Extraction of environmental information from large aerosol data sets through combined application of cluster and factor analyses

To reveal useful environmental information which is contained in large analytical data sets, an approach, based on the successive application of hierarchical cluster analysis and factor analysis, is proposed. Estimation criteria to determine the most suitable number of clusters and/or factors, are discussed and the interpretation of the cluster and factor analyses results is performed using visual techniques. The data sets were obtained by scanning electron microscope-energy-dispersive X-ray analysis of individual North Sea aerosol particles.