Application of fingerprint-based multivariate statistical analyses in source characterization and tracking of contaminated sediment migration in surface water

Environ Pollut. 2013 Aug:179:224-31. doi: 10.1016/j.envpol.2013.04.028. Epub 2013 May 18.

Abstract

This study investigates the suitability of multivariate techniques, including principal component analysis and discriminant function analysis, for analysing polycyclic aromatic hydrocarbon and heavy metal-contaminated aquatic sediment data. We show that multivariate "fingerprint" analysis of relative abundances of contaminants can characterize a contamination source and distinguish contaminated sediments of interest from background contamination. Thereafter, analysis of the unstandardized concentrations among samples contaminated from the same source can identify migration pathways within a study area that is hydraulically complex and has a long contamination history, without reliance on complex hydrodynamic data and modelling techniques. Together, these methods provide an effective tool for drinking water source monitoring and protection.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Environmental Monitoring / methods*
  • Fresh Water / chemistry
  • Geologic Sediments / chemistry*
  • Metals, Heavy / analysis
  • Multivariate Analysis
  • Polycyclic Aromatic Hydrocarbons / analysis
  • Water Pollutants, Chemical / analysis*
  • Water Pollution, Chemical / statistics & numerical data*

Substances

  • Metals, Heavy
  • Polycyclic Aromatic Hydrocarbons
  • Water Pollutants, Chemical