Maritime Surveillance 1

INCOIS issuing the oil spill advisory for Indian waters since 2011, during oil spill incidents, mock drills, contingency planning of various offshore industries and during severe meteorological conditions. The oil spill trajectory prediction system includes the trajectory prediction and validation as its major components. The execution includes, the feed of the details of the spill such as date, time, location, quantity and type of oil spilled. Forecasted wind and current forcings are the driving parameters. The validation of the predicted trajectory was made utilizing the datasets from Envisat, Radarsat, Sentinel- 1, and Landsat. Execution and validation of various case studies since 2010 were elaborately discussed in this paper. The deviation analysis was carried out and the performance of the system was improved by many folds. Thus this paper explains the significance of the earth observation satellites in validating the predicted trajectory, thereby improving the performance of the system. As a growth of which, an operational version of  Online Oil Spill Advisory system  was launched during April 2015 and widely used all over the nation.

Operational detection and characterization of oil spills over oceans have received considerable attention due to their impact on marine ecosystem from an environmental and political point of view. Space-borne Synthetic Aperture Radar (SAR) has been used as a major tool for operational oil spill detection activity and has been attracting significant research interest. Exploitation of SAR polarimetric features for oil spill detection is relatively new and those properties have not been used for operational services until now. In the last decade, a number of semi-automatic and automatic techniques have been proposed in order to differentiate oil spill and ‘look-alike’ spots based on single pol SAR imagery. However these techniques suffer from a high miss-classification rate which is undesirable for operational services. In addition to that, small operational spillages from offshore platforms are often ignored as it appears insignificant on traditional ‘ScanSAR’ (wide) images. In order to mitigate this situation a major focus of research in this area is the development of semi-automated/automated algorithms based on polarimetric images to distinguish oil spills from ‘look-alikes’. This paper describes the development of an automated Near Real Time (NRT) oil spill detection processing chain based on quad-pol RADARSAT-2 (RS-2) and quad-pol TerraSAR-X (TS-X) images, wherein we use polarimetric features (e.g. Lexicographic and Pauli Based features) to characterize oil spills and ‘look-alikes’. Numbers of TS-X images have been acquired over known offshore platforms along with near coincident (spatially and temporally) RS-2 acquisition. A total number of 10 polarimetric feature parameters were extracted from different types of oil (e.g. crude oil, emulsion etc) and look-alike spots and divided into training and validation dataset. Extracted features along with their second order statistics were then used for training and validation of a pixel based Artificial Neural Network (ANN) classifier. Initial performance estimation was carried out for the proposed methodology in order to evaluate its suitability for NRT operational service. Mutual information contents among extracted features were assessed and feature parameters were ranked according to their ability to discriminate between oil spill and look- alike. Polarimetric features such as Scattering Diversity proved to be more discriminative than other polarimetric features.

Marine slicks is one of the most common features on the sea surface at low to moderate wind speeds. A significant part of marine slicks is due to surface films of anthropogenic origin that appear on the water surface as a result of accidental or deliberate oil spills. The slick geometry is an important characteristic that can be potentially used to identify, e.g., oil spills in radar or optical images of the sea surface. The shape of oil slicks depends on different factors determining film spreading, which are surface and interfacial tension coefficients, near surface currents and wind waves. Theory of oil slick spreading under the action of surface tension forces  was developed, e.g. by Fay (1969). According to this theory oil spills instantly poured on an unbounded water surface are circular with the radius growing with time as t^3/4 . However, it is known from everyday observations and satellite data, that oil spills are usually stretched along the wind. Currently, however, there is a lack of data of systematic experiments with slicks, and the very physical mechanisms of slick spreading are still not well understood.

This paper presents results of controlled experiments with oil spills, and a possible physical mechanism of the slick asymmetry is discussed.

Field studies of the oil spills evolution were carried out in different water bodies (on theBlack seaand on the Gorky Water Reservoir) and at different environmental conditions. Oleic acid was used to deploy films on the water surface. Characteristics of oleic acid films, in particular, the surface tension coefficient as functions of surfactant concentration were studied in laboratory conditions before the experiments. Different methodologies were used to study oil spreading, namely: a) photographic observations from a high coast of oil slicks poured from a Platform, b) observations of oil spills with X-band marine radar mounted on a lighthouse, c) satellite observations using radar imagery of Envisat ASAR and TerraSAR-X. In the satellite experiments several slicks were prepared consecutively when spilling a given volume of oil at predetermined times before the satellite overpass.

It is obtained that oil slicks are roughly elliptical, the long axis grows with time approximately as in Fay’s theory, while the short axis grows slower. As a result, accordingly, a long-to-short axis ratio for the oil spills slowly increases with spreading time, and depends on wind velocity. A new effect of slick compression in the cross wind direction at large stages of slick evolution was revealed.

A physical mechanism of slick deformation due to mean surface currents induced by surface wind waves is proposed. Namely, surface stresses associated with mean currents induced by oblique propagating surface waves increase in the area of slicks due to enhanced wave damping thus resulting in film compression in the cross wind direction. The currents at upwind and downwind sides contribute to the slick drift. Theoretical estimates of slick compression are consistent with experiment.                

The work was supported by RFBR (Project 14-05-00876, № 15-35-20992, 15-45-02650), and the Program RAS ”Radiophysics”.

Synthetic aperture radars (SARs) are frequently applied for oil spill detection because they yield images independent of the time of the day and of cloud coverage. However, identifying mineral oil in SAR images of the sea surface is not straightforward. Often black features visible on SAR images, which are believed to be radar signatures of mineral oil, turn out to be radar signatures of “oil spill look-alikes”. The reason is that mineral oil floating on the sea surface becomes visible on SAR images as dark areas, but dark areas can also originate from 1) natural (biogenic) surface films which are produced by plankton or fish, 2) low winds which are often encountered in the lee of islands or coastal mountains, 2) cold upwelling water which changes the stability of the air-sea interface, 3) rain cells, 4) divergent flow regimes associated, e.g., with internal waves or tidal flow over sand banks, 5) river outflows, 6) dry fallen sand banks, 7) turbulent water as encountered in ship wakes, and 8) grease ice. The main challenge in developing oil spill detection algorithms using SAR images is to discriminate between sea areas covered with mineral oil and biogenic surface films. In this paper we focus on this issue and compare Envisat ASAR images with Chl-a maps generated from MODIS data. Sea areas of enhanced biological activity are usually associated with biogenic surface films which damp the short surface waves responsible for the radar backscattering as strongly as mineral oil films Criteria for the discrimination are presented. Another issue discussed in this paper is the applicability of polarimetric SAR for discrimination between oil spills and biogenic surface films, which are only one molecular layer thick. We argue that this is only possible when the mineral oil film (or oil/water emulsion) is sufficiently thick (typically thicker than few millimeters) such that the radar can sense the dielectric constant of the oil (small) or the dielectric constant of the oil emulsion. But usually, the thickness of oil spills is quite small such that polarimetric SAR data does not help in the discrimination. Probably the findings of other investigators that polarimetric parameters, such as the mean scattering angle, can be used in the discrimination process, result from the poor signal-to-noise ratio of the SARs used in their investigations. This interpretation is supported by data acquired by the the Jet Propulsion Laboratory’s (JPL) Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR), which has an extremely high signal to noise ratio, during the Gulf Oil Spill Campaign in 2010-2012.

The deep water horizon blowout in the Gulf of Mexico resulted in one of the largest accidental oil disasters in U.S. history. More than 200 million gallons of oil emitted into the Gulf of Mexico and the petroleum hydrocarbons were released from the reservoir through the wellbore for 87 days causing an oil spill of national significance. The rig exploded off the Louisiana coast on April 20, 2010. The oil spill caused significant damage to the environment and to the marine habitats. The damages associated with the oil spill include oiled and dead wildlife, polluted marshes, and lifeless deep water corals. In response to the Deepwater Horizon Oil Spill disaster in Gulf of Mexico, NASA acquired several synthetic aperture radar (SAR) and hyperspectral imagery (HSI) and have made them available to the scientific research community for analyzing impacts of the oil spill.

In this study, we propose to use the L-band quad-polarized radar data acquired by Unmanned Aerial Vehicle Synthetic Aperture Radar (UAVSAR) and HSI from the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) optical sensor. The objective of this paper is to apply feature level fusion techniques on radar and hyperspectral imagery for detecting the vegetation stress on the shores of Wilkinson Bay, Louisiana as a result of Gulf of Mexico disaster.

Synthetic Aperture Radar technology, due to its high spatial resolution and penetration capability, is a good choice to identify problem areas with oil spill. The radar backscatter data is capable of identifying oil and delineating this from water on the ocean surface and also delineate dead / stressed vegetation from healthy vegetation. Hyperspectral imaging has been widely in use because of the wide spectral and spatial information capabilities, which enables discriminating between similar objects. Polarimetric and textural features extracted from SAR data will be fused with features selected based on the principal components from hyperspectral imagery will be applied as input to the Support Vector Machine classifier for mapping vegetation stress. The results will be presented with overall classification accuracies along with classification maps of the study area.