ML20134D820
| ML20134D820 | |
| Person / Time | |
|---|---|
| Site: | 07000139 |
| Issue date: | 09/30/1996 |
| From: | Brackett R ENVIRONMENTAL PROTECTION AGENCY |
| To: | Chabot D ENGELHARD CORP. |
| References | |
| NUDOCS 9610310020 | |
| Download: ML20134D820 (19) | |
Text
hso er4%
UNITED STATES ENVIRONMENTAL PROTECTION ACENCY 70gg g
(g
)
REGION I J.F. KENNEDY FEDERAL BUILDING. BOSTON. M ASSACHUSETTS 02203-2211 y
l September 30, 1996 Donald P. Chabot Project Coordinator Engelhard Corporation 30 Taunton Street Plainville, MA 02762 Re:
Engelhard Consent Order:
RCRA Docket # I-92-1051 Final Comments, RCRA Facility Investication Phase I Interim Report, Engelhard Corooration, Plainville, MA, dated August 31, 1995.
Dear Mr. Chabot:
EPA - New England has completed its review of Engelhard's RCRA Facility Investigation, Phase I Interim Report.
EPA's final comments are enclosed.
Engelhard shall submit to EPA a response to these comments within 45 days.
If you have any questions please call me at (617) 573-5744.
Sincerely,
~
Robert W.
Brackett RCRA Corrective Action Section Enclosure cc:
Jeff Chormann, MADEP-Boston Gregg Hunt, MADEP-Lakeville Richard.Turtil, NRC Jeffrey Barrett, Plainville Toxic Waste / Radiation Watch Grp Plainville Public Library Information Repository Bill Burt, Plainville Board of Health Thomas S.
Brown, Engelhard-NJ Rob Powell, Environ l
l MLIO 9610310020 960930 PDR ADOCK 07000139
, to PDR C
7 l
n.
f IN*
PRINTED ON RECYCLED PAPE8 Faq pimVOdn
t EPA Final Comments, RCRA Facility Investigation Phase I Interim Report for Engelhard Corporation, dated August 31, 1995.
I.
GENERAL COMMENT
S 1.
The Final RFI Report should include a current private well survey of downgradient residential homes, including the depth of the well point, if possible.
The locations of the wells should be included on a map which shows the Engelhard facility, the contaminant plume, and the residential homes.
2.
The Final RFI Report should contain an executive summary, I
which could be distributed to citizens on the site mailing list.
l l
3.
The Phase I Interim Report (Report) includes a conceptual site model shown on Figure 5.1.2.
The Final RFI Repcrt should l
include a conceptual site model presented as a 3-dimensional l
model showing sources, release mechanisms, contaminant fate and transport, exposure pathways, and potential receptors.
4.
Since the Report was submitted, Engelhard has installed l
additional wells on and of f-site which show that contaminants are present deep within portions of the bedrock.
Engelhard should revise Section 6 of the Report to include two sampling rounds of these newly installed wells.
Engelhard should ensure that all results, assumptions and interpretations presented within the Final RFI Report integrate the results of these additional studies.
5.
The Report does not include an evaluation of the groundwater flow rate, as required on Page D-5 of the RFI Proposal.
This requirement is essential to determining fate and transport of site contaminants, and should be addressed in the Final RFI Report.
l 6.
The Phase I Human Health and Environmental Risk Assessment l
references EPA Region III risk assessment guidance documents, but does not include a reference section.
The Final RFI Report must include a reference section, and ensure that all documents referenced throughout the report are included.
I j
In general, Engelhard should not use Region III documents in the risk assessment. Only one Region III document, the August 1995 Risk Update, is accepted in Region I as a screening tool.
l Any other Region III documents are not accepted in Region I and are not national policy.
Engelhard should use the Health and Environmental Risk Assessment SOW (Appendix III of the Consent Order), as guidance (this SOW has been updated).
7.
The Phase I Human Health Risk Assessment does not discuss whether the site is fenced, or if there are any current and ongoing security measures at the site that will restrict access.
The Final RFI Report should include a discussion of l
~
l the security measures at the facility and whether trespassers I
may gain access and be exposed to the contaminants at the site.
j 8.
It is unclear how assessment and measurement endpoints were selected and related to available site information.
Although the discussion on assessment and measurement endpoints is l
referenced to the Framework for Ecolocical Risk Assessment (EPA /630/R-92/001),-
no discussion relating the proposed i
assessment and measurement endpoints to conditions at the site is provided.
9.
Engelhard should revise Section 6 of the Report to include l
analysis for PCBs in surface water samples.
l 10.
Relatively high concentrations of VOCs and metals were i
detected in ground water samples near the Caccipaglia l
residence.
In its response to these comments, Engelhard should verify that there are no private wells on. the l
Caccipaglia property and evaluate whether the ground water l
contaminant concentrations pose a current health threat.
II.
SPECIFIC COMMENTS l
t 11.
Section 2.1.2.
The Report states that several other observation and extraction wells have been constructed as part of the stabilization program.
All data, including locations, i
well construction diagrams / boring
- logs, water level measurements, etc, for these wells, should be included and discussed in the Final RFI Report.
12.
Section 2.1.2.1.
The Report indicates that ground water samples' collected from several wells were excessively turbid.
However, the Report does not indicate the cause for this l
condition or whether a.
remedy for the condition is i
recommended. Engelhard should discuss how this situation will l
be rectified during the Phase II investigation.
13.
Section 2.1.5.
The Report briefly discusses the anomalous j
water levels in MW27, MW6, and MW28 and the fact that there is l
an upward hydraulic gradient in the MWO4 and MWO6 clusters.
Are these unexpected results related to each other? The area around MW28 is paved.
It seems unlikely that surface runoff is causing a localized ground water mounding around the casing, or that surf ace water is leaking into the well.
This needs to be addressed in the Final RFI Report.
l 14.
Section 2.1. 6.
During the March and May 1995 sampling events, EPA's oversight contractor observed that Engelhard's I
consultants had difficulty with high turbidity in some wells and keeping a
consistent rate of flow in pumps when l
purging / sampling wells.
In some cases, the rate of 500 ml/ min t
i 2
4 was exceeded for overburden wells.
Engelhard must address these problems of high turbidity and variable flow rates.
15.
Section 2.1.9.1.
The Report provides total VOC ground water contaminant distribution maps, but does not include maps for individual constituents.
However, it is clear from the text discussion that contaminant distribution is quite different for various constituents.
An understanding of individual contaminant distribution in groundwater is important to distinguishing potential groundwater contaminant sources. The Final RFI Report should include contaminant contour maps for the five major VOCs detected, and any other contaminants, such as metals which have been detected above MCLs, as appropriate.
16.
Section 2.1.9.1.
The paragraph at the top of page 2-12 indicates that the presence of PCE in background groundwater quality samples is considered " anomalous."
The Report then describes three reasons why the PCE contamination should be considered anomalous.
However, the Report does not consider any site conditions that could result in the occurrence of PCE in this location, including differential flow of dense non-aqueous phase liquids (DNAPL) within groundwater relative to the potentiometric surface, or an unknown contaminant source.
Does Engelhard believe that this is a sampling problem of cross contamination?
Were background wells sampled first?
Engelhard should further justify the assertion that PCE occurrence is anomalous in these groundwater samples, taking into account a fuller spectrum of contaminant distribution possibilities.
Include a discussion of blank results (e.g.,
trip, equipment) and sample duplicate data (if available), the potential for cross-contamination, and laboratory induced contamination.
The following information is contained in the EPA aerial photographic Site Analysis Report for the Engelhard facility prepared in April 1990.
A photo of this area taken on October 10, 1980 shows excavations, mounded material, and a horizontal and vertical tank in the area.
A photo taken on April 7, 1981 shows additional excavations and mounded material.
A March 22, 1986 photo shows 3 drums and two possible horizontal tanks within the fenced area.
17.
Section 2.1.10.
The Final RFI Report should include the location of the former dry well (where PCBs were found) on a map.
18.
Section 2.1.10.
The Report indicates that only five wells will be monitored for PCBs in Phase II groundwater sampling events.
However, if any of these wells are paired with an overburden or bedrock well that was not recommended for
- sampling, these wells should be included regardless of 3
l' l
l l
analytical results to ensure that the full vertical extent of l
contamination in what appears to be a limited plume area, is monitored and understood.
Revise Section 6 of the Report to indicate that groundwater collected from MW-06 will be i
analyzed for PCBs in Phase II.
19.
Section 2.2.2.
This section identifies the background soil l
samples at the facility, and states that " boring SB-43 through l
SB-45 collected during Phase I, will be combined with previous l
sampling data collected at SB-3 and SB-4 to establish a l
database for the natural concentrations of metal in soil for l
the Facility."
Samples collected at SB-3 and SB-4 were not validated and cannot be used in the risk assessment.
In addition, as noted in the above comment regarding air photos, the area around SB-4 and SB-44 contained suspected waste materials in the past.
Engelhard should revise Section 6 of the Report to propose additional background locations for l
soils.
20.
Section 2.2.3.
It is unclear whether Figure 2.2.1 presents the proposed or actual soil boring locations.
For example, the text of the Report states that SB-32 was moved from Building 9 to Building 1, but Figure 2.2.1 shows boring SB-32 to be within Building 9.
The Final RFI Report should show the actual soil boring locations.
21.
Figure 2.2.9.
Analyses of soil sample SB-RD33 detected 242 ppm Hg at a depth of 10 feet.
This is not shown on Figure 2.2.9.
22.
Section 2.2.4.
The Report indicates that the monitoring well screen slot size was appropriate for overburden wells.
- However, Section 2.1 indicates that turbidity within unspecified wells was exceptionally high, which would imply that well construction is inappropriate.
Engelhard should l
discuss the potential origin of turbidity problems, and indicate whether bedrock wells have encountered this problem.
l If so, Engelhard should propose modifications to future well l
construction design to address this concern.
1 23.
Section 2.2.5.
Table 2.2.1 presents the soil sainpling i
l program, including chemical analyses performed.
The table l
indicates that the background samples were analyzed for metals only.
It is assumed that other hazardous constituents were I
omitted from analyses because Engelhard considered any l
occurrence of these constituents (e.g.,
all organic constituents, including chlorinated organics and BTEX) to be anomalous,
- and, therefore, no
" background" values were necessary.
The Final RFI Report should clarify this point.
i 24.
Section 2. 2. 7.
Engelhard compares concentrations detected in soils to the soil performance standards (for Stabilization) to 4
l justify not performing additional borings to delineate the i
extent of contamination.
The Stabilization performance standards assume an industrial scenario, and do not consider risks to ecological receptors.
If Media Protection Standards are ultimately set at levels below the Stabilization performance standards, the collection of additional data by Engelhard may be warranted.
l 25.
Section 2.2.7.
Engelhard should clarify the purpose of
{
l performing SPLP analysis in boring SB-27.
This sample was taken from below a concrete slab where precipitation is not likely to be present.
26.
Section 2.2.7.
Pg. 2-21.
The maximum detected value is not a conservative estimate of background.
27.
Section 2.2.7.
The Report indicates on page 2-20 that ground l
water quality data were used to evaluate the potential impact 4
of soil contamination where no proposed soil performance standards were available.
- However, this approach is not i
universally appropriate, as time of contamination, depth and soil conditions could all-impact contaminant distribution in j
the vadose zone.
That is, high soil contaminant concentrations without concurrent downgradient groundwater contamination does not conclusively indicate that groundwater contamination will not occur in the future.
28.
Section 2.2.7.
Sub AOC A-12.
The boring is located
(
downgradient of the former UST.
The approved RFI Proposal
)
stated the samples were to be collected from just below ground surface, between ground surface and the water table, and from just above the water table.
- However, samples were only collected at the 2 and 5 foot elevations.
Engelhard should explain this apparent departure from the approved proposal.
l 29.
Section 2.2.7.
Soil Borings SB-24 (AOC-5), SB-35 (AOC-29) and i
-SB-36(B) (AOC-30) were to be drilled to bedrock as proposed in j
Section D.4.2 of the Phase I RFI Proposal.
However, the soil l
boring log, referred to in this section and presented in Appendix B, does not indicate that bedrock was encountered in any of these borings.
Engelhard should discuss why the borings were not advanced to bedrock and why no sample was collected, as required in the Proposal.
1 30.
Section 2.2.7.
AOC 30.
Engelhard should provide greater detail on how this AOC will this be further evaluated in Phase II.
31.
Section 2.2.7.
Former Dry Well.
This area should be considered part of AOC 16.
It is subject to RCRA.
Two soil j
borings, SB-RD33 and SB-RD33A, were drilled at the Former Dry j
Well.
However, in Appendix B, a soil boring log is presented 5
for only SB-33.
Verify that this log is for SB-RD33, and include the soil boring log for SB-RD33A in the Final RFI Report.
32.
Section 2.3.1.
Figure 2.3.1.
implies that SW-11 is a background surface water sample.
Engelhard should make clear that this is not a background sample or background location.
33.
Section 2.3.5.2.
This section states that the occurrence of elevated metal concentrations within surface water samples j
collected in the first round from the south embayment may be due to disturbance of sediment from the sediment sampling j
program conducted in this same area. As a matter of protocol, i
surface water samples should always be collected prior to any i
sediment samples in order to prevent this from occurring.
I Indicate whether surface water or sediment samples were collected first at the south embayment during the first round
)
of sampling.
If the sediment samples were collected first, 2
thus potentially compromising surface water sample quality, revise Section 6 to ensure that the additional sampling to verify the metals occurrence will be conducted prior to sediment sampling in the same area.
34.
Section 2.3.5.2.
Concentrations of metals in background samples of surface water are relatively high.
The Report states "The background samples are in remote areas from the Facility, and may reflect the effect of nearby runoff from roadway surf aces. "
SWO9 is also located near an autobody shop.
Measured chloride concentrations and specific conductance values are generally higher in these areas than in i
the other samples, which indicates likely impacts from these
)
off-site sources.
Background samples should not only be collected from areas not affected by the facility, but should also be taken from areas not near or impacted by a suspected source of contamination.
Revise Section 6 of the Report to propose alternate background surface water sample locations.
35.
Section 2.4.6.
Background
sediment samples contained relatively high levels of some metals.
As noted above in comment #47 regarding background surface water locations, i
l these elevated background sediment sample concentrations are likely due to off-site sources, and are thus not suitable as background locations.
Engelhard should revise Section 6 of the Report to include collection and analysis of alternate background sediment samples.
36.
Section 2.4.6.1.2.
The Report indicates that only three sediment samples were analyzed for PCBs, and states " Sediment sample location SD-21 was located approximately 10 feet offshore of the former outlet of the drainage ditch..."
This sample had elevated levels of PCBs.
Revise Section 6 of the Report to collect additional sediment samples in the south 1
6 l-I
._____._____-_.__._._.___-._-.._...m.
embayment to better define the extent of PCB contamination.
37.
Section 2.4.7.
The Report commits to the collection of six sediment samples for metals analyses during Phase II, stating that "there is some uncertainty as to the reliability of the concentrations (of metals in all sediment samples) as originally reported by the laboratory. " Justification for the collection of only six samples, however, is not provided, and the strategy for sample site collection is not included or referenced.
Revise Section 6 of the Report to include this information and indicate the specific criteria that will be used to determine if additional sampling is necessary, should the re-sampling indicate that the Phase I data are unreliable, 38.
Section 2.4.7.1.
Engelhard should provide an explanation of why additional delineation of lead is not warranted in sediments in Turnpike Lake west of AOC A.
According to Figure l
2.4.2, lead was detected at a concentration of 130 ppm at l
SD08.
i l
39.
Section 2.4.7.1.
The Report indicates that data collected
" provide sufficient delineation of the extent of impact for i
the purpose of this investigation".
Engelhard proposes to collect one additional sample to further delineate the extent of contamination (assuming that the data are valid). However, the rationale behind selecting the single sample location is not provided, particularly since contaminant concentration in sediment between other sample locations (e.g.,
SD-15 and SD-
- 13) also requires further delineation.
Provide the rational for collecting only one additional sample.
40.
Section 2.4.7.2.
The third sentence of this section states that "the highest levels of (some) metals (e.g.,
cadmium) in the stream system appears to be associated with an impounded area east of the Drive-in property... ". However, the location of this impounded area is not included.
The Final RFI Report j
should include more information on the impounded area,
~
including its specific location, and describe this feature in greater detail (e.g., how it acts as a sediment trap / filter).
In addition, relatively high concentrations of metals were detected in sediments at several locations in the unnamed streams downstream from Turnpike Lake.
Revise Section 6 of the Report to include collection of additional sediment samples downstream of SD-01 to determine the extent of sediment contamination.
41.
Section 2. 4. 7. 2.
The Report states that SD-03 is a background sediment location.
However, it is unclear whether or not this stream is impacted by the facility.
The figures do not show i
i the origins of the stream.
Engelhard should evaluate this further in the Final RFI Report.
I 1
7
l l
42.
Section 2.5.4.1.
For Prey Fish, Engelhard should not compare l
the PCB concentrations to the FDA tolerance level.
43.
Section 2.5.1.
The first full paragraph on page 2-49 states l
that the poor fishing success in Turnpike Lake and the Old 1
Mill Pond was due to lower than normal water temperatures.
Samples were collected May 16 and 17, 1995.
- However, Engelhard proposes to collect additional samples (page 2-54) during May 1996, stating "May 1996 is believed to be the next opportunity to collect fish tissue samples at a time when water temperatures are warm enough".
Since May proved to be an inappropriate time to sample during 1995, it is not possible to predict whether more favorable conditions will exist in May 1996.
Revise Section 6 of the Report to ensure that the fish sampling program will be conducted at an optimal l
- period, and provide specific temperature parameters and conditions that define this period.
44.
Section 2.5.5.
The Report states that the availability of fish is limited in the old Mill Pond and the private impoundment.
EPA believes that the poor fishing results are more likely due to inappropriate fishing methods, and possibly also the time of year the fishing was attempted.
Unfortunately, both in Turnpike Lake and in Mill Pond and the l
private impoundment to a lesser extent, the backpack shocker l
was not effective.
Since the fish tissue sampling will carry
)
significant weight in the ecological risk assessment (given l
the routes of
- exposure, the contaminant types and concentrations in south erebayment sediment), all targeted fish j
must be caught and analyzed.
The bottom fish should be targeted using gillnets that are set overnight.
Trout lines can be used in the lake as well.
These methods are recommended by the Massachusetts DEP Fish Toxics Monitoring l
Program.
A backpack shocker is only suitable for the private impoundment.
Revise Section 6 of the Report to include fish i
sampling in Turnpike Lake, Old Mill Pond, and the private j
impoundment, as necessary to achieve the program outlined in the approved RFI Proposal and Technical Memorandum No.
4.
45.
Section 2.5.5.
The Report indicates that since PCB values in fillet samples did not exceed the FDA tolerance level of 2 I
ppm, PCB sampling in other locations of Turnpike Lake is not proposed for Phase II.
Since the first sampling event was inadequate, any stations or trophic levels of fish should not be omitted in the Phase II work.
The FDA tolerance level has no bearing on the ecological risk assessment.
For health l
risks, the 10-6 risk level is 4x10-4 ppm for cancer effects and concentrations above 0.03 ppm may produce immunological or reproductive effects.
Revise Section 6 of the Report to include PCB analysis for all fish samples.
46.
Section 5.1.1.2.
The first full paragraph on page 5-6 states 8
=
i that "The maximum and minimum values presented in Tables 5.1.1 through 5.1.10 may change with the collection of additional data for Phase II of the RFI.
These values are provided only I
as an indication of the concentration range reported for the CPCs [ Constituents of Potential Concern]. The methods used to evaluate the data for the selection of exposure point concentrations are not included in this document and will be included in the Second Interim Deliverable."
This statement implies that while the Phase I values may not be used as exposure point concentrations, they are being used for the selection of CPCs.
Engelhard should make clear that if Phase II sampling indicates that concentrations of chemicals not formerly chosen as CPCs (based on Phase I sampling data) now meet the CPC criteria, these chemicals will also be included as CPCs in the risk assessment.
i 47.
Section 5.1.1.2.
The Report states that surface soil data l
from three earlier investigations will be used in the risk assess. ment.
Only data which has been validated can be used in the risk assessment.
l 48.
Section 5.1.1.2.
The Report states "The two large buildings are assumed to remain in their current locations."
This i
should not be assumed, as Engelhard is considering demolishing the buildings. The baseline risk assessment should assume the buildings and pavement are not present.
Soil samples from under the buildings should be considered surficial soil for future exposures.
49.
Section 5.1.1. 2.
The second full paragraph on page 5-6 states that, "The use of the Facility is assumed to remain industrial in the future."
Appendix III, Section 2 of the Engelhard Consent Order states that
" Region I policy dictates the evaluation of a future residential land use unless other scenarios are approved by the Region I office."
l 50.
Section 5.1.1. 2.
The third full paragraph on page 5-6 states that " Surface soil data includes soil samples collected only from 0 - 2 feet below ground surface (bgs).
Subsurface soils are represented by data collected from 2 feet or more bgs."
Region I considers surface soils to be O to 1 foot, and subsurface soils to be 1 to 10 feet or to the depth of a typical foundation.
Compositing should only be done across a single soil bore at the chosen depth, i.e. 0-1 foot composited or 1-10 feet composited.
Soils from different locations should not be composited.
51.
Section 5.1.1.2.
The third full paragraph on page 5-6 states that "For the current soil exposures, borings collected under pavement or building slabs were eliminated from inclusion in j
the data sets for surface and subsurface soil.
This is based on the assumption that direct contact with those soils is an 3
9 I
incomplete exposure pathway."
For current risks, it is acceptable to consider the slabs will be in place.
For future risks, Engelhard should evaluate exposures without pavement or building slabs.
52.
Section 5.1.1.2.
Ground water.
The risk assessment should.
consider the state ground water classification.
The portion of Area I located east of Tavnton Street is not part of the facility. For purposes of the risk assessment, Engelhard must consider that a future resident on the east side of Route 152 can install a drinking well in the plume.
For the final remedy, EPA considers that a resident can install a well anywhere there is not a land use restriction.
53.
Section 5.1.1.3.
The first paragraph of this section states that " Metals are generally found at low concentrations in many environmental media, including soil, sediment, surface water and groundwater.
Only those inorganic chemicals detected at levels above naturally occurring background concentrations will be included as CPCs in the human health risk assessment."
Comparisons of detected concentrations with background samples should not be used to eliminate COCs.
Engelhard should include any chemical that represents a health risk in the risk i
assessment. Comparisons with background concentrations appear in the risk characterization or uncertainty sections.
The onsite measurement which should be used is the 95% upper confidence limit of the mean.
To demonstrate that a chemical is a background contaminant, the chemical should be compared using a statistical test with the appropriate on-site measure.
A second option it to screen COCs using Region III residential concentrations.
54.
Section 5.1.1.3.
Pg. 5-10.
The Report states that chemicals in ground water must be detected at least once above the MCL and above background concentrations in order for it to be included in the risk assessment.
As stated above, Engelhard must use a statistical test to drop a chemical from the risk assessment or screen COCs using the Region III numbers.
Similarly, for fish, Engelhard cannot eliminate COCs based on comparisons with background.
55.
Section 5.1.2.1.
Engelhard should consider the onsite worker and construction worker scenarios.
This should be a standard construction worker scenario which assumes construction of new facilities on the property.
56.
Section 5.1.2 1.2.
Engelhard must assume a private potable well in Area I for off site.
The Final remedy could include a deed restriction to prevent wells in this area, but this scenario must be carried through the risk assessment.
57.
Section 5.1.2.2.2.
Engelhard must evaluate inhalation of 10
l l
vapors seeping into off-site (Areas 1
&2) homes from l
contaminated ground water. The former residential area across the street from Engelhard which is now owned by Engelhard should be considered residential in the future scenario.
58.
Section 5.1.2.2.4. and 5.1.2.2.5.
Engelhard should include a qualitative discussion of recreational users exposed in the south embayment of Turnpike Lake within the property boundary.
There is no fence keeping people out of this area.
59.
Section 5.1.2.5.1.
Engelhard must use the on-site construction worker who is exposed to surface and subsurface soils, ground water, and vapors or particulates.
l 60.
Section 5.1.2.5.2.
The residential scenario should include consumption of ground water.
61.
Section 5.1.2.4.2.
The first paragraph of the Int ru si.ve Worker subsection states that
" ENVIRON will assume that exposures to site constituents may occur for a total of ten days per year for a period of one year.
This is based on the assumption that exposure occurs five days per week for two weeks during a short-term intrusive project."
Engelhard is considering demolishing some or all of the buildings.
It is possible that new buildings may eventually be constructed on site.
Engelhard should use standard "on site worker" and
" construction worker" scenarios.
The utility worker scenario is only used when either of the others is not appropriate.
i l
\\
l 62.
Section 5.1.2.4.2.
The first paragraph of the Residential Adult and Child subsection states that " Consistent with USEPA l
Region I recommendations, an exposure frequency of 150 days per year is assumed for direct contact pathways for both the child and adult residents due to the northeastern climate (i.e.,
snow cover). "
It is unlikely that the ground is covered with snow for seven months of the year in this area.
Furthermore, EPA risk assessment guidance (EPA, 1989a, 1989b) commands an exposure frequency of 350 days per year for the residespial scenario.
Provide a discussion of the regional meteol, hJ cal data that supports the use of 150 days per i
i year.
63.
Section 5.1. 2. 4. 2.
Engelhard considers the recreational child and adult will have an exposure frequency of ten days per year assuming that these individuals visit the lake two days per month for five months each year.
Considering that so many people live in the vicinity, this may be a low estimate.
Engelhard should make inquiries about the recreational use.
It is reasonable to assume once a week canoeing, wading, or fishing.
Alternatively, Engelhard could conduct a survey.
64.
Section 5.1.2.4.3.
Soil and Sediment Incestion.
Engelhard 11
should use the high end and central tendency values for soil consumption for the on-site worker and residents. (August 1994 Risk Update).
Use 480 mg/ day for the construction worker scenario (not utility worker).
The high end value for the on-l site workers is 100 mg/ day, the central tendency value is 50 j
mg/ day.
65.
Section 5.1.2.4.3.
Incestion of Fish.
Engelhard needs site i
specific data (for example, a survey) to alter the ingestion of fresh-water finfish in the risk assessment.
EPA will accept the 25% fraction as a central tendency, and 100% for the high end.
66.
Section 5.1.2.4.3.
Dermal Contact with soil and Sediment.
The dermal exposure of the construction worker should include head, hands, forearms and lower legs.
(See page 8-10 of the dermal guidance).
For the construction worker, use 5,800 cm2 for high end and 5,000 cm2 for central tendency for exposed l
skin surface.
Clothing is not a barrier to soil exposure.
For adult skin surface area assuming residential exposure, use l
the high end and central tendency values in the dermal guidance, pg. 8-20, table 8-6.
For the skin surface area of the child, Engelhard should have a high end and central i
tendency estimate.
The 2,000 cm2 value is ok as a high end i
estimate.
67.
Section 5.1.2.4.3.
Dermal Contact with Soil and Sediment.
EPA could not find the " default assumptions" about dermal l
absorption factors.
EPA Region I
only quantitatively evaluates dermal absorption from soil for a few chemicals:
PCBs, dioxin, and cadmium.
For PCB's, the absorption factor is 6%, for cadmium it is 1%.
An EPA work group is developing
[
additional absorption factors.
Until this information is l
available, the risk assessment should evaluate dermal exposures to other contaminants in soil qualitatively.
l 68.
Section 5.1.2.4.3.
Dermal Contact with Water.
EPA Region I restricts the quantitative assessment of dermal absorption to the chemicals listed in Table 5-3, p.
5-9 and 5-10 of the dermal guidance.
The on-site construction worker has dermal contact with contaminated ground water. The of f-site resident has dermal contact with ground water.
(If this contact is limited or
- typical, it does not need to be evaluated quantitatively in the risk assessment.
Region I Supplemental Guidance to RAGS, June 1989.)
For the skin surface area for the recreationally exposed adults and children, please refer to tables 8-3 and 8-4 in the dermal guidance.
Regarding permeability coefficient values for compounds in aqueous media, use the values in Table 5-3 of the dermal guidance for quantitative risk assessment.
If there are additional COCs at the site, Engelhard should discuss their dermal absorbance qualitatively.
12 i
l
69.
Section 5.1. 2. 4. 3.
Dermal Contact with Water. Top of P. 5-32.
Engelhard should use the central tendency and high end residential inhalation values for adults provided in the August 1994 Risk Update.
Engelhard can use indoor measurements (preferably) or modeling to estimate' the inhalation risk from vapors from ground water contamination.
Inhalation risks from all other household uses of water (showering, dishwashing, etc) are assumed to be equivalent to the risks from ingestion of the contaminated ground water (August 1995 Risk Update).
70.
Section 5.2.1.
Provide a figure of the " Site" as described in this section.
This could be accomplished by labeling the
" Site" boundaries and identifying the features discussed in this section on an existing figure (e.g.,
Figure 2.1.1).
l Features to identify include the adjacent lands owned by private landowners, the Engelhard eastern property boundary, the former drive-in theater property and the Natural Resources Trust.
l
- 71., Section 5. 2. 2. 2.
The last paragraph of this section indicates that sediment guideline values were calculated for organic chemicals for which no values were available in the literature.
The ecological risk assessment should summarize the calculation method and/or provide an example calculation.
l 72.
Section 5.2.2.2.1.
Further justification for screening cut beryllium as a constituent of potential environmental conce rn (CPEC) is warranted.
Indicate why the maximum concentration was not used to evaluate the compound's viability as a CPEC, l
rather than the median value (s).
Discuss, where appropriate, l
the potential toxic effects that beryllium could have on l
ecological receptors.
In addition, provide the specific l
concentration of beryllium as part of this discussion, rather than simply stating that the element is present in
... relatively low concentrations... ". Revise the forthcoming Phase II RFI Report to include this information.
73.
Section 5.2.2.2.2.
Further justification for screening out beryllium as a CPEC is warranted.
Indicate why the maximum concentration was not used to evaluate the compound's viability as a
- CPEC, rather than the median value(s)
Discuss, where appropriate, the potential toxic effects that beryllium could have on ecological receptors.
In addition, provide the specific concentration of beryllium as part of this discussion, rather than simply stating that the element is present in "...relatively low concentrations...".
Revise the forthcoming Phase II RFI Report to include this information.
74.
Section 5.2.2.2.3.
Provide further justification for screening out chloroethene.
Discuss the potential effects 13
- --.- -~ -.-- -.- - -
i l
that exposure to chloroethene could have on ecological receptors.
As part of the justification, define the terms
"...relatively low concentrations..."
l 75.
Section 5.2.2.3.
Engelhard should discuss why volatile organics were analyzed at only 13 of the 23 locations considered.
l 76.
Section 5.2.2.3.
Include a summary of the development of the soil benchmarks to be used to screen chemicals detected in j
surface soils as part of the Phase II ecological risk l
assessment.
Describe the derivation of each benchmark.and I
illustrate the method with a sample calculation.
L 77.
Section 5.2.2.3.
Engelhard should provide further justification for screening out methylene chloride.
Discuss the potential effects that methylene chloride could have on ecological receptcis.
Illustrate how screening out methylene chloride represents a conservative approach when no background sample.results are available.
l l
78.
Section 5.2.3.
The May 17, 1995 survey of the Engelhard site l
provided an extensive list of " species observed" for the site.
l Engelhard should include the list below (Table 1) of l
additional species that have been observed on the adjacent property and in Turnpike Lake. These observations are part of l
a ten year observation list compiled by the Natural Resources l
Trust (Marsha Benes, 1995).
These species would be expected on the Engelhard property as well.
This data will be helpful.
in supplementing the May 17, 1995 wildlife observations.
TABLE 1.
l Soecies to be added to Table 5.2.1 of Potential Receptors l
Avian species:
Scolocax minor Woodcock Bonasa umbellus Ruffed grouse Otus asio Eastern screech owl i
Aix sponsa Wood duck l
Chordeiles minor Common night hawk Circus cyaneus Marsh hawk Accioiter striatus Sharp shinned hawk Falco soarverius Sparrow hawk /American kestrel Botaurus lenticinosus American bittern Ardea herodias Great blue Heron Branta canadensis Canada goose Hylocichla mustelina Wood thrush Mniotilta varia Black & white warblers Certhia americana Brown creeper i
Mimus colvolottus Mockingbird i
j 14
_... ~ -. -.
j Bombycilla cedrorum Cedar waxwing Dumetalla carolinensis Catbird Carduelis tristis American goldfinch Toxostoma rufym Brown thrasher Cardinalis cardinalis Cardinal Picoides villosus Hairy woodpecker Schrvracicus varius Yellow bellied sapsucker Colaotes auratus Northern flicker i
Pheucticus ludovicianus Rose-breasted grosbeak Reculus satraca Golden kinglet Mammalian species:
)
Mechitis meohitis skunk Lutra canadensis otter Marmota monax woodchuck Eotesicus fuscus Large brown bat Didelchis viroiniana opposum Mustela soo.
weasal Microtus soo.
vole species Fish species:
Ictalurus nebulosus Horned pout / Brown bullhead Perca flavescens Yellow perch Esox americanus Pickerel i
Lecomis auritus Esox americanus l
Lecomis macrochirus bluegill Microoterus salmoides large mouthed bass Notemiconus crvsoleucas golden shiner Lecomis cibbosus pumpkinseed 1
l Pomoxis nioroma black crappie Catostomus commersoni white sucker American eel Reptile and Amphibian species:
Notochtalmus vividenscens Eastern newt Chrysemys picta Painted turtle Chelvdra seroentina snapping turtle l
Natrix sioedon Water snake Bufo americanus American toad i
Hyla crucifer Spring peeper Rana catestiana Bullfrog Rana cioiens Leopard frog Rana clamtans Green frog Northern two-lined salamander Musk turtle Carter snake Common milksnake-l List of species not included in May 17, 1995 site visit i.
observations by Environ. Source:
Marsha Benes, Natural Resource i
15
...__.m Trust, Plainville, MA.
Personal communication.
l l
79.
Section'5.2.3. Engelhard should clarify what method they will use to evaluate exposure to terrestrial and wetland plants in the second interim deliverable.
l 80.
Section 5.2.3.
Otter or fox may be the most appropriate j
mammalian species to assess since these have been observed at the site.
81.
Section 5.2.4.
For the third assessment endpoint, unimpaired l
l aquatic food chains for species inhabiting, or utilizing the south embayment of Turnpike Lake and downstream impoundment is questionable.
EPA believes that there is one fish community in Turnpike Lake and so a statement inferring that the fish that inhabit or utilize the south embayment can somehow be i
separated from the other fish is not appropriate to make.
The measurement endpoint that accompanies this third assessment endpoint is more concrete.
It is acceptable to compare fish L
tissue _ concentrations from Turnpike Lake to a " background" i
waterbody.
The statement of the third assessment endpoint should be revised.
82.
Section 6.0.
Section 2.0, Phase I Facility Investigation, commits or alludes to the performance of numerous tasks as part of Phase II that are not reflected in Section 6.
Engelhard should revise Section 6 to include the following:
e Re-evaluate well construction of MW-28 due to anomalously high water levels
- and, if appropriate, install a
l replacement well; e
Soil sampling for PCBs in the area of AOC-7; l
e Soil sampling below the slab underlying AOC -30 relative to VOCs; and, e
Install staff gauges at SW-03, SW-04 and SW-05 prior to initiation of surface water sampling in Phase II, and L
coordination of surface water level measurement with l
groundwater level measurement.
l
[
TABLES l
83.
Table 2.1.13.
The water detection limits for antimony (60 l
ppb) are significantly above the MCL of 10 ppb.
During the Appendix IX sampling
- event, antimony was detected at i
concentrations of 28 ppb in MWO1 (a background well) and 3 6. 9 ppb in MW26.
For the Phase II sampling events, Engelhard should use an analytical method which can achieve lower i
detection limits.
l 16
84.
Table 2.4.11.
Several metals (antimony, mercury, and silver) have detection limits higher than the NOAA effects concentrations.
Engelhard should use analytical methods with detection limits which are below the NOAA effects levels.
85.
Table 2.2.6.
Why is the proposed soil standard higher for hexavalent chromium than for total chromium?
86.
Table 5.1.1 and following tables. The current and future site i
data used in the health risk assessment should be the same.
EPA Region I does not believe that presently available models give adequate predictions of future concentrations of site contaminants.
Therefore, future risk calculations are based on currently measured concentrations of chemicals in various media.
Somewhere in the risk assessment, Engelhard should list all of the chemicals Engelhard is dropping as COCs and why.
There should be one set of COCs for each media for ecological and human populations separately.
I Concentrations in ground water are based on wells in the plume of contamination, not on all wells on the site.
4 When developing a list of COCs, PCBs are regarded as Total PCBs and not divided by Aroclor.
Therefore, if all of the PCBs together have more than a 10-6 risk, they all should be retained as COCs (see IRIS).
87.
Table 5.1.21 presents the exposure assumptions that will be used in the risk assessment.
The surface water ingestion rate for the recreational child is. presented as 0.05 liters / day.
-According to EPA risk assessment guidance (EPA, 1989a) the units for this value are liters / hour.
Make the appropriate changes to the table, or provide the appropriate conversion factor.
88.
Table 5.2.5.
Engelhard should justify why " background" sediments were not analyzed for organics.
APPENDIX A - PHASE I FIELD INVESTIGATION 89.
Section A.1.
The last paragraph of the section states that drill cuttings and cutting fluids were stored on-site until analytical results were available to characterize the wastes for disposal.
The Final Report should discuss the final disposition of these wastes.
90.
Section A.1.1.1.
The second paragraph of this section, states i
that for all overburden soil samples, OVM readings did not l
exceed background levels by more than 9.5 ppm.
However, for a
17
-n
,ne a
w-
,-,n
---y
}
MW-32A, OVM readings were as high as 30 ppm, as shown the monitoring well log in Appendix B.
91.
Section A.3.3.
The text at the bottom of page A-10 and the top of page A-11 indicates that sample bottles were filled in the listed order.
- However, A.T.
Kearney oversight personnel observed on January 18, 1995 that Engelhard contractor personnel were not following the listed order when collecting the sample at MW-15.
When questioned, the field personnel did not seem to be familiar with the sampling order requirements from the Sampling and Analysis Plan.
This indicates that the required sampling order may not have been followed up to that time.
Engelhard should address this concern and discuss the potential ramifications of not following the required order of sample containers.
92.
Section A.4.1.
Section 4.2 (pp. I-17 to I-18) of the Phase I RFI Proposal's Sampling and Analysis Plan indicates that soil samples from each spoon were to be collected, placed in plastic bags and screened with an OVM.
Samples were then to be selected and submitted for laboratory analyses based on the OVM readings.
- However, based on the first and third paragraphs of Section A.4.1, samples were only screened with an OVM upon opening of the split spoon prior to sample handling.
This approach would yield significantly lower levels than the headspace analysis approach described in the Sampling and Analysis Plan.
Engelhard should discuss why the approach was changed and discuss the potential implications of this change.
93.
Section A.4.2.
Section 4.2 (p.
I-16) of the Phase I RFI Proposal's Sampling and Analysis Plan indicates that soil samples collected with a split spoon sampler were to be contained in brass sample tubes.
However, Section A.4.1 of the Report states that samples were placed in laboratory-grade glass jars.
It further states that " soil was spooned from random points over the length of the soil material recovered in the split spoon."
This random sampling approach could potentially mean that individual layers of contamination or
" hot spots" may not have been sampled from within each split spoon.
Engelhard should discuss why the proposed approach was not implemented and the likelihood of hot spots being overlooked in the sampling method used.
l 18