Total Air Pollution And Space-Time Modelling

  • S. De Iaco
  • D. E. Myers
  • D. Posa
Part of the Quantitative Geology and Geostatistics book series (QGAG, volume 11)

Abstract

Two general problems occur in the analysis of air pollution data; multiple contaminants and a dependence on both spatial location and time of observation. Principal Component Analysis (PCA) provides a tool for removing the interdependence of the contaminant concentrations, in addition an analysis of the principal components, eigenvectors and eigenvalues provides additional insight into the dispersion and occurrence of the pollution plume. New models for space-time variograms and techniques for modelling them have been introduced by De laco, Myers and Posa.

Hourly average concentrations for nitric oxide (NO), nitrogen dioxide (NO 2) and carbon monoxide (CO) measured at 30 stations in 1999 in the Milan district, Italy, were used for the analysis. These were converted to daily averages and PCA was applied to each of the 365 data sets (3 contaminants and 30 stations). The eigenvectors of the correlation matrices were used to generate principal components, which can be considered as measures of Total Air Pollution (TAP) in lieu of the separate contaminant concentrations. These components were treated as samples from unobserved variates defined over space and time. Space-time variograms were fitted to these new variates using the product sum model.

Although linked in these analyses, the principal components and their associated eigenvectors as well as the scores for each station vs the space-time variogram models provide two different pictures of the spatial and temporal dispersion of the contaminants as well as their interaction at different times of the year.

Keywords

Dioxide Covariance Carbon Monoxide Kriging Cokriging 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Davis, B. and Greenes, (1983), Estimation using distributed multivariate data: an example with coal quality, Math. Geology 15, 2, 287–300.CrossRefGoogle Scholar
  2. De Cesare, L., Myers, D. E. and Posa, D., (1997), Spatial Temporal Modeling of SO 2 in the Milan District, Geostatistics Wollongong’96, E.Y. Baafi and N.A. Schofield (eds), Kluwer Academic Publishers, 1031–1042.Google Scholar
  3. De Cesare, L., Myers, D.E. and Posa, D., (2000), A FORTRAN program for Space-Time Modeling, submitted.Google Scholar
  4. De Cesare, L., Myers and Posa, D., (2001), Est imating and modelling Space-Time Correlation Structures, Statistics and Probability Letters, 51, 1, 9–14.MathSciNetMATHCrossRefGoogle Scholar
  5. De Iaco, S., Myers, D.E. and Posa, D., (2001), Space-Time analysis using a general product-sum model, Statistics and Probability Letters, 52,1, 21–28.MathSciNetMATHCrossRefGoogle Scholar
  6. Deutsch, C. V. and Journel, A. G. (1997). GSLIB: Geostatistical Software Library and User’s Guide, Oxford Univ. Press, New York.Google Scholar
  7. Myers, D.E., (1994), The Linear coregionalization and simultaneous diagonalization of the variogram matrix function. Sciences de la Terre, 32, 125–139.Google Scholar
  8. Myers, D.E. and Carr, J., (1984), Cokriging and Principal Component Analysis: Bentonite Data revisited. Sciences de la Terre, 21,65–77.Google Scholar
  9. Rouhani, S. and Myers, D.E., (1990), Problems in Space-Time Kriging of Hydrogeological data. Math. Geology, 22, 611–623.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2001

Authors and Affiliations

  • S. De Iaco
    • 1
  • D. E. Myers
    • 2
  • D. Posa
    • 3
    • 4
  1. 1.Facoltà di EconomiaUniversità. “G. D’Annunzio”di ChietiChietiItaly
  2. 2.Dept of MathematicsUniversity of ArizonaTucsonUSA
  3. 3.Facoltà di EconomiaUniversità. di LecceLeeceItaly
  4. 4.Istituto per Ricerche di Matematica Applicata (CNR)BariItaly

Personalised recommendations