REMOTE-SENSING OF CARBON-MONOXIDE VEHICLE EMISSIONS

The University of Denver's remote sensor for automobile exhaust emissi ons has been used to obtain valid model year (defined here as the year of first registration) and emissions data from roadside locations in Middlesbrough and London. Mean carbon monoxide emissions have been plo tted against model year to determine if there is a relationship betwee n mean pollutant emissions and the age of the vehicle. The results ini tially showed a large degree of scatter and no inference, concerning a ge and emissions, could be made. This can be largely attributed to the older vehicles in the fleet. Older vehicles were then omitted from an additional investigation providing a much stronger relationship with R(2) values of 0.88 being calculated for London and 0.77 for Middlesbr ough. Analyses of model year fleet emissions were also undertaken. It was demonstrated that old vehicles do not contribute significantly to fleet emissions. Pre-1983 registered vehicles contributed only 9% of t otal fleet emissions in Middlesbrough and only 21% of total fleet emis sions in London. It was also shown that the vast majority of fleet emi ssions come from a small number of highly polluting new vehicles (quin tile 5). Quintile 5 for model year 1989 represented 140 vehicles (2% o f the measured fleet) which contributed to 10.6% of fleet emissions at the Middlesbrough site.