An Analysis Of Continuous Black Carbon Concentrations In Proximity To An Airport And Major Roadways
Abstracts: ISEE 20th Annual Conference, Pasadena, California, October 12-16, 2008: Symposium Abstracts
Silent Spring Institute, Boston, MA, USA.
Abstracts published in Epidemiology have been reviewed by the organizations of Epidemiology. Affliate Societies at whose meetings the abstracts have been accepted for presentation. These abstracts have not undergone review by the Editorial Board of Epidemiology.
In the vicinity of major airports, multiple source types may contribute to ambient pollutant concentrations, and determining the relative contributions of these sources is a key step in developing appropriate mitigation strategies. Regression analyses of continuous monitoring data can potentially provide these source contribution outputs, but only if real-time source and meteorological data can be incorporated. In this study, we conducted an analysis of continuous black carbon (BC) concentrations collected near T.F. Green Airport in Warwick, Rhode Island over a total of 14 months (July 2005 to September 2006). Sampling was conducted using aethalometers at five sites in Warwick, four in close proximity to the airport at various wind directions and a fifth site approximately three miles east-southeast of the airport. To determine predictors of BC concentrations, we considered both meteorological data and airport operations data, with data combined on an hourly basis. One-minute average BC concentration data were averaged and corrected for potential loading effects. To capture the joint effects of wind speed and direction, vectors of wind speed in the east-west direction and north-south direction were created. Since the temporal autocorrelation was rather high (one-hour lag Spearman correlation ((s) = 0.73), time-series bootstrap methods were used in combination with the LME models to obtain regression estimates. Conditional probability functions indicate that higher BC concentrations tend to occur during periods of low/calm wind speeds. Approximately 40% of the upper decile of the concentration distribution at each of the sampling locations proximate to the airport occurs during when the wind speeds drop below 3 m/s. Regression models indicate joint effects of wind speed and direction consistent with contributions from the airport and defined major roads, and demonstrate statistical significance for airport activities. Our statistical approach provides insight about the joint influence of source activities and meteorology on continuously measured concentrations and provides a template for analyses at complex urban sites.© 2008 Lippincott Williams & Wilkins, Inc.