AAAR 35th Annual Conference October 17 - October 21, 2016 Oregon Convention Center Portland, Oregon, USA
Abstract View
Does it Blend? Combining Observations and Model Output to Determine Health Implications of Wildfire Smoke Exposure
WILLIAM LASSMAN, Bonne Ford, Ryan Gan, Gabriele Pfister, Sheryl Magzamen, John Volckens, Emily Fischer, Jeffrey R. Pierce, Colorado State University
Abstract Number: 519 Working Group: Aerosol Exposure
Abstract Exposure to high concentrations of particulate matter (PM) present during acute pollution events is associated with adverse health effects. While many anthropogenic pollution sources are regulated in the United States, emissions from wildfires are difficult to characterize and control. With wildfire frequency and intensity in the western U.S. projected to increase, it is important to determine more-precisely the effect that wildfire emissions have on human health.
Three tools are frequently used to determine population-level exposure to ambient air pollution: (1) in-situ monitors that provide exact surface concentrations but with limited spatial resolution, (2) satellite observations that provide broad spatial coverage limited surface-level detail, and (3) chemical transport models (CTMs) that provide spatially continuous surface information but with limitations in model accuracy. Wildfires present challenges to all three of these approaches: wildfires are rapidly evolving and highly heterogeneous, making accurate modelling of air quality challenging; the altitude of a biomass-burning plume is dependent on both the fire properties and the meteorological conditions, making it challenging to determine surface concentrations from satellite observations; and the steep concentration gradients of wildfire smoke may be too narrow for even a dense surface site network to resolve.
In this presentation, we will show that combinations of the three approaches can capitalize on their individual strengths while mitigating their weaknesses. We use the Weather Research and Forecasting model with Chemistry (WRF-Chem), MODIS and GOES satellite observations, and in-situ PM monitors. We blend these observations together to produce surface-concentration estimates that are more accurate than the individual approaches alone. Finally, we present an application of our exposure model in a health effects analysis to estimate the association of smoke exposure on cardiopulmonary hospitalizations during the 2012 Washington State wildfire season.