Modeling indoor particulate exposures in inner-city school classrooms.
Modeling indoor particulate exposures in inner-city school classrooms.
复制标题
DOI：
10.1038/jes.2016.52
复制
发表时间：
2017-09
影响因子：
4.5
通讯作者：
Phipatanakul W
中科院分区：
文献类型：
--
作者：
Gaffin JM;Petty CR;Hauptman M;Kang CM;Wolfson JM;Abu Awad Y;Di Q;Lai PS;Sheehan WJ;Baxi S;Coull BA;Schwartz JD;Gold DR;Koutrakis P;Phipatanakul W
室外空气污染渗透到建筑物中，并导致室内总暴露。我们调查了市中心学校教室室内和室外颗粒物的关系。内城学校哮喘研究调查了美国东北部以教室为基础的环境暴露对哮喘学生的影响。采用混合效应线性模型来确定室内PM2.5和BC与其对应的室外浓度之间的关系，并建立了预测这些污染物暴露的模型。采用室内外硫磺比作为室外细颗粒物的渗透因子。来自136个教室(30栋校舍)的199个样本的一周PM2.5和BC浓度与在同一时间段内在中央监测点测得的平均浓度进行了比较。混合效应回归模型发现了显著的随机截获和坡度效应，这表明：1)教室中存在重要的PM2.5源；2)室外PM2.5颗粒物的穿透因学校而异；3)特定地点的PM2.5水平(由模型推断)与中央监测点的观测结果不同。除了缺乏室内来源外，BC也发现了类似的结果。硫磺调整模型的拟合预测对观察到的室内污染物水平是中等预测的(样本相关性：PM2.5：R2=0.68，BC；R2=0.61)。我们的结果表明，PM2.5有重要的课堂来源，这一点因学校而异。此外，使用这些混合效应模型，可以准确地预测中央站点测量可用但室内测量不可用的日期的教室暴露。
Outdoor air pollution penetrates buildings and contributes to total indoor exposures. We investigated the relationship of indoor to outdoor particulate matter in inner-city school classrooms. The School Inner City Asthma Study investigates the effect of classroom-based environmental exposures on students with asthma in the northeast United States. Mixed-effects linear models were used to determine the relationships between indoor PM2.5 and BC and their corresponding outdoor concentrations, and to develop a model for predicting exposures to these pollutants. The indoor-outdoor sulfur ratio was used as an infiltration factor of outdoor fine particles. Weeklong concentrations of PM2.5 and BC in 199 samples from 136 classrooms (30 school buildings) were compared to those measured at a central monitoring site averaged over the same timeframe. Mixed effects regression models found significant random intercept and slope effects, which indicate that: 1) there are important PM2.5 sources in classrooms; 2) the penetration of outdoor PM2.5 particles varies by school, and 3) the site-specific outside PM2.5 levels (inferred by the models) differ from those observed at the central monitor site. Similar results were found for BC except for lack of indoor sources. The fitted predictions from the sulfur-adjusted models were moderately predictive of observed indoor pollutant levels (Out of sample correlations: PM2.5: r2 = 0.68, BC; r2 = 0.61). Our results suggest that PM2.5 has important classroom sources, which vary by school. Furthermore, using these mixed effects models, classroom exposures can be accurately predicted for dates when central site measures are available but indoor measures are not available.