PUBLIC HEALTH ASSESSMENT
FORT DEVENS-SUDBURY TRAINING ANNEX
SUDBURY, MIDDLESEX COUNTY, MASSACHUSETTS
APPENDIX 6: STA GROUNDWATER WATERSHED CHARACTERIZATION
The water table at STA is generally shallow as indicated by swamps, bogs, and waterholes. The surficial aquifer is composed of unconsolidated outwash (glacial drift deposited by meltwater streams beyond an active glacier or "loose rocks") underlying the lowlands of the site. Because of the high permeability of its deposits, the surficial aquifer constitutes the principal aquifer. The thickness of the outwash averages about 50 feet with a maximum of 100 feet (1). Depth to the water table is less than 15 feet, with the groundwater gradient following the area topography (2). The bedrock is overlain by till, a compact, unsorted mixture of clay, silt, gravel, and boulders. The thickness of the till ranges from a few inches to at least 40 feet in thickness. The thickness of the bedrock cannot be accurately defined, but one formation (Marlboro) is more than 2,500 feet in thickness.
Regional wells, in the most favorable areas, have yields of from 100 to 150 gallons per minute (gpm). Groundwater divides (the crest of the water table with flow going in opposite directions on either side) correspond to surface water divides in the unconfined surficial aquifer. A divide transects Area A10, with flows reported to be in a westerly direction toward White Pond. Flows on the east side of this divide are eastward toward Stearns Mill Pond and Hop and Marlboro Brooks. Groundwater flow in the northern portion of STA is toward or parallel to the flow of Taylor and Honey Brook, except in the far southwest where flow is toward Boon and White Ponds (3). Groundwater recharge areas (areas of water movement into the aquifer) are shown in Figure 4 (see Appendix 1).
Groundwater and surface water flow at STA is further characterized into 7 distinct watersheds (see Figure 3 in Appendix 1) (3). Study areas within each watershed are listed in Table 6 (see Appendix 2). Potential transport of environmental contaminants in groundwater and surface water from these study areas is influenced by watershed characteristics. Evaluating such factors is essential for determining potential exposure pathways at STA. A general description of the watersheds follows.
WATERSHED lA: TAYLOR BROOK ABOVE HONEY BROOK
The Watershed lA area encompasses the watershed of Taylor Brook above
its junction with Honey Brook. It receives flow from off-site areas, including
Taylor Brook from its source to Cutting Pond, from Vose Pond, and the northern
parts of the Annex adjacent to the Town of Maynard which were formerly part
of the Annex. Puffer Pond itself is included in this group both because it receives
discharge of groundwater from under the sites west of it, and because it is
to be the site of a bioaccumulation study during which sediment, surface water,
and fish tissue will be analyzed.
WATERSHED lB: TAYLOR BROOK BELOW HONEY BROOK
This part of the Taylor Brook watershed receives runoff and groundwater
flow from two major groups of sites. These are Al, A2, P42, and P45 around Taylor
Brook, and P11, P13, P23, and P26 which discharge to Honey Brook. This area
covers much of the north end of the Annex with the exception of the small areas
draining directly to the Assabet River (Watersheds 3 and 4). All the sites within
this area lie on areas of sandy glacial outwash within short distances of wetland
or surface streams, and the well-drained soils and shallow water tables would
imply rapid impacts on groundwater, adjoining wetlands, surface water, and sediments,
if contamination is present.
WATERSHED 2: HOP BROOK DRAINAGE
A small part of the Annex adjacent to Sudbury Road and the Massachusetts
Firefighting Academy drains south into White Pond, which has no outlet. White
Pond is used as a municipal water supply source by the Town of Maynard. The
pond discharges underground south to Hop Brook. The entire area of the detached
portion of STA where all Watershed 2 sites are located also drains into Hop
Brook, although A11, A12, P36, and P37, and possibly parts of P28 may discharge
first to Marlboro Brook, a small tributary of Hop Brook. Hop Brook discharges
to Sudbury River, unlike most of the Annex, which drains to Assabet River.
WATERSHED 3: UNNAMED TRIBUTARY 1 TO ASSABET RIVER (NORTH END)
A small unnamed tributary of the Assabet River starts at a wetland
just north of White Pond Road and west of the former Massachusetts Fire Fighting
Academy (study area P13) and discharges directly to the Assabet River. The entire
tributary is only 2,000 feet in length, but it runs between or adjacent to a
number of sites to be investigated including study areas A8, P9, P10 and the
newly listed P57 at the site of the demolished Building S449.
WATERSHED 4: UNNAMED TRIBUTARY 2 TO ASSABET RIVER (NORTHWEST END)
This drainage area is difficult to delineate, as the small unnamed
tributary originated in a large area of marsh with very little topographic relief,
south of the hill on which the U.S. Air Force Weather Radar Laboratory is sited.
Relative volumes of flow suggest that this tributary drains much less of this
marsh than does Honey Brook to the east. The two sites within this watershed
are either definitely draining towards this brook (Study Area A6), or, most
probably discharging to groundwater that enters this brook (study area P22).
To further characterize groundwater flow patterns in this and other STA watersheds,
a facility-wide groundwater model is planned flow model using MODFLOW, a 3-dimensional
USGS model. It will help establish groundwater flow patterns, volumes of groundwater
flow off site through various media, and the relative proportions of flow. It
is also capable of providing confirmation of probable mass transport of contaminants
from specific sites to supplement individual site studies.
WATERSHED 5: BOONS POND DRAINAGE
Part of the main portion of the Annex on its southwest side is very
flat and has extensive wetlands. Some of these appear to discharge via groundwater
to Lake Boon, southwest of the site. Lake Boon, in turn, discharges over a dam
into the Assabet River.
WATERSHED 6: RUN BROOK (WILLIS POND) DRAINAGE
On the east side of the main part of STA, some sites are located in
areas that drain into a large area of swamp within which are Crystal Lake (formerly
Bottomless Pond) and Willis Pond. This area drains toward Run Brook, which in
turn drains into Hop Brook and the Sudbury River. Three sites, P1, P2, and P3,
lie on or adjacent to a hill of ground moraine (till), which slopes and drains
toward the swamp.
Reference:
Comparison values for ATSDR public health assessments are contaminant concentrations in specific media used to select contaminants for further evaluation. The values provide guidelines for estimating the dose at which health effects might be observed. Comparison values and the units used to quantitate contaminant concentrations that appear in the Environmental Contamination and Other Hazards and the Public Health Implications sections of this public health assessment are listed and described in the following paragraphs.
|
Comparison Values
|
|
| * CREG | = Cancer Risk Evaluation Guide |
| * EMEG | = Environmental Media Evaluation Guide |
| * MCL | = Maximum Contaminant Level (µg/L) |
| * MRL | = Minimal Risk Level (mg/kg/day) |
| * RfD | = Reference Dose (mg/kg/day) |
| * RMEG | = Reference Dose Media Evaluation Guide |
|
Units
|
|
| * ppm | = milligrams per liter (mg/L water) milligrams per kilogram (mg/kg soil) |
| * ppb | = micrograms per liter (µg/L water) micrograms per kilogram (µg/kg soil) |
| * kg | = kilogram |
| * mg | = milligram |
| * µg | = microgram |
| * L | = liter |
Cancer Risk Evaluation Guides (CREGs) are estimated contaminant concentrations that would be expected to cause no more than one excess cancer in a million (10E-6) persons exposed over a lifetime. CREGs are calculated from EPA's cancer slope factors.
Environmental Media Evaluation Guides (EMEGs) are based on ATSDR minimal risk levels (MRLs) and factor in body weight and ingestion rates. An EMEG is a contaminant concentration in water, soil, or air at which daily human exposure is unlikely to result in adverse noncancerous effects.
Maximum Contaminant Levels (MCLs) are contaminant concentrations that EPA deems protective of public health (considering the availability and economics of water treatment technology) over a lifetime (70 years) at an exposure rate of 2 liters of water per day (for an adult).
A Minimal Risk Level (MRL) is an estimate of daily human exposure to a chemical (in mg/kg/day) that is likely to be without an appreciable risk of deleterious effects (noncancer) over a specified duration of exposure. MRLs are calculated using information from human and animal studies and are reported for acute (< 14 days), intermediate (15-364 days), and chronic (> 365 days) exposures. MRLs are published in ATSDR Toxicological Profiles for specific chemicals.
EPA's Reference Dose (RfD) is an estimate of the daily exposure to a contaminant unlikely to cause adverse noncancerous health effects.
Reference Dose Media Evaluation Guides are derived by ATSDR from EPA oral reference doses and factor in body weight and ingestion rates. An RMEG is a contaminant concentration in water or soil at which daily human exposure is unlikely to cause adverse noncancerous health effects.
The Food and Drug Administration (FDA) action level is the maximum allowable level of poisonous and deleterious substances in human food and animal feed.
APPENDIX 8: ATSDR HEALTH CONSULTATION
Health Consultation
For the
Fort Devens Sudbury Training Annex
Sudbury, Massachusetts
Prepared by
Federal Programs Branch
Division of Health Assessment and Consulation
November 23, 1992
Background and Statement of Issues
A site tour of the Sudbury Training Annex (STA) NPL site was conducted by staff members from ATSDR's Fedeal Programs Branch (Lorna Bozeman, Gary Campbell, and Richard Collins) and Regional Operations (Susanne Simon) on October 7, 1992 (1). During that site tour, ATSDR staff members saw physical hazards that warrant prompt public health action.
Discussion
Abandoned building T104 and T106 are unsecured (no doors), appear structurally unsound -- metal building with holes in the roof and walls -- and contain abundant debris (barbed wired, broken glass, scrap metal, and pieces of asbestos insulation that have falled off the walls and ceiling). Building T106 is a shed-type structure; one end is completely open. Part of the broken-down, chain-link fence remains along one side of T106. Building T104 is accessible through a garage door-size opening. Because the buildings are not maintained, they will continue to deteriorate. In addition to the physical hazards associated with the buildings, the nature and extent of environmental contamination at these sites are currently unknown. However, the Army plans to evaluate these areas during the remedial investigation of STA (3).
Health assessors have seen evidence of unauthorized use of the buildings and area. The walls of both buildings have graffiti, the windows are shattered, and bullet holes are in the sheet metal. Motorcycle tire tracks and remnants of fires set on the cement floor of building T106 were also seen. The buildings are 0.3 mile from the nearest residence. A dirt road provides direct access to the area from Moore Road in Sudbury (Figure 1). The Army has placed a cable and cement barricade across the road, but trespassers can easily enter the area on foot, or on motorcycles or off-road vehicles. During the site tour, we saw a person unload a motorcycle from a van at the access point. Approximately two weeks later, the Installation Restoration Program manager from STA visited the site and said that two children were skateboarding in T106 when he arrived (4).
Restricting access to several areas on post that are currently being investigated for possible contamination or that have physical hazards has been a chronic problem at STA. Attempts to limit access by placing cement barriers or cables across road have been ineffective (5). Although some contaminated areas appear secure (enclosed by 8-foot, chain-link fences), perimeter fencing is generally not well maintainted at STA. The Army is currently installing additional fencing along the northern part of STA, bordering the towns of Hudson and Stow; that action should further restrict access.
Conclusions
Recommendations
|
Lorna Bozeman, MS
Richard Collins, MSEH, RS |
References
Figure 1. Fort Devens Sudbury Training Annex - South Post
There are three general forms of mercury: elemental, inorganic, and organic. The chemical form influences the absorption and disposition of mercury in the body. Elemental mercury is not well absorbed from the gastrointestinal tract, so it is not a major source of mercury poisoning outside of occupational settings where workers may be exposed to mercury vapor in the air. Only about 7 percent of inorganic mercury in food is absorbed while organic mercury is nearly completely absorbed. The toxic manifestations of inorganic mercury are renal whereas those for organic mercury are neurological (1).
Food is the main source of exposure to mercury in non-occupationally exposed populations. Fish and other seafood account for most of this exposure. Mercury in food is predominately methylmercury (organic) and is toxic to the nervous system. Exposure to forms of mercury other than methylmercury seldom reaches toxic concentrations in air, water, and food.
Two episodes of environmental mercury contamination showed the behavioral effects of prenatal exposure to methylmercury. In a village on the Minimata Bay in Japan, villagers were exposed to methylmercury through the ingestion of contaminated fish during the 1950s. Cerebral palsy and mental retardation were widespread. Congenital malformations were rarely seen. The neurological problems seen in the villagers are called "Minimata Disease". Another outbreak occurred in Iraq in 1971 when seed grain was accidentally treated with a methylmercury fungicide. The exposure here was to a much higher concentration and over a shorter period of time than that at Minimata Bay. The children who were exposed prenatally had impaired mental development and problems with speech and motor function (2).
Exposure to both elemental mercury or methylmercury can affect the central nervous system. Elemental mercury exposure can result in depression and emotional instability. Methylmercury exposure can result in changes in vision, decrease in coordination, and impaired taste, smell, and hearing. Inorganic mercury primarily affects the kidneys causing a condition similar to acute tubular necrosis. The inorganic form can also irritate the gastrointestinal tract causing ulceration (2).
References:
1. Doull J, Klaassen CD and Amdur MO. Toxicology: The Basic Science of Poisons. New York: Macmillan Publishing Co., Inc. 1980.
2. Tarcher AB, ed. Principles and Practice of Environmental Medicine. New York: Plenum Medical Book Company. 1992.
APPENDIX 10: ATSDR LETTER RECOMMENDING PRIVATE WELL SAMPLING
April 22, 1994
Department of the Army
Headquarters Fort Devens
ATTN: AFZD-EM, Box 19
Mr. Thomas Strunk
Fort Devens, Massachusetts 01433-5190
Dear Mr. Strunk:
The Agency for Toxic Substances and Disease Registry (ATDSR) recommends that the Army sample private drinking water wells on Dawes Road in Stow.
New groundwater data for Fort Devens Sudbury Training Annex study areas P31/P58 suggest that site contaminants could migrate to off-site drinking water wells. These data are reported in the March 1994 Ecology and Environment Phase II Site Investigations Report.
Three new monitoring wells were installed between study areas P31/P58 and private drinking water wells on Dawes Road. Groundwater samples collected from these wells in September and December 1993 contained metals at concentrations above drinking water standards. Unfiltered samples contained maximum levels of 91 ppb arsenic (the maximum contaminant level is 50 ppb) and 43 ppb lead (the action level is 15 ppb). Arsenic levels in filtered groundwater samples also exceed the standard. Trace levels of pesticides and the explosive 1,3-dinitrobenzene were also detected. No volatile organic compounds were found.
The six homes on Dawes Road are about 500 feet from study areas P31/P58 and are in the direction of groundwater flow. Water from the residential wells probably is not filtered before use. To determine if people are drinking water containing these chemicals, we recommend that you analyze duplicate unfiltered water samples from the private wells. Based on findings from P31/P58, we suggest that analyses include TCL organics and pesticides, TAL metals, herbicides and explosives.
When the sampling is completed, please submit the entire data package to ATSDR for review. If contaminated levels in any water samples exceed primary drinking water standards, provide alternate water to all residents (6 homes) as a precautionary measure until data can be fully evaluated. ATSDR will review data and make recommendations regarding public health actions. If the private wells are not contaminated, quarterly sampling of P31/P58 monitoring wells should continue because increases in contaminant levels would pose a greater threat to off-site drinking water wells.
If you need additional information, please contact Ms. Susanne Simon, ATSDR Regional Representative, Region I, (617) 860-4619, or Ms. Lorna Bozeman, Environmental Health Scientist, (404) 639-6070. We appreciate your ongoing efforts to address public health issues at the Annex.
| Sincerely yours, Sally E. Shaver |
cc:
Diana Feireisel, USAEC
C. Todd Jones, USAEHA
Robert Lim, EPA
Debra Darby, MDEP
Jeff Purvis, MDPH
Stow Board of Health
Four Town FOCUS
APPENDIX 11: COMMENTS ON STA PUBLIC HEALTH ASSESSMENT
Listed below are comments received from the public regarding the December 1993 Public Health Assessment for the Sudbury Training Annex. We have not included comments on accuracy of stated facts. If the accuracy of a statement was questioned, the statement was verified or corrected.
GENERAL STATEMENT
Many of the questions raised by the public concerned contamination levels. We emphasize that human exposure, not contamination levels, is the focus of a health assessment. Our conclusion is that people were not being exposed to contaminants at levels that could pose health concerns. Most people were not being exposed at all, either through groundwater, soil, or the food chain. Even those people who trespassed onto Annex property would not have come into contact with contaminated soil often enough to cause health consequences.
ENVIRONMENTAL
Comment 1: We recognize that ATSDR's role in the CERCLA process is solely to identify sites that may present an actual past or present threat to public health. We also recognize the significant differences between the Public Health Assessment and the Risk Assessment, which is used to support the selection of a remedy. Both the Public Health Assessment and the Risk Assessment require identification of sources, pathways, and receptors. The existing environmental data permit identification of contaminant sources, and the demographics of the area allow identification of potential receptors, both of which are discussed in the Public Health Assessment. In order to show public health impact, however, the sources and the receptors must be linked by pathways. With a few exceptions, this has not been done by ATSDR. In order to establish a completed pathway, it is necessary to show, through analytical data or modeling, that a contaminant is present in the environmental medium (soil, air, fish, sediment, etc.) that the receptor contacts.
The public health assessment does present pathways analysis. Little potential exists for people to contact contaminated environmental media on the Annex; therefore, there is limited public health impact. Public health actions to reduce the few ways that people could continue to be exposed are to restrict trespassing, and post Puffer Pond as catch and release fishing. Those actions have been implemented. The other public health concern is metals in groundwater at study areas P31/P58. We have recommended that the Army sample off-post private wells closest to those study areas.
Comment 2: The environmental data from the Remedial Investigation at the Annex identify numerous sources of toxic materials, but the environmental data are not always adequate to characterize pathways that may be important at the site. The investigations at the Annex have been focused on discrete areas of suspected contamination, with minimal effort to characterize the entire site or to look at relationships between the contaminated sources. Although we realize that this was not the intent of the Remedial Investigations conducted at this site, we feel that ATSDR should have acknowledged the data gaps and the inadequacy of the existing data to characterize pathways.
The sampling strategy used to characterize environmental contamination at the Annex is appropriate. Sampling is not warranted in areas where historical activities would not lead to environmental contamination. In areas where more data are needed to evaluate potential pathways, broader sampling is planned or has been carried out.
Comment 3 (summary): What schedule will be used to review new data on the Annex? How will ATSDR obtain notice of this data? We would like a clarification of this statement.
New data, primarily environmental, relating to public health questions at the Annex are reviewed when they become available. A Memorandum of Understanding requires that the Army provide ATSDR with data as it is available. Through ATSDR regional representatives, EPA also provides copies of split sample data for review. Please refer to the public health action plan in this document for further activities.
Contaminants
Comment 4 (Land Use): We were told in the past that no chemicals were stored any longer on the Annex. Which bunkers contain these materials and are they being stored compliant with federal law?
The comment refers to the sentence, "The Army currently leases some of the bunkers to state and federal agencies for storage of equipment, supplies, chemicals, and herbicides (14).", which is somewhat misleading. Annex bunkers are used for storage of equipment and supplies and the sentence has been reworded to reflect that use. Only a few bunkers contain any chemicals, and those chemicals are limited to an isolated jug of lubricating oil, or a drum of thinner, etc. Bunker 303 was used to store Ureabor, but that herbicide is no longer stored on the Annex. The Army is further investigating areas in and around bunkers where there is a potential for chemical contamination. For specific information about materials stored in Annex bunkers, the reader is referred to pages 8-8 to 8-22 of OHM Remediation Services Corporation, Final Site/Remedial Investigation Report, December 31, 1993.
Federal law pertaining to storage of chemicals is part of CERCLA as amended by SARA, Title III, Emergency Planning Community Right to Know. There is no requirement for storage conditions, but the community Local Emergency Planning Committee (LEPC) must be informed if the quantity of a specific chemical being stored exceeds the threshold planning quantity listed in 40 CFR Part 355.
Comment 5 (summary): The specific study areas with the designation of "indeterminate health assessment" should be defined.
Study areas P2, A6/P22, A9/P12, and P31/P58 were classified as indeterminate public health hazards because limited groundwater data were available. Providing specific study area numbers in the summary will provide little clarification to the reader. Details concerning groundwater issues relating to those study areas are presented in sections entitled Environmental Contamination and Other Hazards and Pathways Analyses (pp 20-23, 35-36).
Comment 6: The second paragraph of the summary should specify what materials were used in R&D.
The summary is an overview of the entire public health assessment and not meant to be a detailed section. Specific information about research and development activities are presented in Appendix 2 (Tables 1 and 2) and in Appendix 4: Site History, Description, and Observations.
Comment 7 (background): A description of the types of materials and wastes should be included.
Page 3 Paragraph 2 refers the reader to Table 1 and specific references where more detailed information is provided. Table 2 also lists additional information related to types of materials and wastes.
Comment 8 (summary): Herbicides should be added to the list of contaminants found at the Annex.
We have added herbicides and explosives to the summary. Specific information about these substances are discussed in the Environmental Contaminants and Other Hazards section of the public health assessment.
Comment 9: ATSDR did not address the issue of elevated levels of DDT in the Annex.
We include DDT among the chemicals detected on the Annex at levels of potential health concern. To prevent exposure to contaminants on the Annex, we currently recommend that people refrain from trespassing. In coordination with EPA and Massachusetts Department of Environmental Protection, the Army is further evaluating the nature and extent of contamination at the Annex. Areas contaminated with chemicals at levels of health concern will be cleaned up before the property is reused.
Comment 10: In the section "Environmental Contamination and Other Hazards", and Tables 4,5,8,9, Comparison Values are provided only for those chemicals that exceeded them. Even among these it is not totally clear what value, out of those possible for a chemical, was chosen. For example it is not clear why an RMEG of 30 ppm was chosen for DDT and a CREG of 2 ppm was chosen for DDE, since DDT is also a carcinogen with a slope factor. The Comparison Value for Arsenic is not listed, presumably because no soil concentration exceeded it. However, A9 is known to have an area of excess arsenic that will be further explored. This is not in the table for surface soil, and arsenic usually shows a risk even at background levels on many sites.
Recommendation: Confirm that appropriate comparison values are being used in the public health assessment. These are understood to be screening levels, which are usually conservative, so that a chemical with any potential for health effect on a site can be further explored. The fact that arsenic does not show up at all, and DDT has a value of 30 ppm places some doubt on the conservatism in the assessment. Alternatively, provide an explanation as to how values are chosen.
All of the comparison values used for the public health assessment are overly conservative considering potential exposure scenarios for the Annex. Based on former and current Annex land use, possible exposures are most likely short term (trespassers). For that reason, comparison values for acute exposure were used when available. Often, however, there were no comparison values for acute exposure (e.g., DDE and arsenic). For such chemicals, we selected what we considered the most appropriate comparison value, for the exposure scenario. In some cases, comparison values for chronic exposure (i.e., daily exposure over a lifetime), such as the CREGs, were used for screening. Clearly, such comparison values are overly protective for short term exposures that might be experienced by trespassers.
With regard to future land use, human exposures are not a concern because contaminated areas of the Annex will be cleaned up to levels protective of public health.
Groundwater
Comment 11 (Groundwater Contamination): ATSDR states that the Army is installing wells between P43A & B, this is incorrect.
The cited information is from the June 1993 Ecology and Environment, Inc., Technical Plan Addenda, Phase II Site Inspections, Remedial Investigations. That document states that monitoring wells will be placed between the alleged site(s) (Study Areas P43A and P43B) and the municipal well. However, those plans were modified in later versions of the Technical Plan (October 1993)/January 1994). Rather than install new monitoring wells, existing wells were resampled. The revised public health assessment reflects changes in groundwater monitoring for study areas P43A and P43B.
Comment 12: What is meant by 'explosives' in monitoring wells? Are these nitrate breakdown products? The term 'explosives'- implies a hazard that does not exist in groundwater.
Groundwater from monitoring well DM9A contained the explosives 1,3,5-Trinitrobenzene and 2,4,6-Trinitrotoluene. The reader is referred to page 19 and Table 6 of the public health assessment.
Comment 13: The groundwater pathway through on-site water supply notes that selenium, iron, and manganese were the only contaminants detected in the on-site potable wells. Have these 'contaminants' been linked to a site source? If not, they should not be evaluated as site-related contaminants. Iron and manganese are common in shallow wells. They are not highly toxic but they might be indicators of leachate from buried metal. Has this been evaluated?
Selenium, iron, and manganese are discussed because they were detected in groundwater from on-site wells that were previously used as potable water sources. The concentrations of the metals exceeded drinking water standards. Based on the findings of environmental investigations at the Annex, it is likely that these substances are naturally occurring, rather than indicators of leachate from buried metal as you suggest. For clarity, the use of the word contaminant in reference to those metals has been changed to metals.
Comment 14: The Onsite Water Supply section discusses the exposure doses of manganese that are expected to be safe, or to have a health effect, by oral route. However, EPA provides different health criteria for food (or soil) manganese intake and that from water. The 2.5-5 mg/day amount from the NRC is based on dietary manganese. The IRIS discusses manganese in water and gives the following levels as the NOAEL and LOAEL: 81.6-252.6 and 1600-2300 µg/L.
Recommendation: Differentiate between dietary and water oral ingestion of manganese when discussing health effects levels.
Because the calculated oral exposure dose of manganese is so low, little information is gained by differentiating between dietary and water oral ingestion of manganese. The oral exposure dose from water of 0.88 mg/day is conservative. It was calculated using the highest concentration of manganese measured in on-post potable wells. Moreover, it assumes that 100% of the ingested manganese is absorbed. The 0.88 mg/kg oral dose is three times less that the safe and adequate intake value of 2.5 mg/day. The dose is also more than 1000 times less than the neurological effects level of 980 mg/kg (59).
Comment 15: The local populace appears to be very concerned with the contamination of groundwater in the RI sites A7 and A9, among others. The public health assessment does not indicate that any groundwater contamination exceeded comparison values in A7 and that only the two explosives exceeded them in A9. The Health Risk Assessment shows on-site groundwater risks exceeding EPA criteria for these two sites. For example, DCE on A9 shows a "maximum risk of 7E-5.
Recommendation: Reevaluate whether groundwater contamination from monitoring wells on these RI sites is below comparison values for all but the two explosives. Considering the local concern over the RI sites, and their groundwater contamination, perhaps some more discussion of what is above comparison values, but also their limited extent of migration, would be helpful.
We do not believe that groundwater contaminants at RI sites A7 and A9 have affected drinking water supplies. Although the health risk assessment shows on-site groundwater risks from these two sites, no one is drinking the on-site groundwater. We evaluated whether groundwater contaminants from these RI sites could migrate toward drinking water wells. The potential for site contaminants to affect drinking water supplies in the area is low. The nearest potable wells are about a quarter mile from A9. Additional groundwater monitoring was recommended and has been done but the results are not yet available. We will review the new groundwater data and make further public health recommendations if needed.
Comment 16: The taste and odor problems reported in Maynard Well No. 3 are not associated with the Annex. Modeling of groundwater flow in the area of the well should be done in conjunction with the continuing studies to determine whether a potential pathway exists between the contaminated Annex groundwater and the zone of influence of the well. ATSDR should note that there is no evidence of a pathway, and should make recommendations regarding future use of this well.
The taste and odor problems reported in Maynard Well No. 3 are the reasons why the well is currently closed. For future use, the well must comply with state drinking water standards applicable for public water systems. We state that the drinking water supply for the towns of Maynard, Hudson, and Sudbury have not been affected by contamination from the Annex. Phase II groundwater data supports the conclusion. The bis (2-ethylhexyl) phthalate (BEHP) previously detected in groundwater from monitoring wells nearest Maynard Well No. 3 has been attributed to improper sample handling either in the field or the laboratory and is not site related. Maynard Well No. 3 did not contain BEHP for any of the Phase II sampling rounds.
Comment 17: The absence of any environmental sampling and analysis in off-site private wells is of concern. ATSDR should recommend sampling of these wells as well as placement of monitoring wells between the private wells and the areas of known contamination. This type of study will permit a more confident assessment of potential future health impact.
Monitoring wells are already in place between private wells and areas of known or potential groundwater contamination. We have recommended sampling of private wells in areas where contaminated groundwater on-post could affect off-site private wells.
Comment 18: In its Initial Release of the public health assessment, ATSDR recommended the Army conduct sampling of Whites Pond and private wells. The Army never did this, yet, ATSDR states in the December 93 public health assessment that "The Army has implemented or is addressing all previous ATSDR recommendations." This is untrue. Please explain this inconsistency.
The statement is true. The Army has sampled White Pond and no contamination was detected in either sediment or surface (52). The sampling is cited in both the initial release (pp 28, 45) and public comment release (pp 25) of the public health assessment. In addition, the Army has installed new monitoring wells in areas where drinking water supplies are hydraulically downgradient of study areas with incomplete sampling information. If groundwater monitoring data indicates that contaminants could affect off-site water supplies, we will recommend sampling those water supplies.
On March 30, 1994, we received new groundwater data for study areas P31 and P58. Metals (arsenic, lead, aluminum, iron, and manganese) exceed drinking water standards. The explosive 1,3-dinitrobenzene was detected in one monitoring well. Traces of pesticides were also detected. Volatile organic compounds were not detected. We called the Army on April 1, 1994 and recommended off-post sampling of private wells on Dawes Road. The Army is already considering that action. A followup letter recommending private well sampling was sent to the Army on April 22, 1994 (see Appendix 11).
Comment 19: Conspicuous by its absence is any discussion of the 'spring' located off Track Road in Maynard. The Maynard Board of Health posted this 'spring' several years ago to discourage residents from obtaining water there.
The Track Road spring was not discussed because water samples collected in June and October 1992 did not contain chemicals at levels of health concern. The reader is referred to the data summary for samples FWGW12 in Table 8-11, Site/Remedial Investigation Report, Fort Devens Sudbury Training Annex. July 1993 and January 1994 (4,15). To address the concern, a brief discussion of these findings has been added to the public health assessment in the section entitled Environmental Contamination and Other Hazards.
Soil
Comment 20: The Maynard Water transmission line is not located in areas of 'known contamination'. The statement that workers repairing lines 'could' be exposed if breaks occurred in areas of known contamination is speculative. ATSDR should also note in its response to citizen concerns about the water line that the line is located approximately four feet below the ground surface, and at least four feet above the groundwater surface. The transmission line is under pressure, and lines under pressure will leak out and not in. There is no evidence that soils under or adjacent to the line are contaminated.
On October 6, 1992, we met with personnel from the Maynard Department of Public Works and discussed the location of the water transmission line. On a map provided at that meeting, the water line is shown near Annex study areas P31, P40, A5, P11, P13, P10, and A8. For that reason we state in the public health assessment that . . . "Workers who repair the pipe could be exposed to chemical contamination if breaks in the water main occur in areas with known contamination." We have also added the comment that . . . " The Maynard Board of Selectmen state that the Maynard water transmission line is not located in areas of known contamination. The transmission line is under pressure, and lines under pressure will leak out and not in. The water distribution line is located approximately four feet below the ground surface, and at least four feet above the groundwater surface." The measures currently employed for repairing and flushing the water main are appropriate. However, we maintain that repair work should be coordinated with Annex Site Health and Safety personnel. Those personnel are knowledgeable about specific chemical hazards in various areas on post. If breaks in the water transmission line occur in such areas, then analysis of water from the Maynard distribution point should include site-specific chemicals.
Comment 21: My concern is that I have not found information which says that the water which the people of Maynard are drinking has been tested and has been found safe to drink. The assumption appears to be that the standard state mandated tests performed by the Maynard water department are sufficient to insure water quality. My assumption is that state mandated tests would not be designed to test for the exotic compounds which might come from the Annex.
On page 47, a community member asks: "b. The Maynard water line from White Pond to Summer Hill crosses STA (Sudbury Training Annex). Breaks in the line could contaminate the water supply." The answer is sufficient to address the question of "breaks" in the line and cautions about hazards to workers. The answer does not address the possibility of contamination from small, undetectable leaks in the line. Considering the topology of STA, it is highly unlikely that the line runs up hill from White's pond all the way across STA. Small leaks at seams in downgradient areas could siphon in contamination when negative pressures are experienced inside the line. Given that harmful levels of toxics described in the public health assessment are measured in parts per billion, it would not take a very large leak to introduce significant contamination in areas where the line runs under or close to toxic sites. I strongly urge ATSDR to recommend that the Maynard water supply be monitored where it exits STA.
Siphoning of contaminants into the water transmission line will not occur. As previous discussed in comment 20, the water transmission line is under pressure, and lines under pressure will leak out and not in. Further, the Maynard Board of Selectmen state that the water transmission line is not in contaminated areas of the Annex. As a precaution, however, ATSDR recommended that workers coordinate repairs to the water transmission line with Annex Site Health and Safety personnel. Those personnel are familiar with study areas on the Annex where there is chemical contamination. If breaks in the water line do occur in such areas, we recommended testing water at the Maynard distribution point for specific chemicals detected in those study areas.
Comment 22: The environmental characterization data that the Public Health Assessment relies upon are not complete for each potential pathway identified by ATSDR. Of primary concern is the relative absence of surface soil quality data, particularly in areas of the Annex where trespassers may contact surface soils. Of equal concern is the scarcity of soil quality data in the vicinity of the Town of Maynard water transmission line from White's Pond. The Public Health Assessment identifies soil in this area as a potential pathway (for water department workers repairing the line), but there are no data to permit quantification of potential risks to these workers. There is also a lack of adequate analytical data for stream sediments, although these are identified as potential pathways in the Assessment.
Discussions of surface soil data focus on Annex study areas where trespassing is known to occur, where community members expressed specific concerns, or in areas with public access. We recognize that there are other areas on the Annex where trespassers could contact surface soil that have not been discussed. The discussion makes the point that there are areas of the Annex that currently have surface soil contamination and physical hazards. The nature and extent of the that contamination is being evaluated. Clean up measures protective of public health will be implemented before the property is suitable for reuse. In the interim, people can avoid potential exposure to areas of surface soil contamination and physical hazards, by NOT trespassing.
We do not know if the water transmission line on the Annex is in areas of chemical contamination. Although the Maynard Board of Selectmen state that it is not, a map, provided in a meeting with the Maynard Department of Public Works (October 6, 1992), shows the water main near some Annex study areas. As a precaution we recommend that workers who repair breaks in the line coordinate with Site Health and Safety personnel. The safety personnel are familiar with specific areas of chemical contamination and can ensure that repair workers avoid potential chemical exposure while working on post.
Evaluation of available stream sediment data was included in the December 1993 public health assessment. We also stated that more sampling was planned in Phase II investigations. The Phase II data now available shows metals and pesticides in some on-post stream sediments and the Army plans to expand sampling in those areas. The Phase II data do not change the public health implications of the surface water and sediments pathway as previously stated in the December 1993 public health assessment. These are that people would not come into contact with surface water or sediment often enough (given the limited nature, both concentration and extent, of contamination) to cause adverse health consequences.
Comment 23: The Surface Soil Contamination section appears to state that only certain sites out of all the areas of concern on the installation were evaluated in the public health assessment. Even if this is appropriate for a public health assessment, one site that went on to an RI is not listed as having been evaluated. This is A4.
Recommendation: Even if A4 was shown to be lacking in significant risk, it should have been on the list as having been evaluated. If all the sites of concern, and data so far collected on them, were not evaluated in the public health assessment, state why. Also state if these other sites will be evaluated in the future, before the land is turned over to other public uses.
We have reviewed data for all Annex study areas and continue to review new information relating to public health issues. Although area A4 was not specifically discussed, we have recommended that people not trespass on the Annex because there is surface soil contamination and physical hazards in many of the on-site study areas; A4 among them. We walked over A4 during our October 1992 site visit. The area is vegetated and contains some construction debris. Cleanup is planned for Annex study areas where surface soil contaminants are at levels of health concern. Environmental cleanup will be completed before the Army releases Annex property for reuse.
Fish
Comment 24 (Land Use): Please state the date that Puffer Pond was posted. Also, permits for fishing are not obtained at the Annex gate.
The catch and release fishing policy at Puffer Pond was implemented in April 1992 and was stated on page 27 of the December 1993 public health assessment. According to Fort Devens Physical Security personnel, the Annex has been closed to recreational use since the summer of 1993. Permits to fish are no longer available. The public health assessment has been revised.
Comment 25: What is the significance of mercury in one fish from Puffer Pond? Mercury is a widespread contaminant in inland water bodies in the northeast, especially those that have lost buffering capacity due to acid precipitation. If the mercury in Puffer Pond fish cannot be linked to a contaminant source on the Annex, this metal should not be considered a site-related contaminant.
We do not state that mercury is a site-related contaminant. The significance of the mercury is that it was detected in one chain pickerel at a concentration exceeding the Food and Drug Administration (FDA) action level. For that reason, the pond has been posted catch and release fishing only. We concur that the posting is an appropriate public health action at this time. Because of the limitation of previous fish sampling, additional sampling has been done in both Puffer Pond and the off-post Minister's Pond. The results of that study, which is being done by OHM Corporation, are not yet available.
The mercury phenomenon may be widespread throughout the state in low alkalinity water bodies (Whitney Pond, Quabbin Reservoir, Echo Lake, and Walden Pond). Mercury may be available to fish from the substrates of these lakes or through atmospheric deposition (64, 99,66,64, 99).
Comment 26: Is there any evidence that mercury or other contaminants have migrated from the Annex to Puffer Pond? On page 33, the report notes that contaminants could not migrate to Vose Pond, Cutting Pond, or Crystal Lake, but nothing in the RI or the Assessment indicates that contaminants in fact can migrate to Puffer Pond.
Mercury is discussed in the response to Comment 25. There is some question about the source of arsenic detected in surface water and sediments from Puffer Pond. The arsenic could be from natural sources, general sources such as the use of arsenic-based herbicides on the Annex, or from the bunker area west of the pond. Further investigations are planned and should reveal potentially site-related sources of arsenic to Puffer Pond (40).
Comment 27 (summary): ATSDR neglects to address Taylor Brook and the Assabet River, both are downstream of Puffer Pond. If fish in Puffer Pond are considered a hazard, then the public may be at risk if they fish in Taylor Brook or the Assabet River. These should also be posted.
Taylor Brook and the Assabet River are not mentioned in the summary. However, those, and other area surface waters, are discussed elsewhere in the document. Sampling of on-post surface water and sediment has been done and the findings are discussed in the Environmental Contamination and Other Hazards section of the public health assessment. Fish sampling in the Assabet River and Taylor Brook has not been done in connection with investigations for the Annex. However, the Massachusetts Department of Environmental Protection did sample fish in the Assabet River in 1985. Five stations were sampled between Maynard and Shrewbury. The highest concentration of mercury detected in fish tissue was 0.3 ppm, which is below the FDA action level of 1 ppm (98).
Tissue from a single chain pickerel in Puffer Pond contained mercury at a concentration exceeding the FDA action level. Although fish data for Puffer Pond are limited, the Army posted the pond as catch and release fishing only. Because chain pickerel are territorial fish, remaining in fairly limited areas, they are likely to stay in Puffer Pond. Therefore, it is not necessary to post areas downstream. The Army has done more sampling in Puffer Pond, but the results of the fish study are not yet available.
Although elevated concentrations of arsenic do appear in some on-post areas, arsenic levels are close to background in samples taken at the exit point of Taylor Brook from the Annex. (40) The highest arsenic concentration detected in sediment samples from the Assabet River was 140 ppm and appears to be related mainly to sources upstream of the Annex. Arsenic does not biomagnify in the food chain and much of the arsenic in fish is an organic form that is essentially nontoxic.
Surface water posting is recommended when fish sampling data indicate that a fish consumption advisory is needed. Massachusetts has a fish toxics group that conducts such evaluations in the state. The fish toxics group includes members of the Massachusetts Department of Environmental Protection (Office of Watershed Management and Office of Research and Standards), the Massachusetts Department of Public Health, and the Department of Fish, Wildlife and Environmental Law Enforcement.
Concerns regarding fish contamination in surface waters off-post should be referred to the Massachusetts fish toxics group. The Massachusetts Department of Environmental Protection, Office of Research and Standards, in Boston, is the point of contact for the fish toxics group. The office telephone number is 617-292-5570. Data relating to Annex watersheds do not suggest that site contaminants have migrated to off-post portions of Taylor Brook or the Assabet River.
Comment 28: The presence of arsenic in Assabet River sediments should be evaluated in light of the historical use of arsenicals as pesticides in apple orchards.
As discussed in response to comment 27, elevated levels of arsenic in surface water and sediment do not necessarily lead to arsenic contamination in fish. Arsenic does not biomagnify in the food chain and much of the arsenic in fish is an organic form that is essentially nontoxic.
Comment 29: Explain how the AWQC were used to evaluate the potential for bioaccumulation in fish. The AWQC were developed using bioaccumulation data among other factors, but they do not provide a measure of bioaccumulation. A better estimate would be the octanol-water partition coefficient (KoW) or the Bioconcentration Factor (BCF).
By definition Ambient Water Quality Criteria (AWQC) are designed to provide adequate protection of aquatic life and its uses from acute and chronic toxicity to animals, toxicity to plants, and bioaccumulation by aquatic organisms (56). Bioaccumulation includes not only bioconcentration (the uptake of contaminants from water only) but also predators accumulating toxic substances by consuming prey.
TRESPASSING
Comment 30 (summary): People trespass on the Annex regularly. ATSDR should recommend that the Army post the site on the fencing they are reportedly installing.
A community member brought up the concern at the January 20, 1994 ATSDR public meeting at Maynard High School. ATSDR referred the posting request to the Army in a letter sent February 16, 1994.
Comment 31 (summary): Fences alone do not discourage people from entering the site if they are not aware of the fact that it may be hazardous to their health. We would have hoped that ATSDR would have at minimum recommended the Army undertake an education program to inform residents living adjacent to the Annex of the hazards of the site. None of the agencies have seen fit to communicate with the people living near the site via a mailing.
The Army held public meetings in January and April of 1994 to discuss the Annex. ATSDR also held a public meeting in January 1994 to discuss the hazards associated with the Annex. During that meeting we recommended that people refrain from trespassing on the Annex because there are localized areas of surface soil contamination and physical hazards.
Comment 32: There are inadequate analytical data for surface soils in all areas of the Annex that could be accessed by trespassers.
We emphasize that the Army has improved Annex fencing and increased patrols to discourage trespassing. To prevent exposure to surface soil contaminants and physical hazards on the Annex, people should not trespass.
Surface soil sampling on the Annex is biased toward areas with evidence of contamination or a history of activities that might have lead to chemical contamination. Such a sampling strategy is appropriate and adequate. In study areas where contaminants have been detected at levels of health concern, further characterization is ongoing. Once the nature and extent of chemical contamination around these particular sites is known, appropriate cleanup can be implemented.
Comment 33: Conclusion 4 notes that there has been 'limited' human exposure to contaminants. This has not been shown. There has been possible exposure to Annex contaminants, most likely by trespassers and people who drank from the Track Road spring, but ATSDR has not shown any evidence of exposure, 'limited' or otherwise.
It is unlikely that anyone has been exposed to contaminants at levels of health concern even if they trespassed on Annex property. "Limited" refers to the small number of trespassers who could contact areas with surface soil contamination on the Annex. Such exposure, if it occurs at all, would be short term. It also is not very likely that people are eating sufficient quantities of contaminated fish from Puffer Pond to result in adverse health effects. With regard to the Track Road spring, samples from June and October 1992 did not contain chemical contaminants at levels of health concern.
Comment 34 (summary): Hunting that takes place on the site is not addressed. Although it is illegal to hunt on the Army's land, we are concerned that the animals may eat fish from the pond or otherwise be contaminated. Again, posting the site as contaminated would help reduce this risk to the hunters.
As pointed out in the comment, hunting on the Annex is illegal. The Army has discouraged trespassing, including illegal hunting, by improving perimeter fencing and increasing the frequency of on-post patrols. On-post contamination is not widespread and, considering the size of the Annex, game are not likely to consistently feed in contaminated areas. Moreover, the types of animals that might eat fish from the pond (probably raccoons) are not consumed by people.
HEALTH OUTCOME DATA
We emphasize again that the focus of a public health assessment is whether human exposure occurred or is occurring. If exposure did not occur or is not occurring, as is the case with the Annex, then there is no question of health effects relating to exposure.
Comment 35: I have only lived here 5 years, yet I have heard artillery fire in the Annex, so particles could carry on the wind. Also dirt on contaminated soil is easily picked up during the summer and fall and with the wind ride through the air to reach surrounding people. I am concerned with wind and I think you should be working backwards. Gather data first and search for the problem. Look for patterns. Track peoples health first. There is enough data for you to gather.
We do not think that people in residential areas are being exposed to contaminated dust from the Annex. Although there is some soil contamination on the Annex, the contaminants are mainly associated with localized spills and are not widespread. The wind would not carry contaminants very far because much of the Annex is wooded. Moreover, the on-post training areas are fairly remote and not close to residential areas.
It is not effective to work backwards from health outcome data to identify a disease in a community and then attempt to find a point of exposure. Diseases described in registry information or medical records may have multiple causes including infectious agents, genetic factors, household chemicals, industrial chemicals or the cause may be entirely unknown. Little or no information may be available in the data bases about the individual case in terms of location or lifestyle. To work backwards from health information to a list of chemicals at a particular installation would introduce too many uncontrolled variables into the analysis to get accurate results. Instead, the public health assessment focuses on whether people are contacting environmental contaminants at levels of health concern. We do not think that such exposures are occurring at the Annex.
Comment 36: The Public Health Assessment finding of no apparent public health hazard is based primarily on the absence of increased cancer incidence in the four towns between 1982 and 1986.
The finding of no apparent public health hazard is based on no one being exposed to site contaminants. Annex study areas with chemical contamination are being further investigated and will be cleaned up before the property is available for reuse.
Comment 37: Our concern is that cancer incidence may not be the most appropriate measure of health status in the four towns. ATSDR has based its entire Assessment of a site that has been in operation since the 1940s on four years of cancer incidence data from the early 1980s.
ATSDR based its public health assessment on little or no evidence (or likelihood) of human exposure to site contaminants.
Comment 38 (summary): All available health outcome data was not reviewed.
ATSDR reviewed cancer incidence data for 1982-1986 and cancer mortality information for 1987-1988 because those data were available to us at the time the initial release public health assessment was written (1991-92). When the public comment release of the public health assessment was written (1992-93) no additional health outcome data were reviewed because new environmental data indicated little likelihood of human exposure. The statement in the summary has been modified for clarity.
Comment 39 (Natural Resource Use): ATSDR did not evaluate all health outcome data. ATSDR only viewed 4 years of the 12 that are available, that is only 1/3 of available data. This should be corrected.
We did not state that all health outcome data were reviewed. The specific data that we reviewed is listed in the public health assessment. For clarity, a sentence has been added to the paragraph.
Comment 40: The Health Outcome Data Evaluation section states that if the "range included 100, then the difference is due to chance." This should state "probably due to chance".
In addition, the opposite problem should be explained more clearly. That is, what if multiple analyses are done, as they are here, and some turn up statistically significant. Some concern has already been raised at a public meeting over the "excess: pancreatic cancer in Sudbury, and liver cancer in Maynard. After this issue is raised in the public health assessment, it is not fully explained, only generalized in the statement on page 43 that states, "this analysis of the cancer incidence data around the Annex does not indicate any pattern of occurrence that is of increased public health concern at the present time." A member of the public may wonder how both conclusions are compatible - i.e., excess cancer and no excess cancer.
In addition, concern was raised over whether this should be followed up in some way. The conclusions, recommendations and plans sections of the public health assessment do not address this.
Recommendation: Either the "excess" cancer that is shown needs to be fully explained so that the public understands why there is no need for follow up, or some sort of follow up should be discussed.
An expanded discussion regarding cancer of the liver and pancreas is provided in the Health Outcome Evaluation Section (p. 43).
Comment 41: We recognize that Massachusetts has had a cancer registry only since 1982 but there are other sources of health outcome data. In addition, ATSDR should have used the 1982-1988 Cancer Registry Data, which has been available since July 1991. If six years of cancer incidence data are used, rates of colo-rectal cancer in Maynard and Stow and pancreatic cancer in Sudbury are significantly higher than expected. There are other differences that can be evaluated using six years' incidence data, but for most of the reported cancers, the total numbers of cases in the four towns are too low to evaluate statistically, even when the incidence data from the four towns are combined.
We are not suggesting that the differences in cancer incidence over the six years is due to the Sudbury Annex, but are questioning the use of such limited data, particularly when no attempt was made to correlate any of the cancer outcomes to the contaminants present at the site. The ATSDR report could be summed as follows: "We didn't see increases in cancer in the four towns between 1982 and 1986, so we conclude that there are no impacts from the site."
Even with additional years of registry data, the limitations of that type of information do not allow specific conclusions about causes of the disease. Exposure pathway information must be considered. At the Annex, no evidence of exposure was seen; therefore, there is no reason to conclude the site has had an impact on the public health.
Comment 42: The Public Health Assessment makes no attempt to correlate contaminants found on the Annex with health outcome. There is no discussion of whether any of the site contaminants is associated with the types of cancer addressed in the health outcome discussion. If the contaminants that receptors could have been exposed to through complete pathways are not associated with the cancers that were discussed, ATSDR cannot use the absence of increased cancer incidence as evidence that there is no public health impact from the Annex.
We evaluated environmental information and analyzed the potential for people to be exposed to contaminants as a basis for determining the health impact of the Annex. The Public Health Assessment makes no attempt to correlate contaminants found on the Annex with health outcome (cancer or noncancer), because there is little evidence, or likelihood, of human exposure.
Comment 43: Based upon the analysis conducted in this report, we also suggest that it is appropriate to state that the health outcome data "do not suggest that the Annex contributed to increased cancer incidence between 1982 and 1986". There is no basis for the broader statement in this report.
The conclusion has been modified.
Comment 44: As ATSDR knows, health impacts other than cancer can be associated with exposure to environmental contaminants. Common contaminants such as mercury, lead, xylene, toluene, and numerous others can cause harmful non-cancer effects in exposed persons. In the Assessment ATSDR addressed the number of children in special education classes in response to community concerns, and concluded that there were no significant differences between observed and expected in the four towns. This was the only non-cancer impact that was given anything but cursory consideration in the Assessment.
The Maynard Board of Health reviewed the Massachusetts Department of Public Health death statistics for 1990, the latest year for which they are available. Age-adjusted death rates could not be calculated because the death statistics are not available in a format that permits assessment of each individual case, and therefore age at death.
However, the Board calculated Crude Death Rates for the six causes of death occurring with the greatest frequency in the four towns, with the exception of motor vehicle accidents and other accidental injury. The six causes as listed by the MDPH were heart disease, all cancers, stroke, pneumonia and influenza, chronic obstructive pulmonary disease, and liver disease and cirrhosis. The calculated Crude Death Rates were compared to calculated Crude Death rates for the state as a whole in 1990. With the exception of significantly increased death rates from pneumonia and influenza in Maynard and heart disease in Hudson, all rates for the four towns were significantly lower than those for Massachusetts. The presence of older than expected populations in Maynard and Hudson may explain the causes of death in these towns, but, as noted, we were unable to calculate age-adjusted rates. With the exception of Hudson, there were significantly fewer deaths from cancer in the four towns. In Hudson, the rate is slightly lower than that of the state. Three towns had very low deaths from liver diseases other than cancer in 1990 and Maynard had none.
We are not suggesting that any of these deaths are related to the site. Our concern is that ATSDR did not review Crude Death Rates as part of its evaluation. These data could be inspected for every year since the Annex began operation. Other sources of health outcome data include fetal death certificates, birth certificates, death certificates and school health records. We recognize that these records are often incomplete, and that death certificates may not include contributing and coexisting illnesses, but a review of other health outcome data would have provided additional confidence in the findings of the Public Health Assessment.
We did not examine the crude death rates because they are extremely limited in the information they provided. The major factor influencing the risk of death is age. Crude rates, which refer to the total population, may obscure the fact that subgroups of the population exhibit significant differences in risk. In addition, the conditions examined above, such as heart disease and COPD (Chronic Obstructive Pulmonary Disease), are chronic, multifactoral diseases. Therefore, it is not possible in a general review to sort out effects from different lifestyle factors as opposed to environmental factors. The death rates do not adequately reflect the rate of new cases. Therefore, crude death rates offer no useful site-specific information.
With regard to death statistics data, individual cases are not needed to calculate age-adjusted rates. The rates are determined by the number of cases in each age group and the age structure for a standard population.
As stated, special education rates were examined to provide information about a specific community concern. Other outcome data were not examined because there is no evidence of exposure of public health significance and no other specific community concerns.
Comment 45: We are concerned that the public health assessment does not answer the citizens' questions about the potential health impacts of living near the Sudbury Annex. To this end, we recommend that ATSDR should review health outcome data other than cancer for the four towns, and should include data from the past 50 years in this review.
It is our opinion that if there is an impact on public health from the contaminants at the Annex, that impact would have been evident in the past when active operations were occurring. There were more residents using private wells in the past and also residents were more likely to have trespassed on the site. An air pathway may have been of concern when the site was active. It is our opinion that limiting a health outcome study to four years in the early 1980s is not appropriate, and that if there has been a health impact, it will be evident from death certificates and other available health outcome data records.
When there is no exposure there will be no impact on the public health. It is not appropriate to examine records from the past for multifactoral diseases and attempt to make an association, particularly when past exposure is not a public health issue.
Comment 46: The toxicity profile for mercury included as Appendix 10 should reference the latest edition of Doull, et al. The fourth edition was published in 1991.
Doull, et al. fourth edition 1991 is the appropriate reference for the information cited in Appendix 10.
Comment 47 (Demographics): Median age should be calculated for all towns.
The median age for each town is listed in Table A-1.
Comment 48 (Demographics): As the Annex has been on the NPL since 1989 and ATSDR has been studying it since 1991, we are puzzled that ATSDR was unable to determine who lived at the Capehart Housing and how many children. If it was difficult to obtain this information in the summer, why couldn't it be obtained during the other seasons?
Specific information regarding Capehart Housing residents varies over time but the occupancy rate is virtually %100. The information presented in the December 1993 public health assessment reflects the residential status information that was available during the summer of 1993; the time period when the inquiry was made. Updated demographic information on Capehart Housing residents is not needed because they are NOT being exposed to site contaminants.
Comment 49 (Land Use): The dates on the leases should be listed.
Army leases for the Annex vary over time. For details regarding property leases, the reader is referred to Chapter 3 of the Draft Phase II Site Investigations Report, Volume 1, March 1994 prepared for the Army by Ecology and Environment, Inc.
FUTURE LAND USE
Comment 50: The public health assessment does not address the future use of the Annex, perhaps the most important issue the communities face at this time.
Suggested reuse is open land for recreational use, but no agreement has been reached at this time. The Army plans to clean up study areas on the Annex to levels protective of public health. The Army will retain control of the property until site remediation is completed.
Comment 51 (Recommendations): This section is inadequate and does not address the future use of the Annex. ATSDR recommends restricting access, but this is unrealistic for the future.
At this time, plans for land reuse for the Annex are speculative. We recommend restricting access to the site because of current site conditions. Once environmental cleanup proceeds and plans for future land use are clearly identified, we will modify public health recommendations as needed.
GENERAL
Comment 52 (Site Visit): ATSDR did meet with FOCUS and some members of other community groups on October 5, 1991; however to define the Public Availability Session as a meeting with area residents is misleading. The only people who attended this meeting were notified by FOCUS. ATSDR made no attempt to advertise this meeting so that "the public" could attend. We would like to see an accurate description of the number of people who attended these sessions" and their specific expressed concerns.
ATSDR held several meetings with community members. A meeting with representatives of area community groups was held on October 5, 1992, not 1991. That meeting was arranged between ATSDR and community group representatives and was not advertised. Approximately 10 people attended. Public Availability Sessions (PAS) were held the following day, October 6, 1992 from 10:00am - 2:30pm and 4:30pm-8:00pm at the Capt. T. Dustin Alward Massachusetts Firefighting Academy. Press releases announcing the PAS were sent to all area newspapers, and local radio and television stations. The Maynard Beacon and the Daily Sun (Marlboro-Hudson Daily Paper) printed articles announcing the meeting. ATSDR coordinated advertising with representatives of FOCUS, the Lake Boone Association, and the Organization for the Assabet River. Those groups requested fliers announcing the PAS (250 were sent). Additional fliers (50) were mailed to representatives of local government for the four towns (boards of selectmen, health departments, etc). Approximately 8 people attended. Community health concerns are listed on pages 12-13 and addressed on pages 45 - 52.
Comment 53. Also, at the October 1991 meeting with FOCUS, ATSDR committed verbally to producing a public health assessment by November of 1991. The public health assessment was not produced until more than a year later. What data was evaluated after this meeting?
The meeting was held in October 1992, not October 1991. In that meeting, Dr. Gary Campbell, ATSDR Army Unit Chief, stated that he could not speculate when the public comment public health assessment would be released. Data that were evaluated after the meeting are listed in the References and Appendix 3 of the December 17, 1993 public health assessment (public comment release).
Comment 54: In the section entitled "A note of explanation" ATSDR states that the public health assessment meets the statutory criteria set out in CERCLA. This is incorrect.
The actual statement reads . . . "This document represents the Agency's best efforts, based on currently available information, to fulfill the statutory criteria set out in CERCLA section 104 (i)(6) within a limited timeframe." That statement is true. We are addressing public health issues at all sites proposed on the National Priority List.
Comment 55 (Recommendations): Although ATSDR recommends "education of health professionals", they then state that "an ATSDR staff member participated in the Massachusetts Department of Public Health Grand Rounds professional health education series. No additional health follow-up activities are indicated at this time." At the Public Availability Session on January 20, 1994, ATSDR representatives stated that the staff member was from Atlanta. We would like to know how an ATSDR staff member residing in the Atlanta area has helped "educate" local health professionals. This is absurd and is another example of what is wrong with ATSDR's process. This does not serve the public.
A physician from ATSDR headquarters in Atlanta flew to Massachusetts to present a grand rounds at the Framingham General Hospital on exposure to volatile organic compounds. Informational handouts were provided at the session and telephone numbers of additional resources were provided. The Massachusetts Department of Public Health selects the format of the health professional education because they have the expertise in determining what are the best mechanisms for meeting those needs in the state. ATSDR provides support to that program as requested and needed.
Comment 56: Why has ATSDR not been in contact with the towns local emergency response teams? They are very concerned about the potential of spills or contamination during clean up activities. As the lead public health agency, ATSDR should be in touch with these professionals.
Emergency response activities relating to environmental cleanup at the Annex are coordinated by site remediation contractors. The reader is referred to the following documents for specific information about those activities; 1. OHM's Site-Specific Health and Safety Plan for Site and Remedial Investigations at the Fort Devens Sudbury Training Annex dated April 13, 1992 and 2. Ecology and Environment's Health and Safety Plan Addendum, Phase II Site Inspections, Remedial Investigations dated January 1994.
Comment 57: There are gaps in the data upon which the public health assessment conclusions are based. The logic behind the overall conclusion, that no public health hazard exists, appears questionable in my mind for the following reasons:
1: Adequate community-specific health outcome data are limited at best;
2: Few people have been exposed, given that the area is restricted, so logically, it would be difficult, if not impossible, to determine the degree of public risk, let alone document negative health outcomes.
I question whether any adequate baseline data is available to determine whether negative health outcomes exist or are due to environmental exposure over time. Nor is there a mechanism to identify and track individuals who may have been exposed. The necessary data to substantiate a conclusion of "no public health hazard" simply does not exist.
Given limited data and the lack of appropriate measures of mortality and morbidity linked to environmental exposure, I am surprised at the conclusion indicating no public health hazard for the Annex. I am concerned about surface and groundwater contamination. I think the general public is unaware of the process underway to test the site for possible health hazards as well as the process involved in eventual land use. I would recommend that a "public forum" be held in Maynard, with sufficient notice given to the general public, at which ATSDR would:
We concur that evaluating health outcome data in relation to the few exposures that might have occurred at the Annex is unproductive. Instead, our approach is to identify ways in which people could be exposed to site contaminants and take public health actions to eliminate those exposures.
The conclusion of no apparent public health hazard is appropriate because it is unlikely that people were exposed to site contaminants at levels of health concern. This conclusion is based on a review of environmental data and historical activities at the Annex.
The potential for current exposure to site contaminants is limited to the few people who trespass on post. To avoid contacting contaminants and physical hazards on the Annex, we have recommended that people not trespass.
Future land use is not a public health concern because contaminated areas will be cleaned up before the property is reused, so people will not come into contact with contaminants.
We have already described at both public meetings and in the health assessment the methods used in the health assessment process. In particular we have emphasized that the assessment focuses on whether or not people have come into contact with contaminants, in other words, on whether there are completed exposure pathways. We will continue to address public health issues at the site as needs arise.
