Premium
PATTERN RECOGNITION OF PARTICULATE‐BOUND POLLUTANTS SAMPLED DURING A LONG TERM URBAN AIR MONITORING SCHEME
Author(s) -
ARMANINO CARLA,
RODA ALDO,
IUS ADRIANO,
CASOLINO M. C.,
BACIGALUPO M. A.
Publication year - 1996
Publication title -
environmetrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.68
H-Index - 58
eISSN - 1099-095X
pISSN - 1180-4009
DOI - 10.1002/(sici)1099-095x(199611)7:6<537::aid-env231>3.0.co;2-#
Subject(s) - particulates , term (time) , environmental science , pollutant , air pollutants , air pollution , air monitoring , environmental engineering , chemistry , physics , organic chemistry , quantum mechanics
Ambient atmospheric particulate samples were collected daily from the northern Italian city of Bologna. Sampling commenced in October 1991 and was completed in March 1993. Benzo(a)pyrene (BaP), a polycyclic aromatic hydrocarbon (PAH) derived from combu stion, and the metal species of Pb, Cr, V, Ni and Cd were determined in each particulate sample, with corresponding meteorological data being logged throughout this entire period. By using multivariate statistical analysis, patterns in that data set were explained. Principal component analysis and cluster analysis were applied to the chemical variables; groups of similar samples were found. The existence of the clusters, identified in the chemical variable domain, was tested in the domain of the atmospher ic parameters by Uneq, a class modelling method based on quadratic discriminant analysis; the dependence of air pollutant concentrations on atmospheric condition was verified. The influence of car traffic limitations on the concentrations of benzo(a)pyren e and metals was evaluated by principal component analysis; a reduction of these concentrations was made evident in the multivariate plots.