PUBLIC HEALTH ASSESSMENT
AIR AND GROUNDWATER PATHWAYS ANALYSES
AGANA POWER PLANT
September 22, 2004
PCBs, when released to aquatic systems, tend to accumulate in sediments. Because they are not very volatile or soluble, they tend to persist in sediments for long periods of time. At some sites, for example, PCBs have been found to remain in river sediments for several decades. Although sediments clearly retain PCBs, various processes gradually remove small fractions of PCBs from contaminated sediments over long time scales. For example, aquatic organisms can accumulate trace amounts of PCBs from sediments, and some bacteria are known to degrade PCBs. PCBs can also partition from sediments into water, and some of the dissolved PCBs can then evaporate into the air. However, only limited evaporation is expected to occur given the chemical and physical properties of PCBs.
When characterizing the extent to which PCBs actually evaporate from Agana Swamp, ATSDR first searched all available site documents and contacted the Navy and GEPA to determine if any sampling studies measured airborne levels of PCBs. No such studies have been performed. To provide perspective on the exposure issues at Agana Swamp, ATSDR reviewed findings from other sites where aquatic environments are contaminated with PCBs. Data from other locations cannot indicate exactly what exposure levels might occur near Agana Swamp, but experiences from these locations do provide insights on the likely exposure levels. The following discussion summarizes our current understanding of the extent to which PCBs evaporate from aquatic systems, based on a review of data collected at selected sites with large volumes of sampling data:
- Hudson River, New York. The Hudson River is arguably one of the most extensively studied sites with PCB-contaminated sediments. Several industrial sources contaminated the river's sediments, but two capacitor manufacturing plants are believed to account for the majority of the contamination. These plants discharged up to 1,300,000 pounds of PCBs directly into the river (EPA 2000). Even though these discharges largely ceased in the 1970s, recent sampling events continue to detect PCB contamination in sediments at concentrations up to 1,352 ppm. Furthermore, site-related contamination is believed to extend more than 100 miles downstream from the two plants (EPA 1998b). Average PCB concentrations in fish tissues of several species (e.g., carp, brown bullhead, largemouth bass) continue to be greater than 10 ppm on a wet weight basis (EPA 2002a). Thus, the Hudson River clearly contains considerably greater PCB contamination than does the Agana Swamp.
When evaluating the Hudson River, EPA found that inhaling PCBs that evaporate from the river water pose virtually no health risks, both for cancer and non-cancer endpoints (EPA 2000). EPA's modeling analysis estimated that average air concentrations of PCBs along the banks of the most highly contaminated stretches of the Hudson River would be 0.00021 g/m3 (EPA 2000). This estimated concentration is more than an order of magnitude lower than 0.0031 ug/m3, the risk-based concentration published by EPA Region 3 that ATSDR considered as part of its initial toxicity screen. In summary, estimated airborne PCB levels along the Hudson River are not at levels of health concern.
- Fox River, Wisconsin. The Fox River flows through an industrial region in Wisconsin. State regulators have estimated that the local industrial sources released as much as 825,000 pounds of PCBs directly to the river (WDNR 1998). Sediments in the most heavily contaminated stretch of the Fox River have an average PCB concentration of 21 ppm and a maximum concentration of 770 ppm (The RETEC Group, Inc., 2002). This river's sediments are considerably more contaminated than the sediments in Agana Swamp, which have a maximum PCB concentration of 9 ppm. Additionally, relatively recent fish tissue sampling from parts of the Fox River have shown average PCB contamination in fish tissue (walleye) to be 2.74 ppm (WDNR 2002)-levels much higher than have been observed in fish from Agana Swamp.
A human health risk assessment prepared for the Wisconsin Department of Natural Resources examined potential exposures to PCBs from various pathways, including inhalation of PCBs that evaporate from the Fox River and Green Bay. The risk assessment found that inhalation exposures are far lower than levels of health concern; further, the inhalation exposures accounted for less than 0.001% of the total additional cancer risks estimated for the site (The RETEC Group, Inc., 2002). Overall, these findings again suggest that evaporation of PCBs from aquatic systems-even the more highly contaminated river systems-is generally not of health concern to local residents.
- New Bedford Harbor, Massachusetts. For many years, industrial facilities discharged wastes, some containing PCBs, into New Bedford Harbor. In the 1980s, EPA discovered that a large section (or "hot spot") in the harbor had sediment concentrations of PCBs greater than 4,000 ppm (EPA 2002b). Extensive dredging and ambient air monitoring has already occurred at this site. According to EPA, "no airborne PCB levels were detected that posed a health risk to cleanup workers or area residents during the five months of excavation work" (EPA 2003c). This result suggests that PCBs are not readily released from sediments, even when sediments are being dredged.
- Review of other sites. ATSDR also considered findings from many other sites with PCB-contaminated sediments. These include the Housatonic River (Massachusetts), the Sheboygan River (Wisconsin), the Kalamazoo River (Michigan), and the St. Lawrence River (New York-Canada border). Similar findings regarding air inhalation exposures were reported for all of these sites.
- Background information on dusts from roads. EPA has conducted extensive research on the amount of airborne dust particles that are released when motor vehicles drive on both dirt roads and paved roads. The agency's most recent publication on this matter indicates that dust emissions from roadways depend on several factors, including the amount of motor vehicle traffic, the average weight of motor vehicles driving on roadways, local road conditions, and frequency of rainfall (EPA 1998d). To a first approximation, EPA's publication indicates that dust emissions from unpaved roads are extremely limited on days that have at least 0.01 inch of rainfall. According to recent data provided by the National Weather Service, the weather station at Guam International Airport received more than 0.01 inch of rainfall more than 90% of the days for the most recent and complete year of record (NCDC 2003). Given this percentage of days with rains, EPA's model would predict that air emissions from the roads in the vicinity of the former APP are relatively low. However, ATSDR could not use EPA's model to estimate actual dust emissions, because a detailed inventory of the motor vehicles, roadway characteristics, and transportation patterns is not available for Guam. ATSDR does not view this lack of information as a critical data gap, because the other lines of evidence considered for this issue (see below) provide further insights on air contaminants that might be released from the local roadways.
- Air pollution measurements at Guam. GEPA and other parties conducted many types of air sampling at Guam and reported their results to a nationwide EPA database named the Aerometric Information Retrieval System, or AIRS. ATSDR recently accessed the entire history of air pollution measurements reported in AIRS for monitoring locations on Guam. Overall, more than 6,500 separate air quality measurements from Guam are documented in the database.
The most relevant pollutant for evaluating exposure to wind-blown dust is particulate matter-airborne particles and water/fluid droplets that come from a variety of sources. One subset of particulate matter that has been well researched is PM10, airborne particles that are smaller than 10 microns. These particles are smaller than a human hair is wide, too small to be seen by the naked eye. Researchers have studied PM10 because particles of this size can be inhaled and can deposit deep in the human lung.
The AIRS database includes nearly 400 sampling results for PM10 collected on Guam. These samples were collected between 1989 and 1991 using EPA-approved sampling equipment. No sampling has occurred since 1991. The available PM10 monitoring data were collected at four locations, one each in Agana, Dededo, Mangilao, and Piti. The precise locations of these former sampling stations are shown in Figure 8. Most of the stations were located in or near heavily traveled areas, and the Agana station was located in a populated area approximately 1 mile west of (and downwind from) the former APP. At each location, 24-hour average PM10 samples were generally collected once every 6 days. Such a monitoring schedule is routinely followed by particulate monitoring stations across the country. No monitoring data are available from these stations for the summer of 1989, a gap that might have resulted from a large typhoon that hit the island at that time. Table 6 summarizes the monitoring data from these four stations.
|Monitoring Station||Years of Operation||Number of Samples Collected||Average PM10 Concentration ( µg/m3)||Highest PM10 Concentration (µg/m3)|
|Piti (Cabras Island)||1989 and 1991||118||49.5||123|
At all four stations, the average PM10 concentrations were lower than EPA's annual average health-based standard (50µg/m3) and the highest concentrations were lower than EPA's corresponding 24-hour average standards (150µg/m3). Moreover, the sampling data collected nearest the former APP (i.e., at the Agana station) were considerably lower than EPA's air quality standards. These observations indicate that, between 1989 and 1991, airborne dusts in and near heavily traveled neighborhoods on Guam were not at levels of health concern. This finding likely applies today, assuming that traffic patterns and the types of motor vehicles used in these areas have not changed dramatically since air samples were previously collected.
- Estimated PCB concentrations in airborne dust. The previous analysis suggests that vehicles driving on dirt roads and paved roads covered with dust do not raise airborne particulate matter levels at Guam to levels of health concern. In addition to evaluating potential particulate matter exposures, ATSDR considered whether residents are exposed to PCBs that might be present in the dusts from the local roads. We evaluated this scenario because residents, noting that surface soils at some locations near the former APP contain PCBs, wondered if the PCBs might become airborne, either in wind-blown dust or due to vehicles traveling over dusty surfaces.
ATSDR is not aware of any air sampling for PCBs that has occurred in the vicinity of the former APP. Although ATSDR would prefer to base its conclusions on sampling data, we note that one can estimate exposures to PCBs in dusts by making assumptions about the PCB content of the particulate matter that was sampled. For example, at the four particulate monitoring stations considered, the highest long-term average PM10 concentration measured was 49 µg/m3. The PCB content of this particulate matter would have to be at least 65 ppm for the total PCB air concentration in the particles to reach 0.0031 µg/m3 (the health-based screening value that ATSDR uses to identify sites that need more detailed evaluation of chronic inhalation exposures to PCBs). The highest surface soil concentration of PCBs measured in 38 samples collected from yards near the former APP was only 0.55 ppm (Earth Tech 1999b). Therefore, it seems unlikely that the PCB content of the airborne particulate matter would be higher than the screening value ATSDR uses.
- Contaminants that naturally form when vegetation burns (i.e., emissions not related to the former APP). Researchers have long established that elevated exposures to smoke from fires can cause acute respiratory health effects (e.g., cough, lung irritation, shortness of breath) that typically go away after exposure stops. These effects have been reported among people who breathe elevated levels of smoke from fires observed at locations throughout the United States and are not unique to smoke from Agana Swamp fires. Whether people suffer these effects depends on the amount of smoke that people inhale and their general health status. Residents can prevent smoke-related health effects by reducing their exposure to smoke from wildfires, perhaps by staying indoors when fires occur, leaving smoke-affected areas, or avoiding strenuous activity during fires (Refer to Appendix G for additional information on how to reduce smoke-related exposures).
- Air emissions of contaminants related to the former APP. ATSDR recognizes that the fires in Agana Swamp might cause trace amounts of contaminants to be released from sediments. The pollutants generated from burning vegetation, though, are believed to far outweigh the APP-related contaminants that might evaporate from sediments. Based on this observation and estimated exposure levels for PCBs, ATSDR does not believe that the swamp fires expose residents to APP-related contaminants at levels associated with adverse health effects.
- Air emissions of contaminants associated with wastes not related to the former APP in the Agana Swamp. ATSDR has learned that various parties have disposed of additional waste materials (e.g., junk piles) in the Agana Swamp. Although combustion of these materials likely releases contaminants into the air, no detailed inventory is available that documents the amount of wastes present. Thus, ATSDR cannot characterize the potential air quality impacts. Nonetheless, by taking the measures mentioned in the first bullet item, Guam residents would dramatically reduce the health risks (if any) associated with inhaling contaminants generated by the burning wastes.
The remainder of this section presents our detailed analyses of the three types of contaminants listed above. It is important to note that only the second type of contaminant is directly associated with wastes formerly discharged by APP.
- Contaminants that naturally form when vegetation burns (i.e., emissions not related to the former APP). Water levels and amounts of vegetation in Agana Swamp vary throughout the year. The entire swamp area is typically submerged in water during the rainy season. During the dry season, however, standing water at times is observed only in low-lying areas and in the main channel of the Agana River. The vegetation at this time of year can become quite dry, making the Agana Swamp vulnerable to fires.
Fires in Agana Swamp have varied in intensity, frequency, and duration. ATSDR consulted with various sources to understand how often the fires occur and how long they typically last, but we could not locate a complete history of all fires in the swamp. However, two data sources provided insights on these events. First, the United Nations Environmental Programme has an ongoing initiative to document the status of wetlands in the Pacific Islands. This initiative-the Protected Areas Programme-includes a report on the Agana Swamp (UNEP 2001). According to this report, four "major fires" occurred in the swamp between 1991 and 2001. No further information is given on these fires or the frequency with which smaller fires occurred. Figure 9 is a wetland map of the Agana Swamp area. ATSDR has contacted other parties at Guam to obtain information on the frequency, duration, and intensity of fires in Agana Swamp, but none of these parties have provided detailed insights on these issues.
During wildfires, dry vegetation burns and releases various pollutants to the air; wet vegetation can also burn if a fire gets hot enough. Pollutants released during fires include particulate matter, carbon monoxide, formaldehyde, acrolein, and benzene. These pollutants are naturally formed from burning vegetation and are released during virtually every wildfire, not just those that occur in Agana Swamp The amount of these and other pollutants released depends primarily on the area burned and the fuel and moisture content of the vegetation (EPA 1996). EPA has published equations that allow one to estimate the amount of pollution released in smoke from fires; actual emission rates are difficult to predict accurately, because they ultimately depend greatly on site-specific conditions.
ATSDR did not locate any studies that were designed specifically to measure air quality impacts from fires at Agana Swamp. Ample evidence, though, demonstrates that people can experience various adverse health effects following elevated exposures to smoke from fires. Specific health effects or incidents observed among exposed populations include increased emergency room visits for asthma, bronchitis, and chest pain (CDC 1999); worsening of lower respiratory symptoms, increased cough, and difficult breathing (CDC 2000); and increased hospitalizations for various circulatory and respiratory illnesses (CDC 2001). It is important to note that the studies reporting these adverse health effects all examined moderate to large forest fires that lasted weeks or months. Because these fire conditions are not consistent with those that occur at Agana Swamp, it is difficult to predict whether or not adverse health effects would occur among Guam residents exposed to smoke from swamp fires. Nonetheless, it is conceivable that residents would suffer adverse health effects if they experience elevated and prolonged exposures to concentrated smoke plumes.
Although we cannot predict whether or not fires in the Agana Swamp will cause adverse health effects, Guam residents can help ensure that no adverse health effects result from inhaling smoke by following two prudent public health measures on days when swamp fires occur. First, they can greatly reduce their exposure levels by avoiding contact with the smoke from the fires, perhaps by staying indoors or leaving smoke-affected areas during the fires. Second, they can reduce exposures by not engaging in strenuous activity when the fires occur. Both exposure reduction measures are particularly important for people who might be sensitive to the effects of smoke, such as children, the elderly, and people with respiratory or circulatory conditions. For further insights on reducing exposures, ATSDR will stock the record repositories with extra copies of a recent EPA fact sheet titled "How Smoke From Fires Can Affect Your Health" (EPA 2003d).
- Air emissions of contaminants related to the former APP. No direct evidence (e.g., an air sampling study) is available to evaluate whether fires in Agana Swamp cause appreciable amounts of APP-related contaminants in sediments to become airborne, but indirect evidence suggests that only a small proportion of PCBs are likely released from sediments during fires. First, because the organic material in vegetation provides the majority of fuel for the swamp fires (rather than the largely inorganic sediments), ATSDR fully expects that air emissions from these fires will be dominated by the pollutants that naturally form when vegetation burns. Second, PCB-contaminated sediments in submerged swamp areas are expected to remain in the swamp during fires, but the fate of PCBs in exposed sediments is not known. Third, the fact that PCB contamination still remains in the Agana Swamp sediments after several major fires have occurred provides indirect evidence that these fires do not cause a large proportion of the sediment-bound contaminants to become airborne. Although none of these individual observations prove that fires do not release a large proportion of the PCB inventory in the Agana Swamp, all three observations combined suggest that the amounts released likely account for only a small fraction of the PCBs in the swamp.
ATSDR also considered toxicity data for PCBs when addressing this community concern. If fires in Agana Swamp indeed release PCBs to the air, only short-term or acute exposures would likely result. The available toxicity data for PCBs suggests that it is highly unlikely that such short-term exposure doses might cause adverse health effects. Results from 11 laboratory animal studies, for instance, indicate that rats, mice, guinea pigs, and rabbits had no toxic effects following short-term inhalation exposures to PCBs at levels up to 9 µg/m3 (ATSDR 2000c). ATSDR doubts that PCB concentrations near the former APP would reach these levels, even during swamp fires. The reason for our doubt is that particulate matter concentrations observed downwind from major forest fires in the United States have been shown to reach levels ranging from 100 to 400 µg/m3. The PCB content in the particulate matter would have to be at least 2% in order for the airborne PCB concentrations to reach 9 &mciro;g/m3. It is highly unlikely that the particulate matter would contain this much PCBs, given that the highest sediment concentration of PCBs is only 0.0009%. Therefore, ATSDR does not believe fires in the Agana Swamp would cause APP-related sediment contamination to be released at levels of public health concern.
For these reasons, ATSDR concludes that fires in Agana Swamp will not cause residents to be exposed to PCBs at levels associated with adverse health effects. Residents who remain concerned about this issue can greatly reduce their potential exposures to PCBs, if any such exposures occur, by adhering to the recommendations we made to avoid coming into contact with smoke and avoid engaging in physical activity on days when swamp fires occur.
- Air emissions of contaminants associated with wastes not related to the former APP in the Agana Swamp. ATSDR has learned that various parties have disposed of additional waste materials in the Agana Swamp. Community members have stated that wastes might include junk piles, municipal waste, and sewage. During fires in Agana Swamp, combustion of these waste items might release contaminants into the air. ATSDR cannot fully evaluate these releases, because no quantitative information is available on the composition and volume of the miscellaneous waste materials in the swamp. Although we cannot make definitive statements about the emissions from these burning wastes, ATSDR notes that residents can dramatically reduce the health risks (if any occur) associated with this source by taking prudent exposure reduction measures during fires, such as avoiding contact with smoke and remaining indoors while fires burn.
During past site visits to Guam, ATSDR met with local residents, the mayor of Mongmong, and representatives from government of Guam agencies to discuss a variety of environmental concerns. One ongoing concern in the community involves unearthing and coming into contact with buried wastes that may pose chemical or physical hazards. Much of the village of Mongmong is located on the old 5th Field Marine Depot, a former military installation. The residents have expressed concerns about the chemical warfare materials from chemical identification and training kits discovered at this former site and the potential of other buried wastes to be unearthed in the Mongmong area (refer to Appendix F for a list of potential sources of unearthed wastes in the Mongmong area). There is little information about the amount of material buried and the locations of burial from military and non-military sources including battlefield contamination.
In addition to the buried canisters identified at the former DOD site, dozens of small glass vials filled with a substance reported by the news media as "white phosphorus" were uncovered in January 2001 from a residential area in the village of Toto. This prompted an emergency response by local agencies. Various government of Guam agencies were involved in removal of the vials from the area (Variety News Staff 2001). The glass vials were later identified as calcium hypochlorite, which was used as a method for purifying water (USACE 2003).
What are GEPA and USACE doing about buried wastes?
ATSDR has consulted with GEPA and the U.S. Army Corps of Engineers (USACE) regarding the concerns about unearthing buried wastes and has made recommendations, which were documented in ATSDR's initial health consultation in September 2000. The current status of each of ATSDR's recommendations is presented below, along with the corresponding contact information if additional information is requested.
- In the 2000 Health Consultation, ATSDR recommended that GEPA identify potential responsible parties for the burial sites to facilitate obtaining additional information for the purpose of public health evaluations.
GEPA is developing a request for proposal to identify "pre" Comprehensive Environmental Response, Compensation, and Liability Information System (CERCLIS) hazardous waste sites. These are previously undiscovered sites, or suspected areas of concern that were never listed in the CERCLIS database and that require confirmation of the presence of hazardous materials. GEPA intends to hold three public meetings in locations that represent the entire island of Guam (north, central, and southern portion). These meetings will give Guam's residents an opportunity to express concerns about any suspected sources of buried wastes that are not currently being addressed through the Superfund program. According to GEPA, a final list of buried waste sites, based on community involvement and archival research, is scheduled to be compiled by October 2004. As part of the process of addressing community concerns and cleaning up the buried wastes, GEPA will conduct searches and gather information to identify, to the extent possible, potentially responsible parties (Walter Leon Guerrero, GEPA, personal communication, May 19, 2003).
For additional information regarding this initiative: Contact GEPA at:
P.O. Box 22439 GMF
Barrigada, Guam 96921 U.S.A.
Telephone: (671) 475-1658/9
- In the 2000 health Consultation, ATSDR recommended that the USACE implement archival searches to identify units and commands using the area in and around the swamp, including the APP. Initial archival searches should be followed by targeted archival searches to determine possible disposal locations for waste, munitions, and other material of potential concern. All archival search information should be provided to GEPA and added to local archives including the University of Guam.
The USACE's Honolulu District has conducted and completed three archival and records searches on formerly used defense sites on Guam. They are Defense Environmental Restoration Program, Guam, by Bruce Karolle and Thomas McGrath, dated 1985; Site Inventory List for Potential Defense Environmental Restoration Project, Island of Guam, Mariana Islands, by Wilson Okamoto and Associates, dated 1992; and Fifth Field Depot, Mongmong, Guam GIS-Based Historical Photographic Analysis, by the Topographic Engineering Center, dated September 2000. At the request of former Congressman Robert Underwood (served in office from1992-2002), the Honolulu District took another comprehensive inventory of potential formerly used defense sites in Guam in March 2002. A General Accounting Office (GAO) report stated that in 2002, in response to congressional concerns, the Corps budgeted $500,000 for an island-wide archival search in Guam to identify formerly used defense sites with evidence of potential chemical warfare material (GAO 2002).
Former and retired military personnel may also contact the Department of Veterans Affairs-Veterans Health Administration (VHA) regarding health concerns from military-related occupational exposures. VHA provides a broad spectrum of medical, surgical, and rehabilitative care to its customers. To find out if you are eligible for benefits, how to apply, and what it will cost, complete an application form online at http://www.va.gov/health_benefits/. If you have a question or need additional help, you may call the VA Health Benefits Service Center toll-free at 1-877-222-VETS
Current civilian DOD employees not presently in a medical surveillance program should contact the medical treatment facility that has their medical records.
Former civilian DOD employees should consult with their private health care providers. They may request their occupational medical records that have been archived by their former military medical treatment facility. They may contact the Department of Labor's Workers' Compensation Commission for correct procedures to document potential exposures and to request compensation or medical evaluation and treatment. For former employees living on Guam, please contact:
Department of Labor
Workers' Compensation Commission
Government of Guam
Post Office Box 9970 Tamuning, Guam 96931-9970
Telephone: (671) 647-4205; fax: (671) 649-4922
For former employees living off island, an overview of state workers' compensation laws is available at the U.S. Department of Labor website, http://www.dol.gov/. A comprehensive directory of workers' compensation administrators for the United States and Canada can be found at http://www.comp.state.nc.us/ncic/pages/wcadmdir.htm.
According to Navy representatives, chlordane was detected in some of the soil removed from the Drainage Outfall Area, however, it does not appear to be related to APP. Chlordane-containing soils were limited to the area around an old typhoon damaged home on the western edge of the PCB removal area. Chlordane has been detected in water samples collected by GWA as part of routine monitoring of supply wells used for the community's drinking water. The highest concentration detected was 0.62 ppb in February 2001. This is below EPA's MCL of 2 ppb for chlordane (GWA 2003).
People typically receive the highest exposures of chlordane from living in homes that were treated with chlordane for termites. The most common source of low-level exposure, though, is chlordane-contaminated food (ATSDR 1994). If you know or suspect that chlordane was used in your home or on your property you can have your property tested to determine whether chlordane is still present and at what level. Tests are also available to measure chlordane in your blood, but they are expensive and not generally included as part of routine blood tests for physical examinations. There is no evidence that Mongmong residents' exposure to chlordane was any different than other residents on the island and ATSDR does not anticipate health problems from past routine use of chlordane for pest control. However, individual residences may have localized areas of chlordane higher than levels typically found around foundations of homes. ATSDR does not recommend routine testing for chlordane around homes unless a health care provider believes that an individual has been exposed to unusually high levels of chlordane.