Abstract

Ambient monitored data at Santiago, Chile, are analyzed using box models with the goal of assessing contributions of different economic activities to air pollution levels. The period analyzed is 1990-2000, characterized by the introduction of air pollution emissions standards, shift to unleaded gasoline and compressed natural gas, and steady growth of the private and public fleet and the associated fuel consumption growth. The box models explicitly include the seasonal behavior of meteorological variables; the results show that dispersion conditions in fall and winter seasons are 20-30% of the summertime values. This result explains the poor air quality in those seasons and shows that significant emissions reductions are required in order to improve air quality in wintertime. Emissions of CO, NOx and SO2 are estimated from data on fuel consumption in the city; the estimated parameters are thus fleet-average or industry-average emission factors. In terms of contributions to ambient concentrations, older cars and diesel vehicles are the major contributors to CO and NOx impacts, with more than 60% and 50%, respectively. Ambient concentrations of SO2 are largely dominated by stationary sources, although long range contributions are not negligible. By contrast, CO and NOx pollution is dominated by local sources within the city boundaries. The box models can be used for forecasting purposes, and they can predict annual average concentrations within 20% of the observed values. The methodology requires data on ambient air quality measurements and fuel consumption statistics, and produces quantitative results, which can be combined with economic models to analyze environmental regulation and public policies. © 2002 Elsevier Science Ltd. All rights reserved.