Formatted Title
Sampling Strategies and Models to Understand Urban Neighborhood Vapor Intrusion Exposure Pathways and Risk
Background/Objectives
Brownfields, which contain subsurface hazardous chemicals, represent a heterogeneous yet ubiquitous exposure for many Americans. One common type of contaminant found in brownfields, volatile organic compounds (VOCs), impacts health within buildings through the vapor intrusion exposure route. To address this issue, the Center for Leadership in Environmental Awareness and Research (CLEAR), a new National Institute of Environmental Health Sciences Superfund Research Program center, is assessing screening and risk modeling methods for understanding vapor intrusion exposure pathways in urban neighborhoods – with an emphasis on cities with aging infrastructure like Detroit and vapor intrusion impacts on preterm birth rates. Specifically, we aim to create resilient buildings and healthy communities using combined field and modeling approaches to the VOC vapor exposure issue.
Approach/Activities
Our team is developing geospatial models and IoT sensors coupled with field observations, such as phytoscreening and soil gas flux, to reduce VOC exposure. Phytoscreening, the chemical analysis of plant tissue to provide evidence for belowground contamination, offers a cost-effective, minimally invasive approach in contaminant transport applications. Field screening methods coupled with VOC sensors enable neighborhood-to-city scale assessment of vapor intrusion risk. In addition, many brownfields have been subject to detailed examination, but utilization of this data is difficult due to data structure and historic reporting limits. To utilize these data, we are developing natural language processes (NLP) to extract data of spatially distributed VOC types, units, and concentrations to ensure the database is human and machine-readable.
Results/Lessons Learned
Current fieldwork is evaluating the use of plants to assess background VOC levels in urban environments. Targeted VOCs were found in plant material, but ongoing fieldwork is evaluating the differences in plant type and plant tissue type for VOC screening. The NLP datasets are taking systematic approaches and focusing on recent reports, heavily contaminated locations, and locations with high population density. The accuracy of NLP will be validated from random samples. By leveraging this NPL technology, both field and modeling efforts are informed by a database of spatially distributed VOC concentrations. This city-scale approach represents a new direction in the vapor intrusion exposure pathway.