A REFLECTIVE SPECTROSCOPY INVESTIGATION OF GASOLINE, DIESEL, AND JET FUEL-A IN LABORATORY AND COLD WEATHER ENVIRONMENTS: IMPLICATIONS FOR AGE MODELS, DETECTION AND QUANTIFICATION

Fuel spills are major environmental events which are difficult to monitor and detect in real time owing to the size of spills and the variability of shape of the plumes over time. Hyperspectral remote sensing offers a way to observe and quantify fuel spills essentially at all scales from hand held detectors to satellite platforms, provided an adequate library of reflective spectra exist for comparison. Study of the evaporation of fuels on substrates under conditions varying in temperature and humidity using spectral imagery could potentially provide increased accuracy and quicker detection of the effects of fuel spills and enable age estimates for fuel spill events. For this project a suite of gasoline, diesel and jet fuel-A was investigated in ambient laboratory conditions and outdoor cold weather conditions over periods of several weeks to investigate variation of spectra overtime. Fuel was applied to four common substrates, along with Ottawa sand. Extensive data were collected and analyzed with the goal of supplementing previous spectral library work. Notable absorption features were produced at 1200, 1400, 1725, and 2250 nm along with other minor features on the VIS/NIR/SWIR spectrum. Reflectance spectra varied with each sample due to physical features in the substrates. Spectra of gasoline samples were all similar to each other, while the jet fuel A and diesel spectra shared similarities but were distinct from gasoline. More pronounced absorption features were observed on lighter colored substrates (e.g., Ottawa sand). The samples contaminated with gasoline evaporated quicker than diesel and jet fuel-A samples in the climate controlled and cold weather experiments. Results suggest that spectral imagery discrimination between water and hydrocarbon materials can be accomplished, it is possible to develop age estimation tools however temperature factors and substrate type likely play a role. Recently published peer-review literature indicate that potential linkages to mathematical models of evaporation flux could exist, and such linkages may enable more quantitative studies from imagery.