Indoor/outdoor PM10 and PM2.5 in Bangkok, Thailand

Twenty-four-hour averaged PM10 and PM2.5 concentrations were obtained by us ing 4-liter-per-minute-pumps and impactors in microenvironments of a busy s hopping district and a university hospital campus. In both areas, most peop le live directly adjacent to their worksites - minimizing the need to measu re commuting exposure as part of total daily exposure. Co-located samplers were set in indoor microenvironments, the near-ambient zone of the househol ds, and at nearby streetside central ambient monitoring stations. Smoking a nd use of other indoor PM sources were recorded daily via questionnaires. C onsistent with previous studies, smoking and the use of charcoal stoves inc reased indoor particulate matter levels. The sampled air-conditioned hospit al area had substantially lower particle concentrations than outdoors. A si mple total exposure model was used to estimate the human exposure. The aver aged ratios of co-located PM2.5/PM10 concentrations in various microenviron ments are reported for each location. A single daily indoor average PM10 co ncentration for all households measured in a given sampling day is calculat ed for correlation analysis. Results showed that day-to-day fluctuations of these calculated indoor PM10 levels correlated well with near-ambient data and moderately well with ambient data collected at the nearby central moni toring site. This implies that ambient monitors are able to capture the dai ly variations of indoor PM levels or even personal exposure and may help ex plain the robust association of ambient PM levels and health effects found in many epidemiological studies. Absolute PM exposures, however, were subst antially underestimated by ambient monitors in the shopping district, proba bly because of strong local sources.