Exploring geochemical data using compositional techniques: a practical guide

Abstract

John Aitchison revolutionised in 1982 our way of approaching geochemical data focusing on their relative nature. In this perspective, the investigation of single variables is meaningless due to the entangled structure that links all the parts of a composition. Starting from that time, several developments have characterized the debate within the scientific community, both from the applied and the theoretical point of view. The consequence was that the number of papers where compositional data are consistently and coherently managed increased exponentially. The exploratory phase of compositional data is a very important step in data analysis and modeling. It not only helps to clarify the available sample data structure but also determines the base to develop models to predict time and space changes. Real chemical data along the course of the river Tevere (Tiber) (Italy) and its tributaries are taken to illustrate how compositional techniques help explore compositions and detect patterns and outliers in the data

JJE and VPG were supported by the Ministerio de Ciencia e Innovación, Spain, under the projects “CODA-GENERA” (Ref. PID2021-123833OB-I00) and “CONBACO” (Ref. PID2021-125380OB-I00). AB and CG acknowledge the support of the National Biodiversity Future Center (NBFC) and National Centre for HPC, Big Data and Quantum Computing to the University of Florence, Department of Earth Sciences, funded by the Italian Ministry of University and Research, PNRR, Missione 4 Componente 2, “Dalla ricerca all'impresa”, Investimento 1.4, Projects CN00000033 and CN00000013