Search

Bayesian additive regression tree (BART) is a recent statistical method that combines ensemble learning and nonparametric regression. BART is constructed under a probabilistic framework that also allows for model-based prediction uncertainty quantification. We evaluated the application of BART in predicting daily concentrations of four fine particulate matter (PM2.5) components (elemental carbon, organic carbon, nitrate, and sulfate) in California during the period 2005 to 2014. We demonstrate in this paper how BART can be tuned to optimize prediction performance and how to evaluate variable importance. Our BART models included, as predictors, a large suite of land-use variables, meteorological conditions, satellite-derived aerosol optical depth parameters, and simulations from a chemical transport model. In cross-validation experiments, BART demonstrated good out-of-sample prediction performance at monitoring locations (R2 from 0.62 to 0.73). More importantly, prediction intervals associated with concentration estimates from BART showed good coverage probability at locations with and without monitoring data. In our case study, major PM2.5 components could be estimated with good accuracy, especially when collocated PM2.5 total mass observations were available. In conclusion, BART is an attractive approach for modeling ambient air pollution levels, especially for its ability to provide uncertainty in estimates that may be useful for subsequent health impact and health effect analyses. [ABSTRACT FROM AUTHOR]

Recent regulatory and legislative developments in the state of California (e.g., Assembly Bill (AB) 1647 and AB 617) will increase the extent of air quality monitoring in communities near petroleum refineries and at facility fencelines. This work reports results over a three-year period for currently-installed community and fenceline air quality monitoring for the Chevron refinery in Richmond, California, USA. This paper presents the most comprehensive air quality dataset that has been published to date for a community near a petroleum refinery, including concentration of different air toxics (e.g., benzene, toluene, xylenes, etc.) and criteria air pollutants (e.g., ozone (O3), sulfur dioxide (SO2), and particulate matter (PM)). Instrumental techniques such as ultra-violet differential absorption spectroscopy (UV-DOAS), laser-based spectroscopy and real-time gas chromatography (GC) were used in the community and fenceline monitors. From 2015 to 2017, measured concentrations at community monitors near the Richmond refinery were generally below California thresholds for acute and chronic health (only two exceedances were observed for 8-h average benzene concentrations during the three-year monitoring period). Although more detailed speciation for volatile organic compounds (VOCs) and fine particulate matter (PM2.5) would be needed to confirm certain source profile identities, preliminary application of source apportionment methods indicates the prevalence of typical urban emission profiles, such as from traffic, in the measured community data. [ABSTRACT FROM AUTHOR]