Murray et al.: new method on PM2.5 Bayesian Ensemble Models

We develop a method to combine PM2.5 estimated from satellite-retrieved aerosol optical depth (AOD) and chemical transport model (CTM) simulations using statistical models. While most previous methods utilize AOD or CTM separately, we aim to leverage advantages offered by both data sources in terms of resolution and coverage using Bayesian ensemble averaging. Our approach differs from previous ensemble approaches in its ability to not only incorporate uncertainties in PM2.5 estimates from individual models but also to provide uncertainties for the resulting ensemble estimates. In an application of estimating daily PM2.5 in the Southeastern US, the ensemble approach outperforms previously developed spatial-temporal statistical models that use either AOD or bias-corrected CTM simulations in cross-validation (CV) analyses.

Publication Link


Vu et al.: Developing an Advanced PM2.5 Exposure Model in Lima, Peru

We developed a machine learning model to estimate daily PM2.5 levels in Lima Peru. This is the first advanced model to incorporate satellite remote sensing data as well as variables from chemical transport and forecast models to predict daily PM2.5 levels in South America. Our results indicates that concentrations are low in the coast and rises with elevation up to the Andes Mountains due to prevailing coastal winds. This model also provides historical daily levels for epidemiological studies in a rapidly developing urban center.



Vu, B.N.; Sánchez, O.; Bi, J.; Xiao, Q.; Hansel, N.N.; Checkley, W.; Gonzales, G.F.; Steenland, K.; Liu, Y. Developing an Advanced PM2.5 Exposure Model in Lima, Peru. Remote Sens. 201911, 641.


Bi et al. Impacts of snow and cloud covers on satellite-derived PM2⁠.5 levels

We developed an AOD gap-filling model with the consideration of snow and cloud covers and a PM2.5 prediction model based on the gap-filled AOD to estimate full-coverage and high-resolution PM2⁠.5 in New York State. It indicated the significance of snow fraction in the AOD gap-filling process and demonstrated the ability of the AOD-based prediction model to reflect detailed PM2.5 emission patterns and small-scale terrain-driven features.

Bi, J., Belle, J. H., Wang, Y., Lyapustin, A. I., Wildani, A., & Liu, Y. (2019). Impacts of snow and cloud covers on satellite-derived PM2.5 levels. Remote Sensing of Environment, 221, 665–674.

Xiao et al: An ensemble machine-learning model to predict historical PM2.5 concentrations in China from satellite data

We developed a machine learning approach to estimate historical PM2.5 levels in China. It overcame several previous issues related to satellite-based PM2.5 exposure estimates such as data missingness due to cloud cover, model stability across a large spatial domain, and robust hindcasting capability beyond the model fitting period. 

ES&T Link


PubMed Link


Wang et al.: A Bayesian Downscaler Model to Estimate Daily PM2.5 levels in the Continental U.S

We proposed a statistically reliable and interpretable national modeling framework based on Bayesian downscaling methods to be applied to the calibration of the daily ground PM2.5 concentrations across the conterminous United States using satellite-retrieved aerosol optical depth (AOD) and other ancillary predictors

IJERPH Link (open access)