Responses to comments received on the two Public Comment Drafts of the Bofors-Nobel Public Health Assessment during the Public Comment Periods ending March 29, 1991, and April 15, 1995, are presented below. The comments received during the first period, received from two citizens representing two area environmental associations, have been numbered 1 through 39. These comments and responses were included in the Interim Public Health Assessment released on March 24, 1992, and some have been updated to reflect more current information. The comments and responses from the Public Comment Period ending April 15, 1995, begin on page 13. The questions are presented as received; in some cases, similar questions are addressed with a single response.

1. Comment: Available data used in preparation of this health assessment is possibly too old to identify area health concerns.

Response: The health assessors are limited to the existing data and information available at the time of the writing of the assessment and are not able, for example, to conduct sampling to generate new data. While reviewing environmental agency documents during ongoing investigations we can, and do when necessary, make recommendations to close gaps in the data.

We solicit the health concerns of the community while conducting research for the assessment; at public meetings, during the public review of the draft; and even after a final assessment has been distributed. Whenever new information is available for a site or new questions are raised, we will consider them for inclusion in current or future assessments of the site.

2. Comment: The validity of the analysis and conclusions drawn were dependent on the reliability of referenced information. MDPH assumed that adequate quality assurance and quality control measures were followed with regard to chain-of-custody, laboratory procedures, and data reporting.

Response: We do not have the ability to personally check that the procedures and measures set out in the quality control and quality assurance plans were adhered to, but we are reasonably confident that the enforcement agencies responsible for oversight of these matters were satisfied that proper procedures were followed. We rely on the preparers of the documents to provide information necessary for analysis of results. We review draft documents that are part of the Remedial Investigation/Feasibility Study for inconsistencies, omissions, and errors. We direct questions, comments, and suggestions to the enforcement agencies before the documents become final.

3. Comment: Increased cumulative dose of toxins from Bofors/Nobel, SCA Independent Landfill, Thermo-Chem (All on NPL) not considered.

Response: The most plausible common pathway by which human exposure to contaminants from the Bofors-Nobel, SCA Independent Landfill, and Thermo-Chem NPL sites involves Big Black Creek. Contaminated groundwater and leachate from all three sites may discharge into the creek. A Muskegon County waste water treatment plant is also in the creek's watershed, upstream of Bofors. The MDNR performed several biological surveys of the creek between 1967 and 1986, and Battelle performed a biological and chemical survey of the creek in 1982 (Reference 5), under contract to Bofors Lakeway. Any local declines of water quality or biological diversity in the creek found in these surveys were attributed to discharges from the wastewater treatment plant or from Bofors. The surveys found no apparent adverse effects in the area from SCA and Thermo-Chem. A 1980 MDNR biological evaluation of leachate from SCA found no apparent effect on the creek, though one of the drains at the landfill was severely affected.

4. Comment: More testing should be done for area residents, and plant workers who live outside the area. It is probable that workers who are most exposed have not been tested for track-out.

Response: We agree that more testing should be done for current and past plant employees as indicated by our referring the site to the Agency for Toxic Substances and Disease Registry (ATSDR) for consideration of followup health activity. The evidence we have doesn't suggest that residents in the vicinity of the plant have been exposed to significant levels of contamination. We have included in our recommendations a request for further investigation of off-site contamination. Such information would help us discover the existence of any additional pathways by which people may have been or may be exposed in the future to contaminants. We are willing to accept new information and data whenever available. As significant new information is received, the assessment and recommendations will be updated.

5. Comment: Track-out probably occurs via Sun Chemical employees. Personnel from area businesses (gas stations etc.) have mentioned that they can tell the workers from Sun Chemical because they and their clothing are yellow. Lomac provides DCB to Sun Chemical for production of yellow-dye [properly, a yellow pigment]. They are located next to each other.

Response: MDCH⁽¹³⁾ considered expanding the follow-up health activities to include other industries where workers and their families may have been exposed to similar contaminants. ATSDR, who is funding the study, has decided that the study must initially be limited to the three Superfund sites targeted, Bofors, Anderson Development in Adrian, and Roto-Finish in Portage. If, at a later date, it is determined that the study group should be expanded, Sun Chemical workers and their families will be recommended for inclusion.

6. Comment: DCB contamination was found up to 1/2 mile from the site. I live 2-1/2 miles downwind, and wonder if testing has been done further than 1/2 mile away.

Response: In 1980, soil testing was conducted as far as three-fourths of a mile from the site. According to the 1980 sampling data, the point at which samples did not show detectable levels of DCB in soil east of the site was near the intersection of Wolf Lake Road and Evanston Road.

7. Comment: The creek should be further analyzed for contamination especially since trout are being stocked. People get the impression that the creek is safe.

Response: In the Off-Site Contamination section, we state that "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-Nobel Site." The reference from which this is taken is the February 1990 RI/FS document (Reference 1), pages 4-81 through 4-87. Included in that section are sample results from creek sediments that also did not detect site-related contaminants. The implications for trout will be addressed in response to comment 9.

8. Comment: Other cancers incidence data (besides bladder cancer) were not analyzed for the two zip code areas nearest to the Bofors site. MANY area residents have expressed concern that the many different types of cancer they have, or have caused deaths of area residents can be attributed to the water, air, or ingestion of game hunted in the area.

Response: The incidence of bladder cancer, in particular, was among the first investigated because that type of cancer is a plausible health outcome from exposure to significant levels of DCB, benzidine, and aniline over time.

We have evaluated the incidence of all cancers for three zip code areas surrounding the site and compared the findings to county, state, and national statistics. We used Michigan Cancer Registry data for the years 1985 to 1988. We have also evaluated the data available from death certificates from 1980 to the present. Our evaluations are included in the "Health Outcome Data Evaluation" section of this draft.

9. Comment: The statement, "Fish samples from 1982 and 1987 were not found to be contaminated." is vague. Contaminated with what? Where were the fish samples taken? No contaminants were detected? At any levels?

Response: In 1982, Battelle/Columbus Laboratories, working for Bofors-Lakeway, carried out a biological and chemical assessment of Big Black Creek. They sampled water, sediment, fish, and other aquatic life at seven stations, two upstream and five downstream of the Bofors-Nobel site. The overall conclusion was that the aquatic ecosystem of the creek was not significantly affected by the Bofors-Nobel site. Fish, specifically mudminnows, sculpin, and a grass pickerel, from three stations downstream of the site were analyzed for DCB, benzidine, and aniline. One sample made up of five whole mudminnows combined contained 10 ppb benzidine, 12 ppb DCB, and 6 ppb aniline. No other fish sampled contained detectable levels of benzidine or DCB (4 ppb and 12 ppb, respectively), and they all contained aniline, up to 6 ppb. Aniline was detected at similar levels in a quality control sample, indicating that the levels detected may be an artifact of the analytical method. Sediment collected from the same location as the five mudminnows contained as much as 51,900 ppb DCB, 16,200 ppb benzidine, and 124 ppb aniline. Sediment from locations where the other fish samples were collected contained up to 13 ppb DCB, 7 ppb benzidine, and 8 ppb aniline. Analysis of water from the creek found no detectable DCB or benzidine and up to 0.6 ppb aniline. As with the fish samples, aniline was detected at similar levels in a quality control water sample, and any detection may have been an artifact.

In May, 1987, MDNR personnel collected 15 fish from Big Black Creek at U.S. 31, 5 miles downstream from the Bofors-Nobel site. The fish were collected shortly before the creek was dosed with rotenone by the MDNR to clear the creek for stocking with brook trout. The fish collected included 3 brown trout, 5 white suckers, and 7 carp. The trout were between 7.5 and 9.5 inches in length, and were probably about 2.5 years old. The suckers were between 15 and 18 inches long, and the carp 19.5 to 25.5 inches long. The fish were analyzed for the MDNR's standard array of contaminants: mercury, dieldrin, hexachlorobenzene, octachlorostyrene, PCBs, chlordane and various related compounds, heptachlor and its epoxide, polybrominated biphenyls (PBBs), aldrin, lindane, terphenyl, toxaphene, DDT and related compounds, mirex, beta-BHC, pentachlorostyrene, hexachlorostyrene, and heptachlorostyrene. Low concentrations of many of these chemicals were found in some fish sampled but not at levels considered by the MDCH to be of health concern.

The 1987 analysis included none of the primary contaminants at the Bofors-Nobel site. Benzidine and DCB are not on the MDEQ⁽¹⁴⁾/MDCH's list of compounds of concern for fish contamination, nor is the laboratory MDEQ uses able to analyze for those compounds. The relevant section of the health assessment was revised in the interim release to more accurately reflect the situation. No new fish contamination data have become available for this revision.

10. Comment: How can MDNR personnel control visitors to the site when the site boundary is not completely fenced?

Response: The property line of the Bofors-Nobel site, now owned by Lomac, does extend south beyond Big Black Creek. Surface soil sampling during the RI detected contaminants mainly within the portion that is now completely surrounded by fence, that is, the property north of Big Black Creek (see Figure 1). Some samples from areas adjacent to the fence line contained contaminants, suggesting that there might be contaminated surface soil in some locations outside the fence. We have included a recommendation for further study of Big Black Creek and its floodplain (land bordering the creek) in the site area.

11. Comment: The creek is not fenced to adequately eliminate the potential for trackout via wildlife, etc. Deer hunters and small game hunters could be contaminated via this pathway.

Response: We would expect that physical track-out would be minimal; however, if animals ingested or absorbed the chemical and their flesh became contaminated, significant exposure to hunters and their families would be possible. Previous studies have indicated that terrestrial wildlife do not bioconcentrate chemical contaminants to the degree that fish and other aquatic organisms do. Aromatic amines such as aniline, benzidine, and dichlorobenzidine are less persistent than most chemical contaminants commonly found in fish tissue and have a lower affinity for fat. Thus, it would be very unlikely that any of those compounds would bioaccumulate to a significant level in the wildlife inhabiting the site.

12. Comment: I would like to request follow-up health activities for current and past manufacturing site workers and their families. More studies & follow-up activities should be performed for area residents also.

Response: We concur with the need for follow-up health activities for current and past manufacturing site workers and their families, and have initiated such a study with funding support from ATSDR. Based on limited off-site soil test results, we are including area residents living within a 0.5 mile radius of the site. We have also recommended that additional off-site soil testing be conducted to better determine the extent of potential community exposure.

13. Comment: Has it been determined why there is a limited number of births? Are people sterile?

Response: The quote from the NIOSH study in the Health Outcome Data section that begins with "The limited number of reproductive events preclude any determination..." does not imply that the people studied had fewer children than a similar group of unexposed people. Rather, it is saying that the total number of people that participated in the study is too small a population to statistically prove or disprove a link between exposure to oryzalin and reproductive hazards.

14. Comment: How is it known that VOC air contamination is "probably due to plant operations at the Lomac facility", and how do you know it is not expected to pose a public health concern to the community?

Response: The chemicals found in the air are used in the Lomac operation. Lomac's reports to the U.S. EPA Toxic Chemical Release Inventory report air releases of toluene and tetrachloroethylene, the chemicals found at highest concentrations in the air on the site. The highest concentrations were found around the Lomac plant. Though some of the chemicals were also found in the soil of the lagoons area, the concentrations found in the air were much higher than would be expected from evaporation from the concentrations found in the soil. Based on these considerations, the plant is the most likely source for the airborne volatile organic chemicals.

The highest levels were found on the plant property, and much lower levels were found downwind of the site. The off-site levels are within health-related guidelines. The on-site levels are within workplace standards.

15. Comment: How is it known that the purge well system is working?

Response: Through the measurement of water levels in monitor wells it has been determined that overall groundwater flow throughout the site is deflected towards the purge well system, as shown on the map of the potentiometric surface (Figure 3). The potentiometric surface reflects the pressure on the groundwater. Groundwater flows from areas of higher pressure to lower pressure, just as surface water flows from higher ground levels to lower ones. As viewed on a map of contours of equal potentiometric surface, groundwater flows approximately perpendicular to those contours, again from higher to lower. The potentiometric map shows a deep depression, a low-pressure area, in the vicinity of the purge wells, indicating that most of the groundwater at the site eventually will flow to the purge wells, where it will be pumped out. The depression is entirely surrounded by higher-pressure areas, preventing flow elsewhere.

16. Comment: Why is it assumed that chlorinated solvents and arsenic ingestion of groundwater at levels found are unlikely to have adverse health effects?

Response: Our statement regarding these contaminant levels was in error, and has been revised. Arsenic and chlorinated solvents are present in the groundwater at levels that may produce adverse health effects through chronic ingestion exposure. However, that affected groundwater is not now being used by anyone as a water supply. We are adding to our recommendations the need for institutional controls over future use of groundwater in the area between the site and the creek.

17. Comment: During a "block parents" meeting at Oakridge Community Education Building, two Lomac officials brought 2 glass jars of DCB and yellow pigment to show parents and school officials. They were there to do some sort of community outreach and let the public know what they "do" at Lomac. Is this an example of how they control track out of toxic chemicals? I guess by showing this powdery substance to a bunch of moms, our fears over it's danger was supposed to be eliminated. They went on to tell us that "DCB is what keeps our plant HEALTHY." How many Lomac workers live in the immediate area? I believe their employees are so afraid of losing their jobs that they are willing to sacrifice the health of the community, their families and themselves.

Response: We recognize the community's concern on this topic.

18. Comment: I obtained the Health Assessment for review through the Walker Library. The repository for the Bofors/Nobel Superfund Site is located at Egelston Township Hall, Apple Ave., Egelston Twp., Muskegon MI 49442. I was not able to view the health assessment through Egelston Township Hall, because when I requested it the township supervisor was not there. No one else in the office knew anything about it or was able to get it for me. I question whether a health assessment was ever sent there. The people most affected by the site live in Egelston Township. If there are interested people, I doubt whether they would travel 15 miles to Walker Public Library in North Muskegon to take a look at these documents. Could you please make sure information on Thermo-Chem, SCA Independent Landfill, and Bofors/Nobel is sent to Egelston Township Hall as well as other area repositories.

Response: A health assessment was sent to the Egelston Township Hall at the same time it was sent to the Egelston Township Library, the Hackley Public Library, and the Walker Branch of the Muskegon County Library. The document sent to the Township Hall has since been moved, together with the U.S. EPA Superfund documents, to the Egelston Township Library next door, at 5402 Apple Avenue. The Library has varying daytime and evening hours and will have further extended hours when it moves to a new building in August 1991. For particular information regarding the library hours, you may call (616) 788-2308. In addition to the repositories, we mailed copies of the assessment to individuals and groups whom we felt would be able to help make it accessible to the community. We apologize for the inconvenience anyone may have had obtaining a copy.

19. Comment: A description of the assumed limits of the contamination plume(s) would be helpful in this section. The two site maps included in the appendix do not appear to indicate the extent of the contamination as determined through any previous studies. Area citizens reading this document would be able to determine more adequately their potential for exposure to the hazards of the site.

Response: Your suggestion for clarifying this part of the health assessment is a good one. The RI document contains several plume maps each showing the estimated distribution of one particular contaminant. To reprint all of them or try to overlay them would be confusing. We have drawn a map (Figure 4) showing the estimated extent of 3,3'-dichlorobenzidine contamination in the groundwater and added it to the final draft. This chemical is representative of the most widespread hazardous substances in the groundwater at the site.

20. Comment: Has there been subsequent U.S. FDA testing of agricultural products irrigated with water from the wastewater plant? What levels are allowed in these products?

Response: To our knowledge, there has been no subsequent FDA testing of agricultural products irrigated with water from the wastewater plant. The "allowed levels" of chemicals in agricultural produce that you refer to are limited to residues of pesticides and herbicides only. There are no established standards for the types of chemicals of concern related to this site, nor are fruits and vegetables routinely tested for them.

21. Comment: The 1980-81 soil sampling survey and track-out study of DCB demonstrated the ability of this contaminant to migrate far from the site. Were the contaminated off-site soils removed or were the areas fenced?

Response: The contaminated off-site soils were not removed nor were the areas fenced. We have included a recommendation for further characterization of identified off-site contamination in the Recommendations section of the health assessment in order to detect areas of possibly higher concentration levels that may not yet have been discovered.

22. Comment: Does the site have 24-hour security?

Response: The site does not have 24-hour security.

23. Comment: The statement that, "The community around the Bofors site has shown occasional concern about conditions at the site . . ." needs to be qualified. Conversations with area residents indicate that information about the site and related health effects has been limited. In particular, information at the repository at Egelston Township is not always accessible. Residents attempting to locate information at the Walker Memorial Library have found that documents have been signed out. Press coverage of the site and its hazards has been minimal.

Response: Our statement was simply intended to document that the community has at times expressed concerns regarding this site. We agree that information about the site and health effects has been limited. Regarding the particular comment about access to information, please see response number 18. The documents that were available in the Walker Memorial Library were circulated in error in the past. That problem has been corrected, and we apologize for the inconvenience this may have caused.

When we release a Public Comment Draft or a final health assessment, we issue press releases to as many representatives of the media in the area of the site as possible. For example, for the Bofors-Nobel Health Assessment, we sent releases to nine press offices representing the major radio, television, and newspaper media serving Muskegon County.

24. Comment: The account of the NIOSH study raises many questions. In addition, the study is outdated. There must be more recent and conclusive information.

Response: We have contacted the author of this comment for more details. We concur that more current information is needed as reflected in our referral and acceptance of this site for follow-up health activity consideration. We recommend that anyone with concerns about the NIOSH study call us with particular questions so that we can determine if they are best directed to the Division of Health Risk Assessment in the Michigan Department of Public Health Bureau of Environmental and Occupational Health, or to NIOSH.

25. Comment: Are advisories restricting recreational activities in order for Big Black Creek and Mona Lake?

Response: There is currently a fish consumption advisory in effect for Mona Lake. Carp from the lake should not be eaten because of contamination with polychlorinated biphenyls (PCBs). The source of the PCBs has not been determined. The statewide general advisory for inland lakes, impoundments, and reservoirs because of mercury contamination also applies to Mona Lake. People are advised not to eat more than one meal per week of fish of the following kinds and sizes from any inland Michigan lake, impoundment, or reservoir: rock bass, perch, or crappie over 9 inches in length; largemouth bass, smallmouth bass, walleye, northern pike, or muskie of any size. Nursing mothers, pregnant women, women who intend to have children, and children under age 15 should not eat more than one meal per month of the fish species listed above (see Reference 14).

There is no evidence available for the need to further restrict recreational activities at Big Black Creek and Mona Lake. The Michigan Department of Public Health has asked the Fish and Wildlife Contaminant Advisory Committee (an interagency advisory committee) to consider including Big Black Creek in the next round of sampling. They will decide to accept or reject the suggestion based on contaminant data from the surface water body and sediments, and the availability of laboratory services capable of testing for the contaminants of concern.

26. Comment: Why were only 12 of the 27 "Potential Compounds of Concern" considered in the Remedial Investigation? Will this lead to an underestimation of risk?

Response: A question regarding the criteria by which an agency selects "Potential Compounds of Concern" during a Remedial Investigation is best directed to that agency. Please contact U.S. EPA Regional Project Manager John Fagiolo at the Region V office in Chicago at (312) 886-0800. For the purposes of the public health assessment, ATSDR has developed a contaminant screening method to select contaminants of concern which we feel confident will not underestimate risk, described in brief on pages 9-11. MDCH and ATSDR would be happy to discuss this process in more detail with anyone interested.

27. Comment: Have all the potential sources of contamination been determined?

Response: According to the data available to date, all the primary sources of contamination appear to be within the Bofors-Nobel site boundaries. Regardless of the actual source of the contamination, the remediation will be applied to all known contaminated media.

28. Comment: Have the 1989 and 1990 air monitoring measured other contaminants than DCB?

Response: Only DCB and particulate emissions from the Lomac operation are regularly monitored by the Air Quality Division (AQD) of the MDEQ, through data reported by the company. AQD controls the emission of other chemicals, including toluene and tetrachloroethylene, from Lomac, as well as at other plants in the state, through a permitting system, without monitoring. The maximum rate of loss to the air of various chemicals and the emissions control systems that must operate are specified in permits issued by AQD. The permissible emission rates are designed to have minimal effect outside the plant boundaries. AQD keeps track of the emissions by monitoring use of the chemicals in the plant. They also test the required emissions control systems at intervals. AQD personnel have reported that one of the toluene emission control systems at Lomac is not working properly. The penalties specified in their permit will be levied until the system is repaired.

29. Comment: The statement, "Fish samples from 1982 and 1987 were not found to be contaminated." is vague. Contaminated with what? Where were the fish samples taken? No contaminants were detected? At any levels?

Response: Please see the response to comment 9.

30. Comment: The documented lack of quality assurance and control suggests the necessity of gathering more valid data.

Response: The health assessment noted the lack of quality assurance and quality control (QA/QC) information, particularly on data from prior studies. The data from past studies may only be available through secondary sources, and, if the documentation of the QA/QC is not available for them, it cannot be produced now. In the absence of evidence to the contrary, the Remedial Investigation, our primary source for data representing site conditions as they are at present, is believed to have followed appropriate QA/QC practices. Please see the response to comment 2.

31. Comment: How is it known that the clay layer is continuous? At a Whitehall Superfund site, that same hypothesis was advanced for years until it was found that the clay layer in question was not continuous and had potentially allowed contaminants to reach the lower aquifer.

Response: The RI states that the clay layer is believed to be continuous, but does not indicate that hypothesis was tested. The layer was found throughout the site from soil borings and well log records, but its thickness was determined only in one deep till aquifer well on-site and found to be 140 feet at that location. No testing of the deeper sandstone aquifer was reported in the RI. We added a recommendation to the Interim Public Health Assessment addressing the need to investigate the deep aquifer and the adequacy of the till (clay) layer in confining it from the contaminated aquifer.

32. Comment: What percentage of the groundwater flow is intercepted by the purge well system?

Response: Please see the response to comment 15.

33. Comment: Why is it considered unlikely that exposure to the arsenic and chlorinated solvents in the groundwater would have significant impacts?

Response: Please see the response to comment 16.

34. Comment: Why is it that the volatile organic chemicals found in the air are not expected to pose a public health concern to the community?

Response: Please see the response to comment 14.

35. Comment: Are there any studies available from which to determine the synergistic effect of the contaminants?

Response: There is a great deal of scientific information to document the occurrence of interactive effects from simultaneous exposure to two or more chemicals. Such interactions may include additive, antagonistic, synergistic, and other interactive effects. However, there is very little information that can be used to develop exposure evaluations that can be adjusted to account for the potential effects of exposure to other chemicals. Even if such chemical-specific information were available, it would be impossible to develop a set of evaluation guides that could account for the vast array of chemicals, at varying proportions, that may be found at most sites.

36. Comment: The potential for exposure to the contaminants to cause cancer is a very serious concern. Will non-cancer health effects be manifested also? If so, what?

Response: Exposure to a relatively large quantity of any of the chemicals of concern usually is required for non-cancer health effects to arise. Since the most contaminated areas of the site are fenced off and no one is using the contaminated groundwater, site-related exposure at hazardous levels is unlikely. The health assessment includes brief descriptions of the non-cancer health impacts from exposure to some of the chemicals of concern. That section has been revised and expanded in the final draft.

37. Comment: Comparing the bladder cancer incidence data of the two zip code areas nearest to the Bofors site to those of Muskegon County as a whole is not appropriate. Cancer, many different kinds, is simply a very common condition of residents in this county.

Response: Please see the response to comment 8.

38. Comment: There could be many more [recommendations] listed in this [Recommendations] section, for example:

a. Continuing and widening the track-out study of DCB.

b. Gathering evidence of cancer and non-cancer health effects by surveys and interviews (extensive).

c. Conducting tests to determine if wildlife exposure has occurred and contaminants are accumulating in the food chain.

d. Gathering more recent and valid health data.

e. Urging immediate follow-up health studies, or assessments, activities, etc.

Response: We believe that your recommendations "a", "d," and "e" are either underway or are included in the follow-up health activities that we are recommending to ATSDR for consideration. For particular information regarding "d," see the response to comment 8. Regarding recommendation "c," see the response to comment 11. As for your recommendation "b," we have applied for and received funding support from ATSDR to conduct cancer screening in workers, their families, and community residents living within 0.5 mile of the site. This study will include interviews with all participants. See also our responses to comments 5 and 12.

39. Comment: We, in Muskegon County, are in the midst of a serious environmental health crisis. The common occurrence of cancer in our communities is becoming a very frequent topic of conversation. It must be acknowledged by local and state health officials.

Response: We are continuing to examine cancer morbidity and mortality for Muskegon County and geographically smaller local areas. We have evaluated the incidence of cancer for Muskegon County in the past, and as new data become available, we will continue to try to identify divergences in actual disease incidence from expected disease incidence. To date we have not seen any data that leads us to the conclusion that cancer is occurring any more frequently in Muskegon County than in comparable populations elsewhere. As mentioned earlier, the revision of the draft health assessment following the public review draft contains an expanded section to clarify these comparisons.

Comments on the Draft Public Health Assessment (PHA) received between March 1, 1995, and April 15, 1995.

Comments from the Bofors-Nobel Potentially Responsible Parties (PRP) Committee:

General Comments (For space considerations, only the headers and excerpts from the comments are cited)

The PHA is based on limited and outdated data which are not indicative of current site conditions. ... The PHA does not indicate, however, that as part of the of the consent agreement, Bofors-Nobel, Inc., removed contaminated creek sediments in 1985.

Response: If there is more extensive or recent data available, ATSDR and MDCH would appreciate a citation or a reference. Historical exposures are relevant to the health hazard posed by the site, since the major contaminants of concern include benzidine, classified by the U.S. EPA as a known human carcinogen (Class A), and DCB, classified by the U.S. EPA as a probable human carcinogen (Class B2). Cancer may not appear until decades after exposure to a carcinogen. The Remedial Investigation Report (Reference 1), the primary source for the information contained in the RI, did not include any information on the sediment removal in 1985. The assessors have obtained information on the removal from the MDEQ.

It is essential that the assessment of the potential risk of exposure to the site, as it exists today, be presented to put the current community risk into perspective. The characterization of the public health risks posed by the site as it exists today would fulfill the goal of the document as stated in the PHA Foreword: "to find out if people are being exposed to hazardous substances and, if so, whether that exposure is harmful and should be stopped or reduced." The current limited site access and operating groundwater purge well system with groundwater treatment (which forms a hydraulic barrier and prevents contaminated groundwater from migrating offsite) should be the focus of the PHA rather than the conditions that existed in the past which have already been modified to reduce the potential risks to human health and the environment. The community needs to be presented with accurate information that presents the potential health risks of the site but does not raise concern of a public health threat where none exists. The basis for any potential future actions to reduce the public health risk should be based on the current risk posed by the site not on the past risk.

Response: The current restrictions on access to the site and groundwater treatment system do provide some limitation on human exposure to the contaminated media at the site. However, past potential human exposures to contaminants are considered in the assessment because the health impacts may continue long after the exposures are stopped.

The document would be more beneficial to the public if the risks from the operating production facility and the Superfund site were separated.

Response: The Superfund site includes the operating production facility. Any potential human health risks in the vicinity of a Superfund site are within the purview of an ATSDR Health Assessment.

The Michigan Department of Public Health failed to recognize all potential sources of contaminants of concern ("COCs") in the community. The Michigan Department of Public Health relied on the Toxic [Chemical] Release Inventory ("TRI") to identify sources of emissions to groundwater, surface water, and air in the community which may have contributed to environmental contamination in the area. Nine facilities were identified in the TRI but all were located approximately three or more miles west of the site. Sun Chemical, a manufacturing facility located immediately adjacent to the site was not identified in the TRI. The absence of this facility in the database, as well as a foundry and metal fabricating facility located within ½ mile of the site, calls into question the reliability of using the TRI to identify other potential sources of COCs in the surrounding community.

Response: ATSDR and MDCH health assessors use the TRI as one of several resources to identify contaminants of concern and sources of contamination. Therefore, the completeness of the TRI database is not critical to the health assessors' use of it. The coverage of the TRI is known to be limited, in that only facilities conducting certain industrial activities handling more than certain quantities of chemicals are required to report. The original TRI search found ten facilities, including Lomac, Inc., in the two zip codes nearest to the site. More recent searches of the database found reports from several additional facilities, including one located across the road from the Bofors site. The original TRI search, by zip codes, did not find the reports from Sun Chemical because the TRI reports from the facility included an out-of-date zip code.

The potential public health risks posed by the site, as presented in the PHA, are based upon extremely conservative and unrealistic exposure assumptions. ... The use of the highest recorded site concentrations results in an estimate of unrealistically high site risks. ... Assuming [pica] behavior as the norm in children for determining the potential public health risk of a site is unrealistic, especially in light of the fact that it is unlikely that the site will ever be developed for residential use. ... The PHA also raises concerns of potential risk where the likelihood of exposure is highly improbable or impossible.

Response: The methods and assumptions used in the PHA are prescribed by the ATSDR. The maximum detected concentrations of contaminants in the environment provide a conservative estimate of exposure when the available data is limited. ATSDR EMEGs and RMEGs (cited in the full comment), calculated assuming pica behavior, were used to select contaminants for further consideration. The ultimate decision on whether the site poses a health hazard is made using a more likely exposure scenario.

The PHA does not address the potential risk to the workers or community associated with implementation of remedial action.

Response: In the Public Health Assessment process, ATSDR assesses the health risks associated with sites as they are, and does not address the risks of remediation or residual risks after a proposed remedial action is complete. An assessment may include recommendations that the remediation of a site address certain contaminated media, but the precise remediation technology to be employed is left to the environmental agencies to select. ATSDR and MDCH do consult with the environmental agencies, when requested, on the health impact of remediation technologies, but this is separate from the Public Health Assessment process.

Past, present, and future risks associated with the site should be addressed separately for the PHA to be useful in communicating the potential risks at the site to the public and developing appropriate actions to reduce potential future risks. It is appropriate to use past site conditions and data to estimate risks associated with past exposures. It is, however, inappropriate to use this outdated data as a basis for determining the current potential site risk and conclude that the site continues to pose a public health risk when current site conditions are much different from past conditions. Separating the discussions of past operations, current operations, and future implementation of remedial actions will more accurately present the public health concerns at the site and allow for the development of appropriate actions to reduce these concerns.

Response: Because of the long latency period associated with cancer, and the presence of known and probable human carcinogens on the site, historical data is relevant to current human health risks associated with historical exposure to the site. Available information does not indicate that the conditions on the site have recently changed in such a way to significantly reduce the health hazard, except for the reduction in contaminant levels in the groundwater from the groundwater purge and treatment system.

Specific Comments

Comment 1: It is stated that organic chemicals have been found in various environmental media, including surface water. (Summary, ¶ 1) There is no data presented in the PHA that surface water contains, or has in the past contained, organic chemicals.

Response: The first analysis on record of surface water in Big Black Creek that was capable of identifying organic chemicals was in 1982. Groundwater contamination at the site was identified in 1976, and the company began installation of the purge well system at that time. Discharge of contaminated groundwater before the purge wells were activated and the incidents of overflow from the lagoons to the creek in 1965 and 1975 probably resulted in contamination of the creek water with organic chemicals. The presence of DCB and benzidine in creek sediments in 1982 strongly suggests that other, more water-soluble organic chemicals associated with them in materials on the site would have reached the creek, but would have been washed away with time when the contaminant transport was interrupted by the purge wells.

Comment 2: It is indicated that the site poses a public health hazard. (Summary, ¶ 2) The sentence following the conclusion states that past exposure to site-related contaminants may have occurred, but does not indicate that any current exposure pathways exist which support the conclusion that the site poses a public health hazard presently or in the future. Contaminated sediments and potentially contaminated fish have been removed from Big Black Creek. A groundwater withdrawal system is in operation which prevents contaminated groundwater from migrating offsite or to the creek and the site is fenced to restrict access to the site.

Response: The criteria for placing a site into the "public health hazard" category during ATSDR's Public Health Assessment process include "Evidence that exposures have occurred, are occurring, or are likely to occur in the future." (Reference 64, Table 8-1) The conclusion section has been revised to reflect the changes in the site situation over time.

Comment 3: ... due to extensive contamination of the groundwater, soil, surface water, and sediment ... (Page 2, ¶ 1) There is no data presented in the PHA which indicates that surface water, past or present, is contaminated.

Response: See our response to Comment 1.

Comment 4: An accidental release of hydrogen chloride gas occurred at the Lomac facility in April 1990 which resulted in the evacuation of approximately 1300 residents. (Page 3, ¶ 5) This release was related to the operating facility and not the LOU (Lagoon Operable Unit) and thus is coincidental to the Superfund site.

Response: ATSDR attempts to address all significant potential public health concerns present in the vicinity of Superfund sites.

Comment 5: Construction on the upgraded groundwater treatment facility began in February 1993. (Page 4, ¶ 2) Construction of this groundwater treatment facility was complete and groundwater treatment initiated in September 1994. The facility is currently operating and successfully treating the groundwater to concentrations below detection limits for COCs.

Response: The assessment has been revised to reflect the current status of this facility.

Comment 6: The description of the site conditions and security measures are based on site visits conducted by Michigan Department of Public Health ("MDPH") representatives in 1989 (onsite) and 1991 (site perimeter vicinity). (Page 5, ¶ 1) The assessment of site conditions is based on observations made 6 years ago onsite and 4 years ago of the site vicinity. A more recent site visit would be appropriate to determine if current site conditions are consistent with past conditions.

Response: John Filpus of the MDPH visited the site and its vicinity on July 11, 1995. His observations support those made during the earlier visits.

Comment 7: The stated population within a one-mile radius of the site is 200. Approximately 6,500 people resides within a three-mile radius of the site. (Page 5, ¶ 2) These numbers are identical to the population estimates in the 1990 Preliminary Health Assessment and the 1992 Interim Health Assessment. It is unlikely that the population has remained static in this area over a five-year period. An indication of population trends (increasing or decreasing) in the area would assist in determining the potential current and future health risks posed by the site in relation to the past risks.

Response: The assessors have re-estimated the population within one mile of the site, based on Table 6-4 in the RI report (1). The population within 3 miles has been re-estimated from 1990 U.S. Census data. The original estimate for the 1-mile population was taken from a 1987 U.S. EPA Hazard Ranking System analysis, based on a count of houses from the U.S. Coast and Geodetic Survey (U.S.G.S.) topographic maps of the area (65). The U.S.G.S. map including the site had been updated in 1981, but the 1-mile radius carried over onto an adjacent map, which had not been updated since 1972. The RI report does not give a reference for their count of residences, but assuming it was carried out during the RI in 1989, and dating the original estimate to 1980, the population growth in the area during that decade would be over 10 percent per year. Projecting that growth over the succeeding 5 years would give a population estimate within 1 mile of the site of approximately 800 persons. However, without more data on the population within the area, this is an extremely unreliable estimate. The U.S. Census block group that includes the site covers 16.8 square miles, extending some 5 miles west of the site and 6 miles north, and housed 1,784 persons in the 1990 Census. In areas of low population density, one-mile and three-mile radii are too small areas to reliably use the U.S. Census data.

Comment 8: It is stated that all private wells servicing Sun Chemical and the machine shop across Evanston Avenue from the Lomac facility have recently tested free of contamination. However, as a precaution due to the Bofors site, these operations provide bottled water to their employees. (Page 6, ¶ 3) A previous statement in the same paragraph indicates that bottled water has been supplied to onsite workers since the 1970's due to water quality complaints. Is the use of bottled water a precaution recommended by the MDPH, an individual facility decision based on the potential for contamination from the Bofors-Nobel site, or due to poor drinking water quality unrelated to the site? The documented groundwater flow direction is to the south (see Page 4, ¶ 4) and thus onsite contamination would not be expected to impact the wells located west and north of the site.

Response: According to MDPH Division of Water Supply/MDEQ Division of Drinking Water and Radiological Protection⁽¹⁵⁾ personnel, it was an individual facility decision, not a recommendation by the MDPH/MDEQ.

Comment 9: A discussion of a series of complaints about worker health and safety in the operating facility is presented. (Page 7, ¶ 2) These complaints were filed prior to the ownership and operation of the facility by Lomac. The NIOSH study cited was performed in 1985 when the facility was operated by Bofors-Nobel. The studies do not refer to current conditions at the Lomac operating facility. Worker health and safety issues at the operating facility are coincidental to the presence of a Superfund site.

Response: Because of the presence of carcinogenic chemicals on the Bofors site, historical exposures are relevant to the public health hazard pose by the site. The MDPH cited the Lomac plant for violations of worker health and safety requirements in 1989 and 1993, after Lomac, Inc., took over the plant and its operations (19, 20, 21). As of the latest MDPH inspection in 1995, the company is in compliance with health and safety requirements. Lomac and the MDPH/Michigan Department of Labor (MDL)⁽¹⁶⁾ have come to an agreement on a 4-year monitoring program to control DCB exposure (22).

Comment 10: It is indicated that the Toxic Chemical Release Inventory ("TRI") contained entries for nine facilities but all were located approximately 3 or more miles west of the site. (Page 8, ¶ 3) Sun Chemical is located immediately adjacent to the west of the site and should be included in the TRI.

Response: The assessors have conducted a new search of the TRI and have extensively revised that section of the Health Assessment to include data through 1993 and reports from Sun Chemical. The initial search of the TRI missed the reports from Sun Chemical because the TRI reports for the facility included an out-of-date zip code.

Comment 11: The RMEGS, ATSDR health-based comparison values for the site, were computed assuming chronic exposure of a child assuming pica behavior for soil ingestion. (Page 8, ¶ 5) The RMEGS is incorrectly defined; it should be ATSDR Reference Dose Media Evaluation Guides. As previously stated in the General Comments section, the consideration of pica behavior, the abnormal consumption of non-food materials, such as soil, is extremely conservative and is inappropriate to consider as the norm for children under the age of five. This assumption, in conjunction with the assumption that the future use of the site will be residential, greatly overestimates the potential risk posed by the site.

Response: The expansion of the abbreviation has been corrected. ATSDR EMEGs and RMEGs, calculated assuming pica behavior, were used to select contaminants for further consideration. The final decisions on the health hazard of the site were based on more realistic models of behavior.

Comment 12: "DCB was found at 8 ppb in one upgradient sample." (Page 10, ¶ 2) Has the source of the DCB in the upgradient monitoring well been identified or any explanation provided for the presence of DCB in an upgradient well? Is this an anomaly or possibly related to cross-contamination during sampling or laboratory analysis? Has the presence of DCB in this well been confirmed by multiple sampling and analysis?

Response: That well was MW-48, which has consistently been reported to contain low concentrations of DCB (see Table 12). The RI report indicates that little QA/QC information is available for the samplings before their work, and they could not make any determination as to whether the values are influenced by field sample contamination, laboratory artifacts, or instrument imprecision near the detection limit (Reference 1, Appendix B-3). One proposed source for this DCB content is deposition on the soil near the well after air transport from the Bofors plant.

Comment 13: A discussion of the concentration decrease in purged water is presented based on data collected between 1982 and 1986. (Page 11, ¶ 2) Is more recent data available on the purge well concentrations to evaluate trends?

Response: The assessors have obtained more recent purge well data from the MDEQ and the PRP contractors and revised the relevant section of the assessment.

Comment 14: "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." (Page 11, ¶ 3) Based on a description of the sampling location in the referenced Air Quality Division report (Reference 2), this sample was collected from residue on the pavement near the DCB production area. The statement that this concentration was associated with surface soil is in error. The pavement from this area was replaced in December 1979, approximately 6 months prior to the sampling event. The concentration detected appears to be associated with activities earlier in 1980 at the operating facility (Bofors Lakeway, Inc.) and not from previous site activities or the LOU.

Response: The reference identifies the sample as soil, and it makes no difference as to the potential for exposure to the contaminants whether the soil is on top of pavement or not. Historical exposures are relevant because of the several of the contaminants of concern at the site are carcinogens.

Comment 15: A discussion of the air sampling performed on the site and at the perimeter of the site is presented. (Page 12 ¶ 3 & 4) There is no information on the location of the monitors relative to the LOU, the operating facility, or the Sun Chemical facility. Individual days of detectable concentrations of DCB cannot be evaluated and the implications interpreted without additional information on wind direction, wind speed, and other meteorological conditions. The onsite xylenes concentrations may be compared to the appropriate occupational exposure limit. Comparisons may be made of the ambient air perimeter data for DCB, on an annual average (not a daily basis), to the Michigan air ARARs.

Response: The wind direction and meteorological conditions will affect the health evaluation of measured ambient air concentrations only because they determine what population might be exposed. We are concerned whenever available ambient air concentrations exceed available health-based environmental standards. Extrapolating from the available data to an annual average, the average concentration of DCB measured in perimeter air samples collected during 1989 at the three stations for which we have data was above the Michigan Acceptable Ambient Air Concentration.

Comment 16: A discussion of the western extent of the plume boundary is presented. (Page 13, ¶ 1) Is the data sufficient to conclude that MW-51 defines the western boundary of the groundwater plume? Is there another potential source of contamination (toluene, xylenes, acetone, and lead) which could account for the contamination detected in MW-51, especially since the major site COCs were not detected in this well?

Response: According to the RI report, water from wells in clusters MW-4, MW-5, and OW-105, northwest, north by west, and south by west of MW-51, respectively, contained no detectable organic chemicals or lead. MW-4 and MW-5 are upgradient of MW-51. Barring a very unusual configuration of contamination in the groundwater, the evidence presented in the RI report supports their conclusion that MW-51 is the western boundary of the contaminant plume and does not indicate that there is any other source of the contamination in this well cluster.

Comment 17: The well sampling data for the private well of the individual diagnosed with bladder cancer is presented. (Page 13, ¶ 2) No statement is made which indicates whether the contaminants detected in the initial round of well sampling could be associated with bladder cancer or if the detection of the chemicals in the well was in error. A statement should be presented that provides sufficient information as to the importance, and any relationship, of the bladder cancer diagnosis and the well sampling data.

Response: Neither toluene nor bis(2-ethylhexyl)phthalate has been linked to bladder cancer. The latter chemical has been linked to liver cancer in laboratory animals. That the chemicals were not found at any level in a second sample a month later casts doubt on the validity of the first detections. The assessment has been revised appropriately to make these conclusions clear.

Comment 18: "Private industrial wells north and west of the site have not shown any contamination." (Page 13, ¶ 3) As presented in Comment 8, it is indicated that as a precaution due to the Bofors site, bottled water is provided to the facilities west and north of the site. The lack of data demonstrating well contamination and the documented groundwater flow direction to the south do not support the need for bottled water at the facilities west and north of the site.

Response: The operators of the facilities north and west of the site have provided their employees with bottled water on their own initiative.

Comment 19: A discussion of the offsite soil sampling data for DCB is presented. (Page 14, ¶ 2) The reference for the 2.2 ppm DCB is surface soil is in error; the reference should be (2).

Response: That is correct. The assessment has been revised to more accurately reflect the early soil sampling results.

Comment 20: "...a track-out study found DCB in the homes of 9 Bofors employees." (Page 14, ¶ 2) Page 3 ¶ 2 states that the study found DCB in the homes of 10 employees participating in the study.

Response: One employee had measurable DCB in his urine, but none was found in other material from his home. The passages in question have been revised to clarify the situation.

Comment 21: The discussion states that urine from at least 12 workers were analyzed for DCB and 10 had measurable concentrations present. (Page 14, ¶ 3) Page 3 ¶ 2 states that DCB was found in all the employees. These statements, even with the footnote on page 14, seem to be in conflict as to the frequency of detection of DCB in the workers' urine.

Response: The reference on page 3 was to the 10 employees whose houses also contained DCB contamination. The passages in question have been revised for clarity.

Comment 22: "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 if organic enrichment have occurred." (Page 14, ¶ 4) The statement that the slime growth can indicate that some form of organic enrichment has occurred is not very definitive, especially since the growth was observed 2-5 years following the release from the site. No data is presented in the PHA indicating the presence of organic constituent concentrations in the surface water. Generally, slime growth is primarily due to enrichment of surface water with nitrogen and phosphate compounds. Sewage treatment plants and enriched non-point surface runoff are the major contributors to slime growth normally detected in surface water.

Response: Health Assessments cite and evaluate all available data on environmental conditions at and near the site that might have an effect on public health, whatever the source for any detected degradation.

Comment 23: Information is presented regarding the sediment concentrations in 1982 and 1990. (Page 15, ¶ 1) A discussion of the removal of contaminated sediments in 1985, pursuant to the consent agreement, should be included.

Response: The RI report (Reference 1) does not mention any sediment removal from the creek in 1985. The assessors have obtained information on the removal from the MDEQ, and included the information in the revised assessment.

Comment 24: Air sampling data from downwind of the Lomac facility is presented. (Page 15, ¶ 5) See Comment 15.

Response: See our Response to Comment 15.

Comment 25: A discussion is presented of the groundwater contamination detected in wells that served the plant site and the potential for exposure. (Page 17, ¶ 1 & 2) The discussion indicates that contamination had been detected in wells serving the plant and that ingestion, dermal contact, and inhalation of volatile chemicals could have occurred during use of the water. The provision of bottled water is also discussed. However, the discussion focusses on the past use of the water and past concentrations of COCs and provides no indication of the existence of current or future risk at the site. The statements present in these two paragraphs do not support the conclusion that the site poses a public health hazard as opposed to past conditions having posed a hazard.

Response: As mentioned earlier, the criteria for placing a site into the "public health hazard" category include "Evidence that exposures have occurred, are occurring, or are likely to occur in the future." (Reference 64, Table 8-1) As we discussed in the PHA, more recent sampling of the water from the plant has not consistently found contamination, however, evidence of past exposure does qualify for the category selected.

Comment 26: Information provided on Page 4 ¶ 4, as well as in this paragraph, states that Big Black Creek acts as a local groundwater discharge area and that groundwater south of the creek flows north to the creek. This is graphically presented in Figure 3 of the PHA. (Page 17, ¶ 3) No explanation is provided for the presence of the occasional, low level detection of site-related contaminants in the monitoring wells south of the creek. The lack of sufficient QA/QC data for much of the sampling data calls into question the validity of these results and the significance of this data in the evaluation of the risk posed by the site.

Response: No explanation for the MW-52 data was offered in the source material. A somewhat conjectural explanation has been added to the PHA. MDCH does not consider absence of QA/QC information sufficient grounds to invalidate reported data.

Comment 27: "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 found to be contaminated." (Page 18, ¶ 1) As discussed above (Comment 17), the relevance of the sampling performed on the private well south of the site (resident with diagnosed bladder cancer) should be discussed. The statement on Page 18 implies that the sampling of this well does not indicate the presence of contamination while the statement on Page 13 indicates contamination is present.

Response: The follow-up sampling of the residential well in question did not confirm the presence of the contaminants detected in the first sampling.

Comment 28: "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." (Page 18, ¶ 2) This statement supports the contention that the use of the extremely conservative exposure assumptions and health-based values in the PHA are inappropriate and unrealistic given the current and likely future site conditions.

Response: MDCH takes a conservative approach by discussing the potential effects of unlimited exposure first and then discussing the situational and institutional limitations on the more likely exposure.

Comment 29: Health and safety issues and worker track-out of DCB into homes in discussed. (Page 18, ¶ 3) The health and safety practices are based on the 1981 study of the Bofors-Nobel facility. Historic activities are not relevant to whether the site currently poses a health risk.

Response: Worker health and safety issues have been raised in the recent past as well. As mentioned before, the MDPH cited Lomac, Inc. for violations of worker health and safety requirements in 1989 (19, 20) and 1993 (21). The assessment has been revised to include the most recent citations. As of the latest MDPH inspection in 1995, the company is in compliance with health and safety requirements. Lomac and the MDPH/MDL have come to an agreement on a 4-year monitoring program to control DCB exposure (22).

Comment 30: A discussion of the potential exposure to contaminated game and fish is presented. (Page 19, ¶ 1) Information is presented regarding the species of fish caught and analyzed for the COCs and the methodology for the analysis. As stated in the General Comments section, the fish species sampled is not a normal source of food and the whole body was analyzed for DCB content rather than the fish flesh which is normally consumed as a dietary item. There is no evidence presented that consumption of contaminated fish from Big Black Creek has occurred. It is also stated that fish contamination may have been associated with sediment contamination. There is no information presented on the sediment removal that occurred in 1985 prior to the controlled fish kill and restocking of the creek with game fish in 1987. These activities at the creek have resulted in an incomplete exposure pathway for human ingestion of contaminated fish. The implication that this potential exposure pathway still exists at the site is not supported by the data.

Response: We agree that the historical fish sampling and analysis has only limited relevance for human exposure to site-related contaminants by consumption. There is no data available on DCB contamination of edible portions of commonly-consumed species. In view of the available data showing that some fish species were contaminated with DCB in the past, in the absence of data on commonly-consumed species, it is not appropriate to eliminate the possibility of consumption. The RI report (Reference 1) does not refer to any sediment removal from the creek in 1985. The assessors have obtained information on the removal from the MDEQ, and included the information in the revised assessment.

Comment 31: Information is presented on the potential exposure to contaminated surface water and sediments. (Page 19, ¶ 2) There is no data provided in the PHA which indicates that organic chemicals have been detected in the creek. One of the potential exposure pathways presented is the use of the creek for recreational purposes which would result in dermal contact, incidental ingestion, and/or inhalation of volatile contaminants. The contaminated sediments have been removed from the creek and the most recent sediment sampling did not detect the presence of COCs. Area residents would not be expected to use the creek as a source of household water supplies.

Response: See our response to Comment 1.

Comment 32: It is stated that dilution in Big Black Creek, Mona Lake, and Lake Michigan would probably reduce concentrations of any contaminants in surface water to below detectable levels before the contamination would reach the Muskegon municipal water intake. (Page 19, ¶ 2) There is no data in the PHA indicating detectable concentrations of COCs have ever been or are currently present in Big Black Creek. Significant dilution would occur as water from the creek is discharged into Mona Lake and subsequently Lake Michigan. The statement in the PHA implies that there is a potential concern for impacts on the Muskegon municipal water system that is not supported by the data.

Response: See our response to Comment 1. The sentence has been revised to indicate that the dilution would also be to levels not of public health concern.

Comment 33: "As discussed in the Pathways Analyses section, the 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." (Page 20, ¶ 1) It is not evident from the information presented in the Pathways Analyses section (which implies that virtually all potential pathways are complete) how the determination was made that only these two pathways associated with site contamination are complete. A more detailed discussion is required of the analysis so that the basis of selecting a complete pathway is clear to the reader. Statements at the end of each section in the Pathways Analyses section indicating whether the potential pathway discussed is considered complete would assist in putting the site risks in perspective and would clarify the decision making process.

Response: The section has been revised to clarify this matter.

Comment 34: "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." (Page 20, ¶ 2) The stated exposure factors are based upon a residential use scenario. As stated earlier in the PHA (Page 18, ¶ 2), residential use of the site is highly unlikely until extensive remediation has been performed. To assess the potential risk of exposure based on an unrealistic site use and maximum contaminant concentrations grossly overestimates the potential risk posed by the site. Dissemination of information based on overestimates of the site risk does a disservice to the community and raises potential public health concerns where none may exist.

Response: See our response to Comment 28.

Comment 35: A discussion is presented of the potential risks associated with ingestion of groundwater containing DCB. (Page 20, ¶ 3) The value of discussing the relative doses to humans and animals is unclear. It is a generally accepted practice to apply safety factors and uncertainty factors to animal exposure data to derive acceptable human exposure concentrations. The discussion of exposed individuals ingesting sufficient groundwater to result in a dose associated with adverse effects in laboratory animals is irrelevant and confusing. The important issue is the potential of groundwater ingestion to increase the risk of adverse health effects in humans based on accepted methodologies for determining risk. The discussion of the potential risk associated with ingestion of groundwater should include the caveats of the conservative exposure assumptions in order to put the potential estimated risk into perspective.

Response: As mentioned in the cited paragraph of the PHA, there are neither MRLs or RfDs available for DCB. Because neither the ATSDR Division of Toxicology or the relevant branch of the U.S. EPA has made public the appropriate safety and uncertainty factors, it is extremely difficult to apply "accepted methodologies for determining risk" to these chemicals. Rather than make otherwise unfounded assumptions, the assessors compared the estimated exposure doses directly with the experimental data.

Comment 36: "Air concentrations at the Lomac plant perimeter have exceeded this value on some occasions." (Page 21, ¶ 1) This discussion refers to concentrations of DCB detected in ambient air at the Lomac plant perimeter. It is inappropriate to compare 24 hour sampling data to the Michigan ambient air criteria established for carcinogenic compounds. The ambient air criteria are based on an annual average concentration in the ambient air. In order to determine if a facility is in compliance with the criteria, an annual average ambient air concentration must be calculated. Exceedance of the screening level on an individual day is not necessarily indicative of noncompliance or unacceptable exposure concentrations in the community. (Note: Michigan Act 348, Rule 230-232, designates the criteria for carcinogenic compounds as Initial Risk Screening Levels for an increased cancer incidence of one in one million and Secondary Risk Screening Levels for increased cancer incidence of one in one hundred thousand.)

Response: We use available data and health-based environmental standards. Extrapolating from the available data to an annual average, the average concentration of DCB in the air over the two year period in 1989 and 1990 discussed in the text was above the Michigan Acceptable Ambient Air Concentration. (Note: There is not sufficient information available in the literature to evaluate the cancer risk from exposure to DCB.)

Comment 37: Exposure to contaminated surface water and sediments is discussed. (Page 21, ¶ 4) No data is presented in the PHA indicating that the surface water is contaminated. Contaminated sediments were removed from Big Black Creek in 1985 and sediment sampling conducted subsequent to the sediment removal does not indicate the presence of contamination. The data does not support the statement that the potential exists for exposure to contaminated surface water and sediments.

Response: See our response to Comment 1.

Comment 38: Ingestion of contaminated fish is discussed. (Page 21, ¶ 5) As previously discussed in the General Comments and Comment 30, ingestion of contaminated fish is not a complete exposure pathway based on past remedial activities performed on the creek.

Response: As discussed under Comment 30 above, the data available to the assessors when the assessment was drafted suggested that the exposure pathway could be complete but there was no data available on the most probable pathway for human exposure.

Comment 39: A discussion of the potential risks associated with exposure to benzidine is presented. (Page 22, ¶ 1-3) The comments regarding comparison to animal doses and ingestion of fish are similar to those stated above in Comment 35 and Comment 38.

Response: See our responses to Comment 35 and 38.

Comment 40: A discussion of the risks associated with exposure to tetrachloroethylene is presented. (Page 22, ¶ 4) A brief discussion of the dose likely to be obtained by a child ingesting groundwater is presented followed by a statement that no children are using the water at the site. This discussion should be deleted since the exposed population does not exist at the site and will likely not exist in the future. It is confusing and misleads the reader to include irrelevant information in the PHA. It is also inappropriate to compare onsite ambient air concentrations (Lomac operating facility) to the MRL (Minimum Risk Levels) based on a residential exposure scenario. A statement should be included that regular consumption of groundwater containing the maximum detected concentration of tetrachloroethylene is not plausible due to the bottled drinking water supplied to the operating factory workers and the absence of potable water supply wells located downgradient of the impacted area but upgradient of the operating purge well system.

Response: See our response to Comment 28. The tetrachloroethylene was detected in measurements of outside, ambient air, and if there were residences near the site, there is nothing to prevent the contamination detected in the air on the site from reaching the occupants.

Comment 41: A discussion of the potential exposure to other volatile organic chemicals through ingestion of groundwater is presented. (Page 23, ¶ 1-2) The discussion focuses on regular ingestion of the groundwater even though it has been demonstrated that ingestion of the site groundwater is not plausible.

Response: See our response to Comment 28.

Comment 42: A discussion of ambient air monitoring data is presented. (Page 23, ¶ 3) It is inappropriate to compare the operating facility onsite ambient air concentrations to the Michigan ambient air screening levels. The screening levels were established for at or beyond the property line assuming a lifetime of residential exposure. The onsite ambient air concentrations would more appropriately be compared to occupational exposure limits. The statement that the ambient air concentrations of volatile organic chemicals are probably due to plant operations at Lomac is speculative.

Response: See our response to Comment 40. The bases for the attribution of the volatile chemicals to the Lomac facility are as follows: 1) Lomac has filed reports with the U.S. EPA Toxic Chemical Release Inventory (TRI) that the facility has released some of the chemicals in question to the air, 2) the RI air sampling program found those contaminants in and downwind of the Lomac facility but not upwind of the facility, and 3) the wind direction at the time of the sampling was from the south, where there are no other sources for the chemicals.

Comment 43: The toxicity and potential risk of exposure to three butyl benzene isomers is presented. (Page 23, ¶ 4) The inability to verify the presence of these compounds in samples collected prior to and subsequent to the sampling event in which they were detected, and the noted lack of QA/QC data on much analytical data, may warrant the deletion of these compounds as potential COCs.

Response: We have chosen to retain the discussion of the chemicals in the text for the following reasons: We have expanded our discussion of the potential health impacts of the chemicals in the nearby residential well, which also were not confirmed in later samples, as a response to Comment 17, therefore it would be inconsistent to delete these references. These data were provided to the MDPH by Lomac, Inc., who should be able to obtain the relevant QA/QC information.

Comment 44: Potential risks associated with ingestion of groundwater containing arsenic and lead are discussed. (Page 24, ¶ 2 & 3) As previously stated in the General Comments section, when the risk of potential exposure is very low due to the unlikely scenario of ingesting groundwater, a statement explaining the likelihood of exposure should appear before the discussion of the worst case scenario.

Response: See our response to Comment 28.

Comment 45: The incidence of bladder cancer in the Muskegon area is discussed. (Page 25, ¶ 3-5) The data used to compare the bladder cancer rate in Muskegon to age-specific nationwide data is outdated (6 to 8 years old).

Response: The assessors have obtained updated cancer rate information from the MDPH Office of the State Registrar and Center for Health Statistics and the section has been revised.

Comment 46: The greatest exposure to the community has been to the operating facility workers and their families due to insufficient industrial safety precautions and measures to prevent track-out. (Page 27, ¶ 1) This determination is based upon historic site conditions and fails to address current conditions at the operating facility. There is no information presented in the PHA that indicates that conditions that existed at the Bofors-Nobel facility at the time of the track-out study still exist at the Lomac facility. Moreover, the industrial hygiene and safety issues are coincidental to the presence of the Superfund site.

Response: Historical potential exposures are relevant because several of the contaminants of concern are carcinogens, whose impact on human health typically have long latency periods. The MDPH cited Lomac, Inc. for violations of worker health and safety requirements in 1989 (19, 20) and 1993 (21). As of the latest MDPH inspection in 1995, the company is in compliance with health and safety requirements. Lomac and the MDPH/MDL have come to an agreement on a 4-year monitoring program to control DCB exposure (22). ATSDR Health Assessments consider all potential public health hazards in the vicinity of Superfund sites.

Comment 47: "Future exposure to these chemicals may occur via contaminated groundwater, should the site be converted to residential use before the remediation of the groundwater contamination now in process is completed." (Page 27, ¶ 2) The ROD indicates that the groundwater treatment system will operate in perpetuity. The development of the site for residential use is implausible at this time because of the depth of contamination that exists at the site and the inability to remediate these soils to levels that would be deemed acceptable for residential use by the State of Michigan. This conclusion should be qualified by including this information.

Response: The remediation currently in place does interdict this exposure route. Health Assessments, however, discuss the possible exposures at the site, in addition to the probable ones. Institutional controls, such as those cited and in place at the site, are subject to change, removal, or evasion.

Comment 48: "Workers at the Lomac plant and workers involved in remedial activities should follow appropriate OSHA, NIOSH, and Michigan Occupational Safety and Health Act (MIOSHA) regulations." (Page 28, ¶ 1) As discussed in the General Comments section, additional evaluation should be performed to address the potential risk of exposure to the remedial workers, the Lomac operating facility workers, and the community during the implementation of the remedy specified in the ROD. The construction of two onsite landfill cells and the excavation and placement of one million cubic yards of contaminated soils into the landfill cells presents a significant potential risk of inhalation exposure to the COCs. The potential risk of implementing the ROD remedy is likely to be greater than the current and future potential risk of leaving the site in its current condition (restricted access and operation of the purge well and groundwater treatment systems).

Response: See our response to the sixth General Comment, page 15, above.

Comment 49: "Further off-site sampling should be done to establish the extent of the area affected." (Page 28, ¶ 3) The 1980 Air Quality Division sampling data indicate that the concentrations of DCB were above the detection limit further west of the investigated facilities on the north side of the road than on the south side of the road. This information is 15 years old and is not representative of current facility operations or site conditions.

Response: The first statement is accurate, but is not relevant to the potential for exposure to the DCB that was found in the soil. We agree that the data cited is old, and have recommended that data reflecting current site conditions and facility operations be collected.

Comment 50: "The occasional lead, chromium, and arsenic soil contamination discovered during the RI should be further investigated to determine the source and to develop appropriate remedial actions." (Page 28, ¶ 4) The maximum concentrations of these metals in plant area and lagoon area soils (as presented in Tables 6 and 8 in the PHA) are within the range of background concentrations as specified in the Michigan Environmental Response Act Operational Memorandum #15: Default Type A (background) Cleanup Criteria. The soils are thus within the range of background and do not require further investigation or remediation. The lagoon sludges do exceed the background concentrations for these metals. The remediation of the lagoon sludges through the Total In-situ Containment proposed by the PRP Committee will effectively address the metals present in the lagoon sludges.

Response: The 4,100 ppm of lead and 125 ppm of chromium found in surface soil samples collected during the RI, listed in Table 9 of the PHA, are substantially higher than background levels.

Comment 51: The health actions planned by ATSDR and MDPH are presented. (Page 29, ¶ 1-3) Evaluation of the potential impact of site remediation on workers and the community should be addressed to assure that the implementation of the remedy specified in the ROD will not increase the public health risk to the community as compared to the risks posed by the existing site conditions. Consideration of the risks associated with the implementation of the proposed alternative remedy (Total In-situ Containment), which does not require the excavation and handling of one million cubic yards of contaminated soils, should also be included.

Response: See our response to the sixth General Comment, page 15, above.

Comment 52: A concern was raised regarding the need to assess the continuity of the clay layer beneath the site to assure that adequate protection of the lower aquifer exists. (Page RS-10, #31) The continuity of the clay layer is not necessary for the successful implementation of the Total In-situ Containment remedy because it does not require the slurry wall to be extended into the clay to assure containment of the contamination.

Response: See our response to the sixth General Comment, page 15, above.

Comments submitted by Lomac, Inc.:

General comments (excerpts):

As a member, Lomac also joins in the Comments submitted by the Committee. In that comment, the Committee makes several points regarding the Assessment. Lomac reiterates those points, in particular that the Assessment is based upon extremely conservative and unrealistic exposure assumptions. Moreover, the conclusion that the site poses a risk is belied by the text of the report itself, which states that present institutional controls are sufficient to control potential exposure pathways.

Response: See our responses above to the Committee's comments. Institutional controls, such as those in place at the site, are subject to change, removal, or evasion.

Specific comments:

Page 7, second full paragraph: The complaints made about workers health and safety in the plant in the citations appear to pre-date the 1985 NIOSH study, and therefore would be attributable to BNI operations, not Lomac operations.

Response: The MDPH cited Lomac, Inc. for violations of worker health and safety requirements in 1989 (19, 20) and 1993 (21). As of the latest MDPH inspection in 1995, the company is in compliance with health and safety requirements. Lomac and the MDPH/MDL have come to an agreement on a 4-year monitoring program to control DCB exposure (22).

Page 12, third full paragraph: CREGs are not the appropriate standard by which to judge the air sampling at the Lomac plant area. The comparison leads to the inaccurate conclusion that Lomac is exposing its workers to levels exceeding applicable law.

Response: ATSDR uses CREGs as comparison values to select chemicals for further evaluation. The ultimate decision on the health hazards presented are based on exposure scenarios that are likely to occur on the site. ATSDR comparison values are generally not legal constraints.

Page 17, second paragraph: The Assessment vastly overstates the potential for past exposure to water from one well at the Lomac plant. The sampling event referenced was a one-time event which detected a minute quantity of tetrachloroethylene. The Assessment later concedes that this was not a significant risk.

Response: MDPH takes a conservative approach by discussing the potential effects of unlimited exposure first and then bringing in the situational limitations on the more likely exposure.

Page 18, fourth paragraph: The sampling in workers' homes that detected DCB in house dust, and urine samples from workers and their families, occurred during 1980-81, not during Lomac's operations. The Assessment fails to make this important distinction and is misleading.

Response: That is correct, and the relevant passages have been corrected.

Page 21, third full paragraph: The manufacture of benzidine was halted early in BNI's operational history and has never been produced by Lomac. This paragraph, as well as the third full paragraph on page 22, suggests otherwise.

Response: The passages have been revised to more accurately reflect the situation.

Page 28, first paragraph: As a recommendation, the Assessment states that workers at the Lomac plant [and] workers involved in remedial activities should follow the appropriate OSHA and related regulations. This suggests that Lomac does not follow these regulations and is inaccurate.

Response: A statement that people should follow appropriate regulations is not necessarily meant to say that those addressed do not do so now. The recommendation is directed to workers as part of the public for whom Public Health Assessments are produced.

Comments received from Sun Chemical Company:

Page 5, Section C, paragraph 1: There is a reference to Sun Chemical as a producer of "inorganic and organic pigments and dyes." Sun Chemical does not produce inorganic pigments or dyes. The Sun Chemical Muskegon plant produces only synthetic organic pigments.

Response: The description has been revised to reflect this information.

Page 6, Section C, paragraph 3: The Sun Chemical property is mentioned as having one private well. In fact, Sun Chemical has 4 process wells and several monitoring well heads, which monitor not only our own activities but also check for any possible contaminants entering our property from surrounding properties, as our process depends on consistently pure water.

Response: The description has been revised to reflect this information.

Page RS-2, Comment 5: We also note ... that a commentor refers to the production of yellow-dye from DCB. Sun Chemical does not produce dyes. Dyes are a different chemistry and structure from pigments; dyes are soluble and reactive and color basically by selective absorption, while pigments are insoluble and color essentially by incorporation into the material to be colored while remaining physically separate and unchanged in chemical composition.

Response: MDPH prefers to reprint comments with minimal editing, and then only for clarity. "Dye" and "pigment" are often used interchangeably by lay people, who may be unaware of the technical distinction between the terms. We have added a note to the comment.