Characterization of oil slicks at sea using remote sensing techniques

Oil slicks can be visually detected through remote sensing techniques because of sharp image contrast variations between the oil slicks and surrounding water. These contrast variations are usually due to the dampening of the water surface roughness caused not just by oil, but possibly also by freshwater runoff and biogenic surfactants (also called “biogenic look-alikes”), such as those due to phytoplankton blooms. Floating macroalgae can also alter the texture of the water surface and contribute to look-alikes. Using methodologies we developed and implemented in previous studies of oil spills using hyperspectral optical imagery, we have tested several algorithms for biogenic look-alikes and oil slick characterization from optical and RADAR sensors in order to improve operational monitoring of marine coastal areas for oil pollution. With the opportunity to use imagery acquired over the Deepwater Horizon oil spill in the Gulf of Mexico in 2010, we have demonstrated promising utility of optical imagery to assist in differentiating oil from RADAR look-alikes in low wind situations. We have demonstrated that the interpretation of false positives for oil slicks in RADAR imagery can be adequately assisted by the analysis of optical imagery. Furthermore, oil spill extent and slick thickness can be mapped and characterized using spaceborne imagery. This represents a major improvement over local observations of oil spill for emergency and mitigation actions by improving response time and providing a synoptic view of the impacted areas.

Перевод пока недоступен