PUBLIC HEALTH ASSESSMENT

AIR AND GROUNDWATER PATHWAYS ANALYSES
AGANA POWER PLANT
MONGMONG, GUAM

September 22, 2004

Evaluation of Environmental Contamination, Exposure Pathways, and Public Health Implications

To identify associated past and current public health hazards, ATSDR reviewed the environmental data from U.S. Navy reports at APP as well as information provided by various government of Guam agencies and research institutes. This section evaluates two exposure issues. (In addition, several issues are discussed in the "Community Concerns" section of this report.) For each of these exposures, information is provided on where the contamination was found, the extent of exposures at that location as compared to the extent of exposures found at other locations, and why ATSDR would or would not expect to find a health concern. Appendix C explains the process ATSDR uses to evaluate exposure, Appendix D explains ATSDR's methods and assumptions, and Appendix E includes ATSDR's glossary of terms frequently used throughout this report.

Breathing Contaminants From Stack Discharges From the Former APP

Guam residents asked ATSDR to evaluate whether past stack discharges from the former APP caused local air pollution to reach levels of health concern. ATSDR considered two approaches when evaluating this issue: we evaluated measurements of air pollution collected on Guam, and we assessed the available data for use in a computer model to estimate potential air quality impacts from the former APP's emissions. The available ambient air monitoring data suggest that sulfur dioxide and particulate levels in neighborhoods downwind from the former APP did not reach levels of health concern. However, specific information about power plant operations are insufficient for an evaluation of past emissions. The remainder of this section presents background information on air emissions from power plants and describes the two approaches we took to attempt to evaluate this issue.

Background Information on Air Emissions From Power Plants

During fuel combustion, the chemicals in fuel are broken down into much smaller, simpler chemicals. This process releases energy, which power plants attempt to harness. Complete combustion of fuels would result in air emissions that consist largely of water vapor, carbon dioxide, and simple inorganic compounds. However, the fuel combustion in boilers, engines, and furnaces is never complete, and the combustion process releases a wide range of byproducts to the air. These byproducts include dioxin and polycyclic aromatic hydrocarbons (PAHs).

EPA and other researchers have studied the amounts of chemicals that power plants typically release to the air. The composition and quantities of air emissions are determined largely by the amount, type, and grade of fuel burned, the engineering design of the power plant, and the air pollution controls in place (EPA 1998c). EPA has published data on typical emission rates from power plants for the following contaminants: particulate matter, sulfur oxides, nitrogen oxides, carbon monoxide, selected organic compounds, and trace elements. These emissions occur from combustion units that burn many types of fuels, including diesel fuel. According to EPA's data, pollutants that combustion units emit in largest quantities include particulate matter (see Text Box below), nitrogen oxides, and sulfur oxides (EPA 1998c). As the following discussion describes, ATSDR considered pollutants identified in EPA's fuel combustion studies when evaluating the nature and extent of stack discharges from the former APP.

Background Information on Particulate Matter

For nearly 20 years, EPA has closely monitored the levels of particulate matter in the air that people breathe. Particulate matter is airborne solid particles, liquid droplets, and aerosols. Many health studies have shown that the size of airborne particles is closely related to potential health effects among exposed populations. As a result, EPA and public health agencies tend to focus on the size of airborne particles when evaluating levels of air pollution. This public health assessment also classifies air concentrations of particulate matter by their size. In this document, particulate matter is classified into two categories:

Total suspended particulates (TSP) refers to a wide range of solid particles and liquid droplets found in ambient air, and typically is measured as particles having aerodynamic diameters of 25 to 40 microns or less. EPA's health-based National Ambient Air Quality Standards (NAAQS) regulated ambient air concentrations of TSP up to 1987. Those standards required annual concentrations of TSP to be less than 75 µg/m³ and 24-hour average concentrations to be less than 260 µg/m³. Many different industrial, commercial, mobile, and natural sources emit TSP to the air. TSP includes a wide mixture of particles, including large particles that generally do not enter the human lung.

Particulate matter smaller than 10 microns (PM₁₀) refers to the subset of TSP comprised of particles smaller than 10 microns in diameter. With research showing that PM₁₀ can penetrate into sensitive regions of the respiratory tract, EPA stopped regulating airborne levels of TSP in 1987, and began regulating ambient air concentrations of PM₁₀. EPA continues to regulate levels of PM₁₀ today. The NAAQS require annual average concentrations to be less than 50 µg/m³ and 24-hour average concentrations to be less than 150 µg/m³. Typical sources of PM₁₀ include, but are not limited to, windblown dust, grinding operations, and dusts generated by motor vehicles driving on roadways.

ATSDR notes that environmental agencies continue to research how different size fractions of particulate matter correlate with adverse health effects. For instance, EPA has also developed NAAQS for fine particulates smaller than 2.5 microns (PM2.5). However, no ambient air measurements for this subset of particulate matter have been collected at Guam

Operations and Emissions Data Specific to the Former APP

The former APP used 10 large piston engines to generate electricity. These engines reportedly burned diesel fuel. Information on the composition of fuel burned, amount of fuel burned, amount of power generated, size of population served, extent of air pollution controls, stack testing data, and air permitting status, could not be found by the Navy or Guam EPA. Without this information, ATSDR cannot estimate the amount of combustion byproducts (including dioxin) that the former APP emitted.

Estimated Air Concentrations Based on Computer Modeling

Although the previous review of monitoring data offers insights on potential air quality impacts from the former APP, the available monitoring data do not completely characterize past exposures: monitoring only occurred at a small number of discrete locations and was performed for only two pollutants (sulfur dioxide and particulates [i.e., TSP and PM₁₀]). These are common air contaminants that environmental regulatory agencies have measured over the past three decades. ATSDR is unable to conduct air modeling for more contaminants unless certain information about the power plant is available from the Navy (e.g., the composition of fuel burned, amount of fuel burned, extent of air pollution controls, and stack testing data). According to navy representatives, this information is not available.

Review of Ambient Air Monitoring Data

Ambient air monitoring refers to the measurement of the air that people may breathe. Ambient air monitoring typically occurs at fixed locations. This does not perfectly represent actual exposures, because people are mobile and move from location to location during the day. Ambient air monitoring is still a useful metric for evaluating inhalation exposures, though, because monitoring offers a direct measure of what is in the air and does not have the uncertainties of modeling analyses.

GEPA has conducted ambient air monitoring at various locations on Guam since the early 1970s. Figure 8 shows the locations where at least 1 year of consecutive air samples were collected on the island. ATSDR accessed more than 6,500 ambient air monitoring results from Guam that were reported to EPA between 1972 and 1992. GEPA has not conducted monitoring on the island since that time. ATSDR does not view the lack of monitoring data from more recent years as a data gap, because the emissions source of concern-the former APP-stopped its main operations in 1995. The majority of ambient air monitoring data collected at Guam was for sulfur dioxide and particulate matter. These are common air contaminants that environmental regulatory agencies across the country have measured over the past three decades. ATSDR reviewed trends among these monitoring data, given that power plants are known to release large quantities of both pollutants to air. Our review of the sulfur dioxide and particulate matter data follows:

Sulfur dioxide sampling data. When power plants burn fossil fuels, the sulfur that occurs naturally in the fuels is typically converted to sulfur dioxide, which is then released to the air. ATSDR reviewed nearly 2,500 sulfur dioxide monitoring results that GEPA had reported to EPA's AIRS (Aerometric Information Retrieval System) database. Between 1973 and 1988, GEPA collected sulfur dioxide samples from at least six locations (see Figure 8): one each in Agana, Anigua, Dededo, and Mangilao and two in Piti. The Agana and Anigua monitoring stations are approximately 1 mile and 2 miles, respectively from the former APP.

ATSDR used the monitoring results from these stations to characterize potential inhalation exposures to sulfur dioxide that the former APP released. The Anigua station sampled sulfur dioxide between 1974 and 1979, and the Agana station sampled sulfur dioxide between 1981 and 1988. More than 300 sampling results are available for each station. The highest annual average sulfur dioxide concentration for these stations was 12 µg/m³, observed in 1975. This concentration is much lower than 80 µg/m³, EPA's health-based ambient air quality standard for annual average sulfur dioxide levels. Similarly, the highest 24-hour average sulfur dioxide level reported for these stations was 164 µg/m³, which is also lower than EPA's corresponding health-based standard for 24-hour average measurements (365 µg/m³). The sampling data also indicate that sulfur dioxide levels in the Piti area were consistently and considerably higher than the sulfur dioxide levels observed in locations closer to the former APP-i.e., Anigua and Agana. In other words, the sampling data suggest that ambient air levels of sulfur dioxide in neighborhoods west of the former APP did not reach levels of health concern and that air concentrations of sulfur dioxide observed nearest to the former APP were not the highest levels observed on the island.

Although airborne sulfur dioxide may not have been a concern near the former APP, ATSDR has two concerns about the quality of the sulfur dioxide monitoring data that Guam agencies reported to AIRS. First, codes in the AIRS database suggest that some sulfur dioxide concentrations were measured using a "Hi-Vol" device; such a device is generally not used to measure concentrations of sulfur dioxide.² Second, the AIRS database indicates that the measurements were submitted in units of micrograms per cubic meter, while the standard reporting convention for sulfur dioxide measurements is to use parts per million or parts per billion. These concerns raise some questions about the validity of the sampling data. ATSDR cannot address these data quality concerns, because extensive documentation of the past sampling efforts is not available. It is worth noting, however, that agencies that conduct environmental sampling perform data quality checks before submitting their data to AIRS.

In summary, the available ambient air monitoring data suggest that sulfur dioxide levels in neighborhoods from the former APP did not reach levels of health concern; however, the data used to reach this conclusion are of unknown quality. The data trends do indicate that other areas on Guam had considerably higher sulfur dioxide levels than those recorded for neighborhoods downwind from the former APP.

Particulate sampling data. When generating electricity, power plants emit particulate matter-airborne particles and water droplets. These are typically emitted through stacks, and the particles can remain airborne for long periods of time. There are many other sources of particulate matter at Guam, such as motor vehicles, fires, construction activities, and wind-blown dust. Past ambient air monitoring activities measured the total levels of airborne particulate matter at Guam resulting from all of these sources combined.

From 1973 to 1988, GEPA measured ambient air concentrations of "total suspended particulates" (TSP). This is the air pollutant that most environmental regulatory agencies considered in the 1970s and 1980s when focusing on airborne particles. TSP includes a wide range of particle sizes, including some that are too large to enter the human lung. From 1989 to 1991, GEPA measured concentrations of "particulate matter smaller than 10 microns" (PM₁₀). PM₁₀ is the subset of TSP that is most likely to enter the lungs of humans. ATSDR reviewed the TSP and PM₁₀ data to evaluate potential exposures to airborne particles.

At least six monitoring stations at Guam measured ambient air concentrations of TSP between 1973 and 1988 (see Figure 8). From 1973 to 1978, the monitoring stations located in Piti (Cabras Island) and Anigua both recorded geometric mean TSP concentrations higher than EPA's corresponding former health-based air quality standard (75 µg/m³). The Anigua station was located at an auto refinishing shop approximately 2 miles downwind from the former APP; the Piti station was located in an industrial area approximately 5 miles from the former APP. The primary causes for these elevated TSP concentrations are not clear. Emissions from APP likely contributed to the elevated levels, but many other sources undoubtedly contributed as well. These include possible releases from the auto refinishing shop, dusts from roads, and motor vehicle exhaust. Whether the elevated TSP levels present a health concern depends on the sizes of the particles that were sampled, but no sampling data are available from Guam to characterize the particle size distribution.

From 1979 through 1988, on the other hand, all annual average TSP levels observed for the stations closest to the former APP met EPA's former health-based standards. ATSDR also reviewed the PM₁₀ sampling that occurred between 1989 and 1991. All of the PM₁₀ sampling results were safely below EPA's health-based standards, both for short-term exposures and long-term exposures. (Refer to our analysis of dusts from roads in the "Community Concerns" section for a more detailed review of the PM₁₀ sampling data.)

Taken together, the TSP and PM₁₀ sampling data indicate that airborne particulate matter levels at locations west of the former APP likely have not exceeded health-based standards over the last 25 years. Between 1973 and 1978, however, elevated particulate matter levels were observed at multiple places around the island. The available information does not tell us what emissions sources contributed most to the elevated particulate matter levels, nor does it tell us why particulate matter levels in Anigua and Piti decreased considerably after 1978. Furthermore, without information on the particle sizes, ATSDR cannot definitively tell whether the past particulate matter levels were of public health concern. Thus, ATSDR concludes that inhalation exposures to particulate matter at locations downwind from the former APP appear to be greatest in years up to 1978, but the public health implications of these exposures cannot be determined.

Estimated Air Concentrations Based on Computer Modeling

Although the previous review of monitoring data offers insights on potential air quality impacts from the former APP, the available monitoring data do not completely characterize past exposures: monitoring only occurred at a small number of discrete locations and was performed for only two pollutants (sulfur dioxide and particulates [i.e., TSP and PM₁₀]). These are common air contaminants that environmental regulatory agencies have measured over the past three decades. As stated previously, ATSDR is unable to estimate air emission rates for various other combustion byproducts (e.g., dioxin) or conduct air modeling unless certain information about the power plant is available (e.g., the type of fuel burned, amount of fuel burned, extent of air pollution controls, and stack testing data). According to navy representatives, this information is not available.

Groundwater and Drinking Water

ATSDR evaluated groundwater contamination at APP and any potential for contaminants to migrate to the principle aquifer used for drinking water. PCBs and other site-related contaminants were not detected above ATSDR's screening level. ATSDR continues to evaluate additional information pertaining to groundwater underneath APP with characterization still ongoing. On the basis of GEPA's recommendation, the Navy has agreed to install 2 additional monitoring wells down gradient of APP, in the direction of groundwater flow, to better measure any past or current migration of site-related contaminants.

Early environmental investigations at APP identified several spills associated with above-ground storage tanks (ASTs) ranging between 50 and 100 gallons of waste oil, contamination originating from two dry wells (east and west) used for the disposal of waste oils originating from the generator building (Building 1200) at APP, and various other sources of contamination from other small releases during APP's operation. A groundwater investigation was conducted at APP in response to the presence of waste oils and total petroleum hydrocarbons (TPHs) at the two dry wells, at underground storage tank (UST) PWC 38 near POI 1, and at POI 6.

As part of the RI for APP, The Navy installed four onsite groundwater monitoring wells and conducted two rounds of sampling (one in October 2000, the wet season, and the other in April 2001, the dry season). During the groundwater investigation, the Navy calculated the groundwater flow direction to be northwest toward Agana Bay and the Philippine Sea. Recent groundwater measurements from existing monitoring wells indicate that groundwater flow is northwest beneath Agana Power Plant.. The Navy did not install any offsite monitoring wells (Earth Tech 2002). The four onsite monitoring wells were installed at the following locations:

GW-MW01-adjacent to the East Drywell near the southeast corner of Building 1200
GW-MW02-adjacent to the West Drywell near the southwest corner of Building 1200
GW-MW03-between POI-7 and POI-1, approximately 65 feet north of Building 1200
GW-MW04-at the southwest corner of POI-6 (West), approximately 215 feet west of Building 1200

The Navy analyzed a total of 10 water samples (4 samples collected during October 2000; 4 samples collected during April 2001, and 2 duplicate samples) for metals, volatile organic compounds (VOCs), semi-volatile organic compounds (SVOCs), PCBs, and TPH. PCBs were not detected in any of the monitoring wells during the two sampling rounds (Earth Tech 2002). All other site-related contaminants analyzed were either not detected or detected at levels below ATSDR's screening level.

The Navy installed monitoring wells at locations where TPHs in soil or waste oils in dry wells had been identified during previous investigations. The samples collected from these monitoring wells would presumably yield higher contaminant concentrations than samples collected at other locations at APP where no soil contaminants were identified. Therefore, on the basis of the results from the Navy's groundwater investigation, ATSDR concludes that groundwater underneath APP poses no apparent public health hazard.

Groundwater measurements from existing monitoring wells indicate that groundwater flow is northwest beneath Agana Power Plant, contrary to the original idea that the groundwater flowed towards Agana Swamp. Consequently, the Navy has agreed to install 2 additional monitoring wells down gradient of APP, in the direction of apparent groundwater flow, to better measure any past or current migration of site-related contaminants. Sampling data from additional down gradient monitoring wells will help the Navy and GEPA identify whether any existing groundwater contamination is present and evaluate the need for future indoor air evaluations in the unlikely event of detection of elevated levels of soil gas.

Concerns About the Quality of Drinking Water in the Mongmong Area

People who live in the Mongmong area of Guam and obtain drinking water from nearby groundwater sources have not been exposed to chemical contaminants at levels that would cause harm.

What Is a Sole Source Aquifer?

Under the federal Safe Drinking Water Act, the Administrator of the EPA may determine that an underground water supply is the sole or principal source of drinking water for an area which, if contaminated, would create a significant hazard to public health.

EPA defines a sole or principal source aquifer as one that supplies at least 50% of the drinking water consumed in the area overlying the aquifer. These areas can have no alternative drinking water source(s) that could supply all those who depend upon the aquifer for drinking water. For convenience, all designated sole or principal source aquifers are referred to as "sole source aquifers."

Sole Source Aquifer designations are one tool to protect drinking water supplies in areas with few or no alternative sources to the groundwater resource, and where if contamination occurred, using an alternative source would be extremely expensive. The designation protects an area's groundwater resource by requiring EPA to review any proposed projects within the designated area that are receiving federal financial assistance. All proposed projects receiving federal funds are subject to review to ensure that they do not endanger the water source (EPA 2003a).Guam's primary source of drinking water is from groundwater contained in the aquifer beneath the northern half of the island. This aquifer, referred to as the Northern Guam Lens, was designated a "principal (i.e., sole) source aquifer" in 1978 by EPA, and is the primary source of potable groundwater on the island (NRCS 1998; GEPA 1998). Groundwater is pumped from this aquifer into the water distribution system by approximately 105 production wells. According to GWA, about 26 million gallons per day are pumped from the northern lens for potable use (U.S. Department of the Interior 1999; GWA 2000).

Residents in the Mongmong area obtain their drinking water and water for domestic purposes (e.g., cooking and bathing) primarily from municipal supply wells in the general vicinity of Mongmong. According to GEPA, there are no current private well users in the Mongmong area. Other than past use of the Agana Springs, ATSDR has also not identified any other current sources of drinking water in this area (e.g., surface water or springs). Some of the GWA's production wells are located within a 1 mile radius of APP. ATSDR reviewed groundwater data collected by the Navy and drinking water monitoring data provided by GWA to determine whether contaminants related to APP were impacting the drinking water supply.

What was sampled?

Navy Monitoring Data

See previous "Groundwater and Drinking Water" discussion for a summary of the Navy's groundwater monitoring investigation.

Guam Waterworks Authority (GWA) Monitoring Data

There are seven GWA drinking water production wells within 1 mile of APP. Six of these (wells A-5, A-23, A-25, A-29, A-30, and A-31) are between 3/4 and 1 mile southwest of APP. The other well (A-26) is approximately 1/3 mile east-northeast of APP (see Figure 5). These wells are installed at depths ranging between 82 and 204 feet bgs. According to the Navy's limited groundwater flow information (again, groundwater flow was estimated to be in a northwest direction beneath APP), these drinking water supply wells are generally upgradient of APP and would not be directly impacted by any contaminated groundwater identified beneath the site (Earth Tech 2002).

Bacteriological contamination from leaking septic tanks, sewer lines, livestock, and other sources has been detected in some wells during routine monitoring. However, according to GWA, all drinking water supplied by GWA is treated with chlorine and this should eliminate most hazards associated with bacteriological contamination. People should contact GWA or GEPA for information pertaining to sampling and groundwater evaluation programs in the Mongmong area. The water from most of the supply wells in the Mongmong area is routed directly into the distribution system to residents and other users. According to GWA, the drinking water from the supply wells is generally not blended or routed to a large holding tank before being distributed. Any specific questions about drinking water should be directed to GWA (Carmen Sian-Denton, Monitoring Laboratory Administrator, GWA, personal communications, May 13, 2003).

In compliance with the Guam Primary Safe Drinking Water Regulations and the Federal Safe Drinking Water Act, all drinking water from GWA's distribution network is routinely monitored for regulated and unregulated contaminants at the treatment plants and well heads, as it enters into the distribution system (GWA 2000). ATSDR reviewed several years of GWA-provided drinking water monitoring data (1996-2001) for supply wells within 1 mile of APP. ATSDR's review indicated that the drinking water during this period met all Guam and federal safe drinking water standards (SDWSs) for chemical contaminants (GWA 2003).

A small number of water samples contained low concentrations of VOCs (i.e., trichloroethylene [TCE] and tetrachloroethylene [PCE]) and pesticides (i.e., chlordane, dieldrin, and lindane) in some supply wells near APP. None of those chemicals exceeded their maximum contaminant levels (MCLs) (see Appendix E for explanation of MCL) for drinking water. (A summary of the VOCs and pesticides identified during routine drinking water monitoring is presented in Table 1). The source of the VOCs and pesticides is not known. TCE and PCE have been detected in groundwater underneath NAS Agana (i.e., Tiyan) and it is likely that other potential sources of contamination exist around Mongmong. PCBs, detected both onsite at APP and offsite in surface and subsurface soil samples collected by the Navy, were not detected in any of the drinking water supply wells within 1 mile of APP (GWA 2003).

ATSDR evaluated the potential exposure to contaminants in drinking water for residents living in the Mongmong area. The only current source of drinking water identified in the Mongmong area is groundwater (see the next section regarding past use of water from Agana Springs). Although low concentrations of some VOCs and pesticides have been detected during routine water quality monitoring, none of these contaminants have exceeded Guam and federal SDWSs (Table 1). ATSDR concludes that municipal drinking water supplies do not contain chemical contaminants at levels that would pose a health concern. .

Table 1. Summary of VOCs, PCBs1, and Pesticides Detected in Drinking Water From GWA Supply Wells Within 1 Mile of APP
Contaminant	Maximum Concentration (ppb)	Supply Well Number	Date Sampled	MCL or ATSDR CV (ppb)
VOCs
PCE	3.6	A-29	1/06/97	5 (MCL)
TCE	1.7	A-30	1/08/01	5 (MCL)
Pesticides
Chlordane	0.6	A-25	2/13/01	2 (MCL)
Dieldrin	0.2	A-25	9/23/01	2 (Chronic EMEG)
Lindane	0.02	A-25	2/13/01	0.2 (MCL)
Source: Guam Waterworks Authority (GWA) 2003. CV = comparison value; EMEG = ATSDR's environmental media evaluation guide; MCL = EPA's maximum contaminant level for drinking water; PCE = Tetrachloroethylene; ppb = parts per billion TCE = Trichloroethylene; VOC = volatile organic compound ¹ PCBs were not detected in any samples

Was it safe to drink water from Agana Springs when it was used by GWA as a municipal water source?

The Agana Springs, located in the Agana River Valley and directly north of the village of Sinajana, is one of the largest freshwater springs on the island. The Agana Springs was used as a municipal water source from 1937 to 1957, with 3.8 to 9.5 million liters (1 to 2.5 million gallons) of water pumped daily. Agana Springs was abandoned in 1957, primarily because water samples had routinely indicated the presence of bacteriological contamination. This contamination was believed to have originated from the village of Sinajana, which at the time did not have a centralized sewage collection and treatment system in place (Smalley and Zolan 1981;UNEP 2001).

No chemical monitoring data are available for the period when Agana Springs was used as a drinking water source. However, ATSDR reviewed water sampling data collected by WERI between March and September 1980. WERI conducted the sampling and analysis of Agana Spring water in an effort to determine the water's quality as it relates to possible development as a municipal water supply. The sampling analyses included metals and bacteriological monitoring. Concentrations of heavy metals (e.g., lead, mercury, cadmium) met all Guam and federal safe drinking water standards-in most cases, heavy metals were not detected.

Although monitoring for chemical contaminants (e.g., metals, PCBs, and VOCs) had not been conducted for Agana Springs in the past, the absence of heavy metals in the 1980 samples suggests that metals from past industrial sources had not significantly impacted the quality of this drinking water source. Since APP has been in operation since the late 1940's, PCB contamination in Agana Springs during a portion of the time it was used as a drinking water supply cannot be ruled out. However, the persistent chlorinated compounds such as PCBs and many pesticides are not readily soluble in water and, if present, would have been more likely to be detected at higher levels in the sediments. It's very unlikely that these chemicals would be present in high enough concentrations in drinking water supplies to be harmful.

The WERI report concluded that "the chemical and physical quality of Agana Springs water is high,"(i.e., good quality) relative to standards set by GEPA for drinking water, and recommended that the government of Guam utilize Agana Springs as an additional drinking water resource for the island (Smalley and Zolan 1981). Without monitoring data for the period when Agana Springs was used as a source of drinking water, ATSDR cannot make a definitive statement about how safe the water was to drink; however, WERI's conclusions lead ATSDR to believe that the Agana Springs water did not contain chemical contaminants that would have posed a past public health hazard. Although PCBs may have been detected at low levels in the soils and sediment around Agana Springs, ATSDR considers it unlikely that chemical contaminants in water from Agana Springs would have been detected in the past at levels known to be harmful. ATSDR classifies past exposures to Agana Springs water as no apparent public health hazard.

In 1989, the Public Utility Agency of Guam installed two new water wells near the Agana Springs: A-29 and A-30. These two new groundwater wells have yielded an additional 5.7 million liters per day. Although they do not draw directly from surface water from Agana Springs, the groundwater where the wells are screened is likely hydrologically connected with the Springs. According to GEPA, these wells continue to supply drinking water and are monitored routinely. Bacteriological contamination has been detected in these wells and GEPA continues to look for potential sources of bacteriological contamination (Angel Marquez, GEPA, Safe Drinking Water Program, personal communication, May 14, 2003).

Do the injection wells northwest of Agana Swamp (Ponding Basin), in close proximity to GWA drinking water supply wells, threaten the quality of the drinking water supply?

EPA's Underground Injection Control program is responsible for regulatory oversight of injection wells that have the potential to impact an underground source of drinking water. In the health consultation "Mongmong and Agana Power Plant, Guam" released in September 2000, ATSDR recommended that the government of Guam, in consultation with the Guam Water Quality Planning Committee, re-evaluate the water injection control program for its impact on drinking water. The injection wells of interest in the Mongmong area are dry wells 1 through 8, referred to as DW-1 through DW-8. These wells, operated by Guam Department of Public Works and not associated with APP operations, are defined by EPA as Class V injection wells. By definition, Class V wells inject non-hazardous liquid (e.g., storm water runoff, sewage treatment effluent, and agricultural waste) into or above an underground source of drinking water. These types of injection wells are very common in the United States mainland: there are about 500,000 to 685,000 Class V injection wells across the country.

According to GEPA, wells DW-1 through DW-8 are used to control stormwater runoff in the Mongmong area. These eight injection wells have been installed at depths ranging between 250 and 300 feet bgs and are routinely monitored semi-annually (in the dry and rainy seasons) for 20 chemical drinking water quality parameters. The methods of analysis conform to safe drinking water standards. The water sampled from these wells have generally met drinking water standards for chemical contaminants. In the past, there has been record of some substances exceeding the action level (e.g., lead). These wells do occasionally contain bacteriological contamination that exceeds Guam's SDWSs. When this occurs, advisories to boil water are put in place. It is possible that bacteriological contamination from these injection wells is impacting nearby drinking water supply wells (i.e., A23, A25, and A31). GEPA, in cooperation with the Guam Department of Public Works, is evaluating this potential link (Angel Marquez, GEPA, Safe Drinking Water Program, personal communication, May 14, 2003; Susan Marquez, GEPA, personal communication, May 28, 2003).

² Approximately one-third of the sulfur dioxide monitoring results that GEPA reported to AIRS are labeled as being measured using a "Hi-Vol" device, and are therefore of questionable quality. This reporting practice occurred only for monitoring that occurred during the 1970s. All monitoring results reported for the 1980s appear to be correctly coded. Regardless of whether the questionable results are included or excluded from this analysis, all of the measured sulfur dioxide concentrations fall safely below EPA's health-based air quality standards.

Next Section Table of Contents

Agency for Toxic Substances and Disease Registry, 1825 Century Blvd, Atlanta, GA 30345
Contact CDC: 800-232-4636 / TTY: 888-232-6348

Department of Health and Human Services

PUBLIC HEALTH ASSESSMENT

AIR AND GROUNDWATER PATHWAYS ANALYSES AGANA POWER PLANT MONGMONG, GUAM

Evaluation of Environmental Contamination, Exposure Pathways, and Public Health Implications

AIR AND GROUNDWATER PATHWAYS ANALYSES
AGANA POWER PLANT
MONGMONG, GUAM