Multivariate Analysis Techniques in Environmental Science

Abstract

One of the characteristics of environmental data, many of them and the complex\rrelationships between them. To reduce the number variables, different statistical methods\rexist. Multivariate statistics is used extensively in environmental science. It helps ecologists\rdiscover structure and previous relatively objective summary of the primary features of the\rdata for easier comprehension. However, it is complicated in theorical structure and in\roperational methodology.\rIn this chapter some important statistical methods such as Principal component analysis\r(PCA), Canonical correspondence analysis (CCA), Redundancy analysis (RDA), Cluster\ranalysis, and Discriminate function analysis will be explained briefly.\rThis chapter too cover the statistical analysis of assemblage data (species by samples\rmatrices of abundance, area cover etc) and/or multi variable environmental data which\rarise in a wide range of applications in ecology and environmental science, from basic\recological studies (e.g. of dietary composition or population size-structure), through\rcommunity-based field studies, environmental impact assessments and monitoring of largescale\rbiodiversity change, to purely physical or chemical analyses.\rThe use of multivariate analysis has been extended much more widely over the past 20\ryears. Much more is included on techniques such as Canonical Correspondence Analysis\r(CCA) and Non-metric Multidimensional Scaling (NMS) and another technique to include\rorganisms and organism-environment relationships other than vegetation. Spatially\rconstrained data analysis will be introduced and the importance of accounting for spatial\rautocorrelation will be emphasized. Use of the methods within ecology and in\renvironmental reconstruction will also be covered. A study and review of the applica