Microenvironments associated with personal exposure to nitrogen dioxide by time-weighted microenvironmental model

Abstract

Although personal exposure of nitrogen dioxide (NO2) can be estimated by time weighted microenvironmental model, the model does not fully explain personal exposure. There are many factors affecting personal exposure and determination of these factors is critical for reduction of the exposure. The purposes of this study were to estimate the personal exposure of NO2 using microenvironmental model, to develop models to predict personal exposure to NO2, and to identify microenvironments associated with personal exposure to NO2. Personal exposures, residential indoor, workplace indoor and outdoor levels of NO2 were measured in four cities in Korea (Seoul, Daegu, Asan and Suncheon) using a passive sampler for five weekdays. The measurements were conducted in summer and winter during 2008–2009. The average personal NO2 exposures were 20.5 (1.6) ppb in summer and 18.6 (1.8) ppb in winter. The average personal exposures of NO2 and residential indoor levels were the highest in Seoul and the lowest in Suncheon in summer and winter. Personal exposure was significantly different by season. Personal exposure was significantly correlated with residential indoor concentration in all cities. The personal exposure estimated by residential indoor, workplace indoor and outdoor levels explained 45% and 49% of the personal exposure variance in summer and winter, respectively. Relationship between the estimated personal exposure and measured personal exposure was significantly associated with the season and city. The low correlation between the estimated and measured personal exposures of Korean may be due to the longer time spent in other microenvironments in the Korean population. Personal exposure was significantly correlated with the size of the house, number of room in house and working time in summer. Times spent in library, classroom and shopping center for workers, in transportation and restaurant for housewives, and in transportation, library and classroom for students were associated with the difference between measured and estimated personal exposure of NO2. The model can provide reasonable estimation of population exposure with appropriate microenvironmental concentrations and time activity data. However, season- and city-specific models should be considered.