1 
-
Mr-K-27-SS 
12=30 FROM=DPNR DEP STT
of
i. PROTECTION
TOO
ID = 80977454 IB
PAGE
1X7
GOVERNMENT OF THE VIRGIN ISLANDS OF THE UNTIED STATES
DEPARTMENT OF PLANNING AND NATURAL RESOURCES
NISKY SHOPPING CENTER
SUITE 45
CHARLOTTE AMAUE, ST. THOMAS, V.J. OO8O2
Express Mail
Return Receipt Requested
April 27, 1995
Caroline Kwan
USEPA Superfund Envision, Region II
New Yoric/Caribbean Branch
290 Broadway, 20th Floor
New York, New York 10007-1866
Dear Ms. Kwan:
Enclosed are the Virgin Islands Department of Planning and Natural Resources (DPNR)
comments on the March 1995 Draft Feasibility Study (FS) Report for the Tutu Wells
Superfund Site in St. Thomas, USVL These comments should be addressed by Geraghty
& Miller prior to the finalization of the FS and the preparation of the Record of
Decision (ROD).
If you have any questions regarding this matter please contact Mr. Adrian Schottroff or
Mr. David Rosoff of my staff at (SOP) 774-3320.
Sincerely yours,
Post-It* brand fax transmittal memo 7671 # of pages »
Leonard'
Assistant Director
Enclosure
Co.
37-
. . -J_._
? 44-3320
Fax*
3V4-ST4/6
AJ
-0
C-i
-o
^
ted:
*64926*
64926
DPNR COMMENTS ON THE DRAFT FEASIBILITY STUDY FOR THE TUTU
WELLFIEID SUPERFUND SITE IN ST. THOMAS, TJSVI
Although the general presentation of remedial alternatives in the
Draft Feasibility Study (FS) for the Tutu Well site is in
accordance with the remedies discussed during the meeting held in
St. Thomas in February 1995 between the VI Department of Planning
and Natural Resources (DPNR), the United States Environmental
Protection Agency (BPA) and the Potentially Responsible Party
(PRP) Group, significant alterations to the document are
required. These changes are discussed in the following comments.
General Comments: •
Site Wide Remedial Alternative Discussion:
General Presentation of Soil Remedial Alternatives (SRAs) and
Ground-water Remedial Alternatives (GRAs) into Site Wide Remedial
Alternatives (SWRAs) is confusing. The SRAs and GRAs should be
discussed separately and screened down to the final group of
acceptable alternatives for each media. EPA can select the
SRA(s) and GRA of choice from these distinct lists.
incorporation of Source Control Remedies Proposed bv Esso and
Texaco for BTEX plume control; 
• 
•• - 
•- • • • -• -..-•••
A major flaw in the FS is its failure to adequately discuss
source control remedies proposed at the Esso and Texaco service
stations and incorporate the proposed designs of these pump and
treat systems into the ground-water remedy discussions. On page
4-5 of the FS, it is stated that the implementation of Interim
Remedial Measures (IRMs) at the service stations has been
considered to ensure that the IRMs are compatible with the
proposed SWRAs. However, the document does not discuss the
parameters of either of these studies in conjunction with the
proposed site-wide ground-water remediation strategies and the
BTEX source remedies are not depicted on Figure 4-7.
Although detailed discussions of pump and treat specifications
are not required at the- FS stage, some mention of the parameters
of the service station remedies and their effect on the overall
ground-water remedy must be made. It is difficult to review the
proposed well arrays and pump rates proposed by each remedy
without knowledge of the source control well locations, well
depth, pump rates, etc.
Proposed,. Parameters of_ the Ground Water Remedies:
It is unclear how the number and location of recovery wells and
the proposed pumping rates for each remedial scenario were
determined in the FS. It.is standard procedure to estimate the
number of recovery wells and rate of pumping in the FS for
costing purposes, however, final selection of a well array and of
TUT O06 2127
3/7
DPNR COMMENTS ON DRAFT FS
April 27, 1995 
Page 2
pumping rates must be based on detailed modeling of site
hydrologic conditions. The FS should make this clear in the
discussions of the remedial alternatives.
Use of POST Systems:
Most of the remedial alternatives include the use of Point of
Entry Treatment Systems (POETS) on commercial and residential
private wells. Historically, DPNR has found that regulation of
pumping rates and general usage of private wells has been
problematic. DPNR envisions that certain wells that may be
eligible for POET systems within the scope of this FS will be
difficult to maintain control over in the long term. Because of
the difficulty in implementing an effective and controlled
Operation & Maintenance Program for POET systems on private well
in the Virgin Islands, DPNR will not support the selection of
this remedy. However, discussion and eventual screening out of
this alternative due to difficulty of implementability is
appropriate.
Use of POTW: 
' . . . - . -^ -;•-—'~ — ' • • - • • • • - • • •
Based on discussions with the VI Department of Public.Works in
the February meetings, there is no current option for discharge
of any
1 additional. volume to the municipal sewer system in Tutu.
At some time in the future, a new sewage treatment plant with
.expanded capacity will be constructed at Mangrove Lagoon which
will receive the sewage from the Tutu Valley area. However, the
FS cannot realistically utilize this speculative discharge as an
option. Discussion of this option and its eventual removal
through screening in the document is appropriate in case EPA
determines that this discharge alternative is appropriate at some
point in the future.
Soil Disposal: 
v
On Page 4-4 of the FS, the statement is made that if the material
that is excavated is non-hazardous, it will be disposed of at the
Bovoni Dump. No potentially hazardous or contaminated material
from the Tutu site can be disposed of at the Bovoni Dump without
first meeting DPNR's disposal standards. To be determined non-
hazardous, the material would have to under go the testing
protocol required by DPNR. This testing procedure is described
in the attached letter to Geraghty & Miller (attachment 2).
Furthermore, the FS assumes, for costing purposes, that all of
the excavated material will be nonhazardous and will be disposed
of at the Bovoni Dump. This assumption is probably not valid and
should be re-svaluated. A conservative costing estimate that
includes off-island transport and disposal of some of the
excavated material would be more accurate.
TU"i"
• . • 
DPNR COMMENTS ON DRAFT FS
April 27, 1995 Page 3
Treatabilitv Studies for Soil Remediation,;
The proposal of the innovative technologies of soil vapor
extraction (SVE) and bioventing for soil remediation at certain
properties is appropriate. However, it must be recognized that
the use o£ these technologies will require substantive pilot-
scale treatability studies during the Remedial Design. The
geological and soil conditions in the Tutu Valley may not allow
successful application of these technologies in-situ. As a
result, ex-situ venting and vapor extraction should also be
discussed.
Additional Pre-Desicrn Work;
Section 4.5, Pre-Design Activities, provides a vague description
of pre-design work that will be required prior to completion of
the Remedial Design Report. This section must be enhanced with
more detail and amended to include the following additional work:
Additional pre-design soil sampling to define the extent of
the contamination must occur at the Western^Auto property
(determine the source of cdhtami'hatlon through investigation of
the contaminated pipe and gravel layer) , former Laga Building
(both under and around the perimeter of the buildincr) , Tillett
Gardens (in and around area of PCS contamination) , and at
former Home petroleum station (now Kentucky Fried Chicken)^—-Re-
design soil sampling must also occur at each impacted property to
determine the extent of the areas of excavation and/or the
position of the SVE wells.
Additional ground-water sampling to define 'the horizontal •
extent of plume. The ?S claims that the plume is approximately
800 to 1000 feet wide on page 2-19. However, data from the
lateral portions of the plume is scarce and certain lateral wells
that-are potentially impacted by the plume have been left out of
the plume delineation (i.e. the Harthman Racetrack Well which had
0.15 ppb of PCE in the 1994 sampling). Additional investigation
of the Harthman wells and the installation of both eastern and
western perimeter monitoring wells is required to fully delineate
the plume boundaries.
The nature of the plume in the southern portion of the site is
poorly known. Additional ground-water elevation data for both.
the shallow and deep aquifer in the southern part of the site
must be collected. Ground-water elevation data for the KFC
wells, Harthraan wells/ EA wells, Rodriguez well, New Gassett
well, REMW wells, Mathias well, Smith well, Delegarde well, and
the Fort Mylner and Versailles supply wells should be collected.
In addition, monitoring wells should be installed south-east of
Fort Mylner Shopping Center and south-west of the Steele well.
TUT 006 2.129
APR-27-95 12=32 FROM=DPNR DEP STT 
ID.803774541B 
PAGE
DPNR COMMENTS ON DRAFT FS
April 27, 1995 Page 4
The information generated from these wells will aid in the
understanding of the dynamics of the southern portion of the
plume. 
.
Additional investigation of the suspected gasoline
contamination emanating from the vicinity of KFC (formerly Home
petroleum). Existing ground-water data show MTBE and chlorinated
hydrocarbon contamination reaching as far south-east as the
Delegarde well. It is conceivable that all or a portion of this
contamination is coming from the area around KFC. If this is not
the case, and the KFC plume is localized close to the site of
origin, the possibility exists that recovery wells placed
doxvngradient of the site will draw the plume to the south-east
and further contaminate the aquifer. In either case, this
potential source of contamination should be recognized,
investigated and remediated as part of the overall remediation of
the Tutu site.
Detailed ground-water modeling utilizing site specific aquifer
_ data generated from localized pumping tests conducted in
potential areas of recovery must be completed prior to the design
of the pump and treat system. This modeling is critical in
determining location of recovery wells, rates of pumpage, methods
of pulse pumping, zones o£ capture, aquifer drawdown dKith-feime.?••:
versus the-safe yield, etc. To date, modeling performed on the
Tutu Wells site has been limited in scope although a substantial
amount of data has been generated.
Page Specific Comments:
Page 2-22, Paragraph 4: It should be mentioned that the
conclusion that subsurface soil does not exceed "unacceptable"
(word left out of report) carcinogenic or non-carcinogenic risks
is pathway based and is true only under circumstances where the
soil is undisturbed.
Page 2-23: The section on risk does not provide any information
on environmental risks associated with the site contaminants. A
small mention of environmental risk is made on Page 3-3 where the
document states that no ecosystems are potential receptors of
COCs. Based on the CDM Risk Assessment, this does not appear to
be an accurate statement. Potential exposure pathways for
environmental receptors do exist, however, the risk assessment
^ 
determined that they will not result in an unacceptable risk.
Page 3-2, Paragraph 7: 
The second and third sentences in this
paragraph should be removed. Pumping, treating and discharging
may not be wasteful use, since this effort will, over time,
reduce the level of contamination of the acuifer. 
,
TUT OOi
APR-27-9S 12.33 FROM=DPNR DEP STT 
ID. 8097745416 
PAGE
DPNE COMMENTS OH DRAFT FS
April 27, 1995 Page 5
Page 3-4, Paragraph 2: 
The document repeatedly refers to the
current inability to reduce or eliminate DNAPL concentrations in
fractured bedrock aquifers. These statements may not be
incorrect, but repeated caveats and references to the
shortcomings of the best available technologies are not necessary
throughout the document, especially in the remedial goals section
of the report.
Page 3-10, Paragraph 3: The number of wells in the containment
program {stipulated to be between 2 and 4 in this paragraph)
cannot be determined without detailed Pre-design modeling of the
plume and aquifer characteristics.
Page 4-2, Paragraph 2: 
Additional investigation into the oily
gravel and pipe behind the former Western Auto store must be
performed during Pre-design field work,
Pa'ge 4-3, Paragraphs 1 and 2: Additional soil sampling (soil
borings) are required west and north-west of the building and
under the Curriculum Center building to determine the extent of
contamination. This should probably include some sampling within
the building, through the slab, in the vicinity of the pipes
containing the PCE and in the vicinity of the location of the
former PCS "still".
Page 4-7, Paragraph 3: 
It is unclear why the proposed 'r'ecovery
wells will be installed to a depth of 250 feet and"be fully
penetrating. Pumping water from 250 feet would draw high-TDS
water into the well and require a reverse osmosis process to be
utilised in the treatment train to provide potable-quality water.
Page 4-8, Paragraph 2: 
Some discussion must be provided
regarding the disposal of the GAC units which will be considered
a hazardous waste.
Page 4-13, Paragraph 4: Why only use a 1.5-day storage. If
potential potable distribution is a conceivable option, a larger
tank might be more economical plus the extra volume might be
required to hold water pending sampling results needed to verify
water quality before distribution.
Page 4-17, Paragraph 4: Western Auto - SVE near and under
buildings should be considered for a portion of the contamination
there.
Page 4-32, Paragraph 1: The consideration of SRA 5 as the least
effective option for soil remediation seems more based on cost
and less based on effectiveness. The prime criterion for
selecting the SRA should be effectiveness.
TUT OO6 2131
. APR-27-9S 1&-.33 FROM«DPNR DEP STT 
ID=S097745416 
PAGE 
7/7
t
DPNR COMMENTS ON DRAFT PS
April 27, 1995 Page 6
Page 4-33, Paragraph 1: The statement, "There are also no O&M
demands that would be associated with in-situ soil treatment
systems", is false. Follow-up sampling is an operational aspect
that is required to determine if the treatment is effective and
when the treatment is
OO6 2132