Satellite Synthetic Aperture Radar Images Research Articles

Comparison of a massive and diverse collection of ensembles and other classifiers for oil spill detection in SAR satellite images

We present a comparison of the largest collection of classifiers considered until now in the literature, composed by 428 methods belonging to 41 very different families. This collection, much larger than the one in our previous work (Fernandez-Delgado et al. in J Mach Learn Res 15:3133–3181, 2014), includes 320 ensembles (varying the base and meta-classifiers), alongside with Support Vector Machines, Bayesian, Neural Networks, Discriminant Analysis, Multivariate Adaptive Regression Splines, Random Forests, Decision Trees and many others. The classifier comparison is developed on the detection of oil spills on Synthetic Aperture Radar (SAR) images taken from satellites. The SAR images have revealed very useful to surveillance maritime agencies for the detection of regular offshore operational discharges, which, despite is commonly accepted, is one of the biggest causes of hydrocarbon marine pollution, instead of tanker and oil platform catastrophes. After a segmentation of the SAR images to select oil spill candidates, classifiers use the features extracted from these candidates to discard frequent and expensives look-alikes (false positives), caused by natural phenomena. Testing experiments revealed that the RotationForest ensemble of MultilayerPerceptron base classifiers, applying Kernel PCA on the original data, achieves the best accuracy and Cohen \(\kappa\) (87.1 % and 71.0 %, respectively) with a low frequency of false positives (5.13 %).

Mapping landslide phenomena in landlocked developing countries by means of satellite remote sensing data: the case of Dilijan (Armenia) area

ABSTRACTLandslide detection and mapping are essential issues for reducing impact of such natural disasters, and for improving the future built-up expansion and planning strategies, especially in developing countries where a reasonable land-use design is an important concern for sustainable growth and environmental management. Armenia is a landlocked country and its urban development is strongly tied to the improvement of infrastructures, which must takes into account the environmental setting and the slope instability of the area, in order to identify risks and possible damages to settlements and economic activities. The use of satellite-based Earth Observation data has advanced significantly in the last decade and has turned out to be very useful for measuring and monitoring slow-moving surface deformation phenomena with millimetric precision. In this framework, this study aims at providing a remote sensing-based Landslide Inventory Map (LIM) and a Landslide Susceptibility Map (LSM) over Dilijan (Armenia) area, performed within the Secondary Cities Urban Development in Armenia project. In particular, LIM and LSM in the study area were produced by using ground deformation measurements derived from satellite Synthetic Aperture Radar (SAR) data, acquired by ALOS and ENVISAT sensors from 2003 up to 2010, and integrated with photo-interpretation of recent optical images and morphological analysis of Digital Elevation Model (DEM). Given the extensive presence of vegetation in the area of interest, satellite SAR images were processed to produce both SqueeSAR™ and Temporary Coherent Scatterers data, which are PSI (Persistent Scatterer Interferometry) data conceived as evolution of PSInSAR™ approach and particularly suited for non-urban and rural areas characterized by low density of coherent terrain benchmarks over time. Landslide mapping produced through this work identifies the most hazardous landslide-affected and landslide-prone areas around Dilijan city, and can be used for further estimating environmental risks for urban infrastructure development in the area.

Origins and features of oil slicks in the Bohai Sea detected from satellite SAR images

Oil slicks were detected using satellite Synthetic Aperture Radar (SAR) images in 2011. We investigated potential origins and regional and seasonal features of oil slick in the Bohai Sea. Distance between oil slicks and potential origins (ships, seaports, and oil exploitation platforms) and the angle at which oil slicks move relative to potential driving forces were evaluated. Most oil slicks were detected along main ship routes rather than around seaports and oil exploitation platforms. Few oil slicks were detected within 20km of seaports. Directions of oil slicks movement were much more strongly correlated with directions of ship routes than with directions of winds and currents. These findings support the premise that oil slicks in the Bohai Sea most likely originate from illegal disposal of oil-polluted wastes from ships. Seasonal variation of oil slicks followed an annual cycle, with a peak in August and a trough in December.

Building Damage Assessment Using High-Resolution Satellite SAR Images of the 2010 Haiti Earthquake

Damage to individual buildings in an urban area of Port-au-Prince, Haiti, from the 2010 Haiti earthquake was assessed by means of high-resolution synthetic aperture radar (SAR) intensity images and ancillary building footprints. A comparison of pre- and post-event images and a building damage inventory showed that backscattering intensity between images was more significantly changed in collapsed buildings than in less damaged buildings. The linear discriminant function, based on the difference and correlation coefficient of the images was developed to detect collapsed buildings. The result showed that almost 75% of the buildings were correctly detected by discriminant analysis. An accuracy assessment revealed the difficulty of detecting small and congested buildings because the number of image pixels was too small and the buildings were obscured by neighboring buildings and other features in the images.

Natural and unnatural oil slicks in the Gulf of Mexico

When wind speeds are 2–10 m s−1, reflective contrasts in the ocean surface make oil slicks visible to synthetic aperture radar (SAR) under all sky conditions. Neural network analysis of satellite SAR images quantified the magnitude and distribution of surface oil in the Gulf of Mexico from persistent, natural seeps and from the Deepwater Horizon (DWH) discharge. This analysis identified 914 natural oil seep zones across the entire Gulf of Mexico in pre‐2010 data. Their ∼0.1 µm slicks covered an aggregated average of 775 km2. Assuming an average volume of 77.5 m3 over an 8–24 h lifespan per oil slick, the floating oil indicates a surface flux of 2.5–9.4 × 104 m3 yr−1. Oil from natural slicks was regionally concentrated: 68%, 25%, 7%, and <1% of the total was observed in the NW, SW, NE, and SE Gulf, respectively. This reflects differences in basin history and hydrocarbon generation. SAR images from 2010 showed that the 87 day DWH discharge produced a surface‐oil footprint fundamentally different from background seepage, with an average ocean area of 11,200 km2 (SD 5028) and a volume of 22,600 m3 (SD 5411). Peak magnitudes of oil were detected during equivalent, ∼14 day intervals around 23 May and 18 June, when wind speeds remained <5 m s−1. Over this interval, aggregated volume of floating oil decreased by 21%; area covered increased by 49% (p < 0.1), potentially altering its ecological impact. The most likely causes were increased applications of dispersant and surface burning operations.

Updated landslide inventory of the area between the Furiano and Rosmarino creeks (Sicily, Italy)

ABSTRACTA 1:10,000 scale landslide inventory map has been prepared for the area between the Furiano and Rosmarino creeks, in the Nebrodi Mountains (north-eastern Sicily, Italy), a territory highly prone to slope failures, due to the local geological and geomorphological settings and intense rainfall. The landslide inventory database included within the Hydrogeological Setting Plan of the Sicily Region has been used as a starting point for this work. The updated inventory map has been compiled through a combination of conventional approaches (i.e. aerial photo-interpretation and field surveys) and new remote sensing techniques (ground deformation measurements obtained by interferometric analysis of satellite Synthetic Aperture Radar images). The new landslide inventory consists of 566 events, classified according to their typology and state of activity.

Exploitation of Amplitude and Phase of Satellite SAR Images for Landslide Mapping: The Case of Montescaglioso (South Italy)

Pre- event and event landslide deformations have been detected and measured for the landslide that occurred on 3 December 2013 on the south-western slope of the Montescaglioso village (Basilicata Region, southern Italy). In this paper, ground displacements have been mapped through an integrated analysis based on a series of high resolution SAR (Synthetic Aperture Radar) images acquired by the Italian constellation of satellites COSMO-SkyMed. Analysis has been performed by exploiting both phase (through multi-image SAR interferometry) and amplitude information (through speckle tracking techniques) of the satellite images. SAR Interferometry, applied to images taken before the event, revealed a general pre-event movement, in the order of a few mm/yr, in the south-western slope of the Montescaglioso village. Highest pre-event velocities, ranging between 8 and 12 mm/yr, have been recorded in the sector of the slope where the first movement of the landslide took place. Speckle tracking, applied to images acquired before and after the event, allowed the retrieval of the 3D deformation field produced by the landslide. It also showed that ground displacements produced by the landslide have a dominant SSW component, with values exceeding 10 m for large sectors of the landslide area, with local peaks of 20 m in its central and deposit areas. Two minor landslides with a dominant SSE direction, which were detected in the upper parts of the slope, likely also occurred as secondary phenomena as consequence of the SSW movement of the main Montescaglioso landslide.

Analysis of a landslide multi-date inventory in a complex mountain landscape: the Ubaye valley case study

Abstract. In the paper we analyse a multi-date landslide inventory prepared for a mountainous area affected by several landslide types with different degrees of activity, we attempt to quantify the uncertainties associated to the mapping, we measure the evolution of morphological indicators and estimate landslide activity and temporal occurrence. The inventory, covering the period 1956–2010, is prepared for the middle section of the Ubaye valley (southern French Alps) based on the analysis of multi-source documents (geomorphological maps, historical reports of landslide events, field surveys, orthophotographs and SAR (synthetic aperture radar) satellite images). The uncertainties derived from the expert interpretation of different sources of information, the landslide morphological features and the affected land covers are taken into account in relation to the source documents. Morphological indicators are calculated to describe quantitatively the evolution of the landslides (length, area, relative elevation, runout distance). Frequency–area density functions are computed to estimate the changes in the landslide distributions and a Poisson model is used to estimate the probability of reactivation of the observed landslides and the occurrence of new failures. The proposed multi-date inventory and the associated statistics provide additional information to the event catalogue managed by the local policy makers.

Disintegration and acceleration of Thwaites Ice Shelf on the Amundsen Sea revealed from remote sensing measurements

Thwaites Ice Shelf in the Amundsen Sea is one of the biggest ice shelves in West Antarctica and is well known for significant mass changes. In the shear zone between Thwaites Glacier Tongue and its eastern ice shelf, shear stress forced by different flow rates of the ice shelves is causing the ice to break apart. A time series analysis of remote sensing data obtained by Landsat 7 Enhanced Thematic Mapper Plus (ETM+), TerraSAR-X, and airborne synthetic aperture radar (SAR) revealed that the shear zone has extended since 2006 and eventually disintegrated in 2008. We quantified the acceleration of Thwaites Ice Shelf with time by using the feature tracking method. The buttressing loss induced by the extension of the shear zone and progressive disintegration accelerated the flow of Thwaites Glacier Tongue, which in turn increased the shear stress on its eastern ice shelf. We determine causes of disintegration in the newly formed shear zone to be oceanic basal melting and structural weakening induced by Circumpolar Deep Water intrusion beneath the eastern ice shelf since 2000. The structural weakening was examined by using the density distribution of rifts and crevasses on the ice shelf, which were well identified from high-resolution SAR and optical satellite images.

Development of Methods for Detection and Monitoring of Fire Disturbance in the Alaskan Tundra Using a Two-Decade Long Record of Synthetic Aperture Radar Satellite Images

Using the extensive archive of historical ERS-1 and -2 synthetic aperture radar (SAR) images, this analysis demonstrates that fire disturbance can be effectively detected and monitored in high northern latitudes using radar technology. A total of 392 SAR images from May to August spanning 1992–2010 were analyzed from three study fires in the Alaskan tundra. The investigated fires included the 2007 Anaktuvuk River Fire and the 1993 DCKN178 Fire on the North Slope of Alaska and the 1999 Uvgoon Creek Fire in the Noatak National Preserve. A 3 dB difference was found between burned and unburned tundra, with the best time for burned area detection being as late in the growing season as possible before frozen ground conditions develop. This corresponds to mid-August for the study fires. In contrast to electro-optical studies from the same region, measures of landscape recovery as detected by the SAR were on the order of four to five years instead of one.

Persistent Scatterer Pair Interferometry: Approach and Application to COSMO-SkyMed SAR Data

Persistent scatterer interferometry is a widely used technique to detect and monitor slow terrain movements, with millimetric accuracy, from satellite synthetic aperture radar (SAR) data. We have recently proposed a method, named persistent scatterer pair (PSP), aimed at overcoming some limitations of standard techniques. The PSP method is characterized by the fact of exploiting only the relative properties of neighboring pairs of points for both detection and analysis of persistent scatterers (PSs), intended in the general sense of scatterers that exhibit interferometric coherence for the time period and baseline span of the acquisitions, including both point-like and distributed scatterers. Thanks to the pair-of-point approach, the PSP technique is intrinsically not affected by artifacts slowly variable in space, like those depending on atmosphere or orbits. Moreover, by exploiting a very redundant set of pair-of-point connections, the PSP approach guarantees extremely dense and accurate displacement and elevation measurements, both in correspondence of structures and when the backscattering is weak or distributed as in the case of natural terrains. In all cases, the measurements keep the full resolution of the input SAR images. In this work, the qualifying characteristics of the PSP technique are described, and several application examples and validation tests based on COSMO-SkyMed data are reported, which demonstrate the validity of the proposed approach.

Combined Use of Multi-Temporal Optical and Radar Satellite Images for Grassland Monitoring

The aim of this study was to assess the ability of optical images, SAR (Synthetic Aperture Radar) images and the combination of both types of data to discriminate between grasslands and crops in agricultural areas where cloud cover is very high most of the time, which restricts the use of visible and near-infrared satellite data. We compared the performances of variables extracted from four optical and five SAR satellite images with high/very high spatial resolutions acquired during the growing season. A vegetation index, namely the NDVI (Normalized Difference Vegetation Index), and two biophysical variables, the LAI (Leaf Area Index) and the fCOVER (fraction of Vegetation Cover) were computed using optical time series and polarization (HH, VV, HV, VH). The polarization ratio and polarimetric decomposition (Freeman–Durden and Cloude–Pottier) were calculated using SAR time series. Then, variables derived from optical, SAR and both types of remotely-sensed data were successively classified using the Support Vector Machine (SVM) technique. The results show that the classification accuracy of SAR variables is higher than those using optical data (0.98 compared to 0.81). They also highlight that the combination of optical and SAR time series data is of prime interest to discriminate grasslands from crops, allowing an improved classification accuracy.

Shoreline movement monitoring based on SAR images in Shanghai, China

In this article, a divide-and-combine approach and multiscale normalized cut techniques are combined to extract water–land boundaries from synthetic aperture radar (SAR) images. A validation study shows that waterline extraction from SAR using the proposed algorithm is consistent with that derived from Landsat optical remote-sensing imagery. Analysis of six European Remote Sensing-1/2 (ERS-1/2) satellite and Envisat SAR images acquired at spring tides in Lingang New City, a new land-reclamation city in the suburb of Shanghai, China, between 1993 and 2005 shows the shoreline movement of the study region. We find that the shoreline had minimum seaward movement in the region due to natural forces between 1993 and 1996. Between 1996 and 1999, the shoreline moved more in the eastern and slower in the western part due to artificial impelling siltation and deposition of silt in the Yangtze River. Later, between 1999 and 2005, the shoreline significantly changed following commencement of a tidal flat reclamation project.

Change detection of boreal forest using bi-temporal ALOS PALSAR backscatter data

Long-wavelength Synthetic Aperture Radar (SAR) satellite systems have the potential to increase the efficiency of forest mapping and monitoring, which today are based primarily on optical satellite systems. Here, we evaluate the effectiveness of using L-band SAR satellite images to detect and delineate clear-cuts in Swedish boreal forest. A set of computationally efficient techniques are combined in a fully automated unsupervised bi-temporal change detection approach that detects changes in SAR backscatter intensities. For radiometric normalization and initial change classification, we propose an iterative procedure consisting of successive polynomial based histogram matching and thresholding. Recently proposed methods for automatic SAR amplitude ratio thresholding and final change classification are adopted. The latter is a Markov random field based change detection method that exploits both spectral and spatial information from one or multiple SAR polarization channels. The change detection approach was applied to SAR images from the Japanese Advanced Land Observing Satellite (ALOS) Phased Array type L-band Synthetic Aperture Radar (PALSAR) acquired in Fine Beam Dual (FBD) mode (HH- and HV-polarizations) with a pixel size of 20m (path data). Clear-cuts that took place between image acquisitions were clearly detected, and most errors were due to imperfect delineations of clear-cut edges. Pixel-wise clear-cut detection accuracies above 90% could be reached, with false alarm rates of approximately 10% or less. The results indicate that ALOS PALSAR path data are well suited for operational clear-cut detection in Swedish boreal forest.

Monitoring Changes in Rice Cultivated Area from SAR and Optical Satellite Images in Ben Tre and Tra Vinh Provinces in Mekong Delta, Vietnam

The objective of this study was to obtain up-to-date information on land use and to identify long term changes in land use, especially rice, aquaculture and other crops in Ben Tre and Tra Vinh provinces in Vietnam’s Mekong Delta. Long-term changes in land-use of the study area have not been studied using long time series of SAR and optical Earth observation (EO) data before. EO data from 1979–2012 was used: ENVISAT ASAR Wide Swath Mode, SPOT and Landsat imagery. An unsupervised ISODATA classification was performed on multitemporal SAR images. The results were validated using ground truth data. Using the Synthetic Aperture Radar (SAR) imagery maps for 2005, 2009 and 2011 were obtained. Different rice crops, aquaculture and fruit trees could be distinguished with an overall accuracy of 80%. Using available optical imagery the time series was extended from 2005 to 1979. Long-term decrease in the rice acreage and increase in the aquaculture acreage could be detected.

A texture-based fusion scheme to integrate high-resolution satellite SAR and optical images

Image fusion of different types of satellite images is of widespread interest due to the increasing availability of various spaceborne imaging sensors. In this letter, a new multi-sensor image fusion method is presented, where a texture-based fusion rule is used to address the fusion of synthetic aperture radar (SAR) and optical images. At first, the integration of panchromatic (PAN) and SAR image is achieved in wavelet domain by taking weighted average of the corresponding input pixels. The weights are adaptively decided according to the local spatial autocorrelation characteristics of SAR texture image. This integrated image is fused into the multispectral (MS) image through generalized intensity-hue-saturation (GIHS) image fusion method to obtain the final fused image. To quantitatively evaluate and test the performance of the proposed method, comparisons with some existing fusion methods are carried out in this letter. Experimental results demonstrate that the proposed method shows better performance than other fusion methods and can get satisfactory fusion results.

Application of Bimodal Histogram Method to Oil Spill Detection from a Satellite Synthetic Aperture Radar Image

As one of segmentation techniques for Synthetic Aperture Radar (SAR) image with oil spill, we applied a bimodal histogram method to discriminate oil pixels from non-oil pixels. The threshold of each moving window was objectively determined using the two peaks in the histogram distribution of backscattering coefficients from ENVISAT ASAR image. To reduce the effect of wind speed on oil spill detection, we selected ASAR image which satisfied a limit of wind speeds for successful detection. Overall, a commonly used adaptive threshold method has been applied with a subjectively-determined single threshold. In contrast, the bimodal histogram method utilized herein produces a variety of thresholds objectively for each moving window by considering the characteristics of statistical distribution of backscattering coefficients. Comparison between the two methods revealed that the bimodal histogram method exhibited no significant difference in terms of performance when compared to the adaptive threshold method, except for around the edges of dark oil spots. Thus, we anticipate that the objective method based on the bimodality of oil slicks may also be applicable to the detection of oil spills from other SAR imagery.

Oil seeps in the southeastern Black Sea studied using satellite synthetic aperture radar images

One of the tasks being completed while oil pollution monitoring in the Black Sea, was a study of natural seepage. Two known oil seeps are located in the southeastern section, the Georgian sector, from which crude oil enters the marine environment. The archives of European Space Agency and ScanEx Research and Development Center have a considerable quantity of synthetic aperture radar (SAR) images acquired in 1993–2011 by the ERS-1/2, Envisat, and Radarsat-1 satellites, on which seep oil slicks are clearly visible in this region. Processing of the collected SAR images with detected slicks in combination with the geoinformation approach has revealed a link of these slicks with the bottom sources on the local sedimentary structure in the Southeast Black Sea; their analysis provides both new insight into this phenomenon and new information to help understand nature of these oil seeps. On the basis of an analysis of collected SAR images and detailed bathymetric data, information on the source positions on the bottom and estimates of oil volumes entering the sea surface are obtained.

Natural oil seepage at Kobuleti Ridge, eastern Black Sea

Analysis of Advanced Synthetic Aperture Radar satellite images in combination with water column and seafloor investigations documented natural oil seepage from Pechori Mound and Colkheti Seep in 1000–1200 m water depth in the eastern Black Sea offshore Georgia. Hydroacoustic imaging of the water column using multibeam echosounder evidenced numerous gas emissions from both structures. Gas bubbles rose as high as 45 m below sea surface. It is proposed that oil coatings around gas bubbles hamper their dissolution allowing them to reach the sea surface where widespread oil slicks are formed. Slow rise velocities (14 ± 1 cm s−1) of bubbles with radii of 2.6 ± 0.3 mm (n = 101) derived from video data obtained with the remotely operated vehicle MARUM QUEST4000, corroborate the assumption that bubbles are oil-coated. High resolution seafloor mapping with the autonomous underwater vehicle MARUM SEAL5000 at Colkheti Seep revealed a crater-strewn morphology whose formation is explained by frequent rafting of shallow gas hydrate deposits. Satellite imaging of oil slicks on the sea surface above both sites indicates that oil seepage is rather persistent since 2003. An order-of-magnitude estimation of minimum oil seepage rates suggests discharge rates of ∼40 liters per hour from both sites. The data presented are the first comprehensive description of oil seepage in the Black Sea.

MEDSLIK-II, a Lagrangian marine surface oil spill model for short-term forecasting – Part 2: Numerical simulations and validations

Abstract. In this paper we use MEDSLIK-II, a Lagrangian marine surface oil spill model described in Part 1 (De Dominicis et al., 2013), to simulate oil slick transport and transformation processes for realistic oceanic cases, where satellite or drifting buoys data are available for verification. The model is coupled with operational oceanographic currents, atmospheric analyses winds and remote sensing data for initialization. The sensitivity of the oil spill simulations to several model parameterizations is analyzed and the results are validated using surface drifters, SAR (synthetic aperture radar) and optical satellite images in different regions of the Mediterranean Sea. It is found that the forecast skill of Lagrangian trajectories largely depends on the accuracy of the Eulerian ocean currents: the operational models give useful estimates of currents, but high-frequency (hourly) and high-spatial resolution is required, and the Stokes drift velocity has to be added, especially in coastal areas. From a numerical point of view, it is found that a realistic oil concentration reconstruction is obtained using an oil tracer grid resolution of about 100 m, with at least 100 000 Lagrangian particles. Moreover, sensitivity experiments to uncertain model parameters show that the knowledge of oil type and slick thickness are, among all the others, key model parameters affecting the simulation results. Considering acceptable for the simulated trajectories a maximum spatial error of the order of three times the horizontal resolution of the Eulerian ocean currents, the predictability skill for particle trajectories is from 1 to 2.5 days depending on the specific current regime. This suggests that re-initialization of the simulations is required every day.