Oils floating on the sea surface can be observed as “dark” patches on radar images since the backscattered signals from the contaminated area are reduced in two dominant ways. First, oil slicks could damp short gravity and capillary waves on the sea surface responsible for backscattering energy. Second, the oil-covered sea surface permittivity decreases significantly if the oil film is sufficiently thick or mixed with seawater, i.e., oil emulsion. In this article, the geometry of the oil-covered sea surface is accounted for by the damping of sea waves, which is described by the model of local balance (MLB) combined with the sea wave spectrum. The radar backscattering is predicted by the advanced integral equation method (AIEM) model. The reflection coefficients are calculated based on a layered-medium model to analyze the impact of oil thickness and emulsions on the radar scattering. Numerical simulations demonstrate that: 1) the sensitivity to oil thickness and water content of the oil spills increases when the radar frequency increases; 2) the backscattering signals exhibit a nonlinear behavior with respect to oil thickness; and 3) high wind speed can generally narrow the difference between the radar backscattering from the clean and oil-covered sea surface, while the incidence angle has little effect. Numerical simulations are then compared with the multifrequency synthetic aperture radar observations acquired during the Gulf of Mexico Deepwater Horizon (DWH) oil spill accident and the 2011 Norwegian Clean Seas Association for Operating Companies (NOFO) oil-on-water exercise. Comparison results show that it is possible to estimate the oil thickness at reasonably good accuracy.

In this study, the bistatic radar scattering coefficients related to an oil-covered sea surface are predicted by modeling both the oil damping effect on surface roughness–through the advanced integral equation method–and the oil modification on the dielectric properties of the scattering surface. The bistatic scattering is analyzed in the whole upper scattering space under different radar frequencies, incidence angles, wind speeds, and oil thicknesses. Numerical predictions show that the scattering energy of an oil-covered sea surface is generally higher in the forward scattering zone than that in the backward one. In addition, the oil damping effect is the main mechanism ruling the scattering behavior in the backward region. The information related to the bistatic scattering geometry is also explored to retrieve oil thickness, representing one of the key parameters for radar-based marine oil pollution observation. A new index is proposed to quantify the sensitivity of bistatic scattering coefficients to oil thickness in different cases: single-polarization features, dual co-polarization features, namely, the polarization ratio (PR) and the normalized polarization difference index (NPDI), and dual-angular scattering features. Numerical results show that the bistatic scattering coefficients result in an enhanced sensitivity to oil thickness with respect to the monostatic case. The single HH-polarized scattering coefficients show better oil thickness sensitivity in the backward region, while the VV-polarized ones are more sensitive to oil thickness in the forward region. The combination of dual-polarized scattering coefficients significantly improves the oil thickness sensitivity compared to single-polarization radar observations, especially in the forward region. PR outperforms NPDI, even though the latter can suppress the effect of wind speed. The combination of dual-angular observations can significantly increase its sensitivity of oil thickness in the backward region but at the expense of reduced sensitivity in the forward region.

The Chinese Chang’E-7 (CE-7) mission is planned to land in the lunar south polar region, and then deploy a mini-flying probe to fly into the cold trap to detect the water ice. The selection of a landing site is crucial for ensuring both a safe landing and the successful achievement of its scientific objectives. This study presents a method for landing site selection in the challenging environment of the lunar south pole, utilizing multi-source remote sensing data. First, the likelihood of water ice in all cold traps within 85°S is assessed and prioritized using neutron spectrometer and hyperspectral data, with the most promising cold traps selected for sampling by CE-7’s mini-flying probe. Slope and illumination data are then used to screen feasible landing sites in the south polar region. Feasible landing sites near cold traps are aggregated into larger landing regions. Finally, high-resolution illumination maps, along with optical and radar images, are employed to refine the selection and identify the optimal landing sites. Six potential landing sites around the de Gerlache crater, an unnamed cold trap at (167.10°E, 88.71°S), Faustini crater, and Shackleton crater are proposed. It would be beneficial for CE-7 to prioritize mapping these sites post-launch using its high-resolution optical camera and radar for further detailed landing site investigation and evaluation.

In this study, the potential of electromagnetic scattering models to retrieve quantitative parameters of sea oil spills is investigated using an artificial intelligence-based approach. The backscattering coefficient of a slick-covered sea surface is predicted using the advanced integral equation model augmented with the model of local balance, an effective dielectric constant model and a composite medium model to include the effect of an oil slick. Damping ratios, predicted for different oil parameters (namely the oil thickness and seawater volume fraction), are used to train and test a four-layer neural network. Once successfully tested, the neural network is applied to an uninhabited aerial vehicle synthetic aperture radar image collected during the Deep-Water Horizon oil spill accident to retrieve the oil slick thickness and volume fraction of seawater in the oil layer. The inversion results show that the thicker (i. e., 2 – 4 mm) emulsions are located in the south and west of the slick and they are surrounded by thinner (i. e., < 1 mm) oil films. In addition, the seawater volume fraction in the oil slick is found to be about 20% - 30%. Results are contrasted with optical data and previous studies of the same accidental oil spill, showing qualitatively good agreement.

Deep neural networks (DNNs) have achieved great success in various computer vision tasks. However, they are susceptible to artificially designed adversarial perturbations, which limit their deployment in security-critical applications. In this paper, we propose a gradual adversarial training (GAT) method for remote sensing image segmentation. Our method incorporates a domain-adaptive mechanism that dynamically modulates input data, effectively reducing adversarial perturbations. GAT not only improves segmentation accuracy on clean images but also significantly enhances robustness against adversarial attacks, all without necessitating changes to the network architecture. The experimental results demonstrate that GAT consistently outperforms conventional standard adversarial training (SAT), showing increased resilience to adversarial attacks of varying intensities on both optical and Synthetic Aperture Radar (SAR) images. Compared to the SAT defense method, GAT achieves a notable defense performance improvement of 1% to 12%.