Spatiotemporal variation of PM2.5 over the contiguous United States

CrowleyFarenga, Remy Michael John

Permalink

https://hdl.handle.net/2142/101621 Copy

Description

Title

Spatiotemporal variation of PM2.5 over the contiguous United States

Author(s)

CrowleyFarenga, Remy Michael John

Issue Date

2018-07-20

Director of Research (if dissertation) or Advisor (if thesis)

• Di Girolamo, Larry
• Riemer, Nicole

Department of Study

Atmospheric Sciences

Discipline

Atmospheric Sciences

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Date of Ingest

2018-09-27T16:18:00Z

Keyword(s)

• PM2.5
• spaiotemporal
• seasonal
• diurnal
• covariation
• pollution
• particulate matter
• united states

Abstract

PM2.5 is the set of airborne particulate matter (also known as aerosols) with diameter 2.5 m or smaller. Ambient concentrations of these aerosols have been shown to be highly correlated with premature mortality throughout the world. Previous studies on ground-level, in-situ PM 2.5 concentrations have explored the diurnal and seasonal variation, but have been limited to small geographic areas. This research aims to identify the locations, times of day, and months of the year at which people are most likely to be exposed to high PM2.5 concentrations using data from ground monitors across the contiguous United States. K-means clustering was utilized to group of the diurnal and seasonal cycles by their shape in order to explore multi-scale temporal variations across the United States. Two distinct groups of diurnal cycles appeared in summer (June, July, August) and winter (December, January, February), with maxima occurring in the morning (0700 local time in the summer and 0800 in the winter) and the evening (roughly 20:00). Further, we found that the seasonal cycles are also separated into two clusters: one that has a summertime maximum, and the other with distinct winter and summer maxima. County-level spatial heterogeneity of PM2.5 was examined to quantify why monitors in similar areas may appear in different clusters. The correlation (Pearson's r) between monitors varies highly by county, which may be influenced by the placement of monitors with respect to emission sources of PM2.5. The correlation decreased with increasing distance between them, but that the correlation depended more strongly on the specific county in which the monitors are located. We quantified the covariation of PM2.5 with other pollutants to determine if we can separate their effects on human health for epidemiological studies. We found that PM2.5 is often positively correlated (r approximately equal to .5) with CO and NO2, with varying correlation when using different temporally averaged measurements. Lastly, we analyzed the diurnal cycles of clouds at several key locations identified as targets for the planning of the orbit of the Multi-angle Imager for Aerosols (MAIA) mission, which can only measure aerosol ground concentrations on clear days. The diurnal cycles in all locations were clearest in the morning, with higher cloudiness in the afternoon due to convection initiated by the warming ground. We found that the best time to sample PM2.5 concentrations remotely is around 08:00 local time, when there is the greatest chance of clear skies. However, this may lead to a high bias in MAIA's retrievals of PM2.5 concentrations if the intent is to retrieve a daily average concentration of PM2.5, as typically done in epidemiology studies.  

Graduation Semester

2018-08

Type of Resource

text

Permalink

http://hdl.handle.net/2142/101621

Copyright and License Information

Copyright 2018 Remy CrowleyFarenga

Owning Collections