HEALTH CONSULTATION
EASTERN MICHAUD FLATS CONTAMINATION
POCATELLO, BANNOCK AND POWER COUNTIES, IDAHO
ATSDR uses a conservative approach to determine whether levels of air pollution indicate a past, present, or future health hazard. The following discussion describes this methodology, and documents how it was applied to the levels of contamination measured in the EMF study area. The remainder of this section provides an overview of the large volume of data collected in the EMF study area, and appendices to this report present more detailed analyses.
ATSDR generally follows a two-step methodology to comment on public health issues related to air pollution. First, ATSDR obtains representative environmental monitoring data for the site of concern and compiles a comprehensive list of site-related contaminants. Second, ATSDR uses health-based comparison values to identify those contaminants that do not have a realistic possibility of causing adverse health effects. For the remaining contaminants, ATSDR reviews recent scientific studies to determine whether the extent of environmental contamination indicates a public health hazard.
The health-based comparison values used in this report are concentrations of contaminants that the current public health literature suggest are "safe" or "harmless." These comparison values are quite conservative, because they include ample safety factors that account for most sensitive populations. ATSDR typically uses comparison values as follows: If a contaminant is never found at levels greater than its comparison value, ATSDR concludes the levels of corresponding contamination are "safe" or "harmless." If, however, a contaminant is found at levels greater than its comparison value, ATSDR designates the pollutant as a contaminant of concern and examines potential human exposures in greater detail. Because comparison values are based on extremely conservative assumptions, the presence of concentrations greater than comparison values does not necessarily suggest that adverse health effects will occur among exposed populations. More information on the comparison values used in this report can be found in Appendix B.
In the case of particulate matter, however, some scientists argue that adverse health effects can occur among sensitive populations even when ambient air concentrations are lower than the health-based comparison value used in this report (i.e., EPA's actual and proposed National Ambient Air Quality Standards). In other words, levels of contamination below the health-based comparison value might, in fact, not be "safe" or "harmless" to certain sensitive populations. The sidebar on the above reviews additional information on the selection of health-based comparison values for particulate matter, and Section IV.E comments on this issue further.
The following analyses identify air pollutants for the EMF study area (Section IV.B), describe how these pollutants disperse throughout the area (Section IV.C), review site-specific studies that have measured levels of air pollution (Section IV.D), and finally comment on the public health implications of inhalation exposures to air pollution in the EMF study area (Section IV.E).
B. Emissions Data: What Pollutants Are Released to the Air?
To identify site-related contaminants for the EMF study area, ATSDR consulted with EPA, IDEQ, the Shoshone-Bannock Tribes, FMC, and Simplot to obtain reports that characterize air emissions from the two phosphorous processing facilities. The reports ATSDR obtained indicate that either FMC or Simplot, or both facilities, are suspected of emitting at least the following pollutants into the air (Bechtel 1996; Bechtel 1998; FMC 1999a, 1999b, 1999c; IDEQ 1999a; USEPA 1999d):
| Aluminum | Fluorides | Phosphoric acid |
| Ammonia | Hydrogen cyanide | Phosphorous |
| Antimony | Iron | Phosphorous pentoxide |
| Arsenic | Lead | Selenium |
| Barium | Manganese | Silver |
| Beryllium | Mercury | Sulfur dioxide |
| Cadmium | Nickel | Sulfuric acid |
| Chromium | Particulate matter | Zinc |
| Copper | Phosphine |
As an example of emissions data for these facilities, Table 1 presents the air emissions data that FMC and Simplot reported to EPA's Toxic Release Inventory (TRI) for calendar years 1997 and 1998. The TRI database is an important source of "right-to-know" information, or information that people can access about the releases of toxic chemicals in their communities. Because the accuracy of TRI emissions data are not known, ATSDR based its findings of this health consultation on the levels of chemicals that were measured in the ambient air, rather than focusing strictly on the emissions data. It is important to note that a large volume of air quality measurements are available for almost every pollutant listed above and in Table 1, and the evaluations of ambient air monitoring data presented later in this section consider the pollutants that FMC and Simplot emit in greatest quantities.
Also noteworthy is the fact that the TRI data do not show that many different operations at FMC and Simplot emit pollutants to the air. Some pollutants are released from elevated sources, like stacks, and others from ground-level sources, like waste ponds. Several studies have reported estimates of chemical-specific emissions from FMC and Simplot (Bechtel 1996; IDEQ 1999a). Though estimated emission rates are somewhat uncertain, they do provide insight into the relative impacts of various sources on air quality. As Table 2 shows, studies have estimated that, in recent years, FMC and Simplot released 727 and 135 tons of particulate matter to the air in a calendar year, respectively (IDEQ 1999a; USEPA 1999a). The data in Table 2 are interpreted in greater detail below.
It is expected that emission rates from these facilities likely have varied from year to year, as a result of changes in production demands, installation and operation of different pollution controls, use of ores from various sources, and other factors. As examples, particulate emissions from Simplot decreased considerably in 1991, after the facility began to receive ore in a slurry pipeline, instead of by rail car (Bechtel 1996); similarly, particulate emissions from FMC decreased after the facility installed new scrubbers on its calciners in 1992 (Severson 1999), and
FMC is currently implementing controls at many other specific emissions sources. The emissions from these facilities will likely continue to decrease in the future, due to pollution control plans recently adopted by EPA and IDEQ (FR 1999; IDEQ 1999a). In fact, FMC has informed ATSDR that its ongoing emissions controls projects are expected to result in a 67% reduction in particulate emissions.
In addition to FMC and Simplot, other industrial and non-industrial sources throughout the EMF study area release many of the pollutants listed above. For example, the Bannock Paving Company, which was known to emit particulate matter, metals, and other pollutants, operated on a leased portion of the FMC property. These operations reportedly ceased on March 12, 1995, and Bannock Paving Company moved to another location in Pocatello later in the year (Bechtel 1996). Furthermore, aircraft, trains, automobiles, residential wood burning, and agricultural operations all emit particulate matter to the atmosphere (IDEQ 1999a). These other sources, many of which are found throughout Chubbuck, Pocatello, and the Fort Hall Indian Reservation, undoubtedly contribute to air pollution in the EMF study area.
For perspective on the relative amounts of particulate matter released by FMC, Simplot, and other sources, Table 2 presents selected findings from recent emissions inventories for particulate matter (IDEQ 1999a; USEPA 1999a). The table indicates that particulate emissions from FMC and Simplot account for a considerable portion of the overall emissions for the EMF study area. To a first approximation, therefore, emissions from these facilities also account for a considerable portion of the airborne particulate matter in the EMF study area, but the relative impacts of these facilities on air quality certainly vary from location to location. Also noteworthy is the fact that the emissions inventories suggest that FMC might release more than five times more particulate matter to the air than does Simplot.
Though this health consultation evaluates many different pollutants that FMC and Simplot emit, much of this document focuses on the facilities' emissions of particulate matter--a class of pollutants consisting of solid particles and liquid droplets in the air. The sidebar on the following page provides definitions of, and relevant background information for, particulate matter.
C. Meteorological Data: Where Do the Air Emissions Go?
Although the FMC and Simplot facilities have emitted pollutants in varying quantities over the years, it does not necessarily follow that residents have been continuously exposed to the site-related pollutants. Local meteorological conditions determine whether emissions from the facilities rapidly disperse in the air or gradually accumulate to potentially unhealthy levels. To understand how these local conditions affect levels of air pollution, ATSDR reviewed several studies that evaluated how emissions from FMC and Simplot disperse in the atmosphere (Bechtel 1993; USEPA 1999d; OMNI 1991a; TRC 1993). These studies identified many meteorological conditions that affect local air pollution, but two factors--surface wind patterns and stagnation episodes (or inversions)--appear to have the strongest impact on air pollution in the EMF study area:
Surface winds. Not surprisingly, the wind direction plays a very important role on air quality issues in the EMF study area: winds blow emissions from the facilities to "downwind" locations, including parts of Chubbuck, Pocatello, and the Fort Hall Indian Reservation. According to wind direction measurements both at the Pocatello Airport (see Figure 2) and near FMC's main process operations, the prevailing wind direction at locations immediately north of the industrial complex is from the southwest to the northeast (USEPA 1999d; TRC 1993). This wind pattern suggests that emissions from the facilities generally, but not always, blow toward the northeast. Somewhat consistent with this prevailing wind direction is the fact that community members have often reported seeing "a dense brown cloud" extend from near the FMC and Simplot facilities to locations as far as 5 miles to the north (Sho-ban 1989).
Though wind patterns observed at the Pocatello Airport exhibit consistent trends from year to year, prevailing wind patterns are considerably different at other locations in the EMF study area. For instance, a meteorological station operated near the Simplot facility has frequently observed winds blowing from the southeast to the northwest--a wind direction rarely observed at the Pocatello Airport (Bechtel 1996). Moreover, prevailing wind patterns in the Portneuf River valley, where the city of Pocatello is located, are also expected to have a strong southeasterly component, due largely to influences from local terrain (TRC 1993). In fact, IDEQ recently observed a prevailing southeasterly wind pattern at its meteorological monitoring station near downtown Pocatello (IDEQ 1999a).
Two studies have reported noteworthy associations between certain wind conditions and levels of air pollution at locations downwind of the FMC and Simplot facilities. More specifically, roughly 75% of the highest PM10 concentrations measured by IDEQ at locations northeast of the FMC and Simplot facilities have occurred when relatively strong winds (i.e., 24-hour average wind speed greater than 9 miles per hour) blow from the southwest (IDEQ 1999a). Further, an ongoing study at the EMF site indicates that the highest concentrations of PM10 at a location directly across the street from the FMC facility are associated with winds blowing from FMC toward the monitors (USEPA 1999d). Section IV.D comments on these studies further.
Stagnation conditions (inversions). Some of the highest levels of air pollution in the EMF study area have occurred during stagnation conditions (IDEQ 1999a). In fact, a particularly severe stagnation episode occurred in December 1999, as ATSDR was preparing an earlier release of this health consultation. In general, these stagnation conditions, which are characterized by a calm atmosphere, light and variable winds, little or no precipitation, and near ground-level inversions, are typically observed in the winter, but they are observed infrequently. In fact, in some years, stagnation episodes have not occurred at all in the EMF study area. During the infrequent stagnation periods, however, emissions from FMC, Simplot, and other local sources become trapped in the lowest levels of the atmosphere. When stagnation conditions persist or are severe, air pollution throughout this area can reach potentially unhealthy levels.
Some researchers have characterized the specific meteorological conditions that are associated with the infrequent inversions. For instance, IDEQ has reported that the wintertime inversions generally occur on days with "temperatures near or below freezing; relative humidities above 70 percent; and multi-day meteorologically stagnant conditions" (IDEQ 1998b). Consistent with this observation, EPA has reported that the inversions occur primarily during "very specific and rare meteorological conditions--cold stagnant winter days with relative high humidity" (USEPA 1999a). As discussed in greater detail in Section IV.D, the aforementioned stagnation conditions are a major factor in the infrequent pollution episodes, or days when airborne particulate matter in much of Chubbuck, Pocatello, and the Fort Hall Indian Reservation reach unusually high levels.
It should be noted that ATSDR has reviewed several dispersion modeling studies (studies that simulate the transport of emissions in the atmosphere) for the EMF study area (Bechtel 1993; IDEQ 1991; OMNI 1991b; TRC 1993). Though these studies provide insight into levels of air pollution in locations where monitoring has not been conducted, the dispersion modeling results can be highly uncertain and are limited by the accuracy of critical inputs, particularly the emission rates from the phosphate processing plants. Perhaps the only consistent finding among these studies, however, is that modeled concentrations of PM10 are highest in the immediate vicinity of FMC and Simplot and that trace levels of site-related contaminants are predicted to occur throughout the EMF study area, including at locations in the cities of Chubbuck and Pocatello, at the Fort Hall Agency, and in unincorporated areas between these locations.
Though ATSDR considered conducting its own dispersion modeling analysis for the EMF study area, the Agency eventually decided to abandon such efforts after learning of the difficulties EPA encountered with modeling emissions from FMC. As evidence of this, EPA has recently reported that ". . .despite repeated efforts of EPA, with the assistance of the Tribes, IDEQ, and affected industry, the air quality models initially selected and approved by EPA for use in the Power-Bannock area PM10 non-attainment area, have continued to fail well-established performance criteria in the vicinity of the FMC facility. . ." (USEPA 1999a). For this reason and many other reasons, ATSDR decided that dispersion modeling results for the EMF site would undoubtedly be extremely uncertain and might possibly raise more questions than they would answer. As a result, the conclusions in this health consultation are based entirely on trends and patterns among the large volume of available air monitoring data, which, as mentioned previously, characterize air concentrations of the pollutants that FMC and Simplot emit in greatest quantities.
D. Ambient Air Monitoring Data: What Are the Levels of Air Pollution?
This section reviews the results of relevant ambient air monitoring studies, or studies of the air that people breathe. Since various organizations have measured levels of air pollution in the EMF study area over the past 25 years, a large volume of ambient air monitoring data are available for review for many locations in the EMF study area. To illustrate this, Figure 3 indicates the locations of the monitoring stations operated by IDEQ and the Shoshone-Bannock Tribes. Further, Appendix A of this report includes ATSDR's review of 12 different air monitoring studies conducted in this area.
Since each study has a limited scope, no single study is sufficient for understanding how levels of air pollution have changed throughout the EMF study area over the years. Combining the results from the many studies, however, provides an extensive and consistent account of air quality in this region. More specifically, the collective weight-of-evidence from these studies indicates the following general trends in air quality:
ATSDR's more detailed findings regarding the ambient air monitoring data are presented below, classified by pollutant. Selected supporting calculations are documented in appendices, as noted. The findings are based only on ambient air monitoring data collected from 1975 to the present. Without extensive data available for earlier years, ATSDR cannot make firm conclusions about levels of air pollution in the EMF study area prior to 1975.
The following discussion does not comment on whether the ambient air monitoring data trends indicate health hazards. Such analyses can be found in the "Public Health Implications" section, or Section IV.E.
Overview of Exposures to Particulate Matter: The Area of Impact. As a brief summary of the Agency's findings regarding exposures to particulate matter, Figure 4 indicates the area where ATSDR believes concentrations of PM10 or PM2.5, either over the short term (24-hour average) or the long term (annual average), have exceeded health-based comparison values at least one time between 1975 and the present. ATSDR derived the area of impact in Figure 4 from the following observations:
Overall, ATSDR believes the area of impact shown in Figure 4 is a best estimate of the areas where levels of airborne particulate matter (whether PM10 or PM2.5, whether over the short term or the long term) have exceeded health-based standards at some time between 1975 and the present. Given the fact that elevated concentrations of particulate matter have occurred throughout this area as recently as December 1999, ATSDR believes that elevated concentrations will likely occur in the future unless the main emissions sources in the area are reduced. As documented above and in the Appendices to this report, marking the boundaries of the area of impact in Figure 4 involves considerable uncertainty.
Recognizing this, ATSDR emphasizes that the boundaries should be viewed as a defensible estimate of the actual region were concentrations have exceeded health-based standards, and the boundary shown might understate or overstate the actual area over which concentrations reached potentially unhealthy levels. In other words, some residents who live outside the shaded region in Figure 4 might have been, and continue to be, exposed to levels of particulate matter higher than relevant health-based standards, and some residents who live within the shaded region might not have been exposed to such levels.
Many different emissions sources are believed to contribute to the elevated levels of particulate matter in the EMF study area, but emissions from FMC and Simplot undoubtedly account for a considerable portion of the air pollution in this area, especially in areas immediately downwind of the facilities. A detailed source apportionment study, however, is not included in the scope of this health consultation.
More information on the short-term and long-term concentrations of PM10 and PM2.5 in the EMF study area follows:
PM10. The results of the many air quality studies performed in the EMF study area show that ambient air concentrations of PM10 have varied both with time and with location. The following discussion comments on these temporal and spatial variations by answering two basic questions about airborne levels of PM10 near the EMF site. The questions address 24-hour average concentrations separate from annual average concentrations of PM10, since health-based standards have been developed for both exposure durations. Responses to the following questions are a critical input to the "Public Health Implications" section of this document:
At what locations were 24-hour average PM10 concentrations higher than corresponding health-based comparison values? The weight-of-evidence from the ambient air monitoring studies suggests that 24-hour average concentrations of PM10 throughout Chubbuck and Pocatello and in parts of the Fort Hall Indian Reservation periodically exceeded health-based standards (i.e., 150 ug/m3) and have the potential to do so in the future. As noted earlier in this report, elevated concentrations near FMC and Simplot are generally associated with strong southwesterly winds that blow emissions toward the monitors, and elevated levels in the Portneuf Valley are generally associated with stagnation conditions, during which emissions from FMC and Simplot and many other sources appear to affect air quality.
The exceedances were clearly most frequent and most severe in the immediate vicinity of the FMC and Simplot facilities. Specifically, EPA has reported that 24-hour average PM10 concentrations measured at a location on the Fort Hall Indian Reservation north of FMC and south of Interstate 86 exceeded 150 ug/m3 up to 21 days in 1996 and 20 days in 1997 (USEPA 1999a), but the exact spatial extent of this poor air quality is not known. Similarly, according to IDEQ's monitoring data, the number of days with PM10 concentrations above health-based standards also varied from year to year: in some years, no exceedances were observed in Chubbuck and Pocatello at all; in other years, however, as many as 6 exceedances likely occurred (IDEQ 1999a). Exceedances of PM10 air quality standards occurred in Pocatello as recently as December 31, 1999--a finding that is based on data that IDEQ recently released to ATSDR (IDEQ 2000).
Appendix C.1 presents the evidence ATSDR considered in reaching its conclusion regarding 24-hour average concentrations of PM10. Note, ATSDR considers the lack of monitoring data on the Fort Hall Indian Reservation at locations north of Interstate 86 an important data gap that needs to be filled.
At what locations were annual average PM10 concentrations higher than corresponding health-based comparison values? The weight-of-evidence suggests that, in at least one year between 1975 and the present, annual average PM10 concentrations exceeded EPA's health-based comparison value (50 ug/m3) in parts of Chubbuck, Pocatello, and the Fort Hall Indian Reservation. The frequency with which annual average levels exceeded health-based standards appears to decrease with distance from the EMF site.
Air monitoring studies sponsored by FMC, Simplot, and EPA all indicate that annual average PM10 concentrations have exceeded EPA's health-based standard in an area immediately north of FMC (Bechtel 1995; Hartman 1999; USEPA 1999a). ATSDR believes these studies, taken together, suggest that concentrations of PM10 likely exceeded the annual average air quality standard in a small area for at least the last 6 years, and probably longer. According to EPA, trends in the ambient air monitoring data "point conclusively to FMC as the source" of the elevated PM10 concentrations in the area between FMC and Interstate 86 (USEPA 1999d). Note, it is not known how far north of the facilities concentrations exceeded health-based standards.
In addition to the data collected in the vicinity of FMC and Simplot, IDEQ's monitoring data suggest (1) that annual average PM10 concentrations at the Pocatello Sewage Treatment Plant might have exceeded 50 ug/m3 in as many as 12 years between 1975 and the present, and (2) that annual average PM10 levels at Chubbuck School might have exceeded this level in 3 years or fewer during this same time frame. On the other hand, ATSDR does not believe that such elevated annual average levels occurred at either Garret and Gould or Idaho State University. As Appendix C.3 explains, these estimates are based, in part, on extrapolations of TSP monitoring data and therefore are somewhat uncertain. Appendix C.2 presents the evidence ATSDR considered in reaching its conclusion.
PM2.5. Though ambient air concentrations of particulate matter have been measured extensively throughout the Pocatello area, few studies have measured concentrations of fine particles, also known as PM2.5. Nonetheless, the available PM2.5 monitoring studies characterize the size distribution of airborne particles typically observed in the EMF study area. Knowledge of the particle size distribution, coupled with the PM10 and TSP measurements made over the years, provides insight into what PM2.5 concentrations might have been during times when this pollutant was not actually measured.
Responses to the following two questions summarize ATSDR's findings regarding the levels of PM2.5 that likely occurred in the EMF study area between 1975 and the present. Like the questions in the review of PM10 concentrations, the following questions address 24-hour average and annual average concentrations separately. Responses to the following questions are a critical input to the "Public Health Implications" section of this document:
At what locations were 24-hour average PM2.5 concentrations higher than corresponding health-based comparison values? To date, 24-hour average ambient air concentrations of PM2.5 have been measured at several locations, including across the street from FMC and at the Pocatello Sewage Treatment Plant, Chubbuck School, Idaho State University, and Garret and Gould. The most extensive PM2.5 monitoring effort conducted in the EMF study area to date has shown that 24-hour average ambient air concentrations of PM2.5 across the street from FMC frequently exceeded health-based comparison values (i.e., 65 ug/m3) between October 1996 and September 1998 (USEPA 1999d). It is reasonable to believe that these exceedances occurred at this location prior to October 1996, even though monitoring was not conducted during this time. It is not known how far north these elevated PM2.5 concentrations occur.
In addition to the data collected across the street from FMC, IDEQ has measured 24-hour average PM2.5 concentrations above health-based standards at all four of its monitoring stations. Since some of the elevated PM2.5 concentrations occurred as recently as December 1999, ATSDR believes it is possible that elevated PM2.5 levels will continue to occur in the future unless sources of this pollutant are reduced. Unlike the trend observed across the street from FMC, the elevated 24-hour average concentrations of PM2.5 in Chubbuck and Pocatello appear to occur infrequently, primarily during stagnation episodes or inversions.
Appendix D.1 presents the evidence ATSDR considered in reaching its conclusion regarding 24-hour average concentrations of PM2.5. The lack of extensive PM2.5 monitoring data on the Fort Hall Indian Reservation at locations north of Interstate 86 is an important data gap that needs to be filled.
At what locations were annual average PM2.5 concentrations higher than corresponding health-based comparison values? The available monitoring data suggests that annual average levels of PM2.5 were highest in the immediate vicinity of the EMF study area, with levels gradually decreasing with downwind distance. For instance, the most extensive PM2.5 monitoring study to date has shown that annual average concentrations of this pollutant have exceeded, and continue to exceed, 15 ug/m3 at locations immediately north of FMC. Based on a limited set of data collected by IDEQ in 1998 and 1999, annual average concentrations of PM2.5 currently do not exceed health-based standards throughout Chubbuck and Pocatello.
The weight-of-evidence suggests that, in the years before the PM2.5 monitoring studies were conducted, annual average PM2.5 concentrations likely exceeded EPA's health-based comparison value (15 ug/m3) in much of Chubbuck and Pocatello and in parts of the Fort Hall Indian Reservation. As Appendix D.2 explains, this finding is based primarily on extrapolations of PM10 monitoring data, using defensible estimates of PM2.5/PM10 ratios. In other words, this finding is somewhat uncertain since it is based on estimated--not measured--concentrations of PM2.5.
Appendix D.2 presents the evidence ATSDR considered in reaching its conclusion regarding annual average concentrations of PM2.5.
Ionic species in particulate matter. Since studies have linked inhalation exposure of acid aerosols to an increased incidence of adverse health effects among sensitive populations, ATSDR obtained and reviewed ambient air monitoring data for several ionic species. These data were found for ammonium, chloride, fluoride, nitrate, potassium ion, and sulfate (Bechtel 1996; IDEQ 1999b). Of these species, the highest peak concentrations observed to date were for ammonium (42.75 ug/m3), nitrate (27.15 ug/m3), and sulfate (83.9 ug/m3) (IDEQ 1999b). Interestingly, these three peak concentrations all occurred at the Idaho State University monitoring station--the IDEQ station located furthest from the FMC and Simplot facilities.
The fact that the highest concentrations of these ions occurred far from FMC and Simplot does not necessarily imply that emissions from these facilities contributed little to the measured levels. To the contrary, the data trends are consistent with the hypothesis that emissions from the two facilities accounted for a considerable portion of the measured concentrations. For example, IDEQ has estimated that emissions of sulfur dioxide from FMC and Simplot account for more than 93% of the total emissions of sulfur dioxide in the EMF study area (IDEQ 1999d). Since sulfur dioxide emissions are a precursor to ambient sulfate ions, and since FMC and Simplot clearly emit more sulfur dioxide to the air than all other sources in the area combined, it is reasonable to assume that the peak concentrations of sulfate at Idaho State University can be attributed, to a large extent, to emissions from the phosphate processing plants. Moreover, given the fact that it takes time for airborne sulfur dioxide to react and form sulfates, it is not surprising that the highest sulfate concentrations have been observed at the monitoring station located furthest from FMC and Simplot. Regardless of the source of these ions, however, Section IV.E evaluates whether these elevated concentrations present a public health hazard.
Though never measured at the levels observed for ammonium, nitrate, and sulfate, fluoride was consistently detected in air samples, particularly those collected in close vicinity to Simplot, a known source of fluoride emissions. For example, the RI reported that the highest concentrations of fluoride were measured at the three stations located around the perimeter of Simplot. The highest concentrations for these stations were 13.14 ug/m3, 11.29 ug/m3, and 10.92 ug/m3; average concentrations were not reported for these stations (Bechtel 1996). All of the samples from IDEQ's network that were selected for chemical analyses had concentrations lower than those measured during the RI. The "Public Health Implications" section reviews the fluoride concentrations in greater detail.
ATSDR reviewed the available monitoring data for the two remaining ionic species (chloride and potassium ion), but both species were measured at considerably lower levels than the other ionic species discussed above. More specifically, concentrations of chloride and potassium ion in the 72 valid samples collected were all less than 2.0 ug/m3. A brief toxicological evaluation is presented for these ions in the "Public Health Implications" section.
Phosphorous compounds (phosphorous, phosphate, phosphine, phosphorous pentoxide). Since both FMC and Simplot process vast quantities of phosphorous every year, ATSDR carefully examined the measured ambient air concentrations of various forms of phosphorous. To date, ambient air monitoring studies conducted in the EMF study area have measured levels of total phosphorous in particulate matter as well as levels of phosphate ion (PO43-). However, no studies have characterized ambient air concentrations of phosphorous pentoxide--a pollutant known to be emitted by FMC (Bechtel 1993). Though ATSDR identified emissions estimates and dispersion modeling results for phosphorous pentoxide, the lack of ambient air monitoring data appears to be due to the lack of approved sampling and analytical methods for this compound. As a result, the actual levels of phosphorous pentoxide that people might have breathed, and might continue to breathe, are not known.
ATSDR does not consider this a critical data gap in the health consultation, however, since phosphorous pentoxide is known to react rapidly in air to form phosphate ion (USEPA 1999b). Due to this reaction, phosphorous pentoxide emitted by FMC will partly, if not entirely, transform to phosphate ion by the time the emissions reach residential areas. Thus, ATSDR believes evaluating ambient air concentrations of total phosphorous and of phosphate ion will adequately address the community concerns regarding emissions of phosphorous pentoxide.
Not surprisingly, concentrations of total phosphorous were consistently found to be highest in areas closest to FMC and Simplot.(2) For example, according to the RI, average concentrations of total phosphorous at a monitoring location immediately north of FMC were more than five times higher than average concentrations measured at any of the six other monitoring locations (Bechtel 1996). The magnitude of total phosphorous concentrations also varied with time: sometimes phosphorous was not detected in 24-hour average samples, and other times it was detected at concentrations as high as 26.8 ug/m3 (Bechtel 1996; USEPA 1999d). The highest long-term average concentration of total phosphorous reported to date is 5.45 ug/m3, at a location immediately north of FMC and based on nearly 1 year of routine sampling (Bechtel 1996). Though neither ATSDR nor EPA have published health-based comparison values for total phosphorous, the "Public Health Implications" section of this report carefully reviews available toxicological data for this metal.
ATSDR also reviewed data available on concentrations of phosphate ion, which were measured by IDEQ using ion chromatography. Data trends for phosphate ion were quite similar to those discussed above for phosphorous. However, because these measurements were not conducted routinely, representative average concentrations of phosphate ion cannot be calculated and compared to the average phosphorous concentrations. Nonetheless, the sporadic measurement of phosphate ion concentrations provides some insight into the magnitude of concentrations that have been observed in the area. More specifically, at the Pocatello Sewage Treatment Plant, 38 24-hour average measurements of phosphate ion have been made over a 5-year period, of which, half had phosphate ion concentrations between 10 and 50 ug/m3 and the other half had phosphate ion concentrations lower than this range (IDEQ 1999b). Of the more limited phosphate ion measurements at IDEQ's three other monitoring stations, which are all in residential neighborhoods, no concentrations of phosphate ion were found to exceed 10 ug/m3. The "Public Health Implications" section of this report comments on the significance of these measurements.
Finally, ATSDR obtained and reviewed emissions and monitoring data for phosphine, an inorganic form of phosphorous that is released from FMC's on-site waste management ponds (Bechtel 1998b; FMC 1999a, 1999b, 1999c, 1999d, 2000a). Unlike the data available for the other chemicals emitted by FMC and Simplot, no off-site ambient air monitoring data are available for phosphine, thus greatly limiting ATSDR's ability to evaluate past and current exposures. Nonetheless, ATSDR has learned that FMC has developed "pond management standards" that include provisions for emissions monitoring, fenceline air monitoring, and "a response action plan to ensure that the public will not be exposed to phosphine . . . levels that exceed federal guidelines" (Bechtel 1998). These management standards reportedly have been reviewed and approved by both EPA and the Shoshone-Bannock Tribes (Bechtel 1998). ATSDR reviewed a limited set of phosphine sampling data that FMC collected, which indicated that phosphine concentrations measured at the facility fenceline using an OSHA-approved sampling and analytical method ranged from nondetect to 101 ppb (Bechtel 1998). Subsequent continuous measurements have shown phosphine concentrations at the edge of on-site ponds to range from nondetect to 2,310 ppb (FMC 1999a, 1999b, 1999c), and measurements of phosphine air concentrations at the facility fenceline on four occasions have reportedly exceeded 1.0 ppm: 1.90 ppm on October 6, 1999; 1.10 ppm on October 23, 1999; 2.50 ppm on November 15, 1999; and 3.16 ppm on November 16, 1999 (FMC 1999d, 2000). These fenceline measurements were collected using "hand-held monitors and Draegers" and not using methods approved by federal agencies (OSHA has an approved phosphine sampling method). ATSDR reviewed additional phosphine monitoring data, but they were collected using a hand-held device that is known to report "false positive" detects for phosphine and, thus, are not included in this health consultation. Section IV.E reviews the significance of the measured phosphine concentrations, but ATSDR notes that the available data for this pollutant are limited.
Metals and other inorganics. Several ambient air monitoring studies have measured concentrations of metals and other inorganics in particulate matter at various locations in the EMF study area (Bechtel 1995; IDEQ 1999b; USEPA 1999d). Combined, these studies characterize airborne levels of more than 40 metals and other inorganics--most of which are emitted by either FMC or Simplot, or by both facilities. Table 3 lists these elements and summarizes how the measured concentrations compared to health-based comparison values. The table classifies the metals and other inorganics into three categories:
As noted earlier, and described in detail in Appendix B, when ambient air concentrations of a given pollutant exceed corresponding comparison values, this situation does not necessarily suggest that adverse health effects will occur, but it rather suggests that concentrations of the pollutant should be evaluated in greater detail to make conclusions on public health implications. As a critical input to the toxicological evaluations presented later in this report, the following list describes in greater detail the extent to which concentrations of 8 metals exceeded health-based comparison values. The "Public Health Implications" section of this report comments on the significance of the following trends.
Note, in the summaries below, results from three different studies were considered for identifying the maximum concentrations of metals and other inorganics (Bechtel 1995; IDEQ 1999b; USEPA 1999d). Since one of these studies (IDEQ 1999b) did not routinely analyze filters for chemical composition, ATSDR used only the data from the Remedial Investigation and the Fort Hall Source Apportionment Study to comment on average concentrations of metals and other inorganics.
Aluminum. Though concentrations of aluminum were measured in several air monitoring studies, only two monitoring locations (monitoring station 6 from the RI, see Appendix A.2, and the "Primary" station in the Fort Hall Source Apportionment Study, see Appendix A.4) reported concentrations of aluminum greater than the metal's most conservative health-based comparison value (3.7 ug/m3). The average concentrations(3) of aluminum in PM10 at these stations (0.15 ug/m3 and 0.85 ug/m3), however, were considerably lower than the comparison value. Concentrations of aluminum measured at all other monitoring stations were also considerably lower than the comparison value as well.
Arsenic. Three air monitoring studies indicated that concentrations of arsenic have recently, and frequently, exceeded the most conservative health-based comparison value (0.0002 ug/m3) (Bechtel 1995; IDEQ 1999b). Average concentrations in PM10 measured during the RI ranged from 0.000502 to 0.00127 ug/m3 (Becthel 1995). Moreover, at the "Primary" station in the Fort Hall Source Apportionment Study, the highest annual average concentration of arsenic in PM10 was 0.0012 ug/m3. This study clearly showed that the elevated metals concentration at the "Primary" station were caused primarily by emissions from FMC. Concentrations measured both in the immediate vicinity of the EMF site and in nearby residential areas, therefore, were found to be higher than the most conservative health-based comparison value.
Barium. Of the numerous reported concentrations of barium that ATSDR reviewed, only one concentration--from a sample collected by IDEQ at the Pocatello Sewage Treatment Plant in 1991--exceeded the corresponding most conservative health-based comparison value. This one concentration (0.57 ug/m3) was only marginally higher than the corresponding comparison value (0.51 ug/m3). At all other monitoring locations, every concentration of barium reported was considerably lower than the comparison value.
Beryllium. Of the many studies that measured ambient air concentrations of metals, only the RI measured ambient levels of beryllium (Bechtel 1995). As Appendix A.2 shows, every concentration of beryllium measured at six of the seven monitoring locations in this study was lower than the corresponding health-based comparison value (0.0004 ug/m3). Station 2, on the other hand, which was located immediately north of FMC in an unpopulated area, had a single concentration in TSP higher than this comparison value. The average concentration of beryllium in PM10 at this station (0.000179 ug/m3), however, was lower than the health-based comparison value.
Cadmium. Every study that has conducted speciated particulate monitoring in the EMF study area has reported both highest and average concentrations of cadmium at levels exceeding the most conservative health-based comparison value (0.0006 ug/m3). This trend was observed for every monitoring station in the RI (see Appendix A.2), for the Shoshone-Bannock monitors (see Appendix A.4), and for the IDEQ air monitoring network (see Appendix A.9). The highest average cadmium concentration in PM10 (0.035 ug/m3) was observed at the "Primary" station in the Fort Hall Source Apportionment Study; the cadmium detected at this station was shown to originate primarily from FMC's emissions (USEPA 1999d). The levels of cadmium measured at stations closer to the FMC and Simplot facilities were consistently higher than the levels measured at stations further from the industrial complex.
Chromium. Three studies have routinely analyzed particulate filters to measure concentrations of chromium (Bechtel 1995; IDEQ 1999b; USEPA 1999d). Interpreting these ambient air monitoring data, however, is complicated by the fact that chromium is often found in two different states (hexavalent and trivalent). These states have entirely different implications from a toxicological perspective. As an initial screening, ATSDR compared the measured concentrations of chromium to the most conservative health-based comparison value for the metal, which happens to be for the hexavalent state (0.00008 ug/m3). This initial screening found that highest and average concentrations of chromium at every sampling location, whether in residential neighborhoods or in close proximity to the EMF study area, exceeded the comparison value for hexavalent chromium. The highest average concentration of total chromium in PM10 (0.029 ug/m3) was observed at the "Primary" station in the Fort Hall Source Apportionment Study. Moreover, concentrations of chromium at locations along the perimeter of FMC and Simplot were consistently higher than those at downwind monitoring locations.
Manganese. Concentrations of manganese were measured in three studies, but only a small subset of the concentrations reported in two of these studies exceeded the corresponding health-based comparison value (0.04 ug/m3). As Appendix A.2 describes, data collected during the RI indicate that ambient air concentrations of manganese in TSP exceeded the comparison value on at least one occasion at six of the seven monitoring locations, including at the two monitoring stations near residential neighborhoods. At all seven monitoring stations, however, the average concentrations of manganese in PM10 were notably lower than the comparison value. Consistent with this trend, monitoring data collected by IDEQ indicate that concentrations of manganese in PM10 generally exceeded the health-based comparison value on days when particulate concentrations were high, but the IDEQ data are insufficient for calculating average concentrations. In the Fort Hall Source Apportionment Study, manganese never exceeded its comparison value in the fine fraction of particulate matter; in the coarse fraction, however, one sample had a manganese concentration (0.067 ug/m3) greater than the comparison value. The average levels of manganese in PM10 during the Fort Hall Source Apportionment Study were lower than the comparison value.
Vanadium. Ambient levels of vanadium in the vicinity of the EMF study area have been routinely measured during three different sampling efforts. Two of the sampling efforts never detected the metal at levels higher than the most conservative comparison value (0.2 ug/m3). The RI, on the other hand, reported several concentrations in TSP at levels higher than the comparison value, but only in unpopulated areas in the immediate vicinity of FMC and Simplot. At all seven monitoring stations that operated during the RI, average concentrations of vanadium in PM10 were lower than the comparison value.
Sulfur Dioxide. For more than 20 years, IDEQ has measured ambient air concentrations of sulfur dioxide in the EMF study area. Specifically, IDEQ monitored sulfur dioxide levels at the Pocatello Sewage Treatment Plant from 1977 to the present and at Garret and Gould from 1994 to the present. Overall, every annual average concentration of sulfur dioxide at both monitoring locations was less than EPA's health-based air quality standard (an annual average concentration of 0.03 ppm). However, a subset of 24-hour average concentrations of sulfur dioxide at the Pocatello Sewage Treatment Plant were higher than EPA's corresponding health-based standard (a 24-hour average concentration of 0.14 ppm) at least once a year, but not more than six times a year, from 1977 to 1985 (IDHW 1988).(4) Since IDEQ's sulfur dioxide monitoring prior to 1994 was limited to one sampling location, however, the area over which elevated sulfur dioxide concentrations occurred in the past is not known, but is likely limited to the immediate vicinity of the monitors at the Pocatello Sewage Treatment Plant. Since 1985, concentrations of sulfur dioxide measured by IDEQ have not exceeded health-based comparison values. Therefore, the data suggest that 24-hour average concentrations of sulfur dioxide exceeded health-based standards in a limited geographic area periodically between 1977 and 1985, but not again since. The "Public Health Implications" section of this report puts the past elevated concentrations of sulfur dioxide into perspective.
Other pollutants. In addition to the pollutants listed above, ATSDR obtained and reviewed information characterizing ambient air concentrations of other pollutants. However, most air quality studies conducted in the Pocatello area have focused on particulate matter, and relatively few studies have measured concentrations of other pollutants, like volatile organic compounds. Nonetheless, recent reports by IDEQ indicate that concentrations of carbon monoxide, nitrogen dioxide, and ozone in Power and Bannock Counties are lower than EPA's corresponding health-based standards (IDEQ 1998a).
More specifically, IDEQ has conducted fairly extensive sampling for nitrogen dioxide at its Garret and Gould monitoring station in Pocatello (see Figure 3). Over the course of 5 years of sampling (from 1994 to 1999), annual average concentrations of nitrogen dioxide were always roughly one-third of EPA's health-based NAAQS of 0.053 ppm. Further, IDEQ has measured ozone concentrations in the EMF study area, but only during special studies conducted in the winter months, when ozone levels are typically at their lowest. All ozone concentrations measured during these studies were less than half of EPA's one-hour average health-based standard of 0.120 ppm, but the extent and timing of sampling are extremely limited.
Finally, ATSDR gathered data on air quality measurements of hydrogen cyanide, a chemical released to the air primarily by the waste-management ponds at FMC. The data obtained by ATSDR indicate that monitoring for hydrogen cyanide has been performed only within the FMC property boundary, and no off-site monitoring data are available. The limited on-site data suggest that air concentrations of hydrogen cyanide at the FMC fenceline range from nondetects to as high as 430 ppb (Bechtel 1998). More recent monitoring at on-site locations along the perimeter of the waste management ponds has revealed hydrogen cyanide concentrations ranging from nondetects to 990 ppb (FMC 1999a, 1999b, 1999c, 1999d, 2000). FMC continues to monitor emissions and off-site transport of hydrogen cyanide as part of its "pond management plan," which both EPA and the Shoshone-Bannock Tribes have approved. Though implementation of this plan provides some level of comfort that off-site concentrations of hydrogen cyanide do not reach levels of health concern, ATSDR notes that only limited monitoring data are available to support such a conclusion.
Extensive information on pollutants other than those listed above are not readily available for the EMF study area. However, the previous summary reviews air quality data for a very large subset of pollutants released by FMC and Simplot, especially those released in greatest quantities.
E. Public Health Implications (Adult and Children's Health): Are the Levels of Air Pollution Unhealthy?
This section evaluates the public health implications of the levels of air pollution in the EMF study area. In general, the ambient air monitoring data described in the previous section indicate that a large segment of the population throughout the EMF study area have, at some time since 1975, been exposed to some site-related air contaminants, including PM10, PM2.5, and the various constituent of these airborne particles (e.g., metals, fluorides, phosphoric acid, sulfuric acid). This section provides a public health context to the exposures that have occurred to individuals who live near the EMF study area, including residents of Chubbuck, Pocatello, and the Fort Hall Indian Reservation. It is important to note that ambient air monitoring levels are used in this health consultation as a surrogate for exposure in the EMF study area. Actual individual exposure to air pollutants is determined by a complex interplay between human activity, including the locations where time is spent, housing characteristics (as they influence penetration of outdoor pollutants), and other factors.
This section opens by providing relevant background information on the many studies that have been conducted in other parts of the country to determine public health implications associated with exposures to particulate matter. Following this general background discussion are detailed health evaluations for the following six categories of site-related contaminants:
For contaminants that are believed to have reached levels that might be associated with adverse health effects, the following discussion identifies populations that are believed to be at the greatest risk. For reference, Appendix B explains some of the health-based comparison values and guidelines that were used to evaluate the public health implications of exposures in the EMF study area. It is important to note that there is some scientific debate regarding the levels of PM2.5 or PM10 that are considered protective for all segments of the population. Threshold concentrations for PM 2.5 or PM 10 (i.e., a level below which no adverse health effects are likely) have not been established within the scientific literature.
As a result, EPA's PM10 standard and proposed PM2.5 standard may not be protective of all sensitive subpopulations, though it is generally believed that the proposed annual PM2.5 standard is protective of the general population and probably many of the sensitive subpopulations. However, when establishing the PM2.5 standards, EPA intended for the annual average and 24-hour levels to work as a dual standard. That is, the 24-hour standard alone does not protect against short-term health effects but the two standards working in concert are protective. Therefore, EPA set a value of 40 ug/m3 (termed an air quality index, or AQI) as a rough surrogate for the general level of protection provided by the two standards in combination. For more information regarding EPA's use of AQIs, see the notice in the Federal Register, Volume 64, No. 149, page 42542, Wednesday, August 4, 1999.
The following evaluation of the public health implications of exposures to PM incorporates the understanding that there are no currently established PM thresholds and the understanding of the dual nature of the PM2.5 standards.
Relevant Background Information on Health Implications of Exposures to PM and Related Constituents. Over the past 20 years, numerous investigators have researched the public health implications of inhalation exposures to PM. The following discussion reviews this large volume of research, which provided a basis for much of the evaluations presented later in this section.
Prior to 1987, EPA enforced health-based standards that regulated ambient air concentrations of total suspended particulates, or TSP. By 1987, a growing amount of research had shown that the particles of greatest health concern were actually PM10, which, at the time, were shown to be capable of penetrating into sensitive regions of the respiratory tract. Consequently, EPA and the states took action in 1987 to monitor and regulate ambient levels of PM10. Since 1987, hundreds of additional studies (mostly epidemiological) have been published on the health effects of PM. These studies generally suggest that adverse health effects in children and other sensitive populations have been associated with exposure to particle levels well below that allowed by EPA's PM10 standard (USEPA 1997). Moreover, it is generally believed that fine particles (PM2.5) can penetrate into the lungs more deeply than PM10 and that fine particles are more likely to contribute to adverse health effects than coarse particles (i.e., particles larger than 2.5 microns, but smaller than 10 microns).
According to the various studies on PM, many health effects were found to be associated with PM2.5 exposures or with PM2.5 exposures coupled with exposures to other pollutants (USEPA 1997). A partial list of these health effects follows:
These studies indicate that elderly, infants, and persons with chronic cardiopulmonary disease, influenza, or asthma, are most susceptible to mortality and serious morbidity effects from short-term acutely elevated exposures. Others are susceptible to less serious health effects such as transient increases in respiratory symptoms, decreased lung function, or other physiological changes. Chronic exposure studies suggest relatively broad susceptibility to cumulative effects of long-term repeated exposure to fine particulate pollution, resulting in substantive estimates of population loss of life expectancy in highly polluted environments (Pope 2000). It is important to note that susceptibility may also be dependent on a number of exposure factors, including duration of exposure. The degree to which an added particle burden may impact an individual will likely be affected by their age, health status, medication usage, and their overall susceptibility to PM inhalation exposures. Certainly, one factor that may promote increased risk in the older population is that, over their lifespan, they may have had more exposure and hence more opportunity to accumulate particles or damage their lungs (USEPA 1996). Current epidemiological research does not provide conclusive evidence of an association between exposure to PM, in general, and cancer. However, since PM is made up of various constituents, depending on the source(s), there are likely to be chemicals included in PM that are potential carcinogens.
For reasons above, EPA proposed revisions to its PM standards in 1997 to include a primary (health-based) annual average PM2.5 standard of 15 ug/m3 and a 24-hour PM2.5 standard of 65 ug/m3 (USEPA 1997). EPA's scientific review concluded that fine particles are a better surrogate for those components of PM most likely linked to mortality (death) and morbidity (disease) effects at levels below the previous standard, while high concentrations of coarse fraction particles are linked to effects such as aggravation of asthma (USEPA 1997).(5)
The body of scientific knowledge used to set the health-based PM2.5 standard consisted primarily of epidemiological studies of communities exposed to elevated levels of PM--communities like those in and around the EMF study area. These epidemiological studies found consistent associations between exposure and adverse health effects both for short-term or acute PM exposure scenarios (i.e., usually measured in days) and for long-term or chronic exposure scenarios (i.e., usually measured in years) (USEPA 1996). Chronic exposures are best measured using annual average PM2.5 levels (concentrations above 15 ug/m3) for one or several years; whereas, acute exposures are best measured by using the 24-hour average PM10 and PM2.5 levels (concentration above 150 ug/m3 and 65 ug/m3, respectively). It should be noted that the epidemiological studies indicate increased health risks associated with PM exposures, either alone or in combination with other air pollutants.
PM-related increases in individual health risks are small, but likely significant from an overall public health perspective because of the large numbers of individuals in susceptible risk groups that are exposed to ambient PM (USEPA 1996). Although the epidemiological data provide support for the associations mentioned above, an understanding of the underlying biological mechanisms has not yet emerged (USEPA 1996). Much of the toxicological findings related to PM are derived from controlled exposure studies in humans and laboratory animals. These studies have most extensively focused on acidic aerosols (a subclass of PM), namely sulfuric acid aerosols and various sulfates and nitrates, and have included characterization of acid aerosols effects on pulmonary mechanical functions, lung particle clearance mechanisms, and other lung defense mechanisms (USEPA 1996). Controlled human exposures to PM constituents other than acid aerosols are limited. Laboratory animal studies and occupational exposure studies provide information on other PM substances, including metals, diesel emissions, crystalline silica, and other miscellaneous particles. Human exposure studies of particles other than acid aerosols generally provide insufficient data to draw conclusions regarding health effects (USEPA 1996). A recent study (Godleski, et al. 2000), funded by the Health Effects Institute (HEI), an independent and unbiased source of information, supported by both public and private sources, found that concentrated airborne particles had adverse effects on the electrical regulation of the heart in dogs with a pre-existing heart condition, while the impact on normal dogs was not clear. Moreover, biological evidence indicates that urban combustion particles can penetrate past the primary defense mechanisms of the lung, can elicit inflammatory changes in the lung and systematically (throughout the body), contain a constituent (soluble transition metals) that by itself can be demonstrated to produce lung damage, can produce electrocardiogram changes including arrhythmia (heart irregularities), and can kill animals with pre-existing heart and lung disease (Schwartz 1999). Human studies have also reported inflammatory changes, including systemic changes, and changes to cardiovascular risk factors (Schwarz 1999). Although scientific evidence has provided some clues into the biological mechanisms of how PM may elicit adverse health effects in animals an humans, clear evidence of the exact mechanisms has not emerged.
In summary, the weight-of-epidemiological evidence suggests that ambient PM exposure has affected and continues to affect the public health of U.S. populations. However, a great deal of uncertainty remains regarding many issues related to the overall scientific inquiry into the health effects of PM (USEPA 1996). Moreover, several viewpoints currently exist on how best to interpret the epidemiological data: one sees PM exposure indicators as surrogate measures of complex ambient air pollution mixtures and reported PM-related effects represent those of the overall mixture; another holds that reported PM-related effects are attributable to PM components (per se) of the air pollution mixture and reflect independent PM effects; and yet another suggests that PM can be viewed both as a surrogate indicator as well as a specific cause of health effects. Whichever the case, reduction of PM exposure would be expected to lead to reductions in the frequency and severity of PM-associated health effects (USEPA 1996).
PM2.5 and PM10 Exposures. ATSDR estimates that at least 53,710 persons have been exposed at some time between 1975 and the present to potentially unhealthy levels of either PM10 or PM2.5. This finding is based on census data and the area of impact shown in Figure 4. Of this exposed population, ATSDR estimates that at least 12,129 persons (that is, 6,619 children 6 years and younger and 5,510 adults aged 65 and older) are in subpopulations that may be sensitive to the effects of exposure to PM. It is important to note that it is likely that these estimates either overstate or understate the actual population exposed to unhealthy levels of PM. As indicated in Figure 4, since levels of air pollution were not measured at locations north of the EMF study area, ATSDR cannot establish the northern extent of the area of impact.
The health concerns expressed by community members in the EMF study area (i.e., increased incidence of asthma, upper respiratory illness, and heart disease) are reasonably consistent with adverse health outcomes reported in the epidemiological research for both acute and chronic exposures to PM2.5 and PM10 above health-based standards. However, the consistency between the concerns and the epidemiological studies does not suggest that any given incident of these health outcomes is caused solely by inhalation exposures to PM2.5 or PM10. Rather, causality of any given disease is usually a result of multiple factors. For example, smoking is a strong risk factor for many lung and heart diseases. Therefore, smokers comprise another population group at likely increased risk for PM-related health effects (USEPA 1996).
The following discussion first evaluates the increased risks from exposures to PM2.5 (annual averages) based on results from chronic mortality epidemiological studies and then evaluates the increased risks from exposures to PM2.5 and PM10 (24-hour maximum values) based on results from acute mortality and morbidity epidemiological studies. The ambient air concentrations of PM reported in these epidemiological studies is compared to estimated and measured levels of PM in the EMF study area. The discussions present a qualitative evaluation of the data collected in the EMF study area and should provide context for understanding the risk of adverse health effects to persons exposed in the EMF study area.
Chronic Exposures to Annual Average PM2.5 Levels. Two large cohort studies, the Harvard Six-City Study (Dockery 1993) and the American Cancer Society Study (ACS) (Pope 1995), found an association between excess mortality in adults and increasing PM2.5 concentrations in various cities and metropolitan areas of the United States (not including the Pocatello area). More specifically, the Harvard Six-City Study showed a 31% increase in mortality for every 25 ug/m3 increase in PM2.5, and the ACS study showed a 17% increase in mortality for every 25 ug/m3 increase in PM2.5. The reported ranges of annual average PM2.5 for the Harvard Six-City Study (HSCS) and the ACS study were 11-30 ug/m3 (mean) and 9-34 ug/m3 (median), respectively, for the least to the highest levels of PM2.5 in a given city during the study period. The risks calculated above were based on the excess mortality between the least to the most polluted cities (USEPA 1996).
Given the importance of the HSCS and ACS studies, HEI funded a study to re-analyze the results of the HSCS and ACS studies. The first major conclusion of the re-analysis study was that the original results of these two studies was of high quality and that the independent analysis of the data produced essentially the same results as the original studies. Moreover, the study tested the original results against a range of alternative variables and analytic models without substantially altering the original findings of an association between indicators of PM air pollution and mortality. In addition, an association between sulfur dioxide and mortality was observed and persisted when other possible confounding variables were included; furthermore, when sulfur dioxide was included in models with fine particulates or sulfate, the associations between these pollutants and mortality diminished. The study found relatively robust associations of mortality with fine particles, sulfates, and sulfur dioxide. The final interpretation by the researchers, related to their expanded analysis of the data, suggested that increased risk of mortality may be attributable to more than one component of the complex mix of ambient air pollutants in urban areas of the United States (Krewski, et al. 2000).
These and other chronic exposure studies, taken together, suggest that there may be increases in mortality in disease categories that are consistent with long-term exposure to airborne particles and that at least some fraction of these deaths reflect cumulative PM impacts above and beyond those exerted by acute exposures events (USEPA 1996). Also important is the fact that the Harvard Six-City Study and the ACS study controlled for subject-specific information regarding other relevant risk factors (such as cigarette smoking, occupational exposure, etc.); thus, these studies appear to provide reliable information about the effects of long-term exposures to PM (USEPA 1996). Moreover, the findings of an independent re-analysis by the HEI of these studies only serves to strengthen the conclusions of the original study and to show they were sound science. Overall, the weight-of-epidemiological data suggests long-term, repeated PM exposure has been associated with increased population-based mortality rates as well as increased risk of mortality in broad-based cohorts or samples of adults and children. Chronic exposures studies of PM suggest rather broad susceptibility to cumulative effects of long-term repeated exposure. There is no evidence that increased mortality risk is unique to any well-defined susceptible subgroup (Pope 2000).
Based on the epidemiological evidence, the extensive monitoring data available, and the estimates of historic levels of PM2.5, the community residing in the area of impact (see Figure 4); that is, in the populated areas northeast of FMC and Simplot (i.e., between the Pocatello Sewage Treatment Plant and Chubbuck School monitoring stations), may have experienced adverse health effects similar to those reported in the literature from chronic exposures to PM2.5 during several years between 1975 and 1993. Chronic exposures and the resulting increased risk of adverse health effects to those residing in Pocatello during this same time frame are also elevated but are likely to be less than those experienced by persons living in areas between Chubbuck and the Pocatello Sewage Treatment Plant. As previously indicated, the numerous studies on PM suggest that the elderly, individuals with pre-existing heart or lung disease, children (not included in Harvard Six-City Study or ACS Study), and asthmatics are the most at risk for adverse health effects from chronic exposure to PM2.5.
The epidemiological evidence, results of monitoring data from the EMF study area from 1994 to present (annual average PM10), and subsequent estimates of PM2.5 levels, indicate that exposure to PM during this time frame within the area of impact were likely to result in only minimal risks for adverse health effects for the general public and for probably many sensitive subpopulations. However, as previously indicated, there is no clear threshold level for PM. Therefore, some hypersensitive segments of the subpopulations residing in the EMF study area may have experienced adverse health effects from their long-term PM exposure during the 1994 to present time frame.
Persons living on the Fort Hall Indian Reservation, especially areas of the reservation nearest to the FMC and Simplot facilities, have likely been and are still being exposed to annual average levels of PM2.5 and PM10 above levels of health concern; however, the actual levels and areal extent of this exposure cannot be determined because of the lack of monitoring data north of the facilities and north of Interstate 86.
Acute Exposures to 24-Hour Average PM2.5 and PM10 Levels. Early indications that fine particles are likely important contributors to observed PM-mortality and morbidity (disease) effects came from evaluations of past serious air pollution episodes in Britain and the United States. The more severe episodes were characterized by several days of calm winds, during which large coarse particles rapidly settled out of the atmosphere and concentrations of fine mode particles dramatically increased (USEPA 1996). These meteorological conditions have been reported on numerous occasions in the EMF study area since 1975, the most recent being a severe 6-day inversion at the end of December 1999.
Most of the epidemiological studies of PM to date have focused on acute exposures (usually daily) and their association with various health end points; such as, mortality counts, hospitalizations, symptoms, and lung function. Unfortunately, until recently (following the promulgation of the new proposed PM2.5 standards), there have been very little daily monitoring of fine particles, and most of the studies used other methods of measuring particulate concentrations (Pope 2000). The table on the following page provides a summary of the epidemiological evidence of health effects of acute exposure to PM (Pope 2000).
Summary of Epidemiological Evidence of Health Effects of Acute Exposure to PM Air Pollutants (Adapted from Pope 2000) Health End Points Observed Association with PM Episodes of death and hospitalizations Elevated respiratory and cardiovascular mortality and hospitalizations. Mortality (death) Elevated daily respiratory and cardiovascular mortality counts. Effects persisted with various approaches to control for time trends, seasonality, and weather. Near-linear associations with little evidence of threshold. Hospitalization and other health-care visits Elevated hospitalizations, emergency room visits, and clinic/outpatient visits for respiratory and cardiovascular disease. Effects generally persisted with various approaches to control for time trends, seasonality, and weather. Symptoms/lung function Increased occurrence of lower respiratory symptoms, cough, and exacerbation of asthma. Only relatively weak associations with respiratory symptoms. Small, often significant declines in lung function. The results of a major study in the United States that evaluated the association of short-term exposures to PM10 and other pollutants, as related to mortality and morbidity (as measured by hospitalizations), was released in 2000 (Samet, et al. 2000). HEI's National Morbidity, Mortality, and Air Pollution Study (NMMAPS) used several new and innovative approaches to overcome some of the limitations of previous studies of daily exposures to air pollutants and its relationship to death and hospitalizations. The approach used was to characterize the effects of PM10 alone or in combination with gaseous air pollutants in a consistent way, in a large number of cities, using the same statistical approach. The study looked at the effects of PM10 and other pollutants on mortality in the 20 and 90 largest U.S. cities. In addition, the study looked at morbidity, as measured by daily PM10 effects on hospitalization among those 65 years of age and older, in 14 U.S. cities. The HEI concluded that the study has made substantial contribution in addressing major limitations of previous studies. The results of the 20 and 90 city mortality studies were generally consistent with an average approximate 0.5% increase in overall mortality for every 10 ug/m3 increase in PM10 measured the day before death. This effect was slightly higher for deaths due to heart and lung disease than for total deaths. The PM10 effect on mortality also did not appear to be affected by other pollutants in the model. The 14-city hospital admission study of persons 65 years or older indicated that there was a consistent approximate 1% increase in admissions for cardiovascular diseases and about a 2% increase in admissions for pneumonia and COPD for each 10 ug/m3 increase in PM10 (Samet, et al. 2000).
The results of these epidemiological studies suggest that the maximum 24-hour levels of PM10 and PM2.5 in the EMF study area between 1975 and the present (see Table A-1) have exceeded concentrations, on numerous occasions, that are associated with adverse health effects. The monitoring data and estimates suggest that the highest levels were detected either near the FMC and Simplot facilities or in the City of Pocatello. These data indicate that the population of Pocatello, because of the meteorological conditions that trap pollutants in the Portneuf Valley during inversion conditions, was at a higher risk of adverse health effects from acute levels of PM10 and PM2.5 than was the population of Chubbuck. However, this did not hold true during the December 1999 inversion, when the maximum PM2.5 levels for the same day (12/29/99), detected in Pocatello (119 ug/m3 at Garrett and Gould) and in Chubbuck (110 ug/m3 at Chubbuck School) were not considerably different. The risks of combined chronic and acute adverse health effects for other years, during the1975 to present time frame, for persons residing in Chubbuck and between the Pocatello Sewage Treatment Plant and Chubbuck would not be considered minimal.
According to the epidemiological literature, some of the adverse health effects associated with the range of maximum 24-hour levels of PM10 and 2.5 in the EMF study area, including the levels detected during the December 1999 inversion, are increased total acute mortality, increased hospital admissions for the elderly (>65 years) for lung and heart disease, chronic obstructive pulmonary disease (COPD), pneumonia, ischemic heart disease, and increased respiratory symptoms (i.e., increased cough and decreased lung function) (USEPA 1996). Overall, the PM risk estimates from total mortality epidemiological studies suggest that an increase of 10 ug/m3 in the 24-hour average PM10 level (or an increase of 5-6 ug/m3 in PM2.5) is associated with increased risks of adverse health effects of 0.5-1.5% (Pope 2000), with even higher risks possible for elderly sub-populations and for those with pre-existing respiratory conditions (USEPA 1996). Moreover, the levels of PM 2.5 detected in the Chubbuck and Pocatello areas, during the December 1999 inversion, were about 2 to 3 times higher than the AQI set by EPA (see previous discussion on the meaning of the AQI).
Persons living on the Fort Hall Indian Reservation, especially areas of the reservation nearest to the FMC and Simplot facilities, may have been and may still be exposed to maximum 24-hour levels of PM10 and PM2.5 above levels of health concern; however, the actual levels and areal extent of this exposure cannot be determined because of the lack of monitoring data north of the facilities (north of Interstate 86).
Sulfate Exposures. Some chronic epidemiological studies have shown that the annual mean levels of sulfate (SO4-2), a subset of fine PM, to be associated with increased mortality in adults, increased bronchitis in children, and decreased lung function in children (USEPA 1996). The two main studies (the Harvard Six-City Study and the ACS study) indicated that every 15 ug/m3 increase in annual average sulfate concentrations was associated with increases of 46 and 10%, respectively, in adult mortality (USEPA 1996). As previously indicated, annual average concentrations for sulfate ion in the EMF study area are not available for comparison to the levels found in epidemiologic studies associated with chronic adverse health effects.
Acute epidemiologic studies have associated sulfate exposures with increased hospitalizations and increased respiratory symptoms. The range of sulfate concentrations for these studies was 2-49 ug/m3. The five highest 24-hour sulfate ion concentrations detected at the IDEQ monitoring stations ranged from 18-73 ug/m3 for the STP monitor, 13-32 ug/m3 for the Chubbuck School monitor, 25-67 ug/m3 for the Garret and Gould monitor, and 26-84 ug/m3 for the ISU monitor. Based on these data and the results of the three epidemiological studies found in the literature, it can be reasonably assumed that persons, especially certain sensitive sub-populations residing in parts of Chubbuck and Pocatello, may have experienced an increased risk of adverse health effects during some of these days.
Acid Aerosol Exposures (including ionic species other than sulfates). Studies of past episodes of air pollution suggest that both acute and chronic health effects are associated with inhalation exposures to strongly acidic PM. For example, studies of historical pollution episodes, notably the London Fog episodes of the 1950's and early 1960's, indicate that acute exposures to extremely elevated levels of acid aerosols may be associated with excess human mortality. Studies evaluating present-day U.S. levels of acid aerosols have not found associations between acid aerosols and acute and chronic mortality, but the series of hydrogen ion (H+) data used may not have spanned a long enough time frame to detect H+ associations. However, several morbidity studies have associated H+ concentrations with increased bronchitis and reduced lung function in children and an increase in respiratory hospital admissions (USEPA 1996). Furthermore, based on animal studies, it is known that sulfuric acid aerosols exert their action throughout the respiratory tract, with the site of deposition dependent upon the particle size and the response dependent on mass and number concentration of specific deposition sites (USEPA 1996). However, the animal studies on acid aerosols provide no evidence that ambient acidic PM components contribute to mortality and essentially no quantitative guidance as to ambient acidic PM levels at which mortality would be expected to occur in either healthy or diseased humans. Furthermore, the effects seen in these animal studies were at acid levels that exceed worst-case ambient concentrations by more than an order of magnitude (USEPA 1996).
Several acids, such as, sulfuric acid, phosphoric acid, and hydrofluoric acid, are know to be released from the phosphate plants. In addition, phosphorous pentoxide (a signature constituent of the FMC emissions) and sulfur dioxide can be transformed in the atmosphere into phosphoric acid and sulfuric acid, respectively. All of these acids are considered potential respiratory irritants. The concentrations of ammonium ion present in filter samples is indicative of the elevated levels of ammonia being released in the EMF study area. It is possible, under certain conditions, that the levels of ammonia will neutralize all or some of the acids present in the ambient air thus ameliorating their potential respiratory effects. Because hydrogen ion data are quite limited in the EMF study area, a more definitive conclusion regarding the acidic nature of the ambient air in the EMF study area and resulting health implications cannot be made.
The presence of other ionic species, such as chloride and potassium ions, detected in the filter samples may be indicative of other acidic, basic, or other species (salts) that were present in the ambient air. Since the concentrations of these ions present in the EMF study area are relatively small, however, it cannot be determined from the available data if they contribute more or less to the overall acidity of the ambient air or are part of metallic or other salts that may have more important toxicological implications.
Exposures to Metals and Inorganics. The chemical analyses of filter samples performed during the RI, by the IDEQ, and by the Sho-Ban Tribe, present results for the elemental forms of metals and other inorganics. Therefore, the public health implications of exposure to the metals and other inorganics detected must be made on this basis. However, it is likely that the elements detected and presented in Table 3 were part of various compounds (either salts or covalently bound organic species of metals) which may be more or less toxic than the elemental species. However, it is important to note that scientific evidence indicates that different metallic salts show similar toxicity, whereas, more differences are found between elemental species with different valence states or metals covalently bonded to organic species. In some cases, the public health implications for these elements cannot be determined due to the paucity of studies for the elemental species. For example, the elements calcium, magnesium, and sodium were detected from filter samples; however, they were likely in the ambient air in the form of various salts formed with other elements. The public health implications of these metallic compounds cannot be determined, since the true forms of the metals in ambient air are not known. In some cases, the toxicity of the metallic compounds in ambient air may be greater (or less) than the elemental metal detected on a filter sample. Therefore, the toxicological evaluation of the individual elements below may overstate or understate the toxicological significance of exposure to metallic compounds in the ambient air. Acceptable analytical methods for determining the concentrations of metallic compounds in air have not been developed.
The public health implications of silicon, bromine, carbon, and chloride ion cannot be determined because they usually form other compounds of varying toxicological properties. For example, silicon in its crystalline forms has different toxicological significance than silicon in its amorphous form. The carbon fraction of ambient particulate matter consists of both elemental and organic carbon. Elemental carbon, also know as carbon black or graphitic carbon, has a chemical structure similar to impure graphite and is emitted directly into the atmosphere predominantly during combustion. Organic carbon is either emitted directly by sources or can be formed in the atmosphere by chemical reactions of hydrocarbons. Soot is commonly represented as elemental carbon, black carbon, or light absorbing carbon measured by thermal/optical or optical absorption techniques; however, soot has no firmly established definition (USEPA 1996).
The following discussion evaluates the public health implications of exposure to the eight metals that were detected above health-based comparison values: aluminum, arsenic, barium, beryllium, cadmium, chromium, manganese, and vanadium. As indicated above, only the public health implications of the elemental forms of these metals can be evaluated; these elemental forms are different from the species that may have been present in the ambient air. Furthermore, as previously indicated, the calculation of average annual metals concentrations and the reporting of 24-hour maximum levels were possible from the RI and Sho-Ban data. However, for the IDEQ data, only the maximum 24-hour levels were reported.
Aluminum. Elemental aluminum has not been classified as to its carcinogenicity. The average concentrations of aluminum detected at the RI and Sho-Ban monitors were all below levels of public health concern. However, the maximum level of aluminum detected at the Sho-Ban monitors (5.55 ug/m3) was above the chronic health comparison value (3.7 ug/m3) for non-carcinogenic health effects. The maximum level is more appropriately compared to levels in the literature that have caused adverse health effects because of short-term or acute exposures. The maximum levels of aluminum detected were compared to animal and human studies in the literature. Based on this evaluation, the levels detected in the EMF study area were about 540 and 1,260 times lower than the no-observed-adverse-effect level (NOAEL) and lowest-observed-adverse-effect level from animal studies (ATSDR 1999a); therefore, adverse health effects from short-term exposure to aluminum is not likely based on the available data. The maximum concentrations of aluminum detected at monitors located in residential areas were below health-comparison values.
Arsenic. EPA has classified arsenic as a human carcinogen via the inhalation route. Based on the highest average concentration of arsenic detected during the RI, exposure to arsenic would result in a no apparent increase risk of cancer. The maximum 24-hour level detected was compared to studies in the literature that investigated the non-carcinogenic effects of exposure to arsenic in animals and humans. Based on this comparison, the levels of arsenic in air were about 18,000 and 40,000 times lower than the NOAEL and the lowest-observed-adverse-effect level (LOAEL), respectively (ATSDR 2000a). Based on this analysis, it is unlikely that adverse health effects would result from short-term exposure to the levels detected in the EMF study area.
Barium. No studies were found in the literature regarding carcinogenic effects in humans or animals after inhalation exposure to barium (ATSDR 1992a). The average concentrations of barium detected during the RI were well below the chronic health comparison value for all monitoring stations. However, the maximum level detected for the IDEQ analysis of selected filter samples was slightly above the chronic health comparison value of 0.51 ug/m3 for non-carcinogenic health effects. Although there are not many studies in the literature for inhalation effects after exposure to barium, maximum levels of barium detected in the EMF study area were well below levels likely to result in adverse health effects from short-term exposures (ATSDR 1992a).
Beryllium. Beryllium is classified by EPA as a probable human carcinogen via the inhalation route. All of the average concentrations of beryllium detected during the RI were below the health-based comparison value for carcinogenic health effects. The maximum level of beryllium detected during the RI was at least 400,000 times lower than the lowest acute LOAEL for respiratory and other effects in animals (ATSDR 2000b). Therefore, adverse health effects from short-term exposure to the levels of beryllium detected in the EMF study area are not likely to occur.
Cadmium. EPA has classified cadmium as a probable human carcinogen via the inhalation route. Based on the highest average concentration of cadmium detected from samples taken during the RI and for the Sho-Ban monitoring, chronic exposure to cadmium would result in no apparent increased risk of cancer. The maximum level of cadmium detected during the RI, for the Sho-Ban monitoring, or during IDEQ's selective filter sampling, were evaluated to determine potential non-carcinogenic health effects from acute exposures to cadmium. Based on this evaluation, the maximum levels of cadmium found in residential areas of the EMF study were at least 3,900 and 6,700 times lower than the lowest NOAEL and LOAEL, respectively, for less serious health effects found in animal studies (ATSDR 1999b). For non-residential areas (near the FMC facility), the maximum levels of cadmium were at least 400 and 690 times lower than the lowest NOAEL and LOAEL, respectively, for less serious health effects found in animal studies (ATSDR 1999b). Moreover, for these same non-residential areas, the maximum levels of cadmium were at least 1,600 and 16,300 times lower than the lowest NOAEL and LOAEL, respectively, for serious respiratory effects found in animal studies (ATSDR 1999b). Based on this analysis alone, exposure to cadmium detected in the EMF study area is not likely to result in adverse health effects. However, there are some uncertainties with this evaluation related to cadmium and other metals. Please see the summary of the health effects of exposure to metals below for more details of these uncertainties.
Chromium. EPA considers hexavalent chromium to be a human carcinogen via the inhalation route; whereas, trivalent chromium has not been shown to be a carcinogen. Since the results from the RI are reported as total chromium, the concentrations of hexavalent chromium and trivalent chromium in the EMF study area are not known. Clearly, however, the relative quantity of hexavalent chromium cannot exceed the total chromium levels. Therefore, as a worst-case scenario of exposure, this analysis assumes that all of the total chromium reported is hexavalent chromium--a highly conservative assumption.
The resulting evaluation of the levels of chromium detected in residential areas (monitoring stations # 3 and #4) for their carcinogenic health effects, indicate a no apparent increased risk of cancer. In addition, if the highest average level of total chromium detected in non-residential area (Sho-Ban monitors next to FMC) were evaluated for its carcinogenic health risks, the resulting analysis would indicate a low risk of cancer. However, it is likely that the actual risks are lower because all of the chromium is probably not predominantly in the hexavalent form.
For acute non-carcinogenic health effects, the maximum total chromium concentration detected in residential areas would be about 57 times lower that the lowest LOAEL for less serious respiratory effects in studies of humans exposed to hexavalent chromium (ATSDR 2000c). However, when compared to studies of animals exposed to the less toxic trivalent chromium, the maximum exposure levels in residential areas is about 25,000 times lower than the lowest LOAEL for less serious respiratory health effects (ATSDR 2000c). The maximum total chromium concentration detected in non-residential areas of the EMF study area was from the Sho-Ban monitors. This level is about 10 times lower than the lowest LOAEL for less serious respiratory effects in humans exposed to hexavalent chromium (ATSDR 2000c). However, when compared to studies of animals exposed to the less toxic trivalent chromium, the maximum exposure levels in non-residential areas is about 4,500 times lower than the lowest LOAEL for less serious respiratory health effects (ATSDR 2000c).
For chronic non-carcinogenic health effects, the average concentration of total chromium detected in residential areas would be about 90 times lower than the lowest LOAEL for less serious respiratory effects in humans exposed to hexavalent chromium (ATSDR 2000c). However, when compared to studies of humans exposed to the less toxic trivalent chromium, the maximum exposure levels in residential areas is about 3,300 lower than the lowest NOAEL for renal effects and about 90,000 times lower than the lowest LOAEL for less serious respiratory health effects (ATSDR 2000c). The maximum total chromium concentration detected in non-residential areas of the EMF study area was from a sample from an RI monitor near the FMC and Simplot facilities. This level is about 115 times lower than the lowest LOAEL for less serious respiratory effects in humans exposed to hexavalent chromium (ATSDR 2000c). However, when compared to studies of humans exposed to the less toxic trivalent chromium, the maximum exposure levels in non-residential areas is about 4,300 times lower than the lowest NOAEL for renal effects and about 114,000 time lower than the lowest LOAEL for less serious respiratory health effects (ATSDR 2000c).
The actual hexavalent chromium levels in ambient air in the EMF study area are undoubtedly much lower than the total chromium levels used in the above evaluation. In this analysis, the actual estimates of health risk are likely closer to the estimates for studies in which humans and animals were exposed to the less toxic trivalent chromium. Therefore, persons living in populated and non-populated areas of the EMF study are not likely to experience adverse non-carcinogenic health effects from their short- or long-term exposures to chromium.
Manganese. No studies were found in the literature regarding carcinogenic effects in humans or animals after inhalation exposure to manganese (ATSDR 2000d). For non-carcinogenic health effects, the maximum level detected in the EMF study area (at the Sewage Treatment Plant) was compared to animal and human studies in the literature. Based on this evaluation, the maximum level detected in the EMF study area were about 11,600 times lower than the NOAEL for short-term adverse respiratory health effects found in animal studies (ATSDR 2000). Based on this evaluation, the levels of manganese detected in the EMF study are not likely to result in adverse health effects.
Vanadium. No studies were found in the literature regarding carcinogenic or chronic non-carcinogenic effects in humans or animals after inhalation exposure to vanadium (ATSDR 1992b). For short-term non-carcinogenic health effects, the maximum levels detected in the EMF study area were compared to animal and human studies in the literature. Based on the this evaluation, the maximum vanadium levels were about 75 times lower than the LOAEL for less serious respiratory effects in humans (i.e., bronchial irritation) (ATSDR 1992b). However, the maximum concentration detected was at the monitoring station located near the site perimeter and not in residential areas. Moreover, recent sampling at the site perimeter did not indicate that the levels of vanadium were above acute health-comparison values. The maximum levels detected in residential areas were below health comparison values. Based on this evaluation, it is unlikely that exposures to vanadium in populated areas of the EMF study would result in acute adverse health effects.
Summary of Metals Exposures. Although the above evaluation did not indicate a public health concern for individual metals, there is some uncertainty with this analysis. Current science provides little evidence as to whether the mix of these air contaminants may increase or decrease their toxicological effects because of cumulative exposures. Some of the metals (e.g., cadmium) were detected at levels in the fine fraction that were similar or greater than levels found in highly urbanized areas of the United States (ATSDR 1999). In addition, many of the metals detected in the EMF study area are transition metals.As indicated above, there is growing biological evidence that indicates that urban combustion particles (i.e., fine PM) can penetrate past the primary defense mechanisms of the lung, can elicit inflammatory changes in the lung and systematically (throughout the body), contain a constituent (soluble transition metals) that by itself can be demonstrated to produce lung damage, can produce electrocardiogram changes including arrhythmia (heart irregularities), and can kill animals with pre-existing heart and lung disease (Schwartz 1999). The extent to which the above evaluation of exposures to metals in the EMF study area is able to capture these concerns is not known. However, the epidemiological evidence (presented above) does indicate that PM, a measure of a mix of contaminants present in air, including all the metals detected in the EMF study area, is a good surrogate measure for estimating the short-term and long-term adverse cardiopulmonary health effects from exposure. From this standpoint, ATSDR evaluated and made definitive public health statements regarding the cumulative health effects of the exposure to the mix of metal contaminants present in the EMF study area as measured by PM.
Sulfur Dioxide Exposures. As previously indicated, annual average concentrations of sulfur dioxide at the Pocatello Sewage Treatment Plant have been below EPA's annual health-based standard since this monitoring station's inception. However, some 24-hour measurements of sulfur dioxide have exceeded EPA's health-based standard. In addition, the levels of sulfur dioxide detected at the STP during the period 1977-1985 exceeded ATSDR's Minimal Risk Level (MRL) of 0.01 ppm at least once a year during that period. Moreover, the maximum levels detected for these years indicate that levels of sulfur dioxide were 17-24 times higher than the MRL. Furthermore, ATSDR considers a concentration of sulfur dioxide of 0.1 ppm to be a minimal LOAEL (ATSDR 1998d). Available human controlled exposure studies indicate that sensitive asthmatics may respond to concentrations of sulfur dioxide as low as 0.1 ppm. Healthy non-asthmatics respond to higher concentrations of sulfur dioxide (greater than or equal to 1.0 ppm). Factors that have been shown to exacerbate the respiratory effects of sulfur dioxide include exercise and breathing of dry or cold air. Animal data support the human data on respiratory effects of sulfur dioxide (ATSDR 1998d).
As previously indicated, the only potentially unhealthy levels of sulfur dioxide measured in the EMF study area were detected at the Pocatello Sewage Treatment Plant during the years 1977 to 1985. Sulfur dioxide levels at this location did not exceed health-based comparison values from 1986 to the present, neither did sulfur dioxide levels at Garret and Gould between 1994 and 1999. Based on the available data, ATSDR suspects that the higher levels of sulfur dioxide from 1977 to 1985 were confined to areas in the immediate vicinity of the Pocatello Sewage Treatment Plant; however, ATSDR cannot rule out the possibility that certain sensitive individuals (i.e., asthmatics) were not exposed to sulfur dioxide at levels of health concern some time during this period. For these individuals, exposure to elevated levels of sulfur dioxide, along with elevated PM exposures, could increase the risk for adverse respiratory health effects. Since 1985, the levels of sulfur dioxide detected at the STP have been below levels of public health concern.
Potential Exposure to Phosphine and Hydrogen Cyanide from FMC. Phosphine, a colorless gas with a characteristic fish- or garlic-like odor, is a severe respiratory irritant. Gastrointestinal, respiratory, and central nervous system (CNS) effects have been noted in workers exposed to mean concentrations less than 10 ppm (Jones 1964). EPA has insufficient information to classify phosphine as to its potential as a human carcinogen (USEPA 1999b). NIOSH has a recommended exposure limit (REL) for phosphine of 0.3 ppm (300 ppb) and a short-term exposure limit (STEL) of 1 ppm (1,000 ppb) (NIOSH 1994). The RELs are time-weighted average (TWA) concentrations for up to a 10-hour workday during a 40-hour workweek, and the STEL is a 15-minute TWA exposure that should not be exceeded anytime during the workday (NIOSH 1994). As previously noted, FMC has measured some phosphine concentrations at the ponds at levels above the STEL. However, the public health implications of these environmental levels in relation to the on-site workers is beyond the scope of this health consultation. Using OSHA-approved methods, the maximum level of phosphine detected at the fence line was 101 ppb--an average of the fence line concentrations was not available. Based on limited animal studies reported by EPA (USEPA 1999b), short-term exposures (less than one year) to phosphine at the maximum levels detected at the fence line are not likely to result in adverse respiratory health effects. The effects of chronic exposures (greater than one year) to phosphine are still unknown (USEPA 1999b). However, additional sampling for phosphine at the fence line using other, less reliable, methods have on several occasions indicated that phosphine levels may have exceeded the STEL. These measured concentrations, if correct, suggest that a passerby, offsite worker (not FMC or Simplot), or other individual in the area might suffer from adverse health effect if exposed to the peak levels of phosphine for as little as 15 minutes.
Based on available data and knowledge of site-conditions, current exposures to the non-worker public would probably only be on an infrequent basis and for only a short duration. Therefore, based on limited environmental and scientific data alone, the occasional visitor to the area around the FMC site would not experience any adverse respiratory health effects from exposure to phosphine at 101 ppb. However, fence line and possibly off-site concentrations of phosphine may have been higher in the past and may have reached levels of public health concern (i.e., above the STEL) in the recent past, but the methods used may be unreliable. Therefore, the complete public health implication of off-site exposures to phosphine cannot be determined based on available data. Because of the toxicity of phosphine, continued operation of FMC's Pond Management Plan is needed to ensure that emissions do not reach levels of health concern to the off-site non-worker public. Moreover, more monitoring at the fence line, using OSHA-approved methods, is needed.
The maximum concentration of hydrogen cyanide (HCN) detected at the ponds was 990 ppb or 0.990 ppm. This level is almost five times lower than NIOSH's STEL (4.7 ppm)--NIOSH has not established a TLV-TWA guidance for HCN (ATSDR 1997b). The concentration of HCN at the fence line was compared to the lowest LOAELs reported in ATSDR's toxicological profile (ATSDR 1997e). The maximum HCN concentration at the perimeter is about 15, 100, and 140 time below the lowest chronic, intermediate, and acute LOAEL, respectively. Therefore, based on the current site conditions, where it is likely that current exposures to the non-worker public would be on an infrequent bases and for only a short time, it is not likely that adverse respiratory health effects would occur from exposure to the maximum HCN level detected at the fence line. However, fence line and possibly off-site concentrations of HCN may have been higher in the past. Therefore, the complete public health implication of off-site exposures to HCN cannot be determined based on available data. Because of the toxicity of HCN (albeit not as toxic as phosphine), continued operation of FMC's Pond Management Plan is needed to ensure that emissions do not reach levels of health concern to the off-site non-worker public.
Next Section Table
of Contents
