Regression and stochastic models for air pollution. II: Application of stochastic models to examine the links between ground-level smoke concentrations and temperature inversions

Abstract In the contex of air pollution modelling, Box-Jenkins transfer function modelling has been used mainly for forecasting. Here the technique is used as a tool to assess the relative importance of various meteorological variables on surface smoke concentrations. Emphasis is given to temperature inversions. We demonstrate that this methodology is more appropriate than the regression approach. Box-Tiao intervention modelling is also used to examine the effect of extreme events, and the presence or absence of certain meteorological conditions, on surface smoke concentrations. The results of the transfer function modelling show that inversions do explain some of the variation in smoke surface concentrations but surface wind speed is a better predictor. Some of the results from Box-Tiao modelling were unexpected, possibly due to the simplicity of the models used, but the potential of the methodology was in little doubt. We make some suggestions for modifications which will allow such models to perform better.

[1]  F. B. Smith,et al.  Meteorological aspects of the transport of pollution over long distances , 1978 .

[2]  Richard A. Berk,et al.  Applied Time Series Analysis for the Social Sciences , 1980 .

[3]  R. Robson,et al.  On the theory of plume trapping by an elevated inversion , 1983 .

[4]  E. W. Hewson,et al.  Some Restrictive Meteorological Conditions to be Considered in the Design of Stacks , 1962 .

[5]  K. Wilson,et al.  Linear stochastic models for forecasting daily maxima and hourly concentrations of air pollutants , 1975 .

[6]  J. Colls,et al.  Correlations between acoustic sounder dispersion estimates, meteorological parameters and pollution concentrations , 1981 .

[7]  T. Davies,et al.  A five-year climatology of elevated inversions at Hemsby (UK) , 1992 .

[8]  R. Simpson,et al.  Forecasting peak ozone levels , 1983 .

[9]  Vertical Temperature Structure During the 1966 Thanksgiving Week Air Pollution Episode in New York City , 1972 .

[10]  Paul M. Mather.  Computational Methods of Multivariate Analysis in Physical Geography , 1976 .

[11]  D. Rubinfeld,et al.  Econometric models and economic forecasts , 2002 .

[12]  P. Brimblecombe,et al.  Chemical Composition of Snow in the Remote Scottish Highlands , 1988 .

[13]  P H Merz,et al.  Aerometric data analysis--time series analysis and forecast and an atmospheric smog diagram. , 1972, Atmospheric environment.

[14]  G. Box,et al.  On a measure of lack of fit in time series models , 1978 .

[15]  Roch Roy,et al.  On long term air quality trends and intervention analysis , 1982 .

[16]  C. Granger,et al.  Forecasting Economic Time Series. , 1988 .

[17]  George E. P. Box,et al.  Intervention Analysis with Applications to Economic and Environmental Problems , 1975 .

[18]  P. C. Manins.  Partial penetration of an elevated inversion layer by chimney plumes , 1979 .

[19]  David P. Chock,et al.  Time-series analysis of Riverside, California air quality data , 1975 .

[20]  Terence C. Mills,et al.  Time series techniques for economists , 1990 .

[21]  R. Stull.  An Introduction to Boundary Layer Meteorology , 1988 .

[22]  Paolo Zannetti,et al.  Simulation of Transformation, Buoyancy and Removal Processes by Lagrangian Particle Methods , 1985 .