Isolation of homogeneous water masses by the data of the Skagerrak experiment using cluster-analysis

By tradition water masses in the Skagerrak are isolated on the basis of hydrological data about their origin and by common oceanological parameters, like temperature and salinity, to which presently nitrate concentration is added in addition. The purpose of this work is to show, that the given problem may be solved using as a powerful method of multivariate statistics as cluster-analysis by K-means. The advantage of this method is a possibility to use all available hydrochemical data for identification and classification of water masses, to decide between them the most informative data and based on them to obtain more reliable and objective findings. Furthermore, analysis by K-means facilitates the estimation of statistical relationships between hydrochemical regime of particular water masses and their hydrobiological characteristics, that is one of a number of objectives specified by ICES in the SKAGEX project.