Evaluation of the changes in the Madrid metropolitan area influencing air quality:Analysis of 1999-2008 temporal trend of particulate matter

摘要

During the 1999-2008 period a statistically significant downward trend in the concentrations of SO_2, NO_x, CO and PM2.5 was determined at most of the urban and urban-background monitoring sites in the Madrid metropolitan area. However, no statistically significant trend was detected in PM10 concentrations at any urban and rural site and in NO2 concentrations at the urban sites. The reduction in the annual coal consumption and in the number of gasoline vehicles and the use of particle filters for diesel engines in this area, have contributed to this reduction of SO_2, CO and fine PM concentrations, but the increase of the diesel fleet is probably the cause of an increase of the NO2/NO_x ratio and the lack of a decreasing trend of NO_2 concentrations. Chemical characterisation and receptor modelling analysis results showed that mean contributions to PM10 and PM2.5 levels attributed to anthropogenic sources throughout the 1999-2008 period, decreased at the Madrid urban areas mainly by reductions in their carbonaceous and SO_4~(2-) contents. However, the contribution of mineral dust to PM10 bulk levels did not significantly decrease at the urban sites. Thus, strategies aimed at reducing levels of PM10 and PM2.5 concentrations focussing mainly on road traffic and coal consumption, can be insufficient to reduce the number of days exceeding the PM10 Daily Limit Value (50 μg m~3) at some metropolitan areas, such as Madrid, where crustal content due to anthropogenic activities and natural phenomena is elevated even when the African dust contribution during episodic days is excluded. Changes in the structure of the metropolitan area such as the increase of the Madrid car fleet and of its proportion of diesel vehicles, could also contribute to keep or increase the concentration of other pollutants (NO2) which resulted in increases of the NO_~-3 contribution to PM10 and PM2.5 levels.