细颗粒物与关键组分的个体暴露评估模型构建与暴露特征研究

In China，PM2.5 has become a primary challenge of public health and the chemical constituents of PM2.5 play significant roles on adverse health effects. Therefore, it has been a priority/necessary to assess the health risk of PM2.5 and its key components. However, interpretation of findings from epidemiological studies has been hampered by uncertainties in exposure measurements. Traditional ambient air pollution measurements, which basically relied on fixed-site monitoring, had limitations on reflecting the spatial variations of target pollutants and the impacts of personal activities. Thus, exposure misclassification may exist as a consequence. This study aims to establish a hybrid model to characterize personal exposure to PM2.5 and its key constituents in three aspects: 1) predict spatial surfaces and temporal variation of PM2.5 with fine spatial and temporal resolutions at a city-wide scale in an outdoor exposure assessment model incorporating Land-Use Regression model and satellite data. ; 2) assess home indoor PM2.5 concentrations by estimating the residential infiltration of outdoor PM2.5 via modeling, and explore the characteristics of home indoor PM2.5 exposure, and 3) establish hybrid model by incorporating outdoor and indoor exposure models through personal activity variables. Based on constructing the hybrid model, this study is expected to assess personal exposure to PM2.5 and its constituents from outdoor to indoor microenvironments and to explore the major sources and potential factors which may affect the exposure characteristic of PM2.5. The results will improve the accuracy in air pollution exposure assessment when estimating the associations of air pollution exposure and health risks. Moreover, this study has implications on design strategies for pollution control on PM2.5 and its key constituents and for their public health prevention.

PM2.5污染的健康危害已成为我国面临的主要公共卫生挑战。其危害程度与负载的化学组份密切相关。开展PM2.5及关键组分的健康研究已成为应对这一挑战的迫切需求。传统暴露评价方法忽视了PM2.5空间异质性与人群活动模式差异，易造成暴露错分；模型方法则提供了有效的手段。本研究拟从三个层次开展PM2.5与关键组分的暴露评估方法构建和暴露特征研究: 1)基于LUR模型和卫星数据，建立具有高时空分辨率的室外暴露评价模型，模拟城市内PM2.5与关键组分的时空动态变化规律；2)通过渗透系数数值模拟方法模拟家庭室内环境目标污染物浓度，探讨其室内环境的暴露特征; 3)在上述模型基础上，纳入人群活动模式特征变量，构建从室外到室内的综合暴露评估模型，探索影响暴露特征的主要来源和相关因素。本研究的研究结果将有助于填补我国在领域的空白，提高暴露-健康效应关系结果的准确性，对制定防控措施和公共卫生预防策略具有重意义。

PM2.5污染的健康危害已成为我国面临的主要公共卫生挑战。其危害程度与负载的化学组成有密切关系。开展PM2.5及关键组分的健康研究已成为应对这一挑战的迫切需求。传统暴露评价方法，忽视了PM2.5空间异质性与人群活动模式差异，造成暴露错分，致使健康研究的开展受到了制约。.本研究从三个层次开展PM2.5与关键组分的暴露评估方法构建和暴露特征研究: 1)基于LUR模型和卫星数据，建立具有高时空分辨率的室外暴露评价模型，模拟城市内空气污染的时空动态变化规律，模型解释了大气污染物空间变异度的65%~80%；2) 选取人群活动时间最长的家庭住宅作为典型场所，开展室内外颗粒物同步观测，通过数值模拟的方法构建了住宅PM2.5渗透系数模型，模型可解释居民住宅渗透系数60%以上的变异度，并识别出影响居民室内环境暴露特征的关键因素主要包括空调使用总时长等活动模式变量，以及楼层、建筑年龄等建筑特征变量; 3)在个体采样的基础上，纳入人群活动模式特征变量，最终构建从室外到室内的综合暴露评估模型。最终模型可解释个体PM2.5和BC暴露变异度的75%，模型交叉验证 R2均达0.66以上，显示了整体预测模型的稳健性。最终完成从室外-室内-个体三水平暴露评价模式的构建。基于上述模型研究，并利用颗粒物主要成分谱的分析和受体模型源解析，本研究还发现了城市内个体细颗粒物暴露主要来源于与人为活动排放相关的源，贡献最大的是工业和机动车排放。.总的来说，本研究建立了更准确有效的、适用于大气流行病学研究的个体暴露评价方法，提高了暴露评价的准确性；本研究的研究结果最主要的成果转化是将来为队列研究提供更可靠的暴露评价，可进一步提高暴露健康研究分析的因果推断效力。本研究的结果同时还初步阐明了细颗粒物的暴露特征和主要来源，为空气污染相关的疾病负担评估和风险管理提供证据支持，为政府部门优先控制污染源提供决策建议。