Quantitative comparison of PM 2.5 aerosol measuring devices
Michael Chin
Master of Public Health (M.P.H.), Drexel University
May 2009
DOI:
https://doi.org/10.17918/etd-3100
Files and links (1)
pdf
Chin_Michael_20091.55 MB
PDF Access upon request, Email title, URL, or DOI to archives@drexel.edu
Abstract
Aerosols--Sampling Air--Sampling PM2.5 Public Health
Objective: The purpose of this study was to examine the agreement of measurements among three different PM2.5 air sampling instruments: the TSI Inc. model 8520 DUSTTRAK aerosol monitor (DUSTTRAK) (TSI Inc., Shoreview, Minnesota), the SKC Inc. Sioutas Cascade Impactor (SCI) (SKC Inc., EightyFour,Pennsylvania), and the Model 200 Personal Environment Monitor (PEM) 2.5um Impactor (SKC Inc., EightyFour,Pennsylvania). These devices provide similar information but with different methodologies. Methods: Three air-sampling instruments, with controls, were utilized into 16 residential homes within the metropolitan Philadelphia area (PA) to determine 8-hr PM2.5 concentration from each home. Temperature, CO₂, relative humidity and weather conditions were recorded at sampling commencement and completion. The sampling was conducted from February 22, 2009 to May 4, 2009. Linear regressions and Pearson correlations were used as the principle statistical methods to analyze the results. A significanfor all statistical tests.ce level of 0.05 was used Results: The geometric means for SCI, PEM and DUSTTRAK were 16.17 ug/m3 (GSD 2.08), 10.23 mg/m3 (GSD 2.35), and 11.89 mg/m3 (GSD 2.23), respectively. Statistical analysis indicated that the highest correlation was observed between loge DUSTTRAK PM2.5 concentrations and loge PEM PM2.5 concentrations (R2 = 0.80) and the regression model p-value was <0.0001. A moderate correlation was observed between loge DUSTTRAK and loge SCI PM2.5 concentrations and between loge SCI and loge PEM PM2.5 concentrations with R2 value of 0.52 and 0.48, respectively. The regression model p-value for DUSTTRAK and SCI was <0.002. The regression model p-value for SCI and PEM was <0.003. Conclusion: The results suggest that there were good overall agreement between DUSTTRAK and PEM on the PM2.5 concentration measurements, whereas, SCI PM2.5 measurements had a moderate agreement with both DUSTTRAK and PEM PM2.5 measurements. The results also suggest that SCI will overall overestimate PM2.5 concentration measurements in a given study conditions reported by the DUSTTRAK and PEM. Further investigations are needed in comparing these devices with the federal reference method instruments.
Metrics
12 File views/ downloads
31 Record Views
Details
Title
Quantitative comparison of PM 2.5 aerosol measuring devices
Creators
Michael Chin - DU
Contributors
Hernando Perez (Advisor) - DU
Awarding Institution
Drexel University
Degree Awarded
Master of Public Health (M.P.H.)
Publisher
Drexel University; Philadelphia, Pennsylvania
Resource Type
Thesis
Language
English
Academic Unit
School of Public Health (2002-2015); Drexel University