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

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.

[1]  R. Yokelson,et al.  In situ measurements of trace gases, PM, and aerosol optical properties during the 2017 NW US wildfire smoke event , 2018, Atmospheric Chemistry and Physics.

[2]  A. Russell,et al.  Review of Acellular Assays of Ambient Particulate Matter Oxidative Potential: Methods and Relationships with Composition, Sources, and Health Effects. , 2019, Environmental science & technology.

[3]  Michelle C. Kondo,et al.  Meta-Analysis of Heterogeneity in the Effects of Wildfire Smoke Exposure on Respiratory Health in North America , 2019, International journal of environmental research and public health.

[4]  Alexei Lyapustin,et al.  Impacts of snow and cloud covers on satellite-derived PM2.5 levels. , 2019, Remote sensing of environment.

[5]  H. Tian,et al.  Development of a REgion‐Specific Ecosystem Feedback Fire (RESFire) Model in the Community Earth System Model , 2019, Journal of Advances in Modeling Earth Systems.

[6]  D. Jaffe,et al.  US particulate matter air quality improves except in wildfire-prone areas , 2018, Proceedings of the National Academy of Sciences.

[7]  W. Cascio Wildland fire smoke and human health. , 2018, The Science of the total environment.

[8]  Long Huan,et al.  Mechanisms for ozone-initiated removal of biomass burning products from the atmosphere , 2018 .

[9]  J. Balmes,et al.  A Review of Community Smoke Exposure from Wildfire Compared to Prescribed Fire in the United States , 2018 .

[10]  Qingyang Xiao,et al.  Full-Coverage High-Resolution Daily PM2.5 Estimation using MAIAC AOD in the Yangtze River Delta of China , 2018 .

[11]  D. DeMarini,et al.  Mutagenicity and Lung Toxicity of Smoldering vs. Flaming Emissions from Various Biomass Fuels: Implications for Health Effects from Wildland Fires , 2018, Environmental health perspectives.

[12]  N. Fann,et al.  The health impacts and economic value of wildland fire episodes in the U.S.: 2008-2012. , 2018, The Science of the total environment.

[13]  A. Laskin,et al.  Characterization of Light-Absorbing Oligomers from Reactions of Phenolic Compounds and Fe(III) , 2017 .

[14]  Y. Tesfaigzi,et al.  Wildfire smoke exposure and human health: Significant gaps in research for a growing public health issue. , 2017, Environmental toxicology and pharmacology.

[15]  Qingyang Xiao,et al.  Full-coverage high-resolution daily PM 2.5 estimation using MAIAC AOD in the Yangtze River Delta of China , 2017 .

[16]  J. Randerson,et al.  Global fire emissions estimates during 1997–2016 , 2017 .

[17]  J. C. D. Silva,et al.  Density Functional Theory Calculation of the Absorption Properties of Brown Carbon Chromophores Generated by Catechol Heterogeneous Ozonolysis , 2017 .

[18]  W. Cascio,et al.  Community Vulnerability to Health Impacts of Wildland Fire Smoke Exposure. , 2017, Environmental science & technology.

[19]  M. Guzman,et al.  Oxidation of Substituted Catechols at the Air-Water Interface: Production of Carboxylic Acids, Quinones, and Polyphenols. , 2017, Environmental science & technology.

[20]  Michael D. Moran,et al.  Blending forest fire smoke forecasts with observed data can improve their utility for public health applications , 2016 .

[21]  Michael Brauer,et al.  Critical Review of Health Impacts of Wildfire Smoke Exposure , 2016, Environmental health perspectives.

[22]  Jeremy D. Smith,et al.  Phenolic carbonyls undergo rapid aqueous photodegradation to form low-volatility, light-absorbing products , 2016 .

[23]  A. Laskin,et al.  Molecular transformations of phenolic SOA during photochemical aging in the aqueous phase: competition among oligomerization, functionalization, and fragmentation , 2015 .

[24]  J. Jimenez,et al.  Understanding sources of organic aerosol during CalNex-2010 using the CMAQ-VBS , 2015 .

[25]  A. Tonkin,et al.  Impact of Fine Particulate Matter (PM2.5) Exposure During Wildfires on Cardiovascular Health Outcomes , 2015, Journal of the American Heart Association.

[26]  S. Freitas,et al.  A review of approaches to estimate wildfire plume injection height within large-scale atmospheric chemical transport models , 2015 .

[27]  G. Pfister,et al.  Spatiotemporal prediction of fine particulate matter during the 2008 northern California wildfires using machine learning. , 2015, Environmental science & technology.

[28]  M. Bell,et al.  A systematic review of the physical health impacts from non-occupational exposure to wildfire smoke. , 2015, Environmental research.

[29]  M. Guzman,et al.  Catechol oxidation by ozone and hydroxyl radicals at the air-water interface. , 2014, Environmental science & technology.

[30]  I. Annesi-Maesano,et al.  Quantifying wildfires exposure for investigating health-related effects , 2014 .

[31]  G. Yarwood,et al.  1.5-Dimensional volatility basis set approach for modeling organic aerosol in CAMx and CMAQ , 2014 .

[32]  N. Fann,et al.  Forecast-based interventions can reduce the health and economic burden of wildfires. , 2014, Environmental science & technology.

[33]  H. Timonen,et al.  Influence of background particulate matter (PM) on urban air quality in the Pacific Northwest. , 2013, Journal of environmental management.

[34]  Gary L. Achtemeier,et al.  Modelling smoke transport from wildland fires: a review , 2013 .

[35]  John Michalakes,et al.  WRF-Fire: Coupled Weather–Wildland Fire Modeling with the Weather Research and Forecasting Model , 2013 .

[36]  D. Jaffe,et al.  Ozone production from wildfires: A critical review , 2012 .

[37]  Michael Brauer,et al.  Estimated Global Mortality Attributable to Smoke from Landscape Fires , 2012, Environmental health perspectives.

[38]  B. Hubbell,et al.  Estimating the National Public Health Burden Associated with Exposure to Ambient PM2.5 and Ozone , 2012, Risk analysis : an official publication of the Society for Risk Analysis.

[39]  R. Mathur,et al.  WRF-CMAQ two-way coupled system with aerosol feedback: software development and preliminary results , 2011 .

[40]  Daniel Krewski,et al.  Lung Cancer and Cardiovascular Disease Mortality Associated with Ambient Air Pollution and Cigarette Smoke: Shape of the Exposure–Response Relationships , 2011, Environmental health perspectives.

[41]  M. Brauer,et al.  Three Measures of Forest Fire Smoke Exposure and Their Associations with Respiratory and Cardiovascular Health Outcomes in a Population-Based Cohort , 2011, Environmental health perspectives.

[42]  PM2.5 measurements in wildfire smoke plumes from fire seasons 2005–2008 in the Northwestern United States , 2011 .

[43]  J. Mandel,et al.  Coupled atmosphere-wildland fire modeling with WRF 3.3 and SFIRE 2011 , 2011, 1102.1343.

[44]  F. L. Herron-Thorpe,et al.  Evaluation of a regional air quality forecast model for tropospheric NO 2 columns using the OMI/Aura satellite tropospheric NO 2 product , 2009 .

[45]  D. Sullivan,et al.  The BlueSky smoke modeling framework , 2008 .

[46]  Susan M. O’Neill,et al.  Enhancement and evaluation of the AIRPACT ozone and PM2.5forecast system for the Pacific Northwest , 2008 .

[47]  G. Powers,et al.  A Description of the Advanced Research WRF Version 3 , 2008 .

[48]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[49]  Joseph Hilbe,et al.  A Handbook of Statistical Analyses Using R , 2006 .

[50]  Yongqiang Liu Variability of wildland fire emissions across the contiguous United States , 2004 .

[51]  P. Stolpman,et al.  Environmental Protection Agency , 2020, The Grants Register 2022.

[52]  G. Briggs,et al.  A Plume Rise Model Compared with Observations , 1965 .