논문 상세보기
pp.151-161
The assessment of personal exposure is a critical component in population-based epidemiologic studies of air pollution. This study was conducted to apply and compare the four exposure estimation methods of individual-level to air pollution concentration in a cohort including 2,283 subjects in Gwangyang, Korea. Individual-level exposure of air pollution were estimated using multiple approaches, including average across all monitors, nearest monitor, and spatial interpolation by inverse distance weighting and kriging. The mean concentrations of PM10, NO2, SO2, CO, O3 by four exposure estimation methods were slightly different but not significantly different from each other. Cross-validation showed that kriging was more accurate than other exposure estimation methods because kriging has probably predicted individual exposure levels equivalent to residential locations after estimating the parameters of a model according to the spatial surface of air pollution concentration. These data support that spatial interpolation methods may provide better estimates than selecting the value from the nearest monitor and averaging across values from all monitors by reflecting spatial attributes of air pollution on personal level.
1. 서 론
2. 자료 및 방법
2.1. 연구대상
2.2. 대기오염 측정망 자료
2.3. 개인노출추정 방법
2.4. 교차검증(cross-validation)
2.5. Passive sampler 측정치와 추정치와의 비교
2.6. 통계분석
3. 결 과
3.1. 연구지역의 대기오염물질 특성
3.2. 개인노출 추정치
3.3. 오염물질 추정방법별 상관성 분석
3.4. 교차 검증(Cross-validation)
3.5. 개인노출 예측치와 passive sampler 실측치의 비교
4. 고 찰
5. 결 론
참 고 문 헌
