Atmospheric aerosols over a southwestern region of Texas

Paul Chiou, Wei Tang, Che Jen Lin, Hsing Wei Chu, Rafael Tadmor, T. C. Ho

Research output: Contribution to journalArticlepeer-review


Speciated samples of PM2.5 were collected at the Big Bend site from July of 2003 to June 2006 and the McDonald Observatory site from July of 2003 to August of 2005 in southwestern Texas, respectively, by the US Environmental Protection Agency. A total of 175 samples for the Big Bend site and 105 samples for the McDonald Observatory site with 52 species were measured; however, 30 and 32 species from the Big Bend and McDonald Observatory sites, respectively, were excluded because of too much below-detection-limit data. Due to the laboratory change about November 1 of 2004 and possible analytical artifacts, phosphorous was excluded as well. Among the species excluded, 31 species are common to both sites. The two data sets were analyzed by positive matrix factorization to infer the sources of PM observed at the two sites. The analysis resolved five source-related factors for Big Bend and four for McDonald Observatory. Sulfate-rich secondary aerosol, coal burning, motor vehicle/road dust, and a mixed factor were identified as common sources to both sites. The other factor identified for Big Bend is related to soil. Sulfate mainly exists as ammonium salts. The sulfate-rich secondary aerosols account for about 62% and 66% of the PM2.5 mass concentration at the two sites, respectively. The highest concentration of Si associated with Ca, Fe,SO2-4, and organic carbon at the two sites was possibly attributed to the coal-fired power plants in the region. Basically, the factor of sulfate and coal burning at the two sites showed similar chemical composition profiles and seasonal variation that reflect the regional characteristics of these sources. The regional factors of sulfate, coal burning, and soil showed predominantly low-frequency variations; however, the area-related and/or local factors showed both high and low frequency variations. The motor vehicle/road dust and the mixed factors were likely to be area-related and/or local source.

Original languageEnglish
Pages (from-to)645-659
Number of pages15
JournalEnvironmental Modeling and Assessment
Issue number5
StatePublished - 1 Jan 2009
Externally publishedYes


  • Airborne particulate matter
  • Conditional probability function
  • Correlation coefficient
  • Factor contribution
  • Factor profiles
  • PM
  • Positive matrix factorization
  • Receptor modeling
  • Time series analysis

ASJC Scopus subject areas

  • Environmental Science (all)


Dive into the research topics of 'Atmospheric aerosols over a southwestern region of Texas'. Together they form a unique fingerprint.

Cite this