Reducing the susceptibility of SAR-based oil spill detection to low-wind look-alikes

Marine pollution caused by oil spills — the release of petroleum hydrocarbons — poses a major threat to marine environments. This has become increasingly true also in the Arctic, where reduced sea ice extent has made Arctic oceans more accessible for industry and tourism-related ship traffic. To minimize the environmental impacts of an oil spill, it is important to identify the source and monitor the extent and progression of a spill. In this paper, Lipschitz-Regularity (LR) transformation and two-dimensional stationary wavelet transform (2D-SWT) are integrated into a previously developed image analysis workflow [1] for application to oil spill detection. To describe the performance of the developed approach under controlled conditions, we have applied our method to simulated SAR data of wind driven oceans containing oil spills of various properties. The simulated SAR data starts from a SAR raw signal simulator, which includes both an ocean and an oil slick model. Several crude oil types of varying slick thickness can be simulated and wind speed as well as sensor wavelength can be varied in the simulator. In addition to simulated data, we also used our method on several real-life real life oil spill scenarios using a series of L-band ALOS PALSAR images, and X-band TerraSAR-X images acquired during the Deep-Water Horizon spill in the Gulf of Mexico in 2010 and C-band Sentinel-1A images acquired from the Persian Gulf in 2017. From our analysis, we concluded that the LR and 2D-SWT have distinct advantages in oil spill detection and lead to high performance spill mapping results.