ACPD Atmospheric Chemistry and Physics Discussions ACPD 1680-7375 Copernicus GmbH Göttingen, Germany 10.5194/acpd-6-8485-2006 Using discriminant analysis as a nucleation event classification method Mikkonen S. 1 Lehtinen K. E. J. 1 2 Hamed A. 1 Joutsensaari J. 3 Facchini M. C. 4 Laaksonen A. 1 Department of Physics, University of Kuopio, P.O. Box 1627, 70211 Kuopio, Finland Finnish meteorological institute, Kuopio Unit, P.O. Box 1627, 70210 Kuopio, Finland Department of Environmental Sciences, University of Kuopio, P.O. Box 1627, 70211 Kuopio, Finland Istituto di Scienze dell’Atmosfera e del Clima – CNR, Italy Via Gobetti 101, 40129 Bologna, Italy 07 09 2006 6 5 8485 8510 This is an open-access article ditributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. This article is available from http://www.atmos-chem-phys-discuss.net/6/8485/2006/acpd-6-8485-2006.html The full text article is available as a PDF file from http://www.atmos-chem-phys-discuss.net/6/8485/2006/acpd-6-8485-2006.pdf

More than three years of measurements of aerosol size-distribution and different gas and meteorological parameters made in Po Valley, Italy were analysed for this study to examine which of the meteorological and trace gas variables effect on the emergence of nucleation events. As the analysis method, we used discriminant analysis with non-parametric Epanechnikov kernel, included in non-parametric density estimation method. The best classification result in our data was reached with the combination of relative humidity, ozone concentration and a third degree polynomial of radiation. <I>RH</I> appeared to have a preventing effect on the new particle formation whereas the effects of O<sub>3</sub> and radiation were more conductive. The concentration of SO<sub>2</sub> and NO<sub>2</sub> also appeared to have significant effect on the emergence of nucleation events but because of the great amount of missing observations, we had to exclude them from the final analysis.