USVI Public Records

Document text

ATTACHMENT 2 SUMMARY REPORT OF TANK CLOSURE AND SUBSEQUENT INVESTIGATIONS BEHIND WESTERN AUTO TUT OO6 06 21 *64633* 64633 1.0 INTRODUCTION 1.1 Background This summary report describes the activities that occurred as part of the closure, removal, and subsequent investigations associated with the two underground storage tanks (LIST) located on the Four Winds Plaza property near the rear service bays of the Western Auto Store. The southernmost tank had a 750 gallon capacity. It was reported that prior to 1982 Western Auto performed automotive oil changes and temporarily stored the spent oil in this tank. The other UST had a 375 gallon capacity. It was reported that this UST may have been used to store diesel fuel for a back-up electrical generator. Generators are typically used on St. Thomas when electrical power is out on the island. The two tanks were used by Western Auto for approximately two years between 1979 and 1981. The USTs were removed from the ground and properly disposed in October 1993. A Closure Report for the USTs was prepared by ENSR Consulting and Engineering (ENSR) and submitted to the Department of Planning and Natural Resource (DPNR) in June of 1994. Subsequent to the submission of this report, which documented the proper closure of the USTs, the following additional activities have occurred: • Characterization and disposal of the wastes generated by the UST closure program; • A second extended Gore Sorter Survey was performed by ENSR; and • Installation of a soil boring and groundwater monitoring well in the immediate vicinity of the former waste oil tank were completed as part of the Tutu well Remedial Investigation (Rl) field program. A summary of these additional activities are presented in this report in order to clarify the results and conclusions documented in the Rl report dated August, 1994. 1.2 Objectives The stated objective of the DPNR at the conclusion of USTs removal program was to delineate the nature and extent of contamination discovered in the shallow gravel layer above the two USTs. ENSR presented a plan to DPNR in November 1993 and January 1994 outlining a work plan for a Remedial Investigation plan. The objective of this document is to summarize the results and findings of the UST closure and post-closure activities. 2-2 TUT OO6 0622 2.0 SITE BACKGROUND 2.1 Site Description and History The Four Winds Plaza property is owned by the Four Winds Plaza Partnership. A variety of commercial and retail companies lease property on the Four Winds Plaza property (i.e., electronic stores, furniture stores, show stores, food stores, and drug stores). Western Auto is one of the companies that presently, and since construction of the plaza in 1978, leases a portion of property on the Four Winds Plaza property. Western Auto is a retail automotive store which sells automotive parts and accessories. Prior to construction of the plaza and under Western Autos* lease agreement, Western Auto prepared a standard set of tank installation specifications for a waste oil tank and floor drain collection system with oil/water separation. These specifications were provided to the landlord who was responsible for installation of the required leasehold improvements. Neither the waste oil tank nor the oil/water separator system were installed based on the specifications prepared by Western Auto (see Section 2.2.8 of Closure Report for Underground Storage Tanks; ENSR, 1994). Z2 Summary of Geologic Setting In general, asphalt and concrete covered the area behind the Western Auto store, where the USTs were buried. The asphalt and concrete was underlain by about 6-12 inches of natural soil consisting of silty day and cobbles. Next, a coarse (approximately 3/4 inch) gravel layer extended another approximately 2-4 inches until native soils (mostly fine textured silt and clay) were encountered. The USTs were encased in this layer of dense native clay below the gravel. The gravel layer extended the entire distance west of the service bays and then north under the building. A mixture of water and oil was present in the entire gravel layer, confined to that layer by the underlying layer of dense day. Phreatic groundwater was not encountered within the 6.5 foot depth of the tank excavation. It is therefore suspected that the gravel layer is charged with either surface water runoff from the higher elevations to the north or spring water (artesian flow). 2.3 Previous Investigations Investigations directly related to the UST area behind Western Auto indude: • A soil gas survey performed by Target Environmental Services in September 1992; • Soil borings and sample collection performed by Caribbean HydroTech in November and December 1992 • UST dosure, induding sample collection and analysis, two Gore Sorber Surveys and two PVC pipe tracings between October 1993 and July 1994 2-3 TUT 006 2.4 Summary of Investigative Rndlngs Activities associated with the excavation of the two USTs, including investigative field studies, were presented in the Closure Report for Underground Storage Tanks (ENSR, 1994) and in two reports by W.L Gore And Associates (W.LGore, 1993; 1994). These reports contained important findings regarding the past and present extent of contamination at the site and its source. The major findings and conclusions of these reports are summarized below. Two USTs, located on Four Winds Plaza property near the rear service bays of the Western Auto store, were removed from the ground and properly disposed of during the period from October 19-29, 1993. Prior to excavation, the contents of the tanks were removed and placed in approved containers for subsequent characterization and disposal. Upon removal, the tanks were placed on plastic sheeting and their insides thoroughly rinsed with clean soapy water. The tanks were then removed from the site to the St Thomas Landfill where they were cut into pieces and landfilled (see Section 2.2.9 and Appendix F of Closure Report for Underground Storage Tanks; ENSR, 1994). During the excavation, it was determined that the USTs were encased in a layer of dense native day, a low permeability soil. Samples of the day were analyzed and found to have a saturated hydraulic conductivity ranging from 1.3 X 10* cm/sec to 4.2 X 10"* cm/sec. These results are indicative of low to impermeable soils. Above the day soil was a 2-4 inch layer of gravel. The gravel was overlain by a 6-12 inch layer of native fill and a 4 inch pad of concrete (see Figure 4 of Closure Report for Underground Storage Tanks; ENSR, 1994). During excavation of the USTs, petroleum hydrocarbon contamination was discovered in the soils above the USTs, primarily in the gravel layer. The gravel layer contained a mixture of water and oil which was confined to the layer by the underlying dense day. Phreatic groundwater was not encountered within the 6.5 foot depth of the excavation. It is therefore suspected that the source of water in the gravel layer is from the higher elevations to the north and that the gravel layer has acted as a conduit for hydrocarbons migrating from a source under the Four Winds building. A 4 inch diameter PVC pipe, running through the gravel layer, was determined to be an additional potential source of the petroleum hydrocarbon contamination in the gravel. The pipe was unintentionally punctured during excavation of the waste oil UST, allowing a mixture of water and a black oily substance to flow out of the pipe. Row from the pipe was not under pressure, but seemed to be gravity fed. The pipe was not associated with either UST and was found to extend away from the Western Auto store, under an adjacent section of the Four Winds building. According to representatives of Four Winds Plaza, this pipe led from an underground spring located to the north of Western Auto to a dstem located to the south of Western Auto. As excavation activities continued, oil was observed to leak from this pipe at several other locations. Oil contamination discovered in the gravel is different than oils found in the USTs. A petroleum hydrocarbon fingerprinting analysis was performed to identify the types of petroleum products present in the soil and gravel, in the waste oil tank and in the PVC pipe and to determine the relationships between them. The results indicate that the oil in both the gravel layer and the PVC pipe was a mixture of heavy and mid-range petroleum product constituents while oil in the waste oil tank was comprised primarily of heavy range constituents. These results are corroborated 2-4 TUT 006 O624 by a PAH ratio analysis of the same samples. Ratios of individual PAHs in samples collected from the waste oil tank display a statistically significant difference from samples collected in the soil and gravel above the tank. Post-excavation sampling indicated that relatively low levels of total petroleum hydrocarbons remain in the soil. In addition to the USTs, approximately 85 cubic yards of soil from the tank graves and contaminated gravel were removed from the site and replaced with dean fill. Prior to backfilling, all excavations were lined with Visqueen to prevent the clean fill from contact with potential contamination. All material that was visibly contaminated with oil, had petroleum odor or had headspace photoionization detector readings greater than 25 units were removed. This included all soil in the vicinity of each soil sample collected during the excavation except for the confirmatory samples (WOT-1 through WOT-4 and DT-1 through DT-4). The extent of gravel and soil excavation during the tank closure is shown as the cross-hatched area in Figure 1. The excavated soils were temporarily stored on site until the proper method of disposal was determined through laboratory analysis. 2.5 Characterization and Disposal of Waste from UST Closure Prior to the removal of the USTs, the contents of each tank were pumped out of the tanks into Department of Transportation (DOT) approved 55-gallon drums. The tanks and associated piping were then removed and thoroughly rinsed with soapy water prior to transport to the St. Thomas landfill. At the landfill, the tanks were cut into pieces and landfilled. All excavated contaminated soil and gravel were stockpiled on-site in roll-off boxes lined with 10-mil plastic sheeting, and later transferred into a lined bulk storage container constructed of wood. The contaminated soil was excavated from the tank graves. The contaminated gravel was excavated from above the tank graves and in the area connecting the two tanks. The drummed wastes and soil pile generated during the UST closures were sampled in February and March of 1994. Of the twenty-four drums of waste generated, twenty-two were sampled for laboratory analyses. Two of the drums, which contained large chunks of concrete and used personal protective equipment (PPE), were not sampled. Table 1 presents information on the samples collected and Table 2 presents the analytical results of the samples. One sample of the oil/water mix, WO-1, slightly exceeded the Toxidty Characteristic Leaching Procedure (TCLP) threshold for lead (5.0 ppm). Thus, any drums containing the oil/water mix were classified as hazardous waste. A total of seven drums were classified as hazardous. This is a conservative procedure because the anomalously high lead value was detected in only one of the oil/water mix samples and it only slightly exceeded the TCLP threshold value. ENSR disposed of all stockpiled wastes by October 1994. DPNR approved, in a letter dated July 27,1994 to ENSR, of disposal of the waste as follows: • Seven drums (drum numbers 1 through 6 and 8 were transferred by a Eastern Chemicals Waste Company frailer for subsequent shipment to the Chem Met Services facility in Wyandott, Michigan as a hazardous waste. The hazardous waste drums were transported under EPA identification number VI00000569509. 2-5 TUT OO6 O625 Seventeen drums (drum numbers 7 and 9 through 24) were transferred to a Clear Ambient Service Company, Inc. truck and taken to the PROTECO facility in Penuelas, Puerto Rico as a nonhazardous waste. The soil pile, consisting of approximately 85 cubic yards of contaminated soil and gravel, did not exceed the action level of 50 mg/kg of total petroleum hydrocarbons (TPH) proposed in the Assessment/Remediation Work Ran. Therefore, the soil pile was disposed of at the St. Thomas landfill on August 29, 1994. 2-6 TUT 006 O626 ^///////////////////A PUMP ROOM TO /-FORMER WASTE-OIL LIST 1 750 G'LLON WOT-I K\\\\^ FORMER DIESEL UST 375 GALLON ^-DISCHARGE Pipr LEGEND f^ REMOVED ASPHALT SOIL AREA WOT-1 • SAMPLING LOCATIONS PIPE LOCATIONS SAMPLE DEPTH * RESULT APPROXIMATE 200 GALLON DIESEL TAN. (ABOVE CROUND) FIGURE 1 NOTE: LOCATION AND ORIENTATION OF LINES ARE APPROXIMATE r • io'-cr ENSR ENGINEERING AUTO^FOyR^WlNDS PLAZA TANK HOLE SAMPLE RESULTS *^^^^^^^^^l**x^^^^^^^^^^^f***£^^^^^^^^^^^ K.P.B. I 2/94 | 7218-001-2CX TUT OO6 O627 Table 1. Information on Samples Collected WO-1 WO-2 WO-3 WO-» SP-2 SP-3 SP-4 SP-5 |^ftliii|iii^^Bil On/Water Mix Aqueous Soil Oil/Water Mix SoU Soil Soil Soil ^iii^PPi^l*lM^r^ilJMi^ii^^liilii|pia^^ lliiii|pi|j|§i|||||:Jl;:;:;:;:S :SSSS»:T*P* "*********?;: ;:•.;.>:.•.: :: TCLP(Metate) TCLP (Metals) TCLP (Metals, VOC, SVOC) TCLP (VOC, SVOC, Pest/Herb) TPH 8015 Modified TPH 8015 Modified TPH 8015 Modified TCLP (Metals) i*/l*rtHc««; tPH • tow Mz&mimtttiW&xxmMZiSf 2-8 TUT OO6 0628 Table 2. Analytical Results SamptoNo. WO-1 WO-2 WCW WO-4 SP-2 SP-3 SP-4 SP-5 i:|iip||piil ililillillllllli^ Ba: 0.296 Cd: 0.002 Cr 0.013 Pb: 5.401 Ba: 0.915 Cr 0.009 Pb: 1.737 As: 0.107 Ba. 0.251 Cr 0.025 Pb: 0.040 Se: 0.068 Ag: 0.051 NA NA NA NA Ba. 0.801 Cd: 0.002 Cr 0.166 Pb: 0.493 Se: 0.071 NA NA ND ND NA NA NA ND iirf: ^iiiii;^-'"^' ^'"rt ''•' i^^-^^^-^^iiS^^ii^^^^^^^^iii-i^^ii'i NA NA ND ND NA NA NA ND IIRiiiiii NA NA NA ND NA NA NA ND liiiiiiiiiii NA NA NA ND 5.4 0.7 6.3 NA ::£&£::::: £:^::::i::^:::^ :?:iiri$ihlff liiiiiijrtjji^^ ::;:•: Si; N'; :;:;;|;»<9<rbiiiiii!ii:i£ii6^^ 2-9 TUT OOS O629 3.0 NATURE AND EXTENT OF CONTAMINATION 3.1 Gore Sorter Soil Screening Surveys 3.1.1 Study Methodology The Gore-Sorber Screening Survey is a passive soil vapor collection technique. The screening modules consist of approximately 100 mg of hydrophobic granular materials and resins specifically designed for collecting organic vapors. The sorbent containers and insertion /retrieval cords are constructed of Gore-Tex expanded polytetrafluoroethylene (ePTFE). The dominant mechanism of analyte transport into the modules is vapor transfer through the microporous walls of the ePTFE insertion cord and sorbent container. For this study, the modules were inserted into pilot holes created by an electric rotary hammer auger or by a slide hammer. The locations of the modules are depicted in maps contained in the two Gore-Sorber reports, dated December 21, 1993 and August 1,1994. For both studies the modules were deployed for about 14 days. The modules were then retrieved, sealed in glass vials, and shipped on ice to W.L Gore Analytical Laboratory. Standard operating procedures for Gore-Sorber Screening modules covering the installation, retrieval, and shipping of the modules are attached. At the lab, the modules were subjected to capillary gas chromatography with mass spectrometry to analyze the module for semivolatile and volatile organic compounds. Gore's lab incorporated an automated thermal desorption unit for analyte recovery from the sorbent. The data generated by the analyses are tabulated and presented by overlaying contaminated patterns detected onto CAD maps of the site. 3.1.2 Initial Survey On November 15-17, 1993, ENSR personnel deployed a total of 20 passive sorbent collection modules (sorbers) in 1-inch diameter drilled holes to depths of approximately 24-36 inches. The survey was conducted in the vicinity of the Western Auto store. The sorbers were installed and left in place for two weeks. The sorbers were then removed from the ground by ENSR personnel and shipped to the laboratory for analysis. One deployed sorber, GS-19, could not be found during the retrieval process and therefore was not analyzed. The results indicated that, while near background levels exist in the former locations of the USTs, a potential source of heavy molecular weight hydrocarbon constituents (as represented by CieH2o)is present to the north of the Western Auto store, directly beneath the shopping center buildings. Lighter hydrocarbons may have a source in that region as well, as represented by C, and CB Cycloalkanes. The results of the C,6HX hydrocarbon and the C8 and C8 Cycloalkanes survey are presented on maps, prepared by W.L Gore & Associates and contained in the Gore report, of December 21, 1993, along with the tabulated data from the survey. 2-10 TUT OO6 O630 The results for some analytes were reported in micrograms of analyte per sorber while other analytes had results reported as total ion counts. This outcome was not unexpected. The analytes that were included in Gore's "standard' analyte list were reported in micrograms because the total ion counts could be related to mass for those compounds. For the others, results were reported as total ion counts. This method of quantitation is perfectly adequate for evaluating the relative distribution of these compounds in the soil vapor as well as for visually depicting the results on maps. Additional sorbers were proposed to better characterize the source of the contamination. This study was carried out and is described below. 3.1.3 Extended Survey On July 14-15, 1994, ENSR personnel deployed a total of 26 modules (sorbers) in 1-inch diameter drilled holes to depths of approximately 36-40 inches below ground surface. The survey not only encompassed the vicinity of the Western Auto store, but also extended further to the south, north, east, and west (see the maps in the Qore-Sorber report of August 11,1994). The sorbers were installed and left in place for two weeks. The sorbers were then removed from the ground by ENSR personnel and shipped to the laboratory for analysis. The results were consistent with the results of the earlier survey and indicated a source of hydrocarbons to the soil gas underneath the former Barker's store building to the north of the Western Auto store. The tabulated results and color displays of the results of several analytes overlain onto site maps are provided in the Gore-Sorber report dated August 11,1994. Results for the heavier hydrocarbon (C^H^ hydrocarbon) and the lighter hydrocarbon (C, and Ce Cydoalkanes) have been plotted and show maximum abundance in the soil vapor underneath the Cost-U-Less store. Tridecane results are also presented and correlate very well with those of the other two compounds. Three soil samples were collected as part of the soil gas survey. They were collected using a manually operated Geoprobe device at approximately 2.5' depth at locations 101,103, and 112. Samples were analyzed for volatile organic compounds and for semivolatile compounds. The results were non-detect for all target analytes. It should be noted that each of the locations selected for soil sampling exhibited non-detect or low concentrations of contaminants in the soil vapor (as determined by the analysis of the modules) and that the soil samples were collected from a level above the modules because it was not possible to drive the Geoprobe down to the level of the modules. Also, the sorbent in the modules act essentially as perfect sinks for organic compounds and hence can concentrate organic compounds from soil vapors. Soil, on the other hand, attains an equilibrium with contaminants in the soil vapor and in this case did not exhibit detectable levels of contaminants. Furthermore, the soil is representative of contamination that exists in the soil sample itself while the modules passively collect vapors from a volume that extends some distance in all directions from the pilot hole depending on soil permeability, porosity, texture, moisture content, etc. In summary, two passive soil gas surveys were performed at Four Winds Plaza. Results from both surveys were consistent and indicated very little (i.e. near background values) soil vapor contamination in the vicinity of the tank graves just west of the Western Auto store and very little soil vapor contamination under Western Auto's service bays or the store itself. Rather, the 2-11 TUT 006 0631 surveys indicated and confirmed the location of maximum soil vapor contamination to be under the former Barker's store to the north of the Western Auto store. 3£ Summary of Post Excavation Sampling Following the removal of the USTs and excavation of surrounding soils, confirmatory samples of the dense day that comprised the side walls and bottom of each tank hole were collected. The samples were analyzed for total petroleum hydrocarbons and the results, shown in Figure 1, are representative of the extent of contamination that remains in the soil. TPH concentrations in the waste oil tank grave were less than 7500 mg/kg and concentrations in the diesel tank grave were less than 1700 mg/kg. The samples were collected at the center of each side wall and from the bottom of each excavation. The depth of sample collection is an indicator of the source of TPH contamination. The concentration of TPH in the shallow gravel (12-18 inches below ground surface) are significantly higher than both the LIST grave results (« 80-96 inches bgs). It is probable that a source for TPH contamination measured in the UST tank areas is the gravel layer, or the 4-inch PVC oil bearing pipe installed in the gravel. 3.3 Summary of Post Closure Sampling On March 31,1994, five months following the completion of tank closure activities, one shallow monitoring well (MW-24) was installed in the vicinity of the former waste oil tank excavations (see Rgure 1-3 in the Geraghty & Miller Rl report for precise well location). The installation was performed by Geraghty & Miller as part of the Phase II Remedial Investigation. The borehole was advanced to a depth of 18 feet (7 feet into bedrock) and was constructed with a double casing through the overburden. ENSR observed the well installation and collected split soil samples from the soil boring. Analysis of soil samples collected from the boring indicate that far lower levels of BTEX, BNAs and TPH currently exist at depth than were present in the shallow gravel layer before excavation. The highest petroleum hydrocarbon concentrations were measured in sample MW-24-2, collected between 4 to 6 feet below ground surface at the level of the base of the waste oil tank. This sample exhibited a TPH concentration of 500 mg/kg, while during the tank closure TPH concentrations were as high as 21000 mg/kg in the gravel layer above the tank. If the hydrocarbon contamination had originated from the waste oil tank, concentrations would have been higher near the base of the tank than above it. Duplicate groundwater samples were collected from MW-24 on June 3,1994. Groundwater at this location generally remains at about 8 feet below ground surface. The samples were analyzed for VOCs and SVOCs and found to be within EPA Drinking Water Guidelines. None of the compounds found during tank closure activities or post closure soil sampling were present in the underlying groundwater at significant concentrations. 2-12 TUT O06 O63 4.0 MIGRATION PATHWAYS AND EXPOSURE POTENTIAL Following the removal of the USTs and approximately 85 cubic yards of soil and gravel, the low levels of petroleum hydrocarbons remaining in the soil behind Western Auto will remain at the site. The hydrocarbons could only be transported via groundwater On soluble form) or via wind (when sorbed to soil particles). Given the geologic conditions at the site, neither of these migration pathways are available to the remaining hydrocarbons. It has been shown that previous to the excavation, the heaviest levels of contamination were confined to the gravel layer near the ground surface by the underlying dense day. Laboratory analysis of the day has determined that it has a very low permeability (10"6 to 10* cm/sec). In addition, there is virtually no infiltration of precipitation due to the overlying asphalt pavement (to be replaced). The lack of infiltration minimizes fluctuation of the water table, which is 6 to 7 feet below the gravel layer and never rises high enough to contact these contaminants. It has also been shown that the lower concentrations of hydrocarbons present in the day below the gravel layer have had no significant impact on groundwater quality. The pavement behind the Western Auto store also eliminates any danger of contaminants being carried from the site via wind. Some areas, which were excavated during the tank dosure have not yet been repaved, however, the contaminated soils have been removed from these areas and replaced with dean fill. Repaving of these areas is scheduled for the near future. Given the relative immobility of hydrocarbons remaining at the site and the lack of an exposure pathway, the risk of human exposure can be considered negligible. There is currently no evidence of impacts to groundwater and little risk that it will be impacted due to the hydrologic constraints represented by the dense day and the low water table. Finally, there is no risk of ingestion of soil, or exposure to soil blown from the site, due to the isolating effect of the pavement. 2-13 JUT 006 0633