PUBLIC HEALTH ASSESSMENT
ALBION-SHERIDAN TOWNSHIP LANDFILL
SHERIDAN TOWNSHIP, CALHOUN COUNTY, MICHIGAN
Contaminants of concern at this site were selected from those chemicals for which the concentration in at least one environmental medium exceeded a health-based comparison value. Inclusion of a chemical as a contaminant of concern does not imply that human exposure to the chemical will inevitably result in health effects. The inclusion merely triggers further evaluation of the exposure to the chemical by the health assessors. Lifetime exposure to chemical concentrations at or below the appropriate comparison values for a chemical should not result in more than 1 case of cancer in 1 million people exposed or any increase in noncancer health effects. Comparison values used in this assessment include the following, developed by the Agency for Toxic Substances and Disease Registry (ATSDR) and the Environmental Protection Agency (EPA):
ATSDR environmental media evaluation guides (EMEGs)
ATSDR cancer risk evaluation guides (CREGs)
ATSDR reference dose media evaluation guides (RMEGs), computed from the EPA reference dose (RfD) for chronic exposure of a child, assuming pica behavior for soil ingestion4
EPA Lifetime Health Advisories (LTHAs)
EPA Safe Drinking Water Act maximum contaminant levels (MCLs)
If no comparison values for a chemical in a medium exist, or no CREG is available for a carcinogen, the chemical is retained as a contaminant of concern. A list of the contaminants of concern for this site is given in Table 2.
To identify facilities that might contribute to the environmental contamination in the area of the Albion-Sheridan Township Landfill (ASTL) site, the Michigan Department of Public Health (MDPH) searched the Toxic Chemical Release Inventory (TRI) database for 1987 through 1992. The EPA compiles the TRI from reports provided by industries.
No entries for the ASTL were found in the TRI, either as source or as recipient of hazardous materials. The landfill closed in 1980, and the transfer station on the site closed in 1985, before the TRI started collecting data.
The TRI contained entries for five other facilities with the same postal ZIP Code (49224) as the ASTL site, including the Brooks Foundry, 0.5 mile west of the ASTL, mentioned previously under Demographics, Land Use, and Natural Resource Use and shown in Figure 1. The TRI entries for the Brooks Foundry were from 1987 only, and listed transport to other disposal facilities of cadmium, lead, and nickel but no environmental release of the metals. A second facility, approximately 1 mile west of the ASTL site across Clark Street from the McGraw-Edison National Priorities List (NPL) site and south of the Clark Street Wellfield (Figure 1), reported releases to the air of formaldehyde, ammonia, phenol, and ammonium sulfate solution. A third facility reported air releases of sulfuric acid from an address near the center of Albion, approximately 1.5 miles west of the site. The remaining two facilities--one reporting air and water releases of copper and manganese, air releases of methylenebis(phenylisocyanate), naphthalene, phenol, and 1,2,4-trimethylbenzene, and off-site transfers but no environmental releases of sulfuric acid, and the other reporting air releases of antimony, barium, cadmium, chromium, lead, and zinc compounds--gave addresses on the west side of Albion, 2 miles or more to the west of the ASTL site. None of the releases from any of the five facilities are likely to contribute to environmental contamination in the vicinity of the ASTL site.
For this assessment, the site is defined to be the entire original landfill property, all five parcels shown in Figure 2.
In 1980 and 1981, the Michigan Department of Natural Resources (MDNR) collected groundwater samples from three on-site monitoring wells that draw water from the upper aquifer--one north of the landfill, one at the southeast corner of the landfill, and one south of the landfill. These samples were found to contain elevated concentrations of iron, sodium, potassium, lead, magnesium, calcium, ammonia, total chromium, copper, and zinc (see Table 3) (18). The highest concentrations were frequently found in samples from the southeast corner well. However, later water-level measurements suggest that the groundwater in the area flows to the southwest, indicating that flow from any known waste area to this well would be across the groundwater gradient.
During the remedial investigation (RI) in 1992, the contractors constructed 13 additional monitoring wells within or along the site boundaries in 4 clusters--1 on the north end, 1 on the south end, and 2 along the west boundary. The well clusters generally included three wells, one screened in the shallow glacial aquifer, one in the upper, weathered layer of the sandstone bedrock, and one deeper within the bedrock. One cluster on the east side included four wells, one in the weathered sandstone and three deeper within the bedrock. The results of analyses of samples collected from these wells in late 1992 and early 1993 are listed in Table 4 (1, 3).
Organic chemicals were detected very rarely, and at concentrations marked as estimated because they were below the contract required detection limits. Of the chemicals listed in Table 4, a very low concentration of heptachlor was reported in a weathered bedrock well north (upgradient) of the landfill, and di-n-octyl phthalate was reported in the deeper bedrock well in the same cluster. The weathered bedrock well north of the site also contained a trace (an estimated 1 part per billion [ppb], below comparison values) of di-n-butyl phthalate, and a deeper bedrock well on the south end of the site contained a trace (1 ppb, below comparison values) of di(2-ethylhexyl)phthalate.
The EPA's RI contractors sampled the private well serving the residence on the south side of the site in October 1992. Arsenic and manganese were reported in the sample at concentrations above the comparison values (Table 5) (3). The concentrations of these metals were far below those found in monitoring wells around the landfill and were comparable to those found in off-site and upgradient residential wells.
The EPA's RI contractors also constructed three wells screened within the landfill to sample leachate, that is, water seeping through the landfilled material. When the contractors attempted to sample these wells in December 1992 and again in March 1993, only one contained any water. Concentrations of contaminants of concern found in the samples from this well are listed in Table 6 (1, 3). These samples contained low concentrations of various organic chemicals as well as elevated concentrations of various metals. Heptachlor was not detected, although heptachlor epoxide, a product of the environmental degradation of heptachlor, was reported at extremely low concentrations.
In 1992, the EPA's RI contractors collected surface soil samples at 11 locations on the site, including 6 locations on the landfill cap, 3 on the north end of the site, and 2 in the residential area in the southwest corner of the site (Parcel D). These samples included the top 6 inches of soil from the sampling locations. ATSDR prefers surface soil samples to be no more than 3 inches deep, to properly evaluate the potential hazard from contamination on the surface. The concentrations of contaminants of concern in the samples collected within the site fence are summarized in Table 7. The concentrations of contaminants of concern found in the residential area are summarized in Table 8 (3). Di-n-octyl phthalate was found in only one sample from the residential area, and the reference indicated that the chemical was also found in a laboratory blank (quality control) sample.
Sludge and Drum Contents
Sludge samples taken in 1980 contained high concentrations of nickel, lead, cyanide, chromium, and cadmium (see Table 9) (18). During a 1989 preliminary investigation, four samples were collected of on-site barrel contents. Three of the samples had flash points of less than 75 F, falling into the category of hazardous wastes because of their ignitability. The fourth sample contained a variety of volatile organic compounds (VOCs). In addition to the identified VOCs listed in Table 9, the sample included an estimated 53 parts per million (ppm) of 23 tentatively identified VOCs, at estimated concentrations between 1 and 8 ppm (5). None of the identified chemicals exceeded available soil comparison values, although no comparison value is available for exposure to 1,1,1-trichloroethane or any of the tentatively identified chemicals through soil.
While installing the leachate wells in the landfill, the EPA's RI contractors collected samples of the material in the landfill. Concentrations found of contaminants of concern in these samples are summarized in Table 10 (3).
No ambient air sampling has been performed at the site. During the field work for the RI, the contractors monitored the air in the workers' breathing space using a nonspecific organic vapor analyzer as part of their worker health and safety plan. When the borehole for a monitoring well within the landfill reached 10 feet below the surface, the analyzer indicated that more than 10,000 ppm of organic gases and vapors were in the air. The instrument indicated the vapors were above the lower explosive limit. Work on that boring was stopped for the day. The following morning, the vapors from the boring contained between 6,000 and 10,000 ppm organic vapors. The workers poured water into the boring, which helped reduce the organic vapor level in the vented gas to between 50 and 100 ppm, and work resumed. Organic vapors continued to be detected, although the concentrations decreased as the depth increased. The odor of landfill gases was noticed during installation of several subsidence monuments--structures used to monitor subsidence of the landfill cap surface--and during installation of a fence post near the northeast corner of the site.
For worker safety during the June 1994 test-pitting activities, the contractors sampled the ambient air with a photoionization detector (PID) and an airborne particle detector. The PID generally reported no organic chemicals in the air except during drum removal operations, when transient readings as high as 20 ppm were detected. The workers occasionally reported garbage odors, but not necessarily at the same times that the PID detected high concentrations. The particle detector one day detected as much as 2.67 mg/m3 during on-site activities, though on other days it didn't exceed 0.3 mg/m3. Workers on the site also wore organic vapor monitors, which were meant to adsorb and retain organic chemicals. After full workday exposures, none of the monitors had adsorbed any benzene, the one chemical they were analyzed for as a surrogate for all organic chemicals (19).
There is no record of any sampling or testing of consumable plants or animals that might live on the site.
In 1992, the EPA's RI contractors constructed 18 monitoring wells near the site. These included four three-well clusters, one to the northeast and three to the southwest; a two-well cluster, shallow glacial and deeper bedrock, to the east; and four shallow glacial wells in the wetlands south and southwest of the site. Concentrations of contaminants of concern found in samples collected from these wells in December 1992 and March 1993 are summarized in Table 11 (1, 3).
As with the results of samples from on-site wells, organic chemicals were found very rarely in these samples and at concentrations marked as estimated because they were below the contract required detection limits. Di-n-octyl phthalate was only found off-site in samples from three upgradient wells, the glacial and deeper bedrock wells from a cluster northeast of the site and the bedrock well east of the landfill. The samples from the wells to the northeast also contained traces (estimated concentrations less than 3 ppb, below comparison values) of butyl benzyl phthalate, di-n-butyl phthalate, and diethyl phthalate, which were not found in samples from any other off-site well. These phthalates are commonly used as plasticizers, are nearly ubiquitous in the environment, and are also common laboratory or sampling contaminants. A sample from the weathered bedrock well in the same cluster contained a trace of xylenes (a total of 4 ppb [estimated], below comparison values), the only well sample to contain any xylenes.
In 1988 and 1989, the MDPH and the Calhoun County Health Department sampled 21 private wells west of the ASTL site to analyze for VOCs as part of the study of the Brooks Foundry site. At least one sample from three of the wells contained at least one VOC. A residential well approximately 0.4 mile west of the ASTL site contained 40 ppb fluorotrichloromethane (LTHA 2,000 ppb), 2 ppb 1,1-dichloroethane, and 14 ppb 1,1,1-trichloroethane (LTHA 200 ppb) in one sampling, and 15 ppb 1,1,1-trichloroethane in a second. The Brooks Foundry well contained 1 ppb 1,1,1-trichloroethane and 2 ppb fluorotrichloromethane in one sampling, but no detectable VOCs in a second. The third well, approximately 0.8 mile southwest of the ASTL site, contained 1 ppb toluene (RMEG 2,000 ppb) and was only sampled once (5). None of the VOCs were detected at levels that exceed available comparison values. No comparison values are available for 1,1-dichloroethane, but the chemical was only detected once and its presence was not confirmed by a later sampling. The detection of these chemicals in these wells is not associated with the ASTL site.
In October 1992, the EPA's RI contractors collected samples from eight residential wells near the site and one of the Amberton Village supply wells. Sampling included three of the wells sampled in 1989 that had then contained no detectable VOCs. Concentrations found of contaminants of concern are listed in Table 12 (3). The available data show no evidence of a pattern of contamination in the residential wells in the site area. No VOCs were found in any of these wells. Heptachlor was reported in samples from three wells, two approximately 0.25 mile southwest (downgradient) from the site and the Amberton Village well northeast (upgradient) from the landfill. The contractors collected duplicate samples from the Amberton Village well; one contained 0.011 ppb of heptachlor, and the other contained no heptachlor at a detection limit of 0.01 ppb. The heptachlor concentration reported from one of the downgradient wells was flagged as an estimate at 0.011 ppb, although there was no qualifier on the reported concentration from the other well (0.022 ppb). One of the downgradient wells that contained heptachlor in 1992 had also been sampled in 1989; however, the sample collected in 1989 was not analyzed for heptachlor. Three other residential wells located near the two that reportedly contained heptachlor, including one located between the two, did not contain any of the chemical at a detection limit of 0.01 ppb.
Seven of the wells contained manganese at concentrations between 61.4 and 185 ppb. Two downgradient wells, the same ones from which the samples that were reported to contain heptachlor were drawn, did not contain manganese at a detection limit of 1 ppb. Wells located north of the landfill generally contained higher concentrations of manganese (83.3 to 185 ppb) than those located southwest of the landfill (not detected at 1 ppb to 104 ppb). Two of the wells that were also sampled in 1989 did contain manganese in the 1992 sampling (83.3 ppb and 104 ppb); however, no analysis of any trend is possible since the 1989 samples were not analyzed for manganese.
In the past, two of the Clark Street wells serving the Albion municipal water supply system, approximately 1 mile west of the ASTL site (Figure 1), have contained low levels (1-2 ppb, CREG 3 ppb) of trichloroethylene (TCE), but none of the chemical has been found in either well since 1984. The Clark Street wells are approximately 1,500 feet from the McGraw-Edison NPL site, a site of soil and groundwater contamination with TCE. Since 1984, VOCs have been detected only twice in these wells:
- Clark Street Well #1 contained 7 ppb of dichlorodifluoromethane in March 1989 (RMEG 2,000 ppb), with none detected in follow-up samples in April and May or subsequent quarterly samples.
- Clark Street Well #3 contained 4.1 ppb of trihalomethanes (MCL 100 ppb) in one sample collected in May 1992. No organic chemicals have been found in the well in subsequent quarterly samples (20).
Another Albion municipal water supply wellfield, the Brownswood field located approximately 2.5 miles west of the ASTL site, contained traces of TCE in a 1980 sampling. The detection was not confirmed in later sampling. These wells have recently (since 1988) contained traces of methyl tert-butyl ether (maximum 18 ppb in November 1988, LTHA 40 ppb), benzene (maximum 1 ppb in November 1988, CREG 1 ppb), and toluene (maximum 1 ppb in August 1990, RMEG 2,000 ppb). The contamination has not been detected since the city stopped pumping the wells on a regular basis, only using them at high-demand periods (20). None of this contamination in the municipal wells is likely to be associated with the ASTL site. Additionally, at least three sites are known to have leaking underground gasoline storage tanks near the Brownswood and Albion Street wells. A leaking gasoline tank was removed in September 1989 from a location less than 130 feet from one of the Clark Street wells. To date, no gasoline components have been found in water samples from the Clark Street Wellfield. Because of the various contamination problems in the area, Albion's municipal wells are on a quarterly monitoring schedule for VOCs.
Because of the proximity of the Amberton Village subdivision's wells to the ASTL site, MDPH and the Jackson County Health Department have regularly sampled them and analyzed for metals and VOCs for the past 4 years. No detectable levels of organic contaminants have been found, and the metals concentrations have not been considered of health concern. The MDPH has not conducted analyses for heptachlor, reported in one 1992 EPA sample, on Amberton Village water samples (20). The heptachlor concentration reported by EPA is substantially below the level of detection used by the MDCH laboratory (0.1 ppb) (21).
The EPA's RI contractors collected surface soil samples from 6 locations outside the site, 4 approximately 1,000 feet west-southwest of the site (and upwind to the prevailing winds) to provide background data and 2 approximately 100 feet from the site boundaries, 1 southeast and 1 northwest. The concentrations of contaminants of concern in these samples are summarized in Table 13 (3). The maximum concentrations listed in the table were primarily found in the sample close to the site boundary to the northwest. The concentrations in this sample exceeded the background values, though rarely by more than a factor of 3. The concentrations of aluminum and some polycyclic aromatic hydrocarbons (PAHs) in this sample also exceeded those found in surface soil samples collected on the landfill (compare Table 7). The concentrations of PAHs found in this sample were generally comparable to those found in background agricultural soils (Reference 22, Table 5-2).
In October 1992, EPA's RI contractors collected water samples from six locations along the Kalamazoo River and six locations in the adjoining wetlands. These included six locations (three each in the river and wetlands) upstream of the site, with the remainder downstream of the site. Two samples, midstream and near-shore, were collected from each location on the river. Analyses of these samples found no noticeable impact on the river (Table 14) or wetlands (Table 15) that could be attributed to the site. No significant increase was found in contaminant concentrations downstream of the site compared with concentrations found upstream (3).
As indicated in the tables, each surface water sample collected was divided into two samples, and one of each pair of samples was filtered before analysis for metals, to separate dissolved metals from those in suspended solids. Chromium and zinc were detected in filtered samples from the river downstream of the site, but not in the corresponding unfiltered samples. As noted in the tables, barium (upstream and downstream river samples), calcium (downstream river samples), magnesium, and sodium concentrations were also frequently reported to be higher in filtered samples than in unfiltered ones, though only slightly so.
In October 1992, EPA's RI contractors collected sediment samples from six locations in the Kalamazoo River and eight locations in the adjacent wetlands. The samples were collected from the same locations as the surface water samples mentioned previously, plus two additional locations in the wetlands, one upstream and one downstream from the site. The two wetlands locations where only sediment samples were collected did not have any surface water at the time of the sampling. Two samples, midstream and near-shore, were collected from each location on the river. Analyses of these samples did not show any significant impact from the site on the river (Table 16) or wetlands (Table 17). Chemical concentrations in downstream samples were not significantly increased compared with those in upstream samples (3).
There is no record of sampling and analysis of biological materials from areas near the site for evidence of site-related contaminants.
In July 1987, MDNR collected fish from several places along the Kalamazoo River to analyze for polychlorinated biphenyls (PCBs) as part of its investigation of PCB contamination of the river. The collection site closest to the ASTL site was in the Ceresco Impoundment, approximately 20 miles downstream from the ASTL site. There, MDNR collected and analyzed 9 carp (lengths between 17.3 inches and 25.6 inches), one largemouth bass (14.4 inches), and one smallmouth bass (15 inches). The highest PCB concentration found was in one of the carp, 0.24 ppm in a skin-off fillet sample. The largemouth bass collected contained 0.02 ppm, the smallmouth 0.04 ppm, both in skin-on fillet samples (23). The PCB concentrations found in these fish are commonly found in fish from Michigan waters, even in areas with no known PCB source. It is highly unlikely, given the distance from the site the fish were collected, that the PCBs detected in the fish are related to the ASTL site.
In preparing this public health assessment, the Michigan Department of Community Health (MDCH) 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. Any qualifications noted in the sources for the data cited for this assessment are discussed with the data.
Much of the environmental data, marked J in the tables, was noted in the sources as estimates, most commonly reflecting a value reported below the contract required detection limit for the analytical technique, but above the detection limit of the instrument used. These results indicate that the chemicals were present in the samples, but the reported values should only be considered to be estimates.
Chromium and zinc were detected in filtered water samples from the Kalamazoo River downstream of the site, but were not detected in the corresponding unfiltered samples. Other filtered surface water samples contained concentrations of metals slightly higher than those in the corresponding unfiltered samples. These findings are therefore questionable, and may be the result of laboratory error or artifacts from the filtration technique used.
The heptachlor detections in the samples from residential wells have been brought into question, because follow-up samples have not been taken to confirm the chemical's presence. Of the three samples from the residential wells with positive detections of heptachlor, one (0.011 ppb) was flagged as estimated and a second (also 0.011 ppb) was a duplicate sample of one in which no heptachlor was detected (detection limit 0.01 ppb). It is also unusual to find heptachlor as the only organic contaminant in such a situation, and no heptachlor epoxide, a common environmental decomposition product (24). In addition, the heptachlor detection limit of 0.01 ppb cited in the RI report might be too low to be reliable. One EPA contract laboratory managed to attain that precision in tests of the detection methods, but the overall average method detection limit (MDL) among all the laboratories consulted was 0.04 ppb. EPA adopted an upper control limit to the MDL of 0.088 ppb. The heptachlor levels reported in the RI investigation were substantially below these MDL values. MDCH laboratory personnel use a detection limit for heptachlor in water of 0.1 ppb, 10 times the detection limit cited in the RI report and 5 times the maximum concentration detected in the water supplies (21).
Until the site was fenced in the summer of 1992, public access to the site was not restricted. In a February 1989 site visit by MDPH, footprints and snowmobile tracks were observed leading onto the site. During the fall 1989 visits, considerable evidence of hunting and target practice was noted in the form of spent shells on the ground surface. During the removal action in the summer of 1990, which also included installation of gates and fences at several entrances to the site, personnel for the contractor supervising the action for EPA found that one of the fences had been cut and a lock from one of the gates had been vandalized and removed. Fresh all-terrain vehicle tracks were present near the cut in the fence. The fence was then repaired and the lock replaced (6). Most of the site was fenced as part of the RI field work in the fall of 1992.
On-site physical hazards include a deep borrow pit in the northwest quadrant of the site and partially buried scrap metal littered around the uneven terrain. A large pile of construction debris was seen in the north end of the site during the MDPH 1992 site visit.
Anaerobic decomposition of organic material can produce methane gas. Excavations into the landfill resulted in the release of organic vapors at potentially explosive concentrations. These landfill gases could pose some threat of fire or explosion if they should collect in sufficient concentration. The chance of this happening at this site is reduced by the current porous sand and gravel cover over the landfill. Any gases generated in the wastes are likely to diffuse quickly through the cover to the surface and disperse in the atmosphere. However, persons excavating the landfill may encounter high concentrations of flammable gas, risking fire or explosion unless appropriate precautions are taken.
The drummed waste found on the site in 1989 included material with a flash point below 75 F, which would lead regulatory agencies to classify the material as ignitable. Some drums were removed from the site in the summer of 1990, but the MDNR found 200 to 400 drums buried on the site in June 1994 (4). Any drums that contain wastes similar to those found in 1989 may pose a fire hazard.
To determine whether nearby residents are exposed to contaminants migrating from the site, the Agency for Toxic Substances and Disease Registry 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 present or have in the past been present. An exposure pathway is considered a potential exposure pathway if one or more of these elements are missing; that is, at least one of the five elements is missing. An exposure pathway can be eliminated from consideration if one of the elements is not present and could never be present. The most important exposure pathways at this site are discussed in the following sections.
Surface soil on the site contains various contaminants at levels potentially of human health concern. People on the site can come into contact with this soil and absorb contaminants through their skin. A person can also incidentally ingest soil that adheres to his or her hands. Soil can be picked up by the wind or kicked up by a vehicle as dust, which a person on the site can then inhale. There were signs of traffic on the site, all-terrain vehicles, before the site was fenced in 1992. The new fence should deter further traffic on the site. In 1984, Michigan Department of Natural Resources (MDNR) personnel inspected the landfill cover, saw no evidence of direct contact hazard, and concluded the site did not require fencing (25). At that time, the Michigan Department of Public Health (MDPH) agreed with their decision (26). During the fall 1989 site investigations, leaking drums were discovered on the site and preliminary testing indicated the drums contained ignitable hazardous wastes and volatile organic compounds (5). These drums did pose a direct contact and fire hazard. The most accessible drums were removed for off-site disposal in 1990. Three adults living in the residence on the south end of the site still have access to their parcel of the site property, although a fence now blocks their access to the bulk of the landfill. From the available information, the number of people who might have had access to the landfill cannot be estimated.
Surface soil samples collected near the site contained metals and polycyclic aromatic hydrocarbons at concentrations potentially of human health concern. The limited data available do not indicate the source of these contaminants, and the concentrations found are generally comparable to background levels. Access to the area is not restricted, and signs of use have been seen. The area where the highest concentrations of chemicals were found in off-site soil samples is near a point where access to the landfill itself occurred in the past. With the available information, the number of people who might have been exposed in this manner cannot be determined.
Contaminants originating on the site can reach the Kalamazoo River and its associated wetlands via discharge of contaminated groundwater or surface runoff. Once in the river and wetlands, the contaminants will partition into the water, sediment, and air, and river or wetland biota may accumulate some of the contaminants. People using the river for recreation could be exposed to the contaminants by direct contact with water or sediments, incidental ingestion of water or sediments, or consumption of biota, especially fish, taken from the river. The river is extensively used for recreation, including swimming, fishing, and boating. There is no record of any municipal or private water supply using water from the river. Analysis of water and sediment from the river and wetlands near the site does not show any discernible contamination attributable to the site. The nearest place to the site along the river where biota have been sampled for analysis was Ceresco, 20 miles downstream. Low concentrations of polychlorinated biphenyls (PCBs) were detected in fish collected at Ceresco, but there is no indication that the PCBs at that point are related to the Albion-Sheridan Township Landfill site.
Rainwater can filter through the landfill cover and into the wastes below. In the wastes, the water can dissolve chemicals from the waste materials, then filter through the soil below the landfill into groundwater aquifers. Contamination with various metals in the groundwater at and near the site has been documented. Organic chemicals in the wastes have been found in leachate samples, but have not been reliably found in deeper or off-site groundwater.
The chemicals could move with the groundwater and reach residential or municipal wells. People using these wells for their household water supply can be exposed to chemicals in the water through ingestion, dermal contact, or inhalation of volatile chemicals secondary to household use.
The existing off-site private wells downgradient of the site are sufficiently far from the site that any groundwater carrying contamination from the site would probably discharge to the Kalamazoo River long before it would reach the wells. The well serving the residence on the south end of the site does not appear to be affected by contamination from the site (compare Table 5 with Tables 3, 4, 11, and 12). Future development of the area immediately southwest of the site might include wells for residential use drilled into the contaminant plume. That potential residents would drill such wells, totally disregarding the situation in the vicinity of their property, seems highly unlikely.
Arsenic has been reported in water from residential wells near the site at concentrations estimated to range from 1.1 to 1.7 ppb. Arsenic, however, is a naturally occurring element that is frequently found at much higher levels than those found in Michigan groundwater, particularly in association with the Marshall Sandstone Formation, which forms the upper bedrock in the vicinity of this site. MDPH sampling of 250 water supply wells in Calhoun and Jackson counties has shown a range of arsenic concentrations from none detected to 22 ppb, with an average of 1.6 ppb. Concentrations of arsenic vary widely based on depth of the well and length of time the well is pumped before sampling. Higher levels of arsenic were found in some monitoring wells on or near the site (up to an estimated 126 ppb) than in residential wells. The concentrations found in the residential wells are probably not site-related because upgradient wells contained concentrations comparable to those downgradient of the site. Arsenic concentrations in private wells are far below the current maximum contaminant level (MCL) of 50 ppb.
Heptachlor was tentatively identified in three residential wells and one upgradient on-site monitoring well but was not detected in leachate samples collected within the landfill. Maximum levels reported in residential well samples were 20 times lower than the EPA MCL (0.4 ppb) for drinking water and far below the limit of detection for most analytical laboratories.
Manganese was reported in landfill leachate samples (832 ppb), upgradient monitoring wells (maximum of 363 ppb), downgradient monitoring wells (maximum of 411 ppb), and in residential wells (maximum of 63 ppb in the on-site residence and 185 ppb in off-site residential wells). Like arsenic, manganese is a naturally occurring element and concentrations will vary by depth and geographic region. Because similar concentrations of manganese were found in upgradient and downgradient samples, it is unlikely that the concentrations of the metal reported in residential wells were related to the site.
Samples from every residential well sampled for the remedial investigation were reported to
contain either arsenic, heptachlor, or manganese at concentrations that exceeded a comparison
value. The total population using these wells is approximately 200 persons. The potential health
effects from exposure to these chemicals through the drinking water are explored in the following
The only known completed pathway for human exposure to contaminants from the Albion-Sheridan Township Landfill (ASTL) site is for a person coming onto the site and coming into direct contact with contaminated surface materials. Current and future access by unauthorized personnel to the landfill itself is strongly deterred by the site fence. Access would have been more frequent before the site fence was installed, although it would probably not have been very common. The occupants of the residence on the south end of the site have had regular and frequent access to their parcel of the property. It is not known how frequently they went onto the landfill proper before the fence was installed. Surface soil samples collected near the landfill contained concentrations of polycyclic aromatic hydrocarbons (PAHs) and metals potentially of human health concern, but not significantly above background levels. People who visit the site vicinity are therefore not likely to incur a significantly increased health risk.
This evaluation covers two exposed populations: residents of the house on the south end of the property and trespassers onto the landfill proper.
In addition, people living near the site and using private wells for their water supply might be exposed to arsenic, heptachlor, and manganese at concentrations above comparison values; however, the presence of heptachlor was not confirmed, and the presence of any of these chemicals in the residential wells may not be related to the ASTL site. Aluminum, calcium, iron, magnesium, potassium, and sodium were also detected in the residential well samples, but no comparison values are available for these metals. The detection of these metals in samples from the private wells may not be related to the ASTL site.
The exposure doses for the on-site residents are calculated using the following assumptions (27):
|      Weight:||70 kg|
|      Water consumption:||2 L/day|
|      Soil ingestion (incidental):||100 mg/day
|      Weight:||10 kg|
|      Water consumption:||1 L/day|
|      Soil ingestion (incidental):||200 mg/day|
|           (pica behavior):||5,000 mg/day|
As a conservative assumption, we assumed that the soil containing the highest contaminant concentrations to which on-site residents are exposed is the soil around the residence on the site (Table 8). Their primary water supply is their well, with composition assumed to be that found in the remedial investigation (RI) sampling (Table 5). No children currently live in the residence on the site; however, children subject to pica behavior were included in this evaluation to address the
Trespassers on the landfill would probably not include very young children, those subject to pica behavior. Trespassers would also not be likely to visit the site more than 2 days a week. We assume, to be conservative, that trespassers encounter soil containing the highest concentrations of contaminants found in surface soil samples from the landfill (Table 7).
The primary benchmarks against which exposures are evaluated for their potential for causing noncancer adverse health effects are the minimal risk levels (MRLs), developed by the Agency for Toxic Substances and Disease Registry (ATSDR), and reference doses (RfDs) and reference concentrations (RfCs), developed by the Environmental Protection Agency (EPA). It is generally accepted that a person exposed to a dose of a chemical less than an MRL, RfD, or RfC is not likely to experience noncancer adverse health effects. The derivation of MRLs, RfDs, and RfCs from the observed threshold exposures includes safety factors to allow for different responses between species and between individuals. However, these values may not be protective for individuals who are hypersensitive to chemical exposures, including the very young, the very old, individuals whose bodies are under stress from illness, and individuals who have an allergic response to the chemical.
Threshold exposures from which MRLs, RfDs, and RfCs are derived may also be cited if none of the derived values are available. The threshold exposures include lowest-observed-adverse-effect levels (LOAELs) and no-observed-adverse-effect levels (NOAELs). In a given experiment, with exposure route, species, and health effect specified, the LOAEL is the lowest exposure at which the effect was observed, and the NOAEL is the highest exposure at which no effect was observed.
For chemicals that may cause cancer, the risk associated with an exposure is evaluated separately from noncancer health risks, using published potency factors, which relate the chance of contracting cancer to the dose of the chemical. For this assessment, the risk of cancer is considered significant if 1 extra case of cancer is likely to develop among 1 million people subject to the exposure over their lifetimes.
No MRLs or RfDs are available for exposure to aluminum. Most aluminum compounds are considered virtually nontoxic. Patients suffering from Alzheimer's disease sometimes have abnormally high concentrations of aluminum in their brains, but whether the aluminum causes the disease is not known. Patients suffering from renal failure who were treated with aluminum-containing compounds accumulated the metal in their brains and bones, experiencing dementia and bone diseases. The dementia in these cases does not include other indicators of Alzheimer's disease. Rats who were fed aluminum-containing compounds had smaller offspring. Consumption of water containing the aluminum concentrations found on and near the site would not be likely to result in a dose of aluminum equaling the LOAEL. No one, not even a child prone to pica behavior, is likely to ingest enough soil or sediment containing the aluminum concentrations found on or near the site to attain the LOAEL. No available evidence links aluminum exposure to cancer. EPA has proposed a secondary maximum contaminant level (SMCL), based on non-health-related criteria such as color, odor, and taste, for aluminum of 50 ppb (28). One of the samples from an upgradient residential well exceeded this level.
The concentrations of arsenic found in surface soil on and near the ASTL site are within the range of concentrations found in the eastern United States (27). The concentrations in the surface soil, including the background samples (3.9-6.5 ppm), were at or above the high end of the range of concentrations found in sandy topsoil5 in that area of Michigan (29). The concentrations of arsenic reported in water samples from residential wells were substantially below the EPA maximum contaminant level (MCL) for the metal (50 ppb) and were within the range commonly found in groundwater. A child subject to pica who plays regularly in soil containing the concentration of arsenic found around the residence on the site might ingest enough arsenic from the soil to exceed the MRL for noncancer adverse health effects. No one else is likely to incidentally ingest enough arsenic from the soil near the residence or to ingest enough of the metal from the water in the well to exceed the MRL. The combined dose through incidental ingestion and water consumption is not likely to exceed the MRL. The estimated exposure dose for a child with pica from ingestion of the soil would not be likely to exceed the dose at which hardening and changes in color of the skin have been observed (the LOAEL), although the exposure dose might exceed the highest dose at which no adverse health effects were observed. The combined arsenic dose from pica consumption of soil and regular consumption of water from the well would not be likely to exceed the LOAEL. Ingestion of arsenic has also been linked to cancer of the skin, bladder, liver, kidney, and lungs in epidemiologic studies. EPA has classified arsenic as a known human carcinogen (EPA Class A) (30). Lifetime consumption of water containing the concentrations of arsenic reported in samples from residential wells on and near the site may result in a low increased risk of contracting skin cancer. A person who lives in an area with soil containing the concentrations of arsenic found in the soil in the site vicinity may incidentally ingest enough of the metal to also incur a low increased risk of contracting skin cancer. The combined dose from incidental ingestion of soil and consumption of groundwater is likely to result in a low increased risk of contracting skin cancer. Available information is insufficient to estimate the risk of contracting other cancers through ingestion of arsenic.
No one would be likely to ingest enough barium from the soil around the residence on the site or the water in the well supplying the residence or both sources combined to exceed the RfD. No one is likely to incidentally ingest enough soil from the site to attain a dose of the metal that would approach the RfD. No MRLs are available for barium. No available evidence links barium exposure to cancer (31).
No one who is likely to have access to the site or the sediments of the Kalamazoo River is likely to incidentally ingest enough cadmium to attain the MRL or RfD for noncancer adverse health effects. The evidence linking occupational inhalation of cadmium compounds to increased incidence of lung cancer is weak. Some laboratory animals who breathed cadmium compounds developed lung cancer. EPA has classified cadmium compounds by inhalation as probable human carcinogens (EPA Class B1). Available information is insufficient to determine whether cadmium and its compounds are carcinogenic when ingested (32).
Chromium is rarely found in the environment as the pure metal, but rather combined with other elements in chemical compounds. Chromium combines with other elements primarily in one of two oxidation or valence states, trivalent or chromium(III) and hexavalent or chromium(VI). The toxicity of a chromium compound is very dependent on the valence state of the chromium included. Chromium(III) compounds are considered virtually nontoxic, and chromium(III) is an essential trace element in the diet. Chromium(VI) compounds can irritate the skin and gastrointestinal tract, and can cause liver damage. Inhalation of chromium(VI) compounds has been linked to cancer of the lung and respiratory tract in industrial workers and laboratory animals. EPA has classified chromium(VI) as a human carcinogen by inhalation (EPA Class A). Not enough information is available to assess the carcinogenicity of chromium(VI) by any other exposure route or of chromium(III) by any route of exposure. The available data on the environmental media at the site reports the total chromium concentration, without distinguishing between the valence states. Chromium(VI) tends to be reduced to chromium(III) by naturally occurring chemical reactions in the environment. No one likely to have access to the site is likely to ingest enough chromium from the soil to exceed any chromium RfD. ATSDR has not issued any MRLs for chromium. No one would be likely to ingest enough of the metal from the soil around the residence on the site to attain the RfD for chromium(VI). A child with pica living in the residence might ingest enough chromium from both the soil and the water supply to attain the RfD for chromium(VI), if the chromium concentration in the water (reported as not detected at a detection limit of 10 ppb) is a substantial fraction of the detection limit. The estimated dose would not be more than 15% above the RfD, substantially below any available LOAEL (for enhancement of dermatitis in chromium-sensitive individuals). Available information is insufficient to evaluate the risk of cancer from the chromium at the site (33).
No MRLs or RfDs are available for cobalt. No one is likely to ingest sufficient cobalt from the soil, sediments, groundwater, or surface water at or near the site to attain the LOAEL listed in the toxicological profile for the metal (34). Cobalt sulfate was once added to beer to stabilize the foam, and some heavy beer drinkers developed serious to fatal heart problems. Other factors may have contributed to their heart conditions. No corresponding NOAEL was reported. A child with pica playing in the soil around the on-site residence might experience an exposure dose approximately one-twentieth of the LOAEL, and trespassers in the site vicinity would experience lower doses. No available evidence links ingestion, inhalation, or dermal exposures to cobalt compounds with cancer. Some laboratory animals developed cancer when cobalt compounds were injected into their bodies. The International Agency for Research on Cancer (IARC) has classified cobalt and cobalt-containing compounds as possible human carcinogens (IARC Class 2B). EPA has not classified cobalt and its compounds for carcinogenicity. Available information is not sufficient to evaluate the cancer risk, if any, from exposure to cobalt at this site (34).
No MRLs or RfDs are available for exposure to copper. No one is likely to ingest enough copper from soil, groundwater, surface water, or sediment on or near the site to attain the LOAEL reported in the toxicological profile for the metal, for gastric distress including vomiting and diarrhea after ingestion of large quantities of copper in food or water. No NOAEL was reported in these incidents. A child with pica playing in soil containing the concentration of copper found near the on-site residence might experience an exposure dose approximately one-tenth of the LOAEL. No evidence links exposure to copper with cancer in humans or animals. Both EPA and IARC have determined that copper is not classifiable as to its carcinogenicity (EPA Class D, IARC Class 3) (35).
No MRLs or RfDs are available for exposure to lead. Lead tends to accumulate in the body, and the health effects of any exposure to the metal depend on the subject's previous history of exposure. No one is likely to ingest enough lead from the soils around the on-site residence to attain the LOAEL listed in the toxicological profile for the metal, for changes in the levels of certain enzymes in the blood on intermediate-term exposure. The enzymes that were tracked are involved in the synthesis of hemoglobin. No one is likely to incidentally ingest enough soil from on or near the site to experience a lead dose in excess of the LOAEL. No NOAEL was cited, and because lead accumulates in the body, a long exposure to a low dose may have the same effect as a short exposure to a high dose. Lead can also interfere with the development of the nervous system in fetuses and children. Some experimental animals fed food containing lead developed cancer of the kidneys. EPA has classified lead as a probable human carcinogen (EPA Class B2). Available information is insufficient to evaluate the risk of contracting cancer from exposure to lead at this site (36). The lead concentration in the soil around the residence is much lower than that commonly found in urban areas and is not generally considered to be of health concern.
A child whose primary drinking water supply contained the manganese concentration found in the residential well on the site or in most of the residential wells near the site might ingest enough of the metal to exceed the RfD through ingestion of water.6 ATSDR has not issued any MRLs for manganese. No one is likely to ingest enough soil from the area of the residence to exceed the RfD for manganese through ingestion of soil. No adult is likely to incidentally ingest enough soil or sediment from anywhere on or near the site to exceed the RfD for manganese. A child with pica who visits the site area 2 days a week might ingest enough manganese from the soil to exceed the RfD, though not to exceed the LOAEL for mild neurological effects seen in a study of people whose drinking water contained high levels of manganese. No available evidence links exposure to manganese with cancer (37, 38). EPA has not issued an MCL for manganese in drinking water under the Safe Drinking Water Act. EPA has issued an SMCL for manganese of 50 ppb, equal to the reference dose media evaluation guide (RMEG). Water samples from residential wells near the site do exceed this SMCL.
No MRLs are available for ingestion of nickel and its compounds. EPA has issued an RfD for ingestion of soluble nickel compounds, but not for other nickel compounds. No one is likely to ingest enough nickel from the soils on or near the site to exceed the RfD for soluble compounds or the NOAEL for a transient impairment of vision reported after a single exposure to nickel sulfate in an experiment using human volunteer subjects. A small fraction of the population is allergic to nickel, and once these persons are sensitized by exposure to the metal, they can suffer dermatitis after further contact or ingestion. Inhalation of water-soluble nickel compounds or of nickel subsulfide (Ni3S2) has been linked to cancer in nickel plant workers and laboratory animals. No evidence links ingestion of nickel compounds with cancer. EPA has classified nickel refinery dust and nickel subsulfide as human carcinogens (EPA Class A), but has not classified other nickel compounds. IARC has classified all nickel compounds as human carcinogens (IARC Class 1), and metallic nickel as a possible human carcinogen (IARC Class 2B) (39). Available information is not sufficient to evaluate the risk of cancer due to exposure to nickel at this site.
An MRL has been derived for intermediate-term exposure (between 14 days and 1-year duration) to vanadium by ingestion, but not for chronic exposure, and no RfDs are available for the metal. A child subject to pica who plays in soil containing the concentration of vanadium found in the soil around the on-site residence might ingest enough of the metal to exceed the intermediate-term MRL. No one else is likely to incidentally ingest enough of the soil to exceed the intermediate-term MRL. The child with pica is not likely to ingest enough vanadium to attain the NOAEL (from an experiment on rats) that the MRL was derived from, or a higher NOAEL from a study on human exposure. No evidence links exposure to vanadium to cancer (40). Health effects from the vanadium at or around the ASTL site are not likely.
- Other Metals
Calcium, iron, magnesium, potassium, and sodium commonly occur in soil, groundwater, and surface water in the environment at concentrations similar to those found at and near the site. There are no health-based standards for these metals in environmental media. EPA has issued an SMCL for iron of 300 ppb in drinking water, and none of the residential well samples exceeded this level. Individuals who have received medical advice to restrict their sodium intake might be advised not to consume water containing more than 200 ppm of sodium. None of the residential well samples exceeded this level.
- Di-n-octyl phthalate
Little information is available on the toxic effects of exposure to di-n-octyl phthalate (DNOP). One report describes a worker in an imitation leather plant who developed an asthmatic reaction to the chemical after continuous occupational exposure (41). Laboratory animals fed food containing the chemical developed enlarged livers, although no indications of changes in the organ's function were reported. DNOP has generally been found to be less toxic than the related compound bis(2-ethylhexyl)phthalate (BEHP). The exposure dose of DNOP experienced by anyone from the soil and water on or near the site is not likely to equal the MRL for BEHP and, therefore, is not likely to result in any adverse noncancer health effects. No evidence is available that exposure to DNOP causes cancer; however, one experiment with laboratory animals indicates that ingestion of DNOP can increase the carcinogenic effects of other chemicals (42, 43).
Although there are no MRLs or RfDs for heptachlor, EPA has issued an MCL of 0.4 ppb, with a maximum contaminant level goal of 0. As mentioned previously in the Quality Assurance and Quality Control section, the concentrations of heptachlor reportedly found in water samples from the residential wells near the site were below detection limits that are considered reliable. The following discussion takes the conservative position that the reported concentrations of heptachlor in samples from the residential wells were real detections and accurately measured. The estimated exposure dose experienced by a child whose primary drinking water supply is one of the residential wells near the site that tentatively contained heptachlor would be approximately a factor of 100,000 below the LOAELs identified in animal studies. No NOAEL was reported in the study with the lowest LOAEL, for reproductive failure in rats. Some laboratory animals fed heptachlor developed liver cancer. EPA has classified heptachlor as a probable human carcinogen (EPA Class B2) (24). Lifetime consumption of water containing the concentration of heptachlor tentatively identified in samples from residential wells near the site would result in no apparent increased risk of contracting cancer. The heptachlor concentration reported in samples from the residential wells is approximately one-twentieth of the MCL.
- Polychlorinated Biphenyls (PCBs)
No one who is likely to be on the site is likely to incidentally ingest high enough concentrations of PCBs from the soil or the wastes in the landfill to exceed the MRL for noncancer adverse health effects. Laboratory animals that ingested PCBs suffered liver cancer. EPA has classified all PCBs as probable human carcinogens (EPA Class B2) (44). Lifetime exposure to and incidental ingestion of soil containing the concentrations of PCBs found in surface soil or wastes at the site could result in a low increased risk of cancer. A typical trespasser on the site would incur no apparent increased risk of cancer, since he or she would probably not spend much time on the site.
A person who consumes the average U.S. consumption rate for fish of 12 g/day (Reference 27, Table E.1) of fish containing the maximum concentrations of PCBs found in the carp collected from the Ceresco Impoundment would ingest a dose of PCBs above the MRL, though less than the LOAEL. Such a fish consumer might also incur a low increased risk of contracting cancer from the PCBs in the fish. The Michigan Department of Community Health (MDCH) uses the Food and Drug Administration's action level for PCBs in fish, 2 ppm, as a trigger level for implementing fish consumption advisories (45). None of the sampled fish contained PCBs at or above the trigger level; therefore, the MDCH has not issued any fish consumption advisory for the Kalamazoo River above Battle Creek, 30 miles downstream from the ASTL site (46).
- Polycyclic Aromatic Hydrocarbons (PAHs)
PAHs found on or near the ASTL site include benzo(a)anthracene, benzo(a)pyrene, benzo(b)fluoranthene, benzo(ghi)perylene, benzo(k)fluoranthene, chrysene, dibenzo(a,h)anthracene, indeno(1,2,3-cd)pyrene, 2-methylnaphthalene, naphthalene, and phenanthrene. No MRLs or RfDs have been established for any of these chemicals. It is highly unlikely that anyone would ingest enough soil from the site to attain a dose of any of these chemicals equal to the LOAELs or NOAELs for noncancer health effects determined in laboratory experiments or health studies. Exposure to benzo(a)anthracene, benzo(a)pyrene, benzo(b)fluoranthene, benzo(k)fluoranthene, chrysene, dibenzo(a,h)anthracene, and indeno(1,2,3-cd)pyrene is linked to cancer in experimental animals. EPA has classified these PAHs as probable human carcinogens (EPA Class B2). EPA has classified the other PAHs found at the site as not classifiable with regard to carcinogenicity (EPA Class D) (22, 47). Benzo(a)pyrene is the most potent of the carcinogenic PAHs found in the surface soil around the residence on the site, according to published relative potency values (48). Estimates of the risk based on these relative potency values indicate that no one is likely to ingest high enough concentrations of] PAHs from the soil from the area around the residence to incur any apparent increased risk of contracting cancer. Someone who spends his or her entire life around soil containing the concentrations of carcinogenic PAHs found on the landfill or in adjacent areas might incidentally ingest enough of the chemicals to incur an increased risk of cancer slightly above the level of significance; however, no one is likely to spend enough time in the vicinity of the landfill to ingest enough soil to incur a significantly increased cancer risk.
The MDCH Office of the State Registrar and Center for Health Statistics provided the assessors with cancer incidence data for ZIP Code 49224 for the years 1985 through 1991. The ZIP Code area includes the ASTL site, the city of Albion, Amberton Village subdivision, and surrounding areas in both Calhoun and Jackson counties. Comparing the incidence data from the area with estimates computed using data from the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program7 shows that the cancer incidence in the area has been lower than the national rates (Table 18) (50). This indicates that there is no increased incidence of cancer in the ZIP Code area around the site, although the data are not sufficient to permit detection of possible localized clusters.
Environmental health personnel from the Jackson County Health Department investigated the reports of an excessive incidence of cancer and other illness in the Amberton Village subdivision. They interviewed the Calhoun County Board of Health member who reported the concerns and personnel from the Michigan Department of Natural Resources, Michigan Department of Public Health, and EPA, and did not find any evidence to suggest an environmental cause for the reported cancer or illness incidence (51). MDCH staff have also contacted the Board of Health member, but he could not provide sufficient information for further investigation of the concerns. As mentioned in the preceding section, cancer incidence data from the MDCH Office of the State Registrar and Center for Health Statistics do not indicate any increased incidence of cancer in the ZIP Code area around the site (50).