A trajectory-clustering-correlation methodology for examining the long-range transport of air pollutants

Abstract We present a robust methodology for examining the relationship between synoptic-scale atmospheric transport patterns and pollutant concentration levels observed at a site. Our approach entails calculating a large number of back-trajectories from the observational site over a long period of time and subjecting them to cluster analysis. The short-term component (weather-related variations) of the pollutant concentration time-series data is segregated according to the back-trajectory clusters. Non-parametric statistics are then used to test for “significant” differences in the chemical composition of pollutant data associated with each cluster. Additional information about the spatial and temporal scales of pollutant transport is obtained from the time-lagged inter-site correlation analysis of ozone for a specific cluster. To illustrate the application of this methodology, we examined 5 yr long time-series data of ozone concentrations measured at Whiteface Mountain, NY, Cliffside Park, NJ, and Quabbin Summit, MA. The results provide evidence of ozone transport to these sites, revealing the spatial and temporal scales involved in the transport.

[1]  Trevor D. Davies,et al.  Cluster analysis: A technique for estimating the synoptic meteorological controls on air and precipitation chemistry—Method and applications , 1992 .

[2]  O. J. Dunn.  Multiple Comparisons Using Rank Sums , 1964 .

[3]  Iain D. Currie,et al.  Distribution‐Free Tests , 1989 .

[4]  Rainer Bleck,et al.  Modeling atmospheric transport to the Marshall Islands , 1985 .

[5]  S T Rao,et al.  Determining temporal and spatial variations in ozone air quality. , 1995, Journal of the Air & Waste Management Association.

[6]  A. Tamhane,et al.  Multiple Comparison Procedures , 1989 .

[7]  B. Stunder.  An Assessment of the Quality of Forecast Trajectories , 1996 .

[8]  P. S. Porter,et al.  Detecting Discontinuities in Time Series of Upper-Air Data: Development and Demonstration of an Adaptive Filter Technique , 1996 .

[9]  Trevor D. Davies,et al.  Cluster analysis: A technique for estimating the synoptic meteorological controls on air and precipitation chemistry—Results from Eskdalemuir, South Scotland , 1992 .

[10]  P. Samson,et al.  The influence of atmospheric transport on precipitation chemistry at two sites in the midwestern United States , 1989 .

[11]  Jonathan D. W. Kahl,et al.  Intercomparison of three long-range trajectory models applied to Arctic haze , 1989 .

[12]  William F. Ryan,et al.  Relationship between back trajectories and tropospheric trace gas concentrations in rural Virginia , 1994 .

[13]  J. Kahl.  On the prediction of trajectory model error , 1996 .

[14]  Igor G. Zurbenko,et al.  Separating Different Scales of Motion in Time Series of Meteorological Variables. , 1997 .

[15]  H. Charles Romesburg,et al.  Cluster analysis for researchers , 1984 .

[16]  James P. Lodge,et al.  Visibility and fine particles , 1991 .

[17]  T. Davies,et al.  Extending cluster analysis—synoptic meteorology links to characterise chemical climates at six northwest European monitoring stations , 1995 .

[18]  Douglas A. Wolfe,et al.  Nonparametric Statistical Methods , 1973 .

[19]  Igor G. Zurbenko,et al.  Space and Time Scales in Ambient Ozone Data , 1997 .

[20]  W. W. Daniel.  Applied Nonparametric Statistics , 1979 .

[21]  Jeffery T. McQueen,et al.  Evaluation of model back trajectories of the Kuwait oil fires smoke plume using digital satellite data , 1994 .