PUBLIC HEALTH ASSESSMENT

BOFORS NOBEL INCORPORATED
EGELSTON TOWNSHIP, MUSKEGON COUNTY, MICHIGAN
CERCLIS NO. MID006030373





ENVIRONMENTAL CONTAMINANTS AND OTHER HAZARDS

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. Lifetime exposure to chemical concentrations at or below the appropriate comparison values for a chemical is not expected to result in more than 1 case of cancer in 1 million people exposed or any non-cancer health effects. Comparison values used in this assessment include:

If no comparison values for a chemical in a medium exist, or there is no CREG available for a carcinogen, the chemical is retained for further evaluation.

The MDNR, in their initial evaluation of the site, identified 27 "Potential Compounds of Concern" at the site (see Table 1) (23). The RI contractors report listed 16 chemicals as compounds of concern in at least one environmental medium, 12 from the MDNR list (marked by footnote "b" in the Table) plus 2-methylphenol and three metals: arsenic, chromium, and lead (marked by footnote "d" in the Table) (1). Most of these compounds and several others (marked by footnote "f" in the Table) meet the above criteria for contaminants of concern, and are shaded in the Table.

To identify facilities which might contribute to the contamination of the groundwater, surface water, soil, sediment, and air near the Bofors-Nobel site, the MDCH searched the Toxic Chemical Release Inventory (TRI) data base for 1987 through 1993. The U.S. EPA compiles the TRI from information provided by industries.

Entries for Lomac, Inc. in the TRI included reports of air releases of acetone (1987-93), ammonia (1987-93), aniline (1989-93), chlorine (1988-91), chlorobenzene (1987-93), 3,3'-dichlorobenzidine (1987-93), ethylene glycol (1987-90), hydrochloric acid (1987-93), nitric acid (1987-91), phosphoric acid (1990-93), sulfuric acid (1987-93), tetrachloroethylene (1987-93), thiourea (1987-93), toluene (1987-93), and 1,2,4-trichlorobenzene (1992-93). Lomac, Inc. also reported to the TRI on transfers of the above chemicals, ammonium sulfate solution (1987-89), and sodium sulfate solution (1987) to the Muskegon County Wastewater Treatment Plant and other off-site disposal facilities. They reported no environmental releases of the last two chemicals. Lomac reported that they sent some 1,2,4-trichlorobenzene in 1993 to an incineration facility approximately 0.5 miles west of their plant. The other off-site disposal facilities that Lomac reported using were farther than 30 miles from the Bofors-Nobel site. Lomac also filed reports in 1987 and 1988 on sodium hydroxide solution, based on their inventory of the chemical, with no data on environmental releases or off-site transfers.

Sun Chemical, located west of the Bofors site, reported air releases of ammonia (1990), barium compounds (1987-93), 3,3'-dichlorobenzidine (1987-88), hydrochloric acid (1987-93), and toluene (1991-93). They also reported transfers to the Muskegon County Waste Water Treatment Plant and other off-site disposal facilities of the above chemicals, ethylene glycol (1991), and sodium hydroxide solution (1988), with no environmental release of the last two chemicals. The off-site facilities Sun Chemical used included landfills located 1.5 miles southwest, 5 miles northeast, and 12 miles southeast of their plant and other facilities located in other parts of Michigan and in other states.

The Eagle Alloy facility, located across Evanston Road from the Bofors-Nobel site, reported air releases of chromium, cobalt, copper, manganese, and nickel in 1989, 1990, 1991, and 1993. They also disposed of these metals through the Muskegon County Waste Water Treatment Plant, a Muskegon County Landfill located approximately 3 miles west of the site, and a facility approximately 12 miles southeast of the site. These metals are included as contaminants of concern for this assessment because of their presence in environmental media at the Bofors site.

The TRI contained entries for nine other facilities with the same postal zip code (49442) as the Bofors-Nobel site and seven in the neighboring zip code 49444 (southwest of the site) and one in zip code 49415 (southeast of the site). Sun Chemical gave its zip code as 49443, currently that for Post Office Boxes at the Muskegon Main Post Office, in its TRI reports. Eighteen other facilities listed zip code 49443, including five facilities that gave both zip codes 49443 and 49444 in their TRI reports. The addresses given in the TRI for the twenty-nine other facilities listed in the TRI in zip codes 49442, 49443, and 49444 are all located approximately 3 miles or more west of the Bofors-Nobel site. The one facility listed in the TRI in zip code 49415 is located approximately 10 miles east of the Bofors-Nobel site. Because of the distances involved, the releases from these facilities are not likely to contribute to the environmental contamination at the site.

Unless specifically cited otherwise, all environmental data in this section is taken from the RI report (1).

A. On-Site Contamination

Groundwater

Groundwater quality measurements show that the upper aquifer is heavily contaminated beneath the site. The major source of contamination appears to be the waste lagoons in the southern portion of the site.

In March 1979, MDPH sampled the Bofors-Lakeway water supply (supplied by ten wells on their property) at several locations within the system, in response to an allegation from employees at the plant that DCB might have been getting into the water system (24). A process-water supply well (#1, also known as the SLS well) contained 2.8 parts per billion (ppb) DCB, and water from a shower/locker room in building #3 (also known as the Prowl building) contained 2 ppb DCB. As a result, MDPH ordered the company not to use water from any plant well for drinking (25). The MDPH also noted one place in the complex where the process water and drinking water systems were interconnected, and advised the company to remove the interconnection and to take steps to prevent further cross-connection. The company had been supplying its employees with bottled water in response to complaints about the taste and odor of the water in the Prowl building (24). Return sampling in June, September, and November of 1979 found no detectable DCB, but in January 1980, the same well #1 contained 0.5 ppb DCB, in March 1980, 1 ppb, in June, 10 ppb, and on September 8, 0.3 ppb. DCB was not detected in a sample collected on September 30, 1980 (26, 27, 28). In January 1981, the monitoring program was reduced to annual samples, and the MDPH advised the company to continue to use bottled water, as a precautionary measure (29).

The next records (chronologically) in MDPH/MDEQ files concerning the plant's water supply are of a November 1988, sampling from a new boiler-supply well. No volatile organic chemicals or inorganic chemicals were detected. The analyses did not include either benzidine nor DCB (30). In July 1991, the MDPH Division of Water Supply collected eight water samples from various locations in the plant's system. No DCB was found in any sample (detection limit 10 ppb). Well #1, which had previously contained DCB contamination, was not sampled because it was no longer in service. Two wells, located in the northwest corner of the plant property, which supplied the showers and sinks in two buildings, contained up to 113 ppb tetrachloroethylene. A third well in the same part of the site contained 1.6 ppb chloroform (CREG 6 ppb). These were the only detections of organic chemicals in any of the 8 samples (31).

In March 1992, Lomac sampled the water from the showers in their plant. No organic chemicals were found (DCB detection limit 5 ppb, tetrachloroethylene detection limit 2 ppb) (32). Lomac has collected water samples from their plant system on a quarterly basis (MDPH/MDEQ files include reports from March 1992 through August 1994), and the MDPH also collected samples from the plant water system in September 1993. The company analyzes their quarterly samples for volatile and semi-volatile organic chemicals, the MDPH analyzed their September 1993 samples for pesticides, PCBs, and metals. One of the company's three September 1993 samples, from a well on the north side of their property, contained 1.2 ppb n-butylbenzene, 1.1 ppb sec-butylbenzene, and 1.1 ppb tert-butylbenzene (no comparison values are available for any butylbenzene isomer). No chemicals were detected in earlier samples and no other chemicals were detected in this one. The December 1993, March 1994, and August 1994 samples contained no organic chemicals. The MDPH found no organic chemicals or metals in their September 1993 samples (33, 34, 35, 36, 37). The results of the analyses of the plant water supply are summarized in Table 2. DCB and tetrachloroethylene have both been found in the plant water system at levels above the CREGs for the chemicals (0.03 ppb and 0.7 ppb, respectively). Neither chemical has been detected in recent samples, though the detection limits cited for these chemicals are also above the CREGs.

Contaminant concentrations in water samples from on-site monitoring wells have exceeded comparison values. Groundwater samples collected from monitoring wells in 1986 contained up to 2,000 ppb benzidine, 530 ppb DCB, 2,400 ppb aniline, 6,300 ppb benzene, and 81 ppb tetrachloroethylene. DCB was found at 8 ppb in one upgradient sample. None of the other chemicals were detected in upgradient samples (38). During the RI in 1988, groundwater samples contained up to 1,300 ppb benzidine, 1,900 ppb DCB, 10,000 ppb aniline, 65,000 ppb benzene, 18,000 ppb tetrachloroethylene, 58 ppb lead, and 74 ppb arsenic, plus other contaminants (see Table 3). Metal concentrations above background levels were found in water from only one monitoring well cluster (WC-27 on Figure 2). Figure 4 shows the areal extent of DCB contamination in the groundwater based on data collected during the RI.

Monitoring wells on the site have been sampled on a quarterly schedule. Concentrations of contaminants of concern found in samples collected between June 1992 and November 1993 are summarized in Table 3 (39). Reference 39 does not include any results for metals analyses of water samples collected from monitoring wells in 1993.

The monitoring wells at the site are from 12 to 126 feet deep, and all are screened in the upper (water table) aquifer. The variation in concentrations between wells of different depths in the same cluster were small. The plant supply wells presumably use the upper aquifer, since water in the bedrock aquifer contains high concentrations of minerals and is therefore not usable for potable purposes. There is no information available regarding any sampling of water from the bedrock aquifer at the site.

Concentrations of contaminants of concern found in water from the purge wells during routine monthly sampling by the company during 1986 (23), an MDNR investigation in 1987 (cited in Reference 1), the RI in 1988 (1), and quarterly sampling between June 1992 and November 1993 (39) are summarized in Table 4. Table 5 shows the annual average concentration of six important contaminants of concern in the purged water between 1982 and 1986 (from Table 10 in Reference 23), and in 1992 (39). The concentrations in the purged water generally decreased over time, though this may not reflect the conditions in the groundwater. Purge wells were put into and taken out of service, and the averaging may not have been weighted properly to reflect the amount of water pumped from the wells.

Soil and Sludge

Contaminant concentrations in on-site soils have exceeded comparison values. An investigation by MDNR in 1980 found DCB concentrations in surface soils as high as 320 parts per million (ppm) near the DCB production area (Table 6) (2). Soil and sludge samples collected from the lagoon areas in 1981 revealed DCB concentrations in the sludges up to 670 ppm (Table 7), and in the underlying soils as high as 85 ppm (Table 8). Bofors conducted a study in 1986 which analyzed soils from the production area of the site for purgeable and semi-volatile compounds. Maximum concentrations for selected chemicals included toluene, 14 ppm; tetrachloroethylene, 20 ppm; DCB, 502 ppm; and 2,2'-dichloroazobenzene (DCAB), 1,170 ppm (Table 6) (38).

Samples of sludge from the lagoons collected during the 1988 RI contained up to 11,000 ppm DCB, 12,000 ppm azobenzene, 3,400 ppm benzidine, 6,040 ppm lead, 916 ppm chromium, and 43.8 ppm arsenic (Table 7). Soil from beneath the lagoons contained as much as 500 ppm DCB, 1,000 ppm azobenzene, 180 ppm benzidine, 19.2 ppm lead, 8.8 ppm chromium, and 2.4 ppm arsenic (Table 8). Surface soil sampling during the RI found up to 1.4 ppm benzidine, 10,500 ppm DCB, 38.4 ppm arsenic, 4,110 ppm lead and 125 ppm chromium (Table 9). The highest lead and arsenic concentrations were found in the southwest corner of the site. The concentrations of the various metals were not strongly correlated, in that the maximum concentrations, except for lead and arsenic, were generally not found in the same samples. The second highest lead (934 ppm), arsenic (9.9 ppm) and chromium (115 ppm) concentrations were all found in the same sample, however. The sample with the high DCB concentration was collected from the area of suspected lagoon overflow on the east side of the lagoon area.

Soil samples from a depth of 1.5 feet at visible spill sites contained as much as 1,600 ppm toluene, 30 ppm benzene, and 10 ppm xylenes. Soil from borings in the production area of the site contained up to 1 ppm trans-1,2-dichloroethylene, 20.5 ppm tetrachloroethylene, 58.7 ppm benzene, 962 ppm toluene, 160 ppm DCB, 160 ppm 2-chloroaniline, 160 ppm benzidine, 29.5 ppm lead, 11.4 ppm chromium, and 4.2 ppm arsenic (Table 6).

Surface and subsurface soil from the vicinity of well cluster WC-27 contained high concentrations of metals, as did water from the wells in cluster WC-27. The second highest lead concentration in surface soil (934 ppm) was found adjacent to the cluster. See Figure 2 for the locations of this cluster and other landmarks cited in this paragraph. This cluster is near Lagoon 1, from which sludge samples with the highest metal concentrations were collected. However, water from well LW-4, between Lagoon 1 and WC-27, contained no detectable lead or chromium and 11 ppb arsenic. WC-27 is not downgradient of Lagoon 1; therefore, there is no firm association between the metal contamination in the lagoon and that in the groundwater. Other locations with high measured metal concentrations in the soil were widely scattered around the site.

Air

Air sampling at the Lomac plant area in June and July, 1988, found up to 400 micrograms per cubic meter (µg/m3) of toluene, 690 µg/m3 tetrachloroethylene, 0.54 µg/m3 total xylenes, 5.2 µg/m3 1,1-dichloroethylene, and 3.2 µg/m3 1,1,1-trichloroethane (Table 10). As indicated in the Table, the 1,1-dichloroethylene and tetrachloroethylene concentrations exceeded the corresponding CREGs, and there is no comparison value available for xylenes in air.

Air at four sampling stations along the perimeter of the Lomac plant has been monitored for DCB and for suspended particulates every six days (nominally) since at least 1986. Most of the measurements have detected little or no DCB, except for March 1989 (when 29 µg/m3 was detected), March 1990 (19 µg/m3), and April 1990 (14 µg/m3). There are no available comparison values for DCB in air. The chemical is considered a potential carcinogen.

As part of pre-design investigation of the Bofors site, a contractor for the U.S. EPA and the U.S. Army Corps of Engineers conducted monthly air sampling and analysis from December 1992 through November 1993. The sampling location changed from month to month and was generally downwind of the lagoons area. In April and May 1993, the sampling location was on the east side (downwind) of the existing on-site landfill, and no VOCs were detected in the samples from those two months. None of the samples collected during the year contained DCB or benzidine. Concentrations found of other contaminants of concern are summarized in Table 11 (40).

B. Off-Site Contamination

Groundwater

During the course of the investigations at the Bofors site, approximately 60 monitoring wells, singly or in clusters, have been installed around the perimeter of the site (See Figure 2). Since 1981, well MW-48 (north and upgradient of the site) and individual wells in clusters MW-51 (west of the site) and MW-52 (south of the site, across Big Black Creek) have sporadically contained contamination (Table 12). During the RI in June 1988, water collected from the 44-foot-deep well in the MW-51 cluster west of the site perimeter contained 14 ppb toluene, 12 ppb total xylenes, 8 ppb acetone, and 5.9 ppb lead, marking the westernmost extent of the contaminant plume. No other off-site monitoring well contained any detectable chemical, except for an estimated 2.6 ppb lead in the deepest well (117 feet deep) of the MW-52 cluster, during the RI and no other contaminants of concern were detected in the MW-51 cluster wells.

The organic chemicals occasionally found in water from the MW-52 cluster, south of Big Black Creek, may have been the result of the groundwater from the site passing under the creek. The operation of the purge well system would prevent further migration. It is also possible that the contaminants might have been carried from the site through the air, deposited on the ground near the off-site monitoring wells, then leached into the groundwater, as described above for the DCB contamination in MW-48. The concentrations of inorganic chemicals in the water from the cluster were within typical background ranges.

A private well serving a house approximately 1 mile south of the Bofors site (the residence of the bladder cancer patient mentioned in the Health Outcome Data Section above) was sampled by the MDNR in March and April 1988 and by the Muskegon County Health Department in July 1990. The MDNR samples were analyzed for the U.S. EPA Priority Pollutant List of volatile and semi-volatile organic chemicals plus aniline, benzidine, 2-chloroaniline, and DCB. The March 1988 sample contained 5 ppb toluene and 46 ppb bis(2-ethylhexyl)phthalate and no other chemicals at detectable levels. The presence of these chemicals was not confirmed in the April 1988 sample, which contained no detectable organic chemicals (41). The July 1990 sample was analyzed for semi-volatile organic chemicals, aromatic amines, and standard water quality parameters. None of the organic chemicals or aromatic amines were found and the water quality parameters were within normal ranges (42). The bis(2-ethylhexyl)phthalate concentration did exceed the comparison value (a CREG of 3 ppb), however, the presence of the chemical was not confirmed in a follow-up sample and the chemical is also a common laboratory contaminant. Groundwater at the site flows southward, however, the contamination plume does not extend beyond the site boundaries, being contained by the purge well system, and, under natural conditions, the shallow groundwater is thought to discharge to Big Black Creek (1). Groundwater south of Big Black Creek flows northwards towards the creek, under the influence of the purge system.

Private industrial wells north and west of the site have not shown any contamination (16). Groundwater at the site flows southward, hence these wells are not considered to be in danger of contamination from the site.

Soil

In June and August 1980, the MDNR Air Quality Division collected samples of surface soil(7) from the Sun Chemical property west of the Bofors site and from 25 locations along Evanston Road from 0.6 miles east of the site to 0.6 miles west of the site. One of the samples from Sun Chemical, which was visibly stained yellow, contained 13 ppm of DCB. Samples collected from along Evanston Road 0.3 miles or further northeast of Bofors contained no detectable DCB (detection limit 0.05 ppm). A sample of tree bark from 0.15 miles northeast of Bofors and 125 feet north of Evanston Road contained 11 ppm DCB. Soil collected from the same area contained no detectable DCB. A soil sample collected 50 feet north of the road and 0.15 miles northeast of Bofors contained 2.2 ppm DCB. A sample collected from the north side of the road 0.6 miles west of Bofors contained 0.077 ppm DCB, while one collected from the south side of the road contained no DCB (2).

In late 1980, a MDPH soil survey found a range of 0.010 to 7.12 ppm DCB in surface soil from eight sites sampled east of Bofors. The samples were taken from roadside locations ranging from 360 feet east of the site property line to places on Wolf Lake Road approximately 0.5 miles east of the site. That survey also found as much as 2.2 ppm of DCB in surface soils and road sediments north of and along Evanston Road (3). Another survey conducted by the MDNR and Bofors-Nobel in 1984 found up to 3.01 ppm DCB north of Evanston Road near the site, as much as 0.85 ppm on the north side of the road half of a mile west of the site, 1.18 ppm a quarter mile east and up to 0.07 ppm half of a mile northeast (6).

Household Materials

In early 1981, a track-out study found DCB in the homes of 9 Bofors employees. Samples collected from vacuum cleaner bags had up to 10.5 ppm DCB, and dryer lint contained up to 0.74 ppm (4).

Workers and Family Urine Tests

In January 1981, the MDPH collected urine samples for analysis for DCB from at least 12 Bofors workers and at least 24 of their family members.(8) Ten of the workers had measurable DCB in their urine, between 15 and 281 ppb. Three family members also had measurable DCB in their urine, between 5.6 and 31 ppb (4). Lomac, Inc., collects urine samples from their workers on a quarterly schedule.

Surface Water

The first major reported contamination of Big Black Creek occurred in 1965 when a dike surrounding a lagoon failed, and the spillage reached the surface water. In 1967 and 1970, MDNR surveys recorded the presence of slime growths as far as 3 miles downstream from the site. Slime growths can indicate that some forms of organic chemical enrichment have occurred. The Michigan Water Resources Commission concluded that the creek was being degraded by seepage and spills from the Bofors site. Dikes surrounding the lagoons at the site failed again in 1975.

DCB and benzidine were not detected in stream water samples collected in 1982 (5). Big Black Creek water samples collected by the MDNR in 1986 and 1987, and by the RI contractor in 1988 showed no detectable contaminants associated with the Bofors site.

Sediments

Sediment samples collected from Big Black Creek immediately downstream of the site in 1982 contained up to 16 ppm of benzidine and 52 ppm of DCB (5). Sediment in the area where contamination was found was removed in 1985. The MDNR could not provide any post-removal sampling data to verify the completeness of this removal (43). Sampling from the same locations during the RI in 1990 showed no detectable concentration of these chemicals.

Biota

Mudminnows collected during the 1982 study, from the section of the creek that had contaminated sediment, contained 0.01 ppm benzidine and 0.012 ppm DCB. A grass pickerel and mudminnows from the creek 2 miles downstream of the site and sculpin from the creek 4 miles downstream contained no detectable benzidine or DCB (5).

The MDNR collected 3 brown trout, 7 carp, and 5 white suckers in May 1987 from the creek at the U.S. 31 bridge, approximately 5 miles downstream from the Bofors-Nobel site. Analysis for mercury, PCBs, polybrominated biphenyls, DDT, chlordane, and other pesticides and pesticide residues found no contamination at or above the MDCH's levels of concern (Table 13). These fish were not analyzed for DCB or benzidine, because these chemicals are not included in the array of chemicals the MDNR generally analyzes for in fish tissue (44).

Other than in some barren areas near the lagoons, there is no evidence that vegetation on the site is currently affected by the contamination. A 1978 aerial infrared photograph of the site area indicated stressed vegetation in the valley of Big Black Creek, especially in the area immediately downstream of the site.

Air

Air sampling downwind from the Lomac plant area in June and July, 1988 found up to 12 µg/m3 toluene, 1.7 µg/m3 tetrachloroethylene, and 0.45 µg/m3 total xylenes. Samples collected upwind of the plant contained up to 0.32 µg/m3 toluene and 0.24 µg/m3 total xylenes. The toluene and tetrachloroethylene concentrations were below the comparison values, and there are no comparison values available for xylenes in air (compare with Table 10). The assessors could not document any other off-site air sampling.

C. Quality Assurance and Quality Control

Quality assurance and quality control information was only available for some of the data examined during preparation of this document. In preparing this public health assessment, 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 public health assessment depends upon the reliability of the referenced information. Any qualifications reported in the sources for the cited data are discussed with the data.

D. Physical and Other Hazards

A few piles of debris, consisting mostly of concrete slabs and large metal storage tanks, were observed on the site during MDPH and ATSDR field visits.

Most of the site is fenced and the fence is maintained; however, the fence does not cross Big Black Creek, but follows the north bank. The site includes some land on the south bank, which is not fenced. The MDNR has concluded that the primary contamination is within the fenced area. All visitors to the MDEQ-controlled section of the site must be accompanied by MDEQ personnel. MDEQ personnel report seeing human tracks from trespassers on the site. They have found and repaired a hole cut in the fence, apparently by hunters to provide an exit point for deer from the site property that they can easily monitor (46). Lomac Inc. is responsible for security at the operating plant on the site.

PATHWAYS ANALYSES

To determine whether nearby residents are exposed to contaminants associated with an NPL 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 at the same time. 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.

A. Completed Exposure Pathways

Groundwater

The groundwater at the site is contaminated with DCB and other chemicals. Waste water from the chemical plant on the site was formerly disposed of in lagoons south of the plant. Chemicals in the waste water settled out with the sediments in the lagoons, and water seeping through the sediments carried the chemicals into the groundwater. In addition, spills and leaks during plant operations may have released chemicals to the soils in the plant area. Air releases of DCB and other chemicals and track-out of the chemicals on workers' shoes and clothing also transported the chemicals beyond the plant boundaries, where the chemicals settled on the soil. Rainwater seeping through the soil could carry chemicals from the soil into the groundwater. This is the probable route by which DCB contamination reached monitoring well MW-48, which is north and upgradient of the site.

Use of the groundwater for consumption or other household purposes, such as washing and showering, could result in exposure to the chemicals by ingestion, by dermal contact, or by inhalation of volatile chemicals that volatilize during use of the water. Wells that serve the plant on the site have been contaminated with DCB. The contamination was mainly found in a process-water supply well, though one water sample from a shower/locker room also contained some DCB, possibly through cross-connection between the process water system and the potable water system. When the contamination was found, the company had already been supplying their workers with bottled water for drinking because of worker complaints about the taste and odor of the well water. On MDPH/MDEQ advice, the company has continued to supply bottled water to their employees. This exposure pathway was complete in the past, but the company has taken precautions to eliminate the exposure by this pathway. The process-water well that was contaminated is no longer in use, and the company has implemented additional precautions to prevent process water from reaching the potable water system in their facility. Recent samples of the water from showers in the plant have not contained detectable DCB.

Two wells which supply showers and sinks in two buildings in the Lomac complex contained tetrachloroethylene contamination on one sampling. Later sampling of water from showers in the complex has found no tetrachloroethylene. As of the end of 1993, one of the two wells had been taken out of service, and the company is planning to take the other out of service in the near future (15, 16, 17). Workers using the showers and sinks supplied by these wells could have been exposed to the chemical via dermal contact, incidental ingestion, and inhalation as the chemical volatilizes during use of the water.

The contamination has only been identified in the shallow aquifer, which flows southward, towards Big Black Creek. Data from the 1988 RI indicates that the contamination is mostly south of the plant and lagoons area, though there are earlier reports of contamination in wells north (upgradient) of the plant. The Creek acts as a local groundwater discharge area, with shallow groundwater flow converging at the creek from both north and south, as shown in Figure 3. The purge well system reportedly has been able to prevent the contaminants in the site groundwater from reaching the creek. There have been only occasional, low-level detection of site-related contaminants in monitoring wells on the south side of the creek, and none at all during the most recent sampling cycle in 1988 (Table 12). The presence of the contaminants in the off-site wells is not likely to result in adverse health effects, given the short times and low concentrations are documented. The concentrations of contaminants in the water sampled from the purge wells have decreased with time, as seen in Table 5.

There appears to be a 140-foot-thick clay layer separating the aquifer in the glacial overburden from the water-bearing sandstone bedrock. The RI did not include any investigation of the bedrock aquifer to identify any contamination that might be present and to establish how thoroughly the two aquifers are separated by the clay layer. Potable water supply wells in the area tap only the upper aquifer because the bedrock aquifer contains high concentrations of salts and other minerals.

All residences and industries near the site which may have water supply wells are located up-gradient of the site or far beyond the creek where the site groundwater discharges. As mentioned above in the Off-Site Contamination section, private wells north, west, and south of the site that have been tested have not been confirmed to be contaminated (16, 41, 42).

Future development of the site area for residential purposes, with an on-site well for household use, is highly unlikely to occur before extensive remediation occurs.

Air/Off-Site Soil

Air monitoring at the Lomac plant boundary has occasionally detected emissions of DCB. People in close proximity to the site might inhale the DCB. The DCB in the air or on fugitive dust settles on soil, vegetation, and other surfaces near the site. Soil samples collected in 1980 and 1981 as far as 0.5 miles from the site contained DCB. People walking in the area might come into contact with contaminated soil or vegetation, and thereby be exposed to the chemical by dermal contact or incidental ingestion.

On-Site Workers and Track-Out

Inspections by the MDPH Bureau of Environmental and Occupational Health (BEOH) in 1985 (18) and 1989 (19, 20) have documented improper occupational health and safety practices in the Lakeway/Bofors/Lomac plant, which practices have resulted in workers being exposed to the chemicals used in the processes through dermal contact and incidental ingestion. A MDPH-BEOH return inspection in June 1990 found lower concentrations of DCB in wipe samples from surfaces in the work areas. The MDPH considered the contamination cited in March and October 1989 to be abated (47). A 1993 MDPH-BEOH inspection of the Lomac operation found measurable DCB in the work areas (21). Lomac and the MDPH/Michigan Department of Labor(9) have come to an agreement on a 4-year monitoring program to control DCB exposure. The MDPH inspected the plant again in 1995, and found the company in compliance with health and safety requirements (22).

Sampling in workers' homes in 1981 detected DCB in house dust, indicating that members of the workers' families were probably exposed to the chemical through dermal contact and incidental ingestion. Analyses of urine samples from workers and their family members in 1981 detected DCB, indicating that exposure has occurred (3).

If the workers' houses to which track-out had occurred had not been adequately cleaned, occupants of the houses, including later occupants, might continue to be exposed to DCB.

On-Site Soil and Sludges

The surface soil and sludge on the site contain DCB and other chemicals at levels of concern. People who enter the site without protective clothing could come into contact with contaminated surface material, and be exposed to the chemicals by dermal contact, incidental ingestion, or inhalation of fugitive dust. DCB and other chemicals have been found in the soil on the Lomac facility, where plant workers and other visitors can come into contact with the soil.

The most contaminated areas of the site are fenced, which limits the occurrence of trespass on the site. Access to these areas is limited to authorized personnel only. Authorized personnel are required to wear appropriate protective clothing and will most likely take appropriate precautions to minimize exposure to the chemicals. MDEQ have seen tracks on the site from trespassers (46). Any trespass exposure is likely to be brief and infrequent. Therefore, this is not likely to be a significant exposure pathway.

B. Potential Exposure Pathways

Biota

Consumption of contaminated game harvested near the site or contaminated fish caught downstream from the site is a potential pathway for human exposure. MDPH personnel noted the presence of a tree stand, probably used by a deer hunter, located between the perimeter fence and the creek. MDEQ personnel found a hole cut in the site fence, apparently by hunters to provide an exit point from the site for deer that could be easily monitored (46). There are no reports available of game animals from the site vicinity being analyzed for site-related chemicals. During a visit to the site in May 1991, MDPH personnel observed people fishing in Big Black Creek in the site vicinity, although fishing pressure on the creek appeared to be light. The species of fish, mudminnows, that showed traces of DCB and benzidine in the 1982 study is not normally caught for consumption; the chemical analysis was done on a whole-fish basis, rather than the fillet, and therefore may not be representative of the portion which may be eaten. One grass pickerel collected from the creek 2 miles downstream of the site at the same time was filleted before analysis, and no DCB or benzidine was found. The contamination in the mudminnows may have been associated with the sediment contamination. Although, sediment from the contaminated areas of the creek was removed in 1985, there is no record of follow-up sampling to verify that all the contaminated sediment was removed. The sediment was not found to be contaminated in a 1990 sampling from the same area. Brown trout, carp, and white suckers from the creek were collected for standard chemical analysis in May 1987, with no elevated levels found, but the analysis did not include DCB or benzidine. All fish in Big Black Creek were killed using rotenone by the MDNR later in 1987, and the creek restocked with game fish.

Surface Water and Sediment

Contaminated groundwater discharged to Big Black Creek before the purge system went into operation. Overflow from the lagoons and spillage when the dikes failed also carried contaminated water and sludge into the creek. Surface runoff can carry contaminated soil from the plant area into the creek. Contaminants that tend to bind to soil will bind to the sediments in the creek. People using the creek for recreation might be exposed to contaminants in the water and sediments by dermal contact, incidental ingestion, or inhalation of volatile contaminants. Some early studies of the creek indicated that the creek water quality was impaired, attributed to impacts from the Bofors site. In 1982, DCB and benzidine were found in creek sediments downstream of the site. More recent analysis of the water and sediments from the creek found no contaminants related to the Bofors site. MDPH personnel visiting the site saw few signs of recreational use of the creek. The nearest drinking water intake downstream of the Bofors site on record is in Lake Michigan, 10 miles away, supplying the Muskegon municipal water system. Dilution in Big Black Creek, Mona Lake, and Lake Michigan would reduce the concentrations in the surface water of any contaminant associated with the Bofors site below detectable levels and below levels of health concern before the contamination would reach the Muskegon intake. The MDPH Water Supply Division/MDEQ Division of Drinking Water and Radiological Protection has no record of any report of contamination in the Muskegon municipal water system from their routine monitoring program (15).

PUBLIC HEALTH IMPLICATIONS

A. Toxicological Implications

This section discusses the toxicology of the major contaminants of concern found in various media at the site. As discussed in the Pathways Analyses section, the primary completed pathways for human exposure to the contaminants of concern at the site are through use of the groundwater (at the on-site plant in the past) and off-site exposure to airborne contaminants through inhalation and dermal contact. In addition, workers at the plant have been exposed to the chemicals at times through improper workplace practices, and the workers' families have been exposed through chemical track out from the plant on the workers' clothing to their homes. Remedial workers and trespassers in the lagoons area might also be exposed to contaminated surface soil, but the exposures are not likely to be significant.

The estimates of doses used in this section are based on an adult weighing 70 kilograms, drinking 2 liters of water per day, and incidentally ingesting 100 milligrams of soil per day. Concentrations used are the highest found in the surface soil and groundwater samples from the site. Estimated exposure doses are compared with ATSDR-developed Minimum Risk Levels (MRLs) and U.S. EPA-developed Reference Doses (RfDs). Doses below the MRLs and RfDs are not generally expected to result in non-cancer adverse health effects. The risks of developing cancer are evaluated separately from the exposure doses and U.S. EPA-developed cancer potency factors. The risk of developing cancer is considered significant if it is estimated that more than one additional case of cancer would occur in a population of one million people exposed over their lifetimes.(10)

DCB

There are no MRLs or RfDs for DCB. Animal testing has linked consumption of DCB with liver damage and neurological effects. Regular consumption of water containing the DCB concentration found in the groundwater at the site is not likely to result in a dose equal to those at which adverse health effects have been observed in laboratory animals. No one is likely to incidentally ingest enough DCB from the soil on the site to obtain the doses at which adverse health effects have been observed in laboratory animals. Laboratory animals who were fed DCB suffered increased incidence of cancers of the bladder, liver, and other organs. The U.S. EPA has classified DCB as a probable human carcinogen (U.S. EPA Class B2) (48). Lifetime consumption of water containing the concentrations of DCB found in the water system of the Lomac plant might result in a significantly (more than 1 in a million) increased risk of developing cancer. At the time this contamination was discovered, the company was already supplying bottled water for their employees' use. There is no information available on how long the employees may have been drinking DCB-contaminated water at the plant. Lifetime consumption of water containing the maximum concentration of DCB found in the groundwater at the site (in samples from monitoring wells) might result in a significantly increased risk of developing cancer. Lifetime exposure through incidental ingestion of soil containing the concentration of DCB found in the plant area or near the site might result in a significantly increased risk of developing cancer. The Occupational Safety and Health Administration (OSHA) regulations and NIOSH guidelines (29 CFR 1910.1007, cited in Reference 48) for DCB prescribe minimal worker contact with the chemical, use of closed or isolated systems, and stringent decontamination procedures due to the chemical's probable carcinogenicity. Lomac Inc. monitors their employees' medical condition as part of a safety and health program.

There are no federal guidelines that specify allowable DCB concentrations in air. OSHA and NIOSH prescribe the lowest attainable levels for occupational settings (48). The Michigan Acceptable Ambient Air Concentration is 0.002 µg/m3 (Reference 1, Table 6-10). Air concentrations at the Lomac plant perimeter have exceeded this value on some occasions.

DCB may degrade when exposed to sunlight under certain environmental conditions (Reference 48, p. 49). However, one product of the photodegradation reaction is benzidine, which is relatively stable in light and is a known human carcinogen.

Lomac still produces DCB. A health hazard exists for workers and their families if adequate measures are not followed to minimize worker exposure to DCB and track-out of the chemical into their homes. A similar hazard exists for remediation workers and their families.

Access to the most highly contaminated areas of the site is adequately restricted. The potential, however, does exist for exposure to contaminated soil, wetlands surface water and wetland sediments in the area between the fence and the creek. There is documented DCB contamination beyond the site boundaries. None of the creek is within the fence, and analysis of creek sediment samples has historically shown contamination. However, more recent sediment sampling found no detectable DCB contamination.

A person who ate an average of 6.5 grams per day of fish containing the maximum concentration of DCB found in fish from Big Black Creek for his or her lifetime would ingest enough DCB to incur a significantly increased risk of contracting cancer. The one fish species (mudminnows) that contained a detectable concentration of DCB is not a food fish and the sample analyzed did not represent a portion that would be likely to be eaten. The detected fish contamination was limited to an area where sediments were contaminated. More recent sampling has not found sediment contamination. It is unlikely that anyone would have eaten enough contaminated fish to ingest a dose of DCB of health concern.

Benzidine

Regular consumption of water containing the concentrations of benzidine found in water from monitoring wells on the site would result in a dose that exceeds the MRL for non-cancer endpoints. Exposure to benzidine may cause bladder damage. Animal studies indicate that benzidine may also contribute to liver, kidney, and blood damage. The dose from consumption of the groundwater is less than that at which adverse health effects have been observed in laboratory animals. Benzidine is a proven animal and human carcinogen (U.S. EPA Class A) (49).

A person who ate an average of 6.5 grams per day of fish containing the maximum concentration of benzidine found in fish from Big Black Creek for his or her lifetime would ingest enough benzidine to incur a significantly increased risk of contracting cancer. The one fish species (mudminnows) that contained a detectable concentration of benzidine is not a food fish and the sample analyzed did not represent a portion that would be likely to be eaten. The detected fish contamination was limited to an area where sediments were contaminated. More recent sampling has not found sediment contamination. It is unlikely that anyone would have eaten enough contaminated fish to ingest a dose of benzidine of health concern.

Benzidine production at the Bofors plant was halted in the early 1970s, though DCB production continues. DCB can decompose into benzidine in the environment, so exposure to benzidine might continue.

Tetrachloroethylene (Perchloroethylene or PCE)

A child whose primary drinking water supply contained the maximum concentration of PCE found in the water system of the Lomac plant could ingest enough of the chemical to exceed the RfD, though not enough to exceed the doses at which adverse non-cancer health effects have been observed in studies of animals or humans. However, no children were using this water, and the workers were supplied with bottled water for drinking. The U.S. EPA had classified PCE as a probable human carcinogen (U.S. EPA Class B2), linked to liver cancer in tests on experimental animals, but the classification has been withdrawn, pending further review (50). Workers who used the showers in the period when the PCE was present in the water system are not likely to have ingested or inhaled enough of the chemical to incur a significantly increased risk of developing cancer. The on-site air concentration of PCE has exceeded the MRL in air, but the reported off-site air concentration was much below the MRL (50). The concentration of PCE in water from monitoring wells on the site has exceeded the U.S. EPA Safe Drinking Water Act Maximum Contaminant Level, and regular consumption of water containing this concentration could result in a dose in excess of the MRL. However, the exposure dose would not be likely to exceed the doses at which adverse health effects have been observed in human or animal studies.

Butylbenzenes

There is very little information available as to the toxicity of the three butylbenzene isomers found in samples collected in 1993 from the Lomac plant showers. Exposure to tert-butylbenzene altered the cellular chemistry in rat liver cells, though the dosages and health implications of these results were not presented. There is no information available on health effects from exposure to the other isomers (51). The exposure doses to butylbenzenes from showering with the water are not likely to be high enough to be of health concern. The butylbenzenes were only detected in one sample, and not in samples collected three months before or after, so any exposure would have lasted only a short time.

Contaminants in residential well

One sample of water from a private well south of the Bofors site contained toluene and bis(2-ethylhexyl)phthalate, but a sample collected a month later did not contain any of the chemicals. The presence of these chemicals in the water from the well is not considered to be confirmed. No one is likely to ingest enough toluene or bis(2-ethylhexyl)phthalate (BEHP) from water containing the concentrations found in the private well south of the site to exceed the RfDs for non-cancer adverse health effects. Toluene has not been linked with cancer (52). Some laboratory animals whose food or water contained BEHP developed liver cancer. The U.S. EPA has classified BEHP as a probable human carcinogen (U.S. EPA Class B2) (53). A person whose water supply over his lifetime contains the concentration of BEHP found in the private well south of the site would not incur any apparent increased risk of contracting cancer.

Other Volatile Organic Chemicals

No other volatile organic chemicals have been detected at concentrations above the comparison values in water supply wells at or near the Bofors site. Several volatile organic chemicals have been detected in water from monitoring wells on the Bofors site, and we discuss them here to evaluate the potential health hazard from use of a well drilled into the contaminated aquifer at some future time before the groundwater remediation is complete.

Regular consumption of water containing the concentrations found in samples from monitoring wells on the site of benzene, 1,2-dichloroethane, and vinyl chloride would result in doses that exceed the MRLs for non-cancer endpoints published in the ATSDR Toxicological Profiles for these chemicals (54, 55, 56). Groundwater concentrations of benzene, chlorobenzene, 1,2-dichloroethane, toluene, trichloroethylene, and vinyl chloride were above either established or proposed Safe Water Drinking Act Maximum Contaminant Levels (compare Table 1 and Table 3). Chlorinated solvents are known to adversely affect the nervous system and the liver. However, it is unlikely that exposure at the concentrations found would have significant non-cancer adverse health effects, since the probable exposure doses are less than the levels at which adverse health effects have been observed in humans or laboratory animals.

Benzene and vinyl chloride are classified as proven (Class A) carcinogens, with exposure to benzene linked to increased incidence of leukemia and vinyl chloride linked to liver cancer. 1,2-Dichloroethane and trichloroethylene are classified as probable (Class B2) carcinogens, linked to liver cancer in tests on experimental animals. Exposure to the proven or probable carcinogens in the groundwater at the site is not likely to result in a significantly increased risk of contracting cancer because the probable exposures to the chemicals were of limited duration (52, 54, 55, 56, 57, 58).

Concentrations of volatile organic chemicals measured in the air at or near the Bofors site are at least 500 times lower that the OSHA Permissible Exposure Levels for those chemicals. The concentrations are within the corresponding State of Michigan Acceptable Ambient Air Concentrations, except the on-site PCE concentration (Reference 1, Table 6-10). Off-site air contaminant levels were one to two orders of magnitude less than those found on-site. The volatile organic chemicals found in the air are probably due to plant operations in the Lomac facility rather than the lagoons area, and are not expected to pose a public health concern to the community.

Metals

There is no report that concentrations of metals in water samples from supply wells on or near the Bofors site have exceeded comparison values. Concentrations of arsenic and lead in some samples from monitoring wells on the site were above either established or proposed Safe Water Drinking Act Maximum Contaminant Levels (compare Table 1 and Table 3).

Regular consumption of water containing the on-site groundwater concentrations of arsenic would result in a dose that exceeds the MRL for non-cancer endpoints. Arsenic is a systemic poison which usually first affects the digestive tract. However, consumption of water containing the maximum arsenic levels found in the groundwater at the site is not likely to result in a dose that exceeds those at which adverse non-cancer health effects have been observed. Arsenic is classified as a proven (Class A) carcinogen, linked to cancer of the skin and other organs in epidemiological studies (59). Lifetime consumption of water containing the maximum concentration of arsenic found in the groundwater at the site would result in a significantly increased risk of contracting skin cancer. There is not enough data available to evaluate the risk of contracting cancer of other organs from arsenic ingestion. Current and future use of the contaminated groundwater at the site is extremely unlikely. A September 1993 sampling of the water system at the Lomac plant found no detectable arsenic. The elevated arsenic concentrations in the groundwater were found in a very few of the monitoring wells on the site, indicating that the arsenic was probably a localized occurrence.

Lead affects the central nervous system and young, developing children are especially sensitive. Lead is classified as a probable (Class B2) carcinogen (60). A September 1993 sampling of the water system at the Lomac plant found no detectable lead. Only one monitoring well contained a lead concentration above the U.S. EPA Proposed Action Level of 15 ppb. The same well also contained high concentrations of arsenic. High surface soil concentrations of lead were found in several limited areas within the restricted area of the site. Children are not likely to be on the site.

B. Health Outcome Data Evaluation

The NIOSH study mentioned previously concluded that "there was a potential hazard from exposure to" DCB. Skin problems were common among workers exposed to DNCB, but not statistically significantly more than among unexposed workers. Other conclusions included these (7):

The summary report on the NIOSH study (7), did not include much information that would assist in our evaluating their conclusions. The size of the cohort studied and the age distribution were not clearly stated and are unavailable to us at this writing. The authors presented their bladder cancer data without comment as to the significance of ten to twelve(11) bladder cancer cases among the cohort, variously described as 66 to 124 workers.

Besides the bladder cancer cases found among workers at the Bofors plant, a resident living approximately 1 mile south of the site was diagnosed with invasive bladder cancer in 1987. The resident had not been employed at Bofors or any of the other chemical plants in the area. The report was confirmed following a review of physician records. According to the Michigan Cancer Registry, cancer of the bladder ranked fifth in the state among primary invasive cancer sites; 1,475 new cases were reported in 1987. However, in the affected resident's age group (30-34), there were only five new cases in Michigan that year (the population in that age group was approximately 800,000) (61).

Bladder cancer incidence data for 1985 through 1993 were analyzed for the three zip code areas nearest the Bofors site (49442, including the site and land to the north, 49444, southwest of the site, and 49415, south and southeast of the site). The number of observed cases was compared to the number expected based on age-specific annual rates derived from Michigan state-wide cancer incidence statistics (See Table 14, Table 15, and Table 16).(12) The incidence for residents of zip code 49442 in 1993 was statistically significantly higher than the state-wide incidence (at a 5% probability level). A similar comparison was made for invasive cancers of all sites for residents of the target zip code areas with the state-wide statistics (Table 17, Table 18, and Table 19). The incidence and rate for zip code 49415 in 1987 were statistically significantly higher (at a 5% probability level) than the state-wide results (63).

The Michigan Inpatient Data Base was reviewed for hospital discharge statistics with any mention of bladder cancer for the years 1983-87. Discharge rates, per 100,000 population, for two Bofors-area zip codes (49444 and 49442) were compared with the discharge rate for Muskegon County as a whole. The rates for the target areas were not significantly different from those for the county as a whole (see Table 20). The data should be interpreted with caution because (1) they may include multiple hospitalizations of single individuals; (2) they include any hospitalization involving bladder cancer and therefore do not reflect either incidence or prevalence; and (3) the rates are not adjusted to reflect age/sex/race differences.

Although the bladder cancer statistics cited above do not indicate a history of excess cases of bladder cancer in the community surrounding the Bofors site, the 1993 data from zip code 49442 does raise some concern. However, these analyses did not focus on the plant workers and their families who are most likely to be at risk. The 124 workers and their families would be a cohort of approximately 500 people, which would be negligible in the 74,000 estimated total population of the two zip code areas in question. In addition, the workers may live outside the areas evaluated.

ATSDR, MDCH, and local health departments are initiating a health study of workers, their families, and exposed community members in relation to the Bofors-Nobel site and two other facilities in Michigan where chemicals similar to DCB were manufactured or used. No data is yet available from this study.

C. Community Health Concerns Evaluation

Responses to comments received by the MDPH from citizens following the release of a draft of this public health assessment in 1991 are included in the Responsiveness Summary appended to this public health assessment. Some responses have been updated since the March 1992 Interim Release of this public health assessment to reflect more current information. The MDPH released a revised draft of this Public Health Assessment for public comment on March 1, 1995. The comment period lasted until March 31, 1995, and was extended to April 15, 1995, at the request of a representative of a Potentially Responsible Party (PRP) for the site. The Responsiveness Summary at the end of this document also includes responses to comments on the later draft.




Next Section       Table of Contents