Evaluation of a Probabilistic Exposure Model Applied

The probabilistic National Ambient Air Quality Standards (NAAQS) Exposure Model applied to carbon monoxide (pNEM/CO) was developed by the U.S. Environmental Protection Agency (EPA) to estimate frequency distributions of population exposure to carbon monoxide (CO) and the resulting carboxyhemoglobin (COHb) levels. To evaluate pNEM/CO, the model was set up to simulate CO exposure data collected during a Denver Personal Exposure Monitoring Study (PEM) conducted during the winter of 1982-1983.

This paper compares computer-simulated exposure distributions obtained by pNEM/CO with the observed cumulative

relative frequency distributions of population exposure to CO from 779 people in the Denver PEM study. The subjects were disaggregated into two categories depending upon whether they lived in a home with a gas stove or an electric stove. The observed and predicted population exposure frequency distributions were compared in terms of 1-hr daily maximum exposure (1DME) and 8-hr daily maximum moving average exposure (8DME) for people living in homes with gas stove or an electric stove. For 1DME, the computer-simulated results from pNEM/CO agreed most closely within the range of 6-13 ppm, but overestimated occurrences at low exposure (<6 ppm) and underestimated occurrences at high exposure (>13 ppm). For 8DME, the predicted exposures agreed best with observed exposures in the range of CO concentration between 5.5 and 7 ppm, and over-predicted occurrences below 5.5 ppm and under-predicted occurrences above 7 ppm.     

Comparison of breath CO,CO exposure,and Coburn model predictions in the U.S. EPA Washington-Denver (CO) study

The U.S. EPA studied the carbon monoxide (CO) exposures and resulting breath CO concentrations of 625 non-smoking persons in Washington, D.C., and 454 non-smokers in Denver, CO, in the winter of 1982–83. Mean population-weighted breath concentrations were 5.1 ± 0.2 (SE) ppm in Washington and 7.2 ± 0.2 ppm in Denver. These values were correlated with the preceding personal air CO exposures (Spearman rank correlation coefficient r_s > 0.5, P < 0.0001) but not with the outdoor concentrations (r_s < 0.2). However, the breath measurements did not agree very closely with the personal exposures according to the current (Coburn) model relating alveolar CO to ambient CO. One reason for the discrepancy may have been the slight observed negative bias displayed by the personal monitors. A method of using the breath measurements to arrive at an improved estimate of personal exposures has been developed and applied. The method leads to an upward revision of exposure estimates: about 10% of the Washington target population of 1.22 million non-smokers are estimated to have exceeded the EPA 8-h ambient standard of 9 ppm during the winter of 1982–83, well above the 3.5% indicated by the personal monitor measurements.     

Alteration in Carboxyhemoglobin Concentrations During Exposure to 9 ppm Carbon Monoxide for 8 Hours at Sea Level and 2134 m Altitude in a Hypobaric Chamber

Seventeen non-smoking young men served as subjects to determine the alteration in carboxyhemoglobin (COHb) concentrations during exposure to 0 or 9 ppm carbon monoxide for 8 hours (CO) at sea level or an altitude of 2134 meters (7000 feet) in a hypobaric chamber. Nine subjects rested during the exposure and 8 exercised for 10 minutes of each exposure hour at a mean ventilation of 25 L (BTPS). All subjects performed a maximal aerobic capacity test at the completion of their respective exposures. Carboxyhemoglobin concentrations fell in all subjects during their exposures to 0 ppm CO at sea level or 2134 m. During the 8-h exposures to 9 ppm CO, COHb rose linearly from approximately 0.2 percent to 0.7 percent. No significant differences in uptake were found whether the subjects were resting or intermittently exercising during their 8-h exposures. COHb levels attained were similar at both sea level and 2134 m. Maximal aerobic capacity was reduced approximately 7-10 percent consequent to altitude exposure during 0 ppm CO exposures. These values were not altered following exposure for 8 h to 9 ppm CO in either the resting or exercising subjects.     

Laboratory and field evaluation of measurement methods for one-hour exposures to O3, PM2.5, and CO.

While researchers have linked acute (less than 12-hr) ambient O3, PM2.5, and CO concentrations to a variety of adverse health effects, few studies have characterized short-term exposures to these air pollutants, in part due to the lack of sensitive, accurate, and precise sampling technologies. In this paper, we present results from the laboratory and field evaluation of several new (or modified) samplers used in the "roll-around" system (RAS), which was developed to measure 1-hr O3, PM2.5, and CO exposures simultaneously. All the field evaluation data were collected during two sampling seasons: the summer of 1998 and the winter of 1999. To measure 1-hr O3 exposures, a new active O3 sampler was developed that uses two nitrite-coated filters to measure O3 concentrations. Laboratory chamber tests found that the active O3 sampler performed extremely well, with a collection efficiency of 0.96 that did not vary with temperature or relative humidity (RH). In field collocation comparisons with a reference UV photometric monitor, the active O3 sampler had an effective collection efficiency ranging between 0.92 and 0.96 and a precision for 1-hr measurements ranging between 4 and 6 parts per billion (ppb). The limits of detection (LOD) of this method were 9 ppb-hr for the chamber tests and approximtely 16 ppb-hr for the field comparison tests. PM2.5 and CO concentrations were measured using modified continuous monitors--the DustTrak and the Langan, respectively. A size-selective inlet and a Nafion dryer were placed upstream of the DustTrak inlet to remove particles with aerodynamic diameters greater than 2.5 microm and to dry particles prior to the measurements, respectively. During the field validation tests, the DustTrak consistently reported higher PM2.5 concentrations than those obtained by the collocated 12-hr PM2.5 PEM samples, by approximately a factor of 2. After the DustTrak response was corrected (correction factor of 2.07 in the summer and 2.02 in the winter), measurements obtained using these methods agreed well with R2 values of 0.87 in the summer and 0.81 in the winter. The results showed that the DustTrak can be used along with integrated measurements to measure the temporal and spatial variation in PM2.5 exposures. Finally, during the field validation tests, CO concentrations measured using the Langan were strongly correlated with those obtained using the reference method when the CO levels were above the LOD of the instrument [approximately 1 part per million (ppm)].     

Laboratory and Field Evaluation of Measurement Methods for One-Hour Exposures to O3, PM25, and CO

ABSTRACT

While researchers have linked acute (less than 12-hr) ambient O₃, PM₂₅, and CO concentrations to a variety of adverse health effects, few studies have characterized short-term exposures to these air pollutants, in part due to the lack of sensitive, accurate, and precise sampling technologies. In this paper, we present results from the laboratory and field evaluation of several new (or modified) samplers used in the “roll-around” system (RAS), which was developed to measure 1-hr O₃, PM₂₅, and CO exposures simultaneously. All the field evaluation data were collected during two sampling seasons: the summer of 1998 and the winter of 1999.

To measure 1-hr O₃ exposures, a new active O₃ sampler was developed that uses two nitrite-coated filters to measure O₃ concentrations. Laboratory chamber tests found that the active O₃ sampler performed extremely well, with a collection efficiency of 0.96 that did not vary with temperature or relative humidity (RH). In field collocation comparisons with a reference UV photometric monitor, the active O₃ sampler had an effective collection efficiency ranging between 0.92 and 0.96 and a precision for 1-hr measurements ranging between 4 and 6 parts per billion (ppb). The limits of detection (LOD) of this method were 9 ppb-hr for the chamber tests and ~16 ppb-hr for the field comparison tests.

PM_2.5 and CO concentrations were measured using modified continuous monitors—the DustTrak and the Langan, respectively. A size-selective inlet and a Nafion dryer were placed upstream of the DustTrak inlet to remove particles with aerodynamic diameters greater than 2.5 um and to dry particles prior to the measurements, respectively. During the field validation tests, the DustTrak consistently reported higher PM_2.5 concentrations than those obtained by the collocated 12-hr PM_{2 5} PEM samples, by approximately a factor of 2. After the DustTrak response was corrected (correction factor of 2.07 in the summer and 2.02 in the winter), measurements obtained using these methods agreed well with R² values of 0.87 in the summer and 0.81 in the winter. The results showed that the DustTrak can be used along with integrated measurements to measure the temporal and spatial variation in PM_{2 5} exposures. Finally, during the field validation tests, CO concentrations measured using the Langan were strongly correlated with those obtained using the reference method when the CO levels were above the LOD of the instrument [~1 part per million (ppm)].     

Predicting particulate (PM10) personal exposure distributions using a random component superposition statistical model

This paper presents a new statistical model designed to extend our understanding from prior personal exposure field measurements of urban populations to other cities where ambient monitoring data, but no personal exposure measurements, exist. The model partitions personal exposure into two distinct components: ambient concentration and nonambient concentration. It is assumed the ambient and nonambient concentration components are uncorrelated and add together; therefore, the model is called a random component superposition (RCS) model. The 24-hr ambient outdoor concentration is multiplied by a dimensionless "attenuation factor" between 0 and 1 to account for deposition of particles as the ambient air infiltrates indoors. The RCS model is applied to field PM10 measurement data from three large-scale personal exposure field studies: THEES (Total Human Environmental Exposure Study) in Phillipsburg, NJ; PTEAM (Particle Total Exposure Assessment Methodology) in Riverside, CA; and the Ethyl Corporation study in Toronto, Canada. Because indoor sources and activities (smoking, cooking, cleaning, the personal cloud, etc.) may be similar in similar populations, it was hypothesized that the statistical distribution of nonambient personal exposure is invariant across cities. Using a fixed 24-hr attenuation factor as a first approximation derived from regression analysis for the respondents, the distributions of nonambient PM10 personal exposures were obtained for each city. Although the mean ambient PM10 concentrations in the three cities varied from 27.9 micrograms/m3 in Toronto to 60.9 micrograms/m3 in Phillipsburg to 94.1 micrograms/m3 in Riverside, the mean nonambient components of personal exposures were found to be closer: 52.6 micrograms/m3 in Toronto; 52.4 micrograms/m3 in Phillipsburg; and 59.2 micrograms/m3 in Riverside. The three frequency distributions of the nonambient components of exposure also were similar in shape, giving support to the hypothesis that nonambient concentrations are similar across different cities and populations. These results indicate that, if the ambient concentrations were completely controlled and set to zero in all three cities, the median of the remaining personal exposures to PM10 would range from 32.0 micrograms/m3 (Toronto) to 34.4 micrograms/m3 (Phillipsburg) to 48.8 micrograms/m3 (Riverside). The highest-exposed 30% of the population in the three cities would still be exposed to 24-hr average PM10 concentrations of 47-74 micrograms/m3; the highest 20% would be exposed to concentrations of 56-92 micrograms/m3; the highest 10% to concentrations of 88-131 micrograms/m3; and the highest 5% to 133-175 micrograms/m3, due only to indoor sources and activities. The distribution for the difference between personal exposures and indoor concentrations, or the "personal cloud," also was similar in the three cities, with a mean of 30-35 micrograms/m3, suggesting that the personal cloud accounts for more than half of the nonambient component of PM10 personal exposure in the three cities. Using only the ambient measurements in Toronto, the nonambient data from THEES in Phillipsburg was used to predict the entire personal exposure distribution in Toronto. The PM10 exposure distribution predicted by the model showed reasonable agreement with the PM10 personal exposure distribution measured in Toronto. These initial results suggest that the RCS model may be a powerful tool for predicting personal exposure distributions and statistics in other cities where only ambient particle data are available.     

Residential Indoor,Outdoor, and Workplace Concentrations of Carbonyl Compounds: Relationships with Personal Exposure Concentrations and Correlation with Sources

Abstract

Personal 48-hr exposures of 15 randomly selected participants as well as microenvironment concentrations in each participant’s residence and workplace were measured for 16 carbonyl compounds during summer–fall 1997 as a part of the Air Pollution Exposure Distributions within Adult Urban Populations in Europe (EXPOLIS) study in Helsinki, Finland. When formaldehyde and acetaldehyde were excluded, geometric mean ambient air concentrations outside each participant’s residence were less than 1 ppb for all target compounds. Geometric mean residential indoor concentrations of carbonyls were systematically higher than geometric mean personal exposures and indoor workplace concentrations. Additionally, residential indoor/outdoor ratios indicated substantial indoor sources for most target compounds. Carbonyls in residential indoor air correlated significantly, suggesting similar mechanisms of entry into indoor environments. Overall, this study demonstrated the important role of non-traffic-related emissions in the personal exposures of participants in Helsinki and that comprehensive apportionment of population risk to air toxics should include exposure concentrations derived from product emissions and chemical formation in indoor air.     

Vehicle Occupant Exposure to Carbon Monoxide

Abstract

This paper focuses on the auto commuting micro-environment and presents typical carbon monoxide (CO) concentrations to which auto commuters in central Riyadh, Saudi Arabia were exposed. Two test vehicles traveling over four main arterial roadways were monitored for inside and outside CO levels during eighty peak and off-peak hours extending over an eight month period. The relative importance of several variables which explained the variability in CO concentrations inside autos was also assessed. It was found that during peak hours auto commuters were exposed to mean CO levels that ranged from 30 to 40 ppm over trips that typically took between 25 to 40 minutes. The mean ratio of inside to outside CO levels was 0.84. Results of variance component analyses indicated that the most important variables affecting CO concentrations inside autos were, in addition to the smoking of vehicle occupants, traffic volume, vehicle speed, period of day and wind velocity. An increase in traffic volume from 1,000 to 5,000 vehicles per hour (vph) increased mean CO level exposure by 71 percent. An increase in vehicle speed from 14 to 55 km/h reduced mean CO exposure by 36 percent. The number of traffic interruptions had a moderate effect on mean concentrations of CO inside vehicles.     

The Exposure to Carbon Monoxide of Occupants of Vehicles Moving in Heavy Traffic

Carbon monoxide and hydrocarbons were sampled at operator’s nose height inside vehicles moving in moderate to heavy traffic in six cities. The samples were integrated over 20-30 minutes by collection in Mylar bags. Carbon monoxide and hydrocarbons were analyzed by infrared and flame ionization, respectively, with instruments at the Continuous Air Monitoring Program (CAMP) station in each city. Detector tubes for carbon monoxide were also used to determine 5-min concentrations at suspected high points in the field. Estimates of traffic density were made. Three types of traffic arteries were considered: (7) heavily traveled, wide expressways, (2) main city streets with moderately rapid vehicular traffic, and (3) center city streets with slow-moving traffic. Integrated half-hour CO concentrations obtained within the vehicles while in traffic were generally considerably higher than the concurrent concentrations measured at the CAMP sites. In-traffic CO values in all cities sampled exceeded 30 ppm in at least 10% of the integrated samples. The range of city averages was 21–39 ppm carbon monoxide and the range of individual integrated samples was 7–77 ppm of carbon monoxide.     

Personal carbon monoxide exposures of preschool children in Helsinki, Finland—comparison to ambient air concentrations

The associations of personal carbon monoxide (CO) exposures with ambient air CO concentrations measured at fixed monitoring sites, were studied among 194 children aged 3–6 yr in four downtown and four suburban day-care centers in Helsinki, Finland. Each child carried a personal CO exposure monitor between 1 and 4 times for a time period of between 20 and 24 h. CO concentrations at two fixed monitoring sites were measured simultaneously. The CO concentrations measured at the fixed monitoring sites were usually lower (mean maximum 8-h concentration: 0.9 and 2.6 mg m⁻³) than the personal CO exposure concentrations (mean maximum 8-h concentration: 3.3 mg m⁻³). The fixed site CO concentrations were poor predictors of the personal CO exposure concentrations. However, the correlations between the personal CO exposure and the fixed monitoring site CO concentrations increased (−0.03–−0.12 to 0.13–0.16) with increasing averaging times from 1 to 8 h. Also, the fixed monitoring site CO concentrations explained the mean daily or weekly personal CO exposures of a group of simultaneously measured children better than individual exposure CO concentrations. This study suggests that the short-term CO personal exposure of children cannot be meaningfully assessed using fixed monitoring sites.     

Spatial and Temporal Variability of Ozone Exposure in Forested Areas of the United States and Canada: 1978—1988

Ambient O₃ exposures have reduced growth rates of tree genotypes in some areas of the United States. For characterizing O₃ exposures in forested areas, data from primarily population-oriented sites have been used. It has been speculated that exposures calculated from population-oriented sites provide estimates greater than those that would actually be experienced in the majority of forested areas. Accordingly, we compared 1988 O₃ data from three remote forested sites with data from several population-oriented monitoring sites in and around the mid? and southern Appalachian Mountains. The number of hours ≥0.08 ppm was lower at the remote forested sites than at the nearby population-oriented locations. In addition, we characterized the temporal variability of O₃ exposures in forested regions of the United States and Canada for the period 1978-1988. We found that the years of highest O₃ exposure in the eastern United States during 1978-1988 were 1978, 1980, 1983, and 1988, with 1988 being the worst year in four of seven eastern forest regions. In 1988, the Whiteface Mountain summit site (1483 m) experienced approximately 10 percent more hourly average concentrations ≥0.08 ppm than in the second highest O₃ exposure year (i.e., 1979). Consistently throughout the 11-year period, the highest O₃ exposures at the Whiteface Mountain site occurred during the late evening and early morning hours, with the result that the longterm 7-h (0900-1559h) exposure index could not distinguish those years in which the highest exposures occurred from those in which the lowest occurred. Similar to the Whiteface Mountain site, two high-elevation Shenandoah National Park sites experienced their highest O₃ exposures in 1988. With the exception of 1986, the lower elevation site (Dickey Ridge) consistently experienced more frequent occurrences of hourly average concentrations ≥0.08 ppm than the higher elevation site (Big Meadows).     

Outdoor/Indoor/Personal ozone exposures of children in Nashville, Tennessee

An ozone (O3) exposure study was conducted in Nashville, TN, using passive O3 samplers to measure six weekly outdoor, indoor, and personal O3 exposure estimates for a group of 10- to 12-yr-old elementary school children. Thirty-six children from two Nashville area communities (Inglewood and Hendersonville) participated in the O3 sampling program, and 99 children provided additional time-activity information by telephone interview. By design, this study coincided with the 1994 Nashville/Middle Tennessee Ozone Study conducted by the Southern Oxidants Study, which provided enhanced continuous ambient O3 monitoring across the Nashville area. Passive sampling estimated weekly average outdoor O3 concentrations from 0.011 to 0.O30 ppm in the urban Inglewood community and from 0.015 to 0.042 ppm in suburban Hendersonville. The maximum 1- and 8-hr ambient concentrations encountered at the Hendersonville continuous monitor exceeded the levels of the 1- and 8-hr metrics for the O3 National Ambient Air Quality Standard. Weekly average personal O3 exposures ranged from 0.0013 to 0.0064 ppm (7-31% of outdoor levels). Personal O3 exposures reflected the proportional amount of time spent in indoor and outdoor environments. Air-conditioned homes displayed very low indoor O3 concentrations, and homes using open windows and fans for ventilation displayed much higher concentrations.     

Tetrachloroethylene Emissions and Exposure in Dry Cleaning

ABSTRACT

Tetrachloroethylene (PCE) emissions and the exposure of workers in six commercial and three industrial dry-cleaning establishments that use dry-to-dry machines were determined. The personal samples and area samples [8-hr time-weighted average (TWA) and short-term exposure] were collected with charcoal tubes and passive monitors. The temporal variation of PCE concentration in the workplace air was monitored using a Fourier transform infrared analyzer (FTIR). The PCE emission rates were determined by multiplying the average PCE concentration in the room and the total airflow rate in the room. The PCE emissions were related to the cleaning rate in units of kg/hr.

The operators' mean TWA exposure in commercial shops and industrial establishments was 28 (4.1 ppm) and 32 mg/m³ (4.6 ppm), and the pressers' exposure was 3.4 (0.5 ppm) and 7.7 mg/m³ (1.1 ppm), respectively. The customer service personnel had the lowest TWA exposure with a mean value of 0.8 mg/m³ (0.1 ppm). The highest peak concentration (2300 mg/m³; 334 ppm) was observed during cleaning of the lint and button trap, during which operation respirators were used. The PCE emission rates ranged from 4 to 118 g/hr corresponding to emission factors (mass of solvent evaporated per mass of cleaned cloths) of 0.3–3.6 g/kg. The workers' exposure to PCE was below the occupational limit values in the United States [according to the American Conference of Governmental Industrial Hygienists (ACGIH)] and in Finland. The outdoor PCE emissions were clearly below the limit values given in the European Union volatile organic compound (VOC) directive requirements.     

Epidemiological Bases for Possible Air Quality Criteria for Carbon Monoxide

Exposures to adequate environmental levels of CO will increase COHb concentrations in human subjects. The amount of this increase is reasonably predictable, and must be considered in relation to exposure to CO in inhaled cigarette smoke as well as to occupational and domestic exposures. The increase in body COHb will result in some degree of impairment of tissue oxygenation.

Methods for estimating COHb levels in large populations are relatively simple. The assumption that an exposure to 30 ppm CO for eight hours will produce on the average, an increase in COHb of 5%, has been substantiated by available data.

Exposure for five hours to between 10 and 12 ppm of CO has been shown to increase the COHb levels in nonsmokers by at least 0.5%. Such an increase adds appreciably to the body burden of COHb in those who do not already have such a body burden from cigarette smoking. Longer exposures could have produced a somewhat greater increase.

Apart from increases in COHb, three possible effects have been a source of major consideration in epidemiologic studies. The first is the production of some persistent toxic reaction. This possibility has been examined with respect to occupational exposure, and the evidence for the occurrence of such a condition is insufficient.

The possible contribution of ambient community CO exposure to the mortality of persons hospitalized with myocardial infarction has been investigated. The evidence suggests that daily average CO values in excess of about 10 ppm may be associated with an increase in mortality in hospitalized patients with myocardial infarction. Substantiation of this impression will require a study of the prognosis of myocardial infarction patients in relationship to COHb levels measured at admission to the hospital.

Finally, in two studies, persons driving motor vehicles which were involved in accidents had higher COHb levels than "control" populations. Controls were not ideal, however. Possible mechanisms by which CO might affect the ability to drive a motor vehicle is suggested in the available data on CO effects upon visual sensitivity, psychological test performance and accurate estimation of time intervals. As little as 2 percent COHb can produce these effects in laboratory studies, and the available epidemiologic information confirms that such an increase in COHb levels among drivers might influence the frequency of accidents.

Specific areas where research is indicated to clarify uncertainties relating to health effects of CO are: 1. The increment in COHb which can be produced by exposures to an average of 20 ppm CO for an eight hour period and the increment which can be produced by 15 ppm for such a period and by 10 ppm for up to twenty-four hours.

2. The relationship of ambient CO levels and of COHb levels to the survival of hospitalized patients with myocardial infarction.

3. The prognostic significance with respect to cardiovascular conditions of elevated levels of COHb.

4. The relationship, if any, between ambient CO and COHb levels and the occurrence of motor vehicle accidents when weather and driving conditions, cigarette smoking, alcohol and drug use, and other factors are adjusted and controlled.

     

Total Human Exposure: Basic Concepts,EPA Field Studies,and Future Research Needs

Historically, environmental regulatory programs designed to protect public health have monitored pollutants only in geophysical carrier media (for example, outdoor air, streams, soil). Field studies have identified a gap between the levels observed in geophysical carrier media and the concentrations with which people actually come into contact: their daily exposures. A new approach—Total Human Exposure (THE)—has evolved to fill this gap and provide the critical data needed for accurately assessing public health risk. The THE approach considers a three-dimensional "bubble" around each person and measures the concentrations of all pollutants contacting that bubble, either through the air, food, water, or skin. Two basic THE approaches have emerged: (1) the direct approach using probability samples of populations and measuring pollutant concentrations in the food eaten, air breathed, water drunk, and skin contacted; and (2) the indirect approach using human activity pattern-exposure models to predict population exposure distributions. Using the direct approach, EPA has conducted over 20 field studies for pollutants representing four groups—volatile organic compounds, carbon monoxide, pesticides, and particles—in 15 cities in 12 states. The indirect modeling approach has been applied to several of these pollutants. Additional research is needed in a great variety of areas. Even from the few projects completed thus far, the THE approach has yielded a rich new data base for risk assessments and has provided many surprises about the relative contribution of various pollutant sources to public health risk.     

Carbon Monoxide Exposures Inside an Automobile Traveling on an Urban Arterial Highway

Carbon monoxide (CO) exposures were measured inside a motor vehicle during 88 standardized drives on a major urban arterial highway, El Camino Real (traffic volume of 30,500-45,000 vehicles per day), over a 13-1/2 month period. On each trip (lasting between 31 and 61 minutes), the test vehicle drove the same 5.9-mile segment of roadway in both directions, for a total of 11.8 miles, passing through 20 intersections with traffic lights (10 in each direction) in three California cities (Menlo Park, Palo Alto, and Los Altos). Earlier tests showed that the test vehicle was free of CO intrusion. For the 88 trips, the mean CO concentration was 9.8 ppm, with a standard deviation of 5.8 ppm. Of nine covariates that were examined to explain the variability in the mean CO exposures observed on the 88 trips (ambient CO at two fixed stations, atmospheric stability, seasonal trend function, time of day, average surrounding vehicle count, trip duration, proportion of time stopped at lights, and instrument type), a fairly strong seasonal trend was found. A model consisting of only a single measure of traffic volume and a seasonal trend component had substantial predictive power (R² = 0.68); by contrast, the ambient CO levels, although partially correlated with average exposures, contributed comparatively little predictive power to the model. The CO exposures experienced while drivers waited at the red lights at an intersection ranged from 6.8 to 14.9 ppm and differed considerably from intersection to intersection. A model also was developed to relate the short-term variability of exposures to averaging time for trip times ranging from 1 to 20 minutes using a variogram approach to deal with the serial autocorrelation. This study shows: (1) the mass balance equation can relate exterior CO concentrations as a function of time to interior CO concentrations; (2) CO exposures on urban arterial highways vary seasonally; (3) momentary CO exposures experienced behind red lights vary with the intersection; and (4) an averaging time model can simulate exposures during short trips (20 minutes or less) on urban arterial highways.     

Exposure of Chronic Obstructive Pulmonary Disease Patients to Particulate Matter: Relationships between Personal and Ambient Air Concentrations

ABSTRACT

Most time-series studies of particulate air pollution and acute health outcomes assess exposure of the study population using fixed-site outdoor measurements. To address the issue of exposure misclassification, we evaluate the relationship between ambient particle concentrations and personal exposures of a population expected to be at risk of particle health effects.

Sampling was conducted within the Vancouver metropolitan area during April-September 1998. Sixteen subjects (non-smoking, ages 54-86) with physician-diagnosed chronic obstructive pulmonary disease (COPD) wore personal PM_{2 5} monitors for seven 24-hr periods, randomly spaced approximately 1.5 weeks apart. Time-activity logs and dwelling characteristics data were also obtained for each subject. Daily 24-hr ambient PM₁₀ and PM_2.5 concentrations were measured at five fixed sites spaced throughout the study region. SO₄ ^2-, which is found almost exclusively in the fine particle fraction and which does not have major indoor sources, was measured in all PM_{2 5} samples as an indicator of accumulation mode particu-late matter of ambient origin.