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.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Davis, B. and Greenes, (1983), Estimation using distributed multivariate data: an example with coal quality, Math. Geology 15, 2, 287–300.
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.
De Cesare, L., Myers, D.E. and Posa, D., (2000), A FORTRAN program for Space-Time Modeling, submitted.
De Cesare, L., Myers and Posa, D., (2001), Est imating and modelling Space-Time Correlation Structures, Statistics and Probability Letters, 51, 1, 9–14.
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.
Deutsch, C. V. and Journel, A. G. (1997). GSLIB: Geostatistical Software Library and User’s Guide, Oxford Univ. Press, New York.
Myers, D.E., (1994), The Linear coregionalization and simultaneous diagonalization of the variogram matrix function. Sciences de la Terre, 32, 125–139.
Myers, D.E. and Carr, J., (1984), Cokriging and Principal Component Analysis: Bentonite Data revisited. Sciences de la Terre, 21,65–77.
Rouhani, S. and Myers, D.E., (1990), Problems in Space-Time Kriging of Hydrogeological data. Math. Geology, 22, 611–623.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer Science+Business Media Dordrecht
About this paper
Cite this paper
De Iaco, S., Myers, D.E., Posa, D. (2001). Total Air Pollution And Space-Time Modelling. In: Monestiez, P., Allard, D., Froidevaux, R. (eds) geoENV III — Geostatistics for Environmental Applications. Quantitative Geology and Geostatistics, vol 11. Springer, Dordrecht. https://doi.org/10.1007/978-94-010-0810-5_4
Download citation
DOI: https://doi.org/10.1007/978-94-010-0810-5_4
Publisher Name: Springer, Dordrecht
Print ISBN: 978-0-7923-7107-6
Online ISBN: 978-94-010-0810-5
eBook Packages: Springer Book Archive