Transboundary contribution to urban air pollution

Abstract

Elevated particulate matter concentrations in urban locations have normally been associated with local traffic emissions. Recently it has been suggested that such episodes are influenced to a high degree by PM10 sources external to urban areas. The aim of this report is three fold: 1) the further corroboration of the above hypothesis, 2) the application of methods other than linear regression to the analysis, 3) the mathematical modeling of the transboundary process. To further corroborate this hypothesis, linear regression was sought between PM10 concentrations measured at eight urban sites in the UK, with particulate sulphate concentration measured at two rural sites, for the years 1993-1997. Analysis of the slopes, intercepts and correlation coefficients indicate a possible relationship between urban PM10 and rural sulphate concentrations. The value of linear regression as an analysis tool in such cases is discussed and it is shown that an analysis of the sign of the rate of change of the urban PM10 and rural sulphate concentrations provides a more realistic method of correlation. Two novel approaches employing the application of filter theory and of the analysis of the standard deviation of hourly correlation coefficients are shown to give basically the same results as the PM10- sulphate linear correlation but using only urban PMl 0 data. The results indicate a major influence on urban PM10 concentrations from the eastern side of the United Kingdom. Linear correlation was also sought using PM10 data from nine urban sites in London and nearby rural Rochester. Analysis of the magnitude of the gradients and intercepts together with episode correlation analysis between the two sites showed the effect of transported PM10 on the local London concentrations. The report also presents methods to estimate the influence of rural and urban PM10 sources on urban PM10 concentrations and to obtain a rough estimate of the transboundary contribution to urban air pollution from the PM1o concentration data of the urban site.