Identifying the Sources of Chemical Contamination After the Fact

In many environmental contamination events it is necessary to reconstruct the source term, and differentiate this source from other potential sources and the baseline condition, without the benefit of any analytical data for the source. This has been made possible with methods such as polytopic vector analysis (PVA), and self-organising maps (SOMs). PVA is an unmixing approach that generates a model with an increasing number of end members (sources) that at some stage explain the environmental data; when coupled with SOMs, the underlying factors that control the composition can be seen. The combination of these techniques has enabled the elucidation of the source chemical profile, along with those of the other contributors. In many cases, one end member will be the ubiquitous baseline condition, and this may be a sensible approach to reconstruct the concentration, chemical makeup, and relative contributions of sources to contaminated environments.