In many scientific branches, and especially the environmental sciences, the need for data treatment and evaluation is increasing continuously. Therefore, the interest and application range of multivariate analysis have expanded accordingly with a view of structuring, interpreting, and evaluating complex data bases. In this context, the present volume attempts to review the state-of-the-art in multivariate analysis and give an overview of the various fields of application in ecology, environmental chemistry, toxicology, risk analysis, and structure-activity relationship (SAR) studies. The first chapters of the book are focusing on fundamental aspects of multivariate analysis. The second part contains a number of case studies, with the application of basic methods such as principal component analysis, correspondence factor analysis, discriminant analysis, and neural networks. This book reflects and combines the lectures organized in June 1991 in the frame of the Eurocourse programme at JRC Ispra under the sponsorship of the Institute for the Environment. It was the intention of the course and the resulting publication to promote and stimulate the application of multivariate analysis in environmental sciences and related areas. F. Geiss Director Environment Institute JRC Ispra vii LIST OF CONTRIBUTORS AND LECTURERS Auda Y.: Maison de 1'0rient Mediterraneen, CNRS, Universite Lyon 2, 7 rue Raulin, 69007 Lyon, France. Bacou A.M.: Centre d'Ecologie Fonctionnelle et Evolutive, CNRS, BP 5051, 34033 Montpellier CEDEX, France. Banco G.: Centre d'Ecologie Fonctionnelle et Evolutive, CNRS, BP 5051, 34033 Montpellier CEDEX, France. Barny M.H.: Commissariat a l'Energie Atomique, IPSN, BP 6,92265 Fontenay-aux­ Roses CEDEX, France.

InhaltsangabePresentation of correspondence analysis and multiple correspondence analysis with the help of examples.- Multiple factor analysis: a method to analyse several groups of variables measured on the same set of individuals.- Principal component and correspondence analyses with respect to instrumental variables: an overview of their role in studies of structure-activity and species-environment relationships.- Basic procedures in hierarchical cluster analysis.- Interpretation of hierarchical clustering.- Graphical techniques for multidimensional data analysis.- Factor analysis and risk perception.- The use of similarity and clustering techniques for the prediction of molecular properties.- Multivariate analysis of the input and output data in the fugacity model level I.- Multivariate analyses in genetic toxicology.- A structure-biodegradability relationship model by discriminant analysis.- Multivariate image analysis in chemistry: an overview.- Chemical multivariate image analysis: some case studies.- Multilayer neural networks applied to structure-activity relationships.