Quantifying Disparities in Air Pollution Exposures across the United States Using Home and Work Addresses.

While human mobility plays a crucial role in determining ambient air pollution exposures and health risks, research to date has assessed risks on the basis of almost solely residential location. Here, we leveraged a database of ∼128-144 million workers in the United States and published ambient PM2.5 data between 2011 and 2018 to explore how incorporating information on both workplace and residential location changes our understanding of disparities in air pollution exposure. In general, we observed higher workplace exposures relative to home exposures, as well as increased exposures for nonwhite and less educated workers relative to the national average. Workplace exposure disparities were higher among racial and ethnic groups and job types than by income, education, age, and sex. Not considering workplace exposures can lead to systematic underestimations in disparities in exposure among these subpopulations. We also quantified the error in assigning workers home instead of a weighted home-and-work exposure. We observed that biases in associations between PM2.5 and health impacts by using home instead of home-and-work exposure were the highest among urban, younger populations.

A nested machine learning approach to short-term PM2.5 prediction in metropolitan areas using PM2.5 data from different sensor networks.

Many predictive models for ambient PM2.5 concentrations rely on ground observations from a single monitoring network consisting of sparsely distributed sensors. Integrating data from multiple sensor networks for short-term PM2.5 prediction remains largely unexplored. This paper presents a machine learning approach to predict ambient PM2.5 concentration levels at any unmonitored location several hours ahead using PM2.5 observations from nearby monitoring sites from two sensor networks and the location's social and environmental properties. Specifically, this approach first applies a Graph Neural Network and Long Short-Term Memory (GNN-LSTM) network to time series of daily observations from a regulatory monitoring network to make predictions of PM2.5. This network produces feature vectors to store aggregated daily observations as well as dependency characteristics to predict daily PM2.5. The daily feature vectors are then set as the precondition of the hourly level learning process. The hourly level learning again uses a GNN-LSTM network based on daily dependency information and hourly observations from a low-cost sensor network to produce spatiotemporal feature vectors capturing the combined dependency described by daily and hourly observations. Finally, the spatiotemporal feature vectors from the hourly learning process and social-environmental data are merged and used as the input to a single-layer Fully Connected (FC) network to output the predicted hourly PM2.5 concentrations. To demonstrate the benefits of this novel prediction approach, we have conducted a case study using data collected from two sensor networks in Denver, CO, during 2021. Results show that the utilization of data from two sensor networks improves the overall performance of predicting fine-level, short-term PM2.5 concentrations compared to other baseline models.