No abstract

An attempt was made to resolve major sources of PCB contamination to the sediments of the Milwaukee Harbor Estuary. A principal component analysis model (PCA) with nonnegative constraints was developed and applied to three dated sediment cores: VC 6 in the Kinnickinnic River, VC 9 in the Harbor Entrance Channel, and VC 12 in the Outer Harbor. The model operates on eight common PCB congeners. Prior to the analysis of real data, the PCA model was examined for suitability using a simulated data set. The data set was created by Monte Carlo simulation and was conceived to reflect PCB Aroclor sources affected by random variations in transport and particle association. The model appears to successfully reconstruct the original profiles. In the actual data set, Aroclor 1016 and 1242 were major contributors to PCBs in cores VC 9 and VC 12. There was no Aroclor 1016 in VC 6. Aroclor 1254 and 1260 were of minor significance, but were identified as sources in all three cores. Both factor loadings and factor scores were examined and were found to give similar outcomes with respect to source identification. Results from the PCA model were also supported by the analysis by a chemical mass balance (CMB) model. Distinct source profiles and number of data points included in the analysis are of prime concern when the model is used for real world cases.

No abstract