1：Experimental Design and Method Selection:

The study investigates the potential of CP from three RCM SAR modes for oil spill detection. CP parameters are simulated for each of the tested RCM SAR modes and investigated for oil spill detection. The most effective CP parameters for oil spill detection in each mode are extracted and used for classification of oil spills and lookalike using the Random Forest (RF) classification algorithm.

2：Sample Selection and Data Sources:

The test site for this study is located in the North Sea offshore Norway. A large-scale oil spill exercise at the Frigg field was conducted in June

3：Crude Oil (CO), Emulsion (EM) which is mixture of oil and water, and plant oil to be treated as lookalike (LA) were spilled during the experiment. Oil spills were captured using two RADARSAT-2 fine quad-pol SAR images. List of Experimental Equipment and Materials:

20 A RCM data simulator, developed at the Canada Center for Mapping and Earth Observation, was used to simulate RCM CP SAR data.

4：Experimental Procedures and Operational Workflow:

The simulator calculates and generates 23 CP SAR parameters for each mode. Samples of oil spill and LA were collected and used for separability estimation in each CP parameter using the Kolmogorov-Smirnov (K-S) nonparametric distance. The correlation between the identified CP parameters was also analyzed by estimating the Spearman correlation coefficient between parameters.

5：Data Analysis Methods:

The extracted subset of parameters of each mode was used for the classification of EM and LA using the RF classification algorithm.