Applications of Remote Sensing Data in Air Pollution Modeling and Utilization of Model-Derived Exposure Estimates in Epidemiological Studies Restricted; Files & ToC

Abstract

Air pollution models rely heavily on regions with sufficient ground monitors for calibration. Recent advances in remotes sensing techniques have been successfully implemented in air pollution modeling in regions with adequate monitoring networks. This dissertation aims to implement remote sensing techniques in a low- and middle-income (LMIC) setting where limited ground monitoring measurements exist. The first aim of this dissertation is to develop a satellite derived PM2.5 (particulate matter with an aerodynamic diameter of 2.5 micrometer or less) exposure model to estimate PM2.5 at 1 km resolution from 2010 to 2016 in Lima, Peru. Estimates from this model is subsequently used in a study to investigate the association between PM2.5 and asthma in Lima to bridge the gaps in knowledge regarding air pollution studies in a LMIC setting where daily exposure often exceeds permissible standards. The next aim of this dissertation is to implementing remote sensing techniques in modeling a major wildfire event that requires finer spatial and temporal resolution data. The second aim of this dissertation is to build a machine learning model that incorporates low-cost sensors and the Synthetic Minority Over-sampling TEchnique (SMOTE) to artificially inflate extreme values in the training dataset to model the Camp Fire event in California in 2018. The methods and results from this aim will inform the necessary steps to improve model performance in modeling extreme events. Finally, the last aim of this dissertation is to utilize exposure estimates from a machine learning model to investigate the association between total PM2.5, smoke PM2.5, and non-smoke PM2.5, and serval cardiovascular diseases (CVDs) including acute myocardial infarction, arrhythmia, heart failure, ischemic heart disease, stroke, and total CVD. Results from this epidemiological study will provide more literature on the association between air pollution, both ambient and from wildland fire sources, and CVD outcomes.

Table of Contents

This table of contents is under embargo until 20 August 2023

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	Environmental Health
Degree	Ph.D.
Submission	Dissertation
Language	English
Research Field	Health Sciences, Public Health Remote Sensing Environmental Health
Keyword	SMOTE California Asthma AOD Random Forest Air Pollution Remote Sensing PM2.5 low-cost sensors CVD
Committee Chair / Thesis Advisor	Yang Liu, Emory University
Committee Members	Matthew Strickland, University of Nevada, Reno Kyle Steenland, Emory University Howard Chang, Emory University Ana Rappold, U.S. EPA

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	File download under embargo until 20 August 2023	2021-08-10 00:02:35 -0400	File download under embargo until 20 August 2023

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions