PUBLIC HEALTH ASSESSMENT
OSSINEKE, ALPENA COUNTY, MICHIGAN
Chemicals of concern at the site were selected by comparison of concentrations in the various media with health-based comparison values, including the following:
- ATSDR Environmental Media Evaluation Guides (EMEGs)
ATSDR Reference Dose Media Evaluation Guides (RMEGs), derived from the U.S. EPA Reference Dose (Chronic) by ingestion (RfD), based on a child exposure and pica behavior for soil ingestion.(1)
ATSDR Cancer Risk Evaluation Guides (CREGs)
U.S. EPA Drinking Water Health Advisories (Lifetime).
U.S. EPA Safe Drinking Water Act Maximum Contaminant Limit Goals (MCLGs)
U.S. EPA Safe Drinking Water Act Maximum Contaminant Limits (MCLs)
EMEGs and RMEGs are concentrations calculated so that a person exposed to a medium containing these concentrations under very conservative assumptions (child exposure, with pica behavior for soil exposure) is not likely to experience a dose of the chemical in excess of the Minimal Risk Level (MRL) or RfD, respectively. MRLs, developed by the ATSDR, and RfDs, developed by the U.S. EPA, are levels of exposure at which no adverse non-cancer health effects are expected to occur. CREGs are concentrations calculated from the U.S. EPA slope factors, such that if 1 million people are exposed for their lifetimes (70 years) to an environmental medium containing a carcinogen at a concentration equal to the CREG, one additional case of cancer might result above that experienced by an equivalent population not exposed to the carcinogen. The U.S. EPA considers that MCLGs represent levels at which no known or anticipated adverse health effect should occur, with an adequate margin of safety. MCLs are regulatory concentrations that the U.S. EPA considers protective of human health (considering the availability and economics of water treatment technology) over a lifetime (70 years) consumption of 2 liters of water per day.
Chemicals whose concentration in some medium exceeds one of these comparison values, or for which comparison values are not available, are retained as contaminants of concern and listed in Table 1 in Appendix B.
To identify facilities that could contribute to the environmental contamination near the Ossineke Ground-water Contamination site, the MDPH searched the Toxic Chemical Release Inventory (TRI) database for 1987 through 1993. The TRI is compiled by the U.S. EPA from chemical release information provided by industries. Searching the TRI for facilities in the Ossineke zip code (49766) did not find any listings.
Unless stated otherwise, environmental data in this section is taken from the RI report (3).
For the purposes of this public health assessment, the site is considered to be bounded by the south side of Nicholson Hill Road, the back lot lines of the residences along the east side of Alphonse Street, Old Ossineke Road, and the north side of the laundromat and its lagoon (see Figure 2).
Groundwater Monitoring Wells
The MDNR installed 14 monitoring wells in the upper aquifer around Area 1 at the site in 1985 (7). During the RI in 1989 and 1990, U.S. EPA contractors installed 20 additional monitoring wells in the water table aquifer, and resampled 10 of the 14 wells from the 1985 sampling. Two of the older wells could not be located and two others could not be used. Eighteen of the new wells were located around or downgradient of Area 1, one was located east of the laundry, and the last east of the lagoon. The maximum concentrations of contaminants of concern in water samples from the monitoring wells in 1985 and during the RI are listed in Table 2. The RI contractors also sampled the groundwater in soil borings (Table 3) and by "in-situ water-table sampling". "In-situ water-table sampling" (so named in the RI) was done by driving a hollow rod into the ground to below the water table, pulling the rod back a short distance, then sampling the water that flowed into the hole. The concentrations of contaminants of concern found in these samples are listed in Table 4.
Data from monitoring wells, soil borings, and in-situ groundwater sampling shows that a plume consisting primarily of benzene, ethylbenzene, toluene, and xylenes (BETX) extends in a north-northeasterly direction from the center of Area 1 (Figure 2). Figure 2, a contour map of the total concentration of benzene, toluene, ethylbenzene, and xylenes from in-situ groundwater sampling in 1989, shows several discrete areas of very high concentrations within the plume. Each area of high concentration corresponds to the former location of an underground storage tank. Polynuclear aromatic hydrocarbons (PAHs), such as benzo(a)anthracene, benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(a)pyrene, benzo(g,h,i)perylene, chrysene, indeno(1,2,3-cd)pyrene, and phenanthrene, were only found in water from one monitoring well on the western edge of the study area (Table 2). The monitoring wells near the laundry and lagoon contained low levels of 1,1-dichloroethane (up to 0.39 ppb), 1,2-dichloroethane (1.5 ppb), and 1,1,1-trichloroethane (0.84 ppb). Only three other monitoring wells, in two clusters north of Area 1, contained any 1,2-dichloroethane, up to 0.84 ppb, and none of them contained any 1,1-dichloroethane or 1,1,1-trichloroethane.
In the 1989-90 sampling for the RI, lead was found in the groundwater at concentrations as high as 304 parts per billion (ppb) in unfiltered samples and 46.3 ppb in filtered samples. Organic lead (so identified in the RI report) was found very occasionally, at up to 1.1 ppb. The 29,900 ppb concentration of copper measured in one unfiltered sample appears to be an outlier, that is, an unusually high measurement that is presumed to be uncharacteristic of the area. The second highest copper concentration found was 453 ppb.
Groundwater Residential Wells
Residential wells in the site area have been extensively sampled and analyzed by county and state health departments, the MDNR, and U.S. EPA since 1977 (Table 5). Of 62 private residential or commercial wells sampled (including 12 replacement wells in the deeper aquifer) 27 have contained some detected contamination, 12 with at least one contaminant at a concentration above the comparison value. All the contaminated residential wells were in the shallow aquifer and were less than 30 feet deep. The deeper replacement wells contained no detectable contamination. The contaminated residential wells have since been abandoned, sealed, and replaced with deeper wells.
The RI contractors collected water samples from 6 residential wells (shallow, water-table aquifer) near the site in 1989 and analyzed the samples for benzene, toluene, xylenes, ethylbenzene, PCE, and 1,2-dichloroethane. No detectable levels of these chemicals were found in any of these samples. The contractors also sampled three residential wells near the laundromat in 1990. Two of these contained traces (less than 0.5 ppb) of 1,1-dichloroethane, 1,2-dichloroethane, and 1,1,1-trichloroethane, and less than 1 ppb of toluene. The presence of the last-named contaminant may have been a result of field or laboratory contamination, according to the RI report. The third well did not contain any detectable contaminants.
Surface water and sediment
The MDNR collected surface water samples from the laundromat washwater pond in 1979, the U.S. EPA Field Investigation Team (FIT) collected water samples from the pond in 1987, and the RI contractors collected water samples from the pond and from a shallow transient pond between the laundromat and the washwater pond in 1989. The major contaminants of concern detected in these samples were benzene and PCE, as much as 100 ppb and 1,200 ppb, respectively, in 1979, but none has been detected in later samples (Table 6). The high zinc concentration noted in a filtered sample (149 ppb) appears to be a laboratory artifact. The corresponding unfiltered sample contained 47.8 ppb zinc, and the highest zinc concentration in an unfiltered sample was 73.4 ppb. One would expect that the highest concentrations of inorganic chemicals would be found in unfiltered water samples.
The U.S. EPA FIT in 1987 and the RI contractors in 1989 each collected sediment samples from the washwater pond. Several contaminants of concern were found (Table 7). Sediment collected in 1989 from the transient pond contained no detectable site-related contaminants.
A sediment sample collected in 1989 from a depression behind the Marathon gas station contained 0.180 parts per million (ppm) benzoic acid and 0.110 ppm butylbenzylphthalate. Neither of these chemicals is present anywhere on the site at a concentration in excess of its comparison value (benzoic acid RMEG 8,000 ppm in soil; butylbenzylphthalate RMEG 400 ppm in soil).
The depth of the "surface soil" samples collected for the RI is not specified in the Work Plan (8) or the final RI report (3). ATSDR guidance recommends that data from the top 3 inches of soil be provided, with the top 6 inches being acceptable, for accurate assessment of health hazards from contact with the soil. There is not enough information available to judge the adequacy of the RI "surface soil" data for this purpose.
The RI contractors collected a total of 15 surface soil samples from 5 locations: 4 on the site and 1 was off-site to provide background values. As shown in Figure 3, three of the locations were behind the Marathon gasoline station, one was behind the laundromat, and the background location was near Sanborn School, southwest of the site. The contractors collected two grab samples and one composite sample from each location. The maximum concentrations of contaminants of concern found are listed in Table 8. Only three samples, one from behind the gasoline station and two from behind the laundromat, contained any significant concentrations of volatile organic chemicals (VOCs). Of the contaminants of concern, the samples from behind the laundromat contained only trichloroethylene and toluene. None of the contaminant concentrations found approach the comparison values for the chemicals, except for 1,1,1-trichloroethane, for which there is no comparison value available for soil (Table 8). Surface soil sample analysis for inorganic chemicals during the RI did not meet laboratory protocols and therefore the results were not considered acceptable by the RI contractor and were not included in the RI report.
The RI contractors drilled 16 borings on and around the site to sample subsurface soil and to provide information on the area's geology. The borings were from 16 to 40 feet deep, and soil samples were collected at 5 foot intervals above the water line in each boring. The concentrations of contaminants of concern found are listed in Table 9. All the PAHs were found in samples from one boring, near the western edge of the study area. No significant concentrations of PAHs were found anywhere else on the site. The monitoring well mentioned above in which the water contained PAHs was adjacent to this boring. However, water collected from this boring did not contain any detectable PAHs.
Analysis of subsurface soil samples collected during the RI found no inorganic chemicals above background levels, except that organic lead was found in subsurface soil samples at up to 1,500 ppb (estimated, below the contract required detection limit but above the instrument detection limit). Organic lead was detected in 8 of 19 samples collected, primarily from borings at or immediately downgradient of suspected areas of gasoline contamination. Two samples from a boring upgradient of the source area, in the southwest corner of the intersection of Nicholson Hill Road with U.S. 23, did contain organic lead, estimated at the instrument detection limit of 500 ppb.
The 1985 MDNR hydrogeological survey of the site included sampling of soil gas from approximately 80 locations around the site area. The gas samples were analyzed using a photoionization detector, which is a general organic vapor detector that does not analyze for individual chemical species. The survey report (9) gives only relative levels, without specifying the concentrations detected. The MDNR identified two areas of high levels, twice or more their background levels, one in approximately the same location as the central area of high concentration in Figure 2 and one near the corner of Alphonse and Larose Streets.
The RI contractors collected samples of soil gas from 36 locations across the site area. The maximum concentrations of chemicals found are listed in Table 10. Benzene, heptane, 1,2-dichloroethane, ethylbenzene, m- and p-xylenes, and nonane were also analyzed for but were not detected at a detection limit of 1 µg/L. Only decane (24 locations) and phenol (16 locations) were found at more than two locations. The highest decane concentrations were found along Alphonse Street, in the same area that elevated levels of organic vapors were found in the 1985 MDNR survey. Phenol was primarily detected on the periphery of the site, and there was little correlation between phenol levels and decane levels. Only 8 samples included both chemicals at detectable levels. The soil gas composition does not correlate well with the contamination plumes found in the groundwater shown in Figure 2. At the location of the one soil gas sample that contained detectable toluene, no toluene and between 1 and 10 ppb of the other BETX compounds were detected in the in-situ groundwater. At one of the two locations where o-xylene was detected in the soil gas, no o-xylene and less than 1 ppb of other BETX compounds were detected in the in-situ groundwater. Several soil gas samples collected along the periphery of the site area contained decane concentrations (above 10 µg/L) that were substantially higher than levels found closer to the source areas. This suggests there may be other off-site sources for the decane in the soil gas.
The RI contractors collected air samples from basements of structures around the site in two phases, one set that used an on-site mobile laboratory (CSL) for analysis, and one set for which the samples were sent off to another laboratory (CLP). The CSL analyzed the air samples from 16 basements in the site area (10) for benzene, heptane, 1,2-dichloroethane, toluene, octane, tetrachloroethylene, xylenes, nonane, decane, and phenol. Only decane and phenol were detected at measurable levels, and in the samples from only 7 of the basements (detection level 1 µg/L). The maximum levels found are listed in Table 11. There was no significant pattern in which locations had detectable levels of chemicals. Three locations where chemicals were detected were isolated, in that locations near them had no detectable levels. These locations included three of the four highest decane and phenol concentrations found. The other four locations where chemicals were detected were more or less contiguous properties, three along the south side of LaRose Street and one east of Alphonse Street south of LaRose. Locations on the north side of LaRose Street contained no detectable chemicals. The two central locations of the four had lower decane concentrations than the end locations. This suggests there may be multiple sources for these chemicals. The basement air concentrations did not correlate well with the soil gas concentrations. With few exceptions, the decane concentrations in the basement air were either substantially higher or lower than those in soil gas sampled nearby. The high decane concentrations in basements were found along U.S. 23, with lower concentrations along Alphonse Street. The reverse was true for soil gas sample results.
The CLP laboratory analyzed air samples for VOCs from 13 basements in the site vicinity, 10 within the area affected by the groundwater contamination (including 9 which were also sampled in the CSL sampling) and 3 to the south or west for background values. Phenol, decane, and other hydrocarbons that the CSL analyzed for were not included in the CLP analysis. The contaminants found in the samples from the site area at levels that exceeded comparison values are listed in Table 11. Benzene, tetrachloroethylene, toluene, and xylenes were detected in CLP samples but not in CSL samples from the same locations, possibly due to the difference in detection levels (as low as 0.0002 µg/L for the CLP vs. 1 µg/L for the CSL). Three of the locations that had detectable levels of chemicals in the CSL sampling were not included in the CLP sampling, including the two with the highest detected decane and phenol concentrations. No pattern was apparent in these results to relate them to the contamination in other media at the site. Benzene, carbon disulfide, and toluene were found in a field blank sample at higher concentrations than were found in the basement samples, raising questions about the validity of detections of these chemicals in the CLP basement air samples.
Surface water and sediments
The RI contractors collected surface water and sediment samples from the wetlands east of the laundromat in 1989 (SW001-003 and SD001-003 in Figure 3). The concentrations of contaminants of concern found in water from the wetlands are listed in Table 12. The maximum reported lead and zinc concentrations in filtered samples (23.7 ppb and 10,900 ppb, respectively) appear to be sampling or laboratory artifacts. The concentrations are higher than those found in any unfiltered sample from the wetlands (maximum 20.7 ppb and 98.7 ppb, respectively). The unfiltered sample from the same location as the filtered sample with the high concentrations contained 2.8 ppb lead (with blank contamination noted) and 53.7 ppb zinc. The other filtered samples from the wetlands contained less than 2 ppb lead and 9.4 ppb zinc. Maximum concentrations of contaminants of concern found in the wetlands sediment samples are listed in Table 13. The sediment samples were analyzed for organic chemicals only. As seen in Table 13, the concentrations of chemicals detected were below available comparison values. No comparison values are available for benzyl alcohol or cis-1,2-dichloroethylene. The total 1,2-dichloroethylene concentration was reported, without identifying which isomer was present.
The RI contractors also collected two water samples from the South Branch of the Devil's River, one where Nicholson Hill Road crosses the river, approximately 0.3 mile west of U.S. 23, and one approximately 0.5 miles downstream from that sample, approximately 0.4 miles northwest of the laundromat (SW008 and SW007 in Figure 3, respectively). The sample from SW008 contained no detectable chemicals, and the sample from SW007 contained 3 ppb 1,1,1-trichloroethane, 3 ppb trichloroethylene and 4 ppb 2-butanone. These were the only contaminants of concern detected, and the contractors described them as being "of unknown origin" (Reference 3, p. 4-11). These are probably not related to the site, since the groundwater at the site flows to the north-northeast and the sampling point on the river is northwest of the site. Surface runoff from the site is more likely to run to the east into the neighboring wetlands rather than west across the highway and other obstacles between the site and the river. None of the concentrations found in the river water exceeded the comparison values.
During the RI, the contractors collected air samples from three basements located to the south or west of the site to obtain information on background values. These samples were analyzed by the CLP laboratory. The contaminants found in these samples at levels that exceeded comparison values are listed in Table 14. Comparing the values in Table 14 with those in Table 11 shows that toluene, xylenes, acetone, chloroform, carbon tetrachloride, trichloroethylene, and 4-methyl-2-pentanone were detected in the background samples at concentrations comparable to (higher than 75% of) those found in the site area. The presence of these chemicals in the basement air may more likely be related to occupant activities, household products, and building materials in the basements than to the environmental contamination in the area. Benzene, carbon disulfide, and toluene were found in a field blank sample at higher concentrations than were found in the basement samples, raising questions about the validity of detections of these chemicals in the CLP basement air samples.
In preparing this Health Assessment, the MDPH relied on the information provided in the referenced documents and assumed that adequate quality assurance and quality control measures were followed with regard to chain-of-custody, laboratory procedures, and data reporting. The validity of the analysis and conclusions drawn for this Health Assessment is determined by the reliability of the referenced information.
The identification of concentrations as estimates and other quality assurance/quality control comments cited in this report were taken from the source documents. The measurements identified as estimates were either above instrument detection limits but below reliable method detection limits, or some quality control parameter was outside acceptable limits. We have also noted a few instances of apparent laboratory artifacts that were not noted in the sources, specifically lead and zinc concentrations in filtered surface water samples that were higher than those in the corresponding unfiltered samples. We added "B" qualifiers to the basement air concentration tables (Table 11 and Table 14) though the RI Report (Reference 3) did not, to indicate those concentrations that did not exceed twice the concentration found in a field blank sample listed in the RI Report. The concentrations of benzene, carbon disulfide, and toluene in the field blank were higher than those in the samples from the basements.
There were several shortcomings in the design of the sampling and analysis conducted during the RI. The water samples from residential wells were analyzed for a limited subset of the chemicals that other groundwater samples were analyzed for. The second round of basement air sampling omitted the locations that had the highest concentrations of chemicals in the first round. Surface water and sediment samples from the Devil's River downgradient (northeast) of the site would have been desirable.
There are no known physical hazards associated with this site. There is no restriction on public access to any part of the site.
To determine whether nearby residents are exposed to contaminants migrating from the site, ATSDR evaluates the environmental and human components that lead to human exposure. An exposure pathway contains five major elements: a source of contamination, transport through an environmental medium, a point of exposure, a route of human exposure, and an exposed population.
An exposure pathway is considered a completed pathway if there is evidence that all five of these elements are or have in the past been present. A pathway is considered a potential pathway if one or more of these elements is not known to be or have been present, but could be or have been. An exposure pathway can be eliminated from consideration if one of the elements is not present and could never be present. The following sections discuss the most important exposure pathways at this site.
Groundwater in the Shallow Aquifer
Groundwater in the shallow, water-table aquifer at the site is contaminated with volatile organic chemicals. The contamination has reached several residential wells in the site area, as shown by analysis of water samples from the wells. Residents used water from these wells for household purposes, including drinking, cooking, and washing. These residents and visitors to their homes were exposed to the contaminants in the water by ingestion, by dermal contact, and by inhalation of volatile contaminants secondary to household use. All known contaminated wells have been removed from service and replaced with new, deeper, uncontaminated wells, either by the householders themselves or by the MDPH. One resident continues to use water from his shallow well for bathing and bottled water for drinking because the deeper aquifer under his property does not supply sufficient water for household use. There are also anecdotal reports that suggest that wells that had once been replaced have been brought back into use (12).
Figure 2 shows that low levels of contamination have been found in the groundwater throughout the site area. The area where the spills actually occurred is zoned for commercial use, while that to the north and east, where the contaminant plume has extended is zoned for residential use (13). There are vacant properties in the area, and no community water supply is available. Hence, any new residence will require a new well. The Michigan Public Health Code requires that wells for drinking water be screened at depths of at least 25 feet below the surface. The affected aquifer is 30 feet deep in places, so it is possible to install a legal drinking water well that taps the contaminated aquifer. The Alpena County Health Department has advised all well drillers working in the site area to follow the well construction specifications that MDNR requires for wells replaced under Act 307, that is, the wells are to be screened in the lower aquifer and cased solidly into the confining layer. However, the County has no residential well-permitting program to enforce this advice.
The household water supplies for some Ossineke residents contained contaminants from 1977 until early 1987, when the last contaminated residential wells were replaced with uncontaminated ones. It is not known when the contamination first reached any private well, because there is no record of sampling that predates the first complaints in 1977. Once the contamination was identified and publicized, various actions to reduce the exposure to the contamination were implemented. The local health department advised against use of water from the contaminated aquifer as soon as the contamination was verified in 1977, though there is no record of the response to this advisory. Area residents began replacing the contaminated wells with wells into the deeper, uncontaminated aquifer in 1979, though the last replacement well was not completed until 1987. Some residents used bottled water for drinking and food preparation before their wells were replaced. Table 1-1 in Reference 3 lists 10 residences, 3 businesses, and 1 government facility in the site area that have had their wells replaced. Based on the average of 3.8 person per residence, the exposed population was approximately 40 persons.
Volatile components of gasoline leaking from underground storage tanks can evaporate into the spaces between soil particles. Volatile components of groundwater contaminants could also contribute to the chemical content of soil gas. The soil gas could diffuse through the soil and enter a basement through cracks in the walls or floor. People, residents or visitors, using the basement could inhale the soil gas, and be exposed to the chemicals. There are volatile contaminants in the groundwater, some of which have been found in the soil gas, and various chemicals have been found at levels of health concern in the air of basements on and near the site. The evidence is not strong that the scenario described above actually occurs, however. The environmental data are not well correlated spatially, there is a suggestion of multiple sources, and some of the chemicals were found in similar concentrations in the air of basements some distance from the source areas on the site. The presence of many of these chemicals in residential air at concentrations comparable to those found in the site area has been well documented (14). In addition, a field blank sample contained several of the chemicals at concentrations comparable to those found in the basement air samples, raising doubts about the reliability of the reported concentrations. The air from 16 of the 20 basements sampled in the RI contained a detectable amount of at least one chemical. Because all 13 houses sampled in the CLP round contained chemicals at detectable levels, including 3 background locations and 5 that had not contained any detectable chemicals in the CSL round, it is conservative to assume that the air in any basement in the area might contain chemicals. Figure 1-2 in Reference 3 lists 24 residences within the site area. Including the 3 background locations, assuming 3.8 inhabitants per residence, there are approximately 100 people exposed via this pathway.
Surface Water and Sediments
Contaminants could migrate via the groundwater and discharge into the wetlands, into the Main Branch of the Devil's River, or into Lake Huron downgradient from the site. The washwater pond has no outlet but contaminants in its water may migrate by infiltration to the groundwater or evaporation into the air. Sediments in the surface water bodies may retain the chemicals, however, and release them slowly to the water.
None of the affected surface water bodies near the site the washwater pond, the transient pond, and the wetlands east of the site is used as a water supply source. Direct contact with and incidental ingestion of contaminated water are the most probable pathways for human exposure to contaminants in these water bodies.
The wetlands downgradient of the site have virtually no standing surface water. It is possible that hunters and hikers in the wetlands could be exposed to chemicals transported from the site through direct contact with and incidental ingestion of contaminated water and sediment. These exposures are not of public health concern considering the low levels of contaminants likely to be present and the probable low frequency of exposure to humans via this pathway.
The South and Main Branches of the Devil's River and Lake Huron are used for recreation, including swimming, fishing, canoeing, and boating. People engaged in such activities may experience dermal contact with and incidental ingestion of the water and/or sediments. The RI investigators found low concentrations of 1,1,1-trichloroethane, trichloroethylene, and 2-butanone in water from the South Branch of the Devil's River at a point upstream of the probable discharge area for groundwater from the site. The source for these chemicals in the River is not known, and the concentrations found are below comparison values.
Groundwater from the site may discharge into these water bodies, and carry site-related contaminants into the water and sediments. People using the River or Lake may thereby be exposed to site-related contaminants through exposure to water or sediments. However, dilution and evaporation on discharge should reduce contaminant concentrations below levels of concern. No site-related contaminants have been detected in the wetlands. The RI did not include any sampling of the River in the area where groundwater from the site is likely to discharge or of the Lake.
The nearest public water supply intakes on Lake Huron are located at Alpena, 10 miles north of the site, and East Tawas, 45 miles south. Over these distances, dilution and evaporation would prevent contaminants related to this site from reaching the intakes in significant quantities. According to MDPH Division of Water Supply personnel, no contamination of any sort has been detected in the Alpena city water supply during routine sampling to monitor the quality of the water supply (15).
Groundwater in the Deeper Aquifer
There is some potential that contaminated water could migrate from the shallow aquifer to the deeper one, through natural cracks or borings for wells in the deeper aquifer. Once in the deeper aquifer, the contaminants could migrate to residential wells. People using these wells could then be exposed to the contaminants through ingestion, dermal contact, or inhalation. There is no evidence that this pathway is or has been complete, and there is little possibility that it would become complete in the future. The deeper confined aquifer has shown no sign of contamination. Available information indicates that the clay layer separating the aquifers is continuous beneath the site. This lower aquifer is also under artesian pressure, which should reduce the possibility of downward flow and prevent contamination of the lower aquifer. If the casing of a well that passes through the upper into the lower aquifer should fail, the artesian pressure should prevent migration of contaminated water from the upper aquifer into the well.
Volatile chemicals, such as benzene, toluene, xylenes, and PCE, evaporate quickly from any surface water body. The concentration of PCE in the washwater pond decreased greatly between 1979 and 1987. Volatile chemicals that evaporate in the open air usually disperse rapidly. The weather in the vicinity of the site is not known for temperature inversions or other conditions that would inhibit dispersion and allow gases to accumulate in the ambient air.
Inhalation exposure to chemicals released directly in outdoor air is not likely to reach levels of concern because of dilution of the chemicals in the air and absorption to the surface soil and water.
The County Health Department has received no further complaints concerning gasoline odors in residences since those in 1982, which have been attributed to a specific incident of gasoline spillage.
This assessment evaluates the health hazard from exposure to chemicals in the environment by comparing the amounts of the chemicals an individual is likely to be exposed to (the exposure dose) with various health guidelines. The primary guidelines used for consideration of non-cancer health effects are the Minimal Risk Levels (MRLs) developed by the ATSDR and the Reference Doses (RfDs) and Reference Concentrations (RfCs) developed by the U.S. EPA. Exposure to chemicals at levels below the MRLs, RfDs, or RfCs is not considered likely to result in adverse non-cancer health effects. The MRLs, RfDs, and RfCs may not be protective for hypersensitive individuals, such as the very young, the very old, those whose bodies are under stress, or those with an allergic reaction to the chemical.
To evaluate the risk of contracting cancer from an exposure to a chemical, we use the U.S. EPA's slope factors with the exposure dose to estimate the increased chance of cancer occurring over a lifetime in a population experiencing the dose. We call the increased risk significant if we estimate that, in a population of one million people experiencing the exposure dose, one (or more) additional cancer cases will result.
The most important completed exposure pathway at this site is household use, including consumption, of contaminated groundwater. As described in the Pathways Analysis section above, the duration of the exposure varied widely, and is not entirely certain. In addition, the concentrations of contaminants in any one well varied sharply from one sampling to the next.
This toxicological evaluation considers a reasonable worst case scenario, that is, an individual whose primary water supply contained the highest concentrations of contaminants found in the groundwater for 10 years. We have selected the concentrations from the full set of data available, including residential wells, monitoring wells, and "in-situ" sampling. The contaminated residential wells and the monitoring wells were screened in the same aquifer, and are geographically interspersed, with no known obstructions to groundwater flow between them. The person considered to be using the water is an adult, weighing 70 kilograms (155 lbs.) and drinking 2 liters (2 quarts) of water per day. Additionally, a child, weighing 10 kilograms (22 lbs.) and drinking 1 liter (1 quart) of water per day is considered as representing a more sensitive population. Exposure to chemicals volatilizing from water during household use is assumed to equal the exposure from drinking water. The maximum concentration of volatile chemicals in the air of a residence is calculated from the model for a shower in Reference 16.
Dermal contact with chemicals in the groundwater during bathing is another possible exposure route. Standards are not available to evaluate the potential health impact from dermal exposure to most chemicals.
To estimate long-term exposure to air in a basement, this evaluation assumes an adult spending 2 hours a day in the basement. This is an original estimate, based on the assessor's personal experience. The time an individual spends in the basement of his residence varies widely, from negligible time to 8 hours or more per day. The latter represents the case of a finished basement that includes a bedroom. There is no information readily available on how the residents used the basements where the air samples were collected. There is similarly no information on whether the basements and the rest of the houses were cross-ventilated. As a conservative measure, the highest detected chemical concentrations in basement air on and near the site are used, regardless of the location and regardless of whether blank contamination may have affected the sample.
Recreational use of the South Branch of the Devil's River could result in exposure to the chemicals that were found in the water. However, the concentrations found of the chemicals were below comparison values. Exposure to contaminated water during recreation, through dermal contact and incidental ingestion, is likely to be infrequent. It is highly unlikely that exposure to the river water will result in any adverse health effects connected to the chemicals found in the water.
The contaminants of concern that were found most frequently or have the greatest potential for public health impact at this site include benzene, toluene, ethylbenzene, xylenes, styrene, tetrachloroethylene, trichloroethylene, 1,2-dichloroethane, 1,1,1-trichloroethane, and lead. We discuss these individually below, to give a general overview of the health hazard from the site. Other contaminants of concern listed in Table 1 were found at the site infrequently, and at comparatively low levels, and we will not discuss specific health effects related to exposure to them here.
Benzene, Ethylbenzene, Toluene, Xylenes (BETX)
There are no ingestion MRLs, RfDs, or RfCs available for benzene. The benzene concentration in the groundwater at the site did exceed the MCL. The benzene concentration in the air during a shower using the most contaminated groundwater at the site would exceed the MRL for exposure of less than 14 days duration. The exposure doses by ingestion are less than the levels at which adverse health effects have been seen in humans or laboratory animals. Long-term, sustained breathing of air containing the maximum benzene concentration which might be in the air during a shower using the most contaminated groundwater at the site has resulted in damage to the blood, immune system, and nervous system in humans and laboratory animals. The amount of benzene breathed in during a single, 10-minute shower once per day is less than the daily doses at which adverse health effects have been observed, though it is higher than the level at which one can be confident that no non-cancer adverse health effects are likely to occur. The benzene concentration in the air of one basement on the site exceeded the MRL for exposure lasting less than 14 days. The concentration was much less than that at which adverse effects have been observed in laboratory animals or humans. Pure liquid benzene is an irritant to the skin and eyes. Such effects are unlikely from skin contact with solutions of benzene at the concentrations found in the groundwater at the site (17).
Studies have documented that workers who inhaled benzene in their workplace developed more leukemia than other workers. Leukemia and cancer of the Zymbal gland, mouth, breast, liver, and other organs has also been reported in laboratory animals exposed to benzene in their food. The International Agency for Research on Cancer (IARC), the U.S. Department of Health and Human Services National Toxicology Program (NTP), and the U.S. EPA all consider benzene a known human carcinogen (IARC Class 1, U.S. EPA Class A). Lifetime consumption of or regular showering in water containing the highest benzene concentration in the groundwater at the site might result in a substantially increased risk of contracting leukemia. A person who spends two hours a day for a lifetime breathing air containing the highest concentration of benzene found in basements near the site might incur a significantly increased risk of contracting leukemia.
Anyone whose primary drinking water supply contains the maximum concentration of toluene found in the groundwater at the site might ingest enough toluene to exceed the RfD for non-cancer health effects. The exposure dose by ingestion is below the dose at which adverse health effects have been observed in humans, though mice exposed to approximately the same dose of toluene for an intermediate length of time experienced some developmental effects. The concentration in the air in a shower using the most contaminated groundwater might substantially exceed the MRL for short-term exposure. Mild intoxication and decreased manual dexterity and visual perception have been observed in humans who breathed air containing similar concentrations of toluene to that which could be experienced in the shower.(2) However, the observed effects were the results of 6 to 6.5 hour exposures, and a 10-minute shower is not likely to be a sufficient exposure to result in these effects. The concentrations found in basement air were substantially below the MRL for acute exposure and all levels at which adverse health effects have been observed in humans or experimental animals. Pure liquid toluene is an irritant to the skin and eyes. Such effects are unlikely from solutions of toluene at the concentrations found in the groundwater at the site. There is no available evidence that exposure to toluene causes cancer (18).
A child whose primary drinking water supply contains the maximum concentration of ethylbenzene found at the site might ingest enough of the chemical to slightly exceed the RfD. The dose an adult would experience would not be likely to exceed the RfD. The expected dose is far below that at which adverse health effects have been observed in laboratory animals (21).
The ethylbenzene concentration in the air of a shower using that water might exceed the MRL for chronic exposure. The calculated air concentration does not exceed that at which adverse health effects have been observed in laboratory animals, and the daily dose from one ten-minute shower is less than the daily dose from continuous exposure to air containing the MRL. There is no evidence available that exposure to ethylbenzene causes cancer. It is unlikely that anyone would suffer adverse health effects due to the ethylbenzene in the groundwater, if it is used for drinking. Pure liquid ethylbenzene is an irritant to the skin and eyes. Such effects are unlikely from solutions of ethylbenzene at the concentrations found in the groundwater at the site.
Anyone whose primary drinking water supply contains the maximum concentration of xylenes found in the groundwater on the site might ingest enough of these compounds to exceed the RfD. However, no one is likely to ingest enough xylenes through the water to equal the doses at which adverse health effects have been observed in humans or experimental animals (22).
The xylene concentration in the air of a shower using the most contaminated groundwater at the site might exceed the concentrations at which eye irritation and neurological effects have been observed in humans on short-term exposure. The xylene concentration found in the air of basements near the site is far below that at which adverse health effects have been observed in humans or experimental animals. Xylenes, either pure isomers or mixtures, can be irritating to the skin. It is not likely that such effects would occur from a water solution of the concentrations found at the site.
There is no evidence available that exposure to xylenes causes cancer.
No one who drinks groundwater from the site is likely to ingest enough styrene to exceed the MRL for non-cancer health effects. The air of a shower using water containing the maximum concentration of styrene found at the site might contain a concentration of the chemical slightly less than the concentration that caused respiratory irritation in laboratory animals that breathed it. The concentration found in basement air near the site was substantially below that at which adverse health effects have been observed in humans or laboratory animals. Pure liquid styrene is an irritant to the skin and eyes. Dilute aqueous solutions, such as those found at this site, are not likely to cause these effects (23).
Some laboratory animals that inhaled or ingested styrene developed cancer of the breast (inhalation) or lung (ingestion). Styrene is currently under review by the U.S. EPA for classification as a possible (Class C) or probable (Class B2) carcinogen. Based on proposed U.S. EPA slope factors, lifetime consumption or use for showers of water containing the maximum concentration of styrene found in groundwater at the site may pose some significantly increased cancer risk. The styrene concentration found in the air of basements at and near the site would not pose any significantly increased cancer risk.
No one who uses the groundwater on the site as a drinking water supply is likely to ingest enough tetrachloroethylene (also known as perchloroethylene or PCE) to exceed the MRL for non-cancer adverse health effects on intermediate-duration exposure. The exposure dose from using water containing the maximum PCE concentration found in the groundwater is substantially below the lowest doses at which any adverse non-cancer health effects have been seen in laboratory animals. The PCE concentration in the air of a shower using the water from the site is also not likely to attain the MRL for acute inhalation exposure, though it might exceed the MRL for intermediate-duration inhalation exposure. The estimated shower-air concentration is far below those at which adverse non-cancer health effects have been observed in humans or experimental animals. The PCE concentration found in basement air near the site is much less than the MRLs for all exposure durations (24). Therefore, no non-cancer adverse health effects due to PCE exposure are likely to occur at this site, based on the available information.
Some laboratory animals who inhaled or ingested PCE developed liver cancer, though there is no direct evidence that exposure to PCE has caused cancer in humans. The IARC has classified PCE as a possible human carcinogen (Class 2B) and the NTP considers PCE as "reasonably anticipated to be a carcinogen". The U.S. EPA had classified PCE as a probable human carcinogen (Class B2), but is reviewing the classification. Lifetime consumption of water containing the maximum PCE concentration found in groundwater at the site may result in a significantly increased risk of cancer. A person who takes showers using water from the site is not likely to inhale enough PCE to incur a significantly increased cancer risk. Breathing air containing the PCE concentration found in the basements on or near the site for two hours per day is not likely to result in a significantly increased cancer risk. The combined inhalation risk from showering with water from the site and two hours breathing the basement air is not likely to result in a significantly increased cancer risk. Pure liquid PCE or highly concentrated solutions can irritate the skin and eyes on contact. It is not likely to have such effects at the concentrations found in groundwater at the site.
No one drinking the groundwater from the site area is likely to ingest enough trichloroethylene (TCE) to exceed the MRL for intermediate duration exposure. The exposure dose from ingestion is substantially below the levels at which adverse non-cancer health effects have been observed in laboratory animals. The estimated TCE concentration in the air in a shower using water from the site area would be substantially below the concentrations at which adverse non-cancer health effects have been observed in humans or experimental animals. The TCE concentration found in the air in basements near the site is substantially below the concentrations at which adverse non-cancer health effects have been observed in humans or experimental animals. Highly concentrated solutions of TCE can irritate and desiccate the skin. Dermal effects have not been reported for dilute aqueous solutions (25). Therefore, no non-cancer adverse health effects are likely from TCE exposure at this site, based on the data reviewed.
Some laboratory animals who inhaled or ingested TCE developed liver cancer, though there is no direct evidence that exposure to TCE has caused cancer in humans. The IARC has classified TCE as a possible human carcinogen (Class 2B) and the NTP considers TCE as "reasonably anticipated to be a carcinogen". The U.S. EPA had classified TCE as a probable human carcinogen (Class B2), but is reviewing the classification. Lifetime consumption of water containing the maximum TCE concentration found in groundwater at the site may result in a significantly increased risk of cancer. However, the TCE concentration in the air of showers using water from the site or that found in air from basements near the site is not likely to produce a significantly increased cancer risk. The combined increased cancer risk due to inhalation of TCE for someone who both showers with water from the contaminated aquifer and spends more than two hours per day breathing the basement air over a lifetime might be significantly increased.
There are no MRLs, RfDs, or RfCs for ingestion or inhalation of 1,2-dichloroethane. No one who consumes water containing the maximum concentration of 1,2-dichloroethane found in groundwater at the site is likely to ingest a dose of the chemical comparable to that at which adverse health effects have been observed in experiments using animals. The concentration of 1,2-dichloroethane in the air of a shower using that water might attain levels at which mice who breathed air containing the level for 3 hours experienced some decreased immune system response. It should be noted that other species did not experience a similar response when breathing much higher concentrations of 1,2-dichloroethane.
One epidemiological study showed an association between 1,2-dichloroethane in the drinking water supply and increased colon and rectal cancer incidence, however, there were probably other chemicals present in the water. Experimental animals who consumed 1,2-dichloroethane in their food developed increased numbers of cancers at various sites, including the lungs, liver, spleen, pancreas, adrenal gland, stomach, and mammary glands. The U.S. EPA has classified 1,2-dichloroethane as a probable human carcinogen (U.S. EPA Class B2). A person whose drinking water supply contained the maximum concentration of 1,2-dichloroethane found in the groundwater at the site for 10 years could incur a significantly increased risk of contracting cancer, based on the slope factor derived from the animal studies cited above. A person who regularly showered in this water for 10 years could incur a significantly increased risk of contracting cancer from inhaling the chemical, based on a slope factor extrapolated from the animal studies on ingestion of the chemical (26).
No one consuming groundwater or incidentally ingesting soil at the site is likely to ingest a dose of 1,1,1-trichloroethane approaching those at which non-cancer health effects have been observed in studies on humans or experimental animals. Neither the concentrations in air found in basements near the site nor in a shower using the groundwater at the site approaches the levels at which non-cancer health effects have been observed in studies on humans or experimental animals. The combined dose from drinking water, incidentally ingesting soil, and inhalation is far below the dose at which non-cancer health effects have been observed in studies on humans or experimental animals. There is no evidence available to link exposure of 1,1,1-trichloroethane to cancer (27).
Some people who participated in two health studies cited in the literature (References 28, 29, cited in Reference 30) experienced decreased activity of certain enzymes involved in blood production after a few days of consuming a daily dose of lead equal to that which could result from a child's regular use of water containing the maximum concentration found in unfiltered groundwater at the site. No one who drinks filtered groundwater from the site is likely to ingest enough lead from the water to exceed the doses at which adverse health effects were observed in those studies. Lead accumulates in the body, so long-term exposure to low doses may cause adverse health effects similar to those resulting from high doses. Exposure to lead can damage the central nervous system and kidneys and can interfere with reproduction and development (30).
Some laboratory animals who ingested some lead compounds developed kidney cancer. There is some limited evidence linking exposure to lead to cancer in workers in lead production plants. The IARC has classified lead as a possible human carcinogen (Class 2B). The NTP considers that lead should be reasonably anticipated to be a carcinogen. The U.S. EPA has proposed that lead should be classified as a probable human carcinogen (U.S. EPA Class B2) (30). There is not enough information on the cancer potency of lead available to estimate the increased cancer risk due to the lead in the groundwater at the site.
Polynuclear Aromatic Hydrocarbons
Of the polynuclear aromatic hydrocarbons (PAHs) found at the site, benzo(a)anthracene, benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(a)pyrene, and indeno(1,2,3-c,d)pyrene are classified as probable human carcinogens (U.S. EPA Class B2) (31, 32). PAHs were found at significant concentrations in only one location, and the highest concentrations were found at 5 to 7 feet below ground level. Surface samples (ground surface to 2 feet deep) contained no carcinogenic PAHs at detectable levels and other PAHs were measured at levels within background ranges (see Table 5-5 in Reference 31). PAHs, as products of incomplete combustion, are commonly found in the environment. The PAHs at the site are inaccessible, and should pose little public health concern.
No one who would consume water from the residential wells on the site or surface water from the wetlands is likely to ingest sufficient phenol to exceed the RfD. There are no RfC or inhalation MRLs for phenol. The phenol concentrations in the air from basements near the site, or that in air in a shower using water from residential wells near the site, are below the concentrations at which adverse non-cancer health effects have been observed in studies on experimental animals. There is not sufficient evidence available to determine whether exposure to phenol causes cancer (33).
No one is likely to ingest enough sediment or surface water from the wetlands or surface soil or sediment from the site to attain a dose of 2-butanone (also known as methyl ethyl ketone) to exceed the RfD. There are no inhalation MRLs or RfCs for the chemical. The 2-butanone concentration found in basement air is substantially below those at which adverse non-cancer health effects have been observed in studies of humans or laboratory animals. There is no evidence that exposure to 2-butanone causes cancer (34).
Chemicals in Soil Gas and Basement Air
Acetone, carbon disulfide, carbon tetrachloride, chloroform, 4-methyl-2-pentanone, and methylene chloride were only detected in basement air samples. None of these were detected at concentrations at which adverse non-cancer health effects have been observed in studies of humans or experimental animals (35, 36, 37, 38, 39, 40). Experimental animals who ingested carbon tetrachloride developed liver cancer (35). Experimental animals who ingested chloroform developed cancer of the liver or kidneys (36). Experimental animals who breathed air containing methylene chloride developed cancer of the liver (37). All three of these chemicals have been classified by the U.S. EPA as probable human carcinogens (U.S. EPA Class B2). The maximum chloroform and methylene chloride concentrations found in the basement air might pose a significantly increased risk of cancer to a person spending 2 hours a day breathing such air. The maximum carbon tetrachloride concentration would pose an increased cancer risk slightly below significance to someone in the basement 2 hours per day, though longer daily exposures might result in significantly increased cancer risk.
Decane was only detected in the soil gas and basement air samples. There is little published health information or health standards specifically for decane. Decane is one of a class of compounds called aliphatic hydrocarbons, which consist of a chain of carbon atoms with hydrogen atoms attached. Decane is a 10-carbon chain. These chemicals can have neurotoxic effects on people exposed to them. In general, the heavier the compound in this series, the less toxic it is, so we can extrapolate an upper limit on toxicity from data on lighter compounds. The OSHA Threshold Limit Value (TLV) for nonane, the 9-carbon member of the series, is 200 ppm averaged over an 8-hour workday. One percent of the TLV, 2 ppm for nonane, is commonly considered protective for residential exposure. The maximum decane concentration in the air of basements on the site exceeded 3 ppm. Using the TLV for nonane as a conservative assumption, the decane concentration in the basement air should pose no significant health risk for short-term exposure, though there may be some concern if a person lives and sleeps in the basement. The decane concentrations were substantially below those at which adverse health effects have been observed in studies of exposure to nonane.
Sanborn Township is sparsely populated. In 1990, the population was 2,196 with 12.5% of the population age 65 or older, and a median age of 30.4. For the state as a whole, 9.8% of the population was age 65 or older, and the median age was 28.7. Thus, age adjustment would be essential in any township-state comparisons. Because of the small population, however, age-adjusted rates are not considered reliable.
The Cancer Registry in the MDPH Office of the State Registrar and Center for Health Statistics lists cases according to the zip code of the patient, and cannot provide data by township. Sanborn Township includes small portions of several very large zip code areas and therefore, evaluation of cancer incidence rates by zip code in relation to the site is not appropriate.
Cancer mortality information was obtained from the MDPH Office of the State Registrar and Center for Health Statistics for Sanborn Township, by year, according to age, sex, and cancer site. However, because of the small population of the area, the census does not provide age breakdowns for the population suitable for computation of age adjusted rates for the years since 1980. Even if the population data were available, the small number of cases each year make it unlikely that age-adjusted comparisons could be carried out with any degree of accuracy. The National Center for Health Statistics does not consider a rate computation to be reliable unless it is based on at least 20 cases. The MDPH Office of the State Registrar and Center for Health Statistics will not calculate a rate unless it is based on at least 6 cases.
Table 15, Table 16, and Table 17 in Appendix B summarize the information on cancer deaths for Sanborn Township during 1985-91 (41). Using statewide cancer mortality rates, we would expect about 2.5 to 3 cancer deaths a year, on the average, in a population the size of Sanborn Township (2,196), or between 15 and 20 deaths over a seven year period. Sanborn Township has a higher proportion of population age 65 and older than the State as a whole does, therefore the Township could be expected to have a higher rate of cancer than the state does. Therefore, the actual seven-year total of 38 cancer deaths may not be excessive. Of the 38 deaths, 24 were persons age 65 or older (Table 15).
The 38 deaths involved at least 15 different cancer sites (Table 16). Cancer of the lung, bronchus, or trachea had the largest number (12) followed by cancer of the breast, pancreas, prostate, and no site specified (3 each). These are also among the leading causes of cancer deaths for the state as a whole (1988 mortality data).
This section addresses the health concerns as expressed by members of the community around the site.
1. There appears to be a large number of relatively young persons in the neighborhood who have had cancer. Could the contamination be responsible?
In response to the expressed community concerns about cancer, the preparers of this assessment obtained cancer mortality data for the Township (see above). As mentioned in the Health Outcome Data Evaluation section above, cancer incidence information for the Township is not available. Table 15 shows that 4 people in Sanford Township under the age of 54 died of cancer in the 7-year period between 1985 and 1991. The population in the Township is small and there has not been a large enough number of cancer cases reported for it to be possible to draw any statistical conclusions from the data. These 4 cases involved 4 different cancer sites: colon, lung, sinus, and leukemia. Exposure to benzene has been linked to leukemia, and exposure to styrene and 1,2-dichloroethane has been linked to lung cancer. Again, with the small population and small number of cases, it is impossible to say if any exposure to contaminants the individuals might have experienced is related to their cancers.
2. I have had cancer. Could drinking the contaminated water have had any effect on my current health condition?
It is extremely difficult to determine whether a specific case of cancer is connected with an exposure to a chemical in the environment. There are a large number of other factors that contribute to whether an individual develops cancer, and in most cases it is impossible to separate out and identify the contribution of any one factor in a specific case. Further discussion with the individual who raised this concern did establish that he lived on the west side of U.S. 23, outside the known limits of the contamination plume and substantially across the groundwater gradient from the plume. His well was probably not contaminated and he was probably not exposed to site-related contaminants via any of the pathways discussed above. Site-related contaminants probably did not contribute to his health problems.
The MDPH released a draft of this Public Health Assessment for public comment on December 14, 1994. The comment period lasted until January 13, 1995. The MDPH received no comments in this period.