An Application of Multivariate Techniques in Plankton Study of a Freshwater Body in the Niger Delta

Abstract

We utilized the principal components analysis (PCA) and hierarchical cluster analysis (HCA) to organize and interpret numerical abundances of phyto- and zoo-plankton biotypes of the middle course of the Imo River in a southeastern locality in Nigeria. PCA was used for data reduction while HCA was used to reveal natural groupings within data set of numerical abundances. Phytoplankton taxa abundance was dominated by bacillariophyceae (diatoms) (53.25%), while zooplankton was dominated by Cladocera (25.87%). Two plankton taxa PCs which accounted for about 76.30% variability in original 14 variables correlated most with Rotifera and fish eggs and larvae. HCA revealed chrysophyceans, euglenophyceans, cyanophyceans and chlorophyceans, as well as Crab larvae, fish eggs and larvae, beetle larvae and Copepoda forming the first and richest phytoplankton and zooplankton clusters, respectively. Results reveal that both extraction and clustering outputs utilized underlying criteria (such as seasonality and climatic variability) rather than numerical abundances in their classifications.

Keywords: Plankton taxa, PCA, HCA, multivariate analysis, numerical abundance