United Statei
Environmental Protection
Agency
Office of
Emergency end
Remedial Retoonie
EPA/BOO/B03-86/023
May 1866
1EPA Superfund
Record of Decision
Millcreek, PA
-------
,. 1111'0" T NO.
EPA/ROD/R03-86/023
.. TITLI AND SUIT.TLI
SUPERFUND REOORD OF DECISION
. illcree k, PA
TECHNICAL REPORT DATA
(Plust nlld IIItt1lJctiOllt 011 th, "vt", btf~' co,""I"illl)
12. 3. "1"7'$ Ar~$I~~ 707 5/15
5. IIII'O"T DATI
May 7, 1986
I. 'l"I'OIllMI~G OIllGA~IZATION COOl
7. AUTHOIIIIS)
.. '11III'OIllM'NG OIilGANIZATfON IIII'OIllT 1100.
-
8. '1 III 1'0 III..I"'G OlllGAHIZATfON HAMI ANO AOOIllI55
10. 'IIIOGIIIAM II.IMINT ~O.
Same as box 12.
11. I;cHT""I;T'~"A~T NC.
12. S'ONSOIIIING AGENCY NAME AND AOOIIIISS
U.S. Environmental Protection Agency
401 M Street, s.w.
Washington, D.C. 20460
13. TY'I 01' IIIE'Ol'lT ANO '11'1'00 COVUIED
Final ROD Report
1.. S'ONSOllltNG AGENCY COOlE
800/00
15. SU'I'LEMINTAIIIY NOT IS
,.. AISTIIIACT
The Millcreek site is a 84.5 acre tract of land located in Millcreek Tbwnship, Erie
County, Pennsylvania. ~~ site was once a 75-acre freshwater wetland. During the past
40 years, all but 4 acres have been filled with foundry sand and industrial and
mun1cipal waste. The site qperated as an unpermitted active landfill during this time..
For the past 15 years, unknown parties bulk disposed halogenated volatile solvents'~
80ils in the eastern portion of the site. This disposal has resulted in significant
ground water contamination both on- and off site. Unit cancer risk calculations reveal
;hat offsite ground water contamination exceeds 10-2 cancer risk levels adjacent to
the eastern portion of the site. In adaition, Region Ill's Remedial Investigation.
discovered extensive soil, sediment, and surface water contamination. The major clases
of compounds detected included: polychlorinated biphenyls (PCBs), polynuclear aromatic
hydrocarbons (PABs), phthalates, volatile organics, phenols and metals such as lead and
coppe r .
The selected remedial action for this site includes: excavation and consolidation of
contaminated soil and sediments under a RCRA cap to meet proposea criteria; site
grading; placing a soil cover over remaining low level contaminated soils not exceeaing
the proposed criteria; construction of surface water management basins and ditches;
revegetation of soil cover and cap; installation of additional monitoring wells;
(Se~ Attached Sheet)
1.
I.
OISCIII'I'TOI'I'
KIY WO"OS AND OOCUMINTANALYS'S
b.'OIENTII"'''S'OI'IN ENDED TE"MS
C. COSAT. Fieldl(iroup
Record of Decision
Millcreek, PA
Contaminated Media: gw, sediments, soil, sw,
wetlands
Key contaminants: Heavy Metals, PCBs,
Phenols, Polynuclear Aromatic Hydro-
carbons (PAH), Trichloroethylene (TCE),
.--
'l:g..TIII.8UT'ON STATEMENT
"0 SIiCU".TY CLASS (ThU R.,po,r/
None
20. SICUIIIITY CLASS (Tllil POPI
None
2'. NO. 01' PAG.E~
43 )t'G~
22. 'III,CE
I'. ,- 2220.1 (I... .-n)
....I:VIOu. I:DITION " O.'OIoI:TI:
-------
EPA/ROD/R03-86/023
Millcreek, PA
16.
ABSTRACT (continued)
construction of a flood retention basin on property owned by Millcreek
Township: pumping and treating of contaminated ground water: additional
sampling and well installation and ground water monitoring. Total capital
cost estimates for the selected remedial alternative vary from. $12,000,000
to $18,000,000 with an estimated baseline cost of $15,000,000. For these
estimates, capital costs included all costs associated with excavat~on,
regrading, revegetating,capping and ground water pumping and treating for
two years. Additional sampling and monitoring wells will be considered as
part of the design. .Design is estimated to cost approximately $1,000,000
and will be funded entirely by trust fund monies. Total present worth cost
for O&M is estimated to be $1,763,000. .
I. 'l....
-------
RECORD OF DEC!SION
REMEDIAL ALTERNATIVE SELECTION
SI ce :
Mll}creek Sice, Erie County, Pennsylvania
DOCUMENTS REVIEWED
I am baaing my decision primarily on the following documents des-
cribing the analysis of cost-effectiveness of remedial alternatives for
the M111creek site:
- Millcreek Remedial Investigation (NUS Corporation, August, 1985)
- Millcreek Feasibility Study (NUS Corporation, August, 1985)
Technical Support Documents prepared by EPA legion III staff
to establish groundwater protection goa18, 8011 criteria, aod
aedi.ent cr1teria.
- Staff summaries and recommendations
- Summary of Remedial Alternatives Selection
- Responsiveness Summary
DESCRIPTION OF SELECTED REMEDY
- 50i1 excavation and consolidation under a RCIA cap to meet
proposed soil criteria (criteria to be reevaluated during design).
- Sediment excavation and consolidation under a RCRA cap to meet
proposed sediment criteria (criteria to be reevaluaced during design).
- Si te gradi ng .
- 5011 cover over remaining low level contaminated soils not exceeding
criteria.
- Construction of surface water management basins and ditches.
- Revegetation of soil cover and cap.
- Installation of additional monitoring wells.
- Construction of flood retention basin on properey owned by
Killcreek Township.
- Pumping and treating of contaminated groundwater.
- Design of the remedy which will require additonal sampling and well
installation.
- Operation and mai" ~nance will be implemented by the State of.
Pennsylvania on the RCRA cap, flood retention basins, surface
. .
"
-------
water management systems and monitoring systems six ~onths after
construction of these systems. The groundwater pumping and treatment
program, and associated monitoring, wtll be operated as a source
control remedial action for a period of at least two years and
will be eligible for Trust Fund monies.
DECLARAnONS
Consistent with the Comprehensive Environmental Response, Compensation.
and Liability Act of 1980 (CERCLA), and the National Contingency Plan (40
CFR Part 300), 1 have determined that the described selected alternative.
provides adequate protection of public health, welfare, and the environment.
The Commonwealth of ~ennsylvania has been consulted and agrees wtth the
approved reMdy.
1 have also determined that the action being taken is.appropriate
when balanced against the availability of Trust Fund monies for .use at other
site.. In addition, the on-site secure disposition is aore cost-effective
than other remedial actions. and is necessary to protect public health,'
welfare or the environment.
~/ Ig~
I te
-\
I
i
I
" .
III
-------
- - --- -------
COMMoNwEALTH OF PlNNSYLVANIA
DEPARTMENT OF ENVIRONMENTAL REsouRCES
'-t 0M08 10. 2083
H8m_.,., '....t8ytqn1a 17120
AprU 2ft I'.'
. a'~
. 'INNSYLVANIA
aEAR
..... .. W... Ml
-------
SITE DESCRIPTION AND SUMMARY
OF REMEDIAL ALTERNATIVE SELECTION FOR THE MILLCREEK SITE
Site Location and Description
The Millcreek site is a 84.5 acre tract of land located in Millcreek
Township, Erie County, Pennsylvania. It is .ituated approximately two
milel welt of the city of Erie as shown in Figure 1. The property i.
presently owned by Millcreek Township (4 acrel), Ralph Riehl. Jr. (57 acres)
JOleph Hals1 (13.5 acres), and James Sitter (10 acres). .
The topography ~f the site is relatively flat except for leveral
isolated mounds of foundry sand and debril. Flood potential maps of the
area show that the site is located within a 100-year and 500-year flood
zone. Flooding occurs frequently, though, ealt of the site in a residential
area during heavy !a1nl.
The site is bordered to the north, northeast, and northwest by
residential areas. A commercial trucking fir. borderl tbe 8ite to the
ealt, and. children's baseball field to the vest. Erie International
Airport il located about 2000 feet west of the 8ite. At lealt 2,000
people work or reside within a.2500 foot radius from the center of the
s1 tee
Ground water for drinking water purposes i8 utilized by municipal
wells located about 1200 feet south (hydraulically upgradient) of the site
boundary. Ground water is not presently utilized downgradient of the site.
Sometime during the past 15 years, unknown parties bulkdispoled of
halogenated volatile 101vent8 in soils in the ealtern portion of the
site. This dispolal has resulted in significant ground water contamination
both onsite and offsite. Unit cancer r~sk calculations reveal that
offsite ground water contaadnation exceeds 10-2 cancer risk levels
adjacent to the eastern portion of the site. There are presently no
State, County, or Municipal restrictions against ground water use in the
site area. . .
Shallo~ground water discharges to a stream (Marshalls Run) east of
site during high water table conditions (spring, summer). Marshalls
Run discharges to Lake Erie 1.2 miles downstream from the site. Aquatic
life (fish and macroinvertebrates) are abundant near the mouth of Marshalls
Run near Lake Erie.
In addition to identifying volatile orga~ic compound contamination
in ground water, Region III's Remedial Investigation discovered extensive
soil and sediment contamination. The major classes of compounds aetected
included: polychlorinated biphenyls (PCBs), polynuclear aromatic hydro-
carbons (PNAs), phthalates, volatiles, phenols, and metals such as lead
and copper. .
The perimeter of the site is deciduous forest, while the central,
southern, and southwestern portions are composed of fill material. A
wetland of about 4 acres lies on the southern perimeter .of the site.
I.
-------
LAKE ERIE
__ _. K
•ASC MAP 8 A PORTION Cf TX U.1S.S. SWANVlLLf. M QUAQRANQLE (7.9 MMUTE SCMICS, I9S7, PMQTOflffVISED l»€9
IfTS). COWTOJW WTCTVW. O'.
FIGURE /
LOCATION MAP -
MILLCREEK SITE. MILLCREEK TWR, PA
SCALE: l"»2000'
2-
CORPORAT1C
A Hailiburton Compa
-------
Debris such a8 junk cars, and abandoned machinery are strewn through-
out the site along with numerous drums of foundry sand and slag.
Site History
The site vaa once a 75 acre freshwater wetland.
40 years, though, all but 4 acres have been filled In
industrial and 8Unicipal waste. The site operated as
active landfill during this time.
Duriog the past
with foundry sand,
an unpermitted
During the paat 10 years, waste oils containing high concentrations
of PCBs were bulk disposed in site fill, along with phthalates, phenols,
polycyclic aromatic hydrocarbons (PNA8), and heavy metals. These con-
taminants and tentattve1y identified compounds (TICs) were detected in
site fill during the lemedia1 Investigation (II).
In April, 1981, thePennsylvaDia Depart8ent of EnvlroD8ental lesources
(PADER) discovered dumping of drums in the central portion 0' the site. .
The drums were later sampled and found to contain trtchloroethene (TCE).
The PADEI later closed the site in 1981. In August, 1982, the Erie
County Health Department (ECHD) discovered drums on the surface of the
site while investigating a natural ga. well fire on the Balm! portion of
the site. .
In Noveaber, 1982, EPA dispatched its Environmental
(EIT) to conduct drum, soil, sediment, ground water, and
s..pling at the site to evaluate potential health risks.
1983, EPA conducted. planned removal of 75 liquid filled
contained vaste 011s, solvents, and antifreeze.
lesponse Team
surface water
In November,
drU1l8 which
The four property owners, except for Mil1creek Township, owned their
property at the time of filling. In 1973, the Sitter brothers purchased
the Sitter portion, and from 1974 to 1979 filled it in with foundry sand.
In 1981, Millcreek TQVftship purchased a 4 acre parcel of land from Mr.
Riehl for the purpose of constructing a flood control structure. PADER
denied a permit for cop~truction in 1982 pending the results of EPA's
RIIFS study..
Personnel Who worked on the Riehl pr~perty indicated that from 1977
to 1979, an unknown amouDt of nonhalogenated solvents and ink vastes,
300 drums a year of polyester resins, 6,600 gallons a year of caustics,
3,000 drums total of paint wastes, and 180 drums a month of slag were
disposed of at the site. Most liquid disposal is believed to have oc-
curred by bulk aethodl. The operators allo ran a metals reclaiming
facility in the eastern portion of the site and constructed a de~ pond
to supply water for foundry sand washing. .
-2- a..
-------
Current Site Status
EPA Region 111 completed a Remedial Investigation/Feasibility Study
(Rl/FS) at.Millcreek in August, 1985. Data collected in the Rl and in
previous studies done by EPA's Environmental Response Team (ERT), Penn-
sylvania Department of Environmental Resource. (PADER), and the Erie
County Health Departme~t (ECBD) vere u.ed to describe the nature and extent
of contamination. Additional .011, sediment, .urface water, and ground
water .amples will be collected during design.
Pathways and receptors are described in detail along with known or
. suspected risks posed by contaminants in the Remedial Investigation
and Feasibility Study Reports and in the EPA Region 111 Technical.
Support Documents.
The following is a brief .U1III2Iary of the types and concentrations of
contaminants detected in .oil, .ediment, ground water and .urface water:
. 5011
- elevated levels of P9lychlorinated biphenyls (PCBs) vere
detected in the eastern and south central portions of the
site. Concentrations of PCBs were found up to 31 "mg/kg vet
weight.
- elevated
detected
portion.
weight.
levels of polYnuclear aromatic hydrocarbons (PNAB) were
through out the site, especially in the southwestern
Concentrations of PNAs vere found to 539 up mg/kg wet
- elevated levels of phthalates vere also detected throughout
the site. The predominant area of contamination vas found
to be in the southern portion of the site. Concentrations
were found up to 72 mg/kg wet weight.
- elevated levels of phenols were detecte~ fn the southern and
eastern portion of the site fn concentrations up to 7 mg/kg
wet weight.
- volatiles were detected in the south central portion of the
site in concentrations up to 6 mg/kg. Volatiles are also
believed to be concentrated in the eastern portion of
the stte as indicated from monitoring well data. Soil
concentrations in this area viII be deterained during design
since this portion of the site did not undergo test pitting
or soil boring during the Rl. .
- metals were also detected throughout the site at various
concentrations. Two metals of concern, copper and lead, were
found in concentrations up to 20,500 and 2,375 ../kg,wet
weight.
-3-
-------
- High concentrations of tentatively identified compounds (TICs)
were det.cted throughout site soil. Most TICS are believed to
be hydrocarbon derivitives pOlsibly resulting from the bulk
disposal of oil. The 11 contains a complete list of a11 TICs
detected in soil. The list probably exceeds 1000 compounds.
TICs were detected in concentrations over 1000 mg/kg. Because
toxicological ,information on many of the TICs i8 sparse, the
11 risk assessment only conlidered Hazardous Substance List
(RSL) compounds present in loil. The risk posed, by TICs
will continue to be evaluated during predesign and design.
. Sediment
- Except for volatiles, which were not detected, many of the
same compounds detected in soil were detected In sediments
of the wetland in the southern portion of the site, In ditches
within and on the perimeter of the lite, and In Marlhalls
Run bordering the eastern portion of the site.
- PCBs were detected In concentrations up t~ 1.50 mg/kg wet weight,.
- phthalates were detected in concentrations up to 5.0 mg/kg
wet weight.
- phenols were detected up to 0.99 mg/kg wet weight.
- metals such as lead and copper were detected In concentrations
up to 0.67 and 6.61 mg/kg wet weight respectively.
- TICs were found In concentrations up to 115 mg/kg wet weight.
. Ground Water
- Except for the metals, manganese and iron, elevated levels of
detectable ground water contaminants were restricted to the
eastern portion of the site.
- Volatiles were detected in concentrations over 30 mg/l. The
list below Outlines the most frequently occurring volatile
organic compounds (VaCs) and their corresponding maximum
concentration detected during the 11.
Compound
I-1-dich10roethane (l~l-DCA)
1,1-dich10roethene (l,l-DCE)
1,2-dlch1oroethene (1,2-DCE)
trichloroethene (TCE)
l,l,l-trichloroethane (l,l,l-TCA)
vinyl chloride (VCM)
Max'. Cone. ( ug 11 )
21»0
16
29,000
300
960
220
-4-
-------
. .
- phthalates were also detected In ground water. Dlethyl
phthalate was found In concentrations up to 41 ug!l while
T dl-n-butyl phthalate was detected at a concentration of 21
ug!I.
- iron and .anganese were detected In concentrations up to 20,800
and 1,920 ug!l respectively. These represent filtered (0.45
UIII) samples.
TICs wer, also detected in jround water in the ea~tern portion
of the .ite. A total of 16 TICs were Identified and present In
concentration. over 1000 ug/l. ..
. Surface Water
- Marshall'. Run and drainage ditches throughout the lite were
dry during the R1 .0 evaluations are ba.ed on prevlou. .ampling
attempt. and .ampling in the wetland located In the .outhern
portion of the site which 11 Wet throughout the year.
- Volatiles were detected in Marshall'. Run.during the 1982 ERT
investigation. The lilt below .ummarlze. VOCs and corresponding
concentrations.
-.
voc
Concentration (ug/l)
Vo.
1.1,I-TCA
18
93
Metals were detected In the wetland in
of the site and in Marshalls Run. The
elevated levels of metals detected and
tratlons. .
the southern portion
list below summarizes
corresponding concen-
Metals
Concentrations (ug/I)
Copper
tron
Hanganese
Lead
Zinc
AlulllinUIII
Mercury
Nickel
Tin
Cadmium
9.560
21,600
1,580
1,940
6.270
6,270
0.81
386
385
3.7
-5-
-------
Alternatives Evaluation
This section will briefly define the public health and environmental
objectives of remediation; screening methods to determine appropriate
re..dial technologies; and specific alternativel considered. The
Fea8ibility Study contains a more in-depth analYlil of these discussions.
. Pub~ic health and environmental remediation objectives:
- prevent onsite air dispersal of particles containing
potentia~ly hazardous substances.
- prevent direct dermal contact wi th potentially hazardous
lubltances.
- prevent offlite tranlport of co~taainated loi1 and lediment
via erolion or Itora transport.
remediate offltte,ground water contamination to ground water
protection goals. Tentative levels established for cost
estimating purposes are outlined in Table 1.
- reaediate loi1 contamination to safe soil 1eve11 capable of
preventing future ground water contamination. Tentative
levell established for cost estimating purposes are in Table 2.
- remediate sediment contamination capable of causing an impact
on aquatic life or wildlife in the wetlands and Harshalls Run.
Tentative levels established for COlt estimating purposes are
outlined in Table 3.
- remediate potential surface water contamination by remediattng
ground water, soil and sediment contamination. Tentative levels
used for cost estimating purposes are outlined in tables 1, 2,
and 3. .
I
I -
The tentative soil and sediment criteria and the groundwater protec-
tion loals were based on a site-specific risk analyses presented in the
EPA Region 111 technical support documents. The exception is the soil
criteria for PCBs, which was 'based on a consensus policy for residential
areas proposed to EPA by a committee of environmental organizations and
industry groups which 1s under consideration by EPA for use as the basts
for a TSCA PCB policy. will use 10 ppm unless additional i~foraation
becomes available during design which would require the use of a lower
number. A site specific analysis for the other .compounds was necessary
because there are no existing regulations for those compounds.
-6-
-------
Table 1
Ground Water Protection Goals
Compound
Organics
(ug/l)
*vinyl chloride
*trichloroethene
*1-2-d1chloroethene
*1.2-dlchloroethane
1. 1. I-trichloroethane
*1.1-d1chloroethene
*chloroform
*benzeae
xylene
toluene
ethyl benzene
phenols
phthalates
0.015 (10-6 UCR)
1.8 (adjusted 10-6 UCR)
70 (adjusted 10-6 UCR)
0.95 (10-6 UCR)
22 (RA)
0.24 (10~ UCR)
0.19 (10-6 UCR)
0.70 (10-6 .UCR)
440 (HA)
2000. (RA)
680 (BA)
300 (taste)
3 (aquatic life)
aaee/Neutral
Inorganics
lead
copper
arsenic
cadmium
chromium III
chromium vt
mercury
z1 nc
nickel
iron
manganese
HCN
NH3(unionized)
11 (acjuatic
27 (aquatic
50 (HA)
3 (aquatic
341 (aquatic
11 (aquatic
0.012 (aquatic
710 (aquatic
150 (HA)
300 (taste)
50 (taste)
5 (aquatic
128. (aquatic
Ufe)
life)
life)
life)
life)
life)
life)
life)
11 fe)
* Carcinogens
o UCR - Unit cancer risk
o HA - Health Advisory Level
o For Inorganics. assume 260'ug/l CaC03.hardness. pH-7.5. and T-15°C
-7-
-------
Compound
*vinyl c:hloride
*tr1c:hloroethene
*1-2-dic:hloroethene
*l,2-dic:hloroethane
*l,l-dic:hloroethene
*chlorofor1l
*benzene .
l,l,l-trichloroethane
xylene
toluene
ethyl benzene
pheno 18
Base/Neutral
phtb.lates
*PNAa
*PCBs
. * Carci nogens
Compound
phenols
phthalates
*PNAs
*PCBs
* Carcinogens
Table 2
Soil Criteria For Or~anics
Criteria (ug/kg dry weight)
(10
(10
594
(10
<10
(10
(10
540
41,926
1,783
26,396
9,000
338,000
2,940
10,000
Table 3
(Detection Limit)
(Detection Limit)
00-6 UCR)
(10-6 UCR)
Sediment Criteria For Organics
Criteria (ug/kg dry weight)
-8-
1843
7183
1730
40 (background)
-------
Compound
volatUes
phenols
phthalates
*PNAs
*PCBs
Inorganics
lead
copper
arsenic
cad IIi um
chtomi UIII I I I
chrollium VI
mercury
dnc
nickel
iron
cyanide
8llllIIonia (unionzed)
Table 4
Surface Water Criteria Goals (ug/l)
Concentration
1000
2560
3
0.03
0.005
11
27
190
3
341
11
0.012
710
197
1000
5
128
(aquatic Ufe)
(aquatic Ufe)
(aquatic Ufe)
(wildlife and human
(background)
health)
(aquatic level)
(aquatic 18vel)
(aquatic level)
(aquaUc level)
(aquatic level)
(aquatic level)
(aquatic level)
(aquatic level)
(aquatic level)
( aqua ti c level)
(aquatic level)
(aquatic level)
Assume:
Ca C03 hardness. 260 mg/l. pH . 7.S and T . IS.C
* Carcino~ens
~.
~--
-9-
-------
Compliance with ground water protection goals, soil, lediment, and
surface water criteria will be determined in an additive fa.hion separately
for carcinogens and ~on carcinogens. The criteria outlined in Tables 1
through 4 will be in celculations of fractions of risk posed by each
contaminant. The individual fractions cannot exceed unity.
. Factors Used in Screening Remedial Technologies
- Technical Criteria
. applicability to site conditions (geology,
toPography, etc.)
. applicability to waste characteristics
. effectiveness and reliability
. implementability (construction). operation.
and maintenance)
- Environmental and Public Health Criteria
. except for risk posed by direct contact or
atmospheric dispersal of contaminants. criteria
to protect human health and the environment are
presented in Tables 1 through 4.
- Cost Criteria
. increased cost offering no greater reliability or
effect! veness
. increased cost offering no greater protection of
public health or. environment as established by
criteria
- Institutional Criteria (Compliance with other environmental laws)
. TSCA
. RCKA
. <:VA
. NPDES
. etc.
For a detailed analysis of technologies screened out see Section 11 of
the FS.
-10-
-------
Technulog1es Considered 1n Detail Include:
~round water remed1ation
. no action
. monitoring
. pumping wi th injection
ground water treatment
. no action
. monitoring
. flow equalization
. precipitation of aetals
. filtering
. air .tripptng of volatiles
. CAC filtering of exhaust gases
associated with air .tripping
.urface water remediation
. .ame technologies a..ground vater tr..tsent
.oil reaediation
. no action
. covering
. capping
. excavation
soil treatment
. offsite disposal
. onsite di.posal under RCRA cap
sediment remediation
. same technologies as soil
sediment treatment
. .ame technologies as s01l
-11-
-------
o
Alternatives Considered In Detail.
Five alternatives incorporacing the technologies considered in
detail were evaluated for remedial action. These 5 alternatives vere:
1 )
2)
No action with ground water and surface monitoring.
Grading, surface vater diversion, soil cover, revegetation, and
ground water and surface water monitoring.
3)
4)
Alternative 2 vtth ground vater pumping and treatment.
Alternative 3 with capping of soils exceeding criteria iq Table 2 and
dredging of sediments exceeding criteria in Table 3 vtth incorporation
under onsite cap .areas. .
Alternative 3 vtth excavation or dredging of soils and sediments
exceeding criteria which would be incorporated under. RCRA cap
constructed in the central portion of the aite.
Excavation of conta~nated 1011 to background levels vtth offsite
disposal in a RCRA regulated facility vas not considered in detail be-
cause of the high costS associated vtth chis option. tf i~ la assumed
that the average depth of foundry sand is 7 feet, 35 acres of land are
contaminated, and the average offsite disposal cost per cubic yard is
$300, this option would cost about 118 million dollars. .
5)
Alternative 1: No action vtth monitoring of ground water and surface
water, would function as a detection system to varn of increasing
contaminant concentrations In ground vater or surface vater. This
alternative i8 not appropriate because:
For Ground Water
o
present offsite ground vater contamination exceeds levels considered
. safe for human ingestton. Present offsite unit cancer risk equals
10-2.
.
EPA Region 111 calculations show that active restoration vtll remediate
ground water to safe levels much more rapidly than natural
restoration.
.
ground water pumping is technically fea8ible and °a well accepted
practice to reduce ground vater contaminant levels.
o
EPA'policy requires remediation of offsite contamination.
For 50tl
o
air dispersal of contaminated 80i1 particles presents a potential
human health impact.
-12-
-------
For Soil
.
air dispersal of contaminated soil particles present. a potential
human health impact
.
not remediating soil belov criteria outlined in Table 2 could,
cause continued or potential future ground water contamination.
.
present site conditions could cause contaminated .oi1s to be
. carried ofts! te duri ng heavy .toru by 8urface runoff.
.
present 80il contaminant level. may cause a direct contact risk
through dermal contact and ingestion by children.
For Sediment
.
not remediating .ediment belov criteria outlined in Table 3
may cause continued .urface vater contaadnation by de80rption.
For Surface Water
o
not remediating sediment and 8urface soil vill cause continued
surface vater contamination through de.orption of organics and
surface runoff from soil. Ground vater discharge to 8urface
vater could also impact .urface water quality. .
Cost
The FS estimates that this alternative would C08t about $1,500,000
over a 30 year pertod. Monitoring would ,include. analysis of
present monitoring vells and selected surface water locations.
Alternative 2: Grading, soil cover. revegetation. surface water management,
and monitoring of ground vater and surface vater.
In this alternative, the site would be graded to prepare a 80il cover.
Exposed 801id va.te or 81ag drums would be buried during grading activities.
The soil cover would consist of 18 inches of borrov material below 6 Inches
. of top .oil. All expo8ed area8 would be covered vtth .oil. Three or
four storm vater runoff ponds would be constructed, along vtth ero~lon
central benches and surface vater diversion ditches.
-13-
-------
The advantages of this alternative include:
.
elimination of public health risk caused by atmospheric dispersal
of contaminated soil.
.
reduction or elimination of erosion and surface water runoff
containing contaminated particles thus increasing surface water
quality. .
.
decrease in percolation of rainfall through the unsaturated
zone thus reducing the contaminant migration through. soil.
.
elimination of direct contact ri.k.
Disadvantages associated with this alternative include:
Costs
.
reduced, but continued flow of contaminants through the
unsaturated zone, thu. impacting ground water. Remediation
to propo.ed .oil :evels in Table 2 i. necessary to eliminate
future ground water. contamnation.
no remediation of present ground water contam~nant which may
pose s risk to future downgradient users or aquatic life or
wildlife living in or subsisting in Marshall's Run. Ground
remediation to ground water goals is necessary to prevent
future risk to. human health, aquatic Ufe or wildlife.
.
water
.
no remediation of sediment, thus impacting surface water quality
from the desorption of contaminants. Excavation to Table 3
levels is necessary to prevent future risks to aquatic life
and wildl1 fe.
.
in general, this alternative provides greater protection, but
still Is insufficient to prevent risks posed by contaminants.
Capital and operation and maintenance costs were obtained from the FS.
Capital co.~. include:
stormwater basin construction. $1,226,000 .
- .otl grading, clearing, cover, and revegetation. $2,064,000
offsite disposal of exposed drums containing solid wastes. $37,000
Total Capital Costs. $3,000,000
Operation and maintainance costs obtained' from the PS, including
monicoring:
. $1,700,000
Therefore, total costs are estimated to be $5,000,000
-14-
-------
Al terna ti ve 3:
Alternative 2 plus ground water remediation.
Ground water remediation would consist of up to 24 months of pumping
with some effluent being reinjected t,o upgradient ground water to increase
flow velocity. The construction and placement of reinjection and pumping
wells i5 illustrated in the FS. It is estimated that 435 gallons per minute
of effluen~ would be discharged to Marlhall'l Run over the tvo year period.
Effluent must be discharged to Karshall'l Run inltead of in a POTW
becaule Hillcreek Townlhip'l POTW sewer IYltem il at clpacity and the
closest lerver Iystem in the city il allo near capacity and would require
an additional pipeline.
Marshall'l Run prelently undergoes severe floodin~ on Harper Road at
the eastern boundary of the site during heavy rainstorms. Thil dilcharge
viII cause more frequent flooding and vorse flooding during rain Itorms
which could eventually cause onsite flooding. To remedy thil, lituation,
it is recommended that a flood retention basin be constructed along the
eastern border of the lite. Since Hillcreek Townl~ip has already purchased
land to construct a flood retention basin, ,nd since loil contamination
is present on Hillcreek Township'l property above loil criteria levels,
this loil would have, to be excavated anyway.
Excavation vould have to proceed to lafe ledi88nt levell, lince balin
s01 Is would be in di rect contact vi th lurfa~e vater. ' If lediment contami nant
levels are suitable after ground water remediation, the flood control
basin could be seeded to also function as a wetland tG .ome reltored area
lost to previous filling.
The level of ' ground water treatment viII
permit Which viII be developed during design.
is assumed that treatment will be extensive:
air Itripping and granular activated carbon.
be determined by a NDPES
For cOlting purpose., it
precipitation, filtration,
The advantages and disadvantages of Alternative 3 are the lame as
Alternative 2, except ground water risks would be reduced vith this alter-
native, While the ground water pumping and treatment program is 1n operation.
However, a disadvantage would be that cQntinued'long term contamination
of the ground water and surface water would occur because contaminants In
the soil and sediments would remain in place and continue to l~ach.
Cost
T6tal costs are the same as Alternative 2 with the exception of
ground vater pumping and treatment and the construction of a flood retention
basin which would cost an additional $3,012,000 and 5500,000 respectively.
Therefore, total costs -
$5,036,000 + $3,072,000 + $500,000
. $8,608,000 or about $8,600,000.
Ground water pumping and treatment and construction of the flood retention
basin are considered capital costs. Operation and maintenance costs would
vary vith the duration of pumping. If pumping occurs over a maximum of
2 years, 0 & H costs would be,similar to Alternative 2.
-15-
-------
Alternative 4:
Alternative 3 with capping
Alternative 4 is the same as Alternative 3 except it would require
capping over areas exceeding soil criteria and excavation of lediment.
exceeding sediment c~iterta. This alternative would require additional
sampling of soil and sediment to determine all areas on .ite exceeding
established criteria. Excavated lediment" would be incorporated under
one of several of the areas onsite. Also, since the flood retention
basin area contains contaminated loils, excavated loil would have to be
incorporated under one of .dveral of the capped areas.
Advantages
This alternative would hydraulically isolate soil contaminants in the
unsaturated zone exc4eding soil criteria, and thu. provide a high degree
of protection. Sediment exceeding criteria would allo be excavated,
surface water would also be adequately protected. Overall, this alternative
is technically feasible.
Di.a4vantages
The predominant disadvantage" of thil alternative is that most soils
exceeding the soil criteria levels lie in the eastern portion of the
site bordering Karshall'. Run. The eastern portion of the lite lies in a
flood plain and therefore capping in this area would not be in compliance
wtth RCRA regulation. Flooding could damage the cap by .couring. Also,
areas of .oil contaadnation exceeding criteria could be scattered throughout
the site and thus cause capping at many areas around the site. This
could cause difficulty in monitoring the capped areas for poslible seepage.
It would be more effective to cap one area and install additional monitoring
wells to ensure ground water compliance.
Costs
If it is assumed that an area 1200 feet by 500 feet would be capped
adjacent to Marshall's Run along with an additional 250,000 ft2 throughout
the site, and that contaminated soil from the retention basin and sediment
from the wetland and Harshall's Run is incorporated under one or more of
the caps, a total capped area of at least 1,000,000 ft2 or about 7 acres
would be required. Past experience has shown caps to cost about $400,000
per acre, so the capped area woul~ cost at least $2,800,000. Excavation
of sediments could cost an additional $500,000 to $700,000 based on the
potential excavation of all sediments in the wetland on Harshall's Run and
additional sampling and analysis could cost $450,000. Additional sampling
would be considered as part of design and thus would be funded as such.
Therefore, capital costs could approach
$6,900,000 + $2,800,000 + $700,000 + $450,000
- $10,850,000
o & H costs are exp~cted to be about $1,700,000 over a 30 year period.
Total costs are expected to be about $12,550,000.
-16-
-------
Al ternati ve 5:
Alternative 3 with Excavation
Alternative 5 would be the same .s Alternative 4 except that 80ils
and sediments exceeding established criteria would be excavated or dredged
and consolidated under an onstte, RCRA cap. The cap would be outside the
flood plai n .area. As wi th Al ternaU ve 4, Al ternathe 5 provides a high
degree of protection, but does not have Alternative 4's dis.dvantages.
A comprehensive ground water monitoring system would be established both
upgradient and downiradient of the capped area and involve additional
installation of monitoring veIls. Alternative 12 in the FS most closely
resembles Alternative 5, therefore, total costs should be similar except
for the required additional lampling and tnstallation of monitoring wells.
Additional sampltng could cOlt about $ 450,000. The additio~al
monitoring wells could cost approximately $60,000..
The Concept of Alternative 5 is as follows:
1) Pump ground water for an initial period of time determined during
design not to exceed 2 years. After the tnitial period has passed,
groundwater protection g081s will be reevaluated to determine their
technical fea8ibility. . At that time, pumping could continue at the same
or new loals or be discontinued if the desilned loals were ..t. This
strategy il necessary because the effectiveness of pumpinl to reduce
VOC contaminants to ug/l levels over a long time period is unknown.
The proposed goals for the first period are outlined in Table 1. These
loals will be reevaluat~d during design to ensure technical feasibility
and protection of human health and the environment.
'.-'
. J.40.
2) Treat ground water to levels consistent with NPD!S permit
standards. Treatment would include the construction of a flood retention
basin on Killcreek Township's property for flow equalization and prevention
of onsite and offsite flooding. Ground water may also be treated for
inorganic and organic removal by precipitation, filtration, air. shipping
and granular activated carbon. Cost estimates assumed extensive treatment.
The NPDES standards will consider technical feasibility and protection of
aquatic life and humans or wildlife which may ingest aquatic life.
3) Excavate 50i1 to proposed levels outlined 1n Table 2 and incorporate
under an onsite RCRA ~ap. Region 111 will reevaluate the criteria during
design. Criteria for PCBs and PAHs are in Table 2. The criteria presented
in this ROD are for cost esttmates on1y.
Technologies to reduce contaminant levels to 80il criteria will be
considered to decrease the volume of excavated 80ils and their associated
costs prior to excavation.
As previously explai ned, so11 cri ceria are calculated usinl many
variables, two of which are the area and concentration within that area of
contaminants. Further sampltng will be required during design to lain
additional information on the areal extent of soil contamination, especially
in the eastern part of the site where elevated concentrations of volatiles
are expected to be present.
-17-
-------
4) Excavate sediment to proposed levels outlined in Table 3 and
incorporate under an.onsite RCRA cap. As with loil criteria, lediment
criteria in Table 3 are for cost estimates only. Region III will also
reevaluate lediment criteria during design. .
5) Any drums found during lampling or earth moving activitiel will
be laap1ed, either individually or al a compolite, as appropriate. Drums
containing lolid non-RCRA waste will be consolidated under the cap. Any
Uquid filled drwu or drums containing RCRA hazardous waltel will be
disposed of off-site.
6) Place loi1 cover over remaining areas of lite not exceeding loll
criteria to protect againlt atmospheric dispersion of contaminated loi1.
7)
Grade and revegetate loi1 and cap areas.
8) Construct 8urface water management basins to control run on,
run off and erosion.
9) Install additional monitoring wells around the cap and other
areas on site to detect possible future releasel.
Total capital costs ranges for 10-4, 10-5, and 10-6 UCR
values are estimated to be 9-11, 10-12, and 12-18 million dollarl. EPA
has chosen a 10-6 UCR value which is consistent with policy and EPA's
long-term ground. water protection strategy. Therefore, including additional
sampling and monitoring wells, COlts are expected to range from 12 million
to 18 million. These costs are estimatel based on the soil criteria
developed by EPA Region III.
Figure 2 illustrates a conceptual sketch of this alternative.
Recommended Alternative
Alternative 5 is the only alternative complying with other environmental
laws and remediating the lite to safe ground water, surface vater, soil,
and sediment levels to protect human health, aquatic life, and wildlife.
Based on our evaluation of the cost-effectiveness of each of the proposed
alternatives, the comments received from the public, the state, and
potentially reponsible parties, information from the RIIFS and Region III
technical support documents, Region III recommends that Alternative 5 be
implemented. .
-18-
-------
II :• ' \ - I ,
'•I ' ) ~ ' ) ,
.1 I, } /
I I < 1
i
TCMPOAAMV
KWATEMN« MO
STOMMVATCM
NlMOFFMNb
APPROXIMATE
u»m or SITE
-------
Consistency With Other Environmental Laws
The reco-.ended alternative was evaluated to determine consistency
with applicable or relevant and appropriate environmental laws.
The transport and offsite disposal of drums wtll have to comply
with all applicable I~ regulationl regarding the transport and
disposal of ha~ardous wastes.
The surface water discharge from the groundwater treatment facility
will comply with NPDES discharge require~nts.
The cap will be designed and constructed to comply with the RCRA
capping requirements of 40 CrR 1264.310(a).
EO 11988. Floodplain Hanagement. will be complied with through the
construction of a flow equalization and prevention basin. which will
8inimize the impact of this action on flood hazards. Additonally.
contaadnated .oill and lediments will b. re80ved fro. the floodplain
and the loils will b~ consolidated and capped outside the floodplain.
EO 11990. Protection of Yetlands. will be complied with through
the construction of the flow equalization and prevention basin which
can be constructed and revegetated to serve a. a wetland providing
the benefical uses described in EO 1l99~. In addition. contaminated
sediment will be removed from the remaining wetlands on-site. The
action to remove the sediments will be designed to minimize the harm
such action will have on the wetlands.
The basin and the run-off/run-on control system will be designed to
meet the requirements of the RCRA regulations in 40 CFR 1264.301~c),
(d), (e).
The establishment of 80il and sediment criteria. consolidation of
material exceeding these criteria under a RCIA cap and a ground water
monitoring program to verify that the ground water protection goals are
met. complies with the CERCLA policy for consistency with the Safe
Drinking Water Act and RCRA regarding ground water contamination and RCRA
Closurel Soil Contamination Requirements. This CERCLA policy i8 described
in the "preamble to the National Contingency Plan published in the Federal
Re~hter on November 20. 1985. on pages 47922-47923.
. The recommended alternative i. also con.istent with the EPA's forth-
coming Superfund groundwater strategy as discussed In a March 24. 1986,
memorandum froe J. Winston Porter. Assistant Administrator for Solid
Waste and Emergency Response. to James Seif. Regional Administrator.
Region III.
- 19 -'
-------
Costs
The projected costs were developed 1n accordance w1th EPA pol1cy for
estimating costs within a reasonable'range (-30% to +50%) of the actual
implementation costs. Total projected present worth costs range 'from
$14.800.000 to $20.800.000 with a baseline estimate of $17.800.000.
Design Costs -
Additional sampling and monitoring vells will be considered a8 part
of the design. Desi~n is estimated to cost approximately $1.000.000.
Design will be funded 100% by trust fund monies.
Capi tal Cos t-
Capital cost estimates vary from $12,000,000 to $18,000,~00, with an
estimated baseline cost of $15,000,000. For these estimates, capital
costs included all costs associated with excavation. regrading. revegetating,
capping and groundwater pumping and ~reating for tvo years (although pumping
may be needed for a period beyond tvo years, tvo years vas selected for
cost estimating purposes). Trust fund monies will be used to pay for
90% of these C08tS and the State of Pennaylvania will finance 10% of
these costs.
Operation aqd Kaintenance (0 & H)-
Total present worth costs for 0 & H i8 $1,763,000.
The components of the recommended alternative that may
require operation and maintenance are:
- RCRA Cap
- Surface water management systems
- Flood retention basin
- Monitoring (excluding that necessary to monitor the effectiveness
of the pumping and treatment program while it 15 being financed
by the trust fund).
. The above listed items vill be considered normal operation and
maintenance and will be the responsibility of the State of Pennsylvania
six months 8ubsequentto completion of construction.
The ground water pumping and treatment program will be considered part
of the approved remedy for a period of at least tvo years. If targets
are not reached after tvo years of remedial activity the Regional Administrator
vill determine if it is technically feasible to reach those targets. If
further pumping and treatment are required, this will also be con&idered
as part of the approved remedy and eligible for Trust Fund monies with
90% of the program financed with trust fund money and 10% financed with
State money.
-20-
-------
1-
Schedule
Ap prove ROD
Start Predesign
Coaplete Predesign
Award Superfund lAG to
U.S. Aray Corps for Design
Start Design
Coaplete Design
Award Superfund
tODtract for Construction
Start Construction
Coaplete Construction
- - --iC---
4/86
6/86 .
12/86
1/87
3/87
12/87
2/88
5/88
12/88
\..
-21-
-------
]'ah~~ Descriptions of Alternatives Considered at the Hillcreek Site
Al ternatl ve
Al to I: No Action
with Surface Water and
Ground Water Monitoring
Alt. 2: Grading, 8011
Cover, Revegetation,
Surface Water ManaRe-
ment, and Monitoring
~
~
~
Alt. J: Alt. 2 Plus
Cround Water Reaedl-
a tI on .
Alt. 4: Alt. J plus
Capping of Areas'Ex-
ceedlng Soil Criteria
A't. S: Alt. ) plus
F.xcAvatlon of 501ls
and Sedl8ent ExceedlnR
Criteria with On Site
Olr I under a RCRA
CAt
Total Cost
x 1000
I,SOO
S,OOO
8.600
12.S00
14.800
to
20.800
Public Health and
Environmental Goals
Does not meet ACL, soil, sediment, or
surface water criteria or goals.
Site would continue to pose a potential
rlak to public health, aquatic life. and
wtldl~fe. this alternative would do
little but warn of worsening conditions.
Would eliminate risk posed by direct con~
tact and Inhalation of contaminated par-
ticles. Would decrease .future surface
water contamination by preventing runoff.
this alternative. though. would not reaedl-
ate ground water. soil, and sediment con-
tamination to acceptable levels. and thus
provlde8 Inadequate protection to receptors.
Offers saae benefits previously described
In Alt. 2 plus reaedlatlon .of ground water.
Thl8 alternative, though, would not eliminate
the potential for future ground water and
surface water contamination caused by 8011
and sediment levels exceeding criteria.
By capping areas exceeding soil criteria and
excavating sedl.ent exceeding sediment cri-
teria, this alternative would provide 8 high
degree of protection. Conta.lnated sediment
would be Incorpor,ted under an onBlte cap
area. A major disadvantage of thts alterna-
tive Is that the pred081nant area of capplnR
lies In a flood plain.
Would provide a hlRh deRree of protection
aeettng a1\ criteria and 8tandarlle estab-
II shed.
Compliance With Other
.Envlronmental Laws
Would not co.ply with
RCRA closure requlre-
aents npr with EPA's
current ground water
protection strategy.
Saae as above.
Would coaply with EPA's
ground water protection
strategy and RCRA's closure
policy for ground water
contamination. It would
not coaply. with RCRA's clo-
sure requirements for con-
taminated eo I I and sediment
Would not comply with RCRAs
closure requirements since
capping would be required
In a flood plain.
eoaplles with all appli-
cable envlron.eantal laws.
-------
!-
Millcreek Responsiveness Summary
A responsiveness summary usually accompanies a Record of Decision
(ROD) to provide EPA an opportunity to respond to comments aade or
submitted by citizens. environmental groups. or Potentially Relponsible
Partie. (PRPs) on the Remedial Investigation/Feasibility Study (RI/FS)
or other previous studies conducted by Federal or State agenciel used to
formulate a remediaL action at a CERCLA site. The responsivenell summary,'
along with public meetings. informs concerned citizens and PRPs of EPA's
assessment of data collected and recommended means of remediating any
risk posed to public health or the environment.
EPA. Region III. conducted a public meetin~ on September. 11. 1985.
to inform citizens of the findings from the RI/FS and to propose a
recommended remedial alternative. The topici discussed during the public
meeting are outlined in Appendix RS-l.
Since any comments submitted by citizens or PRPs should be a matter
of public record. all correspondence received by EPA il. contained in
Appendix RS-2.
PRPs' Comments
The PRPs' comments will first be summarized and responded to since
their comments are the most extensive and in-depth. To aid in the technical
review of the RI/FS, the PRPI consulted with the IT and Environ Corporations.
.
Ground Water
The first major topic il that EPA's RI/FS does not have sufficient
data to properly characterize and asses. the risk posed by ground water
contamination, and that there is no need for active remediation because
ground ~ter is not utilized downgradient of the site. Specifically,
thePRPs claim that: .
. temporal (time-variant) characterization of groundwater quality
cannot be assessed with present data.
. insufficient data to evaluate the extent to which natural
restoration will occur.
. downgradient ground water is not currently being used and Is
unlikely to be used in the future because of the proximity of Lake Erie
and the current distribution system servicing residents of Millcree~ Township.
. the upgradlent public water supply (Yoder Wells) is unlikely to be
affected because of hydrogeological factors and the fact that the 1I811.s .
only use 1 part well water for 3 parts Lake Erie water in their distribution
. system.
-22-
-------
. downgradient use Is rendered-even less likely because the Hillcreek
Township supervisors are willing to pass an ordinance restrictin8 any
future use.
. EPA has been confronted previously with similar decisiona at
other aites (2 within legion IIli Wade and Drake Chemi~al) at which it
has decided not to take active ground water remedial ..asurea. -
. available data indicate that ground water contamination Is
decreaaing with time and Is thus already naturally reatoring Itaelf.
. -
. EPA's analysis of biodegradation of TCE into vinyl chloride which
would in~rea8e the risk of concern from potential ing.ation 1a incorrect.
There do not appear to be any temporal or spatial trends in ground
water data which Indleate that biodegradation i8 occurring. In addition,
current literature indicates thst biodegradation of TCE ahould not occur in
conditions present in Millcreek ground water.
RespoDae:
The decision whether to pursue active ground vater remedial measurea
Involves not only a technical rlak assesament, but a policy decl.lon on
EPA's part to protect currently unused aquifer.. Beca~a. of the latter
concern, Region III delayed the i8suance of the IOD In order to obtain
clear guidance from EPA-HDQ5. EPA Is currently uncertain about the long
term effectiveness of Inatitutional controLl aa a meana of r.atrlctlng
ground water uae. As a reault, the Superfund program generally doe a not
encourage ground water remedies with long time-frames. EPA'a preference
Is for rapid restoration as ..y be achieved by pumping ground vater
(e.g. 2 years) as opposed to long term restoration vhlcb ..y take in excess
of 100 years as calculated in Appendix A.
In response to specific comments-, Region III collected additional
ground water samples In December, 1985,to better characterize temporal
trends in the eastern portion of the site. Sampling points were restricted
to monitoring wells In the eastern portion of the site since this is the
only area where contaminants of concern were detected.
With the December VOC sampling Region III.believes that aufflcient
data Is available to generally characterize the temporal ground water
trends onstte.
In regard to the possibility that the upgradient public supply veIls
will be affected by the site, EPA could conceive of a problem during a
proionged severe drought. EPA will recommend periodic upgradlent monitoring
well sampling If these conditions should occur.
In regard to the decision of the Millcreek Township supervisors to
restrict future downgradient use, refer ~o their recent letter to the
Regional Administrator in Appendix R5-2.
In regard to other ROD decisions such as Drake and Wade, a decision
on Drake has been deferred and -ground water at Wade was not remediat~d
because it lies adjacent to a major river (Delaware) where the areal
-23-
-------
extent of the aquifer was very restricted. Therefore, at this si te there
is a set of conditions distinct from Wade in both dimension and time of
contamination, and in the technical feasibility of ground water restoration.
As previously mentioned, during the December resampling, EPA sampled
two SUMp pumps on Harper Drive. In one sump pump hole, 15 ugll of vinyl
chloride was dectected. Subsequent CLP analysis detected 190 ugll of
vinyl chloride. The owner of the residence explained to the EPA project
manager that he experiences periodic basement flooding during high water
table conditions. The year-around presence of water in his sump hole
indicates that his b~.ement is very close to the ground-water table as are
other basements in the area. There is . possibility of volatilization
of volatile organic chemicals such as l,l,l-trichloroethane, vinyl chloride,
and l,2-dichloroethene from the water table into people's living space
near the site. . EPA i~tends to sample this basement air for volatiles
where vinyl chloride vas detected to confirm or deny this possibility.
EPA agrees that chlorination of drinking water raises the VCR
associated with ingestion, but this matter Is irrelevent to the
deci~ion to pursue active ground water ..asures. Chlorinating .an already
contaminated ground water supply would aggravate an already potentially
harmful situation. Also, ground water contamination may last for decades,
whereas an alternate means of disinfection could be utilized in a shorter
period of tiMe. Also, chlorination at present offers a benefit to consUmeri,
whereas chemical VOC contamination offers none.
Finally, in regard to TCE biodegradation, the following response is
provided. Although there is considerable controversy conc*rning the
dehalogenation of short chain chlorinated aliphaties, specifically trichlo-
roethene (TCE), and 1,1,1-trichloroethane (l,l,l-TCA) in environmental
. matrices, a growing body of evidence seems to indicate that biologically
controlled dehalogenation can occur under favorable field or laboratory
conditions. The Environ Corporation presented a summary of some available
data that indicating that degradation occurs during reducing, anaerobic
conditions in ground water and in aqueous laboratory samples. Recently,
a number of researchers from EPA, Ecology and Environment, Inc. and the
University of Missouri showed that anaerobic degradation of TCE can also
occur in soil (Environ Sci-Technicol, 1985 - 19, 277-279). TCE and
l,l,l-TCA have also been observed to undergo dehalogenation under reducing
anaerobic conditions in ground water. near municipal landfills and solvent
recovery facilities (Management of Uncontrolled Hazardous Waste Sites,
November 1984, Page 217). It is interesting that at one solvent recovery
site, the ground water table was shallow and within 15 feet of the surface
as is Millcreek.
Environ stated that biodegradation of TCE in ground water at the
Millcreek site is highly improbable because ground water at the site is
probably anaerobic and monitoring data show no clear patterns sugiestive
of degradation. Region III disagrees wtth the latter statement and
believes that monitoring data strongly suggests that conversion of TCE
in 1,2-DCE and VCM is occurring. This finding is explained in more
detail in Appendix A. Degradation is resulting hehalogenation which
could conceivable be caused by hydroyses but is more Uke to be caused by
biodegradation. When one compares the concentration of TCE to its breakdown
-24-
-------
products in umoles/l. a clear pattern emerges which not only suggests
biodegradation but a probable source area for TCE.
On the_question of the absence of a reducing environment, an assump-
tion should not be sade that just because an aquifer is Ihallow that it
will alwaYI be aerobic.. Besides. the literature indicates that although
degradation occurl 80re rapidly during anaerobic conditions it ..y also
occur in aerobic conditions at a much reduced rate. legion 111 attempted
to deteradne the reducing potential in ground water with an Eh Heter but
was unsuccessful due to the malfunctioning of the instrument. EPA did.
however, obtain aealurements of dissolved oxygen which varied from 1.60
to 4.30 ppm onsite. Most values were around 2.0 ppm. This indicates a
low oxygen but not an anaerobic environment. It is interesting though that
the concentrations of dissolved iron and manganese are very high in
onaite ground water and that ground water seepage to drainage ditches
appears as an orange tainted ooze. Orange leepage near landfill and
swampy areas are indicative of iron and ..nganese oxidation occurring
when vater containing reduced iron and ..nganese come in contact with
air. Dislolved iron and ..nganese are typically found in reduced foras
when these two inorganici are detected in high concentrationl in.ground
water. Therefore, the ground water at Hillcreek could be in a reducing
environment. "
o.
Even if the ground vater is only slightly reduced, degradation
could still occur at slower rates. One point that must be emphasized
from laboratory studies il that laboratory tests are not sensitive enough
to detect biodegradation rate slower than 0.001 days -1. This is pointed
out several times in Environ literature review. It must be remembered
that ground water is an extremely slov medium and that persiltent contami-
nants such as TCE typically persist for decad~s. Region III calculated
the biodegradation rate of TCE into 1.2.-DCE and 1.2-DC! into VCM based on
their predicted mass balances in the aquifer. The degradation rate for the
foraer vas estimated at 0.0003 days -1 whUethe degradation rate for the
latter vas estimated at 0.000008 days -1. While these appear as
extremely low rates under laboratory conditions. they are meaningful in
.site ground water when reactions occur in years or decades intead of
weeks in laboratory conditions. The point that TCE degradation can
slowly proceed under weakly reducing or even anaerobic conditions is
supported by a number of researchers as evidence in Environ literature
survey." Another indication that biodegradation is occurring perhaps even
as secondary metabolic reaction. is the presence of lov concentrations of
TIC hydrocarbons (1 ppm) in ground vater in the eastern part of the site.
Some studies have shown that microbes can use other organics in ground
vater as a carbon source and metabolize halogenated alkenes and alkanes
by cometabolism.
Thus in conclusion, Region III believes it has sufficient ev~dience
to document degradation of TCE and perhaps also l,l,l-TeA. The literature
provided by Environ provides for a better understanding of the process,
but does not disprove that TCE. and 1.2-DCE are undergoing biodegradation.
at this site.
-25-
-------
o
Soil Contamination
The other major area where IT submitted comments 15 on 80il.contami-
nation, specifically the mobility of PCBs, PABs, and metals. The following
summarizes their comments.
. EPA has no scientifically valid data to assess VOC lubsurface
loil contamination since aost VOC data was rejected during validation.
. A scientifically sound determination as to the need for soil
removal to prevent ground water contamination cannot. be made because
analy.es of waste fill and natural soils were not differentiated In the
RI and no leachate data from direct field measurements or column tests
exist. The existing data suggest soil removal is not necessary at this
site.
. Adequate surface soil data does not exist to allow a delineation
of "clean" VI. "contaminated" loil at the site.
. All highly 8O~11e VOC .ay have already entered the ground vater
from the 80il column.
o EPA's method of determining soil criteria from partition coefficients
and completely reversible linear isotherms is not a Icientificslly valid
approach. A better method is using batch or column desorption test or
monitoring ground water samples to show that metals, PABs, and PCEs are not
migrating to the ground water.
. EPA has developed PCB soil criteria at other sites such as Lehigh
Electric at 10 to 50 ppm yet the PCB criteria at tne Hillcreek is much
lower.
Responses
Only a minor fraction of the VOC soil contamination data was rejected
during validation. For the most part, surface and subsurface VOC data are
believed to be accurate with the exception of the detection of acetone and
methylene chloride. These two contaminants are ~ommon field and laboratory
contaminants. The distribution pattern of acetone and methylene chloride
in soil indicates that they are probably blank contaminants.
In regard to the determination of the need for soil removal to
protect ground water, Region III will conduct additional field
and laboratory work especially with more hydrophobic chemicals
like PCBs and PARs.
In regard to PCBloil contamination, surface soil is obviously
contaminated with moderate levels of PCBs. Host PCB contaminated"soil
seem to be on Hillcreek Township's property and in the south central
portion of the site near the existing wetland. Test pit samples indicate
-26-
-------
PCBs 1n foundry. land at var10us depths. Test pit 016 conta1ned PCB-1248
1n natural soil at a concentration of'357 ug/kg. PCBs were detected in
two borehole samples, one of which 1s ofhice directly southeast of the
site where previous bulk dilpolal 1s suspected. In both boringl, PCBs
were detected in natural s0111 at depths up to 10.5 feet.
Test pit and bor1ng data 1s very limited, but there is an 1ndication
that PCBs have migrated 1n low concentrationl in natural lubsoil. Boring
20A indicates that PCB-1254 has migrated to the aquifer.
The fact that PCBI have not been detected 1n .onitoring wells is
expected. Host lIOn 1 tor1ng _11s on81 te 'and 1n sUlpected PCB .contai nated
areas 11e below 111ty or clayey residual.oil which would 11gnificantly
attenuate PCB migration to the aquifer. Very few .onitoring welll 8Oaitor
perched water zones lying above clayey or lilty residual s01l 1n foundry
sand. PCBs could be concentrated in thele areas. Another reason why
PCBs vere not detected in ground water 11 that the CLl detection 11mits
are too high. Depending on the ilomer class, the aqueous detection limits
for PCBs are either 0.5 or 1 ug/l. PCBI could cause long term health
effectl in the part per trillion range.
Therefore, it is inaccurate to Itate that PCBs have not .igrated and
are ~ot containating ground water at levels which .ay be of concern.
Likewise, it is allo 1naccurate to Itate that PCBs will not .ig~ate in the
future. Data needed to confirm or deny the presence of PCBs in ground
water will be gathered during design.
It is not generally recognized that hydrophobic compounds vill, in
fact, migrate in moderate or low concentrations. PCBs were probably bulk
disposed with solvents or ~tlswhich could have carried them through the
80il column without signiffcant adlorption. PCBs could then be relealed
from aoil via desorption or molecular diffulion. Th11 occurrence il commonly
observed in creosote pi ts or coal tar spUls. . It should be recognized
also, that dissolved organic carbon competes with organic carbon in soil
for hydrophobic adsorption sites. Hydrophobic compounds, especially
phthalates, are commonly observed in landfill leachate where DOC can exceed
1000 mg/l. It should be noted that garbage was disposed throughout
the site and in some areas where PCBs were detected. Garbage would
naturally generate DOC and serve as 'an adsorbent for PCBs.
To confirm the presence of PCBs and other hydrophobic molecules in
subsurface water, additional monitoring wells or lysimeters need to be
installed and anayls1s performed with lover detection limits. These
activities will be recommended dur1ng design along with desorption tests
to aid in determining lafe, residual PCB level I in foundry land and s01l.
In regard to PAH loil contamination, PAHa have been detected in
surface soil throughout the site at concentrattons exceedtng 800 ppm tn
soil. The highest concentrations leem.to be located in the louthern and
western portions of the site. The distribution pattern of PAH contamination
indicates that they are the result of either massive and widespread bulk
disposal or were used with foundry sand as a binding agent. Phenols are
also commonly used in foundries and were detected with PAHs in many instances.
It appears that for most parts of the site, PAH contam1~at10n was from the
-27-
-------
latter mechanism. IT suggest that p~ contamination could be from
coal powered locomotives which were comTllon in the past. Region UI,
though, believes that this is unlikely 8ince PAHs are distributed in
moderate concentrations throughout the 80il column (a condition
that would be unlikely because of strong hydrophobic binding) and because
the highest concentrations of PAHa were detected on property which was
only filled in 15 or 20 years ago.
The well boring samples may provide some indication of the transport
of PAHs to the underyling residual 80il. While this is important in res-
pect to the potential for aquifer contamination it must be remembered
that PABa are distributed fairly evenly in foundry sand, and.local perched
water tables commonly encroach 1 or 2 feet into foundry sand. Perched
water conditions may enhance the ability of PARs to desorb from foundry
sand and eventually m~grate to the aquifer. It is believed that PAHs
were not detected in ground water for the lame reasons that PCBs were
not.
Boringl provide the best opportunity to observe any possible
migration. In B-21A, seveTal PABs were detected below CLP detection
limits (commonly about 1500-2000 ug/kg) at 9-10.5 feet. This may be
significant since the fill only extends to 5.5 feet. The same situation
was observed in boring 18. In boring 18, PABs were detected in foundry
sand at 3-4.5 feet at a concentration of 539,510 ug/kg. In the same
boring at 4.5-6 feet, PAHs were detected but at concentrations below the
CLP detection limit. Residual soil was found at 5.5 feet.
In regaTd to inadequate sampling data being available to delineate
areas of contaminated soil, suf'f1c1ent data is available to arrive at an
ordeT of magnitude cost estimates (+50% to -30%) of potential remedial
alternatives. Region III agrees that additional lampling must be done
during desigD to refine the, estimates. Soil cTiteria is based on the
physical, biological and chemical properties of the compound and soil to
which it is adsorbed and on the area and-depth of contamination. Only
in this way can an estimate be made of contaminants percolating to the
perched or real water table. These are tentative criteria which will be
redeveloped during design based on additional sampling, field, and labo-
ratory desorption tests. .
In regard to all VOCs having already left the soil column, test pit
samples Teveal that low to moderate levels of VOCs still exist in soil In
some areas. For instance. in test pit 7 at 2.5 feet, 733 ug/kg and 307
ug/kg of vinyl chloride and 1,2-dichloroethene were detected. Additional
sampling viII be conducted during design to more precisely define the
full extent of VOC contaminated soil.
In regard to selecting lower PCB levels than had' already been chosen
in previous RODs, Region III is not bound to previous soil criteria
selected by other Regions or within Region III itself. The proposed PCB
cr.iteria is a health based number for residential areas. This is
appropriate because the site is frequented by hunters and children and
borders residential areas.
-28-
-------
Marshall's Run
. There is no current information on conditions in Marshall'. Run.
Thus the need. if any, for relledial action to limit surface water exposure
cannot be determined.
Re.ponse
Both the onsite pond and Karshall'. Run support wildlife and migratory
fowl. Hunters use the .wamp area duriag .pring and fall for huating
waterfowl. Red tail fox and other wildlife have been .een onsite.
Wildlife and fowl ingest surface water and phytoplankton suc~ as duck
weedirow1ag in .urface vater. Al.o aquatic life .uch as .hiaer. and
811a11 bass were found adjacent to the site in Marshall's Run... Rainbow
trout and other cold water ftsh .pecies use Karshall'. Run for .pawning
at the mouth of Lake Erie. Extsting data .hoVlthat the exi.ting wetland
and Marshall's Run are viable pathways to wildlife receptors..
Air
-
. No organtc vapor or particulates analysell of air s..ples were.
conducted. Thus, no data exist to justify.eliadaatlng or reducing potential
air exposure routes.
Response
EPA's consultant, the NUS. Corporation, conducted volatUe air .oaitoriag
during field reconnaissance test pitting and .cnitoriag well drilling.
During tellt pitting, the OVA meter registered over 1000 ppm thus signifying
the presence of volatil~s in high concentration. EPA did not collect.
tenax air samples though for subsequent analysh aor samples for particulate
analY8is. EPA's asseSllment of the risk posed by Inhalation of particulates
is based on.the presence of known carcinogens in high concentrations in
soil and the possibility of air dispersion of soil.
On windy days, especially in areas with little vegetative cover, significant
at r di spersal of dusts m.y occur.
-29-
-------
Response to Comments Other than PRPs
Erie County Department of Health
C01ll1llents:
Agree with proposed remedial action alternative except that 50il
exceeding determined criteria should be removed offsite instead of
incorporated onsite under an impermeable cap because it would still be
exposed to ground water.
Response:
50il exceeding determined criteria would be hydraulically isolated
from la~eral ground water flow and percolating rainfall because soil
would be placed above ground and then capped. .
Millcreek Township
C01llDents:
Township favors the construction of a storm water retention
basin on its parcel of land on the aite to reduce offsite flooding.
Township acquired its parcel after the site was no longer being used
as a landfill.
Township is concerned that placing a cap over contamiriated soil
will fail to contain hazardous substances within the site.
Township is concerned that Yoder Wells be protected and that
adequate IDOnitoring be maintained to ensure a safe water supply.
Response:
EPA will contruct a storm water retention basin onsite if it is
needed to equalize ground water discharge to Marshall's Run, or is more
cost-effective than backfilling the PCB contaminated areas with clean
soil. A retention basin may also be needed to prevent erosion of PCB-
containing residual soils during flooding conditions since PCB con-
taminated &01ls are within a 100 year flood plain.
EPA's enforcement personnel are aware of conditions involving purchase
of property formerly owned by Ralph Reihl.
The cap will eliminate lateral ground
rainwater percolation through contaminated
isolation is not technically feasible, but
should be in low concentrations and dilute
health by the time they migrate offsite.
water flow and greatly reduce
soil. Complete hydraulic
contaminants present in seepage
to levels protective of public'
-30-
-------
EPA will vork with the Erie County Department of Health to establish
a periodic sampling strategy for the 'Yoder Wells. Monitoring welll both
upgradient and downgradient of the site will allo be sampled periodically
to deeermine if offaiee migration of ground water contaminants t. occurring.
If ehe Yoder Wells become threatened by ehe lite, EPA or PADEI will
expeditiously take action to remedy the situation.
Norehwest Cieizens for a Clean Environment Comments:
Concerned ehat capping will not effectively eilidnate the threat
po.ed by the site. Request that EPA provide documentation that
capping is an effective means to ilolate vastes. .
Beiieves that additional monitoring veIls should be placed beyond
che landfill.
Believes that ground vater pumping and treatment should be established
on a cycle of two year pumping to purify water.
lequese .ore frequene aonieoring vell .ampling.
Signs and a fence .hould be lnatalled.
What specific responsibility vill.be ..signed to agencies for operation
and maintenance.
Rl/FS .tate. thae a health .urvey is being developed and the citizens
co.adttee would like to be kept informed of its progres..
Requese a mon ehorough vri teen explanation of EPA's ineent to
develop site specific criteria.
lequest that EPA specify responsibilities of EPA, DEI, Erie County
Health Department, and Millcreek Township during cleanup and monitoring.
lequest that a t~me schedule be presented for cleanup.
Want responsible parties at fault to be prosecuted and bear the full
costs of cleanup.
ravor ehe con.eruceton of a holding ba.in.
Concerned that recommended alternative may prove more cOltly in the
long run.
'If better technologies are available at a later time for stte
remedtation that they be applied at the .ite.
-31-
-------
Response
Capping has proved to be an effective method of greatly reducing
rainwater infiltration at hazardous wastes facilities and is an
established engigeering practice. Capping must be strictly evaluated
in this context and viewed as part of the overall cleanup strategy.
By excavating contaminated soils capable of causing continued ground
water contaadnation, EPA will greatly reduce the leachate generation
caused by lateral ground water flow. By incorporating this contaminated
soil underneath a clay cap wi th a pet1leabiU ty of no more than 10-7
emIl, EPA will ensure leachate generation caused by rainwater percolation
is kept at a minimum. Subsequent dilution in ground wat.r will result
in acceptable levels offsite.
EPA is evaluating the need for additional monitoring wells to ensure
that any future releaae would be quickly recognized and remediated if
necessary. Sampling of monitoring wells il an essential mechanism
in determining the success of cleanup activity and thus EPA places great
importance in the location and sampling frequency of welli.
EPA intends to install warning signs around the site when funds become
available. Also, EPA is again evaluating the need for a fence around
the site and will make a decision when funds beco.e available.
EPA's Remedial Investigation/Feasibility Study consultant believes
that removing most of the volatile organic contaminants from the
ground water and pore .paces. within the ground water ..yrequire
removing 12 pore volumes of water which could take two yeare in
time. If monitoring welle indicate that ground water remediation is
accomplished in less than two years, pumping wells may be turned off.
EPA will establish acceptable levels of groundwater contamination
and then pump to meet those levels.
Frequency of monitoring well sampling will be established to detect
a ground water release as soon as possible.
It will be EPA's specific responsibility to develop a remedial action
alternative and implement it through the Army Corps of Engineers.
It is then DER's responsibility to provide future maintenance
of the remedial action and sampling of the monitoring wells. The
Erte County Department of Health will be responsible to periodically
sample the Yoder Wells.
A health study has not been initiated at this site. Available data
indicates that lufficient information is not available to warrant an
epidemiological study. EPA may however request the Centers for
Dieease Control CCDC) aeeietance in conducting a study if additional
soil sampling during design reveals information which may warrant it.
-32-
-------
A more thorough explanation of spi1 criteria is contained in
Appendix B.
A cleanup schedule is provided in the ROD.
Concerning the long-term cost of cleanup, EPA has attempted to be
conservative in its calculations to minimize any possible rilk in
the future and thus reduce any possible future costs.
In regard to the use of better technology at a later date. EPA may
consider this option if unforeseen events occur which produce a
significant risk to public health or the environment and it i.
apparent that the technology originally uled il not alleviating
this risk. .
Parent Teacher ASlociation - Tracey Elementary School
Comments:
Ask that PRPs be prolecuted to the full extent of the law.
Ask that d.te area be co.pletely fenced in.
If a storm water retention basin i. constructed it should be fenced
in if it could cause a potential hazard.
Wish for more extensive monitoring of the Yoder Welli.
If contami nants are at r - di spersed duri ng lite activities, every
precaution should be taken to prevent any polsible health risks.
-
Response:
EPA is actively negotiating with the PRPs for implementation of the
remedial design and action. If the PRPs refuse to implement the
above, it ts EPA's opinion that they are liable for costs of Superfund
actions implemented at the site and are subject to the cost recovery
provisions of Section 107 of CERCLA.
If a storm water retention basin is constructed EPA will ensure that
it does not present an acute or chronic health risk.
EPA will work with the Erie County Department of Health to ensure
that the wells are periodiCally lampled.
EPA will evaluate the possible risks posed by air dispersal of soil
during remedial action. EPA will take sufficient precaution, to
ensure that air dispersal is kept to a minimum.
-33-
-------
[',
,I
.~
~ ~
1-4 0."'"
o ~ cu .
~ u cu
.~ cu cu t~
"0 ~~
CU~tI)U
1-4~~~
U......;> Q)
1-4"O~"t:S
t8Q)~tI)
~ Q)......
S.>~ S
Q) 0 .sQ) ~
~ ~ ......
..-4 ~
~ ~Q)
o Q) ..-4 ~
~,.c~~
~,....... Q) ~
1-4 ,....... 'C ..-4
::S.z::! 1-4 ~
~ ~ 0 0
Q)~
~~1-4 ..
.~ Q) ::s ~
S s e.p
~ Q) ~..-4
Q)" U bD "C
~~E~
~~~ 0
OQ)~U
~~a::"t:S
tI)~~Q)
Q) ~ ...... bD
o . Q) C'j.
"C"C"t:SS
~~C'jC'j
~tES"C
~z Q) tI) ~
~ ...... ..-4
Reproduced by NTIS
National Technical Information Service
U.S. Department of Commerce
Springfield, VA 22161
This report was printed specifically for your
order from our collection of more than 2 million
technical reports.
For economy and efficiency, NTIS does not maintain stock of its vast
collection of technical reports. Rather, most documents are printed for
each order. Your copy is the best possible reproduction available from
our master archive. If you have any questions concerning this document
or any order you placed with NTIS, please call our Customer Services
Department at (703)487-4660.
Always think of NTIS when you want: .
. Access to the technical, scientific, and engineering results generated
by the ongoing multibillion dollar R&D program of the U.S. Government.
. R&D results from Japan, West Germany, Great Britain, and some 20 .
other countries, most of it reported in English.
NTIS also operates two centers that can provide you with valuable
information:
. The Federal Computer Products Center - offers software and
data files produced by Federal agencies.
. The Center for the Utilization of Federal Technology - gives you
access to the best of Federal technologies and laboratory resources.
For more information about NTIS, send for our FREE NTIS Products
and Services Catalog which describes how you can access this U.S. and
foreign Government technology. Call (703)487-4650 or send this
sheet to NTIS, U.S. Department of Commerce, Springfield, VA 22161.
Ask for catalog, PR-827. .
Name
Address
Telephone
- Your Source to U.S. and Foreign Government
Research and Technology.
------- |