Absolute principal component analysis of atmospheric aerosols in Mexico City

Abstract

A receptor model is presented based on absolute principal component analysis (APCA) of elemental concentrations in atmospheric aerosols from Mexico City during the Summer of 1995. Elemental contents on samples collected with a Stacking Filter Unit of the Davis design was carried out using Particle Induced X-ray Emission (PIXE). The sampling device allowed the separation of particles with mean aerodynamic diameters ranging from 2.5 μm to 15 μm (coarse fraction) and smaller than 2.5 μm (fine fraction). Sampling was divided into morning, afternoon and night periods, with higher concentrations being found during the morning. Seasonal variation is observed when comparisons with other studies are carried out. The application of APCA allowed identification of four sources for each fraction, with a soil-derived dust predominance in the coarse one. The influence of meteorological parameters is studied using cluster analysis, showing that during the morning there is a transport of pollutants from the west towards the sampling site, while the night transport corresponds to soil-derived dust from the north.