IV. Evaluation of the Drinking Water Quality

Drinking Water Supplies:

The public water system on the island of Vieques is currently supplied with water from the main island of Puerto Rico. Any residents or businesses purchasing water from Compania de Aguas are getting water that was collected and treated on the main island of Puerto Rico, then piped into the distribution system through an underwater pipeline. This water originates in the mountains of the main island of Puerto Rico and is not affected by activities at the bombing range on Vieques.

Prior to having water piped from Puerto Rico in 1978, the water distribution system on Vieques was supplied from groundwater wells on the island of Vieques. These wells, located in the Esperanza and Resolucion valleys, are within localized aquifers and are not connected to the groundwater on the eastern end of the island. Increasing water demands of the communities on Vieques and the increased salinity of these water supplies (because of salt water intrusion) mandated the need for a better water supply for residents of Vieques. The use of these wells was therefore, phased out when the decision was made to supply water from the main island of Puerto Rico by pipeline.

Prior to the existence of the current public water distribution system and the installation of the Esperanza and Sun Bay well fields in the early 1960s, water was supplied by smaller private groundwater wells and by rainfall collection systems. Some of these sources may still be used today to augment water supplies in some households and businesses.

Each source of water is addressed below in the form of answers to questions. The answers given are based on the most current epidemiologic, toxicologic, and medical information available. When available, chemical analysis of the water is evaluated to determine whether there is any indication that the water has been impacted by range activities and whether the water is safe to drink and use in the home. The current public water supply and various groundwater wells have been sampled and analyzed by PRDOH, USGS, an environmental firm hired by the Navy, and EPA. Sampling summaries are provided in Appendix B and a complete list of chemicals tested by each agency is shown in Appendix C. Earlier sources of water use are more difficult to address, particularly when limited chemical analysis is available or when there is uncertainty about where the samples were taken. At this time, data do not exist to permit a full evaluation of the use of rainwater collection systems on the island; this water source will be more fully addressed through additional environmental sampling on Vieques and through computer modeling of air dispersion to evaluate the potential for airborne contaminants to affect these systems.

Question 1: Is the current public water supply safe to drink?

Answer:

The water supplied by pipeline to the island of Vieques is safe to drink and is suitable for all home and business use.

Recent chemical analysis of water in the public water supply system indicates that the water has not been impacted by bombing activities at the LIA.

There is no completed pathway between the LIA and the OB/OD area and the public water supply system for the island of Vieques.

Discussion:

Drinking Water Supply:

Most of the residents of Vieques receive their drinking water supply from the mainland of Puerto Rico. An underwater pipeline was built in 1977 from the mainland to Vieques and provides 800,000 gallons of drinking water per day to the residents of Vieques (Cherry and Ramos 1995). The source of water is the Rio Blanco (i.e., White River), which originates in the Yunque Rainforest in the mountains of the main island of Puerto Rico. Compania de Aguas, a company hired by the Puerto Rico Aqueduct and Sewer Authority (PRASA) to maintain and operate the water supply system, is responsible for distributing drinking water to the residents of Vieques. Treated water from the Rio Blanco Filtration Plant in Naguabo, Puerto Rico, is distributed to the Arcadia storage tank in the NASD area on Vieques through the underwater pipeline. This above ground storage tank has an engineered cover and is not susceptible to atmospheric deposition. The water is further treated by chlorination just before it reaches the Arcadia storage tank. Drinking water from the Arcadia storage tank is then supplied to all the other distribution and storage tanks on the island, which in turn supply water to both residents and Navy personnel. Figure 4 graphically displays the locations of the tanks in the public water supply system.

Sampling Summary:

The water provided from the mainland was sampled for many different types of chemicals, including explosives, metals, other inorganics compounds, herbicides, pesticides, polychlorinated biphenyls (PCBs), volatile organic compounds (VOCs), semi-volatile organic compounds (SVOCs), and radionuclides. Various tanks and taps within the PRASA public water supply system on Vieques were sampled by EPA and by the Puerto Rico DOH at different times in 1999 and 2000 (See Appendix B for a summary). All public storage tanks and a tap that is representative of the public drinking water supply on the island were included in the studies. Water from the Rio Blanco Filtration Plant in Puerto Rico, which provides the water piped to Vieques, was sampled for explosives by EPA. Additionally, in September 1999, an environmental consulting firm working for the Navy sampled the storage tank maintained by the Navy (NASD tank), splitting samples with EPA to confirm the findings. A sampling summary is provided in Appendix B and a complete list of chemicals tested by each agency is shown in Appendix C.

Water Quality:

Explosive-related chemical compounds were not detected in any of the samples, using current EPA approved analytical techniques. As disclosed in Table 2, the water from the public water supply system includes the usual, naturally occurring assemblage of metals and inorganic constituents and a few VOCs that are byproducts of water chlorination. Table 2 indicates the chemicals that were found and how much was present.

Evaluation of the Impact of Water Quality on Public Health:

The metals and other inorganic constituents found in the water from the public supply system reflect the geologic rock types, weathering, and soil formation occurring in the Rio Blanco watershed on the mainland of Puerto Rico. All these metals were present at concentrations well below any level of health concern. Barium, fluoride, aluminum, iron, manganese, and zinc were below the appropriate EPA drinking water standards (see Table 2). Additionally, the other inorganic compounds--chloride, sulfate, nitrate, and nitrite--were well within the applicable water standards. Although ammonia, boron, and strontium do not have maximum contaminant levels (MCLs, see text box for definition), the levels found in the Vieques drinking water are not a health concern. The small amounts one would be expected to ingest from the water is well below recommended maximum doses of these chemicals according to ATSDR's comparison with the minimal risk levels (MRLs) and EPA's reference doses (RfDs). Please refer to Appendix D for further details concerning how ATSDR estimated exposure doses and determined health effects.

The remaining metals do not have EPA or ATSDR standards for comparison. However, the levels detected for the remaining metals are also very low. A person drinking this water every day would consume a very small amount of each of these metals. For example, the essential nutrients, magnesium and potassium, do not have EPA MCLs. In fact, the U.S. Food and Drug Administration (FDA) recommends people consume a minimum amount of these metals to stay healthy--400 milligrams (mg) per day for magnesium and 3,500 mg per day for potassium. Based on magnesium and potassium detected in the drinking water, Vieques adults drinking 3 liters of the water per day would consume 11.1 mg of magnesium and 3.9 mg of potassium--5.5 mg and 2.0 mg, respectively, for children drinking 1.5 liters per day. These amounts are far below the recommended intake of these essential nutrients and are in no way harmful to adults or children. In fact, the presence of these metals in the drinking water aids in supplying the body with two essential nutrients, although the contribution to the diet is very low.

All of the VOCs found in the water were also below their corresponding MCLs. ATSDR estimated exposure doses for VOCs by using the highest detected concentrations and assuming people drank 3 liters of water a day (1.5 liters for children) for 70 years (6 years for children). The resulting doses for both children and adults were orders of magnitude lower than levels believed to cause adverse health effects.

The trihalomethanes (THM) found in the drinking water--including total THM, chlorodibromomethane, chloroform, and dichlorobromomethane--are known byproducts of drinking water disinfection and are most likely disinfection byproducts from the chlorination process rather than environmental contaminants. According to ATSDR dose calculations, the levels detected in the public water supply system for Vieques were far lower than levels known to cause adverse health effects. The presence of these compounds in the water at these low levels does not appreciably increase the risk of cancer or other adverse health effects.

Question 2: Is the groundwater on Vieques safe to drink?

Answer:

Water from the public supply wells in the Esperanza aquifer is safe to drink, according to all available analyses. The high total dissolved solids (TDS) and salinity may impact taste but are not a health hazard. When drinking from these wells, however, residents need to be aware of their additional sodium intake. Although the supply wells were not sampled for explosives, samples from other Esperanza aquifer wells are not contaminated by explosives.

There are no health hazards from exposure to explosives or their byproducts by drinking groundwater on Vieques. Although it was not possible to test the Esperanza drinking water supply wells for explosives, testing for explosives in other wells in that aquifer that are located closer to the LIA did not disclose the presence of those compounds. In addition, no other wells on the island detected the presence of explosives.

The water from Well 3-7, a shallow, private drinking water well, is not safe to drink because of elevated nitrate plus nitrite levels. Children, especially infants, and pregnant women should not drink water from this well. PRDOH has issued a health advisory on this well and notified local users. The levels of nitrate plus nitrite most likely result from agricultural pollution and are not conclusive evidence of explosive contamination.

Discussion:

Groundwater Resources for Drinking Water:

The fresh water supply on Vieques is limited by annual rainfall and the water stored in the alluvial aquifers. Prior to piping drinking water from the main island of Puerto Rico, the public water distribution system was supplied by pumping groundwater from wells in the Esperanza aquifer. Navy wells in the Camp Garcia area pump from connected alluvial deposits (Torres-Gonzalez 1989), which for the purposes of this report are considered the same alluvial system. Minor usage of the remaining aquifers on Vieques includes two Navy supply wells and approximately 14 individually owned wells across the center of the island, many of which were dry or unused when surveyed by the USGS in 1991 (Cherry and Ramos 1995). Figure 5 graphically displays the locations of the groundwater wells discussed in this PHA.

Water Use of the Esperanza Aquifer:

Several wells that pump from the Esperanza aquifer, including the Sun Bay and the B wells, were installed in the 1960s. PRASA operated these wells in the valley to provide drinking water for the residents of Vieques. During operation, the well field yielded an average of 425,000 gallons of water per day. Increased water production caused saline water to intrude into the wells near the coast. Chloride concentration, as a measure of salination, was shown to increase from around 100 mg per liter to as high as 300 mg per liter from 1973 to 1977 (Torres-Gonzalez 1989). This increase in salinity, as well as the natural limitation on the volume of water that could be pumped from the aquifer, were key factors in the decision to shut down the Esperanza valley well field in 1978.

Since groundwater management was initiated by PRASA in 1977, the Esperanza aquifer has nearly recovered to pre-developed conditions (CH2MHILL and Baker 1999). If properly maintained and utilized, the Esperanza valley well field could provide an alternate water source in the case of an emergency (Cherry and Ramos 1995). A two-dimensional groundwater model developed by USGS indicates that the aquifer can yield approximately 300,000 gallons of water per day during the wet season and 200,000 gallons per day during the dry season.

Until recently, Compania de Aguas maintained three Sun Bay wells that were used during emergencies (e.g., hurricanes, drought, or electrical outage) when the water supply from the mainland was interrupted. These wells are located in the Esperanza valley aquifer and are now reportedly closed. In the case of an emergency, the Sun Bay wells could be safely utilized, at least for a limited period of time. Four other wells, the B wells, are still available for emergency use.

In the past, the Navy also used groundwater wells in the eastern portion of this alluvial aquifer for its drinking water supply. Until 1998, Camp Garcia personnel received their drinking water from U.S. Marine Well 6 and Navy Well 14. Together, these two wells pumped 84,000 gallons of water per day, twice a week. However, in 1998, because of deterioration and maintenance problems, the Navy decided to bring drinking water from the NASD tank by tanker truck to Camp Garcia rather than continue to pump from these two wells. Navy Well 14 is now used to supply water to cattle.

Water Use of Other Alluvial Valleys:

Several former Navy wells were identified by USGS within the NASD in the Resolucion valley, along the northwest coast of the island (USGS 1997). Their former use is unclear for all but one of the wells--Navy Well 17 is known to be a former supply well (Cherry and Ramos 1995). None of the wells are currently usable.

A small unnamed alluvial valley north of Ensenada Honda also has one known well. A small (13 feet deep) hand-dug well was reportedly used by the Navy in this remote portion of the island. This well is apparently no longer in use, since it had chloride concentrations of over 4,000 mg per liter (Cherry and Ramos 1995).

Water Use of Other Groundwater Resources on Vieques:

The balance of the wells across the center of the island draw water from the upper portion of the weathered bedrock, from isolated patches of alluvial deposits, or from channels and faults in sedimentary rocks (e.g., limestone). Groundwater use is limited on Vieques because only minor aquifer recharge occurs from precipitation due to the high evapotranspiration rate (estimates suggest about 90% of the precipitation is lost to evapotranspiration, Torres-Gonzalez 1989).

Six wells were reportedly still in use in 1991 (Cherry and Ramos 1995). One well is used agriculturally to water livestock and three wells are used for drinking water and other domestic purposes. The last two wells (Well 2-3 and Well 3-7) are only used as an emergency water supply when the pipeline is not in operation. Well 2-3 is located in a remote area of the island, with access restricted by 80 to 100 feet of dense vegetation. Well 3-7 is located in a residential area where people fill containers with water from the well and take them home (EPA 1999b).

Sampling Summary:

In May 1995, PRDOH sampled the three Sun Bay wells and the four abandoned B wells on Vieques for VOCs, SVOCs, inorganics, metals, herbicides, pesticides, and PCBs. The USGS sampled five former Navy wells on NASD in November 1996 for inorganics. Three of these wells were additionally tested for VOCs and organochlorine pesticides. In August 1999, an environmental consulting firm hired by the Navy sampled the former supply well (Navy Well 17) on NASD property and the former drinking water well (Navy Well 14) in Camp Garcia for VOCs, SVOCs, inorganics, metals, herbicides, pesticides, PCBs, and explosives. In September 1999, EPA sampled the three Sun Bay wells and two emergency drinking water wells (Well 2-3 and Well 3-7) for VOCs, SVOCs, inorganics, metals, and explosives. Based on a review of the data, EPA determined that the explosives data were unusable; therefore, in January 2000, EPA returned and re-sampled the two emergency water wells for explosives and nitrate plus nitrite. The three Sun Bay wells were not re-sampled because they had been closed. See Appendix B for a summary of sampling by each agency and Appendix C for a complete list of chemicals tested by each agency.

Additional sampling was performed on monitoring wells. Monitoring wells are not used to supply drinking water to the public; rather, they are used to sample water quality in the aquifer. Because these wells are not used as a water source, the results at these wells are described only to help characterize the groundwater quality. In August 1999, the environmental consulting firm hired by the Navy installed 11 monitoring wells along EMA's western boundary and sampled them for explosives. Of the 11 monitoring wells installed along EMA's western boundary, two were placed to allow sampling in the alluvial deposits in the Esperanza aquifer. The remaining wells sample water within the weathered bedrock across the center of the island.

Water Quality:

Water Quality of the Esperanza Aquifer:

Water samples from the Esperanza aquifer contained metals, high levels of total dissolved solids (TDS), and high salt (sodium and chloride) [Table 3]. These results are consistent with earlier reports of high salinity in the Esperanza valley wells (Torres-Gonzalez 1989). In general, this aquifer is not a high quality water supply. Several of the metals found in the Sun Bay wells and the TDS in all wells are above their secondary maximum contaminant levels (SMCLs, see text box for definition). TDS--including boron, iron, manganese, and sodium--are commonly found in groundwater. The presence of these metals is directly related to the high levels of TDS in the water and probably reflects the natural geology of the island. Igneous and volcanic rocks, which make up the bedrock of Vieques, are a common source of iron and manganese (USGS 1997). Very small amounts of chloroform and total trihalomethanes were found in one of the B wells and in Navy Well 14. Di-n-butylphthalate was also detected in Navy Well 14.

Neither the Sun Bay wells nor the B wells were sampled for explosives by any agency. Currently, the Sun Bay wells are closed and there are no plans to reopen those wells; leaving them unavailable for re-sampling. Therefore, the only groundwater samples in the Esperanza aquifer available to characterize the potential for explosives contamination are the Navy wells that were sampled in 1999. Navy Well 14 was sampled for 16 explosive compounds and none were detected above the method detection limits (Table 3) [Baker 1999]. The Navy has also installed and sampled two monitoring wells in the alluvial deposits at EMA's western boundary (Well NW-8 and Well RCRA-4). As was the case with the old supply well, water samples from these wells did not contain any of the explosive compounds tested. All three of the Navy wells are located closer to the LIA than the Sun Bay wells and B wells and would detect any potential groundwater contamination from the LIA before it reached the Sun Bay wells and B wells.

Water Quality of Other Alluvial Valleys:

The Resolucion valley is the only other alluvial valley sampled. Navy Well 17 and four other former Navy wells in NASD were sampled by the Navy and USGS for the presence of contaminants. As was the case with water samples from the Esperanza aquifer, the Resolucion water samples had high TDS and salinity (sodium and chloride) with some trace metals (e.g., barium, calcium, and iron) [Table 4]. No explosives were detected above the method detection limits. Traces of benzene, chloroform, toluene, di-n-butylphthalate, and siloxane were found at levels very close to the method detection limits.

Water Quality of Other Groundwater Resources on Vieques:

Two of the six shallow water wells potentially in use by the general public were sampled by EPA--Well 2-3 and Well 3-7. In addition to expected trace metals (barium, boron, manganese, and zinc), these two wells had high TDS and a slightly higher salinity (chloride and sodium) than what was found in the Esperanza aquifer (Table 5). TDS does exceed the SMCL and chloride is close to the SMCL. The combined nitrate plus nitrite for Well 2-3 was similar to that of the Sun Bay wells, but Well 3-7 contained much higher levels, up to 12,600 parts per billion (ppb). This combined nitrate plus nitrite exceeds the MCL of 10,000 ppb. When EPA re-sampled Well 3-7, the nitrate plus nitrite levels were found to be considerably lower (1,700 ppb). However, after reviewing quality control data, EPA determined that the second sampling event probably underestimated the concentrations of those samples.

Di(2-ethylhexyl)phthalate (22 ppb in Well 3-7) was the only organic compound found in these wells. Although it was not detected in the laboratory quality control samples, EPA's laboratory noted that di(2-ethylhexyl)phthalate is a common laboratory contaminant and that the presence of di(2-ethylhexyl)phthalate is "most likely due to contamination of the sample during the collection and analysis of the samples" (EPA 1999b). Di(2-ethylhexyl)phthalate is a plasticiser in many common materials found both in the home and in the laboratory. Unless the well is known to contain synthetic materials that may contain di(2-ethylhexyl)phthalate, this result should be considered introduced by the laboratory.

Evaluation of the Impact of Water Quality on Public Health:

Explosives:

Currently, there are no health hazards from exposure to explosives or their byproducts by drinking groundwater on Vieques. No explosives were found in any of the groundwater wells tested (Baker 1999; CH2MHILL and Baker 1999; EPA 2000). Although it is theoretically possible that traces of explosives were present at levels below the limits of detection, no health effects would be expected at such low levels. The explosives reportedly used on the LIA include TNT, RDX, tetryl, HMX, and ammonium picrate (Navy 1990).

The limits of detection for TNT, RDX, tetryl, HMX, and TNT degradation products are well below any level of concern for noncancer health effects, an indication that these adverse health effects would not be expected even if contaminants were present below detection limits. Additionally, dose calculations for water intake over a lifetime exposure, at the detection limits indicate that these levels of intake are below levels of concern for cancer health effects. The potential cancer effects of RDX, TNT, and dinitrotoluenes (TNT breakdown products) can be more thoroughly evaluated than HMX or tetryl because more is known about those compounds. Please refer to Appendix D for further details concerning how ATSDR estimated exposure doses and determined health effects.

Very low levels of RDX, tetryl, ammonia, and nitrate plus nitrite may have been present in drinking water samples taken by the Navy in 1978. The validity and utility of the data is uncertain because of the small number of samples collected and the description of the location or media represented by the samples. Regardless, the concentrations of explosive compounds reported in drinking water in the past were well below levels considered harmful to human health and any potential past exposure to these compounds would not have posed a public health hazard. Please refer to Appendix E for a discussion and evaluation of this sampling event.

Nitrate / Nitrite:

Nitrate plus nitrite (measured together as nitrogen) was elevated in Well 3-7 at 12,600 ppb. This well was not sampled for nitrate and nitrite separately; the nitrogen found could be from either nitrate, nitrite, or both. The detected level of nitrogen is above the MCL for either nitrate, nitrite, or both together (10,000 ppb, 1,000 ppb, or 10,000 ppb, respectively). Excessive levels of nitrate/nitrite in drinking water can cause serious short- and long-term health effects for children because nitrite interferes with oxygen uptake in blood (EPA 1995). The greatest concern is for infants or pregnant women who may be drinking water with elevated levels of nitrate or nitrite. Nitrate can be converted in an infant's intestines to nitrite. The nitrite can then interfere with the oxygen-carrying capacity of the child's blood. Older children and adults do not convert as much nitrate into nitrite, and therefore, nitrate is of less concern if older children and adults are exposed. Nitrite, on the other hand, is of equal concern for all ages, since it does not need to be transformed in the intestines to actively interfere with oxygen uptake in the blood.

Both adults and children drinking all their water each day from Well 3-7 would consume more nitrite than recommended each day, assuming that all of the nitrogen in the water represented nitrite. However, it is unlikely that all the nitrogen is attributable to nitrite because (1) the elevated nitrate/nitrite levels in this well are most likely a product of agricultural chemicals, which contain nitrate, not nitrite and (2) of the eight wells on the island that were sampled for nitrite, nitrite was never detected, and there is therefore, no indication that large amounts of nitrite would be expected in this well. If all the nitrogen was an indication of the presence of nitrate, there would be no hazard for adults and older children. However, infants might be at risk, especially if they are drinking formula made with this water. Finally, the two samples taken at this well had very different levels of nitrogen--12,600 ppb and 1,700 ppb. If the levels in the well are fluctuating seasonally or because of other factors, an increase in the nitrates in the well could pose a hazard for older children and pregnant women. Therefore, the levels of nitrite/nitrate in this well may pose a public health hazard. Children, especially infants, and pregnant women should avoid drinking this water. PRDOH has issued an advisory and has personally informed the residents that water from Well 3-7 is not safe for consumption.

Other Inorganics:

All the groundwater sources sampled have high levels of TDS, high salinity (sodium and chloride), and elevated levels of naturally occurring metals. Iron and manganese found in the Sun Bay and NASD wells and the TDS in all groundwater wells are above EPA's SMCL. Levels detected above the SMCL may affect the taste, odor, or color of the water or result in secondary health effects, such as discoloration of teeth. Chemicals found at levels above the SMCL do not indicate a health hazard; several of these elements are actually essential nutrients for the body. Although the supply wells in the Esperanza aquifer were used for less than 20 years, the discussion will assume a lifetime use in order to be protective of any individuals who may still be using groundwater as the primary source of drinking water in their home. Please refer to Appendix D for further details concerning how ATSDR estimated exposure doses and determined health effects.

The detected concentrations of TDS are well within levels EPA would consider to be usable (EPA considers water with more than 10,000,000 ppb of TDS to be unusable for drinking). In fact, some people may tolerate and/or actually enjoy the taste of water with high levels of TDS, especially if that is what they are used to drinking (EPA 1984). Included among the TDS are iron, manganese, sodium, and boron.

Iron, manganese, and sodium are important minerals that maintain basic life functions.

Iron is used by the body to make hemoglobin, which transports oxygen in the blood from the lungs to other areas of the body that need oxygen. It also helps increase the body's resistance to stress and disease (ANR 2000). FDA's Daily Value for iron is 18 mg.

ATSDR's calculated daily intake of iron from the Sun Bay wells was about one third of the Daily Value for adults and less than one fifth of the Daily Value for children. Iron was not detected in the other drinking water wells. The additional iron that would be received by drinking water from the Sun Bay wells is not harmful. The reason that an SMCL is set for iron is not that the iron can cause adverse health effects, but rather that water with high levels of iron may have an unpleasant taste to some people and may stain material that is washed in water with high iron concentrations.

Manganese is an antioxidant that helps produce energy for the body. The average amount of manganese in a normal diet is about 1 to 5 mg a day (ATSDR 1997). However, very high levels of manganese in the diet may cause harmful effects. Therefore, ATSDR conservatively compared probable intake levels (i.e., ATSDR calculated doses) of manganese to available health guidance values and found that the detected concentrations were below levels that would likely result in adverse health effects due to drinking water.

Sodium is another essential nutrient used by the body to control blood pressure and volume. In addition, sodium helps the muscles and nerves function properly. However, too much sodium can cause high blood pressure. ATSDR calculated daily intakes of sodium according to the highest detected concentration in drinking water wells. Adults who drink 3 liters of water from Well 3-7 receive an additional 687 mg of sodium each day they drink from that well. For comparison, residents who drink 3 liters of water from the public water supply system are ingesting only an additional 30 mg of sodium a day. FDA's Daily Value for sodium is 2,400 mg (USDA and USDHHS 2000). Residents drinking water from a well with elevated sodium levels should be aware that their sodium intakes are increasing and should modify their diets accordingly. People who already have elevated blood pressure or who are on a sodium-restricted diet should avoid drinking water from wells with elevated sodium levels. Determining whether the additional sodium intake will adversely affect a person's health is strictly an individual determination based on that person's diet and health status. Residents concerned about their intake of sodium, should seek advice from their physician.

Another element detected in the water is boron, which occurs naturally in rocks. It is naturally released to the environment through the weathering process of rocks that contain boron. Various organs in the human body can be harmed if large amounts of boron are consumed (e.g., greater than 4,000,000 ppb) [ATSDR 1992]. ATSDR calculated doses of boron from drinking 3 liters of water (1.5 liters per day for children) on a daily basis from the well with the highest concentration. ATSDR compared the calculated intake levels to available health guidance values and found that the conservative doses were well below levels that would be likely to cause adverse health effects.

Organics:

Two wells in the Esperanza aquifer (one of the B wells and Navy Well 14) contained traces of one organic compound, chloroform. The levels of chloroform were well below EPA's MCL. Additionally, the calculated intakes, even over a lifetime of usage, are well below a level that would produce any adverse health effects. Please refer to Appendix D for further details concerning how ATSDR estimated exposure doses and determined health effects.

Although the source of the benzene and toluene in Navy Well 17 is unknown, the levels are far below the MCL. A lifetime of water usage from this well would not result in drinking enough of these compounds to produce any adverse health effects.

The two former Navy supply wells (Navy Well 14 and Navy Well 17) had di-n-butylphthalate reported in the samples. Di(2-ethylhexyl)phthalate was present in Well 3-7. As components of many plastics and common laboratory contaminants, these chemicals are probably not actually present in the drinking water. However, even assuming their presence, the levels found are not of health concern. A lifetime exposure to the maximum amount found in these samples does not increase the risk of any adverse health effects.

Well Construction:

EPA is concerned that the construction of Well 2-3 and Well 3-7 is inadequate to protect the water from surface runoff. ATSDR evaluated the chemical data gathered from these wells and found only the nitrate plus nitrite level in Well 3-7 to pose a health concern. ATSDR agrees with EPA that the poor well construction could potentially lead to contamination in the future.

Question 3: Is water from rainfall collection systems safe to drink?

Answer:

Additional data need to be gathered in order to answer this question adequately. Historical data may not exist to provide a full answer to this question for past exposures.

Discussion:

Drinking Water Supply:

Some private residences may have rainfall collection systems to supplement their drinking water supply. In 1995, it was reported that the rainfall collection systems were a second source of freshwater, although the authors did not specify if the basins were used for drinking water supplies (Cherry and Ramos 1995). It is ATSDR's understanding that most of these basins have been converted into closed tanks that are supplied with public water. However, some residents may still use rainwater from rainfall collection systems in addition to the public water supply. In a 1998 sanitation survey for Vieques, 541 cisterns and tanks were identified (DOH 1998). These tanks may be used for water storage from the public water system, although some may be used in conjunction with rainfall collection systems.

Sampling Summary:

No sampling has been conducted to analyze drinking water from rainfall collection systems. In a 1978 study (see the discussion for Question 4), it was reported that one of the drinking water samples was diluted with rainwater (presumably from the use of rainfall collection system). This one sample point does not provide sufficient data to determine the impact on water quality from potential aerial dispersion of LIA-generated contaminants.

Water Quality:

Because no data are available on the rainfall collection systems, the associated water quality cannot be determined at this time. However, collected rainfall could contain substances from several different sources. Any dust, debris, or chemicals that might collect on the roof between rainfalls could be washed into the collection container unless provisions are made to ensure that the initial rainfall runoff, that rinses potential contaminants off the roof or collection surface, is not collected in the storage container. Otherwise, the water quality in the collection basin would reflect what was in the dust or other contaminants that fell on the collection surface between rainfalls.

ATSDR is in the process of evaluating whether airborne chemicals may travel across Vieques during Navy operations at the LIA. Winds on Vieques generally blow from east to west, potentially carrying dust and chemicals in the direction of populated areas. However, no data are available to quantify what chemicals might be transported across the island. Soil samples from the LIA are currently being analyzed to provide information on what chemicals are at the surface of the LIA. Historical information about weapons usage is also being gathered and an air dispersion model is in development to help provide estimates about how far chemicals may travel.

The seasonal African dust storms are also a major source of dust on both Vieques and the main island of Puerto Rico. The volume of dust as well as the metals that may be contained in the dust will also be evaluated. Finally, local use of pesticides over the years may have resulted in pesticides being deposited on collection areas and washed into collection basins.

Evaluation of the Impact of Water Quality on Public Health:

This potential impact on public health cannot be evaluated at this time because ATSDR does not have information about the location, use, or extent of contamination in rainfall collection systems. This question will be evaluated once data become available.