Optical remote sensing of oil in the marine environment : an investigation into the optical properties of surface and dispersed oil and the means by which pollution may be detected using airborne optical instruments at visible and near-infrared wavelengths

Byfield, Valborg (1998) Optical remote sensing of oil in the marine environment : an investigation into the optical properties of surface and dispersed oil and the means by which pollution may be detected using airborne optical instruments at visible and near-infrared wavelengths. University of Southampton, Doctoral Thesis.

Abstract

Remote sensing has played an increasing role in the routine monitoring of oil pollution and in support of the operational response to major oil pollution incidents. This study develops the technique of optical measurement for the detection of oil in the Marine Environment.

A theoretical model is proposed, which relates upwelling radiance from surface oil to the optical properties of the oil in question, to the thickness of the oil layer, and to a number of ancillary environmental parameters. It is used to interpret the results of laboratory experiments in artificial and natural light, and ultimately as a tool in the analysis of airborne optical data of surface oil in the field, including the Sea Empress oil spill.

Laboratory experiments showed that the thickness of surface oil may be determined using spectral ratios, and the result compared well with the predictions made by the theoretical model. Using the peak to near-infrared ratio, relative thickness estimates can be made from remote sensing data, without extensive data processing. Absolute thickness measurements are more complex, and require the knowledge of a number of environmental parameters.

Both the laboratory and airborne data show that classification of oils into broad groups is possible using spectral analysis. However, the number of environmental parameters that must be considered makes this a complex task for field data.

The model predicts that sheen detection will be most reliable in regions of the spectrum where the sub-surface signal is low, such as the violet to deep blue and the near-infrared. This is confirmed by the laboratory experiments in natural light, and by the airborne data from the field experiments. When water-leaving radiance is high in the near-infrared, sheen detection may be more difficult, although it should still be possible in the violet to deep blue.

The theoretical model and the field data suggest that dispersed oil may be detected if concentrations are sufficiently high. The presence of suspended sediment or high concentrations of planktonic algae will, however, make this task more difficult, and success depends on a good knowledge of the background conditions.

This record has no associated files available for download.

Contributors

Download statistics