application of multivariate statistical methods to optimize water quality monitoring network with emphasis on the pollution caused by fish farms

background: one of the key issues in determining the quality of water in rivers is to create a water quality control network with a suitable performance. the measured qualitative variables at stations should be representative of all the changes in water quality in water systems. since the increase in water quality monitoring stations increases annual monitoring costs, recognition of the stations with higher importance as well as main parameters can be effective in future decisions to improve the existing monitoring network. methods: sampling was carried out on 12 physical and chemical parameters measured at 15 stations during 2013-2014 in haraz river, northern iran. results: the results of the measurements were analyzed using multivariate statistical analysis methods including cluster analysis (ca), principal component analysis (pca), factor analysis (fa), and discriminant analysis (da). according to the ca, pca, and fa, the stations were divided into three groups of high pollution, medium pollution, and low pollution. conclusion: the research findings confirm applicability of multivariate statistical techniques in the interpretation of large data sets, water quality assessment, and source apportionment of different pollution sources.