Abstract
A SAR system can provide a primary source of critical data during an oil spill emergency, because of its insensitivity to weather and lightning conditions. Information on oil location, spreading, volume, and vulnerable ecosystems, is critical to oil spill management. However, the major challenges in offshore oil spill monitoring with SAR include the presence of look-alikes, oil quantity estimation, and identification of oil type. These are problematic to the successful implementation of oil spill response strategies with SAR. Many proposed techniques targeted at improving discrimination of oil spills and lookalikes failed to consider and contextualize the peculiarity of hydrocarbon oil characteristics and weathering processes such as the gradual thinning of an oil spill, which is detectable, observable, and measurable with the SAR system. As an oil spill spreads on the water surface the oil thickness reduces, thus, increasing radar reflectivity. In this thesis, therefore, a mathematical OSB-Model, that defines the relationship between radar backscatter and oil spreading is proposed. The model establishes the underlying scientific principle that characterizes radar backscattering and spreading behaviour of an offshore hydrocarbon oil spill based on empirical analysis such as the Isle of Wight controlled oil spill image dataset. In addition, using regression analysis, two best-line-of-fit models were identified and validated. The oil spill BCP curve model discriminates hydrocarbon oil spills and lookalikes’ backscattering behaviour, while SCP distinguishes the spreading behaviour of crude from diesel oil. 75% of the BCP graphs of the oils were tested and accepted at a 0.05 significance level as the logarithmic model of hydrocarbon oil spills. In terms of spreading rate, however, 50% of the SCP graph samples of each of the oil types was accepted as power and logarithmic curve models that characterized crude oil and diesel oil, respectively. The BCP and SCP techniques will significantly help in offshore oil spill monitoring, particularly during oil spill emergency operations.