Assessment of air pollution sources in an industrial atmosphere using principal component and multilinear regression...

Aerosol samples collected in the industrial area of Estarreja, Portugal, were used to assess the source classes responsible for the particulate levels observed in the local atmosphere. Principal Component Analysis was applied separately to the concentrations of aerosol constituents and meteorological variables to obtain the number of Principal Components and to verify the influence of weather conditions on ambient air quality. The technique led to the conclusion that soil and transport emissions represent important aerosol sources even in this industrial environment. The quantitative contribution of each source class was calculated using Multilinear Regression Analysis; 37% of the aerosol mass had a soil origin, 8% was from sea spray, 18% resulted from transport emissions and 24% contained ammonium salts. Twelve percent of total suspended particle (TSP) mass could not be explained by any of the six Principal Components retained. Ammonium salts and two other minor Principal Components seem to result mainly from industrial emissions. More specific information about the contribution of each particular source was not possible with this technique.