UC Berkeley

To effectively control health risks associated with an airborne contaminant we need to understand when, where, why, and how much humans come into contact with the contaminant. To answer these questions, the temporal and spatial variability in levels of species must be evaluated in relation to the locations of humans in space and time. Characterizing human exposure through the measurement of pollutant levels within occupied microenvironments where people spend time is particularly important for species that have sharp gradients owing to rapid environmental processing. This is especially true if the pollutant dynamics are influenced by the presence or activities of the occupants themselves.

This dissertation investigates inhalation exposures to two dynamic air pollutants in two important settings: ultrafine particles (UFP) in residences and ozone in aircraft cabins. New field data were acquired and observed pollutant trends were modeled to assess the importance for indoor concentrations and exposures of outdoor levels, ventilation characteristics, indoor sources, pollutant dynamics, human factors, and control strategies. Study findings can be applied to assess the risk associated with each exposure scenario and to suggest conditions under which interventions are likely to have the greatest public health impact.

In the first part of the dissertation, residential exposures to ultrafine particles were characterized and governing factors explored on the basis of field data collected from single-family houses in California. During the field study, time-resolved particle number (PN) concentrations were monitored indoors and outdoors over a multi-day period, and information was acquired concerning occupancy, source-related activities, and building operation. Technological challenges have limited prior efforts to acquire time-resolved data on UFP from homes under normal occupied conditions, data that are potentially important for understanding total daily exposures to ultrafine particles as people spend a majority of their time in their own homes.

Results showed levels of ultrafine particles in houses to be highest when residents were present and awake, mainly due to their cooking and other activities that constituted episodic indoor sources. On average, the contribution to residential exposures from indoor episodic sources was 150 percent of the contribution from particles of outdoor origin. A previously unstudied continuous indoor source, unvented pilot lights, caused baseline particle levels to be significantly elevated in houses where present. Particle control devices — a filter or an electrostatic precipitator — were successful at mitigating exposure by reducing the persistence of particles indoors. We found that, owing to the importance of indoor sources, variations in the infiltration factor, and the influence of human behavior patterns on indoor UFP levels, residential exposures to ultrafine particles could not be characterized either by ambient levels or by average indoor levels alone.

The source characterization and exposure apportionment results from the study of ultrafine particles in residences were applied to quantify inhalation intake fractions (iF) for ultrafine particles emitted from indoor sources. Intake fraction is an exposure metric that quantifies the mass of pollution inhaled by all exposed persons per mass of pollution released. As such, iF estimates encapsulate the exposure effectiveness of a source under the exposure conditions considered. The analysis presented is one of only a few iF investigations focused on UFP and is also the first semi-empirical iF investigation for indoor sources to rely on experimental data resolved at the level of individual occupants and source-events.

For the continuous source (unvented pilot lights) and the episodic source events observed during the monitoring period at all study sites, estimated intake fractions ranged from 0.7 × 10^-3 to 16 × 10^-3, consistent with previous estimates for contaminants released indoors. House-specific factors such as the volume and number of residents, and occupant-specific factors such as breathing rates and time-activity patterns, had a significant influence on iF. Particle loss rates and occupancy patterns did not vary markedly among source types. Consequently, source type did not have a significant, independent influence on intake fractions.

In the second part of the dissertation, ozone levels in airplane cabins and factors that influence them were studied on commercial passenger flights. Ozone levels in passenger aircraft had not been the subject of a full-scale time-resolved monitoring effort since 1980, when U.S. Federal Aviation Regulations limiting ozone in cabin air were adopted. Studies conducted prior to 1980 were in need of an update because, in the past three decades, the operating conditions of commercial aircraft have changed significantly. Moreover our understanding of ozone's reactions with cabin surfaces, including human surfaces, and of the health risks associated with exposure to ozone and ozone oxidation byproducts has grown. Findings on in-cabin ozone need to be interpreted in light of the new findings.

To close this knowledge gap real-time ozone data were collected within the cabins of commercial passenger aircraft on 76 flight segments. Sample mean ozone level, peak-hour ozone level, and flight-integrated ozone exposures were highly variable across U.S. domestic segments, with ranges of <1.5 to 146 ppb, 3 to 275 ppb, and <1.5 to 488 ppb-hour, respectively. On planes equipped with ozone catalysts, the mean peak-hour ozone level was substantially lower than on planes not equipped with catalysts. For aircraft with catalysts, levels were higher on transoceanic flights than on domestic routes. In addition, within the transoceanic sample, ozone levels were lower on newer aircraft, a pattern that may be explained by differences in converter efficiency. Seasonal variation on domestic routes without converters was modeled by a sinusoidal curve, predicting peak-hour levels approximately 70 ppb higher in Feb-March than in Aug-Sept. The temporal trend was broadly consistent with expectations, given the annual variation in tropopause height. Episodically elevated (>100 ppb) ozone levels on domestic flights were associated with winter-spring storms that were linked to enhanced exchange between the lower stratosphere and the upper troposphere.

As in-cabin ozone originates outside, findings from the field study were supplemented with an analysis of atmospheric ozone levels collected through the Measurement of Ozone and Water Vapor by Airbus In-service Aircraft (MOZAIC) monitoring campaign. Temporal and spatial trends in ozone levels encountered by aircraft were investigated by analyzing the data from all MOZAIC flights between Munich and three U.S. destinations. In a finding that reinforced the in-cabin study results, the MOZAIC analysis showed ozone levels cycled through the year, and positive outliers to the mean cycle were spatially and temporally consistent with known patterns of stratosphere-to-troposphere exchange. A spatial analysis showed that, for the routes surveyed, there was no monotonic increase in atmospheric ozone with latitude. On average, ozone levels increased with altitude, though the relationship between altitude and ozone was highly variable within and between flights. The spatial analysis also showed that even in domestic US airspace ambient ozone concentrations greater than 100 ppb were routinely encountered. This result illustrated the potential benefit of equipping all U.S. passenger aircraft — not just the ones designed for transoceanic travel, as is standard practice — with ozone catalysts.