Machine Learning-Based Integration of High-Resolution Wildfire Smoke Simulations and Observations for Regional Health Impact Assessment

• Type of Material: Article
• Authors: Yufei Zou, University of Washington; Susan M. O'Neill, US Forest Service; Narasimhan K. Larkin, US Forest Service; Ernesto C. Alvarado, University of Washington; Robert Solomon, University of Washington; Clifford Mass, University of Washington; Yang Liu, Emory University; M. Talat Odman, Georgia Institute of Technology; Huizhong Shen, Georgia Institute of Technology
• Language: English
• Date: 2019-06-02
• Publisher: MDPI
• Publication Version: Final Published Version
• Copyright Statement: © 2019 by the authors. Licensee MDPI, Basel, Switzerland.
• License: Creative Commons Attribution 4.0 International
• Final Published Version (URL): http://dx.doi.org/10.3390/ijerph16122137
• Title of Journal or Parent Work: International Journal of Environmental Research and Public Health
• ISSN: 1661-7827
• Volume: 16
• Issue: 12
• Grant/Funding Information: M.T.O. was supported by the NASA Applied Sciences Program (NASA grant number NNX16AQ29G) while H.S. was supported by the USA Environmental Protection Agency (EPA grant number R835880).; This research was funded by the NASA Health and Air Quality Applied Sciences Team (HAQAST) project (NASA grant number NNH16AD18I) under the agreement FS 17-JV-11261987-044 between the University of Washington and the USA Forest Service Pacific Northwest Research Station.; US Environmental Protection Agency is the primary funding source, with contracting and research support from the National Park Service.
• Keywords: ORGANIC AEROSOL; TRANSPORT; PARTICULATE MATTER; AIR-QUALITY; Environmental Sciences & Ecology; fire smoke modeling; Life Sciences & Biomedicine; SYSTEM; machine learning-based data fusion; OXIDATION; PM2.5 air pollution; EMISSIONS; Science & Technology; EXPOSURE; health impact assessment; WILDLAND FIRE SMOKE; VOLATILITY; Public, Environmental & Occupational Health; Environmental Sciences
• Research Categories: Health Sciences, Public Health

Abstract

Large wildfires are an increasing threat to the western U.S. In the 2017 fire season, extensive wildfires occurred across the Pacific Northwest (PNW). To evaluate public health impacts of wildfire smoke, we integrated numerical simulations and observations for regional fire events during August-September of 2017. A one-way coupled Weather Research and Forecasting and Community Multiscale Air Quality modeling system was used to simulate fire smoke transport and dispersion. To reduce modeling bias in fine particulate matter (PM2.5 ) and to optimize smoke exposure estimates, we integrated modeling results with the high-resolution Multi-Angle Implementation of Atmospheric Correction satellite aerosol optical depth and the U.S. Environmental Protection Agency AirNow ground-level monitoring PM2.5 concentrations. Three machine learning-based data fusion algorithms were applied: An ordinary multi-linear regression method, a generalized boosting method, and a random forest (RF) method. 10-Fold cross-validation found improved surface PM2.5 estimation after data integration and bias correction, especially with the RF method. Lastly, to assess transient health effects of fire smoke, we applied the optimized high-resolution PM2.5 exposure estimate in a short-term exposure-response function. Total estimated regional mortality attributable to PM2.5 exposure during the smoke episode was 183 (95% confidence interval: 0, 432), with 85% of the PM2.5 pollution and 95% of the consequent multiple-cause mortality contributed by fire emissions. This application demonstrates both the profound health impacts of fire smoke over the PNW and the need for a high-performance fire smoke forecasting and reanalysis system to reduce public health risks of smoke hazards in fire-prone regions.