SDMS Document 
115574 
C D M 
F E D E R A L 
P R O G R A M S 
C O R P O R A T I O N 
• 
a s u b s i d i a r y 
o f 
C a m p 
D r e s s e r 
& M c K e e 
I n c , 
September 18, 1995 
Ms. Sherrel Henry 
Work Assignment Manager 
U.S. Environmental Protection Agency 
290 Broadway 
New York, New York 10007-1866 
PROJECT: 
ARCS II Contract No. 68-W9-0024 
Work Assignment 078-2PN7 
DOCUMENT NO.: 
7720-078-LR-CHMH 
SUBJECT: 
Dear Ms. Henry: 
Technical Review of PR? Data Summary Report 
Virgin Island Chemical Site 
St. Croix, U.S. Virgin Islands 
COM FEDERAL PROGRAMS CORPORATION (CDM Federal) has completed its technical review of the 
Draft Data Summary Report for the Virgin Island Chemical Site in St. Croix, U.S. Virgin Islands submitted 
on August 15, 1995. The report was prepared by Harding Lawson Associates (HLA) on behalf of the Island 
Chemical Company, the Potentially Responsible Party (PRP). 
The document was reviewed for its technical competency and to ensure that the objective ^ soecified in the 
Remedial Investigation Work Plan, prepared by HLA (HLA, 1994), were achieved duriii^ "^ A's field 
investigation. 
GENERAL COMMENTS 
1. 
At the time of project planning, HLA anticipated the direction of shallow groundwater flow 
to be southeast towards the Caribbean Sea and subsequently designed the monitoring well 
network in respect to that flow direction. However, after collecting two rounds of water 
level measurements, it has been determined that shallow groundwater flow is to east-
northeast. The report does not adequately address the difference between the assumed 
groundwater flow direction and the observed groundwater flow direction. The report also 
does not address the interaction, if any, between the shallow and deeper aquifer systems. 
3 0 5 2 6 2 
111 Fulton Street, Suite 710 New York, NY 10038 212 393-9634 
Printed on Recycled Paper 
CDM FEDERAL PROGRAMS CORPORATION 
a subsidiary of Camp Dresser & McKee Inc. 
Ms. Henry 
September 18, 1995 
Page 2 
4. 
5. 
Due to the actual flow direction, groundwater quality downgradient of many areas of concem(Areas 
B, D, E, and F) may not have been adequately characterized. On-site conditions and potential 
contaminant pathways cannot be thoroughly evaluated without this data. Therefore, several 
Remedial Investigation (RI) objectives have not been completely achieved by HLA's field program. 
These RI objectives include: evaluation of on-site conditions for additional areas of potential 
environmental concern; identify potential contaminant pathways; and to generate data for use in the 
human health assessment and for comparison with Applicable and Appropriate Requirements 
(ARARs). 
During agency review of the work plan (June 2,1994), EPA requested that HLA collect a 
background soil sample so on-site soil quality conditions could be compared to local soil 
conditions. In HLA's written response, dated June 24, 1994, it was stated that on-site soil 
conditions would be compared against a background sample collected in 1991 by NUS on 
behalf of EPA. HLA did not include the 1991 background soil results in the Draft Summary 
Report. Instead HLA used analytical data collected by the United States Geological Survey 
(USGS) from locations throughout St. Croix. Information on the sample depths, matrix, 
locations and analytical methods used by the USGS must be provided by HLA in order to 
determine whether a comparison between the two data sets is appropriate. 
EPA's Risk Assessment Guidance for Superfiind, Volume 1, Human Health Evaluation 
Manual, December 1989 requires that total (unfiltered) metals groundwater analyses be used 
when performing a human health assessment. " 
the use of filtered samples for estimating 
exposure is very controversial because these data may underestimate chemical 
concentrations in water from an unfiltered tap. Therefore, data from unfiltered samples 
should be used to estimate exposure concentrations." HLA should present total metals data 
in the Data Summary Report and use it when conducting the site risk assessment. 
For interpretation purposes, the soil and groundwater data should be provided in both 
tabular and graphic form. 
It would be helpfiil for the interpretation of data to provide a topographic survey of the 
property which illustrates all property boundaries, above and below grade structures, and 
on-site discharge points for drainage ditches and subgrade piping. All below 
g r a d e 
structures should also include existing and historical structures, as well as, their 
corresponding surface and invert elevations. 
The report does not discuss the depth to bedrock, at or in the vicinity, of the site. 
The objectives of the investigation outlined in the HLA work plan were designed to fill data 
gaps identified following the review of historical documents. These data gaps were outlined 
in Section 4.1 (Data Needs) of the work plan. Based on CDM Federal's review of the 
305263 
Printed on Recycled Paper 
CDM FEDERAL PROGRAMS 
a subsidiary of Camp Dresser & tVlcKee Inc. CORPORATION 
Ms. Henry 
September 18, 1995 
Page 3 
report, many of these data gaps still remain unresolved. These include: a more detailed 
understanding of site geology and hydrogeology and their implications on contaminant 
migration; seasonal variations in groundwater flow; the extent of potential impacts to soil 
and groundwater quality; information on sewage treatment plant operations and discharges 
near the well fields; existence of a third septic tank; possible source of chloroform detected 
in on-site 
and nearby wells; effect on high chloride and total dissolved solids on 
groundwater chemistry; and location of contaminated container burning area. 
8. 
The report does not address data needs or RI objectives which were not resolved during this 
investigation and does not present recommendations for future resolution. 
9. 
The work plan stated (Section 5.5.3) that during rehabilitation of the on-site production 
wells, the wells would be screened for volatile organic compounds (VOCs) using field 
instrumentation and a grab sample of the sediment would be obtained for inspection. This 
data is not presented in the report text or appendices and there is no discussion of the 
findings. 
10. 
Section 5.5.5.2 of the work plan stated that "One or more continuous water level recorders 
will be installed in selected wells to supplement quarterly measurement data". Again, no 
discussion is provided in the text as to which wells are being continuously monitored and 
how the preliminary results of those readings compare with those levels presented in this 
report. This is important because the field data show groundwater flow in a different 
direction than what was originally assumed . 
11. 
Groundwater contamination has been detected in previous samples from the deep on-site 
production wells and in wells located southeast of the property. At a minimum, additional 
samples from the production wells should be obtained for comparison with the shallow 
monitoring well data. This is important since on-site water level measurements indicate that 
groundwater flow in the deeper zone is to the southeast. Also since contamination has been 
detected in both the shallow and deeper horizons some effort should have been extended to 
determine the degree of interaction between the upper and lower aquifers. It should also be 
determined what lithologic conditions exist at the deeper screened interval. 
12. 
Round 2 groundwater sampling occurred on May 9 and 10, 1995. Based on the text, a significant 
rainfall event (1.85 inches) occurred on the afternoon of May 9, 1995 which provided a sufficient 
influx of water which temporarily reversed the direction of groundwater flow and raised the water 
level elevations several feet. The report does not discuss the potential implications this large influx 
of water may have on the results of groundwater sampling (i.e., potential dilution of contaminant 
levels detected). The report should provide a discussion of this event's impacts on groundwater flow 
and the Round 2 sampling results. 
305264 
Printed on Recycled Paper 
CDM 
FEDERAL 
PROGRAMS 
a subsidiary of Camp Dresser & McKee Inc. CORPORATION 
Ms. Henry 
September 18, 1995 
Page 4 
SPECIFIC COMMENTS 
1. 
Section L2, Page 1, Second Bullet: The second objective of this investigation was to 
identify the existence of previously unidentified source areas on-site. Based on the work 
plan, at a minimum, four additional potential areas of concern remain to be evaluated. First 
is the unlocated septic tank; second, is the location of the "contaminated container burning 
area"; third is further subsurface evaluation of the soils (not sediment) at the effluent 
discharge point for the pipe fi-om the process pit; and fourth is the discharge point for Drain 
Lines # 1 and 2. The justification for each of these areas is summarized as follows: 
Unlocated Septic Tank - Given the reported chemical manufacturing processes which 
occurred at the facility, the potential for accidental discharge existed and, therefore, should 
be evaluated. 
Contaminated Container Burring Area - The nature of this operation needs to be described 
and the location of the area investigated to determine the potential for historical discharges. 
Effluent Discharge Point - Sediment sampling conducted in the area reveals the presence 
of low level contaminants in the sediment within the River Gut at the effluent discharge 
point for the pipe fi-om the former process pit area. The work plan (Section 3.3) indicates 
that the pit was used to collect cooling water and any spills which may have occurred. 
Reportedly, water which accumulated within the process pit was sampled to determine if 
it was contaminated. If it was determined that the water was clean, it was then discharged 
to the River Gut. It is implied that water not discharged was recirculated in the system. No 
description of the sampling procedures, results of sampling or disposal procedures or 
documentation has been provided. Since sludge and water collected from the base of the 
pit during cleaning was disposed of as contaminated material, this unit is of particular 
concern. Given the nature of the stream and the volatility of the compounds disposed of in 
the process pit, the potential exists for the shallow volatization of compounds within the 
sediment and for the infiltration of contaminated waters to the shallow groundwater. 
Effluent Discharge Point for Drain Lines #1 and 2 - The results of samples obtained from 
the Drain Lines #1 and 2 revealed elevated levels of volatile organic compounds. For the 
same reasons outlined above (i.e., volatilization in the shallow sediments and vertical 
infiltration of contaminated water in the River Gut) the exclusion of sampling in the River 
Gut appears insufficient to eliminate the area as a potential concern. 
2. 
Section L2, Page 1, Third Bullet: The third objective of the investigation was to identify potential 
contaminant pathways. 
Although several potential pathways were identified, a thorough 
305265 
Printed on Recycled Paper 
CDM FEDERAL PROGRAMS 
a subsidiary of Camp Dresser & McKee Inc. 
CORPORATION 
Ms. Henry 
September 18, 1995 
Page 5 
examination of the effects of the lithology on vertical migration of contaminants requires further 
clarification. 
3. 
Section L2, Page 2, Fifth Bullet: The fifth objective of the investigation was to determine the 
direction of shallow groundwater flow. Although flow has been determined to be in a radiating 
pattern, generally trending to the northeast - east, no hypotheses has been provided for the variation 
from the assumed direction of flow (east - southeast towards the Caribbean Sea). This is important 
given that the newly installed wells were installed to monitor groundwater quality downgradient of 
suspected sources. Based on these findings, several of the wells may not be properly located to 
accomplish this task. Since the intermittent River Gut provides recharge to the water table that 
causes temporary impact to localized flow, fiirther clarification of its effects on contaminant 
migration is warranted. 
4. 
Section 2.3.1, page 4, Paragraph 3: CDM Federal also collected a split sample from boring 
location SBD3 (0-2'). 
5. 
Section 2.3.3, Page 7, Paragraph 1: The results of the field screening and observation activities 
that were specified in the work plan have not ben presented. It is unclear whether these tasks have 
been performed. This information must be provided. 
6. 
Section 3.2, Page 10: As per the HLA work plan, data validation activities were conducted 
according to the USEPA Contract Laboratoty Program Functional Guidelines for Inorganic/Organic 
Data Review (1994). It should be noted that if Region 2 validation guidelines had been followed, 
the evaluation of blank contamination would have resulted in a different qualification of associated 
samples. 
7. 
Section 4.2.2.1, Page 12, Paragraph 1: HLA should note that ethylbenzene was also detected in 
SBB1 (24-26') and that carbon disulfide was also detected in SBF3. 
8. 
Section 4.2.2.1, Page 13, Paragraph 1: HLA should note that an acetone concentration of 19,000 
ppb was detected in sample SBBl (6-8'). 
9. 
Section 4.2.2.3, Page 13, Paragraph 1: HLA should note that gamma-chlordane and endosulfan 
-II were detected in SBF3 (0-1'). 
10. 
Section 4.4, Page 14: Groundwater data in this section is presented in a very confusing manner. 
Since limited analytes were detected in the four wells, HLA should present the data for each 
sampling round separately. 
11. 
Section 4.4.1, Page 15, Paragraph 4: The values for toluene have been reversed. Toluene was 
detected at 5 ppb in MW-1 not in MW-5 as reported. 
Printed on Recycled Paper 
305266 
CDM FEDERAL PROGRAMS CORPORATION 
a subsidiary of Camp Dresser & McKee Inc. 
Ms. Henry 
September 18, 1995 
Page 6 
12. 
Section 4.4.4, Page 17: See Specific Comment No. 6 regarding data validation qualifiers. 
13. 
Section 5.2.2.1, Page 24: Limited sampling and organic analyses have be performed in the area 
of the above ground tanks where HLA noted stained soil. Analytical results of samples collected 
in the tank farm area indicated elevated levels of ethylbenzene, xylenes, acetone, and toluene in the 
soil. Additionally, elevated levels of volatile organic compounds were detected in the site 
groundwater. Since volatile organic compounds were detected at elevated levels in the groundwater 
and only limited organic soil analysis has been performed, it is premature for HLA to conclude that 
volatile organic compounds in the soil do not represent an area of environmental concern. 
Additional characterization of the tank farm area, especially the area that HLA defines as the "source 
of ethylbenzene in. the groundwater", must be conducted before any definitive conclusions on soil 
quality can be drawn. 
14. 
Section 5.2.2.4, Page 25, Paragraph 1: See General Comment No. 2 regarding background soil 
concentrations. It should be noted that beryllium, antimony, arsenic, and cadmium were not detected 
in the EPA background sample. The EPA sample was analyzed using the inorganic Contract 
Laboratory Program method and detection limits. 
15. 
Section 5.2.2.4, Page 25, Paragraph 2: Beryllium was detected in both the soil and groundwater 
at concentrations above screening criteria. Its presence in the soil may represent a potential concern 
when a better characterization of downgradient groundwater quality has been achieved. 
16. 
Section 5.3.2.1, Page 26, Paragraph 1: The report presents data that the concentrations of 
ethylbenzene £ind xylene have decreased between the two sampling rounds. However, the potential 
reasons for the decrease in concentration levels are not discussed. It is a concern that the sampling 
results may not be indicative of actual concentrations given the potential impacts of the rainfall 
event covering the 24 hour period of May 9 and 10. See General Comment #12. 
17. 
Section 5J.2 J , Page 27, Paragraph 1: Pesticides were detected in the soil and in the groundwater 
at concentrations above water quality screening criteria. HLA has also documented the potential for 
on-site pesticide use. Pesticide compounds may represent a potential concern when a better 
characterization of downgradient groundwater quality has been achieved. 
18. 
Section 5.3.2.4, Page 27, Paragraph 1: See General Comment No. 3 regarding the use of total 
metals in human health assessments. 
19. 
Section 5.3.2.4, Page 28, Paragraph 1: See Specific Comments No. 6 and No. 15 regarding the use 
of data validation qualifiers and beryllium detections, respectively. 
20. 
Section 5J.2.4, Page 28, Paragraph 2: While groundwater beneath the site is not used for potable 
purposes, the elevated concentrations of iron and manganese could affect downgradient, offsite 
3 0 5 2 6 7 
Printed on Recycled Paper 
CDM FEDERAL PROGRAMS CORPORATION 
a subsidiary of Camp Dresser & McKee Inc. 
-7 
Ms. Henry 
September 18, 1995 
Page 7 
receptors. Groundwater flow direction and contaminant migration pathways must be clarified and 
defined before the elevated concentrations of these analytes can be dismissed as a non-concern. 
21. 
Section 6.0, Page 29: Before any final recommendations for further action can be made, HLA must 
address the following: 
• 
Groundwater flow directions in the shallow and deep aquifers must be clarified; 
• 
Groundwater quality conditions in shallow and deep aquifers must be discussed; 
• 
The potential interaction between the shallow and deep aquifers and the potential 
impacts on contaminant migration must be evaluated; 
• 
The tank farm soils that are visibly stained and that have been defined as a 
potential source area must be fiirther characterized; 
• 
Downgradient groundwater quality must be addressed and characterized; 
• 
Potential impacts of the May 9 and 10 rainfall events on the Round 2 groundwater 
sampling results must be evaluated; 
• 
Contaminant migration pathways and offsite receptors must be defined and 
identified; and 
• 
The specified objectives of the investigation must be fiilfilled and all identified 
data gaps addressed. 
22. 
Figure 4-1 and Appendices G & H: In comparing soil boring and well construction logs provided 
in Appendices G & H with lithologies presented on Figure 4-1, the lithologies presented do not 
appear to match. A reexamination of the lithologies presented should be made. It would also be 
helpful if additional representative cross-sectional views are provided. This is of particular concern 
given potential lithologic impacts on contaminant migration. 
305268 
Printed on Recycled Paper