PUBLIC HEALTH ASSESSMENT
DARTMOUTH, BRISTOL COUNTY, MASSACHUSETTS
The Re-Solve National Priorities List (NPL) site is a 6-acre area situated in the town of Dartmouth, Massachusetts. During the years 1956-1980 the Re-Solve Company distilled industrial solvents on-site. Waste materials from this process were disposed of by burning solvents in four on-site lagoons and spreading waste oils in various portions of the site. Elevated levels of polychlorinated biphenyls (PCBs) and volatile organic compounds (VOCs) were measured in soils throughout the site in 1981, and it was added to the NPL in December 1982.
This site is characterized as a past and present public health hazard primarily due to the likely past and present exposure to PCBs through the consumption of PCB contaminated fish and eels from area water bodies. Based upon results of fish sampling that detected PCBs in area eel, the Massachusetts Department of Public Health issued and updated an advisory in 1986 and 1994 recommending that people not eat the fish and eel caught in these affected waters. The potentially exposed population; therefore, include area fishers who may be consuming contaminated fish or eel.
A past, and possibly current, completed pathway of concern is the ingestion of lead contaminated water from several residential wells in Dartmouth. The elevated levels of lead detected in these wells is not considered to be site related because lead is not a contaminant of concern at the site, and there is no discrete groundwater plume from the site to these wells. However, sampling detected in 1986 measured lead in these waters at levels as high as 241 ppb. Because subsequent sampling of these wells has not included analyses for lead, it is unknown whether this exposure is ongoing. The potentially exposed population includes the residents served by the eight residential wells contaminated with lead and Re-Solve workers who may have consumed contaminated waters. There is a special concern for children who may be living in these residences because they are particularly sensitive to the health effects of lead exposure.
There are several potential exposure pathways associated with this site. Exposure to VOCs and PCBs in groundwater could occur in the future if, in the absence of adequate remediation, wells are developed for bathing, laundering, or drinking purposes. Institutional controls are currently in effect prohibiting any use of on-site groundwater until federal and state water quality standards are met. Area surface waters also represent a potential pathway of exposure in the past, present, and future due to the presence and possible ingestion of VOCs in water and/or dermal contact with PCBs in sediment. However, the location and quality of these contaminated waters are not conducive to swimming; therefore, significantly limiting this pathway. Dermal contact or ingestion of PCB or VOC contaminated surface soils represent potential exposure pathways for the past. However, completion of on-site and limited off-site soil remediation has considerably decreased their contaminant levels.
Analysis of incidence data for cancers potentially related to site contaminants did not indicate significant elevations for the period 1982-1990 in the census tract encompassing the site. The Massachusetts Lead Poisoning Prevention Program did not report any cases of childhood poisoning between 1991 and 1993 for the town of Dartmouth.
Area citizens have expressed an overall interest in the impact exposure to site-related contaminants may have on citizen health. The safety of those individuals living closest to the site was questioned by local health officials. The safety of surface water near the site and the fish caught from these waters were also questioned. Finally, concern was expressed regarding both the safety and effectiveness of the proposed and ongoing remediation efforts chosen by EPA.
Recommendations made in this public health assessment suggest re-monitoring of the eight residential wells with past lead contamination as well as the remainder of the wells within a one-mile radius of the site, and post remedial monitoring of soils and groundwater. This need will be addressed as part of the Management of Migration (MOM) portion of the remedy for the site in which the Settling Parties will monitor 16 residential wells around the site (focus on residents abutting the site and downgradient/crossgradient from the site) for VOCs on an annual basis beginning in September 1995. Upon attainment of EPA's groundwater cleanup standards, the treatment process will be shutdown and groundwater quality monitored for an additional 3 years to document compliance with the Consent Decree.
In addition, the settling parties will begin an annual fish monitoring program in 1996 in which various species of fish will be collected from Cornell Pond and the Copicut River and monitored for PCBs on an annual basis. Based upon review of past fish sampling, and associated toxicological estimates included in this health assessment, this fish monitoring program is strongly supported. Furthermore, EPA and MDPH have posted fish warning signs around Cornell Pond and Copicut River, and conducted a public meeting explaining the fish health advisory.
Two Remedial Investigation/Feasibility Studies (RI/FS) of the site concluded that on-site soils with PCB levels >25 ppm and sediments with PCB levels >1 ppm should be excavated from the site. The source control remedy including excavation and treatment of contaminated soils utilized an innovative technology called X*TRAX (a low-thermal desorption process). The X*TRAX process began on June 21, 1993 and was completed on July 21, 1994. Confirmatory soil sampling by both the remedial contractor and EPA's oversight contractor confirmed that all contaminated soils were excavated and properly treated. In addition, approximately 1,500 cubic yards of sediments contaminated with PCBs at concentrations greater than 1 ppm in the Northern and Eastern Wetlands were excavated and back-filled on-site. The Wetlands were then backfilled and graded with clean off-site soil, and re-vegetated. The site is currently graded with 18 inch crushed stone and secured with a fence. The groundwater treatment design is in its final stages, and construction of a groundwater treatment facility is expected to begin in 1996 (EPA, 1995).
The data and information developed in the public health assessment for the Re-Solve site was evaluated by ATSDR and MDPH to determine the need for follow-up health actions. Because of the probable past and present exposure to site-related contaminants, the Health Activities Recommendation Panel (HARP) determined that follow-up health actions are indicated including community education regarding the consumption of PCB-contaminated fish and potential ingestion of lead containing well water. The MDPH will continue to monitor the cancer incidence rate for the town of Dartmouth through the Massachusetts Cancer Registry at the MDPH. The MDPH will provide education to the Dartmouth community to assist in their understanding of potential exposures.
The Re-Solve National Priorities List (NPL) site is situated on a 6-acre parcel of land in North Dartmouth in southeastern Massachusetts. North Hixville Road borders the site to the west, while wetlands and small waterways comprise the remaining three boundaries. A small wetland area is located in the northeast corner of the site which is drained by an unnamed tributary that comprises the remainder of the site's northern and eastern boundaries. This brook flows easterly and then southerly where it joins Carol's Brook which flows easterly and constitutes the site's southern boundary (Figure 1).
Between 1956 and 1980, Re-Solve Inc. operated as a waste chemical reclamation facility on this site. A variety of hazardous materials including solvents, waste oils, organic liquids and solids, acids, alkalies, inorganic liquids and solids, and PCBs were handled by Re-Solve (U.S. EPA 1992). Waste materials produced during this period were disposed on-site. Four unlined, on-site lagoons were used for burning volatiles and old tires (ATSDR 1987). Waste oils were spread on various portions of the site and cooling water from these processes was discharged to a shallow on-site cooling pond in the eastern portion of the site (U.S. EPA 1987a). The Re-Solve site was added to the National Priorities List in December 1982; thereby becoming eligible for federal assistance as part of the Superfund Program.
The Remedial Investigation and Feasibility Study (RI/FS) for the site was initiated in the fall of 1982 and completed in June 1983 (CDM 1983). Chemical analyses of the air, surface water, groundwater, soil, lagoon waste, and sediment from the site assessed the extent of on-site source contamination and were used to evaluate remedial alternatives. Approximately 3,100 cubic yards of lagoon wastes and 3,900 cubic yards of contaminated soils were identified as on-site sources of contamination. Elevated levels of PCBs, as well as other organics such as phenols, trichlorobenzene, bis(2-ethylhexyl phthalate), isophorene, ethylbenzene, toluene, and o-xylene, were detected in on-site soils.
According to the Record of Decision (ROD) signed in July, 1983 remedial action for the site would include the excavation, treatment, and transportation off-site of 7,000 cubic yards of source material (U.S. EPA 1992).
To assess the extent of contamination that had migrated beyond the boundaries of the site, an Off-Site RI/FS was initiated by the EPA in September 1983. This investigation included the installation of 45 groundwater monitoring wells at 25 locations; surface water and sediment sampling; soil borings; test pit excavations; and lagoon depth analyses. This report confirmed the major sources of contamination that had been identified in the On-Site RI/FS.
During this series of off-site investigations from September 1984 to May 1985, EPA removed 15,000 cubic yards of contaminated lagoon wastes and soils, and discovered additional on-site contamination (U.S. EPA 1987a) (CDM 1985). A supplemental RI, including a soil boring program, surface water monitoring program, and sediment analysis, was conducted to determine the nature and extent of contamination in on-site soils and to supplement information that had been presented in the Off-Site RI.
This supplemental RI for the Re-Solve site was initiated in September 1985 and completed in February 1987 (CDM 1987). Eight fish samples from two sampling stations at the Copicut River and Cornell Pond were collected and sampled by EPA as part of this RI. The Massachusetts Department of Public Health issued an advisory in August 1986 recommending that residents not eat eels from the Copicut River, based upon the elevated PCB levels detected in these samples.
A brief health assessment was prepared by the Agency for Toxic Substances and Disease Registry (ATSDR) in 1987 to assess the public health impact of the Re-Solve Site. It concluded that a potential health threat exists for persons who may contact or ingest surface soils contaminated with PCBs and VOCs or contact surface water and groundwater contaminated with PCBs and VOCs. The health assessment included the following recommendations:
- Consider testing soil samples for polychlorinated dibenzodioxins and polychlorinated dibenzodifurans.
- Take appropriate action to prevent contaminants from the Re-Solve site from migrating into the ground or surface water.
- Prevent consumption of potable water when the concentration of lead is greater than 50 ppb (the action level for lead has since been reduced to 15 ppb)
- Prevent consumption of potable water when the concentration of any VOC exceeds EPA Water Quality Criteria.
- Provide information for the pattern of consumption of fish from the river and ponds to be used in assessing the presence or absence of a health hazard.
The 1987 FS for the Re-Solve site described and evaluated the cleanup options, or remedial alternatives, which would address contamination at the site. The ROD, including a source control and management of migration component, was signed by the EPA on September 24, 1987. The source control component of the ROD estimated that 22,500 cubic yards of PCB contaminated soils be excavated and treated on-site to meet the 25 ppm PCB health based cleanup level for on-site soils. The cleanup level for contaminated sediments in the wetland resource areas north and east of the site and the unnamed tributary is 1 ppm PCBs.
The Source Control Remedy to address these cleanup levels required the excavation and treatment of contaminated soils through the X*TRAX system, a low-thermal desorption process. The X*TRAX process began on June 21, 1993, and was completed on July 21, 1994 at the ReSolve Site. RUST Remedial Services, Inc. excavated and treated over 38,000 cubic yards of PCB contaminated soil and sediments.
These PCB contaminated soils were treated to levels well below the Agency's established cleanup level of 25 ppm. The X*TRAX process on average treated the PCB contaminated soil to below 5 ppm. RUST Remedial Services collected approximately 900 soil samples to confirm that all contaminated soils were excavated and properly treated. EPA's oversight contractor collected an additional 150 samples to confirm all of RUST's sampling results.
In addition, approximately 1,500 cubic yards of sediments contaminated with PCBs greater than 1 ppm in the Northern and Eastern Wetlands were excavated, and backfilled on-site. The wetlands were then backfilled and graded with clean off-site soil, and re-vegetated.
The Source Control Remediation (excavation of soils/sediments, restoration of wetlands, grading of treated soils, capping with 18 inches of gravel and demobilizing from the site) was completed in December 1994. Now that soil remediation is complete, the management of migration (MOM) component, which will address groundwater contamination, can begin. This component is currently in its design stages: the 95% Design Report was reviewed in early June 1995 and construction of the treatment facility is expected to begin in 1996 (MDEP 1995).
As the MOM portion begins, the 16 residential wells around the site will be monitored for VOCs on an annual basis beginning in September 1995. A significant amount of monitoring will occur for all on-site monitoring wells prior to the groundwater treatment plant's operation and throughout the plant's operation to monitor contaminant concentrations, trends, and the effectiveness of the system. The baseline data collected before operation startup will be comprehensive (inorganics, VOCs, Semi-VOCs) under EPA's Total Analytical List (TAL) and Total Compounds List (TCL). The operation monitoring will focus primarily on VOCs monitoring. Upon attainment of the EPA's groundwater cleanup standards, the treatment process will be shut down and groundwater quality monitored for an additional three years (U.S. EPA 1995). The site is completely enclosed by a locked fence that is intermittently covered with barbed wire and written warnings.
A site visit was conducted by William Strohsnitter, Environmental Analyst for the Massachusetts Department of Public Health (MDPH), on July 16, 1992. The site was adequately secured by the company conducting remediation. A municipal police officer was stationed at the entrance during working hours to ensure that no unauthorized personnel entered the site and that those who were permitted to enter immediately wore head and eye protection. It was also reported that the site was secured by a private company during the non-working hours. The majority of the site's perimeter was secured by 8-foot-high barbed wire fencing. The area in the northeast corner was not secured due to the presence of extensive wetlands and brush. The thick vegetation and swampy ground surface made site access at that point difficult. There was no evidence of site trespassing noted during the site visit.
Piles of excavated soils were noted in the center of the site, which were incompletely covered by plastic sheeting. The fencing to the west, bounding residential property, was woven with plastic sheeting. Treatment of extracted waters was ongoing the day the site was visited. A faint organic odor was noted when pumping activities began in the early afternoon. These odors diminished with time and increased distance from the treatment station. There were seven residences noted one-quarter of a mile north along Hixville Road. A small vegetable (corn) field was noted at the residence adjacent to the site's entrance to the immediate north, 75 feet from the site's western boundary.
The Re-Solve site project manager for the MDEP visited the site in May 1995, and some site conditions have changed since the MDPH's site visit in July 1992. Currently, all soil remediation equipment has been moved off-site, and there is no remediation at this time as groundwater remediation is in the design stages. The site is completely enclosed by a locked fence that is intermittently covered with barbed wire and written warnings. Guards are no longer in place at the site. All soil on-site contaminated with PCB concentrations greater than 25 ppm were excavated, treated, and graded back on-site. All soils were treated well below the soil clean-up level, and an additional 18 inches of crushed stone were graded on-site. Within the perimeter, one small area which contains an active pump has been enclosed in another fence to prevent trespassing (MDEP 1995).
According to the 1990 Census the population of Dartmouth is 27,244, which represents a 13.7% increase over that recorded in 1980 (23,966). The site is located in a sparsely populated section of Hixville, a village of Dartmouth. According to the 1980 Census, 326 people reside within a one-mile radius of the site, 35% of which are less than 18 years of age and 5% are greater than 64 years of age. There are currently 12 residences located on North Hixville Road which extends 1.25 miles from Hixville center to the Fall River boundary. One residence is situated externally to the site's southwestern corner immediately north of the entrance to the site and 75 feet from the site's western boundary. According to the Board of Health for Dartmouth, the water needs for this residence, as well as other residences within a one-mile radius of the site, are all served by private wells (Dartmouth BOH 1995). Furthermore, as there is no municipal water supply line that extends to this area, municipal water is not available to these residences. An automobile salvage yard and a small car dealership are also situated on North Hixville Road north of the site.
The area encompassing an approximate two-mile-radius semicircle north of the site is predominately wooded areas and wetlands. The Copicut Reservoir, which serves the drinking water needs for the city of Fall River, is situated to the north of the Fall River-Dartmouth boundary 4,000 feet north of the site. This surface water body drains into the Copicut River which flows southerly approximately 150 feet east of the site. The unnamed tributary which bounds the site to the north and east flows into the Copicut River which flows southerly approximately 150 feet east of the site. The Copicut eventually drains into Cornell Pond approximately one-quarter mile southeast of the site.
Fishing is reportedly ongoing in both Cornell Pond and the section of the Copicut River near the site despite the posted signs warning against such actions (MDPH 1992). It is not known if fishing occurs in the Copicut Reservoir, although it is prohibited due to its status as a water supply. The New Bedford Rod and Gun Club, which stocks the nearby wooded areas with pheasants and rabbits, is situated approximately 1,700 feet northeast of the site. It is reported that hunting, fishing, and target shooting are ongoing in these areas north (and upstream) of the site.
The land surrounding the site is not used extensively for vegetable growth or farming. There is one farm approximately three quarters of a mile northeast of the site where both dairy cattle and feed for these animals are raised. It is not known how many of these residences near the site grow vegetables for consumption. Corn stalks were, however, observed in the yard of the residence closest to the site.
Available cancer incidence data (1982-1990) from the Massachusetts Cancer Registry (MCR) of MDPH were reviewed to determine if there is an indication of an increased incidence of cancer in the areas surrounding the site. Massachusetts law mandates that all hospitals within the Commonwealth report to the MCR all incident cancer cases diagnosed within their facilities.
Although the lead detected in residential well water is not considered site related, its presence at elevated levels in nearby residential wells warranted the evaluation of relevant health outcome data. Therefore, data from the Childhood Lead Poisoning Prevention Program (CLPPP) within the MDPH were also reviewed in order to address the possibility of environmental lead exposure affecting the prevalence of childhood lead poisoning in the area. Universal screening of blood lead levels is mandatory for Massachusetts children from 9 months to 4 years of age. According to the CLPPP a lead poisoned child is defined as one with blood lead levels of 25 µg/dL or greater. The Massachusetts state action level, however, is 20 µg/dL (CLPPP 1995).
Members of the Dartmouth Health Department and Conservation Commission as well as members of a citizen environmental group were asked if they or their associates had any health concerns connected with the site. Citizens expressed concerns on possible health effects from site-related exposures. Specifically, the safety of those individuals living closest to the site was questioned by local health officials. The safety of surface water near the site and the fish caught from these waters was also questioned. Finally, concern was expressed regarding both the safety and effectiveness of the proposed and ongoing remediation efforts chosen by EPA. These concerns will be evaluated in the Community Health Concerns Evaluation section.
This section evaluates available environmental data at the site. Site concentrations are compared with those levels to which exposure has been associated with minimal risk of adverse health effects. These comparison values include calculations from Reference Doses (RfD), Reference Dose Media Evaluation Guides (RMEG), Environmental Media Evaluation Guides (EMEGs), EPAs Drinking Water Lifetime Health Advisories (LTHA) and Maximum Contamination Levels (MCLs) standards established by the Safe Drinking Water Act. These values are derived from animal and human studies for which non-cancer adverse health effects were investigated. Cancer Risk Evaluation Guides (CREGs) are used for compounds which are known or suspected to be carcinogenic based on either animal or human studies. CREG levels are those for which exposure is associated with an excess cancer risk of one in a million. The presence of contamination on the site does not necessarily indicate that exposure to site-related contaminants is ongoing. The health risks of such exposure are reported in subsequent sections of this public health assessment. Some contamination levels did not exceed their respective comparison (or guidance) values.
A soil boring program was conducted in 1985 as part of the Supplemental RI. Results of soil sampling (0-2 feet BLS) are presented in Table 1. Extensive PCB contamination was detected. PCB concentrations greater than 50 ppm were found in 17 of 40 samples with the majority detected in the area where waste oils had been spread during solvent reclamation activities at the site. The maximum concentration of PCBs measured in this sampling program was 3,800 ppm which is significantly above ATSDR's EMEG of 10 ppm for PCBs in soil. A range of volatile organic compounds and chemical concentrations were detected throughout the site area. Trichloroethylene concentrations included levels from 3.6 to 740 ppm. The CREG for trichloroethylene is 60 ppm. Other VOCs detected included 12,000 ppm methylene chloride, which has a CREG of 90 ppm, and 2.4 ppm 1,1,1-trichloroethane.
This 1985 soil boring program also evaluated subsurface soils (2-10 feet BLS). Although PCB concentrations were generally above 1,000 ppm, PCB levels greater than 10,000 ppm were detected as far down as 14 feet BLS in the former oil spreading area. Further analysis of these soil samples indicated that they contained a mixture of PCB congeners including Aroclor 1242, 1248, and 1254. As seen in upper level soils, subsurface samples also included VOCs such as trichloroethylene and tetrachloroethylene. Trichloroethylene was measured in 470 and 1,200 ppm in the former oil spreading area at depths between 4 and 14 feet BLS. The CREG for trichloroethylene in soil is 60 ppm. Subsurface soils from the former oil spreading area also included 400 ppm of tetrachloroethylene at a depth of 12 to 14 feet BLS. The CREG for tetrachloroethylene is 10 ppm. The results of these subsurface sample measurements are summarized in Table 2.
More extensive surface soil monitoring was conducted in 1990, during the design phase of the selected soil remediation. It more clearly defined the amount of soil excavation that would be necessary in order to meet cleanup levels. Possible comparisons between these two investigations (1985 and 1990) are limited because only a few of the areas where surface soils (0-1 feet BLS) were sampled in 1985 were again monitored in 1990. Of the 191 samples collected, 106 contained PCBs at concentrations above the 25 ppm cleanup level and 124 at concentrations above the current ATSDR comparison value of 10 ppm. The maximum PCB concentration measured in this sampling program (9,200 ppm) was collected near the foot of the cooling pond in the eastern section of the site. Furthermore, the eastern boundary, which was not sampled in 1985 also contained significantly elevated concentrations of PCBs (2,740 ppm). This sampling program did, however, confirm the presence of extensive PCB contamination of soils in the oil spreading and former lagoon areas. Along the western boundary of the site, the maximum PCB concentration was detected in soils from this former oil spreading area. The majority of surface soil samples collected from the western boundary did not detect PCBs. The detection limit range used for these samples was 0.5 to 3.0 ppm.
During the 1992 pilot study of the X*TRAX desorption system required by the ROD, feed soils were analyzed for the presence of PCDDs and PCDFs which are known to be by-products of PCB combustion. Site records indicate that PCB combustion did in fact take place during Re-Solve Inc. solvent reclamation activities; however, no soil monitoring to this point had included analyses for PCDDs and PCDFs. Results of this analysis, using 2,3,7,8-TCDD as a toxicological equivalent, detected 2,3,7,8-TCDD in a range from 0.23 to 0.35 µg/kg (PRC 1992). The ATSDR Environmental Media Evaluation Guide for residential soils is 1 µg/kg.
As stated above, soil remediation was completed at the site in 1994. Soil remediation successfully lowered the levels of PCBs in on-site soils to below 25 ppm with an average that is below 5 ppm PCBs (U.S. EPA 1995). Portions of off-site soils were also remediated to 1 ppm PCBs.
Groundwater Monitoring Wells
In 1985, the most extensive on-site ground water contamination was detected in the eastern and southeastern sections of the site. Hydrogeologic investigations have shown these areas to be hydraulically downgradient of the former disposal areas and in the immediate vicinity of the former cooling pond, indicating possible VOC migration through groundwater. Groundwater sampled near the ground surface appeared to contain more contamination than that sampled from bedrock. One exception to this trend was noted in off-site bedrock groundwater monitored east of the Copicut River. PCBs were detected in groundwater sampled beneath the former oil spreading area (Table 3A).
Groundwater on the site was again monitored in 1989 and the results are reported in Table 3B. The most apparent changes in groundwater contaminant levels occurred in the former lagoon area where vinyl chloride levels were detected at 11,200 ppb and 21,000 ppb in overburden and bedrock groundwater wells respectively. No vinyl chloride was detected in these wells during the monitoring that was conducted in 1985. Levels of trichloroethylene measured in these two wells in 1989 were greatly reduced to none detected in the overburden well and 250 ppb in the bedrock well. No specific groundwater monitoring for PCBs was conducted in 1989.
On-site air monitoring was conducted in 1985. The results of this investigation are presented in Table 4. An organic vapor analyzer was used to record volatile emissions, and a PM 10 monitor and aerosol monitor were used to monitor dust levels. The organic vapor analyzer and dust aerosol monitor continuously recorded data and were linked to an alarm system which alerted workers when volatile organic levels exceeded 5.0 ppm and dust levels exceed 5.0 mg/m3. If these levels were exceeded, then the Settling Parties would evaluate the cause and as necessary correct any problems or possibly shut down activities at the site. During the life of the on-site remedial activities, these levels were exceeded twice during operations around the summer of 1993. The two elevated perimeter monitoring readings were evaluated and a Health Assessment conducted. It was concluded that no impacts occurred beyond the perimeter boundary of the Site. In addition, charcoal tubes and PM 10 filter monitors were installed at the perimeter of the fence to trap organic contaminants and dust particles to the air. The contents of these tubes and filters were analyzed through an approved laboratory and no quantifiable levels of organic compounds or dust were detected (U.S. EPA 1995).
Table 5 shows that elevated PCB levels were detected in surface soils monitored off-site south of the unnamed tributary. Other off-site areas where PCBs were detected in surface soil include that immediately northwest of the former oil spreading area and to southeast of the cooling pond. Of the 11 off-site surface soil samples monitored, three contained PCB levels greater than 50 ppm. No extensive PCB contamination was detected in off-site soils at sub-surface levels.
Monitoring of soils, immediately outside the northwestern site boundary, one to two feet BLS was again conducted in 1990. Of the 14 samples taken, PCBs were detected in nine. The maximum PCB level detected was 152 ppm, 50 feet from the western boundary of the site. PCB contamination was also detected 50 feet further west where levels were as high as 115 ppm.
Groundwater Monitoring Wells
Early site investigations determined that the groundwater from the site is generally moving to the east and southeast. Furthermore, the most extensive VOC contamination detected in off-site ground water was found in that sampled east and southeast of the site and west of the site-associated surface waters. These contaminated waters were contained in the aquifer situated close to the ground surface (overburden). The contamination levels in these waters diminished sharply east of the site-associated surface water bodies. This is likely due to the Copicut River which has characteristics of a hydraulic boundary in regards to groundwater flow. Consequently, the overburden contaminant migration plume discharges, almost entirely, into the Copicut River and the unnamed tributary which is also in the path of the plume (U.S. EPA 1987a). Elevated PCB levels were detected in groundwater sampled beneath the wetlands north of the site. No other PCBs were detected in the groundwater sampled near the site.
Contamination was also detected in bedrock groundwater. The bedrock VOC levels were less than those detected in the overburden system. As previously stated, the exception to this trend is in those wells situated immediately east of the Copicut River. In this area, the VOC levels detected in bedrock groundwater were higher than those detected in the overburden. Bedrock and overburden groundwater sampled further from the site contained little or no contamination. The results of the off-site groundwater monitoring are reported in Table 6A.
Groundwater east of the site was remonitored in 1989. Vinyl chloride levels remained elevated in bedrock wells located southeast of the site adjacent to the Copicut River. Vinyl chloride levels were again detected in the overburden well near the juncture of Carol's Brook and the Copicut River. With the exception of the waters drawn from the well at the northeastern corner of the site, there were no marked reductions of contaminant levels in groundwater east of the site. The results of the off-site groundwater monitoring conducted in 1989 are presented in Table 6B.
Groundwater Private Wells
Significant monitoring of private well water has been conducted at the Re-Solve site; however, there is limited continuity in the wells chosen for sampling and the compounds for which the waters have been analyzed. In December, 1981 the Department of Environmental Quality and Engineering (DEQE) (now Massachusetts Department of Environmental Protection (MDEP)) sampled waters from six private wells along North Hixville Road and Old Fall River Road. None of the wells contained contaminants at levels that violated the National Priority Drinking Water Standard's Maximum Contaminant Levels. An EPA contractor sampled 6 residential wells along North Hixville Road and Old Fall River Road including the Rego home (PW-48) just north of the site on North Hixville Road in January 1983. Only three of the homes sampled in 1981 were resampled in 1983. All of the samples were analyzed for VOCs, refractory organic compounds, trace metals, oil and grease, and total halogenated compounds. Analytical results showed that well PW-48 contained 6.1 ppb trichloroethylene, and 2-hexanone was measured below analytical detection limits. No organic contaminants were detected in the other wells (U.S. EPA 1987a).
A more extensive sampling of residential wells was conducted as part of the 1987 Supplemental RI from 1985 to 1986. Fifty-six upgradient and downgradient wells located within a one-mile radius of the site and deriving their waters from overburden and bedrock aquifers, were evaluated. The investigation included three sampling rounds: November 1985-January 1986; March 1986 - May 1986; and November 1986. Despite their proximity and similar depths, these wells showed little repetition of contaminants during these three sampling rounds. The results of this residential well water investigation are located in Table 7A.
Significantly elevated lead levels (26 - 241 ppb) were measured in eight of the wells evaluated in this investigation. These contaminated wells are located both upgradient and downgradient of the site and over a wide range of distances. Therefore, this contamination is not considered site-related. The Action Level for lead in drinking water is 15 ppb. Other contaminants detected at significant levels include n-nitrosodiphenylamine (27 ppb), benzene (5 - 9 ppb), 1,2-dichloroethane (6 ppb), methylene chloride (180 ppb), and di-n-octylphthalate (7 ppb).
A residential well sampling program was initiated in late 1992 by an EPA contracted consultant and has since become an annual program, with samples being collected in September 1993 and September 1994. The 1992 samples were analyzed for volatile organics and general water quality parameters (e.g., nitrates, total dissolved solids, chloride). Because all of the general parameters were within normal ranges, analysis of samples after 1992 did not include these chemicals. Therefore the remaining sampling rounds were analyzed for PCBs, VOCs, and semi-VOCs (Table 7B). Because none of these waters were evaluated for lead content, it is not possible to determine whether the wells previously deemed lead-contaminated remain as such.
All of the wells sampled during the 1985-1986 investigation were not included in this annual monitoring program. Therefore comparisons of contamination levels over time are limited. During the 1985-1986 investigation, well PW-26 contained 9 ppb of benzene in the first sampling round. Subsequent rounds in this investigation of well PW-26 did not detect benzene. Furthermore, benzene was not detected in the waters sampled from this well in the September 1994 investigation. Similarly, well PW-25 contained 27 ppb of n-nitrosodiphenylamine in 1985-1986 investigations, but did not contain n-nitrosodiphenylamine in later annual monitoring. Well PW-29 contained 6 ppb of 1,2-dichloroethane in early investigations, but more recent monitoring of this well has not detected any further contamination.
In 1993, Well PW-48, immediately north of the site was found to have 6.0 ppb of trichloroethylene, which has a CREG of 3 ppb. In response to this finding, and as a precaution, in December 1993, a water treatment system capable of removing this compound was installed at this residence. The system has been operating since then and is maintained on a 6 month schedule by a local water service company. Because of their proximity to the site, Well PW-0, PW-01 and PW-14 also had treatment systems installed. Water sampled from these wells is collected prior to entering the treatment system. See Figure 6 for the location of these wells (ENSR, 1994). Also in 1993, a low level of toluene (0.6 ppb) was detected in well PW-1 and trace amounts of methylene chloride were detected in several of the monitored residential wells.
Residential well data in 1994 indicated the presence of chloroform ranging from 1.7 to 13 ppb in almost all residential wells tested, and 3.6 ppb in their trip blank sample. Chloroform has a CREG of 6 ppb. The fact that chloroform was detected in ENSR's trip blank and not in EPA's split samples suggests that ENSR's chloroform contamination was the result of contaminated sample bottles and therefore is not likely present in these residential wells (U.S. EPA 1995).
As part of the 1985 Off-Site RI/FS, 26 samples were collected from area water bodies to evaluate their existing and potential for future contamination. These samples were taken from the wetlands north of the site, the Copicut River, the unnamed tributary, Carol's Brook, and the pond area south of the site. Surface water contamination was limited for the most part to the wetland north of the site, the unnamed tributary, and the Copicut River downstream from the unnamed tributary. Sampling stations located progressively downstream from the site area on the Copicut River showed progressively decreasing VOC concentrations. Both elevated VOCs and PCBs were detected in the pooled surface waters in the wetlands immediately north of the site.
The 1987 Supplemental RI also investigated surface waters from the wetlands north of the site, the unnamed tributary, Carol's Brook, and the Copicut River. Sampling included downstream portions of the Copicut River in Cornell Pond, the Shingle Island River, and Noquochoke Lake. Results indicated that VOC contamination extended from the wetlands north of the site, through the unnamed tributary and Copicut River, and downstream to Cornell Pond. Although PCBs were not found in the Copicut River, waters from Carol's Brook and the wetlands to the north did contain PCBs. The results of surface water monitoring are reported in Table 8.
Sediment contamination was investigated in conjunction with the RI/FS in the fall of 1982 through June 1983. Sediment contamination was detected in the pond immediately north of the site, the unnamed tributary to the east of the site, Carol's Brook, and the Copicut River. The nature of this contamination and the levels of contaminants were a function of proximity to the site and local drainage patterns.
In this 1982-1983 investigation the pond sediments were contaminated with organic contaminants. PCBs were measured at levels greater than 45 and 386 ppm. Sediment collected in and immediately adjacent to the unnamed tributary contained PCBs at concentrations from .425 to 41 ppm. Sampling stations located downstream of the Copicut River, which drains southern portions of the site, also contained sediments contaminated with PCBs.
An evaluation of area sediments was conducted in January 1984 in conjunction with the 1985 Off-Site RI/FS. This investigation assessed the extent of off-site sediment contamination. An additional round of sampling was conducted in January 1985, to focus specifically on PCB contamination. This first round of sampling included sediments collected from the wetlands north of the site, the Copicut River, the unnamed tributary, Carol's Brook, and the pond area to the south of the site. These samples were analyzed for VOCs, refractory organic compounds, total organic halogenated compounds, and inorganic compounds. Results indicated that the wetland immediately off-site to the north was heavily contaminated with nonvolatile organic compounds. This contamination was dominated by PCB levels at concentrations as high as 32 and 60 ppm at two of the sampling stations. Sediments from the unnamed tributary contained PCB levels from .11 to 22 ppm as well as phthalates (720 ppm), trichlorobenzene (9.2 ppm), and phenol compounds (totalling 210 ppm). Sampling stations located downstream of both Carol's Brook (.27 and .42 ppm) and the Copciut River (.44 ppm) contained PCB contaminated sediments. The results of analyses of samples collected in March 1982 and January 1985 indicated that there was widespread PCB contamination in local surface water sediments.
The 1985 supplemental sediment sampling contained 51 sampling locations including the Copicut River, Cornell Pond, Carol's Brook, the unnamed tributary, and the wetland north of the site. Sediment at each location was measured at multiple depths in six inch intervals. Results indicated that the entire wetland area north of the site had PCB contamination above 1 ppm to a depth of 1-1.5 feet (Table 9A). Only 4 of the 27 samples analyzed showed concentrations below 1 ppm. The unnamed tributary contained PCB contaminated sediments to a depth of one foot as far as 300 feet downstream from the outlet of the wetland. Less contamination was detected further downstream. The stretch of the Copicut River from Carol's Brook to Cornell Pond, and Carol's Brook southeast of the site did contain PCB contaminated sediments; however, Cornell Pond itself appeared to be free from PCB contamination.
Select sediments near the site were re-monitored between June 1990 and April of 1992. The highest PCB levels were again detected in wetland sediment zero to six inches BLS north of the site at 22 ppm. PCB levels in sediment monitored at this time from the unnamed tributary and the Copicut River were markedly lower than those detected in the same area in 1984. The later monitoring was conducted at depths six to twelve inches BLS whereas monitoring of sediment from these water bodies in 1984 was conducted at zero to six inches BLS. The results are presented in Table 9B. Since completion of sediment remediation at the site in 1994, all sediment concentrations are below 1 ppm PCBs.
The 1987 Supplemental RI also included an evaluation of area aquatic organisms to assess the contamination of fish from water bodies contiguous to the Re-Solve site. Eight fish samples were collected from two sampling stations downstream of the site in the Copicut River and Cornell Pond. Analyses of fish samples indicated PCB concentrations from 0.26 to 1.1 ppm. American eels contained up to 20 ppm PCBs. The action level for PCBs established by the Food and Drug Administration (FDA) is 2 ppm. Up to 4.2 ppm mercury concentrations were also detected in this sampling of area fish. The Action Level for mercury in fish is 1.0 ppm. In 1986, as a result of detection of elevated PCBs in fish collected by the EPA from the Copicut River, the MDPH issued a Public Health Fish Consumption Advisory which recommended that people refrain from eating American eel caught from the Copicut River.
In July 1988 MDEP conducted additional sampling further downstream and at an upstream control site. The sampling also increased the number and types of species sampled. The DEP took samples from the Copicut River, at the Copicut Reservoir as the control site, and at Cornell Pond and Noquochoke Lake along Shingle Island River. The two rivers converge and the Shingle Island River continues for approximately 0.75 miles at which point it meets Noquochoke Lake.
A total of 35 fish samples consisting of eight different species (largemouth bass, chain pickerel, American eel, pumpkinseed, yellow perch, brown bullhead, black crappie, and bluegill) were collected from Cornell Pond and Noquochoke Lake. The fish were filleted and analyzed with skin removed. The average concentration of PCBs detected in fish was low (0.2 ppm) and well within the FDA Action Level for PCBs of 2 ppm (Table 10). The highest PCB concentrations were detected in american eels. The american eels collected from Cornell Pond and Noquochoke Lake had PCB concentrations of 0.79 ppm and 5.3 ppm, respectively. The elevated PCB levels detected in the Noquochoke Lake eel are consistent with concentrations detected in eels collected upstream and closer to the Re-Solve, Inc site. These elevated levels served as the basis for the 1994 update of the 1986 advisory.
Metals detected in fish tissue included aluminum, chromium, copper, iron, lead, mercury, manganese, nickel, and zinc. The concentrations of metals are similar to average concentrations calculated from the statewide database and, with the exception of mercury, do not vary greatly between the two water bodies sampled in this 1988 investigation. Chromium, lead, and zinc levels are higher than levels usually encountered in this state, but they were not cited as contaminants of concern for the Re-Solve Site (MDPH 1994). Mercury was detected in fish at low levels in Cornell Pond (ave. 0.28 ppm) whereas levels in Noquochoke Lake were higher and averaged 0.8 ppm, with a range between 0.31 and 1.4 ppm.
The data in this section are the result of investigations conducted prior to and during soil removal operations which were ongoing at the site in 1986 and 1994. It is possible that contaminant levels in environmental media have since subsided. When more recent monitoring data become available, an addendum to this health assessment may be prepared to characterize the current health hazards associated with the site.
The monitoring for PCB contamination in ground water that was conducted in 1985 was not repeated in 1989. This data is essential to the accurate characterization of the migrating compounds that were detected in the ground water sampling conducted in 1985. The conclusions presented in this Public Health Assessment are based on the data reviewed. The conclusions presented are dependent upon the quality of the data provided.
The entire site, as well as an interior portion containing a pump, is enclosed by fencing with intermittent barbed wire; therefore, there is minimal possibility that trespassers could enter the site and sustain injury. There is currently no remediation on-site and all excavation and treatment equipment has been removed.
To identify possible facilities that could contribute to the air, surface water and soil contamination at the Re-Solve Inc. site, the TRI database was searched. TRI is developed by the USEPA from the chemical release information provided by certain industries. No off-site releases of hazardous material were reported in Dartmouth from 1987 to 1990.
To determine whether nearby residents are exposed to contaminants from the site, the factors influencing human health were evaluated. The pathway analysis consists of five elements: a source of contamination, transport through an environmental medium, a point of exposure, a route of human exposure, and an exposed population.
Pathways are identified as either completed or potential. Completed pathways consist of the five elements and indicate that exposure to a contaminant has occurred in the past, is currently occurring, or will occur in the future. Potential pathways, however, exist when at least one of the five elements is missing. Potential pathways indicate that exposure to a contaminant could have occurred in the past or could occur in the future. Pathways can be eliminated if at least one of the five elements is missing and will never be present.
The extent of public health risk depends not only on the presence of exposure but also on the toxicity of the contaminants and the dosages to which the receptor population is exposed. These two factors are discussed later in the Public Health Implications section.
Private Water Supply-Past
More than 56 different private residential wells surrounding the Re-Solve site have been monitored for contamination since 1981. Early site investigations concluded that groundwater from the site is generally moving to the east and southeast. Furthermore, the Copicut River has characteristics of a hydraulic boundary to groundwater flow. Consequently, the overburden contaminant migration plume discharges, almost entirely, into the Copicut River and the unnamed tributary which is also in the path of the plume. Due to these flow direction patterns, the wells to the southeast of the site are considered to be directly downgradient of the site and the wells to the east of the site are not considered directly downgradient. Figure 2 shows the location of the residential wells evaluated in relation to the site.
Of the wells included in the annual sampling program, PW-2, PW-25, and PW-48 are north and upgradient of the site. In other words, groundwater is generally moving away from these wells towards the site. Several of the wells included in the annual sampling program are located southeast and east of the site, and thus are considered downgradient. This group of wells includes PW-1, PW-14, PW-17, PW-13, and PW-39. Finally, PW-43, PW-9, and PW-26, all included in the annual monitoring, are located southwest of the site and are not downgradient of the site.
Although trichloroethylene was detected in soils, groundwater and surface water on-site, the exact mechanism for the migration to PW-48, if one exists, is not known. Hydrogeologic investigations have provided some possible routes of contaminant migration. Ground water at the site migrates through the overburden at the site from the northwest to the east and southeast. This residence where trichloroethylene was detected; however, is situated 200 feet upgradient and northwest of the majority of on-site contamination disposal. Lateral upgradient migration may be possible because continued disposal of contaminants may temporarily alter hydraulic gradients and subsequent contaminant migration patterns. It is not certain whether this process, called "mounding," occurred at the site.
Nevertheless, the presence of trichloroethylene in this well indicates a route of exposure for those residents who used the well as a water supply. However, exact routes of exposure to trichloroethylene in these wells is uncertain since it is not known if the waters are solely used for laundering and bathing or for drinking and cooking as well. Trichloroethylene, which is a volatile organic compound, can volatilize from the water during bathing or showering due to high temperatures. Subsequently, residents have an increased potential for exposure to trichloroethylene through inhalation.
Contamination levels have subsided in this well. In 1985 trichloroethylene was present in the waters from this well below analytical detection limits. Furthermore, the addition of a water treatment system, in 1993, capable of removing this contaminant has reduced the possibility of exposure via private ground water at this residence. Therefore, current exposure to trichloroethylene through this pathway is not believed to be occurring.
Elevated levels of methylene chloride were measured in waters drawn from a residential well situated in bedrock approximately one mile south of the site (PW-44). It is unlikely that this contamination is the result of methylene chloride emanating from the site because elevated levels were detected in one on-site monitoring well at the northern edge of the site, but not at significant levels south of the well. There is no established contaminant plume from the site to this well. Furthermore, methylene chloride was only detected once during the seasonal series of residential monitorings that were conducted between 1985 and 1986. Subsequent sampling (1992-1994) has only detected trace levels of methylene chloride during 1993 well monitoring. Finally, methylene chloride, a common laboratory contaminant, often appears during the analyses of environmental samples that may not actually contain this compound. It is unlikely that this contamination is a result of methylene chloride emanating from the site, and continuous exposure to methylene chloride may not be ongoing.
In 1985, elevated levels of benzene were detected in two private wells south of the site (PW-26 and PW-30). Neither of these wells is considered downgradient to the site. Nevertheless, the presence of benzene in these waters represents a completed exposure pathway in the past for those residents using the waters drawn from these wells. Subsequent sampling rounds have not detected benzene in either well nor in any of the annual sampling conducted between 1992 and 1994. Therefore, current benzene exposure via the ingestion of contaminated well water is not believed to be occurring.
Significantly elevated lead levels were detected during sampling of residential wells between 1985 and 1986. Some of these wells are located northwest and upgradient of the site and others are situated from 750 to 2500 feet downgradient of the site. Lead is not an established contaminant of concern at this site, and there is no discrete lead contamination plume associated with on-site groundwater; therefore, the lead detected in these wells is not believed to be site related.
While groundwater may not be contaminated with lead, some service pipes to buildings, and some plumbing inside homes contain lead. In older homes, inside plumbing was made of lead or contained lead-based solder because it was inexpensive. Newly built homes contain new services made of iron and steel, but many old lead services continue in use. Lead may then enter drinking water primarily as a result of the corrosion or wearing away of these materials containing lead.
Subsequent sampling of private well water did not include analyses for lead. The presence of lead in the water in the past represents a completed exposure pathway for residents using the contaminated waters. Due to the lack of more recent data, it is unclear whether or not these waters are still contaminated with lead, and whether exposure is still occurring.
In 1985, private well PW-25, located north of the site and private well PW-50, located nearly 1 mile south of the site, contained n-nitrosodiphenylamine. Because n-nitrosodiphenylamine was not detected in any on- or off-site monitoring well at the time of this well water investigation, a definite migration pattern has not been determined. Subsequent sampling through 1994, has not detected n-nitrosodiphenylamine in either of these wells. Therefore, exposure to n-nitrosodiphenylamine is not believed to currently be occurring via ingestion of these waters.
Residential well monitoring between 1985 and 1986 also detected elevated levels of 1,2-dichloroethane in PW-29, located 2,000 feet south east of the Re-Solve site. However, 1,2-dichloroethane was not detected in the remaining rounds of this groundwater investigation nor the annual monitoring program between 1992 and 1994. Therefore, exposure to 1,2-dichloroethane is not believed to be occurring via ingestion of these waters.
The potential for groundwater migration may have been reduced as a result of the extensive soil remediation that has been conducted at the Re-Solve site. During the initial remedial activity through May 1985, the EPA removed 15,000 cubic yards of contaminated lagoon wastes and soils. The Record of Decision for the site that was signed in September, 1987 required the further excavation and treatment of 22,500 cubic yards of PCB contaminated soils and 3,000 cubic yards of PCB contaminated sediments. The chosen treatment system, the X*TRAX process, successfully reduced PCB concentrations in 38,000 cubic yards of on-site soils to less than 25 ppm, and 1,500 cubic yards of sediments in the Northern and Eastern Wetlands were cleaned to a PCB concentration of 1 ppm. Confirmatory soil sampling conducted by the both the remedial contractor and EPA's oversight contractor indicate that all contaminated soils were excavated and properly treated.
The 1987 Record of Decision also included a management of migration component that will address ground water contamination associated with the Re-Solve site. Ground water migrating beyond the waste management area will be treated to meet Maximum Contaminant Levels; however, EPA estimates that it will be necessary to operate this facility for ten years in order to attain these cleanup levels. Finally, the ROD required that the settling parties continue to monitor groundwater for at least three years to ensure that cleanup is complete. Currently, this groundwater remediation is in its design stages. However, this treatment, in conjunction with completed soil remediation, acts to reduce the potential for future contaminant migration through ground water.
Fish--Past and Present
PCB contamination was detected in aquatic organisms caught from surface waters near the site. Because PCBs have been detected in on-site soils, groundwater, and sediments monitored from the site area it is believed that this contamination of fish is site related. Aquatic organisms living within the sediment may take up these PCBs from the environment. Larger fish who consume these plants and plankton accumulate PCBs in their lipids (or fats). These fish are in turn consumed by carnivorous aquatic organisms or humans which will also store PCBs.
The highest levels of PCBs were detected in eels. Dwelling in stream beds, eels are in frequent contact with PCB-contaminated sediment, if present. There they feed on plants and decayed matter also located in sediment. This contact with PCB contaminated environmental media coupled with their high fat content predisposes them to accumulation of PCBs. Subsequently, the levels measured in eels were approximately 10 times those detected in finfish monitored in site-associated waters. It is reported that some individuals do not heed the warnings against fishing posted at Cornell Pond and along the Copicut River (MDPH 1992). These individuals, who may consume the fish they catch in these areas, are at risk of exposure to PCBs via ingestion of contaminated fish.
It cannot currently be determined when eels can again be safely consumed since no remediation currently exists that will have immediate impact on the PCB levels of these organisms. PCB-contaminated sediments of levels greater than 1 ppm were dredged from the wetlands and the unnamed tributary. It is possible that the dredging of these sediments will reduce the exposure of aquatic organisms to PCBs, and the PCB levels in these organisms will subside over time.
In 1988, mercury was detected in fin fish caught upstream from the site at the Copicut Reservoir; however, the mercury levels in fish subsided with increasing distance from the reservoir. Since those fish with the highest levels were detected up-stream of the site, it is unlikely that the low mercury levels detected on-site are impacting those detected in fish caught in site-related waters or the reservoir upstream the site. Because fishing is currently prohibited in reservoir waters, exposure to mercury via ingestion of these contaminated fish is extremely unlikely.
It is possible that recent state regulatory actions may further decrease exposure to contaminants in fish and eels. Based upon the results of the 1988 Copicut/ Shingle Island Rivers Fish Toxics Monitoring, the MDPH updated the advisory for the Copicut River and Cornell Pond in May 1994. This updated advisory was based upon the presence of PCB concentrations in american eels from the Copicut River, and mercury in finfish from Cornell Pond. The same advisory was placed on Noquochoke Lake in May 1994. This advisory recommends:
- Children under 12, pregnant women, and nursing mothers should refrain from consuming any fish from Noquochoke Lake or the Copicut River, including Cornell Pond, in order to prevent exposure of the developing fetuses and young children to mercury and PCBs.
- The general public should refrain from consumption of american eels caught from Noquochoke Lake and the Copicut River, including Cornell Pond, and from the consumption of largemouth bass from Noquochoke Lake.
- The general public should limit consumption of all other Noquochoke Lake fish and largemouth bass from Cornell Pond to two meals per month.
Despite these advisories, exposure to PCBs via ingestion of fish caught in surface waters near the site is, in all likelihood, ongoing since local officials have reported fishing activities in these waters (MDPH 1992). It is not certain, however, whether fish caught from these water bodies are subsequently consumed. Furthermore there is uncertainty whether solely finfish or if other aquatic organisms, such as eels, are also being ingested. Due to the presence of an extensive Portuguese community, among whom eels may be considered a delicacy, it is highly likely that eel are being fished and eaten.
It is likely that in the past on-site workers were exposed to soil bound contaminants through incidental ingestion of and dermal contact with soils, or inhalation of airborne particles. Trespassers on the site in the past may have incurred lower exposures. Currently the likelihood of unauthorized personnel entering the site and becoming exposed to on-site soil contaminants is reduced due to the presence of fencing around the perimeter of the site.
It is reported that tires and disposed solvents were burned in the lagoons between 1956 and 1974 (EPA, 1985). Combustion of PCBs could generate toxic polychlorinated dibenzodioxins and dibenzofurans (PCDD/F). Inhalation of these compounds may have occurred during this time. These compounds were detected at low levels in feed samples monitored during the pilot study of the X*TRAX desorption system.
It is not possible to assess the extent of exposure to surface soils that may have occurred after 1981 since the site was covered with unspecified amounts of material at that time. It is possible that this action may have mitigated exposure to soil contamination at surface level, but it may also have resulted in an underestimate of soil contamination present during monitoring that was conducted after 1981. An emergency removal action in which 15,000 cubic yards of contaminated soil was excavated and transported off-site for disposal was conducted in 1985. As a result of this removal action, contaminants that were at or near the land surface at that time became inaccessible to human contact. Soil contaminants that were well below land surface prior to the excavation; however, with the resultant change in grade, may have become more available for human contact. Elevated PCB levels were still detected in soils monitored after the completion of the initial removal action.
Since 1994, all soil remediation has been completed, and there have been no exposures to on-site PCB or VOC contaminated soils. A significant number of post excavation soil samples were collected on-site above the seasonal low groundwater elevation. All soils unexcavated on-site above the seasonal low groundwater elevation were below the EPA cleanup standard of 25 ppm. As indicated in the Remedial Investigation for this site, soils contaminated with VOCs were located within the soils contaminated with PCBs (i.e., PCBs were widespread throughout the site while VOCs were localized). As a result, PCBs were targeted as driving the clean-up level. Because the PCB contaminated soils were treated through a low thermal desorption process, the VOCs were also removed from the contaminated soils. Therefore, there are no present or future PCB or VOCs exposures to on-site soils.
Groundwater--Present and Future
Contaminants disposed of in the oil-spreading area, the lagoons, and the cooling pond have migrated through on-site media. This is evidenced by the detection of PCBs and VOCs in groundwater from monitoring wells determined to be hydraulically downgradient to these areas. Alone, PCBs are not environmentally mobile due to their low water solubility and their tendency to adhere to soils. These compounds have, however, migrated downward and laterally. PCBs will exhibit altered migration patterns such as these in the presence of VOCs. PCBs are readily soluble in many of the VOCs which were also disposed of at the site. VOCs, in turn are relatively soluble in water. VOCs, therefore, will remove PCBs which have bound to soil and carry them through groundwater. This is occurring on the site as indicated by the detection of PCBs at levels exceeding their water solubility in filtered groundwater samples taken from the site.
Due to their relatively high water solubility, VOCs are readily mobile in groundwater and have been detected in groundwater hydraulically downgradient from the disposal areas on the site. Groundwater monitoring has indicated that these contaminants have migrated both east and southeast in both the bedrock and overburden systems. Soil VOC contamination continues to be a source for groundwater VOC contamination as evidenced by the elevated VOC levels detected in wells drilled where on-site lagoons were once situated.
Comparison between groundwater contaminant levels detected in 1985 and those in 1989 indicates that vinyl chloride was present in some wells in 1989 and not earlier at these wells in 1985. Larger chlorinated compounds such as trichloroethylene and tetrachloroethylene undergo biological breakdown to smaller compounds such as vinyl chloride. Since increased vinyl chloride levels were detected in wells where trichloroethylene and tetrachloroethylene levels decreased over time, it is possible that this breakdown process is occurring in site-related waters.
The majority of on-site overburden groundwater discharges to the unnamed tributary with very little discharge to Carol's Brook at the southern end of the site. This is especially apparent during periods of elevated ground- water flow. During periods of low groundwater flow, hydrologic investigations have demonstrated that on-site groundwater discharges directly to the Copicut River east of the unnamed tributary. Contaminant migration has been observed to follow groundwater flow. For example, elevated VOC levels have been detected in both the unnamed tributary and the Copicut River at points where groundwater discharge is expected to occur. Very little VOC contamination has been detected in overburden water wells drilled east of the Copicut River lending further evidence that the unnamed tributary and the Copicut River intercept most of the VOC contamination migrating via groundwater from the site. Some contamination was, however, detected in waters drawn from a bedrock well drilled east of the Copicut River. It is possible that contamination that has migrated downward to the bedrock groundwater system may be channeled underneath these two surface water bodies and carried further downgradient.
While the fate of contaminants migrating through fissures is uncertain, there is only a small possibility that they could be contaminating private wells drilled in bedrock. Furthermore, there are no residences situated within the groundwater contamination plume associated with the site, which has been determined to flow in an easterly and southeasterly direction. Contaminants, which were detected in on-site monitoring wells and in private wells as far as 2,000 feet downgradient to the site, were not detected in monitoring wells between these points. In addition, private wells near the site were monitored numerous times and none of these wells showed elevated concentrations during these recent investigations.
In residential well water investigations conducted from 1985 - 1986, and subsequent annual monitoring from 1992-1994, organic contamination was detected in some wells within a one mile radius of the site. While the source of the contamination is not certain, exposure to organics detected in these private wells occurred to their users in the past. The addition of treatment systems on four of these wells has likely controlled the exposure of these residents to VOC-contaminated private ground water. Samples at these four residences are taken before they enter the treatment systems. Furthermore, the completion of soil remediation and the imminent implementation of groundwater remediation has significantly limited the potential for exposure to groundwater contaminants in the future. Ingestion of on-site contaminants in ground water could occur in the future, if in the absence of adequate remediation, wells are developed for bathing, laundering, or drinking purposes. Institutional controls are currently in effect prohibiting any use of on-site groundwater until federal and state water quality standards have been met. Finally, monitoring of these waters will be ongoing after the remediation has been deemed complete.
Surface Water--Past, Present, and Future
The highest levels of VOCs have been detected in surface water sampled from both the unnamed tributary and the Copicut River in areas determined to be hydraulically downgradient to those where elevated VOC levels were detected in groundwater. Surface water flow studies also indicate that 95% of the water flow in the unnamed tributary is attributable to groundwater discharge. Surface water VOC contamination has also been shown to increase one thousand fold when the flow in this surface water body doubles. This indicates that groundwater discharge is responsible for both the surface water flow and the levels of VOC contamination detected in these waters. As stated previously, during those periods when ground water flow (and elevation) is low, contaminated groundwater discharges directly to the Copicut River. It is anticipated that the proposed removal of contaminated soil and VOCs from on-site ground water will mitigate the migration of VOCs towards these surface water bodies and lead to their subsequent contamination. While PCB migration in groundwater has been observed, it has not extended as far downgradient as either the unnamed tributary or the Copicut River. It was determined that the PCB contamination that was detected in surface water sampled from these water bodies was the result of sediment contamination.
The unnamed tributary may be accessible from areas south and east of the site. It is, however, not enough to support swimming activities and therefore ingestion of contaminants from the tributary is unlikely. Waders at either the Copicut or the unnamed tributary can sustain exposure to surface water contaminants via dermal contact. It is not certain whether wading activities are actually ongoing in these waters.
Sediments--Past, Present, and Future
Sediments have been contaminated with PCBs either through direct deposition or from surface soil runoff from the lagoons or oil spreading area to the wetlands north of the site. Pooled water, which accounts for only 5% of the water flow in the unnamed tributary, transports PCB contaminated sediments downstream to the unnamed tributary. This is evidenced by the decreasing PCB levels in sediment with increasing distance from the wetlands. Individuals who wade through these contaminated sediments risk PCB exposure via dermal contact. It is uncertain, however, to what extent this is occurring.
The possibility exists that, in the past, individuals on or near the site may have been exposed to VOC contamination emanating from surface soils. The extent of such exposure cannot be assessed in the absence of actual monitoring for these compounds in the areas of concern.
In this section the health risks posed to the public as a result of possible exposure to site contaminants are evaluated. For those contaminants present in environmental media at levels greater than their media specific comparison values, dose calculations were conducted to approximate the dose exposed individuals may have receive(d). The estimated doses are then compared to ATSDR established Minimal Risk Levels (MRL), Lowest Observed Adverse Effect Levels (LOAELs) and No Observable Adverse Effect Levels (NOAELs) and Cancer Slope Factors. An MRL is an estimate of the daily exposure of a human being to a chemical that is likely to be without an appreciable risk of deleterious effects (noncarcinogenic) over a specified period of exposure. For those exposures resulting in significant doses, ATSDR Toxicological Profiles were reviewed to determine what health effects are associated with these exposures.
Available health data pertinent to the site are also presented in this section, and the possible impact of environmental exposure on disease rates is discussed. Finally, citizen concerns specifically voiced to the public health officials are addressed.
Estimates of PCB exposures received via the ingestion of contaminated fish and eel were calculated. The maximum and median contaminant concentrations in fish collected from the 1988 Copicut/Shingle Island Rivers survey were used for these calculations. That is, the maximum and median PCB concentrations detected in fish for each site-related water body evaluated were used in these exposure calculations. Because only one American eel was collected for each of these water bodies, their respective contaminant concentrations were used in a separate exposure calculation representing consumption of eel. For comparison purposes, exposures were calculated for various groups (i.e., recreational fishers, children, and fish eaters) based upon their different fish consumption rates. The exposure factor used in these calculations was based upon a worst case scenario which assumed that individuals have been consuming fish contaminated by this site since the beginning of solvent reclamation activities to the present. This represents a span of 39 years of exposure. Table 11 presents sample calculations.
A review of ATSDR's Toxicological Profile for PCBs was conducted to determine which, if any, health effects are associated with these PCB exposure through ingestion. A series of studies has tried to associate women's consumption of PCB contaminated fish, during and before pregnancy, with adverse neurodevelopmental effects in infants. Some concern, however, has been raised regarding the study limitations that may have obscured the interpretation of the results (ie, co-exposure and lifestyle habits). One such study evaluated the birth weight, length, gestational age, and indices of neurological development in 313 infants. Of these infants, 242 were born to mothers who had consumed moderate to large quantities of Lake Michigan fish sometime during their lives, and 71 were born to mothers who did not consume Lake Michigan fish. Mean fish consumption, estimated by recall, and duration of consumption were 6.7 kg/year and 15.9 years, respectively. The range of median concentrations of PCBs in some cooked Lake Michigan fish have been reported to be 168 ppb to 3,012 ppb (0.168-3.012 ppm) (ATSDR 1992).
Using these consumption parameters, calculations were used to estimate the amount of PCBs these study recipients received. The exposures associated with consumption of contaminated fish and eel from the water bodies in the Re-Solve site area were then compared to those of the Michigan study population. Consumption of PCB contaminated fish from Lake Michigan resulted in exposure estimates from 1.0 x 10-5 mg PCBs/kg/day to 1.8 x 10-4 mg PCBs/kg/day. The exposures associated with consumption of PCB contaminated fish from the Re-Solve area fall within this range of values. Great caution should be taken when comparing such values directly due to the vast number of factors involved in exposure. For example, the samples collected in the Copicut/Shingle Island Rivers were not cooked before analysis, but the Michigan data represents the consumption of cooked fish. Nevertheless, the exposure estimates associated with the Re-Solve site are comparable to the exposure estimates for the participants of this Lake Michigan study.
Analysis of these Lake Michigan infants suggests that fish consumption only during pregnancy does not predict either birth size or gestational age. However, it is related to retardation in neuromuscular development. Further examination of this population found that consumption of contaminated fish was positively correlated with impaired autonomic development, increased number of abnormally weak reflexes, and altered behavioral responses (ATSDR 1992).
Along with this human data, chronic developmental data for PCB exposure in animals include studies in monkeys. These studies suggest that chronic oral exposure to PCBs is associated with low birth weight, a dose-related increase in early abortions, and neurobehavioral deficits. These health outcomes were observed at approximate doses of 0.01 mg/kg/day. Children aged 0-9 years with fish consumption rates in the upper 95th percentile and consuming fish with the 1986 maximum contaminant concentrations are the only receptor group, of those considered in relation to this site, that had an exposure dose equal to or greater than this value (1.8 x 10-2 mg/kg/day). All other groups considered had exposure doses at least an order of magnitude less.
From this review of human and animal data regarding chronic ingestion of PCBs, developmental effects are a possible result of the ingestion of PCB-contaminated fish from water bodies in the Re-Solve site area.
Several immunological parameters were examined in female rhesus monkeys after long-term exposure to Aroclor 1254 at doses ranging from 0.005 to 0.08 mg/kg/day. All doses tested induced a significant and dose-related decrease in antibody levels (IgG and IgM). Based on these changes in immunoglobulin levels, the dose level of 0.005 mg/kg/day is considered the Lowest Observed Adverse Effect Level (LOAEL) for ingestion of PCBs (ATSDR 1992).
From this LOAEL, a chronic oral MRL of 0.00002 mg/kg/day was calculated through the division of the above dose (0.005 mg/kg/day) by an uncertainty factor of 300. This uncertainty factor accounts for the extrapolation from animals to humans, and for the variability of humans. All of the dose estimates, for the various receptor groups considered, representing consumption of PCB contaminated eel collected from the Re-Solve site area, were above this MRL. Furthermore, the majority of the exposure estimates based upon median fish contaminant concentrations from the Re-Solve site area are also above this value. Although this suggests that adverse immunological health effects could be expected from this consumption of contaminated fish, there are several issues that are involved in this analysis that limit the risk. First, these exposure calculations are merely estimates that are based on a worst case scenario. This assumed scenario maximizes exposure estimates, where the actual exposure may be less. Furthermore, the MRL is based on a single study in monkeys that suggests immunological effects due to PCB consumption. These factors combined suggest that there is a limited risk to human health due to the consumption of PCB contaminated fish and eel from Re-Solve area water bodies.
Based upon animal studies, the EPA has classified PCBs as a probable human carcinogen. The estimated past exposures to PCBs in fish should not have resulted in a significant increased risk of cancer associated with the consumption of fish from water bodies near the site.
PCBs were detected at elevated levels in both on-site and off-site soils, and exposure calculations for various receptor groups were conducted. Soil remediation, which was completed in 1994, significantly reduced PCB levels on-site. During remediation the potentially responsible parties also elected to remediate soils along the northern access road and the Reed's property to 1 ppm. Calculations were done to estimate the exposure associated with past ingestion of both on-site and off-site soils.
Trespassers to the site were the first group of receptors considered. For exposure calculations it was assumed that on-site industrial activities would inhibit young children from accessing the site regularly for play. Furthermore, the security and fencing during soil remediation would further prevent access. Therefore, child and adult exposures to on-site soils was considered acute rather than chronic. For all of these calculations the soil ingestion rate was assumed to be 50 mg/day for adults and 75 mg/day for children. The maximum PCB concentrations detected in 1985 and 1990 investigations were used in these calculations; therefore, they represent a worst case scenario. The results of these acute exposure calculations for on-site trespassers are presented in Table 12. The lowest acute ingestion NOAEL based upon an animal study is 0.5 mg/kg/day. The lowest acute ingestion LOAEL, also from an animal study, is 1.0 mg/kg/day. Both adult and child trespassers to the site had approximate exposures well below the acute NOAEL's and LOAEL's for ingestion of PCBs; therefore, there would be no increased risk of health effects associated with past exposures to on-site soils for these groups.
The land encompassing a one-mile semicircle to the north of the site is predominately wooded and wetland areas. It is more likely that these were accessed for play than the actual on-site property. Therefore exposure calculations were conducted for another group of potential receptors who may have accessed these areas. Similar to the above calculations, these exposure estimates are based on a worst case scenario which assumes that PCBs were present in area soils since the beginning of Re-Solve activities until the completion of soil remediation (38 years). The results of these exposure calculations are presented in Table 12. These expsoure estimations are well below the chronic LOAEL for ingestion of PCBs which is 0.005 mg/kg/day. The child's estimated exposure is slightly above the 0.00002 mg/kg/day MRL established for PCBs; however, the extremely conservative assumptions employed in these calculations as well as the determination of the MRL itself do not indicate an increased risk of adverse health effects due to children's ingestion of off-site soils.
On-site Re-Solve workers and remedial workers may have been exposed on a more long-term basis. Re-Solve Co. began site activities in 1956 and discontinued activities in 1980. Therefore, the worst case scenario allows for 24 years of exposure to Re-Solve workers. Due to winter conditions the potential for soil ingestion is not constant throughout the year; however, for these worst case calculations year round exposure was assumed. On-site soil and sediment remediation was conducted for approximately two years at the Re-Solve site. Therefore, this second group of on-site workers had a different set of exposure parameters. The results of these chronic exposure calculations are also presented in Table 12.
The exposures to Re-Solve and on-site remedial workers are below the chronic LOAEL for ingestion of PCBs which is 0.005 mg/kg/day. The approximate exposure for Re-Solve workers is 6.3 x 10-4 mg/kg/day and for on-site remedial workers is 1.3 x 10-4 mg/kg/day. The approximate exposure for Re-Solve workers is slightly above ATSDR's MRL. However, all expsoures above MRL's do not result in health effects and this calculation represents a worst case scenario because extensive soil contamination was not present since the first day of Re-Solve solvent reclamation activities, and it is unlikely that many workers were employed by Re-Solve for the entire 24 years of site activities. Therefore, for Re-Solve workers, past exposure to on-site soils should not result in an increased risk of health effects. Because the estimated exposure for on-site remedial workers is even less than that for Re-Solve workers one would also not expect an increased risk of health effects for this group based upon their ingestion of PCB contaminated soils.
Recent conditions at the site significantly limit exposure via ingestion of contaminated soils. The site was covered with 18 inches of crushed stone, and with a fence for some time thus significantly limiting access to the site for both adults and children. Furthermore, the level of PCBs in soils has been successfully decreased by soil remediation, both on-site and areas off-site, which was completed in 1994.
It is uncertain whether dermal contact with PCB-contaminated sediments occurred in the past, or is currently on-going. It was possible for an individual to access the wetlands and unnamed tributary where PCB levels in sediments were highest. A rash-like condition known as chloracne has been associated with individuals who have come in dermal contact with PCBs (ATSDR 1992). Currently, completed sediment remediation has significantly decreased the level of PCBs in sediment to less than 1 ppm; therefore limiting the potential for exposure to PCB-contaminated sediments.
3. Drinking Water
Estimates of exposure via ingestion of contaminated well water were also conducted. ATSDR Toxicological Profiles and other sources were used for benzene, lead, methylene chloride, n-nitrosodiphenylamine, 1,2-dichloroethane, and trichloroethylene. These specific contaminants were evaluated because they were measured at levels greater than their respective comparison values (i.e., MCL, RMEG, and CREG). Exposure calculations were conducted using the following criteria: child and adult daily intake of water was assumed 1 and 2 L/day respectively, and child and adult body weight was assumed 10 and 70 kg respectively. The exposure factor was calculated using a reasonable worst case scenario which assumes that contaminants were present in these residential wells from the beginning of on-site activities until the point at which they were last detected. For all of the contaminants considered in this section, except for lead, this exposure duration was assumed to be 30 years (1956-1986). Because lead is not considered site-related and may still be present, a life-time exposure was assumed in these dose calculations.
It is unknown whether the residents used this water source for ingestion. Because this site is rural and there are no municipal water supply lines within reach, it is highly probable that the residents used this well as their only water source (Dartmouth BOH 1995).
Exposure to benzene may have occurred in the past due to its presence, at elevated levels (5 and 9 ppb), in two residential wells in close to the site (PW-30 and PW-26). These levels were detected during multiple rounds of EPA sampling conducted from November 1985 through November 1986. During these three rounds, benzene was not consistently present in either well. Furthermore, during sampling of well PW-30, one sample contained 5 ppb benzene while in the duplicate sample only a trace of benzene was identified. Benzene was not detected in subsequent residential well sampling between 1992 and 1994, however these specific wells were not included in these investigations.
Based upon these past benzene levels in well water, exposure calculations estimate that adults received an exposure of 1.1 x 10-4 mg/kg/day and children received an exposure of 3.9 x 10-4 mg/kg/day. There is limited information available regarding the non-cancer health effects associated with ingestion of benzene, and there are no ATSDR MRLs available for this contaminant. EPA has classified benzene as a human carcinogen (ATSDR 1993a). However, past exposures is not expected to have resulted in a significant increased risk of cancer. Recent monitoring has not detected benzene; therefore current exposure to benzene via ingestion of residential well water is not a health concern.
The November 1985 residential well monitoring of PW-29 detected 6 ppb of 1,2-dichloroethane. Exposure calculations estimated an exposure of 7.4×10-5 mg/kg/day for adults and 2.6×10 -4 for children. These exposures are well below the NOAELs for hepatic, renal, and reproductive effects; all of which are based on animal studies. There is no chronic oral MRL for 1,2-dichloroethane; however, the EPA has classified it as a class B2 carcinogen. Therefore, 1,2-dichloroethane is a probable human carcinogen. Nevertheless, the exposures associated with this site are not likely to have resulted in a significant increased risk of cancer due to the past ingestion of 1,2-dichloroethane contaminated well water. Based upon recent well monitoring, 1,2-dichloroethane is no longer present in this well. Therefore, present exposure to this contaminant via ingestion of residential well water is not a health concern.
Lead is not a contaminant of concern for this site, and there is no discrete contaminant plume associated with on-site groundwater. However, exposure to lead may have occurred via the ingestion of water from wells PW-1, 2, 3, 4, 7, 10, 11, 26, and 29 which contained lead concentrations from 22 ppb to 241 ppb in November 1985. Because a completed exposure pathway exists for these residents some further consideration of possible health effects is warranted.
The general population, including children, is exposed to lead primarily through the oral route, such as the ingestion of contaminated soils, water, dusts, and paint chips (ATSDR 1989). The discussion of lead in this health assessment focuses on ingestion of lead contaminated well water; however, these other routes of lead exposure also contribute to the total amount of lead absorbed and stored in the body. Therefore, the amount of lead residents are receiving from contaminated well water, represents only a small portion of their total lead exposure.
Lead has been shown to affect virtually every organ and/or system in the body in both humans and animals. The most sensitive target organs of lead appear to be the nervous system (particularly in children), the hematopoietic system, and the cardiovascular system. Lead has also been shown to be carcinogenic in animals. No MRLs have been developed for lead because a threshold has not yet been defined for the most sensitive effects of lead (i.e., neurotoxicity).
Although the database concerning lead-induced health effects contains a great deal of human data, the exposure data are generally expressed in terms of absorbed dose, measured as levels of lead in the blood. Furthermore, dose-effect data in terms of external exposure levels (such as that used in this health assessment), or milligrams per kilogram per day doses of lead by a single route of exposure are not generally available for humans.
The lack of a clear threshold for health effects and the need to consider multiple routes of exposure makes evaluating the risks from exposure to lead in the environment very difficult. In addition, variable factors such as an individual's absorption potential of lead, and age and nutritional status prevent the development of generic guidance that can be applied to all populations. Numerous studies have attempted to correlate environmental lead levels with blood lead levels. Slope factors have been calculated which attempt to predict increases in blood lead (µg/dL) per unit lead concentration in environmental media.
Studies correlating water concentration and blood lead are difficult to compare due to the wide range of water lead concentrations in the studies. However, the most reliable slope factors for adults estimate a blood lead level of 0.06 µg/dL per µg lead/L water. Similar data are also available for infants however there are two slope factors: 0.26 µg/dL blood per µg/L water at water lead levels below 15 µg/L (15 ppb) and 0.04 µg/dL blood per µg/L water at water lead levels above 15 µg/L (15 ppb).
Using these conversion factors, blood lead levels can be approximated for residents affected by contaminated well water. The maximum lead concentration detected in residential well water from the Re-Solve site area was 241 µg/L. The blood lead level that corresponds with adult ingestion of these waters is 14.5 µg/dL, and child ingestion is related to 9.6 µg/dL. These blood lead levels are only a portion of the total lead body burden; therefore, the actual blood lead levels are likely to be higher.
In March 1986 a drinking water sample was taken at PW-44 and 180 ppb of methylene chloride was found. Exposure calculations determined that the adult exposure was 2.2 x 10-3 mg/kg/day and the child exposure was 7.7 x 10-3 mg/kg/day. No studies are available regarding respiratory, cardiovascular, gastrointestinal, musculoskeletal, or dermal/ocular effects in humans or animals following oral exposure to methylene chloride. However, hematological, hepatic and renal effects have been observed in rats and mice after oral exposure to methylene chloride (ATSDR 1993b).
Chronic ingestion of methylene chloride in drinking water has been associated with histological alterations of the liver of rats exposed to very high dose levels. Such changes were not observed after exposure to levels of 6 mg/kg/day. The chronic MRL was calculated using the results of this study. The division of this dose by an uncertainty factor, which accounts for the extrapolation from animals to humans and the variability of humans, resulted in a conservative chronic oral MRL of 0.06 mg/kg/day. The above estimates of dose, based upon a worst case scenario, are well below this conservative comparison value. Therefore, there is not an increased risk of non-cancer health effects associated with this past exposure to methylene chloride via the ingestion of contaminated well water. Recent monitoring has not detected elevated levels of methylene chloride; therefore, there is no present health concern due to methylene chloride contaminated well water.
Animal studies provide suggestive evidence that ingestion of methylene chloride may increase the risk of liver cancer. The EPA reviewed the available data on the carcinogenic effects of methylene chloride and concluded that there was borderline evidence for carcinogenicity and classified methylene chloride as a probable human carcinogen. Past exposures that may have been associated with the ingestion of contaminated residential well water are not expected to have resulted in a significant increased risk of cancer. Follow-up well monitoring conducted in November 1986, September 1993, and September 1994 detected trace levels of methylene chloride. However, the consulting laboratory found this compound in its control samples (ENSR 1994). Therefore, the presence of trace methylene chloride levels in these subsequent field studies is believed to have been the result of laboratory contamination. There is no current health concern associated with the ingestion of methylene chloride via residential well water.
Field sampling of residential wells PW-25 and PW-50 detected n-nitrosodiphenylamine levels of 27 and 31 ppb respectively. Exposure calculations estimate an adult exposure of 3.8 x 10-4 mg/kg/day and a child expsoure of 1.3 x 10-3 mg/kg/day. No studies are available regarding systemic effects in humans after oral exposure to N-nitrosodiphenylamine; therefore only animal data is available for consideration of health effects associated with chronic oral exposure. From these available animal studies, several NOAELs and LOAELs, above 100 mg/kg/day doses, were established. The estimates of n-nitrosodiphenylamine dose are considerably below these comparison levels; therefore, an increased risk of non-cancer adverse effects would not be likely due to the past ingestion of N-nitrosodiphenylamine contaminated well water.
No studies are available regarding cancer in humans after oral exposure to N-nitrosodiphenylamine. However, the EPA has classified this compound as a probable human carcinogen based upon available animal data. Nonetheless, it is unlikely that a significant increase in the risk of cancer will occur from this past exposure.
Subsequent monitoring did not detect N-nitrosodiphenylamine in this well; therefore, current exposure to this compound via contaminated well water is not believed to be occurring.
Elevated levels of trichloroethylene (6.1 ppb) were detected in the private well of a residence situated to the immediate northwest of the site (PW-48) when monitored in 1983; therefore, past exposure to trichloroethylene may have occurred to residents served by this well.
Calculations estimate that the adult expsoure of trichloroethylene was 7.5 x 10-5 mg/kg/day and the child expsoure was 2.6 x 10-4 mg/kg/day. Animal studies indicate adverse hepatic, renal, and neurological effects from chronic oral exposure to trichloroethylene (ATSDR 1993c). However, the LOAELs and NOAELs associated with these studies are well above the exposure through Re-Solve area well water. Therefore, there is no increased risk of non-cancer health effects associated with these past exposures to trichloroethylene via ingestion of contaminated well water.
Subsequent samples taken in three rounds from November 1985 to November 1986 did not detect trichloroethylene. In a September 1992 well monitoring, trichloroethylene was found in well PW-48 and at a concentration below the level of concern (0.6 ppb). TCE was not detected in the subsequent September 1994 monitoring. Furthermore, in 1993 the Re-Solve Site Group installed a water treatment system at this residence that is capable of removing this compound. The system is still in operation and maintained regularly. Therefore, current exposure to trichloroethylene via ingestion of contaminated well water is not a health concern.
The link between oral exposure to trichloroethylene and the incidence of cancer in humans is controversial. Furthermore, trichloroethylene is under consideration by the EPA for classification as either a probable or possible human carcinogen based upon available animal studies. However, it is not likely that a significant increase in the risk of cancer would have occurred due to past exposure to trichloroethene in residential wells.
Other Onsite Groundwater Contaminants
There were elevated levels of contaminants detected on the site to which the surrounding population is not currently exposed. If land usage in the area changes in the future (e.g., increased site accessibility or removal of institutional control on ground water use) health hazards could exist as a result of exposure to site contaminants. However, since remediation of ground water is proposed, the likelihood of future exposure to contaminants contained in on-site ground water is unlikely.
Cadmium was detected in on-site ground water at elevated levels. Human studies have demonstrated the possibility of kidney dysfunction development with regular ingestion of cadmium levels detected in on-site ground water (ATSDR 1991). Vinyl chloride, which has been detected in on-site ground water, has been established as a human carcinogen. Occupational studies have shown an elevated rate of hepatic angiosarcoma in workers who are regularly exposed to vinyl chloride vapors. In addition, an increased number of liver tumors were observed among laboratory animals exposed to vinyl chloride via ingestion. Ingestion of vinyl chloride contaminated water at the levels found in on-site ground water has also been associated with increased risk of liver dysfunction in animal studies (ATSDR 1989). Trans-1,2 dichloroethylene was detected in on-site ground water at levels which, when regularly ingested, have been associated with a reduction in red blood cell count (ATSDR, 1994). Also regular ingestion of on-site methyl ethyl ketone levels, according to animal studies, may modify liver structure and function (U.S. EPA 1987b). Hazardous exposure to vinyl chloride, cadmium, t-1,2-DCE or methyl ethyl ketone is not currently on-going at the site. The probability of future exposure to these contaminants will be reduced when plans to remediate on-site ground water are completed.
1. Cancer Incidence
As part of the Re-Solve Health Assessment, the MDPH Community Assessment Unit (CAU) investigated cancer incidence in Census Tract 6531 in the town of Dartmouth. Standardized Incidence Ratios (SIRs) were calculated for the nine-year period 1982-1990, for cancers of the bladder, kidney, leukemia, and liver. Cancers were selected for investigation based upon contaminants of concern. Contaminants of concern included polychlorinated biphenyls, vinyl chloride, trichloroethylene, benzene, and methylene chloride.
For this assessment, emphasis was placed on census tract 6531 which includes the Re-Solve Superfund site (Figure 3).
Cancer incidence data were obtained from the Massachusetts Cancer Registry of the MDPH, Bureau of Health Statistics, Research and Evaluation. In order to determine whether an elevated rate of cancer exists in census tract 6531 in Dartmouth, cancer incidence data were adjusted by age and sex and analyzed to compare the actual (or observed) number of cancer cases to the number that would have been expected based on the statewide cancer incidence experience.
Reliable population data is necessary to calculate incidence rates. The population figures for the state of Massachusetts and census tract 6531 in Dartmouth were obtained by comparing the 1980 and 1990 federal census and calculating the percentage of increase or decrease in total population. The percentage of change was applied to the intercensal year through straight line interpolation. For the purpose of consistency and because of the gap between census years, the assumption is made that change occurs at a constant rate throughout the ten year period.
An SIR is an estimate of the occurrence of disease in a population in relation to what might be expected if the population had the same cancer experience as some larger population designated as "normal" or average. This normal population is usually selected to be the state as a whole. Specifically, an SIR is the ratio of the observed number of cancer incident cases to the expected number of cases. An SIR equal to 100 indicates that the number of cancer cases occurring in the population being evaluated is equal to the number of cases expected in a "normal" population. An SIR greater than 100 indicates that more cancer cases occurred than expected; and an SIR less than 100 means that fewer cases occurred than expected. Accordingly, an SIR of 150 is interpreted as an excess of 50% of the cases over the expected number; and an SIR of 90 indicates 10% fewer cases than expected.
Caution should be exercised, however, when interpreting an SIR. The interpretation of an SIR depends on both size and stability. Two SIRs can be the same size, but not have the same stability. An SIR of 150 based on two expected cases and three observed cases indicates a 50% excess of cancer. The excess is attributed to only one excess case, and that case may have occurred by chance alone. Conversely, an SIR of 150 based on 200 expected cases and 300 observed cases show the same 50% excess in cancer, but because the SIR is based on a greater number of cases, the estimate is more stable. It is unlikely that 100 excess cases would occur by chance alone.
In order to determine if the observed number of cases is significantly different from the expected number, or if the difference may be due to chance, a 95% confidence interval is calculated. A 95% confidence interval is the range of estimated SIR values that has a 95% probability of including the true SIR for the population. If the range does not include the value 100, that means the study population is significantly different from the "normal" population. If the range excludes 100, and the observed SIR is greater than 100, there is a significant excess in the number of cancers. Similarly, if an observed SIR is less than 100, and the range excludes 100, the number of cancer cases is significantly lower than expected. If the range includes 100, that means the true SIR may be 100, and it cannot be concluded, with sufficient confidence, that the observed SIR reflects a true cancer excess (or deficit). Statistical significance is not assessed if fewer than five cancer cases are observed.
Place of residence at the time of diagnosis was examined for each cancer case (by primary site) to assess any possible geographic clustering of cases. In addition, the available occupational information and smoking status of individuals diagnosed with specific cancer types was also evaluated. This evaluation was accomplished by using existing cancer registry data.
For males and females combined, cancer incidence in census tract 6531 in Dartmouth was about what would be expected based upon the statewide cancer incidence experience. These data are presented in Table 14.
Census Tract Analysis
Bladder cancer in males and females combined occurred about as expected (14 observed versus 13.7 expected). The incidence of male bladder cancer was less than expected (8 observed versus 10.4 expected). An elevation was observed among females for bladder cancer (6 observed versus 3.6 expected). This elevation, however, was due to an excess of approximately two cases and was not statistically significant.
Kidney cancer among males and females combined in census tract 6531 occurred slightly more than expected (8 observed versus 6.9 expected). This elevation was however, based on an excess of approximately one case and was not statistically significant. The number of kidney cancer cases among males was about as expected, (4 observed versus 4.3 expected). A slight elevation was observed among females (4 observed versus 2.7 expected). This elevation is based on less than two excess cases and is not statistically significant.
Analysis of leukemia incidence for census tract 6531 for males and females overall revealed that leukemia occurred slightly more often than expected (7 observed versus 5.9 expected). This elevation was based on one excess case and was not statistically significant. Six of the seven cases occurred among males (6 observed versus 3.5 expected) indicating an elevation of slightly more than two cases and is not statistically significant.
Liver cancer for males and females overall occurred at approximately the expected rate (2 observed versus 1.4 expected). Both cases were observed among males (2 observed versus 1.1 expected).
Place of residence at the time of diagnosis was examined for individual cases of bladder, kidney, leukemia, and liver cancers in Dartmouth census tract 6531 which contains the Re-Solve Site. Analysis of the geographic distribution of cancer incidence revealed that cancer cases were by and large evenly distributed within the census tract relative to the center of population. Bladder cancer cases, however, appeared to be concentrated in an area of CT 6531 not located near the Re-Solve site.
Smoking status at the time of diagnosis was known for only 16 of the 31 (52%) cancer cases (bladder, kidney, leukemia, liver) diagnosed in Dartmouth CT 6531 for 1982-1990. Five of these 16 cases (31%) reported that they smoked at the time of diagnosis. Of the 14 bladder cancer cases reported, 36% (n=5) reported an unknown smoking status, 29% (n=4) reported being a smoker at the time of diagnosis, and 36% (n=5) reported as never having smoked. Among the eight kidney cancer cases, 63% (n=5) reported an unknown smoking status, one case reported being a current smoker, and two cases reported as being former smokers. Of the seven leukemia cases, 43% (n=3) reported as never having smoked, and 57% (n=4) had an unknown smoking status. Liver cancer cases comprised only 8.7% of the total cancer cases reported (n=2). One case reported as never having smoked, and the smoking status of the other case was unknown.
The lack of occupational information did not allow for a determination of what role, if any, occupation may have played in the development of these cancers.
Cancers, in general, have a variety of associated risk factors which are assumed to be related to the etiology (development) of the disease. Many cancers are believed to be related largely to life-style factors such as cigarette smoking, diet, and alcohol consumption. Epidemiologic studies of humans and laboratory animals have related several cancers to adverse chemical exposures in the work place or in an individual's environment. Other factors associated with cancer are socioeconomic status, heredity/genetics, race, and geography. Review of scientific and medical literature provides known or assumed risk factors for the cancers of interest.
It is difficult to determine the potential for a relationship of the cancers observed with environmental factors since such a small number of cases was available for evaluation. Further information should be evaluated relative to occupation, smoking status, and residential history for the bladder cancer cases that appear to be grouped away from the site.
2. Childhood Lead Poisoning Prevalence
Data on childhood lead poisoning is maintained by the Childhood Lead Poisoning Prevention Program within the MDPH. Data for fiscal years 1991 and 1993 were reviewed. In 1991, 1,061 children between the ages of 6 months and six years of age were screened for lead poisoning in Dartmouth. A lead poisoned child is one whose blood lead levels are 25 µg/dL or greater (CLPPP 1995). No cases of childhood lead poisoning were observed during this time period. One case of childhood lead poisoning was observed among the 1,050 children from Dartmouth screened in 1993. These rates are well below those observed for the entire state of 3.0 and 3.1 cases per thousand children screened for fiscal years 1991 and 1993 respectively.
We have addressed each of the community health concerns as follows:
The Local Board of Health raised concerns for the safety of residents nearest to the site.
Limited environmental monitoring has been conducted at the residences nearest the site, thus limiting the determination of whether exposure to site-related contaminants is on-going. Hydrogeologic investigations conducted on-site have suggested that the residences north and west of the site may be hydraulically upgradient to the site and the possibility of migration of on-site contaminants toward these residences may be small. In 1983, the waters drawn from the private well (PW-48) serving the residence immediately west of the site were found to contain trichloroethylene, which was also detected on-site. This residence was also determined to be situated hydraulically upgradient to the site and a mechanism for migration of on-site contaminants towards this residence could not be firmly established. The installation of water treatment systems by the Re-Solve Site Group at four of the closest residences may have, in all likelihood, however, reduced these residents' potential for exposure to any VOC contamination via private well water. Considering the short distance of some of these residences from the site, regular environmental monitoring has been recommended. Some of these nearby residents have been regularly monitored by either the EPA and/or the settling parties. Furthermore, as part of the MOM, 16 residential wells will be monitored annually beginning in September 1995.
Concerns about the safety of past and proposed remediation were posed by the local environmental group.
It is believed that no hazardous emissions emanated from the thermal extraction unit at the site. The soils were heated in a closed and oxygen-reduced unit so that little, if any, hazardous by-products of PCB combustion were formed. Gases, formed by the indirect heating of soils, were cooled to a temperature at which they condensed to a liquid. At all times the products of thermal extraction were contained in closed chambers until they were removed to an approved, off-site disposal facility.
Risks associated with ingestion of fish caught from Cornell Pond and Copicut River near the site.
Elevated levels of PCBs have been detected in eels caught down stream of Cornell Pond. These PCB levels exceeded FDA tolerance levels. In addition, PCBs have been detected in other species of fish caught from site-associated surface waters. Although there is little data available for the effects of oral exposure to PCBs on humans, animal studies have shown an association between regular ingestion of PCBs and possible immune system dysfunction. Human studies have not yet demonstrated these associations (ATSDR, 1992). Studies of developmental toxicity in humans exposed to PCBs suggest that maternal PCB consumption of contaminated fish is positively correlated with lower birth weight, smaller head circumference, and shorter gestation. In 1994, the Massachusetts Department of Public Health updated a 1986 advisory recommending that residents not eat eels from the Copicut River due to the elevated PCB levels indicating that:
- Children under 12, pregnant women, and nursing mothers should refrain from consuming any fish from Noquochoke Lake or the Copicut River, including Cornell Pond, in order to prevent exposure of the developing fetuses and young children to mercury and PCBs.
- The general public should refrain from the consumption of American eels caught from Noquochoke Lake and the Copicut River, including Cornell Pond, and from the consumption of largemouth bass from Noquochoke Lake.
- The general public should limit consumption of all other Noquochoke Lake fish and largemouth bass from Cornell Pond to two meals per month.