Assessment of High-Resolution PM2.5 Exposures and Changes in PM2.5 Cardiorespiratory Disease Associations Over Time Público
Bi, Jianzhao (Summer 2020)
Abstract
Fine particulate matter with aerodynamic diameters less than 2.5 micrometers (PM2.5) is one of the six criteria air pollutants defined by the National Ambient Air Quality Standards. Numerous epidemiological studies have shown the associations between long-term and short-term exposure to PM2.5 and increased risks of cardiovascular and respiratory diseases. Understanding the accurate distribution of ground PM2.5 concentrations is of growing importance for studying the acute and chronic health effects of PM2.5. Measurements from satellites (aerosol optical depth, AOD) and low-cost air pollution sensors have been increasingly utilized in improving the estimation of ground PM2.5 concentrations due to their extensive spatiotemporal coverage. However, an important issue influencing the effective use of satellite AOD retrievals is the large proportion of non-random missing data caused by snow and cloud cover. This study examined the impacts of snow and cloud cover on AOD and PM2.5 and made full-coverage PM2.5 predictions with the consideration of these impacts. In addition, little has been done to incorporate low-cost sensor PM2.5 measurements in large-scale PM2.5 exposure modeling. This study conducted spatially varying calibration and developed a down-weighting strategy to optimize the use of low-cost sensor data in PM2.5 estimation. Finally, although PM2.5 is a complex mixture composed of different chemical components, it is commonly treated as a single pollutant to assess its health effects given that ambient air regulations focus on PM2.5 (and not its components). This study examined temporal changes in the risk of emergency department visits for cardiovascular diseases and asthma associated with short-term increases in ambient PM2.5 concentrations. By generating the improved PM2.5 exposures and examining the contribution of PM2.5 components to its overall toxicity, this study sought to broaden the application of satellite and low-cost sensor observations to PM2.5 exposure assessment and to provide new information for the health effects of PM2.5 mixture.
Table of Contents
1 Background and Significance . . . . . 1
2 Description of Aims . . . . . 6
3 Manuscript I: Impacts of Snow and Cloud Covers on Satellite-Derived PM2.5 Levels . . . . . 8
3.1 Abstract . . . . . 8
3.2 Introduction . . . . . 9
3.3 Data and Methods . . . . . 11
3.3.1 Study Areas . . . . . 11
3.3.2 PM2.5 Measurements . . . . . 11
3.3.3 MAIAC AOD Data . . . . . 12
3.3.4 MODIS Cloud and Snow Fractions . . . . . 13
3.3.5 Meteorological Data . . . . . 13
3.3.6 Land-Use Variables . . . . . 14
3.3.7 Data Matching . . . . . 14
3.3.8 AOD Gap-Filling Model . . . . . 14
3.3.9 PM2.5 Prediction Model . . . . . 15
3.3.10 Comparison With Another Gap-Filling Method . . . . . 17
3.4 Results . . . . . 18
3.4.1 Descriptive Statistics for MAIAC AOD Missingness . . . . . 18
3.4.2 AOD Gap-Filling by Random Forests . . . . . 19
3.4.3 PM2.5 Prediction by Random Forests . . . . . 20
3.4.4 The Importance of Gap-Filled AOD With Snow Cover Parameter . . . . . 24
3.5 Discussion . . . . . 25
3.6 Conclusions . . . . . 28
4 Manuscript II: Incorporating Low-Cost Sensor Measurements Into High- Resolution PM2.5 Modeling at a Large Spatial Scale . . . . . 29
4.1 Abstract . . . . . 29
4.2 Introduction . . . . . 29
4.3 Data and Methods . . . . . 32
4.3.1 Study Domain and Modeling Strategy . . . . . 32
4.3.2 Data . . . . . 33
4.3.3 PurpleAir PM2.5 Calibration and Weighted PM2.5 Modeling . . . . . 35
4.4 Results . . . . . 41
4.4.1 PurpleAir PM2.5 Calibration . . . . . 41
4.4.2 Weighted PM2.5 Modeling . . . . . 45
4.5 Discussion . . . . . 48
4.6 Conclusions . . . . . 52
5 Manuscript III: Temporal Changes in Short-Term Associations Between Cardiorespiratory Emergency Department Visits and PM2.5 in Los Angeles, 2005 to 2016 . . . . . 53
5.1 Abstract . . . . . 53
5.2 Introduction . . . . . 54
5.3 Data and Methods . . . . . 56
5.3.1 Study Population . . . . . 56
5.3.2 Air Pollution and Weather . . . . . 56
5.3.3 Emissions and Time Periods . . . . . 57
5.3.4 Statistical Analysis . . . . . 58
5.3.5 Sensitivity Analysis . . . . . 61
5.4 Results . . . . . 62
5.4.1 PM2.5 Concentrations . . . . . 62
5.4.2 Emergency Department Visits Data . . . . . 64
5.4.3 Relative Risks Associated With PM2.5 Total Mass . . . . . 65
5.4.4 Relative Risks Associated With PM2.5 Components . . . . . 67
5.4.5 Relative Risks Associated With the Remaining PM2.5 Mass . . . . . 69
5.4.6 Sensitivity Analysis . . . . . 70
5.5 Discussion . . . . . 70
5.6 Conclusions . . . . . 74
6 Conclusions and Future Directions . . . . . 76
7 Appendices . . . . . 78
7.1 Manuscript I Supplemental . . . . . 78
7.2 Manuscript II Supplemental . . . . . 83
7.2.1 Quality Control for PurpleAir PM2.5 Measurements . . . . . 83
7.2.2 Evaluation of PurpleAir PM2.5 Measurements . . . . . 83
7.2.3 Nonlinearity of PurpleAir Systematic Bias . . . . . 84
7.2.4 Validation of Scale Factor . . . . . 84
7.3 Manuscript III Supplemental . . . . . 97
References . . . . . 107
About this Dissertation
| School | |
|---|---|
| Department | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| Palabra Clave | |
| Committee Chair / Thesis Advisor | |
| Committee Members |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Assessment of High-Resolution PM2.5 Exposures and Changes in PM2.5 Cardiorespiratory Disease Associations Over Time () | 2020-06-08 21:12:37 -0400 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|