Source identification of underground fuel spills by pattern recognition analysis of high-speed gas chromatograms

Pattern recognition methods have been used to classify high-speed gas chromatograms of weathered and unweathered jet fuels. A total of 228 neat jet fuel samples representing common aviation fuels sold in the United States were characterized by 85-peak gas chromatograms. Discriminants were developed by parametric and non-parametric pattern recognition procedures that correctly classified the gas chromatograms of neat jet fuels according to fuel type (JP-4, Jet-A, JP-7, JPTS, or JP-5), and these discriminant functions were successfully used to classify gas chromatograms of jet fuels which had undergone weathering in a subsurface environment. This approach for identification of weathered fuels was taken because the physical and chemical interactions of jet fuel components with the subsurface environment are not yet fully understood.