Multidimensional Minimum Euclidean Distance Approach Using Radar Reflectivities for Oil Slick Thickness Estimation.

The need for oil spill monitoring systems has long been of concern in an attempt to contain damage with a rapid response time. When it comes to oil thickness estimation, few reliable methods capable of accurately measuring the thickness of thick oil slick (in mm) on top of the sea surface have been advanced. In this article, we provide accurate estimates of oil slick thicknesses using nadir-looking wide-band radar sensors by incorporating both C- and X-frequency bands operating over calm ocean when the weather conditions are suitable for cleaning operations and the wind speed is very low (<3 m/s). We develop Maximum-Likelihood dual- and multi-frequency statistical signal processing algorithms to estimate the thicknesses of spilled oil. The estimators use Minimum-Euclidean-Distance classification problem, in pre-defined multidimensional constellation sets, on radar reflectivity values. Furthermore, to be able to use the algorithms in oil-spill scenarios, we devise and assess the accuracy of a practical iterative procedure to use the proposed 2D and 3D estimators for accurate and reliable thickness estimations in oil-spill scenarios under noisy conditions. Results on simulated and in-lab experimental data show that M-Scan 4D estimators outperform lower-order estimators even when the iterative procedure is applied. This work is a proof that using radar measurements taken from nadir-looking systems, thick oil slick thicknesses up to 10 mm can be accurately estimated. To the best of our knowledge, the radar active sensor has not yet been used to estimate the oil slick thickness.