Smoke Plume Estimation from Satellite Images

• Activity Number: 1
• Type: Invited
• Date/Time: Sunday, August 1, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Charmaine Dean and Victoria Wan*+ and John Braun and Sarah Henderson
• Companies: Simon Fraser University and Simon Fraser University and The University of Western Ontario and The University of British Columbia
• Keywords: mixture model ; smoke plume modeling ; smoke health effects ; data sharpening ; satellite imagery ; smoke exposure

Abstract

Identifying accurate measures of exposure to forest fire smoke is important for assessment of health risk. Estimating exposure from air quality monitors is challenging because of the sparseness of the monitoring networks in remote areas. However, satellite imagery can provide estimates of fire location and visual information on smoke plumes. This talk will discuss statistical techniques for obtaining estimates of forest fire smoke plumes. We use classification and smoothing mechanisms on data from satellite imagery in order to develop automated processes for identifying exposure. The aim is to identify a high-resolution approach that provides a reliable estimate of smoke and which provides a better idea of the spatial distribution of smoke from fires than is currently available.