III. Background

ATSDR's initial approach to evaluating air quality issues at Vieques involved gathering background information on several important topics, such as specific health concerns, site history, local demographics, and meteorology. The following discussion reviews the information collected on these and other topics, which are important background material for ATSDR's technical analyses, as documented in the "Evaluation of Air Quality Issues" section (Section V).

The remainder of this section primarily presents facts and observations about Vieques, without any analysis or interpretation. Later sections of this PHA document ATSDR's interpretation of the background information presented below.

A. Site Description and Land Use

Vieques is the largest offshore island that is part of the Commonwealth of Puerto Rico. Vieques is 20 miles long, 4.5 miles at its widest point, and about 33,000 acres (or 51 square miles) in area. Figure 1 shows the location of Vieques and surrounding islands. As the figure shows, the nearest island to Vieques is the main island of Puerto Rico, which is approximately 7 miles west of Vieques; the island of Culebra is roughly 9 miles north of Vieques; and St. Thomas, St. John, St. Croix, and other islands within the U.S. Virgin Islands are all at least 20 miles from Vieques, generally to the northeast and southeast. Therefore, Vieques is several miles removed from sources of air pollution on any other island in the Caribbean Sea.

The detailed map in Figure 2 conveys critical background information on land use in Vieques. The figure depicts the island in three separate sections, each of which is described in greater detail below:

• Former NASD Lands (or West Vieques). Figure 2 labels the western portion of Vieques as "former NASD lands," which is also commonly referred to as West Vieques. Prior to May, 2001, these 8,200 acres were Navy property and were known as the Naval Ammunition Support Detachment (NASD). Most of this land is undeveloped, and Navy operations there were limited. The Navy land uses at NASD included ammunition storage, a rock quarry, communication facilities, and Navy support buildings (IT Corporation 2000). In May, 2001, the Navy transferred most of the former NASD lands to various parties, including the island of Vieques, the Puerto Rico Conservation Trust, and the U.S. Department of the Interior. The Navy retained about 100 acres of the former NASD lands to continue operating communication facilities (Navy 2001).
• Residential Lands. Figure 2 labels the central portion of Vieques as "residential lands." This part of Vieques spans approximately 7,000 acres and, prior to May, 2001, bordered Navy property both on the west and the east. It now borders Navy property only on the east. This section of the island houses the entire residential population of Vieques, mostly in the towns of Esperanza and Isabel Segunda. Section III.B describes the demographics of this population in greater detail. Many different land uses are found in this central portion of the island, including residential, agricultural, commercial, and industrial. The industrial land uses, however, are extremely limited, as Section III.E indicates.
• Current Navy Property. Until May 1, 2003, the Navy owned the lands that make up roughly the eastern half of Vieques. As Figure 2 shows, these lands were further divided into two sections. First, the Eastern Maneuver Area (EMA) previously spanned approximately 11,000 acres located immediately east of the residential lands. The Navy used the EMA periodically for various combat activities, such as conducting shore landing exercises and firing at small arms ranges⁽¹⁾ (CH2MHILL and Baker 1999; IT Corporation 2000). The EMA also included Camp Garcia, where Marine Corps and Navy personnel were temporarily stationed at Vieques. Typically, no more than 100 Navy personnel resided at Camp Garcia, but this number increased during training exercises. Sources of air pollution within the EMA were few, and included the small arms firing ranges, wind-blown dust, mobile source emissions, (e.g. vehicles) and releases that occur from sustaining the population in Camp Garcia (e.g., emissions from generators and small boilers, vehicle refueling and maintenance, and other small scale operations).

  East of EMA is the second section of land formerly owned by the Navy, which was called the Atlantic Fleet Weapons Training Facility (AFWTF). AFWTF spanned approximately 3,500 acres (TAMS 1979). As Figure 2 shows, AFWTF was further divided into three smaller sections of land:

  • The western portion of AFWTF was formerly known as the Surface Impact Area. This land is heavily vegetated and almost completely undeveloped, except for dirt roads that pass through the area, a few observation posts and towers, and a larger observation post (OP-1) located on Cerro Matias, near the easternmost portion of this land. Prior to 1978, parts of the Surface Impact Area were used as impact zones for artillery fire.
  • The middle portion of AFWTF is the Live Impact Area (LIA), more commonly referred to as the bombing range. This land spans roughly 900 acres. During military exercises, both aerial bombardment and naval surface fire often took place here. The overwhelming majority of ordnance impacted the LIA, but some bombs and surface fire projectiles landed in the waters near the LIA. The land at the LIA is sparsely vegetated, and did not contain any structures except for "targets" that the Navy periodically placed. The targets were few in number, and included objects such as tanks, small airplanes, and trailers. Section III.D includes much more detailed information about the Navy's bombing practices on Vieques.
  • The eastern tip of AFWTF is the Punta Este Conservation Zone, which has been set aside to preserve sensitive habitats (e.g., turtle nesting areas). No Navy operations took place on this small piece of land.

Not shown in Figures 1 or 2 are terrain features of Vieques, which are important to consider when evaluating how contaminants move through the air. The highest point on the western half of Vieques is Monte Pirata (987 feet above sea level), and the highest point on the eastern half is Cerro Matias (450 feet above sea level), where OP-1 was located. Other than these peaks, the terrain at Vieques includes low rounded hills and an east-west ridge that runs through the residential lands. The average elevation of Vieques is approximately 250 feet above sea level (Cherry and Ramos 1995; Torres-Gonzalez 1989).

B. Demographics

ATSDR examines demographic data, or information on the local population, to determine the number of people who are potentially exposed to environmental contaminants, as well as the presence of any sensitive populations, such as women of childbearing age, children, and the elderly.

Table 1 summarizes demographic data for Vieques, according to the 1990 and 2000 US Census. As the census data show, the population of Vieques increased from 8,602 to 9,106 residents between 1990 and 2000. These figures include both those who live in the residential lands and those who live on Navy property. Table 1 also specifies the number of residents who fall into three potentially sensitive populations: women of childbearing age, children, and the elderly. The table indicates that the percentage of elderly residents in Vieques increased by 2% between 1990 and 2000; ATSDR also notes that the percentage of elderly residents in Vieques (14% in 2000) is notably higher than the percentage of elderly residents living in all of Puerto Rico (11.2% in 2000). ATSDR has received anecdotal accounts suggesting that the population of Vieques is not highly mobile and that many people are lifelong residents of the island, but the site reports that ATSDR has obtained to date do not quantify population mobility trends. ATSDR considered all of the previous demographic figures and observations when evaluating potential exposures among the Vieques residents.

As noted previously, most of the residents at Vieques live in the two largest towns on the island, Isabel Segunda and Esperanza. Although these towns are located relatively close to the Navy property line, they are several miles removed from the LIA. Specifically, the nearest point on residential lands to the geographic center of the LIA is approximately 7.9 miles (or 12.7 kilometers). Therefore, air contaminants from the LIA dispersed over a distance of at least 7.9 miles before they reached the residential populations of Vieques. This was a key issue when evaluating air pollution, as Section V describes further.

C. Climate and Prevailing Winds

The climate and prevailing wind patterns of a given location affect how contaminants move through the air. Annual climatological summaries for Vieques, provided by the National Climatic Data Center (NCDC), indicate that the annual average temperature at Vieques ranged from 77.9 to 80.0 degrees Fahrenheit over a recent 10-year period, with only modest fluctuations in monthly average temperature (NCDC 1985-1994). Annual precipitation totals were more variable, ranging from 42.91 inches in 1991 to 57.07 inches in 1993 (NCDC 1985-1994).

Regarding prevailing wind patterns, a large body of literature reports that trade winds in the Caribbean, which consistently blow from east to west, dominate the meteorology in Puerto Rico. This trend is consistent with wind speed and wind direction data collected at the US Naval Station Roosevelt Roads–the meteorological station closest to Vieques that submits hourly observations of wind speed and wind direction to NCDC. ATSDR obtained more than 10 years of hourly meteorological data for this station. Figure 3 summarizes the hourly wind speed and direction data, in a format known as a wind rose. Wind roses display the statistical distribution of wind speeds and directions in a single plot. The data in Figure 3 demonstrate that the prevailing wind direction at Roosevelt Roads, and presumably in Vieques, is indeed from east to west. In fact, the hourly data provided by NCDC indicate that winds blow from east to west ⁽²⁾ about 75% of the time. This trend is consistent with the influence of trade winds.

Referring to Figures 1 and 2, an easterly wind direction (i.e., winds blowing from east to west) would blow contaminants generated at the LIA toward the residential area of Vieques. This observation, however, does not indicate what levels of air contamination previously occurred. Only sampling data or modeling analyses can provide insights into this issue, as Section V discusses.

D. Navy Operational History

The Navy first began acquiring land on Vieques in 1941 and ceased operations on the eastern half of the island on May 1, 2003. Between 1941 and 2003, a wide range of military training exercises have taken place on Vieques, with the type and intensity of exercises varying from year to year. As a result, the amounts of contaminants released to the air also have changed with time. The following paragraphs note key time frames that ATSDR has defined for purposes of evaluating the extent to which the military training exercises released contaminants into the air. ATSDR's evaluation of air quality issues (see Section V) is based on these time frames.

• 1941 to the early 1970s: Limited military training activities at Vieques. Several reports (e.g., Navy 1971; Rabin 2001; TAMS 1979) indicate that the Navy first acquired land on Vieques in 1941, and continued to acquire lands on the island for several years. Of the many reports ATSDR has reviewed, two suggest that military training exercises first began on Vieques in 1947 (Navy 1971; TAMS 1979), though exercises took place in other parts of the Caribbean before that time. Regardless of the exact date when exercises began at Vieques, ATSDR notes that exercises in the late 1940s were apparently limited to ship-to-shore and land-based exercises, which occurred "only a few weeks a year" (Navy 1971). Military training exercises on Vieques apparently became more frequent in the early 1950s, but these were still limited to ship-to-shore and land-based exercises (Navy 1971; TAMS 1979).

  None of the reports ATSDR has obtained documents exactly when the first air-to-ground exercises took place on Vieques. One report suggests that the Navy first established air-to-ground bombing targets on Vieques in 1960, with actual air-to-ground exercises occurring thereafter (TAMS 1979). Though the early history of air-to-ground exercises on Vieques is not entirely clear, various accounts (e.g., TAMS 1979; Navy 1977) indicate that air-to-ground bombing activity prior to 1971 was far more intense on the island of Culebra than on the island of Vieques. The frequency and intensity of air-to-ground bombing on Vieques gradually increased in the early 1970s, as the Navy slowed and eventually stopped all military training activities on Culebra by 1975.

  ATSDR distinguishes between the time with limited military training activities at Vieques (i.e., from 1941 to the early 1970s) and the time with the most extensive use of the bombing range (i.e., from the early 1970s to April 19, 1999) for purposes of evaluating exposures, as Sections IV and V explain further. Note again that ATSDR has defined these time frames specifically for this PHA and no firm dates mark the transition between this time frame and the one described below.

• The early 1970s to April 19, 1999: Most extensive use of the bombing range, including with "live" bombs. By several accounts, the frequency and intensity of military training exercises on Vieques increased considerably in the early 1970s, after all Navy operations at the island of Culebra ceased. Moreover, air-to-ground exercises with live bombs occurred most frequently from the early 1970s through the 1990s. This period of extensive use of the bombing range ended on April 19, 1999, when two 500-pound bombs were accidentally dropped near an observation post overlooking the LIA, killing a civilian guard.

  Figures 4 and 5 summarize the extent to which the Navy and other parties⁽³⁾ have conducted military training exercises on Vieques between 1983 and 1999–the time frame for which the most complete range utilization statistics are available (Navy 1999). As Figure 4 shows, range utilization statistics indicate that the Navy and other parties conducted exercises on Vieques between 159 and 228 days per year, with the total number of days not varying considerably from one year to the next. (Note, ATSDR reviewed range utilization statistics for 1974-1999, as our response to Comment #19 in Appendix E indicates.)

  Though these usage statistics provide some insights into the number of days when military training exercises took place, the weight of ordnance used during these exercises is a much better indicator of the amount of contaminants that might be released into the air. The graph in Figure 5 illustrates how the total tons of ordnance used at Vieques, as well as the tons of high explosives within this ordnance, have changed from year to year. The range utilization statistics (Navy 1999) suggest that, on average, 1,862 tons of ordnance were used at Vieques annually between 1983 and 1998. This annual amount of ordnance used, on average, contained 353 tons of high explosives. In later sections of this PHA, ATSDR uses these average range utilization statistics to estimate air pollution levels that might have occurred on Vieques during the time when live bombing took place. Refer to Appendix D for the specific inputs that ATSDR considered in its modeling analysis of emissions from military training exercises.

  In addition to researching the usage of ordnance at Vieques, ATSDR considered the extent to which the ordnance was used for different categories of training exercises, namely the proportions used for air-to-ground, ship-to-shore, and land-based activities. Of these activities, air-to-ground bombing accounted for the greatest proportion of high explosives used at Vieques: according to two different reports addressing two different time frames of exercises, 94% of the high explosives used at Vieques were reportedly used for air-to-ground bombing exercises, with ship-to-shore and land-based exercises accounting for the remaining 6% of high explosives (TAMS 1979; IT 2000). These figures indicate that ordnance fired from fixed wing aircraft accounted for the largest portion of air emissions that occur during military training exercises.

  Later sections of this PHA consider the chemical make-up of the various ordnance used at Vieques, as well as the contaminants that might be released after these items impact the LIA.

• April 20, 1999 to May 2000: No military training exercises take place. After the bombing accident occurred on April 19, 1999, the Navy immediately ceased all bombing operations and reviewed the accident and the need for conducting future military training activities on Vieques. After these reviews were completed, President Clinton issued a directive in January 2000 that allowed military training exercises to resume on Vieques, but only using "non-explosive ordnance" (which this document refers to as "practice bombs") and for no more than 90 days per year. No military training exercises took place on Vieques for approximately 13 months, between April 1999 and May 2000.
• May 2000 to May 1, 2003: Military training exercises resume, but only with practice bombs. Starting in May 2000, the Navy resumed its military training exercises on Vieques. These exercises included air-to-ground, ship-to-shore, and land-based activities, but only with practice bombs and other non-explosive ordnance. The Navy completed several military training exercises in 2001, with the main exercises spanning the following dates: February 11 to February 15; April 27 to May 1; June 12 to June 29; August 2 to August 8; and September 21 to October 13. Therefore, in 2001, potential exposures associated with military training exercises using practice bombs occurred on less than 50 days. ATSDR also reviewed range utilization statistics for the exercises that occurred in 2002 and 2003. All military training exercises at Vieques officially ceased on May 1, 2003, when the Navy turned its lands over to the U.S. Department of Interior.
• Specific uses of the LIA that have concerned residents. In addition to concerns about the Navy's more routine uses of the LIA for various military training exercises, residents of Vieques have expressed concern about sporadic uses of specific materials, primarily depleted uranium and chaff, and other activities associated with managing the range, most notably open burning and open detonation of unused waste munitions and unexploded ordnance. ATSDR has obtained the following information on these specific materials and activities:
  • Depleted uranium. During a February 19, 1999, training exercise, ammunition with depleted uranium penetrators was inadvertently loaded aboard two U.S. Marine Corps aircraft that were training at Vieques (NRC 2000). The pilots fired 263 rounds of this ammunition on the LIA during the exercise. The Navy has since worked to identify and recover all detectable depleted uranium penetrators. As of September 2001, the Navy reported having recovered 116 equivalent units, leaving 147 equivalent units not recovered (Higgins 2001). ATSDR has identified no other accounts of depleted uranium usage at Vieques.
  • Chaff. Some residents have expressed concern regarding the Navy's past use of chaff during military training exercises. Chaff is fine aluminum-coated glass fibers that the military has used for many years to confuse radar signals, thus allowing aircraft to operate without being easily detected. The most significant metallic constituents of chaff are aluminum and silicon, though chaff also contains trace amounts of other metallic elements (Naval Research Laboratory 1999).

    The Navy used chaff during military training exercises only with permission from the AFWTF Commanding Officer, and the Navy prohibited chaff from being released directly over the island of Vieques and over the warning and restricted areas that extend several miles from the Vieques shoreline. Though ATSDR has identified several sources indicating that the Navy used chaff at Vieques, none of these sources documents the exact quantities of chaff that were previously used.

  • Open burning and open detonation. Over the years, the Navy used open burning and open detonation to treat two types of wastes: (1) unused waste munitions and (2) unexploded ordnance collected during range clearance activities. The amounts treated differ between these two types of wastes. First, reports the Navy submitted to EPA's Biennial Reporting System indicate that the amounts of unused waste munitions treated in a given year has greatly varied, from zero pounds (in 1993, 1995, 1999) to 30.945 tons (in 1997). Second, an analysis of air emissions from various range management operations indicates that the Navy typically treated 21 tons of unexploded ordnance in open detonation pits per year (IT 2000). This figure is based on waste management statistics for 1998.

The analyses of potential or completed exposure pathways (see Section IV) and evaluations of air quality issues (see Section V) review the public health implications of the different activities described in this section.

E. Other Sources of Air Contaminants

When evaluating the air exposure pathway, ATSDR not only considers emissions from the sources of concern, but also emissions from other sources in the area. This is because residents ultimately are exposed to air contaminants from all local sources, not just those from one or two. At many sites, in fact, air emissions from sources throughout a community far exceed those from a particular site of concern.

When identifying air emissions sources at a given location, ATSDR typically first accesses EPA's Toxic Release Inventory (TRI), a publicly accessible database that documents amounts of toxic chemicals that certain industrial and military facilities release to the environment. As shown in Table 2, which documents the TRI data available for Vieques, only one industrial facility on the island used hazardous chemicals in large enough quantities to trigger TRI reporting. The TRI data for this facility suggest that its air emissions were relatively low, especially when compared to data reported by facilities on the national level. Observations made during ATSDR's site visits (see Section III.F) confirm that industrial operations on Vieques are extremely limited. There are no power plants, chemical manufacturing plants, or other heavy industrial operations on the island.

Though few large industrial sources of air pollution are found on Vieques, numerous small sources of air emissions exist in and near the residential lands. Key among these are transportation sources, including motor vehicles, a small airport, and local ship traffic. Other small-scale sources include gasoline stations, auto refinish shops, construction activities, and a landfill. ATSDR has not identified a representative emissions inventory for the island from any references, thus the exact extent of emissions from these sources in residential lands is not known. Potential impacts of local emissions sources, other than the Navy bombing range, are discussed further in Section V.

In addition to expressing concerns about emissions from the military training exercises, some residents of Vieques asked ATSDR to evaluate the public health implications of exposure to emissions from "African dust storms." These dust storms occur when strong winds blow over the Sahara desert in Africa and carry large quantities of dusts in the upper air winds to locations thousands of miles away, such as the Caribbean islands and the southeastern United States. Many researchers have documented this phenomenon, including those working for the US Geological Survey (USGS), the National Aeronautics and Space Administration (NASA), and the National Oceanic and Atmospheric Association (NOAA) (e.g., Griffin et al. 2001; Taylor 2002).

Some researchers have estimated that these dust storms release as much as one billion tons (1,000,000,000 tons) of dust to the air each year (Moulin et al.1997). This dust is composed of minerals commonly found in the soils and contains many naturally occurring elements, such as lead, iron, mercury, and beryllium. Recent studies have indicated that the dust storms also carry bacteria, fungal spores, and possibly viruses (Griffin et al. 2001). These storms reportedly have the greatest effect on Caribbean air quality during the months of June through October.

To date, community concerns about the African dust storms have fallen into two general categories: Is exposure to the material in African dust unhealthy? What are the relative impacts of emissions sources thousands of miles from Vieques (such as African dust storms) and sources on the island itself (such as emissions from the LIA, motor vehicles, and the limited local industry)? To address these concerns, ATSDR researched many articles on African dust storms published in the scientific literature and consulted with several authors of these studies. ATSDR's interpretations on this issue are documented in Section VI.

F. ATSDR Involvement at Vieques

Since receiving the petition in 1999 to evaluate public health issues at Vieques, ATSDR has worked extensively to characterize and respond to community needs. Many activities to date have provided ATSDR's health assessors critical perspective for evaluating the local air quality issues. Following is a summary of ATSDR's past involvement with this site:

• Site visits. Teams of ATSDR scientists, health educators, and community involvement specialists have conducted more than 10 visits to Vieques since 1999. These visits were conducted for many reasons, such as working with community members to identify health concerns, training nurses on environmental health issues, and identifying sources of air contaminants throughout the island. On two site visits, ATSDR air quality specialists conducted surveys–both on land and by air–of the Navy property. During the land surveys, the specialists extensively toured the EMA and AFWTF, including a driving and walking tour of the LIA.
• Community involvement. Defining community concerns is an essential step in the public health assessment process. To define specific health issues of concern, ATSDR has met several times with residents of Vieques and worked closely with various local individuals and organizations (e.g., elected officials, physicians, nurses, school educators, fishermen, leaders of women's groups). During these meetings, ATSDR also inquired about the most effective ways the agency can provide public health information to the community.
• Health education. Another essential part of the public health assessment process is to design and implement activities that promote health and provide information about hazardous substances in the environment. ATSDR identified health education needs specific to Vieques by conducting a needs assessment in 2001. ATSDR's health education staff have since been developing and offering numerous training sessions and courses on relevant environmental health issues. For instance, ATSDR has facilitated training courses for physicians and nurses and has conducted education sessions on cancer for parents and high school students. Future health education efforts will address specific topics of concern pertaining to Vieques.

The previous list reviews ATSDR's activities while working at Vieques. In addition, ATSDR has invested considerable effort assessing this site's environmental health issues. Most of this work has been conducted at ATSDR's headquarters in Atlanta and is documented in the PHAs listed in Section II.

G. Quality Assurance and Quality Control

To prepare this PHA, ATSDR reviewed and evaluated information provided in the documents listed in the Reference section. The environmental data presented in this PHA are from reports produced by many parties, including ATSDR, EPA, and others. The limitations of these data have been identified in the associated reports, and they are restated in this document, as appropriate. After reviewing the studies conducted to date, ATSDR determined that the quality of environmental data available in the site-related documents for Vieques is adequate to make public health decisions. Appendix C presents ATSDR's specific conclusions regarding the quality of the air sampling studies that have been conducted on Vieques and indicates how the agency factored the findings from these different studies into this document's conclusions.

ATSDR also used an extensive review process for quality control purposes. The review involved numerous parties, including ATSDR scientists, lead authors of several studies cited in this report, and internationally recognized experts in the field of air quality issues and dispersion modeling. To date, all reviewers have agreed that the approaches ATSDR used to evaluate this site are scientifically sound and the available sampling data support this document's conclusions.