AAAR 32nd Annual Conference
September 30 - October 4, 2013
Oregon Convention Center
Portland, Oregon, USA
Abstract View
Spatiotemporal Modeling of Indoor Aerosol Mass Concentration
KIRSTEN KOEHLER, John Volckens, Kirk Lake, Colorado State University
Abstract Number: 476 Working Group: Portable and Inexpensive Sensor Technology for Air Quality Monitoring
Abstract Technological advancements in exposure assessment have recently increased the accuracy, reliability, and affordability of portable, direct-reading monitors. These monitors can rapidly assess exposures to occupational or other indoor hazards. By coupling an estimated exposure with a known location, contour plots of the hazard concentration over space, known as concentration maps, can be created. These maps are used to assess the spatial variability of hazards and show great potential to help identify and mitigate sources of exposure. Using direct-reading instruments, we generated a measurement dataset of aerosol mass concentration at high levels of spatial and temporal resolution at an engine testing facility. Instruments were deployed as static monitors (n=10), located both near and at a distance from known sources, and as a roving monitor (n=1) that traversed the facility along predetermined routes at 1 m spatial resolution. Data from this high-resolution measurement campaign were used to generate a reference concentration map. The reference map accounted for both spatial and temporal variation using spatiotemporal Kriging, which takes into account the correlation of measurements in both time and space. The reference concentration map provided a best estimation of the true spatiotemporal distribution of aerosol mass concentration in the facility and was used to evaluate simpler statistical approaches (i.e., Traditional and Bayesian Kriging) that do not account for temporal variation. We assessed the reliability of the roving monitor data, a method gaining popularity in occupational exposure assessment, to generate a time weighted average concentration map. We also considered the benefit of including static monitor data to improve time weighted average concentration map accuracy and representativeness. This work has resulted in recommendations for the occupational health community to generate concentration maps using direct-reading instrumentation.