Outdoor Air Pollution
Authors
The information in this chapter was prepared by the following individuals:
Description of criteria air pollutants, except lead:
o Gerry Hiatt, Ph.D., EPA Region IX, San Francisco
Description of hazardous air pollutants:
o Will Humble, R.S., M.P.H., Risk Assessment and Environmental Epidemiology, Arizona Department of Health
Criteria Air Pollutants Except Lead
Scope
Several epidemiological studies have suggested an association between health effects and air pollution in metropolitan areas.
This section estimates the number of persons and severity of effects that may be caused by outdoor criteria air pollutants in Arizona. The contaminants considered here include the EPA criteria air pollutants:
o Particulate matter less than 10 microns in size (referred to as PM-10 particles)
Nitrogen oxides
Sulfur dioxides
Carbon monoxide
Ozone
Lead
This section discusses all the criteria air pollutants except lead, which is presented in another chapter.
Fine Particulates
Particulates are mixtures of solid and liquid droplets of material that vary in size and origin. Since only very small particles (less than 10 microns in size) can be inhaled into the respiratory tract, they are the most biologically threatening to humans. Particulates of this size are referred to as PM-10 particles.
Measured and reported by air pollution monitors, PM-10 particles in urban environments are often composed of soot, acid condensates, and sulfate and nitrate particles.
Health Effects
A report prepared for the American Lung Association (Chestnut, 1995) evaluated the health benefit of achieving a lower PM-10 standard. The current EPA standard for PM-10 is 150 microns/m3 for a 24-hour period, and 50 microns/m3 for an annual average.
The lower level assessed in the report was 50 microns/ m3 for a 24-hour period, and 30 microns/m3 for an annual average. These levels are in effect in California.
Similar to the Committee's approach, the American Lung Association report assumed a 1.5% increase in mortality for every 10 microns/m3 increase in PM-10 as a dose response estimate. The study evaluated PM-10 levels in Arizona and concluded that achieving the reduced levels would result in a reduction of 37 premature deaths in Arizona.
However, the study only evaluated the reduction in deaths if the California standard was achieved. It did not consider the benefits of reducing the PM-10 levels to background levels; it assumed that the lower standard was a health effect threshold. Evidence suggests, however, that the association between premature death and the concentration of PM-10 exists at levels well below the current standard.
Exposure and Effects in Arizona
The causal link between exposure to fine particulates and morbidity involving hospital visits and health effects is well established. Nonetheless, it is exceedingly difficult to estimate the actual number of these events in Arizona.
Estimates of the percentage increase of each of these conditions, however, may be made using dose-response information provided in an article by Dockery and Pope (1994). Table 13.1 applies the dose-response numbers from Dockery and Pope to annual PM-10 concentrations in order to estimate the percentage increase of various events and conditions in various Arizona metropolitan areas. The only category listed in the table for which the ADHS has numerical estimates is hospital admissions for respiratory diagnoses.
Table 13.1 Percentage Increase in Event or Condition from Fine Particulate Exposure
------------------------------------------------------------------------------------------------------------------------------------ | Location | Average Concentration | Population | Hospital Admissions | Asthma Episodes | Lower Respiratory | Cough | | | | | | | | | | | (microns/g/m) | | (Respiratory) | | Symptoms | | ------------------------------------------------------------------------------------------------------------------------------------ | | | | (%) | (%) | (%) | (%) | ==================================================================================================================================== | Phoenix Metro | 43 | 2,285,000 | 3 | 13 | 13 | 5 | ------------------------------------------------------------------------------------------------------------------------------------ | Tucson Metro | 24 | 428,000 | 2 | 7 | 7 | 3 | ------------------------------------------------------------------------------------------------------------------------------------ | Yuma | 40 | 58,000 | 3 | 12 | 12 | 5 | ------------------------------------------------------------------------------------------------------------------------------------ | Flagstaff | 26 | 48,000 | 2 | 8 | 8 | 3 | ------------------------------------------------------------------------------------------------------------------------------------ | Nogales | 50 | 20,000 | 4 | 15 | 15 | 6 | ------------------------------------------------------------------------------------------------------------------------------------ | Bullhead City | 34 | 20,000 | 3 | 7 | 7 | 4 | ------------------------------------------------------------------------------------------------------------------------------------ | Douglas | 39 | 15,000 | 3 | 12 | 12 | 5 | ------------------------------------------------------------------------------------------------------------------------------------ | Kingman | 16 | 15,000 | 1 | 5 | 5 | 2 | ------------------------------------------------------------------------------------------------------------------------------------ | Payson | 48 | 9,000 | 4 | 14 | 14 | 6 | ------------------------------------------------------------------------------------------------------------------------------------
As a first step in estimating annual deaths from particulates in Arizona, the number of hospital admissions for respiratory diagnoses in Arizona were determined.
o During 1993, the total number of hospital admissions with respiratory disease listed as the primary diagnosis in Arizona (ICD 485-496) totaled 21,806 in a population of 3,986,774. PM-10 concentrations are for 1991 (ADEQ, 1992).
o Areas outside the analyzed areas were assumed to have the same level of particulates as Prescott (17 microns/m3).
Application of the formula used in this study results in an estimated 600 hospital admissions attributable to fine particulate matter per year in which respiratory diseases (ICD 485-496) were listed as the primary cause of admission.
Mortality
Numerous studies have found associations between PM-10 pollution and mortality.
Many of the studies correlate episodes of extremely high concentrations of particulates with increased mortality. Recent studies, however, have found correlations between increased PM-10 pollution at lower levels and mortality from non-malignant respiratory diseases and cardiopulmonary diseases.
While none of the epidemiological studies prove a causal effect, when taken together, the studies indicate that a causal association exists, particularly among the elderly and those already suffering from a cardiopulmonary or respiratory disorder.
Some of the most robust relationships between PM-10 and mortality from cardiovascular and respiratory diseases are found in studies of air quality that correlate mortality and large temporary increases in particulate levels (increases in excess of 100 microns/m3). Schwartz and Dockery (1992) found that a 100 microns/ m3 increase in PM-10 concentrations led to an increase in mortality from obstructive pulmonary disease of 19%, and an increase in mortality from cardiovascular disease of 10%.
In a recent study of six U.S. cities by Dockery et al. (1993), a correlation was found between mortality from cardiopulmonary causes and particulate pollution levels. The study correlates increased mortality with non-episode periods of particulate pollution. While this study develops mortality rate ratios for each of the cities that indicate particulates as a risk factor for mortality, it does not determine the mortality per microns/m3 of particulates.
A literature review by Dockery and Pope (1994) includes four studies that provide a breakdown of mortality by cause-of-death categories. These studies found that cardiovascular deaths had effect estimates of between 0.8% and 1.8% increases in cardiovascular mortality per each increase in 10 microns/m3 in PM-10 concentration, with a weighted mean increase of 1.4%.
These studies also found that respiratory deaths had effect estimates of between 1.5% and 3.7% increases in respiratory mortality per each increase in 10 microns/m3 in PM-10 concentration, with a weighted mean increase of 3.5%.
To estimate the number of deaths in Arizona that may be attributed to particulate pollution, ADHS staff selected a method derived from Dockery and Pope (1994) which uses recent published research, information from the ADHS hospital discharge database, and air quality data from the ADEQ.
As a first step in estimating annual deaths from particulates in Arizona, the number of deaths from cardiovascular and respiratory diseases in Arizona and Maricopa County were determined by using death certificates.
o During 1993, the total number of registered deaths of residents in Maricopa County from cardiovascular and respiratory diseases (ICD 401-440, 485- 496) totaled 8,574 in a population of 2,285,200.
o The total number of registered deaths of residents outside Maricopa County from these causes totaled 7,854 in a population of 1,662,000.
Mortality rates for cardiovascular and respiratory diseases were calculated for Maricopa County directly. Mortality rates for the rest of the state were calculated by removing Maricopa County data from the Arizona totals.
Locations with fewer than one predicted death were included in the remainder category. The particulate concentration in the remainder category is considered background (17 microns/m3) and is representative of levels found in Prescott.
The death rate from cardiovascular and respiratory diseases was determined using the ADHS database. Particulate concentrations are annual averages for all monitoring stations in the air basin of interest (ADEQ, 1992).
Application of the formula used in this study results in an estimated 963 annual premature deaths from inhalation of particulates in Arizona. Uncertainties in the estimate include the method used to derive the dose-response number and the assumption that there is no threshold effect. The distribution of these deaths is shown in Table 13.2 on the next page.
Table 13.2 Total Estimated Annual Deaths from Fine Particulate Matter in Arizona
----------------------------------------- | Location | Annual Estimated | | | Deaths from PM-10 | ========================================= | Maricopa County | 667 | ----------------------------------------- | Tucson | 88 | ----------------------------------------- | Yuma | 20 | ----------------------------------------- | Flagstaff | 11 | ----------------------------------------- | Nogales | 8 | ----------------------------------------- | Bullhead City | 6 | ----------------------------------------- | Douglas | 5 | ----------------------------------------- | Kingman | 2 | ----------------------------------------- | Payson | 3 | ----------------------------------------- | Remainder | 153 | ----------------------------------------- | Total | 963 | -----------------------------------------
Nitrogen Oxides
Health Effects
Nitrogen oxides are reactive molecules including nitric oxide (NO), nitrogen dioxide (NO2) and nitrogen tetroxide (N2O4).
High temperature combustion of fossil fuels in motor vehicles is the primary source of nitrogen oxides in outdoor air. Nitrogen dioxide photo-dissociates into nitrogen oxide (NO), making an oxygen atom available to combine with an oxygen molecule to form ozone.
Nitrogen oxides are relatively insoluble in water, and therefore cause minimal upper respiratory effects.
Nitrogen dioxide is detectable by smell at between one and three parts per million (ppm), while health effects like mucus membrane occur at about 13 ppm.
Some studies in asthmatics indicate that exposure to between 0.2 and 0.3 ppm of nitrogen dioxide may enhance broncoconstriction. These studies have not had consistent results (Sullivan, 1992).
More serious health effects from nitrogen oxides are possible, but they occur during occupational or extraordinary circumstances.
The NAAQS for nitrogen oxides is 0.053 ppm (100 microns/m3) as an annual average. Concentrations of nitrogen oxides at this level would not be expected to result in any adverse health effects. Ozone, however, may form in the lower atmosphere as a result of nitrogen dioxide pollution, causing health effects caused by exposure to ozone.
Exposure and Effects in Arizona
The Phoenix metropolitan area has the highest level of NO2 in Arizona, with an average annual concentration of about 30 microns/m3 (ADEQ, 1992). Therefore, no adverse health effects would be expected to occur in Arizona as a direct result of exposure to nitrogen oxides.
Sulphur Dioxide
Health Effects
Sulphur dioxide is a very irritating, colorless, soluble gas with a strong, pungent odor and taste.
When sulphur dioxide comes into contact with water, it forms sulfuric acid, which irritates the eyes, mucus membranes, and skin. Sulphur dioxide is detectable by the nose at 0.5 ppm.
Instant mucus membrane irritation occurs at about 6 ppm. Other symptoms that may be experienced at about 6 ppm include eye irritation, runny nose, cough, shortness of breath, and a choking sensation (ADEQ, 1992).
The primary source of sulphur dioxide in outdoor air is the combustion of sulphur- containing fuels such as those used in some power plants.
In Arizona, the highest levels of sulphur dioxide are found in the smelter towns of San Manuel, Hayden, Winkelman, and Miami. These sources and other sources of sulphur dioxide are regulated by the EPA under the Clean Air Act. Under that act, the NAAQS for sulphur dioxide is 0.03 ppm (80 microns/m3) as an annual average, and 0.14 ppm (365 microns/m3) over a 24-hour period. Concentrations of sulphur dioxide at these levels were not expected to result in any adverse health effects.
Exposure and Effects in Arizona
The Hayden-Winkelman area has the highest level of sulphur dioxide in Arizona, with an average annual concentration of between 10 and 44 microns/m3.
The 24-hour maximum concentration of sulphur dioxide in Hayden in 1991 was 342 microns/m3. The only violation of sulphur dioxide standards during 1991 was in San Manuel (367 microns/m3). Annual and 24-hour maximum readings in the rest of the state are less than these concentrations. Therefore, no adverse health effects would be expected to occur in Arizona as a result of exposure to ambient levels of sulphur dioxide.
Carbon Monoxide
Health Effects
Carbon monoxide (CO) is an odorless, colorless gas that is readily absorbed by the lungs when inhaled. When inhaled, it attaches to hemoglobin in red blood cells, which forms carboxyhemoglobin (COHb), resulting in the inability of oxygen to attach to the hemoglobin. When this occurs, tissues suffer from lack of oxygen which causes a number of adverse health effects. These health effects begin as the number of hemoglobin binding sites available for oxygen decreases.
The blood's ability to carry oxygen is reduced at COHb levels between 2-30%. Carboxyhemoglobin levels up to 20% may result in headache upon exertion (Sullivan, 1992). At levels of 30-40%, dizziness may develop. At levels greater than 50%, seizures and coma may result. At high concentrations in poorly ventilated areas, carbon monoxide can cause coma and death.
Sources of carbon monoxide are regulated by the EPA under the Clean Air Act. Under that act, the NAAQS for carbon monoxide is 9 ppm (10 mg/m3) as an eight- hour average, and 35 ppm (40 mg/m3) over a one-hour period.
Exposure to concentrations of carbon monoxide at these levels, for these time periods, would be expected to result in carboxyhemoglobin levels of approximately 2% (Raub, 1989). Carboxyhemoglobin levels in this range may create subtle health effects such as experiencing exhaustion more quickly during exercise, but would not be expected to result in more obvious health effects such as headache or dizziness.
Exposure and Effects
The primary source of carbon monoxide in ambient air is automobile exhaust, and its highest concentrations in Arizona are found in the metropolitan Phoenix and Tucson areas.
There was one violation of the NAAQS for carbon monoxide in Arizona during 1993, a violation that occurred in a portion of Phoenix bounded by 24th St., 75th Ave., Van Buren, and Bethany Home Road (ADEQ, 1994).
Average carbon monoxide levels for the Phoenix area were not available. Yearly maximum eight-hour averages at many Phoenix sampling locations, however, are well below the 9 ppm standard.
The threshold carboxyhemoglobin level that produces acute health effects such as headache is approximately 20% COHb (Sullivan, 1992). Exposure to concentrations of carbon monoxide at the NAAQS levels would be expected to result in carboxyhemoglobin levels of approximately 2% (Raub, 1989).
Therefore, few noticeable adverse health effects would be expected to occur as a result of exposure to the concentrations of carbon monoxide found in Phoenix and Tucson air. If health effects occur on non-attainment days, the contribution made by carbon monoxide would be subtle health effects such as more rapid exhaustion during exercise.
Health effects that may result from conventional air pollutants in Arizona are measured in this section by estimating the number of minor restricted activity days (MRADS) from ozone, hospital admissions for respiratory diseases initiated by particulates, and annual premature deaths from particulates. The analysis produced the following results:
o About 1,600,000 person MRADs per year
o About 609 hospital admissions for respiratory disease from PM-10 per year
o About 963 premature respiratory and cardiovascular deaths per year from PM-10
Ozone
Health Effects
Ozone (O3) is a naturally-occurring and man-made gas with an electrical-like odor. It is the main oxidant in photochemical smog.
Ozone usually forms when nitrogen dioxide (NO2) photo-dissociates into nitrogen monoxide (NO), making an oxygen atom available to combine with an oxygen molecule (O2) to form ozone. Since the reaction is catalyzed by sunlight, ozone forms well on warm, sunny days.
Ozone is an irritant that primarily affects the respiratory tract. Symptoms of exposure include irritation of the skin, eyes, and mucus membranes. Other symptoms may include an increased respiratory rate, shallow breathing, cough, bronchitis, and pulmonary edema (Sittig, 1991). Neurological symptoms may include fatigue, dizziness, and headache.
The NAAQS for ozone is 0.12 parts per million (ppm) per hour, as an average.
At a normal breathing rate, concentrations in excess of 0.3 ppm may cause tightness of the chest, dry throat, wheezing, and irritation of the throat and lungs. At higher breathing rates such as those achieved through exercise, these effects may occur in susceptible persons at levels lower than the NAAQS of 0.12 ppm (Sullivan, 1992). Some epidemiological studies have suggested that ozone concentrations found in urban air may provoke asthmatic episodes in susceptible persons (Sullivan, 1992).
At any breathing rate, concentrations in excess of 0.5 ppm may result in headache, drowsiness, loss of coordination, and accumulation of fluid in the lungs. Levels in excess of 10 ppm may result in immediate, severe irritation of the lungs, continual coughing, and severe chemical pneumonia. Death may occur from prolonged exposures at 2 ppm or short-term exposures in excess of 10 ppm (Sittig, 1991).
Exposures and Effects in Arizona
The regions of Arizona that periodically approach or exceed the NAAQS for ozone include the Phoenix, Tucson and Yuma metropolitan areas.
Health effects from exposure to ozone in Arizona were assessed by using a variation on a formula developed by Hall et al. (1992). The analysis assumes a threshold for irritant effects and reduced activity of 0.08 ppm ozone. This concentration was considered appropriate since health effects from ozone exposure have been documented as low as 0.08 ppm (Hall, 1992).
The formula used in this study calculates the number of MRADs from outdoor ozone exposure. MRADs are events in which human activity is reduced, but not severely restricted. Health effects that may occur during MRADs include sore throat, mild cough, headache, chest discomfort, and eye irritation.
Applying Arizona data from 1993 (ADEQ, 1995) results in an estimate of about 1,600,000 MRADs per year in Arizona as a result of exposure to outdoor ozone. Each of these MRADs would be expected to reduce the activity or production of affected persons.
Hazardous Air Pollutants
Scope
Hazardous air pollutants (HAPs) is a legal category of air pollutants that includes over 180 chemicals.
The Committee's assessment of human health risks from inhaling airborne HAPs was based primarily on the results of a monitoring study of a limited series of HAPs conducted in the Phoenix metropolitan area in 1987 by the ADEQ. During this study, airborne levels of 24 volatile organic compounds (VOCs) were measured at 14 sites within the Phoenix urban area and at one distant site in rural Arizona.
VOC exposure concentrations in this study were calculated from the maximum detected value for each chemical, incorporating an adjustment to an annual average value based on the observed relationship between the maximum detected carbon monoxide (CO) level and the annual average CO level in Phoenix.
These average annual VOC values were used in constructing both Average and Reasonable Maximum Exposure scenarios. The two scenarios differed in exposure assumptions concerning length and duration of exposure and inhalation rate.
The sampling locations represented a mix of residential and industrial/commercial areas in metropolitan Phoenix. Air samples were collected during a three-month period on days when atmospheric dispersion was expected to be limited due to meteorological conditions. Therefore these data are probably representative of typical conditions averaged across an urban Arizona environment.
Cancer risks were determined using the "linear zero threshold model." This model assumes that there is no minimum threshold level below which no cancer risk occurs. In this model, cancer risk is calculated from exposure estimates and a potency factor. The potency factor is based upon the potential for a particular chemical to cause cancer.
Non-cancer health risks were estimated by comparing measured air concentrations of a particular chemical with levels known to produce toxicity. A hazard index of 1.0 or less means that there is no significant risk for non-cancer health risks.
Results
The results of the above calculations are summarized in Table 13.3, where the Hazard Index is defined as the ratio of measured air concentrations to threshold toxicity levels. A Hazard Index of 1.0 indicates no significant health risk.
Table 13.3 Estimated Cancer Risk from Select Hazardous Air Pollutants in Arizona
---------------------------------------------------------------------------------------------------------------- | | Individual Lifetime Risk | | Estimated Number of | | | | | | Cancers in Arizona, Per Year | | ---------------------------------------------------------------------------------------------------------------- | | Average | Reasonable | Average | Reasonable | | | Exposure | Maximum | Exposure | Maximum | | | | Exposure | | Exposure | ================================================================================================================ | 24 VOCs | | | | | ---------------------------------------------------------------------------------------------------------------- | Urban | 1.8 x 10-5 | 7.7 x 10-5 | 0.66 | 2.79 | ---------------------------------------------------------------------------------------------------------------- | Rural | 6.0 x 10-7 | 5.2 x 10-6 | 0.01 | 0.08 | ---------------------------------------------------------------------------------------------------------------- | Arsenic | | | | | ---------------------------------------------------------------------------------------------------------------- | Smelter towns | 5.2 x 10-6 | 2.2 x 10-5 | 0.0 | 0.01 | ---------------------------------------------------------------------------------------------------------------- | Rural | 1.8 x 10-7 | 7.7 x 10--7 | 0.00 | 0.01 | ---------------------------------------------------------------------------------------------------------------- | Total | -- | -- | 0.7 | 2.9 | ----------------------------------------------------------------------------------------------------------------
Urban Areas
Human health risk assessment of these 24 VOCs indicates that potential individual excess lifetime cancer risks are in the range of 1.8x10-5 to 7.7x10-5. For the urban population of Arizona (2,534,800 people), this corresponds to between 0.7 and 2.8 potential excess cancer cases per year from ambient exposure to these VOCs.
Assessment of the potential for non-cancer toxicity from exposure to these 24 VOCs did not indicate a significant hazard potential.
The total hazard index ranged between 0.3 and 0.5, but it should be noted that it is only appropriate to sum hazard indices for individual chemicals when they all produce the same type of toxicity or primarily affect the same organ system.
The primary contributor to this hazard index was tetrachloroethylene (perchloroethylene), for which the hazard index ranged 0.2 to 0.35. A hazard index less than 1.0, as in the present case, indicates there is no significant risk of non-cancer toxicity, even for susceptible individuals in the exposed population.
In the Committee's judgment, the Phoenix VOC study may have assessed only one- third to one-half of all HAPs present in Arizona's urban air.
Among the chemical classes not included were aldehydes, such as formaldehyde, and toxic metals like arsenic.
Therefore, to estimate the overall annual risks from inhalation of HAPs in urban environments, the VOC-based risk estimates are multiplied by two- to three-fold. The resulting estimates of cancer risk from ambient exposure to all hazardous air pollutants range between one to eight potential excess cancers per year.
Because different chemicals (especially those in different classes) produce different types of non-cancer toxicity, it is less appropriate to apply a similar correction factor to the non-cancer risk assessment to account for chemicals not included in the monitoring study.
Rural
The Phoenix air toxics study also provided limited information on exposure to VOCs in rural environments.
The 24 VOCs in the study were monitored at the Boyce Thompson Arboretum to represent a background measurement for the Phoenix study. The Arboretum is located approximately four miles southwest of Superior (population 3468) and 54 miles east of Phoenix.
A human health risk assessment was performed on the Arboretum VOC data in a manner similar to that for urban Arizona. Since these "rural" data represent only one geographic area, average and maximum ambient concentrations were calculated differently from the urban analysis.
The average exposure concentration was derived similar to the urban analysis, using an averaging factor derived from a ratio of the carbon monoxide (CO) levels during the study and the annual average CO level. The Reasonable Maximum Exposure scenario was based on the VOC levels actually measured at the Arboretum.
As with the urban analysis, the two scenarios also differed in exposure assumptions concerning length and duration of exposure and inhalation rate.
Human health risk assessment of the 24 VOCs studied indicates that, for the rural population of Arizona, this corresponds to between 0.01 and 0.08 potential excess cancer cases per year from ambient exposure to them.
Assessment of the potential for non-cancer toxicity from exposure to these 24 VOCs did not indicate a significant hazard potential for residents of rural Arizona.
The total hazard index, summed across all 24 VOCs, ranged 0.03 to 0.1. It should be noted, however, that it is only appropriate to sum hazard indices for individual chemicals when they all produce the same type of toxicity or primarily affect the same organ system.
A hazard index less than 1.0, as in the present case, indicates there is no significant risk of non-cancer toxicity, even for susceptible individuals in the exposed population.
As with the urban assessment, the Committee decided that this VOC study addressed only one-third to one-half of the total HAPs that may be present in the rural environment. Therefore, the VOC-based estimates of cancer risk were multiplied by two to three-fold to estimate the overall annual cancer risks from inhalation of all HAPs in rural Arizona.
The resulting estimates of cancer risk from ambient exposure to all hazardous air pollutants in the rural Arizona environment range between 0.02 to 0.24 potential excess cancer cases per year. Similar to the urban analysis, such a correction factor was not deemed appropriate for the non-cancer risk assessment.
These conclusions about potential human risks assume that rural VOC exposures can be approximated by the measurements taken at the Arboretum. However, the Arboretum data may significantly overestimate or underestimate average rural Arizona exposures to these chemicals.
Because the Arboretum is located near a highway, ambient concentrations of some VOCs, especially benzene, may be greater than for rural Arizona in general, thus overestimating exposures to VOCs.
On the other hand, the Arboretum may be in a more pristine, less populated area than most of rural Arizona; therefore, VOCs exposures may be underestimated by the Arboretum data.
Special Populations
Ambient air risks to special populations can be assessed in the following ways.
Industrial/Commercial Areas
A subset of the Phoenix urban air toxics data can be used to provide an estimate of potential health risks to people living adjacent to industrial/commercial facilities. Some of the air samples in the study were collected in industrial/commercial locations where exposure to VOCs is expected to be higher than for purely residential areas. In fact, most of the maximum air concentrations observed in the study were recorded at industrial/commercial locations.
Since the average urban air exposures were calculated by applying a correction factor (0.485) to the maximum VOC levels observed in the study, an estimate of potential risks from living near industrial/commercial facilities can be derived by eliminating this correction factor. The resulting risk estimates probably represent an upperbound to the actual risks since it is unlikely that any individual will be exposed the sum of these maximum VOC levels for the 30 to 70 year exposure periods assumed in the assessment.
The resultant individual cancer risks are estimated to be within the range of 3.5x10-5 to 1.4x10-4 for VOCs. Multiplying by two- to three-fold, to account for additional exposures to non-VOC chemicals, yields individual lifetime excess cancer risk estimates for exposure to all HAPs in the range 7x10-5 to 4.2x10-4.
Calculation of an annual population cancer risk requires an estimate of the number of Arizonans living in the proximity of such facilities, which is not available. It is likely, however, that this population risk will be less than the population risk estimate for the average Arizona urban population.
In a similar fashion, a non-cancer hazard potential can be estimated for residing adjacent to a primarily industrial/commercial area. Eliminating the correction factor from the non-cancer assessment yields a range of hazard index estimates from 0.6 to 1.8.
A hazard index greater than 1.0, as in the present analysis, indicates a potential risk of toxicity in some individuals exposed under the conditions of the assessment. This corresponds to individuals exposed to all of the VOCs in the study for six hours per day during light to moderate activities, such as most domestic work, climbing stairs, and making minor home repairs while outdoors.
A hazard index greater than 1.0 means that the reference dose(s) for the chemical(s) are exceeded for the exposures being assessed. As the frequency and/or magnitude of exposures exceeding the reference dose increases, the probability of toxicity increases. However, it should not be categorically concluded that all exposures greater than the reference dose will cause toxicity.
Conclusion
HAPs measured in the ADEQ study pose minimal risk to most Arizonans. The aggregate hazard index for urban dwellers was 0.3 - 0.5, and was less than 0.1 for rural populations. Hazard indices greater than one indicate threshold health affects. The Committee estimates that no more than one to eight cancers per year are caused by measured HAPs.
Measured HAPs did not include aldehydes. The health effect of aldehydes may be significant.
A very rough estimate of effects of HAPs on people living near industrial facilities shows that the individual cancer risk to measured HAPs may be as high as 4x10-4, and the cumulative hazard index may exceed one, which indicates that there is some non-cancer health risks. The Committee could not quantify these risks more accurately.