The application of hyperspectral image techniques on MODIS data for the detection of oil spills in the RSA

摘要

Oil spills pose a serious threat to the sensitive marine ecosystem of the RSA. The study aims to detect and identify oil spills using remote sensing data provided by ROPME MODIS receiving station. MODIS data of confirmed incidents of oil spills via in-situ observations were processed to produce radiometrically corrected LIB data. Algal mats were further eliminated as look-alike, when the distinct oil pattern was not visible in the standard MODIS algorithm for Chlorophyll a. Shape analysis based on the operators' prior knowledge of the region was also used as a method for discriminating oil from other look-alikes. Oil spills exhibit different levels of contrast in relation to the viewing angle geometry and sun position. The Spectral Contrast Shift (SCS) is an empirical relationship that was derived to identify sea surface patterns including oil spills using the maximum and minimum spectral radiance values at the 250m spatial resolution bands. Results were combined with GIS based information of oil platform locations and daily tanker routes to aid interpretation and improve the probability for an accurate identification of oil spills, and avoiding false positives.