PUBLIC HEALTH ASSESSMENT
NEW MARLBOROUGH, BERKSHIRE COUNTY, MASSACHUSETTS
The Hatchery Road site in Hartsville, Massachusetts was the site of a release of gasoline to soil and groundwater on the property of a landscaping and excavation company. Former tenants of 8 Hatchery Road, whose residence was downgradient of the release, requested an investigation into the contamination of their drinking water and possible health risks resulting from drinking their contaminated well water.
The Massachusetts Department of Public Health (MDPH) has a cooperative agreement with the Agency for Toxic Substances and Disease Registry (ATSDR) to conduct Public Health Assessments (PHAs) at sites in Massachusetts. For this PHA, environmental sampling data from October 1998 (when sampling first began at the site) to June 2000 have been reviewed.
The Hatchery Road site is located within the village of Hartsville in the Town of New Marlborough, at the intersection of Route 57 (Hartsville/New Marlborough Road), Hatchery Road, and Hartsville Road (MT 1999a). Hartsville Village is a residential community with single and multi-family homes. According to New Marlborough assessor's map, the Konkapot River in New Marlborough runs east-west and is at the north of Hatchery Road. Figures 1 and 2 show the site location.
On October 16, 1998, the Massachusetts Department of Environmental Protection (MA DEP) was notified of a gasoline release at the Mount Everett Landscaping and Excavating Company (MT 1999a). On October 22, 1998, MA DEP sent a letter to the owner of the Mount Everett property notifying him of his responsibilities as a potentially responsible party (PRP) for the Hatchery Road site (MA DEP 1998, MT 1999a). An environmental consulting firm was hired by the Mount Everett property owner to address the release, determine the extent of contamination and conduct remedial activities (MT 1999a).
The landscaping company's building was constructed about 25 years ago (approximately in 1973) (MT 1999a). According to representatives from MA DEP and the environmental consultants, preliminary evaluations indicate that the gasoline spill may have happened sometime in 1997 or 1998 (MT 1999a, MA DEP 1999). One soil sample and three groundwater samples were collected at the site in the area where the gasoline release might have happened and were sent to a laboratory to age date the type of material present. The laboratory report indicated that the release occurred within two years prior to December 1998 when the carbon treatment system was first installed for 1553 Hartsville/New Marlborough Road and 1368 and 1352 Hartsville/Mill River Road (MT 1999a).
Following the initial environmental sampling to identify the area and extent of groundwater contamination, a series of immediate response actions (IRAs) were conducted to prevent further contamination and potential exposure. These activities included intensive water sampling and installation of carbon filtration systems, on-site treatment of the contaminated groundwater and provisions for an alternative water supply for residents whose wells were contaminated. In August 2001 a response action outcome completion report (RAO) was filed with the MA DEP, signifying that levels of oil or hazardous materials were below Massachusetts' guidelines and standards (MT 2001).
Household water sampling initially focused on the following downgradient properties near the spill: (1)
- 8 Hatchery Road
- 1534 Hartsville/New Marlborough Road (house and garage)
- 1352 Hartsville/Mill River Road
- 11 Hatchery Road
Subsequently, two additional downgradient properties were identified for further evaluation:
- 1553 Hartsville/New Marlborough Road
- 1368 Hartsville/Mill River Road
Figure 2 shows the locations of these properties.
Water sampling from private wells on the above properties, starting in October 1998, showed that the drinking water supply wells were contaminated with gasoline constituents. Residents for 8 Hatchery Road, 1534 and 1553 Hartsville/New Marlborough Road, and 1352 and 1368 Hartsville/Mill River Road were advised to use bottled water as an alternative water source shortly after the results came back for these samples (MA DEP 1999).
- The private well at 8 Hatchery Road was sampled twice in October 1998 and four times in November 1998.
- The original well 1352 was sampled once in October 1998 and weekly from November 1998 to May 1999 and in July 1999. In May 1999, a bedrock well was installed at 1352 Hartsville/Mill River Road and was sampled three times in June 1999 and monthly from July 1999 to January 2000 except for September 1999 and December 1999.
- Well 1368 was sampled almost every week from November 1998 to July 1999 and an additional eight times from July 1999 to February 2000.
- Well 1553 was sampled weekly from November 1998 to July 1999 and an additional 17 times from July 1999 to March 2000.
- Wells at 1534 garage and house were sampled once in October 1998 and December 1998; twice in January 1999; weekly from February 1999 to July 1999; biweekly from July 1999 to December 1999; and monthly from January to March 2000.
- Private drinking water wells from approximately 13 other houses in the area (also located downgradient of the source) were sampled either once in November 1998 or monthly from November 1998 to January 2000 except for August 1999 and December 1999.
Results from these samples will be discussed in the "Environmental Contamination and Other Hazards" section.
In late 1998 and early 1999, activated carbon treatment systems, also called point of entry treatment systems, were installed at private drinking water wells of the downgradient residences that have been sampled since October 1998:
- 1553 Hartsville/New Marlborough Road; 1368 and 1352 Hartsville/Mill River Road on December 2, 1998;
- 1534H (house) Hartsville/New Marlborough Road on January 27, 1999; and
- 1534G (garage) Hartsville/New Marlborough Road on February 18, 1999.
To address groundwater impact in the immediate area of the water supply well at 8 Hatchery Road, the environmental consultant designed and, with the approval of MA DEP, installed a groundwater treatment system on-site. The treatment system was completed on December 4, 1998 and began operation on December 7, 1998. With this treatment system, the treated groundwater was discharged to a leaching galley upgradient of the well and adjacent to the landscaping building (MT 1999e).
The landscaping company's building includes a main building and a smaller shed-style addition at the rear. The rear addition consists of the boiler room/restroom and a storage area. The building was first used as a horse stable, then as a retail farm equipment operation, and finally as a maintenance shop for the landscaping company. Besides landscaping, the company business includes general excavation, septic system installation, maintenance services, and underground storage tank removal (MT 1999a).
The landscaping company used three storage tanks on the site: one 300-gallon aboveground tank storing gasoline, one 1,000-gallon underground tank storing diesel fuel, and one 550-gallon underground tank storing No.2 heating oil. These tanks were installed shortly after the construction of the building. The gasoline tank and the No. 2 heating oil tank were installed at the southwest corner of the landscaping building while the diesel tank was installed at the northwest corner of the building (see Figure 3).
Currently, all three tanks have been removed: the gasoline tank in September 1998, the No. 2 heating oil tank in November 1998, and the diesel tank in January 1999. When the gasoline tank was removed, a new 500-gallon aboveground gasoline storage tank was installed along the southerly wall of the salt shed.
There are two commercial buildings in Hartsville in addition to the landscaping company. A vacant auto body shop is located immediately to the west of the intersection of Route 57 and Hatchery Road. The second commercial property, where high performance racecars were built, is located in the garage of the 1534 Hartsville/New Marlborough property (MT 1999a). See Figure 2 for locations of these properties. There are some residences located between the site and the Konkapot River.
MDPH staff visited the site on May 11, 1999, with representatives of MA DEP, Mount Everett Landscaping and Excavation Company and the environmental consultant. In addition to being briefed by MA DEP and the environmental consultant about investigative and remedial work occurring at the site, MDPH also observed the area for evaluation of potential exposure opportunities. At the time of the site visit, except for the rental property at 8 Hatchery Road and the auto body shop, all other properties were occupied. During the site visit, MDPH staff did not observe any evidence of activities taking place at the site by children. Also during the site visit, some heavy equipment as well as excavated soil and drums containing contaminated groundwater awaiting treatment were observed on the landscaping company property. Although there was no fence or locked gate surrounding these equipment and soil pile and drums, there was no evidence of trespassing on the landscaping company property.
MDPH also conducted a community meeting at Hartsville on August 24, 1999. The purpose of this meeting was to gather additional site information and to elicit health and/or environmental concerns from the community members to address in this PHA.
According to MA DEP, at the time of this health assessment, the final site clean-up actions have been implemented, all sources of contamination have been eliminated, all levels are below drinking water standards, and no further action is necessary.
Hartsville is a village within the town of New Marlborough. Approximately 170 people live within one-half mile of the site (MT 1999e). Table 1 contains demographic data for the town of New Marlborough (U.S. Census 2000).
Residents expressed concerns about the possible health effects from opportunities for exposure to contaminants in their drinking water. Health complaints included potential long-term effects (e.g., cancer) from past opportunities for exposures. This PHA will evaluate available cancer incidence data for the town of New Marlborough. Cancer types to be evaluated will include those that have been associated in the scientific literature with the contaminants of concern, which are identified in the following sections. These concerns will be further discussed in section C of the "Discussion". An explanation of methods of determining cancer incidence rates can be found in Appendix A.
To evaluate whether a site poses an existing or potential hazard to an exposed or potentially exposed population, health assessors review all available on-site and off-site environmental contamination data for all media (e.g., soil, surface water, groundwater, air). The quality of the environmental data is discussed in the Quality Assurance and Quality Control section. Physical conditions of the contaminant sources and physical hazards, if any, are discussed in the Physical and Other Hazards section. A plain language glossary of environmental health terms can be found at the end of this document (see Appendix B).
On October 16, 1998, MA DEP was notified of a release at the Mount Everett Landscaping and Excavation Company (MT 1999a). Following notification, MA DEP approved immediate response actions (IRAs) at the site, including the sampling and analysis of water from nearby private drinking water wells, assessment activities such as the installation of soil borings and monitoring wells, and the collection of soil and groundwater samples to determine the source of contamination (MT 1999a). Drinking water, groundwater, indoor air, subsurface soil, and surface waters were sampled as part of the above investigations.
Health assessors use a variety of health-based screening values, called comparison values, to help decide whether compounds detected at a site might need further evaluation. Comparison values include environmental media evaluation guides (EMEG), reference dose media evaluation guides (RMEG), and cancer risk evaluation guides (CREG). These comparison values have been scientifically peer reviewed or were derived from scientifically peer-reviewed values and published by ATSDR and/or EPA. Health assessors also use the maximum contaminant levels (MCLs) for drinking water established by the EPA (or MMCLs established by MA DEP) as comparison values. Under the Safe Drinking Water Act, MCLs are protective of public health "to the extent feasible," i.e., they are established with consideration for public health, economic impact and technological feasibility of implementation.
If the concentration of a compound exceeds its comparison value, adverse health effects are not necessarily expected. Rather, these comparison values help in selecting compounds for further consideration. For example, if the concentration of a chemical in a medium (e.g., soil) is greater than the EMEG for that medium, the potential for exposure to the compound should be further evaluated for the specific situation to determine whether noncancer health effects might be possible. Conversely, if the concentration is less than the EMEG, it is unlikely that exposure would result in noncancer health effects. EMEG values are derived for different durations of exposure according to ATSDR's guidelines. Acute EMEGs correspond to exposures lasting 14 days or less. Intermediate EMEGs correspond to exposures lasting longer than 14 days to less than one year. Chronic EMEGs correspond to exposures lasting one year or longer. CREG values are derived assuming a lifetime duration of exposure. RMEG values also assume chronic exposure. All the comparison values (i.e., CREGs, EMEGs, RMEGs) are derived assuming opportunities for exposure in a residential setting.
Summary data tables are provided in the following discussion of on-site contamination. For the most part, data tables are provided only for those environmental media with contaminants that have been detected at concentrations that exceed their respective comparison values.
The site encompasses the Mount Everett Landscaping Building and all residences in Hartsville in the area downgradient of this building to the Konkapot River. Since this is a small area, the whole area is considered on-site and hence, all data available for the site are summarized in this section.
One soil sample collected at a 7-9 ft-depth during the installation of MW-1 on October 22, 1998 was analyzed for methyl tertiary-butyl ether (MTBE), benzene, and other volatile petroleum hydrocarbons (VPHs). While MTBE and benzene were not detected in this sample, several other VPHs were detected but at levels below their respective screening values.
Two rounds of subsurface soil excavation occurred at the site from November 13 to November 23, 1998 and on April 15, 1999 (MT 1999c) (see Figure 4). During the excavation period at the area west of the Mt. Everett Landscaping building, 21 soil samples were collected from four sidewalls and the bottom of the excavated area at depths ranging from 8-14 ft and analyzed for MTBE, benzene, and other VPHs. MTBE, benzene, and three other VPHs were detected at levels exceeding their screening values. Out of these 21 samples, MTBE was detected in 16 samples at levels ranging from 0.099 parts per million (ppm) to 88.7 ppm with a mean of 5.95 ppm; benzene detected in four samples at levels ranging from 0.449 ppm to 13.6 ppm with a mean of 0.76 ppm; naphthalene detected in 15 samples at levels ranging from 0.056 ppm to 91.3 ppm with a mean of 4.99 ppm; C5-C8 aliphatics detected in 15 samples at levels ranging from 0.062 ppm to 1,400 ppm with a mean of 84.57 ppm; and C9-C10 aromatics detected in 14 samples at levels ranging from 0.077 ppm to 1,290 ppm with a mean of 74.08 ppm. Other VPHs were also detected in these samples but at levels below their respective screening values.
On December 14, 1998, two soil samples were collected from the southwest section of the rear addition to the landscaping building at depths of 9.5 ft to determine if contaminants had migrated beneath the southwest corner of the building. These samples were analyzed for MTBE, benzene and other VPHs. MTBE, benzene, and two other VPHs were detected at levels exceeding their screening values: MTBE at 3.33 ppm and 40.4 ppm with a mean of 21.87 ppm; benzene at 2.7 ppm and nondetectable with a mean of 1.36 ppm (using half the method detection limit of 0.05 ppm to calculate the mean); C5-C8 aliphatics at 146 ppm and 219 ppm with a mean of 182.5 ppm; and C9-C10 aliphatics at 336 ppm and 386 ppm with a mean of 361 ppm. Other VPHs were also detected in these samples but at levels below their respective screening values. Since VPHs were detected in these two soil samples, it appeared that the contaminants had migrated to beneath the southwest corner of the landscaping building or the southwest section of the rear addition to that building.
After additional soil excavation was conducted from underneath the southwest section of the rear addition of the landscaping building on April 15, 1999, four soil samples were collected from three sidewalls and the bottom of the excavated area at depths of 7 ft or 9 ft and analyzed for MTBE, benzene, and other VPHs. In these four samples, MTBE was detected at levels ranging from 0.185 ppm to 0.722 ppm with a mean of 0.38 ppm. Other VPHs were also detected but at levels below their respective screening values.
Five rounds of surface water sampling were conducted in the area of the site at the Konkapot River and at an unnamed brook that flows along the southerly edge of the Mount Everett Landscaping property, downgradient of the release (MT 1999b). For five sampling rounds, a total of 15 samples from the Konkapot River were collected in January, April, July, and October of 1999, and January 2000. These samples were collected at three locations of Konkapot River: upgradient of the site at the confluence of the Konkapot River and the outfall of Lake Buel, immediately downgradient of the site (either from the river or a wetland area adjacent to the site), and downgradient where Crosby Road crosses the Konkapot River (MT 2000a). Out of 15 samples, one was found to contain gasoline-related compounds (i.e., benzene, C5-C8 aliphatics, and C9-C12 aliphatics). This sample was immediately downgradient of the site and was collected in the first sampling round. The detected levels for the aliphatics were both below their comparison values. The benzene level was 1.45 parts per billion (ppb), slightly exceeding its CREG but less than its MCL or MMCL (see Table 2). MTBE, naphthalene, C5-C8 aliphatics and C9-C12 aliphatics were also detected in two to three other samples in the third round but all at levels lower than their comparison values. Those compounds were not detected in any samples collected in the second, fourth, and fifth rounds.
Five samples were also collected in the unnamed brook in February, April, July, October 1999 and January 2000. Compounds detected in these samples at the unnamed brook include naphthalene, C5-C8 aliphatics, and C9-C12 aliphatics. However, all detected levels were below their respective comparison values.
A total of six monitoring wells have been installed on the site to assess groundwater conditions and contamination. Groundwater was determined to flow in a west-southwesterly direction. Three monitoring wells (MW-1, MW-2, and MW-3) were installed in October 1998 (MT 1999a), downgradient of the landscaping building, just upgradient of the rental property (see Figure 5). Three additional monitoring wells (MW-4, MW-5, MW-6) were installed in November 1998: MW-4 was installed immediately downgradient of the 300 gallon aboveground storage tank while MW-5 and MW-6 were installed at the parking lot and northeast corner of the landscaping building. Samples were collected from these monitoring wells and analyzed for VPHs using both the targeted VPH analytes and the EPA 524.2 Analysis methods. Based on the wells' locations, MW-1 through MW-4 would be considered downgradient of the source while MW-5 and MW-6 are upgradient of the source (MT 1999a). Table 3 summarized data collected from MW-1 through MW-4 from October/November 1998 to January 2000.
One sample each was collected from MW-1 through MW-4 in late October to November 1998 and analyzed for VPHs. Three more samples were collected from MW-4 in July and October 1999 and January 2000. Sampling data for these monitoring wells showed that four samples collected in October/November 1998 had MTBE detections ranging from 46,200 ppm to 119,200 ppm while three samples collected from MW-4 from 1999 to 2000 had MTBE levels ranging from nondetectable to 3.37 ppm, which were lower than the MA DEP guidelines for MTBE (70 ppb) (MA DEP 2001). For all these seven samples, the average of MTBE concentrations was 53,287 ppb.
Results from these samples also showed that other compounds that had concentrations detected at levels higher than their respective comparison values included benzene, ethylbenzene, naphthalene, toluene, 1,2,4-trimethylbenzene, 1,2,5-trimethylbenzene, xylene and VPH fractions. Benzene levels ranged from nondetectable to 3,950 ppb with a mean of 1,552 ppb. Ethylbenzene levels ranged from nondetectable to 1,855 ppb with a mean of 378 ppb. Naphthalene levels ranged from nondetectable to 479 ppb with a mean of 110 ppb. Toluene levels ranged from nondetectable to 12,200 ppb with a mean of 2,754 ppb. Xylene levels ranged from nondetectable to 9,985 ppb with a mean of 3,540 ppb. VPH fractions ranged from nondetectable to 9,560. In November 1998, 1,2,4-trimethylbenzene was found at 2,300 ppb and 1,2,5-trimethylbenzene was found at 560 ppb, thus exceeding comparison values. One of the seven samples collected from MW?4 were also analyzed for other compounds such as chloroform, bromodichloromethane, n-propylbenzene, and trichloroethene. The results were all nondetectable. However, the detection limits for these compounds exceeded their respective comparison values.
One sample each was collected for two upgradient monitoring wells MW-5 and MW-6 on November 18, 1998. MTBE was not detected in the sample from MW-5 and was detected in the sample collected from MW-6 at a level of 6.81 ppb. Benzene was detected at 0.74 ppb and 6.26 ppb from MW-5 and MW-6, respectively. The maximum benzene detected was slightly higher than the MCL (5 ppb) for this compound. These two samples were also analyzed for other VPHs; however, results were either nondetectable or lower than comparison values.
The sampling of household water used by area residents began in October 1998. Following gasoline-related contamination being found in the water from 8 Hatchery Road (a rental property), the environmental consultant began collecting samples from other houses in Hartsville village to determine the extent of groundwater contamination.
In November 1998, household water from additional houses was also sampled: five residences immediately west of Hartsville/New Marlborough Road (from property address 1553 through 1581); two residences along Hartsville/Mill River Road (property addresses 1359 and 1368); three residences on Hatchery Road (property addresses 32, 39, and 67); and the Mount Everett Landscaping property along Hatchery Road.
A summary of the results of the household water-sampling program is presented in Tables 4a through 4j.
- 8 Hatchery Road
The well with the highest concentrations of gasoline constituents was at 8 Hatchery Road. Table 4a summarizes data collected from this well from October 7 through November 18 1998. The above sampling period represents the time between the first sampling event and the sampling event closest to the time that the tenants moved out. (2) Sampling data showed that five out of five groundwater samples at this well contained MTBE at concentrations ranging from 5,187 ppb to 41,200 ppb with the mean of 21,527 ppb. The minimum, mean and maximum concentrations all exceed comparison values for MTBE. Benzene was also found in these five samples at concentrations above its comparison values, with detected concentrations ranging from 66 ppb to 620 ppb with a mean of 319 ppb. Comparison values for benzene range from 0.6 to 5 ppb (CREG, MCL, and MMCL). Chloroform and trichloroethene were not detected but the samples had method detection limits higher than their respective comparison values. These latter compounds are not components of gasoline, and it would be unlikely for them to be present at levels of health concern. Hence, these will not be further considered for this residence.
For 1,2,4-trimethylbenzene and 1,3,5-trimethylbenzene, comparison values used are taken from EPA. 1,2,4-trimethylbenzene was detected in two out of five samples with the maximum level lower than its comparison value. 1,3,5-trimethylbenzene was detected in one out of five samples with the detected level lower than its comparison value. However, the method detection limits for the nondetectable samples of these two compounds were higher than their respective comparison values. Since the levels detected for these two compounds were lower than their respective comparison values, it is not likely that opportunities for exposure to these two compounds would result in adverse health effects. Thus, 1,2,4-trimethylbenzene and 1,3,5-trimethylbenzene will not be further considered for this residence. The maximum concentration for toluene slightly exceeds one comparison value, but the mean level is well within an acceptable level. Hence, health concerns would not be expected. MTBE and benzene in groundwater are the two compounds that will be evaluated for human exposures at this residence.
Table 4b presents a summary of sampling data for 8 Hatchery Road shortly after the tenants moved out. Three samples were collected for the period of November 25 through December 9, 1998. Table 4b showed that MTBE and benzene levels exceeded their respective screening values for all three samples.
A hand-driven well was installed in the basement of 8 Hatchery Road on February 8, 1999 (MT 1999d). The property was still vacant. The well was installed at the north end of the basement, at a depth of 8 to 9 feet. It was monitored monthly after installation (MT 1999f, MT 2000b). Available sampling results from February 1999 to January 2000 are summarized in Table 4c. This table shows that MTBE was detected in seven out of seven samples from this well, at concentrations ranging from 33 ppb to 18,000 ppb with a mean of 7,188 ppb. Both the mean and maximum concentrations exceed all available comparison values for MTBE. Benzene was also detected in seven of seven samples, at concentrations ranging from 0.61 ppb to 480 ppb with a mean of 121 ppb. Both the mean and maximum concentrations of benzene also exceed its respective comparison values (CREG of 0.6 ppb, MCL and MMCL of 5ppb). Trichloroethene and tetrachloroethene were not detected in samples collected from the hand-driven well. However, the detection limit of 4.6 ppb for these compounds exceeded the CREG values. 1,2,4-trimethylbenzene was detected in one of seven samples, at a concentration (1.3 ppb) below its respective comparison value (12 ppb).
- 1553 Hartsville/New Marlborough Road
The next highest levels of gasoline constituents were found in well samples from 1553 Hartsville/New Marlborough Road. Sampling results from November 3, 1998 to December 2, 1998 (the date that well water treatment began) showed that MTBE was detected in five out of five samples at concentrations ranging from 2,030 ppb to 3,300 ppb with a mean of 2,846 ppb. All detected concentrations exceed drinking water guidelines. Table 4d summarizes these results.
- 1534G Hartsville/New Marlborough Road
Well water samples collected at 1534G (garage) Hartsville/New Marlborough Road showed the next highest levels of gasoline constituents. Water treatment was installed for this well on February 18, 1999. Sampling results from October 19, 1998 to January 27, 1999 are summarized in Table 4e. MTBE was detected in two out of four samples at concentrations of 2,400 ppb and 2,800 ppb with a mean MTBE level of 1,300 ppb. Both the mean and maximum concentrations exceed drinking water guidelines. Benzene and trichloroethene were not detected in the four samples and therefore will not be discussed further for this location. Tetrachloroethene was detected in one out of the four samples and the detected level (0.58 ppb) was below all of its comparison values (ranging from 0.7 to 400 ppb) and therefore, will also not be discussed further for this location.
- 1534H Hartsville/New Marlborough Road
Water samples collected from this residence between October 19, 1998, and January 27, 1999 showed that two out of four samples contained MTBE at levels of 36 ppb and 86 ppb with a mean of 31 ppb. The maximum detected concentration of 86 ppb exceeded drinking water guidelines. These results are summarized in Table 4f. Water treatment was installed for this well on January 27, 1999.
- 1352 Hartsville/Mill River Road
Water samples collected at 1352 Hartsville/Mill River Road from October 30, 1998 through December 2, 1998 (the date that well water treatment began) showed that MTBE was detected in five out of five samples. Table 4g summarizes these results and shows that MTBE concentrations ranged from 375 ppb to 1,500 ppb with a mean of 903 ppb. All detected concentrations exceed drinking water guidelines. Benzene and trichloroethene were not detected in these five samples and therefore will not be discussed further.
When the ownership of the property changed in May 1999, the new owners installed a bedrock well on this property as an alternative water supply. The shallow well was used for monitoring purposes after that. Table 4h shows results of this well point sampling that occurred between June 4, 1999 and June 8, 2000. MTBE was detected in three of nine samples at a maximum concentration of 36 ppb, with a mean of 5.6 ppb. The new bedrock well that was used as the household water supply beginning in May 1999, was sampled on June 6 and 23, 1999 and again on July 14, 1999. No MTBE nor any other compound of concern was detected.
- 1368 Hartsville/Mill River Road
Water samples collected at this residence between November 4, and December 2, 1998 are summarized in Table 4i. This table shows that MTBE was detected in five out of five samples at concentrations ranging from 2.1 ppb to 6.5 ppb with a mean of 3.8 ppb. The maximum MTBE concentration detected is below drinking water guidelines. Benzene was detected in one out of five samples at 1.2 ppb, which is slightly above its CREG of 0.6 ppb and well below the MCL (5 ppb). All other compounds detected were below comparison values and thus would not be of health concern and will not be discussed further.
- Other residences
Well water from additional properties in Hartsville was sampled to determine the extent of groundwater contamination. Table 4j summarizes sampling data for these additional properties. Most of the drinking water supply wells (i.e., wells at property addresses 1526, 1559, 1565, 1574, 1359, 32, and the landscaping building) have been sampled on a monthly basis. Drinking water wells at other residences (i.e., wells at property addresses 1571, 1581, 1477, 39, 56, and 67) were sampled once in November 1998. MTBE was detected in 13 out of 96 samples, with a maximum detected concentration of 45 ppb (at 1559 Hartsville/New Marlborough Road, also known as SHOP-2) and a mean level of 1.3 ppb. Benzene was not detected in any of the samples. Trichloroethene was detected in five out of these 96 samples at a maximum level of 3.2 ppb, which is approximately the same as its comparison value of 0.09 ppb (CREG) and below the MCL and MMCL for trichloroethene of 5 ppb. Tetrachloroethene was detected in one out of these 96 samples at a maximum level of 3.7 ppb. This maximum level is below all comparison values for tetrachloroethene (RMEGs, MCL, and MMCL).
Treated well water
Point of entry treatment systems were installed at the five residences discussed above in December 1998, and in January and February 1999. Tap water from these houses were sampled on a weekly basis following the installation of the treatment systems. Samples were collected pre-carbon, mid-carbon, and post-carbon filter. Based on sampling results for these five residences for pre-carbon, mid-carbon, and post-carbon water summarized in tables 5a through 5e, the water treatment systems appeared to be working effectively in removing compounds from the water supply to the houses. MTBE, the primary contaminant of concern in these samples, ranged from nondetectable to 3,600 ppb prior to filtration and nondetectable to 29 ppb after filtration. Thus, levels of contaminants at the post-carbon point were significantly lower than pre-carbon and also lower than comparison values. Except for methylene chloride detected as a laboratory contaminant for 1553 Hartsville/New Marlborough Road, there were no exceedances of MCLs or guidance values post treatment for drinking water in all these houses.
Indoor air at four residences was screened by the environmental consultant for the presence of volatile organic compounds (VOCs) during water supply well sampling activities (MT 1999a, MT 1999e). A portable OVM (Organic Vapor Meter) was used; its limit of detection was 0.1 ppm organic vapors. The residences screened were all actively occupied at the time of testing. The indoor air screening activity at 8 Hatchery Road in October 1998 was conducted before the water treatment system was installed in December 1998. For the other three residences, indoor air screening activities were conducted at the same time the carbon filters were installed. These included:
- 8 Hatchery Road (October 19, 1998; basement, first floor living room, and first floor bathroom)
- 1352 Hartsville Road (December 2, 1998; basement and first floor living area)
- 1368 Hartsville Road (December 2, 1998; basement and first floor living area)
- 1553 Hartsville/New Marlborough Road (December 2, 1998; basement and first floor living area)
Screening at 8 Hatchery Road showed the presence of organic vapors in the first floor bathroom at 0.4 ppm. Organic vapors were not detected in the other screened areas at 8 Hatchery Road nor at the three other residences included in the sampling.
Because VOCs were detected above the detection limit in the indoor air at 8 Hatchery Road, air quality monitoring was conducted there on November 30, 1998 for six hours. This sampling occurred after the former tenants moved out of 8 Hatchery Road and the house was vacant. Hence, these results may have underestimated the opportunity for exposure, since normal volumes of household water were likely not in use prior to the tests. Four indoor air samples were taken at the following locations: the basement at the foot of the stairs, the first floor front hall entry, the first floor kitchen, and the second floor bathroom. One sample was also taken outdoors on the porch of the house (MT 1999f). These samples were analyzed for VOCs (3).
Results from these indoor and outdoor samples showed that except for benzene, all other VOC concentrations were either nondetect or lower than their respective comparison values. Table 6 shows benzene concentrations in indoor and ambient air, the compound detected at levels exceeding its comparison values. Benzene was detected in all of the four indoor air samples at levels lower than one of its comparison values (intermediate EMEG of 13 microgram per cubic meter [µg/m3]) but higher than the other comparison value (CREG of 0.1 µg/m3). The outdoor sample showed a concentration of 17.2 µg/m3.
Benzene is ubiquitous in the atmosphere (ATSDR 1997). ATSDR's toxicological profile for benzene provides data on benzene measurements in ambient air from cities across the United States. These data showed concentrations ranging from less than 1 µg/m3 to over 200 µg/m3. Median or average values from these various studies ranged from 1 µg/m3 to more than 60 µg/m3. ATSDR also reported on a study conducted by EPA (Total Exposure Assessment Methodology study) where the median level of benzene in 195 homes without smokers was 7 µg/m3. Thus, the benzene levels measured in indoor or outdoor air at 8 Hatchery Road appear to reflect background ambient air levels for the United States. Although the indoor air levels at 8 Hatchery Road were below comparison values, the house was not occupied at the time of sampling. Therefore, it is likely that the actual indoor air levels might be higher than what was found due to volatilization of the VOCs from using the water for household activities if the house was occupied.
For some VPHs such as 1,2,4-trimethylbenzene, 1,3,5-trimethylbenzene, and also trichloroethene and tetrachloroethene, several groundwater nondetectable samples had the method detection limits two- to eight-fold higher than their maximum values and comparison values; hence, a reliable minimum and mean could not be determined. Since a reliable minimum and mean could not be determined for some samples, those samples were not used to make health evaluations. Methylene chloride was also found as a lab containment in numerous samples.
At the time of the site visit in May 1999, there were trucks, tractors and heavy construction equipment at the unfenced area in front of the landscaping building. This is a site of commercial activity, the nature of which requires heavy equipment. While these potential concerns are not the primary focus of the public health assessment, which was to evaluate potential health and exposure concerns related to exposure opportunities to chemicals, this equipment could present a physical hazard to trespassers. There are no other known physical hazards to the general public at the site.
To determine whether nearby residents and people on-site were, are, or could be exposed to contaminants, an evaluation was made of the environmental and human components that lead to human exposure. 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.
Exposure to a chemical must first occur before any adverse health effects can result. Five conditions must be met for exposure to occur. First, there must be a source of that chemical. Second, a medium (e.g., water) must be contaminated by either the source or by chemicals transported away from the source. Third, there must be a location where a person can potentially contact the contaminated medium. Fourth, there must be a means by which the contaminated medium could enter a person's body (e.g., ingestion). Finally, the chemical must actually reach the target organ susceptible to the toxic effects from that particular substance at a sufficient dose for a sufficient time for an adverse health effect to occur (ATSDR 1993).
A completed exposure pathway exists when all of the above five elements are present. A potential exposure pathway exists when one or more of the five elements is missing and indicates that exposure to a contaminant could have occurred in the past, could be occurring in the present, or could occur in the future. An exposure pathway can be eliminated if at least one of the five elements is missing and will not likely be present. The discussion that follows incorporates only those pathways that are important and relevant to the site.
There are some past opportunities for exposure to contaminated well water for people living in properties downgradient of the release. People could have been exposed by ingestion of contaminated drinking water from their wells, by inhalation of volatile contaminants in their water during showering, bathing, and household use of the water, and by absorption of the contaminants through their skin during bathing, showering, and using the water in their homes for cleaning and cooking.
The four private drinking water wells with levels of compounds greater than drinking water standards or health-based guidelines at the time of initial sampling were: 8 Hatchery Road, 1553 Hartsville/New Marlborough Road, 1352 Hartsville/Mill River Road, and 1534 Hartsville/New Marlborough Road (house and garage).
The 8 Hatchery Road property was bought by its current owners in 1986 (New Marlborough Board of Assessors 1999). From March 1st of 1997 to November 15th of 1998, tenants occupied the property (representing a period of 20 ½ months). Therefore, the former tenants could have had opportunities for exposure to the site contaminants for an approximate period of 20 ½ months, if the release occurred early in 1997. Other information regarding occupancy prior to this was not available to MDPH. According to the Phase I Report, the tenants first noticed an odor in their tap water by or about February 1998 (MT 1999e).
The house at 1553 Hartsville/New Marlborough was built approximately in the 1700s and owned by generations of one family since then (New Marlborough Board of Assessors 1999). The house at 1534 Hartsville/New Marlborough was built approximately in 1935 and also owned by one family since then (New Marlborough Board of Assessors 1999). Therefore, it is possible that the owners of these houses and their families had opportunities for exposure to the contaminants for a period of two years (the estimated length of time between the release and discontinued use of their well water). The house at 1352 Hartsville/Mill River was built in approximately the 1880s and owned by one family since then (New Marlborough Board of Assessors 1999) until May 10, 1999 (MT 1999d). It is possible that the former owners of this property also had opportunities for exposure to contaminants in their well water for approximately two years.
Besides the houses mentioned above, one other house (1534 Hartsville/New Marlborough Road) in the area also had at least one sample with MTBE concentrations detected at levels higher than Massachusetts guideline for MTBE in drinking water. Therefore, opportunities for exposures to drinking water at this property will be addressed further.
No surface soil data were available for review. Subsurface soil in the area of the release revealed elevated levels of some compounds such as MTBE. However, direct contact with these soils did not likely occur (i.e., incomplete exposure pathway). It is possible that employees of the Mount Everett Landscaping Company could have come into contact in the past with gasoline contamination in surface soil if the release occurred on the ground surface, as assumed in the Phase I Report. Such exposures would have occurred before the surface soil was excavated and removed as part of the IRA. However, no data are available to evaluate such exposure opportunities.
Although a potential pathway for exposure to the gasoline contaminants in the nearby surface water existed, via groundwater discharge to the Konkapot River, the unnamed brook, or the wetland area, the likelihood of a significant exposure was low. This was because the concentrations of gasoline contaminants detected were below levels of health concern and actual contact with the surface water would also have been expected to be infrequent.
Opportunities for exposure to contaminants in drinking water from private wells were prevented following installation and monitoring of point of entry treatment systems, and are not expected to happen now or in the future because, according to MA DEP, the site clean-up has been completed. Because contamination was present in the groundwater in the vicinity of the household wells for a period of time following site discovery and during use of the point of entry treatment systems, if these systems had failed to operate effectively, exposure could have potentially occurred. However, based on results from the on-going monitoring that was done until a final solution was reached, exposures did not appear to have occurred.
Although workers involved in soil excavation as part of the IRA and Phase I activities could have come in contact with subsurface soil, if appropriate protective clothing and equipment were missing, it is unlikely that area residents or employees of the landscaping company would have had the opportunity to come into contact with soil beneath the ground surface.
Measurements for VOCs in the indoor air at 8 Hatchery Road revealed that benzene concentrations exceeded one comparison value, the CREG. However, the concentrations of benzene in the indoor air were less than the outdoor air concentration measured at the same time, and also were similar to those that have been reported in many studies in cities across the United States (ATSDR 1997). Thus, for purposes of this PHA, opportunities for exposures to benzene in indoor air, based on these sampling results, are not further assessed.
MDPH staff have summarized the available environmental data and exposure pathways for the Hatchery Road site in Hartsville/New Marlborough in this public health assessment. Completed exposure pathway included contact with household water at the residences in Hartsville/New Marlborough. Based on the evaluation presented above, the compounds of concern in private well water were MTBE and benzene.
Opportunities for exposure to MTBE and benzene were primarily via ingestion of water from private drinking water wells, inhalation, and dermal absorption during everyday household activities such as showering, bathing, washing floors and cooking with contaminated water (MA DEP 1995).
In order to evaluate possible public health implications, estimates of opportunities for exposure to compounds (e.g., in soil) must be combined with what is known about the toxicity of the chemicals. ATSDR has developed minimal risk levels (MRL) for many chemicals. An MRL is an estimate of daily human exposure to a substance that is likely to be without an appreciable risk of adverse noncancer health effects over a specified duration of exposure. MRLs are derived based on no- observed-adverse-effect levels (NOAELs) or lowest-observed-adverse-effect levels (LOAELs) from either human or animal studies. The LOAELs or NOAELs reflect the actual levels of exposure that are used in studies. ATSDR has also classified LOAELs into "less serious" or "serious" effects. "Less serious" effects are those that are not expected to cause significant dysfunction or whose significance to the organism is not entirely clear. "Serious" effects are those that evoke failure in a biological system and can lead to illness or death. When reliable and sufficient data exist, MRLs are derived from NOAELs or from "less serious" LOAELs, if no NOAEL is available for the study. To derive these levels, ATSDR also accounts for uncertainties about the toxicity of a compound by applying various margins of safety, thereby establishing a level that is well below a level of health concern.
MTBE is a flammable liquid made from combinations of chemicals like isobutylene and methanol. The 1990 federal Clean Air Act Amendments require fuel oxygenates, such as MTBE or ethanol, to be added to gasoline used in some metropolitan areas to reduce atmospheric concentrations of carbon monoxide or ozone (Squillace et al. 1996). MTBE is the most commonly used fuel oxygenate (Squillace et al. 1996) so most people would come in contact with it when exposed to automobile fuel vapors or exhaust (ATSDR 1998).
The potential for MTBE to cause health effects has been of concern to many state, federal, and international agencies in recent years. In addition, many studies are currently underway to better understand the toxicity of the compound. The following section summarizes some of the toxicological data, previous reviews, and available drinking water quality guidelines that have been derived by state or federal agencies.
For humans, there are limited data for the effects of MTBE, no data on oral exposures, and no data for long-term exposures (EPA 1997). Although more data are available from animal studies, significant data gaps still exist. For example, ATSDR did not derive a minimal risk level (MRL) for MTBE for chronic (i.e., more than one year) exposures because of insufficient data (ATSDR 1998). The EPA Office of Water also concluded that although MTBE may cause noncancer or cancer effects, available data did not support confident quantitative estimates of risk at low exposure levels due to many uncertainties and limitations associated with the toxicity database for MTBE (EPA 1997).
ATSDR did derive an oral MRL of 0.3 milligrams per kilograms per day (mg/kg/day) for intermediate exposures (i.e., one year or less) to MTBE. The intermediate MRL was based on a minimal lowest observed adverse effect level (LOAEL) of 100 mg/kg/day for hepatic effects in rats exposed by gavage to MTBE for 90 days (Robinson et al. 1990; as cited in ATSDR 1998). The study showed significantly decreased blood urea nitrogen levels in both male and female rats at this dose level. In order to derive the intermediate MRL, ATSDR applied a total uncertainty factor of 300 (3 for use of an LOAEL, 10 for extrapolation from animals to humans, and 10 for human variability). These effects were considered by ATSDR to be "less serious" effects, which are not expected to cause significant dysfunction or for which the significance is not entirely clear (ATSDR 1998).
MTBE has also been shown to affect the kidneys, gastrointestinal tract, and nervous system. For example, studies by Chun et al (1992; as cited in ATSDR 1998) and Burleigh-Flayer et al (1992; as cited in ATSDR 1998) indicated neurotoxicity (e.g., hypoactivity, ataxia, lack of startle reflex) at inhalation exposures of 3,000 and 8,000 ppm in rats and mice. The Robinson et al. (1990) study discussed above also showed adverse effects on kidneys in the rats at dose levels of 300 mg/kg/day and higher.
There are some studies in rodent species on the potential for MTBE to cause cancer. The only oral cancer study was conducted by Belpoggi et al (1997), where Sprague-Dawley rats were dosed by gavage at doses up to 1,000 mg/kg/day, 4 days per week, for two years. Exposure caused a dose-related increase in the incidence of combined leukemia and lymphomas in females and an increase in benign testicular tumors in males at the high dose. However, this study has some limitations for its interpretation. The effect of administering the dose via gavage versus via drinking water is uncertain. Also, the study publication did not provide sufficient histopathological information, which led the National Research Council to recommend not using this study in estimating cancer risk until after a more thorough peer review of the study results (EPA 1997).
Two other inhalation studies (Chun et al. 1992; Burleigh-Falyer et al. 1992; as cited in ATSDR 1998) were also conducted (EPA 1997). In Chun et al. (1992), inhalation exposures doses of 3,000 ppm and 8,000 ppm MTBE over 24-months to F344 rats showed an increased incidence of renal tubular cell adenoma and carcinomas, as well as a dose-related increased incidence of testicular tumors among males (ATSDR 1998). The incidence of testicular tumors was not significantly different from historical controls but was different from the controls used in the study (EPA 1997). EPA noted that the consistency of two studies (Belpoggi et al. 1995; Chun et al. 1992; as cited in ATSDR 1998) showing testicular tumors support the conclusion that these tumors may be treatment-related.
Burleigh-Flayer et al (1992; as cited in ATSDR 1998) administered MTBE to CD-1 mice at concentrations up to 8,000 ppm for 18 months. This exposure was associated with a statistically significant increase in the incidence of hepatocellular carcinomas in male mice and of hepatocellular adenomas in female mice. At this time, EPA believes that these two inhalation studies cannot support an attempt to quantitatively estimate potential cancer risks from MTBE based on ingestion of MTBE in drinking water. Additional research is ongoing to determine whether such route-to-route extrapolation may be developed in the future (EPA 1997).
A number of studies are currently ongoing evaluating the potential noncancer and cancer effects from exposures to MTBE (EPA 1999). Currently available data suggest that MTBE poses a potential for carcinogenicity to humans at high doses, but the data do not support attempts to quantify this risk (EPA 1997). No agency that routinely evaluates cancer incidence for chemicals has yet classified MTBE as to its carcinogenicity. These agencies include the EPA, the National Toxicology Program, and the International Agency for Research on Cancer (IARC).
EPA has developed a nonenforceable guideline for MTBE in drinking water based on the ability to taste or smell the compound. The drinking water advisory level is 20-40 ppb, designed to protect against taste and odor effects but also provides a large margin of safety against potential adverse health effects (EPA 1997). EPA also noted that this guideline is more than 10,000 to 100,000 times lower than concentrations associated with observed cancer and noncancer effects in animal studies (EPA 1997). Thus, 20-40 ppb would also not be expected to result in adverse health effects (EPA 1997).
In addition, eleven states, including Massachusetts, have developed health-based drinking water guidelines or standards for MTBE (EPA 1999). Health-based guidance levels or standards range from 13 ppb (California) to 240 ppb (Michigan) (EPA 1999). With the exception of California and Michigan, all other state health-based levels are in the range of 20-70 ppb. The MA DEP advisory for MTBE is 70 ppb and is based on the same study (the study by Robinson et al. 1990) used by ATSDR to derive its intermediate MRL. MA DEP applied a total uncertainty factor of 10,000 to the 100 mg/kg/day dose level observed in the Robinson et al. study (10 for extrapolation from a subchronic to a chronic exposure, 10 for extrapolation from animals to humans; 10 for human variability, and 10 for the uncertainty about MTBE's ability to cause cancer).
Benzene is a colorless liquid with a sweet odor that can be found in air, water, and soil from both human activities and natural processes. Benzene is made mostly from petroleum sources to make other chemicals and some types of rubbers, lubricants, dyes, detergents, drugs, and pesticides. Benzene is also a part of crude oil, gasoline, and cigarette smoke. Natural sources of benzene include volcanoes and forest fire (ATSDR 1997).
Benzene is readily absorbed through three exposure routes: dermal contact, ingestion, and inhalation. Regardless of the route of exposure, once absorbed into the body, benzene is widely distributed and rapidly metabolized and eliminated. Health effects associated with ingestion or inhalation exposure to benzene include hematological (i.e., blood-related) and reproductive effects, which have been observed in both human populations and experimental animals (ATSDR 1997).
Because of conflicting study results, there are no MRLs for the noncancer effects of benzene via oral exposure (ATSDR 1997). In three short-term oral exposure studies in mice of four weeks (intermediate) duration by Hsieh et al. in 1988, neurological effects such as fluctuation of neurotransmitter levels were observed, as well as decreased numbers of erythrocytes, leukocytes, and lymphocytes, all at exposures of 8 mg/kg/day (ATSDR 1997). In other chronic oral exposure studies, reproductive and hematological effects have been observed in rats and mice that were administered 25 mg/kg/day.
To evaluate the noncancer adverse health effects potentially caused by exposure to benzene at this site, a LOAEL level of 8 mg/kg/day was used as the comparison value with the estimated exposure doses. This level is the lowest dose in either animal or human studies that has been associated with adverse health effects (ATSDR 1997).
It has been established that human exposure to commercial benzene or benzene-containing mixtures can cause damage to the hematopoietic system with subsequent manifestation of acute myelogenous leukemia. Some studies have implicated different types of leukemia and multiple myeloma with exposures to benzene. Benzene has also been shown to cause leukemia and lymphoma in laboratory animals (ATSDR 1997). IARC classifies benzene as a Group 1 carcinogen, a human carcinogen due to sufficient human data (IARC 1987). EPA has identified benzene as a Class A carcinogen (human carcinogen) and National Toxicology Program (NTP) lists benzene as a "substance known to be carcinogenic", that is, a substance for which the evidence from human studies indicates that there is a causal relationship between exposure to the substance and human cancer (ATSDR 1997).
This section evaluates opportunities for exposures and possible health effects to residents or former residents at 8 Hatchery Road, 1553 and 1534 Hartsville/New Marlborough Road, and 1352 Hartsville/Mill River Road.
For New Marlborough residents, opportunities for exposure were estimated for several exposure pathways. The estimates included assumptions that children and adults drink 1 and 1.4 liters of water per day (Olin 1999), and weigh 10 kilogram (kg) and 70 kg, respectively.
In addition to ingestion of drinking water, opportunities for exposures to VOCs, such as MTBE, can also occur with other household use of water, such as showering or cooking. Available indoor air data for the residences indicated that VOCs were not elevated in the residences. Nonetheless, to be conservative, MDPH assumed that opportunities for exposures to MTBE or benzene could have included exposures in the past via household uses of the water. For purposes of this PHA, MDPH adopted guidance from the MA DEP (MA DEP 1995) which assumes that for exposure to VOCs in water, the average daily dose due to other household uses of water (e.g., inhalation and dermal exposures) is twice that estimated for opportunities for exposure via ingestion of drinking water. The MA DEP assumption is based on numerous studies by Brown et al. in 1984, Andelman in 1985, Foster and Chrostowski in 1987, McKone in 1987 and 1991 (MA DEP 1995).
Tenants at 8 Hatchery Road lived in the house for 20 ½ months, which will be used as the assumed duration of their exposure to contaminated well water. During this time period, only adults occupied the residence. For other residences in the Hartsville village, the duration of exposure to contaminated well water is two years with the assumption that the gasoline spill occurred within the two years prior to December 1998 when the carbon treatment system was first installed (MT 1999a).
1. Determination of Contaminants Needing Further Evaluation
MTBE and benzene are the two compounds that were found in private well water at levels higher than their respective comparison values and therefore will be further evaluated.
2. Opportunities for Exposure to MTBE
Noncancer adverse health effects
The MRL and LOAEL for MTBE used to compare the estimated exposure doses are 0.3 mg/kg/day and 100 mg/kg/day, respectively. The average MTBE concentrations are used for calculating the estimated exposure doses rather than the maximum because the average concentrations are likely more representative of opportunities for exposure over a two-year period than the maximum concentration for one sampling event. For example for the private well at 8 Hatchery Road, from October 7 through November 18, 1998, five samples showed MTBE concentrations of 5,187 ppb; 14,250 ppb; 17,600 ppb; 29,400 ppb; and 41,200 ppb. The average concentration of these samples is 21,527 ppb, which was used to estimate opportunities for exposures.
- 8 Hatchery Road
The average MTBE concentration was 21,527 ppb at 8 Hatchery Road. The estimated total exposure doses for adults at this residence were 1.29 mg/kg/day. (4) This estimated exposure dose to adult residents is 4 times higher than ATSDR's intermediate MRL for MTBE for adults. However, this estimated dose is less than the lowest reported LOAEL (100 mg/kg/day). Exposures to contaminated water at the site could have presented some noncancer health concerns (e.g., neurological effects) to some individuals living in this property for the past exposure.
- 1553 Hartsville/New Marlborough Road
The average MTBE concentration of 2,846 ppb found in the drinking water well at this property exceeded available state drinking water guidelines. Using this average MTBE concentration for calculation, the estimated total exposure doses for adults and children at this residence were 0.168 mg/kg/day and 0.84 mg/kg/day, respectively. Although the estimated dose for adults is lower than both the MRL and LOAEL, the estimated dose for children is about 3 times higher than MRL but still lower than LOAEL. The LOAEL was based on animal studies. Because the LOAEL was not based on human studies, safety factors are included when evaluating health effects, especially effects that might occur in children.Thus, exposures to contaminated water at the site could have presented noncancer health concerns to children of this residence.
- 1534G Hartsville/New Marlborough Road
The average MTBE concentration of 1,300 ppb found in the drinking water well at this property exceeded drinking water guidelines. Using this average MTBE concentration for calculation, the total estimated exposure doses for adults and children at this residence were 0.078 mg/kg/day and 0.39 m/kg/day, respectively. These estimated exposure doses are lower than both the MRL and LOAEL. Thus, exposures to contaminated water at the site are not likely to have presented noncancer health concerns to people living at this residence.
- 1352 Hartsville/Mill River Road
As mentioned previously in the background section, this property changed ownership in May 1999. For the former owner of this property, the average MTBE concentration of 903 ppb found in the drinking water well at the property exceeds the MMCL value. Using this average MTBE concentration for calculation, the total estimated exposure doses for the former owner at this residence were 0.054 mg/kg/day and 0.27 mg/kg/day, adults and children, respectively. These estimated exposure doses are lower than or the same as both the MRL and LOAEL. Thus, exposures to contaminated water at the site are not likely to have presented noncancer health concerns to the former owners of this property.
Besides the houses mentioned above, one other house (1534 Hartsville/New Marlborough Road) also had at least one drinking water sample with MTBE levels detected higher than Massachusetts MTBE guideline. However, due to the infrequent detection and the large margin of uncertainty factor applied to these guidelines, it is not likely for residents in this house that opportunities for exposure to MTBE-contaminated drinking water would have resulted in adverse health effects.
3. Opportunities for Exposure to Benzene
Noncancer adverse health effects
The only residence with benzene concentrations exceeding available screening or MCL values was 8 Hatchery Road. Thus, opportunities for exposure to benzene for this residence are further evaluated in this section.
As noted earlier, benzene is a human carcinogen. For noncancer effects, no MRL has been established by ATSDR due to conflicting data in the literature. However, for purposes of this evaluation, the lowest available LOAEL for any health effect from either human or animal studies (8 mg/kg/day) was compared with opportunities for exposures to benzene at 8 Hatchery Road to assess the possibility of noncancer health effects.
The average benzene concentration at 8 Hatchery Road was 319 ppb. Using exposure assumptions discussed above, the estimated dose for adults at this residence is 0.02 mg/kg/day (5). This estimated exposure dose is approximately 400 times lower than the lowest available LOAELs for neurological, hematological, and immunological/ lymphoreticular effects, which were based on 4-week studies in mice. These animal studies were for a short duration versus the possible duration of exposure to residents in this house (20 ½ months). Usually, as the duration of exposure in an experimental study gets shorter, the threshold at which one might see an effect gets higher. In addition, when comparing levels at which health effects are seen, additional margins of safety are added to derive exposure doses at which no adverse health effect would be expected. The animal studies discussed here were studies at which effects were seen (i.e., decreased numbers of erythrocytes, leukocytes, and lymphocytes, changes in neurotransmitter levels in the mice), and thus, an additional margin of safety would be added to these exposure doses than would be added if the exposure dose was a level at which no adverse effect was seen. Given the above two considerations (i.e., short duration in the animal studies and LOAELs rather than a NOAEL), it is possible that opportunities for exposure to benzene in drinking water at this residence may have presented some noncancer health concerns (e.g., neurological, hematological, and immunological/lymphoreticular effects).
Cancer adverse health effects
As to possible cancer concerns, the CREG for benzene in drinking water (0.6 ppb) is based on a lifetime exposure (i.e., 70 years or more). However, the former tenants had lived at the rental property for 20 ½ months or slightly less than 2 years. The estimated opportunities for exposure can be combined with EPA's oral slope factor to calculate a cancer risk from exposure to benzene. Using the average benzene concentration of 319 ppb and the assumption that the tenants lived at the rental property for 20 ½ months, the estimated excess lifetime cancer risks for adults are not expected to haveresulted in unusual cancer risks for the residents. (6)
The compounds of concern at this site were MTBE and benzene. MTBE has not been classified as to possible carcinogenicity, but some animal studies have shown increases in testicular cancer, renal cancer, liver cancer, or leukemia and lymphomas. Benzene is considered a human carcinogen, being associated with leukemia and multiple myeloma. The MDPH reviewed the incidence of leukemia, multiple myeloma, liver cancer, testicular cancer, two lymphomas (i.e., Hodgkin's disease and non-Hodgkin's lymphoma), and kidney and renal pelvis cancer in the town of New Marlborough to determine whether these cancer types were elevated.
Cancer incidence data for the years 1995 through 1999 were obtained from the MDPH Massachusetts Cancer Registry (MCR). Data for the 1995-1999 time period are the most current available data from the MCR. To determine whether elevated numbers of cancer cases have occurred, the MCR calculates standardized incidence ratios (SIRs) and 95 percent confidence intervals (95 percent CIs) (MDPH 2001). Because of the instability of incidence rates based on small numbers of cases, SIRs are not calculated when fewer than five cases are observed. A detailed explanation of the SIR and 95 percent CIs is presented in Appendix A.
No cases of leukemia, kidney/renal pelvis cancer, multiple myeloma, liver cancer, orHodgkin's disease, and one case each of testicular cancer and non-Hodgkin's lymphomawere reported to the MCR among New Marlborough residents during 1995-1999. Tables 7 through 13 show these incidence results. Based on residential information available from the MCR, neither of these two cases lived in the area of the gasoline release being addressed in this public health assessment. Thus, the available cancer incidence data do not suggest an unusual pattern or thatenvironmental exposures were likely to have played a primary role in their development.
ATSDR and MDPH, through ATSDR's Child Health Initiative, recognize that the unique vulnerabilities of infants and children demand special emphasis in communities faced with contamination of their environment. Children are at a greater risk than adults from certain kinds of exposure to hazardous substances emitted from waste sites. They are more likely exposed because they play outdoors and because they often bring food into contaminated areas. Because of their smaller stature, they might breathe dust, soil, and heavy vapors close to the ground. Children are also smaller, resulting in higher doses of contaminant exposure per body weight. The developing body systems of children can sustain permanent damage if certain toxic exposures occur during critical growth stages. Most importantly, children depend completely on adults for risk identification and management decisions, housing decisions, and access to medical care.
MDPH evaluated health concerns with relation to opportunities for exposure to contaminated drinking water for each house where children might have lived during the likely exposure period. Results of this can be found in Section B of the Discussion.
- Based on the available data, the wells at four properties (i.e., 8 Hatchery Road; 1553, 1534G and 1534H Hartsville/New Marlborough Road; 1352 Hartsville/Mill River Road) had concentrations of MTBE greater than health-based screening values. The well at 8 Hatchery Road had concentrations of benzene greater than health-based screening values.
- Based on the assumed exposure duration of 20 ½ months and available sampling data for 8 Hatchery Road, the possibility exists that past exposures to MTBE and benzene in the drinking water could have resulted in some adverse noncancer health effects (e.g., neurological effects) for the former tenants of the property.
- With the exposure duration of 20 ½ months, the estimated excess lifetime cancer risks from opportunities to exposure to benzene for adults at 8 Hatchery Road are not expected to haveresulted in unusual cancer risks for residents living there.
- Given an estimated exposure duration of two years and available sampling data, the opportunities for past exposures to MTBE in the drinking water of 1553 Hartsville/New Marlborough residence could have resulted in some adverse noncancer health effects (e.g., neurological effects) for residents of this property.
- Given an estimated exposure duration of two years and available sampling data, the opportunities for past exposures to MTBE in the drinking water of the other Hartsville residences (1352 Hartsville/Mill River, 1534G, 1534H, 1368 Hartsville/New Marlborough and others) would not be expected to have resulted in noncancer adverse health effects.
- With regard to additional cancer concerns that may have beenpresented by MTBE, this issue is under active discussion within the scientific community. At the time of this PHA no classification with regard to carcinogenicity has been determined by any agency or organization that provides these types of classifications (e.g., NTP, EPA, IARC). In addition, because no slope factor has been developed, it is not possible to quantitatively estimate cancer risks. This introduces some uncertainty into this health assessment at this time with regard to cancer concerns.
- Monitoring for the contaminated drinking water wells before and after the installation of the carbon treatment units showed that MTBE levels at the post-carbon stage were below comparison values. This indicates that the treatment systems worked well in reducing contaminant levels in drinking water. Therefore, following the installation and demonstrated effectiveness of these point of use treatment systems, and subsequent completion of site clean-up, opportunities for exposure to contaminants in drinking water (via ingestion, inhalation, and dermal contact through showering or other household activities) are not likely to have occurred and, thus, adverse health effects would not be expected.
- Cancer incidence information available from the Massachusetts Cancer Registry for the years 1995 through 1999, the most current years for which such information is available, was examined for the Town of New Marlborough. The available cancer incidence data for those cancers reported in human or animal studies in relation to MTBE (i.e., testicular, kidney/renal pelvis, liver, leukemia, lymphoma) and benzene (i.e., leukemia, multiple myeloma) do not suggest an unusual pattern or that environmental exposures were likely to have played a primary role in their development.
ATSDR requires that one of five conclusion categories be used to summarize findings of health consultations and health assessments. These categories are: 1) Urgent Public Health Hazard, 2) Public Health Hazard, 3) Indeterminate Public Health Hazard, 4) No Apparent Public Health Hazard, 5) No Public Health Hazard. A category is selected from site-specific conditions such as the degree of public health hazard based on the presence and duration of human exposure, contaminant concentration, the nature of toxic effects associated with site-related contaminants, presence of physical hazards, and community health concerns.
Based on ATSDR's public health hazard category classification, the drinking water wells at 8 Hatchery Road and 1553 Hartsville/New Marlborough Road posed a public health hazard in the past. At 8 Hatchery Road, a no apparent public health hazard exists currently. Under current conditions, 1553 Hartsville/New Marlborough Road would also pose no apparent public health hazard. These categories are assigned to these properties because human exposure to contaminated media has occurred in the past, but the exposures currently are below a level of health hazard (ATSDR 1993).
Based on ATSDR's public health hazard category classification and available scientific information, the drinking water wells on other properties evaluated in Hartsville pose no apparent public health hazard. However, it is important to note that a great deal of uncertainty exists in the toxicological database for MTBE, especially for long-term exposure via drinking water. Properties that had MTBE levels exceeding drinking water guidelines or standards were estimated to have opportunities for exposures at levels at least 100 times lower than any level shown to cause adverse health effects in animals or humans. Uncertainties with regard to cancer have been discussed above.
- MDPH, upon request, will offer to conduct individual medical record evaluations for those who suspect they have illnesses that may have been related to opportunities for exposure to contaminants in drinking water. This may be important given the current uncertainties in the literature related to health effects associated with exposure to MTBE.
- Upon request, MDPH will offer to review individual medical records for those who suspect they have illness that may have been related to opportunities for exposure to contaminants in drinking water.
- Upon request, MDPH will evaluate any additional environmental data to address public health concerns.
1 Legal departments at both ATSDR and MDPH approved use of actual addresses for this public health assessment.
2 On March 24, 1999, MDPH was notified in writing from a resident of this house that the residents had lived at this address from March 1, 1997 to November 15, 1998.
3 The VOCs that were analyzed for four indoor air samples taken at 8 Hatchery Road were MTBE, vinyl chloride, 1,1-dichloroethene, methylene chloride, chloroform, 1,2-dichloroethane, 1,1,1-trichloroethane, benzene, toluene, trichloroethene, tetrachloroethene, m&p-xylene, o-xylene, trans-1,2-dichloroethene, cis-1,2-dichloroethene, and ethylbenzene.
6 Cancer Risk=(Exposure Dose) x (EPA's oral slope factor) x (Exposure Duration)
Cancer Risk (benzene)=(0.02 mg/kg/day) x (0.055 mg/kg/day-1) x (0.024)= 2.6 x 10-5