 as prepared by GZA
is a valid assumption.
-------
-16-
Comment;
Page 33 - Paragraph 3, attempts to describe possible
analytical problems associated with the CLP (i.e., Contract
Laboratory Procedure) test procedures used for the PCS
determination. If appropriate spikes were used and
appropriate spike recoveries were within acceptable limits,
the test results should be valid. If the test results were
not valid, the data should not be relied upon.
Response:
The test results were accepted as being valid. It
should be noted that the actual quantity of material
to be cleaned up was based upon the analytical data showing
PCBs present at greater than the 10 ppm action level.
Comment:
Recommended remedial alternatives, far exceed, in scope
and cost, those necessary to protect human health and the
environment and are inconsistent with National Contingency
Plan.
f.
Response;
The recommended alternatives are in accordance with the
requirements with the Comprehensive Environmental Response
Compensation and Liability Act (CERCLA). Specifically,
Section 121(d) of CERCLA requires the following: "Remedial
actions in which treatment which permanently and
significantly reduces the volume, toxicity or mobility of
the hazardous substances, pollutants, and contaminates is a
principal element, are to be preferred over remedial actions
not involving such treatment. The off-site transport and
disposal of hazardous substances or contaminated materials
without such treatment, should be the least favored
alternative remedial action where practicable.treatment
technologies are available." It is possible that there are
inconsistencies with the National Contingency Plan since
revision of the plan as required under CERCLA is yet to be
completed. Thereby, both the National Contingency Plan and
CERCLA as amended must be referenced, with the CERCLA
provisions taking preference,' in order to make the proper
interpretations. It is the Agency's position that our
interpretations are in accordance with CERCLA.
-------
-17-
Comment;
Recommended remedial alternatives employing site
excavation present the highest exposure risk to workers ar.d
the community.
Response:
This comment is only correct in the context that the
remedial alternatives involving excavation present the
highest exposure risfc during construction. Engineering
controls will be employed during construction to minimize
risk to both workers and the community. The engineering
controls implemented would be such that the actual exposure
to workers and the community would not present an actual
health risk.
Comment t
Information currently available indicates that
significant error is contained iff the cost estimates for
Option 5A, resulting in a substantial understatment of the
true cost of this alternative. The commehtor also appended
information regarding the cost of known incinerators (e.g.
the operating cost for one mobile incinerator was in the
range of $300 to $1,000 per ton and for the'other it was in
the range of $300 to $400 per ton of contaminated soil).
Response:
The estimated cost for Option 5A, Excavation and
Thermal Treatment, as presented in the Addendum PS is
approximately $15 million. The amount of material estimated
for thermal destruction is approximately 50,000 tons. These
dollar amounts back figure to an estimated cost of
approximately $300 per ton for the thermal treatment option.
It should be noted that the $300 per ton cost figure Is
within the range as presented by the commentor.
Comment; .
The Citizen Participation Specialist for HYSDBC, Region
5, informed those present at the public meeting that if they
felt they needed additional time to submit comments (i.e.,
the end of official comment period was two days after the
meeting) that they should make such requests in writing to
the Department.
Responset
The general public did not request an extension to the public
comment period. In response to a request from the Potentially
Responsible Parties, however, the comment period was extended fret
December 18, 1987 to January 15, 1988.
-------
-18-
Commer.t:
Alcoa's letter of December 29, 1987 addressed to Elena
Kissel, USEPA, dealt with the Addendum Feasibility Study
("Addendum FS") for the York Oil site in Moira, New York.
The purpose of the letter was; 1) to clarify AlCOA's
comments on the Addendum FS that were presented to USEPA and
the New York State Department of Environmental Conservation
("NYSDEC") at a meeting held in Albany on December 17, 1987,
and 2) to voice ALCOA'S concern it was not advised that
USEPA and the NYSDEC were considering thermal treatment as
an alternative remedial action for the York Oil site. ALCOA
specifically mentioned a November 25, 1987. letter received
from Elena Kissel which stated that "the preferred remedial
action for the York Oil site at this time includes site
excavation, on-site solidification of soils, thermal
destruction of tank and other oils ..." ALCOA also requested
a 30 day extension in the public comment period..
Response:
ALCOA's reliance on the letter dated November 25, 1987 as
precluding other'remedies is not supportable. The Addendum
FS clearly outlines the thermal "treatment option. The
November-25 letter in no way indicated that USEPA had made a
final selection of a remedy. The selection of a remedy is
not finalized until a Record of Decision ("ROD") is signed
by the USEPA's Regional Administrator.
-1 » = ' " ..""' . :..-..-
ALCOA was given an extension in the comment period. On
December 31, 1987, Elena Kissel telephoned, the- attorneys'
for ALCOAy Reynolds and "Kenneth Pierce, to inform them that
the public comment period would be extended to allow
comments to be submitted to USEPA on or before January 15,
1988. . ".!=.-
Comment:-"- - -'---
Reynolds-requested a 30 day extension in the public comment
periods for the-York'Oil Addendum FS. Reynolds request was
based on its-belief that USEPA'«."apparent shift" from the
position outlined in a November 25, 1987 letter from USEPA
to Reynolds, effectively denied Reynolds an opportunity to
comment on the remedy. This "apparent shift" wa* from
USEPA's selection of solidification .to, .thermal treatment as
a remedy for the York Oil site. ....
-------
-19-
Resoonse;
Reynolds claim that it was denied an effective opportunity
to comment on the Addendum PS because USEPA shifted its
remedy from solidification to thermal treatment at the York
Oil site is not supportable. The Addendum FS clearly
outlines the thermal treatment option. The November 25,
1987 letter in no way indicated that USEPA had made a final
selection of a remedy for the York Oil site. The selection
of a remedy is not finalized until a Record of Decision
.("ROD") is signed by USEPA's Regional Administrator.
Reynolds request for a thirty day extension in the public
comment period was denied. Reynolds attorney had been
contacted on December 31, 1987 and had been given an
extension until January 15, 1988.
Comment;
Franklin County is concerned that if USEPA selects
solidification as the remedy for the York Oil site, that the
county "will forever hold title to this unattractive plot of.
land." The Franklin County legislature passed a resolution
(No. 5) on January 7, 1988 specifically stating that the
county's order of preference for remedial alternatives at
the York Oil'site is first, the removal of all hazardous
materials from the site, second, thermal treatment and last,
solidification.
The county also inquired as to what future liability the
site would pose to Franklin County and whether or not the
USEPA intends -to relieve the county of all liability
associated with hazardous material stored in a solid state
on county owned property.
Response;
The selection of a remedy for the York Oil site will be made
by the NYSDEC and the USEPA on the basis of criteria
outlined in the Comprehensive Environmental Response,
Compensation and Liability Act ("CERCLA"). Section 121 of
CERCLA provides that the remedial action selected must be
protective of human health and the environment, must be cost
effective and must utilize permanent solutions and
alternative treatment technologies or resource recovery
technologies to the maximum extent practicable.
-------
-20-
The potential of liability to Franklin County in the current
action or a future action under CERCLA, is governed under CERCLA
Section 107(b), which outlines defenses to liability, and Section
101(20)(A)(iii) which excludes from the definition of owner or
operator, "any facility, title or control" of which was conveyed
due to bankruptcy, foreclosure, tax delinquency, abandonment,, or
similar means to a unit of state or local government ..".
The potential liability of Franklin County once the""remedial
project is complete and the site is removed frc:r the National
Priorities List is not addressed in CEFCLA. UGSPA has deemed
the solidification remedy to be a permaoant treatment option,
however, since the solidified soil will remain on-site, the
"Review" provisions specified in Section 121(c) requiring review
of remedial actions no less of ten "than evecy five years, would
be applicable.
-------
EXHIBIT FV
-------
OlWJ A.«WOO. M 0
STATE OF NEW YORK
DEPARTMENT OF HEALTH
Corning Tower The Governor Nelson A. Rockefeller Empire State Plaza Albany, New York 12237
December 30, 1987
Mr. James Ludlam
NYS Department of Environmental Conservation
50 Wolf Road
Albany, New York 12233-0001
Dear Mr. Ludlam:
Please find attached our responses to specific and general comments
raised by both Alcoa'and Reynold's Aluminum regarding the Addendum
Feasibility Study for the York Oil Site, (T) Moira, Franklin County.
.-...-->..; ..» .-:._. ..-,,.- -
Please forward these to Dan Steenberge for inclusion in the
Responsiveness Summary."" ^ -:....- - . :
Ifryou have any further"1 questions ;-pl«ase call me
Sincerely,'
' <-*
f 458-6310.
Gary A. Li twin
Program- Research Specialist III
Bureau of Environmental Exposure
Investigation^
JAN 1 3 1980
D«pt of Environmwtal Conservation
REGIONAL ENGINEER REGION 5
RAY BROOK, NEW YORK 12977
-------
Responses to Aloca's Comments
#3 The levels of PCB's in the water used for a risk assessment are
orders of magnitude above the PCB solubility limit. However, the
water is carrying free phase oil, and the sediments in the wetland
contain up to 210 ppm. This shows that the water is-carrying PCB'
at high levels regardless if it is dissolved or not. The exposure
levels used are appropriate.
#25 The low level (0-10 ppm) PCB's will be managed to eliminate
exposure. This does not mean that they will receive the same
treatment as the-higher levels of PCB. However, it is not
reasonable to leave these low level materials exposed. Some sort of
management is required, whether it is removal, covering them, or
some other method.
#33 (1st Part) The worst case soil sample from 20 feet below grade
contained 3T.5 mg/kg PCB. The worst case surface soil on-site
contained PCB's at 17 mg/kg. The deep and surface levels are of the
same order of magnitude. Since, the levels are close, it is not
unreasonable to use the worst case soil sample found-regardless of
where it is found. Even if.the,surface sample is used, it would not
make a significant difference in the risk assessment. Furthermore,
'' the-distrubance of the soil structure during remediation, may make
the deep soil levels more relevant.' (2nd Part) See response to
"comment #3i --..;> ......
' ~;":1 ' - ' -'-.-^c
#35 See response- to, comment 133.
v
#36 See response to comment #3.3. .
'i V . 4 -
"* -- : :
#37 See response to comment . 13.__ _
#40 The soil sampling was done in order to define the extent of
contamination. This necessitated a large number of non-detect or
low level samples, in^prder to define all of the isopleths. An
average level of all the samples is1" not appropriate^ (Part 2) See
response to-comment.. 133* _
' '"'" "-'- : '* - 'I :.r.s.
#41 See response to comment #3.
#42 See-response-^p.,7comment #3.
143 While the wpxkAr,s. jajt the town garage may be exposed to other sources
of hazardous compounds, the purpose1'of the risk, assessment is to
determine. .Jth exp.ojgu^e f^om the site.
~ "f~ *-*.'
#44 See response to comment.#33,
#45 The report does not limit its treatment of drinking water to
on-nite. Considering the obvious releases from the site, the
consideration of the use of water for drinking and other uses is
appropriate.
-------
-2-
#46 See responses to comments $3 and #33
152 See repsonse to comment #51.
155 See response to comments #3 an #33.
-------
Reynolds
Draft Response to Comment 3
Documentation for the statements about' tlie human carcinogenicity
of Aroclors 1260, 1254 and 1248 was not provided, therefore, the
scientific basis for these statements cannot be evaluated. However,
the statement that Aroclors 1254 and 1260 "are not considered to be
carcinogenic to animals* is incorrect. Two studies showed the
carcinogenicity of Aroclor 1260 in rats (Ximbrough et al., 1974;
Nor back and Weltman, 1985) and a third study showed the
carcinogenicity of Aroclor 1254 In rats (Ward, 1985). In addition,
the US EPA (1987) considers these compounds, as well as Kaneclor SCO,
..... - ,.. ... . _.:». ....:.-*". : ' ':
Clophen A-60 and Clophen' X'X^ ^9 be, jinimalL.^car.cinoj^ns. Similarly,
both the International Agency for Research on Cancer (IARC, 1982) and
the National Toxicology Program (NTP, 19 85 reconsider PCBs,, as a group
of compounds that a rV carcinogenic in animal if.'"
Furthermore," the; isr EPA +l9Sf, IARC (» a$f the _WTP (1985)
"" -^.
all consider the evidence on the" human caVcinogenicity of PCBs to be
inadequate (conclusions, on their, caxcinogenicity., cannot be made on
» . - . .. .' ".-Jig:-..... . . . *..* ---- u c. . . . - . -- .-.-.= 3 . . «.'
the basis of the available dita^ .""Ifonse^'uehtfly, given the positive
animal data and the inadequate^ .human data, the OS EPA considers PCBs
to be probable human carcinogens, the OTP considers PCBs. to be a
..-;;. :" : sifl . .... . '. -* - -
and human '
-------
Alcoa '
Draft Response to Comment 5
Documentation for the statement that only those "isomers of PCS
with 60% or greater chlorine content are suspected carcinogens*' was
not provided, therefore, _.tne.. scientific ba'sis for the statement
cannot be evaluated. However* the statement is .incorrect.
Undoubtly, Dr. Roberta .was discussing the carcinogenicity of
commerical -mixtures of Aroclors_as he concluded .^hat only those
mixtures (i^«.f Aroclors) with a chlorine content (by weight) of
greater than.60*.art carcinogenic in animals,,,He could not be
discussing isomers because the carcihogenic^agerits in commerical*
mixtures of PCBs have not been^identifjed^.g .r - .-=. .= ,,
' ~ -.-.; .-: < , a r c| ^ -...-. t- - . » . ^ _. _ .
However,-the-opinion of Dc-Boberts regarding the carcinogenicity*
of Aroclors is not held by the US EPA (1987) (see response-to-comment
3 of Reynolds). The evidence is clear that Clophen A-30 (a
commerical mixture with only 30% chlorine) and Aroclor 1254 (a
commerical mixture with 54% chlorine) are carcinogenic in rats
(Schaeffer et al., 1984; Kimbrough et al., 1975).
Aroclors 1242 and 1248 are not "PCS isomers* but commerical
mixtures containing a large number of PCS congeners.
In their most recent review of PC3 toxicity, the OS EPA (1987)
concluded there is no information regarding which constituents of any
PCS mixture are the active carcinogenic moieties, therefore/ Aroclor
1260 will be assumed to be representative of all mixtures and the
cancer potency factor of 7.7 (mgAg/day)"1 derived from studies
with Arcolor 1260 will apply to all PCS mixtures.
-------
. Draft Response to Comment >
Documentation to support the estimate that the gastrointestinal
absorption factor for PCBs in drinking water should be 15-30% was not
provided, therefore/ the scientific basis for the estimate cannot be
evaluated. However, limited evidence in humans indicates PCBs are
absorbed in the gastrointestinal tract (quantitative evidence on the
amount absorbed is unavailable) and animal data, including that for
rhesus monkeys, indicate that over 90% of ingested PCBs are absorbed
(Albro and Fishbein, 1972; Allen et al., 1974a, Morales et al.,
1979).
Draft Response to Comment 34 ......
* - 5 .- -. - " *: S ' 3 ' * _ - * " "" ' ' . " ' ' :'-.
Although the toxicities of specific Aroclors was presented in a-
recent US EPA (1987), review on. .PCBs, the available evidence is not
sufficient to conclude that PCBs with less-than 60% chlorine are of
less concern than those with .more than 60% chlgrine, ...
Available evidence, indicates that exposure .to -Aroclor 1016 (41%
chlorine), Aroclor 1248 (48% chlorine) and Arocloc 1254 (54%
chlorine) can induce noarcarcinAgenic, systemic toxicity and -
reproductive toxicity in ^rhesus monkeys (a valuable surrogate animal
model for humans) at relatively low-doses (Barsotti aj»d Van Miller,
1984; Barsotti et al^r 1976; Allen and Barsotti, 1976; Tryphonas et
al./ 1986a,b). In addition, these are no good data on the
non-carcinogenic toxicity of Aroclor 1260 (60% chlorine) in rhesus
monkeys and studies in rats indicate that Aroclor 1260 is less toxic
to rats that Aroclor 1254 (Linder et al., 1974).
-------
See response-to-comment 3 of Reynolds (second paragraph).
The use of "It" to refer to the large number of commerical
mixtures of PCBs is incorrect. "" ""." ---.s- -*
Draft Response to Comment 51
See response-to-comment 6
Draft Response to Comment 53 . ... .... .
The potency, fact or _ of PCBs .is correctly, presented,, as 4.34
(mgAg/day)"1. .The statement - "This seems.high-^or a. suspected.,
carcinogen." - reflects a.misunderstanding regarding the, difference-
between qualitative evidence of ,carcinogenicit£~(ISj-it a carcinogen?)
and q ua nt it at iv«,_ evidence of carcinogenicity (What is the. dose
required to induce a carcinogenic response24^ = -.. -
*.
The potency factor of 4.34 (mg/kg/day)"1 was taken from the
table found in many US-:EPA--Health- Assessment Documents ranking the
carcinogenic potencies of 5-5 cheaicals. -A more recent estimate is
7.7 (mgAg/day)~l (US EJ>A,^l»ft7)^. Both potency factors were
calculated from data obtained-from animal-studies. - _. --
H v
Draft Response to Comment S41-- : :" - - ..-
There is no predetermined r is Jt level that is -considered
regulatory agencies as^aeetfptable^" In general/ risks greater"than
IQ- . io*7 warren* consideration for regulatory action. The
-------
determination that a given level of risk is acceptable *'.'' respect
to clean-up criteria should be made on a case-by-case basis, after
all the factors of the case are carefully assessed and after the
concerns of all interested parties are addressed. An initial ~
-determination that risk levels of 10'4 to 10~7 may be of public
health concern is consistent with this approach
kgb-12
-------
A0946
REFERENCES
Albro, p.W. and L. Pishbein. 1972. Intestinal absorption of poly-
chlorinated biphenyls in rats. Bull. Environ. Contain. Toxicol.
1: 26-31. .
Allen, J.R. and D.A. Barsotti. 1976. The effects of transplacental
and mammary movement or PCSa on infant rhesus monkeys.
Toxicology, it 331-3-40. . ,, ,
Allen, J.R., D.H. Norbacfc and I.C. Hsu. 1974a. Tissue modification
in monkeys as related to absorption, distribution and excretion
of polychlocinated biphenyls. Arch. Environ.- Contain.. Toxicol. 2:
. 86-95. ,.;. '"..
Barsotti, D.A. and J.P. van Miller. . .1984. Accumulation- of<-«=
commercial polychlorinated biphehyl mixtutV (Aroclor 1016) in
adult rhesus 'monkeys and their nursing, infants, Toxicology. 3Qs
"31-44."" " '". . ., ,.i ..
Barsotti, D.A. , R.J. Marlar and J.RV Allen. 1976. Reproductive
dysfunction^ in rhesus monkeys exposed to low levels of
. polychl or in* ted biphenyls. Pppd' Cpsnet. Toxicol. I1& 99-103.
-.. . -... - . - ,-, ...... r .. . . .. ... .. - . ---
.International Agency^ for Research on Cancer (IARC)._ 1982. IARC
. . Monographs on the;«BvAluation. of , Carcinogenic Risk of Chemicals to
Humans. (ARC Monographs Supplement 4." Ly on. Prances World
'Health Organization*^- .,,.;,.^ :
<
-------
-2-
Norback, D.B. and R.B. Weltman. 1985. Polychlorinated biphenyl
induction of hepat©cellular carcinoma in the Sprague-Dawley rat.
Environ* Health Perspeet. &&t 97*105.
Office of Science and Technology Policy (OSTP). 1985. Chemical
carcinogens: A review of the science and its associated
principles. Fed. Register. ££_t 10371-10442.
Schaeffer, B.» H. <3reim anc^W. Goessner. 1984. Pathology of chronic
polychlorinated biphenyl (PCB) feeding in rats. Toxicol. Appi;
Pharnacol. 25* 278-288.
.x
0.5. Environmental Protection Agency (DSCPA). 1987. Drinking Water
Criteria Document for Polychlorinated Biphenyl. ECAC-CIN-414.
Final.
Ward, J.M. 1985. Proliferative lesions of the glandular stomach and
liver in P344 rats fed diets containing Aroclor 1254. Environ.
Health Perspeet. £Qt 89-9$. . . ..;.... r .
-..:i::-. T, i..t7iT'atn c ..;>>:/, 5
Tl: ". .. .:-: - ~ ",iifK .-nyr:;nf t:#
------- |