"U:;ited States
Environmental Protection
Agency
Office of
Emergency and
Remedial Response
EPA/ROD/R03-90/101
September 1990
&EPA Superfund
Record of Decision:
Keystone Sanitation Landfill, PA
-------
50272-101
REPORT DOCUMENTATION 11. REPORT NO. 1 2. 3. Recipient's Accession No.
PAGE EPA/ROD/R03-90/101
4 Title and Subtitle 5. Report Date
SUPERFUND RECORD OF DECISION 09/30/90
Keystone Sanitation Landfill, PA
6.
First Remedial Action
7. Author(a) 8. Performing Organizetion Rapt. No.
II. Performing Orga/nlzation Name and Addreaa 10. ProjecllTaakiWork Unit No.
11. Contracl(C) or Grant(G) No.
eC)
(G)
I
12. Sponsoring Organization Name and Addreaa 13. Type o' Report & Period Covered
U.S. Environmental Protection Agency 800/000
401 M Street, S.W.
Washington, D.C. 20460 14.
15. Supplementary Notea
16. Abstract (Umlt: 200 words)
The 40-acre Keystone Sanitation Landfill site, an inactive, privately owned landfill,
is in Union Township, Adams County, Pennsylvania. Surrounding land use is primarily
agricultural with scattered residences. From 1966 to 1990, the unlined landfill
accepted household and municipal wastes as well as industrial and construction debris,
~ncluding phosphorus-contaminated sand, potato sludge, resin sludge, incineration ash,
..md dried latex paint. The volume of non-homogeneous waste at the site currently is
1.7 million cubic yards. In 1982, State investigations revealed onsite ground water
contamination and a contaminated onsite residential well attributable to leachate from
the landfill contents. In 1984, EPA found low-level contamination in nearby
residential wells. As a result of the ground water contamination, EPA ordered Keystone
to install an onsite spray irrigation system and a leachate collection system to
prevent migration of contaminants offsite. This Record of Decision (ROD) addresses
Operable Unit 1 (OUl), the containment of onsite source area and remediation of onsite
contaminated ground water. A subsequent ROD will address offsite ground water
contamination in monitoring and residential wells (OU2). The primary contaminants of
(See Attached Page)
17. Document Anaiyeis L Descriptors
Record of Decision - Keystone Sanitation Landfill, PA
First Remedial Action
Contaminated Media: soil, gw
Key Contaminants: VOCs (benzene, PCE, TCE, vinyl chloride), other organics (acids,
phenols), metals (chromium, lead)
b. Identifiers/Open-Ended Terms
c. COSA TI Field/Group
'A. AvailabiDty Statement 19. Security Class (This Report) 21. No. o' Psges
None 143
20. Security Class (This Page) 22. Price
I None -.
,.,272 (4-77)
(See ANSI-Z39.18)
See Instrucllons on Rsversq
(Formerly NTlS-3S)
Department o' Commerce
-------
EPA/ROD/R03-90-l0l
Keystone Sanitation Landfill, PA
First Remedial Action
rlbstract (Continued)
concern affecting the soil and ground water are VOCs including benzene, PCE, TCE, and
vinyl chloride; other organics including acids and phenols; and metals including
chromium and lead.
The selected remedial action for this site includes excavating approximately 4,150 cubic
yards of contaminated surface soil with consolidation of the soil in the landfill;
placing a cap on the landfill and. subsequent revegetation; installing an active gas
extraction system to collect gases emitted from the landfill; pumping and treatment of
ground water using equalization, flocculation/precipitation, filtration, ion exchange,
air stripping, and filter press technologies, followed by offsite disposal of any sludge
produced by these processes; discharging the treated water onsite to a surface stream;
monitoring ground water, surface water, and sediment; installing an in-home water
treatment system for the onsite resident; and implementing site access restrictions and
institutional controls, including deed, land use, and ground water restrictions. The
estimated present worth cost for this remedial action is $9,156,950, which includes an
annual O&M cost of $217,000 for 30 years.
PERFORMANCE STANDARDS OR GOALS: No chemical-specific cleanup goals were stated in the
ROD. For carcinogenic compounds, cleanup goals were established to reduce the excess
lifetime cancer risk to 10-4 to 10-6. For non-carcinogenic compounds, the goal is a
hazard index (HI) equal to 1 or less. Individual ground water remediation standards are
based on the more stringent of SDWA MCLs or non-zero MCLGs, or State background levels.
If these levels cannot be met, the ROD will be amended.
-------
RECORD OF DECISION
DECLARATION
Site Name and Location
Keystone Sanitation Landfill Site
Adams County, Union Township, Pennsylvania
Operable Unit One
Statement of Basis and Purpose
This decision document presents the selected remedial action
for the Keystone Sanitation Landfill Site (Site) in Adams County,
Union Township, Pennsylvania. It was developed in accordance with
the Comprehensive Environmental Response, Compensation and
Liability Act of 1980 (CERCLA), as amended, 42 U.S.C. Section 9601
et seq. and is consistent, to the extent practicable, with the
National Oil and Hazardous Substances Contingency Plan (NCP) , 40
C.F.R. Part 300. 55 Fed. Reg. 8666 (March 8, 1990). This decision
is based on the contents of the Administrative Record for the Site.
The Commonwealth of Pennsylvania concur with the Selected
Remedy.
Assessment of the Site
Pursuant to duly delegated authority, I hereby determine,
pursuant to Section 106 of CERCLA 42 U.S.C. § 9606 that actual or
threatened releases of hazardous substances from this Site, as
discussed in the Summary of Site Risks on pages 18 through 30, if
not addressed by implementing the response action selected in this
ROD, may present an imminent and substantial endangerment to public
health, welfare or the environment.
Description of the Remedy
The Selected Remedy addresses the principal threat posed by
the Site from the ingestion of groundwater by reducing the risks
to human health and environment through the use of groundwater
extraction and treatment and the installation of an impermeable
cap. It does not provide treatment of the waste as it has been
found to be impracticable. This remedy will prevent continued
infiltration from the wastes to the groundwater and reduce the
-------
.2
concentration and plume of groundwater contamination.
Operable unit will address offsite contamination.
A second
Engineering and institutional controls are used to effect this
remediation through the following components:
o
Installation and maintenance of an impermeable cap and
gas collection system over a 40-acre Site
o
Installation and maintenance of groundwater extraction
wells and treatment plant to reduce the concentrations of
volatile organic compounds and metals in the groundwater
o
Installation and maintenance of a fence around the site
o
Monitoring of the groundwater in monitoring and
residential wells
o
Monitoring of the surface water and sediments
o
provision of a point-of-use treatment system to onsite
residents
o
Initiation of deed restrictions regarding present and
future site activities including limitations on
construction and aquifer use.
This Site will have a second Operable Unit to further study
the groundwater contamination in the offsite monitoring and
residential wells.
Declaration
The Selected Remedy is protective of human health and the
environment, attains Federal and State requirements that are
applicable or relevant and appropriate for this remedial action
and is cost-effective. This remedy satisfies the statutory
preference for remedies that employ treatment that reduces
toxicity, mobility or volume as a principal element and utilizes
perm~nent solutions and alternative treatment (or resource
recovery) technologies to the maximum extent practicable for this
Site. However, be~ause treatme~t of the pr~ncipal threatz of the
-------
found not to be practicable as a principal element of the remedy
since the size of the landfill and the nonhomogeneity of the wastes
preclude a remedy in which 'the contaminants could be treated in a
cost-effective manner.
Because this remedy will' result in hazardous substances
remaining on-site above health-based levels, a review shall be
conducted as required by Section 121 (c) 0 f CERCLA 42 U. S . C. ~
9621 (c), within five years after commencement of the remedial
action, and every five years thereafter as required, to ensure that
the remedy continues to provide adequate protection of human health
and the environment.
r£L 3c? ~
, -( ~~-
Edwin B. Erickson
Regional Administrator
Region III
9lS0/9u
Da e
-------
KEYSTONE SANITATION COMPANY SITE
RECORD OF DECISION
Table of contents
Page
site Name, Location and Description
1
site History
3
Enforcement Activities
4
Highlights of Community participation
5
Scope and Role of Response Action
5
Summary of site Characteristics
5
Nature and Extent of contamination
10
Summary of site Risks
18
Description of Alternatives
29
Comparative Analysis of Alternatives
38
The Selected Remedy
43
statutory Determinations
49
-------
Tables
1.
2.
3.
4.
S.
6.
7.
8.
9.
10.
KEYSTONE SANITATION COMPANY SITE
List of Tables
contaminants in the soil
contaminants in the Surface Water
and Sediments
Contaminants in the Groundwater
Cancer Potency Factors
References Doses
Lifetime Excess Cancer Risk
by Environmental' Exposures
Routes
Noncarcinogenic Risks by
Exposure Routes
Noncarcinogenic Risks by
Environmental Media
Estimated Costs for Five Alternatives
Costs of Selected Remedy
Page
13
14
lS
23
24
2S
26
27
42
48
-------
~---
I
KEYSTONE SANITATION COMPANY
List of Figures
Figures
Page
1.
site Location Map
2
2.
Generalized Stratigraphic Map.
7
3.
Geologic Cross section
8
4.
Lineation Due to schistosity
9
5.
Potentiometric surface Map
11
6.
Groundwater Exceeding ARARs
17
7.
Area of Attainment
32
8.
Alternative 4 - Plan View
35
9.
Remedy Flow Diagram
45
..~ 4
-------
RECORD OF DECISION
KEYSTONE SANITATION LANDFILL SITE
Decision Summary
SITE NAME. LOCATION AND DESCRIPTION
The Keystone Sanitation Landfill Site (Keystone site or Site) is
an inactive privately owned facility (Keystone Sanitation Company
or Keystone), permitted by Pennsylvania Department of Environmental
Resources (PADER) to receive household and municipal wastes and
certain types of industrial and construction debris. The landfill
is located on a 40 acre tract of land in Union Township, Adams
County, Pennsylvania, southwest of Hanover, Pennsylvania and is
approximately 800 feet north of the Pennsylvania (PA)-Maryland (MD)
border. (See Figure 1 for Site location). It is fenced on the
eastern and western sides and bounded by Line Road to the south and
Clouser Road to the northwest.
The landfill which operated from 1966 to April 1990 is situated on
top of a ridge. It is built in a horse-shoe shape with the owner
of the landfill residing in a home on the inside edge of the
landfill property. There are approximately 36 residents within a
1-mile radius of the Keystone Site and approximately 700 residents
within a 3-mile radius of the site. Littlestown, Pennsylvania is
the closest town to the Site. It has a population of approximately
3,000 and is located 3 miles north of the Site. Some residences are
located near the landfill, but the predominant land use is
agricultural, not residential. Residents in the area of the Site
utilize domestic wells to obtain their water supply.
The topography of the area consists of gently rolling hills and
valleys formed by the northeasterly tending elongated valleys and
ridges. Most surface water flows northward to an unnamed perennial
tributary of Conewago Creek located 100 feet north of the Keystone
site. A smaller quantity of runoff flows southward into an unnamed
tributary to Piney Creek, located about 2000 feet south of the Site
in the State of Maryland.
The landfill was constructed as a renovating base landfill without
a liner or leachate treatment or collection system. Depth of
wastes average 30 feet. Fractured bedrock of the Marburg Schists
underlies the Site overlain by varying thicknesses of silty clay
soil which was used for constructing the base of each cell and for
daily, intermediate and final cover. Its maximum elevation is
approximately 700 feet with a vertical relief of approximately 100
feet within a 2000 foot radius of the site. A perennial grass cap
is growing over the Site with the exception of the two newly closed
cells. .
-------
----- -,
'0 ..,
Q .
. .",
... ..,
... +
~
z
~
C
2
Z
2
... c.
C i
u
° ~
....J ~
>
II!
Z
Z
IW
&.
'-
i.
SCALE
I: 37,820
DATE.
JULY 1990
.10
°lu
u~
en ~
2°
~
CIIZ
QC
C U
1
I
REM V
SITE LOCATION MAP
KEYSTONE SANITATION LANDFILL SITE
LITTLESTOWN, PENNSYLVANIA
FIGuR
1
I
I
C.C.JOHNSON ~ MAe HOTRA,P.C.
2
-------
SITE HISTORY
Since 1966 (pre-dating the establishment of PADER in 1970),
the Keystone Site has been used asa sanitary landfill. It
is estimated that 300 to 376 tons per day of waste have been
disposed at the site. In 1974, five monitoring wells (K1,
K2, K3, K4, and K5) were installed by Keystone at the landfill
perimeter to monitor groundwater ,quality.
In 1982, all facilities permitted by PADER were required to monitor
groundwater for volatile organic compounds (VOCs). A sample taken
by PADER in November 1982 from Keystone monitoring well K1 revealed
VOC contamination in the groundwater. subsequent testing of the
onsite residential well and the nearby Mundorf Spring revealed that
they also contained VOCs.
In April 1984, an EPA Field Investigation Team (FIT) performed a
site investigation in response to citizen complaints of illegal
dumping and groundwater contamination and to assess the site I s
eligibility for inclusion on the EPA National Priorities List (NPL)
established pursuant to Section 105 of CERCLA, 42 U.S.C. ~ 9605.
Sample results from the PADER and EPA FIT investigations confirmed
that some residential wells contained low levels of VOC
contamination.
In August 1984, as result of the VOC contamination, Keystone
installed a spray irrigation system in the most contaminated area
to prevent the migration of contaminants offsite and to remove VOCs
from the groundwater. Water from Keystone monitoring well Kl was
pumped to a series of sprayers located at the edge of the landfill,
within the radius of influence of the well., In addition, a
leachate collection system was installed on the south side of the
landfill along Line Road. The leachate collection system consists
of two perforated pipes and a storage tank. The storage tank is
pumped periodically and the leachate disposed of offsite.
In the spring of 1985, the State of Maryland installed a series of
monitoring wells at the Maryland border to monitor potential
contaminant migration into Maryland. Low levels of VOC
contamination have been consistently detected in Maryland well No.2
(MD2). However, no Maryland residential wells have been determined
to be contaminated as a result of the landfill.
The Keystone Site was placed on the NPL in July 1987. In July
1987, the Potentially Responsible Parties (PRPs) were asked to
perform the Remedial Investigation (RI) and Feasibility Study
(FS) for the site. Negotiations failed to obtain cooperation from
the PRPs to do the RI/FS and EPA assigned the RI/FS tasks to REM
V, a federal government contractor.
3
. '~.'''., , ..'
-------
Also in July 1987, Keystone signed a Consent Adjudication (CA) with
PADER. The intent of the CA was to provide data for the
development of an onsite groundwater remediation plan and to design
and Implement the plan. Keystone was required by the CAto analyze
and summarize previously collected water quality data, determine
the effectiveness of the existing spray irrigation system, install
3 additional monitor wells, abate groundwater contamination at the
site perimeter, prevent offsite groundwater contamination, and
treat the contaminated offsite.. .
EPA and PADER recognized that two separate studies (the CA and the
RI/FS) were being conducted on the Keystone Site. Therefore, it
was agreed that the RI/FS would focus on obtaining offsite data
while the CA study would focus on obtaining onsite data.
The RI/FS field activities began in the spring of 1989 and were
completed in the winter of 1990. The objectives of the RI were:
to determine the nature and extent of hazardous substances,
pollutants or contaminants at the site: to determine the impact of
these hazardous substances on public health, welfare and the
environment: to determine the extent to which sources of
contaminants can be adequately identified and characterized: to
gather sufficient information to determine the necessity for
remedial action: and to provide data in order to evaluate and
estimate costs for .remedial alternatives during the FS.
The purpose of the FS was to develop a range of cost effective
remedial alternatives which are protective of human health and the
environment and comply with applicable or relevant and appropriate
requirements (ARARs).
The RI/FS was finalized on July 20, 1990 and released to the public
along with a Proposed Remedial Action Plan. A 60 day public
comment period followed the release of these documents.
ENFORCEMENT ACTIVITIES
In November 1987, approximately 235 parties were sent 104 (e)
Information Request Letters and General Notice Letters to inform
the PRPs of their potential liability with respect to remedial
actions at the Keystone Site. In July 1988, 21 parties were issued
Special Notice Letters pursuant to section 122(e) of CERCLA, 42
U.S.C. ~ 9622(e), to request their participation in the RI/FS
activities. A meeting was held in August of 1988 to negotiate with
the PRPs. Negotiations failed to obtain cooperation from any of
the PRPS and EPA funded the RI/FS activities. An additional PRP
search is currently being finalized for this site.
4
-------
HIGHLIGHTS OF COMMUNITY PARTICIPATION
Community involvement at this site is very high with a mailing list
of over 600 people. There are at least four organized citizen
groups, two townships, two borough, and two State (MD and PA) and
1 County environmental agencies concerned about the pUblic health
and environmental effects of tJ:1e .landfill. These parties have
rigorously interacted with the EPA and the State governments.
In accordance with Sections 113(k)(2)(B)(i-v) and 117(a)(2) of
CERCLA, the public was kept informed and given an opportunity to
participate in the Keystone Site activities. A public meeting was
held on March 29, 1988 to discuss the workplan for the RI/FS.
Numerous comments received on the RI/FS workplan were considered
in the revisions to the Workplan. Several visits and numerous
phone calls have been made to the community to keep them informed.
In addition, two Fact Sheets on the RI/FS activities and the
Proposed Plan have been distributed to the people on the mailing
list.
On July 20, 1990, the Proposed Plan and the RI/FS were placed for
publ ic viewing in the four information repositories: The Adams
County Public Library; the Hanover Public Library; the Carroll
County Public Library and the St. Mary's United Church of Christ.
The notice of availability was placed in the Gettysburg Times and
the Hanover Evening Sun on July 20, 1990. A public comment period
was originally to be held from July 20, 1990 to August 20, 1990.
This comment period was extended to September 20, 1990 at the
request of the public. The originally scheduled pUblic meeting
for August 15, 1990 was rescheduled and held on September 13, 1990.
This rescheduling was requested by the public to allow them time
to review the information and prepare comments for the meeting.
SCOPE AND ROLE OF RESPONSE ACTION
The scope and role of the response action is to prevent current
and future exposure to the contaminated soils, to reduce
contaminant migration into the groundwater and to prevent migration
of the contaminated groundwater to uncontaminated areas. This
response action addresses the principal threat at the Site from
groundwater contamination through an impermeable cap and
groundwater extraction and treatment. A second Operable Unit will
address the offsite contamination concerns.
SUMMARY OF SITE CHARACTERISITICS
The Keystone site located in the upland section of the western
Piedmont Physiographic Province is bounded on the east by the
Coastal Plain Province and on the west by the Blue Ridge Province.
5
-------
Soil near the Keystone site is typified by a variety of well-
drained loams and silt loams containing rock fragments, and silt-
and clay-size micas derived from the underlying saprolitic zone.
The soil is described as a highly micaceous silty clay with some
rock fragments, but containing no visible relict rock structures.
The Keystone Site is underlain by Precambrian Schist equivalent to
the Babylon Phyllite of the Marburg formation (Marburg Schist) in
Maryland. The Marburg is overlain by saprolith, a clayey soil with
relict rock structures produced by in place weathering . of the
underlying schist. A generalized stratigraphic column of Keystone
is presented in Figure 2, along with a cross section of the Site
in Figure 3.
The steeply dipping Marburg Schist is a Precambrian Age
metasediments consisting of a. fine-grained, finely laminated
grayish-green to grayish-blue schist which is predominantly
composed of chloride, muscovite, quartz and albite. Considerable
porosity is demonstrated in some quartz zones. Fractures were
seen at the site in some thick quartz zones encountered and mostly
associated with loose, fractured vuggy quartz. Most fractures TNere
annealed either with quartz or chlorite. Fracture zones ranged
from a tenth of an inch to approximately one foot in thickness and
usually contained crushed schist with small amounts of clayey
gouge. .
The saprolith overlying the Marburg Schist is produced from the
in-placed weathering of the schist. The saprolith is of variable
thickness and ranges from thin veneer on ridge tops to 30 feet in
thickness in drainage areas and valley bottoms. Saprolith at the
site consists of reddish-brown, brownish-gray and pale grayish-
green clay and heavily weathered schist locally containing fresh
quartz veinlets and fragments. Weathered schist which was
. significantly more competent than the overlying saprolith, was
encountered in one of the boreholes from approximately 30 to 50
feet in depth. The weathered schist is a silvery grayish-green
finely laminated rock that displays reddish brown to pale yellow-
brown iron staining on foliation and fracture surfaces.
Geologic structures at the Keystone Site include faults, and
fractures, and small-scale folding within the Marburg Schist. The
dominant orientation of the foliation of the Marburg Schist
parallels the regional NE-SW trend. (See Figure 4 for schistosity
structure.) Fracture trace analysis at the site confirmed two
major structural trends: a regional sChistosity striking between
N 35 E and N 60 E, which is pervasive in the area of the landfill;
and a local structural trend oriented almost due north and south
which is apparent in the area immediately south of the landfill.
These structural trends suggest that the groundwater flow may be
locally deflected along preferential transport pathway parallel to
the northeast-southwest schistosity in all directions from the
site. To the south of the landfill, flow may also be affected by
6
-------
ERA PERIOD LITHOLOGY DESCRIPTION
-
I o' I :. " -I' - ~~ ~
...
o -" I'~~II
'" -, -
/ /
/
/
1 ~ // /
Ct
c.. i !
.... ....
~ ~ ~
... ~ /
~ i
\i ~ /
SOIL - PALE YEU,OWI$H 8ROWN, COHESIVE,
FROM A FEW INCHES TO . 'EET THICK,
~
SAPROLITH - "ALE YELLOWISH IROWN TO
YILLOWISH OREY WEATHIRED
ICHIIT. CONTAINING RESIDUUM 0'
LEIS WEATHERED SCHIST AND OUARTZ
'RAOMINTS. ORADUAL CONTACT WITH
UNDIItLYING SCHIlt. RANGIS FROM
A 'EW INCHES TO 50 nET THICK.
SCHIST-GItATIIM OltllN TO ORATISH ILUE
. SCMIIT CO"~SID 0' CHLOItITE,
MUICOVITI, QUARTZ AND ALIITE,
'INIL. Y L.AM IN ATID WITH CALCAREOUS
AND OUARTZ -RICH ZONIS. "NILY
DI881111NATID 'YltITI 'S COMMON,
~
....
-
~
-
.
~
;
-.:
L
.- 8CALE REM Y
AS SHOWN GENERALIZED STRAtiGRAPHIC COLUMN
D ATE KEYSTONE SAN'T~TION '..4NDFU.'- SITE
APRU..1990 UNION TOWNSH!'!8, PEHHSYLVANIA
C.C.JOHN80N I MALHOTRA,P.C. 7
FIGURE
2
.
-------
00
END FONT Or CftOM MCTKM,
ON CHAMQC OF OMBCTION
CMANQf Or UTHOLOOV CONTACT
NO MTOAMATION AtNUJAMjC
KEYSTONE
UNOTILL SITE
MW-MOMTOft WCU,
SCALE
AS SHOWN
REM V
KEYSTONE SITE GEOLOGIC CROSS SECTION
KEYSTONE SANITATION LANDFILL SITE
UNION TOWNSHIP PENNSYLVANIA
ON & MALHOTRA.P.C.
-------
I ~
N
~
,
~"/
"I.'~
..,./
~~~
\,..~~~~
..'
+... ~c,
~~.~
'\ to,/, +~
~~
..~ .
/
/
/
/
/'
./
/
./
./
/
,/ PENNSYLVANIA
-
MARYLAND
-"W.FS
MW- FO .
;t .,If ...w - GO
MW-(jl ~w -AC
. MW -(jS ~w - >\:
~W -AS
. I iI
SOURCE: BERG ET AL, 1180.
I
x
J
;/]/
)f
Jf
)1('
x
LEGEND
- - - LITHOLOGIC CONTACT
~ ;t LINEATION ORIENTATION
. MONITOR WELL OR WELL CLUSTER
o
I
1000 2000
. .
f£lT
3000
J
SCALE
AS SHOWN
DATE
APRIL. 1990.
REM V
LINEATIONS DUE TO SCHISTOSITY,
STRUCTURE OR COMPOSITIONAL LAYERING
KEYSTONE SANITATION LANDFILL SITE -
UNION TOWNSHIP PENNSYLVANIA
MALHOTRA.P.C. .
FIG,-~E
1-
C.C.JOHNSON &
9
-------
a local north-south secondary structural trend which is parallel
to the slope of the land surface as a result of the northeast-
tending bedrock ridge. . .
Groundwater at the Keystone site occurs in both the shallow
saturated saprolith (including the weathered schist) and in the
underlying unweathered schist. However, shallow wells completed
in saturated saprolith were found to produce significantly more
water than deep wells in unweathered schist. This can be
contributed to the weathering which has increased the secondary
porosity of material, closer to the surface, allowing the saprolith
to yield more water.
Two zones of low yield exist at the Keystone Site. One zone
is deep, while the other is near the surface. In deep wells, low
yields are the result of low secondary porosity associated with
low permeability and poorly developed fracture networks in the
schist below the weathered zone. In the upper zone from the
surface to several feet below the surface, soil permeability is
lower and secondary porosity is not a factor. The soils on the
upper zone are silty clay with some fragmented rock, and
restricts movement and contaminant migration in the shallow
saturated saprolith.
Groundwater discharges to the surface waters were not documented
in the RI investigation. However , given the topographic and
potentiometric relation shown in Figure 5, a hydraulic connection
can be expected.
The volume of wastes in the landfill have been estimated to be
approximately 1.7 million cubic yards of non-homogeneous wastes.
Very few records exist on the content of the cells in the landfill.
The RI and CA investigations did not investigate the contents of
the cells. It has been reported that in addition to the municipal
waste accepted, various industrial solid wastes including
phosphorus-contaminated sand, potatoe sludge, resin sludge,
incineration ash from a wastewater treatment plant and dried latex
paint were accepted. According to the Keystone owner and the
manager, hazardous wastes have never knowingly been disposed at the
Site.
NATURE AND EXTENT OF CONTAMINATION
The Keystone site consists of an estimated 1.7 million cubic yards
of nonhomogeneous waste. This is considered the source of
contamination which has been detected in the soils, surface water,
sediments and groundwater on and adjacent to the Site. The route
of contaminant migration has been via the groundwater either
through groundwater discharge or the spray irrigation system.
Potentially exposed populations would include residents on or
adjacent .to the Site, workers at the Site and trespasser or
visitors to the site. In addition, environmental areas adjacent
10
-------
KEYSTONE
LANDFILL.
GftOUMMMffCft
(DASHfD
WHCMf
APPROXIMATE)
GAOUNOWATEA
FLOW
DIHCC7IOM
REM V
POTENTIOMETRIC SURFACE MAP
DATE
JUNE I99O
KEYSTONE SANITATION LANDFILL SITE
UNION TOWNSHIP, PENNSYLVANIA
-------
to the site would be potentiallY exposed to the contaminants.
RI field activities included fracture trace analysis, surface
geophysics, drilling and coring, borehole geophysics, packer tests,
pump tests, monitor well installation and collection of Contract
Laboratory Programs (CLP) samples which were analyzed by an EPA
approved laboratory. CLP sampling included soil, surface water and
sediments, groundwater and residential wells. An environmental
survey was also conducted for i~pacts on local aquatic and land
habitats.
There are currently eight monitoring wells onsite which have been
installed by Keystone. In the RI, a total of 20 monitoring wells
and one observation well were installed. This included four
clusters consisting of a shallow, intermediate, and deep wells,
three cluster wells consisting of shallow and deep wells and two
single intermediate depth wells... The shallow wells were installed
to monitor the water table in the unconsolidated saprolith and
weathered bedrock at depths from 25 to 45 feet. The intermediate
wells were installed in consolidated bedrock where significant
water bearing zones were found at depth between 83 to 137 feet.
In some locations, water was not available at the intermediate
locations. The deep wells were screened at the deepest apparent
water bearing zone at depths from 152 to 252 feet.
Soil sampling was conducted in six general areas and included 24
locations at two sample depths. Three soil samples (SL-19,SL-20,
SL-21) were designated as background.
VOC contamination of the soil was confined to a few locations in
the spray irrigation area. VOCs were not found in off-site soil
samples. Several base neutral/acid extractables (BNAs) compounds
and a few pesticides were found in the spray irrigation area. All
TAL metals except cadmium, silver and cyanide were detected in
onsite soil samples. Several metals (arsenic, chromium, copper,
lead and zinc) detected onsite were also detected at similar
concentration ranges offsite and in the core samples. The
contaminants detected and their range of concentration can be seen
in Table 1.
Eighteen surface water and ten sediment samples were taken
during the RIjFS. Surface water and sediment samples taken at SW-
1, SW-3, SD-1 and SD-3 were designated as background samples.
Contaminants detected in the surface water include three VOCs found
in samples from two surface water sampling locations that receive
runoff and sediment from the landfill surface water. Also detected
were five inorganics (mercury, cyanide, lead, chromium, copper and
zinc) which exceeded the Ambient Water Quality criteria (AWQC), 40
CFR Part 131. BNAs, pesticides, and PCBs were not detected in
surface water samples. In the sediments, two BNAs and nine
inorganics were detected. The contaminants detected in the surface
water and sediments and their range of concentration can be seen
in Table 2. .
12
-------
KEYSTONE SANITATION COMPANY SITE TABLE 1
Compounds/Analytes Detected in Soil samples
Compounds/Analyte
Onsite
Ranqe of Concentration
Volatiles (VOCs}
1,1 Dichloroethane
1,2 Dichloroethene(total)
1,1,1 Trichloroethane
Tetrachloroethene
BNAs/Pesticides
Naphthalate
Acenaphthen*
Phenanthrene
Anthracene
Fluoranthene
Benzo(~~~nthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
Indeno(1,2,3-cd)pyrene
Dibenzo(a,h) anthracene
Benzo(q,h,i)perylene
Pentchlorphenol
Benzoic Acid
Dimethylphthalate
Diethylphthalate
Butylbenzylphthalate
Di-n-octylphthalate
N-~itrosodiphenylam1ne
Heptachlor epox1de
Dieldrin
4,4'DDE
4,4'DDT
4-chloro-3-..thyl phenol
Inoraanic8
Antimony
Arsenic
Beryllium
Chromium
Copper
Lead
Mercury
Nickel
Selenium
Silver
Zinc
2J
6J
2J-9
43
19J
33J
19J-160J
14J-120J
14J-200J
36J-190J
20J-89J
23J-200J
13J-160J
21J-180J
110J
160J
100J
73J
28J-240J
68J-88J
15J-160J
10J-140J
120J
6.8J
12J-14J
96J
(6.0)-(6.3)L
(0.6)-4.8L
(0.5)-1.5
13.4-22.6
11.2J-43.3
9.2-80
0.11-1.2
(6.2)-29.3
(0.81)J
32.5-106
13
Offsite
Ranqe of Concentration
17J-36J
14J-63J
19J-35J
22J-31J
19J
23J-130J'
26J
35J
5J-28J
2.3J-4.1J
2.2J-12J
76
8.7.L
(1.4)J-86L
(0.54)-1.6
10.7-32.2
8.3J-38.2J
5.1J-102J
0.10-0.13
(6.1)-29.1
(0.47)J
4.2"
'~3. 7-149
-------
KEYSTONE S~ITATION COMPANY
TABLE 2
Range ot concentrations detected in Surface Water and Sediments
Compounds/Analytes
Surface Water
ug/l
Sediments
uq/kq
Volatiles (VOCs)
1,1,1 Trichloroethane
Dichlorotlouremethane
1,1 Dichloroethane
2J
0.4
0.3
BNAs
Benzoic Acid
Bis (2-ethylhexyl) phthalate
110J
52J-130J
Inorqanics
Arsenic
Beryllium
Cadmium
Chromium
Copper
Cyanide
Lead
Mercury
Nickel
Selenium
Zinc
13.4
(12.0]
10.0-18.0
(1.1]-18.9K
0.30-7.8
[2.1]-13.3
(0.46]-2.1
2.7J-11.1J
12.7J-48.6J
9.5-31.7
[2.1]-[5.0]
[1.3]
[9.4]-67.7
16.6J-200J
12.7J-53.6J
(0.25]-[0.28]K
62.7J-206J
14
-------
KEYSTONE SANITATION COMPANY
TABLE 3
Range 'of Concentration of
Compounds/Analytes Detected in the Monitor Wells
Compounds/Analyte Onsite Offsite .
uq/l ug/l
Volatiles(VOCs)
Vinyl Chloride 4J-40
Chloroethane 3J-18J
1,1 Dichloroethene 5-16
1,1 Dichloroethane 4J-71
1,2 Dichloroethene(Total) 6-140 3J
1,1,1 Trich~oroethane 3J-52 20J
Trichloroethene 3J-57
Carbon Disulfide 7-8
Benzene 7J-14
Tetrachloroethene 8-48 11-15J
Dichlorodifluoromethane 4J-48
BNAs/Pesticides
Benzvl alcohol 3J
Benzoic Acid 3L 3J-4J
Diethylphthalate 2J
Bis(2ethylhexyl)phthalate 3J 3J-7J
N-nitrosodiphenylamine 4J
Indeno(1,2,3cd)pyrene 7J
Dibenzo(a,h) anthracene 4J
Aldrin 0.021J-0.16
4,4'DDT 0.04J-0.35
Gamma Chlordane 0.012J
Inoraanics
Arsenic [1.2]-21.2L [1.3]-[8.1]
Cac:lmium [2.8]
Chromium [6.4]-658 [4.0]-117
Copper [23.4]-1300J [24.3]-156
Lead [2.0]-13.7 [3.0]-25.1J
Mercury O.2J-0.4J
Nickel [24.6]L-1040 [17.3]-89.7
Selenium [1.0]L-[4.7]
Zinc 26.0J-98J [26.6]-77.6J
15
-------
Groundwater samples were taken from both the onsite and offsite
wells. The onsite wells included all Keystone wells (except K2) and
RI wells MW-HS and MW-HD which are near K1. Background samples were
obtained from wells MW-IS, MW-II, MW-ID, MW-FS and MW-FD which are
located to the northeast of the Site.
Groundwater sampling revealed the detection of ten VOCs in
onsite monitoring wells at concentrations up to 140 ug/1. Vinyl
chloride, 1,1-dichloroethene, trichloroethene, and benzene were
detected above the Federal Maximum contaminant Levels (MCLs), 40 CFR
141.11-141.16. Of these four contaminants, vinyl chloride and
benzene are known human carcinogens. Low concentrations of four
VOCs, carbon disulfide, 1,1,1-trichloroethane, tetrachloroethene
and total 1,2-dichloroethene, were found in from offsite
monitoring wells. No VOCs detected offsite exceeded Federal MCLs.
VOC contaminated was detected at a maximum depth of 150 feet below
the Keystone Site (MW-HD) and at shallow (25-35 feet) and at
intermediate depths (65-85 feet) in offsite wells MW-BS and MD2,
respectively.
Low concentrations (up to 3ug/1) of three BNAs, benzoic acid,
diethylphthalate, and bis(2-ethylhexyl)phthalate, were found in
onsite monitoring well K3. BNAs occurred in offsite wells in
concentrations up to 20 ug/l with a maximum frequency of detection,
of 6 out of 26 samples. No pesticides or PCBs were detected in the
onsite wells but pesticides were detected in 6 out of 26 offsite
wells.
Several metals were detected in the onsite monitoring wells at
concentrations up to 1300 ug/1. Chromium was detected above
Federal MCLs (50ug/ I) at 658 ug/l in onsi te well K5. Metals
detected in from offsi te monitor wells were found randomly
distributed around the site. Chromium and mercury exceeded Federal
MCLs at MW-AS and K-8 respectively. contaminants detected in the
onsite and offsite wells and their range of concentration can be
seen in Table 3. Also Figure 6 shows contaminants concentration
exceeding applicable or relevant and appropriate regulations
(ARARs) .
Fifteen residential wells surrounding the site including the one
onsite residential well were sampled. Wells RW-13 and RW-15 which
are located directly north of the site were considered background
wells. For the residential sampling, five VOCs, one BNA and three
metals were detected in the one onsite residential well RW-1. No
Federal MCLs were exceeded but the PA Water Supply criteria was
exceeded for tetrachloroethene and total 1,2 dichlororethene. No
VOCs were detected in the offsite residential wells. Three BNAs,
two pesticides and several metals were detected in offsite
residential wells but none which exceeded Federal MCLs, 40 CFR
141.11-141.16 or PA Water Supply Criteria, 25 PA Code ~~ 93.1
et. seq.. .
16
-------
REM v
GROUNDWATER CONTAMINANT CONCENTRATIONS
EXCEEDING ARARI
KEYSTONE SANITATION LANDFILL SIT!
LITTLESTOWN, PENNSYLVANIA
C.C.JOHNSON . MALHOTAA.P.C. 17
/
.
.
.
~.
",
MW-'$~
MW- ,D"
o
.
"W'OO (
MW . G"":
MW.GS~
KEYSTONE ~ANOFI~~
~.
MW-'~
",w."
MW-'O
~
1.a-o,.a1.,..,a.ft8
'" '01.'''' fteft8
'.","'1.''''~
u
"
n
l.a.Oi"l.'..~8ft8
Trical.,..u.e
,."."1.''''~
"",1 CIIl.,W8
.,
"
u
.
.
LINE "0.
1110-'
MO-'
II 0- 4
..2
..,8\1",
. to.S
......
Oi~.C'..IIR~8G8R8 4
lndeft..y,eM ,
1eft..c.-,),.",l8ft8 .
T.","I.r"'~
u
(
~..
...J~
,
"
"-
110ft. Ift'rl C..o., V,,"" IW",Y
cal,..1&. AND rlDDAI. C I'" ,
..IIIMY IICL8 V.I U'ID TO caDn
".11 rlOUUI
I
.
MW- CS .
IIW- ~l L-
"w".~:A- '..
. "
~ I
"W"DI8 \
.
I
.
( ALL VALUES IN ",/L)
SCALI
bot
\
I
O'
OA.TE
JULY 1"0
;...
... "
~
N
,
11
...
,
1 ..I.I-~'''I.ro8\'''''. at
,."..al.,.-caene 11
1.I-D'8fta.r..~ I
1 .1. a." i8lal.reftbMe 11:
~''''l.,''cMM I I
,.,....1.,..'11" J . I
......
l.l-0,..I.r..,II8ft8
1.1-0'.1.'"'''.'''
a.I.I."".II8,..,a8ft8
"'8911.'''''''''.
,-"..al.,..'''8ft8
,,,,,1 Qle,'"
u
a
140 '
II
!.
..
tC
,
I
. MO.'
. MO-'
. .. 0 -.,
-'
J:W!!R.
..11.- MONITORING Wh.~
.-IIO-MARYLANO
.- Ie -lClyaTON I
. .'HALLOW
I -INTIR MIO'ATI
o -OIl'
...-........-IUR'ACI WATE"
FIGU~e
6
-------
SUMMARY OF SITE RISKS
The purpose of the risk assessment performed for the Keystone Site
was to assess if actual. or threatened releases of hazardous
substances pose potential risks to exposed individuals under
current or possible future exposure circumstances. In order to
accomplish this, the risk assessment focused on the following: (1)
the contaminants detected during the RI; (2) the potential or
actual potential exposure pathways; (3) the potential or actual
human receptors; (4) the toxicity information on contaminants of
concern; and (5) the risk characterization information. The
following paragraphs will discuss these issues and provide a
summarized rational for the Selected Remedy.
contaminant Identification Information
At the Keystone Site, the groundwater, surface water, surface soil
and sediments were considered media to which a human population may
be exposed. Each of these media were analyzed for various organic
and inorganic contaminants. contaminants of concern were selected
for each of the media sampled based on the frequency of occurrence
of contaminants, concentrations found on-site relative to those
found offsite, and concentrations relative to background (primarily
for inorganic in soil, sediment, and groundwater).
The selected contaminants of concern for the surface soil medium
are listed below:
o
Inorganics:
silver.
antimony, manganese, mercury ,selenium, and
o
Volatile Organic Compunds (VOCs): 2-hexanone,
dichloroethane,1,2-dichloroethene(total), 1,1,1-
trichloroethane, trichloroethene, 4-methyl-2- penta none ,
and tetrachloroethene.
1,1-
o
Semi-volatile Organic Compounds (SVOCs): Benzoic acid,
dimethylphthalate, diethylphthalate, phenanthrene,
anthracene, di-n-butylphthalate, fluoranthene,
pyrene, butylbenzylphthalate, chrysene, di-n-
octylphthalate, benzo(b)fluoranthene,
benzo(k)fluoranthene, benzo(a)pyrene, indeno(1,2,3-
cd)pyrene, dibenz(a"h)anthracene, benzo(g,h,i)peryle~e.
and bis(2-ethylhexyl)phthalate.
o
Pesticides: dieldrin.
The selected contaminants of concern for the surface water med: -
are listed below:
18
-------
o
Inorganics: barium, chromium, copper,
mercury, selenium, vanadium, and zinc.
lead, manganese;
o
VOCs: 1,1 dichloroethane,
tetrachloroethene.
1,1,1- trichloroethane, and
The selected contaminants of concern for the sediments are
listed below:
o Inorganics: barium, beryllium, lead, manganese, and silver.
o
VOCs: methylene chloride, acetone, and 4-methyl-2-
pentanone.
o
SVOCs: bis(2ethylhexyl)phthalate.
The selected contaminants of concern for the groundwater
medium are listed below:
o
Inorganics: antimony, beryllium, barium, cadmium,
chromium, cobalt, lead, manganese, mercury, nickel,
selenium, and vanadium.
o
VOCs: vinyl chloride, chloroethane, acetone, carbon
disulfide, l,l-dichloroethane, l,l-dichloroethene,
I, 2-dichloroethene (total) , 1, I, I-trichloroethane,
trichloroethene, benzene, tetrachloroethene, and
dichlorofluoromethane.
o
SVOCs: benzoic acid, diethylphthalate, chrysene,
bis(2-ethylhexyl)phthalate, indeno(1,2,3-cd)pyrene,
dibenz (a, h) anthracene, and benzo(g,h,i)perylene.
o
Pesticides: aldrin, and 4,4'-DDT.
The selected contaminants of concern for the groundwater/
residential well medium are listed below:
o
Inorganics: antimony, beryllium, barium, cadmium,
chromium, copper, cobalt, lead, manganese, mercury,
nickel, selenium, vanadium, and zinc.
o
VOCs: vinyl chloride, chloroethane, acetone, benzene,
carbon disulfide, l,l-dichloroethene, 1,2-
dichloroethene(total), 1, 1, I-trichloroethane,
trichloroethene, tetrachloroethene, 4-methyl-2-pentanon~.
and dichlorofuoromethane.
o
SVOCs: benzoic acid, 2-chlorophenol, diethylphthalate,
chrysene, bis(2-ethylhexyl)phthalate,indeno(1,2,3-
cd)pyrene, di-n-butyphthalate, dimethylphthalate,
benzo(g,h,i)perylene, and dibenz(a,h)anthracene
19
-------
The contaminants of concern for each medium were evaluated
for the following criteria: frequency of detection; range of
sample quanti tation levels for nondetected analytes; range of
detected concentrations; background levels; 95% confidence mean
concentrations for onsite and affected by the site sampling
locations; maximum concentrations detected at onsite and affected
by the site sampling locations; and calculated exposures.
Considering all these evaluation criteria, the following list of
contaminants of concern were developed for the Keystone Site.
o
Inorganics: antimony, beryllium, barium, cadmium,
chromium, copper, cobalt, lead, manganese, mercury,
nickel, selenium, silver, vanadium, and zinc.
o
VOCS: 2-hexanone, vinyl chloride7 chloroethane,
methylene chloride, acetone, benzene, carbon disulfide,
1,1-dichloroethene, 1,1-dichloroethane, 1,2-
dichloroethene(total), 1,1,1-trichloroethane,
trichloroethene, tetrachloroethene, 4-methyl-2-pentanone,
and dichlorofluoromethane.
o
SVOCs: benzoic acid, 2-chlorophenol,diethylphthalate,
chrys~ne, bis (2-ethylhexyl) phthalate, phenanthrene,
anthracene, di-n-butylphthalate, fluoranthene,
dimethylphthalate, pyrene, butylbenzylphthalate, and
di-n-octylphthalate.
Exposure Pathways
Exposure pathways identified at the Keystone site have been
considered for current and future use populations and include the
following:
- Dermal contact and ingestion of surface soils and exposed
wastes~ .
- Inhalation of suspended particles generated from soils and
exposed wastes.
- Dermal contact and ingestion of sediments and surface
water.
- Inhalation during showering, dermal contact with, and
ingestion of.
The most likely potential exposed populations under current
and future conditions are Keystone employees, trepassers,
onsite remedial workers, and nearby residents. Since onsite
residents and workers experience greater exposure at the site, the
quantitative risk assessment was performed to evaluate onsite
adults.
20
-------
A reasonable maximum exposure (RME) was used to characterize
the site risks. It consists of. conservative estimates of
following input parameters;
the
o
Exposure concentrations based on calculated 95%
confidence mean.
o
Ingestion and inhalation rates on skin surface area
exposed.
o
Exposure frequency (events/year or hours/year).
Exposure duration.
Exposure averaging time.
o
o
G)
Meterological conditions conductive to generation of
windborne dusts and volatile organic vapors.
o
Volatility of contaminants for the scenario of using
contaminated for showering.
Using this RME approach, conservative estimates of the
subchronic and chronic daily intakes (CDI) were developed.
The CDI will assume a 70 year lifespan, a body weight of 70
kg., an exposure duration of 30 years, an exposure frequency
depending on the media (250-365 days/year), the exposure
concentration, and other factors depending on the model used
specifically to assess the dose by that specific medium (e.g.,
ingestion of soils, dermal contact of soils). More
information on the specific models can be found in the Risk
Assessment in the RI.
Toxicity Assessment
Using the CDI, a quantitative assessment for cancer risks and
quantitative measure of potential for adverse noncarcinogenic
health effects resulting from contaminant exposure at the Site is
derived. This is referred to as a toxicity assessment. In the
toxicity assessment, cancer potency factors (CPFs) were identified
for potential carcinogencic contaminants, and reference doses
(RfDs) were identified for chemicals exhibiting noncarcinogencic
effects.
CPFs have been developed by EPA's carcinogenic Assessment Group
for estimating excess lifetime cancer risks associated with
exposure to potentially carcinogencic chemicals. CPFs, which are
expressed in units of (mg/kg-day) -, are multiplied by the estimated
intake of a potential carcinogencic chemicals, in mg/kg-day, to
provide an upper-bound estimate of the excess lifetime cancer risks
associated with exposure at that intake level. The term "upper
bound" reflects the conservative estimate of the risk calculated
21
-------
from the CPF. Use of this approach makes an underestimation of the
cancer risk highly unlikely. CPFs are derived from the results of
human epidemiological studies for chronic::: animal bioassays to which
animal-to-human extrapolation and uncertainty factors have been
applied. .
Reference doses (RfDs) have been developed by EPA for indicating
the potential for adverse health effects from exposure to chemicals
exhibiting noncarcinogenic effects. RfDs, which are expressed in
units of mgjkgjday, are estimates of lifetime daily exposure levels
for humans, including sensitive individuals. Estimated intakes of
chemicals from environmental media (e.g., the amount of a chemical
ingested from contaminated drinking water) can be compared to the
RfD. RfDs are derived from human epidemiological studies or animal
studies to which uncertainty factors have been applied (e.g., to
account for the use of animal data to predict effects on humans.)
These uncertainty factors help ensure that the RfDs will not
underestimate the potential for adverse noncarcinogenic effects to
occur.
The CPFs and RfDs for the contaminants of concern are contained in
Tables 4 & 5.
For the Keystone Site, excess lifetime cancer risks are determined
by multiplying the daily intake level of chemical from
environmental media with the CPF. These risks are probabilities
that are generally expressed in scientific notation (e.g., lxlO-
6 or lE-06). An excess lifetime cancer risk of lE-6 indicates
that, as a plausible upper bound, an individual has a one in a
million chance of developing cancer as result of site-related
'exposure to a carcinogen over a 70-year lifetime under the specific
exposure conditions at a site. The estimated lifetime excess
cancer risks by exposure routes and environmental media are shown
in Table 6.
Potential concern for noncarcinogenic effects of a single
contaminant in a single medium is expressed as the hazard quotient
(HQ) (or the ratio of the estimated intake derived from the
reference dose). By adding the HQs for all contaminants within a
medium or across all media to which a given population may be
reasonably be exposed, the Hazard Index (HI) can be generated. The
HI provides a useful reference point for gauging the potential
significance of multiple contaminant exposure within a single
medium or across media. A HI less than one or equal to one
indicates there is no significant risk of adverse health effects.
The estimated HI or noncarcinogenic risk by exposure routes is
shown on Table 7 and by environmental media on Table 8.
The results of the estimated excess lifetime cancer risks indicates
that the total cancer risks for the Site is 5E-04. This value is
considered outside the target range of lE-04 to lE-06 which is used
by EPA for determining the need for a remedial action at the site.
22
-------
TABLE 4
CANCER POTENCY FACTORS
Ch_icil
C.rcine-
. genic Slope
Flctor
Cq-)
Ckg-cUlIIII)
....-..................... ...........
.erylll...
ClChi U8
01 rOIl 8 &8
L-
IUcket
4.3E+00
NA
. NA
4.0E-02
NA
......................................
1,1-0Ichloroethlne
1,1-0icftloroethene
lena..
Methylene chloride
Tetrachloroethene
Trichloroethene
Vinyl QLori.
9.11-02
6.01-01
2.91-02
7.51-03
5.11-02
1_11-02
2.31.00
......................................
lenzo< I)Pyr.,.
ienzO
-------
TABLE 5
CHRONI~ REFERENCE DOSES
Cont.- rwnt of Cone.""
...........................
Ant i.""
'Irh.-
..ryll h.-
CacIIIt YI
Chta8t YI
CobaLt
C~r
L..
...".....
Mercury .
Ntckel
S.l8fttYi
SiLver
vlftldt YI
ztnc
Chronic
. ..f.renc:. .
D.. C 110)
C../k,.dIy)
........... .
U
0.0001
U
U
U
U
U
U
o .OOCD
U
U
U
U
U
U
-.........................................
1,1,1.Trtch1oroeth8n8
1,1-oteh10r0ethan8
1,1.0teh10r0eth8n8
1,Z.Oteh1ortethen8 (totl1)
Z........
,._tltyt .,-pent.,.,.
Ac.d-
.ena..
Ca"*, CUIUUt.
Chl0r0et"-
Otch10r0dtf1uor088thane
MethyLene en\ort.
T.tr8Ch10r0eth8n8
Trtch10r0et"-
"f",C. Cfttort- .
0.3
0.1
IA
IA
IA
0.02
IA
u
IA
.
O.GS
0.857
IA
IIA
IIA
...-............-.._....-..~.............
2-08.."'101
Ant""'''
IenIOtc Actd
leftIO(a)~."
leNoC It) f \uorlftthIN
l8nIoCI,..,t)peryLen8
l8nIoCk)f1uorlftt"..
1t'8CZ-lthy1heay1)phthl1It.
1uty1b8n1y1phthallt.
CIt~..
D'~C.,,,)~"rlCene
Dt.t~1phth81at8
Dt..thytphthatat8
D.-~lutytph~t.t.
Dt-n-octytphthatat.
'1 YOrantMn8
I ndIM( 1 ,2,3-cd)'Yr"
"'8ft8ftth....
,."...
u
U
IIA
IA
IIA
IIA
IIA
U
IIA
IIA
IIA
IIA
IIA
IIA
IIA
IIA
IIA
IIA
IIA
............................ .............
4,4'.DOT
ALdrtn
Dt",,'n
IIA
IA
IA
-------
TOTAL CARCINOGENIC 1USK BY EXPOSVRJ RO\JTE (ADULTS)
KEYSTONE SrrE, PENNSYLVANIA
................._................._..........~........-.......................-..................-..
. L I fl' ,.. 1_' C:..... ."1 'e'c~ Coftc't~len to
CDI--IorIt.,) Llfeet.. I.c... C8fte.~ I.e.
-.................................. ...................................
Ch.fcat 1"""1- CGllDI I"I8CII IMat.- COI8tMd
Ol'lt 0'".' elorl Tot" Of'., OI".L tfen Totl\
........................... ....-... ........ -....... ........ ...----.
........ ........ ........
'.ryLL 'WI .-05 91.09 11-08 61-05 11.721 0.001 0.001 11.731
CIOII".- ItA U 21.01 21.01 IIA U 0.001 0.001
CtI... h.- ItA ItA 41.06 4£.06 U -. 0."1 0.7"
LI'" 51.06 11.09 IIA 11.06 0.'" 0.001 IA 0.'"
IIlck.L ItA IIA 21.01 21.07 Ii IIA 0.041 0.04'
............---..................... ........ ........ ........ ........ ........ ........ ........
1,'.Of~'Of08thene 11-01 31.08 IIA "-OS 2.6\1 0.0'1 IA 2.611
1,1.DfchLorOtth~ ".115 81.08 11.115 11.0. 7.361 0.011 14.021 Z'.S91
181"1"" 21-06 Sf.09 21.06 31.06 0.121 0.001 0.301 0.'21
MethyLene ~lOl'lda 01.00 SI.'O 01.00 SI.10 IIA 0.001 IIA O.COS
Tltr8CftlorOtt~1n8 6£-06 ZI.OI 41-07 61.06 1.011 0.001 0.071 1.151
T r' cIt lOl'08tlMna ".06 11-09 21.06 11-06 0.221 0.001 0.»1 0.561
vt~1 CMOI'lda JI-CM. 51-07' Sf-OS SI-o. 51. 2ft 0.101 6.271 58.611
.....---............................ ........ .......- ........ ........ ........ ........
........
8tMoCI)'Vrena NA ItA IIA IIA IA IIA IA IIA
l~oCb)'\wo'lftthene ItA ItA Ii MA U ItA IIA ItA
1tft100..~,I)'eryL8ft8 IIA IIA IA ItA U IIA NA IIA
len.oCk)Plwol'lftthene Ii IIA IIA ItA IIA IIA IIA ItA
bl.(Z.EehyLheayl)~thal.t. 91.07 51.07 IIA 11.06 O. ,.. 0.101 Ii 0.261
Chrysene MA IIA IIA IIA IIA IIA ItA IIA
D'benIC8,h)Aftthrec8ft8 IIA IIA IlIA IIA IIA IIA ItA 8IA
1~1,Z.S-~),,~ IIA U IIA IIA 8IA U ItA IIA
"""",'hUM IIA 8IA IIA IIA IIA IIA IIA IIA
.................................... ........ ........ ........ ...-.... ........ ........ ........
'," .DOT 41-Of 1..10 21-08 41.07 0.011 0.001 0.001 0.-
'Ld~'" ".06 91.01 JI-" ..06 '.191 0.001 0.011 1.'61
D..L.f" 71-08 ".07 61-09 "-01 0.0'1 0.121 0.001 O. In
........................... ........ ........ ........ ........ ........ ........ ........
........
TOT Ai. ".06 ..06 1..06 51-06 ".IA. O.HI " ..,. 100.001
[[[~............
TOTALCBRONIC CARCINOGENIC RISK BY ENVIRONMlNrAL
MEDIA (ADVLTS)
TABLE 6
L U" ,. ..... C8N:.. ..,..
CDf.....,,,,,..,)
[[[
"ream cen'l'flllut'''' to
L.fte'f. be... ~.r "tel
............................................
............................................
........
........
Sutt.. 0,..... Cc8l~
W8t'" IMIII' Tou\
...........................
........
........
........
IUrfIC8 ....... CGIIIbtNd Sutf1C8
MIter w'''' Tote' 101\
........
~.fc:.'
IUrfece
ioU
...,-
8M
........
........
"1')'\ 1 fill
CadII' WI
CI\'-'~
Ltad
,
ft-,.
M
M
11.10
IA
M.as
a-OI
41-06
...
a-"
..-.
II..
....
.-06
8-01
..
IA
II
..
51-"
M
IIA
IIA
11-08
IIA
........ ........
..., .
8M
IA
M
M
IA
0_001
0.-
M
IIA
0.-
III
IA
IA
IA
0.001
M
1\.411
0.001
O. "61
0.511
0.041
" .411
0.001
0.761
0.57'1
0.01.1
[[[
............................................
1,'.DfChLorOtth8n8
1,'.O'chlOl'OIeh8n8
hNlM
"'t"Y"- cM",f.
T.tI'ICh\or..thene
T I'f ctlloroet"ene
v.~\ 01\..,.
3I.Ot
M
U
IIA
41-.
'..10
IA
"-01
IA
IA
IIA
71-01
IIA
IIA
tI.as
11.06
K-Q6
IA
,,-.
..06
..04
11..
11..
JI-06
sa.10
".06
.-06
.-06
IIA
IIA
IA
11-10
U
IIA
IA
[[[
lenIo(8)'Yr"
ltnIo
-------
TOTAL CHRONIC NONCARCINOGENIC RISK BY EXPOSURE ROUTE (ADULTS)
KEYSTONE SITE. P1!:.NNSYL VANIA
.TADLE 7
fJLlr C!YIUM1.WI'
DATI, 09/01190
CUM.ATIW .111 IT EX'OSUAt .auT1l . ADULT CN.OIue IIAZAID IQIX
.-..........-.-.--........-.-.---.-................-.-.........................................------
Milira J"" . Chranl C AGAt I '.re."t Contrftlwtfon to
(Dt...ton'...) ICI..rd I,.... . Ctll'OtItc AOJ\t
.................-................. -.-.--...................-.........
INlltl' CGlDtnec8 Im.t.. Contt nacI
Ch..fcal Ol'll D.,.l Uon TOI.' Ol'll D....L tton TotaL
........................... ..---... ........ ..-..... ........ ........ ........ ........ -........
AnU...,. 2.900 0.000 ItA 2.900 St.... 0.0'1 IA 59.691
a'l'h. O.J17 O.GOO 0.0\9 0.S96 ., . 761 0.001 0.181 '.U.I
..,.."l L h. 0.D07 0.000 I8A O.GO? 0.141 0.001 ItA o. ",I
C8d1af 108 0 . '" 0.000 ItA 0.141 2.- 0.001 ItA t...
CtlI'.flo8 0.466 0.000 I8A 0.46' 9.SSI 0.001 IIA 9.551
CobiLt I" IIA ItA I8A ... ItA IIA IIA
I Copper O.U' 0.000 u 0.111 2.6ft 0.001 I8A 2.691
LMd I8A I8A IIA ItA ItA tIA U 8IA
M."..... O. '" 0.000 0.012 0.201 S.taI 0.001 0.261 4.111
MercYl'y 0.029 0.000 ItA 0.029 0.'91 0.001 11& 0.591
Mfck.' 0.281 0.000 ItA 0.Z81 '.791 0.001 IIA '.191
S.'lfth. 0.025 0.000 IIA 0.021 0.121 0.001 IIA 0.521
IUwl' 0.000 0.000 ItA 0.000 0.001 0.001 ItA 0.001
VeNd". 0.013 0.000 IIA 0.- 1.111 0.001 I8A 1.111 :
ZIM O.CWOO 0.000 tIA O.M 0.121 0.001 u 0.821
..........-------.............-..... .......- ........ ........ ........ ........ ........ ........
1,1,1.TI'I~Loroe~ent O.OOJ 0.000 0.001 O.OCIS 0.061 0.001 0.021 0.071
1,1'D'ch\oroethent 0.004 0.000 0.- 0.- 0.- 0.001 0.011 0.'"
1,"Of~loroet~ 0.017 0.000 IIA 0.011 0.161 0.001 IIA 0.561
'.2.DI~tol'oel~ (ToI.') O.oao 0.000 IIA 0.010 0.411 0.001 ItA 0.421
Z'MO"" 11& tIA tIA IIA tIA tIA IA IIA
4-Me~yl.Z.'lftt8nan8 .0.002 0.000 0.004 0.- 0.061 0.001 0.091 0.'21
AC8t- D.DOS 0.000 ItA 0.005 0.1. 0.001 ... 0.101
""1- IIA IIA IIA IA IIA IIA IIA IA
C8rbon O'au'f'" 0.001 0.000 IIA 0.001 0.8 0.001 IIA 0.031
Ch \0N8t"- ItA 11& ItA 11& u IIA MA 11&
Dtch\orod'f'OUr'OI8tft- 0.001 0.000 0." 0.- 0.021 0.001 0.091 0.'"
MethyLent Ch\ol"" 0.000 0.000 0.000 0.000 IA 0.001 IA 0.001
'oc'achLo~thefte 0.027 0.000 IIA 0.017 0.561 0.001 IIA 0.561
T'IcfIlo~CMN ItA IIA IIA IIA IIA IIA ItA ItA
Yfft¥L CMol'f. ItA IA IIA IA ItA IIA tIA ItA
.............-...................... ....-... ........ ........ ........ ........ ........ ........
a.Chlol'..,.l 0.- 0.001 IA 0.- O.OD 0.. tIA 0.-
""thl'lCent tIA .. ItA .. ItA IIA IA IA
I""ole Acl~ 0.000 0.000 IIA 0.- 0,. 0.001 ItA 0.001
hNo(I)PyI'- IIA ItA IIA IA IIA ItA ItA 11&
lenzo(b)f h_ent'" IA IIA tIA ItA IIA M ItA ItA
IInIOCI,h.t)Plrylene .. IA IIA IA .. IIA IIA tIA
I""OCk)'h_-- IIA tIA IIA ... .. IA IIA MA
b'.(2..~~~~llt. 0.007 0.- IIA 0.011 0.11S 0.091 u 0.241
Iwtyllllftay '.C. 0.000 0.000 u 0.000 0.001 0.001 ItA 0.001
Ch~ IIA IIA IIA IIA IA IIA ItA I8A
D f MNCI,""""""- tIA ItA U IIA IA ItA U ItA
Dfothyl~~'~ 0.000 0.000 IIA 0.000 0.- 0.001 ItA 0.001
Of.thy' "'1M'''' IIA ItA IIA U M IA IIA ItA
D'.~lwtyt~th.l.c. 0.- 0.001 IA 0.001 O.~I 0.021 IA 0.061
DI-n.Octyl "thll.,. M ItA IIA tIA IA ItA tIA U
, \uorlfttMne u ItA IIA IIA tIA IIA ItA IIA
1nden8(1.I.J.od)~ IA tIA IA 1M tIA IIA tIA IA
fItIlNfttItr- IIA IIA ItA IA IIA tIA IIA IIA
""'8ft8 tIA tIA U IA IIA U IIA U
........................-.-......... ........ ........ ........ ........ ...-.... ........ ........
4,4'.ooT 0.- 0.000 U 0.006 O. "1 0.001 IIA 0.'"
AL.I" 0.. 0.000 tIA O.ISS 0.721 0.001 IIA 0.721
DI.hlr'" 0.000 0.- ItA 0.- 0.001 O.OU - ItA 0.041
.............-............- ........ ........ ........ ........ ........ ........ ........ ........
TOTAL 4." 0.010 0.066 4.15f ".191 0.201 0.911 '00.001
.....................-.-.-...............................-.................-.-.......................
26
-------
TOTAL CHRONIC NONCARCINOCENIC RISK BY ENVIRONMENTAL
MEDIA (ADULTS)
KEYSTONE SITE, PENNSYLVANIA
TABLE 8
FILEl
DAT! i
09/02/90
ftJK IT CMVUMMMUTAL NOIA
AOULT CMONIC NA2A*0 INOfX
Cheafcal
Antlnony
larfui
•eryl I lu»
C»c*fu»
ChrealuB
Cobalt
Capper
Load
"•nfoneao
Mercury
Nickel
selenlua
Stiver
venedlujR
2fne
1,1.1*TrtcMoroethano
1,1-01eMoreetMne
1,1-Oiaileroetheno
1,2-Ofemereethene (Total)
~Ms»anone
•ethyl •2*Mntanom
TfStt
lontono
Carbon olowlffdo
CMoroattiane
OlehlarodlfleuroMthant
Methylene Chloride
TotracM oroethane
Trlcflloroethone
vinyl CMorlde
2-Chlerophenol
Anthrecan*
•eniefe Acid
Sent o< a Gyrene
lento< b) 1 1 uoranthena
•arueif.h, 1 )*erylene
SenioriuorantfcefNi
SutylbansylptithalaM
Chrytene
Olbeni
-------
The primary cancer risk comes from the ingestion of groundwater,
and can be largely attributed to the presense of vinyl chloride
detected at the value of 40 ug/l in well K-1 and a value of 4 ug/l
at well HS. Both of these wells are onsite wells.
The results of the noncarcinogenic indices present a chronic Hazard
Index for the Site of 4.862. This value exceeds the accepted value
of one or less than one. The vaiue of the HI is attributed to the
presence of antimony, barium, manganese and chromium in the
groundwater. These contaminants pose a threat to human health
through either ingestion of the groundwater or inhalation of an
aerosolized mist of groundwater.
In summary, both the Lifetime Excess Cancer Risks (5E-04) and the
Hazard Index (4.862) exceeded EPA's accepted values of 1E-04 to
1E-06 and less than 1 respectively. As indicated by this, in its
current condition, the site presents an imminent and substantial
endangerment to public health as set forth in Section 106 of
CERCLA, 42 U.S.C. ~ 9606. Therefore, a remedial action is required
to reduce the risks to human health.
Environmental Risks
An environmental risk assessment was conducted in the RI to assess
the potential impacts to nonhuman receptors with the contaminants
of concern at ~he Keystone Site. Potentially exposed populations
(receptors) were identified and combined with information on
exposure and toxicity on the contaminants of concern from the Site
to derive the estimates of impact.
The potential receptors identified at the Site included terrestial
plants, terrestial wildlife and aquatic life. For each of these
receptors, exposure pathways were examined and are as follows:
Potential RecDtors EXDosure Pathwavs
Terrestial Plants Uptake from surface soils
Terrestial Wildlife
Aquatic Life
Ingestion of surface soils
Ingestion of food
Direct contact with water
Direct contact with sediments
Ingestion of sediments
Ingestion of food
For the terrestial plant and wildlife, it was determined that t~,P
exposures to the contaminants of concern at the Site would not ha'..'
a significant impact.
For the fish and aquatic invertebraes at the Site, surface wa~. '
toxici ty values were available from the Ambient Water Qua 1 ~ .
28
-------
criteria (AWQC) and the Lowes;t Observable Effect Concentration
(LOEC). The AWQCestablishes criteria for chromium, copper, lead,
mercury, selenium and zinc. The LOEC is defined as the lowest
concentration associated with a toxic effect and is used in the
abs~nce of AWQC for barium~ manganese, 1,1,1 trichloroethane, and
tetrachloroethene. For the Keystone Site, chromium, copper,
cyanide, lead, zinc and mercury exceeded the AWQC. Of these
contaminants, mercury was ident~fied as a possible concern due to
its ability to bioaccumulate in fathead minnow. However, this
bioaccumulation is associated with methylmercury and not inorganic
mercury. As we do not currently know the speciation of mercury at
the Site, it is not possible to state whethe~ these surface water
concentrations are adversely impacting aquatic life at the Site.
For the amphibians at the Site, limited data was available on metal
toxicity to amphibian species. Therefore, the assessment
was restricted to VOCs and amphibians. From this assessment, it
was determined that only concentrations of 1,1,1 trichloroethane
exceeded the 1% mortality rate for frogs. The 1% mortality rate
would be considered significant if frogs were an endangered species
or have inherently low reproductive success. No endangered species
of frogs have been identified at the Site nor has there been an
indication of low reproductive success among frogs. Therefore, it
can be concluded that adverse impacts to amphibian species at the
Keystone Site are unlikely.
In summary, the environmental assessment determined the following:
o
No adverse impact was determined to exist for the terrestial
wildlife and terrestial plant populations at the site from
the contaminants of concern.
o
AWQC was exceeded for chromium, copper, cyanide, leatl, zinc
and mercury. The presence of mercury in the form of
methylmercury may adversely impact the aquatic life on the
surface water. .
o
Additional sampling will be conducted to determine the
speciation of mercury and monitor the levels of contaminants
in the surface water and sediments.
DESCRIPTION OF ALTERNATIVES
The remedial objectives of the Keystone Site are to prevent curre~~
and future exposure to the contaminated soils, to redu~~
contaminate migration into the groundwater and to prevent migrati~~
of contaminated groundwater to uncontaminated area. To addres:
these objectives, remedial alternatives were developed t~~
contaminant source control and groundwater remediation.
Under the statutory requirements of CERCLA section 121, 42 U.s.
9 9621, the alternatives must be protective of human health ~.
29
-------
the environment, comply with applicable or relevant and appropriate
requirements and be cost effective. It is also required that
permanent solutions and alternative treatment technologies or
resource recovery technologies be considered to the maximum extent
practicable. This is to satisfy EPAs preference for treatment that
reduces toxicity, mobility and volume of the site related
contaminants.
At the Keystone Site, certain technologies and p~ocess options were
screened out due to the waste characteristics and Site conditions
which made them technically and/or economically infeasible. The
Keystone Site consists of approximately 1.7 million cubic yards of
nonhomogeneous waste. Technologies such as biological treatment,
sOlidification/stabilization, soil vapor extraction were not
technically feasible for the nonhomogeneous waste content. Other
technologies such as thermal treatment and offsite disposal were
considered not economically feasible.
The Feasibilty Study should be referred to for further discussion
on the technical impracticability of treatment technologies to this
Site.
Five alternatives were given final consideration. With the
exception of the No Action alternative required by the NCP
and the Limited Action alternative, these alternatives all consider
groundwater extraction and treatment. Two of the alternatives also
consider containment of the waste. Pursuant to CERCLA 121(c) 42
U.S.C. ~ 9621(c), all alternatives will include a review every five
years as wastes are being left onsite.
In the following discussion of the alternatives, the costs are
presented as present worth costs for a period of thirty years with
a five percent discount. The present worth costs consist of
construction and operation and maintenance (O&M). Comparison of
the costs is included in Table 9.. .
Alternative 1 - No Action
Under the National oil and Hazardous Substances Contingency Plan,
40 CFR ~ 300.430(e) (6) 55 Fed. Req. 8849 (March 8, 1990), the no
action alternative must be considered as a baseline action against
which all other alternatives are compared. It provides no remedial
action and no reduction of risk posed by the surifical soils and
grounswater at the Keystone site. Because wastes are being left
onsite, a five year effectiveness review would be required. The
only costs associated with this alternative is the present worth
cost for the five year review which is $20,900. The no action
alternative would not comply with the ARARs for the site.
Alternative 2 - Limited Action
The Limited Action alternative provides no remedial action and no
reduction of the risks posed by surficial soi1 and groundwater at
30
-------
the Keystone site. Only one remedial action objective (to prevent
direct contact with surface soils having contaminant concentrations
in excess of accepted background levels, or which would pose an
excess cancer risk) would be attained by this alternative.
This alternative.would provide the minimum measures necessary to
limit exposures of nearby residents to site contaminants. A fence
would be installed at the perimeter of the property to physically
restrict access and subsequently limit the potential for exposure
to on-site .contaminants. No measures limiting exposure to
contaminants beyond the perimeter fence, or off-site, would be
included. A galvanized steel, chain-link fence approximately 6,800
feet long would be constructed around the landfill.
The Limited Action alternative also includes monitoring (for an
assumed period of 30 years, per OSWER Directive 9355.3-01) to
determine variations in future contaminant migration. Also,
restrictions on future property use, which would include aquifer
use limitations and construction restrictions would be enacted. A
point of use treatment unit would be provided to the onsite
residents.
The present worth cost associated with this alternative is
$720,000. This cost was incorrectly stated in the Proposed Plan
as $719,000. These costs are for fencing, a sign, a water treatment
unit, long term monitoring (30 years) and a five year review. The
Limited Action alternative would not comply with applicable state
and Federal ARARs pertaining to landfill closure and groundwater
contaminant levels.
Alternative 3 - Groundwater Treatment
Alternative 3 provides for groundwater remediation in the form of
groundwater collection, treatment and discharge. Groundwater
treatment would meet the objectives of preventing contaminated
groundwater from migrating offsite and to reduce contaminant levels
contained in the migrating plume to an established cleanup goal.
This alternative includes extraction and treatment of groundwater
from the unconsolidated soil and saprolith zone located above
bedrock. Because of the lack of extractable water, no groundwater
would be extracted from within the bedrock. The thickness of the
groundwater aquifer above bedrock is estimated to vary from 5 to
15 feet. The point of compliance would be the area of attainment
which would be slightly downgradient of the landfill itself (See
Figure 7). The area of attainment may change during the Remedial
Design phase.
A total of 20 extraction wells would be used to collect
groundwater. The wells would be spaced about 200 feet apart along
the south, east and north edges of the landfill and would capture
contaminated groundwater originating under the landfill. The wells
31
-------
SITE
•OUNDAftf
ESTIMATED
GftOUNDWATEft
ATTAINMCKT
AftCA
•GAOUNOWATEN
COMTOUftS
(DASHED
WMEME
APMIOXIMATE
SOUftCf ' OCC 6ROUNOWATEM CONfOUM
, FCBftUAJtY
SCALE
tf 900'
REM V
AREA OF ATTAINMENT
DATE
JULY 199O
KEYSTONE SANITATION LANDFILL SITE
LITTLESTOWN, PENNSYLVANIA
C.C.JOHNSON A MALHOTRA.F.C.
32
-------
would be spaced closer together in the low-lying areas (gullies)
and further apart in the higher'elevations of .the Site. The exact
number and placement of the wells will be determined in the
Remedial Design phase. The perimeter of the landfill is the most
efficient location for the wells, since the groundwater moves from
the center to the edge of landfill, and the extraction wells would
only enhance this radial flow. .
Groundwater would be pumped and treated until adequate aquifer
flushing has occurred to reduce groundwater contamination to meet
the remediation goals for groundwater at the Site. These goals
are the Federal MCls, or Non-zero Maximum Contaminant Levels Goals
(MCLGs) as specified in the Safe Drinking Water Act 42 U.S.C. 9
300f et. sea. (SDWA) and/or the Pennsylvania "background" ARAR
specified in 25 PA Code 99 264.90 - 264.100 in particular, by 25
PA Code ~~ 264.97(i) (j) and 264.100(a) (9) which requires
groundwater to be remediated to background, whichever are lower.
The Commonwealth of Pennsylvania also maintains that the
requirement to remediate to background is found in other legal
authorities. See" The Selected Remedy" for further discussion.
Actual treatment time required would be determined by monitoring
contaminant reductions after the treatment begins. An estimated
cleanup time to meet PA Water Supply Criteria is 46 years. No
estimate of time has been made to reach background levels.
The treatment associated with the groundwater pumping would remove
volatile organic compounds and metals. It may consist of
equal ization, flocculation/precipitation , filtration, ion exchange,
air stripping and a filter press. Any sludge produced would be
disposed in an appropriate offsite disposal facility in accordance
with 40 CFR 268.1 - 268.50, 25 PA Code ~~ 75.259-75.282 and 25 PA
Code 75.21-75.38. Final selection of the process units would be
determined during the Remedial Design phase.
The groundwater pump and treat alternative (#3) would be capable
of producing an effluent which meets PA Water Quality Standards
established in 25 PA Code ~ 9 93.1 et. sea., the Federal Water
Quality criteria established in 51 Fed. Rea. 43665 and the
substantive requirements of a PA National Pollution Discharge
Elimination System (NPDES) permit set forth in 25 PA Code 9~ 92.1
et. sea. Emissions from the air stripper would meet the
requirements of the Clean Air Act (Part D) (42 USC sections 7401-
7642) and OWSER Directive 9355.0~28. It would also be in
accordance with Pennsylvania Air Pollution Control Act, 25 PA Code
99 127.1 et.sea. that requires best available technology be
implemented to minimize emissions. As this alternative does not
provide the required closure for the municipal landfills, it would
not comply with PA Municipal Waste Regulations, 25 PA Code 99
273.234 for municipal landfill closure.
Other components of this Alternative 3 include: the installation
of fence to limit access and prevent exposure of nearby residents:
33
-------
institutionals control to restrict placement of wells in the area
to prevent ingestion by humans and interference with the efficiency
of the groundwater extraction system; deed restrictions to prevent
exposure by residents; long term groundwater monitoring and supply
of treated water to onsite resident. The implementation time would
be four months for the construction of the treatment plant and the
installation of the extraction wells and collection system. A five
year effectiveness review would also be conducted as wastes are
being left onsite. The estimated present worth cost for the
Alternative 3 for 30 years of groundwater extraction and treatment
is $3,937,000. The Proposed Plan incorrectly stated the present
worth cost as $4,234,000.
Alternative 4 - Groundwater Treatment and Impermeable Cap
Alternative 4 includes groundwater collection, treatment, and
discharge as described for Alternative 3 in addition to an
impermeable cap. It will also include fencing, groundwater
monitoring, and the institutional/administrative control of
Alternative 2. Groundwater would be pumped and treated until
adequate aquifer flushing has occurred to reduce groundwater
contamination to meet the remediation goals for groundwater at the
Site. These goals are the Federal MCls or Non-zero MCLGs as
specified in the Safe Drinking Water Act, U.S.C. 300f eta sea.
(SDWA) and/or the pennsylvania "background" ARAR specified in 25
PA Code ~~ 264.90 - 264.100 in particular, by 25 PA Code ~~
264.97(i), (j) and 264.100(a) (9) which requires groundwater to be
remediated to background, whichever are lower. The Commonwealth
of Pennsylvania also maintains that the requirement to remediate
to background is found in other legal authorities. See "The
Selected Remedy" for further discussion. Actual treatment time
required would be determined by monitoring contaminant reductions
after the treatment begins. An estimated cleanup time to meet PA
Water Supply Criteria is 46 years. No estimate of time has been
made to reach background levels.
The point of compliance would be the area of attainment
demonstrated in Figure 7. This area of attainment may change
during the Remedial Design phase. An impermeable cap would provide
containment of onsite waste and soils. The migration of
contaminants from the wastes to groundwater would be reduced by
controlling infiltration and reducing leachate generation.
Potential exposures to contaminated surface soils through
inhalation, ingestion, and dermal contact also would be reduced.
A plan of Alternative 4 is shown in Figure 8.
As part of this alternative, surficial soils in the spray
irrigation area containing VOCs would be excavated and deposited
on top of the landfill where they would be covered by the
impermeable cap. The area of these soils is approximately 2.6
acres and would be excavated to a depth of 1 foot, yielding a
volume of approximately 4,150 cubic yards. Upon completion of the
34
-------
SITE
BOUNDARY
GROUND WATER
TREATMENT
PLANT
EXTENT
or CAP
EXTRACTION
WELL/
SOILS
EXCAVATED
FROM SPRAY
IRRIGATION
AREA
SCALE
Cf 500'
DATE
JUNE 1990
REM V
ALTERNATIVE 4 - PLAN VIEW
KEYSTONE SANITATION LANDFILL SITE
UITTLESTOWN, PENNSYLVANIA
FIGURE
8
C.C.JOHNSON A MALHOTRA.P.C.
35
-------
cap, the exposure pathways for' surface and subsurface soils, and
buried wastes would be reduced.
The impermeable cap would be designed to conform to the cover
requirements of 25 PA Code ~~ 273.234 pertaining to Muncipal
Landfill closure. The total area of the cap, including side
slopes, would be approximately 40 acres and would cover a volume
of 1.7 million cubic yards of wastes. The side slopes of the cap
would range from 3 to 15 percent in order to conform with the
existing contours and meet the requirements of 25 PA Code ~ 9
273.234. Depressions and erosion gullies on the landfill would be
restored to grade with surrounding soil, as necessary. The top
layer (and each layer beneath) of the cap would have a slope of at
least 3 percent to enhance surface water runoff. The cap would
require regrading to meet slope requirements and drainage ditches
to collect and transfer runoff away from the landfill. Leachate
generation will be reduced due to a reduction in precipitation
infiltration. A vegetative cover would be placed on top.
An active gas extraction system designed to conform to the
requirements of 25 PA Code ~~ 273.292 would be installed between
the existing landfill and the cap to prevent build up of methane
gas. Gas collection wells installed to the depth of 20 feet and
connected to a blower would collect the gas and send them to a gas
burner. This will be further defined in the Remedial Design phase.
Any sludge produced from the unit processes would be disposed in
an appropriate offsite disposal in compliance with 40 CFR 268.1-
268.50.
A five year
onsite. The
wells and a
months.
review would be conducted as wastes are being left
implementation time to install groundwater extraction
treatment plant plus an impermeable cap would be 14
Alternative 4 would comply with the Federal and state ARAR's for
municipal landfill closure (25 PA Code 99273.234), air emissions
(Clean Air Act (Part D) (42 USC 9!P401-7642); OWSER Directive
9355.0-28; 25 PA Code gg127. et. sea.), surface water discharge
(25 PA Code gg 93.1 et. sea. ; 25 PA Code 9992.1 et. sea. ; 51
CFR 43665), residual sludges (40 CFR 2681.1-268.50; 25 PA Code ~9
75.259-75.282; 25 PA Code 9975.21-75.38), and the "background
quality" requirement for groundwater in 25 PA Code 99 264.90-
264.100 or the MCLs or Non-zero MCLs, whichever is lower. See "The
Selected Remedy" for further discussion. The estimated present
worth cost for Alternative 4 consisting of 30 years of groundwater
treatment and an impeameable cap is $9,156,950. The Proposed Plan
incorrectly stated the present worth cost as $11,215,000.
36
-------
Alternative 5 - Groundwater Treatment and MUltimedia Cap
Alternative 5 provides the fencing, groundwater monitoring and
administrative/institutional controls of Alternative 2. It also
implements and satisfies the previously stated ARARs for the
groundwater collection treatment and discharge of Alternative 3,
plus adding a multimedia Cap. Groundwater extraction and treatment
. would reduce contaminant concentrations to "background quality" as
required in 25 PA Code 99 264.90 ~ 264.100 and in particular, by
99264.97(i), (j) and 264.100(a) (9) and/or the MCLs or NON-zero MCLGs
as specified in the Safe Drinking Water Act, 42 U.S.C. 300f et.
~ (SDWA), whichever are lower. See "The Selected Remedy" for
further STET discussion. The point of compliance would be the area
of attainment demonstrated in Figure 7. This area of attainment
may change during the Remedial Design phase. A multimedia cap
would be used to. provide containment of onsite wastes and soils.
Contaminant migration would be prevented by eliminating
infil tration and minimizing leachate generation. The synthetic
membrane component of the multimedia cap is capable of reducing the
infiltration by nearly 90 percent. This reduction in infiltration
would have a direct effect on reducing leachate generation and the
amount of extractable groundwater for treatment. Potential
exposures to contaminated surface and subsurface soil and wastes
through inhalation, ingestion, and dermal contact would be reduced.
T
The multimedia cap would have a total area of 40 acres and cover
a volume of 1. 7 million cubic yards. It would be designed in
accordance with 25 PA Code 99 273.234 for Muncipal Landfill
closure. The side slopes of the cap would range from 3 to 5
percent and would require regrading. A gas collection system would
be placed between the existing landfill and the multimedia cap.
The cap would consist of a two foot layer of clay overlain by a 40-
mil thick, low density, pOlyethylene synthetic membrane. Next
would be a 1 foot thick granular drainage layer followed by a
filter fabric. On top of the filter fabric, an 18 inch thick
common borrow zone would be placed to protect the underlying layers
and provide a shallow root zone. A 6 inch topsoil zone with
vegetation would be the top layer. All layers of the cap would be
sloped at a minimum of 3 percent to drain surface water.
All previously stated ARARs for Alternative 4 for municipal
landfill closure, air emmisions, surface water discharge,
residual sludges, and water quality would apply and be satisfied
by Alternative 5. The treatment time is assumed to be 30 years in
length. However, with the reduction of treatable volume, the
treatment time may be reduced. Actual treatment time would be
determined by monitoring the groundwater contaminant reduction
after treatment begins. The implementation time to install
groundwater extraction wells and a treatment plant plus an
mul timedia cap would be 14 months. A five year effectiveness
review would be conducted to determine the effectiveness of the
remedy and the need for continuing treatment.
37
-------
The estimated present worth cost for 30 years of groundwater
treatment and a multimedia cap contained in Alternative 5 is
$10,695,000. The Proposed Plan incorrectly stated the present
worth cost as $12,060,000.
COMPARATIVE ANALYSIS OF ALTERNATIVES
Pursuant to 40 C.F.R. Section 300.430 (e) (9) (iii), 55 Fed. Req.
8849, an assessment of the individual alternatives against nine
evaluation criteria specified in this regulation shall be made.
These nine criteria have been categorized into three groups to
weight their importance. The first group is threshold criteria of
which each alterative must meet to be eligible for selection. The
threshold criteria consists of overall protectiveness of human
health and the environment and compliance with ARARs. The second
group is the primary balancing criteria. This is used to evaluate
the performance of each of the alternatives compared to the others.
These criteria include long term effectiveness and permanence,
reduction of toxicity, mobility and volume, short term
effectiveness, implementability and cost. The third group is the
modifying criteria which includes state and community acceptance
that must be considered for the selection of the remedy.
The assessment of the five alternatives against the nine evaluation
criteria and the description of each criterion are as follows:
Overall Protectiveness
Environment
of
Human
Health
and
the
The overall protectiveness criterion evaluates whether or not an
alternative provides adequate protection to human health and the
environment by eliminating, controlling or reducing the current
and potential exposures to levels established as remediation goals.
Alternative 1 and 2 do not provide adequate protection to
human health and the environment as they provide no remedial
action or reduction of the risks.
Alternative 3 provides a degree of protection by using groundwater
treatment. However, it does not provide a mitigative measure ~~
prevent precipitation from infiltrating the wastes and caus i:"'.-;
continued groundwater contamination and generation of leachate.
Al teratives 4 and 5 have no discernable differences in overa: ~
protectiveness. The same degrees of protection of receptors wou: :
be provided by both alternatives. Alteratives 4 and 5 prov :~..
groundwater treatment that would be protective of the peo~ ~ ,.
utilizing it as drinking water. The treatment would also prov: :..
protection to the environment by lessening the groundwa~ ,.:
discharge surface water. Alteratives 4 and 5 also include e..
placement of a cap over the landfill. This action will reduce e.
38
-------
dermal exposure risk and
containment of the wastes.
groundwater
contamination
through
Compliance With
Requirements
Applicable
or
Relevant
and
Appropriate
The compliance with ARARs criterion evaluates whether the
al ternatives would meet all of the applicable or relevant and
appropriate requirements of other environmental statues and/or
provide grounds for involving a waiver. .
Under Section 121(d} of CERCLA, 42 U.S.C. Section 9621(d}, remedial
actions must attain ARARs unless such ARARs may be waived under
CERCLA Section 121(d} (4), 42.U.S.C. Section 9621(d} (4).
Alternatives 1 and 2 do not provide any measures which comply with
ARARs.
Alternative 3 does provide groundwater treatment. It will comply
with the ARARs for "background quality" established in 25 PA Code
99 264.90-264.100 or the MCLs or Non-zero MCLs, whichever are
lower. See "The Selected Remedy" for further discussion. It does
not comply with Municipal Landfill Closure requirements in 25 PA
Code 99 273.234 as it does not provide closure of the landfill.
Alternatives 4 and 5 provide groundwater treatment that will comply
with the ARAR for "background quality" established is 25 PA Code
99 264.90-264.iOO or the MCLs or Non-zero MCLs, whichever are
lower. Alteratives 4 and 5 would also achieve compliance with the
action specific and chemicals specific ARARs associated with these
remedial actions as previously described in the paragraphs
describing the specific alternatives.
Lonq term Effectiveness and Permanence
The long term effectiveness criterion evaluates the long term
protection of human health and the environment over time, once the
remedial action goals have been achieved. It focuses on the
magnitude of residual risk and the adequacy and reliable of
controls of the alternatives.
Al ternatives 1 and 2 provide no long term effectiveness as the':'
provide no remedial action or reduction of risks.
Alternative 3 would be considered minimally effective over the lc~~
term as it does not remediate the source and will require a longe~
treatment period than Alternatives 4 and 5.
Al terna t i ves 4 and 5 prov ide the grea test degrees 0 f long t e ~-
effectiveness through source management with the use of the C~;
and groundwater extraction and treatment. It would reduce."
39
-------
residual risks to the acceptable risk range of 1E-04 to 1E-06
excess cancer risk and less than 1 for the non-carcinogenic hazard
index..
Reduction of Toxicity. Mobility and Volume through Treatment
This criterion evaluates the performance of the alternatives to
reduce the toxicity, mobility and volume of the waste by assessing
the degree of irreversibility. and the types and quantity of
residuals remaining.
Alternatives 1 and 2 do not provide any reduction in toxicity,
mobility and volume.
Alternative 3 would reduce the levels of contaminants in the
groundwater and retard their mobility. However, it would have no
effect on the soils, the wastes and the landfill gases.
Alternatives 4 and 5 reduce the toxicity of the groundwater to
remedial cleanup goal in a irreversible process. The mobility
would be limited through the extraction process. The volume would
be lessened by the installation of the cap and the groundwater
extraction and treatment. The gas collection system would also
irreversibility reduce the volume of methane released from the
Site.
None of the alternatives addresses the CERCLA statutory preference
for treatment of the waste as it has been determined to be
impracticable.
Short-term Effectiveness
Short-term effectiveness evaluates the alternatives against the
period of time needed to achieve protection of human health and
the environment and any adverse impacts that may be posed during
the construction and implementation period, until cleanup goals
are achieved.
Alternatives 1 and 2 would impose the least short term impact to
the community, remedial workers and the environment because no
actions would be taken.
Alternative 3 would impose some minimal impact on the remedial
workers and the environment through the installation of the
extraction wells and treatment plant. Groundwater treatment would
begin 3 to 4 months after plant construction begins.
Alternatives 4 and 5 would take the longest time to achieve
protectiveness due to the estimated construction time for the cap
of 8 months. In turn, it will also impose the largest amount of
impact on the receptors through ground intrusive activities. It
should be noted, however that the activities of Alternatives 4 and
40
-------
5 should not cause greater releases than that which occurred during
landfill operations. Active monitoring and contingencies bar cover
will be used to minimize any emissions and short term impacts from
the Site. .
Implementabilitv .
The implementablibliy evaluation. criterion consists of several
subcomponents, including those which evaluate the compatibility. of
remedial measures with site conditions, availability of materials
and services, ability to undertake further remedial actions if
necessary, and regulatory considerations.
Alternatives 1 and 2 are easily implementable because no
construction or monitoring would be involved. Alternatives 1 and
2 are generally not preferred by agencies when risks are high at
a site, and its implementability, based solely on regulatory
considerations is rated low.
Alternative 3 is fairly easy to implement. Undertaking additional
remedial actions under this alternative is very easy as the waste
is easily accessible. However, the ability to monitor
grounderwater extraction may be difficult because flow patterns in
the unconsolidated and bedrock zones are not fully known.
Alternative 4 would be more difficult to implement. than
Alternatives 1, 2, or 3 but not substantially more difficult. It
would require more effort for the installation of the cap. It
would also be difficult to monitor the effectiveness of the cap
without disturbing the landfill.
Alterative 5 would be marginally more difficult than Alternative
4 due to the cap design and installation requirements.
Cost
The cost evaluation criterion considers the estimated cost for the
capital and operation and maintenance (O&M) of the alternatives on
a present worth basis. (See Table 9)
Alternative 1 has the lowest present worth cost at $20,900 to cover
the five year effectiveness review.
Alternative 2 has a present worth cost of $720,000 to cover
fencing, groundwater monitoring and administrative/ institutional
controls.
Alternative 3 has a present worth cost of $3,937,000 to cover the
groundwater treatment. Of these costs, $2,970,000 is present worth
o & M costs.
Alternative 4 has a present worth cost of $9,157,000 to cover 5
41
-------
TABLE 9
ESTIMATED COSTS OF THE FIVE ALTERNATIVES
Alternative CaDi tal Cost Annual Present Worth
1. No Action 0 $ 7,000 $ 20,900
2. Limited $110,000 $ 45,000 $720,000
Action
3. Groundwater
Extraction' $967,000 $199,000 $3,937,000
Treatment
4. Groundwater
Treatment' $5,906,950 $217,000 $ 9,156,950
Impermeable
Cap
5. Groundwater $7,525,000 $217,000 $10,695,000
Treatment'
Multimedia
Cap
42
-------
treatment and an impermeable cap. Of these costs, approximately
$4,000,000 is the. cap and $3,250,000 is present worth 0 & M costs.
Alternative 5 has the highest present worth cost at $10,695,000 to
cover treatment and a multimedia cap. Of these costs,
approximalety $5,000,000 is the multimedia cap and $3,179,000 is
present worth 0 & M costs.
state Acceptance
The Commonwealth of Pennsylvania has concurred with the selection
of Alternative 4, groundwater treatment with the installation of
an impermeable cap that meets PA Municipal Landfill Closure
requirements.
Community Acceptance
The public meeting for presentation of the RI/FS and the Proposed
Plan was held on September 13, 1990. Community interest was very
high at this Site. Comments received during the public meeting
and the comment period are discussed in the Responsiveness Summary
attached to this Record of Decision.
THE SELECTED REMEDY
From the results of the RI/FS and the baseline risk assessment, it
has been determined that a remedial response action is required at
the Keystone Site to reduce the risk from the groundwater
contamination. It has also been determined that due to the volume
of the waste which is considered the source (1.7 million cubic
yards) and the nonhomegeneity of the waste that is is
impracticable to treat the source. Therefore, the remedial action
response is to contain the waste (source) and treat the
groundwater. Remedial action goals have been established to bring
all risks to within the 10E-4 to 10E-6 risk range for carcinogenic
risks and less than 1 for non-carcinogenic risks. It has also
been established that the groundwater contamination will be
reduced to meet the remediation goals of "background quality" as
required in 25 PA Code ~~ 264.90 - 264.100 and in particular, by
~~ 264.97(i), (j) and 264.100(a) (9) and/or the MCLs or Non-zero
MCLGs as specified in SDWA, whichever are lower. See the next
paragraph for explanation. The Commonwealth of Pennsylvania also
maintains that the requirement to remediate to background is found
on other legal authorities.
Based on the considerations in CERCLA, detailed analysis of the
five alternatives developed and comments from the Commonwealth of
Pennsylvania, the State of Maryland and the public, EPA has
selected a remedy. The remedy EPA has selected is Alternative 4-
groundwater extraction, treatment and discharge and the
installaton of an impermeable cap. The goal of this remedia 1
43
-------
action is to restore groundwater to its beneficial use, which is,
at this site is as a drinking water source. In order to restore
groundwater to its beneficial use, the groundwater treatment
system would operate until Site-specific remediation. goals are
achieved. Thus the groundwater would be remediated until the
contaminant levels reach MCLs, Non-zero Maximum contaminant Levels
Goals (MCLGs) or background, whichever are lower. If the
implementation of the Selected Remedy demonstrates, in
corroboration with hydrogeological and chemical evidence that it
will be technically impracticable to achieve and maintain the
remediation goals throughout the area of attainment, the EPA in
consul tation with the Commonwealth of Pennsylvania intends to
amend the ROD or issue an Explanantion of Significant Differences
to inform the public of alternative groundwater remediation goals.
Groundwater extraction at the Keystone Site may consist of
approximately 20 extraction wells pulling groundwater from the
unconsiderated soil and saprolith zone above bedrock for
treatment. The wells would be spaced about 200 feet apart along
the south, east and north edge of the landfill and would capture
the contaminated originating under the landfill. The Remedial
Design will further define the location, number and spacing of
these wells via a capture zone analysis. Placement of the wells
close to the source will contain the plume and prevent future
pulses of contamination from reaching the residential wells.
Maintenance of the collections system would include replacement of
the pumps every five years and maintenance of the well screens
every three years. The actual treatment time required would be
determined by monitoring contaminant reductions after the
treatment begins. A contaminant reduction curve could be plotted
from actual results, after groundwater extraction and treatment
begins, which could be used to predict the treatment time
required. Based on the current knowledge of the Site and
contaminants of concern, it has been estimated that treatment time
would be approximately 46 years.
Groundwater treatment at the Keystone Site will be designed to
treat a wide range of volatile organic contaminants, naturally
occuring inorganic species and hazardous metals. In addition, the
system must be easy to construct, operate, and maintain as well as
discharge treated groundwater in an environmentally safe manner
and location.
The following unit processes would be required for the
groundwater treatment: Equalization; Flocculation/ Precipitation;
Filtration: Ion Exchange~ Air Stripping: and Fi~ter Press. See
Figure 9 for the flow diagram of the Selected Remedy.
Within this system,
flocculation/precipitation
the
unit
equalization
will be used to
tank
remove
and
the
44
-------
r-u .If",
n.~
~
~
\1'1
- .
..... -
C.C.JOHNION I ...\lHO'''A,P.C.
.
,.....
....,...
1
()
,,,.ca.'.
....
'''''.'''
".....
88(.." ...... .-
~
CII....'C""
,m'l..
.,.
. ....-. ,,,,,p
ICA&.
NONI
DA'.
N&. Y 19M
"...
...
uc......
...
...
...
~
..11.'.' ~
......
llse"a- '0
--..r
,........
, IOUI'I
RIll V
GROUNDWATER TREA1MENT PROC£SS
now DIAGRAM
.IU'ON! '1"""- LlNO'ILL '1'1
ununOWN. rlNttSU"INIA .
~
-------
suspended solids and some dissolved metals. Sludges from these
units will be dewatered in the filter press and disposed in an
appropriate offsite facility as specified in 40 CFR 2681.1-
268.50. The ion exchange unit would .follow to remove the metals
such as copper, chromium, nickel and mercury. Effluent
from the ion exchange unit would be pumped to the air stripper
where VOCs would be removed. The air stripper emmissions would
meet the requirements of the Clean Air Act (Part D) (42 U.S.C. ~
7401-7642), the OWSER Directive 9355.0 - 28 for. an air stripper in
ozone nonattainment areas and the PA Air Pollution Control Act, ~~
127.1 et. sea. requiring air emmisions to be controlled with best
available technology. Effluent from the air stripper would be
discharged from the treatment plant to the unnamed stream flowing
north from the site. The volume of treated groundwater discharged
would be equal to the volume entering the treatment plant with an
expected discharge rate of 20 gpm. It would meet the substantive
requirements of a PA NPDES permit required in 25 PA Code ~~ 92.1
et. sea., the PA Water Quality Standard in 25 PA Code ~~ 93.1 et.
~ and the Federal Water Quality Standards of 51 Fed. Rea.
43665. This discharge will not adversly impact the stream system.
The impermeable cap will be designed to comply with the 25 PA Code
~ ~ 273.234 that pertains to landfill closure requirements. In
these regulations, it states that the cap may be constructed of
clay, a synthetic, or a composition of the two. The Commonwealth
of Pennsylvania has indicated that several problems may occur
wi th the use of a natural clay cap due to the permeabil i ty,
maximum aggregate size requirements and others. The selection of
the capping material will be determined in the Remedial Design
phase of this remedial action. Therefore, it has been concluded
that the cap will include, but not be limited to the following:
o
a cap consisting of a 1 ft. clay layer or a 30 mil.
synthetic liner
a maximum cap permeability of 10.7 cm/sec
o
o
a drainage layer over the cap
o
a 2 ft. soil layer over the drainage layer
o
a minumum surface slope of 3 percent
o
a maximum slope of 15 percent
o
minimization of soil erosion and sedimentation
o
stormwater management based on a 24-hour, 25-year
event
In addition to covering an area of 40 acres and a Volume of 1.7
.million cubic yards, the cap will also cover contaminated
46
-------
surficial soils from the spray irrigation area excavated to back
ground levels and places on top of the landfill. An active gas
extraction system would be installed in compliance with 25 PA Code
~~ 273.292 between the existing landfill and the impermeable cap
to prevent buildup of methane gas. Gas collection wells would be
installed in the landfill to depths of approximately 20 feet. The
gas would be collected through a piping network and sent to a gas
burner. Leachate collection would not be considered necessary as
the contaminated groundwater would be intercepted as it moves
laterally from the Site. Since groundwater would be
collected from the perimeter of the landfill, rather than beneath
the landfill, groundwater levels beneath the landfill are not
expected to drop appreciably. Settlement, due to extraction of
groundwater, is not expected to pose a problem since the landfill
is located on bedrock.
I
In addition to the groundwater treatment and cap, the remedy
selected by EPA has several measures to ensure the protection of
human health and the environment. The measures are as follows:
New fencing will be installed at the perimeter of the property
to restr ict access and to 1 imi t exposure potential to onsi te
contaminants. -
Institutional and administrative controls through ~ use of
deed restrictions on the sale and transfer of the property
would be enacted to insure that future property owners would
not misuse the Site. Restrictions on property use would
include aquifer use limitations and limitations on
construction.
Groundwater would be monitored to determine the effectiveness
of the remedy. Sampling will be conducted quarterly in the
first year for TCL organic and TAL inorganic compounds. After
the first year, sampling will be conducted semi-annually until
the five year review occurs to evaluate the effectiveness of
the remedy.
Residential wells around the Site will be sampled quarterly in
the first year for TCL organic and TAL inorganic compounds.
After the first year, sampling will be conducted annually
until the five year review occurs to evaluate the effectiveness
of the remedy. If two sucessive samples detect MCLs being exceeded,
an additional remedial action will be considered to provide the
affected residents with potable water.
Surface water and sediments will be sampled quarterly in the
first year for TCL organics and TAL inorganic compounds plus
methlymercury. After the first year, sampling will be
conducted semi-annually until the five year review occurs to
evaluate the effectiveness of the remedy. If an anomaly is
47
-------
TABLE 10
THE SELECTED REMEDY COSTS
CAPITAL COSTS
Mobilization Site Preparation
Gas Collection system
Clay Cap
Groundwater Extraction & Treatment
Monitoring
Miscellaneous
Legal Fees (1% of capital Costs)
Engineering & Administrative
Fee (5% of Capital)
Subtotal
Contingency (10% of subtotal)
Total Capital Costs
OPERATION' MAINTENANCE
Cap Inspection & Maintenance
Groundwater Treatment
Long Term Monitoring
Public Health Assessment
Subtotal
Contingency (10%) of Subtotal
Total ANNUAL 0 & M
TOTAL PRESENT WORTH COST
48
161,450
316,000
3,929,000
614,500
11,000
60,000
5,906,950
51,000
255,000
5,397,500
509,000
5.906.950
17,000
139,231
35,000
6,30Q
197,531
19,750
217.000
PRESENT
WORTH COSTS
5,9~6,950
I
3,250,000
9,156,950
-------
seen in two sucessive samples, additional studies will be
conducted to determine the nee.d for remedial action.
The onsite residents would have an in home treatment system
installed for their groundwater.
- A review would be conducted every five years as wastes are being
left onsite. The assessment includes analyzing groundwater samples
for TCL organic and TAL inorganic compounds. A report will be made
to summarize the results and any related health effects.
The estimated net present worth of this remedy is $9,156,950. This
was calculated using the cost associated with construction and the
annual operation and maintenance plus contengenics. A 5% discount
rate was used to convert the annual O & M costs to present worth
costs and this was added to the total of the construction costs.
See costs on Table 10 below. The total construction costs of
$5,906,000 were combined with the present worth O & M costs of
$3,250,00 to derive the total present worth costs of $9,157,000.
The annual O & M costs are $217,000 with the first year O & M being
somewhat larger due to the quarterly sampling requirements of the
first year for the monitoring wells, residential wells, and surface
water and sediments.
STATUTORY DETERMINATIONS
EPA's primary responsibility at Superfund remedial sites is to
undertake remedial actions that achieve adequate protection of
human health and the environment. In addition, Section 121 of
CERCLA, 42 U.S.C. § 9621 establishes several other statutory
requirements and preferences. These specify that when complete,
selected remedial action for a site must comply with applicable or
relevant and appropriate environmental standards established under
Federal and State environmental law unless a waiver is justified.
The Selected Remedy also must be cost effective and utilize
permanent solutions and alternative treatment technologies or
resource recovery technologies to the maximum extent practicable.
Finally, the EPA has a statutory preference for remedial actions
that employ treatment that permanently and significantly reduces
the volume, toxicity and mobility of hazardous substances as their
principal element. The following sections discusses how the
Selected Remedy meets the statutory requirements and preferences
set forth by Section 121 of CERCLA, 42 U.S.C. § 9621.
Protection of Human Health and the Environment.
The baseline risk assessment identified inhalation and ingesticr.
of groundwater as significant exposure pathways having an adverse
effect on human health. The environmental survey in the R:
indicated that although the Ambient Water Quality Criteria is beirr:
exceeded for several metals in the surface water, there is r.
49
-------
impact on the stream system. ,The Selected Remedy would protect
human health and the environment by using groundwater extraction
and treatment to halt the migration of the existing contamination
plume and to reduce the level of contaminants being seen in the
groundwater and surface water. This process is being used in
conjunction with an impermeable cap to contain the wastes in the
landfill which will lessen the volume and the concentration of
contaminants to be removed. .
The current risks associated with these exposure pathway is 1E-04
for inhalation and 4E-04 for ingestion to give a total lifetime
excess cancer risk of 5E-04. By using the Selected Remedy, the risk
level will be reduced to acceptable exposure range of 10-04 to 10-
06 for carcinogenic risk and less than 1 for noncarcinogenic risks.
There are no short term threats associated with the Selected Remedy
that cannot be readily controlled. In addition, no adverse cross-
media impacts are expected from the remedy.
Compliance with Applicable or Relevant and
Appropriate Reauirements
The Selected Remedy of groundwater extraction and treatment and
installation of an impermeable cap over the landfill will comply
with all applicable or relevant and appropriate chemical, location,
and action-specific requirements (ARARs). The ARARs are presented
below:
Chemical Specific ARARs include these laws and requirements that
regulate the release to the environment of specific substances
having certain chemical or physical characteristics or materials
containing specified chemical compounds. For the Keystone site
this includes:
o
Safe Drinking Water Act (42 U.S.C.300f et. sea.) (MCLGs)
Maximum Contaminant Levels (MCLs) (40 CFR 141.11-141.16)
o
Pennsylvania "Background Quality" for remediation of
groundwater as specified in 25 PA Code ~~ 264.90 -264.100,
and in particular by 25 PA Code ~~ 264.97(i),(j) and
264.100 (a) (9) .
o
Pennsylvania Water Quality Standards in 25 PA Code ~~
93.1 et sea. which establishes water quality criteria and
designated water use protection for each stream.
o
Pennsylvania NPDES Regulations in 25 PA Code ~~ 92.1 ~
~
o
Federal Water Quality Standards, 51 CFR 43665
Offsite Disposal requirements in 40 CFR 268.1-268.50
(Land Ban)
o
50
-------
Action specific ARARs are usually technology or activity
requirements or limitations taken with respect to hazardous
and tiggered by a remedial action. . For the Keystone site
are as follows: .
based
waste
these
o
Pennslyvania Air Quality Standard, 25 PA Code ~~
123.1(c} establishes requ~rements for fugitive dusts.
25 PA Code ~~ 127.12 (a) (5) requiring emissions be
reduced to mimumum obtainable levels through the use of
best available technology (BAT).
o
o
OWSER Directive 9355.0-28 for emmission from air stripper
in ozone nonattainment areas.
o
National Emissions Standards for Hazardous Air Pollutants
(NESHAPS) which contains emission standards for vinyl
chloride plants. (61 C.F.R. ~ ~ 61.60 - 61.71)
o
Clean Air Act (Part D) (42 USC Section 7402-7642)
o
Pennslyvania Municipal Landfill Requirements, 25 PA Code
~ 75.1 et. sea. in particular 25 PA Code ~~273.234. which
sets forth municipal landfill cap requirements.
o
25 PA Code ~~ 102.1 et.sea. sets forth requirements
for control of soil erosion and sedimentation resulting
from earth moving activities.
o
25 U.S.C. Parts 1910 and 1926 and 29 CFR Part 1910,
Occupational Health and Safety Act states requirements for
response actions at Superfund Sites.
Cost Effectiveness
The Selected Remedy is cost effective because it has been
determined to provide overall ef~ectiveness proportional to its
costs in reducing the risk from groundwater contamination and
landfilled waste. This remedy at $9,156,950 present net worth cost
is the fourth highest costing remedy. While it is somewhat more
costly than Alternative 1, 2, or 3, the additional costs for
capping represents a degree of protection that is required by
CERCLA for human health and the environment. This remedy provides
a reasonable cost effective solution to a landfill volume of
3,177,000 cubic yards. .
51
-------
utilization of Permanent Solution and Alternative
Technoloaies to the Maximum Extent Practicable
Treatment
The EPA has determined that the Selected Remedy represents the
maximum extent to which permanent treatment technologies can be
utilized in a cost effective manner for the Keystone site. It has
been recognized that the volume of nonhomogeneous wastes (3,177,000
cubic yards) represents a large obstacle in finding a cost-
effective treatment technology. Of the al ternati ves that are
protective of human health and the environment and comply with
ARARs, the EPA, the Commonwealth of Pennsylvania and the State of
Maryland have determined that the Selected Remedy provides the best
balance among tradeoffs of the nine criteria in terms of long term
effectiveness and permanence, reduction in toxicity, mobility and
volume achieved through treatment, short term effectiveness, and
implementability while being cost effective.
In addition, the selection of this remedy has received
concurrence from the Commonwealth of Pennsylvania. The Selected
Remedy is in compliance with the regulatory action that
Pennsylvania may pursue under their Consent Adjudication with
Keystone.
Preference for Treatment as a Principal Element
The statutory preference for a remedy that employs treatment as a
principal element has been determined impracticable at the Keystone
site. Due to the large volume (3,177,000 cubic yards) of
nonhomogeneous wastes, EPA has screened out treatment technologies
and accepted containment and groundwater treatment as a remedy.
This remedy addresses the principal threat at the site from
groundwater contamination by preventing infiltration through the
waste and provides control and treatment of the contaminant plume.
52
-------
RESPONS:IVENESS SUMMARY.
KEYSTONE SAN:ITAT:ION LANDP:ILL
tJN:ION TOWNSH:IP
ADAMS COUNTY, PENNSYLVAN:IA
SEPTEMBER 1990
S:ITE
This Community Relation Responsiveness Summary is divided into
the following sections:
section A:
Section B:
section C:
section D:
overview- A discussion of EPA's preferred remedial
alternative and the public reaction to this
alternative.
Backqround of Community Involvement and Concerns-
A brief history of community's interest in and
involvement with Keystone Landfill site, including
a discussion of concerns raised by community
members and officials during planning activities.
Summary of Public Comments Received Durinq the
Public Comment Period and Aqencv Responses-
A summary of comments followed by EPA responses.
R~maininq Concern- A description of remaining
community concerns that should be considered as the
EPA and Pennsylvania Department of Environmental
Resources (PADER) conduct the Remedial Design and
Remedial Action at the Keystone Sanitation Landfill
Site.
In addition to section A through D, a list of EPA community
Relations activities conducted at the Keystone Sanitation
Landfill Site is included as Attachment A of this Responsiveness
Summary.
-------
A.
OVERVIEW
EPA released the Remedial Investigation/Feasibility study (RI/FS)
and the Proposed Plan for Keystone Sanitation Landfill site in.
union Township, Adams County, Pennsylvania for review and comment
on July 20, 1990. This began the 30 day Public Comment Period
for the release of the Proposed Plan. In the Proposed Plan, the
results of the RI/FS report and the alternatives evaluated in the
FS, including EPA's preferred alternative for remediation of the
Keystone Site were discussed. EPA's preferred remedial action
alternative was the installation of an impermeable cap designed
to meet Pennsylvania Municipal Landfill Closure requirements over
the landfill and a ground water extraction and treatment system
on the eastern and southern perimeters of the Site to treat the
contaminated groundwater while preventing future migration of the
contamination offsite. The groundwater extraction and treatment
system would consist of unit processes to remove the volatile
organic compounds (VOCs) and metals from the ground water and
restore the groundwater to its most beneficial use.
The remediation goals for the groundwater would be the
Pennsylvania "Background Quality" requirement specified in 25 PA
Code ~~ 264.90 - 264.200 and/or the Maximum Contaminant Levels
(MCLs) or Maximum contaminant Level Goals (MCLGs) specified in
Safe Drinking Water Act 42 U.S.C. ~~ 300f et.sea., whichever are
lower. The treated groundwater would be discharged into surface
waters to meet the substantive requirements of a PA NPDES permit
required in 25 PA Code ~~ 92.1 et. sea., and the PA Water
Quality Standard in 25 PA Code ~~ 93.1 et. seq. and the Federal
Water Quality Standards of 51 Fed. Rea. 43665. Any air emissions
would meet the requirements of the Clean Air Act, 42 U.S.C ~~
7401 et. sea., OWSER Directive 9355.0 -28 and PA Air Pollution
Control Act 25 PA Code ~S 127.1 et. sea.. All other applicable
or relevant and appropriate requirements (ARARS) would be met by
the preferred alternative. The preferred alternative also
contained administrative measures such as deed restrictions,
fencing, aquifer use restrictions, supply of treated water to the
onsite residents, as well as residential well monitoring, surface
water and sediment monitoring, groundwater monitoring and a
five-year review requirement. The groundwater remediation period
was estimated for 30 years with a present worth cost of
$9,156,950 for the impermeable cap and the treatment system. It
would take approximately 14 months to implement this alternative.
In response to requests from the public, EPA extended the Public
Comment Period an additional 30 days, for a total of 60 days to
close on September 20, 1990. During this time, EPA held an
informal meeting with elected officials from Union Township,
representatives from citizen groups in Pennsylvania and Maryland,
and a representative from the Maryland Department of the .
Environment to answer questions and receive comments on the
alternatives in the Proposed Plan. As required by sections 113
and 117 of CERCLA, EPA also held a Public Meeting to present the
findings of the RI/FS and Proposed Plan and answer any questions
-------
concerning the Site. This meeting, which was preceded by a press
conference and a briefing for state, federal and local elected
officials, was held on September 13, 1990~ It was originally
scheduled to be held on August 15, 1990 and was rescheduled to
September 13, 1990 in response to requests from the public. The
public was notified of the change in the meeting date via
newspaper, radio and phone calls. At these meetings, EPA
presented the RI/FS report findi~gs, and the Proposed Plan which
was EPA's preferred remedial action alternative. Comments and
questions were received and documented.
JUdging from the comments received during the Public Comment
Period, the Maryland Department of the Environment and PADER
generally support and accept EPA's preferred alternative of an
impermeable cap and ground water extraction and treatment. The
State of Maryland made some suggested modifications concerning
landfill closure and site inspections. Although, citizens,
citizen groups or their consultants stated rejection of EPA's
preferred alternative, some stated that they felt the RI was not
sufficiently comprehensive to design an effective remedy. In
particular, they questioned the ability of EPA to design an
inclusive remedy based on the analysis of one round of sampling
and a data set that was not inclusive of all residential wells
that have historically shown contamination. Residents in the
area requested that additional sampling of residential wells,
testing for contaminants at absolute values, and bottled water to
protect their health be added to the remedial action plan.
Maryland residents and local elected officials were not
supportive of a remedial action that does not include offsite
remediation of Maryland residential wells and Maryland monitoring
wells. Many citizens and representatives of citizen groups
expressed concern with any remedial action plan that would
include cooperation between EPA and PADER or the owners of
Keystone. Community confidence was low in the ability of EPA to
monitor effectively the work of PADER and/or the owners of
Keystone. Local citizen participation in the monitoring effort
was desired.
In addition to the Public Meeting, comments were received on the
RI/FS and Proposed Plan through and after the close of the Public
Comment Period. A large volume of comments were received and
reviewed by EPA. Several studies were submitted by the
potentially Responsible Parties (PRPs) which did not accept the
finding of the RI/FS or support the preferred alternative.
Several PRPs commented that the installation of a cap would
impede the remediation of the groundwater extraction and
treatment system. The PRPs noted that the treatment of metals in
the groundwater was unnecessary as the metals are naturally
occurring and not site-related.
In response to the public concerns over offsite contamination,
especially in the residential wells, and the validity of the
data concerning the need to treat metals in the groundwater, the
preferred alternative was changed from a final ~ction to an
-------
interim action. It was decided that a second operable Unit would
address offsite contamination. Also, in response to the pUblic
concerns over their drinking water; EPA will be working with
members of the communities in both Maryland and Pennsylvania to
develop a sampling plan for the residential wells in the areas.
This sampling, which will use lower detection levels for
contaminants of concern, will occur in the next several months
after the development of the sa~pling plan and the attainment of
funding. It will take into consideration historical data on the
residential wells as requested by the public.
B.
BACKGROUND ON COHKUNITY INVOLVEMENT
Community involvement at the Keystone Sanitation Landfill began
when contamination was discovered at the site in 1983. This, as
well as increased press coverage of the landfill, resulted in the
formation of citizens Urge Rescue of the Environment (C.U.R.E.).
In 1984 the citizens of Carroll County, Maryland formed People
Against contamination of the Environment (P.A.C.E.). Both groups
have spent much time and money reviewing site information and
gathering technical data. The primary community concerns, as
documented in the community relations plan, were as follows:
1. The citizens were concerned that their drinking water was
contaminated by the Keystone Landfill. The citizens felt that
the information on ground water testing did not indicate the true
quality of area drinking water. In addition, at least one
citizen group requested that a health study be done in the
Keystone area to address the citizens' health concerns.
2. citizens expressed distrust for state environmental
agencies, such as the Pennsylvania Department of Environmental
Resources and Maryland's Department of the Environment. They
questioned the accuracy of government released information and
the sincerity of community interaction efforts.
3. Several local resident paid to have their well water tested
privately, and sent the results to EPA. citizens have expressed
anger and disappointment that this data was not referenced by EPA
in further studies that were done.
4. Many citizens and local officials have expressed their
concerns that all potentially responsible parties (PRPs) be
located. They have requested that local industries not be
penalized for following what, at the time, was proper and legal
disposal methods recommended by PADER. They have implied that
those industries that have broken the laws should be held
responsible for cleanup costs.
-------
RESPONSIVENESS SUMMARY
KEYSTONE SANITATION LANDFILL SITE
COMMENTS ON THE RI/FS. REPORTS
The following summarizes and responds to comments received
concerning the RI/FS Proposed Plan, and Administrative Record for
the Response Sanitation Landfill. site (Keystone site or Site).
In some cases, similar comments were received from a number of
commentors. In such cases the comments were addressed only once.
PUBLIC MEETING
1.
How will EPA monitor future work performed by the PRPs or
governmental agencies?
EPA Response: EPA monitors all work performed during the
Remedial Design and Remedial Action through the Remedial
Project Manager, the EPA Technical Support Personnel, EPA
oversight contractors and document reviews. All work must
be in accordance with local, state and federal statues,
regulations and ordinances.
2.
What is the extent of the contaminant plume and will the
Proposed Plan address it?
EPA Response: The RI testing revealed that the extent and
magnitude of offsite groundwater contamination is small.
Some monitoring wells downgradient of the landfill contained
small quantities of contaminants but these were in wells
located close to the landfill. The proposed plan will
prevent the future releases of contaminants to the
groundwater. Ground-water extraction wells and a treatment
plant will reduce the contaminants in the groundwater. The
extraction wells will contain the contaminated plume and
prevent future pulses of contamination from reaching the
residential wells. Levels of offsite contaminants in the
groundwater will decrease as a result of natural processes.
A second operable unit will address offsite contamination.
3.
Will EPA provide bottled water and filters for people in the
area?
EPA Response: The testing conducted for the RI has shown
that offsite residential wells in the area are not affected
by the Site. Only the onsite residents has contaminant
levels which exceeded the PA Drinking Water Standard and
will be supplied with an inhome treatment system for their
drinking water. Monitoring wells placed near the landfill
showed no or low levels of contamination. As the sampling
of the residential wells did not reveal contamination, at
1
-------
this point in time, EPA will not provide bottled water or .
filters. Should future testing reveal problems, we will
reconsider this alternative. Testing of the residential
wells in the area around the site with lower detection.
limits will take place in the near future. If this sampling
indicates a need for bottled water by exceeding the MCLs or
the risk range, EPA will consider it. In addition, pursuant
to the Selected Remedy, the residential wells around the
site will be sampled quarterly in the first year and
annually thereafter. ..
Will EPA obtain an accurate population count of the area?
4.
EPA Response: While the population of the area may have
changed since the RIjFS study began, the findings and
conclusions are not affected by the actual population of the
area. Increasing the number of people in the area would not
change the proposed plan nor influence the calculation of
risk from the site: thus another population count is not
necessary.
5. What is the basis for using analytical results that exceeded
the holding time?
EPA Response: Only data that passes the data validation
process were used in the reports. In this particular case,
if the holding time is exceeded by only a short period of
time, the data is still valid. If the data validation
experts had determined that the data was compromised, it
would have been so noted and would not have been used in the
RIjFS.
6.
The detection limits are too broad to detect most chemicals
in the groundwater.
EPA Response: The detection limits used at the site are the
same that are used across the country to evaluate Superfund
sites. These detection limits are set nationally for the
Contract Laboratory Program by EPA. Lower detection limits
will be used for the sampling conducted on the residential
wells in the near future and during the Design Operations
for detection of contaminants below the MCLs or risk range.
7.
Can you really clean up this site?
EPA Response: The Proposed Plan will effectively remediate
the site. The cap will prevent direct contact with
contaminants that may be at the surface, will prevent
airborne contaminants from migrating, and will reduce the
amount of precipitation entering the landfill. The
groundwater extraction and treatment system will prevent the
migration of contaminants in groundwater and restore
groundwater to it most beneficial use.
2
-------
8.
9.
10.
11.
12.
13.
Will EPA designate the aquifer as a sole-source aquifer?
EPA ReSDonse: EPA cannot "at the stroke of a pen" designate
an aquifer as sole-source. Any state, municipal or local
government, political subdivision or planning entity that
identifies a critical aquifer protection area over which it
has authority may apply to EPA for selection of an area as a
sole-source aquifer.
EPA should not discharge the treated groundwater to surface
waters but should transport it to a treatment center.
EPA ReSDonse: The treated groundwater will meet all ARARs
for surface water discharge. The discharge should have
little impact on the physical or chemical characteristics of
the surface water body. The quantity of groundwater that
will be treated initially is too large to be hauled offsite.
However, once the cap is in place, it is anticipated that
the quantity of groundwater extracted will decrease to such
levels that offsite treatment may be feasible. If this
decrease occurs, EPA with concurrence from the Commonwealth
of Pennsylvania may issue an Explanation of Non-Significant
Difference to change the remedy to offsite treatment of the
groundwater.
Will EPA perform air monitoring and test for methane?
EPA ReSDonse: Air monitoring may be conducted during the
Design Phase if determined to be necessary. It will be
conducted during the construction of the cap, extraction
wells and treatment as part of the Health and Safety Plan
for the Site. During the operations, an active gas
extraction system will be installed between the existing
landfill and the impermeable cap to prevent the buildup of
methane gas.
Could springs located under the landfill affect the remedy?
EPA ResDonse: If a spring is
will mix with the groundwater
groundwater treatment system.
remedy.
Will the pressure of the cap affect the subsurface
conditions?
located under the landfill, it
and be extracted by the
It will not affect the
EPA ReSDonse: The cap may have some subsidence over time.
During the design, this will be taken into consideration and
contingencies in the design will address it.
Has EPA conducted a health study for the community?
The Agency for Toxic Substances and Disease Registry (ATSDR)
conducted a study in November 1988. In addition,-a risk
3
-------
1- -~._------
14.
15.
assessment was performed for the site. The risk assessment
documented the effects of.the contaminants found at the site
under a variety of situations. Both documents can be found
in the Administrative Record located at the Hanover Public
Library.
CITIZEN GROUPS CONSULTANTS AND PUBLIC AGENCIES
We disagree with the conclusion in the reports regarding the
lack of impact to the flora in the area. Aerial photographs
of the Site for the past several years detect on the south
side of the landfill beginning around 1986 to the present,
an area of impact emanating from the landfill in the field
bordered by Line Road, and located on the ridge midway
between the Brown land and Biemiller Road. The pictures
enclosed show close up, the impact to various crops planted
over the past few years. Another area of impact can be seen
by the pictures of the woods located in the southeast corner
of the Keystone Landfill. These were taken in August of
1986 and again in August 1988 when a cluster of trees turned
brown and consequently died. This was in the area of the
spray irrigation system. The last three pictures taken in
July 1985, show water pouring from the landfill to the south
across Line Road and saturating the adjacent field
immediately following a summer storm. This occurred quite
frequently. Again, we call your attention to the adverse
impact on the soybeans. Testing of the soil in that area by
EPA that year detected VOCs and chromium in the soil.
EPA ResDonse: No vegetation was deemed to be adversely
affected as a result of contaminants from the site based on
observations made during the RI. The contaminants found in
the areas identified in the comment are not found in
concentrations that would be toxic to vegetation. In
addition, the location of these areas are such that the
vegetation may be affected by too much water. The constant
watering as a result of the spray irrigation system and
storm water flowing in the drainage in the field may be
having a negative effect on the vegetation.
Another area of our concern involves the EPA location of
surface water sampling sites for both the current study and
the proposed future water sampling to monitor stream
conditions. Because of the dilution factor, the rapid
volatilization of the VOCs particularly in warm weather, and
varying weather conditions, it is difficult to monitor the
off-site contamination. We have found that if the water is
tested at the point where it exits from the ground, the
chances of detecting the contamination are greatly improved,
providing a clearer picture of the problem. Thus, we would
suggest selecting sampling points closer to the discharge
areas. .
4
-------
16.
17.
18.
EPA ReSDonse: The Proposed Plan addresses concerns
regarding the offsite migration of contaminants via surface
water. The system will be designed to intercept
contaminants at the edge of the capped area, thereby
preventing contaminants from reaching any surface water.
The location of future surface water samples to monitor
surface water considerations have not been determined.
Future sample locations will attempt to obtain samples from
groundwater discharges. .
These groundwater exit points are a source to measure
contaminant migration. We wish your comments as to whether
any soil or water sampling points were chosen with this in
mind. The site investigation done in 1984 indicated a
leachate seep along the Brown property land about 200 yards
south of Line Road. The state of Maryland sampling of that
area also indicated leachate contamination. Private
sampling has detected VOC contamination there. Yet, there
were no EPA samplings at all done on this area for the
remedial investigation study. Considering the numerous
discharge areas located downgradient.of the landfill, it
would be prudent to take a closer look at this documented
path of contaminant migration.
EPA ReSDonse: Samples collected in this area included 3
surface water, 2 sediment, 2 residential wells, and 7
groundwater. These samples are sufficient to characterize
the impacts of any contaminants migrating from the site. In
addition, the proposed plan will significantly reduce
releases from the landfill. Additional sampling of the
surface water and sediments will take place during the
Design and Operational Phase.
The Tethys Report clearly shows that there are serious
deficiencies in the analytical data. This illustrates the
need for more testing and analysis before alternatives can
reasonably be decided upon.
EPA ResDonse: No data with serious deficiencies have been
used in the preparation of the RI or FS Reports. All
suspect data were identified by independent data validators
and none of the suspect data were used in the reports.
However, additional samples will be collected prior to and
during the Design Phase of the project.
The failure to include and submit to public comment the
proposed groundwater recovery plan is a serious deficiency
of the study. A groundwater extraction and treatment plan
needs to be developed giving details of the number and
location of recovery wells, their depths, and design before
meaningful comment can be made on this important phase of
the proposed remediation. Spray irrigation must be
eliminated as a means of treatment in view of the serious
5
-------
19.
20.
21.
environmental and health risks associated with such
treatment.
EPA Response: We agree that the details of the g~oundwater
extraction system are important. However, the purpose of
the FS is to present conceptual designs, estimate costs, and
evaluate the alternatives against a given set of criteria.
A detailed design of the selected alternative is made prior
to implementation. The public will have an opportunity to
comment on the design of the selected alternative once the
design is completed. Spray irrigation is not a part of the
proposed plan.
Given the fact that the data do not show conclusively or
with a reasonable degree of certainty that the wells in the
area will not be at risk, even assuming a competent
groundwater remediation plan is developed, the selected
alternative should include the requirement of providing an
alternative public drinking water supply to nearby
residents.
EPA Response: The data show that the wells in the area of
the site are not currently contaminated. The remediation
plan will insure that in the future no contaminants that
could potentially contaminate the groundwater will leave the
site. Additionally, EPA plans to sample residential wells in
the area of the Site using a lower detection level in the
near futur~. If the sampling reveals contaminant above the
MCLs or the risk range, bottled water will be considered.
Sampling of the groundwater, surface water and sediment will
also take place during the Design and Operations.
As a minimum, in view of the serious problems with ground-
water, EPA should insist upon a multi-media cap. The extra
cost is a small percentage of the total and is warranted by
the seriousness of this problem.
EPA Response: No serious problems with the downgradient
groundwater have been identified at the Site. The use of a
single layer cap will provide the protection necessary to
prevent direct contact and infiltration of precipitation.
There is no discernible difference in overall protectiveness
between a multi media cap and a single layer cap at this
site. In addition, the single layer cap may be more
reliable and easier to construct and monitor due to its
simpler construction and fewer layers. The second Operable
Unit will address offsite contamination.
The problem of soil contamination caused by the existing anj
inadequate spray irrigation system should be addressed more
thoroughly.
EPA Response: The soil in the spray irrigation area was
fully characterized. sixteen soil samples were collected ;'
6
-------
22.
23.
24.
25.
this area. Additionally the soil in this area will be
excavated as source remediation, placed on top of landfill
and capped. .
More attention must be paid to analyzing and improving the
existing leachate control system so that periodic outbreaks
of leachate that have been experienced in the past will not
occur in the future. The report does not address this
problem, and it must do so. .
EPA ResDonse: The existing leachate collection system will
not be utilized in the future. Any leachate emanating from
the landfill will be reduced by the cap and collected in the
groundwater extraction system.
Waste and sediment leaving the site have been observed and
documented by numerous citizens, and yet the report ignores
these facts. This problems must be analyzed, and corrected.
EPA ResDonse: All streams in the vicinity of the landfill
were sampled and some contaminants were identified. The
proposed groundwater collection system and placement of an
impermeable cap will be designed to prevent contaminants
from migrating from the landfill and into the areas streams.
No significant contamination problems were detected in the
areas surface water or sediments.
The fai1u~e to adequately characterize the waste which
exists on the site is a serious deficiency. EPA's own early
reports indicated much industrial and hazardous wastes was
disposed of at this site. The failure to assess this
potential for contamination must be rectified before this
study can move forward.
EPA ReSDonse: Early reports indicate that some industrial
waste had been disposed of in the landfill. Character-
ization of the landfill would not provide any additional
information that would affect the selection of a remedy.
The effect of the wastes in the landfill were documented in
the RI Report. The proposed remedy provides sufficient
protection against future releases of many compounds. EPA
guidance does state that sampling of nearby groundwater and
soil is sufficient characterization for large landfills and
waste sampling is not necessary. Also sampling of the waste
could cause contamination by breaching isolated pockets. It
could also be a concern during the boring operations for the
health and safety of the remedial workers, onsite residents
and workers.
The maps, charts and figures were generally well drawn, eas!
to understand, and pertinent. However, a minor problem was
noted on the potentiometric surface map presented at figur.'
3-8 and 4-15, which did not show the measurement points us.' :
to generate it. Also, a discrepancy was found to exist in
7
-------
26.
27.
the 740 ft. contour line along the match lines connecting
the other potentiometric surface map at Plates 1 and 2.
EPA ReSDonse: The water level measurements used to
construct the potentiometer map are given in Table 3-4. The
upper groundwater contour line on Plate 2 should not have
been shown. Because of the topographic high located near
Line Road, the 740 contours on Plate 1 and 2 should not be
considered to be the same continuous contour.
The water and soil chemical quality assessments were
thorough and were sufficient to assess the extent to which
pollutants are migrating from the Site. The few quality
assurance/quality control (QA/QC) problems indicated (such
as the appearance of some common lab contaminants in some of
the samples including blanks) did not hamper the validity of
the assessments. However, some additional attempts to
explain anomalies which were not attributable to the
landfill could have been made, such as pointing out the
agricultural nature of the areas where pesticide residues
were observed. These occurrences might otherwise be
disturbing to residents of the area, to whom the probable
reason for the presence of these compounds might not be
immediately apparent. other examples are the discussion of
the base/neutral- and acid-extractable (BNA) organic
chemicals such as anthracene and fluoranthene, which are
commonly present in asphalt, coal residues and petroleum
products such as gasoline and used lubricating oil; and Bis-
(2-ethylhexyl) phthalate, a plasticizer used in PVC, which
may be observed in wells of low yield such as MD-MW-l and 8
due to the inability to fully purge them prior to sampling.
EPA ReSDonse: The rural nature of the area does result in
potential sources of contaminants other than the landfill.
Where appropriate these other potential sources were
identified.
The Applicable or Relevant and Appropriate Requirements
(ARAR) review did not include Maryland law, although the
planned activities could conceivably impact on Maryland.
Examples of possible infringements could include sediment
control, water pollution and possibly even groundwater
appropriations codes. It is noted that pennsylvania and
Federal laws and regulations appear to be generally
protective of our concerns, and conceptual approval at this
stage will not affect our ability to enforce our codes if
the need arises. However, we do request continued
involvement in the remedial process including review of
design documents and construction activities, particularly
if any changes in the proposed design are introduced.
EPA ResDonse: The State of Maryland will continue to have
an active consultation role in the design and implementation
of the proposed Plan. All Maryland ARARs will be considered
8
-------
28.
29.
30.
31.
32.
for activities which are conducted in operable Unit 2 or
which may impact on the state.
Reference was made in the RI to Keystone being tasked by a
Consent Adjudication with PADER to direct similar
investigations focused mainly on site. To date, a detailed,
technically sound, comprehensive investigation has not been
satisfactorily completed for the landfill site. Is this a
reference to a future study. or to past incomplete studies
already carried out? .'
EPA Response: This was a reference to the Keystone study
submitted to PADER in January 1990.
A previous report prepared by Buchart-Horn concluded that no
deep regional aquifer flow exists, and that fractures in
bedrock close at depths of 25 to 100 feet. This conclusion
was proven to be incorrect. Data collected during the RI
and expressed in graphic form in Figures 4-5, 4-6, 4-8, and
4-13 clearly show fractured zones in the bedrock at depths
as great as 280 feet below the land surface. These
fractured zones would be capable and available for the deep
transmission of groundwater and any contaminants it may
contain.
EPA ResDonse: Nearly all rock contains fractures or
structure capable of containing water. However the
fractures in the bedrock at the Keystone Site are small and
transmit little water. This is evident by the small
quantities of water in the deep wells, the results of the
packer tests, the slow recharge of these wells, and the lack
of contaminants in the samples from the wells.
The RI stated that the Keystone Site is situated on a
drainage divide created by a northeast-trending bedrock
ridge. This statement is not supported by a topographic' map
of the site which is presented in Figure 2-2. In that
figure, the bedrock ridge is clearly trending in a northwest
direction.
EPA ResDonse: The drainage divide does tend to be northwest
east of the site. South of the site the divide tends to be
east-west. Southwest of the site the divide tends to be in
a northeast direction.
It was assumed that the 0.10 and 0.20 inch slot size of the
screens used for well construction as stated on pages 3-19
and 3-20 were in reality actually 0.010 and 0.020 inches in
size.
EPA ResDonse:
The screen sizes were 0.010 and 0.020 inches;
The RI Investigation considered residential well RW-13 and
RW-15 to be background locations, and subsequently. used this
9
-------
33.
34.
consideration in interpreting water chemistry data. Both
RW-13 and RW-15 are clear~y shown on Figures 3-5 and 3-8 as
being hydrogeologically downgradient from the landfill. For
this reason, they should not be considered to be reflective
of background conditions.
EPA ReSDonse: Residential wells RW-13 and RW-15 were
considered to be background locations due to their distance
from the Site and because historically contaminants were not
found in onsite wells located on that side of the Site.
However, when analyzing the data, organic compounds found in
these wells were not considered to be naturally occurring.
During the Design phase, additional sampling will be
conducted to determine a background sample more
representative of the area.
The RI repeatedly stated that all VOCs and inorganic
analyses were successfully analyzed in all soil samples,
surface water and sediment samples, and groundwater samples.
These statements were immediately followed by long lists of
qualifiers and disclaimers regarding the validity of the
data. So many disclaimers were made that the validity of
the entire chemical data bank for all analyses must be
called into question. This is critically important. All of
the contents of both the RI and FS reports hinged on the
validity of the chemical analyses. Since so much of the
data were disqualified, why wasn't a second round of samples
collected for all sampling points and reanalyzed following
better laboratory QA/QC procedures?
EPA ResDonse: The laboratory QA/QC procedures were the
same that are followed at all Superfund sites nationwide.
Few of the data had qualifiers which made the data unusable.
The data that was not useable did not compromise the
investigation or the results and conclusions drawn. To the
. contrary, the qualifiers on the data indicate that the QA/QC
procedures were effective and of the highest quality.
Additional samples will be collected from the monitoring
wells, residential wells, the surface water and sediment
during the design phases.
Offsite soil sampling locations SL-19, SL-20, and SL-21 were
considered in the RI to be background. If one refers to
both the potentiometric map and the topographic map
presented in Figures 3-8 and 2-2, respectively, it becomes
apparent that all three of these locations lie downgradient
from the landfill and should not be considered to be
reflective of background soil conditions.
EPA ResDonse: Because the landfill is the highest point in
the area, all offsite areas lie downslope of the landfill.
However, when surface water drainage patterns on the
landfill, the location of drainage ditches, and the
localized topography are taken into account, the locations
10
-------
35.
36.
37.
38.
of these soil samples are reflective of background
conditions.
Nineteen BNA's were detected in soil samples collected
within the landfill site. Nine BNA's were detected in soils
collected offsite. The RI investigation seemed to be too
quick to rule-out a connection between on site activities
and offsite soil contamination.
EPA Response: Many of the compounds detected offsite were
also detected at the background locations. In addition, no
pattern of detection was evident that would indicate the
landfill as the source of these contaminants (except in the
area of the spray irrigation system). In addition, the
compound detected are often found in chemicals that are used
on farms.
since barium and manganese have been detected in relatively
high concentrations in some local residential wells, why
weren't these elements tested in onsite and offsite soils to
help answer the question of their origin.
EPA Response: Barium and manganese were analyzed for in all
samples and were accounted for in the risk assessment. Both
are naturally occurring in soil and rock. Barium was found
in leachable concentrations in the bedrock.
RI states that offsite residential wells RW2, RW4, and RWll
had the largest number of metals detected. It may be more
than a coincidence that RW4 and RWll are two of the closest
residential wells to the landfill, and that K8 which is
close to RWll was one of two on site wells with relatively
high metal concentrations.
EPA Response: Four metals were detected in each of these
wells. Two of these metals, copper and zinc, were found in
all groundwater, soil and sediment samples. These metals
are considered to be naturally occurring. Three different
metals were detected in the three wells. These metals were
not detected in any of the other residential wells and are
therefore not considered to be Site related.
A significant amount of data has been collected over the
past several years by both local residents, private
laboratories, and the Pennsylvania Department of
Environmental Resources. This data is available and should
have been but apparently wasn't reviewed during the
completion of the RIfFS activities.
EPA Response: These data were reviewed and evaluated during
the RIfFS Workplan preparation and are available in the
Administrative Record. The results of this review was
incorporated in the Workplan and helped to form the basis
11
-------
39.
40.
41.
for the field investigation.
Why does every page in Chapter 7 of the RI contain this
statement "DRAFT: Do not CITE. or QUOTE." This warning
implies that none of the contaminant fate and transport
information or the conclusions based thereon are final,
including the feasibility study results.
EPA Resoonse: At the time that the RI Report was released,
the risk assessment was in draft form and awaiting a final
EPA review. That review has since taken place. The final
risk assessment differs little in its conclusions and does
not affect the results or conclusions of the FS.
It is also believed that the RI was much too limited in
scope. The results contained therein simply represent a
snapshot in time. Since only one set of data was collected,
it is unknown whether the worst contamination problems have
past and things are improving, or whether more problems will
be expressed in the future and the worst is yet to come.
EPA Resoonse: Historical data collected by local residents,
private consultants and the PADER were reviewed and
evaluated during the preparation of the RI/FS Workplan.
Additional samples will be collected. In the interim prior
to design, samples will be taken of the resident~al wells in
the area. During the Design Phase, more samples will be
taken of the monitoring wells, and residential wells.
Soils, surface water and sediments sample may also be
obtained if necessary during the Design. During the
remediation, constant monitoring of the monitoring wells,
residential wells, surface water & sediment will be taking
place. Through the collection of all this data, EPA will be
able to not only monitor the remediation process but make
necessary changes as warranted to reach the remediation
goals.
There should have been an onsite waste characterization
program completed at the landfill, more offsite wells
installed, more residential wells sampled, and more stream
and sediment samples collected. There should have been
several rounds of samples collected from each point, not
just one. Aquatic organisms in nearby streams were
completely omitted from the investigation even though off-
site surface waters were found to be contaminated. Finally,
the very poor laboratory QA/QC procedures which invalidated
so much of the chemical results, especially organic
compounds, have made this entire RI/FS of questionable
value.
EPA Resoonse: The extent of sampling conducted for the RI
provides the data necessary to determine the extent of
contamination, estimate the potential risks posed by the
Site, and evaluate potential remedial alternatives.
Additional sampling will be conducted during the pre-design
12
-------
1-
42.
43.
44.
and design phases of the project. The sampling results did
not show widespread contamination in the streams that would
be detrimental to aquatic' organisms. The laboratory QA/QC
procedures were the same that. are followed at all Superfund
sites nationwide. Few of the data had qualifiers which made
the data unusable. The data that was not useable did not
compromise the investigation or the results and conclusions
drawn. On the contrary, the qualifiers on the data indicate
that the QA/QC procedures were effective and of the highest
quality. . .
PRP CONSULTANTS
REMCOR Comments
Page 1-7; This appears to be the most detailed discussion of
the irrigation system in the entire Remedial Investigation
(RI), yet specific information regarding its operation is
missing (i.e., pumping rate, application area, depth of
leachate collection system.)
EPA ReSDonse: EPA's RI was primarily concerned with how
the system affected contaminant transport via groundwater
and soils, not to investigate the operations of the
irrigation system.
Page 1-8; The RI indicates the first study was conducted by
the U.S. Environmental Protection Agency (EPA) in 1984, yet
monitoring wells have been at the site since 1974. What
data were collected during that period and what do they
indicate? Who installed the wells and why are they still
being sampled?
EPA ReSDonse: Previous investigations were described and
evaluated in the Workplan for the site as indicated in the
RI Reportand are contained in the Administrative Record.
.Monitoring wells were installed at the Site by Keystone in
response to PADER landfill requirements to monitor the
groundwater. Data obtained from these wells indicated the
presence of VOCs in the groundwater. Most recently, these
wells were sampled as a result of a study by Keystone
required by a Consent Adjudication for PADER.
Page 1-8; The EPA presents limited descriptions and data
from previous investigations. However, none of this data is
ever correlated with that generated during the RI. This is
a serious oversight that could have significantly affected
the RI's conclusions. Specifically, no effort was made to
confirm the effectiveness of continuing on-site remedial
efforts.
EPA ResDonse: The objective of the RI was not to determine
the effectiveness of onsite remedial effort but to determine
whether contamination exists and is migrating offsite. The
previous data base was reviewed and used in the development
13
-------
I~
45.
46.
47.
48.
49.
of the Workplan.
Table 2-2 (Page 2-8); Neither Todd, 1980 nor Freeze and
Cherry, 1979 was included in Section 9.0, References.
These references contain much "state-of-the-art" information
and are generally recognized as extremely valuable tools
for this type of study.
EPA Response:
These references were noted in the text.
Page 3-1; No data are provided to evaluate the adequacy and
accuracy of the fracture trace analysis.
EPA Response: The fracture trace analysis was used merely
as a screening technique and was only one of several
techniques used to identify well locations. The photographs
used to develop the fracture trace analysis were noted.
section 3.3 General; The geophysical techniques used in the
investigation are not referenced and should be.
EPA Response: The techniques used for the geophysical
investigations were described in the text and specifically
Appendix A.
section 3.3 General; This section does not provide a clear
description of why and how the areas used in the surface
geophysical investigation were selected.
EPA Response: Literature concerning the area of the Site
showed that groundwater flow in the area of the landfill
generally follows the surface topography. As such, the
surface geophysical investigations were performed in the
valleys where the monitor well locations were proposed in
the RI/FS Workplan. The surface geophysical investigations
were used as an aid in further locating the wells within the
valley areas identified in the Workplan. Areas were
selected by using fracture trace analysis and field recon-
naisance. The objective of the survey was to effectively
place monitoring wells in fracture zones.
section 3.6 General; This section, describing the packer
tests, is a bit unclear because results are included.
Results were not included in the previous sections
discussing surface geophysical technique. More importantly,
none of the results from the tests have been quantified.
Information on zones that accepted water (Page 3-11) is
important, but it is equally important to know how much
water and how rapidly the water was accepted.
EPA Response: None of the zones on which packer tests were
performed accepted large quantities of water. The packer
tests were not intended to provide quantitative data
concerning the aquifer characteristics, but were a screening
14
-------
59.
60.
61.
technique to be used along with borehole geophysics and
observations made during drilling for locating the well
screens. '
Section 3.6 General; No packer test was conducted in Well
EI, which is reported 'to be downgradient of Keystone
Sanitation Company, Inc. (Keystone) Well 2. This would have
provided useful information regarding hydrogeologic
interconnection between the. eastern portion of the landfill
and the Maryland Wells 1,2, and 3. . .
EPA Response: EPA obtained sufficient information to make a
remedial action decision. This decision provides extraction
wells for the eastern portion and towards MD 1, 2, and 3 as
contamination was detected in this area. Additional
sampling and studying of this area will be conducted during
the Design Phase and in the investigation for Operable Unit
2.
Section 3.7 General; Well locations A,B, and C (Sections
3.7.1, 3.7.2, and 3.7.3, respectively) are located fI...along
the major structural trend." This appears to be based on
Figure 4-2. However, are they placed along major
structures (e.g., a major fracture)? If so, the feature is
never presented in figure form. This is unclear. Also, why
were wells at locations A and E considered continqency wells
(Page 3-12)? How could they be located for optimal
effectiveness before analytical data were collected?
EPA Response: All EPA installed monitor wells were placed
via the surface resistivity surveys described in Appendix A
of the R1 report. Although well A and E were to be
considered contingent and drilled if sampling analysis from
the other wells indicated a need, EPA decided to save
mobilization cost and drill all the wells at once.
Page 3-16; Well location D1 is south of the tributary to
Piney Creek. What is the basis on which this location was
selected? The State of Maryland (Page 1-10) indicated that
ground water in the area south of the tributary "could not
be affected by landfill contamination" because of differ-
ences in hydraulic head. No evidence is presented to
support a contrary conclusion; therefore, it appears that
this well was ineffective in meeting the objectives of the
study.
EPA Response: Well D1 was placed south of the tributary to
Piney Creek in order to support the assertion that the
tributary is a groundwater divide. The water level measured
in this well was higher than the levels in cluster C, which
is located north of the tributary, and therefore supports
the contention that the tributary is a groundwater divide.
As stated in the Workplan on p. 5-3-7, Well D-1 was
installed for the purpose of determining the hydra~lic
15
-------
62.
63.
64.
65.
66.
67.
gradient in this area as the Maryland residents were
concerned with contamination from the landfill.
Pages 3-17 and 3-18; Wells at locations F and I are intended
to provide background water quality data. It cannot be
determined if these are appropriate locations based on
Figure 3-8 (potentiometric surface map) because none of the
wells have been located. Likewise, Plates 3 and 5 present
improbable contour lines (i.e., contour lines of the same
elevation that intersect) at the 740-footelevation.
EPA Response:
Please refer to the response for Comment #13.
Page 3-18; All figures indicate that location H is offsite,
yet the text indicates the well is onsite. Where is this
well located?
EPA Response: Well location H is located adjacent to the
site and is considered to be representative of conditions at
Kl.
Page 3-19; No justification is given for constructing every
well of stainless steel. The data indicate that polyvinyl
chloride (PVC) would have sufficed and would most certainly
have been less expensive.
EPA Response: The community insisted on the use of
stainless steel wells during their comments on the Workplan.
There is a concern that PVC wells may leach VOC contaminants
in the ppb range. The use of stainless steel alleviates
this technical concern.
Page 3-2-; The reviewer must assume the three wells
installed in 1987 were Keystone Wells 6,7, and 8 and that
those are the PVC wells. If this is the case, how were
Keystone Wells 1 through 5 constructed (e. e., diameter and
type of casing):
EPA Response: Keystone well construction details are
contained in the Workplan as well as in reports prepared by
Keystone.
Table 3-2 (Page 3-22); Elevations of the Keystone and
Maryland wells are not provided in the text or a figure.
Therefore, correlation of hydrogeologic and analytical data
with wells installed during the RI is diffi9ult, if not
impossible.
EPA ReSDonse: Well elevations were not available for all
the Keystone and Maryland monitoring wells. The measure-
ments from the RI wells were used to develop the potentio-
metric maps.
Section 3.8.1 General; The RI is not clear on why.
.. .
16
-------
I'
l
approximately 37 percent of surface soil samples are
collected onsite when the.Executive Summary (Page ES-2)
indicates the focus of the RI would be offsite~ In
addition, the number of samples collected in the areas of
the residential wells is not sufficient to make any
determination regarding the source of any potential
contamination in those areas. In fact, the effects of two
potential sources of any subsurface contamination, septic
tanks, and underground stor~ge tanks, would probably not
have been identified even if the samples were collected from
immediately above such a structure. Data from these samples
are misleading because of the their extremely limited
nature. Finally, the three background samples appear to be
adjacent to roadways. These are poor background locations
because of the increased possibility of elevated metals and
semivolatile concentrations due to vehicular traffic and
effects of road maintenance operations (e.g., surface
tarring and chemical vegetation control). No rationale was
given for how analytical parameters were chosen. In
addition, factors governing surface runoff such as type of
vegetative cover, moisture content of soil, or irrigation
area were not discussed. A sample of the irrigation water
should have been analyzed for correlation with the soil
data.
EPA Response: Soil samples were collected near residential
wells only if suspect areas were observed when collecting
the water sample. Since no offsite residential well was
considered contaminated, the soil samples collected near the
wells were not extensively evaluated. Background samples
are selected not to be clean but to be representative of the
area. The analytical parameters selected were those
standard for EPA Superfund sites. A sample of the
irrigation water was not taken as K1 was sampled which would
be representative of the irrigation water.
68.
Table 3-3 (Page 3-26); Source and data of maximum
contaminant levels (MCLs) need to be referenced. Have only
final MCLs been used or have MCLs at various stages of
finalization been used (e.g., those designated as proposed,
draft, etc.)? Similarly, "Primary MCL" needs to be defined.
EPA Res90nse: The source of the MCLs is the Federal Safe
Drinking Water Act. (See 42 C.F.R. ~300f et.sea.). Primary
MCLs refers to enforceable standards which apply to specific
contaminants which EPA has determined have an adverse health
effect on humans. These are usually referred to as MCLs.
The MCLs used are the most current promulgated laws.
69.
Page 3-28; No reason has been provided for conducting grain
~ize analysis on three sediment samples. This has not been
adequately described and is confusing.
EPA Response:
contamination in sediments is primarily fo~: :
17
-------
70.
71.
72.
73.
74.
in the fine size fraction. Grain size analysis which
determines the size of sediment grains is a crucial part of
sediment sampling protocol.
section 3.8 General; No rationale was given for the
selection of analytical parameters.
EPA ResDonse: The selection of analytical parameters was
done in the Workplan and was consistent with normal
Superfund site protocol utilizing the Target Compound List
(TCL) for organics and Target Analyte List (TAL) for
inorganics. In addition, dichlorofluoromethane was analyzed
as it had previously been detected in MD and is common
landfill contaminant.
Page 3-29; What types of pumps and bailers were used to
collect the ground water samples, and how were they
decontaminated?
EPA ReSDonse: Stainless steel submersible pumps were used
to purge the wells prior to sampling. Teflon bailers were
used to collect the samples. The decontamination procedures
were as follows:
o
o
o
o
o
Soap and potable water wash
Rinse with potable water
Rinse with Optima grade methanol
Rinse with high purity, analyte free
Air dry overnight when possible
water
Page 3-29; It is unclear whether a water elevation was
determined in the wells from which no samples were collected
due to obstructions (Keystone Well 2 and Maryland Well 7).
If water levels were obtained, a water sample could be
collected by some means (e.g., a narrow bailer).
EPA Res~onse: Water samples could not be collected for K2
and MD7 due to obstructions. However, water levels were
taken in all wells.
Page 3-29: Why were no ground water samples analyzed for
dissolved metals or turbidity? These results could have
been used to assist in determining if well installation
activities were a likely source of any metals contamination.
EPA ResDonse: Answered in other responses. Filterable and
suspended solids were analyzed on the water samples.
Page 3-31: Residential well sampling techniques have not
been referenced.
EPA ReSDonse: Residential well sampling techniques were
presented in the Field Operations Plan for the Site.
18
-------
I
75.
76.
77.
Page 3-33; Why was oniy water level data plotted from the
RI's shallow wells (Figure 3-8 and Plates 3 and 5) used
Using data from the appropriate Keystone and Maryland wells
would have provided a more controlled/accurate represen-
tation of the potentiometric surface. Water levels from the
intermediate and deep wells should have been plotted to
assist in assessing vertical hydraulic conductivity and
ground water flow directions in the bed rock where flow is
presumably controlled by secondary permeability.
EPA Response: Shallow and intermediate wells represent
different flow regimes. A map combining their water levels
is only representative of regional, not local flow.
Page 3-33; The objectives of the pump test at Keystone wells
were to "...determine the aquifer characteristics at the
site and to qualitatively investigate if Keystone WeIll and
the Mundorf Spring were hydraulically interconnected."
Buchart-Horn, Inc. (B-H) presumably conducted a pumping test
on this well earlier, why was another one conducted here?
Was there some reason to question the results of the B-H
test? Similarly, is water from the spray irrigation system
drawn from this well or is this well in the zone of
influence of the irrigation well? In either case, what
measures were taken to ensure that the aquifer was in steady
state?
EPA Response: Precipitation at the Site during EPA pumping
test prevented the use of the test data for determination of
aquifer characteristics. Buchart-Horn's pump (B-H) test was
performed to determine if the aquifer could be suffi-
ciently pumped as a clean-up tool. B-H's test was for
approximately 72 hours, which is sufficient time to reach
steady state conditions. It is standard EPA practice to
conduct a pump test in a RI to obtain information to make
our conclusions. As the test failed, B & H's data was
examined and considered sufficient not to warrant the
expense of another pump test.
Page 3-33; Regarding the objective of determining whether
there is interconnection with the spring, it is questionable
that this could have been determined in clear weather. This
is due to the relatively large drawdown required to affect a
measurable change in water level and flow at the spring.
The determination was virtually impossible when precipita-
tion and surface runoff enter the picture. The spring
itself must be described: is it one seep or a series of
seeps - Has it been influenced by man (e.g., has it been
piped)?
EPA Response: Due to problems with the pump test EPA was
unable to determine the effect of the ground water system on
Mundorf spring. This will be evaluated during the~econd
19
-------
78.
79.
80.
81.
Operable Unit and/or Design.
Page 3-38: No determination regarding the presence of
wetlands has been made. Although.reference has been made to
their delineation here and on Page 4-62, no documentation
has been provided to support any conclusion regarding the
occurrence of wetlands.
EPA Res90nse: A wetland determination was conducted in May
1990 by the U.S. EPA. Although this information was not
contained in the RI report, it is available in the
Administrative Record. Wetlands were found in several areas
near the site and no impact was indicated from the Site.
Figure 4-1 (Page 4-2): This figure is titled "Landfill
Surface Features," yet very few surface features are shown.
EPA Res90nse: This figure indicates the shape of the
landfill and location of major streams and roads, it was not
meant to be a topographic map of the landfill.
Page 4-3: The photographs and maps used to perform the
fracture trace analysis must be referenced, along with their
respective scales. Similarly, it is unclear when
photographs were taken. It is stated on this page that
photographs were taken in August 1985, yet section 3.1 (Page
3-1) states that photographs were taken in April 1989.
EPA Response: Photographs from both years were used:
however the 1985 photographs were the primary source for the
fracture trace analysis. The scales of the photographs are
the same as the figure. This information is found on pages
4-4 to 4-5 in the RI report.
Page 4-5: The discussion of ground water flow paths, as
determined for the fracture trace analysis, is confusing.
Up to this point, the RI has indicated that flow is
controlled by primary permeability (Figure 3-8 and
discussions of well locations in section 3.7). However, in
this section, the RI states that flow may be deflected along
the strike (direction) of schistosity (foliations or layers
of coarse minerals). Do they mean that two flow regimes
exist in the aquifer? For example, the first is in the
saprolite (weathered bedrock), which is controlled by
primary permeability and, to a lesser extent, secondary
permeability. The second regime is in the bedrock where
flow is predominantly controlled by secondary permeability.
EPA Response: This was to indicate that groundwater does
not flow equally in all directions from the landfill due to
its position on a hill. Instead the groundwater flow is
controlled by the regional strike and local structures such
as valleys.
20
-------
82
83.
84.
85.
86.
Page 4-5; References are needed where "typical"
resistivities are presented.
EPA ResDonse:
The reference is noted in Table 4-1.
Section 4.4.3
resistance of
of high water
high compared
General; At location C, the RI states the
the third layer, 600 ohmmeter, is indicative
or weathered.rock. Yet this resistivity seems
with data presented in Table 4-1. .
EPA ResDonse: The discussion of the 600 ohm-m value merely
suggested the it may the result of high water. The value is
in the range of a shale or metamorphic rock, as shown on
Table 4-1.
section 4.5 General; It appears that much of the information
throughout the geology section was derived from the
literature, yet no references are cited. Was an extensive
mineralogical study of the bedrock and overburden conducted
during the RI? If not, the source of the information should
be cited. If it was performed, then methods and techniques
should be described.
EPA ResDonse: The material presented in this section is
based on the work performed at the site during the RI. No
detailed mineralogical study was performed, other that the
core analysis.
Figure 4-5 (Page 4-17); This cross section indicates that
the thickness of the saprolite is relatively uniform.
However, the text on Page 4-19 states "The saprolite is of
variable thickness and ranges from a thin veneer on ridge
tops to 30 feet in thickness in drainage areas and valley
bottoms." This is a confusing discrepancy that must be
explained. .
EPA ResDonse: The cross section was constructed from wells
located in valley areas near the site. The thickness of the
saprolite on ridges is based on other reports describing the
area.
Section 4.5 General; Certain core samples from the bedrock
were analyzed for extraction procedure toxicity and total
metals. However, none of the samples from the saprolite
were analyzed for parameters that could have been used in
calculating contaminant mobility through the media (e.g.,
organic carbon content, cation exchange capacity).
EPA Response: Sampling the groundwater in the deep aquifer
gives a much more realistic answer in determining contami-
nant mobility (i.e., indicates whether there is contami-
nation) than calculating physical parameters.
21
-------
87.
88.
89.
90.
Page 4-48; The RI states, at this point, that the hydrology
of the site is dominated by two regimes. However, no data
or discussion are presented regarding how the two regimes
are interrelated except that well yields decrease with depth
(Figure 4-14). No hydrogeologic parameters have been
determined from this study. The hydraulic conductivities
used in calculating velocity span 5 orders of magnitude and
no reference for this information has been cited.
Similarly, the values used for the area and outflow rate
variables in the specific discharge equations have not been
defined. The reviewer recalculated average ground water
velocities using the equation presented, the range of
hydraulic conductivities, the hydraulic gradient, and a
porosity of 38 percent (instead of 0.38 percent as
presented). The range calculated was 3.16 x 10-4 feet per
day (ft/day) to 32 ft/day. This is significantly different
than the range of 4.4 x 10.4 ft/day and 4.4 ft/day presented
in the RI.
EPA Response: The difference in the calculated values could
be as a result of using different wells to calculate the
gradient. The wells used in the RI were Hand B.
Page 5-7; The theory that the base/neutral and acids in soil
around the irrigation system ". . . may be a result of the
spray system configuration" needs to be further described.
EPA Response: A greater number and higher concentration of
these compound were found in this area than in other areas
where soil samples were collected. The significant differ-
ence was the presence of the spray irrigation system.
Page 5-22; The statement is made that the source of the
metals in offsite surface water was not site-related.
Conversely, it is mentioned that volatile organic compound
(VOC) contamination is site-related. Neither of these
claims are substantiated.
EPA Res~onse: Because there was no discernable pattern of
inorganics detection in groundwater samples from offsite
monitoring wells, the site was not considered to be the
source. Because volatile organics were detected in K1 and
cluster H and are typical constituents of landfills, the
landfill was considered to be the source of these compounds.
section 5 General; General comment - What is a site-related
contaminant? Is there information available that has
sufficiently characterized the material in and beneath the
landfill?
EPA Response: A site related contaminant refers to
contaminants detected onsite and detected in a migration
pattern in an offsite locations indicating it may.have come
22
-------
91.
92.
93.
94.
from the site. No information was obtained during the RIto
characterize the waste.
Table 5-5 (Page 5-25); Range of mercury concentrations (0.2
to 0.4 parts per billion [ppb]) is incorrect. Keystone Well
8 sample contained 3 ppb mercury according to analytical
data.
EPA Response: Table 5-5 and Figure 5-5 are both incorrect
for mercury. The level detected at K8 was 2.5 ppm as stated
in the RI on pages 5-32 and in Appendix E, Volume III.
Corrections will be made to the RI through the issuance of
errata sheets to the Administrative Record for the
appropriate document.
Page 5-30; An inference is made in the last paragraph that
VOC contamination at MD-2 (approximately 1,500 feet south of
the site) is related to the site because it is on the same
structural trend as Keystone Well 1. Therefore, Keystone
Well 2 (from which no sample was collected) should also be
on the same structural "trend." However, no contamination
has been identified in any of the wells identified by the RI
as being downgradient of Keystone Well 2. In fact, Maryland
Well 2 was never mentioned as being downgradient of Keystone
Well 2. If these wells are on a controlling structure, why
wasn't a well placed closer to the site? The RI's
conclusion is not substantiated by the data.
EPA Response: Maryland Well 2
E was placed close to the Site
Maryland Well 2. Page 3-17 of
downgradient of K2.
is downgradient of K2. Well
and is located between K2 and
the RI states it is
Page 5-32: The last sentence in the first paragraph
indicates that the ground water sample for Keystone Well 5
contained 1,3200 micrograms per liter (2ug/l) copper, which
exceeded the "Federal Standard" of 1,000 ug/l. However,
Table 5-6 indicates there is no federal MCL for copper.
Which is correct and what is a "Federal Standard?"
EPA Response: Although there is no primary MCL for copper
usually referenced to as Federal MCL, there is a federal
standard which is the secondary MCL. The secondary MCL is
used to regulate the aesthetic quality of drinking water.
The text should read secondary MCL, not Federal Standard.
Page 5-39: No VOCs were identified in any off-site
residential wells, even though several wells (RW-4, RW-5,
and RW-S) appear to be situated on the "dominant structural
trend." No mention is made of this in any conclusions
relating to the other off-site wells, which were installed
during the RI and appear to be contaminated.
EPA Response:
The RI did note that contamination in the
23
-------
95.
96.
97.
98.
offsite residential wells was not found during the RI.
section 6.0 General; In general, this section lacks
references, particularly Table 6-2.
EPA Response: The primary source of data for section 6.0 is
EPA's "Water Related Environmental Fate of 129 priority
Pollutants, Vol.l". .
Page 6-6; contaminant persistence in the environment is
usually defined by its half-life degradation rates under
various conditions such as hydrolysis, oxidation,
biodegradation, etc. Organic carbon and soil water
partition coefficients usually refer to contaminant
mobility. Persistence and mObility data indicate how long a
contaminant will stay in the environment and the direction
it will move. Unfortunately, these parameters were
discussed generally in qualitative terms rather than
quantitatively.
EPA Response:
No response needed.
Table 6-2 (Page 6-9); No source of the fate and transport
data is cited.
EPA Response: Answered in Response 95.
Page 6-10; The statement is made that "Ground water analysis
has confirmed VOC contamination emanating from the Keystone
site." This is not true. The data show only that VOCs were
present in several samples from off-site wells. The total
lack of hydrogeologic data precludes such a statement from
being made regarding the source, however likely a source the
landfill may be. The RI has given no explanation for the
fact that VOCs were not identified in other downgradient
wells (e.g., MW-EI, MD-l, MD-3, RW-3, RW-5, etc.). It
appears that the RI authors are basing their interpretation
of the site hydrogeology on the analytical data rather than
results from an adequate investigation of aquifer
characteristics. Finally, the underlying deficiency in
making any interpretation of where the source is located is
that it is unclear which direction ground water is flowing
from the site. Does it flow radially from the site, as
indicated from the figures, or does it flow along the
"dominant structural trend," which has not been clearly
defined.
EPA Response: Groundwater flows from the site in several
directions due to its location on a topographic high. The
primary flow paths are valleys originating near the
landfill. This flow regime contrasts to typical situations
where groundwater flows through a site in a single
direction. No explanation can be given for why VOCs were
24
-------
'/
not identified in other downgradient wells without being
speculative. It could be none were present or none were
present at that depth. The RI was to provide facts.
99.
Page 6-11; The section on surface water is unclear.
many springs were identified? Did any of them dry
seasonally? What are their elevations, etc?
How
EPA ReSDonse: Detailed information concerning springs in
the area was not collected~ This information can be
addressed in the second Operable Unit for offsite
contamination.
100. The landfill was identified as a source area based on
groundwater sampling at the perimeter. No sampling within
the landfill was performed during the RI, so no waste
characterization is available (other than historical data).
In order to definitively identify it as a source, additional
sampling and analysis within the landfill is required to
characterize and quantify the landfill material.
EPA ResDonse:
Please refer to comment No. 24
101. As stated in Section 2.2.2.3 (Page 2-4), "chemical-specific
applicable or relevant and appropriate requirements (ARARs)
do not exist for the contaminants present in soil or
sediment at the Keystone Sanitation Company, Inc. site,
based on a review of federal and state regulations."
Direct contact risks have not been calculated for the soils
in the spray irrigation area. Remedial action objectives
should not be established unless there is a risk exceeding
the point of departure.
Risks exceeding the point to departure were not identified
for surface soils. Remedial action objectives and ARARs are
not identified in the absence of risk.
EPA ResDonse: The sampling revealed that a release of
contaminants has occurred as a result of the spray
irrigation system. The remedial action objectives for
in the spray irrigation area are intended to provide
protection of human health and the environment from
potential risks in addition to remediating the release.
102. The list of contaminants of concern for ground water
identified in the remedial action objectives (Section
2.2.3.4) is different from those identified in the risk
assessment (Section 1.5). The risk assessment must clearly
define those contaminants which contribute,to an incremental
increased cancer risk greater than 1 x 10-6. That
information can then be used to identify the remedial action
objectives. This issue is further discussed in constituents
of concern.
soils
25
-------
1-
EPA ReSDonse: Risk is not. the only criterion used to
determine contaminants of concern with regard to
remediation, specifically treatment. The risk assessment
identified contaminants which pose excess cancer risks. The
lists in the FS also include compounds in the groundwater
that may affect the implementation or operation of a
treatment process.
103. A definitive list of the contaminants of concern must be
clearly defined in the risk assessment and adequately
supported by the RI data. This same list can also be used
in the evaluation of remediation alternatives.
EPA ResDonse: The previously collected data was considered
in the development of the Workplanand can be found on this
Administrative Record. Due to questions concerning QA/QC,
it can not be quantitatively used in the remedial decisions.
However, this information is considered important and can be
used to substantiate data from the RI. The objective of the
RI was to determine contaminant migration from the site.
Data generated would determine the need for remedial action
and be used in developing and evaluating remedial
alternatives.
104. Page 2-13, ARARs; Is the Resource Conservation and Recovery
Act (RCRA) an ARAR? This site was a municipal solid waste'
(MSW) lan~fill, not a hazardous waste treatment, storage, or
disposal facility. They should develop their ARARs more
carefully to show why RCRA, etc., are ARARs.
EPA ResDonse: EPA guidance identifies RCRA as an ARAR for
Superfund landfill sites. Therefore, RCRA was noted as a
potential ARAR for this Site. However, the cap design was
based upon PA Municipal Landfill Closure, not RCRA, as RCRA,
is not considered to be applicable to the site. But the
Offsite Land Disposal requirements in 40CFR 268.1-268.50 are
be considered relevant and appropriate for the disposal of
the sludges from the air stripper.
105. Page 2-15, RCRA; Have characteristic hazardous wastes been
identified at the site? Remcor is not aware of any sampling
at the site. The constituents of concern are probably the
result of MSW disposal or pre-RCRA disposal.
EPA ResDonse: The landfill itself was not sampled, and
characteristic wastes were not identified at the Site.
106. Section 2.2.2, General;'overall, the ARAR's discussion and
development are very poor. Direct applicability of the
ARARs has not been properly developed.
EPA ReSDonse: EPA disagrees and feels that all applicabl~
or relevant and appropriate requirements were properly
2.6
-------
identified and used at this site.
107. Sections 2.4 and 2.5, Technology Identification and
Screening; In the identification and screening of technology
types and process options sections of the FS (sections 2.4
and 2.5), five potential feasible options were unjustifiably
rejected. These options should be used as parts of the
developed remedial action alternatives (Table 2-8, Page 2-
59) and further considered in the detailed evaluation of
alternatives. . .
EPA Response: Each of the "five potential feasible options"
addressed in Remcor's comments have been explicitly
discussed in various sections of the FS. To clarify:
o
EPA determined that there was no need to relocate the
onsite residents since the risk is coming from the
ingestion of groundwater which can be treated.
Relocating the On-site Residents - The residents are
the PRPs and may choose to relocate if desired. As an
option, this is considered much more expensive and more
difficult to administer than point-of-use treatment.
o
Subsurface drains - These were not rejected, but were
not used for costing purposes. During the design, this
process option will be further evaluated.
o
Activated Carbon Adsorption - Since only VOCs are a
concern at this point in time, air stripping was
determined to be less costly but equally effective for
VOC removal at the site.
o
Discharge to Groundwater - This is a more costly means
of effluent discharge and not necessary based on
technical considerations.
o
Conventional Excavation - This process option was not
rejected (see page 2-50).
108. section 3.3, Description of Alternatives; Remcor has seven
comments on the selected alternatives (Section 3.3), which
were identified as a result of the screening process. These
comments are related to the technical appropriateness and
feasibility of the alternatives. Most of the comments
relate back to the rejected options identified in Comment ":
EPA Response:
See previous response.
109. Appendix C, Cost Estimates: After a review of the cost
estimate tables shown in Appendix C, Remcor has found the
U.S. Environmental Protection Agency (EPA) cost estimates ~
be seriously flawed. The EPA's costs are significantly
27
-------
underestimated, as discussed below and shown in Table 1 of
this attachment:
o
A contingency of only 10 percent is not adequate for
cost estimates based on preliminary planning of
remediation alternatives. Changes in the scope of work
due to design refinements, estimated quantities, permit
requirements, and site conditions are likely to occur.
Remcor feels a contingency of 20 percent would be more
appropriate. .
o
The capital cost estimate for the single-layer clay cap
is grossly underestimated. Remcor estimates a
construction price of $10.5 million compared to the
original estimate of $4.9 million. The increase in
cost is due to an error in the estimated quantity of
filter fabric and sand, and in the in-place unit costs
for top soil, common borrow, and clay. Details of
Remcor's estimated costs are shown in Table 1.
o
The capital cost estimate for the multimedia cap is
also grossly underestimated. Remcor estimates a
construction price of $10.9 million compared to $6.5
million. This underestimate is due to error on the
same five items as the single-layer cap. Details of
these additional costs are also shown in Table 1.
EPA ReSDonse: The cost estimates proposed by Remcor reflect
a difference in design and in costs of materials. The FS
cost estimate is based on a preliminary design to meet PADER
and/or RCRA regulations, and on discussions with PADER
officials regarding commonly accepted practices for landfill
closures in Pennsylvania. Remcor apparently has increased
the thickness of various layers resulting in an increase in
material cost. The unit costs used for clay and compaction,
topsoil and spreading, and common borrow were obtained from
vendors and costing guides. A 10% contingency is standard
for preliminary designs. As stated in EPA RI/FS guidance,
the FS is expected to provide an accuracy of +50 percent to
-30 percent for the "study estimate" costs.
R. E. Wriaht
110. Resistivity electrode spacing was not sufficient to reliably
permit location of bedrock fracture zones; since the depth
to bedrock ranged from 20 to 40 feet.
EPA ResDonse: R.E. Wright (REWAI) reviewed the Work Plan
for both the Borough of Littlestown (4/14/88) and for Alloy
Rods (4/28/88).. In the former, REWAI had no comments
regarding the geophysics, whereas in the latter, REWAI only
suggested that EPA use resistivity to determine monitor well
locations. Because the use of resistivity for fracture
delineation must be very precise, EPA had its Envtronmental
28
-------
Monitoring Systems Laboratory in Las Vegas acquire the
service of Gregory Hempen,. a geophysicists with the st.
Louis Corps of Engineers to perform the survey. Mr. Hempen
is a CPG with many years of experience in the field of
geophysics and is currently the president of the AEG.
Mr. Hempen used both Wenner and Dipole-Dipole profiling to
detect fracture signatures for monitor well placement.
Although the Wenner arrays ~pacing may have been too large,
the data did show that the water path through the zone was
influenced by bedrock fractures. However the Dipole-Dipole
array was used to confirm this suspicion. The Dipole-
Dipole array depth had a 50-foot spread and multiple depths
so that depth to bedrock was assured.
111. Borehole geophysics did not utilize the temperature logging
required by the Work Plan to recognize the location of
water-bearing zones and determine the presence and direction
of vertical groundwater flow in the boreholes.
EPA Response: Although temperature logging was not
performed during the RI, a suite of other geophysical
packer tests and cores were performed for fracture
delineation.
logs,
112. Historical on-site groundwater quality data was not
integrated into the RI.
EPA ReSDonse: Historical onsite groundwater quality data
was used in the development of the Workplan and included in
section 1 of the RI. The RI was to focus on offsite
problems to determine if contamination was migrating
offsite. This was done as Keystone was already evaluating
the onsite contamination under a Consent Adjudication with
PADER.
113. EPA has arrived at RI results which do not support the
alternative proposed.
EPA ResDonse: The RI has shown that there is enough water
in the weathered zone to support a pump and treat system.
Additionally, Buchart-Horn, Keystone's consultant, has
proposed the same remedy as EPA, pump and treat, in its
January 1990 Supplemental Groundwater Plan.
114. The FS requires more of a clean up than necessary because
EPA did not take into consideration the fact that Keystone's
on site bedrock wells were open for long intervals causing
incorrect determination of specific contaminant migration
patterns in multiple aquifers.
EPA ResDonse: The fact that the K-series wells are open for
long intervals only serves as stronger evidence that ground-
water cleanup is necessary because such long intervals
29
-------
dilute and under-estimate contaminant concentrations. since
other sources of contamination in the area do not exist, it
is possible that onsite clean up may require more, not less,
of an effort.
115. Filtered Samples should have been taken.
EPA Response: EPA agrees and will be resampling the surface
water, residential wells, sediment and groundwater taking
both filtered and non filtered samples. This data will be
used in the Remedial Design.
116. The well heads were not secured properly.
EPA Response: The wells were secured after construction was
completed. It was noted during the public meeting that the
wells may presently be unsecured. EPA will address this
problem and redevelop wells as needed in the future.
117. The pump test data is not valid due to problems conducting
the test.
EPA Response: EPA agrees that-there were problems with the
72 hour pump test, and this was reported in the RI.
However, Buchart-Horn's pump tests, which ran successfully
indicated that remediation of the aquifer via pump and treat
is feasible. This remedy is consistent with that suggested
by EPA.
118. REWA I has questioned EPA's understanding of the site geology
because REWAI purports that EPA has correlated bedrock
foliation to former layering on P 4-20, 2nd paragraph in the
RI.
EPA Response: The 2nd paragraph on p 4-20 in the RI notes
that "Metamorphism has imparted a distinct, continuous
slatey cleavage (foliation) subparallel to the original
sedimentary bedding. Cleavage is subparallel to the
original sedimentary bedding". The narrative does not
purport what REWAI has indicated. The statements have not
indicated that foliation can be correlated to former
bedding.
119. EPA's fracture trace analysis was used only to place wells
in valleys and the aerial photos used were of little use due
to vegetative cover.
EPA Response: EPA did not indicate in the RI that valleys
were used solely as evidence of fracture traces. EPA used
aerial photos to identify potential locations for monitor
wells, then field verified those locations with surface
30
-------
geophysics and examination of natural outcrops and other
surface features. Additionally, the presence of vegetation
can help identify fractures.
120. Problems with EPA's pump test resulted in little
information.
EPA Response: Although there were problems with EPA's pump
test results as noted in the RI, the Buchart-Horn
investigation, which successfully completed pump tests in
1990, clearly indicated that the aquifer could be stressed
due to pumping and that such a remedy is feasible for
Keystone. This information is more useful that K values
within proper orders of magnitude.
121. The RI vacillates between alternative models of groundwater
flow.
EPA Response: Water moves differently in unconsolidated
saprolite and consolidated bedrock. The RI has stated this
as two appropriate models for groundwater flow.
122. No mention was made of whether the aquifer was confined or
unconfined.
EPA Response: EPA used Buchart-Horn's 1990 pump test data
which reported the aquifer as unconfined.
123. The depth of the proposed recovery wells is not correct.
EPA Response: EPA's has indicated that the exact
specifications for the pump and treat system will be
detailed in the Design documents which will be made
available to the public for review and comment.
R. H. Sheppard Co.
124. A sufficient number of
to establish naturally
chromium, and, lead.
background samples were not collected
occurring levels of barium, cadmium,
EPA Response: The three background soil locations, two
background surface water and sediment locations, and two
background groundwater locations are sufficient to establish
levels of naturally occurring inorganics. In addition, rock
from the cores was analyzed for inorganics. General
references concerning typical ranges of inorganics in a
region cannot be used to support contaminant decisions at a
Superfund site.
125. Barium, cadmium, chromium, and lead should not be considered
contaminants of concern in soil.
EPA Response: No remedial action is proposed specifically
as a results of the occurrence of these inorganics.in soil
31
-------
samples.
126. Barium, cadmium, chromium, and lead should not
contaminants of concern in surface water since
. background samples were collected and regional
values should have been used.
be considered
insufficient
background
EPA ReSDonse: These compounds were not found in
concentrations significantly above ARARs in the surface
water. Also see previous response concerning background
samples.
127. Background surface water samples should have been collected
west of the site.
EPA ResDonse: The streams located west of the site are down
slope and originate near the landfill. For this reason,
background samples could not be collected west of the
landfill.
128. Barium should not be considered a contaminant of concern in
sediment as the concentrations found were less than reported
background values for the region.
EPA ReSDonse: Regional background values reported in the
literature are not used to establish background locations at
a particular site. Sufficient background samples were
collected to establish local background conditions.
129. Barium, cadmium, chromium, and lead should not be considered
contaminants of concern in groundwater based on its
presences in the rock and soil, and frequency of detection.
EPA ReSDonse: While metals detected in the groundwater at
offsite areas are not a present concern, metal were detected
in onsite wells. Thus, the nature of landfill waste, and
the concern for future releases require that a treatment
system capable of removing a range of metals from the
groundwater be provided.
130. Barium, cadmium, chromium, and lead should not be considered
contaminants of concern in residential wells based on its
presences in the rock and soil, and frequency of detection.
EPA ResDonse: The residential wells in the area were found
not to be contaminated as a result of the site.
Littletown Hardware
131. Chemical composition of selected core intervals and EP TOX
of selected core intervals were analyzed in the ppm range,
not the ppb range that is required for Water Samples. Thus,
this data is of little utility.
32
-------
I
I
EPA ReSDonse: The EP toxicity conducted on the core samples
show that arsenic and barium may be leachable under some
circumstances. It was a qualitative analysis, not
quantitative. -
132. Water samples for metals analysis were not filtered
to preservation.
prior
EPA ReSDonse: It was not required in the-Workplan.
sampling during the Design and Operation Phases will
of filtered and nonfiltered samples. These analysis
used for the design of the treatment system.
133. During well development, Littlestown Water was used although
the water was sampled the RI did not provide information
regarding the condition of the tank that transported the
water, the length of time the water remained in the tank at
the site, or whether the sample was obtained when the water
was loaded or unloaded. These questions are important if
the tank was constructed of reactive material.
-Future
consist
will be
EPA ReSDonse: The tank was constructed of non-reactive
plastic material.
134. EPA should consider the use of buffered distilled water for
well development as the use of Littlestown water as a
development fluid may have caused invasion into the
formation and formed complexes which remained in the
formation.
EPA ResDonse: The potential for potable uncontaminated
water to cause contamination problems in an aquifer is
minuscule. There is no research in this area that EPA is
aware of that has determined that this is a problem.
Additionally, the purging of wells before sampling prevents
purging of stagnant potentially "reactive" water from
analysis.
135. The sandpack for the monitor wells consisted of pre-washed,
graded 10/20 silica sand. The pre-wash fluid was not
discussed and an analysis of the sand for organic and
inorganic constituents was not presented for the sand
pelletized bentonite seal or cement bentonite grout.
EPA ReSDonse: The sandpack, bentonite, and cement used for
monitor well installation was all GSF approved and NS
standard.
136. Duplicate samples were collected from monitor wells BS and
FS. It is not explained in the RI whether one sample was
collected at each well, preserved and then split into two
samples for analysis or whether two samples were collected
at each location. The suspended solids result from MW-BS is
33
-------
. -
,
3x greater than the duplicate and 2x as great for MW-FS.
EPA ReSDonse: The duplicate samples were collected
independently. Natural variations in the suspended solids
are not considered to be significant.
137. Results for the monitor wells were presented on Table 7-1 in
the RI. This table noted that antimony was detected in 2 of
30 samples, which is inconsistent with that reported on
Table 5-3A in the RI.' . .
EPA ReSDonse: Editorial problems such as these will be
addressed by placing errata sheets in the Administrative
Record with the RI and FS.
138. The computation of background coverage for barium for
monitor wells F and I is incorrect.
EPA ReSDonse:
The comment is noted and will be checked.
139. Table S-3A in the RI does not indicate that general water
quality analysis samples were not taken at MW-AI and MW-GI
which is inconsistent from that stated in the RI on page 3-
31.
EPA ReSDonse: Editorial problems such as these will be
addressed by placing errata sheets in the Administrative
Record with the RI and FS.
140. Page 3-29 of the RI states that all MW's except K2 and MD7
were sampled however, Table 5-3A has no data for MW-MD-1
EPA Response: Editorial problems such as these will be
addressed by placing errata sheets in the Administrative
Record with the RI and FS.
141. Residential well survey results should have included a
description of plumbing materials or location of the tap.
EPA Response:
Comment noted.
142. Residential well samples were collected after a 20
flushing period or until a holding tank volume was
discharged. The methodology used in measuring the
tank discharge was not described.
EPA Response: A standard formula used across the industry
that determines the time required to empty one tank volume
was used.
minute
holding
143. Calculation of a corrosion index for each residential well
would give an indication of the potential for metals
contribution from plumbing materials and further clarify the
origin of metals in water samples. The presence of copper
34
-------
and zinc in all of the samples may be partially the result
of plumbing
EPA ResDonse:
The remedy does not address copper .or zinc.
144. Failure to describe land use practices or conditions that
may influence soil sampling results is scientifically
dishonest, .and a full description and disclosure are not
only important but mandatory. .
EPA ReSDonse: Soil sampling was done to establish how the
Site affected natural soils since clean-up is based upon
site related contamination. Pesticides were not
historically considered to be a problem so an in depth study
of former land use practices for these chemicals was not
necessary to study site cleanup.
145. Drift of spray according to prevailing winds coupled with
surface runoff to nearby drainage channels may explain the
presence of metals in surface water and sediment samples.
EPA ReSDonse:
contamination.
Yes, and this may also indicate site related
146. The pumping of K-1 has caused all of the off-site
contamination.
EPA ResDonse: K1 may have contributed to offsite
contamination.
147. Will EPA place a monitoring well clusters to investigate
ground water flow between K8 and K4, between K4 and K7 and
between K1 and K2.
EPA ResDonse: At this time, it has not been determined
where the wells will be placed. This decision will be made
during the Design Phase after additional studying and
sampling of the area has taken place. The public will be
given an opportunity to comment on the design.
GEOSERVICES
148. Page ES-1 Paragraph 3: C.C. Johnson and Malholtra (CCJM)
states that the RI was implemented to supplement previous
reports. However, data from previous reports were not
included in the RI or the FS. Therefore,the purpose of the
RI was not met by subsequent activity on the project. In
the same paragraph, it is stated that the primary objectives
were to evaluate impacts although existing laboratory data
were not included in the evaluation. The CCJM study relied
completely on samples which were collected during the single
sampling event and which included a limited number.of wells
35
-------
relative to those sampled during previous investigations.
Had CCJM included the existing database in the analysis, the
significance of anomalously high trace metals concentrations
could have been evaluated and subsequent mistakes and
erroneous conclusions founded upon this data could have been
avoided. In addition, had CCJM relied on existing data, the
RI/FS process and subsequent report would have been more
thorough, more accurate and less costly.
EPA Response: The previously existing data base was
evaluated in the Work Plan for the site and is available in
the Administrative Record. A major problem with the
previously existing data base is the lack of information
concerning data validation and QA/QC procedures. This
problem is compounded by the number of sources of the data.
149. Page ES-1 Paragraph 5: It is stated that surficial and
subsurface shallow soils were sampled to determine if
offsite soils had been affected by contaminant migration
from the site. Additionally, it is stated in the same
paragraph that surface water and sediment samples were taken
along with ground water samples to determine the extent of
contamination emanating from the site.
The report concludes that no site related contaminants were
detected in offsite soils near the site or along surface
drainages. Additionally, the report concludes that no
site-related BNAs, pesticides, or PCB were detected in
surface water or sediments, and that metals which were
detected above fresh water chronic water quality criteria
bore no direct connection to the Keystone site. The report
also concluded that no site related pesticides or PCBs were
detected in ground water from off site monitor wells.
Although several metals were detected in off site monitoring
wells, CCJM concluded that their occurrence was random and
apparently not site-related. The report further concluded
that the relative abundance of metals in shallow monitor
wells may result from natural weathering processes. The
report also concluded that there was no indication of metal
contamination emanating from the Keystone site and in nearby
residential wells. Finally, metals contamination was not
detected in any on site surface soils.
Clearly, in accordance with this information, the most
recent soils and groundwater sampling data indicate that no
release or threat of release of metals has occurred, either
to onsite soils, to offsite soils, or to either ground
water, surface water, or stream sediments. In the absence
of any observed release or threat of release, there is no
need to consider design of a remedial system although under
the circumstances, monitoring or other precautionary
measures may be considered prudent. Additionally,
assessment of risks to human populations or to the
environment must be carefully evaluated within the. overall
36
-------
context of the actual site conditions.
EPA ReSDonse: While there is no direct risk to human health
and the environment from the identified metals in offsite .
areas, there is potential for metals migration from the
landfill based on metals found in onsite wells and
characteristic of the landfill in general.
150. Page ES-5/ES-6 Paragraph 8:. The report states that exposure
pathways include groundwater, surface water, and soil routes
by means of inhalation of airborne particles originating
from site soil waste or due to dermal contact. We have
clearly shown in previous sections of the report that metals
concentrations are not present as a dissolved phase but
rather as a particulate phase within the aqueous regime.
Therefore, dermal contact could only be established due to
showering, bathing, or swimming in extremely turbid water.
Additionally, metals concentrations in particulate matter
within the leachate are little different from naturally
occurring metals concentrations in soils and rock materials
which constitute the geologic setting of the region.
EPA ReSDonse: The metal concentration does exist as a
particulate phase within the aqueous regime. These
particulates can be inhaled during showering.
151. Page ES-6 Paragraph 2: Several metals are listed as major
contaminants which include chromium, nickel, cadmium, and
beryllium. These compounds are immobile in the natural
environment due to their existence as a solid phase.
Therefore, trace metals present no more of a potential risk
to human health than does naturally occurring bedrock or
soil particulates derived from the regional bedroc~.
EPA ReSDonse: Responded to previously in Comment No. 150.
Dissolved trace metals can potentially pose a problem at any
Superfund site.
152. Page 1-1: One of the stated objectives of the RI was to
develop a cost effective remedial action plan. Development
of a more than $10 million remedial alternative involving
groundwater extraction, capping, and construction of a water
treatment plant to treat contaminants in which neither a
contaminant release nor threat of release has been observed,
can hardly be regarded as cost effective.
EPA ReSDonse: The remedy
on-site VOC contamination
migration from the site.
documented the release of
site.
primarily addresses the risk from
while preventing any further meta:
As reflected in the RI, EPA has
a hazardous substance from the
153. Page 1-9 Paragraph 1: The CCJM states that EPA/NUS found
elevated levels of lead, chromium, cadmium and mercury in
37
-------
site monitor wells during the 1984 investigation. These
sample results have been conclusively identified as
statistical outliers representing the concentrations of
particulate matter within an unfiltered sample.
Furthermore, subsequent sampling demonstrated that these
measurements were entirely non-reproducible and therefore
unreliable. Therefore, no such elevated levels of lead,
chromium, cadmium, or mercury can be considered to have
been detected when included with the sampling and
preservation protocol which was employed.
EPA ReSDonse: EPA detected the presence of inorganic
contamination in the RI. The remedy in the Proposed Plan
addresses this concern. Additional sampling during the
Design Phase will include both filtered and nonfiltered
samples. It may be determined in the Design Phase based on
new sampling results that metal treatment is not necessary.
154. Page 1-11 Paragraph 3: CCJM states that review of the
previously existing database shows that there is
insufficient information to define the hydrogeologic regime
and to characterize the extent of existing contamination.
This statement immediately follows the conclusions developed
by the state of Maryland in its 1986 report which provides a
number of conclusions regarding the three-dimensional
configuration of the groundwater flow system dominating the
area. The Maryland report provides a detailed discussion of
the vertical and horizontal extent of the flow system
relative to discharge of groundwater derived from Keystone
Landfill.
Additionally, CCJM offers no explanation for omission of the
data from hundreds of ground water samples which had been
collected at the time of report preparation and prior to the
sampling event conducted by CCJM. As will be developed
below, the stated purpose of collecting additional
information (to determine potential contaminant migration
routes) was not served by the subsequent investigations
since CCJM failed to conduct a single successful test of
aquifer characteristics or to provide any meaningful
additional data by which the Keystone ground water flow
system could be further refined.
EPA Response: As indicated in a earlier comment and an
earlier response, the existing data was evaluated and the
quality of the data was questioned. Sufficient information
as reflected in the RIfFS is available to evaluate the
groundwater flow systems in the area of the Keystone site.
155. Page 2-9 Paragraphs 1 and 2: CCJM states that aquifer tests
by Keystone and the State of Maryland, as well as a pumping
test. conducted by Buchart-Horn, were used to determine
aquifer properties or to characterize the hydraulic behavior
of wells completed within the site aquifer at the 6ite.
38
-------
Although CCJM claim that the existing database was
insufficient to define th~ hydrogeologic regime at the site,
the existing database is repeatedly cited to define the site
hydrogeologic regime. This is in contrast to the almost
complete omission of any assessment completed by CCJM. We
therefore question the need for the additional work
completed by CCJM embodied in the RI or, in fact, the need
for-any additional work at the present time.
EPA ResDonse: The study completed by Buchart-Horn was not
sufficient to study the offsite contamination. The RI study
and data was necessary and did provide sufficient
information to make a remedial decision. The pump test data
by B & H further supports this decision.
156. Pages 3-1 to 3-20: An extensive program of surface and
borehole geophysics was conducted, presumably for the
purpose of monitor well siting and delineation of aquifer
characteristics. Whereas no evaluation of the
three-dimensional ground water flow regime or the vertical
distribution of ground water contaminants is presented
anywhere within the report, we see the extensive geophysical
program described in this section to be wholly unnecessary
and blindly extravagant.
EPA Response: The purpose of the surface geophysics was for
monitor well placement, whereas the borehole geophysics were
used for determination of bedrock fracture zones. This type
of work is routinely preformed at Superfund sites in
fractured bedrock.
157. Page 3-20 Paragraph 3: The report states that, although the
wells were fully developed, some wells continued to produce
pale yellowish-brown water indicating a high percentage of
particulate matter. When sampled under the protocol
employed by CCJM, such samples could only produce
anomalously high metals concentrations not representative of
dissolved aqueous phase metals.
EPA ReSDonse: Additional sampling well take place during
the Design Phase which will include filtered and unfiltered
samples. The final design of the treatment system will be
determined by these results.
158. Page 3-28 Paragraph 1: CCJM states that surface water
samples were collected from well-mixed (flowing) reaches of
streams or springs by direct immersion of a sample
container. While CCJM states that some effort was made to
avoid collection of suspended materials by sampling surface
water prior to sampling sediment, streams draining the
region characteristically contained significant quantities.
of suspended materials due to the fine grained nature of the
soils which mantle the region and which are drained by the
streams. Suspended materials are especially present
39
-------
following rainfall events. The presence of suspended
materials in stream water. (particularly those in well
flowing reaches of streams or springs) imparts a
characteristic green or brown color to the streams of
Southeastern Pennsylvania. Again, unfiltered samples
collected from streams draining the area and preserved using
a pH of 2 could only be expected to contain elevated levels
of metals. Actual concentrations of metals is dependent on
the volume of particulate matter collected along with the
sample and processed during laboratory analyses.
EPA Response: The surface and sediment samples of the RI
provided sufficient information for the proper remedy.
Additional sampling of the surface water and sediments will
be done during the design and on a regular basis as
indicated in the Proposed Plan. This information can verify
the presence of metals. The proposed remedy does not reme-
diate the surface water or sediment directly, but indirectly
by lessening the amount of groundwater discharge.
159. Page 3-29 Paragraph 2: During a description of the sample
collection procedure, it is noted that the well cases, was
pumped dry rather than fully purged. cavitation of the
well normally results in a significant increase of the
volume of particulate matter in the well water. This would
again lead to excessive metals concentrations due to sample
collection procedure rather than as an accurate measure of
metals content of groundwater in the vicinity of the well.
EPA ResDonse: The slow recharge rate of the bedrock is the
reason the well was pumped dry. This has the same effect as
purging 3 to 5 well volumes in a more transmissive aquifer.
160. Page 3-33 Paragraph 3: A seventy-two hour constant rate
pumping test was conducted to determine aquifer
characteristics at the site. During the test, response of
water levels in 11 wells and a spring were monitored
continuously. The results of the pumping test were
reported in a subsequent section with the success of the
test characterized by distorted drawdown results. During
the test, only the pumped well and a single observation well
displayed decreasing water levels. Other wells showed an
increase in water levels due to rainfall effects.
Additional comments relative to the inadequacy and folly of
the RI pumping test are included in a subsequent comment.
EPA ReSDonse:
Please refer to comment No. 76.
161.
Page 4-1 Paragraphs 2 and 3: The report states that
precipitation falling on the landfill can flow northward
over ground surface or seep downward into the landfill to
the groundwater below. In following paragraph, it is stated
that no eroded gullies or seeps were visible at the
landfill. It is presumed therefore, that CCJM envisions
40
-------
precipitation draining the landfill as a sheet of water over
the ground surface rather than collecting in any gullies,
seeps or swales. This circumstance is highly unlikely and
actual site conditions provide little potential for
transport of surficial materials from the landfill by sheet
wash. .
EPA ReSDonse: Sheet wash was not implied in the report.
What was noted was that the landfill cover material does not
appear to be eroding.
162. Pages 4-5 to 4-15: Surface geophysics are discussed in
order to identify the three-dimensional structural framework
in the vicinity of the landfill. As would be expected,
discrete fracture zones are present within the area and the
depth to fresh bedrock varies significantly throughout the
area. The summary presented at the end of the section
states that weathered near surface rock may be the dominant
aquifer until major fractured/weathered zones are
encountered. In both cases, the report concluded that major
avenues of groundwater migration are through significantly
weathered zone~ presumably containing an abundance of clay
minerals and iron oxides which are the normal weathering
products of the rocks which underlie the area. Both
materials have been previously cited as primary adsorption
sites for metals adsorption. significant mobility of trace
metals through these materials is, therefore, not possible
under these conditions.
EPA ResDonse: There is a potential for metal to migrate
from the site. This will be further investigated through
the filtered and unfiltered samples taken during the Design
Phase and through the investigation of Operable unit 2.
163. Page 4-16 Figure 4-3: The figure summarizes the geology of
the area indicating that the saprolite is composed of
weathered schist and that the schist itself is composed of
chlorite, muscovite, and finely disseminated pyrite along
with other minerals. Chlorite, muscovite, and pyrite have
been shown to contain extremely high concentrations of trace
metals. These concentrations are high enough to be
expressed as weight percent rather than as ppm. Therefore,
the entire saprolitic sequence must be regarded as a source
of trace metals which are elsewhere identified as
contaminants of concern by CCJM.
EPA ReSDonse: The report indicates that this may be the
case in some instances.
164. Page 4-19: Fracture zones are characterized as ranging in
thickness from 1-inch to approximately 1-foot and containing
crushed schist with small amounts of finely braided rock
material (clayey gouge). Additionally (in the third
paragraph of Page 4-19), saprolith is characterized as a
41
-------
pale yellowish brown clayey soil containing remnant schist
and quartz fragments. In the following sentence it is .
characterized as consisting of reddish-brown, brownish-gray
and pale grayish-green clay and heavily weathered schist.
In all cases, the areas through which groundwater migration
is focused is associated with clay minerals and iron oxides
(imparting the reddish or brownish coloration to the clayey
sequences). Materials underlying the site represent an
enormous volume of potential adsorption sites for any trace
metals which would be transported as a dissolved phase in
ground water migrating through these areas. Furthermore,
not only are adsorption sites found everywhere within the
aquifer, but the aquifer itself represents an enormous
source of trace metals by virtue of the minerals contained
within the sequence.
EPA Response: The comment seems to say that there is a
background presence of metals in the groundwater and at the
same time there are sufficient adsorption sites that no
metal can be present in the groundwater. There is potential
for metals to migrate from the site. This will be further
investigated through the filtered and unfiltered samples
taken in the Design Phase and through the investigation for
the Operable unit 2.
165. Page 4-20 Paragraph 1: The report states that chemical
analyses were performed on rock cores from boreholes and
that these analyses indicated that only barium and chloride
were readily leachable from rock. During bench testing by
GeoServices, Ltd., soils present within the area were shown
to represent a significant source of chromium, lead, and
nickel when leached with a strongly acidic solution such as
the solution in which the samples were preserved. There-
fore, although the minerals are present in the materials
which were cored by CCJM, their own analyses demonstrate
that those minerals are immobile. Finally, trace metals are
present both in the landfill leachate and in virgin soils
within the same naturally occurring minerals and mineral
phases which are present in the rocks underlying the site.
EPA Response: The EP Toxicity testing conducted on the core
samples show that arsenic and barium may be leachable under
some circumstances.
166. Page 4-21 Table 4-2: Chromium and lead were detected in the
two cored intervals at maximum concentrations of 77.6 and
73.5 ppm respectively. These data are consistent with those
reported by the State of Maryland for similar analyses. As
was previously demonstrated, digestion of these samples as a
consequence of the preservation process employed by CCJM
causes anomalously high dissolved metals concentrations in
aqueous samples when particulate matter is included in the
samples which are analyzed.
42
-------
EPA Response: The comment in noted. EPA proposed remedy
does address the metals as they were detected at the site
and offsite in the RI. However, during the Design Phase,
additional sampling of the monitoring wells will take place
to substantiate the need for metal treatment and to aid in
the design of the unit processes.
167. Page 4-22 Table 4-3: Table 4-3 is a summary of the EP
Toxicity analyses of selected core intervals which indicates
that chromium and lead were not present in the leachate. EP
Toxicity analyses of materials deposited in the landfill by
Alloy Rods showed similar results. Again, little difference
exists between the behavior of materials placed in the
landfill by Alloy Rods and naturally occurring materials
which characterize the environment. There is no justifi-
cation, therefore, to conclude that materials placed in the
landfill by Alloy Rods represent any potential threat to
human health or the environment. Such thinking would
require one to conclude that similar threat is present
throughout the region as a result of the chemical
composition of the rocks through which groundwater flow is
concentrated.
EPA Response: The RIjFS or proposed plan makes no
conclusion concerning Alloy Rods involvement with this site.
168. section 4.6 Page 4-20 to 4-28: The entire section is a
discussion of the results of a borehole geophysics survey
conducted on monitoring wells drilled by CCJM. The
discussion is largely academic and duplicates information
concerning water bearing zones which should have been
available from well logs from the geologist who logged the
borings. Normally, water bearing zones are recognized and
logged by the geologist present during well construction
and are recognized by the appearance of additional water
emanating from the hole, the physical characteristics of the
cuttings produced during drilling, and other features.
EPA Response: The small quantity of water encountered in
the deep boreholes and the size and extent of the fractures
prevented the identification of all potential water bearing
zones during the drilling. Borehole geophysics was
performed to assess water bearing fracture zones and its use
was supported by the communities in Pennsylvania and
Maryland.
169. Page 4-29: The comparison of geophysical logs with core
descriptions was based on the relationship between various
geophysical factors to features observed during core
'examination. . No new information was provided by the
application of borehole geophysics and therefore, this
application represents an'unnecessary and extravagant
expense.
43
-------
EPA ReSDonse: The borehole geophysics which is normal
protocol for Superfund provided a relatively inexpensive
means of evaluating subsurface conditions in all of the
boreholes. Borehole geophysics was performed to assess
water bearing fracture zones and its use was supported by
the communities in PA & MD.
170. section 4.7 Hydrogeology: As a general comment to this
section, CCJM concludes that most of the water occurs at
depths generally less than 45 ft and is contained in the
saturated saprolite and weathered schist. Additionally,
CCJM concludes that porosity and hydraulic conductivity
decrease significantly with depth and that little water is
contained within the unweathered schist. Therefore, the
ground water flow system emanating from Keystone Landfill is
strongly controlled by topographic divides (both ridge
crests and valley axes) which limits any recharge migration
regardless of its chemical makeup.
EPA ResDonse: Although the groundwater flow from the
Keystone site is strongly controlled by topographic divide,
it does not prevent recharge from occurring.
171. Page 4-53: An attempt is made to estimate groundwater
velocity in the absence of any direct measurement by CCJM of
porosity or hydraulic conductivity of any of the units
present at the site. CCJM assumes that hydraulic
conductivity ranges between 10-2 to 103 gpd/ft2 (a range of 5
orders of magnitude) and that porosity is 0.38 percent.
Under these conditions, CCJM estimates that groundwater
velocity ranges between 4.4 X 10-4 to 4.4 ft/day. This range
of estimated velocity (between 2-inches and nearly a third
of a mile/year) represents little more than a guess at
groundwater velocity. Such an estimation of the critical
hydraulic parameters upon which any risk assessment or
subsequent remedial system design is based constitutes a
blatant misapplication of scientific method and is
completely inappropriate to remedial system design. This is
especially onerous where the primary stated motivation of
the design is to protect human health and the environment.
EPA ResDonse: A detailed evaluation and determination of
aquifer parameters was not planned at the Site and does not
affect the evaluation of the site as presented. These
parameters will be obtained, if necessary, during the
remedial design phase.
172. Page 4-54 Paragraph 2: CCJM discuss the Buchart-Horn
pumping test and concludes that the storage values reported
by Buchart-Horn are low for unconfined aquifers. On the
contrary, the storativity values reported by Buchart-Horn
are typical for unconfined bedrock aquifer or where flow
through fractures affects the numerical magnitude of the
storativity value. In such situations, storativity is a
44
-------
reflection of a degree to which observation wells are
hydraulically connected to the pumping well, rather than
reflecting the storage capacity of the aquifer.
. EPA Response:
No response needed.
173. Page 4-54 Paragraph 3: The CCJM study further summarizes
the Buchart-Horn report stating that the flow regime of the
pumping well was linear along strike rather than radial
across strike. It is unclear what CCJM is attempting to
indicate by this statement in that the flow system in the
immediate vicinity of a single fracture is linear within a
discrete distance from the fracture, and thereafter becomes
radial. There is no relationship between flow system
configuration and bedrock strike as indicated by this
statement. statements offered by CCJM in this and the
preceding paragraph indicate an extremely limited
understanding of the hydraulic system in the vicinity of a
pumping well in fractured rock terrain. Furthermore, CCJM's
demonstrated lack of understanding of fractured rock
hydraulics is a direct precursor of problems with design of
a groundwater capture system in such terrain.
EPA Response: The statement was meant to indicate that the
aquifer response to the pumping test was reflective of
fractured bedrock.
174. Page 4-154, section 4.7.2.1 - RI Pumping Test: During the
CCJM pumping test, only the pumping well and a single
observation well displayed decreasing water levels due to
pumping. All other wells showed a significant increase in
water levels during the pumping test as a result of a
rainfall event. If the rainfall event was the sole failing
of the test, the test should have been rescheduled in that
the characterization of the hydrogeologic regime was the
stated reason for conducting the remedial investigation.
Data is not presented in the CCJM report concerning the
pumping rate which was employed at the pumping well nor were
any calculations offered in data that was available. It is
therefore difficult to assess what the actual cause of the
pumping test failure was, although based on review of
Buchart-Horn pumping test data results and upon GeoServices,
Ltd. experience within the Wissahickon Formation, the
observation well configuration relative to the discharge
rate at the pumping well was probably too widely spaced to
allow any significant drawdown results to occur in the
observation wells as a result of pumping. Therefore,
failure of the CCJM pumping test was based on an improper
pumping test design as opposed to an untimely rainfall
event.
EPA Response:
EPA has noted the failure of the pu~p test in
45
-------
the RI due to precipitation. To avoid additional costs,
Buchart-Horn pump test data was reviewed and found
sufficient for the needs of the investigation.
175. Page 4-56 Paragraph 2: CCJMpostulates that a hydraulic
connection exists between well K-1 and Mundorf spring on the
basis of topographic and potentiometric relations. The
placement of this conjecture in the section on pumping test
analysis results is curious in that the statement is not
supported by any quantitative measurements developed by
CCJM, especially not by the pumping test completed by CCJM.
Therefore, there is no basis for this statement.
EPA Response: The figure
connection between K1 and
the same valley as Kl, is
downgradient from Kl.
does in fact indicate the
Mundorf Spring. The spring is in
downslope from Xl, and is
The pump test was originally designed to provide
quantitative information regarding the connection between
the Spring and groundwater.
176. Page 4-62: CCJM states that no signs of stressed vegeta-
tion, damaged crops, or extreme environmental concerns were
observed during a site visit to the site and surrounding
area. This is not surprising and is consistent with the
absence of any observed release of contaminants from the
landfill.
EPA Response: The fact that no environmental damage to
flora or fauna was noted during a site visit does not
preclude a possibility of groundwater contamination.
Additionally VOC contamination was noted in the on-site
drinking water well which indicates a release.
177. Page 4-62 Paragraph 4: CCJM reports that the streams are
less than three ft wide and only a few inches deep.
Therefore, risks associated with swimming in streams which
drain the area, dermal contact with stream water, and
ingestion of stream water appear non-existent.
EPA Response: The dermal contact risk from exposure to
surface water was noted in the Risk Assessment as being 2 x
10 -8 .
178. Page 4-65 Paragraph 1: CCJM notes that stream sediment
contains a significant portion of clay sized particles wit~
relatively strong exchange capacities. The sediments also
contain high contents of total organic carbon. Again, any
aqueous phased metals present in stream water would be
quickly and irreversibly adsorbed onto stream sediments, i~
such materials were present.
46
-------
EPA ReSDonse: Although metals may soak onto fine particles,
the rate of sorption depends upon other important water
quality criteria. The resampling of surface water and
sediment that will be conducted in the future should clarify
this issue.
179. Page 5-9 Table 5-1 Entitled "compoundjAnalytes Detected in
On site Soil Samples": CCJM indicates that the range of
chromium on site soil samples fell between 13.4 and 22.6
ppb. Lead ranged between 9.2 and 80 ppb and nickel ranged
between 6.1 and 29.1 ppb. These concentrations should be
evaluated in the context of naturally occurring metals
levels in off-site soils and bedrock as reported by both
CCJM, and the University of Maryland. In all cases, trace
metals concentrations in on- site soils were at least an
order of magnitude less than those in background samples.
Therefore, trace metals in site soils can hardly be
considered to represent a release or to constitute any
threat to the human health or environment. Furthermore, it
is noted that these samples were collected (or include
samples collected) in the spray irrigation area in which
landfill leachate was directly discharged in an attempt to
control VOCs. Therefore, these soils represent worst case
circumstances relative to metals contamination.
EPA ReSDonse: Additional sampling of the groundwater during
the Desigr Phase will obtain both filtered and unfiltered
samples. If the results determines that metals are
naturally occurring, the treatment system design could
change to treat only VOCs.
180. Page 5-12 Paragraph 5: The report states that all Toxic
Analyte List (TAL) metals except cadmium, silver, and
cyanide were detected in onsite soil samples. The report
correctly notes that all metals detected on site were also
detected in core samples. However, the report states that
metals were detected in similar concentration ranges in core
samples, which is inaccurate. Metals present in core
samples were measured at the ppm range while metals present
in soils were measured at the ppb range. Therefore, metal
concentrations in the cores with three orders of magnitude
(1000 times) higher than those in on site soils. (sic)
CCJM's failure to distinguish between ppm and ppb is
symptomatic of the carelessness which characterizes the
entire remedial investigation.
EPA ReSDonse: The report should have noted that the
detected metals in the cores were in ppm; the higher
detection was for qualitative information only.
181. Page 5-13 Paragraph 2: Concentration ranges for offsite
samples are noted to be similar to those for onsite sample~~
Therefore, there is no difference between metals
47
-------
concentrations in soils onsite or offsite indicating that no
release has occurred to spils at either location.
EPA ReSDonse: There does appear .to .be similar concentration
in the onsite and offsite soils. However, it is evident
that a release has occurred to the groundwater. The most
probable source would be the landfilled wastes, not the
soils surrounding the landfill.
182. Page 5-14 Paragraph 4: CCJM notes that surface water
samples were taken from locations with clear flowing water.
The water depths were between 6 and 12 inches. As has been
previously described, stream sediment samples
characteristically contain a significant portion of clay as
well as other fine grained materials. Therefore, even
though the water may have appeared clear during sampling,
the presence of the flowing water indicates that sufficient
entrainment velocity was present to suspend fine grained
particles in the stream water sample. Additionally, in that
these samples were unfiltered, it is obvious that the sample
contained particulate matter. As has been previously
described, particulate matter, especially chlorite and other
clay minerals which are platy in nature and easily
suspended, contain significant concentrations of trace
metals. Assuming that the stream samples were preserved in
a solution of pH 2, it is only natural that metals
concentrations would appear in laboratory analyses.
EPA ResDonse: The proposed remedy does not provide for
remediation of surface waters and sediments. EPA will be
resampling the surface water (filtered & unfiltered) and
sediment in the future because of these concerns and to
clarify the issue.
183. Page 5-18 Section 5.4.3: The report states that chromium,
copper, and lead were all detected at sample location SW-16.
In the case of lead at sample locations SW-11 and SW-15 lead
concentrations were less than 20 ppb. CCJM states that
these concentrations of these metals were above the fresh
water chronic water quality criteria (AWQC). Additionally,
mercury was detected at several sample locations and zinc
was also detected at SW-I6 at nearly 68 ppb. The report
concludes that the pattern of metals occurrence indicates
that the source is not site related. Therefore, the
occurrence of metals as a result of laboratory analyses and
their exceeding the AWQC is completely immaterial to an
evaluation of a release of contaminants to surface waters in
the vicinity of the site. We are gratified to find that, in
at least one case, CCJM has correctly recognized that minor
metals concentrations in a sampling media are not site
related. We do note, however, that CCJM felt it necessary
to qUalify this statement with the modifiers "not likely"
(5-22) and "probably" (5-23). The word "clearly" should be
substituted for both modifiers.
48
-------
EPA Response: The AWQC was exceeded at the site and is of
concern to EPA. A second Operable Unit for offsite
contamination will further investigate this concern. It has
not been totally ruled out that the metals are coming from
Keystone. .
184. Page 5-23 Paragraph 4: CCJM correctly states that Keystone
wells were constructed as Qpen boreholes with 20 ft of
surficial slotted casing. Therefore, wells K-1 through K-8
provide a direct communication between unconsolidated fine
grained landfill materials, soils, and the well bore.
Vigorous pumping of the well in order to purge the well bore
prior to sample collection could result in entrainment of a
significant volume of particulate matter. Again,
preservation in an acidic media results in generation of
dissolved phase metals which are not present within the
groundwater regime.
EPA Response: The wells were not vigorously pumped during
purging, but at a slow rate. It is standard EPA practice to
purge wells at a slower rate than at which they are
developed in order to avoid turbid samples.
185. Pages 5-24 and 5-29; Tables 5-5 and 5-6: These tables
summarize analyses detected in on-site and off-site monitor
wells respectively. We note that the magnitude of maximum
metals concentration is similar in both settings indicating
little difference between sample populations. However, the
presentation of the range of all measured concentrations
overshadows the anomalous nature of metals concentration
maxima. In all cases, the concentration maxima listed by
CCJM represent anomalous and irreproducable (sic) events.
These events are related to inclusion of particulate matter
in samples extracted from the wells and the preservation
protocol employed by CCJM. Therefore, any conclusions based
on these results. must be regarded with extreme caution and
in context of the artificial nature of the analytical
results.
EPA Response: The data collected in the RI is sufficient on
which to base the selection of the remedy. EPA's resampling
activities in the future will include both filtered and
unfiltered samples. This data will be used to determine the
design.
186. Page 5-27; Figure 5.5: Chromium concentrations in excess of
the MCL are indicated at wells MW-AS and well K-S. Well K-5
is a well from which the leachate residuum sample was
collected. Because of the method of casing, the well
induces fine grained soil materials into the sample. This
well contains an abundance of particulate matter when
pumped. Additionally, the low yield of the well results in
cavitation of the pump during the course of sample-
49
-------
collection further leading to introduction of a high volume
of particulate matter. W~ll MW-AS was constructed during
the course of the CCJM study.
EPA ReSDonse: Chromium was detected above MCLs in MW-AS.
This well was properly constructed and was properly pumped
to prevent cavitation of the pump.
187. As noted by the state of Maryland report, and from the
statistical analyses presented in this report, newly
constructed wells typically contain significant volumes of
particulate matter as a result of incomplete well
development following construction. Unfiltered samples
reflect the inclusion of particulate matter in the sample
during laboratory analysis. Finally, mercury is indicated
at well K-8 at a concentration of 3 ppb. Mercury
concentrations in rain water are also within this order of
magnitude. In all three cases, the data are not
reproducible which would be the case if a plume containing
dissolved metals was present at the landfill. These data
further demonstrate that no observed release of metals from
the landfill has occurred.
EPA Response: At this time, there is no data to support
that the release of metal is not from the landfill. The
data collected is sufficient on which to base the selected
remedy. Future sampling activities will include both
filtered and unfiltered samples and will be used in the
Remedial Design.
188. Page 5-31 Paragraph 5: The report states that chromium was
the only metal that exceeds the federal MCL and was
collected at well K-5. We again reference previous
comments regarding the volume of particulate matter present
in well K-5 and the lack of reproducibility of any
analytical results from that location. A filtered sample
from well K-5 collected at the time of leachate residuum
collection yielded a concentration of chromium below
detection limits.
EPA ResDonse: This comment should have said chromium was
the only metal that exceeds MCLs and was collected at K5 and
MW-AS. See response to Comment No. 186.
189. Page 5-32 Paragraph 1: CCJM correctly acknowledged that
metals were randomly distributed in off-site areas. This is
an indication that no metals bearing plume is emanating from
the landfill. High concentrations represent point source
anomalies. These specific metals maxima have been
identified with statistical outliers resulting from
inclusion of particulate matter in water samples.
EPA ResDonse:
Please refer to Comment No. 187.
50
-------
190. Page 5-32 Paragraph 2: The occurrence of mercury, chromium,
and copper at Keystone we~ls K-8 and K-5 has been previously
referenced. These concentrations are due to the fact that
samples were not filtered prior to collection and were
preserved at a pH of 2. Additionally, observed mercury
levels were within the same range of mercury measured in
normal rainfall.
EPA Response: Please refer, to Comment No. 187. The level
of mercury detected exceeded the MCL and is of concern to
EPA.
191. Page 5-33 Paragraph 3: The CCJM report concludes that metal
detected in off-site monitor wells are random and apparently
not site related. The report further concludes that the
relative abundance of metals in shallow monitor wells may
result from natural weathering processes. We agree with
this conclusion but again would omit any indefinite
modifiers in that the mineralogical analyses and chemical
analyses previously described in this report clearly
indicate that metals originate from naturally occurring
materials in the Keystone environment or exist as a solid
phase due to adsorption from solution. In this case, no
difference exists between particles containing metals which
originate from adsorption or metals that originate from
natural processes.
EPA Response:
Please refer to Comment No. 187.
192. Page 5-39 Paragraph 2: CCJM states that chromium, copper,
and zinc were detected in the RW-1 sample. Although RW-1 is
a residential well, it is completed near the geographic
center of the landfill and therefore, can not be considered
as representative of potential residential well conditions.
Any risk assessment based on the chemistry of well RW-1
should be considered as a risk assessment of exposure to
landfill leachate. The well is operated by the owner of the
landfill. Chromium concentration was 3 ppb while copper and
zinc were 26 and 3000 respectively. Considering the
location of the well, metals levels are extremely low,
supporting the non-existent mobility of chromium
demonstrated in previous sections of this report. CCJM
notes that copper and zinc were detected in all residential
well samples. Additionally, the report had previously
stated that off-site occurrence of metals was random and not
site related. Therefore, it is interpreted that the
presence of copper and zinc in RW-1 and in other off site
residential wells reflects the presence copper and zinc
plumbing fixtures rather than copper and zinc in ground
water. Finally, the inclusion of well RW-1 as a residential
well is questioned in that it is similar to the monitoring
wells located at the landfill and is completed in the heart
of the area for which leachate generation is assumed.
51
-------
EPA Response: RW1 was constructed for
and is used for residential purposes.
states, results indicate that organics
and do enter the groundwater. .
residential purposes
As the comment
and inorganics can
193. Page 5-40 Paragraph 5: CCJM states that no indication was
found of metal contamination emanating from the Keystone
site in the nearby residential wells. This further
demonstrates that no release of contaminants has been
observed under any circumstances or by any investigator.
Therefore, it is clear that no direct remedial action for
metals is necessary.
EPA Response: The remedial action for metals is based on
the potential for inorganics to migrate offsite and the
uncertainty of the source of offsite inorganic
contamination. Additional sampling during the Design Phase
will substantiate the need for the inorganic treatment. The
second Operable Unit will address the offsite contamination.
194. Page 6-5 Paragraph 7: CCJM lists five potential contaminant
transport routes. Because of the physical and chemical
nature of metals behavior in the environment, none of the
transport routes postulated by CCJM are realistic. Each one
is examined below:
o CCJM: "Leaching from contaminated landfill wastes
to the underlying ground water." As has been
demonstrated during both mineralogical analyses and
bench testing, metals present within the landfill
environment are strongly bound to hydrous iron oxide
particles or to mineral particles which are present
within the leachate solution. Therefore, metals can
only migrate adsorbed onto a particulate phase. Metal
concentrations present are quickly adsorbed on the
particles of the landfill leachate. However, in the
absence of these particles, metals would be quickly
removed from solution by adsorption onto iron oxides
and clay minerals which have been documented to line
fractures through which bedrock flow is concentrated.
Therefore, no route exists between landfill leachate
and the ground water regime.
o
CCJM: "Volatilization of contaminants from the
landfill waste to the atmosphere." Metals are not
volatile and therefore will not, under any
conditions, volatilize to the atmosphere.
o
CCJM: "Discharge from ground water into nearby
surface water." Again, discharge transport by
ground water could only be as an adsorbed phase on
particulate matter because of the volume of
aquifer between the landfill and any surface water
body to which discharge would occur. Any
52
-------
195.
particulate matter present within the ground water
regime will be filtered out by the aquifer.
Additionally, the extremely slow rate of ground
water migration is insufficient to entrain
particulate matter. Finally, whereas fracture
planes are lined with clay minerals and hydrous
iron oxides as a result of weathering along those
surfaces, any metals present in solution will be
directly adsorbed by those materials limiting
migration in the groundwater regime. .
o
CCJM: "Surface water/sediment interaction." As
opposed to a leaching from sediments to surface
water the reverse process would occur even if
metals were present as a dissolved phase in
surface water. This phenomenon is caused by the
presence of organic materials, clays, and iron
oxides within the stream sediments. These
materials have an extremely strong affinity for
trace metal adsorption and attenuation.
o
CCJM: "Discharge from groundwater to surface
soil." It is assumed that CCJM is referencing
. volatilization of organic compounds from ground
water to overlying surface soils. However, as
previously stated, metals are not volatile. CCJM
may also be referencing discharge from ground
water to soils surrounding springs or seeps.
However, we offer the same comments as referenced
above. No known mechanism could lead to a
discharge of metals from groundwater to surficial
soils.
EPA Response: In response to the bulleted item above: No.
1 - Metal particles in the landfill leachate will mix with
the ground; No.2 - Organic contaminants do volatilize to
the atmosphere; No.3 - The discharge of metals to the
surface water may be limited but is possible and No.4 - The
surface water metals would be adsorbed to the sediments and
No.5 - This could also refer to contaminated groundwater
contacting clean soils in the vadose zone.
Page 6-6 Paragraph 4: CCJM states that clays tend to absorb
anions and that metals such as chromium and arsenic tend to
be mobile in the environment. As demonstrated in our
evaluation of mineralogy and chemical behavior of metals,
hydrous iron oxides bear a strong affinity for anions such
as chromium or arsenic and therefore, results in the
adsorption in these compounds from solution. Finally, it
has been demonstrated that metals are immobile in the
Keystone landfill site environment as well as in the
surrounding ground water regime.
53
-------
""
I
EPA Response:
Please refer to Comment No. 187.
196. Page 6-8 Paragraphs 1 and 43: CCJM states that chromium was
detected in monitor wells and surface water samples, above
federal MCLs and AWQC. Lead was detected in all media
sampled and was above AWQC in three surface water samples.
We question these statements in that CCJM themselves
acknowledged that off-site occurrence in metals in off-site
media is random and not si~e related even though AWQC is
locally exceeded. Finally, we have demonstrated in several
different ways that chromium maxima are limited to monitor
wells improperly completed in fine-grained sediments and
directly open to the landfill (Keystone monitoring wells) or
in recently constructed wells. Again, it has been
conclusively demonstrated that no such chromium release
occurred.
EPA Response: EPA's monitoring wells were not improperly
completed in fine grained sediments. The statement only
notes where detection exceeded standards which are of
concern to EPA. It does not acknowledge the source of the
metals. However, chromium was detected in onsite wells
upgradient from the offsite well detection location.
Although a concern, it cannot be directly attributed to a
source. The second Operable unit will address the offsite
contamination concern.
197. Page 6-11 Paragraph 5: CCJM states that there are
indications that contaminants from the landfill are
migrating into surface water as a result of groundwater
discharge or surface runoff. As previously stated and
conclusively demonstrated, this has never been observed in
any media, including surface water in the vicinity of the
site, or as surface runoff.
EPA Response:
Please refer to Comments No. 158 and 178.
198. Page 6-12 Paragraph 5: CCJM states that contaminants adsorb
onto and desorb from sediments in springs and tributary
channels. Again, no release of metals from surface water or
stream sediments has been observed nor is there any
likelihood of any such a release as previously described.
EPA Response:
Please refer to Comments No. 158 and 178.
CITIZENS COMMENTS
199. citizens are entitled to an equal voice with EPA, in
developing and implementing a cleanup program
EPA Response: In accordance with sections 113 and 117 of
CERCLA, 42 U.S.C.~~ 9613 and 9617 EPA allows the citizens an
opportunity to participate in the developing and
implantation of a cleanup program during the requiTed public
54
-------
comments periods for the RIjFS and the Proposed Plan. In
addition, EPA will also provide the opportunity for the
public to comment on the Remedial Design.
200. Any plan, in order to be acceptable must include offsite
cleanup.
EPA Response: At this time, the RI date did not indicate
offsite contamination that presented a significant health
risk to human health and the environment. . The current
proposal allow the contaminants offsite to naturally be
diluted and degrade while preventing any further
contamination through the use of extraction wells. An
additional study is being planned by EPA to further
investigate the offsite contamination in the second Operable
Unit and the current proposal is considered an interim
remedy as opposed to a final remedy.
203. citizens feel a health survey should be done of the
community.
EPA Response: EPA plans to pursue this request through the
Agency for Toxic Substances and Disease Registry.
204. Can a true line be drawn around 35 acres and say the real
onsite contamination is restricted within these lines?
EPA Response: The answer is no. The site has been defined
as onsite and offsite for the effort of the RI.
205. Why weren't all the wells in a 3 mile radius tested?
EPA Response: Although the population count for the Hazard
Ranking Score was calculated for a three mile radius of the
Site, this is not meant to imply that all people within this
radius will be affected by the site. In any investigation,
the residents closest to the site are sampled with a gradual
widening radius of sampling until no contaminants are
detected. The purpose is to determine if the residents are
being affected by the site, not to determine necessarily if
they have good water quality. Many contaminants which may
be detected in this three mile radius may be from
another source which is not purpose of this investigation.
206. The Risk Assessment was an error, how can you have an
accurate assessment if the data is erroneous?
EPA RESPONSE: The data used in the Risk Assessment was not
erroneous. It was the technical methodology which required
modifications. These modifications did not change the total
risk at the site from the 10-4 risk range and therefore did
not change the remedial decision requiring a. response
action.
207. The RI failed to meet its primary objective regarding
55
-------
offsite contamination and therefore, the development of a
remedial action cannot forth go unless it is an interim
action. .
EPA ReSDonse: The RI did meet its primary objective to
determine if contamination is migrating offsite. Adequate
information was obtained to determine a need for a response
to stop the migration of contaminated groundwater. The
proposed plan could be considered an interim action as a
second operable unit will study further offsite
contamination concerns.
208. Numerous comments were received concerning corrections to
statement made in the RI concerning site history, .
population, size of landfill, and leachate disposal should
be placed in the public record.
EPA RESPONSES: All comments will be placed in the
Administrative Record in the Hanover Public Library.
209. Soil sampling did not adequately evaluate the area north and
northwest of Clouser Rd and west between locations SL-12 and
SL-12 and Sl-17. .
EPA ResDonse: pr.evious EPA investigation revealed
contaminant migrating from east and south of the site so
more samples were taken on these sides. However, adequate
sampling \7as conducted of the north and northwest areas to
characterize them. Additionally, the operable unit 2 will
further investigate this area.
210. Does the presence of contaminants CBNAs)
Residential wells indicate hydrogeologic
between the landfill and the residential
the BNAs come from?
in offsite
communication
wells? W11ere do
EPA ResDonse: No, the presence of BNA in the residential
wells do not indicate hydrogeologic communication between
the residential wells and the landfill. Only 3 BNAs were
detected in 1 onsite well. The source of BNAs in the
residential wells is unknown.
211
The extent of the off groundwater contamination has not bee~
determined.
EPA ResDonse: A second operable unit will address the
offsite groundwater contamination.
212. will EPA honor the request to do sampling in a one mile
radius as soon as possible?
EPA ReSDonse: EPA plans to do additional residential well
sampling as soon as possible. Members of the community hd.'
been asked to assist in the development of the sampling. "
56
-------
determination has been made on the radius of the sampling.
Surface water and sedimen~s will take place during the
Design Phase~
213. Identification of contaminant pathways and receptors has not
been shown.
EPA Response: Both personnel from EPA and EPA's contractor
ICF were involved with the environmental assessment
including a sensitive receptor survey for-this site.
Contaminant pathways and potential receptors were
identified. Data from the RI revealed little impact to the
terrestrial plants and/or wildlife near the site.
214. How can the RI identify pathways and receptors if deeper
residential wells have not been investigated?
EPA Res90nse: The additional residential sampling will take
into consideration this comment concerning sampling the
deeper residential wells. This will also be considered in
the investigation of second operable unit for offsite
groundwater contamination.
215. Why didn't the RI consider the 2 area proposed surface
impoundment?
EPA Response: The focus of the RI was to investigate
contamination from the Keystone Site. The surface
impoundments would be outside the scope of work and
inappropriate for this investigation.
216. Will EPA require Keystone to give us water.
EPA Response: At this time, there has not been determined a
need to supply residents with water. Should it be
determined that such need exists, EPA will consider its
enforcement options in obtaining a PRP response or would
supply the water or supply the water through federal action.
217. will data gap in the nature and extent of contamination be
filled in by a continued RI?
EPA Response: Additional sampling will take place in the
predesign and design phase.
218. Groundwater sampling is not sufficient to monitor offsite
flow and fractures.
EPA Response: The purpose of groundwater sampling was to
assess contamination in discrete fractured intervals.
Sampling does not monitor flow; water level measurements
taken during the RI determine groundwater flow.
Additionally the purpose of sampling is not to monitor
fractures; EPA's wells were carefully placed and drilled l~
57
-------
fractured zones so that proper samples could be obtained.
219. How can the nature of subsurface geologic structure be
complete considering data gaps?
EPA Response: The data gaps noted during the RI were for
groundwater sampling only. Surface geophysics, using
borehole geophysics and packer testing of monitor wells was
much more than sufficient for characterizing the geologic
structure at this site. . . . .
220. Why weren't all of the wells in a 3-mile radius tested? The
onsite area should be enlarged to include the 2 1/2 to 3-
mile radius around the site.
EPA Response: The 3 mile radius population count around a
Superfund site is used by the EPA in the Hazard Ranking
System (HRS) , as a screening tool for a comparison of
population of all sites on the National Priorities List
(NPL) for the purpose of prioritization. However, the "3-
mile radius" does not mean that all of the population within
this distance is affected by the site. The purpose of the
Remedial Investigation (RI) is to identify the extent of
contamination and potentially affected receptors. For most
Superfund sites across the county receptors affected by
ground-water contamination level less than 2 miles from a
site.
221. Why is there not concern over each new residential well
being drilled?
EPA Response: If the local population is concerned about
contamination in new domestic well the local government
should require either testing or a moratorium period of
residential well drilling the area. It is not in EPA's
jurisdiction to prevent the drilling of new wells. However,
the EPA has worked with other counties in Pennsylvania (e.g.
Chester County) in this issue and would be willing to do the
same with the appropriate local governments in the vicinity
of the Keystone site.
222. If Mr. Noel's well at 600 feet is the most contaminated,
should the test wells be considerably deeper?
EPA Response: The residential wells in this area are
drilled very deep in order to obtain water from many
fracture zones as well as to provide a means for storage of
groundwater in the well during periods of non use. Thus,
the contamination found in Mr. Noel's well is not
necessarily from fractures at 600 feet, but from fractures
shallower in the aquifer. The purpose of the monitor well
drilled by the EPA was to assess which discrete fracture
zones were contaminated EPA's monitor wells were drilled to
approximately 300 feet to assess the depth of cont~mination:
58
-------
it is highly
in .fractured
area without
wells.
improbable for contamination to have migrated
intervals greater than 300 feet all across the
'leaving a trace in EPA's deep offsite monitor
223. Can you guarantee the proposal will correct the problem in
writing?
EPA Response: All of EPA's, decisions are documented and
available to the public for review. EPA's proposed plan has
noted the time frame expected for clean-up as well as
contingency plans if problems arise with the remedy.
224. How do you determine community acceptance?
EPA Response: In accordance with Sections 113 and 117 of
CERCLA, information from EPA's public meeting as well as
comments sent in to EPA concerning the RI/FS and Proposed
Plan are carefully evaluated to d~termine community
acceptance. opinions of elected officials and community
groups are also carefully evaluated.
RISK ASSESSMENT COMMENTS
225. Page 7-1 Paragraph 1: CCJM states that risk assessments are
performed to evaluate the impact of the no action remedial
alternative and to assess if actual or threatened releases
pose risks to individuals under current or future exposure
circumstances. As has been demonstrated, no actual release
of hazardous substances from the landfill has been observed
to date, even though monitoring of ground water and other
site media has been conducted for nearly a decade. At the
present time, the landfill is inactive indicating that no
new mechanisms or contaminants will be introduced into the
landfill. Therefore, in that no release has been observed
during the period when the landfill was active, or in the
ensuing time during which monitoring has been conducted,
there is no reason to believe that such a release of metals
could occur. Addi tionally, 'CCJM has failed to identify
plausible routes of metals migration from the landfill or
to cite any plausible mechanism of contaminant release. As
stated by CCJM, the purposes for conducting a risk
assessment are not present at Keystone Landfill. Therefore,
such an assessment, if motivated by these criteria is
unwarranted.
EPA Response: The sampling data collected at and near the
landfill during the RI attest to the fact that a release has
occurred and there is a threat of future release from the'
site.
226. Page 7-1 Paragraph 6: CCJM states that data used in the
59
-------
evaluation were limited to those collected by CCJM. No
reasons were given for excluding the hundreds of chemical
analyses collected over previous investigations. The
limitation of analytical measurements to a single sampling
event eliminates the possibility of evaluating the data for
reproducibility, significance, or in the cQntext of the
temporal framework. As has been demonstrated, evaluation of
specific concentration maxima within the temporal and
spatial framework of the existing database is critical in
developing a reasonably accurate understanding of metals
behavior in the environment and the effects of sample
collection procedures and preservation protocol on reported
values. Additionally, the omission of previously collected
data results in an order of magnitude decrease in the number
of values included in the sample population. This prohibits
a statistically significant evaluation of background levels
or the development of statistical parameters such as
concentration percentile values at specific locations or
within specific populations.
EPA Response: The chemical data to be used for risk
assessment have to meet a certain data quality level per
EPA's guidance; the samples should be analyzed by a EPA
certified laboratory and the data should also be validated.
The data collected by CCJM met this requirement. The
protocols have generally not been followed for data
previously collected making them unusable for the risk
assessment.
227. Page 7-3 Paragraph 3: Background concentration levels of
inorganic contaminants were developed (according to CCJM) by
averaging naturally occurring levels. These average
background concentration levels to identify contaminants as
those of concern. In all cases, regardless of media,
background samples were limited to between two and five
samples. Very little statistical significance can be
ascribed to such a small sample population.
EPA Response: Although the sample set used to define
background concentrations was not large it still has utility
for screening out chemicals present at naturally occurring
concentrations.
228. Page 7-4 Paragraph 2: CCJM states that inorganic analytes
known to be required human trace or macronutrients were not
included in the selection of contaminants of concern.
However, chromium, which is an essential human nutrient, was
included.
EPA Response: Although chromium is an essential nutrient it
also exhibits high toxicity and since acceptable dietary
levels are not known it was conservatively retained for
evaluation in the risk assessment. This is in accordance
with EPA guidance cited on the Risk Assessment on page 5-
60
-------
23. (EPA 1989 p. 5-23).
229. Page 7-5 Paragraph 2: Two of the three samples identified by
CCJM as background samples are referenced as minor hot spots
by CCJM. We fail to understand how a background sample
could also represent a hot spot.
EPA Response: Background sampling stations were chosen on
site specific geographical .and hydrogeological character-
istics at each sampling station; stations.lIleast likely to
be affected by the site" were chosen. This does not mean
the station will be free from contaminants or that the
contaminants are necessarily site related. Accordingly, the
presence of organic chemicals in background samples was not
considered sufficient reason to eliminate those detected at
the site from evaluation in the quantitative risk
assessment, since their presence could be the result of site
contaminants. Moreover, use of data from the two background
sampling stations (SL-20 and SL-21) has resulted in nine
metals being eliminated from consideration in the risk
assessment (Table 7-1).
230. Pages 7-6 to 7-12: Applying an erroneous procedure for
identification of contaminants of concern, CCJM identifies
some 15 metals as contaminants of concern for the Keystone
site even though most have never been observed in excess of
the MCL for ground water. In fact, most of the metals
identified as contaminants of concern by CCJM have not been
previously mentioned in the RI (outside 'of listing of
analytes).
EPA Response: The designation of a chemical as a
contaminant of concern does not imply that it exceeds an MCL
or other indicators of risk. Such measures of risk are not
to be used in selecting chemicals of concern according to
EPA in their risk assessment guidance (EPA 1989); chemicals
of concern are those chemicals that are selected for
evaluation in the risk assessment for their potential to
cause risk at significant levels. In most risk assessments
only a few chemicals of concern contribute significantly to
risk at a site although their presence is indicative of
contamination.
231. Page 7-25 section 7.3: section 7.3 is a discussion of a
selection of exposure pathways in identification of
populations at risk. It is important to recognize in
reviewing this entire section that, where metals have been
observed at elevated levels, they have been present in
either naturally occurring solid phases or adsorbed onto
hydrous iron oxides, present as a solid phase. Therefore,
exposure pathways must include an avenue by which contact is
established with turbid water. Metals concentration in
naturally occurring minerals outside the influence.of the
61
-------
site are similar to those present in particulate matter
suspended in wells completed at the site (for example well
K-S). Exposure to such compounds simply constitutes
exposure to muddy turbid water. This is particularly
important when considering any of the routes involved
including intake of groundwater or surface water by
ingestion: showering or swimming in ground water or surface
water; inhalation of particles; or intake as food. Because
most humans are generally wary of turbid water, exposure to
such discolored or turbid water in any of the aforementioned
circumstances is extremely unlikely. Furthermore, since
metals concentration in solid phases present in landfill
leachate are similar to those present in naturally occurring
materials, exposure to landfill leachate represents no more
of a threat than does exposure to naturally occurring turbid
water, although most people would probably avoid both.
EPA Response: The concern at the site is from
of organic contamination in the groundwater.
contamination in the groundwater results in a S
from the site.
the presence
The orgnaic
x -4 risk
232. Page 7-28: CCJM states that the most likely potential
exposed population includes the family residing on site,
Keystone site employees, trespassers on site, remedial
workers, and nearby residents. Exposure to the family
residing on site and Keystone site employees presumably is
via drinking of water generated from well RW-l which is
completed in the landfill. Although common sense should
preclude use of such a well as a drinking water source, a
more than $10 million remedial alternative is hardly
justified in order to renovate.well RW-1. Additionally,
such an expenditure is clearly not warranted for the sake of
keeping Keystone Sanitation Company in business or to
protect the health of trespassers on site, beyond issuing
warnings, fencing the sites, and taking other precautionary
measures.
Exposure of a trespasser on the site to ground water is
extremely unlikely in that monitoring wells are locked. No
power is supplied to pumps in the wells and access is
otherwise not provided to the site water regime.
Furthermore, site soils are not contaminated with metals (as
documented by CCJM and other studies) and, therefore, dermal
contact with soils by site trespassers is also unlikely.
This unlikelihood is further strengthened by the fact that
the sites is completely covered by grass and bare soils are
not exposed. Finally, whereas no release of contaminated
ground water is likely, nor has such a release been
observed, no route exists between the site and residences
surrounding the site. Therefore, exposure to off site is
simply not possible. Even if such exposure was possible,
monitoring and point source treatment could be employed to
limit such exposure. .
62
-------
EPA Response: The PRP comments that "Exposure of a
trespasser ori the site to groundwater is extremely unlikely
in that monitoring wells are locked...,". The risk
assessment did not indicate that groundwater is a potential
exposure route for trespassers. The total carcinogenic risk
for site trespassers is estimated to be 7X10-7 by exposure
to the surface soil and surface water only.
This comment concerning risks associated with the off-site
residents has been dealt with in another response (see
Responses to Wright Associates, Inc.). In short, the off-
site residents poses essentially the same risks as that for
the site trespassers, i.e., very minimal.
233. Section 7.5-Risk Characterization: The risk assessment was
performed to evaluate the risk to on-site adults(Page 7-43
'Paragraph 8). Additionally, although no on-site ground
water is consumed by residents, on-site ground water samples
were used in the risk assessment when calculating ground
water exposure concentrations. It should, therefore, be
remembered that the risk assessment is based on direct
consumption of landfill leachate.
EPA Response: The risk assessment is not based on direct
consumption of landfill leachate. It is based upon the
consumption of the contaminated groundwater due to leaching
from the landfill.
234. Pages 7-46 to 7-50: These subsections are presented by CCJM
to discuss risks posed by exposure to soils, sediments, and
surface water. However, no release of metals to any of
these media, either on-site or off-site, has been observed.
The identification of risks through these routes indicates
the misapplication of the risk assessment process by CCJM.
EPA Response: As summarized in the Risk Assessment Report
(Table 7-15), groundwater contributed 99.86% of cancer risk.
The impact of soil, sediment, and surface water was minimal.
235. Page 7-52: This discussion is presented to estimate the
risks due to inhalation during showering with contaminated
ground water. This section must be taken into context that
the ground water in question is landfill leachate and that
risks presented in this section refer to showering in
landfill leachate for 30 years for adults and 5 years for
children under 6. Such an assessment is clearly not
reasonable in that showering in landfill leachate is outside
of the realm of conceivable possibilities.
EPA Response: As it has been mentioned, this baseline risks
assessment was performed in accordance with EPA guidance
(EPA 1989). And it should be stated again that groundwater
but not leachate was assessed (see Comment 233).
63
-------
236. Page 7-53: Similarly, an.evaluation is presented for
ingestion of contaminated ground water (drinking landfill
leachate) which is based on consumption of two liters of
landfill leachate every day for 30 years. It is difficult
to imagine a circumstance where an individual would drink
two liters of landfill leachate and happily shower in
landfill leachate for a 30 year period.
EPA ReSDonse:
Please refer to Comment No~ 233.
237. Page 7-55: The quantitative risk characterizations identify
various metals as posing risks to human health. It must be
remembered that risks are based on showering and consuming
liberal quantities of landfill leachate over a 30 year
period. That such an estimate is based on an almost comical
circumstance, reflects failure by CCJM to integrate
analytical measurements and other technical data involving
potential exposure routes to yield a reasonable evaluation
of the risks posed by actual site conditions. The process
of risk characterization employed by CCJM involves a
progressive and consistent distancing of the assessment from
reality by subsequent layers of erroneous assumptions.
Under these circumstances, the entire risk assessment must
be regarded as suspect.
EPA ReSDonse:
Please refer to Comment No. 233.
238. Page 7-77 - site Trespassers: CCJM indicates that adult
trespassers could be exposed to inhalation, ingestion, or
dermal contact with surface soil and surface water. First,
there is no surface water at the site: secondly, it is
likely that should such surface water be present it is
unlikely an adult trespasser would choose to swim in it.
Finally, there is no path between inhalation or ingestion of
soils by an accidental process. Again, the circumstances
cited by CCJM are, at best, ludicrous.
EPA Res90nse: Risks associated with site trespassers are
minimal (see Comment No. 232).
239. Page 7-92 Table 7-26: Metals are identified as chemicals of
concern in the surface soil, surface water, and sediment
environments although no elevated concentrations of any of
the metals cited has been measured in any case in any of the
media cited by CCJM. Identification of these metals as
chemicals of concern defies reality and indicates the
reckless application of the risk assessment process by CCJM.
EPA ReSDonse: Again, surface soil, surface water, and
sediment are not major media of concern. They only
presented a total of 0.14% of cancer risk of the entire Site
64
-------
240. Page 7-1: The term "qualifiers" in the last sentence should
be defined. Which qualifiers were carried through the data
evaluation process?
EPA Response: Qualifiers carried through the data
evaluation process were identified in the Risk Assessment
Report.
241. Page 7-3: Background levels of contaminants need to be
explained in greater detail. Where do the background
samples come from and how were they compared to site
samples?
EPA Response: Background stations and samples were
discussed in great detail throughout this RI report.
see other responses.
Also
242. Page 7-4: The terms "on-site and
"background" are confusing. Are
site? If not, are they not also
affected by the site?"
affected by the site" and
the background samples off
considered "on-site or
EPA Response:
offsite.
Background samples are considered to be
243. Background samples are SL-19, SL-20, and SL-21. However, in
the third paragraph from the bottom, "other minor hot spots
are represented by... SL-20 and SL-21 ..." How can the same
samples be background and minor hot spots simultaneously?
EPA Response: Background sampling stations were chosen
based on site specific geographica~ and hydrogeological
characteristics at each sampling station: stations "least
likely to be affected by the site" were chosen. This does
not mean the station will free from contamination or the
contaminants are site related. Accordingly, the presence of
organic chemicals in background samples was not considered
sufficient reason to eliminate those detected at the site
from evaluation in the quantitative risk assessment, since
their presence could be the result of site contaminants.
Moreover, use of data from the two background sampling
stations (SL-20 and SL-21) has resulted in nine metals being
eliminated from consideration in the risk assessment (Table
7-1).How were the data utilized? Were surface samples and
deep (12 inches to 18 inches) averaged together?
EPA Response:
together.
All surface soil samples were averaged
244. Page 7-6: Frequency of detection should be define.
percentage is "frequency?" .
What
EPA Response: Frequency of detection is the number of
detects versus the total number of samples for that media.
65
-------
245. Page 7-7; The brief discussion of surface water analyses is
confusing and should be elaborated. The level of detail in
this discussion is much less than that for the soil samples,
possibly contributing to the confusion.
EPA Response: Surface water analyses were discussed in
great detail in the previous RI sections.
246. Page 7-9; Were the concentrations of methylene chloride and
acetone 10 times greater than that detected in the blanks?
In addition, Figure 3-1 should be Figure 3-2.
EPA Response:
Yes (see Appendix).
247. Page
Page
be a
Page
7-7; Figure 3-4 should be Figure 3-5.
7-8; Figure 3-4 should be Figure 3-5.
blank page.
7-10; Figure 3-5 should be Figure 3-7.
This should not
EPA Response: Corrections will be made to the document
through errata sheets placed in the Administrative Record.
248. Were the data for shallow, intermediate, and deep samples
all averaged together? The level of detail in this
discussion (residential wells) was less than the surface
soil discussion.
EPA Response:
together.
The data for all the wells were averaged
249. Page 7-11; Again, were all the samples averaged together?
Why were residential well data and ground water data
combined if they were sampled separately?
EPA Response: Because residential wells and monitoring
wells are all located downgradient of the site, they were
grouped together in the calculation of quantitative risk
250. Table 7-1; Why are antimony, silver, and selenium
contaminants of concern when there frequency of detection is
3/19, 1/36, and 2/19, respectively? There are may compounds
in this table that are denoted as compounds of concern with
a low frequency of detection.
EPA Response: For chemicals that were detected, but
infrequently, a conservative approach was used in retaining
infrequently detected chemicals in the risk assessment.
This conservatism in determining chemicals of concern is i~
accordance with current US EPA guidance (Dec. 1989). The
infrequent detection of a chemical at a site does not
automatically imply that its presence should be ignored.
Infrequent detected chemicals that may be present at
toxicologically significant concentrations. .
66
-------
251. Page 7-26; The third paragraph from the bottom should be
contiguous with the second paragraph from the bottom.
EPA Response:
No response needed.
252. Page 7-29; The brief discussion of ingestion of
biota (plants and animals) and food is untrue.
biota and food sampling, uptake into plants and
such as cows, can be estimated using
bioconcentration/bioaccumulation factors available in the
literature. This analysis would be especially important if
the surrounding farmland were used for grazing animals in
which compounds can bioconcentrate in the muscle, fat, and
milk. (It is unclear from the report whether the farmland is
used for crop growth and/or grazing animals.)
contaminated
Even without
animals,
EPA Response: The soil sampling in the field near the spray
irrigation system did not show contaminants at a level of
concern for crops.
253. Section 7.4 General; All acronyms should be defined before
using them to aid the reader. This comment applies to all of
Section 7.4.
EPA Response:
The acronyms were defined when first used.
254. Page 7-30; The following acronyms are not defined:
o
o
o
NOEL- No observed Effect Level
HEA- Health Effects Assessment
CAG- Cancer Assessment Group.
EPA Response:
No response needed.
255. Page 7-31; Which type chromium is actually at the site?
Using the more toxic chromium VI may not adequately
represent the site contaminants. What is the reference for
the 9.6 cubic meters inhalation rate used for employees?
EPA Response: The analyses refer to total
conservative and reasonable to use chronic
hexavalent chromium in estimating the risk
chromium.
chromium, it is
criteria for
caused by
256. Page 7-32; The following acronyms are not defined:
o ADI - Acceptable Daily Intake
o ADL - ???
EPA Response:
ADL is a typo of ADI.
257. Page 7-32; Why is the toxicity value for nickel refinery d'~l'
used in this analysis? Is there refinery dust at the site'
If not, the toxicity value for nickel should be that for
67
-------
inorganic nickel compounds.
EPA ReSDonse: There are only two nickel compounds having
slope factors, namely nickel refinery dust and nickel
subsulfide (HEAST). It is believed that toxicity value for
nickel refinery dust is more appropriate in this case.
258. Page 7-33; The reference for the ingestion reference dose
(RfD) for zinc is incorrect (EPA, 1989d). This is the
EXDosure Factors Handbook reference, which does not contain
any toxicity values.
EPA ResDonse:
The reference should be EPA 1989c.
259. ,Page 7-33; The following acronym is not defined:
o
NAS - National Academy of Science.
EPA ResDonse:
No response needed.
260. Page 7-34; In the second paragraph from the top, the Ames
assay should be described.
EPA ResDonse:
The Ames test is microorganism test.
261. Page 7-34; "Carb Disulfide" should be "carbon disulfide." If
a no observable effect level for inhalation exposure was
determined, then an inhalation RfD was developed. Is the
RfD for carbon disulfide for oral exposure or inhalation
exposure? This discussion is confusing.
EPA ResDonse: No inhalation exposure RfD has been derived
for carbon disulfide. EPA has derived an oral RfD of 0.1
mg/kg/day (HEAST, 1990).
262. Page 7-35; The Following acronyms are not defined:
o
o
o
TWA - Time-Weighted Average
I.p.- intraperitoneal
MTD -
EPA Response:
MTD is median toxic dose.
263. Page 7-36; The first sentence of the fourth paragraph from
the top does not make sense. The terms "A," "H," and "S"
should be defined when modifying uncertainty factors.
EPA ResDonse:
lOA: uncertainty factor of 10 in the
extrapolation of dose levels from animals to
humans.
10H: uncertainty factor of 10 in the
threshold for sensitive humans.
68
-------
10S: uncertainty in the effect of duration
when extrapolating from subcronic to chronic
exposure.
264. Page 7-37; The following acronym is not defined:
o
NCI - National Cancer Institute (NTP is defined when
used in test, why not NCI?)
EPA Response:
No response needed.
265. Page 7-37; The first paragraph refers to EPA, 1986. Which
EPA, 1986 is meant, a, b, or c? For all ethylene/ethene
compounds, the spelling should be consistent, either all
ethylene or all ethene. 1,1,1-trichloroethane should be
1,1,1-trichloroethane. This paragraph is difficult to
follow.
EPA Response:
No response needed.
266. Page 7-38; There is a group within EPA that has been
establishing noncancer toxicity values for the
noncarcinogenic polynuclear aromatic hydrocarbons. These
values have not yet been entered into the IRIS database, but
are verified RfDs and are available over the telephone.
Therefore, toxicity values for anthracene and fluoranthene,
as well as naphthalene and phenanthrene, are available.
EPA Response: RfDs for PAH are under evaluation by USEPA
and furthermore, they were taken out from both IRIS (may
1990) and HEAST (first/second quarter, 1990). Accordingly,
no values for the compounds were available for use.
267. Page 7-39; Why all the white space on this space?
EPA Response:
Reproduction error.
268. Page 7-41; In Section 7.4.4, Migration is spelled mi-
gration.
EPA Response:
No response needed.
269. Page 7-41; The fOllowing acronym is not defined:
o
s.c. - subcutaneous.
EPA Response:
No response needed.
270. Page 7-43; The following acronym is not defined:
o
. FDA Food and Drug Administration.
EPA Response:
No response needed.
69
-------
271. Table 7-3 (Page 7-44); Pennsylvania is spelled incorrectly
(Pennsylvania).
EPA Response:
No response needed.
272. Page 7-45; Why all the white space on this page?
EPA Response:
Reproduction error.
273. Page 7-45; The multiplication signs are missing from all
equations in Section 7.5.1.
EPA Response:
Word processing error.
274. Page 7-47; What is the value for AT?
EPA Response: The values for AT depend upon the exposure
duration for chronic and subchronic exposure [365 days x
exposure duration (years)] and carcinogenic effect (365
days/year x 70 years).
275. Page 7-48; The discussion on the box model is confusing.
Also, a box model may not be appropriate for this analysis.
What does "resussion" mean?
EPA Response: Box model is not the most sophisticated model
to assess the impacts of particulate emissions from site
surfaces due to wind as discussed in the text. However,
because of a general lack of accurate/reliable predictive
air dispersion models for concentrations near or within the
source and site-specific airborne concentrations for the
essential model calibration work, a site-specific model
cannot be established at present. The box model, a
recognized model by USEPA as appropriate, is adopted for a
first-cut approximation for estimating contaminant
concentrations near and within the site as an area source.
The box model is fully discussed in detailed in Section
7.5.1.3. Resussion refers to particle movement in the air.
276. Page 7-49:
H should be approximately 6 feet.
EPA Response:
No response needed.
277. Page 7-51; The assumption that both adults and children
shower 365 days per year is an overestimate (especially for
children under age 6). It is apparent that the EPA did not
use their own guidance when arriving at this assumption
[i.e., data found in the EPA's Exposure Factor Handbook
(1989)]. This raises the question of whether the most
70
-------
realistic and relevant data were used in determining
potential risk.
What is the reference for the percent volatilized for the
various compound types (bottom of Page 7-52)?
EPA Response: Conservative assumptions and best engineering
judgement were made for this _analysis.
278. Page 7-53; Further explain the box model used in determining
the shower air concentration.
For the drinking water pathway, water ingestion of 1.4
liters per day could be used as noted in the Exposure
Factors Handbook (EPA, 1989.
EPA Response: Please refer to Comment No. 276 concerning
the box model. According to "Risk Assessment Guidance for
Superfund; Vol 1; Human Health Evaluation Manual (Part A),
Dec. 1989", 2 liters should be used.
279. Page 7-54; The equations on this page are typed incorrectly.
The summation signs are missing.
EPA Response:
No response necessary.
280. Page 7-55; When summing (adding together of results from a
calculation) hazard indices, only those compounds with the
same end point should be added (i.e., only compounds that
affect only the kidneys or only the liver or only the
central nervous system). Summing all hazard indices can be
done as a screening measure; however, if the value is over
unity, which is 1, the compounds should be separated out
according to target organ (EPA, 1986, "Guidelines for the
Health Risk Assessment of -Chemical Mixtures, "Federal
Reqister. 51:34014-24025). -
EPA Response: When the total hazard index for an exposed
individual or group of individuals exceeds unity, there may
be concern for potential noncancer health effects. For
multiple exposure pathways, the hazard index can exceed
unity even if no single exposure pathway hazard index
exceeds unity and if cOmbining exposure pathways has
resulted in combining hazard indices based on different
chemicals, one may need to consider segregating the
contributions of the different chemicals according to major
effects. This was done in the risk assessment.
281. Table 7-4 is missing from this report.
EPA Response:
No response necessary.
282. It is unlikely that carcinogenic inhalation risks from
inorganic compounds during showering would be 2 to 3 orders
71
-------
of magnitude higher than-the carcinogenic inhalation risks
from 1,1-dichloroethylene .and vinyl chloride. This is also
true for the noncarcinogenic compounds. The hazard indices
for barium and manganese should not be 1 to 3 orders of
magnitude higher than the hazard indices for 1,1-
dichloroethane or 1,1,1-trichloroethane.
EPA Response: In the final document (Sep. 1990), the
inhalation risk has been recaiculated and the risk
associated with inorganics has decreased by 2 orders of
magnitude. However, this recalculation did not reduce the
risk at the site from the 10-4 requiring a remedial action.
283. Table 7-7 (Page 7-60); Lead is not considered a carcinogenic
compound. Phenanthrene is also not a carcinogenic compound.
Neither of these chemicals should be present on this table.
EPA Response: Lead is considered to be carcinogenic
chemical by EPA Region III (see toxicology profile in the
text also).
284. Page 7-77; The table on this page has typographical errors.
EPA Response:
No response needed.
285. Page 7-78; Add a space between section 7.5.4.3 and site
Trespassers.
EPA Response:
No response needed.
286. Page 7-83; In the last paragraph, the phrase "for because"
should be modified.
EPA Response:
No response needed.
287. Page 7-85; As stated in the comment for Page 7-55, all
hazard indices are not additive. Thus, the sentence on
page, "Health effects, whether carcinogenic or
noncarcinogenic, are assumed to be additive for all
contaminants of concern," is untrue.
this
EPA Response: This statement refers to uncertainty
associated with quantitative risk estimates. Please refer
to Comment No. 280 for additional responses.
288. Page 7-93; It is possible to determine a more realistic
scenario pertaining to the amount of site water a deer might
ingest. since the size of the home range of the deer is
known (2 to 3 square miles, as stated in the report) and the
size of the site area is also known, a ratio of the two
areas can be taken to determine the percentage of the site
that is in the home range of the deer.
EPA Response:
It is possible to estimate the perc~ntage of
72
-------
a deer's home range comprised by a site. However, a more
appropriate estimation of,the amount of drinking water
obtained onsite is not possible. Deer do not obtain water
from all available water sources within their home range:
they are more likely to utilize a few select watering spots.
Likewise, they do not equally utilize each of the preferred
spots. Since the preferred watering spots and percentage of.
these spots located at or near the Keystone site are not
known, it was conservative~y assumed for the purposes of
this assessment that a deer obtains 50 percent of its water
from onsite sources.
It should be noted that if a deer obtained 100 percent of
its water from onsite sources, the surface water exposure
concentrations would still be far below the representative
drinking water limits (obtained by dividing the drinking
water limits in the table by a factor of two). Therefore, a
more detailed analysis would not change the conclusions of
the environmental assessment.
289. This comment is in narrative form and questions many of the
assumptions made in the risk assessment. See R.E. Wright
comments, pages 33-36.
EPA Response: See previous comments concerning the analysis
of chromium in the risk assessment.
The choice of children under 6 years of age as site
trespassers is not unreasonable. This group would have
higher potential exposures than teenagers. Even with this
reasonable worst case scenario the total subchronic
noncarcinogenic risk was estimated to be well below 1 (i.e.,
0.019).
In the final risk assessment (Sep. 1990), daily water
consumption for children was modified to 1 liter per day.
The assumption of children to take showers daily 365 per
year is a conservative and reasonable approach since some
families may not have a bath tub available.
Soil at the Keystone site is more clay-like than sandy.
Therefore we believe that the use of the soil adherence
factor for clay is more appropriate than that for potting
soil.
Use of the national average for swimming frequency (7
days/year is inconsistent with EPA guidance (EPA 1989) which
directs that 95 or 95th percentile exposure frequency values
be used for risk calculations.
Although ingestion of surface water without swimming is less
likely, it still may take place.
It is generally true that inorganics will not vaporize from
73
-------
shower water. However, it is possible for inorganics to be
entrained as very small particles in water droplets. In the
final RI report (September 1990) 0.01% of the inorganics in
groundwater was assumed to be available for inhalation.
These compounds .contributed 4% to the total LECR for this
pathway. .
290. This comment is in narrative form and relates to
miscellaneous issues concerning the risk assessment. For
the detailed comments see the R.E Wright comments, pages 36-
39.
EPA ResDonse: Background sampling stations were chosen
based on site specific geographical and hydrogeological
characteristics at each sampling station; stations "least
likely to be affected by the site" were chosen.
Accordingly, the presence of organic chemicals in background
samples was not considered sufficient reason to eliminate
those detected at the site from evaluation in the
quantitative risk assessment, since their presence could be
the result of site contaminants.
For chemicals that were infrequently detected, a
conservative approach was used in retaining them in the risk
assessment. This conservatism in determining chemicals of
concern is in accordance with current USEPA guidance (Dec.
1989). The infrequent detection of a chemical at a site
does not automatically imply that its presence should be
ignored since infrequently detected chemicals may be present
at toxicologically significant concentrations. This can
best be achieved using a conservative approach. The PRP
agreed that vinyl chloride was present at the site but
questioned how widespread it was. Vinyl chloride was
detected in groundwater in 2/45 samples (4 and 40 ug/l) with
SQL ranges from 10 to 50 ug/l. The exposure concentration
was calculated to be 10 ug/l (still below the SQL). In
accordance with USEPA guidance, vinyl chloride needs to be
assessed. Assuming an exposure concentration of 10 ug/l,
the LECR was determined to be 3x10'4 in groundwater for
vinyl chloride. - Even if it were reduced to 5 ug/l, vinyl
chloride would still present an LECR of 1.5X10.4.
Chrysene was detected in monitoring well MW-CD with a
concentration of 4 ug/l at 1/29 samples. It was not
screened out from consideration in the quantitative risk
assessment in groundwater because it was also detected in
surface soil in 9/36 samples (exposure concentration 202
ug/kg). Nevertheless, since no toxicological criteria were
available for chrysene it was not included in the
calculation of risk.
The relatively clean groundwater from offsite residential
wells does not appear to be affected by the onsite
groundwater contamination because of the site specific
74
-------
\..- =---=--
hydrology characteristics. Thus, the risk associated with
groundwater is insignificant. The total risk associated
with the site for offsite residents should be essentially
the same as that for site trespassers.
List of Tables was misprinted for Table 19, 20, and 20.
Deed and zoning restrictions cannot be considered in a
baseline risk assessment. ~ccording to the National
Contingency Plan (40 CFR Part 300), a baseline risk
assessment is an assessment of the risks from the site in
the absence of remedial alternatives (which include
institutional controls such as deed and zoning restrictions.
The effect of such controls may be considered in the
evaluation of alternatives during the feasibility study but
not as part of the baseline risk assessment.
291. This comment is in narrative form and includes various
concerns regarding the selection of contaminants of concern.
in the Risk Assessment (See Semmes, Bowen and Semmes
comments, pages 3-1 through 3-8).
EPA Response: The MCL is not an acceptable criterion to use
in the selection of contaminants of concern (see EPA
guidance on risk assessment, Dec. 1989). Data on the
regional naturally occurring levels of contaminants should
only be used as a reference or indicator in the selection of
contaminants of concern because the variation at different
locations is significant in some cases. However, site
specific background information is preferable and since this
was available, it was used.
A different statistical approach for determining whether a
chemical was elevated above background (EPA Region III
specific method) was employed in the final risk assessment
(September 1990). As a result, barium was identified as a
contaminant of concern in surface water, sediment, and
groundwater; cadmium was identified as a contaminant of
concern in groundwater; and chromium was identified as a
contaminant of concern in surface water and groundwater.
292. What is the definition of surface soils.
EPA Response: The EPA definition of surface soil differs
from site to site (depending on site-specific
characteristics). However, soil to a depth of 18 in. can be
reasonably considered to be surface soil since it is not
uncommon for contaminants to reach 12-18 in. below the
surface offsite through surface run-off.
293. Data with qualifiers should not be used in the risk
assessment.
EPA Response:
According to EPA guidance (EPA, Dece~er,
75
-------
1989), data labelled with the qualifier IIJII should be
included in quantitative risk assessments. Also, data for
MW-MD1 are given in the Appendix.
294. The exposure assumptions presented in the risk assessment
are not realistic.
EPA Response: The choice of children under 6 years of age
as site trespassers is not , unreasonable. This group would
have higher potential exposures than teenagers. Even with
this reasonable worst case scenario the total subchronic
noncarcinogenic risk was estimated to be well below 1 (i.e.,
0.019) (Also see previous responses).
Avoiding lengthy calculations, the risk for onsite workers
(the worst case scenario) was simply adapted to adult site
trespassers to show that there was no significant risk (LECR
7 x 10-7).
Ingestion of surface water could happen regardless of the
size of the stream. Also, the risk values associated with
surface water near the site are insignificant (carcinogenic
risk, 0.004%; chronic effect, 0.002%; and subchronic effect,
0.001% of the total risk).
Generally, inorganics will not vaporize from the water into
the bathroom air during showering, since the shower water
temperature is approximately 50°C. However, it is possible
for inorganics to be entrained (very small particles of
water droplets) in air through elevated temperatures and
showering activities (i.e. shower spray). In the final RI
report, 0.01% of the inorganic entrainment was utilized to
estimate the inhalation risk associated with groundwater
during showering. Inorganics contributed 4% to the LECR of
4 x 10-4.
The units in these equations are correct.
necessary.
No response is
295. The individual chromium species should be identified and
'used for risk assessment.
EPA Response: Since the analyses refer to total chromium,
it is conservative and reasonable to use chronic criteria
for hexavalent chromium in estimating the risk caused by
chromium.
296. These comments relate to several issues related to risk
characterization.
~PA Response: Information on the carcinogenic risk for
ingestion of groundwater is summarized in Table 7-7.
Because ingestion of groundwater and inhalation of
groundwater \vhile showering are two different pathways, tr..
76
-------
were discussed separately in the quantitative risk
assessment.
The risk for cadmium exposure while showering was reduced to
2 x 10-8 in the final risk assessment (September 1990). See
previous responses.
In the final risk assessment, the values for the
carcinogenic risk, chronic effect, and subchronic effect via
chronic inhalation while showering were reduced to 1x 10.4,
0.044, and 0.016, respectively.
It is EPA Region III's policies that when an oral slope
factor is not available, inhalation slope factor could be
employed as a conservative measurement.
Page 7-2; Capitalize "I" in the second bullet from the top
of the page.
Page 7-5; Figure 3-3 should be Figure 3-4.
FEASIBILITY STUDY
297. In light of the extensive comment concerning the
inadequacies and technical errors which plague the entire
RI report by CCJM, a similarly exhaustive review of the
feasibility study will be here. Any remedial system design
based upon the remedial investigation completed by CCJM
could only suffer from the same flaws reviewed in the RI.
Additionally, combined errors in various sections of the
remedial investigation would only serve to compound the
technical inadequacy of the feasibility study.
An evaluation of CCJM remedial system alternatives.was
conducted during the course of numerical model evaluation
and presented in a previous section of this report. No
additional comment is, therefore, necessary to demonstrate
the technical inadequacy of any remedial design based on
this RI. In that CCJM based the treatment system design, in
part, upon limited ground water sampling results in which
metals were entrained in the sample during monitor well
purging, treatment was designed under the erroneous
assumption that particulate matter represents a dissolved
phase which must be accommodated by treatment.
Whereas metals present within the landfill environment are
present as a solid phase, subsequent precipitation, or other
treatment system design elements are unnecessary. In fact,
given the fact that there has been no observed release of
metals from the landfill, and that such a release is not
possible by any means other than direct extraction of
particulate matter during the course of monitoring and
recovery well pumping, the treatment process itself
represents the only possible route by which metal& could
77
-------
escape from the landfill environment to environment outside
of the landfill. If treatment is required, it would only be
required as a by product of VOC treatment which, in light of
the type of compounds present; should be limited to air
stripping.
Relative to the alleged presence of trace metals at the
Keystone Landfill site. the RIjFS firmly supports one and
only one conclusion - the only reasonable remedial action
supported by the evidence is the no-action alternative.
EPA Response: The feasibility study is not designed to
provide a definite, detailed design. Instead it is a
conceptual design used to estimate costs and evaluate
alternatives. The data obtained" during the RI indicate that
a release of hazardous substances has occurred from the
site. The levels of metals and VOCS detected were
sufficient to warrant the treatments of these contaminants
during the remediation. As indicated in the Risk
Assessment, the risk from this site to human health exceed
10-4 risk range and therefore a remedial response is
required to be protective of human health. The no-action
alternative would allow the contaminant plume to further
migrate offsite.
78
-------
D. REMAINING CONCERNS
Several remaining issues were expressed during the Public Comment
Period and are as follows:
1. citizens are concerned that one round of groundwater sampling
was not representative of the problems in the area residential
wells which have historically shown contamination. In addition,
they want EPA to supply them with bottled water.
EPA ResDonse: In response to the citizen concern, EPA will be
conducting another round of residential well sampling in the next
several months. This sampling will use a detection level below
the MCLs or risk levels for the contaminants of concern such as
vinyl chloride. The sampling plan will be developed with
assistance from members of the communities in Pennsylvania and
Maryland using historical data on the residential wells in the
area that they have collected, data from the State of Maryland
sampling events, data from PADER sampling events and data from
the RI. If the sampling results from this next round of
residential well sampling indicates that contaminants are present
above MCLs or acceptable risk levels, EPA will consider
providing water or a treatment system to the affected residents.
~
2. PRPs are concerned that a release of metals to the groundwater
from the site is not possible and has not occurred. Therefore,
metal treatment is not necessary and in some cases, contend that
the site only warrants a No Action remedy.
EPA ResDonse: From the Risk Assessment, it was determined that a
risk of 5 x 10-4 exists at the site from the ingestion of
groundwater. This level of risk requires a remedial action
response and therefore, a No Action remedy would not be
appropriate. Levels of metals have been detected onsite and
offsite above the MCLs. Due to the presence of the metals above
the MCLs, it was determined that metal treatment would be a
component of the groundwater treatment. However, additional
sampling of the groundwater obtaining both filtered and
unfiltered samples will be conducted during the Remedial Design.
The data from these samples will be used in the design of the
groundwater treatment and further substantiate the need for metal
treatment. If the data determines that the metal treatment is
not necessary, EPA will issue an Explanation of Non Significant
Differences to modify the Selected Remedy.
J~~
"
3. The community and local officials want to be kept better
informed and more involved in the Site activities and decisions.
EPA ReSDonse: EPA has recognized the concerns of the community
and the local officials and will keep them informed and involved
in all the site activities and decisions.
-------
..
4.
The citizens want a health'study conducted in the area.
EPA ReSDonse: EPA will make a request to the Agency for Toxic
Substances and Disease Registry (ATSDR) to conduct an additional
health study in the area. ATSDR conducted a health study in
November 1988. In addition, a Risk Assessment will be conducted
in the investigation of the Operable Unit 2 to evaluate risks
specifically to the offsite residential wells.
5. The citizens are concerned over allowing the PRPs to
implement the remedy and any involvement from PADER and the State
of Maryland.
.
~
EPA ResDonse: EPA will be actively involved in the monitoring of
all activities conducted at the site either by the PRP
contractors or federally financed contractors. The monitoring
will involve the Remedial project Manager, the EPA Technical
Support Staff, and EPA's oversight contractor and consist of
direct supervision of site activities, review of the data
collected and interaction with local residents. EPA is also
interested in getting local residents involved in the monitoring
activities as they have indicated their willingness.
6. The citizens and local official do not feel that the study was
comprehensive enough and that additional studies should be
conducted on the Site.
EPA ReSDonse: EPA has changed the Selected Remedy from a final
action ROD to an interim action ROD. A second Operable Unit will
investigate the offsite contamination. In addition, a sampling
of residential wells will take place in the near future in
response to the concern over the need for EPA to supply bottled
water.
.......
-
,",
-------
COMMONWEALTH OF PENNSYLVANIA
DEPARTMENT OF ENVIRONMENTAL RESOURCES
BUREAU OF WASTE MANAGEMENT
One Ararat Boulevard
Harrisburg, PA 17110
(717) 657-4588
Fax t (717) 657-4446
September 27, 1990
Mr. Edwin B. Erickson
Regional Administrator
us EPA, Region in
841 Chestnut Building
Philadelphia, PA 19107
Re: Record of Decision (ROD) Concurrence
Keystone Sanitation Company Site, Union Township, Adams County
I
Dear Mr. Erickson:
The Record of Decision for the final remedy will address
remediation of groundwater contamination at the perimeter by
eliminating or reducing the risks through engineering and
institutional controls.
The major components of the selected remedy include:
1. Installation and maintenance of an impermeable cap and
gas collection system over the 40 acre landfill.
2. Installation and maintenance of groundwater extraction
veils and a treatment plant to reduce the
concentrations of volatile organic compounds and metals
in the groundwater.
3. installation and maintenance of a fence around the
site.
4. Monitoring the groundwater in monitoring and
residential wells.
5. Monitoring surface water and sediments.
6. Initiation of deed restrictions regarding present and
future site activities including limitations on
construction and aquifer use.
7. Conduct a five year review to assess the effectiveness
of the remedy to reduce the risks to human health and
the environment.
IS)
-------
'.. . ---_.}...
£Q'd l\:fJ.OJ.
\ ~
Mr. Erickson, page 2, septem&ai--27, 1990 t'
( '.
I hereby concur with the' £PA'S proposed remedy, with the
followinq conditions:
»
..
EPA wi:ll assure that..tne Department is provided an
opportunity to fully participate in any nS90tiations
wi th responsible parti~s.'
The Depa~ent will be 9iven the opportunity to concur
with decisions related to the design of the remedial
action and to assure compliance with DER design
specific ARARs.
The Department's position is that its desiqnstandards
are ARAR8 pursuant to SARA Section 121, ana we will
reserve our right to enforce those design standards. .
))
»
»
The Department will reserve our riqht and
responsibility to take independent enforceaent actions
pursuant to state and federal law.
This concurrence with the selected remedial action is
not intended to provide any assurance pursuant to SARA
section 104(C)(3). .'
»
»
The Departaent aqr..s with the propose4 reaediation
which provides that "backCJroundw quality is the
objective of the qroundwater reaediation plan. In the
event tha~ EPA ao41tie. its position on the cleanup
standard, and deviates fro. back;roun4 quality a8 the
standard, the Department will withdraw its concurrence.
At that ti.e, EPA must c:1aonstrate the iapractica1:>ili ty
ot achiev1nq backqround quality and CJive the Department
a ..aningful opportunity to recohcur.
If you have any questions rec;ardinq thi8 _tter please cia
not hesitate to contact .. at tb8 above listed number.
.... . 0. .. ~ ", .
-"...'
ar, Ph.D., P.E.,
or Barrisbur9 R8CJion
------- |