Modified polytopic vector analysis to identify and quantify a dioxin dechlorination signature in sediments. 1. Theory

Area of Science:

• Environmental Chemistry
• Geochemistry
• Multivariate Analysis

Background:

• Effective sediment management relies on identifying contamination sources and environmental fate processes.
• Polytopic Vector Analysis (PVA) is a multivariate technique used for environmental forensic investigations.
• Traditional PVA is limited to positive component values, hindering the analysis of reactive end-members.

Purpose of the Study:

• To develop a modified PVA algorithm (M-PVA) capable of resolving dioxin dechlorination fingerprints.
• To adapt PVA for analyzing biotic/abiotic transformations in sediment samples.
• To enable the characterization of reactive end-members with both positive and negative values.

Main Methods:

• Development of a modified Polytopic Vector Analysis (M-PVA) algorithm.
• Application of M-PVA to artificial datasets with known end-member compositions.
• Analysis of dioxin patterns to isolate dechlorination-specific compositional changes.

Main Results:

• The M-PVA successfully reproduced dioxin dechlorination fingerprints with a root mean square error of 28-41% on artificial datasets.
• The dechlorination end-member contribution was overestimated 1-5 fold.
• Model accuracy in reproducing patterns and variability depends on the actual contribution of dechlorination.

Conclusions:

• M-PVA is a significant advancement for identifying dioxin dechlorination processes in sediments.
• Uncertainty analysis is crucial when applying M-PVA to environmental data to differentiate dechlorination variability from error.
• This method facilitates field characterization of fate processes in dioxin-contaminated sediments.