Multifrequency Radar Imagery and Characterization of Hazardous and Noxious Substances at Sea

Abstract

The increase in maritime traffic, particularly the transport of hazardous and noxious substances (HNSs), requires advanced methods of identification and characterization in environmental chemical spills. Knowledge about HNS monitoring using radar remote sensing is not as extensive as for oil spills; however, any progress on this issue would likely advance the monitoring of both chemical and oil-related incidents. To address the need for HNS monitoring, an experiment was conducted in May 2015 over the Mediterranean Sea during which controlled releases of HNS were imaged by a multifrequency radar system. The aim of this experiment was to establish a procedure for collecting evidence of illegal maritime pollution by noxious liquid substances using airborne radar sensors. In this paper, we demonstrate the ability of radar imagery to detect and characterize chemicals at sea. A normalized polarization difference parameter is introduced to quantify both the impacts of released product on the ocean surface and the relative concentration of the substance within the spill. We show that radar imagery can provide knowledge of the involved HNS. In particular, one can distinguish a product that forms a film on the top of the sea surface from another that mixes with seawater, the information that is critical for efficient cleanup operations.

Dual-Polarized L-Band SAR Imagery for Temporal Monitoring of Marine Oil Slick Concentration

“…At the resolution cell scale, an emulsion and a partially covered surface have a similar impact on the Polarization Ratio. In this paper, it is assumed that the slick is homogeneous at the pixel scale (metric in this experiment) and the distinction between covered, partially covered and uncovered surface can be done using a detection method based on, for example, the Polarization Difference [10,14] which is mostly impacted by the sea surface roughness.…”

“…The four SAR scenes investigated in this paper have been projected from the slant range to the ground range geometry ( Figure 6). The spilled area has been automatically detected using the polarization difference (in linear unit) between VV and HH channels [10,12,14,31] and the temporal evolution of the detected area is pictured on Figure 7 in a false color representation: drifting of the MOS Sweeper spill for one hour (from 09:37 to 10:27 UTC) in the morning of 9 June (Figure 7a) and drifting of the spill for four hours (from 09:37 to 13:30 UTC), Figure 7b. The spilled product during the MOS Sweeper exercise has drifted in the North-East direction, under the influence of wind (coming from 250 • , cf.…”

“…A VV polarized image is displayed in Figure 11 together with the experimental PR and the estimated oil concentration maps. A detection mask has first been applied, based on the thresholding of the polarization difference [10,12,14,31]. The ranges of variation of PR values and oil concentrations have been reduced to 0.15-0.35 and 20-80%, respectively, to enhance the visualization.…”

“…The EM frequency is of primary importance when imaging the ocean surface and particularly for remotely sensed slick-covered area [8][9][10]. High-frequency imaging radars (e.g., X-or C-band) are preferable to those operating at lower frequency (e.g., L-band) for mineral oil slick detection [8].…”

“…High-frequency imaging radars (e.g., X-or C-band) are preferable to those operating at lower frequency (e.g., L-band) for mineral oil slick detection [8]. SAR data collected at L-band allows definition of the characteristics of a marine slick along a level ranging from thin surface films to oil-in-water emulsion [10,11]. In any case, the instrument noise is a…”