Air Pathway Evaluation
SIERRA ARMY DEPOT
HERLONG, LASSEN COUNTY, CALIFORNIA
Sierra Army Depot (SIAD) is located in northeastern California, near the border of California andNevada. SIAD currently conducts many military support activities, including receiving, storing,and distributing munitions, explosives, propellants, and other materials. Until August 2001,SIAD also used open burning (OB) and open detonation (OD) to treat, or destroy, explosives,propellants, and other materials that were considered to be waste. The OB/OD waste treatmentoperations generated large plumes of air contaminants that were visible from locations severalmiles from SIAD. These waste treatment operations ceased in 2001, due to a debate on theapplicability of certain air pollution control rules.
Concerned that air emissions from SIAD might be causing adverse health effects among peoplewho live downwind from the installation, U.S. Senator Harry Reid of Nevada submitted apetition in 2000 asking the Centers for Disease Control and Prevention to evaluate the publichealth implications of potential exposures to air contamination from SIAD. The request wasassigned to ATSDR for response. The Agency for Toxic Substances and Disease Registry(ATSDR) has been gathering and evaluating data to address community health concerns. Thishealth consultation presents ATSDR's response to Senator Reid's petition.
To understand community concerns, ATSDR has met with many residents from California,Nevada, the Susanville Rancheria, and the Pyramid Lake Paiute Tribe. We also have gathereddata on air quality and health issues relevant to SIAD from numerous parties, including: the U.S.Environmental Protection Agency, the California Air Resources Board, the California CancerRegistry, the California Department of Toxic Substances and Control, the Nevada State HealthDivision, the Nevada Division of Environmental Protection, and the U.S. Army.
ATSDR believes the data we collected on air emissions, air pollution measurements, and cancer incidence paint a consistent picture of how SIAD's OB/OD waste treatment operations affectedlocal air quality, and public health. ATSDR's two major findings for this site follow:
- Were residents in the vicinity of SIAD, both in California and Nevada, inhalingunhealthy levels of air contaminants released from SIAD's OB/OD waste treatmentoperations? OB/OD waste treatment operations at SIAD released many contaminantsinto the air, including particulate matter, metals, and chemical by-products of the wastetreatment. The emissions blow primarily in the downwind direction, which is toward theeast, and rarely blow west or towards Susanville. Thus, SIAD's air quality impacts inSusanville are believed to be minimal, and quite possibly not detectable. Even for thedownwind directions, these emissions disperse considerably over the distance thatseparates SIAD from locations where people live. ATSDR reviewed findings fromseveral modeling studies and an extensive air sampling study, all of which indicate thatresidents in the area have not been exposed to levels of air pollution that are associatedwith adverse cancer or non-cancer health effects.
- Are there more cases of cancer in the areas around the SIAD, both in California andNevada, than are to be expected? ATSDR used two independent approaches, which relyon different data sets and are rooted in different scientific disciplines, to assess whetherspecific types of cancer are elevated among residents in communities around SIAD. First,our Division of Health Assessment and Consultation conducted an environmental healthevaluation: they evaluated how much cancer-causing chemicals were released to the airand found that residents in the area are not breathing these chemicals at levels of publichealth concern. Second, our Division of Health Studies conducted a health outcome dataevaluation: they reviewed descriptive data analyzed by the Cancer Registry of NorthernCalifornia, Region 6 of the California Cancer Registry, which overall did not suggestevidence of excess cancers based on the cancer types analyzed in small area assessmentsfor census tracts surrounding SIAD. However, the exception to this finding was a slightexcess of leukemias (all types combined) in the Susanville area for the period 1988through 1997–a finding which did not remain statistically significant when persons withpost office boxes or other non-U.S. Postal Service addresses were removed from theanalysis (Section VI.C explains this process). Because Susanville is more than 30 milesupwind from SIAD, it is extremely unlikely that cancers among Susanville residents arecaused by exposure to SIAD's air emissions. The data currently available from the stateof Nevada does not allow for a meaningful analysis of cancer cases among Nevadaresidents who live downwind from SIAD.
ATSDR acknowledges that our findings for many air contaminants are based entirely onmodeling analyses, which have inherent uncertainties. However, we believe the modelinganalyses were rigorously conducted and likely predicted air concentrations that are higherthan were actually observed. Moreover, for almost every contaminant, the estimated long-term average air concentrations were more than 100 times lower than levels of publichealth concern. Given this ample "margin of safety," ATSDR has confidence that basingthis conclusion on modeling analyses is appropriate. Section VII provides additionalinformation on uncertainties associated with this conclusion.
These two distinct approaches both suggest that levels of air pollution resulting fromSIAD's waste treatment operation are most likely not associated with perceived elevatedtypes of cancer among exposed populations; however, ATSDR acknowledges that theindividual approaches used have inherent uncertainties and limitations.
The remainder of this health consultation explains how we reached our two major findings.Sections II through IV provide background information on SIAD and ATSDR's standardapproach to evaluating environmental health issues. Sections V through VIII evaluate specificcommunity concerns and describe how we interpreted the available air sampling, air modeling,and cancer registry data. Sections IX through XI present ATSDR's main conclusions,recommendations, and a plan for future health actions related to SIAD. Detailed technicalanalyses of selected issues are presented in appendices to this health consultation.
The U.S. Army (Army) owns and operates Sierra Army Depot, where many different militarysupport activities take place. Until August 2001, a major operation at SIAD was treatment oflarge quantities of munitions, explosives, and propellants using open burning (OB), opendetonation (OD), and incineration. SIAD's OB/OD treatment capacity was larger than that of anyother military installation in the United States.
Concerned that air emissions from the treatment operations might cause health problems amongnearby residents, U.S. Senator Harry Reid of Nevada wrote two letters to the director of theCenters for Disease Control and Prevention (CDC) in February 2000 requesting a study on thematter. His first letter asked that CDC investigate a perceived cancer cluster in the area aroundSIAD, and his second letter requested that CDC investigate other illnesses and health conditions,including birth defects, respiratory ailments, autoimmune diseases, and attention deficit disorder.The CDC director, who is also the administrator for ATSDR, referred these requests to ATSDRfor further evaluation.
This health consultation presents ATSDR's response to Senator Reid's letters. Specifically, thishealth consultation addresses the following two key questions, which encompass the manyconcerns expressed to ATSDR:
- Are residents in the vicinity of SIAD, both in California and Nevada, inhaling unhealthy levels of air contaminants released from the treatment activities at SIAD?
- Are there more cases of cancer in the areas around the SIAD, both in California andNevada, than are to be expected?
The remainder of this health consultation presents ATSDR's response to these questions.Specifically, the health consultation summarizes the information we collected on SIAD (seeSection III), and then documents how we interpreted this information (see Sections IV through VIII) to reach our main conclusions (see Section IX).
To answer the two key questions, ATSDR first gathered extensive background information onSIAD, the neighboring communities, and the local environmental setting. The remainder of thissection reviews this background information, which ATSDR critically evaluated to understandwhat contaminants SIAD released to the air, how these contaminants moved through the air, andto what levels people may have been exposed. This section focuses on facts and observationsabout SIAD, without analysis or interpretation. Later sections of this health consultation(particularly Sections V and VI) describe how this background information factored intoATSDR's public health evaluations.
Figure 1 shows the location of SIAD, which is in northeastern California, approximately 40 mileseast-southeast of Susanville, California, and 55 miles north-northwest of Reno, Nevada. Theinstallation spans 96,340 acres, all within Lassen County, California. The border betweenCalifornia and Nevada is less than 4 miles from SIAD's easternmost boundary. Most of theoperations at SIAD take place on two major parcels of land, which Figure 2 shows.
- OB/OD Area. The site documents refer to the northern parcel of land shown in Figure 2by several names, including the OB/OD Area, the Upper Burning and Detonation Area,and the Upper Burning Grounds. The remainder of this health consultation refers to thisarea as the OB/OD Area. This area, which spans approximately 5,350 acres, is whereOB/OD activities takes place. Access to this area is restricted.
- Main Depot. The southern parcel of land shown in Figure 2 is known as the Main Depot,which spans 29,950 acres. Many operations occur at the Main Depot, including receiptand storage of munitions, incineration of certain wastes in the Deactivation Furnace, andvarious other military support activities. Access to this part of SIAD is also restricted.
In addition to these parcels of land, SIAD property also includes a large portion of Honey Lake.The shoreline of this lake fluctuates significantly from one month to the next, depending on theseason, temperature, and precipitation levels. No routine operations occur on this part of SIAD.Access to Honey Lake is not restricted.
Several small cities and communities are located within 15 miles of the SIAD property boundary.These include, but are not limited to, Herlong and Doyle, in California, and Flanigan in Nevada.SIAD is situated in a valley, known as Honey Lake Valley, along the eastern slope of the SierraNevada mountains. The average elevation of this valley is approximately 4,200 feet above sealevel. The Main Depot is located on the valley floor, and the OB/OD Area is located in foothillson the northern edge of the valley. The mountains that surround the valley reach much higherelevations, some higher than 7,000 feet above sea level.
The land surrounding SIAD is used for many purposes, including military, agricultural, ranching,recreation, and residential. Zoning restrictions have left much of the land immediately adjacent toSIAD largely undeveloped. Specifically, Lassen County has zoned all land within 1 mile of SIADboundaries for general agricultural uses with a "Public Safety" restriction (DEIR 2000). Thisrestriction requires prospective property owners to obtain special permits before building homes.
There are several additional land use restrictions in the vicinity of SIAD. For instance, the landimmediately west, north, and east of the OB/OD area are federal lands managed by the Bureau ofLand Management (BLM). Most of these lands are part of the Skedaddle Mountain WildernessStudy Area (WSA) (BLM 1997), where land uses are extremely limited. BLM policies, forexample, indicate that "permitted activities in WSAs (except grandfathered and valid existingrights) are temporary uses that create no new surface disturbance, nor involve permanentplacement of structures" (BLM 1995). Therefore, activities such as building homes and off-roaddriving are prohibited in the WSA.
The Skedaddle Mountain WSA extends east from the OB/OD Area to the border betweenCalifornia and Nevada. East of the Skedaddle Mountain WSA is the Dry Valley Rim WSA,which is located entirely within Nevada. Overall, WSA lands extend approximately 7 miles eastof the OB/OD Area. East of the two WSAs and additional BLM lands is the Pyramid Lake IndianReservation, home of the Pyramid Lake Paiute Tribe. The Reservation lands are at least 14 milesfrom where OB/OD operations take place.
Land uses to the south and southeast of the OB/OD area are less restricted. Outside of the "PublicSafety" restriction lands, the land in this direction is primarily used for ranching, whether onprivately owned lands or lands zoned as open space. The Plumas National Forest is located 15miles southwest of the OB/OD Area. These forest lands are used for recreation, grazing, wildlifemanagement, and other purposes.
ATSDR examines demographic data, or information on the local population, not only todetermine the number of people who are potentially exposed to environmental contaminants butalso to evaluate exposures for sensitive sub-populations, such as children, women of childbearingage, and the elderly.
ATSDR compiled demographic data from the U.S. Census. These data indicate that the landimmediately surrounding SIAD is sparsely populated and 11,725 California and Nevada residentslive within 20 miles of SIAD. Of this population, 6% are children aged 6 and younger; 7% areadults aged 65 and older; and 14% are women between the ages of 15 and 44, which ATSDRconsiders to be of childbearing age. A much larger population (270,580 people) lives within 50miles of SIAD. This radius includes the entire Pyramid Lake Indian Reservation and the cities ofReno, Nevada, and Susanville, California.
When researching the demographics of the area, ATSDR also reviewed site documents toidentify the specific populations living closest to the OB/OD operations. In California, thenearest residences are at two ranches, which are located 1.3 and 3.1 miles from where OBoccurred (DEIR 2000). In Nevada, the nearest populated area identified in the site's risk assessment was Flanigan, which is 11.6 miles from where OB occurred (Brown and RootEnvironmental 1996a). As Section V describes, ATSDR considered these distances whenevaluating air quality impacts from the OB/OD operations.
The climate and prevailing wind patterns affect how contaminants move through the air. TheHoney Lake Valley, where SIAD is located, is arid, typically getting 5 to 6 inches of precipitationper year (DEIR 2000). Temperatures in the vicinity of SIAD vary considerably with season. Inthe summer, daily temperatures range from 64 to 100 oF; for comparison, the average temperaturein winter months is 33 ºF (DEIR 2000).
The prevailing wind patterns determine the directions where air emissions from SIAD primarilyblow. Two meteorological monitoring stations, both operated by SIAD, have collected hourlyobservations of surface wind speeds and directions in Honey Lake Valley. These data werecollected according to SIAD's Meteorological Monitoring Plan, which reportedly has beenapproved by the California Air Resources Board (CARB) and meets EPA requirements forregulatory applications involving meteorological monitoring (Tetra Tech NUS 2001).
Figures 3 and 4 summarize the hourly wind speed and direction measurements in a format knownas a wind rose. Wind roses display the statistical distribution of wind speeds and directions in asingle plot. Figure 3 presents this information for the meteorological station near the incineratoron the Main Depot property, and Figure 4 presents data for the station operated at the BreakShack in the OB/OD area. Though specific trends in wind patterns differ between these two windroses, as is common for locations near complex terrain, both figures indicate that surface windsnear SIAD predominantly blow from west to east. At the Break Shack station, for instance, windsdirections during the afternoon hours, when most OB/OD operations occur, were roughly fromwest to east 75% of the time (TetraTech NUS 2000)(1). For this same subset of hours, windsblowing from east to west occurred only 10% of the time.
Though these observations clearly demonstrate that the prevailing pattern of surface winds isfrom west to east, prevailing wind patterns at the surface may differ from those aloft. How winddirections vary with altitude is an important consideration for SIAD, because plumes fromOB/OD operations have been observed to rise more than 2,000 feet above the ground (Brown andRoot Environmental 1996a). One study examined upper air wind patterns by placing surveillance equipment on weather balloons, which observed wind directions during the afternoon hours atseveral different altitudes (Tetra Tech NUS 2000). This study found that prevailing winds at allaltitudes were from directions between the south and west-northwest, and winds blowing fromthe east were again very rare.
Overall, the prevailing wind patterns at SIAD generally blow from west to east, which wouldcarry air emissions from SIAD predominantly toward the state of Nevada. These emissions blowtoward the communities west of the installation infrequently.
Since it was constructed in 1942, SIAD has conducted various military support activities. SIAD'smission has consistently been to receive, store, transport, repair, and treat many types ofmunitions, explosives, propellants, and other materials. Treatment (or destruction) of thesematerials has been achieved primarily through OB, OD, and incineration. Although SIAD hasother sources of air pollution (e.g., storage tanks, motor vehicles, boilers), this healthconsultation focuses exclusively on SIAD's waste treatment operations, which account for themajority of SIAD's air emissions and are most likely to transport long distances.
SIAD operated, and continues to operate, under various environmental regulations and permittingauthorities. Permitting of waste management operations fell under EPA's Resource Conservationand Recovery Act (RCRA). From 1980 to 2003, SIAD operated its OB/OD operations as an"interim status" facility under RCRA. It is not uncommon for certain operations to remain underthis status for many years. SIAD applied for a RCRA permit in the early 1990s. That applicationhad been under various states of review, until SIAD officially withdrew the application in May2003.
Air emissions from SIAD fell under the federal Clean Air Act, among other regulations. SIADfirst applied for a "Title V" air permit in 1996, as required by the 1990 Clean Air ActAmendments. The installation received its first 5-year permit in 1998. In 2001, that permit wasreissued with minor changes. One notable change that occurred since then resulted from a legalsettlement that prevented SIAD from conducting routine OB/OD operations. Since 2001, theinstallation has been permitted to use OB/OD only under emergency situations or for nationalsecurity reasons.
The rest of this section describes key features of SIAD's waste treatment operations:
- What waste materials has SIAD treated? SIAD has treated a wide range of wastematerials, both from military and non-military waste generators. The materials includebombs, warheads, rocket motors, propellant charges, grenades, and mines, all of whichcontain different components (e.g., metal casings, explosive charges, propellants). Thematerials are considered to be "waste" for various reasons: they may have exceeded theirshelf-life, they may not have been built to specifications, or they may have becomeobsolete. Though SIAD has treated a wide range of conventional weapons, it has nottreated nuclear, chemical, or biological weapons, and does not treat radioactive wastes.Site documents state, for example, that "radioactive items, including depleted uraniumrounds, are not treated [in the OB/OD area] under any circumstance" (DEIR 2000).
- How much waste material has SIAD treated? The amount of waste material SIAD treatedvaried from year to year. Until August 2001, the installation's waste treatment permitallowed SIAD to destroy 30,000 tons of waste material in OB/OD operations per year(DEIR 2000), but the actual amounts of waste treated were typically lower. Figure 5 shows the total amount of waste material that SIAD treated per year from 1990 to 2001.
- What waste treatment technologies didSIAD employ? The text box on thefollowing page describes, in generalterms, how OB, OD, and incinerationsdestroys wastes, and the remainder ofthis section explains how theseoperations were specifically applied atSIAD.
- OB waste treatment operations.According to site documents,SIAD first used OB to treatwaste materials in 1950 (DEIR2000), and this practice continued until 2001. The nature and extent of OBactivities peaked during the 1990s. Site documents provide extensive insights onOB activities in more recent years.
- OD waste treatment operations. Like the OB operations, OD operations first began at SIAD in 1950 (DEIR 2000). In the 1950s, only small amounts of ammunition were treated at the installation, and these amounts came from the installation's existing stockpile. In the 1960s and 1970s, the treatment activity increased, and approximately 20 employees worked seasonally on the OB/OD grounds. During this time, however, the installation still treated only waste materials from the installation's existing stockpile. In the early 1980s, SIAD was designated one of the primary demilitarization sites for OB/OD. From the early 1980s through 2001, ammunition started being shipped to SIAD for OB/OD treatment. In the late 1980s, approximately 60 employees worked on the OB/OD grounds (Holsey 2003). The OB/OD treatment activity peaked in the 1990sthe time frame shown in Figure 5.
- Incineration waste treatment operations. Site documents indicate that SIAD hasoperated two incinerators on the Main Depot since 1942. One incinerator treatedwaste materials from 1942 until it was dismantled in the mid-1950s (DEIR 2000).Since little information is available on the design of this incinerator and theamount and types of wastes that it treated, and exposures to air emissions fromthis operation ceased approximately 50 years ago, this health consultation doesnot address this incinerator further.
- When did the waste treatments occur? SIAD treated wastes at various times during theyear, but primarily during the spring, summer, and fall. OB/OD waste treatments werelimited to the daytime hours. SIAD implemented several standard operating proceduresthat further limited when waste treatments would occur. These limitations prohibitedtreatments from occurring during thunderstorms or on days with calm winds, limitedvisibility, or air quality alerts (Brown and Root Environmental 1996a).
The amounts treated in 2001 are the lowest for the time frame considered, because thefacility ceased all treatment operations in August 2001, when Lassen County AirPollution Control District informed SIAD that continued OB/OD operations would beconsidered a violation of the district's air quality regulations. Since August 2001, the onlyOB/OD operations that have been permitted at SIAD are those for emergency purposes ornational security reasons.
Between 1990 and 2000, theinstallation treated an average of19,000 tons of waste material (grossweight, see text box) per year, or 64%of the maximum treatment amountallowed by its permit. Site statistics suggest that OD operations accountedfor the majority of the waste materialtreated, but the relative amount treatedby OB, OD, and incineration changedfrom year to year. Until routine OB/ODoperations ceased in August 2001,SIAD had the largest OB/OD treatmentcapacity of all military installations inthe United States (DEIR 2000).
These activities occurred exclusively in the OB/OD Area (see Figure 2). OBoccurred both in "pits" and "pans," as described below.
OB pans are made of carbon steel and were used primarily for treating smallerquantities of solid propellants, which typically contained nitrocellulose,nitroglycerine, and nitroguanidine in greatest quantities. Most pans are between 15and 20 feet long, and up to 8 feet wide. Before OB operations ceased in 2001,SIAD operated 150 burn pans, which were divided among 30 burn stations.Permit requirements limited SIAD to burn no more than 1,000 pounds of NEW inthese pans during each event. These burns typically lasted 2 minutes or less, after which waste ash was collected and handled according to solid waste managementrequirements.
OB was conducted in pits to treat much larger quantities of propellants–up to160,000 pounds of NEW during a single event. These OB operations, which wereprimarily used to treat large rocket motors, generally lasted up to 10 minutes, butsome took longer due to the large amount of propellants being treated. Theprincipal materials treated in these operations include ammonium perchlorate andnitroglycerine. Residual ash and rocket motor casings left after such OB eventswere collected, recycled, or disposed of according to solid waste managementregulations.
All OD operations took place in the OB/OD Area (see Figure 2) in 14 pits, whichwere dug into the sides of hills. For each detonation, waste material is placed intothe pit and then detonated from a safe distance. SIAD was allowed to conduct upto two detonations per day in each pit, and each detonation could treat no morethan 10,000 pounds of NEW (DEIR 2000). SIAD personnel collected andrecycled scrap metals that remain after the detonations.
SIAD constructed another incinerator, known as the Deactivation Furnace, on theMain Depot. This incinerator treated waste material from 1971 until 1989, afterwhich it was closed for 3 years while SIAD obtained the appropriate operatingpermits. In 1992, the incinerator began operating again, but only to treat wastesclassified as "non-hazardous" by EPA's waste management regulations (DEIR2000). These non-hazardous wastes include small arms munitions, fuzes, somegrenades, and detonators. The incinerator treats far less quantities of wastematerials than the OB/OD operations. In 1996, for instance, SIAD treated 6 tonsof non-hazardous waste munitions by incineration, which accounted for less than0.1% of the total waste material treated at the depot.
The incinerator is equipped with air pollution controls, including a cyclone and abaghouse, that reduce the amount of pollution released to the air. Site documentssuggest that these controls reduce air emissions of several metals (barium,chromium, lead, antimony, beryllium) by more than 99% (DEIR 2000), whileothers (copper, manganese, nickel) are controlled less effectively.
Though these operating procedures likely reduced air quality impacts from OB/ODoperations, recent press accounts reported that SIAD did not always adhere to theprocedures (Reno Gazette-Journal 2000). ATSDR notes that the operating procedureswere internal guidelines, not statutory requirements of air permits. The frequency withwhich the standard operating procedures were not followed is not known.
The previous paragraphs provide relevant background information on the three waste treatmentoperations that SIAD has recently employed. Section V refers to this background informationwhen evaluating the air quality impacts of contaminants released by these operations.
When evaluating the air exposure pathway, ATSDR not only considers emissions from thesources of concern (in this case, OB/OD operations at SIAD), but also considers emissions fromother sources in the area. We do this because community members ultimately are exposed to aircontaminants from all local sources, not just those from one or two.
The area surrounding SIAD is sparsely populated and contains few air emissions sources,especially when compared to urban and industrial settings. The Lassen County Air PollutionControl District has estimated emissions from major industrial sources (DEIR 2000), but most ofthe sources identified are at least 10 miles from SIAD. Air emissions sources in the vicinity ofSIAD are limited, and include wind-blown dust and motor vehicle exhaust.
With one exception, which is discussed in the following paragraph, air emissions from SIAD'sOB/OD operations far exceed those from all other local sources. Recent press accounts haveacknowledged this (Sacramento Bee 2001), citing air emissions data that SIAD submitted toEPA's Toxic Release Inventory (TRI). Specifically, the TRI data EPA originally released forreporting year 1999 indicate that SIAD emitted more air emissions of toxic chemicals than didany other industrial or federal facility in the state of California (EPA 2001). SIAD has sincerevised its air emissions estimates, and the installation's emissions no longer rank among the top100 in the state (EPA 2002a). Section V.A of this health consultation discusses this revision inTRI data in greater detail.
Other than SIAD, one notable emissions source that has been found to affect local air quality iswildfires. SIAD can expect to experience several small wildfires annually in the area surroundingthe installation; these fires are caused either by lightning or human activity. The area typicallyexperiences one or two large wildfires each year at some location within a 60-mile radius of theinstallation (Holsey 2003). According to EPA emissions data, a wildfire in California can releaseover 150 tons of particulate matter to the air per day,(2) in addition to releasing various othercontaminants (EPA 1996). Because wildfires release large amounts of contaminants over shorttime frames, emissions from wildfires often far outweigh those from all other local sourcescombined. This trend was recently observed during an ambient air monitoring program at SIAD,when some of the highest levels of air pollution detected occurred when a wildfire burned out ofcontrol in the nearby Feather River Canyon (TetraTech NUS 2001). Section V.C revisits thisissue.
Overall, during the 1990s, OB/OD operations at SIAD were clearly the dominant local airemissions source over the long term; however, emissions from wildfires during that time stillexceeded those from all other sources near SIAD over short time frames. Since 2001, OB/ODoperations at SIAD have had only minimal air quality impacts in the area.(3) Section V of thishealth consultation evaluates the public health implications of all air emissions sources in thevicinity of SIAD.
Since receiving Senator Reid's letters in February 2000, ATSDR has conducted many activitiesto identify community concerns, understand the local environmental setting, and evaluate airquality impacts from SIAD's waste treatment operations. These activities began with a site tourin June 2000, when ATSDR environmental health scientists and community involvementspecialists visited SIAD, at a time when waste treatment operations were occurring. Our staffmembers met with local community members, tribal government officials from the SusanvilleRancheria in California and the Pyramid Lake Paiute Tribe in Nevada, and other groups. We alsomet separately with representatives of the California Cancer Registry of Northern California andof the Nevada Bureau of Disease Control and Intervention. Following the site visit, ATSDRdeveloped and distributed a fact sheet describing our involvement with the site.
From October 2000 through the present, ATSDR environmental health scientists have beenobtaining and interpreting site documents. Several critical developments have occurred duringthis time, including the release of the only ambient air monitoring study conducted duringOB/OD waste treatment activities (TetraTech NUS 2001) and the cessation of routine OB/ODactivities in August 2001. ATSDR considered information published as recently as August 2002 when preparing this health consultation.
ATSDR reviewed and evaluated information provided in the documents referenced in SectionXIII. The environmental data presented in this health consultation were taken from reports andanalyses produced by many parties, including EPA, SIAD, the California Cancer Registry, theNevada State Health Division, and others. The limitations of these data have been identified inthe associated reports, and they are restated in this document, as appropriate. After reviewing thestudies conducted to date, ATSDR determined that the quality of environmental data available inthe site-related documents for SIAD is adequate to make public health decisions, except asotherwise noted. Sections V and VI present ATSDR's specific conclusions regarding the qualityof the air sampling and modeling studies and cancer registries and describe how these differentstudies' findings factored into our conclusions.
ATSDR also used an extensive review process for quality control purposes and to ensure that ourevaluations are scientifically sound. The review involved numerous parties, including ATSDRscientists, state environmental and health agencies, and lead authors of several studies cited inthis report. Our final health consultation will be issued after we have received and addressed all comments.
This section of the health consultation addresses the inhalation exposure pathway to aircontaminants, focusing specifically on where air emissions from SIAD go and who might comeinto contact with them. Analyzing exposure pathways is important because:
- If people are not exposed to a site's environmental contamination, then the contaminantscannot pose a public health hazard and additional analyses are not necessary.
- If people are exposed to site-related contamination, then further analysis is needed tocharacterize that exposure. Just because exposure occurs does not mean that people willhave health effects or get sick. In fact, for many chemicals, environmental exposures areoften far lower than the exposures that people experience through their diets and perhapsthrough their occupations. Several questions must be answered to understand the publichealth implications of exposure: To what chemicals are people exposed? How often arepeople exposed, and for how long? At what levels are people exposed? These are justsome of the issues that ATSDR considers when assessing whether harmful health effectsmight result from exposure.
The remainder of this section describes how ATSDR assessed inhalation exposures forcommunities near and downwind from SIAD (Section IV.A) and reviews the process ATSDRused to evaluate the inhalation exposures (Section IV.B).
One of the first steps in ATSDR's health consultation process is to identify populations that aredefinitely or potentially exposed to a site's contamination. We do this by reviewing five elementsthat together make up an exposure pathway. These five elements, and how they relate to airemissions from SIAD, follow:
- Source of contamination. A source of contamination must exist in order for exposures tooccur. The OB/OD waste treatment activities at SIAD released large quantities of airpollutants. Thus, a source of contamination clearly existed for this site.
- Environmental media and transport. People cannot be exposed unless contaminantsmove from their source or origin through the environment to an exposure point. A recentair sampling study at SIAD found that emissions from the OB/OD area affected airquality at locations within 5 miles of the installation boundary (see Appendix C.1), but itis difficult to determine exactly how far away actual air quality impacts occurred. Werecognize that most dispersion models predict that emissions sources generally have airquality impacts for many miles downwind, but the magnitude of these impacts often is fartoo small to measure.
- Point of exposure. Exposure cannot occur unless contaminants reach a location wherepeople have access. The main modeling study published for this site predicts that somecontaminants from the OB/OD area can transport to most locations in the Honey LakeValley, so a point of exposure clearly exists.
- Route of exposure. For exposure to occur, people must contact chemicals in acontaminated media, either through inhalation, ingestion, or dermal contact. Inhalationexposures clearly occur if air contamination is present.
- Potentially exposed population. Ultimately, people must come into contact withchemicals at the point of exposure in order for ATSDR to conclude that exposures haveoccurred. Our analyses of demographics in California and Nevada confirm that peoplelive within the 50-mile radius that is the focus of this study, thus the condition of apotentially exposed population is met.
Knowing that residents who live more than 20 miles away from SIAD have expressedconcern about the OB/OD emissions, ATSDR decided to consider all populations withina 50-mile radius as potentially exposed. This decision does not mean that everyone within50 miles of SIAD is actually exposed to the air emissions. It is rather a decision we madeto frame the analyses for this health consultation. We use the term potentially exposedbecause the available air sampling and modeling studies suggest that air concentrationsresulting from SIAD's emissions are quite limited and perhaps not detectable beyond 10miles from the OB/OD area. Thus, this health consultation only examines the possibilitythat residents who live within 50 miles of SIAD are exposed to the OB/OD emissions.
By focusing our analyses on a 50-mile radius, we include among the potentially exposedpopulations residents as far upwind as Susanville, CA, and as far downwind as thePyramid Lake Indian Reservation. On the other hand, we are assuming that residents ofReno, Nevada, are not exposed to the OB/OD air emissions. ATSDR based this decisionon its experience evaluating other sites where residents are concerned about long-rangeatmospheric transport of emissions and on two site-specific observations: Reno is not inthe prevailing downwind direction from SIAD (see Figures 2, 4, and 5), and past airquality problems in Reno have been attributed primarily to emissions sources found in theReno area (see Appendix C.3).
Section V describes how we reviewed the available air sampling and dispersion modeling studiesto assess the health implications of potential exposures to air contaminants released by SIAD.
ATSDR used established methodologies to determine the public health implications of exposureto air contaminants from SIAD. Specifically, we followed a three-step approach when addressingthe potential exposures described previously: (1) identify concentrations of contaminants releasedto the air, (2) select chemicals for further evaluation by screening the concentrations againsthealth-based comparison values, and (3) perform toxicologic evaluations for those contaminantsselected for further evaluation.
The first step involves tabulating air concentrations for site-related contaminants. ATSDR prefersto use actual measurements (i.e., air sampling results) for this step, rather than relying onengineering calculations or predictions from air quality models. This preference results from thefact that air quality models estimate ambient air concentrations, sometimes with great degrees ofuncertainty, while sampling studies measure ambient air concentrations. However, air qualitymodels are critical tools in cases when exposures may have occurred during time frames when,and at locations where, sampling did not. Section V.C reviews our best estimates of exposureconcentrations. Our evaluation includes different averaging times: 24-hour average levels areused to evaluate acute exposures (i.e., those that occur over the short-term), while annual averagelevels are used to evaluate chronic exposures (i.e., those that occur over the long-term). Theambient air concentrations presented in Section V.C draw from the best available information forthis site, which is a combination of measured and estimated air contamination levels.
The second step in evaluating exposure pathways is selecting contaminants for furtherevaluation. This is accomplished by comparing the ambient air concentrations for site-relatedcontaminants to health-based comparison values. Comparison values are developed from thescientific literature concerning exposure and health effects. To be protective of human health,most comparison values have large safety factors built into them. For some chemicals, the safetyfactors are quite large (a factor of 100 or greater). As a result, ambient air concentrations lowerthan their corresponding comparison values are generally considered to be safe and not expectedto cause harmful health effects, but the opposite is not true: ambient air concentrations greaterthan comparison values are not necessarily unhealthy levels of air pollution. Rather, chemicalswith concentrations higher than comparison values require further evaluation. Chemicals withoutpublished health-based comparison values are automatically considered as requiring furtherevaluation. The text box on the following page presents the approach ATSDR used to selectcomparison values for this health consultation.
The final step in the assessment methodology is evaluating the public health implications ofexposure to any contaminants identified as requiring further evaluation. For these contaminants,ATSDR puts the public health implications of exposure into perspective by considering site-specific exposure conditions and interpreting toxicologic and epidemiologic studies published inthe scientific literature. Thus, this step is a state-of-the-science review of what the exposure levels mean in a public health context.
To assess potential air exposures for this site,ATSDR addressed three questions: Whatcontaminants did SIAD release to the air(Section V.A)? How did these contaminantsmove through the air to where people live(Section V.B)? Were residents exposed tocontaminants at levels that might causeadverse health effects (Section V.C)? Theanswers to these questions were importantfactors to consider when addressingcommunity concerns regarding cancer(Section VI), because the exposureevaluations tell us which populations wereexposed, to what chemicals (includingcarcinogens), and for how long.
To characterize air emissions from SIAD, ATSDR first identified contaminants that are releasedto the air and then obtained estimates for the rates at which these emissions occur. An importantfirst step in this process was to understand exactly what happens during typical OB/ODoperations. Figure 6 illustrates what emissions occur during a typical OD event. Referring to thisfigure, the following paragraphs identify the categories of contaminants released to the air, anddescribe qualitatively how emissions may differ between OD and OB.
- Particulate matter (PM). As Figure 6 shows, OD events are explosions at ground level. The explosions release large amounts of energy, which cause some of the nearby soils and fragments from the waste material to eject to the air. Much of the soils and fragments that become airborne fall back to the ground near the OD pit, but some particles (or PM) remain airborne and can transport downwind. Though Figure 6 depicts a typical OD event, PM emissions also occur when rocket motors are burned and during OB events. OB of wastes in burn pans release a smaller quantity of PM, because the energy released during the waste treatment does not cause large quantities of soils to become airborne.
- Metals and inorganic compounds. During an OD event, metals and other inorganic materials may be emitted from several sources. For instance, metallic casings from munitions fragment after a detonation, and some fine particles containing these metal fragments can become airborne. Further, the soil ejected into the air in an OD event contains metals, both naturally occurring metals and those that remain in the soil from past releases. Finally, some mixtures of explosives and propellants contain metals, such as aluminum dust, which can become airborne during an OB/OD event. Table 1 lists the metals that are most commonly emitted from SIAD's OB/OD operations.
- Explosives, propellants, and fillers. The explosives and propellants in the waste material essentially provide the "fuel" for the OB/OD events. In OD, for example, explosives are remotely detonated, which then triggers the chemical reactions that rapidly consume these materials and any other organic fillers in the waste. These chemical reactions release large amounts of energy as the chemical bonds from the explosive and propellant molecules break, thus forming smaller, more stable molecules, such as water and carbon dioxide. Data collected during OB/OD events conducted in controlled settings suggest that more than 99.9% of explosives and propellants in munitions waste are typically destroyed during waste treatment (Bjorklund et al. 1998; Brown and Root Environmental 1996a). In other words, OB/OD destroys almost all of the explosives, propellants, and fillers in waste munitions, but small amounts of these chemicals are released to the air. Table 1 lists the explosives, propellants, and fillers found in largest quantities in the waste material treated at SIAD.
- OB/OD chemical by-products. During OB/OD events, chemical reactions not only consume the explosives, propellants, and fillers, but they also form numerous organic and inorganic chemical by-products. The overwhelming majority of these by-products are relatively benign from a public health perspective. Examples include water vapor, nitrogen, and carbon dioxideall of which are relatively abundant in the atmosphere. However, incomplete combustion of the waste material also generates trace amounts of toxic chemicals. Table 1 lists the OB/OD chemical by-products believed to be released in greatest amounts by SIAD's waste treatment operations.
The chemical composition of the waste being treated largely determines the chemical by-products released from a given event. The materials treated by the majority of OB/OD operations are explosives and propellants, most of which are molecules containing carbon, nitrogen, oxygen, and hydrogen. To a first approximation, the chemical by-products will also be composed of these elements. OB of rocket motors, on the other hand, has notably different emissions because the propellants contain high levels of chlorine and aluminum. ATSDR considered this when evaluating the public health implications of the air emission rates.
Although Table 1 identifies the contaminants that SIAD's waste treatment operations release in greatest quantities, the actual amounts of contaminants released to the airor emission ratesare better indicators of the potential air quality impacts of a given source. ATSDR obtained air emissions data for SIAD from two sources: EPA's Toxic Release Inventory (TRI) and the human health risk assessment SIAD prepared for its air permit application (Brown and Root Environmental 1996a). More information on these data sources follows.
Table 2 presents the air emissions data that SIAD reported to TRI. SIAD's original air emissions data submitted to TRI received local press attention for being higher than those for any other facility in California (Reno Gazette-Journal 2001b); however, SIAD has since submitted revised data and the installation no longer ranks among the state's top 100 polluters. The revised TRI submission had dramatically lower emissions estimates for aluminum, copper, and zinc. Fortunately, ambient air monitoring data are available for these metals, and the measured concentrations (see Section V.C) strongly suggest that SIAD's original TRI submissions are indeed gross overestimates of the installation's actual air emissions. (The text box, on the following page, provides additional information on the recent changes to SIAD's TRI emissions estimates.)
In addition to the emissions data SIAD submitted to TRI, the installation has recently reported air emissions data for nearly 100 air contaminants in a human health risk assessment prepared to accompany a permit application (Brown and Root Environmental 1996a). Table 1 identifies all contaminants with estimated emissions rates greater than 10 pounds per year. Appendix D.1.1 describes how these emission rates were estimated, and presents ATSDR's critical review of the emissions estimation algorithm. ATSDR found that the emissions data reported in the human health risk assessment are based on several conservative assumptions that likely overstate actual emission rates. The remainder of this section addresses the potential air quality impacts from the contaminants listed in Tables 1 and 2 (and others considered in the human health risk assessment).
Although SIAD's waste treatment operations have released numerous pollutants in varyingquantities, it does not necessarily follow that all residents in the area have been continuouslyexposed to the pollutants listed in Table 1. In fact, the air emissions will disperse greatly over thedistance that separates the OB/OD area from the nearest residential locations. Localmeteorological conditions largely determine where SIAD's air emissions blow and how rapidlythey disperse.
As Section III.D explains, the prevailing surface wind direction at SIAD is from west to east, andwinds rarely blow from east to west (see Figures 3 and 4). The prevailing wind directiontherefore blows waste treatment emissions from SIAD over the Skedaddle Mountain WSAtoward the state of Nevada. Winds may also blow SIAD's air emissions in other directions, butthis occurs far less frequently. For instance, wind direction observations suggest that windsduring the afternoon hours, when most OB/OD operations occur, blow from east to west roughly10% of the time. This trend suggests that air emissions from approximately 1 out of every 10days with OB/OD operations will blow toward the California communities of Milford, Wendel,and Susanville. Qualitatively, SIAD's air emissions affect these cities infrequently.
Air models can provide quantitative insights of how SIAD's emissions move through theatmosphere to downwind locations. ATSDR and multiple California agencies critically reviewedmodeling studies conducted by contractors to SIAD. Appendix D summarizes in detail ATSDR'scritical review of the accuracy and validity of the available modeling studies. As Appendix Dindicates, ATSDR believes the modeling studies provide useful information on upper-boundexposure concentrations for residents who live near SIAD. The modeling analyses report thatemissions from SIAD's waste treatment operations have their greatest impacts in theunpopulated, mountainous region, directly east of the OB/OD area (TetraTech NUS 2000). Theresidential locations believed to have the highest air quality impacts are Skedaddle Ranch(California) and Flanigan (Nevada) (Brown and Root Environmental 1996a). The followingsection evaluates whether air pollution at these and other locations reached levels known to beassociated with adverse health effects.
ATSDR thoroughly reviewed available air sampling data and air modeling data to evaluate thepublic health implications of exposures to SIAD's air emissions. Our technical reviews of theavailable sampling and modeling studies are included as Appendix C and D, respectively. Weevaluated potential air quality impacts for nearly 100 chemicals, and considered the bestavailable information on acute (or short-term) exposures and chronic (or long-term) exposures.Detailed findings, organized by category of air pollutant, follow:
- Particulate matter. Modeling and sampling studies have estimated and measured ambient air concentrations of particulate matter in the vicinity of SIAD. Both types of studies characterized particulate matter in terms of PM10, or airborne particles and droplets with diameters smaller than 10 microns. The available data suggests that ambient air concentrations of PM10 do not reach levels of health concern where people live, with elevated short-term levels occurring only in the uninhabited areas immediately east of the OB/OD area.
- Metals and inorganic compounds. Consistent with the approach used to evaluate air concentrations of particulate matter, ATSDR reviewed available modeling analyses and sampling results to assess inhalation exposures to metals and inorganic compounds, considering both long-term and short-term exposures.
- All estimated air concentrations in the dispersion modeling analysis are based on the assumption that SIAD treated the maximum amount of waste materials allowed by its air permit. Actual waste management data, however, indicate that SIAD treated approximately 2/3 of the maximum allowed amounts.
- Emissions data for metals were estimated by assuming that the entire weight of bomb casings vaporize during OD operations, even though considerable amounts of metal fragments fall to the ground following OD events. This assumption caused the air model to overstate air concentrations of metals.
- Ambient air concentrations measured during the recent air sampling study (TetraTech NUS 2001) were consistently lower than the values predicted by the dispersion modeling analysis. For instance, the highest average ambient air concentrations of aluminum and iron out of all sampling locations considered were 0.76 and 0.85 µg/m3, respectively(6). These values are roughly a factor of two or more less than the average levels predicted by the models. This comparison suggests that the concentrations calculated by the dispersion model (Table 3) are higher than actual exposure concentrations.
- With three exceptions, the estimated air concentrations listed in Table 1 are more than 100 times lower than the corresponding health-based comparison values. Therefore, even if the emissions estimates used in the dispersion modeling analysis understate actual emissions by a large factor (as much as 100), the estimated air concentrations would still be lower than levels of public health concern. (As Appendix D explains, however, ATSDR believes the modeling analysis actually overestimated air emissions of metals.)
- Explosives, propellants, and fillers. ATSDR evaluated potential air quality impacts of 18 chemicals that comprise the explosive and propellant charge in the majority of waste materials that SIAD treats. Our evaluation for these chemicals is based entirely on modeling results, because no sampling studies have been conducted to measure actual ambient air concentrations of these contaminants. Although ATSDR would prefer to base these conclusions on measurements, the lack of sampling data is not a critical data gap because OB/OD waste treatment operations destroy the majority of the explosives, propellants, and fillers in the waste materials, rather than causing them to be released into the air. In fact, the large plumes that form following OB/OD events result largely from the energy released when chemical bonds in explosives and propellants are broken.
- OB/OD Chemical By-Products. As explained previously, OB/OD events break explosives and propellants into much smaller molecules, thus releasing large amounts of energy. The majority of chemicals formed during these events are relatively benign. Incomplete combustion of the explosives and propellants form trace amounts of other chemicals, including a wide range of volatile organic compounds and semi-volatile organic compounds. Table 5 lists 55 chemicals that have been identified in OB/OD tests conducted in controlled settings (see Appendix D.1). ATSDR evaluated whether ambient air concentrations of these 55 chemicals reach levels of public health concern near SIAD.
Regarding long-term exposures, a dispersion modeling analysis for the human health risk assessment (see Appendix D.1) predicted that SIAD's waste treatment operations would cause relatively small increases in PM10 concentrations at most places where people live (Brown and Root Environmental 1996a). The model suggests that SIAD's air emissions would cause PM10 levels to increase by 11 µg/m3 at the Skedaddle Ranch in California and by 2 µg/m3 at Flanigan, Nevada. These two locations were predicted to have the greatest air quality impacts from SIAD's emissions. Given that average "background" PM10 levels in remote areas typically fall between 20 and 25 µg/m3 (EPA 2002e), the modeling analysis suggests that actual annual average PM10 levels were likely no greater than 36 µg/m3 (at Skedaddle Range)a level that is considerably lower than EPA's health-based air quality standard (50 µg/m3) for exposures to particles of this size. Moreover, it is important to note that these predicted increases would occur only if SIAD operated at its maximum capacity, which never occurred. As a result, the predicted air quality impacts were likely greater than what was actually observed. In summary, long-term average PM10 concentrations estimated by the dispersion model are not of public health concern.
Results from air sampling studies support the findings of the modeling analysis. First, a recent field study at SIAD (TetraTech NUS 2001) collected nearly 250 PM10 air samples at 12 locations near the installation; three of the sampling locations were designated as being near populated areas (Susanville, Patton Village, Pyramid Lake) and two locations were near Skedaddle Ranch. As Appendix C.1 explains, ATSDR believes that the field study adequately characterized air quality impacts from typical OB/OD operations at places where people live.(4) Based on 22 days with valid air samples, the average PM10 levels were 26 µg/m3 in Susanville, 19 µg/m3 in Patton Village, and 26 µg/m3 at the Pyramid Lake Indian Reservationall lower than EPA's annual average health-based standard. Further, at the two sampling locations closest to Skedaddle Ranch, average PM10 levels were 28 and 35 µg/m3. Therefore, both predicted and measured long-term average concentrations of PM10 in residential locations are not at levels of health concern. It is important to note here that this recent sampling study characterized air quality impacts during what were once typical operation activities at SIAD, and not during a time when OB/OD operations were at a lull. Refer to Appendix C and D for more information on the monitoring and modeling studies that support our finding.
ATSDR also examined the public health implications of estimated and measured short-term ambient air concentrations of PM10. According to the most extensive modeling study for SIAD (Brown and Root Environmental 1996a), the highest 1-hour average PM10 concentration predicted for a residential location was 18,300 µg/m3, at Skedaddle Ranch in California. As Appendix D.2 explains, ATSDR believes this modeling prediction grossly overstates actual ambient air concentrations because the model evaluated a highly unrealistic waste treatment scenario. Specifically, the short-term peak concentrations were calculated assuming that SIAD simultaneously treats waste material in all 14 OD pits and in all 30 OB burn pans, that the amounts of wastes simultaneously treated are the maximum allowed in the installation's air permit, and that this maximum treatment scenario occurs in the hour with the least favorable meteorological conditions for atmospheric dispersion. Because this combination of events is not expected to ever occur at SIAD, ATSDR concludes that the short-term modeling predictions are not representative of actual exposure conditions.
ATSDR's conclusion regarding short-term exposures to PM10 is therefore based on the available sampling results. During the 2000 air sampling study at SIAD, only 2 of the 238 valid 3-hour average PM10 measurements (or 0.008%) exceeded EPA's 24-hour average health-based air quality standard (150 µg/m3). The two elevated PM10 concentrations (156 and 288 µg/m3) were observed in the mountainous, uninhabited lands east of the OB/OD area.(5) The highest PM10 concentration measured in or near residential locations was 100 µg/m3, which is considerably lower than EPA's 24-hour average health-based standard (150 µg/m3). Therefore, the sampling data indicate that actual short-term PM10 levels in or near where people live were lower than the levels of health concern. In fact, as Figure 7 shows, the sampling data strongly suggest that air emissions from local wildfires have a much greater impact on PM10 levels in the residential locations near SIAD than did air emissions from the installation's waste treatment operations.
Overall, the best available data for the site indicate that SIAD's air emissions do not cause PM10 concentrations to reach levels of public health concern over the long term. The data further suggest that PM10 levels over the short term also are not at levels of health concern. This latter observation, however, is based on air samples collected during typical waste treatment operations. Should SIAD in the future be allowed to treat larger amounts of wastes than were treated during the sampling program, further air sampling during waste treatment operations will be needed to ensure that residents are not exposed to PM10 at levels of public health concern. Section X of this health consultation provides additional detail on ATSDR's recommendation for additional air sampling.
Table 3 lists the highest annual average air concentrations of metals and inorganic compounds predicted by SIAD's models for residential areas in California and Nevada. None of the estimated concentrations are higher than health-based comparison values, which suggests that chronic inhalation exposures to the metals and inorganic compounds in SIAD's air emissions are not of public health concern. Although concentrations predicted by air models have inherent uncertainties, several observations assure ATSDR that actual exposures to metals would not cause adverse health effects among nearby residents:
As the three exceptions, predicted concentrations of aluminum, hydrogen chloride, and manganese were all less than a factor of 10 lower than their corresponding health-based comparison values. Ordinarily, this lower margin of safety might cause ATSDR to question whether the modeling analysis is an adequate basis for reaching a conclusion. However, ATSDR notes that these three contaminants were evaluated in the recent air sampling study, and every measured concentration of these contaminants at every sampling station was lower than the health-based comparison values listed in Table 3.
For these reasons, ATSDR has confidence that long-term average inhalation exposures to SIAD's emissions of metals and inorganic compounds, as characterized by model predictions and air quality measurements, are not at levels of public health concern.
ATSDR also assessed whether short-term exposures to elevated levels of metals and other inorganics would be expected to cause adverse health effects among people who live near SIAD. Because ATSDR believes the modeling analysis grossly overstates short-term air quality impacts (see Appendix D.2), we based the acute exposure assessment on the available sampling data for the following analytes: aluminum, barium, cadmium, chloride, copper, iron, lead, manganese, nickel, and zinc. Overall, the recent air sampling study reported more than 200 valid sampling results for each of these metals and other inorganics. With the exception of barium, not a single measured concentration exceeded the health-based comparison value for chronic exposures (i.e., those listed in Table 3), which strongly suggests that none of these chemicals are found at ambient air concentrations that would be acutely toxic to humans.
In the case of barium, the highest ambient air concentration measured out of 238 valid observations was 0.911 µg/m3, or less than a factor of two higher than the lowest health-based comparison value for chronic exposures. Although the effects of inhalation exposures to barium have not been extensively studied, ATSDR does not believe one-time inhalation exposures to 0.911 µg/m3 of barium will be associated with health effects. This judgment is based on an occupational study of workers who used barium-containing welding rods. The study found no evidence of exposure-related health effects among individuals who inhaled average barium levels as high as 4,400 µg/m3 (Zschiesche et al. 1992)or exposure levels more than 1,000 times greater than the highest measured concentration near SIAD.
Overall, our evaluations of air emissions of metals and other inorganic materials indicate that residents who live near SIAD are exposed to trace amounts of these contaminants, but the exposure levels are far below those known to be associated with adverse health effects.
Table 4 lists the estimated annual average concentrations of the 18 chemicals evaluated in SIAD's human health risk assessment (Brown and Root Environmental 1996a). For the 12 chemicals that have published health-based comparison values, the estimated annual average air concentrations were all more than 100 times lower than the most conservative comparison value, including those derived for cancer outcomes. Although modeling results likely do not equal actual exposure concentrations, ATSDR believes it is highly unlikely that the modeling analysis underestimates actual exposures by more than a factor of 100. Thus, ATSDR has confidence that actual exposure concentrations of explosives, propellants, and fillers are not greater than levels of health concern.
No health-based comparison values are available for six of the chemicals listed in Table 4; however, the estimated concentrations of these chemicals are extremely low. With the exception of nitrocellulose, for example, the highest annual average concentrations of the chemicals at any residential receptor are all estimated as being less than 0.001 µg/m3 (or 1 ng/m3). ATSDR has found no evidence in the scientific literature that any of the chemicals would be toxic at these exposure levels; however, we acknowledge that a limited set of toxicity studies have been conducted on these chemicals. In the case of nitrocellulose, the highest estimated exposure concentration at a residential location was 0.0041 µg/m3. These levels are not expected to be toxic to humans, based on the limited ingestion toxicity data available from EPA, which suggest that humans experience toxic effects only if they ingest tremendous amounts of nitrocellulose (i.e., amounts equivalent to 10% of our total diet) (EPA 1987b). For these reasons, ATSDR believes that long-term exposures to the estimated ambient air concentrations listed in Table 4 are not of public health concern.
To assess the public health implications of short-term exposures to the selected explosives, propellants, and fillers, ATSDR reviewed data published in its toxicological profiles, which are available for seven of the chemicals listed in Table 4 (ATSDR 1995a, 1995b, 1995c, 1995d, 1997, 1998, 2000). Limited data are available on acute inhalation toxicity associated with the chemicals that we researched, and most acute toxic effects were reported among occupational cohorts (e.g., employees of ammunition manufacturing plants), who likely were exposed to much greater levels of explosives, propellants, and fillers than were residents who live near SIAD. There is considerable uncertainty associated with our assessment of short-term exposures to explosives, propellants, and fillers, due to the lack of acute toxicity data. ATSDR notes, however, that the acute exposure levels to these chemicals are extremely lowsome much lower than can be measured with highly sensitive measurement devices. In summary, ATSDR finds that short-term exposures to the chemicals listed in Table 4 will be of extremely low magnitudes. We do not believe the exposures would be associated with adverse health effects, but this finding is based on limited acute toxicity data.
To assess long-term exposures, ATSDR compared the estimated concentrations for the chemical by-products of OB/OD events to their most sensitive health-based comparison value, considering both those derived for cancer and non-cancer outcomes. As Table 5 shows, for every chemical considered(7), the highest estimated air concentration in residential locations was lower than its corresponding comparison value. In fact, for the majority of these chemicals, estimated air concentrations were more than 1,000 times lower than health-based comparison values (i.e., levels that would require more detailed toxicologic evaluations in an ATSDR public health evaluation). Given this ample margin of safety, and ATSDR's belief that the modeling analysis includes several assumptions that likely overstate actual ambient air concentrations (see Appendix D.1), we conclude that long-term exposures to the chemical by-products of OB/OD events will not cause adverse health effects at any of the residential locations in the vicinity of SIAD.
To assess acute exposures, ATSDR limited its review to chemicals in Table 5 with estimated annual average air concentrations greater than 0.0001 µg/m3. This decision was made to focus on the chemicals with the higher concentrations, while not considering those chemicals with air quality impacts that would be virtually impossible to measure in an ambient air monitoring study. After examining acute toxicity data for the selected chemicals (ammonia, benzene, carbon monoxide, hydrogen cyanide, nitric oxide, nitrogen dioxide, phenol, styrene, sulfur dioxide, toluene, and xylenes), ATSDR found that none of the estimated short-term concentrations exceeded levels of concern for acute toxicity.
Overall, our review focused on nearly 100 pollutants known to be emitted from OB/OD waste treatment operations. Whether considering reasonable estimates of ambient air concentrations or measured levels of air contamination, we did not identify any air contaminants that residents (both in California and Nevada) would breathe at levels that are harmful. In fact, for most pollutants examined, the estimated air concentrations were orders of magnitude lower than levels of potential health concern, whether for cancer or non-cancer effects.
ATSDR notes, however, that evaluating the potential effects of environmental contaminants onhumans involves some uncertainty. For example, our analyses for many chemicals are basedstrictly on findings of a modeling analysis (Brown and Root Environmental 1996a), and not onmeasured levels of air pollution. However, as Appendix D.1 describes, ATSDR believes themodeling analysis likely overstated actual exposure concentrations. Further, the health effectsthat might result from exposure to complex mixtures of chemicals are largely not understood.Given these uncertainties, ATSDR conducteda separate environmental health evaluationbased on an entirely different set of data (i.e.,cancer registry data) to assess whether localresidents have increased risks of developingcancer. This separate analysis is the subject ofthe next section. To reduce uncertainties inour evaluation, ATSDR recommends thatadditional PM10 sampling occur if SIADconducts OB/OD waste treatment operationsin the future, although we have no knowledgeif these operations will ever resume.
1 Specifically, wind directions between southwestand northwest were considered to be "roughly from west to east," while wind directionsbetween southeast and northeast were considered to be "roughly from east to west."
2 This estimate was calculated for fires that consume1,000 acres of forest in a 24-hour period. Larger fires will have higher emissionrates, and smaller fires will have lower ones.
3 In 2001, SIAD ceased its routine OB/OD operations,but the installation is still allowed to use OB/OD for emergency situations andnational security reasons. Since 2001, SIAD has used OB/OD to treat very smallquantities of ordnance and propellants, such as those believed to be unsafe totransport.
4 ATSDR notes that the contractors who performedthe monitoring study concluded in their summary report that the monitoring maynot have captured the highest air quality impacts from the OB/OD operations.ATSDR believes that conclusion was based on a flawed statistical analysis ofthe data. Appendix C.1 explains why ATSDR disagreeswith the contractor's findings on the monitoring study.
5 These measured PM10 concentrations are 3-houraverage samples collected during the time when OB/OD operations occurred. Itis quite possible that the 24-hour average concentrations at these locationswere considerably lower, because waste treatment occurred only during a few hoursper day. However, 24-hour average PM10 levels were not measured at these samplinglocations.
6 Aluminum and iron were selected for this comparisonbecause they were detected more frequently than all other metals and inorganiccompounds considered. Calculating average concentrations for these metals involvesless uncertainty, since the data sets involve the fewest amount of non detects.
7 Eight chemicals in Table5 do not have health-based comparison values. The estimated air concentrationof methane is well below levels that would be viewed as a physical hazard, andthe estimated concentrations of all other chemicals were lower than 0.00004 µg/m3.Such trace levels would be extremely difficult to measure, even when using highlysensitive field sampling equipment. Despite the limited toxicological data forthese chemicals, ATSDR doubts that exposures to the concentrations listed wouldcause any adverse health effect.