Ambient Air Pollution Estimation Using Bayesian Hierarchical Models Open Access

Abstract

Ambient fine particulate matter less than 2.5 micrometers in aerodynamic diameter (PM2.5) negatively affects various health outcomes. However, the sparsity of existing air quality monitors greatly restricts the spatial-temporal coverage of PM2.5 data, potentially limiting the accuracy of PM2.5-related health studies. Various methods exist to address these limitations by supplementing air quality monitoring measurements with additional data. We aim to contribute to these methods with ambient air pollution estimation using Bayesian models. First, 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. 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. Second, we evaluate the potential impact of differential exposure measurement error in PM2.5 when examining differences in associations among subpopulations defined by spatial regions. In a simulation study, we observe bias when performing stratified analyses by neighborhood-level socioeconomic status measures when exposure granularity is ignored. Finally, we further develop the ensemble approach for PM2.5 using multiple models and improve accuracy of methods by incorporating covariates into the weights. Bayesian estimation is accomplished through data augmentation with parameter expansion. The resulting weights are then used in a Bayesian ensemble averaging framework to combine estimates across data integration techniques.

Table of Contents

1 Introduction 1

1.1 Particulate matter less than 2.5 µm in diameter 2

1.2 Bayesian Spatial Hierarchical Models 3

1.3 Data Fusion 4

1.4 Specific Aims 5

2 A Bayesian Ensemble Approach to Combine PM2.5 Estimates from Statistical Models Using Satellite Imagery and Numerical Model Simulation 7

2.1 Introduction 8

2.2 Methods 11

2.3 Results 18

2.4 Discussion 25

3 Impacts of PM2.5 Exposure Spatial Resolutions on Estimating Neighborhood-Level Socioeconomic Status as an Effect Modifier  28 

3.1 Introduction 29

3.2 Methods 30

3.3 Application to Asthma Emergency Department Data 36

3.4 Simulation for Emergency Department Visit Modeling 40

3.5 Discussion 46

4 Combining Air Pollution Estimates from Multiple Models Using Bayesian Ensemble Averaging 49 

4.1 Introduction 50

4.2 Modeling 51

4.3 Estimation and Inference 53

4.4 Application 58

4.5 Discussion 66

A   Appendix for Chapter 2 68

B   Appendix for Chapter 3 79

C   Appendix for Chapter 4 86

Bibliography 88

About this Dissertation

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Laney Graduate School
Department	Biostatistics
Degree	Ph.D.
Submission	Dissertation
Language	English
Research Field	Health Sciences, Epidemiology Statistics Health Sciences, Public Health
Keyword	exposure assessment spatial modeling air pollution health impact Bayesian modeling
Committee Chair / Thesis Advisor	Chang, Howard, Emory University
Committee Members	Lyles, Robert, Emory University Ebelt, Stefanie, Emory University Waller, Lance, Emory University Liu, Yang, Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Ambient Air Pollution Estimation Using Bayesian Hierarchical Models ()	2021-11-19 16:07:10 -0500	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions