Chapter 8 - Principal Components

Abstract

The concept, procedure, and application of principal components analysis (PCA) in the field of environmental science is delivered in an easy and enjoyable way. PCA is a multivariate method employed to portray the relation among k measured variables and to identify the source of variation in the collected data. PCA works effectively when the number of variables is large or when the selected variables are strongly correlated (positively or negatively). Components are formed from the original variables; PCA tries to identify the source of variation with a few components. A step-by-step and interesting way is given to compute the components and enjoyable interpretation is provided to enable the researchers to comprehend the idea of PCA and how to apply it to environmental science. Examples are delivered from environmental science and tested by principal components analysis with R built-in functions and commands. The particulate matter in the air of selected palm oil mills was studied as a case study, and the second case considers the assessment of surface water quality.