CAREER: Leveraging mobile monitoring, low-cost sensors, and Google Street View imagery to identify and modify street-level determinants of exposure to particulate air pollution

在城市地区，通勤和其他交通时间是造成空气污染的主要原因。暴露量不仅受车辆交通量的影响，而且在很大程度上受城市地区街道和社区设计的影响。该项目的目标是改善空气质量模型预测，以解决自然地理在暴露中的作用。这将通过使用谷歌街景（GSV）图像，低成本传感和移动的测量的新组合来实现。这些模型的结果将为如何最好地改造街道和社区以减少空气污染提供新的证据。这种新方法的另一个社会效益是能够以成本效益高的方式测量城市地区的空气污染。一个智能手机应用程序将被创建作为一个外展工具，以跟踪暴露在华盛顿特区地铁区作为一个测试案例。社区成员、弗吉尼亚理工大学的学生和其他利益相关者将使用智能手机应用程序来促进合作伙伴社区的设计解决方案，以经济有效地减少空气污染并保护人类健康。土地利用回归（LUR）的开发旨在提供高空间分辨率的空气质量估计，无需测量。最近在移动的监测和低成本传感方面取得的进展使城市地区的测量具有前所未有的空间覆盖范围。这些测量的LUR估计值的比较表明，LUR模型不捕捉所有的浓度梯度在城市微环境。这些差异的一个潜在原因是街道一级LUR协变量不包括在传统的数据库中。该项目将测试新兴的移动的监测和低成本传感测量技术是否可以与GSV图像的街道级指标结合使用，以确定以前被忽视的暴露决定因素，这些因素可能已经成熟，可以进行修改。该项目有三个目标来解决这些差距：1）使用颗粒空气污染的移动的监测和低成本传感器网络来开发以前无法获得的实时LUR模型; 2）开发新的基于GSV的城市地区街道特征测量，以确定街道水平的暴露决定因素;以及3）将LUR模型集成到智能手机应用程序中，以创建用于与高中学生和社区合作伙伴协作教学活动的实时曝光工具。该项目将为公众和政策制定者提供重要的新知识，这些知识可以随时应用于其他城市，环境和污染物。基于网络和电话的曝光工具有可能改变空气质量模型的传播和公众使用方式。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Commuting and other time spent in transport in urban areas is responsible for a large share of exposure to air pollution. The amount of exposure is influenced not only by the amount of vehicle traffic, but also in large part by the design of streets and neighborhoods in urban areas. The goal of this project is to improve air quality model prediction to address the role that physical geography plays in exposure. This will be achieved by using a novel combination Google Street View (GSV) images, low-cost sensing, and mobile measurements. Results from these models will provide new evidence on how best to modify streets and neighborhoods to reduce exposure to air pollution. An added societal benefit of this new approach will be the ability to cost-effectively measure air pollution in urban areas. A smartphone app will be created as an outreach tool to track exposure in the Washington DC metro area as a test case. The smartphone app will be used by community members, Virginia Tech students, and other stakeholders to facilitate design solutions in partner communities to cost effectively reduce air pollution and protect human health.Land use regression (LUR) was developed to provide high spatial resolution estimates of air quality at locations without measurements. Recent advances in mobile monitoring and low-cost sensing have enabled unprecedented spatial coverage of measurements in urban areas. Comparisons of these measurements to LUR estimates suggest that LUR models do not capture all concentration gradients in urban micro-environments. A potential reason for these differences is that street-level LUR covariates are not included in traditional databases. This project will test if emerging mobile monitoring and low-cost sensing measurement techniques can be used in conjunction with street-level metrics from GSV images to identify previously overlooked determinants of exposure that may be ripe for modification. The project has three objectives to address these gaps: 1) use mobile monitoring and a low-cost sensor network of particulate air pollution to develop previously unavailable real-time LUR models; 2) develop new GSV-derived measures of street-level features in urban areas to identify street-level determinants of exposure; and 3) integrate the LUR models into a smartphone app to create a real-time exposure tool for collaborative teaching activities with high school students and community partners. This project will provide new knowledge of importance to the public and policy-makers that could be readily applied to other cities, settings, and pollutants. The web- and phone-based exposure tools have the potential to transform how air quality models are disseminated and used by the public.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.