Combining Satellite Imagery and Numerical Model Simulation to Estimate Ambient Air Pollution: An Ensemble Averaging Approach

TL;DR

This paper presents an ensemble averaging method combining satellite aerosol data and numerical air quality models to improve daily PM2.5 estimation accuracy across the Southeastern US.

Contribution

It introduces a Bayesian model averaging approach that leverages both satellite and model simulation data, outperforming methods that use either source alone.

Findings

Ensemble approach improves PM2.5 estimation accuracy.

Outperforms statistical downscalers in cross-validation.

Applicable to other environmental risk estimations.

Abstract

Ambient fine particulate matter less than 2.5 $\mu$m in aerodynamic diameter (PM$_{2.5}$) has been linked to various adverse health outcomes and has, therefore, gained interest in public health. However, the sparsity of air quality monitors greatly restricts the spatio-temporal coverage of PM$_{2.5}$ measurements, limiting the accuracy of PM$_{2.5}$-related health studies. We develop a method to combine estimates for PM$_{2.5}$ using satellite-retrieved aerosol optical depth (AOD) and simulations from the Community Multiscale Air Quality (CMAQ) modeling system. While most previous methods utilize AOD or CMAQ separately, we aim to leverage advantages offered by both methods in terms of resolution and coverage by using Bayesian model averaging. In an application of estimating daily PM$_{2.5}$ in the Southeastern US, the ensemble approach outperforms statistical downscalers that use either AOD or CMAQ in cross-validation analyses. In addition to PM$_{2.5}$, our approach is also highly applicable for estimating other environmental risks that utilize information from both satellite imagery and numerical model simulation.