PETITIONED PUBLIC HEALTH ASSESSMENT

BOVONI DUMP
ST. THOMAS, U.S. VIRGIN ISLANDS

APPENDIX A - Figures

Figure 1: Eastern St. Thomas -- Location of Bovoni Landfill
Figure 2: Demographic Statistics Within One Mile of the Bovoni Landfill
Figure 3: Air Monitoring Locations -- 1995
Figure 4: Groundwater Monitoring Well Locations and Air Sampling Locations -- 1996

Figure 1: Eastern St. Thomas -- Location of Bovoni Landfill

Figure 2: Demographic Statistics Within One Mile of the Bovoni Landfill

Figure 3: Air Monitoring Locations -- 1995

Figure 4: Groundwater Monitoring Well Locations and Air Sampling Locations -- 1996

APPENDIX B - Tables

Table 1: Completed Exposure Pathways

Table 2: Potential Exposure Pathways

Table 3: Groundwater Monitoring Well Data

Table 4: Air Monitoring Data -- 1995

Table 5: Air Monitoring Data -- 1996

Table 6: Sediment Data for the Mangrove Lagoon and Benner Bay

Table 1:

COMPLETED EXPOSURE PATHWAYS

Pathway Name	Source	EnvironmentalMedium	Point of Exposure	Route of Exposure	Exposed Pop.	Time
Air	Bovoni	Air	Onsite Work Areas	Inhalation	Workers	Past, Current, Future
Air	Bovoni	Air	OffsiteResidential Areas	Inhalation	Area Residents	Past, Current, Future

Table 2:

POTENTIAL EXPOSURE PATHWAYS

Pathway Name	Source	Environmental Medium	Point of Exposure	Route of Exposure	Exposed Pop.	Time
Groundwater	Bovoni	Groundwater	Offsite PrivateWells	Inhalation, Ingestion, Dermal	Residents usingPrivate Wells	Past, Current, Future
Cistern Water	Bovoni	Cistern Water	Offsite Cisterns	Inhalation, Ingestion, Dermal	Residents usingCisterns	Past, Current, Future
Sediment	Bovoni	Sediment	Mangrove Lagoon	Dermal, Incidental Ingestion	Recreationalusers of theMangrove Lagoon	Past, Current, Future

Table 3:

Groundwater Monitoring Well Data

Contaminant	Concentration Range ppb Sampling Year-1982/83	Concentration Range ppb Sampling Year-1996	Comparison Value*
			Value ppb	Source
Arsenic	3-4	<10	3	Chronic EMEG (child)
			10	Chronic EMEG (adult)
			50	MCL
Cadmium	7 - 132	<5	7	Chronic EMEG (child)
			10	Chronic EMEG (adult)
			5	LTHA / MCL
Chromium	8 - 200	17	100	LTHA / MCL
Lead	80 - 400	<15	15	EPA Action Level
Mercury	0.2 - 2	<0.2	2	LTHA / MCL
Silver	2 - 100	<10	50	RMEG (child)
			200	RMEG (adult)
			100	LTHA
Fluoride-F	170 - 700	--	None
Nitrate-N	60 - 67,000	--	None
Gross Beta	2.21 - 11.72 (pCi/L)	--	None
Gross Alpha	1.09 - 4.2 (pCi/L)	--	None

* The drinking water comparison values shown are based on chronic, non-cancer effects only, and assumeconsumption of 1-2 liters of water per day. (See Appendix C.) However, according to the information available toATSDR, no one is drinking the groundwater sampled by these monitoring wells.

Source: Reference 3 and Reference 8

Table Symbols:

ppb pCi/L < --	parts per billion pico Curies per Liter below detection limits not reported

Table 4:

Air Monitoring Data -- 1995

Contaminant	Concentration Range ug/m3	Location of Maximum Concentration	Comparison Valuea
			Value ug/m3	Source
1,2-Dichlorobenzene	10 - 50	Smolder Pit	150	RBC
1,2,3-Trichloropropane	110	Burn Pit	1.8 (0.3 ppb)	Acute EMEG
4-Methylstyrene	10 (2 detections)	Upwind	None
a-Methylstyrene	30 - 40	Smolder Pit	260	RBC
Benzene	20 - 670	Smolder Pit	160 (50 ppb)	Acute EMEG
Benzyl Chloride	20 - 140	Downwind	5,200	TLV
Bromochloromethane	20 - 1,500	Smolder Pit	None
Carbon Tetrachloride	20 - 50	Smolder Pit	315 (50 ppb)	Intermediate EMEG
Chlorobenzene	6 - 20	Upwind	21	RBC
Chloroform	10 - 80	Smolder Pit	244 (50 ppb)	Intermediate EMEG
			4,883 (1,000 ppb)	Acute EMEG
Cumene	10 - 30	Smolder Pit	246,000	TLV-TWA b
Hexachloroethane	7 - 120	Smolder Pit	871 (90 ppb)	Intermediate EMEG
			4,840 500 ppb)	Acute EMEG
t-Butyltoluene	10 (3 detections)	Upwind	6,100	TLV
Tetrachloroethylene	20 - 30	Smolder Pit	1,356 (200 ppb)	Acute EMEG
Acenaphthylene	0.261 - 7.86	Smolder Pit	220	RBC
Anthracene	0.0104 - 0.464	Smolder Pit	1,100	RBC
Chrysene	0.0261 - 0.393	Smolder Pit	None
Fluoranthene	0.0261 - 0.416	Smolder Pit	150	RBC
Fluorene	0.0521 - 4.64	Smolder Pit	150	RBC
Phenanthrene	0.0307 - 3.14	Smolder Pit	None
Pyrene	0.0521 - 0.295	Smolder Pit	110	RBC
Benzo(a)anthracene	0.0261 - 0.393	Smolder Pit	None
Benzo(b)fluoranthene	0.0104 - 0.157	Smolder Pit	None
Benzo(k)fluoranthene	0.0104 - 0.157	Smolder Pit	None
Benzo(a)pyrene	0.0261 - 0.393	Smolder Pit	None
Dibenzo(a,h)anthracene	0.104 - 1.57	Smolder Pit	None
Benzo(g,h,i)perylene	0.0417 - 0.629	Smolder Pit	None
Indeno(1,2,3-cd)pyrene	0.0261 - 0.393	Smolder Pit	None
Aldrin	0.0327 - 0.0662	Smolder Pit	250	TLV
Acetaldehyde	1 - 10	Burn Pit	9	RfC
Acetophenone	2 - 6	Burn Pit	49,000	TLV-TWA c
Acrolein	6	Burn Pit	0.02 (0.009 ppb)	Intermediate EMEG
Benzaldehyde	1 - 20	Burn Pit	370	RBC
Crotonal	1 - 5	Burn Pit	None
Formaldehyde	5 - 70	Burn Pit	300	TLV-STEL
Propional	1 - 10	Burn Pit	None
Phosgene	0.01 - 0.75 ppm	Smolder Pit	400	TLV
1,1-Dichloroethane	5.3 (1.3 ppbv)	Burn Pit	520	RBC
1,2-Dichloroethane	40 - 170	Downwind	810 (200 ppb)	Acute EMEG
1,1,2-Trichloroethane	7 - 260	Upwind	55,000	TLV
Trichlorofluoromethane (F-11)	3.7 - 53 (0.7 - 9.8 ppbv)	Smolder Pit	730	RBC
Trichlorotrifluoroethane (F-113)	4.6 (0.60 ppbv)	Downwind	31,000	RBC
Aluminum	10	Burn Pit	3,700	RBC
Arsenic	7.3	Burn Pit	1.1	RBC
Calcium	14.8 - 72.3	Burn Pit	None
Iron	19.4	Burn Pit	1,100	RBC
Magnesium	10	Burn Pit	None
Mercury Vapor	1 - 121	Burn Pit	0.02	Acute EMEG
Nickel	12 - 14.6	Metal Baling	0.1	Intermediate EMEG
Sodium	9.5 - 23.2	Burn Pit	None

^a The comparison values shown are the most relevant ones among those available and do not include any based oncancer effects only. See explanation in the Toxicological Evaluation section.

^b The only other comparison value available for cumene is an RBC based on an inhalation reference dose ofuncertain origin. (EPA 's RfC for cumene has been under review since 1989.) In addition, cumene is actually lesstoxic than benzene, not more so as the RBC would suggest.

^c The only other comparison value available for acetophenone is an RBC based on an inhalation reference dose ofuncertain origin. (EPA 's RfC for acetophenone has been under review since 1991.) In addition, the RBC inquestion is actually below the odor threshold of acetophenone, a chemical which is used to impart an orangeblossom-like odor to perfumes and which does not constitute a health hazard by inhalation.

Source: Reference 1

Table Symbols:

ug/m³ micrograms/cubic meter
ppb parts per billion
ppbv parts per billion volume

Conversion factor for air:

C_ug/m3 = C_ppb X MW_g/mole/24.45

where
C = Concentration
MW = Molecular Weight

Table 5.

Air Monitoring Data -- 1996

Contaminant	Concentration Range ppbv	Location ofMaximumConcentration	Comparison Valuea
			Value ppb	Source
Benzene	294 - 2500	Vent	50	Acute EMEG
Bromomethane	1.10 - 10.9	Vent	50	Intermediate EMEG
			50	Acute EMEG
Carbon Tetrachloride	2.89	Vent	50	Intermediate EMEG
			200	Acute EMEG
Chlorobenzene	20.6 - 122	Vent	6.7 (31 ug/m3)	RBC
Chloroform	3.68 - 34.0	Vent	50	Intermediate EMEG
			1,000	Acute EMEG
1,2-Dichlorobenzene	3.94 - 29.3	Vent	25 (150 ug/m3)	RBC
1,3-Dichlorobenzene	3.26 - 24.7	Vent	53 (320 ug/m3)	RBC
Dichlorodifluoromethane	0.53 - 1.08	Vent	42.5 (210 ug/m3)	RBC
Ethylbenzene	1.31 - 391	Vent	300	Intermediate EMEG
Methylene Chloride	0.68 - 8.59	Upwind	30	Intermediate EMEG
			400	Acute EMEG
1,1,2,2-Tetrachloroethane	7.33 - 39.8	Vent	400	Intermediate EMEG
			1000	Acute EMEG
Tetrachloroethene	4.03	Vent	200	Acute EMEG
1,2,4-Trichlorobenzene	2.03 - 13.7	Vent	28.3 (210 ug/m3)	RBC
Trichloroethene	0.78 - 7.38	Vent	100	Intermediate EMEG
			2000	Acute EMEG
1,2,4-Trimethylbenzene	4.17 - 24.1	Vent	36.6 (180 ug/m3)	RBC
1,3,5-Trimethylbenzene	4.40 - 25.0	Vent	36.6 (180 ug/m3)	RBC
M,P Xylene	6.99 - 102	Vent	300	M-Xylene Acute EMEG
			40	P-Xylene Acute EMEG

^a The comparison values shown are the most relevant of those available and do not include any based on cancereffects only. See explanation in the Toxicological Evaluation section.

Source: Reference 8

Table Symbols:

ug/m³ micrograms per cubic meter
ppb parts per billion
ppbv parts per billion volume

Conversion factor for air:

C_ug/m3 = C_ppb X MW_g/mole/24.45
where
C = Concentration
MW = Molecular Weight

Table 6:

Sediment Data for the Mangrove Lagoon and Benner Bay

Contaminant	Concentration Range ppm	Comparison Valuea
		Value ppm	Source
Cadmium	0.14 - 2.7	40	Chronic EMEG (child)
		500	Chronic EMEG (adult)
Copper	7.6 - 140	82,000	RBC (industrial soil)
		31,000	RBC (residential soil)
Lead	2 - 110	400	EPA Office of Solid Waste
Mercury	0.0002 - 0.23	None
Oil & Grease	92 - 3,600	None
Total Phosphorus	2 - 320	1	RMEG (child)
		10	RMEG (adult)
Total Nitrogen	280 - 4,400	None

^a The soil comparison values shown assume ingestion of 100-200 mg soil per day, and are not strictly applicable tosediment. There are no comparison values specifically designed for sediment exposure.

Source: Reference 4

Table Symbols:

ppm     parts per million

APPENDIX C - Comparison Values

ATSDR comparison values are media-specific concentrations that are considered to besafe under default conditions of exposure. They are used as screening values in thepreliminary identification of site-specific "contaminants of concern". The latter termshould not be misinterpreted as an implication of "hazard". As ATSDR uses thephrase, a "contaminant of concern" is merely a chemical substance detected at the sitein question and selected by the health assessor for further evaluation of potentialhealth effects. Generally, a chemical is selected as a "contaminant of concern"because its maximum concentration in air, water, or soil at the site exceeds one ofATSDR's comparison values.

However, it must be emphasized that comparison values are not thresholds of toxicity. While concentrations at or below the relevant comparison value may reasonably beconsidered safe, it does not automatically follow that any environmental concentrationthat exceeds a comparison value would be expected to produce adverse health effects. The whole purpose behind highly conservative, health-based standards and guidelinesis to enable health professionals to recognize and resolve potential public healthhazards before they can become actual public health consequences. Thus,comparison values are designed to be preventive, rather than predictive, of adversehealth effects. The probability that such effects will actually occur depends, not onenvironmental concentrations alone, but on a unique combination of site-specificconditions and individual lifestyle and genetic factors that affect the route, magnitude,and duration of actual exposure.

Listed and described below are the various comparison values that ATSDR uses toselect chemicals for further evaluation, as well as other non-ATSDR values that aresometimes used to put environmental concentrations into a meaningful frame ofreference. Also listed below are the abbreviations for some of the more common units of measure.

CREG	=	Cancer Risk Evaluation Guides
MRL	=	Minimal Risk Level
EMEG	=	Environmental Media Evaluation Guides
IEMEG	=	Intermediate Environmental Media Evaluation Guides
RMEG	=	Reference Dose Media Evaluation Guide
RfD	=	Reference Dose
RfC	=	Reference Dose Concentration
RBC	=	Risk-Based Concentration
DWEL	=	Drinking Water Equivalent Level
LTHA	=	Lifetime Health Advisory
MCL	=	Maximum Contaminant Level
PRG	=	Permissible Remediation Goal (Action Level)
PEL	=	Permissible Exposure Limit
TLV	=	Threshold Limit Value
ppm	=	parts per million (mg/L water or mg/kg soil)
ppb	=	parts per billion (ug/L water or ug/kg soil)
kg	=	kilogram (1,000 grams)
mg	=	milligram (0.001 grams)
ug	=	microgram (0.000001 grams)
L	=	liter
m3	=	cubic meter, referring to 1,000 liters of air)

Cancer Risk Evaluation Guides (CREGs) are estimated contaminant concentrationsexpected to cause no more than one excess cancer in a million persons exposed overa lifetime. CREGs are calculated from EPA's cancer slope factors or cancer potencyfactors using default values for exposure rates. However, neither CREGs nor CSFscan be used to make realistic predictions of cancer risk. The true risk is alwaysunknown and may be as low as zero.

Minimal Risk Levels (MRL) are estimates of daily human exposure to a chemical (i.e.,doses expressed in mg/kg/day) that are unlikely to be associated with any appreciablerisk of deleterious noncancer effects over a specified duration of exposure. MRLs arecalculated using data from human and animal studies and are reported for acute ( 14days), intermediate (15-364 days), and chronic ( 365 days) exposures. MRLs arepublished in ATSDR Toxicological Profiles for specific chemicals.

Environmental Media Evaluation Guides (EMEGs) are concentrations that arecalculated from ATSDR minimal risk levels by factoring in default body weights andingestion rates.

Intermediate Environmental Media Evaluation Guides (IEMEG) are calculated fromATSDR minimal risk levels; they factor in body weight and ingestion rates forintermediate exposures (i.e., those occurring for more than 14 days and less than 1year).

Reference Dose Media Evaluation Guide (RMEG) is the concentration of acontaminant in air, water or soil that corresponds to EPA's RfD for that contaminantwhen default values for body weight and intake rates are taken into account.

EPA's Reference Dose (RfD) is an estimate of the daily exposure to a contaminantunlikely to cause noncarcinogenic adverse health effects. Like ATSDR's MRL, EPA'sRfD is a dose expressed in mg/kg/day.

Reference Concentrations (RfC) is a concentration of a substance in air which EPAconsiders unlikely to cause non-cancer adverse health effects over a lifetime of chronic exposure.

Risk-Based Concentrations (RBC) are media-specific concentrations derived byRegion III of the Environmental Protection Agency Region III from RfDs, RfC's, orEPA's cancer slope factors. They represent concentrations of a contaminant in tapwater, ambient air, fish, or soil (industrial or residential) that are considered unlikely tocause adverse health effects over a lifetime of chronic exposure.

Drinking Water Equivalent Levels (DWEL) are based on EPA's oral RfD andrepresent corresponding concentrations of a substance in drinking water that areestimated to have negligible deleterious effects in humans at an intake rate of 2 L/dayfor life, assuming that drinking water is the sole source of exposure.

Lifetime Health Advisories (LTHA) are calculated from the DWEL and represents theconcentration of a substance in drinking water estimated to have negligible deleteriouseffects in humans over a lifetime of 70 years, assuming 2 L/day water consumption fora 70-kg adult, and taking into account other probable sources of exposure. In theabsence of chemical-specific data, the assumed fraction of total intake from drinkingwater is 20%. Lifetime health advisories are not derived for compounds consideredpotentially carcinogenic for humans.

Maximum Contaminant Levels (MCLs) represent contaminant concentrations indrinking water that EPA deems protective of public health (considering the availabilityand economics of water treatment technology) over a lifetime (70 years) at an exposurerate of 2 liters of water per day.

Permissible Remediation Goals (PRGs) or Action Levels are chemical- and media-specific levels of contamination which, when exceeded, automatically trigger aregulatory response or remedial action of some kind.

Permissible Exposure Limit (PEL) is an 8-hour, time-weighted average concentrationof a substance in workplace air designed by the Occupational Safety and HealthAdministration (OSHA) to assure, to the extent feasible, that chemical exposures in theworkplace do not impair the health or functional capacity of workers throughout theirworking life. The PEL may be exceeded for brief periods, but the sum of the exposurelevels averaged over 8 hours must not exceed the PEL.

Threshold Limit Value (TLV), according to the American Conference of GovernmentalIndustrial Hygienists (ACGIH), is "the time-weighted average concentrations for anormal 8-hour workday and a 40-hour workweek, to which nearly all workers may berepeatedly exposed, day after day, without adverse effect". Many of ACGIH's TLVswere adopted by OSHA for use as PELs.

TLVs and PELs, which were designed to protect healthy workers, are usually muchhigher than the health-based values of ATSDR and EPA, which were designed toprotect the health of the general population, including the very young and the elderly. ATSDR does not base any of its community health decisions solely or primarily onTLVs or PELs, but these and other non-ATSDR values may be referred to in PublicHealth Assessments or consultations as a means of providing the reader with anexpanded, and perhaps more meaningful, perspective on the concentrations ofcontaminants detected at a site.

Reference

Agency for Toxic Substances and Disease Registry. Health Assessment GuidanceManual. Atlanta: ATSDR, March, 1992.

APPENDIX D - ATSDR Response to Comments

This appendix contains both the comments received during the public comment periodfor the Bovoni Landfill site and ATSDR's response to those comments. The commentshave been numbered and are in italic with ATSDR's response directly below eachcomment. Each comment references specific sections and subsections of the publichealth assessment.

1. BACKGROUND, Site Description and History, second paragraph, second sentence: This statement is quite misleading. Please explain and/or clarify this statement.

During ATSDR's data collecting activities, a variety of individuals stated their beliefs asto how long the underground fire has been burning. These reports conflicted concerningthe actual length of time the landfill has been burning underground. Because ATSDR isnot able to confirm the exact time frame with conclusive data, the sentence cannot befurther clarified. In the main text of the public health assessment, ATSDR has simplystated the various time frames that were reported.

2. BACKGROUND, Site Description and History, third paragraph, second sentence: Is the ATSDR 100% certain that all medical waste is routed to anincinerator at the Roy Schneider Hospital rather than to the landfill? If so, please explain the bases for your assumption.

During the initial site visit in August 1996, ATSDR staff were informed that the medicalwaste incinerator at the Roy Schneider hospital was functioning and that the operatingprocedure was to route medical waste to this incinerator. ATSDR staff were alsoinformed that there was a 'Cease and Desist Order' to stop accepting and disposing ofmedical waste (infectious materials, pathological wastes, etc.), along with otherhazardous waste, at the landfill on January 31, 1996 (45). ATSDR saw no data duringour visit in August 1996 to indicate a procedure was in place to routinely route medicalwaste to the landfill. ATSDR staff spoke with a representative of the Department ofPublic Works on October 7, 1997, who indicated that medical waste is not currentlyaccepted at the landfill (49). ATSDR recognizes that lack of a routine plan does not mean medical waste may not accidentally end up in the landfill.

3. BACKGROUND, Site Description and History, third paragraph, third sentence:
   This is clearly a major oversight of the government by not installing environmentalmonitoring facilities to collect potentially dangerous elements and contaminateswhich may be impacting the health and safety of the public and environment of St.Thomas, U.S. Virgin Islands.

This comment appears to be a statement of position, not a question needing clarification.ATSDR is part of the US Department of Health and Human Services and has no legalauthority regarding landfill operation or monitoring requirements.

4. BACKGROUND, Site Description and History, third paragraph, fourthsentence: This field investigation was totally insufficient and incomplete, becausethere were no collection of off-site groundwater and/or air samples during orbetween landfill surface fires.

The stated purpose of the field investigation was to collect information that could beutilized as a basis for subsequent activities at the landfill. This field investigationfocused on determining the extent of subsurface burning and daily landfill emissions.The field investigation was not conducted to address off-site exposures to groundwateror off-site exposures to air contaminants during above-ground fires. ATSDR has madeas much use of available data as possible, regardless of the purpose for which the datawere collected. This public health assessment clearly indicates the lack of off-site airdata and also indicates those areas where ATSDR believes important data gaps existregarding potential effects on public health.

5. BACKGROUND, Site Description and History, fourth paragraph, fourth and fifth sentences: We also feel strongly that the residents and possibly the touriststhat ingest the lagoon's marine life are contaminating their bodies and adverselyaffecting their health.

ATSDR stated in the Community Health Concerns Evaluation section that we wereunable to locate any fish sampling data from the Mangrove Lagoon area. ATSDR notesthat there does not appear to be a good method available to determine if marine lifesampled in the lagoon are contaminated by the landfill. If fish sampling were conducted,the possibility exists that any fish caught in the lagoon could have been contaminatedelsewhere and entered the lagoon shortly before being caught. Another possibilitywould be that fish caught in the lagoon but not showing contamination may have justrecently entered the lagoon from open waters. To address community concerns anddata gaps regarding the lagoon, ATSDR has added a recommendation in this publichealth assessment to sample sediment in an area of the lagoon adjacent to the landfill. Please refer to the Recommendation section of the public health assessment. If databecome available, ATSDR will evaluate the data for public health significance.

6. BACKGROUND, Demographics: The accuracy of the demographic information is questionable. Please review the data used to compile this information and make necessary corrections.

ATSDR apologizes for errors in the demographic information concerning the totalpopulations for black and white persons within one mile of the site. Please refer to themain text and Figure Two for these changes.

7. BACKGROUND, Natural Resource Use, second paragraph, first sentence: Assuming that the on-site groundwater is contaminated and the groundwatermovement is determined to flow towards the east continuing into the MangroveLagoon, then the residents concerns are quite valid regarding the possiblecontamination of marine life in the lagoon. It is essential for the government toinstall groundwater wells for sampling/testing to determine the groundwaterdirectional flow, speed of flow, total depths, and levels and concentrations ofcontaminants within the lagoon and residential private groundwater wells.

Given the statement that residents are concerned about contamination of marine life inthe lagoon and information that firefighting water has been allowed to runoff from thesurface of the landfill into the lagoon, it appears the appropriate place to sample wouldbe the water or sediment of the lagoon. The unique location of the landfill on apeninsula and the limited elevation of the peninsula (approximately 140 feet or lessabove sea level) make it reasonable to predict that groundwater movement is toward theeast, which would be directly into the lagoon with no intervening potential humanexposure points. If additional groundwater data become available, ATSDR will re-evaluate our hydrogeologic interpretation. As stated previously, ATSDR has added arecommendation to sample sediment in an area of the lagoon adjacent to the landfill. Ifdata become available, ATSDR will evaluate the data for public health significance.

8. ENVIRONMENTAL CONTAMINATION AND OTHER HAZARDS, On-Site Contamination, Groundwater, first paragraph, first sentence: Based on thisAssessment Document, dated July 24, 1997, the 1992-1993 sampling data isantiquated, insufficient/incomplete, lacking off-site groundwater monitoring wellsamplings.

ATSDR notes that only on-site wells were sampled in the data used for the public healthassessment. In addition, the document states that ATSDR's evaluation was limited bythe amount of data available regarding the landfill. ATSDR staff and the commenterparticipated in a conference call on November 5, 1997, to clarify his concerns (43). Thecommenter believed a friend knew of private drinking water wells located on the westside of the peninsula although no further information was provided to ATSDR. ATSDRinformed the commenter that no primary water supply wells were identified on thepeninsula near the landfill which could be anticipated to be impacted by potentialgroundwater contamination from the landfill.

9. ENVIRONMENTAL CONTAMINATION AND OTHER HAZARDS, On-Site Contamination, Groundwater, second paragraph, first sentence: Again, no off-sitegroundwater monitoring wells. No off-site monitoring wells have been installedand sampled. Based on the absence of off-site monitoring wells, ATSDR cannotassure the residents that their private wells and the Mangrove Lagoon have notbeen contaminated. If the Lagoon has been contaminated then the marine life isalso contaminated, which is ingested by the residents and possibly the tourists ontravel to St. Thomas, U.S. Virgin Islands.

ATSDR agrees that these are valid concerns; however, ATSDR has not located anynearby private water supply wells and does not see a hydrogeologic basis to predictcontaminant migration through groundwater to any location other than on the peninsula. ATSDR has already noted concerns about the lagoon, but does not have any data toshow that marine life is likely contaminated at levels of public health concern. As statedpreviously, ATSDR is recommending sediment sampling in the lagoon, but notes thatthere does not appear to be a good method available to determine if marine life sampledin the lagoon are contaminated by the landfill.

10. ENVIRONMENTAL CONTAMINATION AND OTHER HAZARDS, On-Site Contamination, Air, first paragraph, second sentence: An absence of airmonitoring stations off-site and not during an above ground fire, clearly shows thatATSDR cannot, with any reasonable accuracy, identify the true effects of theresidents and workers exposure and health effects caused by the hazardousevents of the landfill.

ATSDR was constrained by the limited available data for the site and stated this in thepublic health assessment. In the absence of any off-site monitoring data, thecommenter's statement is true about it not being possible to ascertain "with anyreasonable accuracy" the concentrations of contaminants in off-site air. Fortunately,however, a useful answer to the qualitative (i.e., imprecise) question ATSDR usuallyaddresses -- "Were exposures likely to have been somewhere above or somewherebelow health effects thresholds" -- does not demand a particularly high degree of"accuracy", as long as the contaminant levels from which the assessor is extrapolatingare low enough. In the case of Bovoni Landfill, on-site air monitoring indicated that allcontaminants, with the possible exception of phosgene and mercury vapor, were belowlevels that might produce adverse health effects. It is only reasonable to expect thatcontaminant levels 500 feet to a mile away from the source will be substantially lowerthan those at the source. From this, ATSDR concluded that contaminants at BovoniLandfill were not likely to produce adverse health effects in nearby residents except,possibly, during major surface fires. If the concentrations of certain respiratory irritants(such as formaldehyde, acrolein, phosgene, and mercury vapor) were significantly higherduring past above-ground fire events than they were during the 1995-96 on-sitesampling event, then they could have exacerbated the effect of smoke inhalation.

There was little that ATSDR could do to compensate for the absence of off-site air dataat Bovoni; however, ATSDR did make use of data located for similar fire events at otherlandfills including two in the Virgin Islands. For instance, the results of off-site airmonitoring at the Susanneberg Landfill on St. John indicated that adjacent anddownwind residential areas were not exposed to acutely toxic levels of air contaminantsbut were continually exposed to relatively small concentrations of smoke and particulates(48). In order to better understand potential risks to residents, ATSDR has alsorecommended that off-site air sampling be conducted downwind during efforts toextinguish the underground landfill fire.

11. ENVIRONMENTAL CONTAMINATION AND OTHER HAZARDS, On-Site Contamination, Air, third paragraph, seventh sentence: Table Five - Air MonitoringData, 1996 - depicts low concentration of contaminants. The results wouldobviously yield low concentration of contaminants due to, the sampling wasconducted in the absences of an above-ground fire event. The data results inTable Five are misleading.

ATSDR clearly states that samples were taken upwind to determine backgroundconcentrations and at vents to determine landfill emissions. Table 5 specifies thatsamples were collected from locations where the underground fire vents to the surfaceand therefore is not misleading. ATSDR recognizes that every fire event may lead todifferent chemical composition of the smoke emitted. To further clarify this samplingevent, ATSDR has added to the main text of the public health assessment that thesampling was not conducted during an above-ground fire event.

12. ENVIRONMENTAL CONTAMINATION AND OTHER HAZARDS, Off-Site Contamination, Sediment: Table Six data results are outdated and are basicallyuseless when ATSDR is attempting to identify the current health hazards to theresidents and landfill workers.

ATSDR acknowledges the limitations of the data in the text. ATSDR's evaluation of thesite includes as much evaluation as possible (from the available data) of past andpotential future exposures in addition to current concerns. As stated previously, ATSDRhas added a recommendation to sample sediment from an area of the lagoon adjacent tothe landfill. If data become available, ATSDR will evaluate the data for public healthsignificance.

13. ENVIRONMENTAL CONTAMINATION AND OTHER HAZARDS, QualityAssurance and Quality Control, second paragraph: Based on this paragraph, it isessential that ATSDR advise Federal, State, and Local Government/ Agencies totake immediate actions to rectify the hazardous dangers of the landfill in order toprotect the health and safety of the public and the environment.

ATSDR's mission is to "prevent exposure and adverse human health effects anddiminished quality of life associated with exposure to hazardous substances from wastesites, unplanned releases, and other sources of pollution present in the environment." ATSDR does have the authority to issue public health advisories in instances where asite poses an 'Urgent Public Health Hazard' that requires rapid intervention to mitigatethe health risks posed by the site. This site does not fit the criteria for an 'Urgent PublicHealth Hazard' based on the available data. However, ATSDR considers that theBovoni Landfill site does pose a 'Public Health Hazard,' primarily because of physicalhazards (unstable earth and the potential for explosions and surface fires) and thecombined effect of such respiratory irritants as smoke, various aldehydes, phosgene,and mercury vapor. Please refer to the Recommendation and Public Health Action Plansections of this public health assessment where ATSDR has provided informationconcerning specific actions to protect public health for this site.

14. PATHWAYS ANALYSES, Completed Exposure Pathways, Air, secondparagraph, first and second sentences: It is essential that off-site air monitoringstations be installed to sample and test for air contaminants, in residential areas, during above-ground landfill fire events and between landfill fire events.

As noted in the response to Comment Number 11, ATSDR recognizes that every fireevent is different and therefore data from a future fire event would not provide anycertainty regarding exposure during past fire events. ATSDR believes it is far moreimportant to public health for the operator of the landfill to make every effort to preventfuture fire events and thus prevent future exposure. This goal is embedded in ATSDR'srecommendations to extinguish the underground fire at the landfill and to moreeffectively control access to the landfill. Further, ATSDR has recommended samplingdownwind during remedial actions in an attempt to address potential exposures duringthis activity.

15. PATHWAYS ANALYSES, Potential Exposure Pathways, Groundwater, first paragraph, fourth and fifth sentences: A thorough investigation must beconducted to determine if there exist private wells in the vicinity and pathway ofthe groundwater flow from the Bovoni Landfill. If there are, all private wells must be tested for contamination.

ATSDR has made efforts to address community concerns about the groundwater supply. Further, ATSDR has indicated that we do not anticipate any potential groundwatercontamination other than on the Bovoni peninsula. ATSDR is not aware of any privatewells on the peninsula.

16. PATHWAYS ANALYSES, Potential Exposure Pathways, Groundwater, first paragraph, sixth sentence: Please identify the following: the range of depth of theon-site wells that the 1996 sampling data was gathered from, the range of depthof the on-site groundwater, the range of depth of the off-site private wells in thevicinity of the landfill, and the range of depth of the groundwater that flows fromthe landfill to the Mangrove Lagoon.

Two on-site wells sampled in 1996 as part of the field investigation at the landfillencountered groundwater at three to four feet. Two other wells were planned, one nearthe entrance of the landfill and one along the southern boundary. For one well the augerwas not able to penetrate rock below eight feet and for the other well, the water tablewas not encountered at greater than 30 feet. From the available data, ATSDR has madethe hydrogeologic determination that groundwater beneath the landfill flows to thelagoon. The depth of active groundwater circulation beneath the landfill would not beexpected to have any impact on the ultimate direction of flow. As stated previously,ATSDR is not aware of any off-site private wells on the landfill peninsula.

17. PATHWAYS ANALYSES, Potential Exposure Pathways, Groundwater, first paragraph, seventh and eighth sentences: We understand that the steeply slopingelevation of the landfill topography controls groundwater movement (at a minimaldepth), at the Bovoni Landfill, and the movement is determined to be toward theeast; But, ATSDR must consider the subsurface as being highly folded, faultedand fractured rock which allows the groundwater to follow these fault zones. Oncethe groundwater reaches a certain depth the steep, sloping elevation of thetopography plays a lesser role in the directional flow of the groundwater. Theuncertainty of subsurface elements affords the groundwater to flow in manydirections, even though the eastern direction may be the main pathway. ATSDR,documented that you had difficulty securing groundwater in the faulted andfractured rock in the Bovoni Landfill area. Based on ATSDR's available data, andthe understanding that the groundwater flow is multi-directional; and with residentsliving in the vicinity, especially in the eastern section near the landfill, with privatewells, then, these ATSDR statements are misleading and totally incorrect.

ATSDR concurs that the site appears to be underlain by a fractured rock hydrogeologicsystem; however, ATSDR sees no hydrogeologic basis for the statement that"groundwater flow is multi-directional." In addition, it should be clarified that ATSDRcited a report by a consulting firm that indicated some wells drilled in the area did notencounter groundwater, but that ATSDR did not perform the work or document the fieldconditions. To summarize ATSDR's position regarding potential groundwatercontamination, given the limited data available at this site, ATSDR does not see humanexposure to landfill contaminants in the groundwater as likely to occur anywhere excepton the Bovoni peninsula and ATSDR is not aware of any private wells operating on theBovoni peninsula.

18. PATHWAYS ANALYSES, Potential Exposure Pathways, Cistern Water, first paragraph, fourth sentence: Based on the facts that no air monitoring stations(on-site and/or off-site of the landfill) were utilized to take samples during surfacefires; and no cistern water sampling data exist, it is obvious that there is a lack ofadequate data to support ATSDR's assumption that "no significant exposurewould be expected from utilizing the cistern waters". In fact it would seem that theopposite would be true, that there does exist the possibility of significant exposurefrom utilizing the cistern waters.

The determination that 'no significant exposure to cistern water would be expected' wasdetailed in the Community Health Concerns Evaluation section of the public healthassessment and is further clarified here. While no on-site or off-site air monitoringstations were utilized to take samples during the fires at the Bovoni Landfill, ATSDR didreview data from similar landfill fires at the Anguilla Landfill on St. Croix and theSusanneberg Landfill on St. John. Sampling of PAHs during the Anguilla Landfill fire didnot reveal any contamination above the method detection limit (44). Sampling of the on-site air at the Susanneberg Landfill revealed PAHs while off-site samples were non-detect (46). PAHs were detected in on-site monitoring events at the Bovoni Landfill. Inreviewing this data, ATSDR determined that since it is likely the levels of PAHs werelower (possibly non-detect) off-site and PAHs are relatively insoluble in water, nosignificant PAH exposure is expected from drinking cistern water.

The levels of VOCs detected during the Anguilla fire were similar to concentrations foundin the smolder pit area and at vents at the Bovoni Landfill (44). Air sampling at theSusanneberg Landfill indicated that exposure and/or inhalation of air contaminantsoutside of the dump premises was at non-detectable levels of the tested aircontaminants (47). Most VOCs have high vapor pressures and relatively low watersolubilities (13). For this reason, the amount of VOCs absorbed by falling rain would besmall. Therefore, the amount of VOC contamination from a landfill fire entering cisternsfrom falling rain is not expected to be significant.

Cistern water sampling was conducted after a fire at the Susanneberg Landfill. No contamination that could be attributed to the Susanneberg Landfill fire was found in thecistern water samples (48). Therefore, it was determined that there is no significanthealth threat with the cisterns that may be associated to the Susanneberg Landfill fire(47). In addition, in 1987 as part of an investigation of the Tutu Wellfield site in east-central St. Thomas (north of the Bovoni Landfill), cistern water samples were analyzedfor selected VOCs. This testing did not reveal any VOC contamination (13). Based onthe above discussion, ATSDR does not consider it likely that cistern water has beensignificantly contaminated as the result of the Bovoni Landfill fires.

19. PATHWAYS ANALYSES, Potential Exposure Pathways, Sediment, first paragraph: The referenced paragraph is totally misleading. First, the samplingwas conducted over 27 years ago, which has no validity relative health hazardscaused by the landfill between the 1970s and the current date. Next, what is thebases for ATSDR's determination that exposure to the public is intermittent? Identify/clarify "quantities of concern" regarding sediments ingested. Finally, canthe ATSDR confirm with 100% accuracy that each individual had only intermittentexposure and each time the person was exposed he or she unlikely ingestedsediments in quantities of concern? Please validate your assumptions.

ATSDR based its health assessment on the highest recorded concentration available foreach contaminant detected. There is no guarantee that even higher, but unmeasured,levels of these contaminants exist somewhere on the island, but the margin for error isgreatly reduced by ATSDR's practice of using only the maximum recordedconcentrations. In any case, ATSDR can only comment on the available information. Asstated previously, ATSDR has added a recommendation for sediment sampling in anarea of the lagoon adjacent to the landfill. ATSDR will evaluate the public healthsignificance of any new data collected from the lagoon.

One hundred percent accuracy is not needed to answer the question of exposures atlevels of public health concern. The assumptions that sediment exposures were"intermittent" and that recreational users of the Mangrove Lagoon were not likely toingest sediments in "quantities of concern" are based primarily on general experience;that is, people would not likely eat much sediment while boating, swimming and fishing.

20. PUBLIC HEALTH IMPLICATIONS, Toxicological Evaluation, fourth paragraph, first and second sentences: There is a second pathway, Groundwater, which wasnot monitored or sampled in off-site residential areas or in the inner MangroveLagoon.

Completed pathways indicate that exposure to the contaminant has occurred, isoccurring, or will occur in the future. As stated previously, ATSDR is not aware of anyprivate drinking water wells in the vicinity of the Bovoni Landfill. Both groundwater andsurface water are defined as potential pathways in this public health assessment. Please refer to the Pathways Analyses section for further clarification.

21. PUBLIC HEALTH IMPLICATIONS, Toxicological Evaluation, fourth paragraph, sixth sentence: This may be true, but at the time of a surface fire event at thelandfill, the lower off-site concentration of contaminants in the air may be at ahigher enough, critical, level which could cause adverse impacts to the health andsafety of the public and the environment. ATSDR's assumption is not only notsafe, but is also misleading.

In the paragraph in question, ATSDR clearly states that its conclusion is based solely onair data from "on-site monitoring performed at worst-case locations (i.e., at vents, burnpits, etc.), and at times when no major landfill fire was raging." Elsewhere in thetoxicological evaluation (Acrolein, second paragraph; Formaldehyde, second paragraph;Phosgene, second paragraph), ATSDR stated the possibility that concentrations ofcertain respiratory irritants may have been higher during landfill fires, and that, acting inconcert with smoke inhalation, they could have contributed to the respiratory irritationreported by workers and nearby residents at those times.

22. PUBLIC HEALTH IMPLICATIONS, Toxicological Evaluation, Contaminants in Air at Bovoni Landfill, 1995 and 1996, three quoted ATSDR statements: 1.) "Theprimary health hazards at the Bovoni Landfill are the respiratory irritants (bothchemicals and smoke) generated by the subterranean fire that occasionallybreaks through to the surface." 2.) "Nearby residents as well as on-site workerswill be at greatest risk during a surface landfill fire, i.e., when concentrations ofchemical respiratory irritants may be further elevated and smoke inhalation willexacerbate their effect." 3.) "Some workers, especially those with prolongedexposure, may also experience respiratory effects between fires." We agree withthese statements.

ATSDR thanks the commenter for this comment which quotes statements made in thepublic health assessment and agrees with ATSDR's evaluation of potential airexposures.

23. PUBLIC HEALTH IMPLICATIONS, Toxicological Evaluation, Contaminants in Air at Bovoni Landfill, 1995 and 1996, General Observation: Based on the factsthat no air monitoring was conducted or sampling collected during landfill surfacefires, it would be understandable that the air monitoring data from 1995 and 1996would show concentrations of individual chemicals (i.e. Acetaldehyde, Acrolein,Arsenic, Benzene, Carbon Monoxide, Ethyl benzene, Formaldehyde, MercuryVapor, Nickel, Trichloroethane, Trichloropropane, and Xylene) below theirrespective thresholds for respiratory irritation. ATSDR's conclusions are missingan important element from the theory, the absences of air monitoring duringsurface fires, which causes ATSDR's assumptions to be flawed.

As stated previously, in many places in the Toxicological Evaluation section (Acrolein,second paragraph; Formaldehyde, second paragraph; Phosgene, second paragraph), aswell as elsewhere in the document, ATSDR clearly states that concentrations of certainrespiratory irritants may have been significantly higher during landfill fires. Acting inconcert with smoke (the primary respiratory irritant), they may have contributed to therespiratory irritation reported by workers and nearby residents at those times. Inaddition, while no air monitoring data are available for the Bovoni Landfill fires, ATSDRdid review data from similar landfill fires and found on-site concentrations ofcontaminants similar to data from the smolder fissures and vents at the Bovoni Landfill. Off-site data collected during these fires indicated concentrations of contaminants belowlevels of health concern. Effects were stated as mostly of the nuisance type (i.e., irritantproperties from the smoke) (48).

24. PUBLIC HEALTH IMPLICATIONS, Community Health Concerns Evaluation, second bullet, third paragraph: If one takes into consideration the landfill waste,such as, waste mercury, mercury-containing medical waste (e.g., pharmaceuticalsand dental amalgams), agricultural wastes (e.g., pesticides), and/or electricalproducts (mercury switches, fluorescent lamps, and dry-cell batteries) drainingdown into the groundwater and flowing into the Mangrove Lagoon contaminatingthe marine life, which would be ingested by the local inhabitants and possibletourist, surely could plausibly cause the related health degeneration (e.g.,Musculoskeletal, Dermatological, and Gastrointestinal) symptoms.

Based on all the relevant environmental data available to ATSDR, none of the estimatedexposures, either individually or in combination, would have been likely to producesystemic toxicity. ATSDR considers that the available data, though incomplete, aresufficient to support this interim conclusion. If new data suggesting otherwise becomeavailable to ATSDR in the future, the Agency will amend this conclusion accordingly. Acute irritation of the eyes and respiratory tract were the only adverse health effectsclearly related to emissions from the Bovoni Landfill, and then only during major surfacefires. Smoke inhalation was the most plausible cause of these effects, although otherrespiratory irritants (phosgene and certain aldehydes) may also have contributed.

25. PUBLIC HEALTH IMPLICATIONS, Community Health Concerns Evaluation, third bullet, last sentences of the second and third paragraphs: Based on the factthat no air monitoring was conducted in residential areas or during above-groundfire events; and no residential cistern waters have been sampled or analyzed; it isclear that ATSDR has no valid, factual data to base these assumptions. ATSDR'sstatements gives the reader a false sense of security.

Please refer to the response to Comment Number 18 for a description of the data utilizedto determine that no significant exposures to VOCs and PAHs would be expected fromdrinking cistern water.

26. PUBLIC HEALTH IMPLICATIONS, Community Health Concerns Evaluation, third bullet, fourth paragraph, fourth and fifth sentences: Based on ATSDR'shygienic recommendations to the area residents, it is obvious that ATSDR thinksthat the residential cistern waters have been contaminated, caused by the BovoniLandfill.

As stated in first sentence of this paragraph, ATSDR does not consider it likely thatcistern water has been significantly impacted by the landfill. Because ATSDR was notable to confirm this conclusion with actual cistern water samples, we considered it goodpublic health practice to put forth recommendations that would reduce any potential forexposure. Please refer to the response to Comment Number 18 for further informationconcerning the cistern water evaluation.

27. PUBLIC HEALTH IMPLICATIONS, Community Health Concerns Evaluation, fifth bullet, second paragraph: Because the ATSDR staff touring the landfill sitedid not notice any visible medical waste on the surface of the landfill does notnecessarily conclude that medical waste is not continuously being dumped at thelandfill. Did ATSDR monitor the disposition of medical waste from cradle to grave?If yes, please explain the monitoring process and identify the overall time frame of the process.

Please refer to the response to Comment Number 2 for a discussion of medical waste and the landfill.

28. PUBLIC HEALTH IMPLICATIONS, Community Health Concerns Evaluation, sixth bullet, third paragraph: In reviewing this public assessment document, itseems ATSDR has made a plurality of public health statements and assumptionswithout substantial data. In this particular case it is essential that ATSDR appointthe appropriate agency(s) to conduct a comprehensive study and evaluation ofcontaminants and marine life within the Mangrove Lagoon.

ATSDR does not have the authority to "appoint" another agency to conduct sampling;however, ATSDR can recommend sampling. As stated previously, ATSDR has added arecommendation to sample sediment in an area of the lagoon adjacent to the landfill.

29. CONCLUSIONS, fifth paragraph, fourth sentence: Please describe the evaluation process, and document the engineering alternatives that will provide a solution to the underground fires.

Please refer to the Public Health Action Plan section of the public health assessment fora description of the current activities being conducted by the Department of PublicWorks. For further information on this issue, please contact the Department of PublicWorks directly.

30. CONCLUSIONS, sixth paragraph, first sentence: This sentence is unfounded and misleading and has no sampling data of facts to validate its conclusion.

Please refer to the response to Comment Number 18 for a description of the data utilizedto determine that no significant exposures to VOCs and PAHs would be expected fromdrinking cistern water.

31. CONCLUSIONS, sixth paragraph, second sentence: This sentence basically contradicts the previous sentence.

Please refer to the response to Comment Number 18 and Comment Number 25.

32. RECOMMENDATIONS, General Observation: ATSDR'srecommendations seems to be a little thin in assertively identifying the pertinentagencies to perform specific functions to rectify the hazardous conditions at theBovoni Landfill.*

    * The commenter further identified several specific functions that are not includedin this public health assessment. During a conference call with the commenter onNovember 5, 1997, ATSDR responded to each of the items identified by thecommenter (43).

The purpose of the Recommendation section of the public health assessment is toidentify actions to protect public health, obtain additional health information, conductpublic health actions, and/or obtain additional site-characterization information. ATSDRis not in a position to "appoint" other agencies to follow our recommendations. ATSDRdoes share copies of our documents with appropriate agencies which may follow-up onour recommendations and ATSDR personally contacts those agencies to emphasize ourconcerns and data needs. When appropriate, the Public Health Action Plan section mayeven be able to identify an agency that has agreed to follow an ATSDRrecommendation.

Table of Contents