Nicholas D. Rose and Timothy Negley shared a presentation on the "Challenges of Using Disparate Data Sets in  Forensic Methods"  at the tenth International Conference on Remediation and Management of Contaminated Sediments. 

Background/Objectives. Statistical methods are a valuable part of an environmental forensic  investigation and can be useful in identifying sources of contamination or allocating costs. Due  to removal of or lack of access to historic sources and/or cost limitations, forensic investigations  may require the use of data sets that were not collected explicitly for this purpose. Combining  data sets collected for different purposes, across multiple years, and by different entities  presents challenges that must be carefully considered to ensure usability and to understand the  limitations of the data. This work discusses some of the challenges involved in using disparate  data sets for environmental forensic applications.  

Approach/Activities. Sediment samples collected from the Elizabeth River in Virginia were  obtained from the National Oceanic and Atmospheric Administration (NOAA) website. These  sediment samples span much of the Elizabeth River and provide insights into potential source  identification and design of future environmental forensic investigations. Because the samples  were collected as part of 20 different studies between 1986 and 2002, they require a data  usability evaluation and cleaning to harmonize the results. To investigate issues in the data set  that could potentially skew the results of an analysis, the team used univariate statistical  techniques and multivariate statistics, including principal component analysis and Kmeans  clustering. This work highlights issues caused by multiple detection limits, censoring of data  sets, and differences in the analytes tested for in different studies.  

Results/Lessons Learned. The preliminary analysis highlights the multiple issues typically  present in environmental data sets when attempting to use multivariate forensics methods. An  analysis of polycyclic aromatic hydrocarbons (PAHs) in the data set, without an evaluation of  usability, results in the identification of four clusters, one of which suggests a unique source of  PAH contamination in Paradise Creek, a tributary of the Elizabeth River. However, a close  evaluation of the data reveals that the cluster suggesting a unique source of PAH contamination  is the result of studies where the detection limits for PAHs were high. Cleaning the data to  account for this issue results in identification of only two clusters, which is consistent with other  lines of evidence from investigations of the potential sources, including spatial location of known  sources, river-mile plots, and previous forensic investigations. Results of the analysis of the  cleaned data set, combined with the other lines of evidence, indicate that PAHs in Paradise  Creek are the result of urban background that is also present in other parts of the Elizabeth  River and not a separate source. Thus, analysis of the uncleaned data set could have resulted  in inconclusive or erroneous conclusions regarding sources of PAH contamination in Paradise  Creek. The results of this work emphasize that detailed evaluation of the data is required when  using disparate data sets to ensure usability and harmonization.