We developed national machine-learning models to provide interpretable results, to predict concentrations of PM2.5 sulfate, nitrate, organic carbon (OC) and elemental carbon (EC) across the conterminous United States from 2005 to 2015 at the daily level.
Author: Yang Liu PhD
Dr. Liu has 20 years of experience on studying the spatial and temporal characteristics of air pollution using remote sensing data from various satellite instruments. His research interests also include satellite aerosol retrieval and product design, the potential impacts of global climate change on public health, GIS and spatial statistics. Dr. Liu has been funded by NASA, CDC, NIH, EPA, HEI, and WHO to apply satellite data in air quality modeling and study the impact of climate change on air quality and human health using remote sensing and model simulations. Specifically, he has been working with the Oak Ridge National Lab to generate statistically and dynamically downscaled high-resolution regional climate projections under various RCPs. Using these simulation results, Dr. Liu conducted research to estimate spatially resolved population risks associated with extreme heat, wildfires, ambient air pollution such as ozone and PM2.5, and infectious diseases that can be attributed to future climate change. He has published 200+ peer-reviewed journal articles and contributed to two books in these areas. He is a member of the NASA MAIA and Terra MISR science team, the NASA Aura science team, a PI member of the NASA Air Quality Applied Science Team (AQAST) and the following Health and Air Quality Applied Science Team (HAQAST). He is a member of the Scientific Steering Committee of the WHO Platform on Air Quality and Health, and a GBD expert on ambient air pollution.
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.
Geng et al. The sensitivity of satellite-based PM2.5 estimates to its inputs: implications to model development in data-poor regions
We conducted comprehensive sensitivity model simulations to investigate the key factors that could affect the performance of satellite-based PM2.5 exposure in data-poor regions including temporal sampling frequency, number of ground monitors, accuracy of the chemical transport model simulation, and different combinations of the additional predictors.
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